- 25 Jul, 2022 40 commits
-
-
Pavel Begunkov authored
Some io_uring-eventfd users assume that there won't be spurious wakeups. That assumption has to be honoured by all io_cqring_ev_posted() callers, which is inconvenient and from time to time leads to problems but should be maintained to not break the userspace. Instead of making the callers track whether a CQE was posted or not, hide it inside io_eventfd_signal(). It saves ->cached_cq_tail it saw last time and triggers the eventfd only when ->cached_cq_tail changed since then. Signed-off-by:
Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/0ffc66bae37a2513080b601e4370e147faaa72c5.1655684496.git.asml.silence@gmail.comSigned-off-by:
Jens Axboe <axboe@kernel.dk>
-
Pavel Begunkov authored
clang complains on bitwise operations with bools, add a bit more verbosity to better show that we want to call io_poll_remove_all_table() twice but with different arguments. Reported-by:
Nathan Chancellor <nathan@kernel.org> Signed-off-by:
-
Pavel Begunkov authored
Don't spin trying to cancel timeouts that are reachable but not cancellable, e.g. already executing. Signed-off-by:
-
Pavel Begunkov authored
io_uring_try_cancel_requests() loops until there is nothing left to do with the ring, however there might be several rings and they might have dependencies between them, e.g. via poll requests. Instead of cancelling rings one by one, try to cancel them all and only then loop over if we still potenially some work to do. Signed-off-by:
-
Pavel Begunkov authored
It's not clear how widely used IOSQE_CQE_SKIP_SUCCESS is, and how often ->flush_cqes flag prevents from completion being flushed. Sometimes it's high level of concurrency that enables it at least for one CQE, but sometimes it doesn't save much because nobody waiting on the CQ. Remove ->flush_cqes flag and the optimisation, it should benefit the normal use case. Note, that there is no spurious eventfd problem with that as checks for spuriousness were incorporated into io_eventfd_signal(). Signed-off-by:
-
Pavel Begunkov authored
Move io_eventfd_signal() in the sources without any changes and kill its forward declaration. Signed-off-by:
-
Pavel Begunkov authored
It's a good idea to first do forward declarations and then inline helpers, otherwise there will be keep stumbling on dependencies between them. Signed-off-by:
-
Pavel Begunkov authored
We don't post events in __io_commit_cqring_flush() anymore but send all requests to tw, so no need to do io_commit_cqring() there. Signed-off-by:
-
Jens Axboe authored
Commit 3a3d47fa9cfd ("io_uring: make io_uring_types.h public") moved a bunch of io_uring types to a kernel wide header, so we could make tracing a bit saner rather than pass in a ton of arguments. However, there are a few types in there that are not really needed to be system wide. Move the cancel data and mapped buffers back to the appropriate io_uring local headers. Signed-off-by: -
Pavel Begunkov authored
We have lots of trace events accepting an io_uring request and wanting to print some of its fields like user_data, opcode, flags and so on. However, as trace points were unaware of io_uring structures, we had to pass all the fields as arguments. Teach trace/events/io_uring.h about struct io_kiocb and stop the misery of passing a horde of arguments to trace helpers. Signed-off-by:
-
Pavel Begunkov authored
Move io_uring types to linux/include, need them public so tracing can see the definitions and we can clean trace/events/io_uring.h Signed-off-by:
-
Pavel Begunkov authored
io_uring/io_uring.h already includes io_uring_types.h, no need to include it every time. Kill it in a bunch of places, it prepares us for following patches. Signed-off-by:
-
Pavel Begunkov authored
With IORING_SETUP_CQE32 ->cqe_cached doesn't store a real address but rather an implicit offset into cqes. Store the real cqe pointer and increment it accordingly if CQE32. Signed-off-by:
-
Pavel Begunkov authored
Deduplicate calls to io_get_cqe() from __io_fill_cqe_req(). Signed-off-by:
-
Pavel Begunkov authored
Deduplicate two trace_io_uring_complete() calls in __io_fill_cqe_req(). Signed-off-by:
-
Pavel Begunkov authored
__io_fill_cqe_req() is hot and inlined, we want it to be as small as possible. Add io_req_cqe_overflow() accepting only a request and doing all overflow accounting, and replace with it two calls to 6 argument io_cqring_event_overflow(). Signed-off-by:
-
Pavel Begunkov authored
__io_get_cqe() is not as hot as io_get_cqe(), no need to inline it, it sheds ~500B from the binary. Signed-off-by:
-
Pavel Begunkov authored
Deduplicate some code and add a helper for filling an aux CQE, locking and notification. Signed-off-by:
-
Hao Xu authored
Add comments to explain why it is always under uring lock when incrementing head in __io_kbuf_recycle. And rectify one comemnt about kbuf consuming in iowq case. Signed-off-by:
Hao Xu <howeyxu@tencent.com> Link: https://lore.kernel.org/r/20220617050429.94293-1-hao.xu@linux.devSigned-off-by:
-
Pavel Begunkov authored
Currently we do two extra spin lock/unlock pairs to add a poll/apoll request to the cancellation hash table and remove it from there. On the submission side we often already hold ->uring_lock and tw completion is likely to hold it as well. Add a second cancellation hash table protected by ->uring_lock. In concerns for latency because of a need to have the mutex locked on the completion side, use the new table only in following cases: 1) IORING_SETUP_SINGLE_ISSUER: only one task grabs uring_lock, so there is little to no contention and so the main tw hander will almost always end up grabbing it before calling callbacks. 2) IORING_SETUP_SQPOLL: same as with single issuer, only one task is a major user of ->uring_lock. 3) apoll: we normally grab the lock on the completion side anyway to execute the request, so it's free. Signed-off-by:
-
Pavel Begunkov authored
Poll cancellation will be soon need to grab ->uring_lock inside, pass the locking state, i.e. issue_flags, inside the cancellation functions. Signed-off-by:
-
Pavel Begunkov authored
Instead of passing around a pointer to hash buckets, add a bit of type safety and wrap it into a structure. Signed-off-by:
-
Pavel Begunkov authored
In preparation for having multiple cancellation hash tables, pass a table pointer into io_poll_find() and other poll cancel functions. Signed-off-by:
-
Pavel Begunkov authored
Add a new IORING_SETUP_SINGLE_ISSUER flag and the userspace visible part of it, i.e. put limitations of submitters. Also, don't allow it together with IOPOLL as we're not going to put it to good use. Signed-off-by:
-
Pavel Begunkov authored
Use io_req_task_complete() for poll request completions, so it can utilise state completions and save lots of unnecessary locking. Signed-off-by:
-
Pavel Begunkov authored
Add a variable for the number of hash buckets in io_ring_ctx_alloc(), makes it more readable. Signed-off-by:
-
Pavel Begunkov authored
Don't allocate to many hash/cancellation buckets, there might be too many, clamp it to 8 bits, or 256 * 64B = 16KB. We don't usually have too many requests, and 256 buckets should be enough, especially since we do hash search only in the cancellation path. Signed-off-by:
-
Pavel Begunkov authored
Get rid of an unnecessary extra goto in io_try_cancel() and simplify the function. Signed-off-by:
-
Pavel Begunkov authored
Instead of using implicit knowledge of what is locked or not after io_poll_find() and co returns, pass back a pointer to the locked bucket if any. If set the user must to unlock the spinlock. Signed-off-by:
-
Hao Xu authored
Add a new io_hash_bucket structure so that each bucket in cancel_hash has separate spinlock. Use per entry lock for cancel_hash, this removes some completion lock invocation and remove contension between different cancel_hash entries. Signed-off-by:
-
Hao Xu authored
We now don't need to set req->refcount for poll requests since the reworked poll code ensures no request release race. Signed-off-by:
-
Pavel Begunkov authored
io_put_kbuf() is huge, don't bloat the kernel with inlining. Signed-off-by:
-
Pavel Begunkov authored
Clean up io_req_task_complete() and deduplicate io_put_kbuf() calls. Signed-off-by:
-
Pavel Begunkov authored
REQ_F_COMPLETE_INLINE is only needed to delay queueing into the completion list to io_queue_sqe() as __io_req_complete() is inlined and we don't want to bloat the kernel. As now we complete in a more centralised fashion in io_issue_sqe() we can get rid of the flag and queue to the list directly. Signed-off-by:
-
Pavel Begunkov authored
io_issue_sqe() from the io_uring core knows how to complete requests based on the returned error code, we can delegate io_read()/io_write() completion to it. Make kiocb_done() to return the right completion code and propagate it. Signed-off-by:
-
Jens Axboe authored
Signed-off-by: -
Pavel Begunkov authored
io_req_task_complete() enqueues requests for state completion itself, no need for REQ_F_COMPLETE_INLINE, which is only serve the purpose of not bloating the kernel. Signed-off-by:
-
Pavel Begunkov authored
All ctx->check_cq events are slow path, don't test every single flag one by one in the hot path, but add a common guarding if. Signed-off-by:
-
Pavel Begunkov authored
Luckily, nnobody completes multi-apoll requests outside the polling functions, but don't set IO_URING_F_COMPLETE_DEFER in any case as there is nobody who is catching REQ_F_COMPLETE_INLINE, and so will leak requests if used. Signed-off-by:
-
Pavel Begunkov authored
There can be only 16 registered rings, no need to allocate an array for them separately but store it in tctx. Signed-off-by:
-
