One of the goals is to reduce the overhead of using ->read_iter()
and ->write_iter() instead of ->read()/->write(); new_sync_{read,write}()
has a surprising amount of overhead, in particular inside iocb_flags().
That's why the beginning of the series is in this pile; it's not directly
iov_iter-related, but it's a part of the same work...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQQqUNBr3gm4hGXdBJlZ7Krx/gZQ6wUCYurGOQAKCRBZ7Krx/gZQ
6ysyAP91lvBfMRepcxpd9kvtuzWkU8A3rfSziZZteEHANB9Q7QEAiPn2a2OjWkcZ
uAyUWfCkHCNx+dSMkEvUgR5okQ0exAM=
=9UCV
-----END PGP SIGNATURE-----
Merge tag 'pull-work.iov_iter-base' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull vfs iov_iter updates from Al Viro:
"Part 1 - isolated cleanups and optimizations.
One of the goals is to reduce the overhead of using ->read_iter() and
->write_iter() instead of ->read()/->write().
new_sync_{read,write}() has a surprising amount of overhead, in
particular inside iocb_flags(). That's the explanation for the
beginning of the series is in this pile; it's not directly
iov_iter-related, but it's a part of the same work..."
* tag 'pull-work.iov_iter-base' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
first_iovec_segment(): just return address
iov_iter: massage calling conventions for first_{iovec,bvec}_segment()
iov_iter: first_{iovec,bvec}_segment() - simplify a bit
iov_iter: lift dealing with maxpages out of first_{iovec,bvec}_segment()
iov_iter_get_pages{,_alloc}(): cap the maxsize with MAX_RW_COUNT
iov_iter_bvec_advance(): don't bother with bvec_iter
copy_page_{to,from}_iter(): switch iovec variants to generic
keep iocb_flags() result cached in struct file
iocb: delay evaluation of IS_SYNC(...) until we want to check IOCB_DSYNC
struct file: use anonymous union member for rcuhead and llist
btrfs: use IOMAP_DIO_NOSYNC
teach iomap_dio_rw() to suppress dsync
No need of likely/unlikely on calls of check_copy_size()
-----BEGIN PGP SIGNATURE-----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=vSwm
-----END PGP SIGNATURE-----
Merge tag 'for-5.20/io_uring-zerocopy-send-2022-07-29' of git://git.kernel.dk/linux-block
Pull io_uring zerocopy support from Jens Axboe:
"This adds support for efficient support for zerocopy sends through
io_uring. Both ipv4 and ipv6 is supported, as well as both TCP and
UDP.
The core network changes to support this is in a stable branch from
Jakub that both io_uring and net-next has pulled in, and the io_uring
changes are layered on top of that.
All of the work has been done by Pavel"
* tag 'for-5.20/io_uring-zerocopy-send-2022-07-29' of git://git.kernel.dk/linux-block: (34 commits)
io_uring: notification completion optimisation
io_uring: export req alloc from core
io_uring/net: use unsigned for flags
io_uring/net: make page accounting more consistent
io_uring/net: checks errors of zc mem accounting
io_uring/net: improve io_get_notif_slot types
selftests/io_uring: test zerocopy send
io_uring: enable managed frags with register buffers
io_uring: add zc notification flush requests
io_uring: rename IORING_OP_FILES_UPDATE
io_uring: flush notifiers after sendzc
io_uring: sendzc with fixed buffers
io_uring: allow to pass addr into sendzc
io_uring: account locked pages for non-fixed zc
io_uring: wire send zc request type
io_uring: add notification slot registration
io_uring: add rsrc referencing for notifiers
io_uring: complete notifiers in tw
io_uring: cache struct io_notif
io_uring: add zc notification infrastructure
...
We want to use all optimisations that we have for io_uring requests like
completion batching, memory caching and more but for zc notifications.
Fortunately, notification perfectly fit the request model so we can
overlay them onto struct io_kiocb and use all the infratructure.
Most of the fields of struct io_notif natively fits into io_kiocb, so we
replace struct io_notif with struct io_kiocb carrying struct
io_notif_data in the cmd cache line. Then we adapt io_alloc_notif() to
use io_alloc_req()/io_alloc_req_refill(), and kill leftovers of hand
coded caching. __io_notif_complete_tw() is converted to use io_uring's
tw infra.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/9e010125175e80baf51f0ca63bdc7cc6a4a9fa56.1658913593.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Overlay notification control onto IORING_OP_RSRC_UPDATE (former
IORING_OP_FILES_UPDATE). It allows to flush a range of zc notifications
from slots with indexes [sqe->off, sqe->off+sqe->len). If sqe->arg is
not zero, it also copies sqe->arg as a new tag for all flushed
notifications.
Note, it doesn't flush a notification of a slot if there was no requests
attached to it (since last flush or registration).
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/df13e2363400682a73dd9e71c3b990b8d1ff0333.1657643355.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Allow zerocopy sends to use fixed buffers. There is an optimisation for
this case, the network layer don't need to reference the pages, see
SKBFL_MANAGED_FRAG_REFS, so io_uring have to ensure validity of fixed
buffers until the notifier is released.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/e1d8bd1b5934e541d90c1824eb4020ae3f5f43f3.1657643355.git.asml.silence@gmail.com
[axboe: fold in 32-bit pointer cast warning fix]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a new io_uring opcode IORING_OP_SENDZC. The main distinction from
IORING_OP_SEND is that the user should specify a notification slot
index in sqe::notification_idx and the buffers are safe to reuse only
when the used notification is flushed and completes.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/a80387c6a68ce9cf99b3b6ef6f71068468761fb7.1657643355.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In preparation to zerocopy sends with fixed buffers make notifiers to
reference the rsrc node to protect the used fixed buffers. We can't just
grab it for a send request as notifiers can likely outlive requests that
used it.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/3cd7a01d26837945b6982fa9cf15a63230f2ed4f.1657643355.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
kmalloc'ing struct io_notif is too expensive when done frequently, cache
them as many other resources in io_uring. Keep two list, the first one
is from where we're getting notifiers, it's protected by ->uring_lock.
The second is protected by ->completion_lock, to which we queue released
notifiers. Then we splice one list into another when needed.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/9dec18f7fcbab9f4bd40b96e5ae158b119945230.1657643355.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add internal part of send zerocopy notifications. There are two main
structures, the first one is struct io_notif, which carries inside
struct ubuf_info and maps 1:1 to it. io_uring will be binding a number
of zerocopy send requests to it and ask to complete (aka flush) it. When
flushed and all attached requests and skbs complete, it'll generate one
and only one CQE. There are intended to be passed into the network layer
as struct msghdr::msg_ubuf.
The second concept is notification slots. The userspace will be able to
register an array of slots and subsequently addressing them by the index
in the array. Slots are independent of each other. Each slot can have
only one notifier at a time (called active notifier) but many notifiers
during the lifetime. When active, a notifier not going to post any
completion but the userspace can attach requests to it by specifying
the corresponding slot while issueing send zc requests. Eventually, the
userspace will want to "flush" the notifier losing any way to attach
new requests to it, however it can use the next atomatically added
notifier of this slot or of any other slot.
When the network layer is done with all enqueued skbs attached to a
notifier and doesn't need the specified in them user data, the flushed
notifier will post a CQE.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/3ecf54c31a85762bf679b0a432c9f43ecf7e61cc.1657643355.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This adds the io_uring_short_write tracepoint to io_uring. A short write
is issued if not all pages that are required for a write are in the page
cache and the async buffered writes have to return EAGAIN.
Signed-off-by: Stefan Roesch <shr@fb.com>
Link: https://lore.kernel.org/r/20220616212221.2024518-13-shr@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This is actually an older issue, but we never used to hit the -EAGAIN
path before having done sb_start_write(). Make sure that we always call
kiocb_end_write() if we need to retry the write, so that we keep the
calls to sb_start_write() etc balanced.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This enables the async buffered writes for the filesystems that support
async buffered writes in io-uring. Buffered writes are enabled for
blocks that are already in the page cache or can be acquired with noio.
Signed-off-by: Stefan Roesch <shr@fb.com>
Link: https://lore.kernel.org/r/20220616212221.2024518-12-shr@fb.com
[axboe: adapt to 5.20 branch]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If we're offloading requests directly to io-wq because IOSQE_ASYNC was
set in the sqe, we can miss hashing writes appropriately because we
haven't set REQ_F_ISREG yet. This can cause a performance regression
with buffered writes, as io-wq then no longer correctly serializes writes
to that file.
Ensure that we set the flags in io_prep_async_work(), which will cause
the io-wq work item to be hashed appropriately.
Fixes: 584b0180f0 ("io_uring: move read/write file prep state into actual opcode handler")
Link: https://lore.kernel.org/io-uring/20220608080054.GB22428@xsang-OptiPlex-9020/
Reported-by: kernel test robot <oliver.sang@intel.com>
Tested-by: Yin Fengwei <fengwei.yin@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
A previous change enabled external users to copy the data before
calling __get_compat_msghdr(), but didn't modify get_compat_msghdr() or
__io_compat_recvmsg_copy_hdr() to take that into account. They are both
stil passing in the __user pointer rather than the copied version.
Ensure we pass in the kernel struct, not the pointer to the user data.
Link: https://lore.kernel.org/all/46439555-644d-08a1-7d66-16f8f9a320f0@samsung.com/
Fixes: 1a3e4e94a1b9 ("net: copy from user before calling __get_compat_msghdr")
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The commit 8bb649ee1d ("io_uring: remove ring quiesce for
io_uring_register") removed the worklow relying on reinit/resurrection
of the percpu_ref, hence, initialization with that requested is a relic.
This is based on code review, this causes no real bug (and theoretically
can't). Technically it's a revert of commit 214828962d ("io_uring:
initialize percpu refcounters using PERCU_REF_ALLOW_REINIT") but since
the flag omission is now justified, I'm not making this a revert.
Fixes: 8bb649ee1d ("io_uring: remove ring quiesce for io_uring_register")
Signed-off-by: Michal Koutný <mkoutny@suse.com>
Acked-by: Roman Gushchin <roman.gushchin@linux.dev>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_recvmsg_multishot_overflow had incorrect types on non x64 system.
But also it had an unnecessary INT_MAX check, which could just be done
by changing the type of the accumulator to int (also simplifying the
casts).
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Fixes: a8b38c4ce724 ("io_uring: support multishot in recvmsg")
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220715130252.610639-1-dylany@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Use atomic_long_try_cmpxchg instead of
atomic_long_cmpxchg (*ptr, old, new) == old in __io_account_mem.
x86 CMPXCHG instruction returns success in ZF flag, so this
change saves a compare after cmpxchg (and related move
instruction in front of cmpxchg).
Also, atomic_long_try_cmpxchg implicitly assigns old *ptr value
to "old" when cmpxchg fails, enabling further code simplifications.
No functional change intended.
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Similar to multishot recv, this will require provided buffers to be
used. However recvmsg is much more complex than recv as it has multiple
outputs. Specifically flags, name, and control messages.
Support this by introducing a new struct io_uring_recvmsg_out with 4
fields. namelen, controllen and flags match the similar out fields in
msghdr from standard recvmsg(2), payloadlen is the length of the payload
following the header.
This struct is placed at the start of the returned buffer. Based on what
the user specifies in struct msghdr, the next bytes of the buffer will be
name (the next msg_namelen bytes), and then control (the next
msg_controllen bytes). The payload will come at the end. The return value
in the CQE is the total used size of the provided buffer.
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220714110258.1336200-4-dylany@fb.com
[axboe: style fixups, see link]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
this is in preparation for multishot receive from io_uring, where it needs
to have access to the original struct user_msghdr.
functionally this should be a no-op.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220714110258.1336200-3-dylany@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
this is in preparation for multishot receive from io_uring, where it needs
to have access to the original struct user_msghdr.
functionally this should be a no-op.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220714110258.1336200-2-dylany@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Match up work done in "io_uring: allow iov_len = 0 for recvmsg and buffer
select", but for compat code path.
Fixes: a68caad69ce5 ("io_uring: allow iov_len = 0 for recvmsg and buffer select")
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220708181838.1495428-3-dylany@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If multishot is not actually polling then return IOU_OK rather than the
result.
If the result was > 0 this will confuse things further up the callstack
which expect a return <= 0.
Fixes: 1300ebb20286 ("io_uring: multishot recv")
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220708181838.1495428-2-dylany@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For recvmsg/sendmsg, if they don't complete inline, we currently need
to allocate a struct io_async_msghdr for each request. This is a
somewhat large struct.
Hook up sendmsg/recvmsg to use the io_alloc_cache. This reduces the
alloc + free overhead considerably, yielding 4-5% of extra performance
running netbench.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Caches like this tend to grow to the peak size, and then never get any
smaller. Impose a max limit on the size, to prevent it from growing too
big.
A somewhat randomly chosen 512 is the max size we'll allow the cache
to get. If a batch of frees come in and would bring it over that, we
simply start kfree'ing the surplus.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In preparation for adding limits, and one more user, abstract out the
core bits of the allocation+free cache.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Just as with io_poll_double_prepare() setting REQ_F_DOUBLE_POLL, we can
race with the first poll entry when setting REQ_F_ASYNC_DATA. Move it
under io_poll_double_prepare().
Fixes: a18427bb2d9b ("io_uring: optimise submission side poll_refs")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/df6920f509c11115aa2bce8b34dc5fdb0eb98920.1657203020.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
recvmsg has semantics that do not make it trivial to extend to
multishot. Specifically it has user pointers and returns data in the
original parameter. In order to make this API useful these will need to be
somehow included with the provided buffers.
For now remove multishot for recvmsg as it is not useful.
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220704140106.200167-1-dylany@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In overflow we see a duplcate line in the trace, and in some cases 3
lines (if initial io_post_aux_cqe fails).
Instead just trace once for each CQE
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220630091231.1456789-13-dylany@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Support multishot receive for io_uring.
Typical server applications will run a loop where for each recv CQE it
requeues another recv/recvmsg.
This can be simplified by using the existing multishot functionality
combined with io_uring's provided buffers.
The API is to add the IORING_RECV_MULTISHOT flag to the SQE. CQEs will
then be posted (with IORING_CQE_F_MORE flag set) when data is available
and is read. Once an error occurs or the socket ends, the multishot will
be removed and a completion without IORING_CQE_F_MORE will be posted.
The benefit to this is that the recv is much more performant.
* Subsequent receives are queued up straight away without requiring the
application to finish a processing loop.
* If there are more data in the socket (sat the provided buffer size is
smaller than the socket buffer) then the data is immediately
returned, improving batching.
* Poll is only armed once and reused, saving CPU cycles
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220630091231.1456789-11-dylany@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
On overflow, multishot poll can still complete with the IORING_CQE_F_MORE
flag set.
If in the meantime the user clears a CQE and a the poll was cancelled then
the poll will post a CQE without the IORING_CQE_F_MORE (and likely result
-ECANCELED).
However when processing the application will encounter the non-overflow
CQE which indicates that there will be no more events posted. Typical
userspace applications would free memory associated with the poll in this
case.
It will then subsequently receive the earlier CQE which has overflowed,
which breaks the contract given by the IORING_CQE_F_MORE flag.
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220630091231.1456789-9-dylany@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Some use cases of io_post_aux_cqe would not want to overflow as is, but
might want to change the flags/result. For example multishot receive
requires in order CQE, and so if there is an overflow it would need to
stop receiving until the overflow is taken care of.
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220630091231.1456789-8-dylany@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Linus Torvalds
Pavel Begunkov
Stefan Roesch
Jens Axboe
Michal Koutný
Dylan Yudaken
Uros Bizjak