From: Dmitry Kadashev <[email protected]>
To: Pavel Begunkov <[email protected]>
Cc: Jens Axboe <[email protected]>, Josef <[email protected]>,
Norman Maurer <[email protected]>,
io-uring <[email protected]>
Subject: Re: "Cannot allocate memory" on ring creation (not RLIMIT_MEMLOCK)
Date: Tue, 22 Dec 2020 18:06:09 +0700 [thread overview]
Message-ID: <CAOKbgA5xZSpMWGfDpetXqVck4fvC9xkmKuWYV8nrpOBqPmCfAQ@mail.gmail.com> (raw)
In-Reply-To: <CAOKbgA7shBKAnVKXQxd6PadiZi0O7UZZBZ6rHnr3HnuDdtx3ng@mail.gmail.com>
On Tue, Dec 22, 2020 at 6:04 PM Dmitry Kadashev <[email protected]> wrote:
>
> On Tue, Dec 22, 2020 at 11:11 AM Pavel Begunkov <[email protected]> wrote:
> >
> > On 22/12/2020 03:35, Pavel Begunkov wrote:
> > > On 21/12/2020 11:00, Dmitry Kadashev wrote:
> > > [snip]
> > >>> We do not share rings between processes. Our rings are accessible from different
> > >>> threads (under locks), but nothing fancy.
> > >>>
> > >>>> In other words, if you kill all your io_uring applications, does it
> > >>>> go back to normal?
> > >>>
> > >>> I'm pretty sure it does not, the only fix is to reboot the box. But I'll find an
> > >>> affected box and double check just in case.
> > >
> > > I can't spot any misaccounting, but I wonder if it can be that your memory is
> > > getting fragmented enough to be unable make an allocation of 16 __contiguous__
> > > pages, i.e. sizeof(sqe) * 1024
> > >
> > > That's how it's allocated internally:
> > >
> > > static void *io_mem_alloc(size_t size)
> > > {
> > > gfp_t gfp_flags = GFP_KERNEL | __GFP_ZERO | __GFP_NOWARN | __GFP_COMP |
> > > __GFP_NORETRY;
> > >
> > > return (void *) __get_free_pages(gfp_flags, get_order(size));
> > > }
> > >
> > > What about smaller rings? Can you check io_uring of what SQ size it can allocate?
> > > That can be a different program, e.g. modify a bit liburing/test/nop.
> >
> > Even better to allocate N smaller rings, where N = 1024 / SQ_size
> >
> > static int try_size(int sq_size)
> > {
> > int ret = 0, i, n = 1024 / sq_size;
> > static struct io_uring rings[128];
> >
> > for (i = 0; i < n; ++i) {
> > if (io_uring_queue_init(sq_size, &rings[i], 0) < 0) {
> > ret = -1;
> > break;
> > }
> > }
> > for (i -= 1; i >= 0; i--)
> > io_uring_queue_exit(&rings[i]);
> > return ret;
> > }
> >
> > int main()
> > {
> > int size;
> >
> > for (size = 1024; size >= 2; size /= 2) {
> > if (!try_size(size)) {
> > printf("max size %i\n", size);
> > return 0;
> > }
> > }
> >
> > printf("can't allocate %i\n", size);
> > return 0;
> > }
>
> Unfortunately I've rebooted the box I've used for tests yesterday, so I can't
> try this there. Also I was not able to come up with an isolated reproducer for
> this yet.
>
> The good news is I've found a relatively easy way to provoke this on a test VM
> using our software. Our app runs with "admin" user perms (plus some
> capabilities), it bumps RLIMIT_MEMLOCK to infinity on start. I've also created
> an user called 'ioutest' to run the check for ring sizes using a different user.
>
> I've modified the test program slightly, to show the number of rings
> successfully
> created on each iteration and the actual error message (to debug a problem I was
> having with it, but I've kept this after that). Here is the output:
>
> # sudo -u admin bash -c 'ulimit -a' | grep locked
> max locked memory (kbytes, -l) 1024
>
> # sudo -u ioutest bash -c 'ulimit -a' | grep locked
> max locked memory (kbytes, -l) 1024
>
> # sudo -u admin ./iou-test1
> Failed after 0 rings with 1024 size: Cannot allocate memory
> Failed after 0 rings with 512 size: Cannot allocate memory
> Failed after 0 rings with 256 size: Cannot allocate memory
> Failed after 0 rings with 128 size: Cannot allocate memory
> Failed after 0 rings with 64 size: Cannot allocate memory
> Failed after 0 rings with 32 size: Cannot allocate memory
> Failed after 0 rings with 16 size: Cannot allocate memory
> Failed after 0 rings with 8 size: Cannot allocate memory
> Failed after 0 rings with 4 size: Cannot allocate memory
> Failed after 0 rings with 2 size: Cannot allocate memory
> can't allocate 1
>
> # sudo -u ioutest ./iou-test1
> max size 1024
>
> # ps ax | grep wq
> 8 ? I< 0:00 [mm_percpu_wq]
> 121 ? I< 0:00 [tpm_dev_wq]
> 124 ? I< 0:00 [devfreq_wq]
> 20593 pts/1 S+ 0:00 grep --color=auto wq
This was on kernel 5.6.7, I'm going to try this on 5.10.1 now.
--
Dmitry Kadashev
next prev parent reply other threads:[~2020-12-22 11:07 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-17 8:19 "Cannot allocate memory" on ring creation (not RLIMIT_MEMLOCK) Dmitry Kadashev
2020-12-17 8:26 ` Norman Maurer
2020-12-17 8:36 ` Dmitry Kadashev
2020-12-17 8:40 ` Dmitry Kadashev
2020-12-17 10:38 ` Josef
2020-12-17 11:10 ` Dmitry Kadashev
2020-12-17 13:43 ` Victor Stewart
2020-12-18 9:20 ` Dmitry Kadashev
2020-12-18 17:22 ` Jens Axboe
2020-12-18 15:26 ` Jens Axboe
2020-12-18 17:21 ` Josef
2020-12-18 17:23 ` Jens Axboe
2020-12-19 2:49 ` Josef
2020-12-19 16:13 ` Jens Axboe
2020-12-19 16:29 ` Jens Axboe
2020-12-19 17:11 ` Jens Axboe
2020-12-19 17:34 ` Norman Maurer
2020-12-19 17:38 ` Jens Axboe
2020-12-19 20:51 ` Josef
2020-12-19 21:54 ` Jens Axboe
2020-12-19 23:13 ` Jens Axboe
2020-12-19 23:42 ` Josef
2020-12-19 23:42 ` Pavel Begunkov
2020-12-20 0:25 ` Jens Axboe
2020-12-20 0:55 ` Pavel Begunkov
2020-12-21 10:35 ` Dmitry Kadashev
2020-12-21 10:49 ` Dmitry Kadashev
2020-12-21 11:00 ` Dmitry Kadashev
2020-12-21 15:36 ` Pavel Begunkov
2020-12-22 3:35 ` Pavel Begunkov
2020-12-22 4:07 ` Pavel Begunkov
2020-12-22 11:04 ` Dmitry Kadashev
2020-12-22 11:06 ` Dmitry Kadashev [this message]
2020-12-22 13:13 ` Dmitry Kadashev
2020-12-22 16:33 ` Pavel Begunkov
2020-12-23 8:39 ` Dmitry Kadashev
2020-12-23 9:38 ` Dmitry Kadashev
2020-12-23 11:48 ` Dmitry Kadashev
2020-12-23 12:27 ` Pavel Begunkov
2020-12-20 1:57 ` Pavel Begunkov
2020-12-20 7:13 ` Josef
2020-12-20 13:00 ` Pavel Begunkov
2020-12-20 14:19 ` Pavel Begunkov
2020-12-20 15:56 ` Josef
2020-12-20 15:58 ` Pavel Begunkov
2020-12-20 16:14 ` Jens Axboe
2020-12-20 16:59 ` Josef
2020-12-20 18:23 ` Josef
2020-12-20 18:41 ` Pavel Begunkov
2020-12-21 8:22 ` Josef
2020-12-21 15:30 ` Pavel Begunkov
2020-12-21 10:31 ` Dmitry Kadashev
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAOKbgA5xZSpMWGfDpetXqVck4fvC9xkmKuWYV8nrpOBqPmCfAQ@mail.gmail.com \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox