IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
You can pass opaque pointers directly.
I also renamed 'va' and 'vb' into more meaningful arguments.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Jessica Yu <jeyu@kernel.org>
Commit 8651ec01daed ("module: add support for symbol namespaces.")
broke linking for arm64 defconfig:
| lib/crypto/arc4.o: In function `__ksymtab_arc4_setkey':
| arc4.c:(___ksymtab+arc4_setkey+0x8): undefined reference to `no symbol'
| lib/crypto/arc4.o: In function `__ksymtab_arc4_crypt':
| arc4.c:(___ksymtab+arc4_crypt+0x8): undefined reference to `no symbol'
This is because the dummy initialisation of the 'namespace_offset' field
in 'struct kernel_symbol' when using EXPORT_SYMBOL on architectures with
support for PREL32 locations uses an offset from an absolute address (0)
in an effort to trick 'offset_to_pointer' into behaving as a NOP,
allowing non-namespaced symbols to be treated in the same way as those
belonging to a namespace.
Unfortunately, place-relative relocations require a symbol reference
rather than an absolute value and, although x86 appears to get away with
this due to placing the kernel text at the top of the address space, it
almost certainly results in a runtime failure if the kernel is relocated
dynamically as a result of KASLR.
Rework 'namespace_offset' so that a value of 0, which cannot occur for a
valid namespaced symbol, indicates that the corresponding symbol does
not belong to a namespace.
Cc: Matthias Maennich <maennich@google.com>
Cc: Jessica Yu <jeyu@kernel.org>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Fixes: 8651ec01daed ("module: add support for symbol namespaces.")
Reported-by: kbuild test robot <lkp@intel.com>
Tested-by: Matthias Maennich <maennich@google.com>
Tested-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Reviewed-by: Matthias Maennich <maennich@google.com>
Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Jessica Yu <jeyu@kernel.org>
This splits the size parameter to swiotlb_tbl_map_single() and
swiotlb_tbl_unmap_single() into an alloc_size and a mapping_size
parameter, where the latter one is rounded up to the iommu page
size.
Suggested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
It was recently discovered that the linux version of waitid is not a
superset of the other wait functions because it does not include support
for waiting for the current process group. This has two downsides:
1. An extra system call is needed to get the current process group.
2. After the current process group is received and before it is passed
to waitid a signal could arrive causing the current process group to change.
Inherent race-conditions as these make it impossible for userspace to
emulate this functionaly and thus violate async-signal safety
requirements for waitpid.
Arguments can be made for using a different choice of idtype and id
for this case but the BSDs already use this P_PGID and 0 to indicate
waiting for the current process's process group. So be nice to user
space programmers and don't introduce an unnecessary incompatibility.
Some people have noted that the posix description is that
waitpid will wait for the current process group, and that in
the presence of pthreads that process group can change. To get
clarity on this issue I looked at XNU, FreeBSD, and Luminos. All of
those flavors of unix waited for the current process group at the
time of call and as written could not adapt to the process group
changing after the call.
At one point Linux did adapt to the current process group changing but
that stopped in 161550d74c07 ("pid: sys_wait... fixes"). It has been
over 11 years since Linux has that behavior, no programs that fail
with the change in behavior have been reported, and I could not
find any other unix that does this. So I think it is safe to clarify
the definition of current process group, to current process group
at the time of the wait function.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Palmer Dabbelt <palmer@sifive.com>
Cc: Rich Felker <dalias@libc.org>
Cc: Alistair Francis <alistair23@gmail.com>
Cc: Zong Li <zongbox@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Cc: GNU C Library <libc-alpha@sourceware.org>
Link: https://lore.kernel.org/r/20190814154400.6371-2-christian.brauner@ubuntu.com
The recent consolidation of the three permission checks introduced a subtle
regression. For timer_create() with a process wide timer it returns the
current task if the lookup through the PID which is encoded into the
clockid results in returning current.
That's broken because it does not validate whether the current task is the
group leader.
That was caused by the two different variants of permission checks:
- posix_cpu_timer_get() allowed access to the process wide clock when the
looked up task is current. That's not an issue because the process wide
clock is in the shared sighand.
- posix_cpu_timer_create() made sure that the looked up task is the group
leader.
Restore the previous state.
Note, that these permission checks are more than questionable, but that's
subject to follow up changes.
Fixes: 6ae40e3fdcd3 ("posix-cpu-timers: Provide task validation functions")
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Link: https://lkml.kernel.org/r/alpine.DEB.2.21.1909052314110.1902@nanos.tec.linutronix.de
If MODULE_ALLOW_MISSING_NAMESPACE_IMPORTS is enabled (default=n), the
requirement for modules to import all namespaces that are used by
the module is relaxed.
Enabling this option effectively allows (invalid) modules to be loaded
while only a warning is emitted.
Disabling this option keeps the enforcement at module loading time and
loading is denied if the module's imports are not satisfactory.
Reviewed-by: Martijn Coenen <maco@android.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Matthias Maennich <maennich@google.com>
Signed-off-by: Jessica Yu <jeyu@kernel.org>
The EXPORT_SYMBOL_NS() and EXPORT_SYMBOL_NS_GPL() macros can be used to
export a symbol to a specific namespace. There are no _GPL_FUTURE and
_UNUSED variants because these are currently unused, and I'm not sure
they are necessary.
I didn't add EXPORT_SYMBOL_NS() for ASM exports; this patch sets the
namespace of ASM exports to NULL by default. In case of relative
references, it will be relocatable to NULL. If there's a need, this
should be pretty easy to add.
A module that wants to use a symbol exported to a namespace must add a
MODULE_IMPORT_NS() statement to their module code; otherwise, modpost
will complain when building the module, and the kernel module loader
will emit an error and fail when loading the module.
MODULE_IMPORT_NS() adds a modinfo tag 'import_ns' to the module. That
tag can be observed by the modinfo command, modpost and kernel/module.c
at the time of loading the module.
The ELF symbols are renamed to include the namespace with an asm label;
for example, symbol 'usb_stor_suspend' in namespace USB_STORAGE becomes
'usb_stor_suspend.USB_STORAGE'. This allows modpost to do namespace
checking, without having to go through all the effort of parsing ELF and
relocation records just to get to the struct kernel_symbols.
On x86_64 I saw no difference in binary size (compression), but at
runtime this will require a word of memory per export to hold the
namespace. An alternative could be to store namespaced symbols in their
own section and use a separate 'struct namespaced_kernel_symbol' for
that section, at the cost of making the module loader more complex.
Co-developed-by: Martijn Coenen <maco@android.com>
Signed-off-by: Martijn Coenen <maco@android.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Matthias Maennich <maennich@google.com>
Signed-off-by: Jessica Yu <jeyu@kernel.org>
Similar to modpost's get_next_modinfo(), introduce get_next_modinfo() in
kernel/module.c to acquire any further values associated with the same
modinfo tag name. That is useful for any tags that have multiple
occurrences (such as 'alias'), but is in particular introduced here as
part of the symbol namespaces patch series to read the (potentially)
multiple namespaces a module is importing.
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Reviewed-by: Martijn Coenen <maco@android.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Matthias Maennich <maennich@google.com>
Signed-off-by: Jessica Yu <jeyu@kernel.org>
In some special cases we must not block, but there's not a spinlock,
preempt-off, irqs-off or similar critical section already that arms the
might_sleep() debug checks. Add a non_block_start/end() pair to annotate
these.
This will be used in the oom paths of mmu-notifiers, where blocking is not
allowed to make sure there's forward progress. Quoting Michal:
"The notifier is called from quite a restricted context - oom_reaper -
which shouldn't depend on any locks or sleepable conditionals. The code
should be swift as well but we mostly do care about it to make a forward
progress. Checking for sleepable context is the best thing we could come
up with that would describe these demands at least partially."
Peter also asked whether we want to catch spinlocks on top, but Michal
said those are less of a problem because spinlocks can't have an indirect
dependency upon the page allocator and hence close the loop with the oom
reaper.
Suggested by Michal Hocko.
Link: https://lore.kernel.org/r/20190826201425.17547-4-daniel.vetter@ffwll.ch
Acked-by: Christian König <christian.koenig@amd.com> (v1)
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The powerpc version only supported 64 bit. Add some
code to switch decoding of fields during runtime so
we can kexec a 32 bit kernel from a 64 bit kernel and
vice versa.
Signed-off-by: Sven Schnelle <svens@stackframe.org>
Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com>
Signed-off-by: Helge Deller <deller@gmx.de>
base was never assigned, so we can remove it.
Reviewed-by: Christophe Leroy <christophe.leroy@c-s.fr>
Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com>
Signed-off-by: Sven Schnelle <svens@stackframe.org>
Signed-off-by: Helge Deller <deller@gmx.de>
It wasn't used anywhere, so lets drop it.
Reviewed-by: Christophe Leroy <christophe.leroy@c-s.fr>
Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com>
Signed-off-by: Sven Schnelle <svens@stackframe.org>
Signed-off-by: Helge Deller <deller@gmx.de>
It's not used anywhere so just drop it.
Signed-off-by: Sven Schnelle <svens@stackframe.org>
Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com>
Signed-off-by: Helge Deller <deller@gmx.de>
We're not using them, so we can drop the parsing.
Signed-off-by: Sven Schnelle <svens@stackframe.org>
Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com>
Signed-off-by: Helge Deller <deller@gmx.de>
Change the order to have a 64/32/16 order, no functional change.
Signed-off-by: Sven Schnelle <svens@stackframe.org>
Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com>
Signed-off-by: Helge Deller <deller@gmx.de>
Right now powerpc provides an implementation to read elf files
with the kexec_file_load() syscall. Make that available as a public
kexec interface so it can be re-used on other architectures.
Signed-off-by: Sven Schnelle <svens@stackframe.org>
Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com>
Signed-off-by: Helge Deller <deller@gmx.de>
Daniel Borkmann says:
====================
The following pull-request contains BPF updates for your *net-next* tree.
The main changes are:
1) Add the ability to use unaligned chunks in the AF_XDP umem. By
relaxing where the chunks can be placed, it allows to use an
arbitrary buffer size and place whenever there is a free
address in the umem. Helps more seamless DPDK AF_XDP driver
integration. Support for i40e, ixgbe and mlx5e, from Kevin and
Maxim.
2) Addition of a wakeup flag for AF_XDP tx and fill rings so the
application can wake up the kernel for rx/tx processing which
avoids busy-spinning of the latter, useful when app and driver
is located on the same core. Support for i40e, ixgbe and mlx5e,
from Magnus and Maxim.
3) bpftool fixes for printf()-like functions so compiler can actually
enforce checks, bpftool build system improvements for custom output
directories, and addition of 'bpftool map freeze' command, from Quentin.
4) Support attaching/detaching XDP programs from 'bpftool net' command,
from Daniel.
5) Automatic xskmap cleanup when AF_XDP socket is released, and several
barrier/{read,write}_once fixes in AF_XDP code, from Björn.
6) Relicense of bpf_helpers.h/bpf_endian.h for future libbpf
inclusion as well as libbpf versioning improvements, from Andrii.
7) Several new BPF kselftests for verifier precision tracking, from Alexei.
8) Several BPF kselftest fixes wrt endianess to run on s390x, from Ilya.
9) And more BPF kselftest improvements all over the place, from Stanislav.
10) Add simple BPF map op cache for nfp driver to batch dumps, from Jakub.
11) AF_XDP socket umem mapping improvements for 32bit archs, from Ivan.
12) Add BPF-to-BPF call and BTF line info support for s390x JIT, from Yauheni.
13) Small optimization in arm64 JIT to spare 1 insns for BPF_MOD, from Jerin.
14) Fix an error check in bpf_tcp_gen_syncookie() helper, from Petar.
15) Various minor fixes and cleanups, from Nathan, Masahiro, Masanari,
Peter, Wei, Yue.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
- Large GICv3 updates to support new PPI and SPI ranges
- Conver all alloc_fwnode() users to use PAs instead of VAs
- Add support for Marvell's MMP3 irqchip
- Add support for Amlogic Meson SM1
- Various cleanups and fixes
-----BEGIN PGP SIGNATURE-----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=PsGr
-----END PGP SIGNATURE-----
Merge tag 'irqchip-5.4' of git://git.kernel.org/pub/scm/linux/kernel/git/maz/arm-platforms into irq/core
Pull irqchip updates for Linux 5.4 from Marc Zyngier:
- Large GICv3 updates to support new PPI and SPI ranges
- Conver all alloc_fwnode() users to use PAs instead of VAs
- Add support for Marvell's MMP3 irqchip
- Add support for Amlogic Meson SM1
- Various cleanups and fixes
If we disable the compiler's auto-initialization feature, if
-fplugin-arg-structleak_plugin-byref or -ftrivial-auto-var-init=pattern
are disabled, arch_hw_breakpoint may be used before initialization after:
9a4903dde2c86 ("perf/hw_breakpoint: Split attribute parse and commit")
On our ARM platform, the struct step_ctrl in arch_hw_breakpoint, which
used to be zero-initialized by kzalloc(), may be used in
arch_install_hw_breakpoint() without initialization.
Signed-off-by: Mark-PK Tsai <mark-pk.tsai@mediatek.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alix Wu <alix.wu@mediatek.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: YJ Chiang <yj.chiang@mediatek.com>
Link: https://lkml.kernel.org/r/20190906060115.9460-1-mark-pk.tsai@mediatek.com
[ Minor edits. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The following crash was observed:
Unable to handle kernel NULL pointer dereference at 0000000000000158
Internal error: Oops: 96000004 [#1] SMP
pc : resend_irqs+0x68/0xb0
lr : resend_irqs+0x64/0xb0
...
Call trace:
resend_irqs+0x68/0xb0
tasklet_action_common.isra.6+0x84/0x138
tasklet_action+0x2c/0x38
__do_softirq+0x120/0x324
run_ksoftirqd+0x44/0x60
smpboot_thread_fn+0x1ac/0x1e8
kthread+0x134/0x138
ret_from_fork+0x10/0x18
The reason for this is that the interrupt resend mechanism happens in soft
interrupt context, which is a asynchronous mechanism versus other
operations on interrupts. free_irq() does not take resend handling into
account. Thus, the irq descriptor might be already freed before the resend
tasklet is executed. resend_irqs() does not check the return value of the
interrupt descriptor lookup and derefences the return value
unconditionally.
1):
__setup_irq
irq_startup
check_irq_resend // activate softirq to handle resend irq
2):
irq_domain_free_irqs
irq_free_descs
free_desc
call_rcu(&desc->rcu, delayed_free_desc)
3):
__do_softirq
tasklet_action
resend_irqs
desc = irq_to_desc(irq)
desc->handle_irq(desc) // desc is NULL --> Ooops
Fix this by adding a NULL pointer check in resend_irqs() before derefencing
the irq descriptor.
Fixes: a4633adcdbc1 ("[PATCH] genirq: add genirq sw IRQ-retrigger")
Signed-off-by: Yunfeng Ye <yeyunfeng@huawei.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/1630ae13-5c8e-901e-de09-e740b6a426a7@huawei.com
ENOTSUPP is not supposed to be returned to userspace. This was found on an
OpenPower machine, where the RTC does not support set_alarm.
On that system, a clock_nanosleep(CLOCK_REALTIME_ALARM, ...) results in
"524 Unknown error 524"
Replace it with EOPNOTSUPP which results in the expected "95 Operation not
supported" error.
Fixes: 1c6b39ad3f01 (alarmtimers: Return -ENOTSUPP if no RTC device is present)
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20190903171802.28314-1-cascardo@canonical.com
Add "gfp_t" support in synthetic_events, then the "gfp_t" type
parameter in some functions can be traced.
Prints the gfp flags as hex in addition to the human-readable flag
string. Example output:
whoopsie-630 [000] ...1 78.969452: testevent: bar=b20 (GFP_ATOMIC|__GFP_ZERO)
rcuc/0-11 [000] ...1 81.097555: testevent: bar=a20 (GFP_ATOMIC)
rcuc/0-11 [000] ...1 81.583123: testevent: bar=a20 (GFP_ATOMIC)
Link: http://lkml.kernel.org/r/20190712015308.9908-1-zhengjun.xing@linux.intel.com
Signed-off-by: Zhengjun Xing <zhengjun.xing@linux.intel.com>
[ Added printing of flag names ]
Signed-off-by: Tom Zanussi <zanussi@kernel.org>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
The problem can be seen in the following two tests:
0: (bf) r3 = r10
1: (55) if r3 != 0x7b goto pc+0
2: (7a) *(u64 *)(r3 -8) = 0
3: (79) r4 = *(u64 *)(r10 -8)
..
0: (85) call bpf_get_prandom_u32#7
1: (bf) r3 = r10
2: (55) if r3 != 0x7b goto pc+0
3: (7b) *(u64 *)(r3 -8) = r0
4: (79) r4 = *(u64 *)(r10 -8)
When backtracking need to mark R4 it will mark slot fp-8.
But ST or STX into fp-8 could belong to the same block of instructions.
When backtracing is done the parent state may have fp-8 slot
as "unallocated stack". Which will cause verifier to warn
and incorrectly reject such programs.
Writes into stack via non-R10 register are rare. llvm always
generates canonical stack spill/fill.
For such pathological case fall back to conservative precision
tracking instead of rejecting.
Reported-by: syzbot+c8d66267fd2b5955287e@syzkaller.appspotmail.com
Fixes: b5dc0163d8fd ("bpf: precise scalar_value tracking")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
The recent change to avoid taking the expiry lock when a timer is currently
migrated missed to add a bracket at the end of the if statement leading to
compile errors. Since that commit the variable `migration_base' is always
used but it is only available on SMP configuration thus leading to another
compile error. The changelog says "The timer base and base->cpu_base
cannot be NULL in the code path", so it is safe to limit this check to SMP
configurations only.
Add the missing bracket to the if statement and hide `migration_base'
behind CONFIG_SMP bars.
[ tglx: Mark the functions inline ... ]
Fixes: 68b2c8c1e4210 ("hrtimer: Don't take expiry_lock when timer is currently migrated")
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20190904145527.eah7z56ntwobqm6j@linutronix.de
Since BUG() and WARN() may use a trap (e.g. UD2 on x86) to
get the address where the BUG() has occurred, kprobes can not
do single-step out-of-line that instruction. So prohibit
probing on such address.
Without this fix, if someone put a kprobe on WARN(), the
kernel will crash with invalid opcode error instead of
outputing warning message, because kernel can not find
correct bug address.
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Acked-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Cc: David S . Miller <davem@davemloft.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Naveen N . Rao <naveen.n.rao@linux.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/156750890133.19112.3393666300746167111.stgit@devnote2
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Thadeu Lima de Souza Cascardo reported that 'chrt' broke on recent kernels:
$ chrt -p $$
chrt: failed to get pid 26306's policy: Argument list too long
and he has root-caused the bug to the following commit increasing sched_attr
size and breaking sched_read_attr() into returning -EFBIG:
a509a7cd7974 ("sched/uclamp: Extend sched_setattr() to support utilization clamping")
The other, bigger bug is that the whole sched_getattr() and sched_read_attr()
logic of checking non-zero bits in new ABI components is arguably broken,
and pretty much any extension of the ABI will spuriously break the ABI.
That's way too fragile.
Instead implement the perf syscall's extensible ABI instead, which we
already implement on the sched_setattr() side:
- if user-attributes have the same size as kernel attributes then the
logic is unchanged.
- if user-attributes are larger than the kernel knows about then simply
skip the extra bits, but set attr->size to the (smaller) kernel size
so that tooling can (in principle) handle older kernel as well.
- if user-attributes are smaller than the kernel knows about then just
copy whatever user-space can accept.
Also clean up the whole logic:
- Simplify the code flow - there's no need for 'ret' for example.
- Standardize on 'kattr/uattr' and 'ksize/usize' naming to make sure we
always know which side we are dealing with.
- Why is it called 'read' when what it does is to copy to user? This
code is so far away from VFS read() semantics that the naming is
actively confusing. Name it sched_attr_copy_to_user() instead, which
mirrors other copy_to_user() functionality.
- Move the attr->size assignment from the head of sched_getattr() to the
sched_attr_copy_to_user() function. Nothing else within the kernel
should care about the size of the structure.
With these fixes the sched_getattr() syscall now nicely supports an
extensible ABI in both a forward and backward compatible fashion, and
will also fix the chrt bug.
As an added bonus the bogus -EFBIG return is removed as well, which as
Thadeu noted should have been -E2BIG to begin with.
Reported-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Acked-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Patrick Bellasi <patrick.bellasi@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: a509a7cd7974 ("sched/uclamp: Extend sched_setattr() to support utilization clamping")
Link: https://lkml.kernel.org/r/20190904075532.GA26751@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
CONFIG_SHELL falls back to sh when bash is not installed on the system,
but nobody is testing such a case since bash is usually installed.
So, shell scripts invoked by CONFIG_SHELL are only tested with bash.
It makes it difficult to test whether the hashbang #!/bin/sh is real.
For example, #!/bin/sh in arch/powerpc/kernel/prom_init_check.sh is
false. (I fixed it up)
Besides, some shell scripts invoked by CONFIG_SHELL use bash-extension
and #!/bin/bash is specified as the hashbang, while CONFIG_SHELL may
not always be set to bash.
Probably, the right thing to do is to introduce BASH, which is bash by
default, and always set CONFIG_SHELL to sh. Replace $(CONFIG_SHELL)
with $(BASH) for bash scripts.
If somebody tries to add bash-extension to a #!/bin/sh script, it will
be caught in testing because /bin/sh is a symlink to dash on some major
distributions.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
A helper to find the backing page array based on a virtual address.
This also ensures we do the same vm_flags check everywhere instead
of slightly different or missing ones in a few places.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Currently the generic dma remap allocator gets a vm_flags passed by
the caller that is a little confusing. We just introduced a generic
vmalloc-level flag to identify the dma coherent allocations, so use
that everywhere and remove the now pointless argument.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Most dma_map_ops instances are IOMMUs that work perfectly fine in 32-bits
of IOVA space, and the generic direct mapping code already provides its
own routines that is intelligent based on the amount of memory actually
present. Wire up the dma-direct routine for the ARM direct mapping code
as well, and otherwise default to the constant 32-bit mask. This way
we only need to override it for the occasional odd IOMMU that requires
64-bit IOVA support, or IOMMU drivers that are more efficient if they
can fall back to the direct mapping.
Signed-off-by: Christoph Hellwig <hch@lst.de>
dma_declare_coherent_memory is something that the platform setup code
(which pretty much means the device tree these days) need to do so that
drivers can use the memory as declared by the platform. Drivers
themselves have no business calling this function.
Signed-off-by: Christoph Hellwig <hch@lst.de>
This function is entirely unused given that declared memory is
generally provided by platform setup code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
CONFIG_ARCH_NO_COHERENT_DMA_MMAP is now functionally identical to
!CONFIG_MMU, so remove the separate symbol. The only difference is that
arm did not set it for !CONFIG_MMU, but arm uses a separate dma mapping
implementation including its own mmap method, which is handled by moving
the CONFIG_MMU check in dma_can_mmap so that is only applies to the
dma-direct case, just as the other ifdefs for it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> # m68k
Add a helper to check if DMA allocations for a specific device can be
mapped to userspace using dma_mmap_*.
Signed-off-by: Christoph Hellwig <hch@lst.de>
While the default ->mmap and ->get_sgtable implementations work for the
majority of our dma_map_ops impementations they are inherently safe
for others that don't use the page allocator or CMA and/or use their
own way of remapping not covered by the common code. So remove the
defaults if these methods are not wired up, but instead wire up the
default implementations for all safe instances.
Fixes: e1c7e324539a ("dma-mapping: always provide the dma_map_ops based implementation")
Signed-off-by: Christoph Hellwig <hch@lst.de>
The comment that says that module_event() is not static is clearly
wrong.
Signed-off-by: Nadav Amit <namit@vmware.com>
Signed-off-by: Daniel Thompson <daniel.thompson@linaro.org>
strncmp(str, const, len) is error-prone.
We had better use newly introduced
str_has_prefix() instead of it.
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Signed-off-by: Daniel Thompson <daniel.thompson@linaro.org>
The play_idle resolution is 1ms. The intel_powerclamp bases the idle
duration on jiffies. The idle injection API is also using msec based
duration but has no user yet.
Unfortunately, msec based time does not fit well when we want to
inject idle cycle precisely with shallow idle state.
In order to set the scene for the incoming idle injection user, move
the precision up to usec when calling play_idle.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Recently device pass-through stops working for Linux VM running on Hyper-V.
git-bisect shows the regression is caused by the recent commit
467a3bb97432 ("PCI: hv: Allocate a named fwnode ..."), but the root cause
is that the commit d59f6617eef0 forgets to set the domain->fwnode for
IRQCHIP_FWNODE_NAMED*, and as a result:
1. The domain->fwnode remains to be NULL.
2. irq_find_matching_fwspec() returns NULL since "h->fwnode == fwnode" is
false, and pci_set_bus_msi_domain() sets the Hyper-V PCI root bus's
msi_domain to NULL.
3. When the device is added onto the root bus, the device's dev->msi_domain
is set to NULL in pci_set_msi_domain().
4. When a device driver tries to enable MSI-X, pci_msi_setup_msi_irqs()
calls arch_setup_msi_irqs(), which uses the native MSI chip (i.e.
arch/x86/kernel/apic/msi.c: pci_msi_controller) to set up the irqs, but
actually pci_msi_setup_msi_irqs() is supposed to call
msi_domain_alloc_irqs() with the hbus->irq_domain, which is created in
hv_pcie_init_irq_domain() and is associated with the Hyper-V chip
hv_msi_irq_chip. Consequently, the irq line is not properly set up, and
the device driver can not receive any interrupt.
Fixes: d59f6617eef0 ("genirq: Allow fwnode to carry name information only")
Fixes: 467a3bb97432 ("PCI: hv: Allocate a named fwnode instead of an address-based one")
Reported-by: Lili Deng <v-lide@microsoft.com>
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/PU1P153MB01694D9AF625AC335C600C5FBFBE0@PU1P153MB0169.APCP153.PROD.OUTLOOK.COM
The supported clamp indexes are defined in 'enum clamp_id', however, because
of the code logic in some of the first utilization clamping series version,
sometimes we needed to use 'unsigned int' to represent indices.
This is not more required since the final version of the uclamp_* APIs can
always use the proper enum uclamp_id type.
Fix it with a bulk rename now that we have all the bits merged.
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Michal Koutny <mkoutny@suse.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Alessio Balsini <balsini@android.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Paul Turner <pjt@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Perret <quentin.perret@arm.com>
Cc: Rafael J . Wysocki <rafael.j.wysocki@intel.com>
Cc: Steve Muckle <smuckle@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Todd Kjos <tkjos@google.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lkml.kernel.org/r/20190822132811.31294-7-patrick.bellasi@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
On updates of task group (TG) clamp values, ensure that these new values
are enforced on all RUNNABLE tasks of the task group, i.e. all RUNNABLE
tasks are immediately boosted and/or capped as requested.
Do that each time we update effective clamps from cpu_util_update_eff().
Use the *cgroup_subsys_state (css) to walk the list of tasks in each
affected TG and update their RUNNABLE tasks.
Update each task by using the same mechanism used for cpu affinity masks
updates, i.e. by taking the rq lock.
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Michal Koutny <mkoutny@suse.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Alessio Balsini <balsini@android.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Paul Turner <pjt@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Perret <quentin.perret@arm.com>
Cc: Rafael J . Wysocki <rafael.j.wysocki@intel.com>
Cc: Steve Muckle <smuckle@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Todd Kjos <tkjos@google.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lkml.kernel.org/r/20190822132811.31294-6-patrick.bellasi@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When a task specific clamp value is configured via sched_setattr(2), this
value is accounted in the corresponding clamp bucket every time the task is
{en,de}qeued. However, when cgroups are also in use, the task specific
clamp values could be restricted by the task_group (TG) clamp values.
Update uclamp_cpu_inc() to aggregate task and TG clamp values. Every time a
task is enqueued, it's accounted in the clamp bucket tracking the smaller
clamp between the task specific value and its TG effective value. This
allows to:
1. ensure cgroup clamps are always used to restrict task specific requests,
i.e. boosted not more than its TG effective protection and capped at
least as its TG effective limit.
2. implement a "nice-like" policy, where tasks are still allowed to request
less than what enforced by their TG effective limits and protections
Do this by exploiting the concept of "effective" clamp, which is already
used by a TG to track parent enforced restrictions.
Apply task group clamp restrictions only to tasks belonging to a child
group. While, for tasks in the root group or in an autogroup, system
defaults are still enforced.
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Michal Koutny <mkoutny@suse.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Alessio Balsini <balsini@android.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Paul Turner <pjt@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Perret <quentin.perret@arm.com>
Cc: Rafael J . Wysocki <rafael.j.wysocki@intel.com>
Cc: Steve Muckle <smuckle@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Todd Kjos <tkjos@google.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lkml.kernel.org/r/20190822132811.31294-5-patrick.bellasi@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The clamp values are not tunable at the level of the root task group.
That's for two main reasons:
- the root group represents "system resources" which are always
entirely available from the cgroup standpoint.
- when tuning/restricting "system resources" makes sense, tuning must
be done using a system wide API which should also be available when
control groups are not.
When a system wide restriction is available, cgroups should be aware of
its value in order to know exactly how much "system resources" are
available for the subgroups.
Utilization clamping supports already the concepts of:
- system defaults: which define the maximum possible clamp values
usable by tasks.
- effective clamps: which allows a parent cgroup to constraint (maybe
temporarily) its descendants without losing the information related
to the values "requested" from them.
Exploit these two concepts and bind them together in such a way that,
whenever system default are tuned, the new values are propagated to
(possibly) restrict or relax the "effective" value of nested cgroups.
When cgroups are in use, force an update of all the RUNNABLE tasks.
Otherwise, keep things simple and do just a lazy update next time each
task will be enqueued.
Do that since we assume a more strict resource control is required when
cgroups are in use. This allows also to keep "effective" clamp values
updated in case we need to expose them to user-space.
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Michal Koutny <mkoutny@suse.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Alessio Balsini <balsini@android.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Paul Turner <pjt@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Perret <quentin.perret@arm.com>
Cc: Rafael J . Wysocki <rafael.j.wysocki@intel.com>
Cc: Steve Muckle <smuckle@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Todd Kjos <tkjos@google.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lkml.kernel.org/r/20190822132811.31294-4-patrick.bellasi@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In order to properly support hierarchical resources control, the cgroup
delegation model requires that attribute writes from a child group never
fail but still are locally consistent and constrained based on parent's
assigned resources. This requires to properly propagate and aggregate
parent attributes down to its descendants.
Implement this mechanism by adding a new "effective" clamp value for each
task group. The effective clamp value is defined as the smaller value
between the clamp value of a group and the effective clamp value of its
parent. This is the actual clamp value enforced on tasks in a task group.
Since it's possible for a cpu.uclamp.min value to be bigger than the
cpu.uclamp.max value, ensure local consistency by restricting each
"protection" (i.e. min utilization) with the corresponding "limit"
(i.e. max utilization).
Do that at effective clamps propagation to ensure all user-space write
never fails while still always tracking the most restrictive values.
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Michal Koutny <mkoutny@suse.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Alessio Balsini <balsini@android.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Paul Turner <pjt@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Perret <quentin.perret@arm.com>
Cc: Rafael J . Wysocki <rafael.j.wysocki@intel.com>
Cc: Steve Muckle <smuckle@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Todd Kjos <tkjos@google.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lkml.kernel.org/r/20190822132811.31294-3-patrick.bellasi@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The cgroup CPU bandwidth controller allows to assign a specified
(maximum) bandwidth to the tasks of a group. However this bandwidth is
defined and enforced only on a temporal base, without considering the
actual frequency a CPU is running on. Thus, the amount of computation
completed by a task within an allocated bandwidth can be very different
depending on the actual frequency the CPU is running that task.
The amount of computation can be affected also by the specific CPU a
task is running on, especially when running on asymmetric capacity
systems like Arm's big.LITTLE.
With the availability of schedutil, the scheduler is now able
to drive frequency selections based on actual task utilization.
Moreover, the utilization clamping support provides a mechanism to
bias the frequency selection operated by schedutil depending on
constraints assigned to the tasks currently RUNNABLE on a CPU.
Giving the mechanisms described above, it is now possible to extend the
cpu controller to specify the minimum (or maximum) utilization which
should be considered for tasks RUNNABLE on a cpu.
This makes it possible to better defined the actual computational
power assigned to task groups, thus improving the cgroup CPU bandwidth
controller which is currently based just on time constraints.
Extend the CPU controller with a couple of new attributes uclamp.{min,max}
which allow to enforce utilization boosting and capping for all the
tasks in a group.
Specifically:
- uclamp.min: defines the minimum utilization which should be considered
i.e. the RUNNABLE tasks of this group will run at least at a
minimum frequency which corresponds to the uclamp.min
utilization
- uclamp.max: defines the maximum utilization which should be considered
i.e. the RUNNABLE tasks of this group will run up to a
maximum frequency which corresponds to the uclamp.max
utilization
These attributes:
a) are available only for non-root nodes, both on default and legacy
hierarchies, while system wide clamps are defined by a generic
interface which does not depends on cgroups. This system wide
interface enforces constraints on tasks in the root node.
b) enforce effective constraints at each level of the hierarchy which
are a restriction of the group requests considering its parent's
effective constraints. Root group effective constraints are defined
by the system wide interface.
This mechanism allows each (non-root) level of the hierarchy to:
- request whatever clamp values it would like to get
- effectively get only up to the maximum amount allowed by its parent
c) have higher priority than task-specific clamps, defined via
sched_setattr(), thus allowing to control and restrict task requests.
Add two new attributes to the cpu controller to collect "requested"
clamp values. Allow that at each non-root level of the hierarchy.
Keep it simple by not caring now about "effective" values computation
and propagation along the hierarchy.
Update sysctl_sched_uclamp_handler() to use the newly introduced
uclamp_mutex so that we serialize system default updates with cgroup
relate updates.
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Michal Koutny <mkoutny@suse.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Alessio Balsini <balsini@android.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Paul Turner <pjt@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Perret <quentin.perret@arm.com>
Cc: Rafael J . Wysocki <rafael.j.wysocki@intel.com>
Cc: Steve Muckle <smuckle@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Todd Kjos <tkjos@google.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lkml.kernel.org/r/20190822132811.31294-2-patrick.bellasi@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
SD_BALANCE_{FORK,EXEC} and SD_WAKE_AFFINE are stripped in sd_init()
for any sched domains with a NUMA distance greater than 2 hops
(RECLAIM_DISTANCE). The idea being that it's expensive to balance
across domains that far apart.
However, as is rather unfortunately explained in:
commit 32e45ff43eaf ("mm: increase RECLAIM_DISTANCE to 30")
the value for RECLAIM_DISTANCE is based on node distance tables from
2011-era hardware.
Current AMD EPYC machines have the following NUMA node distances:
node distances:
node 0 1 2 3 4 5 6 7
0: 10 16 16 16 32 32 32 32
1: 16 10 16 16 32 32 32 32
2: 16 16 10 16 32 32 32 32
3: 16 16 16 10 32 32 32 32
4: 32 32 32 32 10 16 16 16
5: 32 32 32 32 16 10 16 16
6: 32 32 32 32 16 16 10 16
7: 32 32 32 32 16 16 16 10
where 2 hops is 32.
The result is that the scheduler fails to load balance properly across
NUMA nodes on different sockets -- 2 hops apart.
For example, pinning 16 busy threads to NUMA nodes 0 (CPUs 0-7) and 4
(CPUs 32-39) like so,
$ numactl -C 0-7,32-39 ./spinner 16
causes all threads to fork and remain on node 0 until the active
balancer kicks in after a few seconds and forcibly moves some threads
to node 4.
Override node_reclaim_distance for AMD Zen.
Signed-off-by: Matt Fleming <matt@codeblueprint.co.uk>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@surriel.com>
Cc: Suravee.Suthikulpanit@amd.com
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Thomas.Lendacky@amd.com
Cc: Tony Luck <tony.luck@intel.com>
Link: https://lkml.kernel.org/r/20190808195301.13222-3-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
