linux-kernel - Re: [PATCH 2/6] x86: Simplify compat syscall userspace allocation

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: <87zh5vmnvq.fsf@collabora.com>
Date:   Sat, 12 Sep 2020 03:54:01 -0400
From:   Gabriel Krisman Bertazi <krisman@...labora.com>
To:     Christoph Hellwig <hch@....de>
Cc:     luto@...nel.org, tglx@...utronix.de, hpa@...or.com, bp@...en8.de,
        rric@...nel.org, peterz@...radead.org, mingo@...hat.com,
        x86@...nel.org, linux-kernel@...r.kernel.org, kernel@...labora.com
Subject: Re: [PATCH 2/6] x86: Simplify compat syscall userspace allocation

Christoph Hellwig <hch@....de> writes:

> On Sat, Sep 12, 2020 at 03:05:49AM -0400, Gabriel Krisman Bertazi wrote:
>> When allocating user memory space for a compat system call, don't
>> consider whether the originating code is IA32 or X32, just allocate from
>> a safe region for both, beyond the redzone.  This should be safe for
>> IA32, and has the benefit of avoiding TIF_IA32, which we want to drop.
>
> This doesn't look wrong, by why bother (maybe Ccing me on the whole
> seris as you always should instead of sending annoying out of context
> single patches would have told..).

Hi Chris,

Thanks for the quick reply. sorry and I will make sure to cc you for the
rest of the series if this spin again.  The reason is the removal of
TIF_IA32 to reclaim some bits in the ti flags.

If you want to see the rest of it immediately: <https://lkml.org/lkml/2020/9/12/28>

> We will hopefully kill off compat_alloc_user_space in the next few
> merge windows..

I plan to kill TIF_IA32 hopefully in the next merge window, to
facilitate other work I'm doing and I wouldn't like to wait for other
stuff, since this is trivial enough.  Can I get your reviewed-by here?

Thanks,

-- 
Gabriel Krisman Bertazi