linux-kernel - Re: Commit 0be0ee71 ("fs: properly and reliably lock f_pos in fdget

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-=wj_nbDN3piDJBEteUrjyrZCTA+CCk15NfV=5D2xFfGJGw@mail.gmail.com>
Date:   Mon, 25 Nov 2019 14:55:22 -0800
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     "Kenneth R. Crudup" <kenny@...ix.com>
Cc:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: Commit 0be0ee71 ("fs: properly and reliably lock f_pos in
 fdget_pos()") breaking userspace

On Mon, Nov 25, 2019 at 1:30 PM Kenneth R. Crudup <kenny@...ix.com> wrote:
>
> FYI. Don't know if the fault lies with these subsystems or not, but I know
> it's been a goal to keep things working. In both cases, reverting this commit
> fixes both these issues. If you need additional information from me, please
> let me know.

Do you have any way to figure out what file descriptor we have in
those two cases?

It's almost certainly trivial to fix - if I were to know just what
kind of file we're blocking on that just needs to be marked
FMODE_STREAM.

The commit itself is also trivial enough to just revert if we can't
figure it out, but that's why I applied it first thing in the merge
window - to see the problems quickly and hopefully get them fixed. It
obviously showed no problems on my machines, but I have neither vmware
nor KDE upower...

Anyway, here's a TOTALLYT UNTESTED patch that may help pinpoint which
thing it is that causes issues.

It might also be so noisy as to be useless, I didn't think it through
a lot. Mind trying it out if you can't immediately see "ahh, vmware fd
#N is a file of type XYZ"?

                    Linus

View attachment "patch.diff" of type "text/x-patch" (1084 bytes)