linux-kernel - Re: [PATCH 5/8] x86/mce: Avoid tail copy when machine check terminated a copy from user

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] [day] [month] [year] [list]

Date:   Fri, 18 Sep 2020 09:51:25 +0200
From:   Borislav Petkov <bp@...en8.de>
To:     "Luck, Tony" <tony.luck@...el.com>
Cc:     Youquan Song <youquan.song@...el.com>, x86@...nel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 5/8] x86/mce: Avoid tail copy when machine check
 terminated a copy from user

On Thu, Sep 17, 2020 at 02:57:51PM -0700, Luck, Tony wrote:
> On Thu, Sep 17, 2020 at 07:04:06PM +0200, Borislav Petkov wrote:
> > So actually, I'm thinking:
> > 
> > .LMCE_during_user_access:
> > 	mov $-ENODEV, %eax
> > 	ASM_CLAC
> > 	ret
> > 
> > I have no clue which error code we should put there but it should be an
> > error code which tells you not to retry and to back off immediately.
> 
> That does look a lot easier to understand *at this point* in the code.
> 
> But the existing iterator code is not expecting an error code.  Just a
> count of bytes not copied.
> 
> So doing this would mean some surgery on the maze of giant #defines that
> is lib/iov_iter.c

Ok, since the user task is guaranteed to get the SIGBUS, let's just
resort to documenting this properly for now and we can consider more
involved surgery later, if really needed.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette