linux-kernel - Re: [GIT PULL] dlm updates for 5.3

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAHk-=wj00gDz=tX-b5C-xwdogZSaKtRJEDh3SGB69D8W+Wsr2Q@mail.gmail.com>
Date:   Wed, 10 Jul 2019 21:05:21 -0700
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     David Teigland <teigland@...hat.com>
Cc:     Linux List Kernel Mailing <linux-kernel@...r.kernel.org>
Subject: Re: [GIT PULL] dlm updates for 5.3

On Tue, Jul 9, 2019 at 9:57 AM David Teigland <teigland@...hat.com> wrote:
>
> Apart from a couple trivial fixes, the more notable fix makes the dlm
> continuing waiting for a user space result if a signal interrupts the
> wait event.

What? No.

That's not sensible at all.

If wait_event_interruptible() returns -ERESTARTSYS, it means that we
have a signal pending.

And if we have a signal pending, then you can't go back and call
wait_event_interruptible() in a loop, because the signal will
*continue* to be pending, so now your "wait event" becomes a kernel
busy loop.

If you don't want to react to signals, then you shouldn't use the
"interruptible()" version of wait-event.

I'm not pulling this. Because the code looks completely and utterly wrong to me.

Am I missing something? Feel free to educate me and re-submit.

               Linus