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.
The current implementation of splice() and tee() ignores O_NONBLOCK set
on pipe file descriptors and checks only the SPLICE_F_NONBLOCK flag for
blocking on pipe arguments. This is inconsistent since splice()-ing
from/to non-pipe file descriptors does take O_NONBLOCK into
consideration.
Fix this by promoting O_NONBLOCK, when set on a pipe, to
SPLICE_F_NONBLOCK.
Some context for how the current implementation of splice() leads to
inconsistent behavior. In the ongoing work[1] to add VM tracing
capability to trace-cmd we stream tracing data over named FIFOs or
vsockets from guests back to the host.
When we receive SIGINT from user to stop tracing, we set O_NONBLOCK on
the input file descriptor and set SPLICE_F_NONBLOCK for the next call to
splice(). If splice() was blocked waiting on data from the input FIFO,
after SIGINT splice() restarts with the same arguments (no
SPLICE_F_NONBLOCK) and blocks again instead of returning -EAGAIN when no
data is available.
This differs from the splice() behavior when reading from a vsocket or
when we're doing a traditional read()/write() loop (trace-cmd's
--nosplice argument).
With this patch applied we get the same behavior in all situations after
setting O_NONBLOCK which also matches the behavior of doing a
read()/write() loop instead of splice().
This change does have potential of breaking users who don't expect
EAGAIN from splice() when SPLICE_F_NONBLOCK is not set. OTOH programs
that set O_NONBLOCK and don't anticipate EAGAIN are arguably buggy[2].
[1] https://github.com/skaslev/trace-cmd/tree/vsock
[2] d47e3da175/fs/read_write.c (L1425)
Signed-off-by: Slavomir Kaslev <kaslevs@vmware.com>
Reviewed-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull vfs fixes from Al Viro:
"Assorted fixes that sat in -next for a while, all over the place"
* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
aio: Fix locking in aio_poll()
exec: Fix mem leak in kernel_read_file
copy_mount_string: Limit string length to PATH_MAX
cgroup: saner refcounting for cgroup_root
fix cgroup_do_mount() handling of failure exits
Every in-kernel use of this function defined it to KERNEL_DS (either as
an actual define, or as an inline function). It's an entirely
historical artifact, and long long long ago used to actually read the
segment selector valueof '%ds' on x86.
Which in the kernel is always KERNEL_DS.
Inspired by a patch from Jann Horn that just did this for a very small
subset of users (the ones in fs/), along with Al who suggested a script.
I then just took it to the logical extreme and removed all the remaining
gunk.
Roughly scripted with
git grep -l '(get_ds())' -- :^tools/ | xargs sed -i 's/(get_ds())/(KERNEL_DS)/'
git grep -lw 'get_ds' -- :^tools/ | xargs sed -i '/^#define get_ds()/d'
plus manual fixups to remove a few unusual usage patterns, the couple of
inline function cases and to fix up a comment that had become stale.
The 'get_ds()' function remains in an x86 kvm selftest, since in user
space it actually does something relevant.
Inspired-by: Jann Horn <jannh@google.com>
Inspired-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Al Viro root-caused a race where the IOCB_CMD_POLL handling of
fget/fput() could cause us to access the file pointer after it had
already been freed:
"In more details - normally IOCB_CMD_POLL handling looks so:
1) io_submit(2) allocates aio_kiocb instance and passes it to
aio_poll()
2) aio_poll() resolves the descriptor to struct file by req->file =
fget(iocb->aio_fildes)
3) aio_poll() sets ->woken to false and raises ->ki_refcnt of that
aio_kiocb to 2 (bumps by 1, that is).
4) aio_poll() calls vfs_poll(). After sanity checks (basically,
"poll_wait() had been called and only once") it locks the queue.
That's what the extra reference to iocb had been for - we know we
can safely access it.
5) With queue locked, we check if ->woken has already been set to
true (by aio_poll_wake()) and, if it had been, we unlock the
queue, drop a reference to aio_kiocb and bugger off - at that
point it's a responsibility to aio_poll_wake() and the stuff
called/scheduled by it. That code will drop the reference to file
in req->file, along with the other reference to our aio_kiocb.
6) otherwise, we see whether we need to wait. If we do, we unlock the
queue, drop one reference to aio_kiocb and go away - eventual
wakeup (or cancel) will deal with the reference to file and with
the other reference to aio_kiocb
7) otherwise we remove ourselves from waitqueue (still under the
queue lock), so that wakeup won't get us. No async activity will
be happening, so we can safely drop req->file and iocb ourselves.
If wakeup happens while we are in vfs_poll(), we are fine - aio_kiocb
won't get freed under us, so we can do all the checks and locking
safely. And we don't touch ->file if we detect that case.
However, vfs_poll() most certainly *does* touch the file it had been
given. So wakeup coming while we are still in ->poll() might end up
doing fput() on that file. That case is not too rare, and usually we
are saved by the still present reference from descriptor table - that
fput() is not the final one.
But if another thread closes that descriptor right after our fget()
and wakeup does happen before ->poll() returns, we are in trouble -
final fput() done while we are in the middle of a method:
Al also wrote a patch to take an extra reference to the file descriptor
to fix this, but I instead suggested we just streamline the whole file
pointer handling by submit_io() so that the generic aio submission code
simply keeps the file pointer around until the aio has completed.
Fixes: bfe4037e722e ("aio: implement IOCB_CMD_POLL")
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Reported-by: syzbot+503d4cc169fcec1cb18c@syzkaller.appspotmail.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Use inode->i_lock to protect i_size_write(), else i_size_read() in
generic_fillattr() may loop infinitely in read_seqcount_begin() when
multiple processes invoke v9fs_vfs_getattr() or v9fs_vfs_getattr_dotl()
simultaneously under 32-bit SMP environment, and a soft lockup will be
triggered as show below:
watchdog: BUG: soft lockup - CPU#5 stuck for 22s! [stat:2217]
Modules linked in:
CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4
Hardware name: Generic DT based system
PC is at generic_fillattr+0x104/0x108
LR is at 0xec497f00
pc : [<802b8898>] lr : [<ec497f00>] psr: 200c0013
sp : ec497e20 ip : ed608030 fp : ec497e3c
r10: 00000000 r9 : ec497f00 r8 : ed608030
r7 : ec497ebc r6 : ec497f00 r5 : ee5c1550 r4 : ee005780
r3 : 0000052d r2 : 00000000 r1 : ec497f00 r0 : ed608030
Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none
Control: 10c5387d Table: ac48006a DAC: 00000051
CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4
Hardware name: Generic DT based system
Backtrace:
[<8010d974>] (dump_backtrace) from [<8010dc88>] (show_stack+0x20/0x24)
[<8010dc68>] (show_stack) from [<80a1d194>] (dump_stack+0xb0/0xdc)
[<80a1d0e4>] (dump_stack) from [<80109f34>] (show_regs+0x1c/0x20)
[<80109f18>] (show_regs) from [<801d0a80>] (watchdog_timer_fn+0x280/0x2f8)
[<801d0800>] (watchdog_timer_fn) from [<80198658>] (__hrtimer_run_queues+0x18c/0x380)
[<801984cc>] (__hrtimer_run_queues) from [<80198e60>] (hrtimer_run_queues+0xb8/0xf0)
[<80198da8>] (hrtimer_run_queues) from [<801973e8>] (run_local_timers+0x28/0x64)
[<801973c0>] (run_local_timers) from [<80197460>] (update_process_times+0x3c/0x6c)
[<80197424>] (update_process_times) from [<801ab2b8>] (tick_nohz_handler+0xe0/0x1bc)
[<801ab1d8>] (tick_nohz_handler) from [<80843050>] (arch_timer_handler_virt+0x38/0x48)
[<80843018>] (arch_timer_handler_virt) from [<80180a64>] (handle_percpu_devid_irq+0x8c/0x240)
[<801809d8>] (handle_percpu_devid_irq) from [<8017ac20>] (generic_handle_irq+0x34/0x44)
[<8017abec>] (generic_handle_irq) from [<8017b344>] (__handle_domain_irq+0x6c/0xc4)
[<8017b2d8>] (__handle_domain_irq) from [<801022e0>] (gic_handle_irq+0x4c/0x88)
[<80102294>] (gic_handle_irq) from [<80101a30>] (__irq_svc+0x70/0x98)
[<802b8794>] (generic_fillattr) from [<8056b284>] (v9fs_vfs_getattr_dotl+0x74/0xa4)
[<8056b210>] (v9fs_vfs_getattr_dotl) from [<802b8904>] (vfs_getattr_nosec+0x68/0x7c)
[<802b889c>] (vfs_getattr_nosec) from [<802b895c>] (vfs_getattr+0x44/0x48)
[<802b8918>] (vfs_getattr) from [<802b8a74>] (vfs_statx+0x9c/0xec)
[<802b89d8>] (vfs_statx) from [<802b9428>] (sys_lstat64+0x48/0x78)
[<802b93e0>] (sys_lstat64) from [<80101000>] (ret_fast_syscall+0x0/0x28)
[dominique.martinet@cea.fr: updated comment to not refer to a function
in another subsystem]
Link: http://lkml.kernel.org/r/20190124063514.8571-2-houtao1@huawei.com
Cc: stable@vger.kernel.org
Fixes: 7549ae3e81cc ("9p: Use the i_size_[read, write]() macros instead of using inode->i_size directly.")
Reported-by: Xing Gaopeng <xingaopeng@huawei.com>
Signed-off-by: Hou Tao <houtao1@huawei.com>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Users can still control this value explicitly using the
max_session_cb_slots module parameter, but let's bump the default
up to 16 for now.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Get rid of the redundant parameter and rename the function
ff_layout_mirror_valid() to ff_layout_init_mirror_ds() for clarity.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
nfs4_ff_alloc_deviceid_node() guarantees that if mirror->mirror_ds is
a valid pointer, then so is mirror->mirror_ds->ds.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Pass in a pointer to the mirror rather than having to retrieve it from
the array and then verify the resulting pointer.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If we notice that a DS may be down, we should attempt to read from the
other mirrors first before we go back to retry the dead DS.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If the DS is unresponsive, we want to just mark it as such, while
reporting the errors. If the server later returns the same deviceid
in a new layout, then we don't want to have to look it up again.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
In ff_layout_mirror_valid() we may not want to invalidate the layout
segment despite the call to GETDEVICEINFO failing. The reason is that
a read may still be able to make progress on another mirror.
So instead we let the caller (in this case nfs4_ff_layout_prepare_ds())
decide whether or not it needs to invalidate.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
While we may want to skip attempting to connect to a downed mirror
when we're deciding which mirror to select for a read, we do not
want to do so once we've committed to attempting the I/O in
ff_layout_read/write_pagelist(), or ff_layout_initiate_commit()
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
When a read to the preferred mirror returns an error, the flexfiles
driver records the error in the inode list and currently marks the
layout for return before failing over the attempted read to the next
mirror.
What we actually want to do is fire off a LAYOUTERROR to notify the
MDS that there is an issue with the preferred mirror, then we fail
over. Only once we've failed to read from all mirrors should we
return the layout.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
The radix tree would rewind the index in an iterator to the lowest index
of a multi-slot entry. The XArray iterators instead leave the index
unchanged, but I overlooked that when converting DAX from the radix tree
to the XArray. Adjust the index that we use for flushing to the start
of the PMD range.
Fixes: c1901cd33cf4 ("page cache: Convert find_get_entries_tag to XArray")
Cc: <stable@vger.kernel.org>
Reported-by: Piotr Balcer <piotr.balcer@intel.com>
Tested-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
If a layout segment gets invalidated while a pNFS I/O operation
is queued for transmission, then we ideally want to abort
immediately. This is particularly the case when there is a large
number of I/O related RPCs queued in the RPC layer, and the layout
segment gets invalidated due to an ENOSPC error, or an EACCES (because
the client was fenced). We may end up forced to spam the MDS with a
lot of otherwise unnecessary LAYOUTERRORs after that I/O fails.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Fix the memory barriers in nfs4_mark_deviceid_unavailable() and
nfs4_test_deviceid_unavailable().
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If the attempt to instantiate the mirror's layout DS pointer failed,
then that pointer may hold a value of type ERR_PTR(), so we need
to check that before we dereference it.
Fixes: 65990d1afbd2d ("pNFS/flexfiles: Fix a deadlock on LAYOUTGET")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
These really should have been there from the beginning, but we never
noticed because there was enough slack in the RPC request for the extra
bytes. Chuck's recent patch to use au_cslack and au_rslack to compute
buffer size shrunk the buffer enough that this was now a problem for
SEEK operations on my test client.
Fixes: f4ac1674f5da4 ("nfs: Add ALLOCATE support")
Fixes: 2e72448b07dc3 ("NFS: Add COPY nfs operation")
Fixes: cb95deea0b4aa ("NFS OFFLOAD_CANCEL xdr")
Fixes: 624bd5b7b683c ("nfs: Add DEALLOCATE support")
Fixes: 1c6dcbe5ceff8 ("NFS: Implement SEEK")
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Ensure that if we call nfs41_sequence_process() a second time for the
same rpc_task, then we only process the results once.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
If we have to retransmit a request, we should ensure that we reinitialise
the sequence results structure, since in the event of a signal
we need to treat the request as if it had not been sent.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Cc: stable@vger.kernel.org
Merge misc fixes from Andrew Morton:
"2 fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
hugetlbfs: fix races and page leaks during migration
kasan: turn off asan-stack for clang-8 and earlier
hugetlb pages should only be migrated if they are 'active'. The
routines set/clear_page_huge_active() modify the active state of hugetlb
pages.
When a new hugetlb page is allocated at fault time, set_page_huge_active
is called before the page is locked. Therefore, another thread could
race and migrate the page while it is being added to page table by the
fault code. This race is somewhat hard to trigger, but can be seen by
strategically adding udelay to simulate worst case scheduling behavior.
Depending on 'how' the code races, various BUG()s could be triggered.
To address this issue, simply delay the set_page_huge_active call until
after the page is successfully added to the page table.
Hugetlb pages can also be leaked at migration time if the pages are
associated with a file in an explicitly mounted hugetlbfs filesystem.
For example, consider a two node system with 4GB worth of huge pages
available. A program mmaps a 2G file in a hugetlbfs filesystem. It
then migrates the pages associated with the file from one node to
another. When the program exits, huge page counts are as follows:
node0
1024 free_hugepages
1024 nr_hugepages
node1
0 free_hugepages
1024 nr_hugepages
Filesystem Size Used Avail Use% Mounted on
nodev 4.0G 2.0G 2.0G 50% /var/opt/hugepool
That is as expected. 2G of huge pages are taken from the free_hugepages
counts, and 2G is the size of the file in the explicitly mounted
filesystem. If the file is then removed, the counts become:
node0
1024 free_hugepages
1024 nr_hugepages
node1
1024 free_hugepages
1024 nr_hugepages
Filesystem Size Used Avail Use% Mounted on
nodev 4.0G 2.0G 2.0G 50% /var/opt/hugepool
Note that the filesystem still shows 2G of pages used, while there
actually are no huge pages in use. The only way to 'fix' the filesystem
accounting is to unmount the filesystem
If a hugetlb page is associated with an explicitly mounted filesystem,
this information in contained in the page_private field. At migration
time, this information is not preserved. To fix, simply transfer
page_private from old to new page at migration time if necessary.
There is a related race with removing a huge page from a file and
migration. When a huge page is removed from the pagecache, the
page_mapping() field is cleared, yet page_private remains set until the
page is actually freed by free_huge_page(). A page could be migrated
while in this state. However, since page_mapping() is not set the
hugetlbfs specific routine to transfer page_private is not called and we
leak the page count in the filesystem.
To fix that, check for this condition before migrating a huge page. If
the condition is detected, return EBUSY for the page.
Link: http://lkml.kernel.org/r/74510272-7319-7372-9ea6-ec914734c179@oracle.com
Link: http://lkml.kernel.org/r/20190212221400.3512-1-mike.kravetz@oracle.com
Fixes: bcc54222309c ("mm: hugetlb: introduce page_huge_active")
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: <stable@vger.kernel.org>
[mike.kravetz@oracle.com: v2]
Link: http://lkml.kernel.org/r/7534d322-d782-8ac6-1c8d-a8dc380eb3ab@oracle.com
[mike.kravetz@oracle.com: update comment and changelog]
Link: http://lkml.kernel.org/r/420bcfd6-158b-38e4-98da-26d0cd85bd01@oracle.com
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
statx(2) notes that any attribute that is not indicated as supported by
stx_attributes_mask has no usable value. Commit 5f955f26f3d42d ("xfs: report
crtime and attribute flags to statx") added support for informing userspace
of extra file attributes but forgot to list these flags as supported
making reporting them rather useless for the pedantic userspace author.
$ git describe --contains 5f955f26f3d42d04aba65590a32eb70eedb7f37d
v4.11-rc6~5^2^2~2
Fixes: 5f955f26f3d42d ("xfs: report crtime and attribute flags to statx")
Signed-off-by: Luis R. Rodriguez <mcgrof@kernel.org>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
[darrick: add a comment reminding people to keep attributes_mask up to date]
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
In jbd2_get_transaction, a new transaction is initialized,
and set to the j_running_transaction. No need for a return
value, so remove it.
Also, adjust some comments to match the actual operation
of this function.
Signed-off-by: Liu Song <liu.song11@zte.com.cn>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
In jbd2_journal_commit_transaction(), if we are in abort mode,
we may flush the buffer without setting descriptor block checksum
by goto start_journal_io. Then fs is mounted,
jbd2_descriptor_block_csum_verify() failed.
[ 271.379811] EXT4-fs (vdd): shut down requested (2)
[ 271.381827] Aborting journal on device vdd-8.
[ 271.597136] JBD2: Invalid checksum recovering block 22199 in log
[ 271.598023] JBD2: recovery failed
[ 271.598484] EXT4-fs (vdd): error loading journal
Fix this problem by keep setting descriptor block checksum if the
descriptor buffer is not NULL.
This checksum problem can be reproduced by xfstests generic/388.
Signed-off-by: luojiajun <luojiajun3@huawei.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
Ext4 may not free clusters correctly when punching holes in bigalloc
file systems under high load conditions. If it's not possible to
extend and restart the journal in ext4_ext_rm_leaf() when preparing to
remove blocks from a punched region, a retry of the entire punch
operation is triggered in ext4_ext_remove_space(). This causes a
partial cluster to be set to the first cluster in the extent found to
the right of the punched region. However, if the punch operation
prior to the retry had made enough progress to delete one or more
extents and a partial cluster candidate for freeing had already been
recorded, the retry would overwrite the partial cluster. The loss of
this information makes it impossible to correctly free the original
partial cluster in all cases.
This bug can cause generic/476 to fail when run as part of
xfstests-bld's bigalloc and bigalloc_1k test cases. The failure is
reported when e2fsck detects bad iblocks counts greater than expected
in units of whole clusters and also detects a number of negative block
bitmap differences equal to the iblocks discrepancy in cluster units.
Signed-off-by: Eric Whitney <enwlinux@gmail.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
guard_bio_eod() can truncate a segment in bio to allow it to do IO on
odd last sectors of a device.
It already checks if the IO starts past EOD, but it does not consider
the possibility of an IO request starting within device boundaries can
contain more than one segment past EOD.
In such cases, truncated_bytes can be bigger than PAGE_SIZE, and will
underflow bvec->bv_len.
Fix this by checking if truncated_bytes is lower than PAGE_SIZE.
This situation has been found on filesystems such as isofs and vfat,
which doesn't check the device size before mount, if the device is
smaller than the filesystem itself, a readahead on such filesystem,
which spans EOD, can trigger this situation, leading a call to
zero_user() with a wrong size possibly corrupting memory.
I didn't see any crash, or didn't let the system run long enough to
check if memory corruption will be hit somewhere, but adding
instrumentation to guard_bio_end() to check truncated_bytes size, was
enough to see the error.
The following script can trigger the error.
MNT=/mnt
IMG=./DISK.img
DEV=/dev/loop0
mkfs.vfat $IMG
mount $IMG $MNT
cp -R /etc $MNT &> /dev/null
umount $MNT
losetup -D
losetup --find --show --sizelimit 16247280 $IMG
mount $DEV $MNT
find $MNT -type f -exec cat {} + >/dev/null
Kudos to Eric Sandeen for coming up with the reproducer above
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We'll use this for the POLL implementation. Regular requests will
NOT be using references, so initialize it to 0. Any real use of
the io_kiocb ref will initialize it to at least 2.
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This enables an application to do IO, without ever entering the kernel.
By using the SQ ring to fill in new sqes and watching for completions
on the CQ ring, we can submit and reap IOs without doing a single system
call. The kernel side thread will poll for new submissions, and in case
of HIPRI/polled IO, it'll also poll for completions.
By default, we allow 1 second of active spinning. This can by changed
by passing in a different grace period at io_uring_register(2) time.
If the thread exceeds this idle time without having any work to do, it
will set:
sq_ring->flags |= IORING_SQ_NEED_WAKEUP.
The application will have to call io_uring_enter() to start things back
up again. If IO is kept busy, that will never be needed. Basically an
application that has this feature enabled will guard it's
io_uring_enter(2) call with:
read_barrier();
if (*sq_ring->flags & IORING_SQ_NEED_WAKEUP)
io_uring_enter(fd, 0, 0, IORING_ENTER_SQ_WAKEUP);
instead of calling it unconditionally.
It's mandatory to use fixed files with this feature. Failure to do so
will result in the application getting an -EBADF CQ entry when
submitting IO.
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We normally have to fget/fput for each IO we do on a file. Even with
the batching we do, the cost of the atomic inc/dec of the file usage
count adds up.
This adds IORING_REGISTER_FILES, and IORING_UNREGISTER_FILES opcodes
for the io_uring_register(2) system call. The arguments passed in must
be an array of __s32 holding file descriptors, and nr_args should hold
the number of file descriptors the application wishes to pin for the
duration of the io_uring instance (or until IORING_UNREGISTER_FILES is
called).
When used, the application must set IOSQE_FIXED_FILE in the sqe->flags
member. Then, instead of setting sqe->fd to the real fd, it sets sqe->fd
to the index in the array passed in to IORING_REGISTER_FILES.
Files are automatically unregistered when the io_uring instance is torn
down. An application need only unregister if it wishes to register a new
set of fds.
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If we have fixed user buffers, we can map them into the kernel when we
setup the io_uring. That avoids the need to do get_user_pages() for
each and every IO.
To utilize this feature, the application must call io_uring_register()
after having setup an io_uring instance, passing in
IORING_REGISTER_BUFFERS as the opcode. The argument must be a pointer to
an iovec array, and the nr_args should contain how many iovecs the
application wishes to map.
If successful, these buffers are now mapped into the kernel, eligible
for IO. To use these fixed buffers, the application must use the
IORING_OP_READ_FIXED and IORING_OP_WRITE_FIXED opcodes, and then
set sqe->index to the desired buffer index. sqe->addr..sqe->addr+seq->len
must point to somewhere inside the indexed buffer.
The application may register buffers throughout the lifetime of the
io_uring instance. It can call io_uring_register() with
IORING_UNREGISTER_BUFFERS as the opcode to unregister the current set of
buffers, and then register a new set. The application need not
unregister buffers explicitly before shutting down the io_uring
instance.
It's perfectly valid to setup a larger buffer, and then sometimes only
use parts of it for an IO. As long as the range is within the originally
mapped region, it will work just fine.
For now, buffers must not be file backed. If file backed buffers are
passed in, the registration will fail with -1/EOPNOTSUPP. This
restriction may be relaxed in the future.
RLIMIT_MEMLOCK is used to check how much memory we can pin. A somewhat
arbitrary 1G per buffer size is also imposed.
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Similarly to how we use the state->ios_left to know how many references
to get to a file, we can use it to allocate the io_kiocb's we need in
bulk.
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a separate io_submit_state structure, to cache some of the things
we need for IO submission.
One such example is file reference batching. io_submit_state. We get as
many references as the number of sqes we are submitting, and drop
unused ones if we end up switching files. The assumption here is that
we're usually only dealing with one fd, and if there are multiple,
hopefuly they are at least somewhat ordered. Could trivially be extended
to cover multiple fds, if needed.
On the completion side we do the same thing, except this is trivially
done just locally in io_iopoll_reap().
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Some uses cases repeatedly get and put references to the same file, but
the only exposed interface is doing these one at the time. As each of
these entail an atomic inc or dec on a shared structure, that cost can
add up.
Add fget_many(), which works just like fget(), except it takes an
argument for how many references to get on the file. Ditto fput_many(),
which can drop an arbitrary number of references to a file.
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add support for a polled io_uring instance. When a read or write is
submitted to a polled io_uring, the application must poll for
completions on the CQ ring through io_uring_enter(2). Polled IO may not
generate IRQ completions, hence they need to be actively found by the
application itself.
To use polling, io_uring_setup() must be used with the
IORING_SETUP_IOPOLL flag being set. It is illegal to mix and match
polled and non-polled IO on an io_uring.
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a new fsync opcode, which either syncs a range if one is passed,
or the whole file if the offset and length fields are both cleared
to zero. A flag is provided to use fdatasync semantics, that is only
force out metadata which is required to retrieve the file data, but
not others like metadata.
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The submission queue (SQ) and completion queue (CQ) rings are shared
between the application and the kernel. This eliminates the need to
copy data back and forth to submit and complete IO.
IO submissions use the io_uring_sqe data structure, and completions
are generated in the form of io_uring_cqe data structures. The SQ
ring is an index into the io_uring_sqe array, which makes it possible
to submit a batch of IOs without them being contiguous in the ring.
The CQ ring is always contiguous, as completion events are inherently
unordered, and hence any io_uring_cqe entry can point back to an
arbitrary submission.
Two new system calls are added for this:
io_uring_setup(entries, params)
Sets up an io_uring instance for doing async IO. On success,
returns a file descriptor that the application can mmap to
gain access to the SQ ring, CQ ring, and io_uring_sqes.
io_uring_enter(fd, to_submit, min_complete, flags, sigset, sigsetsize)
Initiates IO against the rings mapped to this fd, or waits for
them to complete, or both. The behavior is controlled by the
parameters passed in. If 'to_submit' is non-zero, then we'll
try and submit new IO. If IORING_ENTER_GETEVENTS is set, the
kernel will wait for 'min_complete' events, if they aren't
already available. It's valid to set IORING_ENTER_GETEVENTS
and 'min_complete' == 0 at the same time, this allows the
kernel to return already completed events without waiting
for them. This is useful only for polling, as for IRQ
driven IO, the application can just check the CQ ring
without entering the kernel.
With this setup, it's possible to do async IO with a single system
call. Future developments will enable polled IO with this interface,
and polled submission as well. The latter will enable an application
to do IO without doing ANY system calls at all.
For IRQ driven IO, an application only needs to enter the kernel for
completions if it wants to wait for them to occur.
Each io_uring is backed by a workqueue, to support buffered async IO
as well. We will only punt to an async context if the command would
need to wait for IO on the device side. Any data that can be accessed
directly in the page cache is done inline. This avoids the slowness
issue of usual threadpools, since cached data is accessed as quickly
as a sync interface.
Sample application: http://git.kernel.dk/cgit/fio/plain/t/io_uring.c
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Alter the AFS automounting code to create and modify an fs_context struct
when parameterising a new mount triggered by an AFS mountpoint rather than
constructing device name and option strings.
Also remove the cell=, vol= and rwpath options as they are then redundant.
The reason they existed is because the 'device name' may be derived
literally from a mountpoint object in the filesystem, so default cell and
parent-type information needed to be passed in by some other method from
the automount routines. The vol= option didn't end up being used.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Eric W. Biederman <ebiederm@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Add fs_context support to the AFS filesystem, converting the parameter
parsing to store options there.
This will form the basis for namespace propagation over mountpoints within
the AFS model, thereby allowing AFS to be used in containers more easily.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
