linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Florian Weimer <fweimer@redhat.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@kernel.org>, Darren Hart <darren@dvhart.com>,
	Yi Wang <wang.yi59@zte.com.cn>, Yang Tao <yang.tao172@zte.com.cn>,
	Oleg Nesterov <oleg@redhat.com>,
	Carlos O'Donell <carlos@redhat.com>,
	Alexander Viro <viro@zeniv.linux.org.uk>
Subject: Re: [patch 00/12] futex: Cure robust/PI futex exit races
Date: Mon, 11 Nov 2019 10:48:53 +0100	[thread overview]
Message-ID: <87woc6yc8q.fsf@oldenburg2.str.redhat.com> (raw)
In-Reply-To: <alpine.DEB.2.21.1911081223110.26566@nanos.tec.linutronix.de> (Thomas Gleixner's message of "Fri, 8 Nov 2019 12:51:53 +0100 (CET)")

* Thomas Gleixner:

> pthread_create() returns EAGAIN while the underlying problem is ENOMEM
> which causes this bonkers output:
>
>   error: pthread_create for thread 253 failed: Resource temporarily unavailable
>
> There is nothing temporarily. The process has its address space exhausted.

Thanks for analyzing the failure.  I thought we had already covered
that.  I've fixed the test locally and will submit the changes.  The
fixed test passes, as expected.

I expected that we've fixed all such occurrences of per-CPU thread
creation, but apparently not. 8-(

> That test's output is anyway strange:
>
>  info: Detected CPU set size (in bits): 225
>  info: Maximum test CPU: 255
>
> Interesting how it fits 256 CPUs into a cpuset with a size of 225 bits.

That's an unfortunate side effect of how the CPU set allocation works in
userspace.  The allocation uses a size meaured in bits (which are
rounded up, out of necessity), but the kernel interface is byte-based.
The kernel does not not know that some bits are padding, and happily
writes result data there.  So we get bits back for which no space had
been allocated explicitly.

I hesitated to clean this up because the story on the kernel side was
equally mystifying.  getaffinity requires space in the mask for CPUs
that are currently not present and whose affinity bits are not set.  Due
to firmware bugs, this means that we can cross the magic 1024 bits
boundary (corresponding to the 128 byte legacy mask size), and some
applications will refuse to start. 8-( There was considerable
controversy on the kernel side the last time this came up IIRC.

Thanks,
Florian


      reply	other threads:[~2019-11-11  9:49 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-06 21:55 [patch 00/12] futex: Cure robust/PI futex exit races Thomas Gleixner
2019-11-06 21:55 ` [patch 01/12] futex: Prevent robust futex exit race Thomas Gleixner
2019-11-06 21:55 ` [patch 02/12] futex: Move futex exit handling into futex code Thomas Gleixner
2019-11-07  9:24   ` Peter Zijlstra
2019-11-07  9:38     ` Thomas Gleixner
2019-11-15 18:19   ` [tip: locking/core] " tip-bot2 for Thomas Gleixner
2019-11-16 20:53     ` Borislav Petkov
2019-11-20  9:38   ` tip-bot2 for Thomas Gleixner
2019-11-06 21:55 ` [patch 03/12] futex: Replace PF_EXITPIDONE with a state Thomas Gleixner
2019-11-15 18:19   ` [tip: locking/core] " tip-bot2 for Thomas Gleixner
2019-11-20  9:38   ` tip-bot2 for Thomas Gleixner
2019-11-06 21:55 ` [patch 04/12] exit/exec: Seperate mm_release() Thomas Gleixner
2019-11-15 18:19   ` [tip: locking/core] " tip-bot2 for Thomas Gleixner
2019-11-20  9:38   ` tip-bot2 for Thomas Gleixner
2019-11-06 21:55 ` [patch 05/12] futex: Split futex_mm_release() for exit/exec Thomas Gleixner
2019-11-15 18:19   ` [tip: locking/core] " tip-bot2 for Thomas Gleixner
2019-11-20  9:38   ` tip-bot2 for Thomas Gleixner
2019-11-06 21:55 ` [patch 06/12] futex: Set task::futex state to DEAD right after handling futex exit Thomas Gleixner
2019-11-15 18:19   ` [tip: locking/core] futex: Set task::futex_state " tip-bot2 for Thomas Gleixner
2019-11-20  9:38   ` tip-bot2 for Thomas Gleixner
2019-11-06 21:55 ` [patch 07/12] futex: Mark the begin of futex exit explicitly Thomas Gleixner
2019-11-15 18:19   ` [tip: locking/core] " tip-bot2 for Thomas Gleixner
2019-11-20  9:38   ` tip-bot2 for Thomas Gleixner
2019-11-06 21:55 ` [patch 08/12] futex: Sanitize exit state handling Thomas Gleixner
2019-11-07  9:38   ` Peter Zijlstra
2019-11-15 18:19   ` [tip: locking/core] " tip-bot2 for Thomas Gleixner
2019-11-20  9:38   ` tip-bot2 for Thomas Gleixner
2019-11-06 21:55 ` [patch 09/12] futex: Provide state handling for exec() as well Thomas Gleixner
2019-11-07  9:49   ` Peter Zijlstra
2019-11-07  9:54     ` Thomas Gleixner
2019-11-15 18:19   ` [tip: locking/core] " tip-bot2 for Thomas Gleixner
2019-11-20  9:38   ` tip-bot2 for Thomas Gleixner
2019-11-06 21:55 ` [patch 10/12] futex: Add mutex around futex exit Thomas Gleixner
2019-11-15 18:19   ` [tip: locking/core] " tip-bot2 for Thomas Gleixner
2019-11-20  9:38   ` tip-bot2 for Thomas Gleixner
2019-11-06 21:55 ` [patch 11/12] futex: Provide distinct return value when owner is exiting Thomas Gleixner
2019-11-15 18:19   ` [tip: locking/core] " tip-bot2 for Thomas Gleixner
2019-11-20  9:38   ` tip-bot2 for Thomas Gleixner
2019-11-06 21:55 ` [patch 12/12] futex: Prevent exit livelock Thomas Gleixner
2019-11-15 18:19   ` [tip: locking/core] " tip-bot2 for Thomas Gleixner
2019-11-20  9:38   ` tip-bot2 for Thomas Gleixner
2019-11-07  8:29 ` [patch 00/12] futex: Cure robust/PI futex exit races Ingo Molnar
2019-11-07  8:41 ` Ingo Molnar
2019-11-07  9:40 ` Peter Zijlstra
2019-11-07 15:03 ` Oleg Nesterov
2019-11-07 22:29 ` Florian Weimer
2019-11-07 22:40   ` Florian Weimer
2019-11-08  7:38     ` Florian Weimer
2019-11-08  9:18       ` Thomas Gleixner
2019-11-08 10:17         ` Florian Weimer
2019-11-08 10:37           ` Thomas Gleixner
2019-11-08 11:51             ` Thomas Gleixner
2019-11-11  9:48               ` Florian Weimer [this message]

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=87woc6yc8q.fsf@oldenburg2.str.redhat.com \
    --to=fweimer@redhat.com \
    --cc=carlos@redhat.com \
    --cc=darren@dvhart.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=oleg@redhat.com \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --cc=viro@zeniv.linux.org.uk \
    --cc=wang.yi59@zte.com.cn \
    --cc=yang.tao172@zte.com.cn \
    /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;
as well as URLs for NNTP newsgroup(s).