Both io_poll_wq_wake() and io_cqring_wake() contain the exact same code,
and most of the comment in the latter applies equally to both.
Move the test and wakeup handling into a basic helper that they can both
use, and move part of the comment that applies generically to this new
helper.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a helper for posting 32 byte CQEs in a multishot mode and add a cmd
helper on top. As it specifically works with requests, the helper ignore
the passed in cqe->user_data and sets it to the one stored in the
request.
The command helper is only valid with multishot requests.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/c29d7720c16e1f981cfaa903df187138baa3946b.1750065793.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This function exists in both tctx.h (where it belongs) and in io_uring.h
as a remnant of before the tctx handling code got split out. Remove the
io_uring.h definition and ensure that sqpoll.c includes the tctx.h
header to get the definition.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Inflight marking is used so that do_exit() -> io_uring_files_cancel()
will find requests with files that reference an io_uring instance,
so they can get appropriately canceled before the files go away.
However, it's also called before the mm goes away.
Mark futex/futexv wait requests as being inflight, so that
io_uring_files_cancel() will prune them. This ensures that the mm stays
alive, which is important as an exiting mm will also free the futex
private hash buckets. An io_uring futex request with FUTEX2_PRIVATE
set relies on those being alive until the request has completed. A
recent commit added these futex private hashes, which get killed when
the mm goes away.
Fixes: 80367ad01d ("futex: Add basic infrastructure for local task local hash")
Link: https://lore.kernel.org/io-uring/38053.1749045482@localhost/
Reported-by: Robert Morris <rtm@csail.mit.edu>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
IOU_COMPLETE is more descriptive, in that it explicitly says that the
return value means "please post a completion for this request". This
patch completes the transition from IOU_OK to IOU_COMPLETE, replacing
existing IOU_OK users.
This is a purely mechanical change.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Don't rely on CQ sequence numbers for draining, as it has become messy
and needs cq_extra adjustments. Instead, base it on the number of
allocated requests and only allow flushing when all requests are in the
drain list.
As a result, cq_extra is gone, no overhead for its accounting in aux cqe
posting, less bloating as it was inlined before, and it's in general
simpler than trying to track where we should bump it and where it should
be put back like in cases of overflow. Also, it'll likely help with
cleaning and unifying some of the CQ posting helpers.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/46ece1e34320b046c06fee2498d6b4cd12a700f2.1746788718.git.asml.silence@gmail.com
Link: https://lore.kernel.org/r/24497b04b004bceada496033d3c9d09ff8e81ae9.1746944903.git.asml.silence@gmail.com
[axboe: fold in fix from link2]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
-----BEGIN PGP SIGNATURE-----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=vOm1
-----END PGP SIGNATURE-----
Merge tag 'for-6.15/io_uring-reg-vec-20250327' of git://git.kernel.dk/linux
Pull more io_uring updates from Jens Axboe:
"Final separate updates for io_uring.
This started out as a series of cleanups improvements and improvements
for registered buffers, but as the last series of the io_uring changes
for 6.15, it also collected a few fixes for the other branches on top:
- Add support for vectored fixed/registered buffers.
Previously only single segments have been supported for commands,
now vectored variants are supported as well. This series includes
networking and file read/write support.
- Small series unifying return codes across multi and single shot.
- Small series cleaning up registerd buffer importing.
- Adding support for vectored registered buffers for uring_cmd.
- Fix for io-wq handling of command reissue.
- Various little fixes and tweaks"
* tag 'for-6.15/io_uring-reg-vec-20250327' of git://git.kernel.dk/linux: (25 commits)
io_uring/net: fix io_req_post_cqe abuse by send bundle
io_uring/net: use REQ_F_IMPORT_BUFFER for send_zc
io_uring: move min_events sanitisation
io_uring: rename "min" arg in io_iopoll_check()
io_uring: open code __io_post_aux_cqe()
io_uring: defer iowq cqe overflow via task_work
io_uring: fix retry handling off iowq
io_uring/net: only import send_zc buffer once
io_uring/cmd: introduce io_uring_cmd_import_fixed_vec
io_uring/cmd: add iovec cache for commands
io_uring/cmd: don't expose entire cmd async data
io_uring: rename the data cmd cache
io_uring: rely on io_prep_reg_vec for iovec placement
io_uring: introduce io_prep_reg_iovec()
io_uring: unify STOP_MULTISHOT with IOU_OK
io_uring: return -EAGAIN to continue multishot
io_uring: cap cached iovec/bvec size
io_uring/net: implement vectored reg bufs for zctx
io_uring/net: convert to struct iou_vec
io_uring/net: pull vec alloc out of msghdr import
...
-----BEGIN PGP SIGNATURE-----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=MOve
-----END PGP SIGNATURE-----
Merge tag 'for-6.15/io_uring-rx-zc-20250325' of git://git.kernel.dk/linux
Pull io_uring zero-copy receive support from Jens Axboe:
"This adds support for zero-copy receive with io_uring, enabling fast
bulk receive of data directly into application memory, rather than
needing to copy the data out of kernel memory.
While this version only supports host memory as that was the initial
target, other memory types are planned as well, with notably GPU
memory coming next.
This work depends on some networking components which were queued up
on the networking side, but have now landed in your tree.
This is the work of Pavel Begunkov and David Wei. From the v14 posting:
'We configure a page pool that a driver uses to fill a hw rx queue
to hand out user pages instead of kernel pages. Any data that ends
up hitting this hw rx queue will thus be dma'd into userspace
memory directly, without needing to be bounced through kernel
memory. 'Reading' data out of a socket instead becomes a
_notification_ mechanism, where the kernel tells userspace where
the data is. The overall approach is similar to the devmem TCP
proposal
This relies on hw header/data split, flow steering and RSS to
ensure packet headers remain in kernel memory and only desired
flows hit a hw rx queue configured for zero copy. Configuring this
is outside of the scope of this patchset.
We share netdev core infra with devmem TCP. The main difference is
that io_uring is used for the uAPI and the lifetime of all objects
are bound to an io_uring instance. Data is 'read' using a new
io_uring request type. When done, data is returned via a new shared
refill queue. A zero copy page pool refills a hw rx queue from this
refill queue directly. Of course, the lifetime of these data
buffers are managed by io_uring rather than the networking stack,
with different refcounting rules.
This patchset is the first step adding basic zero copy support. We
will extend this iteratively with new features e.g. dynamically
allocated zero copy areas, THP support, dmabuf support, improved
copy fallback, general optimisations and more'
In a local setup, I was able to saturate a 200G link with a single CPU
core, and at netdev conf 0x19 earlier this month, Jamal reported
188Gbit of bandwidth using a single core (no HT, including soft-irq).
Safe to say the efficiency is there, as bigger links would be needed
to find the per-core limit, and it's considerably more efficient and
faster than the existing devmem solution"
* tag 'for-6.15/io_uring-rx-zc-20250325' of git://git.kernel.dk/linux:
io_uring/zcrx: add selftest case for recvzc with read limit
io_uring/zcrx: add a read limit to recvzc requests
io_uring: add missing IORING_MAP_OFF_ZCRX_REGION in io_uring_mmap
io_uring: Rename KConfig to Kconfig
io_uring/zcrx: fix leaks on failed registration
io_uring/zcrx: recheck ifq on shutdown
io_uring/zcrx: add selftest
net: add documentation for io_uring zcrx
io_uring/zcrx: add copy fallback
io_uring/zcrx: throttle receive requests
io_uring/zcrx: set pp memory provider for an rx queue
io_uring/zcrx: add io_recvzc request
io_uring/zcrx: dma-map area for the device
io_uring/zcrx: implement zerocopy receive pp memory provider
io_uring/zcrx: grab a net device
io_uring/zcrx: add io_zcrx_area
io_uring/zcrx: add interface queue and refill queue
IOU_OK means that the request ownership is now handed back to core
io_uring and it has to complete it using the result provided in
req->cqe. Same is true for multishot and IOU_STOP_MULTISHOT.
Rename it into IOU_COMPLETE to avoid confusion and use for both modes.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/e6a5b2edb0eb9558acb1c8f1db38ac45fee95491.1741453534.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Multishot errors can be mapped 1:1 to normal errors, but there are not
identical. It leads to a peculiar situation where all multishot requests
has to check in what context they're run and return different codes.
Unify them starting with EAGAIN / IOU_ISSUE_SKIP_COMPLETE(EIOCBQUEUED)
pair, which mean that core io_uring still owns the request and it should
be retried. In case of multishot it's naturally just continues to poll,
otherwise it might poll, use iowq or do any other kind of allowed
blocking. Introduce IOU_RETRY aliased to -EAGAIN for that.
Apart from obvious upsides, multishot can now also check for misuse of
IOU_ISSUE_SKIP_COMPLETE.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/da117b79ce72ecc3ab488c744e29fae9ba54e23b.1741453534.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add io_uring opcode OP_RECV_ZC for doing zero copy reads out of a
socket. Only the connection should be land on the specific rx queue set
up for zero copy, and the socket must be handled by the io_uring
instance that the rx queue was registered for zero copy with. That's
because neither net_iovs / buffers from our queue can be read by outside
applications, nor zero copy is possible if traffic for the zero copy
connection goes to another queue. This coordination is outside of the
scope of this patch series. Also, any traffic directed to the zero copy
enabled queue is immediately visible to the application, which is why
CAP_NET_ADMIN is required at the registration step.
Of course, no data is actually read out of the socket, it has already
been copied by the netdev into userspace memory via DMA. OP_RECV_ZC
reads skbs out of the socket and checks that its frags are indeed
net_iovs that belong to io_uring. A cqe is queued for each one of these
frags.
Recall that each cqe is a big cqe, with the top half being an
io_uring_zcrx_cqe. The cqe res field contains the len or error. The
lower IORING_ZCRX_AREA_SHIFT bits of the struct io_uring_zcrx_cqe::off
field contain the offset relative to the start of the zero copy area.
The upper part of the off field is trivially zero, and will be used
to carry the area id.
For now, there is no limit as to how much work each OP_RECV_ZC request
does. It will attempt to drain a socket of all available data. This
request always operates in multishot mode.
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: David Wei <dw@davidwei.uk>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20250215000947.789731-7-dw@davidwei.uk
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In preparation for changing how io_tw_state is passed, introduce a type
alias io_tw_token_t for struct io_tw_state *. This allows for changing
the representation in one place, without having to update the many
functions that just forward their struct io_tw_state * argument.
Also add a comment to struct io_tw_state to explain its purpose.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Link: https://lore.kernel.org/r/20250217022511.1150145-1-csander@purestorage.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
init_once is called when an object doesn't come from the cache, and
hence needs initial clearing of certain members. While the whole
struct could get cleared by memset() in that case, a few of the cache
members are large enough that this may cause unnecessary overhead if
the caches used aren't large enough to satisfy the workload. For those
cases, some churn of kmalloc+kfree is to be expected.
Ensure that the 3 users that need clearing put the members they need
cleared at the start of the struct, and wrap the rest of the struct in
a struct group so the offset is known.
While at it, improve the interaction with KASAN such that when/if
KASAN writes to members inside the struct that should be retained over
caching, it won't trip over itself. For rw and net, the retaining of
the iovec over caching is disabled if KASAN is enabled. A helper will
free and clear those members in that case.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
-----BEGIN PGP SIGNATURE-----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=ufpE
-----END PGP SIGNATURE-----
Merge tag 'for-6.14/io_uring-20250119' of git://git.kernel.dk/linux
Pull io_uring updates from Jens Axboe:
"Not a lot in terms of features this time around, mostly just cleanups
and code consolidation:
- Support for PI meta data read/write via io_uring, with NVMe and
SCSI covered
- Cleanup the per-op structure caching, making it consistent across
various command types
- Consolidate the various user mapped features into a concept called
regions, making the various users of that consistent
- Various cleanups and fixes"
* tag 'for-6.14/io_uring-20250119' of git://git.kernel.dk/linux: (56 commits)
io_uring/fdinfo: fix io_uring_show_fdinfo() misuse of ->d_iname
io_uring: reuse io_should_terminate_tw() for cmds
io_uring: Factor out a function to parse restrictions
io_uring/rsrc: require cloned buffers to share accounting contexts
io_uring: simplify the SQPOLL thread check when cancelling requests
io_uring: expose read/write attribute capability
io_uring/rw: don't gate retry on completion context
io_uring/rw: handle -EAGAIN retry at IO completion time
io_uring/rw: use io_rw_recycle() from cleanup path
io_uring/rsrc: simplify the bvec iter count calculation
io_uring: ensure io_queue_deferred() is out-of-line
io_uring/rw: always clear ->bytes_done on io_async_rw setup
io_uring/rw: use NULL for rw->free_iovec assigment
io_uring/rw: don't mask in f_iocb_flags
io_uring/msg_ring: Drop custom destructor
io_uring: Move old async data allocation helper to header
io_uring/rw: Allocate async data through helper
io_uring/net: Allocate msghdr async data through helper
io_uring/uring_cmd: Allocate async data through generic helper
io_uring/poll: Allocate apoll with generic alloc_cache helper
...
There are two remaining uses of the old async data allocator that do not
rely on the alloc cache. I don't want to make them use the new
allocator helper because that would require a if(cache) check, which
will result in dead code for the cached case (for callers passing a
cache, gcc can't prove the cache isn't NULL, and will therefore preserve
the check. Since this is an inline function and just a few lines long,
keep a second helper to deal with cases where we don't have an async
data cache.
No functional change intended here. This is just moving the helper
around and making it inline.
Signed-off-by: Gabriel Krisman Bertazi <krisman@suse.de>
Link: https://lore.kernel.org/r/20241216204615.759089-9-krisman@suse.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This helper replaces io_alloc_async_data by using the folded allocation.
Do it in a header to allow the compiler to decide whether to inline.
Signed-off-by: Gabriel Krisman Bertazi <krisman@suse.de>
Link: https://lore.kernel.org/r/20241216204615.759089-3-krisman@suse.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Instead of eagerly running all available local tw, limit the amount of
local tw done to the max of IO_LOCAL_TW_DEFAULT_MAX (20) or wait_nr. The
value of 20 is chosen as a reasonable heuristic to allow enough work
batching but also keep latency down.
Add a retry_llist that maintains a list of local tw that couldn't be
done in time. No synchronisation is needed since it is only modified
within the task context.
Signed-off-by: David Wei <dw@davidwei.uk>
Link: https://lore.kernel.org/r/20241120221452.3762588-3-dw@davidwei.uk
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In preparation for adding a new llist of tw to retry due to hitting the
tw limit, add a helper io_local_work_pending(). This function returns
true if there is any local tw pending. For now it only checks
ctx->work_llist.
Signed-off-by: David Wei <dw@davidwei.uk>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/20241120221452.3762588-2-dw@davidwei.uk
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When a DEFER_TASKRUN io_uring is terminating it requeues deferred task
work items as normal tw, which can further fallback to kthread
execution. Avoid this extra step and always push them to the fallback
kthread.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/d1cd472cec2230c66bd1c8d412a5833f0af75384.1730772720.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Rather than store the task_struct itself in struct io_kiocb, store
the io_uring specific task_struct. The life times are the same in terms
of io_uring, and this avoids doing some dereferences through the
task_struct. For the hot path of putting local task references, we can
deref req->tctx instead, which we'll need anyway in that function
regardless of whether it's local or remote references.
This is mostly straight forward, except the original task PF_EXITING
check needs a bit of tweaking. task_work is _always_ run from the
originating task, except in the fallback case, where it's run from a
kernel thread. Replace the potentially racy (in case of fallback work)
checks for req->task->flags with current->flags. It's either the still
the original task, in which case PF_EXITING will be sane, or it has
PF_KTHREAD set, in which case it's fallback work. Both cases should
prevent moving forward with the given request.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Right now the task_struct pointer is used as the key to match a task,
but in preparation for some io_kiocb changes, move it to using struct
io_uring_task instead. No functional changes intended in this patch.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Abstract out a io_uring_fill_params() helper, which fills out the
necessary bits of struct io_uring_params. Add it to io_uring.h as well,
in preparation for having another internal user of it.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In preparation for needing this somewhere else, move the definitions
for the maximum CQ and SQ ring size into io_uring.h. Make the
rings_size() helper available as well, and have it take just the setup
flags argument rather than the fill ring pointer. That's all that is
needed.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We have too many helpers posting CQEs, instead of tracing completion
events before filling in a CQE and thus having to pass all the data,
set the CQE first, pass it to the tracing helper and let it extract
everything it needs.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/b83c1ca9ee5aed2df0f3bb743bf5ed699cce4c86.1729267437.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When the sqpoll is exiting and cancels pending work items, it may need
to run task_work. If this happens from within io_uring_cancel_generic(),
then it may be under waiting for the io_uring_task waitqueue. This
results in the below splat from the scheduler, as the ring mutex may be
attempted grabbed while in a TASK_INTERRUPTIBLE state.
Ensure that the task state is set appropriately for that, just like what
is done for the other cases in io_run_task_work().
do not call blocking ops when !TASK_RUNNING; state=1 set at [<0000000029387fd2>] prepare_to_wait+0x88/0x2fc
WARNING: CPU: 6 PID: 59939 at kernel/sched/core.c:8561 __might_sleep+0xf4/0x140
Modules linked in:
CPU: 6 UID: 0 PID: 59939 Comm: iou-sqp-59938 Not tainted 6.12.0-rc3-00113-g8d020023b155 #7456
Hardware name: linux,dummy-virt (DT)
pstate: 61400005 (nZCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
pc : __might_sleep+0xf4/0x140
lr : __might_sleep+0xf4/0x140
sp : ffff80008c5e7830
x29: ffff80008c5e7830 x28: ffff0000d93088c0 x27: ffff60001c2d7230
x26: dfff800000000000 x25: ffff0000e16b9180 x24: ffff80008c5e7a50
x23: 1ffff000118bcf4a x22: ffff0000e16b9180 x21: ffff0000e16b9180
x20: 000000000000011b x19: ffff80008310fac0 x18: 1ffff000118bcd90
x17: 30303c5b20746120 x16: 74657320313d6574 x15: 0720072007200720
x14: 0720072007200720 x13: 0720072007200720 x12: ffff600036c64f0b
x11: 1fffe00036c64f0a x10: ffff600036c64f0a x9 : dfff800000000000
x8 : 00009fffc939b0f6 x7 : ffff0001b6327853 x6 : 0000000000000001
x5 : ffff0001b6327850 x4 : ffff600036c64f0b x3 : ffff8000803c35bc
x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000e16b9180
Call trace:
__might_sleep+0xf4/0x140
mutex_lock+0x84/0x124
io_handle_tw_list+0xf4/0x260
tctx_task_work_run+0x94/0x340
io_run_task_work+0x1ec/0x3c0
io_uring_cancel_generic+0x364/0x524
io_sq_thread+0x820/0x124c
ret_from_fork+0x10/0x20
Cc: stable@vger.kernel.org
Fixes: af5d68f889 ("io_uring/sqpoll: manage task_work privately")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When an application uses SQPOLL, it must wait for the SQPOLL thread to
consume SQE entries, if it fails to get an sqe when calling
io_uring_get_sqe(). It can do so by calling io_uring_enter(2) with the
flag value of IORING_ENTER_SQ_WAIT. In liburing, this is generally done
with io_uring_sqring_wait(). There's a natural expectation that once
this call returns, a new SQE entry can be retrieved, filled out, and
submitted. However, the kernel uses the cached sq head to determine if
the SQRING is full or not. If the SQPOLL thread is currently in the
process of submitting SQE entries, it may have updated the cached sq
head, but not yet committed it to the SQ ring. Hence the kernel may find
that there are SQE entries ready to be consumed, and return successfully
to the application. If the SQPOLL thread hasn't yet committed the SQ
ring entries by the time the application returns to userspace and
attempts to get a new SQE, it will fail getting a new SQE.
Fix this by having io_sqring_full() always use the user visible SQ ring
head entry, rather than the internally cached one.
Cc: stable@vger.kernel.org # 5.10+
Link: https://github.com/axboe/liburing/discussions/1267
Reported-by: Benedek Thaler <thaler@thaler.hu>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When an io_uring request needs blocking context we offload it to the
io_uring's thread pool called io-wq. We can get there off ->uring_cmd
by returning -EAGAIN, but there is no straightforward way of doing that
from an asynchronous callback. Add a helper that would transfer a
command to a blocking context.
Note, we do an extra hop via task_work before io_queue_iowq(), that's a
limitation of io_uring infra we have that can likely be lifted later
if that would ever become a problem.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/f735f807d7c8ba50c9452c69dfe5d3e9e535037b.1726072086.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Waiting for events with io_uring has two knobs that can be set:
1) The number of events to wake for
2) The timeout associated with the event
Waiting will abort when either of those conditions are met, as expected.
This adds support for a third event, which is associated with the number
of events to wait for. Applications generally like to handle batches of
completions, and right now they'd set a number of events to wait for and
the timeout for that. If no events have been received but the timeout
triggers, control is returned to the application and it can wait again.
However, if the application doesn't have anything to do until events are
reaped, then it's possible to make this waiting more efficient.
For example, the application may have a latency time of 50 usecs and
wanting to handle a batch of 8 requests at the time. If it uses 50 usecs
as the timeout, then it'll be doing 20K context switches per second even
if nothing is happening.
This introduces the notion of min batch wait time. If the min batch wait
time expires, then we'll return to userspace if we have any events at all.
If none are available, the general wait time is applied. Any request
arriving after the min batch wait time will cause waiting to stop and
return control to the application.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a new registration opcode IORING_REGISTER_CLOCK, which allows the
user to select which clock id it wants to use with CQ waiting timeouts.
It only allows a subset of all posix clocks and currently supports
CLOCK_MONOTONIC and CLOCK_BOOTTIME.
Suggested-by: Lewis Baker <lewissbaker@gmail.com>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/98f2bc8a3c36cdf8f0e6a275245e81e903459703.1723039801.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
It's more natural to use ktime/ns instead of keeping around usec,
especially since we're comparing it against user provided timers,
so convert napi busy poll internal handling to ktime. It's also nicer
since the type (ktime_t vs unsigned long) now tells the unit of measure.
Keep everything as ktime, which we convert to/from micro seconds for
IORING_[UN]REGISTER_NAPI. The net/ busy polling works seems to work with
usec, however it's not real usec as shift by 10 is used to get it from
nsecs, see busy_loop_current_time(), so it's easy to get truncated nsec
back and we get back better precision.
Note, we can further improve it later by removing the truncation and
maybe convincing net/ to use ktime/ns instead.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/95e7ec8d095069a3ed5d40a4bc6f8b586698bc7e.1722003776.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This helper will post a CQE, and can be called from task_work where we
now that the ctx is already properly locked and that deferred
completions will get flushed later on.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
All our task_work handling is targeted at the state in the io_kiocb
itself, which is what it is being used for. However, MSG_RING rolls its
own task_work handling, ignoring how that is usually done.
In preparation for switching MSG_RING to be able to use the normal
task_work handling, add io_req_task_work_add_remote() which allows the
caller to pass in the target io_ring_ctx.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This is pretty nicely abstracted already, but let's move it to a separate
file rather than have it in the main io_uring file. With that, we can
also move the io_ev_fd struct and enum out of global scope.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In some ways, it just "happens to work" currently with using the ops
field for both the free and signaling bit. But it depends on ordering
of operations in terms of freeing and signaling. Clean it up and use the
usual refs == 0 under RCU read side lock to determine if the ev_fd is
still valid, and use the reference to gate the freeing as well.
Fixes: 21a091b970 ("io_uring: signal registered eventfd to process deferred task work")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Allowing retries for everything is arguably the right thing to do, now
that every command type is async read from the start. But it's exposed a
few issues around missing check for a retry (which cca6571381 exposed),
and the fixup commit for that isn't necessarily 100% sound in terms of
iov_iter state.
For now, just revert these two commits. This unfortunately then re-opens
the fact that -EAGAIN can get bubbled to userspace for some cases where
the kernel very well could just sanely retry them. But until we have all
the conditions covered around that, we cannot safely enable that.
This reverts commit df604d2ad4.
This reverts commit cca6571381.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
A previous commit removed the checking on whether or not it was possible
to retry a request, since it's now possible to retry any of them. This
would previously have caused the request to have been ended with an error,
but now the retry condition can simply get lost instead.
Cleanup the retry handling and always just punt it to task_work, which
will queue it with io-wq appropriately.
Reported-by: Changhui Zhong <czhong@redhat.com>
Tested-by: Ming Lei <ming.lei@redhat.com>
Fixes: cca6571381 ("io_uring/rw: cleanup retry path")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move the related code from io_uring.c into memmap.c. No functional
changes in this patch, just cleaning it up a bit now that the full
transition is done.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Rather than use remap_pfn_range() for this and manually free later,
switch to using vm_insert_page() and have it Just Work.
This requires a bit of effort on the mmap lookup side, as the ctx
uring_lock isn't held, which otherwise protects buffer_lists from being
torn down, and it's not safe to grab from mmap context that would
introduce an ABBA deadlock between the mmap lock and the ctx uring_lock.
Instead, lookup the buffer_list under RCU, as the the list is RCU freed
already. Use the existing reference count to determine whether it's
possible to safely grab a reference to it (eg if it's not zero already),
and drop that reference when done with the mapping. If the mmap
reference is the last one, the buffer_list and the associated memory can
go away, since the vma insertion has references to the inserted pages at
that point.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Rather than use remap_pfn_range() for this and manually free later,
switch to using vm_insert_pages() and have it Just Work.
If possible, allocate a single compound page that covers the range that
is needed. If that works, then we can just use page_address() on that
page. If we fail to get a compound page, allocate single pages and use
vmap() to map them into the kernel virtual address space.
This just covers the rings/sqes, the other remaining user of the mmap
remap_pfn_range() user will be converted separately. Once that is done,
we can kill the old alloc/free code.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_task_work_pending() uses wq_list_empty() on ctx->work_llist, but it's
not an io_wq_work_list, it's a struct llist_head. They both have
->first as head-of-list, and it turns out the checks are identical. But
be proper and use the right helper.
Fixes: dac6a0eae7 ("io_uring: ensure iopoll runs local task work as well")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
