mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andy Lutomirski <luto@kernel.org>
To: "H.J. Lu" <hjl.tools@gmail.com>
Cc: Andy Lutomirski <luto@kernel.org>,
	Dave Hansen <dave.hansen@intel.com>,
	"Robert O'Callahan" <robert@ocallahan.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	X86 ML <x86@kernel.org>
Subject: Re: xgetbv nondeterminism
Date: Thu, 15 Jun 2017 15:45:32 -0700	[thread overview]
Message-ID: <CALCETrWCSO3R6USiATLU-X0f9mMFGop6rrayTfeBxrx2GPC4ow@mail.gmail.com> (raw)
In-Reply-To: <CAMe9rOp3A4WJjfsT0VyzQEDVua4840XD8Ev66=X5qTae5C+hoQ@mail.gmail.com>

On Thu, Jun 15, 2017 at 3:40 PM, H.J. Lu <hjl.tools@gmail.com> wrote:
> On Thu, Jun 15, 2017 at 3:18 PM, Andy Lutomirski <luto@kernel.org> wrote:
>> On Thu, Jun 15, 2017 at 7:33 AM, Dave Hansen <dave.hansen@intel.com> wrote:
>>> On 06/14/2017 10:18 PM, Andy Lutomirski wrote:
>>>> Dave, why is XINUSE exposed at all to userspace?
>>>
>>> You need it for XSAVEOPT when it is using the init optimization to be
>>> able to tell which state was written and which state in the XSAVE buffer
>>> is potentially stale with respect to what's in the registers.  I guess
>>> you can just use XSAVE instead of XSAVEOPT, though.
>>>
>>> As you pointed out, if you are using XSAVEC's compaction features by
>>> leaving bits unset in the requested feature bitmap registers, you have
>>> no idea how much data XSAVEC will write, unless you read XINUSE with
>>> XGETBV.  But, you can get around *that* by just presizing the XSAVE
>>> buffer to be big.
>>
>> I imagine that, if you're going to save, do something quick, and
>> restore, you'd be better off allocating a big buffer rather than
>> trying to find the smallest buffer you can get away with by reading
>> XINUSE.  Also, what happens if XINUSE nondeterministically changes out
>> from under you before you do XSAVEC?  I assume you can avoid this
>> becoming a problem by using RFBM carefully.
>>
>>>
>>> So, I guess that leaves its use to just figuring out how much XSAVEOPT
>>> (and friends) are going to write.
>>>
>>>> To be fair, glibc uses this new XGETBV feature, but I suspect its
>>>> usage is rather dubious.  Shouldn't it just do XSAVEC directly rather
>>>> than rolling its own code?
>>>
>>> A quick grep through my glibc source only shows XGETBV(0) used which
>>> reads XCR0.  I don't see any XGETBV(1) which reads XINUSE.  Did I miss it.
>>
>> Take a look at sysdeps/x86_64/dl-trampoline.h in a new enough version.
>
> I wrote a test to compare latency against different approaches.   This
> is on Skylake:
>
> [hjl@gnu-skl-1 glibc-test]$ make
> ./test
> move    : 47212
> fxsave  : 719440
> xsave   : 925146
> xsavec  : 811036
> xsave_state_size: 1088
> xsave_state_comp_size: 896
>
> load/store is about 17X faster than xsavec.
>
> I put my hjl/pr21265/xsavec branch at
>
> https://sourceware.org/git/?p=glibc.git;a=summary
>
> It uses xsave/xsave/xsavec in _dl_runtime_resolve.

What is this used for?  Is it just to avoid clobbering argument regs
when resolving a symbol that uses an ifunc, or is there more to it?

  reply	other threads:[~2017-06-15 22:45 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-06-15  5:18 Andy Lutomirski
2017-06-15 14:33 ` Dave Hansen
2017-06-15 22:18   ` Andy Lutomirski
2017-06-15 22:40     ` H.J. Lu
2017-06-15 22:45       ` Andy Lutomirski [this message]
2017-06-15 23:11         ` H.J. Lu
2017-06-15 23:28           ` Andy Lutomirski
2017-06-16  2:17             ` H.J. Lu
2017-06-16  3:05               ` Andy Lutomirski
2017-06-16  4:34                 ` H.J. Lu
2017-06-16 16:01                   ` Andy Lutomirski
2017-06-16 16:17                     ` H.J. Lu
2017-06-16 16:38                       ` Andy Lutomirski
2017-06-16 17:44                         ` H.J. Lu
2017-06-16 17:56                           ` Andy Lutomirski
2017-06-16 18:03                             ` H.J. Lu
2017-06-17  6:21                               ` Andy Lutomirski
2017-06-17 12:51                                 ` H.J. Lu
2017-06-17 16:32                                   ` Andy Lutomirski
2017-06-17 19:48                                     ` H.J. Lu
2017-06-15 23:37     ` Dave Hansen
2017-06-16  2:28       ` H.J. Lu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CALCETrWCSO3R6USiATLU-X0f9mMFGop6rrayTfeBxrx2GPC4ow@mail.gmail.com \
    --to=luto@kernel.org \
    --cc=dave.hansen@intel.com \
    --cc=hjl.tools@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=robert@ocallahan.org \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®