From: Tom Lendacky <firstname.lastname@example.org>
To: Thomas Gleixner <email@example.com>,
Dave Hansen <firstname.lastname@example.org>,
Cc: email@example.com, Andrew Cooper <firstname.lastname@example.org>,
"Edgecombe, Rick P" <email@example.com>
Subject: Re: [patch 3/3] x86/fpu/xsave: Optimize XSAVEC/S when XGETBV1 is supported
Date: Wed, 20 Apr 2022 13:15:14 -0500 [thread overview]
Message-ID: <firstname.lastname@example.org> (raw)
On 4/19/22 16:22, Thomas Gleixner wrote:
> On Tue, Apr 19 2022 at 15:43, Thomas Gleixner wrote:
>> On Thu, Apr 14 2022 at 10:24, Dave Hansen wrote:
>>> On 4/4/22 05:11, Thomas Gleixner wrote:
>>>> which is suboptimal. Prefetch works better when the access is linear. But
>>>> what's worse is that PKRU can be located in a different page which
>>>> obviously affects dTLB.
>>> The numbers don't lie, but I'm still surprised by this. Was this in a
>>> VM that isn't backed with large pages? task_struct.thread.fpu is
>>> kmem_cache_alloc()'d and is in the direct map, which should be 2M/1G
>>> pages almost all the time.
>> Hmm. Indeed, that's weird.
>> That was bare metal and I just checked that this was a production config
>> and not some weird debug muck which breaks large pages. I'll look deeper
>> into that.
> I can't find any reasonable explanation. The pages are definitely large
> pages, so yes the dTLB miss count does not make sense, but it's
> consistently faster and it's always the dTLB miss count which makes the
> big difference according to perf.
> For enhanced fun, I ran the lot on a AMD Zen3 machine and with the same
> test case (hackbench -l 10000) repeated 10 times by perf stat this is
> consistently slower than the non optimized variant. There is at least an
> explanation for that. A tight loop of 1 Mio xgetbv(1) invocations takes
> 9 Mio cycles on a SKL-X and 50 Mio cycles on a AMD Zen3.
I'll take a look into this and see what I find. Might be interesting to
see if the actual XSAVES is slower or quicker, too, based on the input mask.
If the performance slowdown shows up in real world benchmarks, we might
want to consider not using the xgetbv() call on AMD.
> XSAVE is wonderful, isn't it?
next prev parent reply other threads:[~2022-04-20 18:15 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-04-04 12:11 [patch 0/3] x86/fpu/xsave: Add XSAVEC support and XGETBV1 utilization Thomas Gleixner
2022-04-04 12:11 ` [patch 1/3] x86/fpu/xsave: Support XSAVEC in the kernel Thomas Gleixner
2022-04-04 16:10 ` Andrew Cooper
2022-04-14 14:43 ` Dave Hansen
2022-04-25 13:11 ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2022-04-04 12:11 ` [patch 2/3] x86/fpu/xsave: Prepare for optimized compaction Thomas Gleixner
2022-04-14 15:46 ` Dave Hansen
2022-04-19 12:39 ` Thomas Gleixner
2022-04-19 13:33 ` Thomas Gleixner
2022-04-04 12:11 ` [patch 3/3] x86/fpu/xsave: Optimize XSAVEC/S when XGETBV1 is supported Thomas Gleixner
2022-04-14 17:24 ` Dave Hansen
2022-04-19 13:43 ` Thomas Gleixner
2022-04-19 21:22 ` Thomas Gleixner
2022-04-20 18:15 ` Tom Lendacky [this message]
2022-04-22 19:30 ` Thomas Gleixner
2022-04-23 15:20 ` Dave Hansen
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:
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
* 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;
as well as URLs for NNTP newsgroup(s).