Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1727110

Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation

From Byungchul Park <byungchul.park@lge.com>
Newsgroups linux.kernel
Subject Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation
Date 2017-09-06 03:40 +0200
Message-ID <umxtT-2vX-3@gated-at.bofh.it> (permalink)
References (5 earlier) <umjqV-1ux-3@gated-at.bofh.it> <umjKi-1Bt-13@gated-at.bofh.it> <ummoN-3iq-5@gated-at.bofh.it> <umvV7-1o3-3@gated-at.bofh.it> <umwHv-1VB-7@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw

On Wed, Sep 06, 2017 at 08:42:11AM +0800, Boqun Feng wrote:
> > OK. If the workqueue is only user of the weird lockdep annotations, then
> > it might be better to defer introducing the extra state until needed.
> > 
> > But, the 'might' thing I introduced would be necessary if more users
> > want to report deadlocks at the time for crosslocks with speculative
> > acquisitions like the workqueue does, since the recursive-read thing
> > would generate false dependencies much more than we want, while the
> 
> What do you mean by "false dependencies"? AFAICT, recursive-read could

All locks used in every work->func() generate false dependencies with
'work' and 'wq', while any flush works are not involved. It's inevitable.

Moreover, it's also possible to generate more false ones between the
pseudo acquisitions, if real acquisitions are used for that speculative
purpose e.i. recursive-read here, which are anyway real ones.

Moreover, it's also possible to generate more false ones between holding
locks and the pseudo ones, of course, the workqueue code is not the case
for now.

Moreover, it's also possible to generate more false ones between the
pseudo ones and crosslocks on commit, once making crossrelease work even
for recursive-read things.

Whatever...

Only thing I want to say is, don't make code just work, but make code
use right ones semantically for its specific application. Otherwise,
we cannot avoid side effects we don't expect. Of course, these side
effects might be not visible at the moment, IOW, generating false
dependencies might be not problems at the moment, but I just want to
avoid not doing something in the right way, if possible. That's all.

> have dependencies to the following cross commit, for example:
> 
> 	A(a)
> 			ARR(a)
> 			RRR(a)
> 	WFC(X)
> 			C(X)
> 
> This is a deadlock, no?

This is obviously a deadlock.

> In my upcoming v2 for recursive-read support, I'm going to make this
> detectable. But please note as crossrelease doesn't have any selftests

I hope you make the recursive-read things work successfully.

> as normal lockdep stuffs, I may miss something subtle.

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-08-29 11:10 +0200
  [tip:locking/core] locking/lockdep: Untangle xhlock history  save/restore from task independence tip-bot for Peter Zijlstra <tipbot@zytor.com> - 2017-08-29 16:30 +0200
  Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <max.byungchul.park@gmail.com> - 2017-08-29 18:10 +0200
    Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-08-29 20:50 +0200
  Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-08-30 04:20 +0200
    Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-08-30 09:50 +0200
      Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-08-30 11:00 +0200
        RE: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation "Byungchul Park" <byungchul.park@lge.com> - 2017-08-30 11:10 +0200
          Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-08-30 11:20 +0200
            RE: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation "Byungchul Park" <byungchul.park@lge.com> - 2017-08-30 11:40 +0200
          Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-08-30 11:20 +0200
            RE: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation "Byungchul Park" <byungchul.park@lge.com> - 2017-08-30 11:30 +0200
              Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-08-30 13:30 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <max.byungchul.park@gmail.com> - 2017-08-30 15:00 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-08-31 09:30 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-08-31 10:10 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-08-31 10:20 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-08-31 10:40 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-01 04:10 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-09-01 11:50 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-01 12:20 +0200
                RE: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation 박병철/선임연구원/SW Platform(연)AOT팀(byungchul.park@lge.com)          <byungchul.park@lge.com> - 2017-09-01 14:10 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-09-01 14:40 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <max.byungchul.park@gmail.com> - 2017-09-01 16:00 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-09-01 18:40 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-04 03:40 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-04 04:10 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-09-04 13:50 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-05 02:40 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-09-05 09:10 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-09-05 09:20 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-05 11:00 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-09-05 11:40 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-05 12:40 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-05 13:00 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-09-05 15:50 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-06 02:00 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Boqun Feng <boqun.feng@gmail.com> - 2017-09-06 02:50 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-06 03:40 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-07 02:00 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-07 02:20 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-06 02:50 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-09-05 13:00 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-05 13:30 +0200
                Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Byungchul Park <byungchul.park@lge.com> - 2017-09-05 10:40 +0200
              Re: [PATCH 4/4] lockdep: Fix workqueue crossrelease annotation Peter Zijlstra <peterz@infradead.org> - 2017-08-31 10:10 +0200

csiph-web