2019-05-19 15:08:55 +03:00
// SPDX-License-Identifier: GPL-2.0-only
2005-04-17 02:20:36 +04:00
/*
* linux / kernel / fork . c
*
* Copyright ( C ) 1991 , 1992 Linus Torvalds
*/
/*
* ' fork . c ' contains the help - routines for the ' fork ' system call
* ( see also entry . S and others ) .
* Fork is rather simple , once you get the hang of it , but the memory
* management can be a bitch . See ' mm / memory . c ' : ' copy_page_range ( ) '
*/
2019-03-27 15:04:15 +03:00
# include <linux/anon_inodes.h>
2005-04-17 02:20:36 +04:00
# include <linux/slab.h>
2017-02-08 20:51:29 +03:00
# include <linux/sched/autogroup.h>
2017-02-08 20:51:29 +03:00
# include <linux/sched/mm.h>
2017-02-08 20:51:30 +03:00
# include <linux/sched/coredump.h>
2017-02-08 20:51:30 +03:00
# include <linux/sched/user.h>
2017-02-08 20:51:31 +03:00
# include <linux/sched/numa_balancing.h>
2017-02-08 20:51:35 +03:00
# include <linux/sched/stat.h>
2017-02-08 20:51:36 +03:00
# include <linux/sched/task.h>
2017-02-08 20:51:37 +03:00
# include <linux/sched/task_stack.h>
2017-02-05 13:48:36 +03:00
# include <linux/sched/cputime.h>
2019-03-27 15:04:15 +03:00
# include <linux/seq_file.h>
2017-02-03 12:08:30 +03:00
# include <linux/rtmutex.h>
2005-04-17 02:20:36 +04:00
# include <linux/init.h>
# include <linux/unistd.h>
# include <linux/module.h>
# include <linux/vmalloc.h>
# include <linux/completion.h>
# include <linux/personality.h>
# include <linux/mempolicy.h>
# include <linux/sem.h>
# include <linux/file.h>
2008-04-24 15:44:08 +04:00
# include <linux/fdtable.h>
2008-06-30 22:42:08 +04:00
# include <linux/iocontext.h>
2005-04-17 02:20:36 +04:00
# include <linux/key.h>
# include <linux/binfmts.h>
# include <linux/mman.h>
mmu-notifiers: core
With KVM/GFP/XPMEM there isn't just the primary CPU MMU pointing to pages.
There are secondary MMUs (with secondary sptes and secondary tlbs) too.
sptes in the kvm case are shadow pagetables, but when I say spte in
mmu-notifier context, I mean "secondary pte". In GRU case there's no
actual secondary pte and there's only a secondary tlb because the GRU
secondary MMU has no knowledge about sptes and every secondary tlb miss
event in the MMU always generates a page fault that has to be resolved by
the CPU (this is not the case of KVM where the a secondary tlb miss will
walk sptes in hardware and it will refill the secondary tlb transparently
to software if the corresponding spte is present). The same way
zap_page_range has to invalidate the pte before freeing the page, the spte
(and secondary tlb) must also be invalidated before any page is freed and
reused.
Currently we take a page_count pin on every page mapped by sptes, but that
means the pages can't be swapped whenever they're mapped by any spte
because they're part of the guest working set. Furthermore a spte unmap
event can immediately lead to a page to be freed when the pin is released
(so requiring the same complex and relatively slow tlb_gather smp safe
logic we have in zap_page_range and that can be avoided completely if the
spte unmap event doesn't require an unpin of the page previously mapped in
the secondary MMU).
The mmu notifiers allow kvm/GRU/XPMEM to attach to the tsk->mm and know
when the VM is swapping or freeing or doing anything on the primary MMU so
that the secondary MMU code can drop sptes before the pages are freed,
avoiding all page pinning and allowing 100% reliable swapping of guest
physical address space. Furthermore it avoids the code that teardown the
mappings of the secondary MMU, to implement a logic like tlb_gather in
zap_page_range that would require many IPI to flush other cpu tlbs, for
each fixed number of spte unmapped.
To make an example: if what happens on the primary MMU is a protection
downgrade (from writeable to wrprotect) the secondary MMU mappings will be
invalidated, and the next secondary-mmu-page-fault will call
get_user_pages and trigger a do_wp_page through get_user_pages if it
called get_user_pages with write=1, and it'll re-establishing an updated
spte or secondary-tlb-mapping on the copied page. Or it will setup a
readonly spte or readonly tlb mapping if it's a guest-read, if it calls
get_user_pages with write=0. This is just an example.
This allows to map any page pointed by any pte (and in turn visible in the
primary CPU MMU), into a secondary MMU (be it a pure tlb like GRU, or an
full MMU with both sptes and secondary-tlb like the shadow-pagetable layer
with kvm), or a remote DMA in software like XPMEM (hence needing of
schedule in XPMEM code to send the invalidate to the remote node, while no
need to schedule in kvm/gru as it's an immediate event like invalidating
primary-mmu pte).
At least for KVM without this patch it's impossible to swap guests
reliably. And having this feature and removing the page pin allows
several other optimizations that simplify life considerably.
Dependencies:
1) mm_take_all_locks() to register the mmu notifier when the whole VM
isn't doing anything with "mm". This allows mmu notifier users to keep
track if the VM is in the middle of the invalidate_range_begin/end
critical section with an atomic counter incraese in range_begin and
decreased in range_end. No secondary MMU page fault is allowed to map
any spte or secondary tlb reference, while the VM is in the middle of
range_begin/end as any page returned by get_user_pages in that critical
section could later immediately be freed without any further
->invalidate_page notification (invalidate_range_begin/end works on
ranges and ->invalidate_page isn't called immediately before freeing
the page). To stop all page freeing and pagetable overwrites the
mmap_sem must be taken in write mode and all other anon_vma/i_mmap
locks must be taken too.
2) It'd be a waste to add branches in the VM if nobody could possibly
run KVM/GRU/XPMEM on the kernel, so mmu notifiers will only enabled if
CONFIG_KVM=m/y. In the current kernel kvm won't yet take advantage of
mmu notifiers, but this already allows to compile a KVM external module
against a kernel with mmu notifiers enabled and from the next pull from
kvm.git we'll start using them. And GRU/XPMEM will also be able to
continue the development by enabling KVM=m in their config, until they
submit all GRU/XPMEM GPLv2 code to the mainline kernel. Then they can
also enable MMU_NOTIFIERS in the same way KVM does it (even if KVM=n).
This guarantees nobody selects MMU_NOTIFIER=y if KVM and GRU and XPMEM
are all =n.
The mmu_notifier_register call can fail because mm_take_all_locks may be
interrupted by a signal and return -EINTR. Because mmu_notifier_reigster
is used when a driver startup, a failure can be gracefully handled. Here
an example of the change applied to kvm to register the mmu notifiers.
Usually when a driver startups other allocations are required anyway and
-ENOMEM failure paths exists already.
struct kvm *kvm_arch_create_vm(void)
{
struct kvm *kvm = kzalloc(sizeof(struct kvm), GFP_KERNEL);
+ int err;
if (!kvm)
return ERR_PTR(-ENOMEM);
INIT_LIST_HEAD(&kvm->arch.active_mmu_pages);
+ kvm->arch.mmu_notifier.ops = &kvm_mmu_notifier_ops;
+ err = mmu_notifier_register(&kvm->arch.mmu_notifier, current->mm);
+ if (err) {
+ kfree(kvm);
+ return ERR_PTR(err);
+ }
+
return kvm;
}
mmu_notifier_unregister returns void and it's reliable.
The patch also adds a few needed but missing includes that would prevent
kernel to compile after these changes on non-x86 archs (x86 didn't need
them by luck).
[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: fix mm/filemap_xip.c build]
[akpm@linux-foundation.org: fix mm/mmu_notifier.c build]
Signed-off-by: Andrea Arcangeli <andrea@qumranet.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Jack Steiner <steiner@sgi.com>
Cc: Robin Holt <holt@sgi.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Kanoj Sarcar <kanojsarcar@yahoo.com>
Cc: Roland Dreier <rdreier@cisco.com>
Cc: Steve Wise <swise@opengridcomputing.com>
Cc: Avi Kivity <avi@qumranet.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Anthony Liguori <aliguori@us.ibm.com>
Cc: Chris Wright <chrisw@redhat.com>
Cc: Marcelo Tosatti <marcelo@kvack.org>
Cc: Eric Dumazet <dada1@cosmosbay.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Izik Eidus <izike@qumranet.com>
Cc: Anthony Liguori <aliguori@us.ibm.com>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-29 02:46:29 +04:00
# include <linux/mmu_notifier.h>
2005-04-17 02:20:36 +04:00
# include <linux/fs.h>
mm: per-thread vma caching
This patch is a continuation of efforts trying to optimize find_vma(),
avoiding potentially expensive rbtree walks to locate a vma upon faults.
The original approach (https://lkml.org/lkml/2013/11/1/410), where the
largest vma was also cached, ended up being too specific and random,
thus further comparison with other approaches were needed. There are
two things to consider when dealing with this, the cache hit rate and
the latency of find_vma(). Improving the hit-rate does not necessarily
translate in finding the vma any faster, as the overhead of any fancy
caching schemes can be too high to consider.
We currently cache the last used vma for the whole address space, which
provides a nice optimization, reducing the total cycles in find_vma() by
up to 250%, for workloads with good locality. On the other hand, this
simple scheme is pretty much useless for workloads with poor locality.
Analyzing ebizzy runs shows that, no matter how many threads are
running, the mmap_cache hit rate is less than 2%, and in many situations
below 1%.
The proposed approach is to replace this scheme with a small per-thread
cache, maximizing hit rates at a very low maintenance cost.
Invalidations are performed by simply bumping up a 32-bit sequence
number. The only expensive operation is in the rare case of a seq
number overflow, where all caches that share the same address space are
flushed. Upon a miss, the proposed replacement policy is based on the
page number that contains the virtual address in question. Concretely,
the following results are seen on an 80 core, 8 socket x86-64 box:
1) System bootup: Most programs are single threaded, so the per-thread
scheme does improve ~50% hit rate by just adding a few more slots to
the cache.
+----------------+----------+------------------+
| caching scheme | hit-rate | cycles (billion) |
+----------------+----------+------------------+
| baseline | 50.61% | 19.90 |
| patched | 73.45% | 13.58 |
+----------------+----------+------------------+
2) Kernel build: This one is already pretty good with the current
approach as we're dealing with good locality.
+----------------+----------+------------------+
| caching scheme | hit-rate | cycles (billion) |
+----------------+----------+------------------+
| baseline | 75.28% | 11.03 |
| patched | 88.09% | 9.31 |
+----------------+----------+------------------+
3) Oracle 11g Data Mining (4k pages): Similar to the kernel build workload.
+----------------+----------+------------------+
| caching scheme | hit-rate | cycles (billion) |
+----------------+----------+------------------+
| baseline | 70.66% | 17.14 |
| patched | 91.15% | 12.57 |
+----------------+----------+------------------+
4) Ebizzy: There's a fair amount of variation from run to run, but this
approach always shows nearly perfect hit rates, while baseline is just
about non-existent. The amounts of cycles can fluctuate between
anywhere from ~60 to ~116 for the baseline scheme, but this approach
reduces it considerably. For instance, with 80 threads:
+----------------+----------+------------------+
| caching scheme | hit-rate | cycles (billion) |
+----------------+----------+------------------+
| baseline | 1.06% | 91.54 |
| patched | 99.97% | 14.18 |
+----------------+----------+------------------+
[akpm@linux-foundation.org: fix nommu build, per Davidlohr]
[akpm@linux-foundation.org: document vmacache_valid() logic]
[akpm@linux-foundation.org: attempt to untangle header files]
[akpm@linux-foundation.org: add vmacache_find() BUG_ON]
[hughd@google.com: add vmacache_valid_mm() (from Oleg)]
[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: adjust and enhance comments]
Signed-off-by: Davidlohr Bueso <davidlohr@hp.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Reviewed-by: Michel Lespinasse <walken@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Tested-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-08 02:37:25 +04:00
# include <linux/mm.h>
# include <linux/vmacache.h>
2006-10-02 13:18:06 +04:00
# include <linux/nsproxy.h>
2006-01-11 23:17:46 +03:00
# include <linux/capability.h>
2005-04-17 02:20:36 +04:00
# include <linux/cpu.h>
2007-10-19 10:39:33 +04:00
# include <linux/cgroup.h>
2005-04-17 02:20:36 +04:00
# include <linux/security.h>
2008-07-24 08:27:23 +04:00
# include <linux/hugetlb.h>
seccomp: add system call filtering using BPF
[This patch depends on luto@mit.edu's no_new_privs patch:
https://lkml.org/lkml/2012/1/30/264
The whole series including Andrew's patches can be found here:
https://github.com/redpig/linux/tree/seccomp
Complete diff here:
https://github.com/redpig/linux/compare/1dc65fed...seccomp
]
This patch adds support for seccomp mode 2. Mode 2 introduces the
ability for unprivileged processes to install system call filtering
policy expressed in terms of a Berkeley Packet Filter (BPF) program.
This program will be evaluated in the kernel for each system call
the task makes and computes a result based on data in the format
of struct seccomp_data.
A filter program may be installed by calling:
struct sock_fprog fprog = { ... };
...
prctl(PR_SET_SECCOMP, SECCOMP_MODE_FILTER, &fprog);
The return value of the filter program determines if the system call is
allowed to proceed or denied. If the first filter program installed
allows prctl(2) calls, then the above call may be made repeatedly
by a task to further reduce its access to the kernel. All attached
programs must be evaluated before a system call will be allowed to
proceed.
Filter programs will be inherited across fork/clone and execve.
However, if the task attaching the filter is unprivileged
(!CAP_SYS_ADMIN) the no_new_privs bit will be set on the task. This
ensures that unprivileged tasks cannot attach filters that affect
privileged tasks (e.g., setuid binary).
There are a number of benefits to this approach. A few of which are
as follows:
- BPF has been exposed to userland for a long time
- BPF optimization (and JIT'ing) are well understood
- Userland already knows its ABI: system call numbers and desired
arguments
- No time-of-check-time-of-use vulnerable data accesses are possible.
- system call arguments are loaded on access only to minimize copying
required for system call policy decisions.
Mode 2 support is restricted to architectures that enable
HAVE_ARCH_SECCOMP_FILTER. In this patch, the primary dependency is on
syscall_get_arguments(). The full desired scope of this feature will
add a few minor additional requirements expressed later in this series.
Based on discussion, SECCOMP_RET_ERRNO and SECCOMP_RET_TRACE seem to be
the desired additional functionality.
No architectures are enabled in this patch.
Signed-off-by: Will Drewry <wad@chromium.org>
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Reviewed-by: Indan Zupancic <indan@nul.nu>
Acked-by: Eric Paris <eparis@redhat.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
v18: - rebase to v3.4-rc2
- s/chk/check/ (akpm@linux-foundation.org,jmorris@namei.org)
- allocate with GFP_KERNEL|__GFP_NOWARN (indan@nul.nu)
- add a comment for get_u32 regarding endianness (akpm@)
- fix other typos, style mistakes (akpm@)
- added acked-by
v17: - properly guard seccomp filter needed headers (leann@ubuntu.com)
- tighten return mask to 0x7fff0000
v16: - no change
v15: - add a 4 instr penalty when counting a path to account for seccomp_filter
size (indan@nul.nu)
- drop the max insns to 256KB (indan@nul.nu)
- return ENOMEM if the max insns limit has been hit (indan@nul.nu)
- move IP checks after args (indan@nul.nu)
- drop !user_filter check (indan@nul.nu)
- only allow explicit bpf codes (indan@nul.nu)
- exit_code -> exit_sig
v14: - put/get_seccomp_filter takes struct task_struct
(indan@nul.nu,keescook@chromium.org)
- adds seccomp_chk_filter and drops general bpf_run/chk_filter user
- add seccomp_bpf_load for use by net/core/filter.c
- lower max per-process/per-hierarchy: 1MB
- moved nnp/capability check prior to allocation
(all of the above: indan@nul.nu)
v13: - rebase on to 88ebdda6159ffc15699f204c33feb3e431bf9bdc
v12: - added a maximum instruction count per path (indan@nul.nu,oleg@redhat.com)
- removed copy_seccomp (keescook@chromium.org,indan@nul.nu)
- reworded the prctl_set_seccomp comment (indan@nul.nu)
v11: - reorder struct seccomp_data to allow future args expansion (hpa@zytor.com)
- style clean up, @compat dropped, compat_sock_fprog32 (indan@nul.nu)
- do_exit(SIGSYS) (keescook@chromium.org, luto@mit.edu)
- pare down Kconfig doc reference.
- extra comment clean up
v10: - seccomp_data has changed again to be more aesthetically pleasing
(hpa@zytor.com)
- calling convention is noted in a new u32 field using syscall_get_arch.
This allows for cross-calling convention tasks to use seccomp filters.
(hpa@zytor.com)
- lots of clean up (thanks, Indan!)
v9: - n/a
v8: - use bpf_chk_filter, bpf_run_filter. update load_fns
- Lots of fixes courtesy of indan@nul.nu:
-- fix up load behavior, compat fixups, and merge alloc code,
-- renamed pc and dropped __packed, use bool compat.
-- Added a hidden CONFIG_SECCOMP_FILTER to synthesize non-arch
dependencies
v7: (massive overhaul thanks to Indan, others)
- added CONFIG_HAVE_ARCH_SECCOMP_FILTER
- merged into seccomp.c
- minimal seccomp_filter.h
- no config option (part of seccomp)
- no new prctl
- doesn't break seccomp on systems without asm/syscall.h
(works but arg access always fails)
- dropped seccomp_init_task, extra free functions, ...
- dropped the no-asm/syscall.h code paths
- merges with network sk_run_filter and sk_chk_filter
v6: - fix memory leak on attach compat check failure
- require no_new_privs || CAP_SYS_ADMIN prior to filter
installation. (luto@mit.edu)
- s/seccomp_struct_/seccomp_/ for macros/functions (amwang@redhat.com)
- cleaned up Kconfig (amwang@redhat.com)
- on block, note if the call was compat (so the # means something)
v5: - uses syscall_get_arguments
(indan@nul.nu,oleg@redhat.com, mcgrathr@chromium.org)
- uses union-based arg storage with hi/lo struct to
handle endianness. Compromises between the two alternate
proposals to minimize extra arg shuffling and account for
endianness assuming userspace uses offsetof().
(mcgrathr@chromium.org, indan@nul.nu)
- update Kconfig description
- add include/seccomp_filter.h and add its installation
- (naive) on-demand syscall argument loading
- drop seccomp_t (eparis@redhat.com)
v4: - adjusted prctl to make room for PR_[SG]ET_NO_NEW_PRIVS
- now uses current->no_new_privs
(luto@mit.edu,torvalds@linux-foundation.com)
- assign names to seccomp modes (rdunlap@xenotime.net)
- fix style issues (rdunlap@xenotime.net)
- reworded Kconfig entry (rdunlap@xenotime.net)
v3: - macros to inline (oleg@redhat.com)
- init_task behavior fixed (oleg@redhat.com)
- drop creator entry and extra NULL check (oleg@redhat.com)
- alloc returns -EINVAL on bad sizing (serge.hallyn@canonical.com)
- adds tentative use of "always_unprivileged" as per
torvalds@linux-foundation.org and luto@mit.edu
v2: - (patch 2 only)
Signed-off-by: James Morris <james.l.morris@oracle.com>
2012-04-13 01:47:57 +04:00
# include <linux/seccomp.h>
2005-04-17 02:20:36 +04:00
# include <linux/swap.h>
# include <linux/syscalls.h>
# include <linux/jiffies.h>
# include <linux/futex.h>
2008-11-15 21:20:36 +03:00
# include <linux/compat.h>
2011-03-23 02:30:44 +03:00
# include <linux/kthread.h>
2006-12-10 13:19:19 +03:00
# include <linux/task_io_accounting_ops.h>
2005-09-10 00:04:13 +04:00
# include <linux/rcupdate.h>
2005-04-17 02:20:36 +04:00
# include <linux/ptrace.h>
# include <linux/mount.h>
# include <linux/audit.h>
2008-02-07 11:13:51 +03:00
# include <linux/memcontrol.h>
2008-11-23 08:22:56 +03:00
# include <linux/ftrace.h>
2012-05-11 00:01:45 +04:00
# include <linux/proc_fs.h>
2005-04-17 02:20:36 +04:00
# include <linux/profile.h>
# include <linux/rmap.h>
2009-09-22 04:01:57 +04:00
# include <linux/ksm.h>
2005-04-17 02:20:36 +04:00
# include <linux/acct.h>
2017-02-23 02:42:27 +03:00
# include <linux/userfaultfd_k.h>
2006-10-01 10:28:59 +04:00
# include <linux/tsacct_kern.h>
2005-11-07 11:59:16 +03:00
# include <linux/cn_proc.h>
2007-05-24 00:57:25 +04:00
# include <linux/freezer.h>
2006-07-14 11:24:36 +04:00
# include <linux/delayacct.h>
2006-07-14 11:24:44 +04:00
# include <linux/taskstats_kern.h>
2006-09-26 12:52:38 +04:00
# include <linux/random.h>
Audit: add TTY input auditing
Add TTY input auditing, used to audit system administrator's actions. This is
required by various security standards such as DCID 6/3 and PCI to provide
non-repudiation of administrator's actions and to allow a review of past
actions if the administrator seems to overstep their duties or if the system
becomes misconfigured for unknown reasons. These requirements do not make it
necessary to audit TTY output as well.
Compared to an user-space keylogger, this approach records TTY input using the
audit subsystem, correlated with other audit events, and it is completely
transparent to the user-space application (e.g. the console ioctls still
work).
TTY input auditing works on a higher level than auditing all system calls
within the session, which would produce an overwhelming amount of mostly
useless audit events.
Add an "audit_tty" attribute, inherited across fork (). Data read from TTYs
by process with the attribute is sent to the audit subsystem by the kernel.
The audit netlink interface is extended to allow modifying the audit_tty
attribute, and to allow sending explanatory audit events from user-space (for
example, a shell might send an event containing the final command, after the
interactive command-line editing and history expansion is performed, which
might be difficult to decipher from the TTY input alone).
Because the "audit_tty" attribute is inherited across fork (), it would be set
e.g. for sshd restarted within an audited session. To prevent this, the
audit_tty attribute is cleared when a process with no open TTY file
descriptors (e.g. after daemon startup) opens a TTY.
See https://www.redhat.com/archives/linux-audit/2007-June/msg00000.html for a
more detailed rationale document for an older version of this patch.
[akpm@linux-foundation.org: build fix]
Signed-off-by: Miloslav Trmac <mitr@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Paul Fulghum <paulkf@microgate.com>
Cc: Casey Schaufler <casey@schaufler-ca.com>
Cc: Steve Grubb <sgrubb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-16 10:40:56 +04:00
# include <linux/tty.h>
2008-01-24 10:52:45 +03:00
# include <linux/blkdev.h>
2009-03-30 03:50:06 +04:00
# include <linux/fs_struct.h>
2008-04-23 01:38:23 +04:00
# include <linux/magic.h>
perf: Do the big rename: Performance Counters -> Performance Events
Bye-bye Performance Counters, welcome Performance Events!
In the past few months the perfcounters subsystem has grown out its
initial role of counting hardware events, and has become (and is
becoming) a much broader generic event enumeration, reporting, logging,
monitoring, analysis facility.
Naming its core object 'perf_counter' and naming the subsystem
'perfcounters' has become more and more of a misnomer. With pending
code like hw-breakpoints support the 'counter' name is less and
less appropriate.
All in one, we've decided to rename the subsystem to 'performance
events' and to propagate this rename through all fields, variables
and API names. (in an ABI compatible fashion)
The word 'event' is also a bit shorter than 'counter' - which makes
it slightly more convenient to write/handle as well.
Thanks goes to Stephane Eranian who first observed this misnomer and
suggested a rename.
User-space tooling and ABI compatibility is not affected - this patch
should be function-invariant. (Also, defconfigs were not touched to
keep the size down.)
This patch has been generated via the following script:
FILES=$(find * -type f | grep -vE 'oprofile|[^K]config')
sed -i \
-e 's/PERF_EVENT_/PERF_RECORD_/g' \
-e 's/PERF_COUNTER/PERF_EVENT/g' \
-e 's/perf_counter/perf_event/g' \
-e 's/nb_counters/nb_events/g' \
-e 's/swcounter/swevent/g' \
-e 's/tpcounter_event/tp_event/g' \
$FILES
for N in $(find . -name perf_counter.[ch]); do
M=$(echo $N | sed 's/perf_counter/perf_event/g')
mv $N $M
done
FILES=$(find . -name perf_event.*)
sed -i \
-e 's/COUNTER_MASK/REG_MASK/g' \
-e 's/COUNTER/EVENT/g' \
-e 's/\<event\>/event_id/g' \
-e 's/counter/event/g' \
-e 's/Counter/Event/g' \
$FILES
... to keep it as correct as possible. This script can also be
used by anyone who has pending perfcounters patches - it converts
a Linux kernel tree over to the new naming. We tried to time this
change to the point in time where the amount of pending patches
is the smallest: the end of the merge window.
Namespace clashes were fixed up in a preparatory patch - and some
stylistic fallout will be fixed up in a subsequent patch.
( NOTE: 'counters' are still the proper terminology when we deal
with hardware registers - and these sed scripts are a bit
over-eager in renaming them. I've undone some of that, but
in case there's something left where 'counter' would be
better than 'event' we can undo that on an individual basis
instead of touching an otherwise nicely automated patch. )
Suggested-by: Stephane Eranian <eranian@google.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Paul Mackerras <paulus@samba.org>
Reviewed-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <linux-arch@vger.kernel.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-09-21 14:02:48 +04:00
# include <linux/perf_event.h>
2009-07-29 14:15:26 +04:00
# include <linux/posix-timers.h>
2009-11-29 17:34:48 +03:00
# include <linux/user-return-notifier.h>
2010-10-27 01:21:23 +04:00
# include <linux/oom.h>
2011-01-14 02:46:58 +03:00
# include <linux/khugepaged.h>
epoll: introduce POLLFREE to flush ->signalfd_wqh before kfree()
This patch is intentionally incomplete to simplify the review.
It ignores ep_unregister_pollwait() which plays with the same wqh.
See the next change.
epoll assumes that the EPOLL_CTL_ADD'ed file controls everything
f_op->poll() needs. In particular it assumes that the wait queue
can't go away until eventpoll_release(). This is not true in case
of signalfd, the task which does EPOLL_CTL_ADD uses its ->sighand
which is not connected to the file.
This patch adds the special event, POLLFREE, currently only for
epoll. It expects that init_poll_funcptr()'ed hook should do the
necessary cleanup. Perhaps it should be defined as EPOLLFREE in
eventpoll.
__cleanup_sighand() is changed to do wake_up_poll(POLLFREE) if
->signalfd_wqh is not empty, we add the new signalfd_cleanup()
helper.
ep_poll_callback(POLLFREE) simply does list_del_init(task_list).
This make this poll entry inconsistent, but we don't care. If you
share epoll fd which contains our sigfd with another process you
should blame yourself. signalfd is "really special". I simply do
not know how we can define the "right" semantics if it used with
epoll.
The main problem is, epoll calls signalfd_poll() once to establish
the connection with the wait queue, after that signalfd_poll(NULL)
returns the different/inconsistent results depending on who does
EPOLL_CTL_MOD/signalfd_read/etc. IOW: apart from sigmask, signalfd
has nothing to do with the file, it works with the current thread.
In short: this patch is the hack which tries to fix the symptoms.
It also assumes that nobody can take tasklist_lock under epoll
locks, this seems to be true.
Note:
- we do not have wake_up_all_poll() but wake_up_poll()
is fine, poll/epoll doesn't use WQ_FLAG_EXCLUSIVE.
- signalfd_cleanup() uses POLLHUP along with POLLFREE,
we need a couple of simple changes in eventpoll.c to
make sure it can't be "lost".
Reported-by: Maxime Bizon <mbizon@freebox.fr>
Cc: <stable@kernel.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-02-24 23:07:11 +04:00
# include <linux/signalfd.h>
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-13 22:00:11 +04:00
# include <linux/uprobes.h>
2013-05-08 03:19:08 +04:00
# include <linux/aio.h>
2014-04-08 02:39:20 +04:00
# include <linux/compiler.h>
2015-04-16 22:47:50 +03:00
# include <linux/sysctl.h>
kernel: add kcov code coverage
kcov provides code coverage collection for coverage-guided fuzzing
(randomized testing). Coverage-guided fuzzing is a testing technique
that uses coverage feedback to determine new interesting inputs to a
system. A notable user-space example is AFL
(http://lcamtuf.coredump.cx/afl/). However, this technique is not
widely used for kernel testing due to missing compiler and kernel
support.
kcov does not aim to collect as much coverage as possible. It aims to
collect more or less stable coverage that is function of syscall inputs.
To achieve this goal it does not collect coverage in soft/hard
interrupts and instrumentation of some inherently non-deterministic or
non-interesting parts of kernel is disbled (e.g. scheduler, locking).
Currently there is a single coverage collection mode (tracing), but the
API anticipates additional collection modes. Initially I also
implemented a second mode which exposes coverage in a fixed-size hash
table of counters (what Quentin used in his original patch). I've
dropped the second mode for simplicity.
This patch adds the necessary support on kernel side. The complimentary
compiler support was added in gcc revision 231296.
We've used this support to build syzkaller system call fuzzer, which has
found 90 kernel bugs in just 2 months:
https://github.com/google/syzkaller/wiki/Found-Bugs
We've also found 30+ bugs in our internal systems with syzkaller.
Another (yet unexplored) direction where kcov coverage would greatly
help is more traditional "blob mutation". For example, mounting a
random blob as a filesystem, or receiving a random blob over wire.
Why not gcov. Typical fuzzing loop looks as follows: (1) reset
coverage, (2) execute a bit of code, (3) collect coverage, repeat. A
typical coverage can be just a dozen of basic blocks (e.g. an invalid
input). In such context gcov becomes prohibitively expensive as
reset/collect coverage steps depend on total number of basic
blocks/edges in program (in case of kernel it is about 2M). Cost of
kcov depends only on number of executed basic blocks/edges. On top of
that, kernel requires per-thread coverage because there are always
background threads and unrelated processes that also produce coverage.
With inlined gcov instrumentation per-thread coverage is not possible.
kcov exposes kernel PCs and control flow to user-space which is
insecure. But debugfs should not be mapped as user accessible.
Based on a patch by Quentin Casasnovas.
[akpm@linux-foundation.org: make task_struct.kcov_mode have type `enum kcov_mode']
[akpm@linux-foundation.org: unbreak allmodconfig]
[akpm@linux-foundation.org: follow x86 Makefile layout standards]
Signed-off-by: Dmitry Vyukov <dvyukov@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Cc: syzkaller <syzkaller@googlegroups.com>
Cc: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Tavis Ormandy <taviso@google.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Quentin Casasnovas <quentin.casasnovas@oracle.com>
Cc: Kostya Serebryany <kcc@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@google.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Sasha Levin <sasha.levin@oracle.com>
Cc: David Drysdale <drysdale@google.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Kirill A. Shutemov <kirill@shutemov.name>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-23 00:27:30 +03:00
# include <linux/kcov.h>
livepatch: change to a per-task consistency model
Change livepatch to use a basic per-task consistency model. This is the
foundation which will eventually enable us to patch those ~10% of
security patches which change function or data semantics. This is the
biggest remaining piece needed to make livepatch more generally useful.
This code stems from the design proposal made by Vojtech [1] in November
2014. It's a hybrid of kGraft and kpatch: it uses kGraft's per-task
consistency and syscall barrier switching combined with kpatch's stack
trace switching. There are also a number of fallback options which make
it quite flexible.
Patches are applied on a per-task basis, when the task is deemed safe to
switch over. When a patch is enabled, livepatch enters into a
transition state where tasks are converging to the patched state.
Usually this transition state can complete in a few seconds. The same
sequence occurs when a patch is disabled, except the tasks converge from
the patched state to the unpatched state.
An interrupt handler inherits the patched state of the task it
interrupts. The same is true for forked tasks: the child inherits the
patched state of the parent.
Livepatch uses several complementary approaches to determine when it's
safe to patch tasks:
1. The first and most effective approach is stack checking of sleeping
tasks. If no affected functions are on the stack of a given task,
the task is patched. In most cases this will patch most or all of
the tasks on the first try. Otherwise it'll keep trying
periodically. This option is only available if the architecture has
reliable stacks (HAVE_RELIABLE_STACKTRACE).
2. The second approach, if needed, is kernel exit switching. A
task is switched when it returns to user space from a system call, a
user space IRQ, or a signal. It's useful in the following cases:
a) Patching I/O-bound user tasks which are sleeping on an affected
function. In this case you have to send SIGSTOP and SIGCONT to
force it to exit the kernel and be patched.
b) Patching CPU-bound user tasks. If the task is highly CPU-bound
then it will get patched the next time it gets interrupted by an
IRQ.
c) In the future it could be useful for applying patches for
architectures which don't yet have HAVE_RELIABLE_STACKTRACE. In
this case you would have to signal most of the tasks on the
system. However this isn't supported yet because there's
currently no way to patch kthreads without
HAVE_RELIABLE_STACKTRACE.
3. For idle "swapper" tasks, since they don't ever exit the kernel, they
instead have a klp_update_patch_state() call in the idle loop which
allows them to be patched before the CPU enters the idle state.
(Note there's not yet such an approach for kthreads.)
All the above approaches may be skipped by setting the 'immediate' flag
in the 'klp_patch' struct, which will disable per-task consistency and
patch all tasks immediately. This can be useful if the patch doesn't
change any function or data semantics. Note that, even with this flag
set, it's possible that some tasks may still be running with an old
version of the function, until that function returns.
There's also an 'immediate' flag in the 'klp_func' struct which allows
you to specify that certain functions in the patch can be applied
without per-task consistency. This might be useful if you want to patch
a common function like schedule(), and the function change doesn't need
consistency but the rest of the patch does.
For architectures which don't have HAVE_RELIABLE_STACKTRACE, the user
must set patch->immediate which causes all tasks to be patched
immediately. This option should be used with care, only when the patch
doesn't change any function or data semantics.
In the future, architectures which don't have HAVE_RELIABLE_STACKTRACE
may be allowed to use per-task consistency if we can come up with
another way to patch kthreads.
The /sys/kernel/livepatch/<patch>/transition file shows whether a patch
is in transition. Only a single patch (the topmost patch on the stack)
can be in transition at a given time. A patch can remain in transition
indefinitely, if any of the tasks are stuck in the initial patch state.
A transition can be reversed and effectively canceled by writing the
opposite value to the /sys/kernel/livepatch/<patch>/enabled file while
the transition is in progress. Then all the tasks will attempt to
converge back to the original patch state.
[1] https://lkml.kernel.org/r/20141107140458.GA21774@suse.cz
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Acked-by: Miroslav Benes <mbenes@suse.cz>
Acked-by: Ingo Molnar <mingo@kernel.org> # for the scheduler changes
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2017-02-14 04:42:40 +03:00
# include <linux/livepatch.h>
2017-07-14 14:23:09 +03:00
# include <linux/thread_info.h>
2018-08-17 01:16:58 +03:00
# include <linux/stackleak.h>
2019-12-01 04:54:57 +03:00
# include <linux/kasan.h>
2020-04-27 19:00:07 +03:00
# include <linux/scs.h>
2020-09-13 22:09:39 +03:00
# include <linux/io_uring.h>
2021-02-26 02:43:14 +03:00
# include <linux/bpf.h>
2005-04-17 02:20:36 +04:00
# include <asm/pgalloc.h>
2016-12-24 22:46:01 +03:00
# include <linux/uaccess.h>
2005-04-17 02:20:36 +04:00
# include <asm/mmu_context.h>
# include <asm/cacheflush.h>
# include <asm/tlbflush.h>
2009-04-15 03:39:12 +04:00
# include <trace/events/sched.h>
2012-01-11 03:08:09 +04:00
# define CREATE_TRACE_POINTS
# include <trace/events/task.h>
2015-04-16 22:47:47 +03:00
/*
* Minimum number of threads to boot the kernel
*/
# define MIN_THREADS 20
/*
* Maximum number of threads
*/
# define MAX_THREADS FUTEX_TID_MASK
2005-04-17 02:20:36 +04:00
/*
* Protected counters by write_lock_irq ( & tasklist_lock )
*/
unsigned long total_forks ; /* Handle normal Linux uptimes. */
2011-07-27 03:08:39 +04:00
int nr_threads ; /* The idle threads do not count.. */
2005-04-17 02:20:36 +04:00
2019-06-01 08:30:12 +03:00
static int max_threads ; /* tunable limit on nr_threads */
2005-04-17 02:20:36 +04:00
2019-09-26 02:47:27 +03:00
# define NAMED_ARRAY_INDEX(x) [x] = __stringify(x)
static const char * const resident_page_types [ ] = {
NAMED_ARRAY_INDEX ( MM_FILEPAGES ) ,
NAMED_ARRAY_INDEX ( MM_ANONPAGES ) ,
NAMED_ARRAY_INDEX ( MM_SWAPENTS ) ,
NAMED_ARRAY_INDEX ( MM_SHMEMPAGES ) ,
} ;
2005-04-17 02:20:36 +04:00
DEFINE_PER_CPU ( unsigned long , process_counts ) = 0 ;
2006-07-10 15:45:40 +04:00
__cacheline_aligned DEFINE_RWLOCK ( tasklist_lock ) ; /* outer */
2010-03-03 18:46:56 +03:00
# ifdef CONFIG_PROVE_RCU
int lockdep_tasklist_lock_is_held ( void )
{
return lockdep_is_held ( & tasklist_lock ) ;
}
EXPORT_SYMBOL_GPL ( lockdep_tasklist_lock_is_held ) ;
# endif /* #ifdef CONFIG_PROVE_RCU */
2005-04-17 02:20:36 +04:00
int nr_processes ( void )
{
int cpu ;
int total = 0 ;
2009-11-03 13:11:14 +03:00
for_each_possible_cpu ( cpu )
2005-04-17 02:20:36 +04:00
total + = per_cpu ( process_counts , cpu ) ;
return total ;
}
fork: fix error handling in dup_task()
The function dup_task() may fail at the following function calls in the
following order.
0) alloc_task_struct_node()
1) alloc_thread_info_node()
2) arch_dup_task_struct()
Error by 0) is not a matter, it can just return. But error by 1) requires
releasing task_struct allocated by 0) before it returns. Likewise, error
by 2) requires releasing task_struct and thread_info allocated by 0) and
1).
The existing error handling calls free_task_struct() and
free_thread_info() which do not only release task_struct and thread_info,
but also call architecture specific arch_release_task_struct() and
arch_release_thread_info().
The problem is that task_struct and thread_info are not fully initialized
yet at this point, but arch_release_task_struct() and
arch_release_thread_info() are called with them.
For example, x86 defines its own arch_release_task_struct() that releases
a task_xstate. If alloc_thread_info_node() fails in dup_task(),
arch_release_task_struct() is called with task_struct which is just
allocated and filled with garbage in this error handling.
This actually happened with tools/testing/fault-injection/failcmd.sh
# env FAILCMD_TYPE=fail_page_alloc \
./tools/testing/fault-injection/failcmd.sh --times=100 \
--min-order=0 --ignore-gfp-wait=0 \
-- make -C tools/testing/selftests/ run_tests
In order to fix this issue, make free_{task_struct,thread_info}() not to
call arch_release_{task_struct,thread_info}() and call
arch_release_{task_struct,thread_info}() implicitly where needed.
Default arch_release_task_struct() and arch_release_thread_info() are
defined as empty by default. So this change only affects the
architectures which implement their own arch_release_task_struct() or
arch_release_thread_info() as listed below.
arch_release_task_struct(): x86, sh
arch_release_thread_info(): mn10300, tile
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Salman Qazi <sqazi@google.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 01:42:33 +04:00
void __weak arch_release_task_struct ( struct task_struct * tsk )
{
}
2012-05-05 19:05:48 +04:00
# ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR
2006-12-07 07:33:20 +03:00
static struct kmem_cache * task_struct_cachep ;
2012-05-05 19:05:41 +04:00
static inline struct task_struct * alloc_task_struct_node ( int node )
{
return kmem_cache_alloc_node ( task_struct_cachep , GFP_KERNEL , node ) ;
}
static inline void free_task_struct ( struct task_struct * tsk )
{
kmem_cache_free ( task_struct_cachep , tsk ) ;
}
2005-04-17 02:20:36 +04:00
# endif
Clarify naming of thread info/stack allocators
We've had the thread info allocated together with the thread stack for
most architectures for a long time (since the thread_info was split off
from the task struct), but that is about to change.
But the patches that move the thread info to be off-stack (and a part of
the task struct instead) made it clear how confused the allocator and
freeing functions are.
Because the common case was that we share an allocation with the thread
stack and the thread_info, the two pointers were identical. That
identity then meant that we would have things like
ti = alloc_thread_info_node(tsk, node);
...
tsk->stack = ti;
which certainly _worked_ (since stack and thread_info have the same
value), but is rather confusing: why are we assigning a thread_info to
the stack? And if we move the thread_info away, the "confusing" code
just gets to be entirely bogus.
So remove all this confusion, and make it clear that we are doing the
stack allocation by renaming and clarifying the function names to be
about the stack. The fact that the thread_info then shares the
allocation is an implementation detail, and not really about the
allocation itself.
This is a pure renaming and type fix: we pass in the same pointer, it's
just that we clarify what the pointer means.
The ia64 code that actually only has one single allocation (for all of
task_struct, thread_info and kernel thread stack) now looks a bit odd,
but since "tsk->stack" is actually not even used there, that oddity
doesn't matter. It would be a separate thing to clean that up, I
intentionally left the ia64 changes as a pure brute-force renaming and
type change.
Acked-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-06-25 01:09:37 +03:00
# ifndef CONFIG_ARCH_THREAD_STACK_ALLOCATOR
2012-05-05 19:05:41 +04:00
2012-05-05 19:05:41 +04:00
/*
* Allocate pages if THREAD_SIZE is > = PAGE_SIZE , otherwise use a
* kmemcache based allocator .
*/
2016-08-11 12:35:21 +03:00
# if THREAD_SIZE >= PAGE_SIZE || defined(CONFIG_VMAP_STACK)
2016-09-16 08:45:49 +03:00
# ifdef CONFIG_VMAP_STACK
/*
* vmalloc ( ) is a bit slow , and calling vfree ( ) enough times will force a TLB
* flush . Try to minimize the number of calls by caching stacks .
*/
# define NR_CACHED_STACKS 2
static DEFINE_PER_CPU ( struct vm_struct * , cached_stacks [ NR_CACHED_STACKS ] ) ;
2017-05-09 01:56:11 +03:00
static int free_vm_stack_cache ( unsigned int cpu )
{
struct vm_struct * * cached_vm_stacks = per_cpu_ptr ( cached_stacks , cpu ) ;
int i ;
for ( i = 0 ; i < NR_CACHED_STACKS ; i + + ) {
struct vm_struct * vm_stack = cached_vm_stacks [ i ] ;
if ( ! vm_stack )
continue ;
vfree ( vm_stack - > addr ) ;
cached_vm_stacks [ i ] = NULL ;
}
return 0 ;
}
2016-09-16 08:45:49 +03:00
# endif
2016-08-11 12:35:21 +03:00
static unsigned long * alloc_thread_stack_node ( struct task_struct * tsk , int node )
2008-07-25 12:45:40 +04:00
{
2016-08-11 12:35:21 +03:00
# ifdef CONFIG_VMAP_STACK
2016-09-16 08:45:49 +03:00
void * stack ;
int i ;
for ( i = 0 ; i < NR_CACHED_STACKS ; i + + ) {
2017-07-13 00:33:11 +03:00
struct vm_struct * s ;
s = this_cpu_xchg ( cached_stacks [ i ] , NULL ) ;
2016-09-16 08:45:49 +03:00
if ( ! s )
continue ;
2020-12-22 23:00:21 +03:00
/* Mark stack accessible for KASAN. */
kasan_unpoison_range ( s - > addr , THREAD_SIZE ) ;
2019-12-01 04:54:57 +03:00
2017-10-14 01:58:22 +03:00
/* Clear stale pointers from reused stack. */
memset ( s - > addr , 0 , THREAD_SIZE ) ;
2018-04-21 00:55:31 +03:00
2016-09-16 08:45:49 +03:00
tsk - > stack_vm_area = s ;
2019-01-09 02:22:57 +03:00
tsk - > stack = s - > addr ;
2016-09-16 08:45:49 +03:00
return s - > addr ;
}
2018-10-27 01:03:19 +03:00
/*
* Allocated stacks are cached and later reused by new threads ,
* so memcg accounting is performed manually on assigning / releasing
* stacks to tasks . Drop __GFP_ACCOUNT .
*/
2017-07-14 14:23:09 +03:00
stack = __vmalloc_node_range ( THREAD_SIZE , THREAD_ALIGN ,
2016-09-16 08:45:49 +03:00
VMALLOC_START , VMALLOC_END ,
2018-10-27 01:03:19 +03:00
THREADINFO_GFP & ~ __GFP_ACCOUNT ,
2016-09-16 08:45:49 +03:00
PAGE_KERNEL ,
0 , node , __builtin_return_address ( 0 ) ) ;
2016-08-11 12:35:21 +03:00
/*
* We can ' t call find_vm_area ( ) in interrupt context , and
* free_thread_stack ( ) can be called in interrupt context ,
* so cache the vm_struct .
*/
2018-12-22 01:30:54 +03:00
if ( stack ) {
2016-08-11 12:35:21 +03:00
tsk - > stack_vm_area = find_vm_area ( stack ) ;
2018-12-22 01:30:54 +03:00
tsk - > stack = stack ;
}
2016-08-11 12:35:21 +03:00
return stack ;
# else
mm: charge/uncharge kmemcg from generic page allocator paths
Currently, to charge a non-slab allocation to kmemcg one has to use
alloc_kmem_pages helper with __GFP_ACCOUNT flag. A page allocated with
this helper should finally be freed using free_kmem_pages, otherwise it
won't be uncharged.
This API suits its current users fine, but it turns out to be impossible
to use along with page reference counting, i.e. when an allocation is
supposed to be freed with put_page, as it is the case with pipe or unix
socket buffers.
To overcome this limitation, this patch moves charging/uncharging to
generic page allocator paths, i.e. to __alloc_pages_nodemask and
free_pages_prepare, and zaps alloc/free_kmem_pages helpers. This way,
one can use any of the available page allocation functions to get the
allocated page charged to kmemcg - it's enough to pass __GFP_ACCOUNT,
just like in case of kmalloc and friends. A charged page will be
automatically uncharged on free.
To make it possible, we need to mark pages charged to kmemcg somehow.
To avoid introducing a new page flag, we make use of page->_mapcount for
marking such pages. Since pages charged to kmemcg are not supposed to
be mapped to userspace, it should work just fine. There are other
(ab)users of page->_mapcount - buddy and balloon pages - but we don't
conflict with them.
In case kmemcg is compiled out or not used at runtime, this patch
introduces no overhead to generic page allocator paths. If kmemcg is
used, it will be plus one gfp flags check on alloc and plus one
page->_mapcount check on free, which shouldn't hurt performance, because
the data accessed are hot.
Link: http://lkml.kernel.org/r/a9736d856f895bcb465d9f257b54efe32eda6f99.1464079538.git.vdavydov@virtuozzo.com
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Minchan Kim <minchan@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-27 01:24:24 +03:00
struct page * page = alloc_pages_node ( node , THREADINFO_GFP ,
THREAD_SIZE_ORDER ) ;
2011-03-23 02:30:42 +03:00
2019-06-28 22:07:14 +03:00
if ( likely ( page ) ) {
2020-08-07 09:24:57 +03:00
tsk - > stack = kasan_reset_tag ( page_address ( page ) ) ;
2019-06-28 22:07:14 +03:00
return tsk - > stack ;
}
return NULL ;
2016-08-11 12:35:21 +03:00
# endif
2008-07-25 12:45:40 +04:00
}
2016-08-11 12:35:21 +03:00
static inline void free_thread_stack ( struct task_struct * tsk )
2008-07-25 12:45:40 +04:00
{
2016-09-16 08:45:49 +03:00
# ifdef CONFIG_VMAP_STACK
2018-10-27 01:03:19 +03:00
struct vm_struct * vm = task_stack_vm_area ( tsk ) ;
if ( vm ) {
2016-09-16 08:45:49 +03:00
int i ;
2020-08-07 09:21:37 +03:00
for ( i = 0 ; i < THREAD_SIZE / PAGE_SIZE ; i + + )
2020-04-02 07:06:46 +03:00
memcg_kmem_uncharge_page ( vm - > pages [ i ] , 0 ) ;
2018-10-27 01:03:19 +03:00
2016-09-16 08:45:49 +03:00
for ( i = 0 ; i < NR_CACHED_STACKS ; i + + ) {
2017-07-13 00:33:11 +03:00
if ( this_cpu_cmpxchg ( cached_stacks [ i ] ,
NULL , tsk - > stack_vm_area ) ! = NULL )
2016-09-16 08:45:49 +03:00
continue ;
return ;
}
2016-12-13 03:44:14 +03:00
vfree_atomic ( tsk - > stack ) ;
2016-09-16 08:45:49 +03:00
return ;
}
# endif
__free_pages ( virt_to_page ( tsk - > stack ) , THREAD_SIZE_ORDER ) ;
2008-07-25 12:45:40 +04:00
}
2012-05-05 19:05:41 +04:00
# else
Clarify naming of thread info/stack allocators
We've had the thread info allocated together with the thread stack for
most architectures for a long time (since the thread_info was split off
from the task struct), but that is about to change.
But the patches that move the thread info to be off-stack (and a part of
the task struct instead) made it clear how confused the allocator and
freeing functions are.
Because the common case was that we share an allocation with the thread
stack and the thread_info, the two pointers were identical. That
identity then meant that we would have things like
ti = alloc_thread_info_node(tsk, node);
...
tsk->stack = ti;
which certainly _worked_ (since stack and thread_info have the same
value), but is rather confusing: why are we assigning a thread_info to
the stack? And if we move the thread_info away, the "confusing" code
just gets to be entirely bogus.
So remove all this confusion, and make it clear that we are doing the
stack allocation by renaming and clarifying the function names to be
about the stack. The fact that the thread_info then shares the
allocation is an implementation detail, and not really about the
allocation itself.
This is a pure renaming and type fix: we pass in the same pointer, it's
just that we clarify what the pointer means.
The ia64 code that actually only has one single allocation (for all of
task_struct, thread_info and kernel thread stack) now looks a bit odd,
but since "tsk->stack" is actually not even used there, that oddity
doesn't matter. It would be a separate thing to clean that up, I
intentionally left the ia64 changes as a pure brute-force renaming and
type change.
Acked-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-06-25 01:09:37 +03:00
static struct kmem_cache * thread_stack_cache ;
2012-05-05 19:05:41 +04:00
2016-06-25 14:53:30 +03:00
static unsigned long * alloc_thread_stack_node ( struct task_struct * tsk ,
2012-05-05 19:05:41 +04:00
int node )
{
2018-12-22 01:30:54 +03:00
unsigned long * stack ;
stack = kmem_cache_alloc_node ( thread_stack_cache , THREADINFO_GFP , node ) ;
2020-08-07 09:24:57 +03:00
stack = kasan_reset_tag ( stack ) ;
2018-12-22 01:30:54 +03:00
tsk - > stack = stack ;
return stack ;
2012-05-05 19:05:41 +04:00
}
2016-08-11 12:35:21 +03:00
static void free_thread_stack ( struct task_struct * tsk )
2012-05-05 19:05:41 +04:00
{
2016-08-11 12:35:21 +03:00
kmem_cache_free ( thread_stack_cache , tsk - > stack ) ;
2012-05-05 19:05:41 +04:00
}
Clarify naming of thread info/stack allocators
We've had the thread info allocated together with the thread stack for
most architectures for a long time (since the thread_info was split off
from the task struct), but that is about to change.
But the patches that move the thread info to be off-stack (and a part of
the task struct instead) made it clear how confused the allocator and
freeing functions are.
Because the common case was that we share an allocation with the thread
stack and the thread_info, the two pointers were identical. That
identity then meant that we would have things like
ti = alloc_thread_info_node(tsk, node);
...
tsk->stack = ti;
which certainly _worked_ (since stack and thread_info have the same
value), but is rather confusing: why are we assigning a thread_info to
the stack? And if we move the thread_info away, the "confusing" code
just gets to be entirely bogus.
So remove all this confusion, and make it clear that we are doing the
stack allocation by renaming and clarifying the function names to be
about the stack. The fact that the thread_info then shares the
allocation is an implementation detail, and not really about the
allocation itself.
This is a pure renaming and type fix: we pass in the same pointer, it's
just that we clarify what the pointer means.
The ia64 code that actually only has one single allocation (for all of
task_struct, thread_info and kernel thread stack) now looks a bit odd,
but since "tsk->stack" is actually not even used there, that oddity
doesn't matter. It would be a separate thing to clean that up, I
intentionally left the ia64 changes as a pure brute-force renaming and
type change.
Acked-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-06-25 01:09:37 +03:00
void thread_stack_cache_init ( void )
2012-05-05 19:05:41 +04:00
{
2017-06-11 05:50:41 +03:00
thread_stack_cache = kmem_cache_create_usercopy ( " thread_stack " ,
THREAD_SIZE , THREAD_SIZE , 0 , 0 ,
THREAD_SIZE , NULL ) ;
Clarify naming of thread info/stack allocators
We've had the thread info allocated together with the thread stack for
most architectures for a long time (since the thread_info was split off
from the task struct), but that is about to change.
But the patches that move the thread info to be off-stack (and a part of
the task struct instead) made it clear how confused the allocator and
freeing functions are.
Because the common case was that we share an allocation with the thread
stack and the thread_info, the two pointers were identical. That
identity then meant that we would have things like
ti = alloc_thread_info_node(tsk, node);
...
tsk->stack = ti;
which certainly _worked_ (since stack and thread_info have the same
value), but is rather confusing: why are we assigning a thread_info to
the stack? And if we move the thread_info away, the "confusing" code
just gets to be entirely bogus.
So remove all this confusion, and make it clear that we are doing the
stack allocation by renaming and clarifying the function names to be
about the stack. The fact that the thread_info then shares the
allocation is an implementation detail, and not really about the
allocation itself.
This is a pure renaming and type fix: we pass in the same pointer, it's
just that we clarify what the pointer means.
The ia64 code that actually only has one single allocation (for all of
task_struct, thread_info and kernel thread stack) now looks a bit odd,
but since "tsk->stack" is actually not even used there, that oddity
doesn't matter. It would be a separate thing to clean that up, I
intentionally left the ia64 changes as a pure brute-force renaming and
type change.
Acked-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-06-25 01:09:37 +03:00
BUG_ON ( thread_stack_cache = = NULL ) ;
2012-05-05 19:05:41 +04:00
}
# endif
2008-07-25 12:45:40 +04:00
# endif
2005-04-17 02:20:36 +04:00
/* SLAB cache for signal_struct structures (tsk->signal) */
2006-12-07 07:33:20 +03:00
static struct kmem_cache * signal_cachep ;
2005-04-17 02:20:36 +04:00
/* SLAB cache for sighand_struct structures (tsk->sighand) */
2006-12-07 07:33:20 +03:00
struct kmem_cache * sighand_cachep ;
2005-04-17 02:20:36 +04:00
/* SLAB cache for files_struct structures (tsk->files) */
2006-12-07 07:33:20 +03:00
struct kmem_cache * files_cachep ;
2005-04-17 02:20:36 +04:00
/* SLAB cache for fs_struct structures (tsk->fs) */
2006-12-07 07:33:20 +03:00
struct kmem_cache * fs_cachep ;
2005-04-17 02:20:36 +04:00
/* SLAB cache for vm_area_struct structures */
2018-07-21 23:48:51 +03:00
static struct kmem_cache * vm_area_cachep ;
2005-04-17 02:20:36 +04:00
/* SLAB cache for mm_struct structures (tsk->mm) */
2006-12-07 07:33:20 +03:00
static struct kmem_cache * mm_cachep ;
2005-04-17 02:20:36 +04:00
2018-07-22 01:24:03 +03:00
struct vm_area_struct * vm_area_alloc ( struct mm_struct * mm )
2018-07-21 23:48:51 +03:00
{
2018-08-22 07:53:06 +03:00
struct vm_area_struct * vma ;
2018-07-22 01:24:03 +03:00
2018-08-22 07:53:06 +03:00
vma = kmem_cache_alloc ( vm_area_cachep , GFP_KERNEL ) ;
2018-07-27 02:37:25 +03:00
if ( vma )
vma_init ( vma , mm ) ;
2018-07-22 01:24:03 +03:00
return vma ;
2018-07-21 23:48:51 +03:00
}
struct vm_area_struct * vm_area_dup ( struct vm_area_struct * orig )
{
2018-07-22 00:48:45 +03:00
struct vm_area_struct * new = kmem_cache_alloc ( vm_area_cachep , GFP_KERNEL ) ;
if ( new ) {
2020-02-19 22:00:54 +03:00
ASSERT_EXCLUSIVE_WRITER ( orig - > vm_flags ) ;
ASSERT_EXCLUSIVE_WRITER ( orig - > vm_file ) ;
/*
* orig - > shared . rb may be modified concurrently , but the clone
* will be reinitialized .
*/
* new = data_race ( * orig ) ;
2018-07-22 00:48:45 +03:00
INIT_LIST_HEAD ( & new - > anon_vma_chain ) ;
2020-04-07 06:03:39 +03:00
new - > vm_next = new - > vm_prev = NULL ;
2018-07-22 00:48:45 +03:00
}
return new ;
2018-07-21 23:48:51 +03:00
}
void vm_area_free ( struct vm_area_struct * vma )
{
kmem_cache_free ( vm_area_cachep , vma ) ;
}
2016-08-11 12:35:21 +03:00
static void account_kernel_stack ( struct task_struct * tsk , int account )
2009-09-22 04:01:32 +04:00
{
2016-08-11 12:35:21 +03:00
void * stack = task_stack_page ( tsk ) ;
struct vm_struct * vm = task_stack_vm_area ( tsk ) ;
2021-04-30 08:56:02 +03:00
if ( vm ) {
int i ;
2016-08-11 12:35:21 +03:00
2021-04-30 08:56:02 +03:00
for ( i = 0 ; i < THREAD_SIZE / PAGE_SIZE ; i + + )
mod_lruvec_page_state ( vm - > pages [ i ] , NR_KERNEL_STACK_KB ,
account * ( PAGE_SIZE / 1024 ) ) ;
} else {
/* All stack pages are in the same node. */
2020-12-15 06:07:04 +03:00
mod_lruvec_kmem_state ( stack , NR_KERNEL_STACK_KB ,
2020-08-07 09:21:37 +03:00
account * ( THREAD_SIZE / 1024 ) ) ;
2021-04-30 08:56:02 +03:00
}
2009-09-22 04:01:32 +04:00
}
2018-10-27 01:03:19 +03:00
static int memcg_charge_kernel_stack ( struct task_struct * tsk )
{
# ifdef CONFIG_VMAP_STACK
struct vm_struct * vm = task_stack_vm_area ( tsk ) ;
int ret ;
2020-08-07 09:21:37 +03:00
BUILD_BUG_ON ( IS_ENABLED ( CONFIG_VMAP_STACK ) & & PAGE_SIZE % 1024 ! = 0 ) ;
2018-10-27 01:03:19 +03:00
if ( vm ) {
int i ;
2020-08-07 09:21:37 +03:00
BUG_ON ( vm - > nr_pages ! = THREAD_SIZE / PAGE_SIZE ) ;
2018-10-27 01:03:19 +03:00
for ( i = 0 ; i < THREAD_SIZE / PAGE_SIZE ; i + + ) {
/*
mm: memcontrol: Use helpers to read page's memcg data
Patch series "mm: allow mapping accounted kernel pages to userspace", v6.
Currently a non-slab kernel page which has been charged to a memory cgroup
can't be mapped to userspace. The underlying reason is simple: PageKmemcg
flag is defined as a page type (like buddy, offline, etc), so it takes a
bit from a page->mapped counter. Pages with a type set can't be mapped to
userspace.
But in general the kmemcg flag has nothing to do with mapping to
userspace. It only means that the page has been accounted by the page
allocator, so it has to be properly uncharged on release.
Some bpf maps are mapping the vmalloc-based memory to userspace, and their
memory can't be accounted because of this implementation detail.
This patchset removes this limitation by moving the PageKmemcg flag into
one of the free bits of the page->mem_cgroup pointer. Also it formalizes
accesses to the page->mem_cgroup and page->obj_cgroups using new helpers,
adds several checks and removes a couple of obsolete functions. As the
result the code became more robust with fewer open-coded bit tricks.
This patch (of 4):
Currently there are many open-coded reads of the page->mem_cgroup pointer,
as well as a couple of read helpers, which are barely used.
It creates an obstacle on a way to reuse some bits of the pointer for
storing additional bits of information. In fact, we already do this for
slab pages, where the last bit indicates that a pointer has an attached
vector of objcg pointers instead of a regular memcg pointer.
This commits uses 2 existing helpers and introduces a new helper to
converts all read sides to calls of these helpers:
struct mem_cgroup *page_memcg(struct page *page);
struct mem_cgroup *page_memcg_rcu(struct page *page);
struct mem_cgroup *page_memcg_check(struct page *page);
page_memcg_check() is intended to be used in cases when the page can be a
slab page and have a memcg pointer pointing at objcg vector. It does
check the lowest bit, and if set, returns NULL. page_memcg() contains a
VM_BUG_ON_PAGE() check for the page not being a slab page.
To make sure nobody uses a direct access, struct page's
mem_cgroup/obj_cgroups is converted to unsigned long memcg_data.
Signed-off-by: Roman Gushchin <guro@fb.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Link: https://lkml.kernel.org/r/20201027001657.3398190-1-guro@fb.com
Link: https://lkml.kernel.org/r/20201027001657.3398190-2-guro@fb.com
Link: https://lore.kernel.org/bpf/20201201215900.3569844-2-guro@fb.com
2020-12-02 00:58:27 +03:00
* If memcg_kmem_charge_page ( ) fails , page ' s
* memory cgroup pointer is NULL , and
* memcg_kmem_uncharge_page ( ) in free_thread_stack ( )
* will ignore this page .
2018-10-27 01:03:19 +03:00
*/
2020-04-02 07:06:46 +03:00
ret = memcg_kmem_charge_page ( vm - > pages [ i ] , GFP_KERNEL ,
0 ) ;
2018-10-27 01:03:19 +03:00
if ( ret )
return ret ;
}
}
# endif
return 0 ;
}
2016-09-16 08:45:48 +03:00
static void release_task_stack ( struct task_struct * tsk )
2005-04-17 02:20:36 +04:00
{
2021-06-11 11:28:17 +03:00
if ( WARN_ON ( READ_ONCE ( tsk - > __state ) ! = TASK_DEAD ) )
2016-10-31 18:11:43 +03:00
return ; /* Better to leak the stack than to free prematurely */
2016-08-11 12:35:21 +03:00
account_kernel_stack ( tsk , - 1 ) ;
free_thread_stack ( tsk ) ;
2016-09-16 08:45:48 +03:00
tsk - > stack = NULL ;
# ifdef CONFIG_VMAP_STACK
tsk - > stack_vm_area = NULL ;
# endif
}
# ifdef CONFIG_THREAD_INFO_IN_TASK
void put_task_stack ( struct task_struct * tsk )
{
2019-01-18 15:27:30 +03:00
if ( refcount_dec_and_test ( & tsk - > stack_refcount ) )
2016-09-16 08:45:48 +03:00
release_task_stack ( tsk ) ;
}
# endif
void free_task ( struct task_struct * tsk )
{
2020-04-27 19:00:07 +03:00
scs_release ( tsk ) ;
2016-09-16 08:45:48 +03:00
# ifndef CONFIG_THREAD_INFO_IN_TASK
/*
* The task is finally done with both the stack and thread_info ,
* so free both .
*/
release_task_stack ( tsk ) ;
# else
/*
* If the task had a separate stack allocation , it should be gone
* by now .
*/
2019-01-18 15:27:30 +03:00
WARN_ON_ONCE ( refcount_read ( & tsk - > stack_refcount ) ! = 0 ) ;
2016-09-16 08:45:48 +03:00
# endif
2006-06-27 13:54:53 +04:00
rt_mutex_debug_task_free ( tsk ) ;
2008-11-25 23:07:04 +03:00
ftrace_graph_exit_task ( tsk ) ;
fork: fix error handling in dup_task()
The function dup_task() may fail at the following function calls in the
following order.
0) alloc_task_struct_node()
1) alloc_thread_info_node()
2) arch_dup_task_struct()
Error by 0) is not a matter, it can just return. But error by 1) requires
releasing task_struct allocated by 0) before it returns. Likewise, error
by 2) requires releasing task_struct and thread_info allocated by 0) and
1).
The existing error handling calls free_task_struct() and
free_thread_info() which do not only release task_struct and thread_info,
but also call architecture specific arch_release_task_struct() and
arch_release_thread_info().
The problem is that task_struct and thread_info are not fully initialized
yet at this point, but arch_release_task_struct() and
arch_release_thread_info() are called with them.
For example, x86 defines its own arch_release_task_struct() that releases
a task_xstate. If alloc_thread_info_node() fails in dup_task(),
arch_release_task_struct() is called with task_struct which is just
allocated and filled with garbage in this error handling.
This actually happened with tools/testing/fault-injection/failcmd.sh
# env FAILCMD_TYPE=fail_page_alloc \
./tools/testing/fault-injection/failcmd.sh --times=100 \
--min-order=0 --ignore-gfp-wait=0 \
-- make -C tools/testing/selftests/ run_tests
In order to fix this issue, make free_{task_struct,thread_info}() not to
call arch_release_{task_struct,thread_info}() and call
arch_release_{task_struct,thread_info}() implicitly where needed.
Default arch_release_task_struct() and arch_release_thread_info() are
defined as empty by default. So this change only affects the
architectures which implement their own arch_release_task_struct() or
arch_release_thread_info() as listed below.
arch_release_task_struct(): x86, sh
arch_release_thread_info(): mn10300, tile
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Salman Qazi <sqazi@google.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 01:42:33 +04:00
arch_release_task_struct ( tsk ) ;
2016-11-29 20:50:57 +03:00
if ( tsk - > flags & PF_KTHREAD )
free_kthread_struct ( tsk ) ;
2005-04-17 02:20:36 +04:00
free_task_struct ( tsk ) ;
}
EXPORT_SYMBOL ( free_task ) ;
2018-02-01 03:15:51 +03:00
# ifdef CONFIG_MMU
static __latent_entropy int dup_mmap ( struct mm_struct * mm ,
struct mm_struct * oldmm )
{
struct vm_area_struct * mpnt , * tmp , * prev , * * pprev ;
struct rb_node * * rb_link , * rb_parent ;
int retval ;
unsigned long charge ;
LIST_HEAD ( uf ) ;
uprobe_start_dup_mmap ( ) ;
2020-06-09 07:33:25 +03:00
if ( mmap_write_lock_killable ( oldmm ) ) {
2018-02-01 03:15:51 +03:00
retval = - EINTR ;
goto fail_uprobe_end ;
}
flush_cache_dup_mm ( oldmm ) ;
uprobe_dup_mmap ( oldmm , mm ) ;
/*
* Not linked in yet - no deadlock potential :
*/
2020-06-09 07:33:33 +03:00
mmap_write_lock_nested ( mm , SINGLE_DEPTH_NESTING ) ;
2018-02-01 03:15:51 +03:00
/* No ordering required: file already has been exposed. */
RCU_INIT_POINTER ( mm - > exe_file , get_mm_exe_file ( oldmm ) ) ;
mm - > total_vm = oldmm - > total_vm ;
mm - > data_vm = oldmm - > data_vm ;
mm - > exec_vm = oldmm - > exec_vm ;
mm - > stack_vm = oldmm - > stack_vm ;
rb_link = & mm - > mm_rb . rb_node ;
rb_parent = NULL ;
pprev = & mm - > mmap ;
retval = ksm_fork ( mm , oldmm ) ;
if ( retval )
goto out ;
retval = khugepaged_fork ( mm , oldmm ) ;
if ( retval )
goto out ;
prev = NULL ;
for ( mpnt = oldmm - > mmap ; mpnt ; mpnt = mpnt - > vm_next ) {
struct file * file ;
if ( mpnt - > vm_flags & VM_DONTCOPY ) {
vm_stat_account ( mm , mpnt - > vm_flags , - vma_pages ( mpnt ) ) ;
continue ;
}
charge = 0 ;
2018-06-15 01:26:34 +03:00
/*
* Don ' t duplicate many vmas if we ' ve been oom - killed ( for
* example )
*/
if ( fatal_signal_pending ( current ) ) {
retval = - EINTR ;
goto out ;
}
2018-02-01 03:15:51 +03:00
if ( mpnt - > vm_flags & VM_ACCOUNT ) {
unsigned long len = vma_pages ( mpnt ) ;
if ( security_vm_enough_memory_mm ( oldmm , len ) ) /* sic */
goto fail_nomem ;
charge = len ;
}
2018-07-21 23:48:51 +03:00
tmp = vm_area_dup ( mpnt ) ;
2018-02-01 03:15:51 +03:00
if ( ! tmp )
goto fail_nomem ;
retval = vma_dup_policy ( mpnt , tmp ) ;
if ( retval )
goto fail_nomem_policy ;
tmp - > vm_mm = mm ;
retval = dup_userfaultfd ( tmp , & uf ) ;
if ( retval )
goto fail_nomem_anon_vma_fork ;
if ( tmp - > vm_flags & VM_WIPEONFORK ) {
mm: don't prepare anon_vma if vma has VM_WIPEONFORK
Patch series "mm: Fix misuse of parent anon_vma in dup_mmap path".
This patchset fixes the misuse of parenet anon_vma, which mainly caused by
child vma's vm_next and vm_prev are left same as its parent after
duplicate vma. Finally, code reached parent vma's neighbor by referring
pointer of child vma and executed wrong logic.
The first two patches fix relevant issues, and the third patch sets
vm_next and vm_prev to NULL when duplicate vma to prevent potential misuse
in future.
Effects of the first bug is that causes rmap code to check both parent and
child's page table, although a page couldn't be mapped by both parent and
child, because child vma has WIPEONFORK so all pages mapped by child are
'new' and not relevant to parent.
Effects of the second bug is that the relationship of anon_vma of parent
and child are totallyconvoluted. It would cause 'son', 'grandson', ...,
etc, to share 'parent' anon_vma, which disobey the design rule of reusing
anon_vma (the rule to be followed is that reusing should among vma of same
process, and vma should not gone through fork).
So, both issues should cause unnecessary rmap walking and have unexpected
complexity.
These two issues would not be directly visible, I used debugging code to
check the anon_vma pointers of parent and child when inspecting the
suspicious implementation of issue #2, then find the problem.
This patch (of 3):
In dup_mmap(), anon_vma_prepare() is called for vma has VM_WIPEONFORK, and
parameter 'tmp' (i.e., the new vma of child) has same ->vm_next and
->vm_prev as its parent vma. That allows anon_vma used by parent been
mistakenly shared by child (find_mergeable_anon_vma() will do this reuse
work).
Besides this issue, call anon_vma_prepare() should be avoided because we
don't copy page for this vma. Preparing anon_vma will be handled during
fault.
Fixes: d2cd9ede6e19 ("mm,fork: introduce MADV_WIPEONFORK")
Signed-off-by: Li Xinhai <lixinhai.lxh@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Link: http://lkml.kernel.org/r/1581150928-3214-2-git-send-email-lixinhai.lxh@gmail.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-04-07 06:03:33 +03:00
/*
* VM_WIPEONFORK gets a clean slate in the child .
* Don ' t prepare anon_vma until fault since we don ' t
* copy page for current vma .
*/
2018-02-01 03:15:51 +03:00
tmp - > anon_vma = NULL ;
} else if ( anon_vma_fork ( tmp , mpnt ) )
goto fail_nomem_anon_vma_fork ;
tmp - > vm_flags & = ~ ( VM_LOCKED | VM_LOCKONFAULT ) ;
file = tmp - > vm_file ;
if ( file ) {
struct inode * inode = file_inode ( file ) ;
struct address_space * mapping = file - > f_mapping ;
get_file ( file ) ;
if ( tmp - > vm_flags & VM_DENYWRITE )
2020-10-16 06:10:08 +03:00
put_write_access ( inode ) ;
2018-02-01 03:15:51 +03:00
i_mmap_lock_write ( mapping ) ;
if ( tmp - > vm_flags & VM_SHARED )
2020-10-14 02:54:10 +03:00
mapping_allow_writable ( mapping ) ;
2018-02-01 03:15:51 +03:00
flush_dcache_mmap_lock ( mapping ) ;
/* insert tmp into the share list, just after mpnt */
vma_interval_tree_insert_after ( tmp , mpnt ,
& mapping - > i_mmap ) ;
flush_dcache_mmap_unlock ( mapping ) ;
i_mmap_unlock_write ( mapping ) ;
}
/*
* Clear hugetlb - related page reserves for children . This only
* affects MAP_PRIVATE mappings . Faults generated by the child
* are not guaranteed to succeed , even if read - only
*/
if ( is_vm_hugetlb_page ( tmp ) )
reset_vma_resv_huge_pages ( tmp ) ;
/*
* Link in the new vma and copy the page table entries .
*/
* pprev = tmp ;
pprev = & tmp - > vm_next ;
tmp - > vm_prev = prev ;
prev = tmp ;
__vma_link_rb ( mm , tmp , rb_link , rb_parent ) ;
rb_link = & tmp - > vm_rb . rb_right ;
rb_parent = & tmp - > vm_rb ;
mm - > map_count + + ;
if ( ! ( tmp - > vm_flags & VM_WIPEONFORK ) )
2020-10-14 02:54:21 +03:00
retval = copy_page_range ( tmp , mpnt ) ;
2018-02-01 03:15:51 +03:00
if ( tmp - > vm_ops & & tmp - > vm_ops - > open )
tmp - > vm_ops - > open ( tmp ) ;
if ( retval )
goto out ;
}
/* a new mm has just been created */
2018-09-05 01:45:41 +03:00
retval = arch_dup_mmap ( oldmm , mm ) ;
2018-02-01 03:15:51 +03:00
out :
2020-06-09 07:33:25 +03:00
mmap_write_unlock ( mm ) ;
2018-02-01 03:15:51 +03:00
flush_tlb_mm ( oldmm ) ;
2020-06-09 07:33:25 +03:00
mmap_write_unlock ( oldmm ) ;
2018-02-01 03:15:51 +03:00
dup_userfaultfd_complete ( & uf ) ;
fail_uprobe_end :
uprobe_end_dup_mmap ( ) ;
return retval ;
fail_nomem_anon_vma_fork :
mpol_put ( vma_policy ( tmp ) ) ;
fail_nomem_policy :
2018-07-21 23:48:51 +03:00
vm_area_free ( tmp ) ;
2018-02-01 03:15:51 +03:00
fail_nomem :
retval = - ENOMEM ;
vm_unacct_memory ( charge ) ;
goto out ;
}
static inline int mm_alloc_pgd ( struct mm_struct * mm )
{
mm - > pgd = pgd_alloc ( mm ) ;
if ( unlikely ( ! mm - > pgd ) )
return - ENOMEM ;
return 0 ;
}
static inline void mm_free_pgd ( struct mm_struct * mm )
{
pgd_free ( mm , mm - > pgd ) ;
}
# else
static int dup_mmap ( struct mm_struct * mm , struct mm_struct * oldmm )
{
2020-06-09 07:33:25 +03:00
mmap_write_lock ( oldmm ) ;
2018-02-01 03:15:51 +03:00
RCU_INIT_POINTER ( mm - > exe_file , get_mm_exe_file ( oldmm ) ) ;
2020-06-09 07:33:25 +03:00
mmap_write_unlock ( oldmm ) ;
2018-02-01 03:15:51 +03:00
return 0 ;
}
# define mm_alloc_pgd(mm) (0)
# define mm_free_pgd(mm)
# endif /* CONFIG_MMU */
static void check_mm ( struct mm_struct * mm )
{
int i ;
2019-09-26 02:47:27 +03:00
BUILD_BUG_ON_MSG ( ARRAY_SIZE ( resident_page_types ) ! = NR_MM_COUNTERS ,
" Please make sure 'struct resident_page_types[]' is updated as well " ) ;
2018-02-01 03:15:51 +03:00
for ( i = 0 ; i < NR_MM_COUNTERS ; i + + ) {
long x = atomic_long_read ( & mm - > rss_stat . count [ i ] ) ;
if ( unlikely ( x ) )
2019-09-26 02:47:27 +03:00
pr_alert ( " BUG: Bad rss-counter state mm:%p type:%s val:%ld \n " ,
mm , resident_page_types [ i ] , x ) ;
2018-02-01 03:15:51 +03:00
}
if ( mm_pgtables_bytes ( mm ) )
pr_alert ( " BUG: non-zero pgtables_bytes on freeing mm: %ld \n " ,
mm_pgtables_bytes ( mm ) ) ;
# if defined(CONFIG_TRANSPARENT_HUGEPAGE) && !USE_SPLIT_PMD_PTLOCKS
VM_BUG_ON_MM ( mm - > pmd_huge_pte , mm ) ;
# endif
}
# define allocate_mm() (kmem_cache_alloc(mm_cachep, GFP_KERNEL))
# define free_mm(mm) (kmem_cache_free(mm_cachep, (mm)))
/*
* Called when the last reference to the mm
* is dropped : either by a lazy thread or by
* mmput . Free the page directory and the mm .
*/
2018-02-22 01:45:17 +03:00
void __mmdrop ( struct mm_struct * mm )
2018-02-01 03:15:51 +03:00
{
BUG_ON ( mm = = & init_mm ) ;
2018-04-06 02:25:12 +03:00
WARN_ON_ONCE ( mm = = current - > mm ) ;
WARN_ON_ONCE ( mm = = current - > active_mm ) ;
2018-02-01 03:15:51 +03:00
mm_free_pgd ( mm ) ;
destroy_context ( mm ) ;
2019-12-18 20:40:35 +03:00
mmu_notifier_subscriptions_destroy ( mm ) ;
2018-02-01 03:15:51 +03:00
check_mm ( mm ) ;
put_user_ns ( mm - > user_ns ) ;
free_mm ( mm ) ;
}
2018-02-22 01:45:17 +03:00
EXPORT_SYMBOL_GPL ( __mmdrop ) ;
2018-02-01 03:15:51 +03:00
static void mmdrop_async_fn ( struct work_struct * work )
{
struct mm_struct * mm ;
mm = container_of ( work , struct mm_struct , async_put_work ) ;
__mmdrop ( mm ) ;
}
static void mmdrop_async ( struct mm_struct * mm )
{
if ( unlikely ( atomic_dec_and_test ( & mm - > mm_count ) ) ) {
INIT_WORK ( & mm - > async_put_work , mmdrop_async_fn ) ;
schedule_work ( & mm - > async_put_work ) ;
}
}
signals: make task_struct->signal immutable/refcountable
We have a lot of problems with accessing task_struct->signal, it can
"disappear" at any moment. Even current can't use its ->signal safely
after exit_notify(). ->siglock helps, but it is not convenient, not
always possible, and sometimes it makes sense to use task->signal even
after this task has already dead.
This patch adds the reference counter, sigcnt, into signal_struct. This
reference is owned by task_struct and it is dropped in
__put_task_struct(). Perhaps it makes sense to export
get/put_signal_struct() later, but currently I don't see the immediate
reason.
Rename __cleanup_signal() to free_signal_struct() and unexport it. With
the previous changes it does nothing except kmem_cache_free().
Change __exit_signal() to not clear/free ->signal, it will be freed when
the last reference to any thread in the thread group goes away.
Note:
- when the last thead exits signal->tty can point to nowhere, see
the next patch.
- with or without this patch signal_struct->count should go away,
or at least it should be "int nr_threads" for fs/proc. This will
be addressed later.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Alan Cox <alan@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-05-27 01:43:16 +04:00
static inline void free_signal_struct ( struct signal_struct * sig )
{
2010-05-27 01:43:20 +04:00
taskstats_tgid_free ( sig ) ;
2011-01-05 13:16:04 +03:00
sched_autogroup_exit ( sig ) ;
2016-10-08 02:58:54 +03:00
/*
* __mmdrop is not safe to call from softirq context on x86 due to
* pgd_dtor so postpone it to the async context
*/
2016-10-08 02:58:51 +03:00
if ( sig - > oom_mm )
2016-10-08 02:58:54 +03:00
mmdrop_async ( sig - > oom_mm ) ;
signals: make task_struct->signal immutable/refcountable
We have a lot of problems with accessing task_struct->signal, it can
"disappear" at any moment. Even current can't use its ->signal safely
after exit_notify(). ->siglock helps, but it is not convenient, not
always possible, and sometimes it makes sense to use task->signal even
after this task has already dead.
This patch adds the reference counter, sigcnt, into signal_struct. This
reference is owned by task_struct and it is dropped in
__put_task_struct(). Perhaps it makes sense to export
get/put_signal_struct() later, but currently I don't see the immediate
reason.
Rename __cleanup_signal() to free_signal_struct() and unexport it. With
the previous changes it does nothing except kmem_cache_free().
Change __exit_signal() to not clear/free ->signal, it will be freed when
the last reference to any thread in the thread group goes away.
Note:
- when the last thead exits signal->tty can point to nowhere, see
the next patch.
- with or without this patch signal_struct->count should go away,
or at least it should be "int nr_threads" for fs/proc. This will
be addressed later.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Alan Cox <alan@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-05-27 01:43:16 +04:00
kmem_cache_free ( signal_cachep , sig ) ;
}
static inline void put_signal_struct ( struct signal_struct * sig )
{
2019-01-18 15:27:27 +03:00
if ( refcount_dec_and_test ( & sig - > sigcnt ) )
signals: make task_struct->signal immutable/refcountable
We have a lot of problems with accessing task_struct->signal, it can
"disappear" at any moment. Even current can't use its ->signal safely
after exit_notify(). ->siglock helps, but it is not convenient, not
always possible, and sometimes it makes sense to use task->signal even
after this task has already dead.
This patch adds the reference counter, sigcnt, into signal_struct. This
reference is owned by task_struct and it is dropped in
__put_task_struct(). Perhaps it makes sense to export
get/put_signal_struct() later, but currently I don't see the immediate
reason.
Rename __cleanup_signal() to free_signal_struct() and unexport it. With
the previous changes it does nothing except kmem_cache_free().
Change __exit_signal() to not clear/free ->signal, it will be freed when
the last reference to any thread in the thread group goes away.
Note:
- when the last thead exits signal->tty can point to nowhere, see
the next patch.
- with or without this patch signal_struct->count should go away,
or at least it should be "int nr_threads" for fs/proc. This will
be addressed later.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Alan Cox <alan@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-05-27 01:43:16 +04:00
free_signal_struct ( sig ) ;
}
2006-03-31 14:31:34 +04:00
void __put_task_struct ( struct task_struct * tsk )
2005-04-17 02:20:36 +04:00
{
2007-10-19 10:40:38 +04:00
WARN_ON ( ! tsk - > exit_state ) ;
2019-01-18 15:27:29 +03:00
WARN_ON ( refcount_read ( & tsk - > usage ) ) ;
2005-04-17 02:20:36 +04:00
WARN_ON ( tsk = = current ) ;
2020-09-13 22:09:39 +03:00
io_uring_free ( tsk ) ;
2015-10-15 23:41:53 +03:00
cgroup_free ( tsk ) ;
2019-07-16 18:20:45 +03:00
task_numa_free ( tsk , true ) ;
2011-12-22 00:17:03 +04:00
security_task_free ( tsk ) ;
2021-02-26 02:43:14 +03:00
bpf_task_storage_free ( tsk ) ;
2009-09-02 12:13:40 +04:00
exit_creds ( tsk ) ;
2006-09-01 08:27:38 +04:00
delayacct_tsk_free ( tsk ) ;
signals: make task_struct->signal immutable/refcountable
We have a lot of problems with accessing task_struct->signal, it can
"disappear" at any moment. Even current can't use its ->signal safely
after exit_notify(). ->siglock helps, but it is not convenient, not
always possible, and sometimes it makes sense to use task->signal even
after this task has already dead.
This patch adds the reference counter, sigcnt, into signal_struct. This
reference is owned by task_struct and it is dropped in
__put_task_struct(). Perhaps it makes sense to export
get/put_signal_struct() later, but currently I don't see the immediate
reason.
Rename __cleanup_signal() to free_signal_struct() and unexport it. With
the previous changes it does nothing except kmem_cache_free().
Change __exit_signal() to not clear/free ->signal, it will be freed when
the last reference to any thread in the thread group goes away.
Note:
- when the last thead exits signal->tty can point to nowhere, see
the next patch.
- with or without this patch signal_struct->count should go away,
or at least it should be "int nr_threads" for fs/proc. This will
be addressed later.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Alan Cox <alan@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-05-27 01:43:16 +04:00
put_signal_struct ( tsk - > signal ) ;
2021-03-26 20:55:06 +03:00
sched_core_free ( tsk ) ;
2005-04-17 02:20:36 +04:00
if ( ! profile_handoff_task ( tsk ) )
free_task ( tsk ) ;
}
2011-02-01 17:51:46 +03:00
EXPORT_SYMBOL_GPL ( __put_task_struct ) ;
2005-04-17 02:20:36 +04:00
2012-05-05 19:05:40 +04:00
void __init __weak arch_task_cache_init ( void ) { }
2008-03-11 01:28:04 +03:00
2015-04-16 22:47:44 +03:00
/*
* set_max_threads
*/
2015-04-16 22:47:50 +03:00
static void set_max_threads ( unsigned int max_threads_suggested )
2015-04-16 22:47:44 +03:00
{
2015-04-16 22:47:47 +03:00
u64 threads ;
2018-12-28 11:34:29 +03:00
unsigned long nr_pages = totalram_pages ( ) ;
2015-04-16 22:47:44 +03:00
/*
2015-04-16 22:47:47 +03:00
* The number of threads shall be limited such that the thread
* structures may only consume a small part of the available memory .
2015-04-16 22:47:44 +03:00
*/
2018-12-28 11:34:20 +03:00
if ( fls64 ( nr_pages ) + fls64 ( PAGE_SIZE ) > 64 )
2015-04-16 22:47:47 +03:00
threads = MAX_THREADS ;
else
2018-12-28 11:34:20 +03:00
threads = div64_u64 ( ( u64 ) nr_pages * ( u64 ) PAGE_SIZE ,
2015-04-16 22:47:47 +03:00
( u64 ) THREAD_SIZE * 8UL ) ;
2015-04-16 22:47:50 +03:00
if ( threads > max_threads_suggested )
threads = max_threads_suggested ;
2015-04-16 22:47:47 +03:00
max_threads = clamp_t ( u64 , threads , MIN_THREADS , MAX_THREADS ) ;
2015-04-16 22:47:44 +03:00
}
2015-07-17 13:28:12 +03:00
# ifdef CONFIG_ARCH_WANTS_DYNAMIC_TASK_STRUCT
/* Initialized by the architecture: */
int arch_task_struct_size __read_mostly ;
# endif
2015-07-17 13:28:11 +03:00
2019-08-12 09:55:24 +03:00
# ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR
2017-08-16 23:00:58 +03:00
static void task_struct_whitelist ( unsigned long * offset , unsigned long * size )
{
/* Fetch thread_struct whitelist for the architecture. */
arch_thread_struct_whitelist ( offset , size ) ;
/*
* Handle zero - sized whitelist or empty thread_struct , otherwise
* adjust offset to position of thread_struct in task_struct .
*/
if ( unlikely ( * size = = 0 ) )
* offset = 0 ;
else
* offset + = offsetof ( struct task_struct , thread ) ;
}
2019-08-12 09:55:24 +03:00
# endif /* CONFIG_ARCH_TASK_STRUCT_ALLOCATOR */
2017-08-16 23:00:58 +03:00
2015-04-16 22:47:44 +03:00
void __init fork_init ( void )
2005-04-17 02:20:36 +04:00
{
2016-08-08 22:41:52 +03:00
int i ;
2012-05-05 19:05:48 +04:00
# ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR
2005-04-17 02:20:36 +04:00
# ifndef ARCH_MIN_TASKALIGN
2017-01-11 16:17:48 +03:00
# define ARCH_MIN_TASKALIGN 0
2005-04-17 02:20:36 +04:00
# endif
2017-02-18 17:26:45 +03:00
int align = max_t ( int , L1_CACHE_BYTES , ARCH_MIN_TASKALIGN ) ;
2017-08-16 23:00:58 +03:00
unsigned long useroffset , usersize ;
2017-01-11 16:17:48 +03:00
2005-04-17 02:20:36 +04:00
/* create a slab on which task_structs can be allocated */
2017-08-16 23:00:58 +03:00
task_struct_whitelist ( & useroffset , & usersize ) ;
task_struct_cachep = kmem_cache_create_usercopy ( " task_struct " ,
2017-01-11 16:17:48 +03:00
arch_task_struct_size , align ,
2017-08-16 23:00:58 +03:00
SLAB_PANIC | SLAB_ACCOUNT ,
useroffset , usersize , NULL ) ;
2005-04-17 02:20:36 +04:00
# endif
2008-03-11 01:28:04 +03:00
/* do the arch specific task caches init */
arch_task_cache_init ( ) ;
2015-04-16 22:47:50 +03:00
set_max_threads ( MAX_THREADS ) ;
2005-04-17 02:20:36 +04:00
init_task . signal - > rlim [ RLIMIT_NPROC ] . rlim_cur = max_threads / 2 ;
init_task . signal - > rlim [ RLIMIT_NPROC ] . rlim_max = max_threads / 2 ;
init_task . signal - > rlim [ RLIMIT_SIGPENDING ] =
init_task . signal - > rlim [ RLIMIT_NPROC ] ;
2016-08-08 21:41:24 +03:00
2021-04-22 15:27:11 +03:00
for ( i = 0 ; i < MAX_PER_NAMESPACE_UCOUNTS ; i + + )
2016-08-08 22:41:52 +03:00
init_user_ns . ucount_max [ i ] = max_threads / 2 ;
2017-05-09 01:56:11 +03:00
ucounts: Fix regression preventing increasing of rlimits in init_user_ns
"Ma, XinjianX" <xinjianx.ma@intel.com> reported:
> When lkp team run kernel selftests, we found after these series of patches, testcase mqueue: mq_perf_tests
> in kselftest failed with following message.
>
> # selftests: mqueue: mq_perf_tests
> #
> # Initial system state:
> # Using queue path: /mq_perf_tests
> # RLIMIT_MSGQUEUE(soft): 819200
> # RLIMIT_MSGQUEUE(hard): 819200
> # Maximum Message Size: 8192
> # Maximum Queue Size: 10
> # Nice value: 0
> #
> # Adjusted system state for testing:
> # RLIMIT_MSGQUEUE(soft): (unlimited)
> # RLIMIT_MSGQUEUE(hard): (unlimited)
> # Maximum Message Size: 16777216
> # Maximum Queue Size: 65530
> # Nice value: -20
> # Continuous mode: (disabled)
> # CPUs to pin: 3
> # ./mq_perf_tests: mq_open() at 296: Too many open files
> not ok 2 selftests: mqueue: mq_perf_tests # exit=1
> ```
>
> Test env:
> rootfs: debian-10
> gcc version: 9
After investigation the problem turned out to be that ucount_max for
the rlimits in init_user_ns was being set to the initial rlimit value.
The practical problem is that ucount_max provides a limit that
applications inside the user namespace can not exceed. Which means in
practice that rlimits that have been converted to use the ucount
infrastructure were not able to exceend their initial rlimits.
Solve this by setting the relevant values of ucount_max to
RLIM_INIFINITY. A limit in init_user_ns is pointless so the code
should allow the values to grow as large as possible without riscking
an underflow or an overflow.
As the ltp test case was a bit of a pain I have reproduced the rlimit failure
and tested the fix with the following little C program:
> #include <stdio.h>
> #include <fcntl.h>
> #include <sys/stat.h>
> #include <mqueue.h>
> #include <sys/time.h>
> #include <sys/resource.h>
> #include <errno.h>
> #include <string.h>
> #include <stdlib.h>
> #include <limits.h>
> #include <unistd.h>
>
> int main(int argc, char **argv)
> {
> struct mq_attr mq_attr;
> struct rlimit rlim;
> mqd_t mqd;
> int ret;
>
> ret = getrlimit(RLIMIT_MSGQUEUE, &rlim);
> if (ret != 0) {
> fprintf(stderr, "getrlimit(RLIMIT_MSGQUEUE) failed: %s\n", strerror(errno));
> exit(EXIT_FAILURE);
> }
> printf("RLIMIT_MSGQUEUE %lu %lu\n",
> rlim.rlim_cur, rlim.rlim_max);
> rlim.rlim_cur = RLIM_INFINITY;
> rlim.rlim_max = RLIM_INFINITY;
> ret = setrlimit(RLIMIT_MSGQUEUE, &rlim);
> if (ret != 0) {
> fprintf(stderr, "setrlimit(RLIMIT_MSGQUEUE, RLIM_INFINITY) failed: %s\n", strerror(errno));
> exit(EXIT_FAILURE);
> }
>
> memset(&mq_attr, 0, sizeof(struct mq_attr));
> mq_attr.mq_maxmsg = 65536 - 1;
> mq_attr.mq_msgsize = 16*1024*1024 - 1;
>
> mqd = mq_open("/mq_rlimit_test", O_RDONLY|O_CREAT, 0600, &mq_attr);
> if (mqd == (mqd_t)-1) {
> fprintf(stderr, "mq_open failed: %s\n", strerror(errno));
> exit(EXIT_FAILURE);
> }
> ret = mq_close(mqd);
> if (ret) {
> fprintf(stderr, "mq_close failed; %s\n", strerror(errno));
> exit(EXIT_FAILURE);
> }
>
> return EXIT_SUCCESS;
> }
Fixes: 6e52a9f0532f ("Reimplement RLIMIT_MSGQUEUE on top of ucounts")
Fixes: d7c9e99aee48 ("Reimplement RLIMIT_MEMLOCK on top of ucounts")
Fixes: d64696905554 ("Reimplement RLIMIT_SIGPENDING on top of ucounts")
Fixes: 21d1c5e386bc ("Reimplement RLIMIT_NPROC on top of ucounts")
Reported-by: kernel test robot lkp@intel.com
Acked-by: Alexey Gladkov <legion@kernel.org>
Link: https://lkml.kernel.org/r/87eeajswfc.fsf_-_@disp2133
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
2021-08-23 19:12:17 +03:00
set_rlimit_ucount_max ( & init_user_ns , UCOUNT_RLIMIT_NPROC , RLIM_INFINITY ) ;
set_rlimit_ucount_max ( & init_user_ns , UCOUNT_RLIMIT_MSGQUEUE , RLIM_INFINITY ) ;
set_rlimit_ucount_max ( & init_user_ns , UCOUNT_RLIMIT_SIGPENDING , RLIM_INFINITY ) ;
set_rlimit_ucount_max ( & init_user_ns , UCOUNT_RLIMIT_MEMLOCK , RLIM_INFINITY ) ;
2021-04-22 15:27:11 +03:00
2017-05-09 01:56:11 +03:00
# ifdef CONFIG_VMAP_STACK
cpuhp_setup_state ( CPUHP_BP_PREPARE_DYN , " fork:vm_stack_cache " ,
NULL , free_vm_stack_cache ) ;
# endif
2017-08-07 10:12:52 +03:00
2020-04-27 19:00:07 +03:00
scs_init ( ) ;
2017-08-07 10:12:52 +03:00
lockdep_init_task ( & init_task ) ;
2019-04-27 02:22:44 +03:00
uprobes_init ( ) ;
2005-04-17 02:20:36 +04:00
}
2014-04-08 02:39:20 +04:00
int __weak arch_dup_task_struct ( struct task_struct * dst ,
2008-03-11 01:28:04 +03:00
struct task_struct * src )
{
* dst = * src ;
return 0 ;
}
2014-09-12 17:16:17 +04:00
void set_task_stack_end_magic ( struct task_struct * tsk )
{
unsigned long * stackend ;
stackend = end_of_stack ( tsk ) ;
* stackend = STACK_END_MAGIC ; /* for overflow detection */
}
2016-05-24 02:24:05 +03:00
static struct task_struct * dup_task_struct ( struct task_struct * orig , int node )
2005-04-17 02:20:36 +04:00
{
struct task_struct * tsk ;
Clarify naming of thread info/stack allocators
We've had the thread info allocated together with the thread stack for
most architectures for a long time (since the thread_info was split off
from the task struct), but that is about to change.
But the patches that move the thread info to be off-stack (and a part of
the task struct instead) made it clear how confused the allocator and
freeing functions are.
Because the common case was that we share an allocation with the thread
stack and the thread_info, the two pointers were identical. That
identity then meant that we would have things like
ti = alloc_thread_info_node(tsk, node);
...
tsk->stack = ti;
which certainly _worked_ (since stack and thread_info have the same
value), but is rather confusing: why are we assigning a thread_info to
the stack? And if we move the thread_info away, the "confusing" code
just gets to be entirely bogus.
So remove all this confusion, and make it clear that we are doing the
stack allocation by renaming and clarifying the function names to be
about the stack. The fact that the thread_info then shares the
allocation is an implementation detail, and not really about the
allocation itself.
This is a pure renaming and type fix: we pass in the same pointer, it's
just that we clarify what the pointer means.
The ia64 code that actually only has one single allocation (for all of
task_struct, thread_info and kernel thread stack) now looks a bit odd,
but since "tsk->stack" is actually not even used there, that oddity
doesn't matter. It would be a separate thing to clean that up, I
intentionally left the ia64 changes as a pure brute-force renaming and
type change.
Acked-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-06-25 01:09:37 +03:00
unsigned long * stack ;
2018-12-28 11:40:00 +03:00
struct vm_struct * stack_vm_area __maybe_unused ;
2007-10-17 10:25:50 +04:00
int err ;
2005-04-17 02:20:36 +04:00
2016-05-24 02:24:05 +03:00
if ( node = = NUMA_NO_NODE )
node = tsk_fork_get_node ( orig ) ;
2011-03-23 02:30:41 +03:00
tsk = alloc_task_struct_node ( node ) ;
2005-04-17 02:20:36 +04:00
if ( ! tsk )
return NULL ;
Clarify naming of thread info/stack allocators
We've had the thread info allocated together with the thread stack for
most architectures for a long time (since the thread_info was split off
from the task struct), but that is about to change.
But the patches that move the thread info to be off-stack (and a part of
the task struct instead) made it clear how confused the allocator and
freeing functions are.
Because the common case was that we share an allocation with the thread
stack and the thread_info, the two pointers were identical. That
identity then meant that we would have things like
ti = alloc_thread_info_node(tsk, node);
...
tsk->stack = ti;
which certainly _worked_ (since stack and thread_info have the same
value), but is rather confusing: why are we assigning a thread_info to
the stack? And if we move the thread_info away, the "confusing" code
just gets to be entirely bogus.
So remove all this confusion, and make it clear that we are doing the
stack allocation by renaming and clarifying the function names to be
about the stack. The fact that the thread_info then shares the
allocation is an implementation detail, and not really about the
allocation itself.
This is a pure renaming and type fix: we pass in the same pointer, it's
just that we clarify what the pointer means.
The ia64 code that actually only has one single allocation (for all of
task_struct, thread_info and kernel thread stack) now looks a bit odd,
but since "tsk->stack" is actually not even used there, that oddity
doesn't matter. It would be a separate thing to clean that up, I
intentionally left the ia64 changes as a pure brute-force renaming and
type change.
Acked-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-06-25 01:09:37 +03:00
stack = alloc_thread_stack_node ( tsk , node ) ;
if ( ! stack )
fork: fix error handling in dup_task()
The function dup_task() may fail at the following function calls in the
following order.
0) alloc_task_struct_node()
1) alloc_thread_info_node()
2) arch_dup_task_struct()
Error by 0) is not a matter, it can just return. But error by 1) requires
releasing task_struct allocated by 0) before it returns. Likewise, error
by 2) requires releasing task_struct and thread_info allocated by 0) and
1).
The existing error handling calls free_task_struct() and
free_thread_info() which do not only release task_struct and thread_info,
but also call architecture specific arch_release_task_struct() and
arch_release_thread_info().
The problem is that task_struct and thread_info are not fully initialized
yet at this point, but arch_release_task_struct() and
arch_release_thread_info() are called with them.
For example, x86 defines its own arch_release_task_struct() that releases
a task_xstate. If alloc_thread_info_node() fails in dup_task(),
arch_release_task_struct() is called with task_struct which is just
allocated and filled with garbage in this error handling.
This actually happened with tools/testing/fault-injection/failcmd.sh
# env FAILCMD_TYPE=fail_page_alloc \
./tools/testing/fault-injection/failcmd.sh --times=100 \
--min-order=0 --ignore-gfp-wait=0 \
-- make -C tools/testing/selftests/ run_tests
In order to fix this issue, make free_{task_struct,thread_info}() not to
call arch_release_{task_struct,thread_info}() and call
arch_release_{task_struct,thread_info}() implicitly where needed.
Default arch_release_task_struct() and arch_release_thread_info() are
defined as empty by default. So this change only affects the
architectures which implement their own arch_release_task_struct() or
arch_release_thread_info() as listed below.
arch_release_task_struct(): x86, sh
arch_release_thread_info(): mn10300, tile
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Salman Qazi <sqazi@google.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 01:42:33 +04:00
goto free_tsk ;
2005-04-17 02:20:36 +04:00
2018-10-27 01:03:19 +03:00
if ( memcg_charge_kernel_stack ( tsk ) )
goto free_stack ;
2016-08-11 12:35:21 +03:00
stack_vm_area = task_stack_vm_area ( tsk ) ;
2011-07-27 03:08:39 +04:00
err = arch_dup_task_struct ( tsk , orig ) ;
2016-08-11 12:35:21 +03:00
/*
* arch_dup_task_struct ( ) clobbers the stack - related fields . Make
* sure they ' re properly initialized before using any stack - related
* functions again .
*/
tsk - > stack = stack ;
# ifdef CONFIG_VMAP_STACK
tsk - > stack_vm_area = stack_vm_area ;
# endif
2016-09-16 08:45:48 +03:00
# ifdef CONFIG_THREAD_INFO_IN_TASK
2019-01-18 15:27:30 +03:00
refcount_set ( & tsk - > stack_refcount , 1 ) ;
2016-09-16 08:45:48 +03:00
# endif
2016-08-11 12:35:21 +03:00
2012-06-26 05:18:15 +04:00
if ( err )
Clarify naming of thread info/stack allocators
We've had the thread info allocated together with the thread stack for
most architectures for a long time (since the thread_info was split off
from the task struct), but that is about to change.
But the patches that move the thread info to be off-stack (and a part of
the task struct instead) made it clear how confused the allocator and
freeing functions are.
Because the common case was that we share an allocation with the thread
stack and the thread_info, the two pointers were identical. That
identity then meant that we would have things like
ti = alloc_thread_info_node(tsk, node);
...
tsk->stack = ti;
which certainly _worked_ (since stack and thread_info have the same
value), but is rather confusing: why are we assigning a thread_info to
the stack? And if we move the thread_info away, the "confusing" code
just gets to be entirely bogus.
So remove all this confusion, and make it clear that we are doing the
stack allocation by renaming and clarifying the function names to be
about the stack. The fact that the thread_info then shares the
allocation is an implementation detail, and not really about the
allocation itself.
This is a pure renaming and type fix: we pass in the same pointer, it's
just that we clarify what the pointer means.
The ia64 code that actually only has one single allocation (for all of
task_struct, thread_info and kernel thread stack) now looks a bit odd,
but since "tsk->stack" is actually not even used there, that oddity
doesn't matter. It would be a separate thing to clean that up, I
intentionally left the ia64 changes as a pure brute-force renaming and
type change.
Acked-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-06-25 01:09:37 +03:00
goto free_stack ;
2012-06-26 05:18:15 +04:00
2020-04-27 19:00:07 +03:00
err = scs_prepare ( tsk , node ) ;
if ( err )
goto free_stack ;
2014-06-28 02:18:48 +04:00
# ifdef CONFIG_SECCOMP
/*
* We must handle setting up seccomp filters once we ' re under
* the sighand lock in case orig has changed between now and
* then . Until then , filter must be NULL to avoid messing up
* the usage counts on the error path calling free_task .
*/
tsk - > seccomp . filter = NULL ;
# endif
2012-07-31 01:42:31 +04:00
setup_thread_stack ( tsk , orig ) ;
2009-11-29 17:34:48 +03:00
clear_user_return_notifier ( tsk ) ;
2010-12-08 13:05:42 +03:00
clear_tsk_need_resched ( tsk ) ;
2014-09-12 17:16:17 +04:00
set_task_stack_end_magic ( tsk ) ;
kernel: Implement selective syscall userspace redirection
Introduce a mechanism to quickly disable/enable syscall handling for a
specific process and redirect to userspace via SIGSYS. This is useful
for processes with parts that require syscall redirection and parts that
don't, but who need to perform this boundary crossing really fast,
without paying the cost of a system call to reconfigure syscall handling
on each boundary transition. This is particularly important for Windows
games running over Wine.
The proposed interface looks like this:
prctl(PR_SET_SYSCALL_USER_DISPATCH, <op>, <off>, <length>, [selector])
The range [<offset>,<offset>+<length>) is a part of the process memory
map that is allowed to by-pass the redirection code and dispatch
syscalls directly, such that in fast paths a process doesn't need to
disable the trap nor the kernel has to check the selector. This is
essential to return from SIGSYS to a blocked area without triggering
another SIGSYS from rt_sigreturn.
selector is an optional pointer to a char-sized userspace memory region
that has a key switch for the mechanism. This key switch is set to
either PR_SYS_DISPATCH_ON, PR_SYS_DISPATCH_OFF to enable and disable the
redirection without calling the kernel.
The feature is meant to be set per-thread and it is disabled on
fork/clone/execv.
Internally, this doesn't add overhead to the syscall hot path, and it
requires very little per-architecture support. I avoided using seccomp,
even though it duplicates some functionality, due to previous feedback
that maybe it shouldn't mix with seccomp since it is not a security
mechanism. And obviously, this should never be considered a security
mechanism, since any part of the program can by-pass it by using the
syscall dispatcher.
For the sysinfo benchmark, which measures the overhead added to
executing a native syscall that doesn't require interception, the
overhead using only the direct dispatcher region to issue syscalls is
pretty much irrelevant. The overhead of using the selector goes around
40ns for a native (unredirected) syscall in my system, and it is (as
expected) dominated by the supervisor-mode user-address access. In
fact, with SMAP off, the overhead is consistently less than 5ns on my
test box.
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Andy Lutomirski <luto@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20201127193238.821364-4-krisman@collabora.com
2020-11-27 22:32:34 +03:00
clear_syscall_work_syscall_user_dispatch ( tsk ) ;
2005-04-17 02:20:36 +04:00
Kbuild: rename CC_STACKPROTECTOR[_STRONG] config variables
The changes to automatically test for working stack protector compiler
support in the Kconfig files removed the special STACKPROTECTOR_AUTO
option that picked the strongest stack protector that the compiler
supported.
That was all a nice cleanup - it makes no sense to have the AUTO case
now that the Kconfig phase can just determine the compiler support
directly.
HOWEVER.
It also meant that doing "make oldconfig" would now _disable_ the strong
stackprotector if you had AUTO enabled, because in a legacy config file,
the sane stack protector configuration would look like
CONFIG_HAVE_CC_STACKPROTECTOR=y
# CONFIG_CC_STACKPROTECTOR_NONE is not set
# CONFIG_CC_STACKPROTECTOR_REGULAR is not set
# CONFIG_CC_STACKPROTECTOR_STRONG is not set
CONFIG_CC_STACKPROTECTOR_AUTO=y
and when you ran this through "make oldconfig" with the Kbuild changes,
it would ask you about the regular CONFIG_CC_STACKPROTECTOR (that had
been renamed from CONFIG_CC_STACKPROTECTOR_REGULAR to just
CONFIG_CC_STACKPROTECTOR), but it would think that the STRONG version
used to be disabled (because it was really enabled by AUTO), and would
disable it in the new config, resulting in:
CONFIG_HAVE_CC_STACKPROTECTOR=y
CONFIG_CC_HAS_STACKPROTECTOR_NONE=y
CONFIG_CC_STACKPROTECTOR=y
# CONFIG_CC_STACKPROTECTOR_STRONG is not set
CONFIG_CC_HAS_SANE_STACKPROTECTOR=y
That's dangerously subtle - people could suddenly find themselves with
the weaker stack protector setup without even realizing.
The solution here is to just rename not just the old RECULAR stack
protector option, but also the strong one. This does that by just
removing the CC_ prefix entirely for the user choices, because it really
is not about the compiler support (the compiler support now instead
automatially impacts _visibility_ of the options to users).
This results in "make oldconfig" actually asking the user for their
choice, so that we don't have any silent subtle security model changes.
The end result would generally look like this:
CONFIG_HAVE_CC_STACKPROTECTOR=y
CONFIG_CC_HAS_STACKPROTECTOR_NONE=y
CONFIG_STACKPROTECTOR=y
CONFIG_STACKPROTECTOR_STRONG=y
CONFIG_CC_HAS_SANE_STACKPROTECTOR=y
where the "CC_" versions really are about internal compiler
infrastructure, not the user selections.
Acked-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-06-14 06:21:18 +03:00
# ifdef CONFIG_STACKPROTECTOR
2017-07-13 00:36:20 +03:00
tsk - > stack_canary = get_random_canary ( ) ;
2006-09-26 12:52:38 +04:00
# endif
2019-04-23 17:26:36 +03:00
if ( orig - > cpus_ptr = = & orig - > cpus_mask )
tsk - > cpus_ptr = & tsk - > cpus_mask ;
2006-09-26 12:52:38 +04:00
2011-07-27 03:08:39 +04:00
/*
2019-09-14 15:33:58 +03:00
* One for the user space visible state that goes away when reaped .
* One for the scheduler .
2011-07-27 03:08:39 +04:00
*/
2019-09-14 15:33:58 +03:00
refcount_set ( & tsk - > rcu_users , 2 ) ;
/* One for the rcu users */
refcount_set ( & tsk - > usage , 1 ) ;
2006-09-29 12:59:40 +04:00
# ifdef CONFIG_BLK_DEV_IO_TRACE
2006-03-23 22:00:26 +03:00
tsk - > btrace_seq = 0 ;
2006-09-29 12:59:40 +04:00
# endif
2006-04-20 15:05:33 +04:00
tsk - > splice_pipe = NULL ;
net: use a per task frag allocator
We currently use a per socket order-0 page cache for tcp_sendmsg()
operations.
This page is used to build fragments for skbs.
Its done to increase probability of coalescing small write() into
single segments in skbs still in write queue (not yet sent)
But it wastes a lot of memory for applications handling many mostly
idle sockets, since each socket holds one page in sk->sk_sndmsg_page
Its also quite inefficient to build TSO 64KB packets, because we need
about 16 pages per skb on arches where PAGE_SIZE = 4096, so we hit
page allocator more than wanted.
This patch adds a per task frag allocator and uses bigger pages,
if available. An automatic fallback is done in case of memory pressure.
(up to 32768 bytes per frag, thats order-3 pages on x86)
This increases TCP stream performance by 20% on loopback device,
but also benefits on other network devices, since 8x less frags are
mapped on transmit and unmapped on tx completion. Alexander Duyck
mentioned a probable performance win on systems with IOMMU enabled.
Its possible some SG enabled hardware cant cope with bigger fragments,
but their ndo_start_xmit() should already handle this, splitting a
fragment in sub fragments, since some arches have PAGE_SIZE=65536
Successfully tested on various ethernet devices.
(ixgbe, igb, bnx2x, tg3, mellanox mlx4)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>
Cc: Vijay Subramanian <subramanian.vijay@gmail.com>
Cc: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Vijay Subramanian <subramanian.vijay@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-24 03:04:42 +04:00
tsk - > task_frag . page = NULL ;
2015-12-21 20:17:10 +03:00
tsk - > wake_q . next = NULL ;
2021-04-25 02:26:03 +03:00
tsk - > pf_io_worker = NULL ;
2009-09-22 04:01:32 +04:00
2016-08-11 12:35:21 +03:00
account_kernel_stack ( tsk , 1 ) ;
2009-09-22 04:01:32 +04:00
kernel: add kcov code coverage
kcov provides code coverage collection for coverage-guided fuzzing
(randomized testing). Coverage-guided fuzzing is a testing technique
that uses coverage feedback to determine new interesting inputs to a
system. A notable user-space example is AFL
(http://lcamtuf.coredump.cx/afl/). However, this technique is not
widely used for kernel testing due to missing compiler and kernel
support.
kcov does not aim to collect as much coverage as possible. It aims to
collect more or less stable coverage that is function of syscall inputs.
To achieve this goal it does not collect coverage in soft/hard
interrupts and instrumentation of some inherently non-deterministic or
non-interesting parts of kernel is disbled (e.g. scheduler, locking).
Currently there is a single coverage collection mode (tracing), but the
API anticipates additional collection modes. Initially I also
implemented a second mode which exposes coverage in a fixed-size hash
table of counters (what Quentin used in his original patch). I've
dropped the second mode for simplicity.
This patch adds the necessary support on kernel side. The complimentary
compiler support was added in gcc revision 231296.
We've used this support to build syzkaller system call fuzzer, which has
found 90 kernel bugs in just 2 months:
https://github.com/google/syzkaller/wiki/Found-Bugs
We've also found 30+ bugs in our internal systems with syzkaller.
Another (yet unexplored) direction where kcov coverage would greatly
help is more traditional "blob mutation". For example, mounting a
random blob as a filesystem, or receiving a random blob over wire.
Why not gcov. Typical fuzzing loop looks as follows: (1) reset
coverage, (2) execute a bit of code, (3) collect coverage, repeat. A
typical coverage can be just a dozen of basic blocks (e.g. an invalid
input). In such context gcov becomes prohibitively expensive as
reset/collect coverage steps depend on total number of basic
blocks/edges in program (in case of kernel it is about 2M). Cost of
kcov depends only on number of executed basic blocks/edges. On top of
that, kernel requires per-thread coverage because there are always
background threads and unrelated processes that also produce coverage.
With inlined gcov instrumentation per-thread coverage is not possible.
kcov exposes kernel PCs and control flow to user-space which is
insecure. But debugfs should not be mapped as user accessible.
Based on a patch by Quentin Casasnovas.
[akpm@linux-foundation.org: make task_struct.kcov_mode have type `enum kcov_mode']
[akpm@linux-foundation.org: unbreak allmodconfig]
[akpm@linux-foundation.org: follow x86 Makefile layout standards]
Signed-off-by: Dmitry Vyukov <dvyukov@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Cc: syzkaller <syzkaller@googlegroups.com>
Cc: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Tavis Ormandy <taviso@google.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Quentin Casasnovas <quentin.casasnovas@oracle.com>
Cc: Kostya Serebryany <kcc@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@google.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Sasha Levin <sasha.levin@oracle.com>
Cc: David Drysdale <drysdale@google.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Kirill A. Shutemov <kirill@shutemov.name>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-23 00:27:30 +03:00
kcov_task_init ( tsk ) ;
2020-11-18 22:48:43 +03:00
kmap_local_fork ( tsk ) ;
kernel: add kcov code coverage
kcov provides code coverage collection for coverage-guided fuzzing
(randomized testing). Coverage-guided fuzzing is a testing technique
that uses coverage feedback to determine new interesting inputs to a
system. A notable user-space example is AFL
(http://lcamtuf.coredump.cx/afl/). However, this technique is not
widely used for kernel testing due to missing compiler and kernel
support.
kcov does not aim to collect as much coverage as possible. It aims to
collect more or less stable coverage that is function of syscall inputs.
To achieve this goal it does not collect coverage in soft/hard
interrupts and instrumentation of some inherently non-deterministic or
non-interesting parts of kernel is disbled (e.g. scheduler, locking).
Currently there is a single coverage collection mode (tracing), but the
API anticipates additional collection modes. Initially I also
implemented a second mode which exposes coverage in a fixed-size hash
table of counters (what Quentin used in his original patch). I've
dropped the second mode for simplicity.
This patch adds the necessary support on kernel side. The complimentary
compiler support was added in gcc revision 231296.
We've used this support to build syzkaller system call fuzzer, which has
found 90 kernel bugs in just 2 months:
https://github.com/google/syzkaller/wiki/Found-Bugs
We've also found 30+ bugs in our internal systems with syzkaller.
Another (yet unexplored) direction where kcov coverage would greatly
help is more traditional "blob mutation". For example, mounting a
random blob as a filesystem, or receiving a random blob over wire.
Why not gcov. Typical fuzzing loop looks as follows: (1) reset
coverage, (2) execute a bit of code, (3) collect coverage, repeat. A
typical coverage can be just a dozen of basic blocks (e.g. an invalid
input). In such context gcov becomes prohibitively expensive as
reset/collect coverage steps depend on total number of basic
blocks/edges in program (in case of kernel it is about 2M). Cost of
kcov depends only on number of executed basic blocks/edges. On top of
that, kernel requires per-thread coverage because there are always
background threads and unrelated processes that also produce coverage.
With inlined gcov instrumentation per-thread coverage is not possible.
kcov exposes kernel PCs and control flow to user-space which is
insecure. But debugfs should not be mapped as user accessible.
Based on a patch by Quentin Casasnovas.
[akpm@linux-foundation.org: make task_struct.kcov_mode have type `enum kcov_mode']
[akpm@linux-foundation.org: unbreak allmodconfig]
[akpm@linux-foundation.org: follow x86 Makefile layout standards]
Signed-off-by: Dmitry Vyukov <dvyukov@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Cc: syzkaller <syzkaller@googlegroups.com>
Cc: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Tavis Ormandy <taviso@google.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Quentin Casasnovas <quentin.casasnovas@oracle.com>
Cc: Kostya Serebryany <kcc@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@google.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Sasha Levin <sasha.levin@oracle.com>
Cc: David Drysdale <drysdale@google.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Kirill A. Shutemov <kirill@shutemov.name>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-23 00:27:30 +03:00
fault-inject: support systematic fault injection
Add /proc/self/task/<current-tid>/fail-nth file that allows failing
0-th, 1-st, 2-nd and so on calls systematically.
Excerpt from the added documentation:
"Write to this file of integer N makes N-th call in the current task
fail (N is 0-based). Read from this file returns a single char 'Y' or
'N' that says if the fault setup with a previous write to this file
was injected or not, and disables the fault if it wasn't yet injected.
Note that this file enables all types of faults (slab, futex, etc).
This setting takes precedence over all other generic settings like
probability, interval, times, etc. But per-capability settings (e.g.
fail_futex/ignore-private) take precedence over it. This feature is
intended for systematic testing of faults in a single system call. See
an example below"
Why add a new setting:
1. Existing settings are global rather than per-task.
So parallel testing is not possible.
2. attr->interval is close but it depends on attr->count
which is non reset to 0, so interval does not work as expected.
3. Trying to model this with existing settings requires manipulations
of all of probability, interval, times, space, task-filter and
unexposed count and per-task make-it-fail files.
4. Existing settings are per-failure-type, and the set of failure
types is potentially expanding.
5. make-it-fail can't be changed by unprivileged user and aggressive
stress testing better be done from an unprivileged user.
Similarly, this would require opening the debugfs files to the
unprivileged user, as he would need to reopen at least times file
(not possible to pre-open before dropping privs).
The proposed interface solves all of the above (see the example).
We want to integrate this into syzkaller fuzzer. A prototype has found
10 bugs in kernel in first day of usage:
https://groups.google.com/forum/#!searchin/syzkaller/%22FAULT_INJECTION%22%7Csort:relevance
I've made the current interface work with all types of our sandboxes.
For setuid the secret sauce was prctl(PR_SET_DUMPABLE, 1, 0, 0, 0) to
make /proc entries non-root owned. So I am fine with the current
version of the code.
[akpm@linux-foundation.org: fix build]
Link: http://lkml.kernel.org/r/20170328130128.101773-1-dvyukov@google.com
Signed-off-by: Dmitry Vyukov <dvyukov@google.com>
Cc: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-07-13 00:34:35 +03:00
# ifdef CONFIG_FAULT_INJECTION
tsk - > fail_nth = 0 ;
# endif
2018-07-31 19:39:04 +03:00
# ifdef CONFIG_BLK_CGROUP
tsk - > throttle_queue = NULL ;
tsk - > use_memdelay = 0 ;
# endif
fs: fsnotify: account fsnotify metadata to kmemcg
Patch series "Directed kmem charging", v8.
The Linux kernel's memory cgroup allows limiting the memory usage of the
jobs running on the system to provide isolation between the jobs. All
the kernel memory allocated in the context of the job and marked with
__GFP_ACCOUNT will also be included in the memory usage and be limited
by the job's limit.
The kernel memory can only be charged to the memcg of the process in
whose context kernel memory was allocated. However there are cases
where the allocated kernel memory should be charged to the memcg
different from the current processes's memcg. This patch series
contains two such concrete use-cases i.e. fsnotify and buffer_head.
The fsnotify event objects can consume a lot of system memory for large
or unlimited queues if there is either no or slow listener. The events
are allocated in the context of the event producer. However they should
be charged to the event consumer. Similarly the buffer_head objects can
be allocated in a memcg different from the memcg of the page for which
buffer_head objects are being allocated.
To solve this issue, this patch series introduces mechanism to charge
kernel memory to a given memcg. In case of fsnotify events, the memcg
of the consumer can be used for charging and for buffer_head, the memcg
of the page can be charged. For directed charging, the caller can use
the scope API memalloc_[un]use_memcg() to specify the memcg to charge
for all the __GFP_ACCOUNT allocations within the scope.
This patch (of 2):
A lot of memory can be consumed by the events generated for the huge or
unlimited queues if there is either no or slow listener. This can cause
system level memory pressure or OOMs. So, it's better to account the
fsnotify kmem caches to the memcg of the listener.
However the listener can be in a different memcg than the memcg of the
producer and these allocations happen in the context of the event
producer. This patch introduces remote memcg charging API which the
producer can use to charge the allocations to the memcg of the listener.
There are seven fsnotify kmem caches and among them allocations from
dnotify_struct_cache, dnotify_mark_cache, fanotify_mark_cache and
inotify_inode_mark_cachep happens in the context of syscall from the
listener. So, SLAB_ACCOUNT is enough for these caches.
The objects from fsnotify_mark_connector_cachep are not accounted as
they are small compared to the notification mark or events and it is
unclear whom to account connector to since it is shared by all events
attached to the inode.
The allocations from the event caches happen in the context of the event
producer. For such caches we will need to remote charge the allocations
to the listener's memcg. Thus we save the memcg reference in the
fsnotify_group structure of the listener.
This patch has also moved the members of fsnotify_group to keep the size
same, at least for 64 bit build, even with additional member by filling
the holes.
[shakeelb@google.com: use GFP_KERNEL_ACCOUNT rather than open-coding it]
Link: http://lkml.kernel.org/r/20180702215439.211597-1-shakeelb@google.com
Link: http://lkml.kernel.org/r/20180627191250.209150-2-shakeelb@google.com
Signed-off-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Amir Goldstein <amir73il@gmail.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Roman Gushchin <guro@fb.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-08-18 01:46:39 +03:00
# ifdef CONFIG_MEMCG
tsk - > active_memcg = NULL ;
# endif
2005-04-17 02:20:36 +04:00
return tsk ;
2008-03-11 01:28:04 +03:00
Clarify naming of thread info/stack allocators
We've had the thread info allocated together with the thread stack for
most architectures for a long time (since the thread_info was split off
from the task struct), but that is about to change.
But the patches that move the thread info to be off-stack (and a part of
the task struct instead) made it clear how confused the allocator and
freeing functions are.
Because the common case was that we share an allocation with the thread
stack and the thread_info, the two pointers were identical. That
identity then meant that we would have things like
ti = alloc_thread_info_node(tsk, node);
...
tsk->stack = ti;
which certainly _worked_ (since stack and thread_info have the same
value), but is rather confusing: why are we assigning a thread_info to
the stack? And if we move the thread_info away, the "confusing" code
just gets to be entirely bogus.
So remove all this confusion, and make it clear that we are doing the
stack allocation by renaming and clarifying the function names to be
about the stack. The fact that the thread_info then shares the
allocation is an implementation detail, and not really about the
allocation itself.
This is a pure renaming and type fix: we pass in the same pointer, it's
just that we clarify what the pointer means.
The ia64 code that actually only has one single allocation (for all of
task_struct, thread_info and kernel thread stack) now looks a bit odd,
but since "tsk->stack" is actually not even used there, that oddity
doesn't matter. It would be a separate thing to clean that up, I
intentionally left the ia64 changes as a pure brute-force renaming and
type change.
Acked-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-06-25 01:09:37 +03:00
free_stack :
2016-08-11 12:35:21 +03:00
free_thread_stack ( tsk ) ;
fork: fix error handling in dup_task()
The function dup_task() may fail at the following function calls in the
following order.
0) alloc_task_struct_node()
1) alloc_thread_info_node()
2) arch_dup_task_struct()
Error by 0) is not a matter, it can just return. But error by 1) requires
releasing task_struct allocated by 0) before it returns. Likewise, error
by 2) requires releasing task_struct and thread_info allocated by 0) and
1).
The existing error handling calls free_task_struct() and
free_thread_info() which do not only release task_struct and thread_info,
but also call architecture specific arch_release_task_struct() and
arch_release_thread_info().
The problem is that task_struct and thread_info are not fully initialized
yet at this point, but arch_release_task_struct() and
arch_release_thread_info() are called with them.
For example, x86 defines its own arch_release_task_struct() that releases
a task_xstate. If alloc_thread_info_node() fails in dup_task(),
arch_release_task_struct() is called with task_struct which is just
allocated and filled with garbage in this error handling.
This actually happened with tools/testing/fault-injection/failcmd.sh
# env FAILCMD_TYPE=fail_page_alloc \
./tools/testing/fault-injection/failcmd.sh --times=100 \
--min-order=0 --ignore-gfp-wait=0 \
-- make -C tools/testing/selftests/ run_tests
In order to fix this issue, make free_{task_struct,thread_info}() not to
call arch_release_{task_struct,thread_info}() and call
arch_release_{task_struct,thread_info}() implicitly where needed.
Default arch_release_task_struct() and arch_release_thread_info() are
defined as empty by default. So this change only affects the
architectures which implement their own arch_release_task_struct() or
arch_release_thread_info() as listed below.
arch_release_task_struct(): x86, sh
arch_release_thread_info(): mn10300, tile
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Salman Qazi <sqazi@google.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-31 01:42:33 +04:00
free_tsk :
2008-03-11 01:28:04 +03:00
free_task_struct ( tsk ) ;
return NULL ;
2005-04-17 02:20:36 +04:00
}
2007-10-18 14:06:07 +04:00
__cacheline_aligned_in_smp DEFINE_SPINLOCK ( mmlist_lock ) ;
2005-04-17 02:20:36 +04:00
2009-01-07 01:42:47 +03:00
static unsigned long default_dump_filter = MMF_DUMP_FILTER_DEFAULT ;
static int __init coredump_filter_setup ( char * s )
{
default_dump_filter =
( simple_strtoul ( s , NULL , 0 ) < < MMF_DUMP_FILTER_SHIFT ) &
MMF_DUMP_FILTER_MASK ;
return 1 ;
}
__setup ( " coredump_filter= " , coredump_filter_setup ) ;
2005-04-17 02:20:36 +04:00
# include <linux/init_task.h>
2009-09-24 02:57:32 +04:00
static void mm_init_aio ( struct mm_struct * mm )
{
# ifdef CONFIG_AIO
spin_lock_init ( & mm - > ioctx_lock ) ;
aio: convert the ioctx list to table lookup v3
On Wed, Jun 12, 2013 at 11:14:40AM -0700, Kent Overstreet wrote:
> On Mon, Apr 15, 2013 at 02:40:55PM +0300, Octavian Purdila wrote:
> > When using a large number of threads performing AIO operations the
> > IOCTX list may get a significant number of entries which will cause
> > significant overhead. For example, when running this fio script:
> >
> > rw=randrw; size=256k ;directory=/mnt/fio; ioengine=libaio; iodepth=1
> > blocksize=1024; numjobs=512; thread; loops=100
> >
> > on an EXT2 filesystem mounted on top of a ramdisk we can observe up to
> > 30% CPU time spent by lookup_ioctx:
> >
> > 32.51% [guest.kernel] [g] lookup_ioctx
> > 9.19% [guest.kernel] [g] __lock_acquire.isra.28
> > 4.40% [guest.kernel] [g] lock_release
> > 4.19% [guest.kernel] [g] sched_clock_local
> > 3.86% [guest.kernel] [g] local_clock
> > 3.68% [guest.kernel] [g] native_sched_clock
> > 3.08% [guest.kernel] [g] sched_clock_cpu
> > 2.64% [guest.kernel] [g] lock_release_holdtime.part.11
> > 2.60% [guest.kernel] [g] memcpy
> > 2.33% [guest.kernel] [g] lock_acquired
> > 2.25% [guest.kernel] [g] lock_acquire
> > 1.84% [guest.kernel] [g] do_io_submit
> >
> > This patchs converts the ioctx list to a radix tree. For a performance
> > comparison the above FIO script was run on a 2 sockets 8 core
> > machine. This are the results (average and %rsd of 10 runs) for the
> > original list based implementation and for the radix tree based
> > implementation:
> >
> > cores 1 2 4 8 16 32
> > list 109376 ms 69119 ms 35682 ms 22671 ms 19724 ms 16408 ms
> > %rsd 0.69% 1.15% 1.17% 1.21% 1.71% 1.43%
> > radix 73651 ms 41748 ms 23028 ms 16766 ms 15232 ms 13787 ms
> > %rsd 1.19% 0.98% 0.69% 1.13% 0.72% 0.75%
> > % of radix
> > relative 66.12% 65.59% 66.63% 72.31% 77.26% 83.66%
> > to list
> >
> > To consider the impact of the patch on the typical case of having
> > only one ctx per process the following FIO script was run:
> >
> > rw=randrw; size=100m ;directory=/mnt/fio; ioengine=libaio; iodepth=1
> > blocksize=1024; numjobs=1; thread; loops=100
> >
> > on the same system and the results are the following:
> >
> > list 58892 ms
> > %rsd 0.91%
> > radix 59404 ms
> > %rsd 0.81%
> > % of radix
> > relative 100.87%
> > to list
>
> So, I was just doing some benchmarking/profiling to get ready to send
> out the aio patches I've got for 3.11 - and it looks like your patch is
> causing a ~1.5% throughput regression in my testing :/
... <snip>
I've got an alternate approach for fixing this wart in lookup_ioctx()...
Instead of using an rbtree, just use the reserved id in the ring buffer
header to index an array pointing the ioctx. It's not finished yet, and
it needs to be tidied up, but is most of the way there.
-ben
--
"Thought is the essence of where you are now."
--
kmo> And, a rework of Ben's code, but this was entirely his idea
kmo> -Kent
bcrl> And fix the code to use the right mm_struct in kill_ioctx(), actually
free memory.
Signed-off-by: Benjamin LaHaise <bcrl@kvack.org>
2013-07-30 20:54:40 +04:00
mm - > ioctx_table = NULL ;
2009-09-24 02:57:32 +04:00
# endif
}
2019-05-15 01:40:46 +03:00
static __always_inline void mm_clear_owner ( struct mm_struct * mm ,
struct task_struct * p )
{
# ifdef CONFIG_MEMCG
if ( mm - > owner = = p )
WRITE_ONCE ( mm - > owner , NULL ) ;
# endif
}
2014-08-09 01:22:03 +04:00
static void mm_init_owner ( struct mm_struct * mm , struct task_struct * p )
{
# ifdef CONFIG_MEMCG
mm - > owner = p ;
# endif
}
2021-03-13 08:07:15 +03:00
static void mm_init_pasid ( struct mm_struct * mm )
{
# ifdef CONFIG_IOMMU_SUPPORT
mm - > pasid = INIT_PASID ;
# endif
}
mm, uprobes: fix multiple free of ->uprobes_state.xol_area
Commit 7c051267931a ("mm, fork: make dup_mmap wait for mmap_sem for
write killable") made it possible to kill a forking task while it is
waiting to acquire its ->mmap_sem for write, in dup_mmap().
However, it was overlooked that this introduced an new error path before
the new mm_struct's ->uprobes_state.xol_area has been set to NULL after
being copied from the old mm_struct by the memcpy in dup_mm(). For a
task that has previously hit a uprobe tracepoint, this resulted in the
'struct xol_area' being freed multiple times if the task was killed at
just the right time while forking.
Fix it by setting ->uprobes_state.xol_area to NULL in mm_init() rather
than in uprobe_dup_mmap().
With CONFIG_UPROBE_EVENTS=y, the bug can be reproduced by the same C
program given by commit 2b7e8665b4ff ("fork: fix incorrect fput of
->exe_file causing use-after-free"), provided that a uprobe tracepoint
has been set on the fork_thread() function. For example:
$ gcc reproducer.c -o reproducer -lpthread
$ nm reproducer | grep fork_thread
0000000000400719 t fork_thread
$ echo "p $PWD/reproducer:0x719" > /sys/kernel/debug/tracing/uprobe_events
$ echo 1 > /sys/kernel/debug/tracing/events/uprobes/enable
$ ./reproducer
Here is the use-after-free reported by KASAN:
BUG: KASAN: use-after-free in uprobe_clear_state+0x1c4/0x200
Read of size 8 at addr ffff8800320a8b88 by task reproducer/198
CPU: 1 PID: 198 Comm: reproducer Not tainted 4.13.0-rc7-00015-g36fde05f3fb5 #255
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-20170228_101828-anatol 04/01/2014
Call Trace:
dump_stack+0xdb/0x185
print_address_description+0x7e/0x290
kasan_report+0x23b/0x350
__asan_report_load8_noabort+0x19/0x20
uprobe_clear_state+0x1c4/0x200
mmput+0xd6/0x360
do_exit+0x740/0x1670
do_group_exit+0x13f/0x380
get_signal+0x597/0x17d0
do_signal+0x99/0x1df0
exit_to_usermode_loop+0x166/0x1e0
syscall_return_slowpath+0x258/0x2c0
entry_SYSCALL_64_fastpath+0xbc/0xbe
...
Allocated by task 199:
save_stack_trace+0x1b/0x20
kasan_kmalloc+0xfc/0x180
kmem_cache_alloc_trace+0xf3/0x330
__create_xol_area+0x10f/0x780
uprobe_notify_resume+0x1674/0x2210
exit_to_usermode_loop+0x150/0x1e0
prepare_exit_to_usermode+0x14b/0x180
retint_user+0x8/0x20
Freed by task 199:
save_stack_trace+0x1b/0x20
kasan_slab_free+0xa8/0x1a0
kfree+0xba/0x210
uprobe_clear_state+0x151/0x200
mmput+0xd6/0x360
copy_process.part.8+0x605f/0x65d0
_do_fork+0x1a5/0xbd0
SyS_clone+0x19/0x20
do_syscall_64+0x22f/0x660
return_from_SYSCALL_64+0x0/0x7a
Note: without KASAN, you may instead see a "Bad page state" message, or
simply a general protection fault.
Link: http://lkml.kernel.org/r/20170830033303.17927-1-ebiggers3@gmail.com
Fixes: 7c051267931a ("mm, fork: make dup_mmap wait for mmap_sem for write killable")
Signed-off-by: Eric Biggers <ebiggers@google.com>
Reported-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org> [4.7+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-09-01 02:15:26 +03:00
static void mm_init_uprobes_state ( struct mm_struct * mm )
{
# ifdef CONFIG_UPROBES
mm - > uprobes_state . xol_area = NULL ;
# endif
}
2016-10-14 05:23:16 +03:00
static struct mm_struct * mm_init ( struct mm_struct * mm , struct task_struct * p ,
struct user_namespace * user_ns )
2005-04-17 02:20:36 +04:00
{
fork/exec: cleanup mm initialization
mm initialization on fork/exec is spread all over the place, which makes
the code look inconsistent.
We have mm_init(), which is supposed to init/nullify mm's internals, but
it doesn't init all the fields it should:
- on fork ->mmap,mm_rb,vmacache_seqnum,map_count,mm_cpumask,locked_vm
are zeroed in dup_mmap();
- on fork ->pmd_huge_pte is zeroed in dup_mm(), immediately before
calling mm_init();
- ->cpu_vm_mask_var ptr is initialized by mm_init_cpumask(), which is
called before mm_init() on both fork and exec;
- ->context is initialized by init_new_context(), which is called after
mm_init() on both fork and exec;
Let's consolidate all the initializations in mm_init() to make the code
look cleaner.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Christoph Lameter <cl@linux.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-08-09 01:21:56 +04:00
mm - > mmap = NULL ;
mm - > mm_rb = RB_ROOT ;
mm - > vmacache_seqnum = 0 ;
2005-04-17 02:20:36 +04:00
atomic_set ( & mm - > mm_users , 1 ) ;
atomic_set ( & mm - > mm_count , 1 ) ;
2020-12-15 06:05:44 +03:00
seqcount_init ( & mm - > write_protect_seq ) ;
2020-06-09 07:33:25 +03:00
mmap_init_lock ( mm ) ;
2005-04-17 02:20:36 +04:00
INIT_LIST_HEAD ( & mm - > mmlist ) ;
2008-07-25 12:47:41 +04:00
mm - > core_state = NULL ;
2017-11-16 04:35:40 +03:00
mm_pgtables_bytes_init ( mm ) ;
fork/exec: cleanup mm initialization
mm initialization on fork/exec is spread all over the place, which makes
the code look inconsistent.
We have mm_init(), which is supposed to init/nullify mm's internals, but
it doesn't init all the fields it should:
- on fork ->mmap,mm_rb,vmacache_seqnum,map_count,mm_cpumask,locked_vm
are zeroed in dup_mmap();
- on fork ->pmd_huge_pte is zeroed in dup_mm(), immediately before
calling mm_init();
- ->cpu_vm_mask_var ptr is initialized by mm_init_cpumask(), which is
called before mm_init() on both fork and exec;
- ->context is initialized by init_new_context(), which is called after
mm_init() on both fork and exec;
Let's consolidate all the initializations in mm_init() to make the code
look cleaner.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Christoph Lameter <cl@linux.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-08-09 01:21:56 +04:00
mm - > map_count = 0 ;
mm - > locked_vm = 0 ;
2019-02-06 20:59:15 +03:00
atomic64_set ( & mm - > pinned_vm , 0 ) ;
2010-03-06 00:41:39 +03:00
memset ( & mm - > rss_stat , 0 , sizeof ( mm - > rss_stat ) ) ;
2005-04-17 02:20:36 +04:00
spin_lock_init ( & mm - > page_table_lock ) ;
mm: introduce arg_lock to protect arg_start|end and env_start|end in mm_struct
mmap_sem is on the hot path of kernel, and it very contended, but it is
abused too. It is used to protect arg_start|end and evn_start|end when
reading /proc/$PID/cmdline and /proc/$PID/environ, but it doesn't make
sense since those proc files just expect to read 4 values atomically and
not related to VM, they could be set to arbitrary values by C/R.
And, the mmap_sem contention may cause unexpected issue like below:
INFO: task ps:14018 blocked for more than 120 seconds.
Tainted: G E 4.9.79-009.ali3000.alios7.x86_64 #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this
message.
ps D 0 14018 1 0x00000004
Call Trace:
schedule+0x36/0x80
rwsem_down_read_failed+0xf0/0x150
call_rwsem_down_read_failed+0x18/0x30
down_read+0x20/0x40
proc_pid_cmdline_read+0xd9/0x4e0
__vfs_read+0x37/0x150
vfs_read+0x96/0x130
SyS_read+0x55/0xc0
entry_SYSCALL_64_fastpath+0x1a/0xc5
Both Alexey Dobriyan and Michal Hocko suggested to use dedicated lock
for them to mitigate the abuse of mmap_sem.
So, introduce a new spinlock in mm_struct to protect the concurrent
access to arg_start|end, env_start|end and others, as well as replace
write map_sem to read to protect the race condition between prctl and
sys_brk which might break check_data_rlimit(), and makes prctl more
friendly to other VM operations.
This patch just eliminates the abuse of mmap_sem, but it can't resolve
the above hung task warning completely since the later
access_remote_vm() call needs acquire mmap_sem. The mmap_sem
scalability issue will be solved in the future.
[yang.shi@linux.alibaba.com: add comment about mmap_sem and arg_lock]
Link: http://lkml.kernel.org/r/1524077799-80690-1-git-send-email-yang.shi@linux.alibaba.com
Link: http://lkml.kernel.org/r/1523730291-109696-1-git-send-email-yang.shi@linux.alibaba.com
Signed-off-by: Yang Shi <yang.shi@linux.alibaba.com>
Reviewed-by: Cyrill Gorcunov <gorcunov@openvz.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mateusz Guzik <mguzik@redhat.com>
Cc: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-06-08 03:05:28 +03:00
spin_lock_init ( & mm - > arg_lock ) ;
fork/exec: cleanup mm initialization
mm initialization on fork/exec is spread all over the place, which makes
the code look inconsistent.
We have mm_init(), which is supposed to init/nullify mm's internals, but
it doesn't init all the fields it should:
- on fork ->mmap,mm_rb,vmacache_seqnum,map_count,mm_cpumask,locked_vm
are zeroed in dup_mmap();
- on fork ->pmd_huge_pte is zeroed in dup_mm(), immediately before
calling mm_init();
- ->cpu_vm_mask_var ptr is initialized by mm_init_cpumask(), which is
called before mm_init() on both fork and exec;
- ->context is initialized by init_new_context(), which is called after
mm_init() on both fork and exec;
Let's consolidate all the initializations in mm_init() to make the code
look cleaner.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Christoph Lameter <cl@linux.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-08-09 01:21:56 +04:00
mm_init_cpumask ( mm ) ;
2009-09-24 02:57:32 +04:00
mm_init_aio ( mm ) ;
cgroups: add an owner to the mm_struct
Remove the mem_cgroup member from mm_struct and instead adds an owner.
This approach was suggested by Paul Menage. The advantage of this approach
is that, once the mm->owner is known, using the subsystem id, the cgroup
can be determined. It also allows several control groups that are
virtually grouped by mm_struct, to exist independent of the memory
controller i.e., without adding mem_cgroup's for each controller, to
mm_struct.
A new config option CONFIG_MM_OWNER is added and the memory resource
controller selects this config option.
This patch also adds cgroup callbacks to notify subsystems when mm->owner
changes. The mm_cgroup_changed callback is called with the task_lock() of
the new task held and is called just prior to changing the mm->owner.
I am indebted to Paul Menage for the several reviews of this patchset and
helping me make it lighter and simpler.
This patch was tested on a powerpc box, it was compiled with both the
MM_OWNER config turned on and off.
After the thread group leader exits, it's moved to init_css_state by
cgroup_exit(), thus all future charges from runnings threads would be
redirected to the init_css_set's subsystem.
Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Pavel Emelianov <xemul@openvz.org>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Sudhir Kumar <skumar@linux.vnet.ibm.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: Hirokazu Takahashi <taka@valinux.co.jp>
Cc: David Rientjes <rientjes@google.com>,
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Reviewed-by: Paul Menage <menage@google.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 12:00:16 +04:00
mm_init_owner ( mm , p ) ;
2021-03-13 08:07:15 +03:00
mm_init_pasid ( mm ) ;
fork: fix incorrect fput of ->exe_file causing use-after-free
Commit 7c051267931a ("mm, fork: make dup_mmap wait for mmap_sem for
write killable") made it possible to kill a forking task while it is
waiting to acquire its ->mmap_sem for write, in dup_mmap().
However, it was overlooked that this introduced an new error path before
a reference is taken on the mm_struct's ->exe_file. Since the
->exe_file of the new mm_struct was already set to the old ->exe_file by
the memcpy() in dup_mm(), it was possible for the mmput() in the error
path of dup_mm() to drop a reference to ->exe_file which was never
taken.
This caused the struct file to later be freed prematurely.
Fix it by updating mm_init() to NULL out the ->exe_file, in the same
place it clears other things like the list of mmaps.
This bug was found by syzkaller. It can be reproduced using the
following C program:
#define _GNU_SOURCE
#include <pthread.h>
#include <stdlib.h>
#include <sys/mman.h>
#include <sys/syscall.h>
#include <sys/wait.h>
#include <unistd.h>
static void *mmap_thread(void *_arg)
{
for (;;) {
mmap(NULL, 0x1000000, PROT_READ,
MAP_POPULATE|MAP_ANONYMOUS|MAP_PRIVATE, -1, 0);
}
}
static void *fork_thread(void *_arg)
{
usleep(rand() % 10000);
fork();
}
int main(void)
{
fork();
fork();
fork();
for (;;) {
if (fork() == 0) {
pthread_t t;
pthread_create(&t, NULL, mmap_thread, NULL);
pthread_create(&t, NULL, fork_thread, NULL);
usleep(rand() % 10000);
syscall(__NR_exit_group, 0);
}
wait(NULL);
}
}
No special kernel config options are needed. It usually causes a NULL
pointer dereference in __remove_shared_vm_struct() during exit, or in
dup_mmap() (which is usually inlined into copy_process()) during fork.
Both are due to a vm_area_struct's ->vm_file being used after it's
already been freed.
Google Bug Id: 64772007
Link: http://lkml.kernel.org/r/20170823211408.31198-1-ebiggers3@gmail.com
Fixes: 7c051267931a ("mm, fork: make dup_mmap wait for mmap_sem for write killable")
Signed-off-by: Eric Biggers <ebiggers@google.com>
Tested-by: Mark Rutland <mark.rutland@arm.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org> [v4.7+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-08-26 01:55:43 +03:00
RCU_INIT_POINTER ( mm - > exe_file , NULL ) ;
2019-12-18 20:40:35 +03:00
mmu_notifier_subscriptions_init ( mm ) ;
mm: migrate: prevent racy access to tlb_flush_pending
Patch series "fixes of TLB batching races", v6.
It turns out that Linux TLB batching mechanism suffers from various
races. Races that are caused due to batching during reclamation were
recently handled by Mel and this patch-set deals with others. The more
fundamental issue is that concurrent updates of the page-tables allow
for TLB flushes to be batched on one core, while another core changes
the page-tables. This other core may assume a PTE change does not
require a flush based on the updated PTE value, while it is unaware that
TLB flushes are still pending.
This behavior affects KSM (which may result in memory corruption) and
MADV_FREE and MADV_DONTNEED (which may result in incorrect behavior). A
proof-of-concept can easily produce the wrong behavior of MADV_DONTNEED.
Memory corruption in KSM is harder to produce in practice, but was
observed by hacking the kernel and adding a delay before flushing and
replacing the KSM page.
Finally, there is also one memory barrier missing, which may affect
architectures with weak memory model.
This patch (of 7):
Setting and clearing mm->tlb_flush_pending can be performed by multiple
threads, since mmap_sem may only be acquired for read in
task_numa_work(). If this happens, tlb_flush_pending might be cleared
while one of the threads still changes PTEs and batches TLB flushes.
This can lead to the same race between migration and
change_protection_range() that led to the introduction of
tlb_flush_pending. The result of this race was data corruption, which
means that this patch also addresses a theoretically possible data
corruption.
An actual data corruption was not observed, yet the race was was
confirmed by adding assertion to check tlb_flush_pending is not set by
two threads, adding artificial latency in change_protection_range() and
using sysctl to reduce kernel.numa_balancing_scan_delay_ms.
Link: http://lkml.kernel.org/r/20170802000818.4760-2-namit@vmware.com
Fixes: 20841405940e ("mm: fix TLB flush race between migration, and
change_protection_range")
Signed-off-by: Nadav Amit <namit@vmware.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Acked-by: Rik van Riel <riel@redhat.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-08-11 01:23:56 +03:00
init_tlb_flush_pending ( mm ) ;
fork/exec: cleanup mm initialization
mm initialization on fork/exec is spread all over the place, which makes
the code look inconsistent.
We have mm_init(), which is supposed to init/nullify mm's internals, but
it doesn't init all the fields it should:
- on fork ->mmap,mm_rb,vmacache_seqnum,map_count,mm_cpumask,locked_vm
are zeroed in dup_mmap();
- on fork ->pmd_huge_pte is zeroed in dup_mm(), immediately before
calling mm_init();
- ->cpu_vm_mask_var ptr is initialized by mm_init_cpumask(), which is
called before mm_init() on both fork and exec;
- ->context is initialized by init_new_context(), which is called after
mm_init() on both fork and exec;
Let's consolidate all the initializations in mm_init() to make the code
look cleaner.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Christoph Lameter <cl@linux.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-08-09 01:21:56 +04:00
# if defined(CONFIG_TRANSPARENT_HUGEPAGE) && !USE_SPLIT_PMD_PTLOCKS
mm - > pmd_huge_pte = NULL ;
# endif
mm, uprobes: fix multiple free of ->uprobes_state.xol_area
Commit 7c051267931a ("mm, fork: make dup_mmap wait for mmap_sem for
write killable") made it possible to kill a forking task while it is
waiting to acquire its ->mmap_sem for write, in dup_mmap().
However, it was overlooked that this introduced an new error path before
the new mm_struct's ->uprobes_state.xol_area has been set to NULL after
being copied from the old mm_struct by the memcpy in dup_mm(). For a
task that has previously hit a uprobe tracepoint, this resulted in the
'struct xol_area' being freed multiple times if the task was killed at
just the right time while forking.
Fix it by setting ->uprobes_state.xol_area to NULL in mm_init() rather
than in uprobe_dup_mmap().
With CONFIG_UPROBE_EVENTS=y, the bug can be reproduced by the same C
program given by commit 2b7e8665b4ff ("fork: fix incorrect fput of
->exe_file causing use-after-free"), provided that a uprobe tracepoint
has been set on the fork_thread() function. For example:
$ gcc reproducer.c -o reproducer -lpthread
$ nm reproducer | grep fork_thread
0000000000400719 t fork_thread
$ echo "p $PWD/reproducer:0x719" > /sys/kernel/debug/tracing/uprobe_events
$ echo 1 > /sys/kernel/debug/tracing/events/uprobes/enable
$ ./reproducer
Here is the use-after-free reported by KASAN:
BUG: KASAN: use-after-free in uprobe_clear_state+0x1c4/0x200
Read of size 8 at addr ffff8800320a8b88 by task reproducer/198
CPU: 1 PID: 198 Comm: reproducer Not tainted 4.13.0-rc7-00015-g36fde05f3fb5 #255
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-20170228_101828-anatol 04/01/2014
Call Trace:
dump_stack+0xdb/0x185
print_address_description+0x7e/0x290
kasan_report+0x23b/0x350
__asan_report_load8_noabort+0x19/0x20
uprobe_clear_state+0x1c4/0x200
mmput+0xd6/0x360
do_exit+0x740/0x1670
do_group_exit+0x13f/0x380
get_signal+0x597/0x17d0
do_signal+0x99/0x1df0
exit_to_usermode_loop+0x166/0x1e0
syscall_return_slowpath+0x258/0x2c0
entry_SYSCALL_64_fastpath+0xbc/0xbe
...
Allocated by task 199:
save_stack_trace+0x1b/0x20
kasan_kmalloc+0xfc/0x180
kmem_cache_alloc_trace+0xf3/0x330
__create_xol_area+0x10f/0x780
uprobe_notify_resume+0x1674/0x2210
exit_to_usermode_loop+0x150/0x1e0
prepare_exit_to_usermode+0x14b/0x180
retint_user+0x8/0x20
Freed by task 199:
save_stack_trace+0x1b/0x20
kasan_slab_free+0xa8/0x1a0
kfree+0xba/0x210
uprobe_clear_state+0x151/0x200
mmput+0xd6/0x360
copy_process.part.8+0x605f/0x65d0
_do_fork+0x1a5/0xbd0
SyS_clone+0x19/0x20
do_syscall_64+0x22f/0x660
return_from_SYSCALL_64+0x0/0x7a
Note: without KASAN, you may instead see a "Bad page state" message, or
simply a general protection fault.
Link: http://lkml.kernel.org/r/20170830033303.17927-1-ebiggers3@gmail.com
Fixes: 7c051267931a ("mm, fork: make dup_mmap wait for mmap_sem for write killable")
Signed-off-by: Eric Biggers <ebiggers@google.com>
Reported-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org> [4.7+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-09-01 02:15:26 +03:00
mm_init_uprobes_state ( mm ) ;
2005-04-17 02:20:36 +04:00
2014-04-08 02:37:10 +04:00
if ( current - > mm ) {
mm - > flags = current - > mm - > flags & MMF_INIT_MASK ;
mm - > def_flags = current - > mm - > def_flags & VM_INIT_DEF_MASK ;
} else {
mm - > flags = default_dump_filter ;
2005-04-17 02:20:36 +04:00
mm - > def_flags = 0 ;
2014-04-08 02:37:10 +04:00
}
fork/exec: cleanup mm initialization
mm initialization on fork/exec is spread all over the place, which makes
the code look inconsistent.
We have mm_init(), which is supposed to init/nullify mm's internals, but
it doesn't init all the fields it should:
- on fork ->mmap,mm_rb,vmacache_seqnum,map_count,mm_cpumask,locked_vm
are zeroed in dup_mmap();
- on fork ->pmd_huge_pte is zeroed in dup_mm(), immediately before
calling mm_init();
- ->cpu_vm_mask_var ptr is initialized by mm_init_cpumask(), which is
called before mm_init() on both fork and exec;
- ->context is initialized by init_new_context(), which is called after
mm_init() on both fork and exec;
Let's consolidate all the initializations in mm_init() to make the code
look cleaner.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Christoph Lameter <cl@linux.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-08-09 01:21:56 +04:00
if ( mm_alloc_pgd ( mm ) )
goto fail_nopgd ;
if ( init_new_context ( p , mm ) )
goto fail_nocontext ;
2008-02-07 11:13:51 +03:00
2016-10-14 05:23:16 +03:00
mm - > user_ns = get_user_ns ( user_ns ) ;
fork/exec: cleanup mm initialization
mm initialization on fork/exec is spread all over the place, which makes
the code look inconsistent.
We have mm_init(), which is supposed to init/nullify mm's internals, but
it doesn't init all the fields it should:
- on fork ->mmap,mm_rb,vmacache_seqnum,map_count,mm_cpumask,locked_vm
are zeroed in dup_mmap();
- on fork ->pmd_huge_pte is zeroed in dup_mm(), immediately before
calling mm_init();
- ->cpu_vm_mask_var ptr is initialized by mm_init_cpumask(), which is
called before mm_init() on both fork and exec;
- ->context is initialized by init_new_context(), which is called after
mm_init() on both fork and exec;
Let's consolidate all the initializations in mm_init() to make the code
look cleaner.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Christoph Lameter <cl@linux.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-08-09 01:21:56 +04:00
return mm ;
fail_nocontext :
mm_free_pgd ( mm ) ;
fail_nopgd :
2005-04-17 02:20:36 +04:00
free_mm ( mm ) ;
return NULL ;
}
/*
* Allocate and initialize an mm_struct .
*/
2011-07-27 03:08:39 +04:00
struct mm_struct * mm_alloc ( void )
2005-04-17 02:20:36 +04:00
{
2011-07-27 03:08:39 +04:00
struct mm_struct * mm ;
2005-04-17 02:20:36 +04:00
mm = allocate_mm ( ) ;
2011-05-25 04:12:15 +04:00
if ( ! mm )
return NULL ;
memset ( mm , 0 , sizeof ( * mm ) ) ;
2016-10-14 05:23:16 +03:00
return mm_init ( mm , current , current_user_ns ( ) ) ;
2005-04-17 02:20:36 +04:00
}
2016-05-21 02:57:21 +03:00
static inline void __mmput ( struct mm_struct * mm )
{
VM_BUG_ON ( atomic_read ( & mm - > mm_users ) ) ;
uprobe_clear_state ( mm ) ;
exit_aio ( mm ) ;
ksm_exit ( mm ) ;
khugepaged_exit ( mm ) ; /* must run before exit_mmap */
exit_mmap ( mm ) ;
2016-10-08 03:00:08 +03:00
mm_put_huge_zero_page ( mm ) ;
2016-05-21 02:57:21 +03:00
set_mm_exe_file ( mm , NULL ) ;
if ( ! list_empty ( & mm - > mmlist ) ) {
spin_lock ( & mmlist_lock ) ;
list_del ( & mm - > mmlist ) ;
spin_unlock ( & mmlist_lock ) ;
}
if ( mm - > binfmt )
module_put ( mm - > binfmt - > module ) ;
mmdrop ( mm ) ;
}
2005-04-17 02:20:36 +04:00
/*
* Decrement the use count and release all resources for an mm .
*/
void mmput ( struct mm_struct * mm )
{
2006-06-23 13:05:15 +04:00
might_sleep ( ) ;
2016-05-21 02:57:21 +03:00
if ( atomic_dec_and_test ( & mm - > mm_users ) )
__mmput ( mm ) ;
}
EXPORT_SYMBOL_GPL ( mmput ) ;
2017-10-04 02:15:00 +03:00
# ifdef CONFIG_MMU
static void mmput_async_fn ( struct work_struct * work )
{
struct mm_struct * mm = container_of ( work , struct mm_struct ,
async_put_work ) ;
__mmput ( mm ) ;
}
void mmput_async ( struct mm_struct * mm )
{
if ( atomic_dec_and_test ( & mm - > mm_users ) ) {
INIT_WORK ( & mm - > async_put_work , mmput_async_fn ) ;
schedule_work ( & mm - > async_put_work ) ;
}
}
# endif
2015-04-16 22:47:56 +03:00
/**
* set_mm_exe_file - change a reference to the mm ' s executable file
*
* This changes mm ' s executable file ( shown as symlink / proc / [ pid ] / exe ) .
*
2015-04-16 22:47:59 +03:00
* Main users are mmput ( ) and sys_execve ( ) . Callers prevent concurrent
* invocations : in mmput ( ) nobody alive left , in execve task is single
2021-04-23 11:20:25 +03:00
* threaded .
2015-04-16 22:47:56 +03:00
*/
2011-05-27 03:25:46 +04:00
void set_mm_exe_file ( struct mm_struct * mm , struct file * new_exe_file )
{
2015-04-16 22:47:59 +03:00
struct file * old_exe_file ;
/*
* It is safe to dereference the exe_file without RCU as
* this function is only called if nobody else can access
* this mm - - see comment above for justification .
*/
old_exe_file = rcu_dereference_raw ( mm - > exe_file ) ;
2015-04-16 22:47:56 +03:00
2011-05-27 03:25:46 +04:00
if ( new_exe_file )
get_file ( new_exe_file ) ;
2015-04-16 22:47:56 +03:00
rcu_assign_pointer ( mm - > exe_file , new_exe_file ) ;
if ( old_exe_file )
fput ( old_exe_file ) ;
2011-05-27 03:25:46 +04:00
}
2021-04-23 11:20:25 +03:00
/**
* replace_mm_exe_file - replace a reference to the mm ' s executable file
*
* This changes mm ' s executable file ( shown as symlink / proc / [ pid ] / exe ) ,
* dealing with concurrent invocation and without grabbing the mmap lock in
* write mode .
*
* Main user is sys_prctl ( PR_SET_MM_MAP / EXE_FILE ) .
*/
int replace_mm_exe_file ( struct mm_struct * mm , struct file * new_exe_file )
{
struct vm_area_struct * vma ;
struct file * old_exe_file ;
int ret = 0 ;
/* Forbid mm->exe_file change if old file still mapped. */
old_exe_file = get_mm_exe_file ( mm ) ;
if ( old_exe_file ) {
mmap_read_lock ( mm ) ;
for ( vma = mm - > mmap ; vma & & ! ret ; vma = vma - > vm_next ) {
if ( ! vma - > vm_file )
continue ;
if ( path_equal ( & vma - > vm_file - > f_path ,
& old_exe_file - > f_path ) )
ret = - EBUSY ;
}
mmap_read_unlock ( mm ) ;
fput ( old_exe_file ) ;
if ( ret )
return ret ;
}
/* set the new file, lockless */
get_file ( new_exe_file ) ;
old_exe_file = xchg ( & mm - > exe_file , new_exe_file ) ;
if ( old_exe_file )
fput ( old_exe_file ) ;
return 0 ;
}
2015-04-16 22:47:56 +03:00
/**
* get_mm_exe_file - acquire a reference to the mm ' s executable file
*
* Returns % NULL if mm has no associated executable file .
* User must release file via fput ( ) .
*/
2011-05-27 03:25:46 +04:00
struct file * get_mm_exe_file ( struct mm_struct * mm )
{
struct file * exe_file ;
2015-04-16 22:47:56 +03:00
rcu_read_lock ( ) ;
exe_file = rcu_dereference ( mm - > exe_file ) ;
if ( exe_file & & ! get_file_rcu ( exe_file ) )
exe_file = NULL ;
rcu_read_unlock ( ) ;
2011-05-27 03:25:46 +04:00
return exe_file ;
}
2015-04-16 22:49:12 +03:00
EXPORT_SYMBOL ( get_mm_exe_file ) ;
2011-05-27 03:25:46 +04:00
2016-08-23 17:20:38 +03:00
/**
* get_task_exe_file - acquire a reference to the task ' s executable file
*
* Returns % NULL if task ' s mm ( if any ) has no associated executable file or
* this is a kernel thread with borrowed mm ( see the comment above get_task_mm ) .
* User must release file via fput ( ) .
*/
struct file * get_task_exe_file ( struct task_struct * task )
{
struct file * exe_file = NULL ;
struct mm_struct * mm ;
task_lock ( task ) ;
mm = task - > mm ;
if ( mm ) {
if ( ! ( task - > flags & PF_KTHREAD ) )
exe_file = get_mm_exe_file ( mm ) ;
}
task_unlock ( task ) ;
return exe_file ;
}
EXPORT_SYMBOL ( get_task_exe_file ) ;
2011-05-27 03:25:46 +04:00
2005-04-17 02:20:36 +04:00
/**
* get_task_mm - acquire a reference to the task ' s mm
*
2008-07-25 12:47:38 +04:00
* Returns % NULL if the task has no mm . Checks PF_KTHREAD ( meaning
2005-04-17 02:20:36 +04:00
* this kernel workthread has transiently adopted a user mm with use_mm ,
* to do its AIO ) is not set and if so returns a reference to it , after
* bumping up the use count . User must release the mm via mmput ( )
* after use . Typically used by / proc and ptrace .
*/
struct mm_struct * get_task_mm ( struct task_struct * task )
{
struct mm_struct * mm ;
task_lock ( task ) ;
mm = task - > mm ;
if ( mm ) {
2008-07-25 12:47:38 +04:00
if ( task - > flags & PF_KTHREAD )
2005-04-17 02:20:36 +04:00
mm = NULL ;
else
2017-02-28 01:30:10 +03:00
mmget ( mm ) ;
2005-04-17 02:20:36 +04:00
}
task_unlock ( task ) ;
return mm ;
}
EXPORT_SYMBOL_GPL ( get_task_mm ) ;
2012-02-02 05:04:09 +04:00
struct mm_struct * mm_access ( struct task_struct * task , unsigned int mode )
{
struct mm_struct * mm ;
int err ;
2020-12-03 23:12:00 +03:00
err = down_read_killable ( & task - > signal - > exec_update_lock ) ;
2012-02-02 05:04:09 +04:00
if ( err )
return ERR_PTR ( err ) ;
mm = get_task_mm ( task ) ;
if ( mm & & mm ! = current - > mm & &
! ptrace_may_access ( task , mode ) ) {
mmput ( mm ) ;
mm = ERR_PTR ( - EACCES ) ;
}
2020-12-03 23:12:00 +03:00
up_read ( & task - > signal - > exec_update_lock ) ;
2012-02-02 05:04:09 +04:00
return mm ;
}
2012-03-06 02:59:13 +04:00
static void complete_vfork_done ( struct task_struct * tsk )
2012-03-06 02:59:13 +04:00
{
2012-03-06 02:59:13 +04:00
struct completion * vfork ;
2012-03-06 02:59:13 +04:00
2012-03-06 02:59:13 +04:00
task_lock ( tsk ) ;
vfork = tsk - > vfork_done ;
if ( likely ( vfork ) ) {
tsk - > vfork_done = NULL ;
complete ( vfork ) ;
}
task_unlock ( tsk ) ;
}
static int wait_for_vfork_done ( struct task_struct * child ,
struct completion * vfork )
{
int killed ;
freezer_do_not_count ( ) ;
cgroup: cgroup v2 freezer
Cgroup v1 implements the freezer controller, which provides an ability
to stop the workload in a cgroup and temporarily free up some
resources (cpu, io, network bandwidth and, potentially, memory)
for some other tasks. Cgroup v2 lacks this functionality.
This patch implements freezer for cgroup v2.
Cgroup v2 freezer tries to put tasks into a state similar to jobctl
stop. This means that tasks can be killed, ptraced (using
PTRACE_SEIZE*), and interrupted. It is possible to attach to
a frozen task, get some information (e.g. read registers) and detach.
It's also possible to migrate a frozen tasks to another cgroup.
This differs cgroup v2 freezer from cgroup v1 freezer, which mostly
tried to imitate the system-wide freezer. However uninterruptible
sleep is fine when all tasks are going to be frozen (hibernation case),
it's not the acceptable state for some subset of the system.
Cgroup v2 freezer is not supporting freezing kthreads.
If a non-root cgroup contains kthread, the cgroup still can be frozen,
but the kthread will remain running, the cgroup will be shown
as non-frozen, and the notification will not be delivered.
* PTRACE_ATTACH is not working because non-fatal signal delivery
is blocked in frozen state.
There are some interface differences between cgroup v1 and cgroup v2
freezer too, which are required to conform the cgroup v2 interface
design principles:
1) There is no separate controller, which has to be turned on:
the functionality is always available and is represented by
cgroup.freeze and cgroup.events cgroup control files.
2) The desired state is defined by the cgroup.freeze control file.
Any hierarchical configuration is allowed.
3) The interface is asynchronous. The actual state is available
using cgroup.events control file ("frozen" field). There are no
dedicated transitional states.
4) It's allowed to make any changes with the cgroup hierarchy
(create new cgroups, remove old cgroups, move tasks between cgroups)
no matter if some cgroups are frozen.
Signed-off-by: Roman Gushchin <guro@fb.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
No-objection-from-me-by: Oleg Nesterov <oleg@redhat.com>
Cc: kernel-team@fb.com
2019-04-19 20:03:04 +03:00
cgroup_enter_frozen ( ) ;
2012-03-06 02:59:13 +04:00
killed = wait_for_completion_killable ( vfork ) ;
cgroup: cgroup v2 freezer
Cgroup v1 implements the freezer controller, which provides an ability
to stop the workload in a cgroup and temporarily free up some
resources (cpu, io, network bandwidth and, potentially, memory)
for some other tasks. Cgroup v2 lacks this functionality.
This patch implements freezer for cgroup v2.
Cgroup v2 freezer tries to put tasks into a state similar to jobctl
stop. This means that tasks can be killed, ptraced (using
PTRACE_SEIZE*), and interrupted. It is possible to attach to
a frozen task, get some information (e.g. read registers) and detach.
It's also possible to migrate a frozen tasks to another cgroup.
This differs cgroup v2 freezer from cgroup v1 freezer, which mostly
tried to imitate the system-wide freezer. However uninterruptible
sleep is fine when all tasks are going to be frozen (hibernation case),
it's not the acceptable state for some subset of the system.
Cgroup v2 freezer is not supporting freezing kthreads.
If a non-root cgroup contains kthread, the cgroup still can be frozen,
but the kthread will remain running, the cgroup will be shown
as non-frozen, and the notification will not be delivered.
* PTRACE_ATTACH is not working because non-fatal signal delivery
is blocked in frozen state.
There are some interface differences between cgroup v1 and cgroup v2
freezer too, which are required to conform the cgroup v2 interface
design principles:
1) There is no separate controller, which has to be turned on:
the functionality is always available and is represented by
cgroup.freeze and cgroup.events cgroup control files.
2) The desired state is defined by the cgroup.freeze control file.
Any hierarchical configuration is allowed.
3) The interface is asynchronous. The actual state is available
using cgroup.events control file ("frozen" field). There are no
dedicated transitional states.
4) It's allowed to make any changes with the cgroup hierarchy
(create new cgroups, remove old cgroups, move tasks between cgroups)
no matter if some cgroups are frozen.
Signed-off-by: Roman Gushchin <guro@fb.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
No-objection-from-me-by: Oleg Nesterov <oleg@redhat.com>
Cc: kernel-team@fb.com
2019-04-19 20:03:04 +03:00
cgroup_leave_frozen ( false ) ;
2012-03-06 02:59:13 +04:00
freezer_count ( ) ;
if ( killed ) {
task_lock ( child ) ;
child - > vfork_done = NULL ;
task_unlock ( child ) ;
}
put_task_struct ( child ) ;
return killed ;
2012-03-06 02:59:13 +04:00
}
2005-04-17 02:20:36 +04:00
/* Please note the differences between mmput and mm_release.
* mmput is called whenever we stop holding onto a mm_struct ,
* error success whatever .
*
* mm_release is called after a mm_struct has been removed
* from the current process .
*
* This difference is important for error handling , when we
* only half set up a mm_struct for a new process and need to restore
* the old one . Because we mmput the new mm_struct before
* restoring the old one . . .
* Eric Biederman 10 January 1998
*/
2019-11-07 00:55:38 +03:00
static void mm_release ( struct task_struct * tsk , struct mm_struct * mm )
2005-04-17 02:20:36 +04:00
{
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-13 22:00:11 +04:00
uprobe_free_utask ( tsk ) ;
2005-04-17 02:20:36 +04:00
/* Get rid of any cached register state */
deactivate_mm ( tsk , mm ) ;
2006-12-07 07:36:34 +03:00
/*
2016-09-02 02:15:13 +03:00
* Signal userspace if we ' re not exiting with a core dump
* because we want to leave the value intact for debugging
* purposes .
2006-12-07 07:36:34 +03:00
*/
execve: must clear current->clear_child_tid
While looking at Jens Rosenboom bug report
(http://lkml.org/lkml/2009/7/27/35) about strange sys_futex call done from
a dying "ps" program, we found following problem.
clone() syscall has special support for TID of created threads. This
support includes two features.
One (CLONE_CHILD_SETTID) is to set an integer into user memory with the
TID value.
One (CLONE_CHILD_CLEARTID) is to clear this same integer once the created
thread dies.
The integer location is a user provided pointer, provided at clone()
time.
kernel keeps this pointer value into current->clear_child_tid.
At execve() time, we should make sure kernel doesnt keep this user
provided pointer, as full user memory is replaced by a new one.
As glibc fork() actually uses clone() syscall with CLONE_CHILD_SETTID and
CLONE_CHILD_CLEARTID set, chances are high that we might corrupt user
memory in forked processes.
Following sequence could happen:
1) bash (or any program) starts a new process, by a fork() call that
glibc maps to a clone( ... CLONE_CHILD_SETTID | CLONE_CHILD_CLEARTID
...) syscall
2) When new process starts, its current->clear_child_tid is set to a
location that has a meaning only in bash (or initial program) context
(&THREAD_SELF->tid)
3) This new process does the execve() syscall to start a new program.
current->clear_child_tid is left unchanged (a non NULL value)
4) If this new program creates some threads, and initial thread exits,
kernel will attempt to clear the integer pointed by
current->clear_child_tid from mm_release() :
if (tsk->clear_child_tid
&& !(tsk->flags & PF_SIGNALED)
&& atomic_read(&mm->mm_users) > 1) {
u32 __user * tidptr = tsk->clear_child_tid;
tsk->clear_child_tid = NULL;
/*
* We don't check the error code - if userspace has
* not set up a proper pointer then tough luck.
*/
<< here >> put_user(0, tidptr);
sys_futex(tidptr, FUTEX_WAKE, 1, NULL, NULL, 0);
}
5) OR : if new program is not multi-threaded, but spied by /proc/pid
users (ps command for example), mm_users > 1, and the exiting program
could corrupt 4 bytes in a persistent memory area (shm or memory mapped
file)
If current->clear_child_tid points to a writeable portion of memory of the
new program, kernel happily and silently corrupts 4 bytes of memory, with
unexpected effects.
Fix is straightforward and should not break any sane program.
Reported-by: Jens Rosenboom <jens@mcbone.net>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sonny Rao <sonnyrao@us.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-08-07 02:09:28 +04:00
if ( tsk - > clear_child_tid ) {
2016-09-02 02:15:13 +03:00
if ( ! ( tsk - > signal - > flags & SIGNAL_GROUP_COREDUMP ) & &
execve: must clear current->clear_child_tid
While looking at Jens Rosenboom bug report
(http://lkml.org/lkml/2009/7/27/35) about strange sys_futex call done from
a dying "ps" program, we found following problem.
clone() syscall has special support for TID of created threads. This
support includes two features.
One (CLONE_CHILD_SETTID) is to set an integer into user memory with the
TID value.
One (CLONE_CHILD_CLEARTID) is to clear this same integer once the created
thread dies.
The integer location is a user provided pointer, provided at clone()
time.
kernel keeps this pointer value into current->clear_child_tid.
At execve() time, we should make sure kernel doesnt keep this user
provided pointer, as full user memory is replaced by a new one.
As glibc fork() actually uses clone() syscall with CLONE_CHILD_SETTID and
CLONE_CHILD_CLEARTID set, chances are high that we might corrupt user
memory in forked processes.
Following sequence could happen:
1) bash (or any program) starts a new process, by a fork() call that
glibc maps to a clone( ... CLONE_CHILD_SETTID | CLONE_CHILD_CLEARTID
...) syscall
2) When new process starts, its current->clear_child_tid is set to a
location that has a meaning only in bash (or initial program) context
(&THREAD_SELF->tid)
3) This new process does the execve() syscall to start a new program.
current->clear_child_tid is left unchanged (a non NULL value)
4) If this new program creates some threads, and initial thread exits,
kernel will attempt to clear the integer pointed by
current->clear_child_tid from mm_release() :
if (tsk->clear_child_tid
&& !(tsk->flags & PF_SIGNALED)
&& atomic_read(&mm->mm_users) > 1) {
u32 __user * tidptr = tsk->clear_child_tid;
tsk->clear_child_tid = NULL;
/*
* We don't check the error code - if userspace has
* not set up a proper pointer then tough luck.
*/
<< here >> put_user(0, tidptr);
sys_futex(tidptr, FUTEX_WAKE, 1, NULL, NULL, 0);
}
5) OR : if new program is not multi-threaded, but spied by /proc/pid
users (ps command for example), mm_users > 1, and the exiting program
could corrupt 4 bytes in a persistent memory area (shm or memory mapped
file)
If current->clear_child_tid points to a writeable portion of memory of the
new program, kernel happily and silently corrupts 4 bytes of memory, with
unexpected effects.
Fix is straightforward and should not break any sane program.
Reported-by: Jens Rosenboom <jens@mcbone.net>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sonny Rao <sonnyrao@us.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-08-07 02:09:28 +04:00
atomic_read ( & mm - > mm_users ) > 1 ) {
/*
* We don ' t check the error code - if userspace has
* not set up a proper pointer then tough luck .
*/
put_user ( 0 , tsk - > clear_child_tid ) ;
2018-03-11 13:34:26 +03:00
do_futex ( tsk - > clear_child_tid , FUTEX_WAKE ,
1 , NULL , NULL , 0 , 0 ) ;
execve: must clear current->clear_child_tid
While looking at Jens Rosenboom bug report
(http://lkml.org/lkml/2009/7/27/35) about strange sys_futex call done from
a dying "ps" program, we found following problem.
clone() syscall has special support for TID of created threads. This
support includes two features.
One (CLONE_CHILD_SETTID) is to set an integer into user memory with the
TID value.
One (CLONE_CHILD_CLEARTID) is to clear this same integer once the created
thread dies.
The integer location is a user provided pointer, provided at clone()
time.
kernel keeps this pointer value into current->clear_child_tid.
At execve() time, we should make sure kernel doesnt keep this user
provided pointer, as full user memory is replaced by a new one.
As glibc fork() actually uses clone() syscall with CLONE_CHILD_SETTID and
CLONE_CHILD_CLEARTID set, chances are high that we might corrupt user
memory in forked processes.
Following sequence could happen:
1) bash (or any program) starts a new process, by a fork() call that
glibc maps to a clone( ... CLONE_CHILD_SETTID | CLONE_CHILD_CLEARTID
...) syscall
2) When new process starts, its current->clear_child_tid is set to a
location that has a meaning only in bash (or initial program) context
(&THREAD_SELF->tid)
3) This new process does the execve() syscall to start a new program.
current->clear_child_tid is left unchanged (a non NULL value)
4) If this new program creates some threads, and initial thread exits,
kernel will attempt to clear the integer pointed by
current->clear_child_tid from mm_release() :
if (tsk->clear_child_tid
&& !(tsk->flags & PF_SIGNALED)
&& atomic_read(&mm->mm_users) > 1) {
u32 __user * tidptr = tsk->clear_child_tid;
tsk->clear_child_tid = NULL;
/*
* We don't check the error code - if userspace has
* not set up a proper pointer then tough luck.
*/
<< here >> put_user(0, tidptr);
sys_futex(tidptr, FUTEX_WAKE, 1, NULL, NULL, 0);
}
5) OR : if new program is not multi-threaded, but spied by /proc/pid
users (ps command for example), mm_users > 1, and the exiting program
could corrupt 4 bytes in a persistent memory area (shm or memory mapped
file)
If current->clear_child_tid points to a writeable portion of memory of the
new program, kernel happily and silently corrupts 4 bytes of memory, with
unexpected effects.
Fix is straightforward and should not break any sane program.
Reported-by: Jens Rosenboom <jens@mcbone.net>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sonny Rao <sonnyrao@us.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-08-07 02:09:28 +04:00
}
2005-04-17 02:20:36 +04:00
tsk - > clear_child_tid = NULL ;
}
2012-06-01 03:26:21 +04:00
/*
* All done , finally we can wake up parent and return this mm to him .
* Also kthread_stop ( ) uses this completion for synchronization .
*/
if ( tsk - > vfork_done )
complete_vfork_done ( tsk ) ;
2005-04-17 02:20:36 +04:00
}
2019-11-07 00:55:38 +03:00
void exit_mm_release ( struct task_struct * tsk , struct mm_struct * mm )
{
2019-11-07 00:55:39 +03:00
futex_exit_release ( tsk ) ;
2019-11-07 00:55:38 +03:00
mm_release ( tsk , mm ) ;
}
void exec_mm_release ( struct task_struct * tsk , struct mm_struct * mm )
{
2019-11-07 00:55:39 +03:00
futex_exec_release ( tsk ) ;
2019-11-07 00:55:38 +03:00
mm_release ( tsk , mm ) ;
}
2019-04-26 03:11:25 +03:00
/**
* dup_mm ( ) - duplicates an existing mm structure
* @ tsk : the task_struct with which the new mm will be associated .
* @ oldmm : the mm to duplicate .
*
* Allocates a new mm structure and duplicates the provided @ oldmm structure
* content into it .
*
* Return : the duplicated mm or NULL on failure .
2006-02-07 23:59:01 +03:00
*/
2019-04-26 03:11:25 +03:00
static struct mm_struct * dup_mm ( struct task_struct * tsk ,
struct mm_struct * oldmm )
2006-02-07 23:59:01 +03:00
{
2019-04-26 03:11:25 +03:00
struct mm_struct * mm ;
2006-02-07 23:59:01 +03:00
int err ;
mm = allocate_mm ( ) ;
if ( ! mm )
goto fail_nomem ;
memcpy ( mm , oldmm , sizeof ( * mm ) ) ;
2016-10-14 05:23:16 +03:00
if ( ! mm_init ( mm , tsk , mm - > user_ns ) )
2006-02-07 23:59:01 +03:00
goto fail_nomem ;
err = dup_mmap ( mm , oldmm ) ;
if ( err )
goto free_pt ;
mm - > hiwater_rss = get_mm_rss ( mm ) ;
mm - > hiwater_vm = mm - > total_vm ;
2009-09-24 02:57:41 +04:00
if ( mm - > binfmt & & ! try_module_get ( mm - > binfmt - > module ) )
goto free_pt ;
2006-02-07 23:59:01 +03:00
return mm ;
free_pt :
2009-09-24 02:57:41 +04:00
/* don't put binfmt in mmput, we haven't got module yet */
mm - > binfmt = NULL ;
2019-05-15 01:40:46 +03:00
mm_init_owner ( mm , NULL ) ;
2006-02-07 23:59:01 +03:00
mmput ( mm ) ;
fail_nomem :
return NULL ;
}
2011-07-27 03:08:39 +04:00
static int copy_mm ( unsigned long clone_flags , struct task_struct * tsk )
2005-04-17 02:20:36 +04:00
{
2011-07-27 03:08:39 +04:00
struct mm_struct * mm , * oldmm ;
2005-04-17 02:20:36 +04:00
tsk - > min_flt = tsk - > maj_flt = 0 ;
tsk - > nvcsw = tsk - > nivcsw = 0 ;
2009-02-07 02:37:47 +03:00
# ifdef CONFIG_DETECT_HUNG_TASK
tsk - > last_switch_count = tsk - > nvcsw + tsk - > nivcsw ;
2018-08-22 07:55:52 +03:00
tsk - > last_switch_time = 0 ;
2009-02-07 02:37:47 +03:00
# endif
2005-04-17 02:20:36 +04:00
tsk - > mm = NULL ;
tsk - > active_mm = NULL ;
/*
* Are we cloning a kernel thread ?
*
* We need to steal a active VM for that . .
*/
oldmm = current - > mm ;
if ( ! oldmm )
return 0 ;
mm: per-thread vma caching
This patch is a continuation of efforts trying to optimize find_vma(),
avoiding potentially expensive rbtree walks to locate a vma upon faults.
The original approach (https://lkml.org/lkml/2013/11/1/410), where the
largest vma was also cached, ended up being too specific and random,
thus further comparison with other approaches were needed. There are
two things to consider when dealing with this, the cache hit rate and
the latency of find_vma(). Improving the hit-rate does not necessarily
translate in finding the vma any faster, as the overhead of any fancy
caching schemes can be too high to consider.
We currently cache the last used vma for the whole address space, which
provides a nice optimization, reducing the total cycles in find_vma() by
up to 250%, for workloads with good locality. On the other hand, this
simple scheme is pretty much useless for workloads with poor locality.
Analyzing ebizzy runs shows that, no matter how many threads are
running, the mmap_cache hit rate is less than 2%, and in many situations
below 1%.
The proposed approach is to replace this scheme with a small per-thread
cache, maximizing hit rates at a very low maintenance cost.
Invalidations are performed by simply bumping up a 32-bit sequence
number. The only expensive operation is in the rare case of a seq
number overflow, where all caches that share the same address space are
flushed. Upon a miss, the proposed replacement policy is based on the
page number that contains the virtual address in question. Concretely,
the following results are seen on an 80 core, 8 socket x86-64 box:
1) System bootup: Most programs are single threaded, so the per-thread
scheme does improve ~50% hit rate by just adding a few more slots to
the cache.
+----------------+----------+------------------+
| caching scheme | hit-rate | cycles (billion) |
+----------------+----------+------------------+
| baseline | 50.61% | 19.90 |
| patched | 73.45% | 13.58 |
+----------------+----------+------------------+
2) Kernel build: This one is already pretty good with the current
approach as we're dealing with good locality.
+----------------+----------+------------------+
| caching scheme | hit-rate | cycles (billion) |
+----------------+----------+------------------+
| baseline | 75.28% | 11.03 |
| patched | 88.09% | 9.31 |
+----------------+----------+------------------+
3) Oracle 11g Data Mining (4k pages): Similar to the kernel build workload.
+----------------+----------+------------------+
| caching scheme | hit-rate | cycles (billion) |
+----------------+----------+------------------+
| baseline | 70.66% | 17.14 |
| patched | 91.15% | 12.57 |
+----------------+----------+------------------+
4) Ebizzy: There's a fair amount of variation from run to run, but this
approach always shows nearly perfect hit rates, while baseline is just
about non-existent. The amounts of cycles can fluctuate between
anywhere from ~60 to ~116 for the baseline scheme, but this approach
reduces it considerably. For instance, with 80 threads:
+----------------+----------+------------------+
| caching scheme | hit-rate | cycles (billion) |
+----------------+----------+------------------+
| baseline | 1.06% | 91.54 |
| patched | 99.97% | 14.18 |
+----------------+----------+------------------+
[akpm@linux-foundation.org: fix nommu build, per Davidlohr]
[akpm@linux-foundation.org: document vmacache_valid() logic]
[akpm@linux-foundation.org: attempt to untangle header files]
[akpm@linux-foundation.org: add vmacache_find() BUG_ON]
[hughd@google.com: add vmacache_valid_mm() (from Oleg)]
[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: adjust and enhance comments]
Signed-off-by: Davidlohr Bueso <davidlohr@hp.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Reviewed-by: Michel Lespinasse <walken@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Tested-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-08 02:37:25 +04:00
/* initialize the new vmacache entries */
vmacache_flush ( tsk ) ;
2005-04-17 02:20:36 +04:00
if ( clone_flags & CLONE_VM ) {
2017-02-28 01:30:10 +03:00
mmget ( oldmm ) ;
2005-04-17 02:20:36 +04:00
mm = oldmm ;
2021-05-07 04:04:25 +03:00
} else {
mm = dup_mm ( tsk , current - > mm ) ;
if ( ! mm )
return - ENOMEM ;
2005-04-17 02:20:36 +04:00
}
tsk - > mm = mm ;
tsk - > active_mm = mm ;
return 0 ;
}
2007-10-19 10:41:10 +04:00
static int copy_fs ( unsigned long clone_flags , struct task_struct * tsk )
2005-04-17 02:20:36 +04:00
{
2009-03-30 15:20:30 +04:00
struct fs_struct * fs = current - > fs ;
2005-04-17 02:20:36 +04:00
if ( clone_flags & CLONE_FS ) {
2009-03-30 15:20:30 +04:00
/* tsk->fs is already what we want */
2010-08-17 22:37:33 +04:00
spin_lock ( & fs - > lock ) ;
2009-03-30 15:20:30 +04:00
if ( fs - > in_exec ) {
2010-08-17 22:37:33 +04:00
spin_unlock ( & fs - > lock ) ;
2009-03-30 15:20:30 +04:00
return - EAGAIN ;
}
fs - > users + + ;
2010-08-17 22:37:33 +04:00
spin_unlock ( & fs - > lock ) ;
2005-04-17 02:20:36 +04:00
return 0 ;
}
2009-03-30 15:20:30 +04:00
tsk - > fs = copy_fs_struct ( fs ) ;
2005-04-17 02:20:36 +04:00
if ( ! tsk - > fs )
return - ENOMEM ;
return 0 ;
}
2011-07-27 03:08:39 +04:00
static int copy_files ( unsigned long clone_flags , struct task_struct * tsk )
2006-02-07 23:59:02 +03:00
{
struct files_struct * oldf , * newf ;
int error = 0 ;
/*
* A background process may not have any files . . .
*/
oldf = current - > files ;
if ( ! oldf )
goto out ;
if ( clone_flags & CLONE_FILES ) {
atomic_inc ( & oldf - > count ) ;
goto out ;
}
2020-06-03 22:48:55 +03:00
newf = dup_fd ( oldf , NR_OPEN_MAX , & error ) ;
2006-02-07 23:59:02 +03:00
if ( ! newf )
goto out ;
tsk - > files = newf ;
error = 0 ;
out :
return error ;
}
2008-01-24 10:54:47 +03:00
static int copy_io ( unsigned long clone_flags , struct task_struct * tsk )
2008-01-24 10:52:45 +03:00
{
# ifdef CONFIG_BLOCK
struct io_context * ioc = current - > io_context ;
block: make ioc get/put interface more conventional and fix race on alloction
Ignoring copy_io() during fork, io_context can be allocated from two
places - current_io_context() and set_task_ioprio(). The former is
always called from local task while the latter can be called from
different task. The synchornization between them are peculiar and
dubious.
* current_io_context() doesn't grab task_lock() and assumes that if it
saw %NULL ->io_context, it would stay that way until allocation and
assignment is complete. It has smp_wmb() between alloc/init and
assignment.
* set_task_ioprio() grabs task_lock() for assignment and does
smp_read_barrier_depends() between "ioc = task->io_context" and "if
(ioc)". Unfortunately, this doesn't achieve anything - the latter
is not a dependent load of the former. ie, if ioc itself were being
dereferenced "ioc->xxx", it would mean something (not sure what tho)
but as the code currently stands, the dependent read barrier is
noop.
As only one of the the two test-assignment sequences is task_lock()
protected, the task_lock() can't do much about race between the two.
Nothing prevents current_io_context() and set_task_ioprio() allocating
its own ioc for the same task and overwriting the other's.
Also, set_task_ioprio() can race with exiting task and create a new
ioc after exit_io_context() is finished.
ioc get/put doesn't have any reason to be complex. The only hot path
is accessing the existing ioc of %current, which is simple to achieve
given that ->io_context is never destroyed as long as the task is
alive. All other paths can happily go through task_lock() like all
other task sub structures without impacting anything.
This patch updates ioc get/put so that it becomes more conventional.
* alloc_io_context() is replaced with get_task_io_context(). This is
the only interface which can acquire access to ioc of another task.
On return, the caller has an explicit reference to the object which
should be put using put_io_context() afterwards.
* The functionality of current_io_context() remains the same but when
creating a new ioc, it shares the code path with
get_task_io_context() and always goes through task_lock().
* get_io_context() now means incrementing ref on an ioc which the
caller already has access to (be that an explicit refcnt or implicit
%current one).
* PF_EXITING inhibits creation of new io_context and once
exit_io_context() is finished, it's guaranteed that both ioc
acquisition functions return %NULL.
* All users are updated. Most are trivial but
smp_read_barrier_depends() removal from cfq_get_io_context() needs a
bit of explanation. I suppose the original intention was to ensure
ioc->ioprio is visible when set_task_ioprio() allocates new
io_context and installs it; however, this wouldn't have worked
because set_task_ioprio() doesn't have wmb between init and install.
There are other problems with this which will be fixed in another
patch.
* While at it, use NUMA_NO_NODE instead of -1 for wildcard node
specification.
-v2: Vivek spotted contamination from debug patch. Removed.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2011-12-14 03:33:38 +04:00
struct io_context * new_ioc ;
2008-01-24 10:52:45 +03:00
if ( ! ioc )
return 0 ;
2008-01-24 10:54:47 +03:00
/*
* Share io context with parent , if CLONE_IO is set
*/
if ( clone_flags & CLONE_IO ) {
2012-03-06 01:15:25 +04:00
ioc_task_link ( ioc ) ;
tsk - > io_context = ioc ;
2008-01-24 10:54:47 +03:00
} else if ( ioprio_valid ( ioc - > ioprio ) ) {
block: make ioc get/put interface more conventional and fix race on alloction
Ignoring copy_io() during fork, io_context can be allocated from two
places - current_io_context() and set_task_ioprio(). The former is
always called from local task while the latter can be called from
different task. The synchornization between them are peculiar and
dubious.
* current_io_context() doesn't grab task_lock() and assumes that if it
saw %NULL ->io_context, it would stay that way until allocation and
assignment is complete. It has smp_wmb() between alloc/init and
assignment.
* set_task_ioprio() grabs task_lock() for assignment and does
smp_read_barrier_depends() between "ioc = task->io_context" and "if
(ioc)". Unfortunately, this doesn't achieve anything - the latter
is not a dependent load of the former. ie, if ioc itself were being
dereferenced "ioc->xxx", it would mean something (not sure what tho)
but as the code currently stands, the dependent read barrier is
noop.
As only one of the the two test-assignment sequences is task_lock()
protected, the task_lock() can't do much about race between the two.
Nothing prevents current_io_context() and set_task_ioprio() allocating
its own ioc for the same task and overwriting the other's.
Also, set_task_ioprio() can race with exiting task and create a new
ioc after exit_io_context() is finished.
ioc get/put doesn't have any reason to be complex. The only hot path
is accessing the existing ioc of %current, which is simple to achieve
given that ->io_context is never destroyed as long as the task is
alive. All other paths can happily go through task_lock() like all
other task sub structures without impacting anything.
This patch updates ioc get/put so that it becomes more conventional.
* alloc_io_context() is replaced with get_task_io_context(). This is
the only interface which can acquire access to ioc of another task.
On return, the caller has an explicit reference to the object which
should be put using put_io_context() afterwards.
* The functionality of current_io_context() remains the same but when
creating a new ioc, it shares the code path with
get_task_io_context() and always goes through task_lock().
* get_io_context() now means incrementing ref on an ioc which the
caller already has access to (be that an explicit refcnt or implicit
%current one).
* PF_EXITING inhibits creation of new io_context and once
exit_io_context() is finished, it's guaranteed that both ioc
acquisition functions return %NULL.
* All users are updated. Most are trivial but
smp_read_barrier_depends() removal from cfq_get_io_context() needs a
bit of explanation. I suppose the original intention was to ensure
ioc->ioprio is visible when set_task_ioprio() allocates new
io_context and installs it; however, this wouldn't have worked
because set_task_ioprio() doesn't have wmb between init and install.
There are other problems with this which will be fixed in another
patch.
* While at it, use NUMA_NO_NODE instead of -1 for wildcard node
specification.
-v2: Vivek spotted contamination from debug patch. Removed.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2011-12-14 03:33:38 +04:00
new_ioc = get_task_io_context ( tsk , GFP_KERNEL , NUMA_NO_NODE ) ;
if ( unlikely ( ! new_ioc ) )
2008-01-24 10:52:45 +03:00
return - ENOMEM ;
block: make ioc get/put interface more conventional and fix race on alloction
Ignoring copy_io() during fork, io_context can be allocated from two
places - current_io_context() and set_task_ioprio(). The former is
always called from local task while the latter can be called from
different task. The synchornization between them are peculiar and
dubious.
* current_io_context() doesn't grab task_lock() and assumes that if it
saw %NULL ->io_context, it would stay that way until allocation and
assignment is complete. It has smp_wmb() between alloc/init and
assignment.
* set_task_ioprio() grabs task_lock() for assignment and does
smp_read_barrier_depends() between "ioc = task->io_context" and "if
(ioc)". Unfortunately, this doesn't achieve anything - the latter
is not a dependent load of the former. ie, if ioc itself were being
dereferenced "ioc->xxx", it would mean something (not sure what tho)
but as the code currently stands, the dependent read barrier is
noop.
As only one of the the two test-assignment sequences is task_lock()
protected, the task_lock() can't do much about race between the two.
Nothing prevents current_io_context() and set_task_ioprio() allocating
its own ioc for the same task and overwriting the other's.
Also, set_task_ioprio() can race with exiting task and create a new
ioc after exit_io_context() is finished.
ioc get/put doesn't have any reason to be complex. The only hot path
is accessing the existing ioc of %current, which is simple to achieve
given that ->io_context is never destroyed as long as the task is
alive. All other paths can happily go through task_lock() like all
other task sub structures without impacting anything.
This patch updates ioc get/put so that it becomes more conventional.
* alloc_io_context() is replaced with get_task_io_context(). This is
the only interface which can acquire access to ioc of another task.
On return, the caller has an explicit reference to the object which
should be put using put_io_context() afterwards.
* The functionality of current_io_context() remains the same but when
creating a new ioc, it shares the code path with
get_task_io_context() and always goes through task_lock().
* get_io_context() now means incrementing ref on an ioc which the
caller already has access to (be that an explicit refcnt or implicit
%current one).
* PF_EXITING inhibits creation of new io_context and once
exit_io_context() is finished, it's guaranteed that both ioc
acquisition functions return %NULL.
* All users are updated. Most are trivial but
smp_read_barrier_depends() removal from cfq_get_io_context() needs a
bit of explanation. I suppose the original intention was to ensure
ioc->ioprio is visible when set_task_ioprio() allocates new
io_context and installs it; however, this wouldn't have worked
because set_task_ioprio() doesn't have wmb between init and install.
There are other problems with this which will be fixed in another
patch.
* While at it, use NUMA_NO_NODE instead of -1 for wildcard node
specification.
-v2: Vivek spotted contamination from debug patch. Removed.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2011-12-14 03:33:38 +04:00
new_ioc - > ioprio = ioc - > ioprio ;
2012-02-07 10:51:30 +04:00
put_io_context ( new_ioc ) ;
2008-01-24 10:52:45 +03:00
}
# endif
return 0 ;
}
2007-10-19 10:41:10 +04:00
static int copy_sighand ( unsigned long clone_flags , struct task_struct * tsk )
2005-04-17 02:20:36 +04:00
{
struct sighand_struct * sig ;
2009-01-07 01:40:46 +03:00
if ( clone_flags & CLONE_SIGHAND ) {
2019-01-18 15:27:26 +03:00
refcount_inc ( & current - > sighand - > count ) ;
2005-04-17 02:20:36 +04:00
return 0 ;
}
sig = kmem_cache_alloc ( sighand_cachep , GFP_KERNEL ) ;
2020-01-27 20:58:21 +03:00
RCU_INIT_POINTER ( tsk - > sighand , sig ) ;
2005-04-17 02:20:36 +04:00
if ( ! sig )
return - ENOMEM ;
2015-06-30 12:30:54 +03:00
2019-01-18 15:27:26 +03:00
refcount_set ( & sig - > count , 1 ) ;
2018-08-22 08:00:58 +03:00
spin_lock_irq ( & current - > sighand - > siglock ) ;
2005-04-17 02:20:36 +04:00
memcpy ( sig - > action , current - > sighand - > action , sizeof ( sig - > action ) ) ;
2018-08-22 08:00:58 +03:00
spin_unlock_irq ( & current - > sighand - > siglock ) ;
2019-10-14 13:45:37 +03:00
/* Reset all signal handler not set to SIG_IGN to SIG_DFL. */
if ( clone_flags & CLONE_CLEAR_SIGHAND )
flush_signal_handlers ( tsk , 0 ) ;
2005-04-17 02:20:36 +04:00
return 0 ;
}
2006-03-29 04:11:27 +04:00
void __cleanup_sighand ( struct sighand_struct * sighand )
2006-03-29 04:11:17 +04:00
{
2019-01-18 15:27:26 +03:00
if ( refcount_dec_and_test ( & sighand - > count ) ) {
epoll: introduce POLLFREE to flush ->signalfd_wqh before kfree()
This patch is intentionally incomplete to simplify the review.
It ignores ep_unregister_pollwait() which plays with the same wqh.
See the next change.
epoll assumes that the EPOLL_CTL_ADD'ed file controls everything
f_op->poll() needs. In particular it assumes that the wait queue
can't go away until eventpoll_release(). This is not true in case
of signalfd, the task which does EPOLL_CTL_ADD uses its ->sighand
which is not connected to the file.
This patch adds the special event, POLLFREE, currently only for
epoll. It expects that init_poll_funcptr()'ed hook should do the
necessary cleanup. Perhaps it should be defined as EPOLLFREE in
eventpoll.
__cleanup_sighand() is changed to do wake_up_poll(POLLFREE) if
->signalfd_wqh is not empty, we add the new signalfd_cleanup()
helper.
ep_poll_callback(POLLFREE) simply does list_del_init(task_list).
This make this poll entry inconsistent, but we don't care. If you
share epoll fd which contains our sigfd with another process you
should blame yourself. signalfd is "really special". I simply do
not know how we can define the "right" semantics if it used with
epoll.
The main problem is, epoll calls signalfd_poll() once to establish
the connection with the wait queue, after that signalfd_poll(NULL)
returns the different/inconsistent results depending on who does
EPOLL_CTL_MOD/signalfd_read/etc. IOW: apart from sigmask, signalfd
has nothing to do with the file, it works with the current thread.
In short: this patch is the hack which tries to fix the symptoms.
It also assumes that nobody can take tasklist_lock under epoll
locks, this seems to be true.
Note:
- we do not have wake_up_all_poll() but wake_up_poll()
is fine, poll/epoll doesn't use WQ_FLAG_EXCLUSIVE.
- signalfd_cleanup() uses POLLHUP along with POLLFREE,
we need a couple of simple changes in eventpoll.c to
make sure it can't be "lost".
Reported-by: Maxime Bizon <mbizon@freebox.fr>
Cc: <stable@kernel.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-02-24 23:07:11 +04:00
signalfd_cleanup ( sighand ) ;
2014-09-29 01:44:18 +04:00
/*
2017-01-18 13:53:44 +03:00
* sighand_cachep is SLAB_TYPESAFE_BY_RCU so we can free it
2014-09-29 01:44:18 +04:00
* without an RCU grace period , see __lock_task_sighand ( ) .
*/
2006-03-29 04:11:17 +04:00
kmem_cache_free ( sighand_cachep , sighand ) ;
epoll: introduce POLLFREE to flush ->signalfd_wqh before kfree()
This patch is intentionally incomplete to simplify the review.
It ignores ep_unregister_pollwait() which plays with the same wqh.
See the next change.
epoll assumes that the EPOLL_CTL_ADD'ed file controls everything
f_op->poll() needs. In particular it assumes that the wait queue
can't go away until eventpoll_release(). This is not true in case
of signalfd, the task which does EPOLL_CTL_ADD uses its ->sighand
which is not connected to the file.
This patch adds the special event, POLLFREE, currently only for
epoll. It expects that init_poll_funcptr()'ed hook should do the
necessary cleanup. Perhaps it should be defined as EPOLLFREE in
eventpoll.
__cleanup_sighand() is changed to do wake_up_poll(POLLFREE) if
->signalfd_wqh is not empty, we add the new signalfd_cleanup()
helper.
ep_poll_callback(POLLFREE) simply does list_del_init(task_list).
This make this poll entry inconsistent, but we don't care. If you
share epoll fd which contains our sigfd with another process you
should blame yourself. signalfd is "really special". I simply do
not know how we can define the "right" semantics if it used with
epoll.
The main problem is, epoll calls signalfd_poll() once to establish
the connection with the wait queue, after that signalfd_poll(NULL)
returns the different/inconsistent results depending on who does
EPOLL_CTL_MOD/signalfd_read/etc. IOW: apart from sigmask, signalfd
has nothing to do with the file, it works with the current thread.
In short: this patch is the hack which tries to fix the symptoms.
It also assumes that nobody can take tasklist_lock under epoll
locks, this seems to be true.
Note:
- we do not have wake_up_all_poll() but wake_up_poll()
is fine, poll/epoll doesn't use WQ_FLAG_EXCLUSIVE.
- signalfd_cleanup() uses POLLHUP along with POLLFREE,
we need a couple of simple changes in eventpoll.c to
make sure it can't be "lost".
Reported-by: Maxime Bizon <mbizon@freebox.fr>
Cc: <stable@kernel.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-02-24 23:07:11 +04:00
}
2006-03-29 04:11:17 +04:00
}
timers: fix itimer/many thread hang
Overview
This patch reworks the handling of POSIX CPU timers, including the
ITIMER_PROF, ITIMER_VIRT timers and rlimit handling. It was put together
with the help of Roland McGrath, the owner and original writer of this code.
The problem we ran into, and the reason for this rework, has to do with using
a profiling timer in a process with a large number of threads. It appears
that the performance of the old implementation of run_posix_cpu_timers() was
at least O(n*3) (where "n" is the number of threads in a process) or worse.
Everything is fine with an increasing number of threads until the time taken
for that routine to run becomes the same as or greater than the tick time, at
which point things degrade rather quickly.
This patch fixes bug 9906, "Weird hang with NPTL and SIGPROF."
Code Changes
This rework corrects the implementation of run_posix_cpu_timers() to make it
run in constant time for a particular machine. (Performance may vary between
one machine and another depending upon whether the kernel is built as single-
or multiprocessor and, in the latter case, depending upon the number of
running processors.) To do this, at each tick we now update fields in
signal_struct as well as task_struct. The run_posix_cpu_timers() function
uses those fields to make its decisions.
We define a new structure, "task_cputime," to contain user, system and
scheduler times and use these in appropriate places:
struct task_cputime {
cputime_t utime;
cputime_t stime;
unsigned long long sum_exec_runtime;
};
This is included in the structure "thread_group_cputime," which is a new
substructure of signal_struct and which varies for uniprocessor versus
multiprocessor kernels. For uniprocessor kernels, it uses "task_cputime" as
a simple substructure, while for multiprocessor kernels it is a pointer:
struct thread_group_cputime {
struct task_cputime totals;
};
struct thread_group_cputime {
struct task_cputime *totals;
};
We also add a new task_cputime substructure directly to signal_struct, to
cache the earliest expiration of process-wide timers, and task_cputime also
replaces the it_*_expires fields of task_struct (used for earliest expiration
of thread timers). The "thread_group_cputime" structure contains process-wide
timers that are updated via account_user_time() and friends. In the non-SMP
case the structure is a simple aggregator; unfortunately in the SMP case that
simplicity was not achievable due to cache-line contention between CPUs (in
one measured case performance was actually _worse_ on a 16-cpu system than
the same test on a 4-cpu system, due to this contention). For SMP, the
thread_group_cputime counters are maintained as a per-cpu structure allocated
using alloc_percpu(). The timer functions update only the timer field in
the structure corresponding to the running CPU, obtained using per_cpu_ptr().
We define a set of inline functions in sched.h that we use to maintain the
thread_group_cputime structure and hide the differences between UP and SMP
implementations from the rest of the kernel. The thread_group_cputime_init()
function initializes the thread_group_cputime structure for the given task.
The thread_group_cputime_alloc() is a no-op for UP; for SMP it calls the
out-of-line function thread_group_cputime_alloc_smp() to allocate and fill
in the per-cpu structures and fields. The thread_group_cputime_free()
function, also a no-op for UP, in SMP frees the per-cpu structures. The
thread_group_cputime_clone_thread() function (also a UP no-op) for SMP calls
thread_group_cputime_alloc() if the per-cpu structures haven't yet been
allocated. The thread_group_cputime() function fills the task_cputime
structure it is passed with the contents of the thread_group_cputime fields;
in UP it's that simple but in SMP it must also safely check that tsk->signal
is non-NULL (if it is it just uses the appropriate fields of task_struct) and,
if so, sums the per-cpu values for each online CPU. Finally, the three
functions account_group_user_time(), account_group_system_time() and
account_group_exec_runtime() are used by timer functions to update the
respective fields of the thread_group_cputime structure.
Non-SMP operation is trivial and will not be mentioned further.
The per-cpu structure is always allocated when a task creates its first new
thread, via a call to thread_group_cputime_clone_thread() from copy_signal().
It is freed at process exit via a call to thread_group_cputime_free() from
cleanup_signal().
All functions that formerly summed utime/stime/sum_sched_runtime values from
from all threads in the thread group now use thread_group_cputime() to
snapshot the values in the thread_group_cputime structure or the values in
the task structure itself if the per-cpu structure hasn't been allocated.
Finally, the code in kernel/posix-cpu-timers.c has changed quite a bit.
The run_posix_cpu_timers() function has been split into a fast path and a
slow path; the former safely checks whether there are any expired thread
timers and, if not, just returns, while the slow path does the heavy lifting.
With the dedicated thread group fields, timers are no longer "rebalanced" and
the process_timer_rebalance() function and related code has gone away. All
summing loops are gone and all code that used them now uses the
thread_group_cputime() inline. When process-wide timers are set, the new
task_cputime structure in signal_struct is used to cache the earliest
expiration; this is checked in the fast path.
Performance
The fix appears not to add significant overhead to existing operations. It
generally performs the same as the current code except in two cases, one in
which it performs slightly worse (Case 5 below) and one in which it performs
very significantly better (Case 2 below). Overall it's a wash except in those
two cases.
I've since done somewhat more involved testing on a dual-core Opteron system.
Case 1: With no itimer running, for a test with 100,000 threads, the fixed
kernel took 1428.5 seconds, 513 seconds more than the unfixed system,
all of which was spent in the system. There were twice as many
voluntary context switches with the fix as without it.
Case 2: With an itimer running at .01 second ticks and 4000 threads (the most
an unmodified kernel can handle), the fixed kernel ran the test in
eight percent of the time (5.8 seconds as opposed to 70 seconds) and
had better tick accuracy (.012 seconds per tick as opposed to .023
seconds per tick).
Case 3: A 4000-thread test with an initial timer tick of .01 second and an
interval of 10,000 seconds (i.e. a timer that ticks only once) had
very nearly the same performance in both cases: 6.3 seconds elapsed
for the fixed kernel versus 5.5 seconds for the unfixed kernel.
With fewer threads (eight in these tests), the Case 1 test ran in essentially
the same time on both the modified and unmodified kernels (5.2 seconds versus
5.8 seconds). The Case 2 test ran in about the same time as well, 5.9 seconds
versus 5.4 seconds but again with much better tick accuracy, .013 seconds per
tick versus .025 seconds per tick for the unmodified kernel.
Since the fix affected the rlimit code, I also tested soft and hard CPU limits.
Case 4: With a hard CPU limit of 20 seconds and eight threads (and an itimer
running), the modified kernel was very slightly favored in that while
it killed the process in 19.997 seconds of CPU time (5.002 seconds of
wall time), only .003 seconds of that was system time, the rest was
user time. The unmodified kernel killed the process in 20.001 seconds
of CPU (5.014 seconds of wall time) of which .016 seconds was system
time. Really, though, the results were too close to call. The results
were essentially the same with no itimer running.
Case 5: With a soft limit of 20 seconds and a hard limit of 2000 seconds
(where the hard limit would never be reached) and an itimer running,
the modified kernel exhibited worse tick accuracy than the unmodified
kernel: .050 seconds/tick versus .028 seconds/tick. Otherwise,
performance was almost indistinguishable. With no itimer running this
test exhibited virtually identical behavior and times in both cases.
In times past I did some limited performance testing. those results are below.
On a four-cpu Opteron system without this fix, a sixteen-thread test executed
in 3569.991 seconds, of which user was 3568.435s and system was 1.556s. On
the same system with the fix, user and elapsed time were about the same, but
system time dropped to 0.007 seconds. Performance with eight, four and one
thread were comparable. Interestingly, the timer ticks with the fix seemed
more accurate: The sixteen-thread test with the fix received 149543 ticks
for 0.024 seconds per tick, while the same test without the fix received 58720
for 0.061 seconds per tick. Both cases were configured for an interval of
0.01 seconds. Again, the other tests were comparable. Each thread in this
test computed the primes up to 25,000,000.
I also did a test with a large number of threads, 100,000 threads, which is
impossible without the fix. In this case each thread computed the primes only
up to 10,000 (to make the runtime manageable). System time dominated, at
1546.968 seconds out of a total 2176.906 seconds (giving a user time of
629.938s). It received 147651 ticks for 0.015 seconds per tick, still quite
accurate. There is obviously no comparable test without the fix.
Signed-off-by: Frank Mayhar <fmayhar@google.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-12 20:54:39 +04:00
/*
* Initialize POSIX timer handling for a thread group .
*/
static void posix_cpu_timers_init_group ( struct signal_struct * sig )
{
2019-08-21 22:09:04 +03:00
struct posix_cputimers * pct = & sig - > posix_cputimers ;
2010-03-06 00:42:54 +03:00
unsigned long cpu_limit ;
2015-04-28 23:00:20 +03:00
cpu_limit = READ_ONCE ( sig - > rlim [ RLIMIT_CPU ] . rlim_cur ) ;
2019-08-21 22:09:06 +03:00
posix_cputimers_group_init ( pct , cpu_limit ) ;
timers: fix itimer/many thread hang
Overview
This patch reworks the handling of POSIX CPU timers, including the
ITIMER_PROF, ITIMER_VIRT timers and rlimit handling. It was put together
with the help of Roland McGrath, the owner and original writer of this code.
The problem we ran into, and the reason for this rework, has to do with using
a profiling timer in a process with a large number of threads. It appears
that the performance of the old implementation of run_posix_cpu_timers() was
at least O(n*3) (where "n" is the number of threads in a process) or worse.
Everything is fine with an increasing number of threads until the time taken
for that routine to run becomes the same as or greater than the tick time, at
which point things degrade rather quickly.
This patch fixes bug 9906, "Weird hang with NPTL and SIGPROF."
Code Changes
This rework corrects the implementation of run_posix_cpu_timers() to make it
run in constant time for a particular machine. (Performance may vary between
one machine and another depending upon whether the kernel is built as single-
or multiprocessor and, in the latter case, depending upon the number of
running processors.) To do this, at each tick we now update fields in
signal_struct as well as task_struct. The run_posix_cpu_timers() function
uses those fields to make its decisions.
We define a new structure, "task_cputime," to contain user, system and
scheduler times and use these in appropriate places:
struct task_cputime {
cputime_t utime;
cputime_t stime;
unsigned long long sum_exec_runtime;
};
This is included in the structure "thread_group_cputime," which is a new
substructure of signal_struct and which varies for uniprocessor versus
multiprocessor kernels. For uniprocessor kernels, it uses "task_cputime" as
a simple substructure, while for multiprocessor kernels it is a pointer:
struct thread_group_cputime {
struct task_cputime totals;
};
struct thread_group_cputime {
struct task_cputime *totals;
};
We also add a new task_cputime substructure directly to signal_struct, to
cache the earliest expiration of process-wide timers, and task_cputime also
replaces the it_*_expires fields of task_struct (used for earliest expiration
of thread timers). The "thread_group_cputime" structure contains process-wide
timers that are updated via account_user_time() and friends. In the non-SMP
case the structure is a simple aggregator; unfortunately in the SMP case that
simplicity was not achievable due to cache-line contention between CPUs (in
one measured case performance was actually _worse_ on a 16-cpu system than
the same test on a 4-cpu system, due to this contention). For SMP, the
thread_group_cputime counters are maintained as a per-cpu structure allocated
using alloc_percpu(). The timer functions update only the timer field in
the structure corresponding to the running CPU, obtained using per_cpu_ptr().
We define a set of inline functions in sched.h that we use to maintain the
thread_group_cputime structure and hide the differences between UP and SMP
implementations from the rest of the kernel. The thread_group_cputime_init()
function initializes the thread_group_cputime structure for the given task.
The thread_group_cputime_alloc() is a no-op for UP; for SMP it calls the
out-of-line function thread_group_cputime_alloc_smp() to allocate and fill
in the per-cpu structures and fields. The thread_group_cputime_free()
function, also a no-op for UP, in SMP frees the per-cpu structures. The
thread_group_cputime_clone_thread() function (also a UP no-op) for SMP calls
thread_group_cputime_alloc() if the per-cpu structures haven't yet been
allocated. The thread_group_cputime() function fills the task_cputime
structure it is passed with the contents of the thread_group_cputime fields;
in UP it's that simple but in SMP it must also safely check that tsk->signal
is non-NULL (if it is it just uses the appropriate fields of task_struct) and,
if so, sums the per-cpu values for each online CPU. Finally, the three
functions account_group_user_time(), account_group_system_time() and
account_group_exec_runtime() are used by timer functions to update the
respective fields of the thread_group_cputime structure.
Non-SMP operation is trivial and will not be mentioned further.
The per-cpu structure is always allocated when a task creates its first new
thread, via a call to thread_group_cputime_clone_thread() from copy_signal().
It is freed at process exit via a call to thread_group_cputime_free() from
cleanup_signal().
All functions that formerly summed utime/stime/sum_sched_runtime values from
from all threads in the thread group now use thread_group_cputime() to
snapshot the values in the thread_group_cputime structure or the values in
the task structure itself if the per-cpu structure hasn't been allocated.
Finally, the code in kernel/posix-cpu-timers.c has changed quite a bit.
The run_posix_cpu_timers() function has been split into a fast path and a
slow path; the former safely checks whether there are any expired thread
timers and, if not, just returns, while the slow path does the heavy lifting.
With the dedicated thread group fields, timers are no longer "rebalanced" and
the process_timer_rebalance() function and related code has gone away. All
summing loops are gone and all code that used them now uses the
thread_group_cputime() inline. When process-wide timers are set, the new
task_cputime structure in signal_struct is used to cache the earliest
expiration; this is checked in the fast path.
Performance
The fix appears not to add significant overhead to existing operations. It
generally performs the same as the current code except in two cases, one in
which it performs slightly worse (Case 5 below) and one in which it performs
very significantly better (Case 2 below). Overall it's a wash except in those
two cases.
I've since done somewhat more involved testing on a dual-core Opteron system.
Case 1: With no itimer running, for a test with 100,000 threads, the fixed
kernel took 1428.5 seconds, 513 seconds more than the unfixed system,
all of which was spent in the system. There were twice as many
voluntary context switches with the fix as without it.
Case 2: With an itimer running at .01 second ticks and 4000 threads (the most
an unmodified kernel can handle), the fixed kernel ran the test in
eight percent of the time (5.8 seconds as opposed to 70 seconds) and
had better tick accuracy (.012 seconds per tick as opposed to .023
seconds per tick).
Case 3: A 4000-thread test with an initial timer tick of .01 second and an
interval of 10,000 seconds (i.e. a timer that ticks only once) had
very nearly the same performance in both cases: 6.3 seconds elapsed
for the fixed kernel versus 5.5 seconds for the unfixed kernel.
With fewer threads (eight in these tests), the Case 1 test ran in essentially
the same time on both the modified and unmodified kernels (5.2 seconds versus
5.8 seconds). The Case 2 test ran in about the same time as well, 5.9 seconds
versus 5.4 seconds but again with much better tick accuracy, .013 seconds per
tick versus .025 seconds per tick for the unmodified kernel.
Since the fix affected the rlimit code, I also tested soft and hard CPU limits.
Case 4: With a hard CPU limit of 20 seconds and eight threads (and an itimer
running), the modified kernel was very slightly favored in that while
it killed the process in 19.997 seconds of CPU time (5.002 seconds of
wall time), only .003 seconds of that was system time, the rest was
user time. The unmodified kernel killed the process in 20.001 seconds
of CPU (5.014 seconds of wall time) of which .016 seconds was system
time. Really, though, the results were too close to call. The results
were essentially the same with no itimer running.
Case 5: With a soft limit of 20 seconds and a hard limit of 2000 seconds
(where the hard limit would never be reached) and an itimer running,
the modified kernel exhibited worse tick accuracy than the unmodified
kernel: .050 seconds/tick versus .028 seconds/tick. Otherwise,
performance was almost indistinguishable. With no itimer running this
test exhibited virtually identical behavior and times in both cases.
In times past I did some limited performance testing. those results are below.
On a four-cpu Opteron system without this fix, a sixteen-thread test executed
in 3569.991 seconds, of which user was 3568.435s and system was 1.556s. On
the same system with the fix, user and elapsed time were about the same, but
system time dropped to 0.007 seconds. Performance with eight, four and one
thread were comparable. Interestingly, the timer ticks with the fix seemed
more accurate: The sixteen-thread test with the fix received 149543 ticks
for 0.024 seconds per tick, while the same test without the fix received 58720
for 0.061 seconds per tick. Both cases were configured for an interval of
0.01 seconds. Again, the other tests were comparable. Each thread in this
test computed the primes up to 25,000,000.
I also did a test with a large number of threads, 100,000 threads, which is
impossible without the fix. In this case each thread computed the primes only
up to 10,000 (to make the runtime manageable). System time dominated, at
1546.968 seconds out of a total 2176.906 seconds (giving a user time of
629.938s). It received 147651 ticks for 0.015 seconds per tick, still quite
accurate. There is obviously no comparable test without the fix.
Signed-off-by: Frank Mayhar <fmayhar@google.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-12 20:54:39 +04:00
}
2007-10-19 10:41:10 +04:00
static int copy_signal ( unsigned long clone_flags , struct task_struct * tsk )
2005-04-17 02:20:36 +04:00
{
struct signal_struct * sig ;
clone(): fix race between copy_process() and de_thread()
Spotted by Hiroshi Shimamoto who also provided the test-case below.
copy_process() uses signal->count as a reference counter, but it is not.
This test case
#include <sys/types.h>
#include <sys/wait.h>
#include <unistd.h>
#include <stdio.h>
#include <errno.h>
#include <pthread.h>
void *null_thread(void *p)
{
for (;;)
sleep(1);
return NULL;
}
void *exec_thread(void *p)
{
execl("/bin/true", "/bin/true", NULL);
return null_thread(p);
}
int main(int argc, char **argv)
{
for (;;) {
pid_t pid;
int ret, status;
pid = fork();
if (pid < 0)
break;
if (!pid) {
pthread_t tid;
pthread_create(&tid, NULL, exec_thread, NULL);
for (;;)
pthread_create(&tid, NULL, null_thread, NULL);
}
do {
ret = waitpid(pid, &status, 0);
} while (ret == -1 && errno == EINTR);
}
return 0;
}
quickly creates an unkillable task.
If copy_process(CLONE_THREAD) races with de_thread()
copy_signal()->atomic(signal->count) breaks the signal->notify_count
logic, and the execing thread can hang forever in kernel space.
Change copy_process() to increment count/live only when we know for sure
we can't fail. In this case the forked thread will take care of its
reference to signal correctly.
If copy_process() fails, check CLONE_THREAD flag. If it it set - do
nothing, the counters were not changed and current belongs to the same
thread group. If it is not set, ->signal must be released in any case
(and ->count must be == 1), the forked child is the only thread in the
thread group.
We need more cleanups here, in particular signal->count should not be used
by de_thread/__exit_signal at all. This patch only fixes the bug.
Reported-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Tested-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-08-27 01:29:24 +04:00
if ( clone_flags & CLONE_THREAD )
2008-11-24 19:06:57 +03:00
return 0 ;
2010-03-11 02:23:01 +03:00
sig = kmem_cache_zalloc ( signal_cachep , GFP_KERNEL ) ;
2005-04-17 02:20:36 +04:00
tsk - > signal = sig ;
if ( ! sig )
return - ENOMEM ;
2010-05-27 01:43:24 +04:00
sig - > nr_threads = 1 ;
2005-04-17 02:20:36 +04:00
atomic_set ( & sig - > live , 1 ) ;
2019-01-18 15:27:27 +03:00
refcount_set ( & sig - > sigcnt , 1 ) ;
introduce for_each_thread() to replace the buggy while_each_thread()
while_each_thread() and next_thread() should die, almost every lockless
usage is wrong.
1. Unless g == current, the lockless while_each_thread() is not safe.
while_each_thread(g, t) can loop forever if g exits, next_thread()
can't reach the unhashed thread in this case. Note that this can
happen even if g is the group leader, it can exec.
2. Even if while_each_thread() itself was correct, people often use
it wrongly.
It was never safe to just take rcu_read_lock() and loop unless
you verify that pid_alive(g) == T, even the first next_thread()
can point to the already freed/reused memory.
This patch adds signal_struct->thread_head and task->thread_node to
create the normal rcu-safe list with the stable head. The new
for_each_thread(g, t) helper is always safe under rcu_read_lock() as
long as this task_struct can't go away.
Note: of course it is ugly to have both task_struct->thread_node and the
old task_struct->thread_group, we will kill it later, after we change
the users of while_each_thread() to use for_each_thread().
Perhaps we can kill it even before we convert all users, we can
reimplement next_thread(t) using the new thread_head/thread_node. But
we can't do this right now because this will lead to subtle behavioural
changes. For example, do/while_each_thread() always sees at least one
task, while for_each_thread() can do nothing if the whole thread group
has died. Or thread_group_empty(), currently its semantics is not clear
unless thread_group_leader(p) and we need to audit the callers before we
can change it.
So this patch adds the new interface which has to coexist with the old
one for some time, hopefully the next changes will be more or less
straightforward and the old one will go away soon.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Sergey Dyasly <dserrg@gmail.com>
Tested-by: Sergey Dyasly <dserrg@gmail.com>
Reviewed-by: Sameer Nanda <snanda@chromium.org>
Acked-by: David Rientjes <rientjes@google.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mandeep Singh Baines <msb@chromium.org>
Cc: "Ma, Xindong" <xindong.ma@intel.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: "Tu, Xiaobing" <xiaobing.tu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-01-22 03:49:56 +04:00
/* list_add(thread_node, thread_head) without INIT_LIST_HEAD() */
sig - > thread_head = ( struct list_head ) LIST_HEAD_INIT ( tsk - > thread_node ) ;
tsk - > thread_node = ( struct list_head ) LIST_HEAD_INIT ( sig - > thread_head ) ;
2005-04-17 02:20:36 +04:00
init_waitqueue_head ( & sig - > wait_chldexit ) ;
2008-04-30 11:52:52 +04:00
sig - > curr_target = tsk ;
2005-04-17 02:20:36 +04:00
init_sigpending ( & sig - > shared_pending ) ;
2018-07-23 23:20:37 +03:00
INIT_HLIST_HEAD ( & sig - > multiprocess ) ;
2014-08-16 21:40:10 +04:00
seqlock_init ( & sig - > stats_lock ) ;
2015-06-30 12:30:54 +03:00
prev_cputime_init ( & sig - > prev_cputime ) ;
2005-04-17 02:20:36 +04:00
posix-timers: Make them configurable
Some embedded systems have no use for them. This removes about
25KB from the kernel binary size when configured out.
Corresponding syscalls are routed to a stub logging the attempt to
use those syscalls which should be enough of a clue if they were
disabled without proper consideration. They are: timer_create,
timer_gettime: timer_getoverrun, timer_settime, timer_delete,
clock_adjtime, setitimer, getitimer, alarm.
The clock_settime, clock_gettime, clock_getres and clock_nanosleep
syscalls are replaced by simple wrappers compatible with CLOCK_REALTIME,
CLOCK_MONOTONIC and CLOCK_BOOTTIME only which should cover the vast
majority of use cases with very little code.
Signed-off-by: Nicolas Pitre <nico@linaro.org>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <john.stultz@linaro.org>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Cc: Paul Bolle <pebolle@tiscali.nl>
Cc: linux-kbuild@vger.kernel.org
Cc: netdev@vger.kernel.org
Cc: Michal Marek <mmarek@suse.com>
Cc: Edward Cree <ecree@solarflare.com>
Link: http://lkml.kernel.org/r/1478841010-28605-7-git-send-email-nicolas.pitre@linaro.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2016-11-11 08:10:10 +03:00
# ifdef CONFIG_POSIX_TIMERS
2017-01-21 08:09:08 +03:00
INIT_LIST_HEAD ( & sig - > posix_timers ) ;
2007-02-16 12:27:49 +03:00
hrtimer_init ( & sig - > real_timer , CLOCK_MONOTONIC , HRTIMER_MODE_REL ) ;
2005-04-17 02:20:36 +04:00
sig - > real_timer . function = it_real_fn ;
posix-timers: Make them configurable
Some embedded systems have no use for them. This removes about
25KB from the kernel binary size when configured out.
Corresponding syscalls are routed to a stub logging the attempt to
use those syscalls which should be enough of a clue if they were
disabled without proper consideration. They are: timer_create,
timer_gettime: timer_getoverrun, timer_settime, timer_delete,
clock_adjtime, setitimer, getitimer, alarm.
The clock_settime, clock_gettime, clock_getres and clock_nanosleep
syscalls are replaced by simple wrappers compatible with CLOCK_REALTIME,
CLOCK_MONOTONIC and CLOCK_BOOTTIME only which should cover the vast
majority of use cases with very little code.
Signed-off-by: Nicolas Pitre <nico@linaro.org>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <john.stultz@linaro.org>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Cc: Paul Bolle <pebolle@tiscali.nl>
Cc: linux-kbuild@vger.kernel.org
Cc: netdev@vger.kernel.org
Cc: Michal Marek <mmarek@suse.com>
Cc: Edward Cree <ecree@solarflare.com>
Link: http://lkml.kernel.org/r/1478841010-28605-7-git-send-email-nicolas.pitre@linaro.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2016-11-11 08:10:10 +03:00
# endif
2005-04-17 02:20:36 +04:00
task_lock ( current - > group_leader ) ;
memcpy ( sig - > rlim , current - > signal - > rlim , sizeof sig - > rlim ) ;
task_unlock ( current - > group_leader ) ;
2009-03-27 03:06:07 +03:00
posix_cpu_timers_init_group ( sig ) ;
Audit: add TTY input auditing
Add TTY input auditing, used to audit system administrator's actions. This is
required by various security standards such as DCID 6/3 and PCI to provide
non-repudiation of administrator's actions and to allow a review of past
actions if the administrator seems to overstep their duties or if the system
becomes misconfigured for unknown reasons. These requirements do not make it
necessary to audit TTY output as well.
Compared to an user-space keylogger, this approach records TTY input using the
audit subsystem, correlated with other audit events, and it is completely
transparent to the user-space application (e.g. the console ioctls still
work).
TTY input auditing works on a higher level than auditing all system calls
within the session, which would produce an overwhelming amount of mostly
useless audit events.
Add an "audit_tty" attribute, inherited across fork (). Data read from TTYs
by process with the attribute is sent to the audit subsystem by the kernel.
The audit netlink interface is extended to allow modifying the audit_tty
attribute, and to allow sending explanatory audit events from user-space (for
example, a shell might send an event containing the final command, after the
interactive command-line editing and history expansion is performed, which
might be difficult to decipher from the TTY input alone).
Because the "audit_tty" attribute is inherited across fork (), it would be set
e.g. for sshd restarted within an audited session. To prevent this, the
audit_tty attribute is cleared when a process with no open TTY file
descriptors (e.g. after daemon startup) opens a TTY.
See https://www.redhat.com/archives/linux-audit/2007-June/msg00000.html for a
more detailed rationale document for an older version of this patch.
[akpm@linux-foundation.org: build fix]
Signed-off-by: Miloslav Trmac <mitr@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Paul Fulghum <paulkf@microgate.com>
Cc: Casey Schaufler <casey@schaufler-ca.com>
Cc: Steve Grubb <sgrubb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-16 10:40:56 +04:00
tty_audit_fork ( sig ) ;
sched: Add 'autogroup' scheduling feature: automated per session task groups
A recurring complaint from CFS users is that parallel kbuild has
a negative impact on desktop interactivity. This patch
implements an idea from Linus, to automatically create task
groups. Currently, only per session autogroups are implemented,
but the patch leaves the way open for enhancement.
Implementation: each task's signal struct contains an inherited
pointer to a refcounted autogroup struct containing a task group
pointer, the default for all tasks pointing to the
init_task_group. When a task calls setsid(), a new task group
is created, the process is moved into the new task group, and a
reference to the preveious task group is dropped. Child
processes inherit this task group thereafter, and increase it's
refcount. When the last thread of a process exits, the
process's reference is dropped, such that when the last process
referencing an autogroup exits, the autogroup is destroyed.
At runqueue selection time, IFF a task has no cgroup assignment,
its current autogroup is used.
Autogroup bandwidth is controllable via setting it's nice level
through the proc filesystem:
cat /proc/<pid>/autogroup
Displays the task's group and the group's nice level.
echo <nice level> > /proc/<pid>/autogroup
Sets the task group's shares to the weight of nice <level> task.
Setting nice level is rate limited for !admin users due to the
abuse risk of task group locking.
The feature is enabled from boot by default if
CONFIG_SCHED_AUTOGROUP=y is selected, but can be disabled via
the boot option noautogroup, and can also be turned on/off on
the fly via:
echo [01] > /proc/sys/kernel/sched_autogroup_enabled
... which will automatically move tasks to/from the root task group.
Signed-off-by: Mike Galbraith <efault@gmx.de>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Markus Trippelsdorf <markus@trippelsdorf.de>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Paul Turner <pjt@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
[ Removed the task_group_path() debug code, and fixed !EVENTFD build failure. ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
LKML-Reference: <1290281700.28711.9.camel@maggy.simson.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-11-30 16:18:03 +03:00
sched_autogroup_fork ( sig ) ;
Audit: add TTY input auditing
Add TTY input auditing, used to audit system administrator's actions. This is
required by various security standards such as DCID 6/3 and PCI to provide
non-repudiation of administrator's actions and to allow a review of past
actions if the administrator seems to overstep their duties or if the system
becomes misconfigured for unknown reasons. These requirements do not make it
necessary to audit TTY output as well.
Compared to an user-space keylogger, this approach records TTY input using the
audit subsystem, correlated with other audit events, and it is completely
transparent to the user-space application (e.g. the console ioctls still
work).
TTY input auditing works on a higher level than auditing all system calls
within the session, which would produce an overwhelming amount of mostly
useless audit events.
Add an "audit_tty" attribute, inherited across fork (). Data read from TTYs
by process with the attribute is sent to the audit subsystem by the kernel.
The audit netlink interface is extended to allow modifying the audit_tty
attribute, and to allow sending explanatory audit events from user-space (for
example, a shell might send an event containing the final command, after the
interactive command-line editing and history expansion is performed, which
might be difficult to decipher from the TTY input alone).
Because the "audit_tty" attribute is inherited across fork (), it would be set
e.g. for sshd restarted within an audited session. To prevent this, the
audit_tty attribute is cleared when a process with no open TTY file
descriptors (e.g. after daemon startup) opens a TTY.
See https://www.redhat.com/archives/linux-audit/2007-June/msg00000.html for a
more detailed rationale document for an older version of this patch.
[akpm@linux-foundation.org: build fix]
Signed-off-by: Miloslav Trmac <mitr@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Paul Fulghum <paulkf@microgate.com>
Cc: Casey Schaufler <casey@schaufler-ca.com>
Cc: Steve Grubb <sgrubb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-16 10:40:56 +04:00
oom: badness heuristic rewrite
This a complete rewrite of the oom killer's badness() heuristic which is
used to determine which task to kill in oom conditions. The goal is to
make it as simple and predictable as possible so the results are better
understood and we end up killing the task which will lead to the most
memory freeing while still respecting the fine-tuning from userspace.
Instead of basing the heuristic on mm->total_vm for each task, the task's
rss and swap space is used instead. This is a better indication of the
amount of memory that will be freeable if the oom killed task is chosen
and subsequently exits. This helps specifically in cases where KDE or
GNOME is chosen for oom kill on desktop systems instead of a memory
hogging task.
The baseline for the heuristic is a proportion of memory that each task is
currently using in memory plus swap compared to the amount of "allowable"
memory. "Allowable," in this sense, means the system-wide resources for
unconstrained oom conditions, the set of mempolicy nodes, the mems
attached to current's cpuset, or a memory controller's limit. The
proportion is given on a scale of 0 (never kill) to 1000 (always kill),
roughly meaning that if a task has a badness() score of 500 that the task
consumes approximately 50% of allowable memory resident in RAM or in swap
space.
The proportion is always relative to the amount of "allowable" memory and
not the total amount of RAM systemwide so that mempolicies and cpusets may
operate in isolation; they shall not need to know the true size of the
machine on which they are running if they are bound to a specific set of
nodes or mems, respectively.
Root tasks are given 3% extra memory just like __vm_enough_memory()
provides in LSMs. In the event of two tasks consuming similar amounts of
memory, it is generally better to save root's task.
Because of the change in the badness() heuristic's baseline, it is also
necessary to introduce a new user interface to tune it. It's not possible
to redefine the meaning of /proc/pid/oom_adj with a new scale since the
ABI cannot be changed for backward compatability. Instead, a new tunable,
/proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may
be used to polarize the heuristic such that certain tasks are never
considered for oom kill while others may always be considered. The value
is added directly into the badness() score so a value of -500, for
example, means to discount 50% of its memory consumption in comparison to
other tasks either on the system, bound to the mempolicy, in the cpuset,
or sharing the same memory controller.
/proc/pid/oom_adj is changed so that its meaning is rescaled into the
units used by /proc/pid/oom_score_adj, and vice versa. Changing one of
these per-task tunables will rescale the value of the other to an
equivalent meaning. Although /proc/pid/oom_adj was originally defined as
a bitshift on the badness score, it now shares the same linear growth as
/proc/pid/oom_score_adj but with different granularity. This is required
so the ABI is not broken with userspace applications and allows oom_adj to
be deprecated for future removal.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-08-10 04:19:46 +04:00
sig - > oom_score_adj = current - > signal - > oom_score_adj ;
2011-01-14 02:46:05 +03:00
sig - > oom_score_adj_min = current - > signal - > oom_score_adj_min ;
oom: move oom_adj value from task_struct to signal_struct
Currently, OOM logic callflow is here.
__out_of_memory()
select_bad_process() for each task
badness() calculate badness of one task
oom_kill_process() search child
oom_kill_task() kill target task and mm shared tasks with it
example, process-A have two thread, thread-A and thread-B and it have very
fat memory and each thread have following oom_adj and oom_score.
thread-A: oom_adj = OOM_DISABLE, oom_score = 0
thread-B: oom_adj = 0, oom_score = very-high
Then, select_bad_process() select thread-B, but oom_kill_task() refuse
kill the task because thread-A have OOM_DISABLE. Thus __out_of_memory()
call select_bad_process() again. but select_bad_process() select the same
task. It mean kernel fall in livelock.
The fact is, select_bad_process() must select killable task. otherwise
OOM logic go into livelock.
And root cause is, oom_adj shouldn't be per-thread value. it should be
per-process value because OOM-killer kill a process, not thread. Thus
This patch moves oomkilladj (now more appropriately named oom_adj) from
struct task_struct to struct signal_struct. it naturally prevent
select_bad_process() choose wrong task.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Paul Menage <menage@google.com>
Cc: David Rientjes <rientjes@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-22 04:03:13 +04:00
2010-10-28 02:34:08 +04:00
mutex_init ( & sig - > cred_guard_mutex ) ;
2020-12-03 23:12:00 +03:00
init_rwsem ( & sig - > exec_update_lock ) ;
2010-10-28 02:34:08 +04:00
2005-04-17 02:20:36 +04:00
return 0 ;
}
2014-06-28 02:18:48 +04:00
static void copy_seccomp ( struct task_struct * p )
{
# ifdef CONFIG_SECCOMP
/*
* Must be called with sighand - > lock held , which is common to
* all threads in the group . Holding cred_guard_mutex is not
* needed because this new task is not yet running and cannot
* be racing exec .
*/
2014-08-11 07:50:30 +04:00
assert_spin_locked ( & current - > sighand - > siglock ) ;
2014-06-28 02:18:48 +04:00
/* Ref-count the new filter user, and assign it. */
get_seccomp_filter ( current ) ;
p - > seccomp = current - > seccomp ;
/*
* Explicitly enable no_new_privs here in case it got set
* between the task_struct being duplicated and holding the
* sighand lock . The seccomp state and nnp must be in sync .
*/
if ( task_no_new_privs ( current ) )
task_set_no_new_privs ( p ) ;
/*
* If the parent gained a seccomp mode after copying thread
* flags and between before we held the sighand lock , we have
* to manually enable the seccomp thread flag here .
*/
if ( p - > seccomp . mode ! = SECCOMP_MODE_DISABLED )
2020-11-16 20:42:00 +03:00
set_task_syscall_work ( p , SECCOMP ) ;
2014-06-28 02:18:48 +04:00
# endif
}
2009-01-14 16:14:10 +03:00
SYSCALL_DEFINE1 ( set_tid_address , int __user * , tidptr )
2005-04-17 02:20:36 +04:00
{
current - > clear_child_tid = tidptr ;
2007-10-19 10:40:14 +04:00
return task_pid_vnr ( current ) ;
2005-04-17 02:20:36 +04:00
}
2007-10-19 10:41:10 +04:00
static void rt_mutex_init_task ( struct task_struct * p )
2006-06-27 13:54:53 +04:00
{
2009-11-17 16:54:03 +03:00
raw_spin_lock_init ( & p - > pi_lock ) ;
2007-03-17 00:38:34 +03:00
# ifdef CONFIG_RT_MUTEXES
2017-09-09 02:15:01 +03:00
p - > pi_waiters = RB_ROOT_CACHED ;
2017-03-23 17:56:08 +03:00
p - > pi_top_task = NULL ;
2006-06-27 13:54:53 +04:00
p - > pi_blocked_on = NULL ;
# endif
}
2017-09-26 21:06:43 +03:00
static inline void init_task_pid_links ( struct task_struct * task )
{
enum pid_type type ;
2020-12-26 14:40:21 +03:00
for ( type = PIDTYPE_PID ; type < PIDTYPE_MAX ; + + type )
2017-09-26 21:06:43 +03:00
INIT_HLIST_NODE ( & task - > pid_links [ type ] ) ;
}
2013-07-04 02:08:31 +04:00
static inline void
init_task_pid ( struct task_struct * task , enum pid_type type , struct pid * pid )
{
2017-09-26 21:06:43 +03:00
if ( type = = PIDTYPE_PID )
task - > thread_pid = pid ;
else
task - > signal - > pids [ type ] = pid ;
2013-07-04 02:08:31 +04:00
}
2017-02-03 23:37:49 +03:00
static inline void rcu_copy_process ( struct task_struct * p )
{
# ifdef CONFIG_PREEMPT_RCU
p - > rcu_read_lock_nesting = 0 ;
p - > rcu_read_unlock_special . s = 0 ;
p - > rcu_blocked_node = NULL ;
INIT_LIST_HEAD ( & p - > rcu_node_entry ) ;
# endif /* #ifdef CONFIG_PREEMPT_RCU */
# ifdef CONFIG_TASKS_RCU
p - > rcu_tasks_holdout = false ;
INIT_LIST_HEAD ( & p - > rcu_tasks_holdout_list ) ;
p - > rcu_tasks_idle_cpu = - 1 ;
# endif /* #ifdef CONFIG_TASKS_RCU */
rcu-tasks: Add an RCU Tasks Trace to simplify protection of tracing hooks
Because RCU does not watch exception early-entry/late-exit, idle-loop,
or CPU-hotplug execution, protection of tracing and BPF operations is
needlessly complicated. This commit therefore adds a variant of
Tasks RCU that:
o Has explicit read-side markers to allow finite grace periods in
the face of in-kernel loops for PREEMPT=n builds. These markers
are rcu_read_lock_trace() and rcu_read_unlock_trace().
o Protects code in the idle loop, exception entry/exit, and
CPU-hotplug code paths. In this respect, RCU-tasks trace is
similar to SRCU, but with lighter-weight readers.
o Avoids expensive read-side instruction, having overhead similar
to that of Preemptible RCU.
There are of course downsides:
o The grace-period code can send IPIs to CPUs, even when those
CPUs are in the idle loop or in nohz_full userspace. This is
mitigated by later commits.
o It is necessary to scan the full tasklist, much as for Tasks RCU.
o There is a single callback queue guarded by a single lock,
again, much as for Tasks RCU. However, those early use cases
that request multiple grace periods in quick succession are
expected to do so from a single task, which makes the single
lock almost irrelevant. If needed, multiple callback queues
can be provided using any number of schemes.
Perhaps most important, this variant of RCU does not affect the vanilla
flavors, rcu_preempt and rcu_sched. The fact that RCU Tasks Trace
readers can operate from idle, offline, and exception entry/exit in no
way enables rcu_preempt and rcu_sched readers to do so.
The memory ordering was outlined here:
https://lore.kernel.org/lkml/20200319034030.GX3199@paulmck-ThinkPad-P72/
This effort benefited greatly from off-list discussions of BPF
requirements with Alexei Starovoitov and Andrii Nakryiko. At least
some of the on-list discussions are captured in the Link: tags below.
In addition, KCSAN was quite helpful in finding some early bugs.
Link: https://lore.kernel.org/lkml/20200219150744.428764577@infradead.org/
Link: https://lore.kernel.org/lkml/87mu8p797b.fsf@nanos.tec.linutronix.de/
Link: https://lore.kernel.org/lkml/20200225221305.605144982@linutronix.de/
Cc: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Andrii Nakryiko <andriin@fb.com>
[ paulmck: Apply feedback from Steve Rostedt and Joel Fernandes. ]
[ paulmck: Decrement trc_n_readers_need_end upon IPI failure. ]
[ paulmck: Fix locking issue reported by rcutorture. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-03-10 05:56:53 +03:00
# ifdef CONFIG_TASKS_TRACE_RCU
p - > trc_reader_nesting = 0 ;
2020-03-18 02:02:06 +03:00
p - > trc_reader_special . s = 0 ;
rcu-tasks: Add an RCU Tasks Trace to simplify protection of tracing hooks
Because RCU does not watch exception early-entry/late-exit, idle-loop,
or CPU-hotplug execution, protection of tracing and BPF operations is
needlessly complicated. This commit therefore adds a variant of
Tasks RCU that:
o Has explicit read-side markers to allow finite grace periods in
the face of in-kernel loops for PREEMPT=n builds. These markers
are rcu_read_lock_trace() and rcu_read_unlock_trace().
o Protects code in the idle loop, exception entry/exit, and
CPU-hotplug code paths. In this respect, RCU-tasks trace is
similar to SRCU, but with lighter-weight readers.
o Avoids expensive read-side instruction, having overhead similar
to that of Preemptible RCU.
There are of course downsides:
o The grace-period code can send IPIs to CPUs, even when those
CPUs are in the idle loop or in nohz_full userspace. This is
mitigated by later commits.
o It is necessary to scan the full tasklist, much as for Tasks RCU.
o There is a single callback queue guarded by a single lock,
again, much as for Tasks RCU. However, those early use cases
that request multiple grace periods in quick succession are
expected to do so from a single task, which makes the single
lock almost irrelevant. If needed, multiple callback queues
can be provided using any number of schemes.
Perhaps most important, this variant of RCU does not affect the vanilla
flavors, rcu_preempt and rcu_sched. The fact that RCU Tasks Trace
readers can operate from idle, offline, and exception entry/exit in no
way enables rcu_preempt and rcu_sched readers to do so.
The memory ordering was outlined here:
https://lore.kernel.org/lkml/20200319034030.GX3199@paulmck-ThinkPad-P72/
This effort benefited greatly from off-list discussions of BPF
requirements with Alexei Starovoitov and Andrii Nakryiko. At least
some of the on-list discussions are captured in the Link: tags below.
In addition, KCSAN was quite helpful in finding some early bugs.
Link: https://lore.kernel.org/lkml/20200219150744.428764577@infradead.org/
Link: https://lore.kernel.org/lkml/87mu8p797b.fsf@nanos.tec.linutronix.de/
Link: https://lore.kernel.org/lkml/20200225221305.605144982@linutronix.de/
Cc: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Andrii Nakryiko <andriin@fb.com>
[ paulmck: Apply feedback from Steve Rostedt and Joel Fernandes. ]
[ paulmck: Decrement trc_n_readers_need_end upon IPI failure. ]
[ paulmck: Fix locking issue reported by rcutorture. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-03-10 05:56:53 +03:00
INIT_LIST_HEAD ( & p - > trc_holdout_list ) ;
# endif /* #ifdef CONFIG_TASKS_TRACE_RCU */
2017-02-03 23:37:49 +03:00
}
2019-07-28 01:22:29 +03:00
struct pid * pidfd_pid ( const struct file * file )
{
if ( file - > f_op = = & pidfd_fops )
return file - > private_data ;
return ERR_PTR ( - EBADF ) ;
}
2019-03-27 15:04:15 +03:00
static int pidfd_release ( struct inode * inode , struct file * file )
{
struct pid * pid = file - > private_data ;
file - > private_data = NULL ;
put_pid ( pid ) ;
return 0 ;
}
# ifdef CONFIG_PROC_FS
2019-10-14 19:20:32 +03:00
/**
* pidfd_show_fdinfo - print information about a pidfd
* @ m : proc fdinfo file
* @ f : file referencing a pidfd
*
* Pid :
* This function will print the pid that a given pidfd refers to in the
* pid namespace of the procfs instance .
* If the pid namespace of the process is not a descendant of the pid
* namespace of the procfs instance 0 will be shown as its pid . This is
* similar to calling getppid ( ) on a process whose parent is outside of
* its pid namespace .
*
* NSpid :
* If pid namespaces are supported then this function will also print
* the pid of a given pidfd refers to for all descendant pid namespaces
* starting from the current pid namespace of the instance , i . e . the
* Pid field and the first entry in the NSpid field will be identical .
* If the pid namespace of the process is not a descendant of the pid
* namespace of the procfs instance 0 will be shown as its first NSpid
* entry and no others will be shown .
* Note that this differs from the Pid and NSpid fields in
* / proc / < pid > / status where Pid and NSpid are always shown relative to
* the pid namespace of the procfs instance . The difference becomes
* obvious when sending around a pidfd between pid namespaces from a
2021-05-07 04:04:28 +03:00
* different branch of the tree , i . e . where no ancestral relation is
2019-10-14 19:20:32 +03:00
* present between the pid namespaces :
* - create two new pid namespaces ns1 and ns2 in the initial pid
* namespace ( also take care to create new mount namespaces in the
* new pid namespace and mount procfs )
* - create a process with a pidfd in ns1
* - send pidfd from ns1 to ns2
* - read / proc / self / fdinfo / < pidfd > and observe that both Pid and NSpid
* have exactly one entry , which is 0
*/
2019-03-27 15:04:15 +03:00
static void pidfd_show_fdinfo ( struct seq_file * m , struct file * f )
{
struct pid * pid = f - > private_data ;
2019-10-17 13:18:28 +03:00
struct pid_namespace * ns ;
pid_t nr = - 1 ;
2019-10-14 19:20:32 +03:00
2019-10-17 13:18:28 +03:00
if ( likely ( pid_has_task ( pid , PIDTYPE_PID ) ) ) {
2020-05-18 21:07:38 +03:00
ns = proc_pid_ns ( file_inode ( m - > file ) - > i_sb ) ;
2019-10-17 13:18:28 +03:00
nr = pid_nr_ns ( pid , ns ) ;
}
seq_put_decimal_ll ( m , " Pid: \t " , nr ) ;
2019-03-27 15:04:15 +03:00
2019-10-14 19:20:32 +03:00
# ifdef CONFIG_PID_NS
2019-10-17 13:18:28 +03:00
seq_put_decimal_ll ( m , " \n NSpid: \t " , nr ) ;
if ( nr > 0 ) {
2019-10-14 19:20:32 +03:00
int i ;
2019-03-27 15:04:15 +03:00
2019-10-14 19:20:32 +03:00
/* If nr is non-zero it means that 'pid' is valid and that
* ns , i . e . the pid namespace associated with the procfs
* instance , is in the pid namespace hierarchy of pid .
* Start at one below the already printed level .
*/
for ( i = ns - > level + 1 ; i < = pid - > level ; i + + )
2019-10-17 13:18:28 +03:00
seq_put_decimal_ll ( m , " \t " , pid - > numbers [ i ] . nr ) ;
2019-10-14 19:20:32 +03:00
}
# endif
2019-03-27 15:04:15 +03:00
seq_putc ( m , ' \n ' ) ;
}
# endif
2019-04-30 19:21:53 +03:00
/*
* Poll support for process exit notification .
*/
2019-11-20 03:33:20 +03:00
static __poll_t pidfd_poll ( struct file * file , struct poll_table_struct * pts )
2019-04-30 19:21:53 +03:00
{
struct pid * pid = file - > private_data ;
2019-11-20 03:33:20 +03:00
__poll_t poll_flags = 0 ;
2019-04-30 19:21:53 +03:00
poll_wait ( file , & pid - > wait_pidfd , pts ) ;
/*
* Inform pollers only when the whole thread group exits .
* If the thread group leader exits before all other threads in the
* group , then poll ( 2 ) should block , similar to the wait ( 2 ) family .
*/
2020-07-01 15:30:06 +03:00
if ( thread_group_exited ( pid ) )
2019-11-20 03:33:20 +03:00
poll_flags = EPOLLIN | EPOLLRDNORM ;
2019-04-30 19:21:53 +03:00
return poll_flags ;
}
2019-03-27 15:04:15 +03:00
const struct file_operations pidfd_fops = {
. release = pidfd_release ,
2019-04-30 19:21:53 +03:00
. poll = pidfd_poll ,
2019-03-27 15:04:15 +03:00
# ifdef CONFIG_PROC_FS
. show_fdinfo = pidfd_show_fdinfo ,
# endif
} ;
2019-05-15 01:40:46 +03:00
static void __delayed_free_task ( struct rcu_head * rhp )
{
struct task_struct * tsk = container_of ( rhp , struct task_struct , rcu ) ;
free_task ( tsk ) ;
}
static __always_inline void delayed_free_task ( struct task_struct * tsk )
{
if ( IS_ENABLED ( CONFIG_MEMCG ) )
call_rcu ( & tsk - > rcu , __delayed_free_task ) ;
else
free_task ( tsk ) ;
}
mm, oom_adj: don't loop through tasks in __set_oom_adj when not necessary
Currently __set_oom_adj loops through all processes in the system to keep
oom_score_adj and oom_score_adj_min in sync between processes sharing
their mm. This is done for any task with more that one mm_users, which
includes processes with multiple threads (sharing mm and signals).
However for such processes the loop is unnecessary because their signal
structure is shared as well.
Android updates oom_score_adj whenever a tasks changes its role
(background/foreground/...) or binds to/unbinds from a service, making it
more/less important. Such operation can happen frequently. We noticed
that updates to oom_score_adj became more expensive and after further
investigation found out that the patch mentioned in "Fixes" introduced a
regression. Using Pixel 4 with a typical Android workload, write time to
oom_score_adj increased from ~3.57us to ~362us. Moreover this regression
linearly depends on the number of multi-threaded processes running on the
system.
Mark the mm with a new MMF_MULTIPROCESS flag bit when task is created with
(CLONE_VM && !CLONE_THREAD && !CLONE_VFORK). Change __set_oom_adj to use
MMF_MULTIPROCESS instead of mm_users to decide whether oom_score_adj
update should be synchronized between multiple processes. To prevent
races between clone() and __set_oom_adj(), when oom_score_adj of the
process being cloned might be modified from userspace, we use
oom_adj_mutex. Its scope is changed to global.
The combination of (CLONE_VM && !CLONE_THREAD) is rarely used except for
the case of vfork(). To prevent performance regressions of vfork(), we
skip taking oom_adj_mutex and setting MMF_MULTIPROCESS when CLONE_VFORK is
specified. Clearing the MMF_MULTIPROCESS flag (when the last process
sharing the mm exits) is left out of this patch to keep it simple and
because it is believed that this threading model is rare. Should there
ever be a need for optimizing that case as well, it can be done by hooking
into the exit path, likely following the mm_update_next_owner pattern.
With the combination of (CLONE_VM && !CLONE_THREAD && !CLONE_VFORK) being
quite rare, the regression is gone after the change is applied.
[surenb@google.com: v3]
Link: https://lkml.kernel.org/r/20200902012558.2335613-1-surenb@google.com
Fixes: 44a70adec910 ("mm, oom_adj: make sure processes sharing mm have same view of oom_score_adj")
Reported-by: Tim Murray <timmurray@google.com>
Suggested-by: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Christian Brauner <christian.brauner@ubuntu.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Eugene Syromiatnikov <esyr@redhat.com>
Cc: Christian Kellner <christian@kellner.me>
Cc: Adrian Reber <areber@redhat.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Aleksa Sarai <cyphar@cyphar.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Alexey Gladkov <gladkov.alexey@gmail.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Daniel Jordan <daniel.m.jordan@oracle.com>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Bernd Edlinger <bernd.edlinger@hotmail.de>
Cc: John Johansen <john.johansen@canonical.com>
Cc: Yafang Shao <laoar.shao@gmail.com>
Link: https://lkml.kernel.org/r/20200824153036.3201505-1-surenb@google.com
Debugged-by: Minchan Kim <minchan@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-14 02:58:35 +03:00
static void copy_oom_score_adj ( u64 clone_flags , struct task_struct * tsk )
{
/* Skip if kernel thread */
if ( ! tsk - > mm )
return ;
/* Skip if spawning a thread or using vfork */
if ( ( clone_flags & ( CLONE_VM | CLONE_THREAD | CLONE_VFORK ) ) ! = CLONE_VM )
return ;
/* We need to synchronize with __set_oom_adj */
mutex_lock ( & oom_adj_mutex ) ;
set_bit ( MMF_MULTIPROCESS , & tsk - > mm - > flags ) ;
/* Update the values in case they were changed after copy_signal */
tsk - > signal - > oom_score_adj = current - > signal - > oom_score_adj ;
tsk - > signal - > oom_score_adj_min = current - > signal - > oom_score_adj_min ;
mutex_unlock ( & oom_adj_mutex ) ;
}
2005-04-17 02:20:36 +04:00
/*
* This creates a new process as a copy of the old one ,
* but does not actually start it yet .
*
* It copies the registers , and all the appropriate
* parts of the process environment ( as per the clone
* flags ) . The actual kick - off is left to the caller .
*/
2016-06-20 21:42:34 +03:00
static __latent_entropy struct task_struct * copy_process (
2008-07-26 06:45:47 +04:00
struct pid * pid ,
clone: support passing tls argument via C rather than pt_regs magic
clone has some of the quirkiest syscall handling in the kernel, with a
pile of special cases, historical curiosities, and architecture-specific
calling conventions. In particular, clone with CLONE_SETTLS accepts a
parameter "tls" that the C entry point completely ignores and some
assembly entry points overwrite; instead, the low-level arch-specific
code pulls the tls parameter out of the arch-specific register captured
as part of pt_regs on entry to the kernel. That's a massive hack, and
it makes the arch-specific code only work when called via the specific
existing syscall entry points; because of this hack, any new clone-like
system call would have to accept an identical tls argument in exactly
the same arch-specific position, rather than providing a unified system
call entry point across architectures.
The first patch allows architectures to handle the tls argument via
normal C parameter passing, if they opt in by selecting
HAVE_COPY_THREAD_TLS. The second patch makes 32-bit and 64-bit x86 opt
into this.
These two patches came out of the clone4 series, which isn't ready for
this merge window, but these first two cleanup patches were entirely
uncontroversial and have acks. I'd like to go ahead and submit these
two so that other architectures can begin building on top of this and
opting into HAVE_COPY_THREAD_TLS. However, I'm also happy to wait and
send these through the next merge window (along with v3 of clone4) if
anyone would prefer that.
This patch (of 2):
clone with CLONE_SETTLS accepts an argument to set the thread-local
storage area for the new thread. sys_clone declares an int argument
tls_val in the appropriate point in the argument list (based on the
various CLONE_BACKWARDS variants), but doesn't actually use or pass along
that argument. Instead, sys_clone calls do_fork, which calls
copy_process, which calls the arch-specific copy_thread, and copy_thread
pulls the corresponding syscall argument out of the pt_regs captured at
kernel entry (knowing what argument of clone that architecture passes tls
in).
Apart from being awful and inscrutable, that also only works because only
one code path into copy_thread can pass the CLONE_SETTLS flag, and that
code path comes from sys_clone with its architecture-specific
argument-passing order. This prevents introducing a new version of the
clone system call without propagating the same architecture-specific
position of the tls argument.
However, there's no reason to pull the argument out of pt_regs when
sys_clone could just pass it down via C function call arguments.
Introduce a new CONFIG_HAVE_COPY_THREAD_TLS for architectures to opt into,
and a new copy_thread_tls that accepts the tls parameter as an additional
unsigned long (syscall-argument-sized) argument. Change sys_clone's tls
argument to an unsigned long (which does not change the ABI), and pass
that down to copy_thread_tls.
Architectures that don't opt into copy_thread_tls will continue to ignore
the C argument to sys_clone in favor of the pt_regs captured at kernel
entry, and thus will be unable to introduce new versions of the clone
syscall.
Patch co-authored by Josh Triplett and Thiago Macieira.
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Acked-by: Andy Lutomirski <luto@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Thiago Macieira <thiago.macieira@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-06-26 01:01:19 +03:00
int trace ,
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
int node ,
struct kernel_clone_args * args )
2005-04-17 02:20:36 +04:00
{
2019-03-27 15:04:15 +03:00
int pidfd = - 1 , retval ;
2007-10-19 10:41:09 +04:00
struct task_struct * p ;
2018-07-23 23:20:37 +03:00
struct multiprocess_signals delayed ;
2019-06-27 05:22:09 +03:00
struct file * pidfile = NULL ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
u64 clone_flags = args - > flags ;
ns: Introduce Time Namespace
Time Namespace isolates clock values.
The kernel provides access to several clocks CLOCK_REALTIME,
CLOCK_MONOTONIC, CLOCK_BOOTTIME, etc.
CLOCK_REALTIME
System-wide clock that measures real (i.e., wall-clock) time.
CLOCK_MONOTONIC
Clock that cannot be set and represents monotonic time since
some unspecified starting point.
CLOCK_BOOTTIME
Identical to CLOCK_MONOTONIC, except it also includes any time
that the system is suspended.
For many users, the time namespace means the ability to changes date and
time in a container (CLOCK_REALTIME). Providing per namespace notions of
CLOCK_REALTIME would be complex with a massive overhead, but has a dubious
value.
But in the context of checkpoint/restore functionality, monotonic and
boottime clocks become interesting. Both clocks are monotonic with
unspecified starting points. These clocks are widely used to measure time
slices and set timers. After restoring or migrating processes, it has to be
guaranteed that they never go backward. In an ideal case, the behavior of
these clocks should be the same as for a case when a whole system is
suspended. All this means that it is required to set CLOCK_MONOTONIC and
CLOCK_BOOTTIME clocks, which can be achieved by adding per-namespace
offsets for clocks.
A time namespace is similar to a pid namespace in the way how it is
created: unshare(CLONE_NEWTIME) system call creates a new time namespace,
but doesn't set it to the current process. Then all children of the process
will be born in the new time namespace, or a process can use the setns()
system call to join a namespace.
This scheme allows setting clock offsets for a namespace, before any
processes appear in it.
All available clone flags have been used, so CLONE_NEWTIME uses the highest
bit of CSIGNAL. It means that it can be used only with the unshare() and
the clone3() system calls.
[ tglx: Adjusted paragraph about clone3() to reality and massaged the
changelog a bit. ]
Co-developed-by: Dmitry Safonov <dima@arista.com>
Signed-off-by: Andrei Vagin <avagin@gmail.com>
Signed-off-by: Dmitry Safonov <dima@arista.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://criu.org/Time_namespace
Link: https://lists.openvz.org/pipermail/criu/2018-June/041504.html
Link: https://lore.kernel.org/r/20191112012724.250792-4-dima@arista.com
2019-11-12 04:26:52 +03:00
struct nsproxy * nsp = current - > nsproxy ;
2005-04-17 02:20:36 +04:00
2018-02-07 02:39:34 +03:00
/*
* Don ' t allow sharing the root directory with processes in a different
* namespace
*/
2005-04-17 02:20:36 +04:00
if ( ( clone_flags & ( CLONE_NEWNS | CLONE_FS ) ) = = ( CLONE_NEWNS | CLONE_FS ) )
return ERR_PTR ( - EINVAL ) ;
2013-03-13 22:51:49 +04:00
if ( ( clone_flags & ( CLONE_NEWUSER | CLONE_FS ) ) = = ( CLONE_NEWUSER | CLONE_FS ) )
return ERR_PTR ( - EINVAL ) ;
2005-04-17 02:20:36 +04:00
/*
* Thread groups must share signals as well , and detached threads
* can only be started up within the thread group .
*/
if ( ( clone_flags & CLONE_THREAD ) & & ! ( clone_flags & CLONE_SIGHAND ) )
return ERR_PTR ( - EINVAL ) ;
/*
* Shared signal handlers imply shared VM . By way of the above ,
* thread groups also imply shared VM . Blocking this case allows
* for various simplifications in other code .
*/
if ( ( clone_flags & CLONE_SIGHAND ) & & ! ( clone_flags & CLONE_VM ) )
return ERR_PTR ( - EINVAL ) ;
2009-09-24 02:57:20 +04:00
/*
* Siblings of global init remain as zombies on exit since they are
* not reaped by their parent ( swapper ) . To solve this and to avoid
* multi - rooted process trees , prevent global and container - inits
* from creating siblings .
*/
if ( ( clone_flags & CLONE_PARENT ) & &
current - > signal - > flags & SIGNAL_UNKILLABLE )
return ERR_PTR ( - EINVAL ) ;
2012-12-21 07:26:06 +04:00
/*
2013-09-12 01:19:41 +04:00
* If the new process will be in a different pid or user namespace
2015-08-11 02:25:44 +03:00
* do not allow it to share a thread group with the forking task .
2012-12-21 07:26:06 +04:00
*/
2015-08-11 02:25:44 +03:00
if ( clone_flags & CLONE_THREAD ) {
2013-09-12 01:19:41 +04:00
if ( ( clone_flags & ( CLONE_NEWUSER | CLONE_NEWPID ) ) | |
ns: Introduce Time Namespace
Time Namespace isolates clock values.
The kernel provides access to several clocks CLOCK_REALTIME,
CLOCK_MONOTONIC, CLOCK_BOOTTIME, etc.
CLOCK_REALTIME
System-wide clock that measures real (i.e., wall-clock) time.
CLOCK_MONOTONIC
Clock that cannot be set and represents monotonic time since
some unspecified starting point.
CLOCK_BOOTTIME
Identical to CLOCK_MONOTONIC, except it also includes any time
that the system is suspended.
For many users, the time namespace means the ability to changes date and
time in a container (CLOCK_REALTIME). Providing per namespace notions of
CLOCK_REALTIME would be complex with a massive overhead, but has a dubious
value.
But in the context of checkpoint/restore functionality, monotonic and
boottime clocks become interesting. Both clocks are monotonic with
unspecified starting points. These clocks are widely used to measure time
slices and set timers. After restoring or migrating processes, it has to be
guaranteed that they never go backward. In an ideal case, the behavior of
these clocks should be the same as for a case when a whole system is
suspended. All this means that it is required to set CLOCK_MONOTONIC and
CLOCK_BOOTTIME clocks, which can be achieved by adding per-namespace
offsets for clocks.
A time namespace is similar to a pid namespace in the way how it is
created: unshare(CLONE_NEWTIME) system call creates a new time namespace,
but doesn't set it to the current process. Then all children of the process
will be born in the new time namespace, or a process can use the setns()
system call to join a namespace.
This scheme allows setting clock offsets for a namespace, before any
processes appear in it.
All available clone flags have been used, so CLONE_NEWTIME uses the highest
bit of CSIGNAL. It means that it can be used only with the unshare() and
the clone3() system calls.
[ tglx: Adjusted paragraph about clone3() to reality and massaged the
changelog a bit. ]
Co-developed-by: Dmitry Safonov <dima@arista.com>
Signed-off-by: Andrei Vagin <avagin@gmail.com>
Signed-off-by: Dmitry Safonov <dima@arista.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://criu.org/Time_namespace
Link: https://lists.openvz.org/pipermail/criu/2018-June/041504.html
Link: https://lore.kernel.org/r/20191112012724.250792-4-dima@arista.com
2019-11-12 04:26:52 +03:00
( task_active_pid_ns ( current ) ! = nsp - > pid_ns_for_children ) )
return ERR_PTR ( - EINVAL ) ;
}
/*
* If the new process will be in a different time namespace
* do not allow it to share VM or a thread group with the forking task .
*/
if ( clone_flags & ( CLONE_THREAD | CLONE_VM ) ) {
if ( nsp - > time_ns ! = nsp - > time_ns_for_children )
2013-09-12 01:19:41 +04:00
return ERR_PTR ( - EINVAL ) ;
}
2012-12-21 07:26:06 +04:00
2019-03-27 15:04:15 +03:00
if ( clone_flags & CLONE_PIDFD ) {
/*
* - CLONE_DETACHED is blocked so that we can potentially
* reuse it later for CLONE_PIDFD .
* - CLONE_THREAD is blocked until someone really needs it .
*/
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
if ( clone_flags & ( CLONE_DETACHED | CLONE_THREAD ) )
2019-03-27 15:04:15 +03:00
return ERR_PTR ( - EINVAL ) ;
}
2018-07-23 23:20:37 +03:00
/*
* Force any signals received before this point to be delivered
* before the fork happens . Collect up signals sent to multiple
* processes that happen during the fork and delay them so that
* they appear to happen after the fork .
*/
sigemptyset ( & delayed . signal ) ;
INIT_HLIST_NODE ( & delayed . node ) ;
spin_lock_irq ( & current - > sighand - > siglock ) ;
if ( ! ( clone_flags & CLONE_THREAD ) )
hlist_add_head ( & delayed . node , & current - > signal - > multiprocess ) ;
recalc_sigpending ( ) ;
spin_unlock_irq ( & current - > sighand - > siglock ) ;
retval = - ERESTARTNOINTR ;
2021-03-22 18:39:12 +03:00
if ( task_sigpending ( current ) )
2018-07-23 23:20:37 +03:00
goto fork_out ;
2005-04-17 02:20:36 +04:00
retval = - ENOMEM ;
2016-05-24 02:24:05 +03:00
p = dup_task_struct ( current , node ) ;
2005-04-17 02:20:36 +04:00
if ( ! p )
goto fork_out ;
2021-03-26 18:05:22 +03:00
if ( args - > io_thread ) {
/*
* Mark us an IO worker , and block any signal that isn ' t
* fatal or STOP
*/
2021-03-04 22:21:05 +03:00
p - > flags | = PF_IO_WORKER ;
2021-03-26 18:05:22 +03:00
siginitsetinv ( & p - > blocked , sigmask ( SIGKILL ) | sigmask ( SIGSTOP ) ) ;
}
2005-04-17 02:20:36 +04:00
2017-05-09 10:39:59 +03:00
/*
* This _must_ happen before we call free_task ( ) , i . e . before we jump
* to any of the bad_fork_ * labels . This is to avoid freeing
* p - > set_child_tid which is ( ab ) used as a kthread ' s data pointer for
* kernel threads ( PF_KTHREAD ) .
*/
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
p - > set_child_tid = ( clone_flags & CLONE_CHILD_SETTID ) ? args - > child_tid : NULL ;
2017-05-09 10:39:59 +03:00
/*
* Clear TID on mm_release ( ) ?
*/
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
p - > clear_child_tid = ( clone_flags & CLONE_CHILD_CLEARTID ) ? args - > child_tid : NULL ;
2017-05-09 10:39:59 +03:00
2009-06-03 00:39:48 +04:00
ftrace_graph_init_task ( p ) ;
2006-10-17 11:10:33 +04:00
rt_mutex_init_task ( p ) ;
2020-05-25 13:22:41 +03:00
lockdep_assert_irqs_enabled ( ) ;
2008-07-14 14:09:28 +04:00
# ifdef CONFIG_PROVE_LOCKING
2006-07-03 11:24:42 +04:00
DEBUG_LOCKS_WARN_ON ( ! p - > softirqs_enabled ) ;
# endif
2005-04-17 02:20:36 +04:00
retval = - EAGAIN ;
2021-04-22 15:27:11 +03:00
if ( is_ucounts_overlimit ( task_ucounts ( p ) , UCOUNT_RLIMIT_NPROC , rlimit ( RLIMIT_NPROC ) ) ) {
2013-07-04 02:08:29 +04:00
if ( p - > real_cred - > user ! = INIT_USER & &
! capable ( CAP_SYS_RESOURCE ) & & ! capable ( CAP_SYS_ADMIN ) )
2005-04-17 02:20:36 +04:00
goto bad_fork_free ;
}
2011-08-08 19:02:04 +04:00
current - > flags & = ~ PF_NPROC_EXCEEDED ;
2005-04-17 02:20:36 +04:00
2008-11-14 02:39:17 +03:00
retval = copy_creds ( p , clone_flags ) ;
if ( retval < 0 )
goto bad_fork_free ;
2005-04-17 02:20:36 +04:00
/*
* If multiple threads are within copy_process ( ) , then this check
* triggers too late . This doesn ' t hurt , the check is only there
* to stop root fork bombs .
*/
2009-02-06 11:17:19 +03:00
retval = - EAGAIN ;
2020-06-23 07:12:40 +03:00
if ( data_race ( nr_threads > = max_threads ) )
2005-04-17 02:20:36 +04:00
goto bad_fork_cleanup_count ;
2006-07-14 11:24:36 +04:00
delayacct_tsk_init ( p ) ; /* Must remain after dup_task_struct() */
2021-05-26 02:58:49 +03:00
p - > flags & = ~ ( PF_SUPERPRIV | PF_WQ_WORKER | PF_IDLE | PF_NO_SETAFFINITY ) ;
2014-04-08 02:37:27 +04:00
p - > flags | = PF_FORKNOEXEC ;
2005-04-17 02:20:36 +04:00
INIT_LIST_HEAD ( & p - > children ) ;
INIT_LIST_HEAD ( & p - > sibling ) ;
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-23 00:56:52 +04:00
rcu_copy_process ( p ) ;
2005-04-17 02:20:36 +04:00
p - > vfork_done = NULL ;
spin_lock_init ( & p - > alloc_lock ) ;
init_sigpending ( & p - > pending ) ;
2011-12-15 17:56:09 +04:00
p - > utime = p - > stime = p - > gtime = 0 ;
2016-11-15 05:06:51 +03:00
# ifdef CONFIG_ARCH_HAS_SCALED_CPUTIME
2011-12-15 17:56:09 +04:00
p - > utimescaled = p - > stimescaled = 0 ;
2016-11-15 05:06:51 +03:00
# endif
2015-06-30 12:30:54 +03:00
prev_cputime_init ( & p - > prev_cputime ) ;
2012-12-16 23:00:34 +04:00
# ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN
2017-06-29 20:15:10 +03:00
seqcount_init ( & p - > vtime . seqcount ) ;
p - > vtime . starttime = 0 ;
p - > vtime . state = VTIME_INACTIVE ;
2012-12-16 23:00:34 +04:00
# endif
2020-09-13 22:09:39 +03:00
# ifdef CONFIG_IO_URING
p - > io_uring = NULL ;
# endif
2010-04-07 01:34:42 +04:00
# if defined(SPLIT_RSS_COUNTING)
memset ( & p - > rss_stat , 0 , sizeof ( p - > rss_stat ) ) ;
# endif
2007-07-09 20:52:00 +04:00
2008-09-02 02:52:40 +04:00
p - > default_timer_slack_ns = current - > timer_slack_ns ;
psi: pressure stall information for CPU, memory, and IO
When systems are overcommitted and resources become contended, it's hard
to tell exactly the impact this has on workload productivity, or how close
the system is to lockups and OOM kills. In particular, when machines work
multiple jobs concurrently, the impact of overcommit in terms of latency
and throughput on the individual job can be enormous.
In order to maximize hardware utilization without sacrificing individual
job health or risk complete machine lockups, this patch implements a way
to quantify resource pressure in the system.
A kernel built with CONFIG_PSI=y creates files in /proc/pressure/ that
expose the percentage of time the system is stalled on CPU, memory, or IO,
respectively. Stall states are aggregate versions of the per-task delay
accounting delays:
cpu: some tasks are runnable but not executing on a CPU
memory: tasks are reclaiming, or waiting for swapin or thrashing cache
io: tasks are waiting for io completions
These percentages of walltime can be thought of as pressure percentages,
and they give a general sense of system health and productivity loss
incurred by resource overcommit. They can also indicate when the system
is approaching lockup scenarios and OOMs.
To do this, psi keeps track of the task states associated with each CPU
and samples the time they spend in stall states. Every 2 seconds, the
samples are averaged across CPUs - weighted by the CPUs' non-idle time to
eliminate artifacts from unused CPUs - and translated into percentages of
walltime. A running average of those percentages is maintained over 10s,
1m, and 5m periods (similar to the loadaverage).
[hannes@cmpxchg.org: doc fixlet, per Randy]
Link: http://lkml.kernel.org/r/20180828205625.GA14030@cmpxchg.org
[hannes@cmpxchg.org: code optimization]
Link: http://lkml.kernel.org/r/20180907175015.GA8479@cmpxchg.org
[hannes@cmpxchg.org: rename psi_clock() to psi_update_work(), per Peter]
Link: http://lkml.kernel.org/r/20180907145404.GB11088@cmpxchg.org
[hannes@cmpxchg.org: fix build]
Link: http://lkml.kernel.org/r/20180913014222.GA2370@cmpxchg.org
Link: http://lkml.kernel.org/r/20180828172258.3185-9-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Daniel Drake <drake@endlessm.com>
Tested-by: Suren Baghdasaryan <surenb@google.com>
Cc: Christopher Lameter <cl@linux.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Johannes Weiner <jweiner@fb.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Enderborg <peter.enderborg@sony.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vinayak Menon <vinmenon@codeaurora.org>
Cc: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-10-27 01:06:27 +03:00
# ifdef CONFIG_PSI
p - > psi_flags = 0 ;
# endif
2008-07-27 19:29:15 +04:00
task_io_accounting_init ( & p - > ioac ) ;
2005-04-17 02:20:36 +04:00
acct_clear_integrals ( p ) ;
2019-08-21 22:09:06 +03:00
posix_cputimers_init ( & p - > posix_cputimers ) ;
2005-04-17 02:20:36 +04:00
p - > io_context = NULL ;
2018-05-13 04:58:21 +03:00
audit_set_context ( p , NULL ) ;
2007-10-19 10:39:33 +04:00
cgroup_fork ( p ) ;
2005-04-17 02:20:36 +04:00
# ifdef CONFIG_NUMA
2008-04-28 13:13:09 +04:00
p - > mempolicy = mpol_dup ( p - > mempolicy ) ;
2011-07-27 03:08:39 +04:00
if ( IS_ERR ( p - > mempolicy ) ) {
retval = PTR_ERR ( p - > mempolicy ) ;
p - > mempolicy = NULL ;
2014-03-28 11:18:27 +04:00
goto bad_fork_cleanup_threadgroup_lock ;
2011-07-27 03:08:39 +04:00
}
2005-04-17 02:20:36 +04:00
# endif
2011-07-27 03:08:30 +04:00
# ifdef CONFIG_CPUSETS
p - > cpuset_mem_spread_rotor = NUMA_NO_NODE ;
p - > cpuset_slab_spread_rotor = NUMA_NO_NODE ;
2020-07-20 18:55:19 +03:00
seqcount_spinlock_init ( & p - > mems_allowed_seq , & p - > alloc_lock ) ;
2011-07-27 03:08:30 +04:00
# endif
2006-07-03 11:24:42 +04:00
# ifdef CONFIG_TRACE_IRQFLAGS
2020-07-29 14:09:15 +03:00
memset ( & p - > irqtrace , 0 , sizeof ( p - > irqtrace ) ) ;
p - > irqtrace . hardirq_disable_ip = _THIS_IP_ ;
p - > irqtrace . softirq_enable_ip = _THIS_IP_ ;
p - > softirqs_enabled = 1 ;
p - > softirq_context = 0 ;
2006-07-03 11:24:42 +04:00
# endif
2015-05-11 18:52:06 +03:00
p - > pagefault_disabled = 0 ;
[PATCH] lockdep: core
Do 'make oldconfig' and accept all the defaults for new config options -
reboot into the kernel and if everything goes well it should boot up fine and
you should have /proc/lockdep and /proc/lockdep_stats files.
Typically if the lock validator finds some problem it will print out
voluminous debug output that begins with "BUG: ..." and which syslog output
can be used by kernel developers to figure out the precise locking scenario.
What does the lock validator do? It "observes" and maps all locking rules as
they occur dynamically (as triggered by the kernel's natural use of spinlocks,
rwlocks, mutexes and rwsems). Whenever the lock validator subsystem detects a
new locking scenario, it validates this new rule against the existing set of
rules. If this new rule is consistent with the existing set of rules then the
new rule is added transparently and the kernel continues as normal. If the
new rule could create a deadlock scenario then this condition is printed out.
When determining validity of locking, all possible "deadlock scenarios" are
considered: assuming arbitrary number of CPUs, arbitrary irq context and task
context constellations, running arbitrary combinations of all the existing
locking scenarios. In a typical system this means millions of separate
scenarios. This is why we call it a "locking correctness" validator - for all
rules that are observed the lock validator proves it with mathematical
certainty that a deadlock could not occur (assuming that the lock validator
implementation itself is correct and its internal data structures are not
corrupted by some other kernel subsystem). [see more details and conditionals
of this statement in include/linux/lockdep.h and
Documentation/lockdep-design.txt]
Furthermore, this "all possible scenarios" property of the validator also
enables the finding of complex, highly unlikely multi-CPU multi-context races
via single single-context rules, increasing the likelyhood of finding bugs
drastically. In practical terms: the lock validator already found a bug in
the upstream kernel that could only occur on systems with 3 or more CPUs, and
which needed 3 very unlikely code sequences to occur at once on the 3 CPUs.
That bug was found and reported on a single-CPU system (!). So in essence a
race will be found "piecemail-wise", triggering all the necessary components
for the race, without having to reproduce the race scenario itself! In its
short existence the lock validator found and reported many bugs before they
actually caused a real deadlock.
To further increase the efficiency of the validator, the mapping is not per
"lock instance", but per "lock-class". For example, all struct inode objects
in the kernel have inode->inotify_mutex. If there are 10,000 inodes cached,
then there are 10,000 lock objects. But ->inotify_mutex is a single "lock
type", and all locking activities that occur against ->inotify_mutex are
"unified" into this single lock-class. The advantage of the lock-class
approach is that all historical ->inotify_mutex uses are mapped into a single
(and as narrow as possible) set of locking rules - regardless of how many
different tasks or inode structures it took to build this set of rules. The
set of rules persist during the lifetime of the kernel.
To see the rough magnitude of checking that the lock validator does, here's a
portion of /proc/lockdep_stats, fresh after bootup:
lock-classes: 694 [max: 2048]
direct dependencies: 1598 [max: 8192]
indirect dependencies: 17896
all direct dependencies: 16206
dependency chains: 1910 [max: 8192]
in-hardirq chains: 17
in-softirq chains: 105
in-process chains: 1065
stack-trace entries: 38761 [max: 131072]
combined max dependencies: 2033928
hardirq-safe locks: 24
hardirq-unsafe locks: 176
softirq-safe locks: 53
softirq-unsafe locks: 137
irq-safe locks: 59
irq-unsafe locks: 176
The lock validator has observed 1598 actual single-thread locking patterns,
and has validated all possible 2033928 distinct locking scenarios.
More details about the design of the lock validator can be found in
Documentation/lockdep-design.txt, which can also found at:
http://redhat.com/~mingo/lockdep-patches/lockdep-design.txt
[bunk@stusta.de: cleanups]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-07-03 11:24:50 +04:00
# ifdef CONFIG_LOCKDEP
2017-08-07 10:12:52 +03:00
lockdep_init_task ( p ) ;
[PATCH] lockdep: core
Do 'make oldconfig' and accept all the defaults for new config options -
reboot into the kernel and if everything goes well it should boot up fine and
you should have /proc/lockdep and /proc/lockdep_stats files.
Typically if the lock validator finds some problem it will print out
voluminous debug output that begins with "BUG: ..." and which syslog output
can be used by kernel developers to figure out the precise locking scenario.
What does the lock validator do? It "observes" and maps all locking rules as
they occur dynamically (as triggered by the kernel's natural use of spinlocks,
rwlocks, mutexes and rwsems). Whenever the lock validator subsystem detects a
new locking scenario, it validates this new rule against the existing set of
rules. If this new rule is consistent with the existing set of rules then the
new rule is added transparently and the kernel continues as normal. If the
new rule could create a deadlock scenario then this condition is printed out.
When determining validity of locking, all possible "deadlock scenarios" are
considered: assuming arbitrary number of CPUs, arbitrary irq context and task
context constellations, running arbitrary combinations of all the existing
locking scenarios. In a typical system this means millions of separate
scenarios. This is why we call it a "locking correctness" validator - for all
rules that are observed the lock validator proves it with mathematical
certainty that a deadlock could not occur (assuming that the lock validator
implementation itself is correct and its internal data structures are not
corrupted by some other kernel subsystem). [see more details and conditionals
of this statement in include/linux/lockdep.h and
Documentation/lockdep-design.txt]
Furthermore, this "all possible scenarios" property of the validator also
enables the finding of complex, highly unlikely multi-CPU multi-context races
via single single-context rules, increasing the likelyhood of finding bugs
drastically. In practical terms: the lock validator already found a bug in
the upstream kernel that could only occur on systems with 3 or more CPUs, and
which needed 3 very unlikely code sequences to occur at once on the 3 CPUs.
That bug was found and reported on a single-CPU system (!). So in essence a
race will be found "piecemail-wise", triggering all the necessary components
for the race, without having to reproduce the race scenario itself! In its
short existence the lock validator found and reported many bugs before they
actually caused a real deadlock.
To further increase the efficiency of the validator, the mapping is not per
"lock instance", but per "lock-class". For example, all struct inode objects
in the kernel have inode->inotify_mutex. If there are 10,000 inodes cached,
then there are 10,000 lock objects. But ->inotify_mutex is a single "lock
type", and all locking activities that occur against ->inotify_mutex are
"unified" into this single lock-class. The advantage of the lock-class
approach is that all historical ->inotify_mutex uses are mapped into a single
(and as narrow as possible) set of locking rules - regardless of how many
different tasks or inode structures it took to build this set of rules. The
set of rules persist during the lifetime of the kernel.
To see the rough magnitude of checking that the lock validator does, here's a
portion of /proc/lockdep_stats, fresh after bootup:
lock-classes: 694 [max: 2048]
direct dependencies: 1598 [max: 8192]
indirect dependencies: 17896
all direct dependencies: 16206
dependency chains: 1910 [max: 8192]
in-hardirq chains: 17
in-softirq chains: 105
in-process chains: 1065
stack-trace entries: 38761 [max: 131072]
combined max dependencies: 2033928
hardirq-safe locks: 24
hardirq-unsafe locks: 176
softirq-safe locks: 53
softirq-unsafe locks: 137
irq-safe locks: 59
irq-unsafe locks: 176
The lock validator has observed 1598 actual single-thread locking patterns,
and has validated all possible 2033928 distinct locking scenarios.
More details about the design of the lock validator can be found in
Documentation/lockdep-design.txt, which can also found at:
http://redhat.com/~mingo/lockdep-patches/lockdep-design.txt
[bunk@stusta.de: cleanups]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-07-03 11:24:50 +04:00
# endif
2005-04-17 02:20:36 +04:00
2006-01-10 02:59:20 +03:00
# ifdef CONFIG_DEBUG_MUTEXES
p - > blocked_on = NULL ; /* not blocked yet */
# endif
2013-03-24 03:11:31 +04:00
# ifdef CONFIG_BCACHE
p - > sequential_io = 0 ;
p - > sequential_io_avg = 0 ;
# endif
2021-02-26 02:43:14 +03:00
# ifdef CONFIG_BPF_SYSCALL
RCU_INIT_POINTER ( p - > bpf_storage , NULL ) ;
# endif
2009-04-03 18:43:48 +04:00
sched: fix copy_namespace() <-> sched_fork() dependency in do_fork
Sukadev Bhattiprolu reported a kernel crash with control groups.
There are couple of problems discovered by Suka's test:
- The test requires the cgroup filesystem to be mounted with
atleast the cpu and ns options (i.e both namespace and cpu
controllers are active in the same hierarchy).
# mkdir /dev/cpuctl
# mount -t cgroup -ocpu,ns none cpuctl
(or simply)
# mount -t cgroup none cpuctl -> Will activate all controllers
in same hierarchy.
- The test invokes clone() with CLONE_NEWNS set. This causes a a new child
to be created, also a new group (do_fork->copy_namespaces->ns_cgroup_clone->
cgroup_clone) and the child is attached to the new group (cgroup_clone->
attach_task->sched_move_task). At this point in time, the child's scheduler
related fields are uninitialized (including its on_rq field, which it has
inherited from parent). As a result sched_move_task thinks its on
runqueue, when it isn't.
As a solution to this problem, I moved sched_fork() call, which
initializes scheduler related fields on a new task, before
copy_namespaces(). I am not sure though whether moving up will
cause other side-effects. Do you see any issue?
- The second problem exposed by this test is that task_new_fair()
assumes that parent and child will be part of the same group (which
needn't be as this test shows). As a result, cfs_rq->curr can be NULL
for the child.
The solution is to test for curr pointer being NULL in
task_new_fair().
With the patch below, I could run ns_exec() fine w/o a crash.
Reported-by: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Signed-off-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2007-11-10 00:39:39 +03:00
/* Perform scheduler related setup. Assign this task to a CPU. */
sched/deadline: Add SCHED_DEADLINE structures & implementation
Introduces the data structures, constants and symbols needed for
SCHED_DEADLINE implementation.
Core data structure of SCHED_DEADLINE are defined, along with their
initializers. Hooks for checking if a task belong to the new policy
are also added where they are needed.
Adds a scheduling class, in sched/dl.c and a new policy called
SCHED_DEADLINE. It is an implementation of the Earliest Deadline
First (EDF) scheduling algorithm, augmented with a mechanism (called
Constant Bandwidth Server, CBS) that makes it possible to isolate
the behaviour of tasks between each other.
The typical -deadline task will be made up of a computation phase
(instance) which is activated on a periodic or sporadic fashion. The
expected (maximum) duration of such computation is called the task's
runtime; the time interval by which each instance need to be completed
is called the task's relative deadline. The task's absolute deadline
is dynamically calculated as the time instant a task (better, an
instance) activates plus the relative deadline.
The EDF algorithms selects the task with the smallest absolute
deadline as the one to be executed first, while the CBS ensures each
task to run for at most its runtime every (relative) deadline
length time interval, avoiding any interference between different
tasks (bandwidth isolation).
Thanks to this feature, also tasks that do not strictly comply with
the computational model sketched above can effectively use the new
policy.
To summarize, this patch:
- introduces the data structures, constants and symbols needed;
- implements the core logic of the scheduling algorithm in the new
scheduling class file;
- provides all the glue code between the new scheduling class and
the core scheduler and refines the interactions between sched/dl
and the other existing scheduling classes.
Signed-off-by: Dario Faggioli <raistlin@linux.it>
Signed-off-by: Michael Trimarchi <michael@amarulasolutions.com>
Signed-off-by: Fabio Checconi <fchecconi@gmail.com>
Signed-off-by: Juri Lelli <juri.lelli@gmail.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1383831828-15501-4-git-send-email-juri.lelli@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2013-11-28 14:14:43 +04:00
retval = sched_fork ( clone_flags , p ) ;
if ( retval )
goto bad_fork_cleanup_policy ;
2009-05-25 16:45:27 +04:00
2021-04-08 13:35:58 +03:00
retval = perf_event_init_task ( p , clone_flags ) ;
2009-05-25 16:45:27 +04:00
if ( retval )
goto bad_fork_cleanup_policy ;
2011-07-27 03:08:39 +04:00
retval = audit_alloc ( p ) ;
if ( retval )
2014-10-03 03:17:02 +04:00
goto bad_fork_cleanup_perf ;
2005-04-17 02:20:36 +04:00
/* copy all the process information */
shm: make exit_shm work proportional to task activity
This is small set of patches our team has had kicking around for a few
versions internally that fixes tasks getting hung on shm_exit when there
are many threads hammering it at once.
Anton wrote a simple test to cause the issue:
http://ozlabs.org/~anton/junkcode/bust_shm_exit.c
Before applying this patchset, this test code will cause either hanging
tracebacks or pthread out of memory errors.
After this patchset, it will still produce output like:
root@somehost:~# ./bust_shm_exit 1024 160
...
INFO: rcu_sched detected stalls on CPUs/tasks: {} (detected by 116, t=2111 jiffies, g=241, c=240, q=7113)
INFO: Stall ended before state dump start
...
But the task will continue to run along happily, so we consider this an
improvement over hanging, even if it's a bit noisy.
This patch (of 3):
exit_shm obtains the ipc_ns shm rwsem for write and holds it while it
walks every shared memory segment in the namespace. Thus the amount of
work is related to the number of shm segments in the namespace not the
number of segments that might need to be cleaned.
In addition, this occurs after the task has been notified the thread has
exited, so the number of tasks waiting for the ns shm rwsem can grow
without bound until memory is exausted.
Add a list to the task struct of all shmids allocated by this task. Init
the list head in copy_process. Use the ns->rwsem for locking. Add
segments after id is added, remove before removing from id.
On unshare of NEW_IPCNS orphan any ids as if the task had exited, similar
to handling of semaphore undo.
I chose a define for the init sequence since its a simple list init,
otherwise it would require a function call to avoid include loops between
the semaphore code and the task struct. Converting the list_del to
list_del_init for the unshare cases would remove the exit followed by
init, but I left it blow up if not inited.
Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: Jack Miller <millerjo@us.ibm.com>
Cc: Davidlohr Bueso <davidlohr@hp.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Anton Blanchard <anton@samba.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-08-09 01:23:19 +04:00
shm_init_task ( p ) ;
2017-03-24 14:46:33 +03:00
retval = security_task_alloc ( p , clone_flags ) ;
2011-07-27 03:08:39 +04:00
if ( retval )
2005-04-17 02:20:36 +04:00
goto bad_fork_cleanup_audit ;
2017-03-24 14:46:33 +03:00
retval = copy_semundo ( clone_flags , p ) ;
if ( retval )
goto bad_fork_cleanup_security ;
2011-07-27 03:08:39 +04:00
retval = copy_files ( clone_flags , p ) ;
if ( retval )
2005-04-17 02:20:36 +04:00
goto bad_fork_cleanup_semundo ;
2011-07-27 03:08:39 +04:00
retval = copy_fs ( clone_flags , p ) ;
if ( retval )
2005-04-17 02:20:36 +04:00
goto bad_fork_cleanup_files ;
2011-07-27 03:08:39 +04:00
retval = copy_sighand ( clone_flags , p ) ;
if ( retval )
2005-04-17 02:20:36 +04:00
goto bad_fork_cleanup_fs ;
2011-07-27 03:08:39 +04:00
retval = copy_signal ( clone_flags , p ) ;
if ( retval )
2005-04-17 02:20:36 +04:00
goto bad_fork_cleanup_sighand ;
2011-07-27 03:08:39 +04:00
retval = copy_mm ( clone_flags , p ) ;
if ( retval )
2005-04-17 02:20:36 +04:00
goto bad_fork_cleanup_signal ;
2011-07-27 03:08:39 +04:00
retval = copy_namespaces ( clone_flags , p ) ;
if ( retval )
CRED: Inaugurate COW credentials
Inaugurate copy-on-write credentials management. This uses RCU to manage the
credentials pointer in the task_struct with respect to accesses by other tasks.
A process may only modify its own credentials, and so does not need locking to
access or modify its own credentials.
A mutex (cred_replace_mutex) is added to the task_struct to control the effect
of PTRACE_ATTACHED on credential calculations, particularly with respect to
execve().
With this patch, the contents of an active credentials struct may not be
changed directly; rather a new set of credentials must be prepared, modified
and committed using something like the following sequence of events:
struct cred *new = prepare_creds();
int ret = blah(new);
if (ret < 0) {
abort_creds(new);
return ret;
}
return commit_creds(new);
There are some exceptions to this rule: the keyrings pointed to by the active
credentials may be instantiated - keyrings violate the COW rule as managing
COW keyrings is tricky, given that it is possible for a task to directly alter
the keys in a keyring in use by another task.
To help enforce this, various pointers to sets of credentials, such as those in
the task_struct, are declared const. The purpose of this is compile-time
discouragement of altering credentials through those pointers. Once a set of
credentials has been made public through one of these pointers, it may not be
modified, except under special circumstances:
(1) Its reference count may incremented and decremented.
(2) The keyrings to which it points may be modified, but not replaced.
The only safe way to modify anything else is to create a replacement and commit
using the functions described in Documentation/credentials.txt (which will be
added by a later patch).
This patch and the preceding patches have been tested with the LTP SELinux
testsuite.
This patch makes several logical sets of alteration:
(1) execve().
This now prepares and commits credentials in various places in the
security code rather than altering the current creds directly.
(2) Temporary credential overrides.
do_coredump() and sys_faccessat() now prepare their own credentials and
temporarily override the ones currently on the acting thread, whilst
preventing interference from other threads by holding cred_replace_mutex
on the thread being dumped.
This will be replaced in a future patch by something that hands down the
credentials directly to the functions being called, rather than altering
the task's objective credentials.
(3) LSM interface.
A number of functions have been changed, added or removed:
(*) security_capset_check(), ->capset_check()
(*) security_capset_set(), ->capset_set()
Removed in favour of security_capset().
(*) security_capset(), ->capset()
New. This is passed a pointer to the new creds, a pointer to the old
creds and the proposed capability sets. It should fill in the new
creds or return an error. All pointers, barring the pointer to the
new creds, are now const.
(*) security_bprm_apply_creds(), ->bprm_apply_creds()
Changed; now returns a value, which will cause the process to be
killed if it's an error.
(*) security_task_alloc(), ->task_alloc_security()
Removed in favour of security_prepare_creds().
(*) security_cred_free(), ->cred_free()
New. Free security data attached to cred->security.
(*) security_prepare_creds(), ->cred_prepare()
New. Duplicate any security data attached to cred->security.
(*) security_commit_creds(), ->cred_commit()
New. Apply any security effects for the upcoming installation of new
security by commit_creds().
(*) security_task_post_setuid(), ->task_post_setuid()
Removed in favour of security_task_fix_setuid().
(*) security_task_fix_setuid(), ->task_fix_setuid()
Fix up the proposed new credentials for setuid(). This is used by
cap_set_fix_setuid() to implicitly adjust capabilities in line with
setuid() changes. Changes are made to the new credentials, rather
than the task itself as in security_task_post_setuid().
(*) security_task_reparent_to_init(), ->task_reparent_to_init()
Removed. Instead the task being reparented to init is referred
directly to init's credentials.
NOTE! This results in the loss of some state: SELinux's osid no
longer records the sid of the thread that forked it.
(*) security_key_alloc(), ->key_alloc()
(*) security_key_permission(), ->key_permission()
Changed. These now take cred pointers rather than task pointers to
refer to the security context.
(4) sys_capset().
This has been simplified and uses less locking. The LSM functions it
calls have been merged.
(5) reparent_to_kthreadd().
This gives the current thread the same credentials as init by simply using
commit_thread() to point that way.
(6) __sigqueue_alloc() and switch_uid()
__sigqueue_alloc() can't stop the target task from changing its creds
beneath it, so this function gets a reference to the currently applicable
user_struct which it then passes into the sigqueue struct it returns if
successful.
switch_uid() is now called from commit_creds(), and possibly should be
folded into that. commit_creds() should take care of protecting
__sigqueue_alloc().
(7) [sg]et[ug]id() and co and [sg]et_current_groups.
The set functions now all use prepare_creds(), commit_creds() and
abort_creds() to build and check a new set of credentials before applying
it.
security_task_set[ug]id() is called inside the prepared section. This
guarantees that nothing else will affect the creds until we've finished.
The calling of set_dumpable() has been moved into commit_creds().
Much of the functionality of set_user() has been moved into
commit_creds().
The get functions all simply access the data directly.
(8) security_task_prctl() and cap_task_prctl().
security_task_prctl() has been modified to return -ENOSYS if it doesn't
want to handle a function, or otherwise return the return value directly
rather than through an argument.
Additionally, cap_task_prctl() now prepares a new set of credentials, even
if it doesn't end up using it.
(9) Keyrings.
A number of changes have been made to the keyrings code:
(a) switch_uid_keyring(), copy_keys(), exit_keys() and suid_keys() have
all been dropped and built in to the credentials functions directly.
They may want separating out again later.
(b) key_alloc() and search_process_keyrings() now take a cred pointer
rather than a task pointer to specify the security context.
(c) copy_creds() gives a new thread within the same thread group a new
thread keyring if its parent had one, otherwise it discards the thread
keyring.
(d) The authorisation key now points directly to the credentials to extend
the search into rather pointing to the task that carries them.
(e) Installing thread, process or session keyrings causes a new set of
credentials to be created, even though it's not strictly necessary for
process or session keyrings (they're shared).
(10) Usermode helper.
The usermode helper code now carries a cred struct pointer in its
subprocess_info struct instead of a new session keyring pointer. This set
of credentials is derived from init_cred and installed on the new process
after it has been cloned.
call_usermodehelper_setup() allocates the new credentials and
call_usermodehelper_freeinfo() discards them if they haven't been used. A
special cred function (prepare_usermodeinfo_creds()) is provided
specifically for call_usermodehelper_setup() to call.
call_usermodehelper_setkeys() adjusts the credentials to sport the
supplied keyring as the new session keyring.
(11) SELinux.
SELinux has a number of changes, in addition to those to support the LSM
interface changes mentioned above:
(a) selinux_setprocattr() no longer does its check for whether the
current ptracer can access processes with the new SID inside the lock
that covers getting the ptracer's SID. Whilst this lock ensures that
the check is done with the ptracer pinned, the result is only valid
until the lock is released, so there's no point doing it inside the
lock.
(12) is_single_threaded().
This function has been extracted from selinux_setprocattr() and put into
a file of its own in the lib/ directory as join_session_keyring() now
wants to use it too.
The code in SELinux just checked to see whether a task shared mm_structs
with other tasks (CLONE_VM), but that isn't good enough. We really want
to know if they're part of the same thread group (CLONE_THREAD).
(13) nfsd.
The NFS server daemon now has to use the COW credentials to set the
credentials it is going to use. It really needs to pass the credentials
down to the functions it calls, but it can't do that until other patches
in this series have been applied.
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: James Morris <jmorris@namei.org>
2008-11-14 02:39:23 +03:00
goto bad_fork_cleanup_mm ;
2011-07-27 03:08:39 +04:00
retval = copy_io ( clone_flags , p ) ;
if ( retval )
2008-01-24 10:52:45 +03:00
goto bad_fork_cleanup_namespaces ;
2020-06-11 12:04:15 +03:00
retval = copy_thread ( clone_flags , args - > stack , args - > stack_size , p , args - > tls ) ;
2005-04-17 02:20:36 +04:00
if ( retval )
2008-01-24 10:52:45 +03:00
goto bad_fork_cleanup_io ;
2005-04-17 02:20:36 +04:00
2018-08-17 01:16:58 +03:00
stackleak_task_init ( p ) ;
2007-10-19 10:40:07 +04:00
if ( pid ! = & init_struct_pid ) {
2019-11-15 15:36:20 +03:00
pid = alloc_pid ( p - > nsproxy - > pid_ns_for_children , args - > set_tid ,
args - > set_tid_size ) ;
2015-04-16 22:47:38 +03:00
if ( IS_ERR ( pid ) ) {
retval = PTR_ERR ( pid ) ;
fork: free thread in copy_process on failure
When using this program (as root):
#include <err.h>
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <sys/io.h>
#include <sys/types.h>
#include <sys/wait.h>
#define ITER 1000
#define FORKERS 15
#define THREADS (6000/FORKERS) // 1850 is proc max
static void fork_100_wait()
{
unsigned a, to_wait = 0;
printf("\t%d forking %d\n", THREADS, getpid());
for (a = 0; a < THREADS; a++) {
switch (fork()) {
case 0:
usleep(1000);
exit(0);
break;
case -1:
break;
default:
to_wait++;
break;
}
}
printf("\t%d forked from %d, waiting for %d\n", THREADS, getpid(),
to_wait);
for (a = 0; a < to_wait; a++)
wait(NULL);
printf("\t%d waited from %d\n", THREADS, getpid());
}
static void run_forkers()
{
pid_t forkers[FORKERS];
unsigned a;
for (a = 0; a < FORKERS; a++) {
switch ((forkers[a] = fork())) {
case 0:
fork_100_wait();
exit(0);
break;
case -1:
err(1, "DIE fork of %d'th forker", a);
break;
default:
break;
}
}
for (a = 0; a < FORKERS; a++)
waitpid(forkers[a], NULL, 0);
}
int main()
{
unsigned a;
int ret;
ret = ioperm(10, 20, 0);
if (ret < 0)
err(1, "ioperm");
for (a = 0; a < ITER; a++)
run_forkers();
return 0;
}
kmemleak reports many occurences of this leak:
unreferenced object 0xffff8805917c8000 (size 8192):
comm "fork-leak", pid 2932, jiffies 4295354292 (age 1871.028s)
hex dump (first 32 bytes):
ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
backtrace:
[<ffffffff814cfbf5>] kmemdup+0x25/0x50
[<ffffffff8103ab43>] copy_thread_tls+0x6c3/0x9a0
[<ffffffff81150174>] copy_process+0x1a84/0x5790
[<ffffffff811dc375>] wake_up_new_task+0x2d5/0x6f0
[<ffffffff8115411d>] _do_fork+0x12d/0x820
...
Due to the leakage of the memory items which should have been freed in
arch/x86/kernel/process.c:exit_thread().
Make sure the memory is freed when fork fails later in copy_process.
This is done by calling exit_thread with the thread to kill.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: "James E.J. Bottomley" <jejb@parisc-linux.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Chen Liqin <liqin.linux@gmail.com>
Cc: Chris Metcalf <cmetcalf@mellanox.com>
Cc: Chris Zankel <chris@zankel.net>
Cc: David Howells <dhowells@redhat.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guan Xuetao <gxt@mprc.pku.edu.cn>
Cc: Haavard Skinnemoen <hskinnemoen@gmail.com>
Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Helge Deller <deller@gmx.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: James Hogan <james.hogan@imgtec.com>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: Jonas Bonn <jonas@southpole.se>
Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com>
Cc: Lennox Wu <lennox.wu@gmail.com>
Cc: Ley Foon Tan <lftan@altera.com>
Cc: Mark Salter <msalter@redhat.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Michal Simek <monstr@monstr.eu>
Cc: Mikael Starvik <starvik@axis.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Rich Felker <dalias@libc.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Richard Kuo <rkuo@codeaurora.org>
Cc: Richard Weinberger <richard@nod.at>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Steven Miao <realmz6@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 03:00:25 +03:00
goto bad_fork_cleanup_thread ;
2015-04-16 22:47:38 +03:00
}
2007-10-19 10:40:07 +04:00
}
2019-03-27 15:04:15 +03:00
/*
* This has to happen after we ' ve potentially unshared the file
* descriptor table ( so that the pidfd doesn ' t leak into the child
* if the fd table isn ' t shared ) .
*/
if ( clone_flags & CLONE_PIDFD ) {
2019-06-27 05:22:09 +03:00
retval = get_unused_fd_flags ( O_RDWR | O_CLOEXEC ) ;
2019-03-27 15:04:15 +03:00
if ( retval < 0 )
goto bad_fork_free_pid ;
pidfd = retval ;
2019-06-27 05:22:09 +03:00
pidfile = anon_inode_getfile ( " [pidfd] " , & pidfd_fops , pid ,
O_RDWR | O_CLOEXEC ) ;
if ( IS_ERR ( pidfile ) ) {
put_unused_fd ( pidfd ) ;
2019-07-01 17:01:46 +03:00
retval = PTR_ERR ( pidfile ) ;
2019-06-27 05:22:09 +03:00
goto bad_fork_free_pid ;
}
get_pid ( pid ) ; /* held by pidfile now */
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
retval = put_user ( pidfd , args - > pidfd ) ;
2019-03-27 15:04:15 +03:00
if ( retval )
goto bad_fork_put_pidfd ;
}
2011-03-08 15:19:51 +03:00
# ifdef CONFIG_BLOCK
p - > plug = NULL ;
# endif
2019-11-07 00:55:36 +03:00
futex_init_task ( p ) ;
[PATCH] Fix sigaltstack corruption among cloned threads
This patch fixes alternate signal stack corruption among cloned threads
with CLONE_SIGHAND (and CLONE_VM) for linux-2.6.16-rc6.
The value of alternate signal stack is currently inherited after a call of
clone(... CLONE_SIGHAND | CLONE_VM). But if sigaltstack is set by a
parent thread, and then if multiple cloned child threads (+ parent threads)
call signal handler at the same time, some threads may be conflicted -
because they share to use the same alternative signal stack region.
Finally they get sigsegv. It's an undesirable race condition. Note that
child threads created from NPTL pthread_create() also hit this conflict
when the parent thread uses sigaltstack, without my patch.
To fix this problem, this patch clears the child threads' sigaltstack
information like exec(). This behavior follows the SUSv3 specification.
In SUSv3, pthread_create() says "The alternate stack shall not be inherited
(when new threads are initialized)". It means that sigaltstack should be
cleared when sigaltstack memory space is shared by cloned threads with
CLONE_SIGHAND.
Note that I chose "if (clone_flags & CLONE_SIGHAND)" line because:
- If clone_flags line is not existed, fork() does not inherit sigaltstack.
- CLONE_VM is another choice, but vfork() does not inherit sigaltstack.
- CLONE_SIGHAND implies CLONE_VM, and it looks suitable.
- CLONE_THREAD is another candidate, and includes CLONE_SIGHAND + CLONE_VM,
but this flag has a bit different semantics.
I decided to use CLONE_SIGHAND.
[ Changed to test for CLONE_VM && !CLONE_VFORK after discussion --Linus ]
Signed-off-by: GOTO Masanori <gotom@sanori.org>
Cc: Roland McGrath <roland@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Acked-by: Linus Torvalds <torvalds@osdl.org>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Jakub Jelinek <jakub@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-14 08:20:44 +03:00
/*
* sigaltstack should be cleared when sharing the same VM
*/
if ( ( clone_flags & ( CLONE_VM | CLONE_VFORK ) ) = = CLONE_VM )
2016-04-14 23:20:04 +03:00
sas_ss_reset ( p ) ;
[PATCH] Fix sigaltstack corruption among cloned threads
This patch fixes alternate signal stack corruption among cloned threads
with CLONE_SIGHAND (and CLONE_VM) for linux-2.6.16-rc6.
The value of alternate signal stack is currently inherited after a call of
clone(... CLONE_SIGHAND | CLONE_VM). But if sigaltstack is set by a
parent thread, and then if multiple cloned child threads (+ parent threads)
call signal handler at the same time, some threads may be conflicted -
because they share to use the same alternative signal stack region.
Finally they get sigsegv. It's an undesirable race condition. Note that
child threads created from NPTL pthread_create() also hit this conflict
when the parent thread uses sigaltstack, without my patch.
To fix this problem, this patch clears the child threads' sigaltstack
information like exec(). This behavior follows the SUSv3 specification.
In SUSv3, pthread_create() says "The alternate stack shall not be inherited
(when new threads are initialized)". It means that sigaltstack should be
cleared when sigaltstack memory space is shared by cloned threads with
CLONE_SIGHAND.
Note that I chose "if (clone_flags & CLONE_SIGHAND)" line because:
- If clone_flags line is not existed, fork() does not inherit sigaltstack.
- CLONE_VM is another choice, but vfork() does not inherit sigaltstack.
- CLONE_SIGHAND implies CLONE_VM, and it looks suitable.
- CLONE_THREAD is another candidate, and includes CLONE_SIGHAND + CLONE_VM,
but this flag has a bit different semantics.
I decided to use CLONE_SIGHAND.
[ Changed to test for CLONE_VM && !CLONE_VFORK after discussion --Linus ]
Signed-off-by: GOTO Masanori <gotom@sanori.org>
Cc: Roland McGrath <roland@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Acked-by: Linus Torvalds <torvalds@osdl.org>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Jakub Jelinek <jakub@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-14 08:20:44 +03:00
2005-04-17 02:20:36 +04:00
/*
2009-12-16 03:47:16 +03:00
* Syscall tracing and stepping should be turned off in the
* child regardless of CLONE_PTRACE .
2005-04-17 02:20:36 +04:00
*/
2009-12-16 03:47:16 +03:00
user_disable_single_step ( p ) ;
2020-11-16 20:42:02 +03:00
clear_task_syscall_work ( p , SYSCALL_TRACE ) ;
2020-11-16 20:42:03 +03:00
# if defined(CONFIG_GENERIC_ENTRY) || defined(TIF_SYSCALL_EMU)
clear_task_syscall_work ( p , SYSCALL_EMU ) ;
[PATCH] UML Support - Ptrace: adds the host SYSEMU support, for UML and general usage
Jeff Dike <jdike@addtoit.com>,
Paolo 'Blaisorblade' Giarrusso <blaisorblade_spam@yahoo.it>,
Bodo Stroesser <bstroesser@fujitsu-siemens.com>
Adds a new ptrace(2) mode, called PTRACE_SYSEMU, resembling PTRACE_SYSCALL
except that the kernel does not execute the requested syscall; this is useful
to improve performance for virtual environments, like UML, which want to run
the syscall on their own.
In fact, using PTRACE_SYSCALL means stopping child execution twice, on entry
and on exit, and each time you also have two context switches; with SYSEMU you
avoid the 2nd stop and so save two context switches per syscall.
Also, some architectures don't have support in the host for changing the
syscall number via ptrace(), which is currently needed to skip syscall
execution (UML turns any syscall into getpid() to avoid it being executed on
the host). Fixing that is hard, while SYSEMU is easier to implement.
* This version of the patch includes some suggestions of Jeff Dike to avoid
adding any instructions to the syscall fast path, plus some other little
changes, by myself, to make it work even when the syscall is executed with
SYSENTER (but I'm unsure about them). It has been widely tested for quite a
lot of time.
* Various fixed were included to handle the various switches between
various states, i.e. when for instance a syscall entry is traced with one of
PT_SYSCALL / _SYSEMU / _SINGLESTEP and another one is used on exit.
Basically, this is done by remembering which one of them was used even after
the call to ptrace_notify().
* We're combining TIF_SYSCALL_EMU with TIF_SYSCALL_TRACE or TIF_SINGLESTEP
to make do_syscall_trace() notice that the current syscall was started with
SYSEMU on entry, so that no notification ought to be done in the exit path;
this is a bit of a hack, so this problem is solved in another way in next
patches.
* Also, the effects of the patch:
"Ptrace - i386: fix Syscall Audit interaction with singlestep"
are cancelled; they are restored back in the last patch of this series.
Detailed descriptions of the patches doing this kind of processing follow (but
I've already summed everything up).
* Fix behaviour when changing interception kind #1.
In do_syscall_trace(), we check the status of the TIF_SYSCALL_EMU flag
only after doing the debugger notification; but the debugger might have
changed the status of this flag because he continued execution with
PTRACE_SYSCALL, so this is wrong. This patch fixes it by saving the flag
status before calling ptrace_notify().
* Fix behaviour when changing interception kind #2:
avoid intercepting syscall on return when using SYSCALL again.
A guest process switching from using PTRACE_SYSEMU to PTRACE_SYSCALL
crashes.
The problem is in arch/i386/kernel/entry.S. The current SYSEMU patch
inhibits the syscall-handler to be called, but does not prevent
do_syscall_trace() to be called after this for syscall completion
interception.
The appended patch fixes this. It reuses the flag TIF_SYSCALL_EMU to
remember "we come from PTRACE_SYSEMU and now are in PTRACE_SYSCALL", since
the flag is unused in the depicted situation.
* Fix behaviour when changing interception kind #3:
avoid intercepting syscall on return when using SINGLESTEP.
When testing 2.6.9 and the skas3.v6 patch, with my latest patch and had
problems with singlestepping on UML in SKAS with SYSEMU. It looped
receiving SIGTRAPs without moving forward. EIP of the traced process was
the same for all SIGTRAPs.
What's missing is to handle switching from PTRACE_SYSCALL_EMU to
PTRACE_SINGLESTEP in a way very similar to what is done for the change from
PTRACE_SYSCALL_EMU to PTRACE_SYSCALL_TRACE.
I.e., after calling ptrace(PTRACE_SYSEMU), on the return path, the debugger is
notified and then wake ups the process; the syscall is executed (or skipped,
when do_syscall_trace() returns 0, i.e. when using PTRACE_SYSEMU), and
do_syscall_trace() is called again. Since we are on the return path of a
SYSEMU'd syscall, if the wake up is performed through ptrace(PTRACE_SYSCALL),
we must still avoid notifying the parent of the syscall exit. Now, this
behaviour is extended even to resuming with PTRACE_SINGLESTEP.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-09-04 02:57:18 +04:00
# endif
2019-05-15 01:42:34 +03:00
clear_tsk_latency_tracing ( p ) ;
2005-04-17 02:20:36 +04:00
/* ok, now we should be set up.. */
2013-07-04 02:08:32 +04:00
p - > pid = pid_nr ( pid ) ;
if ( clone_flags & CLONE_THREAD ) {
p - > group_leader = current - > group_leader ;
p - > tgid = current - > tgid ;
} else {
p - > group_leader = p ;
p - > tgid = p - > pid ;
}
2012-03-14 22:55:38 +04:00
writeback: per task dirty rate limit
Add two fields to task_struct.
1) account dirtied pages in the individual tasks, for accuracy
2) per-task balance_dirty_pages() call intervals, for flexibility
The balance_dirty_pages() call interval (ie. nr_dirtied_pause) will
scale near-sqrt to the safety gap between dirty pages and threshold.
The main problem of per-task nr_dirtied is, if 1k+ tasks start dirtying
pages at exactly the same time, each task will be assigned a large
initial nr_dirtied_pause, so that the dirty threshold will be exceeded
long before each task reached its nr_dirtied_pause and hence call
balance_dirty_pages().
The solution is to watch for the number of pages dirtied on each CPU in
between the calls into balance_dirty_pages(). If it exceeds ratelimit_pages
(3% dirty threshold), force call balance_dirty_pages() for a chance to
set bdi->dirty_exceeded. In normal situations, this safeguarding
condition is not expected to trigger at all.
On the sqrt in dirty_poll_interval():
It will serve as an initial guess when dirty pages are still in the
freerun area.
When dirty pages are floating inside the dirty control scope [freerun,
limit], a followup patch will use some refined dirty poll interval to
get the desired pause time.
thresh-dirty (MB) sqrt
1 16
2 22
4 32
8 45
16 64
32 90
64 128
128 181
256 256
512 362
1024 512
The above table means, given 1MB (or 1GB) gap and the dd tasks polling
balance_dirty_pages() on every 16 (or 512) pages, the dirty limit won't
be exceeded as long as there are less than 16 (or 512) concurrent dd's.
So sqrt naturally leads to less overheads and more safe concurrent tasks
for large memory servers, which have large (thresh-freerun) gaps.
peter: keep the per-CPU ratelimit for safeguarding the 1k+ tasks case
CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: Andrea Righi <andrea@betterlinux.com>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2011-06-12 04:10:12 +04:00
p - > nr_dirtied = 0 ;
p - > nr_dirtied_pause = 128 > > ( PAGE_SHIFT - 10 ) ;
2011-06-12 05:25:42 +04:00
p - > dirty_paused_when = 0 ;
writeback: per task dirty rate limit
Add two fields to task_struct.
1) account dirtied pages in the individual tasks, for accuracy
2) per-task balance_dirty_pages() call intervals, for flexibility
The balance_dirty_pages() call interval (ie. nr_dirtied_pause) will
scale near-sqrt to the safety gap between dirty pages and threshold.
The main problem of per-task nr_dirtied is, if 1k+ tasks start dirtying
pages at exactly the same time, each task will be assigned a large
initial nr_dirtied_pause, so that the dirty threshold will be exceeded
long before each task reached its nr_dirtied_pause and hence call
balance_dirty_pages().
The solution is to watch for the number of pages dirtied on each CPU in
between the calls into balance_dirty_pages(). If it exceeds ratelimit_pages
(3% dirty threshold), force call balance_dirty_pages() for a chance to
set bdi->dirty_exceeded. In normal situations, this safeguarding
condition is not expected to trigger at all.
On the sqrt in dirty_poll_interval():
It will serve as an initial guess when dirty pages are still in the
freerun area.
When dirty pages are floating inside the dirty control scope [freerun,
limit], a followup patch will use some refined dirty poll interval to
get the desired pause time.
thresh-dirty (MB) sqrt
1 16
2 22
4 32
8 45
16 64
32 90
64 128
128 181
256 256
512 362
1024 512
The above table means, given 1MB (or 1GB) gap and the dd tasks polling
balance_dirty_pages() on every 16 (or 512) pages, the dirty limit won't
be exceeded as long as there are less than 16 (or 512) concurrent dd's.
So sqrt naturally leads to less overheads and more safe concurrent tasks
for large memory servers, which have large (thresh-freerun) gaps.
peter: keep the per-CPU ratelimit for safeguarding the 1k+ tasks case
CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: Andrea Righi <andrea@betterlinux.com>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2011-06-12 04:10:12 +04:00
2013-11-13 18:36:12 +04:00
p - > pdeath_signal = 0 ;
2006-03-29 04:11:25 +04:00
INIT_LIST_HEAD ( & p - > thread_group ) ;
2012-06-27 09:24:13 +04:00
p - > task_works = NULL ;
2005-04-17 02:20:36 +04:00
2020-08-29 16:03:24 +03:00
# ifdef CONFIG_KRETPROBES
p - > kretprobe_instances . first = NULL ;
# endif
2015-06-09 14:32:09 +03:00
/*
* Ensure that the cgroup subsystem policies allow the new process to be
2020-10-16 06:10:28 +03:00
* forked . It should be noted that the new process ' s css_set can be changed
2015-06-09 14:32:09 +03:00
* between here and cgroup_post_fork ( ) if an organisation operation is in
* progress .
*/
2020-02-05 16:26:22 +03:00
retval = cgroup_can_fork ( p , args ) ;
2015-06-09 14:32:09 +03:00
if ( retval )
2020-02-05 16:26:20 +03:00
goto bad_fork_put_pidfd ;
2015-06-09 14:32:09 +03:00
fork: record start_time late
This changes the fork(2) syscall to record the process start_time after
initializing the basic task structure but still before making the new
process visible to user-space.
Technically, we could record the start_time anytime during fork(2). But
this might lead to scenarios where a start_time is recorded long before
a process becomes visible to user-space. For instance, with
userfaultfd(2) and TLS, user-space can delay the execution of fork(2)
for an indefinite amount of time (and will, if this causes network
access, or similar).
By recording the start_time late, it much closer reflects the point in
time where the process becomes live and can be observed by other
processes.
Lastly, this makes it much harder for user-space to predict and control
the start_time they get assigned. Previously, user-space could fork a
process and stall it in copy_thread_tls() before its pid is allocated,
but after its start_time is recorded. This can be misused to later-on
cycle through PIDs and resume the stalled fork(2) yielding a process
that has the same pid and start_time as a process that existed before.
This can be used to circumvent security systems that identify processes
by their pid+start_time combination.
Even though user-space was always aware that start_time recording is
flaky (but several projects are known to still rely on start_time-based
identification), changing the start_time to be recorded late will help
mitigate existing attacks and make it much harder for user-space to
control the start_time a process gets assigned.
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Tom Gundersen <teg@jklm.no>
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-01-08 15:58:52 +03:00
/*
* From this point on we must avoid any synchronous user - space
* communication until we take the tasklist - lock . In particular , we do
* not want user - space to be able to predict the process start - time by
* stalling fork ( 2 ) after we recorded the start_time but before it is
* visible to the system .
*/
p - > start_time = ktime_get_ns ( ) ;
2019-11-07 13:07:58 +03:00
p - > start_boottime = ktime_get_boottime_ns ( ) ;
fork: record start_time late
This changes the fork(2) syscall to record the process start_time after
initializing the basic task structure but still before making the new
process visible to user-space.
Technically, we could record the start_time anytime during fork(2). But
this might lead to scenarios where a start_time is recorded long before
a process becomes visible to user-space. For instance, with
userfaultfd(2) and TLS, user-space can delay the execution of fork(2)
for an indefinite amount of time (and will, if this causes network
access, or similar).
By recording the start_time late, it much closer reflects the point in
time where the process becomes live and can be observed by other
processes.
Lastly, this makes it much harder for user-space to predict and control
the start_time they get assigned. Previously, user-space could fork a
process and stall it in copy_thread_tls() before its pid is allocated,
but after its start_time is recorded. This can be misused to later-on
cycle through PIDs and resume the stalled fork(2) yielding a process
that has the same pid and start_time as a process that existed before.
This can be used to circumvent security systems that identify processes
by their pid+start_time combination.
Even though user-space was always aware that start_time recording is
flaky (but several projects are known to still rely on start_time-based
identification), changing the start_time to be recorded late will help
mitigate existing attacks and make it much harder for user-space to
control the start_time a process gets assigned.
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Tom Gundersen <teg@jklm.no>
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-01-08 15:58:52 +03:00
2013-07-04 02:08:32 +04:00
/*
* Make it visible to the rest of the system , but dont wake it up yet .
* Need tasklist lock for parent etc handling !
*/
2005-04-17 02:20:36 +04:00
write_lock_irq ( & tasklist_lock ) ;
/* CLONE_PARENT re-uses the old parent */
2009-03-03 00:58:45 +03:00
if ( clone_flags & ( CLONE_PARENT | CLONE_THREAD ) ) {
2005-04-17 02:20:36 +04:00
p - > real_parent = current - > real_parent ;
2009-03-03 00:58:45 +03:00
p - > parent_exec_id = current - > parent_exec_id ;
2020-11-07 09:47:22 +03:00
if ( clone_flags & CLONE_THREAD )
p - > exit_signal = - 1 ;
else
p - > exit_signal = current - > group_leader - > exit_signal ;
2009-03-03 00:58:45 +03:00
} else {
2005-04-17 02:20:36 +04:00
p - > real_parent = current ;
2009-03-03 00:58:45 +03:00
p - > parent_exec_id = current - > self_exec_id ;
2020-11-07 09:47:22 +03:00
p - > exit_signal = args - > exit_signal ;
2009-03-03 00:58:45 +03:00
}
2005-04-17 02:20:36 +04:00
livepatch: change to a per-task consistency model
Change livepatch to use a basic per-task consistency model. This is the
foundation which will eventually enable us to patch those ~10% of
security patches which change function or data semantics. This is the
biggest remaining piece needed to make livepatch more generally useful.
This code stems from the design proposal made by Vojtech [1] in November
2014. It's a hybrid of kGraft and kpatch: it uses kGraft's per-task
consistency and syscall barrier switching combined with kpatch's stack
trace switching. There are also a number of fallback options which make
it quite flexible.
Patches are applied on a per-task basis, when the task is deemed safe to
switch over. When a patch is enabled, livepatch enters into a
transition state where tasks are converging to the patched state.
Usually this transition state can complete in a few seconds. The same
sequence occurs when a patch is disabled, except the tasks converge from
the patched state to the unpatched state.
An interrupt handler inherits the patched state of the task it
interrupts. The same is true for forked tasks: the child inherits the
patched state of the parent.
Livepatch uses several complementary approaches to determine when it's
safe to patch tasks:
1. The first and most effective approach is stack checking of sleeping
tasks. If no affected functions are on the stack of a given task,
the task is patched. In most cases this will patch most or all of
the tasks on the first try. Otherwise it'll keep trying
periodically. This option is only available if the architecture has
reliable stacks (HAVE_RELIABLE_STACKTRACE).
2. The second approach, if needed, is kernel exit switching. A
task is switched when it returns to user space from a system call, a
user space IRQ, or a signal. It's useful in the following cases:
a) Patching I/O-bound user tasks which are sleeping on an affected
function. In this case you have to send SIGSTOP and SIGCONT to
force it to exit the kernel and be patched.
b) Patching CPU-bound user tasks. If the task is highly CPU-bound
then it will get patched the next time it gets interrupted by an
IRQ.
c) In the future it could be useful for applying patches for
architectures which don't yet have HAVE_RELIABLE_STACKTRACE. In
this case you would have to signal most of the tasks on the
system. However this isn't supported yet because there's
currently no way to patch kthreads without
HAVE_RELIABLE_STACKTRACE.
3. For idle "swapper" tasks, since they don't ever exit the kernel, they
instead have a klp_update_patch_state() call in the idle loop which
allows them to be patched before the CPU enters the idle state.
(Note there's not yet such an approach for kthreads.)
All the above approaches may be skipped by setting the 'immediate' flag
in the 'klp_patch' struct, which will disable per-task consistency and
patch all tasks immediately. This can be useful if the patch doesn't
change any function or data semantics. Note that, even with this flag
set, it's possible that some tasks may still be running with an old
version of the function, until that function returns.
There's also an 'immediate' flag in the 'klp_func' struct which allows
you to specify that certain functions in the patch can be applied
without per-task consistency. This might be useful if you want to patch
a common function like schedule(), and the function change doesn't need
consistency but the rest of the patch does.
For architectures which don't have HAVE_RELIABLE_STACKTRACE, the user
must set patch->immediate which causes all tasks to be patched
immediately. This option should be used with care, only when the patch
doesn't change any function or data semantics.
In the future, architectures which don't have HAVE_RELIABLE_STACKTRACE
may be allowed to use per-task consistency if we can come up with
another way to patch kthreads.
The /sys/kernel/livepatch/<patch>/transition file shows whether a patch
is in transition. Only a single patch (the topmost patch on the stack)
can be in transition at a given time. A patch can remain in transition
indefinitely, if any of the tasks are stuck in the initial patch state.
A transition can be reversed and effectively canceled by writing the
opposite value to the /sys/kernel/livepatch/<patch>/enabled file while
the transition is in progress. Then all the tasks will attempt to
converge back to the original patch state.
[1] https://lkml.kernel.org/r/20141107140458.GA21774@suse.cz
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Acked-by: Miroslav Benes <mbenes@suse.cz>
Acked-by: Ingo Molnar <mingo@kernel.org> # for the scheduler changes
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2017-02-14 04:42:40 +03:00
klp_copy_process ( p ) ;
2021-03-29 16:18:35 +03:00
sched_core_fork ( p ) ;
2006-02-15 22:13:24 +03:00
spin_lock ( & current - > sighand - > siglock ) ;
2006-03-29 04:11:26 +04:00
2014-06-28 02:18:48 +04:00
/*
* Copy seccomp details explicitly here , in case they were changed
* before holding sighand lock .
*/
copy_seccomp ( p ) ;
rseq: Introduce restartable sequences system call
Expose a new system call allowing each thread to register one userspace
memory area to be used as an ABI between kernel and user-space for two
purposes: user-space restartable sequences and quick access to read the
current CPU number value from user-space.
* Restartable sequences (per-cpu atomics)
Restartables sequences allow user-space to perform update operations on
per-cpu data without requiring heavy-weight atomic operations.
The restartable critical sections (percpu atomics) work has been started
by Paul Turner and Andrew Hunter. It lets the kernel handle restart of
critical sections. [1] [2] The re-implementation proposed here brings a
few simplifications to the ABI which facilitates porting to other
architectures and speeds up the user-space fast path.
Here are benchmarks of various rseq use-cases.
Test hardware:
arm32: ARMv7 Processor rev 4 (v7l) "Cubietruck", 2-core
x86-64: Intel E5-2630 v3@2.40GHz, 16-core, hyperthreading
The following benchmarks were all performed on a single thread.
* Per-CPU statistic counter increment
getcpu+atomic (ns/op) rseq (ns/op) speedup
arm32: 344.0 31.4 11.0
x86-64: 15.3 2.0 7.7
* LTTng-UST: write event 32-bit header, 32-bit payload into tracer
per-cpu buffer
getcpu+atomic (ns/op) rseq (ns/op) speedup
arm32: 2502.0 2250.0 1.1
x86-64: 117.4 98.0 1.2
* liburcu percpu: lock-unlock pair, dereference, read/compare word
getcpu+atomic (ns/op) rseq (ns/op) speedup
arm32: 751.0 128.5 5.8
x86-64: 53.4 28.6 1.9
* jemalloc memory allocator adapted to use rseq
Using rseq with per-cpu memory pools in jemalloc at Facebook (based on
rseq 2016 implementation):
The production workload response-time has 1-2% gain avg. latency, and
the P99 overall latency drops by 2-3%.
* Reading the current CPU number
Speeding up reading the current CPU number on which the caller thread is
running is done by keeping the current CPU number up do date within the
cpu_id field of the memory area registered by the thread. This is done
by making scheduler preemption set the TIF_NOTIFY_RESUME flag on the
current thread. Upon return to user-space, a notify-resume handler
updates the current CPU value within the registered user-space memory
area. User-space can then read the current CPU number directly from
memory.
Keeping the current cpu id in a memory area shared between kernel and
user-space is an improvement over current mechanisms available to read
the current CPU number, which has the following benefits over
alternative approaches:
- 35x speedup on ARM vs system call through glibc
- 20x speedup on x86 compared to calling glibc, which calls vdso
executing a "lsl" instruction,
- 14x speedup on x86 compared to inlined "lsl" instruction,
- Unlike vdso approaches, this cpu_id value can be read from an inline
assembly, which makes it a useful building block for restartable
sequences.
- The approach of reading the cpu id through memory mapping shared
between kernel and user-space is portable (e.g. ARM), which is not the
case for the lsl-based x86 vdso.
On x86, yet another possible approach would be to use the gs segment
selector to point to user-space per-cpu data. This approach performs
similarly to the cpu id cache, but it has two disadvantages: it is
not portable, and it is incompatible with existing applications already
using the gs segment selector for other purposes.
Benchmarking various approaches for reading the current CPU number:
ARMv7 Processor rev 4 (v7l)
Machine model: Cubietruck
- Baseline (empty loop): 8.4 ns
- Read CPU from rseq cpu_id: 16.7 ns
- Read CPU from rseq cpu_id (lazy register): 19.8 ns
- glibc 2.19-0ubuntu6.6 getcpu: 301.8 ns
- getcpu system call: 234.9 ns
x86-64 Intel(R) Xeon(R) CPU E5-2630 v3 @ 2.40GHz:
- Baseline (empty loop): 0.8 ns
- Read CPU from rseq cpu_id: 0.8 ns
- Read CPU from rseq cpu_id (lazy register): 0.8 ns
- Read using gs segment selector: 0.8 ns
- "lsl" inline assembly: 13.0 ns
- glibc 2.19-0ubuntu6 getcpu: 16.6 ns
- getcpu system call: 53.9 ns
- Speed (benchmark taken on v8 of patchset)
Running 10 runs of hackbench -l 100000 seems to indicate, contrary to
expectations, that enabling CONFIG_RSEQ slightly accelerates the
scheduler:
Configuration: 2 sockets * 8-core Intel(R) Xeon(R) CPU E5-2630 v3 @
2.40GHz (directly on hardware, hyperthreading disabled in BIOS, energy
saving disabled in BIOS, turboboost disabled in BIOS, cpuidle.off=1
kernel parameter), with a Linux v4.6 defconfig+localyesconfig,
restartable sequences series applied.
* CONFIG_RSEQ=n
avg.: 41.37 s
std.dev.: 0.36 s
* CONFIG_RSEQ=y
avg.: 40.46 s
std.dev.: 0.33 s
- Size
On x86-64, between CONFIG_RSEQ=n/y, the text size increase of vmlinux is
567 bytes, and the data size increase of vmlinux is 5696 bytes.
[1] https://lwn.net/Articles/650333/
[2] http://www.linuxplumbersconf.org/2013/ocw/system/presentations/1695/original/LPC%20-%20PerCpu%20Atomics.pdf
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Watson <davejwatson@fb.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: "H . Peter Anvin" <hpa@zytor.com>
Cc: Chris Lameter <cl@linux.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Andrew Hunter <ahh@google.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: "Paul E . McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Paul Turner <pjt@google.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Ben Maurer <bmaurer@fb.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: linux-api@vger.kernel.org
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/20151027235635.16059.11630.stgit@pjt-glaptop.roam.corp.google.com
Link: http://lkml.kernel.org/r/20150624222609.6116.86035.stgit@kitami.mtv.corp.google.com
Link: https://lkml.kernel.org/r/20180602124408.8430-3-mathieu.desnoyers@efficios.com
2018-06-02 15:43:54 +03:00
rseq_fork ( p , clone_flags ) ;
2018-07-13 23:30:33 +03:00
/* Don't start children in a dying pid namespace */
2017-11-18 02:30:34 +03:00
if ( unlikely ( ! ( ns_of_pid ( pid ) - > pid_allocated & PIDNS_ADDING ) ) ) {
pid_ns: Fix race between setns'ed fork() and zap_pid_ns_processes()
Imagine we have a pid namespace and a task from its parent's pid_ns,
which made setns() to the pid namespace. The task is doing fork(),
while the pid namespace's child reaper is dying. We have the race
between them:
Task from parent pid_ns Child reaper
copy_process() ..
alloc_pid() ..
.. zap_pid_ns_processes()
.. disable_pid_allocation()
.. read_lock(&tasklist_lock)
.. iterate over pids in pid_ns
.. kill tasks linked to pids
.. read_unlock(&tasklist_lock)
write_lock_irq(&tasklist_lock); ..
attach_pid(p, PIDTYPE_PID); ..
.. ..
So, just created task p won't receive SIGKILL signal,
and the pid namespace will be in contradictory state.
Only manual kill will help there, but does the userspace
care about this? I suppose, the most users just inject
a task into a pid namespace and wait a SIGCHLD from it.
The patch fixes the problem. It simply checks for
(pid_ns->nr_hashed & PIDNS_HASH_ADDING) in copy_process().
We do it under the tasklist_lock, and can't skip
PIDNS_HASH_ADDING as noted by Oleg:
"zap_pid_ns_processes() does disable_pid_allocation()
and then takes tasklist_lock to kill the whole namespace.
Given that copy_process() checks PIDNS_HASH_ADDING
under write_lock(tasklist) they can't race;
if copy_process() takes this lock first, the new child will
be killed, otherwise copy_process() can't miss
the change in ->nr_hashed."
If allocation is disabled, we just return -ENOMEM
like it's made for such cases in alloc_pid().
v2: Do not move disable_pid_allocation(), do not
introduce a new variable in copy_process() and simplify
the patch as suggested by Oleg Nesterov.
Account the problem with double irq enabling
found by Eric W. Biederman.
Fixes: c876ad768215 ("pidns: Stop pid allocation when init dies")
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: Ingo Molnar <mingo@kernel.org>
CC: Peter Zijlstra <peterz@infradead.org>
CC: Oleg Nesterov <oleg@redhat.com>
CC: Mike Rapoport <rppt@linux.vnet.ibm.com>
CC: Michal Hocko <mhocko@suse.com>
CC: Andy Lutomirski <luto@kernel.org>
CC: "Eric W. Biederman" <ebiederm@xmission.com>
CC: Andrei Vagin <avagin@openvz.org>
CC: Cyrill Gorcunov <gorcunov@openvz.org>
CC: Serge Hallyn <serge@hallyn.com>
Cc: stable@vger.kernel.org
Acked-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
2017-05-12 19:11:31 +03:00
retval = - ENOMEM ;
goto bad_fork_cancel_cgroup ;
}
2006-03-29 04:11:26 +04:00
2018-07-23 16:01:10 +03:00
/* Let kill terminate clone/fork in the middle */
if ( fatal_signal_pending ( current ) ) {
retval = - EINTR ;
goto bad_fork_cancel_cgroup ;
}
2019-06-27 05:22:09 +03:00
/* past the last point of failure */
if ( pidfile )
fd_install ( pidfd , pidfile ) ;
2006-03-29 04:11:26 +04:00
2017-09-26 21:06:43 +03:00
init_task_pid_links ( p ) ;
2006-03-29 04:11:07 +04:00
if ( likely ( p - > pid ) ) {
2011-06-17 18:50:38 +04:00
ptrace_init_task ( p , ( clone_flags & CLONE_PTRACE ) | | trace ) ;
2006-03-29 04:11:07 +04:00
2013-07-04 02:08:31 +04:00
init_task_pid ( p , PIDTYPE_PID , pid ) ;
2006-03-29 04:11:07 +04:00
if ( thread_group_leader ( p ) ) {
2017-06-04 12:32:13 +03:00
init_task_pid ( p , PIDTYPE_TGID , pid ) ;
2013-07-04 02:08:31 +04:00
init_task_pid ( p , PIDTYPE_PGID , task_pgrp ( current ) ) ;
init_task_pid ( p , PIDTYPE_SID , task_session ( current ) ) ;
2010-07-13 04:10:36 +04:00
if ( is_child_reaper ( pid ) ) {
2010-03-03 01:51:53 +03:00
ns_of_pid ( pid ) - > child_reaper = p ;
2010-07-13 04:10:36 +04:00
p - > signal - > flags | = SIGNAL_UNKILLABLE ;
}
2018-07-23 23:20:37 +03:00
p - > signal - > shared_pending . signal = delayed . signal ;
2008-10-13 13:37:26 +04:00
p - > signal - > tty = tty_kref_get ( current - > signal - > tty ) ;
prctl: propagate has_child_subreaper flag to every descendant
If process forks some children when it has is_child_subreaper
flag enabled they will inherit has_child_subreaper flag - first
group, when is_child_subreaper is disabled forked children will
not inherit it - second group. So child-subreaper does not reparent
all his descendants when their parents die. Having these two
differently behaving groups can lead to confusion. Also it is
a problem for CRIU, as when we restore process tree we need to
somehow determine which descendants belong to which group and
much harder - to put them exactly to these group.
To simplify these we can add a propagation of has_child_subreaper
flag on PR_SET_CHILD_SUBREAPER, walking all descendants of child-
subreaper to setup has_child_subreaper flag.
In common cases when process like systemd first sets itself to
be a child-subreaper and only after that forks its services, we will
have zero-length list of descendants to walk. Testing with binary
subtree of 2^15 processes prctl took < 0.007 sec and has shown close
to linear dependency(~0.2 * n * usec) on lower numbers of processes.
Moreover, I doubt someone intentionaly pre-forks the children whitch
should reparent to init before becoming subreaper, because some our
ancestor migh have had is_child_subreaper flag while forking our
sub-tree and our childs will all inherit has_child_subreaper flag,
and we have no way to influence it. And only way to check if we have
no has_child_subreaper flag is to create some childs, kill them and
see where they will reparent to.
Using walk_process_tree helper to walk subtree, thanks to Oleg! Timing
seems to be the same.
Optimize:
a) When descendant already has has_child_subreaper flag all his subtree
has it too already.
* for a) to be true need to move has_child_subreaper inheritance under
the same tasklist_lock with adding task to its ->real_parent->children
as without it process can inherit zero has_child_subreaper, then we
set 1 to it's parent flag, check that parent has no more children, and
only after child with wrong flag is added to the tree.
* Also make these inheritance more clear by using real_parent instead of
current, as on clone(CLONE_PARENT) if current has is_child_subreaper
and real_parent has no is_child_subreaper or has_child_subreaper, child
will have has_child_subreaper flag set without actually having a
subreaper in it's ancestors.
b) When some descendant is child_reaper, it's subtree is in different
pidns from us(original child-subreaper) and processes from other pidns
will never reparent to us.
So we can skip their(a,b) subtree from walk.
v2: switch to walk_process_tree() general helper, move
has_child_subreaper inheritance
v3: remove csr_descendant leftover, change current to real_parent
in has_child_subreaper inheritance
v4: small commit message fix
Fixes: ebec18a6d3aa ("prctl: add PR_{SET,GET}_CHILD_SUBREAPER to allow simple process supervision")
Signed-off-by: Pavel Tikhomirov <ptikhomirov@virtuozzo.com>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
2017-01-30 18:06:12 +03:00
/*
* Inherit has_child_subreaper flag under the same
* tasklist_lock with adding child to the process tree
* for propagate_has_child_subreaper optimization .
*/
p - > signal - > has_child_subreaper = p - > real_parent - > signal - > has_child_subreaper | |
p - > real_parent - > signal - > is_child_subreaper ;
2009-12-18 02:27:15 +03:00
list_add_tail ( & p - > sibling , & p - > real_parent - > children ) ;
2006-04-19 09:20:16 +04:00
list_add_tail_rcu ( & p - > tasks , & init_task . tasks ) ;
2017-06-04 12:32:13 +03:00
attach_pid ( p , PIDTYPE_TGID ) ;
2013-07-04 02:08:31 +04:00
attach_pid ( p , PIDTYPE_PGID ) ;
attach_pid ( p , PIDTYPE_SID ) ;
2010-12-08 18:22:55 +03:00
__this_cpu_inc ( process_counts ) ;
2013-07-04 02:08:30 +04:00
} else {
current - > signal - > nr_threads + + ;
atomic_inc ( & current - > signal - > live ) ;
2019-01-18 15:27:27 +03:00
refcount_inc ( & current - > signal - > sigcnt ) ;
2018-07-23 21:38:00 +03:00
task_join_group_stop ( p ) ;
2013-07-04 02:08:30 +04:00
list_add_tail_rcu ( & p - > thread_group ,
& p - > group_leader - > thread_group ) ;
introduce for_each_thread() to replace the buggy while_each_thread()
while_each_thread() and next_thread() should die, almost every lockless
usage is wrong.
1. Unless g == current, the lockless while_each_thread() is not safe.
while_each_thread(g, t) can loop forever if g exits, next_thread()
can't reach the unhashed thread in this case. Note that this can
happen even if g is the group leader, it can exec.
2. Even if while_each_thread() itself was correct, people often use
it wrongly.
It was never safe to just take rcu_read_lock() and loop unless
you verify that pid_alive(g) == T, even the first next_thread()
can point to the already freed/reused memory.
This patch adds signal_struct->thread_head and task->thread_node to
create the normal rcu-safe list with the stable head. The new
for_each_thread(g, t) helper is always safe under rcu_read_lock() as
long as this task_struct can't go away.
Note: of course it is ugly to have both task_struct->thread_node and the
old task_struct->thread_group, we will kill it later, after we change
the users of while_each_thread() to use for_each_thread().
Perhaps we can kill it even before we convert all users, we can
reimplement next_thread(t) using the new thread_head/thread_node. But
we can't do this right now because this will lead to subtle behavioural
changes. For example, do/while_each_thread() always sees at least one
task, while for_each_thread() can do nothing if the whole thread group
has died. Or thread_group_empty(), currently its semantics is not clear
unless thread_group_leader(p) and we need to audit the callers before we
can change it.
So this patch adds the new interface which has to coexist with the old
one for some time, hopefully the next changes will be more or less
straightforward and the old one will go away soon.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Sergey Dyasly <dserrg@gmail.com>
Tested-by: Sergey Dyasly <dserrg@gmail.com>
Reviewed-by: Sameer Nanda <snanda@chromium.org>
Acked-by: David Rientjes <rientjes@google.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mandeep Singh Baines <msb@chromium.org>
Cc: "Ma, Xindong" <xindong.ma@intel.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: "Tu, Xiaobing" <xiaobing.tu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-01-22 03:49:56 +04:00
list_add_tail_rcu ( & p - > thread_node ,
& p - > signal - > thread_head ) ;
2006-03-29 04:11:07 +04:00
}
2013-07-04 02:08:31 +04:00
attach_pid ( p , PIDTYPE_PID ) ;
2006-03-29 04:11:07 +04:00
nr_threads + + ;
2005-04-17 02:20:36 +04:00
}
total_forks + + ;
2018-07-23 23:20:37 +03:00
hlist_del_init ( & delayed . node ) ;
2006-02-15 22:13:24 +03:00
spin_unlock ( & current - > sighand - > siglock ) ;
2014-04-13 22:58:54 +04:00
syscall_tracepoint_update ( p ) ;
2005-04-17 02:20:36 +04:00
write_unlock_irq ( & tasklist_lock ) ;
2014-04-13 22:58:54 +04:00
2005-11-29 00:43:48 +03:00
proc_fork_connector ( p ) ;
sched/uclamp: Add a new sysctl to control RT default boost value
RT tasks by default run at the highest capacity/performance level. When
uclamp is selected this default behavior is retained by enforcing the
requested uclamp.min (p->uclamp_req[UCLAMP_MIN]) of the RT tasks to be
uclamp_none(UCLAMP_MAX), which is SCHED_CAPACITY_SCALE; the maximum
value.
This is also referred to as 'the default boost value of RT tasks'.
See commit 1a00d999971c ("sched/uclamp: Set default clamps for RT tasks").
On battery powered devices, it is desired to control this default
(currently hardcoded) behavior at runtime to reduce energy consumed by
RT tasks.
For example, a mobile device manufacturer where big.LITTLE architecture
is dominant, the performance of the little cores varies across SoCs, and
on high end ones the big cores could be too power hungry.
Given the diversity of SoCs, the new knob allows manufactures to tune
the best performance/power for RT tasks for the particular hardware they
run on.
They could opt to further tune the value when the user selects
a different power saving mode or when the device is actively charging.
The runtime aspect of it further helps in creating a single kernel image
that can be run on multiple devices that require different tuning.
Keep in mind that a lot of RT tasks in the system are created by the
kernel. On Android for instance I can see over 50 RT tasks, only
a handful of which created by the Android framework.
To control the default behavior globally by system admins and device
integrator, introduce the new sysctl_sched_uclamp_util_min_rt_default
to change the default boost value of the RT tasks.
I anticipate this to be mostly in the form of modifying the init script
of a particular device.
To avoid polluting the fast path with unnecessary code, the approach
taken is to synchronously do the update by traversing all the existing
tasks in the system. This could race with a concurrent fork(), which is
dealt with by introducing sched_post_fork() function which will ensure
the racy fork will get the right update applied.
Tested on Juno-r2 in combination with the RT capacity awareness [1].
By default an RT task will go to the highest capacity CPU and run at the
maximum frequency, which is particularly energy inefficient on high end
mobile devices because the biggest core[s] are 'huge' and power hungry.
With this patch the RT task can be controlled to run anywhere by
default, and doesn't cause the frequency to be maximum all the time.
Yet any task that really needs to be boosted can easily escape this
default behavior by modifying its requested uclamp.min value
(p->uclamp_req[UCLAMP_MIN]) via sched_setattr() syscall.
[1] 804d402fb6f6: ("sched/rt: Make RT capacity-aware")
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200716110347.19553-2-qais.yousef@arm.com
2020-07-16 14:03:45 +03:00
sched_post_fork ( p ) ;
2020-02-05 16:26:22 +03:00
cgroup_post_fork ( p , args ) ;
perf: Do the big rename: Performance Counters -> Performance Events
Bye-bye Performance Counters, welcome Performance Events!
In the past few months the perfcounters subsystem has grown out its
initial role of counting hardware events, and has become (and is
becoming) a much broader generic event enumeration, reporting, logging,
monitoring, analysis facility.
Naming its core object 'perf_counter' and naming the subsystem
'perfcounters' has become more and more of a misnomer. With pending
code like hw-breakpoints support the 'counter' name is less and
less appropriate.
All in one, we've decided to rename the subsystem to 'performance
events' and to propagate this rename through all fields, variables
and API names. (in an ABI compatible fashion)
The word 'event' is also a bit shorter than 'counter' - which makes
it slightly more convenient to write/handle as well.
Thanks goes to Stephane Eranian who first observed this misnomer and
suggested a rename.
User-space tooling and ABI compatibility is not affected - this patch
should be function-invariant. (Also, defconfigs were not touched to
keep the size down.)
This patch has been generated via the following script:
FILES=$(find * -type f | grep -vE 'oprofile|[^K]config')
sed -i \
-e 's/PERF_EVENT_/PERF_RECORD_/g' \
-e 's/PERF_COUNTER/PERF_EVENT/g' \
-e 's/perf_counter/perf_event/g' \
-e 's/nb_counters/nb_events/g' \
-e 's/swcounter/swevent/g' \
-e 's/tpcounter_event/tp_event/g' \
$FILES
for N in $(find . -name perf_counter.[ch]); do
M=$(echo $N | sed 's/perf_counter/perf_event/g')
mv $N $M
done
FILES=$(find . -name perf_event.*)
sed -i \
-e 's/COUNTER_MASK/REG_MASK/g' \
-e 's/COUNTER/EVENT/g' \
-e 's/\<event\>/event_id/g' \
-e 's/counter/event/g' \
-e 's/Counter/Event/g' \
$FILES
... to keep it as correct as possible. This script can also be
used by anyone who has pending perfcounters patches - it converts
a Linux kernel tree over to the new naming. We tried to time this
change to the point in time where the amount of pending patches
is the smallest: the end of the merge window.
Namespace clashes were fixed up in a preparatory patch - and some
stylistic fallout will be fixed up in a subsequent patch.
( NOTE: 'counters' are still the proper terminology when we deal
with hardware registers - and these sed scripts are a bit
over-eager in renaming them. I've undone some of that, but
in case there's something left where 'counter' would be
better than 'event' we can undo that on an individual basis
instead of touching an otherwise nicely automated patch. )
Suggested-by: Stephane Eranian <eranian@google.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Paul Mackerras <paulus@samba.org>
Reviewed-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <linux-arch@vger.kernel.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-09-21 14:02:48 +04:00
perf_event_fork ( p ) ;
2012-01-11 03:08:09 +04:00
trace_task_newtask ( p , clone_flags ) ;
2013-10-16 21:39:37 +04:00
uprobe_copy_process ( p , clone_flags ) ;
2012-01-11 03:08:09 +04:00
mm, oom_adj: don't loop through tasks in __set_oom_adj when not necessary
Currently __set_oom_adj loops through all processes in the system to keep
oom_score_adj and oom_score_adj_min in sync between processes sharing
their mm. This is done for any task with more that one mm_users, which
includes processes with multiple threads (sharing mm and signals).
However for such processes the loop is unnecessary because their signal
structure is shared as well.
Android updates oom_score_adj whenever a tasks changes its role
(background/foreground/...) or binds to/unbinds from a service, making it
more/less important. Such operation can happen frequently. We noticed
that updates to oom_score_adj became more expensive and after further
investigation found out that the patch mentioned in "Fixes" introduced a
regression. Using Pixel 4 with a typical Android workload, write time to
oom_score_adj increased from ~3.57us to ~362us. Moreover this regression
linearly depends on the number of multi-threaded processes running on the
system.
Mark the mm with a new MMF_MULTIPROCESS flag bit when task is created with
(CLONE_VM && !CLONE_THREAD && !CLONE_VFORK). Change __set_oom_adj to use
MMF_MULTIPROCESS instead of mm_users to decide whether oom_score_adj
update should be synchronized between multiple processes. To prevent
races between clone() and __set_oom_adj(), when oom_score_adj of the
process being cloned might be modified from userspace, we use
oom_adj_mutex. Its scope is changed to global.
The combination of (CLONE_VM && !CLONE_THREAD) is rarely used except for
the case of vfork(). To prevent performance regressions of vfork(), we
skip taking oom_adj_mutex and setting MMF_MULTIPROCESS when CLONE_VFORK is
specified. Clearing the MMF_MULTIPROCESS flag (when the last process
sharing the mm exits) is left out of this patch to keep it simple and
because it is believed that this threading model is rare. Should there
ever be a need for optimizing that case as well, it can be done by hooking
into the exit path, likely following the mm_update_next_owner pattern.
With the combination of (CLONE_VM && !CLONE_THREAD && !CLONE_VFORK) being
quite rare, the regression is gone after the change is applied.
[surenb@google.com: v3]
Link: https://lkml.kernel.org/r/20200902012558.2335613-1-surenb@google.com
Fixes: 44a70adec910 ("mm, oom_adj: make sure processes sharing mm have same view of oom_score_adj")
Reported-by: Tim Murray <timmurray@google.com>
Suggested-by: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Christian Brauner <christian.brauner@ubuntu.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Eugene Syromiatnikov <esyr@redhat.com>
Cc: Christian Kellner <christian@kellner.me>
Cc: Adrian Reber <areber@redhat.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Aleksa Sarai <cyphar@cyphar.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Alexey Gladkov <gladkov.alexey@gmail.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Daniel Jordan <daniel.m.jordan@oracle.com>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Bernd Edlinger <bernd.edlinger@hotmail.de>
Cc: John Johansen <john.johansen@canonical.com>
Cc: Yafang Shao <laoar.shao@gmail.com>
Link: https://lkml.kernel.org/r/20200824153036.3201505-1-surenb@google.com
Debugged-by: Minchan Kim <minchan@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-14 02:58:35 +03:00
copy_oom_score_adj ( clone_flags , p ) ;
2005-04-17 02:20:36 +04:00
return p ;
2015-06-09 14:32:09 +03:00
bad_fork_cancel_cgroup :
2021-03-29 16:18:35 +03:00
sched_core_free ( p ) ;
pid_ns: Fix race between setns'ed fork() and zap_pid_ns_processes()
Imagine we have a pid namespace and a task from its parent's pid_ns,
which made setns() to the pid namespace. The task is doing fork(),
while the pid namespace's child reaper is dying. We have the race
between them:
Task from parent pid_ns Child reaper
copy_process() ..
alloc_pid() ..
.. zap_pid_ns_processes()
.. disable_pid_allocation()
.. read_lock(&tasklist_lock)
.. iterate over pids in pid_ns
.. kill tasks linked to pids
.. read_unlock(&tasklist_lock)
write_lock_irq(&tasklist_lock); ..
attach_pid(p, PIDTYPE_PID); ..
.. ..
So, just created task p won't receive SIGKILL signal,
and the pid namespace will be in contradictory state.
Only manual kill will help there, but does the userspace
care about this? I suppose, the most users just inject
a task into a pid namespace and wait a SIGCHLD from it.
The patch fixes the problem. It simply checks for
(pid_ns->nr_hashed & PIDNS_HASH_ADDING) in copy_process().
We do it under the tasklist_lock, and can't skip
PIDNS_HASH_ADDING as noted by Oleg:
"zap_pid_ns_processes() does disable_pid_allocation()
and then takes tasklist_lock to kill the whole namespace.
Given that copy_process() checks PIDNS_HASH_ADDING
under write_lock(tasklist) they can't race;
if copy_process() takes this lock first, the new child will
be killed, otherwise copy_process() can't miss
the change in ->nr_hashed."
If allocation is disabled, we just return -ENOMEM
like it's made for such cases in alloc_pid().
v2: Do not move disable_pid_allocation(), do not
introduce a new variable in copy_process() and simplify
the patch as suggested by Oleg Nesterov.
Account the problem with double irq enabling
found by Eric W. Biederman.
Fixes: c876ad768215 ("pidns: Stop pid allocation when init dies")
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: Ingo Molnar <mingo@kernel.org>
CC: Peter Zijlstra <peterz@infradead.org>
CC: Oleg Nesterov <oleg@redhat.com>
CC: Mike Rapoport <rppt@linux.vnet.ibm.com>
CC: Michal Hocko <mhocko@suse.com>
CC: Andy Lutomirski <luto@kernel.org>
CC: "Eric W. Biederman" <ebiederm@xmission.com>
CC: Andrei Vagin <avagin@openvz.org>
CC: Cyrill Gorcunov <gorcunov@openvz.org>
CC: Serge Hallyn <serge@hallyn.com>
Cc: stable@vger.kernel.org
Acked-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
2017-05-12 19:11:31 +03:00
spin_unlock ( & current - > sighand - > siglock ) ;
write_unlock_irq ( & tasklist_lock ) ;
2020-02-05 16:26:22 +03:00
cgroup_cancel_fork ( p , args ) ;
2019-03-27 15:04:15 +03:00
bad_fork_put_pidfd :
2019-06-27 05:22:09 +03:00
if ( clone_flags & CLONE_PIDFD ) {
fput ( pidfile ) ;
put_unused_fd ( pidfd ) ;
}
2007-10-19 10:40:07 +04:00
bad_fork_free_pid :
if ( pid ! = & init_struct_pid )
free_pid ( pid ) ;
fork: free thread in copy_process on failure
When using this program (as root):
#include <err.h>
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <sys/io.h>
#include <sys/types.h>
#include <sys/wait.h>
#define ITER 1000
#define FORKERS 15
#define THREADS (6000/FORKERS) // 1850 is proc max
static void fork_100_wait()
{
unsigned a, to_wait = 0;
printf("\t%d forking %d\n", THREADS, getpid());
for (a = 0; a < THREADS; a++) {
switch (fork()) {
case 0:
usleep(1000);
exit(0);
break;
case -1:
break;
default:
to_wait++;
break;
}
}
printf("\t%d forked from %d, waiting for %d\n", THREADS, getpid(),
to_wait);
for (a = 0; a < to_wait; a++)
wait(NULL);
printf("\t%d waited from %d\n", THREADS, getpid());
}
static void run_forkers()
{
pid_t forkers[FORKERS];
unsigned a;
for (a = 0; a < FORKERS; a++) {
switch ((forkers[a] = fork())) {
case 0:
fork_100_wait();
exit(0);
break;
case -1:
err(1, "DIE fork of %d'th forker", a);
break;
default:
break;
}
}
for (a = 0; a < FORKERS; a++)
waitpid(forkers[a], NULL, 0);
}
int main()
{
unsigned a;
int ret;
ret = ioperm(10, 20, 0);
if (ret < 0)
err(1, "ioperm");
for (a = 0; a < ITER; a++)
run_forkers();
return 0;
}
kmemleak reports many occurences of this leak:
unreferenced object 0xffff8805917c8000 (size 8192):
comm "fork-leak", pid 2932, jiffies 4295354292 (age 1871.028s)
hex dump (first 32 bytes):
ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
backtrace:
[<ffffffff814cfbf5>] kmemdup+0x25/0x50
[<ffffffff8103ab43>] copy_thread_tls+0x6c3/0x9a0
[<ffffffff81150174>] copy_process+0x1a84/0x5790
[<ffffffff811dc375>] wake_up_new_task+0x2d5/0x6f0
[<ffffffff8115411d>] _do_fork+0x12d/0x820
...
Due to the leakage of the memory items which should have been freed in
arch/x86/kernel/process.c:exit_thread().
Make sure the memory is freed when fork fails later in copy_process.
This is done by calling exit_thread with the thread to kill.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: "James E.J. Bottomley" <jejb@parisc-linux.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Chen Liqin <liqin.linux@gmail.com>
Cc: Chris Metcalf <cmetcalf@mellanox.com>
Cc: Chris Zankel <chris@zankel.net>
Cc: David Howells <dhowells@redhat.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Guan Xuetao <gxt@mprc.pku.edu.cn>
Cc: Haavard Skinnemoen <hskinnemoen@gmail.com>
Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Helge Deller <deller@gmx.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: James Hogan <james.hogan@imgtec.com>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: Jonas Bonn <jonas@southpole.se>
Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com>
Cc: Lennox Wu <lennox.wu@gmail.com>
Cc: Ley Foon Tan <lftan@altera.com>
Cc: Mark Salter <msalter@redhat.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Michal Simek <monstr@monstr.eu>
Cc: Mikael Starvik <starvik@axis.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Rich Felker <dalias@libc.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Richard Kuo <rkuo@codeaurora.org>
Cc: Richard Weinberger <richard@nod.at>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Steven Miao <realmz6@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 03:00:25 +03:00
bad_fork_cleanup_thread :
exit_thread ( p ) ;
2008-01-24 10:52:45 +03:00
bad_fork_cleanup_io :
2009-12-04 16:52:42 +03:00
if ( p - > io_context )
exit_io_context ( p ) ;
2006-10-02 13:18:06 +04:00
bad_fork_cleanup_namespaces :
2007-01-31 00:35:18 +03:00
exit_task_namespaces ( p ) ;
2005-04-17 02:20:36 +04:00
bad_fork_cleanup_mm :
2019-05-15 01:40:46 +03:00
if ( p - > mm ) {
mm_clear_owner ( p - > mm , p ) ;
2005-04-17 02:20:36 +04:00
mmput ( p - > mm ) ;
2019-05-15 01:40:46 +03:00
}
2005-04-17 02:20:36 +04:00
bad_fork_cleanup_signal :
clone(): fix race between copy_process() and de_thread()
Spotted by Hiroshi Shimamoto who also provided the test-case below.
copy_process() uses signal->count as a reference counter, but it is not.
This test case
#include <sys/types.h>
#include <sys/wait.h>
#include <unistd.h>
#include <stdio.h>
#include <errno.h>
#include <pthread.h>
void *null_thread(void *p)
{
for (;;)
sleep(1);
return NULL;
}
void *exec_thread(void *p)
{
execl("/bin/true", "/bin/true", NULL);
return null_thread(p);
}
int main(int argc, char **argv)
{
for (;;) {
pid_t pid;
int ret, status;
pid = fork();
if (pid < 0)
break;
if (!pid) {
pthread_t tid;
pthread_create(&tid, NULL, exec_thread, NULL);
for (;;)
pthread_create(&tid, NULL, null_thread, NULL);
}
do {
ret = waitpid(pid, &status, 0);
} while (ret == -1 && errno == EINTR);
}
return 0;
}
quickly creates an unkillable task.
If copy_process(CLONE_THREAD) races with de_thread()
copy_signal()->atomic(signal->count) breaks the signal->notify_count
logic, and the execing thread can hang forever in kernel space.
Change copy_process() to increment count/live only when we know for sure
we can't fail. In this case the forked thread will take care of its
reference to signal correctly.
If copy_process() fails, check CLONE_THREAD flag. If it it set - do
nothing, the counters were not changed and current belongs to the same
thread group. If it is not set, ->signal must be released in any case
(and ->count must be == 1), the forked child is the only thread in the
thread group.
We need more cleanups here, in particular signal->count should not be used
by de_thread/__exit_signal at all. This patch only fixes the bug.
Reported-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Tested-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-08-27 01:29:24 +04:00
if ( ! ( clone_flags & CLONE_THREAD ) )
2011-01-05 13:16:04 +03:00
free_signal_struct ( p - > signal ) ;
2005-04-17 02:20:36 +04:00
bad_fork_cleanup_sighand :
2006-03-29 04:11:27 +04:00
__cleanup_sighand ( p - > sighand ) ;
2005-04-17 02:20:36 +04:00
bad_fork_cleanup_fs :
exit_fs ( p ) ; /* blocking */
bad_fork_cleanup_files :
exit_files ( p ) ; /* blocking */
bad_fork_cleanup_semundo :
exit_sem ( p ) ;
2017-03-24 14:46:33 +03:00
bad_fork_cleanup_security :
security_task_free ( p ) ;
2005-04-17 02:20:36 +04:00
bad_fork_cleanup_audit :
audit_free ( p ) ;
2014-10-03 03:17:02 +04:00
bad_fork_cleanup_perf :
perf: Do the big rename: Performance Counters -> Performance Events
Bye-bye Performance Counters, welcome Performance Events!
In the past few months the perfcounters subsystem has grown out its
initial role of counting hardware events, and has become (and is
becoming) a much broader generic event enumeration, reporting, logging,
monitoring, analysis facility.
Naming its core object 'perf_counter' and naming the subsystem
'perfcounters' has become more and more of a misnomer. With pending
code like hw-breakpoints support the 'counter' name is less and
less appropriate.
All in one, we've decided to rename the subsystem to 'performance
events' and to propagate this rename through all fields, variables
and API names. (in an ABI compatible fashion)
The word 'event' is also a bit shorter than 'counter' - which makes
it slightly more convenient to write/handle as well.
Thanks goes to Stephane Eranian who first observed this misnomer and
suggested a rename.
User-space tooling and ABI compatibility is not affected - this patch
should be function-invariant. (Also, defconfigs were not touched to
keep the size down.)
This patch has been generated via the following script:
FILES=$(find * -type f | grep -vE 'oprofile|[^K]config')
sed -i \
-e 's/PERF_EVENT_/PERF_RECORD_/g' \
-e 's/PERF_COUNTER/PERF_EVENT/g' \
-e 's/perf_counter/perf_event/g' \
-e 's/nb_counters/nb_events/g' \
-e 's/swcounter/swevent/g' \
-e 's/tpcounter_event/tp_event/g' \
$FILES
for N in $(find . -name perf_counter.[ch]); do
M=$(echo $N | sed 's/perf_counter/perf_event/g')
mv $N $M
done
FILES=$(find . -name perf_event.*)
sed -i \
-e 's/COUNTER_MASK/REG_MASK/g' \
-e 's/COUNTER/EVENT/g' \
-e 's/\<event\>/event_id/g' \
-e 's/counter/event/g' \
-e 's/Counter/Event/g' \
$FILES
... to keep it as correct as possible. This script can also be
used by anyone who has pending perfcounters patches - it converts
a Linux kernel tree over to the new naming. We tried to time this
change to the point in time where the amount of pending patches
is the smallest: the end of the merge window.
Namespace clashes were fixed up in a preparatory patch - and some
stylistic fallout will be fixed up in a subsequent patch.
( NOTE: 'counters' are still the proper terminology when we deal
with hardware registers - and these sed scripts are a bit
over-eager in renaming them. I've undone some of that, but
in case there's something left where 'counter' would be
better than 'event' we can undo that on an individual basis
instead of touching an otherwise nicely automated patch. )
Suggested-by: Stephane Eranian <eranian@google.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Paul Mackerras <paulus@samba.org>
Reviewed-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <linux-arch@vger.kernel.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-09-21 14:02:48 +04:00
perf_event_free_task ( p ) ;
2014-10-03 03:17:02 +04:00
bad_fork_cleanup_policy :
2017-08-07 10:12:52 +03:00
lockdep_free_task ( p ) ;
2005-04-17 02:20:36 +04:00
# ifdef CONFIG_NUMA
2008-04-28 13:13:08 +04:00
mpol_put ( p - > mempolicy ) ;
2014-03-28 11:18:27 +04:00
bad_fork_cleanup_threadgroup_lock :
2005-04-17 02:20:36 +04:00
# endif
2006-09-01 08:27:38 +04:00
delayacct_tsk_free ( p ) ;
2005-04-17 02:20:36 +04:00
bad_fork_cleanup_count :
2021-04-22 15:27:11 +03:00
dec_rlimit_ucounts ( task_ucounts ( p ) , UCOUNT_RLIMIT_NPROC , 1 ) ;
2009-09-02 12:13:40 +04:00
exit_creds ( p ) ;
2005-04-17 02:20:36 +04:00
bad_fork_free :
2021-06-11 11:28:17 +03:00
WRITE_ONCE ( p - > __state , TASK_DEAD ) ;
2016-09-16 08:45:48 +03:00
put_task_stack ( p ) ;
2019-05-15 01:40:46 +03:00
delayed_free_task ( p ) ;
2006-01-08 12:04:02 +03:00
fork_out :
2018-07-23 23:20:37 +03:00
spin_lock_irq ( & current - > sighand - > siglock ) ;
hlist_del_init ( & delayed . node ) ;
spin_unlock_irq ( & current - > sighand - > siglock ) ;
2006-01-08 12:04:02 +03:00
return ERR_PTR ( retval ) ;
2005-04-17 02:20:36 +04:00
}
2017-09-26 21:06:43 +03:00
static inline void init_idle_pids ( struct task_struct * idle )
2010-05-27 01:44:11 +04:00
{
enum pid_type type ;
for ( type = PIDTYPE_PID ; type < PIDTYPE_MAX ; + + type ) {
2017-09-26 21:06:43 +03:00
INIT_HLIST_NODE ( & idle - > pid_links [ type ] ) ; /* not really needed */
init_task_pid ( idle , type , & init_struct_pid ) ;
2010-05-27 01:44:11 +04:00
}
}
sched/core: Initialize the idle task with preemption disabled
As pointed out by commit
de9b8f5dcbd9 ("sched: Fix crash trying to dequeue/enqueue the idle thread")
init_idle() can and will be invoked more than once on the same idle
task. At boot time, it is invoked for the boot CPU thread by
sched_init(). Then smp_init() creates the threads for all the secondary
CPUs and invokes init_idle() on them.
As the hotplug machinery brings the secondaries to life, it will issue
calls to idle_thread_get(), which itself invokes init_idle() yet again.
In this case it's invoked twice more per secondary: at _cpu_up(), and at
bringup_cpu().
Given smp_init() already initializes the idle tasks for all *possible*
CPUs, no further initialization should be required. Now, removing
init_idle() from idle_thread_get() exposes some interesting expectations
with regards to the idle task's preempt_count: the secondary startup always
issues a preempt_disable(), requiring some reset of the preempt count to 0
between hot-unplug and hotplug, which is currently served by
idle_thread_get() -> idle_init().
Given the idle task is supposed to have preemption disabled once and never
see it re-enabled, it seems that what we actually want is to initialize its
preempt_count to PREEMPT_DISABLED and leave it there. Do that, and remove
init_idle() from idle_thread_get().
Secondary startups were patched via coccinelle:
@begone@
@@
-preempt_disable();
...
cpu_startup_entry(CPUHP_AP_ONLINE_IDLE);
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210512094636.2958515-1-valentin.schneider@arm.com
2021-05-12 12:46:36 +03:00
struct task_struct * __init fork_idle ( int cpu )
2005-04-17 02:20:36 +04:00
{
2006-07-03 11:25:41 +04:00
struct task_struct * task ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
struct kernel_clone_args args = {
. flags = CLONE_VM ,
} ;
task = copy_process ( & init_struct_pid , 0 , cpu_to_node ( cpu ) , & args ) ;
2010-05-27 01:44:11 +04:00
if ( ! IS_ERR ( task ) ) {
2017-09-26 21:06:43 +03:00
init_idle_pids ( task ) ;
2006-11-25 22:09:34 +03:00
init_idle ( task , cpu ) ;
2010-05-27 01:44:11 +04:00
}
2006-03-29 04:11:07 +04:00
2005-04-17 02:20:36 +04:00
return task ;
}
2019-04-26 03:11:25 +03:00
struct mm_struct * copy_init_mm ( void )
{
return dup_mm ( NULL , & init_mm ) ;
}
2021-03-04 22:21:05 +03:00
/*
* This is like kernel_clone ( ) , but shaved down and tailored to just
* creating io_uring workers . It returns a created task , or an error pointer .
* The returned task is inactive , and the caller must fire it up through
* wake_up_new_task ( p ) . All signals are blocked in the created task .
*/
struct task_struct * create_io_thread ( int ( * fn ) ( void * ) , void * arg , int node )
{
unsigned long flags = CLONE_FS | CLONE_FILES | CLONE_SIGHAND | CLONE_THREAD |
CLONE_IO ;
struct kernel_clone_args args = {
. flags = ( ( lower_32_bits ( flags ) | CLONE_VM |
CLONE_UNTRACED ) & ~ CSIGNAL ) ,
. exit_signal = ( lower_32_bits ( flags ) & CSIGNAL ) ,
. stack = ( unsigned long ) fn ,
. stack_size = ( unsigned long ) arg ,
. io_thread = 1 ,
} ;
2021-03-26 18:05:22 +03:00
return copy_process ( NULL , 0 , node , & args ) ;
2021-03-04 22:21:05 +03:00
}
2005-04-17 02:20:36 +04:00
/*
* Ok , this is the main fork - routine .
*
* It copies the process , and if successful kick - starts
* it and waits for it to finish using the VM if required .
fork: block invalid exit signals with clone3()
Previously, higher 32 bits of exit_signal fields were lost when copied
to the kernel args structure (that uses int as a type for the respective
field). Moreover, as Oleg has noted, exit_signal is used unchecked, so
it has to be checked for sanity before use; for the legacy syscalls,
applying CSIGNAL mask guarantees that it is at least non-negative;
however, there's no such thing is done in clone3() code path, and that
can break at least thread_group_leader.
This commit adds a check to copy_clone_args_from_user() to verify that
the exit signal is limited by CSIGNAL as with legacy clone() and that
the signal is valid. With this we don't get the legacy clone behavior
were an invalid signal could be handed down and would only be detected
and ignored in do_notify_parent(). Users of clone3() will now get a
proper error when they pass an invalid exit signal. Note, that this is
not user-visible behavior since no kernel with clone3() has been
released yet.
The following program will cause a splat on a non-fixed clone3() version
and will fail correctly on a fixed version:
#define _GNU_SOURCE
#include <linux/sched.h>
#include <linux/types.h>
#include <sched.h>
#include <stdio.h>
#include <stdlib.h>
#include <sys/syscall.h>
#include <sys/wait.h>
#include <unistd.h>
int main(int argc, char *argv[])
{
pid_t pid = -1;
struct clone_args args = {0};
args.exit_signal = -1;
pid = syscall(__NR_clone3, &args, sizeof(struct clone_args));
if (pid < 0)
exit(EXIT_FAILURE);
if (pid == 0)
exit(EXIT_SUCCESS);
wait(NULL);
exit(EXIT_SUCCESS);
}
Fixes: 7f192e3cd316 ("fork: add clone3")
Reported-by: Oleg Nesterov <oleg@redhat.com>
Suggested-by: Oleg Nesterov <oleg@redhat.com>
Suggested-by: Dmitry V. Levin <ldv@altlinux.org>
Signed-off-by: Eugene Syromiatnikov <esyr@redhat.com>
Link: https://lore.kernel.org/r/4b38fa4ce420b119a4c6345f42fe3cec2de9b0b5.1568223594.git.esyr@redhat.com
[christian.brauner@ubuntu.com: simplify check and rework commit message]
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
2019-09-11 20:45:40 +03:00
*
* args - > exit_signal is expected to be checked for sanity by the caller .
2005-04-17 02:20:36 +04:00
*/
2020-08-19 13:46:45 +03:00
pid_t kernel_clone ( struct kernel_clone_args * args )
2005-04-17 02:20:36 +04:00
{
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
u64 clone_flags = args - > flags ;
2018-02-07 02:39:30 +03:00
struct completion vfork ;
struct pid * pid ;
2005-04-17 02:20:36 +04:00
struct task_struct * p ;
int trace = 0 ;
2020-08-19 13:46:45 +03:00
pid_t nr ;
2005-04-17 02:20:36 +04:00
2020-06-08 18:28:50 +03:00
/*
* For legacy clone ( ) calls , CLONE_PIDFD uses the parent_tid argument
* to return the pidfd . Hence , CLONE_PIDFD and CLONE_PARENT_SETTID are
* mutually exclusive . With clone3 ( ) CLONE_PIDFD has grown a separate
* field in struct clone_args and it still doesn ' t make sense to have
* them both point at the same memory location . Performing this check
* here has the advantage that we don ' t need to have a separate helper
* to check for legacy clone ( ) .
*/
if ( ( args - > flags & CLONE_PIDFD ) & &
( args - > flags & CLONE_PARENT_SETTID ) & &
( args - > pidfd = = args - > parent_tid ) )
return - EINVAL ;
2008-07-26 06:45:47 +04:00
/*
2011-06-17 18:50:38 +04:00
* Determine whether and which event to report to ptracer . When
* called from kernel_thread or CLONE_UNTRACED is explicitly
* requested , no event is reported ; otherwise , report if the event
* for the type of forking is enabled .
2008-07-26 06:45:47 +04:00
*/
2012-10-23 07:10:08 +04:00
if ( ! ( clone_flags & CLONE_UNTRACED ) ) {
2011-06-17 18:50:38 +04:00
if ( clone_flags & CLONE_VFORK )
trace = PTRACE_EVENT_VFORK ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
else if ( args - > exit_signal ! = SIGCHLD )
2011-06-17 18:50:38 +04:00
trace = PTRACE_EVENT_CLONE ;
else
trace = PTRACE_EVENT_FORK ;
if ( likely ( ! ptrace_event_enabled ( current , trace ) ) )
trace = 0 ;
}
2005-04-17 02:20:36 +04:00
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
p = copy_process ( NULL , trace , NUMA_NO_NODE , args ) ;
gcc-plugins: Add latent_entropy plugin
This adds a new gcc plugin named "latent_entropy". It is designed to
extract as much possible uncertainty from a running system at boot time as
possible, hoping to capitalize on any possible variation in CPU operation
(due to runtime data differences, hardware differences, SMP ordering,
thermal timing variation, cache behavior, etc).
At the very least, this plugin is a much more comprehensive example for
how to manipulate kernel code using the gcc plugin internals.
The need for very-early boot entropy tends to be very architecture or
system design specific, so this plugin is more suited for those sorts
of special cases. The existing kernel RNG already attempts to extract
entropy from reliable runtime variation, but this plugin takes the idea to
a logical extreme by permuting a global variable based on any variation
in code execution (e.g. a different value (and permutation function)
is used to permute the global based on loop count, case statement,
if/then/else branching, etc).
To do this, the plugin starts by inserting a local variable in every
marked function. The plugin then adds logic so that the value of this
variable is modified by randomly chosen operations (add, xor and rol) and
random values (gcc generates separate static values for each location at
compile time and also injects the stack pointer at runtime). The resulting
value depends on the control flow path (e.g., loops and branches taken).
Before the function returns, the plugin mixes this local variable into
the latent_entropy global variable. The value of this global variable
is added to the kernel entropy pool in do_one_initcall() and _do_fork(),
though it does not credit any bytes of entropy to the pool; the contents
of the global are just used to mix the pool.
Additionally, the plugin can pre-initialize arrays with build-time
random contents, so that two different kernel builds running on identical
hardware will not have the same starting values.
Signed-off-by: Emese Revfy <re.emese@gmail.com>
[kees: expanded commit message and code comments]
Signed-off-by: Kees Cook <keescook@chromium.org>
2016-06-20 21:41:19 +03:00
add_latent_entropy ( ) ;
2018-02-07 02:39:30 +03:00
if ( IS_ERR ( p ) )
return PTR_ERR ( p ) ;
2005-04-17 02:20:36 +04:00
/*
* Do this prior waking up the new thread - the thread pointer
* might get invalid after that point , if the thread exits quickly .
*/
2018-02-07 02:39:30 +03:00
trace_sched_process_fork ( current , p ) ;
tracing, sched: LTTng instrumentation - scheduler
Instrument the scheduler activity (sched_switch, migration, wakeups,
wait for a task, signal delivery) and process/thread
creation/destruction (fork, exit, kthread stop). Actually, kthread
creation is not instrumented in this patch because it is architecture
dependent. It allows to connect tracers such as ftrace which detects
scheduling latencies, good/bad scheduler decisions. Tools like LTTng can
export this scheduler information along with instrumentation of the rest
of the kernel activity to perform post-mortem analysis on the scheduler
activity.
About the performance impact of tracepoints (which is comparable to
markers), even without immediate values optimizations, tests done by
Hideo Aoki on ia64 show no regression. His test case was using hackbench
on a kernel where scheduler instrumentation (about 5 events in code
scheduler code) was added. See the "Tracepoints" patch header for
performance result detail.
Changelog :
- Change instrumentation location and parameter to match ftrace
instrumentation, previously done with kernel markers.
[ mingo@elte.hu: conflict resolutions ]
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Acked-by: 'Peter Zijlstra' <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-07-18 20:16:17 +04:00
2018-02-07 02:39:30 +03:00
pid = get_task_pid ( p , PIDTYPE_PID ) ;
nr = pid_vnr ( pid ) ;
2007-10-19 10:40:10 +04:00
2018-02-07 02:39:30 +03:00
if ( clone_flags & CLONE_PARENT_SETTID )
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
put_user ( nr , args - > parent_tid ) ;
2007-10-19 10:39:53 +04:00
2018-02-07 02:39:30 +03:00
if ( clone_flags & CLONE_VFORK ) {
p - > vfork_done = & vfork ;
init_completion ( & vfork ) ;
get_task_struct ( p ) ;
}
2005-04-17 02:20:36 +04:00
2018-02-07 02:39:30 +03:00
wake_up_new_task ( p ) ;
2008-07-26 06:45:47 +04:00
2018-02-07 02:39:30 +03:00
/* forking complete and child started to run, tell ptracer */
if ( unlikely ( trace ) )
ptrace_event_pid ( trace , pid ) ;
2014-06-07 01:36:42 +04:00
2018-02-07 02:39:30 +03:00
if ( clone_flags & CLONE_VFORK ) {
if ( ! wait_for_vfork_done ( p , & vfork ) )
ptrace_event_pid ( PTRACE_EVENT_VFORK_DONE , pid ) ;
2005-04-17 02:20:36 +04:00
}
2018-02-07 02:39:30 +03:00
put_pid ( pid ) ;
[PATCH] pidhash: Refactor the pid hash table
Simplifies the code, reduces the need for 4 pid hash tables, and makes the
code more capable.
In the discussions I had with Oleg it was felt that to a large extent the
cleanup itself justified the work. With struct pid being dynamically
allocated meant we could create the hash table entry when the pid was
allocated and free the hash table entry when the pid was freed. Instead of
playing with the hash lists when ever a process would attach or detach to a
process.
For myself the fact that it gave what my previous task_ref patch gave for free
with simpler code was a big win. The problem is that if you hold a reference
to struct task_struct you lock in 10K of low memory. If you do that in a user
controllable way like /proc does, with an unprivileged but hostile user space
application with typical resource limits of 1000 fds and 100 processes I can
trigger the OOM killer by consuming all of low memory with task structs, on a
machine wight 1GB of low memory.
If I instead hold a reference to struct pid which holds a pointer to my
task_struct, I don't suffer from that problem because struct pid is 2 orders
of magnitude smaller. In fact struct pid is small enough that most other
kernel data structures dwarf it, so simply limiting the number of referring
data structures is enough to prevent exhaustion of low memory.
This splits the current struct pid into two structures, struct pid and struct
pid_link, and reduces our number of hash tables from PIDTYPE_MAX to just one.
struct pid_link is the per process linkage into the hash tables and lives in
struct task_struct. struct pid is given an indepedent lifetime, and holds
pointers to each of the pid types.
The independent life of struct pid simplifies attach_pid, and detach_pid,
because we are always manipulating the list of pids and not the hash table.
In addition in giving struct pid an indpendent life it makes the concept much
more powerful.
Kernel data structures can now embed a struct pid * instead of a pid_t and
not suffer from pid wrap around problems or from keeping unnecessarily
large amounts of memory allocated.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-31 14:31:42 +04:00
return nr ;
2005-04-17 02:20:36 +04:00
}
2012-09-22 03:55:31 +04:00
/*
* Create a kernel thread .
*/
pid_t kernel_thread ( int ( * fn ) ( void * ) , void * arg , unsigned long flags )
{
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
struct kernel_clone_args args = {
fork: prevent accidental access to clone3 features
Jan reported an issue where an interaction between sign-extending clone's
flag argument on ppc64le and the new CLONE_INTO_CGROUP feature causes
clone() to consistently fail with EBADF.
The whole story is a little longer. The legacy clone() syscall is odd in a
bunch of ways and here two things interact. First, legacy clone's flag
argument is word-size dependent, i.e. it's an unsigned long whereas most
system calls with flag arguments use int or unsigned int. Second, legacy
clone() ignores unknown and deprecated flags. The two of them taken
together means that users on 64bit systems can pass garbage for the upper
32bit of the clone() syscall since forever and things would just work fine.
Just try this on a 64bit kernel prior to v5.7-rc1 where this will succeed
and on v5.7-rc1 where this will fail with EBADF:
int main(int argc, char *argv[])
{
pid_t pid;
/* Note that legacy clone() has different argument ordering on
* different architectures so this won't work everywhere.
*
* Only set the upper 32 bits.
*/
pid = syscall(__NR_clone, 0xffffffff00000000 | SIGCHLD,
NULL, NULL, NULL, NULL);
if (pid < 0)
exit(EXIT_FAILURE);
if (pid == 0)
exit(EXIT_SUCCESS);
if (wait(NULL) != pid)
exit(EXIT_FAILURE);
exit(EXIT_SUCCESS);
}
Since legacy clone() couldn't be extended this was not a problem so far and
nobody really noticed or cared since nothing in the kernel ever bothered to
look at the upper 32 bits.
But once we introduced clone3() and expanded the flag argument in struct
clone_args to 64 bit we opened this can of worms. With the first flag-based
extension to clone3() making use of the upper 32 bits of the flag argument
we've effectively made it possible for the legacy clone() syscall to reach
clone3() only flags. The sign extension scenario is just the odd
corner-case that we needed to figure this out.
The reason we just realized this now and not already when we introduced
CLONE_CLEAR_SIGHAND was that CLONE_INTO_CGROUP assumes that a valid cgroup
file descriptor has been given. So the sign extension (or the user
accidently passing garbage for the upper 32 bits) caused the
CLONE_INTO_CGROUP bit to be raised and the kernel to error out when it
didn't find a valid cgroup file descriptor.
Let's fix this by always capping the upper 32 bits for all codepaths that
are not aware of clone3() features. This ensures that we can't reach
clone3() only features by accident via legacy clone as with the sign
extension case and also that legacy clone() works exactly like before, i.e.
ignoring any unknown flags. This solution risks no regressions and is also
pretty clean.
Fixes: 7f192e3cd316 ("fork: add clone3")
Fixes: ef2c41cf38a7 ("clone3: allow spawning processes into cgroups")
Reported-by: Jan Stancek <jstancek@redhat.com>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Dmitry V. Levin <ldv@altlinux.org>
Cc: Andreas Schwab <schwab@linux-m68k.org>
Cc: Florian Weimer <fw@deneb.enyo.de>
Cc: libc-alpha@sourceware.org
Cc: stable@vger.kernel.org # 5.3+
Link: https://sourceware.org/pipermail/libc-alpha/2020-May/113596.html
Link: https://lore.kernel.org/r/20200507103214.77218-1-christian.brauner@ubuntu.com
2020-05-07 13:32:14 +03:00
. flags = ( ( lower_32_bits ( flags ) | CLONE_VM |
CLONE_UNTRACED ) & ~ CSIGNAL ) ,
. exit_signal = ( lower_32_bits ( flags ) & CSIGNAL ) ,
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
. stack = ( unsigned long ) fn ,
. stack_size = ( unsigned long ) arg ,
} ;
2020-08-19 13:46:45 +03:00
return kernel_clone ( & args ) ;
2012-09-22 03:55:31 +04:00
}
2012-10-23 21:17:59 +04:00
# ifdef __ARCH_WANT_SYS_FORK
SYSCALL_DEFINE0 ( fork )
{
# ifdef CONFIG_MMU
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
struct kernel_clone_args args = {
. exit_signal = SIGCHLD ,
} ;
2020-08-19 13:46:45 +03:00
return kernel_clone ( & args ) ;
2012-10-23 21:17:59 +04:00
# else
/* can not support in nommu mode */
2014-01-24 03:55:47 +04:00
return - EINVAL ;
2012-10-23 21:17:59 +04:00
# endif
}
# endif
# ifdef __ARCH_WANT_SYS_VFORK
SYSCALL_DEFINE0 ( vfork )
{
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
struct kernel_clone_args args = {
. flags = CLONE_VFORK | CLONE_VM ,
. exit_signal = SIGCHLD ,
} ;
2020-08-19 13:46:45 +03:00
return kernel_clone ( & args ) ;
2012-10-23 21:17:59 +04:00
}
# endif
# ifdef __ARCH_WANT_SYS_CLONE
# ifdef CONFIG_CLONE_BACKWARDS
SYSCALL_DEFINE5 ( clone , unsigned long , clone_flags , unsigned long , newsp ,
int __user * , parent_tidptr ,
clone: support passing tls argument via C rather than pt_regs magic
clone has some of the quirkiest syscall handling in the kernel, with a
pile of special cases, historical curiosities, and architecture-specific
calling conventions. In particular, clone with CLONE_SETTLS accepts a
parameter "tls" that the C entry point completely ignores and some
assembly entry points overwrite; instead, the low-level arch-specific
code pulls the tls parameter out of the arch-specific register captured
as part of pt_regs on entry to the kernel. That's a massive hack, and
it makes the arch-specific code only work when called via the specific
existing syscall entry points; because of this hack, any new clone-like
system call would have to accept an identical tls argument in exactly
the same arch-specific position, rather than providing a unified system
call entry point across architectures.
The first patch allows architectures to handle the tls argument via
normal C parameter passing, if they opt in by selecting
HAVE_COPY_THREAD_TLS. The second patch makes 32-bit and 64-bit x86 opt
into this.
These two patches came out of the clone4 series, which isn't ready for
this merge window, but these first two cleanup patches were entirely
uncontroversial and have acks. I'd like to go ahead and submit these
two so that other architectures can begin building on top of this and
opting into HAVE_COPY_THREAD_TLS. However, I'm also happy to wait and
send these through the next merge window (along with v3 of clone4) if
anyone would prefer that.
This patch (of 2):
clone with CLONE_SETTLS accepts an argument to set the thread-local
storage area for the new thread. sys_clone declares an int argument
tls_val in the appropriate point in the argument list (based on the
various CLONE_BACKWARDS variants), but doesn't actually use or pass along
that argument. Instead, sys_clone calls do_fork, which calls
copy_process, which calls the arch-specific copy_thread, and copy_thread
pulls the corresponding syscall argument out of the pt_regs captured at
kernel entry (knowing what argument of clone that architecture passes tls
in).
Apart from being awful and inscrutable, that also only works because only
one code path into copy_thread can pass the CLONE_SETTLS flag, and that
code path comes from sys_clone with its architecture-specific
argument-passing order. This prevents introducing a new version of the
clone system call without propagating the same architecture-specific
position of the tls argument.
However, there's no reason to pull the argument out of pt_regs when
sys_clone could just pass it down via C function call arguments.
Introduce a new CONFIG_HAVE_COPY_THREAD_TLS for architectures to opt into,
and a new copy_thread_tls that accepts the tls parameter as an additional
unsigned long (syscall-argument-sized) argument. Change sys_clone's tls
argument to an unsigned long (which does not change the ABI), and pass
that down to copy_thread_tls.
Architectures that don't opt into copy_thread_tls will continue to ignore
the C argument to sys_clone in favor of the pt_regs captured at kernel
entry, and thus will be unable to introduce new versions of the clone
syscall.
Patch co-authored by Josh Triplett and Thiago Macieira.
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Acked-by: Andy Lutomirski <luto@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Thiago Macieira <thiago.macieira@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-06-26 01:01:19 +03:00
unsigned long , tls ,
2012-10-23 21:17:59 +04:00
int __user * , child_tidptr )
# elif defined(CONFIG_CLONE_BACKWARDS2)
SYSCALL_DEFINE5 ( clone , unsigned long , newsp , unsigned long , clone_flags ,
int __user * , parent_tidptr ,
int __user * , child_tidptr ,
clone: support passing tls argument via C rather than pt_regs magic
clone has some of the quirkiest syscall handling in the kernel, with a
pile of special cases, historical curiosities, and architecture-specific
calling conventions. In particular, clone with CLONE_SETTLS accepts a
parameter "tls" that the C entry point completely ignores and some
assembly entry points overwrite; instead, the low-level arch-specific
code pulls the tls parameter out of the arch-specific register captured
as part of pt_regs on entry to the kernel. That's a massive hack, and
it makes the arch-specific code only work when called via the specific
existing syscall entry points; because of this hack, any new clone-like
system call would have to accept an identical tls argument in exactly
the same arch-specific position, rather than providing a unified system
call entry point across architectures.
The first patch allows architectures to handle the tls argument via
normal C parameter passing, if they opt in by selecting
HAVE_COPY_THREAD_TLS. The second patch makes 32-bit and 64-bit x86 opt
into this.
These two patches came out of the clone4 series, which isn't ready for
this merge window, but these first two cleanup patches were entirely
uncontroversial and have acks. I'd like to go ahead and submit these
two so that other architectures can begin building on top of this and
opting into HAVE_COPY_THREAD_TLS. However, I'm also happy to wait and
send these through the next merge window (along with v3 of clone4) if
anyone would prefer that.
This patch (of 2):
clone with CLONE_SETTLS accepts an argument to set the thread-local
storage area for the new thread. sys_clone declares an int argument
tls_val in the appropriate point in the argument list (based on the
various CLONE_BACKWARDS variants), but doesn't actually use or pass along
that argument. Instead, sys_clone calls do_fork, which calls
copy_process, which calls the arch-specific copy_thread, and copy_thread
pulls the corresponding syscall argument out of the pt_regs captured at
kernel entry (knowing what argument of clone that architecture passes tls
in).
Apart from being awful and inscrutable, that also only works because only
one code path into copy_thread can pass the CLONE_SETTLS flag, and that
code path comes from sys_clone with its architecture-specific
argument-passing order. This prevents introducing a new version of the
clone system call without propagating the same architecture-specific
position of the tls argument.
However, there's no reason to pull the argument out of pt_regs when
sys_clone could just pass it down via C function call arguments.
Introduce a new CONFIG_HAVE_COPY_THREAD_TLS for architectures to opt into,
and a new copy_thread_tls that accepts the tls parameter as an additional
unsigned long (syscall-argument-sized) argument. Change sys_clone's tls
argument to an unsigned long (which does not change the ABI), and pass
that down to copy_thread_tls.
Architectures that don't opt into copy_thread_tls will continue to ignore
the C argument to sys_clone in favor of the pt_regs captured at kernel
entry, and thus will be unable to introduce new versions of the clone
syscall.
Patch co-authored by Josh Triplett and Thiago Macieira.
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Acked-by: Andy Lutomirski <luto@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Thiago Macieira <thiago.macieira@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-06-26 01:01:19 +03:00
unsigned long , tls )
2013-08-14 03:00:53 +04:00
# elif defined(CONFIG_CLONE_BACKWARDS3)
SYSCALL_DEFINE6 ( clone , unsigned long , clone_flags , unsigned long , newsp ,
int , stack_size ,
int __user * , parent_tidptr ,
int __user * , child_tidptr ,
clone: support passing tls argument via C rather than pt_regs magic
clone has some of the quirkiest syscall handling in the kernel, with a
pile of special cases, historical curiosities, and architecture-specific
calling conventions. In particular, clone with CLONE_SETTLS accepts a
parameter "tls" that the C entry point completely ignores and some
assembly entry points overwrite; instead, the low-level arch-specific
code pulls the tls parameter out of the arch-specific register captured
as part of pt_regs on entry to the kernel. That's a massive hack, and
it makes the arch-specific code only work when called via the specific
existing syscall entry points; because of this hack, any new clone-like
system call would have to accept an identical tls argument in exactly
the same arch-specific position, rather than providing a unified system
call entry point across architectures.
The first patch allows architectures to handle the tls argument via
normal C parameter passing, if they opt in by selecting
HAVE_COPY_THREAD_TLS. The second patch makes 32-bit and 64-bit x86 opt
into this.
These two patches came out of the clone4 series, which isn't ready for
this merge window, but these first two cleanup patches were entirely
uncontroversial and have acks. I'd like to go ahead and submit these
two so that other architectures can begin building on top of this and
opting into HAVE_COPY_THREAD_TLS. However, I'm also happy to wait and
send these through the next merge window (along with v3 of clone4) if
anyone would prefer that.
This patch (of 2):
clone with CLONE_SETTLS accepts an argument to set the thread-local
storage area for the new thread. sys_clone declares an int argument
tls_val in the appropriate point in the argument list (based on the
various CLONE_BACKWARDS variants), but doesn't actually use or pass along
that argument. Instead, sys_clone calls do_fork, which calls
copy_process, which calls the arch-specific copy_thread, and copy_thread
pulls the corresponding syscall argument out of the pt_regs captured at
kernel entry (knowing what argument of clone that architecture passes tls
in).
Apart from being awful and inscrutable, that also only works because only
one code path into copy_thread can pass the CLONE_SETTLS flag, and that
code path comes from sys_clone with its architecture-specific
argument-passing order. This prevents introducing a new version of the
clone system call without propagating the same architecture-specific
position of the tls argument.
However, there's no reason to pull the argument out of pt_regs when
sys_clone could just pass it down via C function call arguments.
Introduce a new CONFIG_HAVE_COPY_THREAD_TLS for architectures to opt into,
and a new copy_thread_tls that accepts the tls parameter as an additional
unsigned long (syscall-argument-sized) argument. Change sys_clone's tls
argument to an unsigned long (which does not change the ABI), and pass
that down to copy_thread_tls.
Architectures that don't opt into copy_thread_tls will continue to ignore
the C argument to sys_clone in favor of the pt_regs captured at kernel
entry, and thus will be unable to introduce new versions of the clone
syscall.
Patch co-authored by Josh Triplett and Thiago Macieira.
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Acked-by: Andy Lutomirski <luto@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Thiago Macieira <thiago.macieira@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-06-26 01:01:19 +03:00
unsigned long , tls )
2012-10-23 21:17:59 +04:00
# else
SYSCALL_DEFINE5 ( clone , unsigned long , clone_flags , unsigned long , newsp ,
int __user * , parent_tidptr ,
int __user * , child_tidptr ,
clone: support passing tls argument via C rather than pt_regs magic
clone has some of the quirkiest syscall handling in the kernel, with a
pile of special cases, historical curiosities, and architecture-specific
calling conventions. In particular, clone with CLONE_SETTLS accepts a
parameter "tls" that the C entry point completely ignores and some
assembly entry points overwrite; instead, the low-level arch-specific
code pulls the tls parameter out of the arch-specific register captured
as part of pt_regs on entry to the kernel. That's a massive hack, and
it makes the arch-specific code only work when called via the specific
existing syscall entry points; because of this hack, any new clone-like
system call would have to accept an identical tls argument in exactly
the same arch-specific position, rather than providing a unified system
call entry point across architectures.
The first patch allows architectures to handle the tls argument via
normal C parameter passing, if they opt in by selecting
HAVE_COPY_THREAD_TLS. The second patch makes 32-bit and 64-bit x86 opt
into this.
These two patches came out of the clone4 series, which isn't ready for
this merge window, but these first two cleanup patches were entirely
uncontroversial and have acks. I'd like to go ahead and submit these
two so that other architectures can begin building on top of this and
opting into HAVE_COPY_THREAD_TLS. However, I'm also happy to wait and
send these through the next merge window (along with v3 of clone4) if
anyone would prefer that.
This patch (of 2):
clone with CLONE_SETTLS accepts an argument to set the thread-local
storage area for the new thread. sys_clone declares an int argument
tls_val in the appropriate point in the argument list (based on the
various CLONE_BACKWARDS variants), but doesn't actually use or pass along
that argument. Instead, sys_clone calls do_fork, which calls
copy_process, which calls the arch-specific copy_thread, and copy_thread
pulls the corresponding syscall argument out of the pt_regs captured at
kernel entry (knowing what argument of clone that architecture passes tls
in).
Apart from being awful and inscrutable, that also only works because only
one code path into copy_thread can pass the CLONE_SETTLS flag, and that
code path comes from sys_clone with its architecture-specific
argument-passing order. This prevents introducing a new version of the
clone system call without propagating the same architecture-specific
position of the tls argument.
However, there's no reason to pull the argument out of pt_regs when
sys_clone could just pass it down via C function call arguments.
Introduce a new CONFIG_HAVE_COPY_THREAD_TLS for architectures to opt into,
and a new copy_thread_tls that accepts the tls parameter as an additional
unsigned long (syscall-argument-sized) argument. Change sys_clone's tls
argument to an unsigned long (which does not change the ABI), and pass
that down to copy_thread_tls.
Architectures that don't opt into copy_thread_tls will continue to ignore
the C argument to sys_clone in favor of the pt_regs captured at kernel
entry, and thus will be unable to introduce new versions of the clone
syscall.
Patch co-authored by Josh Triplett and Thiago Macieira.
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Acked-by: Andy Lutomirski <luto@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Thiago Macieira <thiago.macieira@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-06-26 01:01:19 +03:00
unsigned long , tls )
2012-10-23 21:17:59 +04:00
# endif
{
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
struct kernel_clone_args args = {
fork: prevent accidental access to clone3 features
Jan reported an issue where an interaction between sign-extending clone's
flag argument on ppc64le and the new CLONE_INTO_CGROUP feature causes
clone() to consistently fail with EBADF.
The whole story is a little longer. The legacy clone() syscall is odd in a
bunch of ways and here two things interact. First, legacy clone's flag
argument is word-size dependent, i.e. it's an unsigned long whereas most
system calls with flag arguments use int or unsigned int. Second, legacy
clone() ignores unknown and deprecated flags. The two of them taken
together means that users on 64bit systems can pass garbage for the upper
32bit of the clone() syscall since forever and things would just work fine.
Just try this on a 64bit kernel prior to v5.7-rc1 where this will succeed
and on v5.7-rc1 where this will fail with EBADF:
int main(int argc, char *argv[])
{
pid_t pid;
/* Note that legacy clone() has different argument ordering on
* different architectures so this won't work everywhere.
*
* Only set the upper 32 bits.
*/
pid = syscall(__NR_clone, 0xffffffff00000000 | SIGCHLD,
NULL, NULL, NULL, NULL);
if (pid < 0)
exit(EXIT_FAILURE);
if (pid == 0)
exit(EXIT_SUCCESS);
if (wait(NULL) != pid)
exit(EXIT_FAILURE);
exit(EXIT_SUCCESS);
}
Since legacy clone() couldn't be extended this was not a problem so far and
nobody really noticed or cared since nothing in the kernel ever bothered to
look at the upper 32 bits.
But once we introduced clone3() and expanded the flag argument in struct
clone_args to 64 bit we opened this can of worms. With the first flag-based
extension to clone3() making use of the upper 32 bits of the flag argument
we've effectively made it possible for the legacy clone() syscall to reach
clone3() only flags. The sign extension scenario is just the odd
corner-case that we needed to figure this out.
The reason we just realized this now and not already when we introduced
CLONE_CLEAR_SIGHAND was that CLONE_INTO_CGROUP assumes that a valid cgroup
file descriptor has been given. So the sign extension (or the user
accidently passing garbage for the upper 32 bits) caused the
CLONE_INTO_CGROUP bit to be raised and the kernel to error out when it
didn't find a valid cgroup file descriptor.
Let's fix this by always capping the upper 32 bits for all codepaths that
are not aware of clone3() features. This ensures that we can't reach
clone3() only features by accident via legacy clone as with the sign
extension case and also that legacy clone() works exactly like before, i.e.
ignoring any unknown flags. This solution risks no regressions and is also
pretty clean.
Fixes: 7f192e3cd316 ("fork: add clone3")
Fixes: ef2c41cf38a7 ("clone3: allow spawning processes into cgroups")
Reported-by: Jan Stancek <jstancek@redhat.com>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Dmitry V. Levin <ldv@altlinux.org>
Cc: Andreas Schwab <schwab@linux-m68k.org>
Cc: Florian Weimer <fw@deneb.enyo.de>
Cc: libc-alpha@sourceware.org
Cc: stable@vger.kernel.org # 5.3+
Link: https://sourceware.org/pipermail/libc-alpha/2020-May/113596.html
Link: https://lore.kernel.org/r/20200507103214.77218-1-christian.brauner@ubuntu.com
2020-05-07 13:32:14 +03:00
. flags = ( lower_32_bits ( clone_flags ) & ~ CSIGNAL ) ,
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
. pidfd = parent_tidptr ,
. child_tid = child_tidptr ,
. parent_tid = parent_tidptr ,
fork: prevent accidental access to clone3 features
Jan reported an issue where an interaction between sign-extending clone's
flag argument on ppc64le and the new CLONE_INTO_CGROUP feature causes
clone() to consistently fail with EBADF.
The whole story is a little longer. The legacy clone() syscall is odd in a
bunch of ways and here two things interact. First, legacy clone's flag
argument is word-size dependent, i.e. it's an unsigned long whereas most
system calls with flag arguments use int or unsigned int. Second, legacy
clone() ignores unknown and deprecated flags. The two of them taken
together means that users on 64bit systems can pass garbage for the upper
32bit of the clone() syscall since forever and things would just work fine.
Just try this on a 64bit kernel prior to v5.7-rc1 where this will succeed
and on v5.7-rc1 where this will fail with EBADF:
int main(int argc, char *argv[])
{
pid_t pid;
/* Note that legacy clone() has different argument ordering on
* different architectures so this won't work everywhere.
*
* Only set the upper 32 bits.
*/
pid = syscall(__NR_clone, 0xffffffff00000000 | SIGCHLD,
NULL, NULL, NULL, NULL);
if (pid < 0)
exit(EXIT_FAILURE);
if (pid == 0)
exit(EXIT_SUCCESS);
if (wait(NULL) != pid)
exit(EXIT_FAILURE);
exit(EXIT_SUCCESS);
}
Since legacy clone() couldn't be extended this was not a problem so far and
nobody really noticed or cared since nothing in the kernel ever bothered to
look at the upper 32 bits.
But once we introduced clone3() and expanded the flag argument in struct
clone_args to 64 bit we opened this can of worms. With the first flag-based
extension to clone3() making use of the upper 32 bits of the flag argument
we've effectively made it possible for the legacy clone() syscall to reach
clone3() only flags. The sign extension scenario is just the odd
corner-case that we needed to figure this out.
The reason we just realized this now and not already when we introduced
CLONE_CLEAR_SIGHAND was that CLONE_INTO_CGROUP assumes that a valid cgroup
file descriptor has been given. So the sign extension (or the user
accidently passing garbage for the upper 32 bits) caused the
CLONE_INTO_CGROUP bit to be raised and the kernel to error out when it
didn't find a valid cgroup file descriptor.
Let's fix this by always capping the upper 32 bits for all codepaths that
are not aware of clone3() features. This ensures that we can't reach
clone3() only features by accident via legacy clone as with the sign
extension case and also that legacy clone() works exactly like before, i.e.
ignoring any unknown flags. This solution risks no regressions and is also
pretty clean.
Fixes: 7f192e3cd316 ("fork: add clone3")
Fixes: ef2c41cf38a7 ("clone3: allow spawning processes into cgroups")
Reported-by: Jan Stancek <jstancek@redhat.com>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Dmitry V. Levin <ldv@altlinux.org>
Cc: Andreas Schwab <schwab@linux-m68k.org>
Cc: Florian Weimer <fw@deneb.enyo.de>
Cc: libc-alpha@sourceware.org
Cc: stable@vger.kernel.org # 5.3+
Link: https://sourceware.org/pipermail/libc-alpha/2020-May/113596.html
Link: https://lore.kernel.org/r/20200507103214.77218-1-christian.brauner@ubuntu.com
2020-05-07 13:32:14 +03:00
. exit_signal = ( lower_32_bits ( clone_flags ) & CSIGNAL ) ,
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
. stack = newsp ,
. tls = tls ,
} ;
2020-08-19 13:46:45 +03:00
return kernel_clone ( & args ) ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
}
2019-06-21 02:26:35 +03:00
# endif
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
2019-06-21 02:26:35 +03:00
# ifdef __ARCH_WANT_SYS_CLONE3
2020-01-02 20:24:13 +03:00
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
noinline static int copy_clone_args_from_user ( struct kernel_clone_args * kargs ,
struct clone_args __user * uargs ,
2019-10-01 04:10:53 +03:00
size_t usize )
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
{
2019-10-01 04:10:53 +03:00
int err ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
struct clone_args args ;
2019-11-15 15:36:20 +03:00
pid_t * kset_tid = kargs - > set_tid ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
2020-04-12 23:26:58 +03:00
BUILD_BUG_ON ( offsetofend ( struct clone_args , tls ) ! =
CLONE_ARGS_SIZE_VER0 ) ;
BUILD_BUG_ON ( offsetofend ( struct clone_args , set_tid_size ) ! =
CLONE_ARGS_SIZE_VER1 ) ;
BUILD_BUG_ON ( offsetofend ( struct clone_args , cgroup ) ! =
CLONE_ARGS_SIZE_VER2 ) ;
BUILD_BUG_ON ( sizeof ( struct clone_args ) ! = CLONE_ARGS_SIZE_VER2 ) ;
2019-10-01 04:10:53 +03:00
if ( unlikely ( usize > PAGE_SIZE ) )
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
return - E2BIG ;
2019-10-01 04:10:53 +03:00
if ( unlikely ( usize < CLONE_ARGS_SIZE_VER0 ) )
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
return - EINVAL ;
2019-10-01 04:10:53 +03:00
err = copy_struct_from_user ( & args , sizeof ( args ) , uargs , usize ) ;
if ( err )
return err ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
2019-11-15 15:36:20 +03:00
if ( unlikely ( args . set_tid_size > MAX_PID_NS_LEVEL ) )
return - EINVAL ;
if ( unlikely ( ! args . set_tid & & args . set_tid_size > 0 ) )
return - EINVAL ;
if ( unlikely ( args . set_tid & & args . set_tid_size = = 0 ) )
return - EINVAL ;
fork: block invalid exit signals with clone3()
Previously, higher 32 bits of exit_signal fields were lost when copied
to the kernel args structure (that uses int as a type for the respective
field). Moreover, as Oleg has noted, exit_signal is used unchecked, so
it has to be checked for sanity before use; for the legacy syscalls,
applying CSIGNAL mask guarantees that it is at least non-negative;
however, there's no such thing is done in clone3() code path, and that
can break at least thread_group_leader.
This commit adds a check to copy_clone_args_from_user() to verify that
the exit signal is limited by CSIGNAL as with legacy clone() and that
the signal is valid. With this we don't get the legacy clone behavior
were an invalid signal could be handed down and would only be detected
and ignored in do_notify_parent(). Users of clone3() will now get a
proper error when they pass an invalid exit signal. Note, that this is
not user-visible behavior since no kernel with clone3() has been
released yet.
The following program will cause a splat on a non-fixed clone3() version
and will fail correctly on a fixed version:
#define _GNU_SOURCE
#include <linux/sched.h>
#include <linux/types.h>
#include <sched.h>
#include <stdio.h>
#include <stdlib.h>
#include <sys/syscall.h>
#include <sys/wait.h>
#include <unistd.h>
int main(int argc, char *argv[])
{
pid_t pid = -1;
struct clone_args args = {0};
args.exit_signal = -1;
pid = syscall(__NR_clone3, &args, sizeof(struct clone_args));
if (pid < 0)
exit(EXIT_FAILURE);
if (pid == 0)
exit(EXIT_SUCCESS);
wait(NULL);
exit(EXIT_SUCCESS);
}
Fixes: 7f192e3cd316 ("fork: add clone3")
Reported-by: Oleg Nesterov <oleg@redhat.com>
Suggested-by: Oleg Nesterov <oleg@redhat.com>
Suggested-by: Dmitry V. Levin <ldv@altlinux.org>
Signed-off-by: Eugene Syromiatnikov <esyr@redhat.com>
Link: https://lore.kernel.org/r/4b38fa4ce420b119a4c6345f42fe3cec2de9b0b5.1568223594.git.esyr@redhat.com
[christian.brauner@ubuntu.com: simplify check and rework commit message]
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
2019-09-11 20:45:40 +03:00
/*
* Verify that higher 32 bits of exit_signal are unset and that
* it is a valid signal
*/
if ( unlikely ( ( args . exit_signal & ~ ( ( u64 ) CSIGNAL ) ) | |
! valid_signal ( args . exit_signal ) ) )
return - EINVAL ;
2020-04-12 23:31:23 +03:00
if ( ( args . flags & CLONE_INTO_CGROUP ) & &
( args . cgroup > INT_MAX | | usize < CLONE_ARGS_SIZE_VER2 ) )
2020-02-05 16:26:22 +03:00
return - EINVAL ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
* kargs = ( struct kernel_clone_args ) {
. flags = args . flags ,
. pidfd = u64_to_user_ptr ( args . pidfd ) ,
. child_tid = u64_to_user_ptr ( args . child_tid ) ,
. parent_tid = u64_to_user_ptr ( args . parent_tid ) ,
. exit_signal = args . exit_signal ,
. stack = args . stack ,
. stack_size = args . stack_size ,
. tls = args . tls ,
2019-11-15 15:36:20 +03:00
. set_tid_size = args . set_tid_size ,
2020-02-05 16:26:22 +03:00
. cgroup = args . cgroup ,
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
} ;
2019-11-15 15:36:20 +03:00
if ( args . set_tid & &
copy_from_user ( kset_tid , u64_to_user_ptr ( args . set_tid ) ,
( kargs - > set_tid_size * sizeof ( pid_t ) ) ) )
return - EFAULT ;
kargs - > set_tid = kset_tid ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
return 0 ;
}
clone3: validate stack arguments
Validate the stack arguments and setup the stack depening on whether or not
it is growing down or up.
Legacy clone() required userspace to know in which direction the stack is
growing and pass down the stack pointer appropriately. To make things more
confusing microblaze uses a variant of the clone() syscall selected by
CONFIG_CLONE_BACKWARDS3 that takes an additional stack_size argument.
IA64 has a separate clone2() syscall which also takes an additional
stack_size argument. Finally, parisc has a stack that is growing upwards.
Userspace therefore has a lot nasty code like the following:
#define __STACK_SIZE (8 * 1024 * 1024)
pid_t sys_clone(int (*fn)(void *), void *arg, int flags, int *pidfd)
{
pid_t ret;
void *stack;
stack = malloc(__STACK_SIZE);
if (!stack)
return -ENOMEM;
#ifdef __ia64__
ret = __clone2(fn, stack, __STACK_SIZE, flags | SIGCHLD, arg, pidfd);
#elif defined(__parisc__) /* stack grows up */
ret = clone(fn, stack, flags | SIGCHLD, arg, pidfd);
#else
ret = clone(fn, stack + __STACK_SIZE, flags | SIGCHLD, arg, pidfd);
#endif
return ret;
}
or even crazier variants such as [3].
With clone3() we have the ability to validate the stack. We can check that
when stack_size is passed, the stack pointer is valid and the other way
around. We can also check that the memory area userspace gave us is fine to
use via access_ok(). Furthermore, we probably should not require
userspace to know in which direction the stack is growing. It is easy
for us to do this in the kernel and I couldn't find the original
reasoning behind exposing this detail to userspace.
/* Intentional user visible API change */
clone3() was released with 5.3. Currently, it is not documented and very
unclear to userspace how the stack and stack_size argument have to be
passed. After talking to glibc folks we concluded that trying to change
clone3() to setup the stack instead of requiring userspace to do this is
the right course of action.
Note, that this is an explicit change in user visible behavior we introduce
with this patch. If it breaks someone's use-case we will revert! (And then
e.g. place the new behavior under an appropriate flag.)
Breaking someone's use-case is very unlikely though. First, neither glibc
nor musl currently expose a wrapper for clone3(). Second, there is no real
motivation for anyone to use clone3() directly since it does not provide
features that legacy clone doesn't. New features for clone3() will first
happen in v5.5 which is why v5.4 is still a good time to try and make that
change now and backport it to v5.3. Searches on [4] did not reveal any
packages calling clone3().
[1]: https://lore.kernel.org/r/CAG48ez3q=BeNcuVTKBN79kJui4vC6nw0Bfq6xc-i0neheT17TA@mail.gmail.com
[2]: https://lore.kernel.org/r/20191028172143.4vnnjpdljfnexaq5@wittgenstein
[3]: https://github.com/systemd/systemd/blob/5238e9575906297608ff802a27e2ff9effa3b338/src/basic/raw-clone.h#L31
[4]: https://codesearch.debian.net
Fixes: 7f192e3cd316 ("fork: add clone3")
Cc: Kees Cook <keescook@chromium.org>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: linux-api@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: <stable@vger.kernel.org> # 5.3
Cc: GNU C Library <libc-alpha@sourceware.org>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Aleksa Sarai <cyphar@cyphar.com>
Link: https://lore.kernel.org/r/20191031113608.20713-1-christian.brauner@ubuntu.com
2019-10-31 14:36:08 +03:00
/**
* clone3_stack_valid - check and prepare stack
* @ kargs : kernel clone args
*
* Verify that the stack arguments userspace gave us are sane .
* In addition , set the stack direction for userspace since it ' s easy for us to
* determine .
*/
static inline bool clone3_stack_valid ( struct kernel_clone_args * kargs )
{
if ( kargs - > stack = = 0 ) {
if ( kargs - > stack_size > 0 )
return false ;
} else {
if ( kargs - > stack_size = = 0 )
return false ;
if ( ! access_ok ( ( void __user * ) kargs - > stack , kargs - > stack_size ) )
return false ;
# if !defined(CONFIG_STACK_GROWSUP) && !defined(CONFIG_IA64)
kargs - > stack + = kargs - > stack_size ;
# endif
}
return true ;
}
static bool clone3_args_valid ( struct kernel_clone_args * kargs )
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
{
2019-10-14 13:45:37 +03:00
/* Verify that no unknown flags are passed along. */
2020-02-05 16:26:22 +03:00
if ( kargs - > flags &
~ ( CLONE_LEGACY_FLAGS | CLONE_CLEAR_SIGHAND | CLONE_INTO_CGROUP ) )
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
return false ;
/*
2021-05-07 04:04:28 +03:00
* - make the CLONE_DETACHED bit reusable for clone3
* - make the CSIGNAL bits reusable for clone3
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
*/
if ( kargs - > flags & ( CLONE_DETACHED | CSIGNAL ) )
return false ;
2019-10-14 13:45:37 +03:00
if ( ( kargs - > flags & ( CLONE_SIGHAND | CLONE_CLEAR_SIGHAND ) ) = =
( CLONE_SIGHAND | CLONE_CLEAR_SIGHAND ) )
return false ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
if ( ( kargs - > flags & ( CLONE_THREAD | CLONE_PARENT ) ) & &
kargs - > exit_signal )
return false ;
clone3: validate stack arguments
Validate the stack arguments and setup the stack depening on whether or not
it is growing down or up.
Legacy clone() required userspace to know in which direction the stack is
growing and pass down the stack pointer appropriately. To make things more
confusing microblaze uses a variant of the clone() syscall selected by
CONFIG_CLONE_BACKWARDS3 that takes an additional stack_size argument.
IA64 has a separate clone2() syscall which also takes an additional
stack_size argument. Finally, parisc has a stack that is growing upwards.
Userspace therefore has a lot nasty code like the following:
#define __STACK_SIZE (8 * 1024 * 1024)
pid_t sys_clone(int (*fn)(void *), void *arg, int flags, int *pidfd)
{
pid_t ret;
void *stack;
stack = malloc(__STACK_SIZE);
if (!stack)
return -ENOMEM;
#ifdef __ia64__
ret = __clone2(fn, stack, __STACK_SIZE, flags | SIGCHLD, arg, pidfd);
#elif defined(__parisc__) /* stack grows up */
ret = clone(fn, stack, flags | SIGCHLD, arg, pidfd);
#else
ret = clone(fn, stack + __STACK_SIZE, flags | SIGCHLD, arg, pidfd);
#endif
return ret;
}
or even crazier variants such as [3].
With clone3() we have the ability to validate the stack. We can check that
when stack_size is passed, the stack pointer is valid and the other way
around. We can also check that the memory area userspace gave us is fine to
use via access_ok(). Furthermore, we probably should not require
userspace to know in which direction the stack is growing. It is easy
for us to do this in the kernel and I couldn't find the original
reasoning behind exposing this detail to userspace.
/* Intentional user visible API change */
clone3() was released with 5.3. Currently, it is not documented and very
unclear to userspace how the stack and stack_size argument have to be
passed. After talking to glibc folks we concluded that trying to change
clone3() to setup the stack instead of requiring userspace to do this is
the right course of action.
Note, that this is an explicit change in user visible behavior we introduce
with this patch. If it breaks someone's use-case we will revert! (And then
e.g. place the new behavior under an appropriate flag.)
Breaking someone's use-case is very unlikely though. First, neither glibc
nor musl currently expose a wrapper for clone3(). Second, there is no real
motivation for anyone to use clone3() directly since it does not provide
features that legacy clone doesn't. New features for clone3() will first
happen in v5.5 which is why v5.4 is still a good time to try and make that
change now and backport it to v5.3. Searches on [4] did not reveal any
packages calling clone3().
[1]: https://lore.kernel.org/r/CAG48ez3q=BeNcuVTKBN79kJui4vC6nw0Bfq6xc-i0neheT17TA@mail.gmail.com
[2]: https://lore.kernel.org/r/20191028172143.4vnnjpdljfnexaq5@wittgenstein
[3]: https://github.com/systemd/systemd/blob/5238e9575906297608ff802a27e2ff9effa3b338/src/basic/raw-clone.h#L31
[4]: https://codesearch.debian.net
Fixes: 7f192e3cd316 ("fork: add clone3")
Cc: Kees Cook <keescook@chromium.org>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: linux-api@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: <stable@vger.kernel.org> # 5.3
Cc: GNU C Library <libc-alpha@sourceware.org>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Aleksa Sarai <cyphar@cyphar.com>
Link: https://lore.kernel.org/r/20191031113608.20713-1-christian.brauner@ubuntu.com
2019-10-31 14:36:08 +03:00
if ( ! clone3_stack_valid ( kargs ) )
return false ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
return true ;
}
2019-09-27 18:28:42 +03:00
/**
* clone3 - create a new process with specific properties
* @ uargs : argument structure
* @ size : size of @ uargs
*
* clone3 ( ) is the extensible successor to clone ( ) / clone2 ( ) .
* It takes a struct as argument that is versioned by its size .
*
* Return : On success , a positive PID for the child process .
* On error , a negative errno number .
*/
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
SYSCALL_DEFINE2 ( clone3 , struct clone_args __user * , uargs , size_t , size )
{
int err ;
struct kernel_clone_args kargs ;
2019-11-15 15:36:20 +03:00
pid_t set_tid [ MAX_PID_NS_LEVEL ] ;
kargs . set_tid = set_tid ;
fork: add clone3
This adds the clone3 system call.
As mentioned several times already (cf. [7], [8]) here's the promised
patchset for clone3().
We recently merged the CLONE_PIDFD patchset (cf. [1]). It took the last
free flag from clone().
Independent of the CLONE_PIDFD patchset a time namespace has been discussed
at Linux Plumber Conference last year and has been sent out and reviewed
(cf. [5]). It is expected that it will go upstream in the not too distant
future. However, it relies on the addition of the CLONE_NEWTIME flag to
clone(). The only other good candidate - CLONE_DETACHED - is currently not
recyclable as we have identified at least two large or widely used
codebases that currently pass this flag (cf. [2], [3], and [4]). Given that
CLONE_PIDFD grabbed the last clone() flag the time namespace is effectively
blocked. clone3() has the advantage that it will unblock this patchset
again. In general, clone3() is extensible and allows for the implementation
of new features.
The idea is to keep clone3() very simple and close to the original clone(),
specifically, to keep on supporting old clone()-based workloads.
We know there have been various creative proposals how a new process
creation syscall or even api is supposed to look like. Some people even
going so far as to argue that the traditional fork()+exec() split should be
abandoned in favor of an in-kernel version of spawn(). Independent of
whether or not we personally think spawn() is a good idea this patchset has
and does not want to have anything to do with this.
One stance we take is that there's no real good alternative to
clone()+exec() and we need and want to support this model going forward;
independent of spawn().
The following requirements guided clone3():
- bump the number of available flags
- move arguments that are currently passed as separate arguments
in clone() into a dedicated struct clone_args
- choose a struct layout that is easy to handle on 32 and on 64 bit
- choose a struct layout that is extensible
- give new flags that currently need to abuse another flag's dedicated
return argument in clone() their own dedicated return argument
(e.g. CLONE_PIDFD)
- use a separate kernel internal struct kernel_clone_args that is
properly typed according to current kernel conventions in fork.c and is
different from the uapi struct clone_args
- port _do_fork() to use kernel_clone_args so that all process creation
syscalls such as fork(), vfork(), clone(), and clone3() behave identical
(Arnd suggested, that we can probably also port do_fork() itself in a
separate patchset.)
- ease of transition for userspace from clone() to clone3()
This very much means that we do *not* remove functionality that userspace
currently relies on as the latter is a good way of creating a syscall
that won't be adopted.
- do not try to be clever or complex: keep clone3() as dumb as possible
In accordance with Linus suggestions (cf. [11]), clone3() has the following
signature:
/* uapi */
struct clone_args {
__aligned_u64 flags;
__aligned_u64 pidfd;
__aligned_u64 child_tid;
__aligned_u64 parent_tid;
__aligned_u64 exit_signal;
__aligned_u64 stack;
__aligned_u64 stack_size;
__aligned_u64 tls;
};
/* kernel internal */
struct kernel_clone_args {
u64 flags;
int __user *pidfd;
int __user *child_tid;
int __user *parent_tid;
int exit_signal;
unsigned long stack;
unsigned long stack_size;
unsigned long tls;
};
long sys_clone3(struct clone_args __user *uargs, size_t size)
clone3() cleanly supports all of the supported flags from clone() and thus
all legacy workloads.
The advantage of sticking close to the old clone() is the low cost for
userspace to switch to this new api. Quite a lot of userspace apis (e.g.
pthreads) are based on the clone() syscall. With the new clone3() syscall
supporting all of the old workloads and opening up the ability to add new
features should make switching to it for userspace more appealing. In
essence, glibc can just write a simple wrapper to switch from clone() to
clone3().
There has been some interest in this patchset already. We have received a
patch from the CRIU corner for clone3() that would set the PID/TID of a
restored process without /proc/sys/kernel/ns_last_pid to eliminate a race.
/* User visible differences to legacy clone() */
- CLONE_DETACHED will cause EINVAL with clone3()
- CSIGNAL is deprecated
It is superseeded by a dedicated "exit_signal" argument in struct
clone_args freeing up space for additional flags.
This is based on a suggestion from Andrei and Linus (cf. [9] and [10])
/* References */
[1]: b3e5838252665ee4cfa76b82bdf1198dca81e5be
[2]: https://dxr.mozilla.org/mozilla-central/source/security/sandbox/linux/SandboxFilter.cpp#343
[3]: https://git.musl-libc.org/cgit/musl/tree/src/thread/pthread_create.c#n233
[4]: https://sources.debian.org/src/blcr/0.8.5-2.3/cr_module/cr_dump_self.c/?hl=740#L740
[5]: https://lore.kernel.org/lkml/20190425161416.26600-1-dima@arista.com/
[6]: https://lore.kernel.org/lkml/20190425161416.26600-2-dima@arista.com/
[7]: https://lore.kernel.org/lkml/CAHrFyr5HxpGXA2YrKza-oB-GGwJCqwPfyhD-Y5wbktWZdt0sGQ@mail.gmail.com/
[8]: https://lore.kernel.org/lkml/20190524102756.qjsjxukuq2f4t6bo@brauner.io/
[9]: https://lore.kernel.org/lkml/20190529222414.GA6492@gmail.com/
[10]: https://lore.kernel.org/lkml/CAHk-=whQP-Ykxi=zSYaV9iXsHsENa+2fdj-zYKwyeyed63Lsfw@mail.gmail.com/
[11]: https://lore.kernel.org/lkml/CAHk-=wieuV4hGwznPsX-8E0G2FKhx3NjZ9X3dTKh5zKd+iqOBw@mail.gmail.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Christian Brauner <christian@brauner.io>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Serge Hallyn <serge@hallyn.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Cc: Jann Horn <jannh@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Adrian Reber <adrian@lisas.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrei Vagin <avagin@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: linux-api@vger.kernel.org
2019-05-25 12:36:41 +03:00
err = copy_clone_args_from_user ( & kargs , uargs , size ) ;
if ( err )
return err ;
if ( ! clone3_args_valid ( & kargs ) )
return - EINVAL ;
2020-08-19 13:46:45 +03:00
return kernel_clone ( & kargs ) ;
2012-10-23 21:17:59 +04:00
}
# endif
2017-01-30 18:06:11 +03:00
void walk_process_tree ( struct task_struct * top , proc_visitor visitor , void * data )
{
struct task_struct * leader , * parent , * child ;
int res ;
read_lock ( & tasklist_lock ) ;
leader = top = top - > group_leader ;
down :
for_each_thread ( leader , parent ) {
list_for_each_entry ( child , & parent - > children , sibling ) {
res = visitor ( child , data ) ;
if ( res ) {
if ( res < 0 )
goto out ;
leader = child ;
goto down ;
}
up :
;
}
}
if ( leader ! = top ) {
child = leader ;
parent = child - > real_parent ;
leader = parent - > group_leader ;
goto up ;
}
out :
read_unlock ( & tasklist_lock ) ;
}
2006-01-12 00:46:15 +03:00
# ifndef ARCH_MIN_MMSTRUCT_ALIGN
# define ARCH_MIN_MMSTRUCT_ALIGN 0
# endif
2008-07-26 06:45:34 +04:00
static void sighand_ctor ( void * data )
2006-03-29 04:11:12 +04:00
{
struct sighand_struct * sighand = data ;
2007-05-17 09:10:57 +04:00
spin_lock_init ( & sighand - > siglock ) ;
2007-09-20 23:40:16 +04:00
init_waitqueue_head ( & sighand - > signalfd_wqh ) ;
2006-03-29 04:11:12 +04:00
}
2005-04-17 02:20:36 +04:00
void __init proc_caches_init ( void )
{
2018-07-16 22:03:31 +03:00
unsigned int mm_size ;
2005-04-17 02:20:36 +04:00
sighand_cachep = kmem_cache_create ( " sighand_cache " ,
sizeof ( struct sighand_struct ) , 0 ,
2017-01-18 13:53:44 +03:00
SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_TYPESAFE_BY_RCU |
2017-11-16 04:35:54 +03:00
SLAB_ACCOUNT , sighand_ctor ) ;
2005-04-17 02:20:36 +04:00
signal_cachep = kmem_cache_create ( " signal_cache " ,
sizeof ( struct signal_struct ) , 0 ,
2017-11-16 04:35:54 +03:00
SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_ACCOUNT ,
2016-01-15 02:18:21 +03:00
NULL ) ;
2007-07-20 05:11:58 +04:00
files_cachep = kmem_cache_create ( " files_cache " ,
2005-04-17 02:20:36 +04:00
sizeof ( struct files_struct ) , 0 ,
2017-11-16 04:35:54 +03:00
SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_ACCOUNT ,
2016-01-15 02:18:21 +03:00
NULL ) ;
2007-07-20 05:11:58 +04:00
fs_cachep = kmem_cache_create ( " fs_cache " ,
2005-04-17 02:20:36 +04:00
sizeof ( struct fs_struct ) , 0 ,
2017-11-16 04:35:54 +03:00
SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_ACCOUNT ,
2016-01-15 02:18:21 +03:00
NULL ) ;
2018-07-16 22:03:31 +03:00
2011-05-29 22:32:28 +04:00
/*
2018-07-16 22:03:31 +03:00
* The mm_cpumask is located at the end of mm_struct , and is
* dynamically sized based on the maximum CPU number this system
* can have , taking hotplug into account ( nr_cpu_ids ) .
2011-05-29 22:32:28 +04:00
*/
2018-07-16 22:03:31 +03:00
mm_size = sizeof ( struct mm_struct ) + cpumask_size ( ) ;
2017-08-16 02:45:00 +03:00
mm_cachep = kmem_cache_create_usercopy ( " mm_struct " ,
2018-07-16 22:03:31 +03:00
mm_size , ARCH_MIN_MMSTRUCT_ALIGN ,
2017-11-16 04:35:54 +03:00
SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_ACCOUNT ,
2017-08-16 02:45:00 +03:00
offsetof ( struct mm_struct , saved_auxv ) ,
sizeof_field ( struct mm_struct , saved_auxv ) ,
2016-01-15 02:18:21 +03:00
NULL ) ;
vm_area_cachep = KMEM_CACHE ( vm_area_struct , SLAB_PANIC | SLAB_ACCOUNT ) ;
2009-01-08 15:04:47 +03:00
mmap_init ( ) ;
2011-06-28 23:41:10 +04:00
nsproxy_cache_init ( ) ;
2005-04-17 02:20:36 +04:00
}
2006-02-07 23:58:58 +03:00
/*
2011-03-23 02:34:09 +03:00
* Check constraints on flags passed to the unshare system call .
2006-02-07 23:58:58 +03:00
*/
2011-03-23 02:34:09 +03:00
static int check_unshare_flags ( unsigned long unshare_flags )
2006-02-07 23:58:58 +03:00
{
2011-03-23 02:34:09 +03:00
if ( unshare_flags & ~ ( CLONE_THREAD | CLONE_FS | CLONE_NEWNS | CLONE_SIGHAND |
CLONE_VM | CLONE_FILES | CLONE_SYSVSEM |
2010-03-03 02:41:50 +03:00
CLONE_NEWUTS | CLONE_NEWIPC | CLONE_NEWNET |
ns: Introduce Time Namespace
Time Namespace isolates clock values.
The kernel provides access to several clocks CLOCK_REALTIME,
CLOCK_MONOTONIC, CLOCK_BOOTTIME, etc.
CLOCK_REALTIME
System-wide clock that measures real (i.e., wall-clock) time.
CLOCK_MONOTONIC
Clock that cannot be set and represents monotonic time since
some unspecified starting point.
CLOCK_BOOTTIME
Identical to CLOCK_MONOTONIC, except it also includes any time
that the system is suspended.
For many users, the time namespace means the ability to changes date and
time in a container (CLOCK_REALTIME). Providing per namespace notions of
CLOCK_REALTIME would be complex with a massive overhead, but has a dubious
value.
But in the context of checkpoint/restore functionality, monotonic and
boottime clocks become interesting. Both clocks are monotonic with
unspecified starting points. These clocks are widely used to measure time
slices and set timers. After restoring or migrating processes, it has to be
guaranteed that they never go backward. In an ideal case, the behavior of
these clocks should be the same as for a case when a whole system is
suspended. All this means that it is required to set CLOCK_MONOTONIC and
CLOCK_BOOTTIME clocks, which can be achieved by adding per-namespace
offsets for clocks.
A time namespace is similar to a pid namespace in the way how it is
created: unshare(CLONE_NEWTIME) system call creates a new time namespace,
but doesn't set it to the current process. Then all children of the process
will be born in the new time namespace, or a process can use the setns()
system call to join a namespace.
This scheme allows setting clock offsets for a namespace, before any
processes appear in it.
All available clone flags have been used, so CLONE_NEWTIME uses the highest
bit of CSIGNAL. It means that it can be used only with the unshare() and
the clone3() system calls.
[ tglx: Adjusted paragraph about clone3() to reality and massaged the
changelog a bit. ]
Co-developed-by: Dmitry Safonov <dima@arista.com>
Signed-off-by: Andrei Vagin <avagin@gmail.com>
Signed-off-by: Dmitry Safonov <dima@arista.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://criu.org/Time_namespace
Link: https://lists.openvz.org/pipermail/criu/2018-June/041504.html
Link: https://lore.kernel.org/r/20191112012724.250792-4-dima@arista.com
2019-11-12 04:26:52 +03:00
CLONE_NEWUSER | CLONE_NEWPID | CLONE_NEWCGROUP |
CLONE_NEWTIME ) )
2011-03-23 02:34:09 +03:00
return - EINVAL ;
2006-02-07 23:58:58 +03:00
/*
2015-08-11 01:35:07 +03:00
* Not implemented , but pretend it works if there is nothing
* to unshare . Note that unsharing the address space or the
* signal handlers also need to unshare the signal queues ( aka
* CLONE_THREAD ) .
2006-02-07 23:58:58 +03:00
*/
2011-03-23 02:34:09 +03:00
if ( unshare_flags & ( CLONE_THREAD | CLONE_SIGHAND | CLONE_VM ) ) {
2015-08-11 01:35:07 +03:00
if ( ! thread_group_empty ( current ) )
return - EINVAL ;
}
if ( unshare_flags & ( CLONE_SIGHAND | CLONE_VM ) ) {
2019-01-18 15:27:26 +03:00
if ( refcount_read ( & current - > sighand - > count ) > 1 )
2015-08-11 01:35:07 +03:00
return - EINVAL ;
}
if ( unshare_flags & CLONE_VM ) {
if ( ! current_is_single_threaded ( ) )
2011-03-23 02:34:09 +03:00
return - EINVAL ;
}
2006-02-07 23:58:58 +03:00
return 0 ;
}
/*
2006-02-07 23:58:59 +03:00
* Unshare the filesystem structure if it is being shared
2006-02-07 23:58:58 +03:00
*/
static int unshare_fs ( unsigned long unshare_flags , struct fs_struct * * new_fsp )
{
struct fs_struct * fs = current - > fs ;
2009-03-30 15:20:30 +04:00
if ( ! ( unshare_flags & CLONE_FS ) | | ! fs )
return 0 ;
/* don't need lock here; in the worst case we'll do useless copy */
if ( fs - > users = = 1 )
return 0 ;
* new_fsp = copy_fs_struct ( fs ) ;
if ( ! * new_fsp )
return - ENOMEM ;
2006-02-07 23:58:58 +03:00
return 0 ;
}
/*
2006-02-07 23:59:02 +03:00
* Unshare file descriptor table if it is being shared
2006-02-07 23:58:58 +03:00
*/
2020-06-03 22:48:55 +03:00
int unshare_fd ( unsigned long unshare_flags , unsigned int max_fds ,
struct files_struct * * new_fdp )
2006-02-07 23:58:58 +03:00
{
struct files_struct * fd = current - > files ;
2006-02-07 23:59:02 +03:00
int error = 0 ;
2006-02-07 23:58:58 +03:00
if ( ( unshare_flags & CLONE_FILES ) & &
2006-02-07 23:59:02 +03:00
( fd & & atomic_read ( & fd - > count ) > 1 ) ) {
2020-06-03 22:48:55 +03:00
* new_fdp = dup_fd ( fd , max_fds , & error ) ;
2006-02-07 23:59:02 +03:00
if ( ! * new_fdp )
return error ;
}
2006-02-07 23:58:58 +03:00
return 0 ;
}
/*
* unshare allows a process to ' unshare ' part of the process
* context which was originally shared using clone . copy_ *
2020-08-19 13:46:45 +03:00
* functions used by kernel_clone ( ) cannot be used here directly
2006-02-07 23:58:58 +03:00
* because they modify an inactive task_struct that is being
* constructed . Here we are modifying the current , active ,
* task_struct .
*/
2018-03-11 13:34:42 +03:00
int ksys_unshare ( unsigned long unshare_flags )
2006-02-07 23:58:58 +03:00
{
struct fs_struct * fs , * new_fs = NULL ;
struct files_struct * fd , * new_fd = NULL ;
2012-07-26 16:15:35 +04:00
struct cred * new_cred = NULL ;
2007-10-19 10:39:54 +04:00
struct nsproxy * new_nsproxy = NULL ;
2008-04-29 12:00:57 +04:00
int do_sysvsem = 0 ;
2011-03-23 02:34:09 +03:00
int err ;
2006-02-07 23:58:58 +03:00
2012-07-26 16:15:35 +04:00
/*
2015-08-11 02:25:44 +03:00
* If unsharing a user namespace must also unshare the thread group
* and unshare the filesystem root and working directories .
2012-07-26 16:15:35 +04:00
*/
if ( unshare_flags & CLONE_NEWUSER )
2013-03-13 22:51:49 +04:00
unshare_flags | = CLONE_THREAD | CLONE_FS ;
2010-03-03 02:41:50 +03:00
/*
* If unsharing vm , must also unshare signal handlers .
*/
if ( unshare_flags & CLONE_VM )
unshare_flags | = CLONE_SIGHAND ;
2015-08-11 01:35:07 +03:00
/*
* If unsharing a signal handlers , must also unshare the signal queues .
*/
if ( unshare_flags & CLONE_SIGHAND )
unshare_flags | = CLONE_THREAD ;
2011-03-23 02:34:09 +03:00
/*
* If unsharing namespace , must also unshare filesystem information .
*/
if ( unshare_flags & CLONE_NEWNS )
unshare_flags | = CLONE_FS ;
2010-03-03 02:41:50 +03:00
err = check_unshare_flags ( unshare_flags ) ;
if ( err )
goto bad_unshare_out ;
2008-04-29 12:00:59 +04:00
/*
* CLONE_NEWIPC must also detach from the undolist : after switching
* to a new ipc namespace , the semaphore arrays from the old
* namespace are unreachable .
*/
if ( unshare_flags & ( CLONE_NEWIPC | CLONE_SYSVSEM ) )
2008-04-29 12:00:57 +04:00
do_sysvsem = 1 ;
2011-07-27 03:08:39 +04:00
err = unshare_fs ( unshare_flags , & new_fs ) ;
if ( err )
2011-03-23 02:34:09 +03:00
goto bad_unshare_out ;
2020-06-03 22:48:55 +03:00
err = unshare_fd ( unshare_flags , NR_OPEN_MAX , & new_fd ) ;
2011-07-27 03:08:39 +04:00
if ( err )
2011-03-23 02:34:09 +03:00
goto bad_unshare_cleanup_fs ;
2012-07-26 16:15:35 +04:00
err = unshare_userns ( unshare_flags , & new_cred ) ;
2011-07-27 03:08:39 +04:00
if ( err )
2008-04-29 12:00:57 +04:00
goto bad_unshare_cleanup_fd ;
2012-07-26 16:15:35 +04:00
err = unshare_nsproxy_namespaces ( unshare_flags , & new_nsproxy ,
new_cred , new_fs ) ;
if ( err )
goto bad_unshare_cleanup_cred ;
2006-10-02 13:18:18 +04:00
2021-04-22 15:27:09 +03:00
if ( new_cred ) {
err = set_cred_ucounts ( new_cred ) ;
if ( err )
goto bad_unshare_cleanup_cred ;
}
2012-07-26 16:15:35 +04:00
if ( new_fs | | new_fd | | do_sysvsem | | new_cred | | new_nsproxy ) {
2008-04-29 12:00:57 +04:00
if ( do_sysvsem ) {
/*
* CLONE_SYSVSEM is equivalent to sys_exit ( ) .
*/
exit_sem ( current ) ;
}
shm: make exit_shm work proportional to task activity
This is small set of patches our team has had kicking around for a few
versions internally that fixes tasks getting hung on shm_exit when there
are many threads hammering it at once.
Anton wrote a simple test to cause the issue:
http://ozlabs.org/~anton/junkcode/bust_shm_exit.c
Before applying this patchset, this test code will cause either hanging
tracebacks or pthread out of memory errors.
After this patchset, it will still produce output like:
root@somehost:~# ./bust_shm_exit 1024 160
...
INFO: rcu_sched detected stalls on CPUs/tasks: {} (detected by 116, t=2111 jiffies, g=241, c=240, q=7113)
INFO: Stall ended before state dump start
...
But the task will continue to run along happily, so we consider this an
improvement over hanging, even if it's a bit noisy.
This patch (of 3):
exit_shm obtains the ipc_ns shm rwsem for write and holds it while it
walks every shared memory segment in the namespace. Thus the amount of
work is related to the number of shm segments in the namespace not the
number of segments that might need to be cleaned.
In addition, this occurs after the task has been notified the thread has
exited, so the number of tasks waiting for the ns shm rwsem can grow
without bound until memory is exausted.
Add a list to the task struct of all shmids allocated by this task. Init
the list head in copy_process. Use the ns->rwsem for locking. Add
segments after id is added, remove before removing from id.
On unshare of NEW_IPCNS orphan any ids as if the task had exited, similar
to handling of semaphore undo.
I chose a define for the init sequence since its a simple list init,
otherwise it would require a function call to avoid include loops between
the semaphore code and the task struct. Converting the list_del to
list_del_init for the unshare cases would remove the exit followed by
init, but I left it blow up if not inited.
Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: Jack Miller <millerjo@us.ibm.com>
Cc: Davidlohr Bueso <davidlohr@hp.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Anton Blanchard <anton@samba.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-08-09 01:23:19 +04:00
if ( unshare_flags & CLONE_NEWIPC ) {
/* Orphan segments in old ns (see sem above). */
exit_shm ( current ) ;
shm_init_task ( current ) ;
}
2006-10-02 13:18:06 +04:00
2013-02-28 05:03:23 +04:00
if ( new_nsproxy )
2007-10-19 10:39:54 +04:00
switch_task_namespaces ( current , new_nsproxy ) ;
2006-02-07 23:58:58 +03:00
2007-10-19 10:39:54 +04:00
task_lock ( current ) ;
2006-02-07 23:58:58 +03:00
if ( new_fs ) {
fs = current - > fs ;
2010-08-17 22:37:33 +04:00
spin_lock ( & fs - > lock ) ;
2006-02-07 23:58:58 +03:00
current - > fs = new_fs ;
2009-03-30 15:20:30 +04:00
if ( - - fs - > users )
new_fs = NULL ;
else
new_fs = fs ;
2010-08-17 22:37:33 +04:00
spin_unlock ( & fs - > lock ) ;
2006-02-07 23:58:58 +03:00
}
if ( new_fd ) {
fd = current - > files ;
current - > files = new_fd ;
new_fd = fd ;
}
task_unlock ( current ) ;
2012-07-26 16:15:35 +04:00
if ( new_cred ) {
/* Install the new user namespace */
commit_creds ( new_cred ) ;
new_cred = NULL ;
}
2006-02-07 23:58:58 +03:00
}
2017-03-07 23:41:36 +03:00
perf_event_namespaces ( current ) ;
2012-07-26 16:15:35 +04:00
bad_unshare_cleanup_cred :
if ( new_cred )
put_cred ( new_cred ) ;
2006-02-07 23:58:58 +03:00
bad_unshare_cleanup_fd :
if ( new_fd )
put_files_struct ( new_fd ) ;
bad_unshare_cleanup_fs :
if ( new_fs )
2009-03-30 15:20:30 +04:00
free_fs_struct ( new_fs ) ;
2006-02-07 23:58:58 +03:00
bad_unshare_out :
return err ;
}
2008-04-22 13:31:30 +04:00
2018-03-11 13:34:42 +03:00
SYSCALL_DEFINE1 ( unshare , unsigned long , unshare_flags )
{
return ksys_unshare ( unshare_flags ) ;
}
2008-04-22 13:31:30 +04:00
/*
* Helper to unshare the files of the current task .
* We don ' t want to expose copy_files internals to
* the exec layer of the kernel .
*/
2020-11-21 02:14:19 +03:00
int unshare_files ( void )
2008-04-22 13:31:30 +04:00
{
struct task_struct * task = current ;
2020-11-21 02:14:19 +03:00
struct files_struct * old , * copy = NULL ;
2008-04-22 13:31:30 +04:00
int error ;
2020-06-03 22:48:55 +03:00
error = unshare_fd ( CLONE_FILES , NR_OPEN_MAX , & copy ) ;
2020-11-21 02:14:19 +03:00
if ( error | | ! copy )
2008-04-22 13:31:30 +04:00
return error ;
2020-11-21 02:14:19 +03:00
old = task - > files ;
2008-04-22 13:31:30 +04:00
task_lock ( task ) ;
task - > files = copy ;
task_unlock ( task ) ;
2020-11-21 02:14:19 +03:00
put_files_struct ( old ) ;
2008-04-22 13:31:30 +04:00
return 0 ;
}
2015-04-16 22:47:50 +03:00
int sysctl_max_threads ( struct ctl_table * table , int write ,
2020-09-05 02:35:49 +03:00
void * buffer , size_t * lenp , loff_t * ppos )
2015-04-16 22:47:50 +03:00
{
struct ctl_table t ;
int ret ;
int threads = max_threads ;
2019-10-07 03:58:19 +03:00
int min = 1 ;
2015-04-16 22:47:50 +03:00
int max = MAX_THREADS ;
t = * table ;
t . data = & threads ;
t . extra1 = & min ;
t . extra2 = & max ;
ret = proc_dointvec_minmax ( & t , write , buffer , lenp , ppos ) ;
if ( ret | | ! write )
return ret ;
2019-10-07 03:58:19 +03:00
max_threads = threads ;
2015-04-16 22:47:50 +03:00
return 0 ;
}
