summaryrefslogtreecommitdiffstats
path: root/io_uring/io_uring.c
AgeCommit message (Collapse)AuthorFilesLines
2023-01-27io_uring: always prep_async for drain requestsDylan Yudaken1-10/+8
Drain requests all go through io_drain_req, which has a quick exit in case there is nothing pending (ie the drain is not useful). In that case it can run the issue the request immediately. However for safety it queues it through task work. The problem is that in this case the request is run asynchronously, but the async work has not been prepared through io_req_prep_async. This has not been a problem up to now, as the task work always would run before returning to userspace, and so the user would not have a chance to race with it. However - with IORING_SETUP_DEFER_TASKRUN - this is no longer the case and the work might be defered, giving userspace a chance to change data being referred to in the request. Instead _always_ prep_async for drain requests, which is simpler anyway and removes this issue. Cc: stable@vger.kernel.org Fixes: c0e0d6ba25f1 ("io_uring: add IORING_SETUP_DEFER_TASKRUN") Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20230127105911.2420061-1-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2023-01-20io_uring/msg_ring: fix remote queue to disabled ringPavel Begunkov1-2/+2
IORING_SETUP_R_DISABLED rings don't have the submitter task set, so it's not always safe to use ->submitter_task. Disallow posting msg_ring messaged to disabled rings. Also add task NULL check for loosy sync around testing for IORING_SETUP_R_DISABLED. Cc: stable@vger.kernel.org Fixes: 6d043ee1164ca ("io_uring: do msg_ring in target task via tw") Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2023-01-05io_uring: fix CQ waiting timeout handlingPavel Begunkov1-3/+3
Jiffy to ktime CQ waiting conversion broke how we treat timeouts, in particular we rearm it anew every time we get into io_cqring_wait_schedule() without adjusting the timeout. Waiting for 2 CQEs and getting a task_work in the middle may double the timeout value, or even worse in some cases task may wait indefinitely. Cc: stable@vger.kernel.org Fixes: 228339662b398 ("io_uring: don't convert to jiffies for waiting on timeouts") Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/f7bffddd71b08f28a877d44d37ac953ddb01590d.1672915663.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2023-01-03io_uring: lockdep annotate CQ lockingPavel Begunkov1-3/+2
Locking around CQE posting is complex and depends on options the ring is created with, add more thorough lockdep annotations checking all invariants. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/aa3770b4eacae3915d782cc2ab2f395a99b4b232.1672795976.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2023-01-03io_uring: pin context while queueing deferred twPavel Begunkov1-1/+7
Unlike normal tw, nothing prevents deferred tw to be executed right after an tw item added to ->work_llist in io_req_local_work_add(). For instance, the waiting task may get waken up by CQ posting or a normal tw. Thus we need to pin the ring for the rest of io_req_local_work_add() Cc: stable@vger.kernel.org Fixes: c0e0d6ba25f18 ("io_uring: add IORING_SETUP_DEFER_TASKRUN") Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/1a79362b9c10b8523ef70b061d96523650a23344.1672795998.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-23io_uring: check for valid register opcode earlierJens Axboe1-2/+3
We only check the register opcode value inside the restricted ring section, move it into the main io_uring_register() function instead and check it up front. Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-21io_uring: finish waiting before flushing overflow entriesJens Axboe1-9/+16
If we have overflow entries being generated after we've done the initial flush in io_cqring_wait(), then we could be flushing them in the main wait loop as well. If that's done after having added ourselves to the cq_wait waitqueue, then the task state can be != TASK_RUNNING when we enter the overflow flush. Check for the need to overflow flush, and finish our wait cycle first if we have to do so. Reported-and-tested-by: syzbot+cf6ea1d6bb30a4ce10b2@syzkaller.appspotmail.com Link: https://lore.kernel.org/io-uring/000000000000cb143a05f04eee15@google.com/ Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-17io_uring: include task_work run after scheduling in wait for eventsJens Axboe1-1/+10
It's quite possible that we got woken up because task_work was queued, and we need to process this task_work to generate the events waited for. If we return to the wait loop without running task_work, we'll end up adding the task to the waitqueue again, only to call io_cqring_wait_schedule() again which will run the task_work. This is less efficient than it could be, as it requires adding to the cq_wait queue again. It also triggers the wakeup path for completions as cq_wait is now non-empty with the task itself, and it'll require another lock grab and deletion to remove ourselves from the waitqueue. Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-15io_uring: use call_rcu_hurry if signaling an eventfdDylan Yudaken1-1/+1
io_uring uses call_rcu in the case it needs to signal an eventfd as a result of an eventfd signal, since recursing eventfd signals are not allowed. This should be calling the new call_rcu_hurry API to not delay the signal. Signed-off-by: Dylan Yudaken <dylany@meta.com> Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Cc: Paul E. McKenney <paulmck@kernel.org> Acked-by: Paul E. McKenney <paulmck@kernel.org> Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Link: https://lore.kernel.org/r/20221215184138.795576-1-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-15io_uring: fix overflow handling regressionPavel Begunkov1-1/+1
Because the single task locking series got reordered ahead of the timeout and completion lock changes, two hunks inadvertently ended up using __io_fill_cqe_req() rather than io_fill_cqe_req(). This meant that we dropped overflow handling in those two spots. Reinstate the correct CQE filling helper. Fixes: f66f73421f0a ("io_uring: skip spinlocking for ->task_complete") Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-14io_uring: ease timeout flush locking requirementsPavel Begunkov1-5/+4
We don't need completion_lock for timeout flushing, don't take it. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/1e3dc657975ac445b80e7bdc40050db783a5935a.1670002973.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-14io_uring: revise completion_lock lockingPavel Begunkov1-2/+14
io_kill_timeouts() doesn't post any events but queues everything to task_work. Locking there is needed for protecting linked requests traversing, we should grab completion_lock directly instead of using io_cq_[un]lock helpers. Same goes for __io_req_find_next_prep(). Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/88e75d481a65dc295cb59722bb1cf76402d1c06b.1670002973.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-13Merge tag 'for-6.2/io_uring-next-2022-12-08' of git://git.kernel.dk/linuxLinus Torvalds1-57/+110
Pull io_uring updates part two from Jens Axboe: - Misc fixes (me, Lin) - Series from Pavel extending the single task exclusive ring mode, yielding nice improvements for the common case of having a single ring per thread (Pavel) - Cleanup for MSG_RING, removing our IOPOLL hack (Pavel) - Further poll cleanups and fixes (Pavel) - Misc cleanups and fixes (Pavel) * tag 'for-6.2/io_uring-next-2022-12-08' of git://git.kernel.dk/linux: (22 commits) io_uring/msg_ring: flag target ring as having task_work, if needed io_uring: skip spinlocking for ->task_complete io_uring: do msg_ring in target task via tw io_uring: extract a io_msg_install_complete helper io_uring: get rid of double locking io_uring: never run tw and fallback in parallel io_uring: use tw for putting rsrc io_uring: force multishot CQEs into task context io_uring: complete all requests in task context io_uring: don't check overflow flush failures io_uring: skip overflow CQE posting for dying ring io_uring: improve io_double_lock_ctx fail handling io_uring: dont remove file from msg_ring reqs io_uring: reshuffle issue_flags io_uring: don't reinstall quiesce node for each tw io_uring: improve rsrc quiesce refs checks io_uring: don't raw spin unlock to match cq_lock io_uring: combine poll tw handlers io_uring: improve poll warning handling io_uring: remove ctx variable in io_poll_check_events ...
2022-12-13Merge tag 'for-6.2/io_uring-2022-12-08' of git://git.kernel.dk/linuxLinus Torvalds1-91/+132
Pull io_uring updates from Jens Axboe: - Always ensure proper ordering in case of CQ ring overflow, which then means we can remove some work-arounds for that (Dylan) - Support completion batching for multishot, greatly increasing the efficiency for those (Dylan) - Flag epoll/eventfd wakeups done from io_uring, so that we can easily tell if we're recursing into io_uring again. Previously, this would have resulted in repeated multishot notifications if we had a dependency there. That could happen if an eventfd was registered as the ring eventfd, and we multishot polled for events on it. Or if an io_uring fd was added to epoll, and io_uring had a multishot request for the epoll fd. Test cases here: https://git.kernel.dk/cgit/liburing/commit/?id=919755a7d0096fda08fb6d65ac54ad8d0fe027cd Previously these got terminated when the CQ ring eventually overflowed, now it's handled gracefully (me). - Tightening of the IOPOLL based completions (Pavel) - Optimizations of the networking zero-copy paths (Pavel) - Various tweaks and fixes (Dylan, Pavel) * tag 'for-6.2/io_uring-2022-12-08' of git://git.kernel.dk/linux: (41 commits) io_uring: keep unlock_post inlined in hot path io_uring: don't use complete_post in kbuf io_uring: spelling fix io_uring: remove io_req_complete_post_tw io_uring: allow multishot polled reqs to defer completion io_uring: remove overflow param from io_post_aux_cqe io_uring: add lockdep assertion in io_fill_cqe_aux io_uring: make io_fill_cqe_aux static io_uring: add io_aux_cqe which allows deferred completion io_uring: allow defer completion for aux posted cqes io_uring: defer all io_req_complete_failed io_uring: always lock in io_apoll_task_func io_uring: remove iopoll spinlock io_uring: iopoll protect complete_post io_uring: inline __io_req_complete_put() io_uring: remove io_req_tw_post_queue io_uring: use io_req_task_complete() in timeout io_uring: hold locks for io_req_complete_failed io_uring: add completion locking for iopoll io_uring: kill io_cqring_ev_posted() and __io_cq_unlock_post() ...
2022-12-07io_uring: skip spinlocking for ->task_completePavel Begunkov1-19/+52
->task_complete was added to serialised CQE posting by doing it from the task context only (or fallback wq when the task is dead), and now we can use that to avoid taking ->completion_lock while filling CQ entries. The patch skips spinlocking only in two spots, __io_submit_flush_completions() and flushing in io_aux_cqe, it's safer and covers all cases we care about. Extra care is taken to force taking the lock while queueing overflow entries. It fundamentally relies on SINGLE_ISSUER to have only one task posting events. It also need to take into account overflowed CQEs, flushing of which happens in the cq wait path, and so this implementation also needs DEFER_TASKRUN to limit waiters. For the same reason we disable it for SQPOLL, and for IOPOLL as it won't benefit from it in any case. DEFER_TASKRUN, SQPOLL and IOPOLL requirement may be relaxed in the future. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/2a8c91fd82cfcdcc1d2e5bac7051fe2c183bda73.1670384893.git.asml.silence@gmail.com [axboe: modify to apply] Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-07io_uring: never run tw and fallback in parallelPavel Begunkov1-2/+10
Once we fallback a tw we want all requests to that task to be given to the fallback wq so we dont run it in parallel with the last, i.e. post PF_EXITING, tw run of the task. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/96f4987265c4312f376f206511c6af3e77aaf5ac.1670384893.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-07io_uring: use tw for putting rsrcPavel Begunkov1-0/+1
Use task_work for completing rsrc removals, it'll be needed later for spinlock optimisations. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/cbba5d53a11ee6fc2194dacea262c1d733c8b529.1670384893.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-07io_uring: complete all requests in task contextPavel Begunkov1-3/+11
This patch adds ctx->task_complete flag. If set, we'll complete all requests in the context of the original task. Note, this extends to completion CQE posting only but not io_kiocb cleanup / free, e.g. io-wq may free the requests in the free calllback. This flag will be used later for optimisations purposes. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/21ece72953f76bb2e77659a72a14326227ab6460.1670384893.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-07io_uring: don't check overflow flush failuresPavel Begunkov1-18/+6
The only way to fail overflowed CQEs flush is for CQ to be fully packed. There is one place checking for flush failures, i.e. io_cqring_wait(), but we limit the number to be waited for by the CQ size, so getting a failure automatically means that we're done with waiting. Don't check for failures, rarely but they might spuriously fail CQ waiting with -EBUSY. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/6b720a45c03345655517f8202cbd0bece2848fb2.1670384893.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-07io_uring: skip overflow CQE posting for dying ringPavel Begunkov1-14/+31
After io_ring_ctx_wait_and_kill() is called there should be no users poking into rings and so there is no need to post CQEs. So, instead of trying to post overflowed CQEs into the CQ, drop them. Also, do it in io_ring_exit_work() in a loop to reduce the number of contexts it can be executed from and even when it struggles to quiesce the ring we won't be leaving memory allocated for longer than needed. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/26d13751155a735a3029e24f8d9ca992f810419d.1670384893.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-07io_uring: dont remove file from msg_ring reqsPavel Begunkov1-1/+1
We should not be messing with req->file outside of core paths. Clearing it makes msg_ring non reentrant, i.e. luckily io_msg_send_fd() fails the request on failed io_double_lock_ctx() but clearly was originally intended to do retries instead. Cc: stable@vger.kernel.org Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/e5ac9edadb574fe33f6d727cb8f14ce68262a684.1670384893.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-12-07io_uring: Fix a null-ptr-deref in io_tctx_exit_cb()Harshit Mogalapalli1-1/+3
Syzkaller reports a NULL deref bug as follows: BUG: KASAN: null-ptr-deref in io_tctx_exit_cb+0x53/0xd3 Read of size 4 at addr 0000000000000138 by task file1/1955 CPU: 1 PID: 1955 Comm: file1 Not tainted 6.1.0-rc7-00103-gef4d3ea40565 #75 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0xcd/0x134 ? io_tctx_exit_cb+0x53/0xd3 kasan_report+0xbb/0x1f0 ? io_tctx_exit_cb+0x53/0xd3 kasan_check_range+0x140/0x190 io_tctx_exit_cb+0x53/0xd3 task_work_run+0x164/0x250 ? task_work_cancel+0x30/0x30 get_signal+0x1c3/0x2440 ? lock_downgrade+0x6e0/0x6e0 ? lock_downgrade+0x6e0/0x6e0 ? exit_signals+0x8b0/0x8b0 ? do_raw_read_unlock+0x3b/0x70 ? do_raw_spin_unlock+0x50/0x230 arch_do_signal_or_restart+0x82/0x2470 ? kmem_cache_free+0x260/0x4b0 ? putname+0xfe/0x140 ? get_sigframe_size+0x10/0x10 ? do_execveat_common.isra.0+0x226/0x710 ? lockdep_hardirqs_on+0x79/0x100 ? putname+0xfe/0x140 ? do_execveat_common.isra.0+0x238/0x710 exit_to_user_mode_prepare+0x15f/0x250 syscall_exit_to_user_mode+0x19/0x50 do_syscall_64+0x42/0xb0 entry_SYSCALL_64_after_hwframe+0x63/0xcd RIP: 0023:0x0 Code: Unable to access opcode bytes at 0xffffffffffffffd6. RSP: 002b:00000000fffb7790 EFLAGS: 00000200 ORIG_RAX: 000000000000000b RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 </TASK> Kernel panic - not syncing: panic_on_warn set ... This happens because the adding of task_work from io_ring_exit_work() isn't synchronized with canceling all work items from eg exec. The execution of the two are ordered in that they are both run by the task itself, but if io_tctx_exit_cb() is queued while we're canceling all work items off exec AND gets executed when the task exits to userspace rather than in the main loop in io_uring_cancel_generic(), then we can find current->io_uring == NULL and hit the above crash. It's safe to add this NULL check here, because the execution of the two paths are done by the task itself. Cc: stable@vger.kernel.org Fixes: d56d938b4bef ("io_uring: do ctx initiated file note removal") Reported-by: syzkaller <syzkaller@googlegroups.com> Signed-off-by: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com> Link: https://lore.kernel.org/r/20221206093833.3812138-1-harshit.m.mogalapalli@oracle.com [axboe: add code comment and also put an explanation in the commit msg] Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-30io_uring: don't raw spin unlock to match cq_lockPavel Begunkov1-1/+1
There is one newly added place when we lock ring with io_cq_lock() but unlocking is hand coded calling spin_unlock directly. It's ugly and troublesome in the long run. Make it consistent with the other completion locking. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/4ca4f0564492b90214a190cd5b2a6c76522de138.1669821213.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-30io_uring: kill io_poll_issue's PF_EXITING checkPavel Begunkov1-2/+0
We don't need to worry about checking PF_EXITING in io_poll_issue(). task works using the function should take care of it and never try to resubmit / retry if the task is dying. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/2e9dc998dc07507c759a0c9cb5d2fbea0710d58c.1669821213.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-25io_uring: keep unlock_post inlined in hot pathPavel Begunkov1-2/+9
This partially reverts 6c16fe3c16bdc ("io_uring: kill io_cqring_ev_posted() and __io_cq_unlock_post()") The redundancy of __io_cq_unlock_post() was always to keep it inlined into __io_submit_flush_completions(). Inline it back and rename with hope of clarifying the intention behind it. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/372a16c485fca44c069be2e92fc5e7332a1d7fd7.1669310258.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-25io_uring: spelling fixDylan Yudaken1-1/+1
s/pushs/pushes/ Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221125103412.1425305-3-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-25io_uring: remove io_req_complete_post_twDylan Yudaken1-1/+1
It's only used in one place. Inline it. Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221125103412.1425305-2-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-25io_uring: allow multishot polled reqs to defer completionDylan Yudaken1-1/+2
Until now there was no reason for multishot polled requests to defer completions as there was no functional difference. However now this will actually defer the completions, for a performance win. Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221124093559.3780686-10-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-25io_uring: remove overflow param from io_post_aux_cqeDylan Yudaken1-4/+8
The only call sites which would not allow overflow are also call sites which would use the io_aux_cqe as they care about ordering. So remove this parameter from io_post_aux_cqe. Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221124093559.3780686-9-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-25io_uring: add lockdep assertion in io_fill_cqe_auxDylan Yudaken1-0/+2
Add an assertion for the completion lock to io_fill_cqe_aux Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221124093559.3780686-8-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-25io_uring: make io_fill_cqe_aux staticDylan Yudaken1-2/+2
This is only used in io_uring.c Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221124093559.3780686-7-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-25io_uring: add io_aux_cqe which allows deferred completionDylan Yudaken1-0/+34
Use the just introduced deferred post cqe completion state when possible in io_aux_cqe. If not possible fallback to io_post_aux_cqe. This introduces a complication because of allow_overflow. For deferred completions we cannot know without locking the completion_lock if it will overflow (and even if we locked it, another post could sneak in and cause this cqe to be in overflow). However since overflow protection is mostly a best effort defence in depth to prevent infinite loops of CQEs for poll, just checking the overflow bit is going to be good enough and will result in at most 16 (array size of deferred cqes) overflows. Suggested-by: Pavel Begunkov <asml.silence@gmail.com> Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221124093559.3780686-6-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-25io_uring: allow defer completion for aux posted cqesDylan Yudaken1-3/+24
Multishot ops cannot use the compl_reqs list as the request must stay in the poll list, but that means they need to run each completion without benefiting from batching. Here introduce batching infrastructure for only small (ie 16 byte) CQEs. This restriction is ok because there are no use cases posting 32 byte CQEs. In the ring keep a batch of up to 16 posted results, and flush in the same way as compl_reqs. 16 was chosen through experimentation on a microbenchmark ([1]), as well as trying not to increase the size of the ring too much. This increases the size to 1472 bytes from 1216. [1]: https://github.com/DylanZA/liburing/commit/9ac66b36bcf4477bfafeff1c5f107896b7ae31cf Run with $ make -j && ./benchmark/reg.b -s 1 -t 2000 -r 10 Gives results: baseline 8309 k/s 8 18807 k/s 16 19338 k/s 32 20134 k/s Suggested-by: Pavel Begunkov <asml.silence@gmail.com> Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221124093559.3780686-5-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-25io_uring: defer all io_req_complete_failedDylan Yudaken1-9/+8
All failures happen under lock now, and can be deferred. To be consistent when the failure has happened after some multishot cqe has been deferred (and keep ordering), always defer failures. To make this obvious at the caller (and to help prevent a future bug) rename io_req_complete_failed to io_req_defer_failed. Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221124093559.3780686-4-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-23io_uring: iopoll protect complete_postPavel Begunkov1-6/+15
io_req_complete_post() may be used by iopoll enabled rings, grab locks in this case. That requires to pass issue_flags to propagate the locking state. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/cc6d854065c57c838ca8e8806f707a226b70fd2d.1669203009.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-23io_uring: inline __io_req_complete_put()Pavel Begunkov1-13/+7
Inline __io_req_complete_put() into io_req_complete_post(), there are no other users. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/1923a4dfe80fa877f859a22ed3df2d5fc8ecf02b.1669203009.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-23io_uring: remove io_req_tw_post_queuePavel Begunkov1-12/+0
Remove io_req_tw_post() and io_req_tw_post_queue(), we can use io_req_task_complete() instead. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/b9b73c08022c7f1457023ac841f35c0100e70345.1669203009.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-23io_uring: hold locks for io_req_complete_failedPavel Begunkov1-1/+5
A preparation patch, make sure we always hold uring_lock around io_req_complete_failed(). The only place deviating from the rule is io_cancel_defer_files(), queue a tw instead. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/70760344eadaecf2939287084b9d4ba5c05a6984.1669203009.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-22io_uring: kill io_cqring_ev_posted() and __io_cq_unlock_post()Jens Axboe1-13/+4
__io_cq_unlock_post() is identical to io_cq_unlock_post(), and io_cqring_ev_posted() has a single caller so migth as well just inline it there. Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-22io_uring: pass in EPOLL_URING_WAKE for eventfd signaling and wakeupsJens Axboe1-2/+2
Pass in EPOLL_URING_WAKE when signaling eventfd or doing poll related wakups, so that we can check for a circular event dependency between eventfd and epoll. If this flag is set when our wakeup handlers are called, then we know we have a dependency that needs to terminate multishot requests. eventfd and epoll are the only such possible dependencies. Cc: stable@vger.kernel.org # 6.0 Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-21io_uring: inline __io_req_complete_post()Pavel Begunkov1-8/+3
There is only one user of __io_req_complete_post(), inline it. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/ef4c9059950a3da5cf68df00f977f1fd13bd9306.1668597569.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-21io_uring: split tw fallback into a functionPavel Begunkov1-10/+15
When the target process is dying and so task_work_add() is not allowed we push all task_work item to the fallback workqueue. Move the part responsible for moving tw items out of __io_req_task_work_add() into a separate function. Makes it a bit cleaner and gives the compiler a bit of extra info. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/e503dab9d7af95470ca6b214c6de17715ae4e748.1668162751.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-21io_uring: inline io_req_task_work_add()Pavel Begunkov1-6/+1
__io_req_task_work_add() is huge but marked inline, that makes compilers to generate lots of garbage. Inline the wrapper caller io_req_task_work_add() instead. before and after: text data bss dec hex filename 47347 16248 8 63603 f873 io_uring/io_uring.o text data bss dec hex filename 45303 16248 8 61559 f077 io_uring/io_uring.o Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/26dc8c28ca0160e3269ef3e55c5a8b917c4d4450.1668162751.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-21io_uring: update outdated comment of callbacksLin Ma1-1/+1
Previous commit ebc11b6c6b87 ("io_uring: clean io-wq callbacks") rename io_free_work() into io_wq_free_work() for consistency. This patch also updates relevant comment to avoid misunderstanding. Fixes: ebc11b6c6b87 ("io_uring: clean io-wq callbacks") Signed-off-by: Lin Ma <linma@zju.edu.cn> Link: https://lore.kernel.org/r/20221110122103.20120-1-linma@zju.edu.cn Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-21io_uring: do not always force run task_work in io_uring_registerDylan Yudaken1-2/+0
Running task work when not needed can unnecessarily delay operations. Specifically IORING_SETUP_DEFER_TASKRUN tries to avoid running task work until the user requests it. Therefore do not run it in io_uring_register any more. The one catch is that io_rsrc_ref_quiesce expects it to have run in order to process all outstanding references, and so reorder it's loop to do this. Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221107123349.4106213-1-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-21io_uring: move kbuf put out of generic tw completePavel Begunkov1-6/+0
There are multiple users of io_req_task_complete() including zc notifications, but only read requests use selected buffers. As we already have an rw specific tw function, move io_put_kbuf() in there. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/94374c7649aaefc3a17808dc4701f25ccd457e25.1667557923.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-17io_uring: fix multishot accept request leaksPavel Begunkov1-1/+1
Having REQ_F_POLLED set doesn't guarantee that the request is executed as a multishot from the polling path. Fortunately for us, if the code thinks it's multishot issue when it's not, it can only ask to skip completion so leaking the request. Use issue_flags to mark multipoll issues. Cc: stable@vger.kernel.org Fixes: 390ed29b5e425 ("io_uring: add IORING_ACCEPT_MULTISHOT for accept") Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/7700ac57653f2823e30b34dc74da68678c0c5f13.1668710222.git.asml.silence@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-11-08io_uring: calculate CQEs from the user visible valueDylan Yudaken1-2/+8
io_cqring_wait (and it's wake function io_has_work) used cached_cq_tail in order to calculate the number of CQEs. cached_cq_tail is set strictly before the user visible rings->cq.tail However as far as userspace is concerned, if io_uring_enter(2) is called with a minimum number of events, they will verify by checking rings->cq.tail. It is therefore possible for io_uring_enter(2) to return early with fewer events visible to the user. Instead make the wait functions read from the user visible value, so there will be no discrepency. This is triggered eventually by the following reproducer: struct io_uring_sqe *sqe; struct io_uring_cqe *cqe; unsigned int cqe_ready; struct io_uring ring; int ret, i; ret = io_uring_queue_init(N, &ring, 0); assert(!ret); while(true) { for (i = 0; i < N; i++) { sqe = io_uring_get_sqe(&ring); io_uring_prep_nop(sqe); sqe->flags |= IOSQE_ASYNC; } ret = io_uring_submit(&ring); assert(ret == N); do { ret = io_uring_wait_cqes(&ring, &cqe, N, NULL, NULL); } while(ret == -EINTR); cqe_ready = io_uring_cq_ready(&ring); assert(!ret); assert(cqe_ready == N); io_uring_cq_advance(&ring, N); } Fixes: ad3eb2c89fb2 ("io_uring: split overflow state into SQ and CQ side") Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221108153016.1854297-1-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-10-27io_uring: unlock if __io_run_local_work locked insideDylan Yudaken1-4/+4
It is possible for tw to lock the ring, and this was not propogated out to io_run_local_work. This can cause an unlock to be missed. Instead pass a pointer to locked into __io_run_local_work. Fixes: 8ac5d85a89b4 ("io_uring: add local task_work run helper that is entered locked") Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221027144429.3971400-3-dylany@meta.com [axboe: WARN_ON() -> WARN_ON_ONCE() and add a minor comment] Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-10-27io_uring: use io_run_local_work_locked helperDylan Yudaken1-2/+1
prefer to use io_run_local_work_locked helper for consistency Signed-off-by: Dylan Yudaken <dylany@meta.com> Link: https://lore.kernel.org/r/20221027144429.3971400-2-dylany@meta.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-01 13:49:24 +0000