io_uring/io-wq: always retry worker create on ERESTART*
PATCH 1/1 io_uring/io-wq: always retry worker create on ERESTART*
If a task has a pending signal when create_io_thread() is called, copy_process() will return -ERESTARTNOINTR. io_should_retry_thread() will request a retry of create_io_thread() up to WORKER_INIT_LIMIT = 3 times. If all retries fail, the io_uring request will fail with ECANCELED.
Commit 3918315c5dc ("io-wq: backoff when retrying worker creation") added a linear backoff to allow the thread to handle its signal before the retry. However, a thread receiving frequent signals may get unlucky and have a signal pending at every retry. Since the userspace task doesn't control when it receives signals, there's no easy way for it to prevent the create_io_thread() failure due to pending signals. The task may also lack the information necessary to regenerate the canceled SQE. So always retry the create_io_thread() on the ERESTART* errors, analogous to what a fork() syscall would do. EAGAIN can occur due to various persistent conditions such as exceeding RLIMIT_NPROC, so respect the WORKER_INIT_LIMIT retry limit for EAGAIN errors.
--- io_uring/io-wq.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
io_uring/io-wq.c +3 -2
@@ -803,15 +803,16 @@ static inline bool io_should_retry_thread(struct io_worker *worker, long err)* Prevent perpetual task_work retry, if the task (or its group) is * exiting. */ if (fatal_signal_pending (current))return false; -if (worker->init_retries++ >=WORKER_INIT_LIMIT ) -return false; + worker->init_retries++;switch (err) {case -EAGAIN : +return worker->init_retries <=WORKER_INIT_LIMIT ; +/* Analogous to a fork() syscall, always retry on a restartable error */ case -ERESTARTSYS :case -ERESTARTNOINTR :case -ERESTARTNOHAND :return true;default :
Signature · 1 line
2.45.2
Quoted 17 lines · Caleb Sander Mateos
If a task has a pending signal when
create_io_thread()is called,copy_process()will return-ERESTARTNOINTR.io_should_retry_thread()will request a retry ofcreate_io_thread()up toWORKER_INIT_LIMIT= 3 times. If all retries fail, theio_uringrequest will fail with ECANCELED.
Commit3918315c5dc("io-wq: backoff when retrying worker creation") added a linear backoff to allow the thread to handle its signal before the retry. However, a thread receiving frequent signals may get unlucky and have a signal pending at every retry. Since the userspace task doesn't control when it receives signals, there's no easy way for it to prevent thecreate_io_thread()failure due to pending signals. The task may also lack the information necessary to regenerate the canceled SQE. So always retry thecreate_io_thread()on the ERESTART* errors, analogous to what afork()syscall would do. EAGAIN can occur due to various persistent conditions such as exceedingRLIMIT_NPROC, so respect theWORKER_INIT_LIMITretry limit for EAGAIN errors.
Quoted 1 line
[...]
Applied, thanks!
[1/1] io_uring/io-wq: always retry worker create on ERESTART*
commit: 777dfd696d3db9b7b08a41c3c03554ce0ba6c94e
Best regards,
Signature · 1 line
Jens Axboe