linux-kernel - Re: [PATCH v5 02/13] x86/xsaves: Rename xstate_size to kernel_xstate_size to explicitly distinguish xstate size in kernel from user space

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]

Date:	Tue, 10 May 2016 19:26:57 +0200
From:	Borislav Petkov <bp@...e.de>
To:	Dave Hansen <dave.hansen@...ux.intel.com>
Cc:	Yu-cheng Yu <yu-cheng.yu@...el.com>, linux-kernel@...r.kernel.org,
	x86@...nel.org, "H. Peter Anvin" <hpa@...or.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...hat.com>,
	Andy Lutomirski <luto@...nel.org>,
	Sai Praneeth Prakhya <sai.praneeth.prakhya@...el.com>,
	"Ravi V. Shankar" <ravi.v.shankar@...el.com>,
	Fenghua Yu <fenghua.yu@...el.com>
Subject: Re: [PATCH v5 02/13] x86/xsaves: Rename xstate_size to
 kernel_xstate_size to explicitly distinguish xstate size in kernel from user
 space

On Tue, May 10, 2016 at 10:08:44AM -0700, Dave Hansen wrote:
> But the kernel never actually stores "user_xstate_size" anywhere or
> really ever even cares about it except when copying in/out of userspace.

Sounds like a reason enough to me.

> "user_xstate_size" is also entirely enumerable in userspace with a
> single cpuid instruction.

So is a lot of other stuff we're dumping in dmesg.

> It's nice to dump out interesting data in dmesg, but I'm curious why you
> think it's interesting.

I think it would be interesting to know what the kernel's idea
is of user_xstate_size. I know, I know, one can follow the code
and figure out what it is but one can say the same about a lot of
other "interesting" data dumped in dmesg. And I'd like to know what
fpu__init_system_xstate_size_legacy() decided. And so I know how many
data is shuffled to/from userspace.

And btw, this message needs more "humanization":

[    0.000000] x86/fpu: xstate_offset[2]:  576, xstate_sizes[2]:  256

That doesn't tell me anything.

Oh and it can be read out from CPUID too.

-- 
Regards/Gruss,
    Boris.

SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg)
--