[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <cab5e409-be9e-f2c7-424f-6b7d031324dc@amd.com>
Date: Wed, 27 May 2020 12:49:06 -0500
From: Tom Lendacky <thomas.lendacky@....com>
To: Borislav Petkov <bp@...en8.de>,
Sean Christopherson <sean.j.christopherson@...el.com>
Cc: Joerg Roedel <joro@...tes.org>, x86@...nel.org, hpa@...or.com,
Andy Lutomirski <luto@...nel.org>,
Dave Hansen <dave.hansen@...ux.intel.com>,
Peter Zijlstra <peterz@...radead.org>,
Thomas Hellstrom <thellstrom@...are.com>,
Jiri Slaby <jslaby@...e.cz>,
Dan Williams <dan.j.williams@...el.com>,
Juergen Gross <jgross@...e.com>,
Kees Cook <keescook@...omium.org>,
David Rientjes <rientjes@...gle.com>,
Cfir Cohen <cfir@...gle.com>,
Erdem Aktas <erdemaktas@...gle.com>,
Masami Hiramatsu <mhiramat@...nel.org>,
Mike Stunes <mstunes@...are.com>,
Joerg Roedel <jroedel@...e.de>, linux-kernel@...r.kernel.org,
kvm@...r.kernel.org, virtualization@...ts.linux-foundation.org
Subject: Re: [PATCH v3 64/75] x86/sev-es: Cache CPUID results for improved
performance
On 5/26/20 4:19 AM, Borislav Petkov wrote:
> On Tue, May 19, 2020 at 10:16:37PM -0700, Sean Christopherson wrote:
>> The whole cache on-demand approach seems like overkill. The number of CPUID
>> leaves that are invoked after boot with any regularity can probably be counted
>> on one hand. IIRC glibc invokes CPUID to gather TLB/cache info, XCR0-based
>> features, and one or two other leafs. A statically sized global array that's
>> arbitrarily index a la x86_capability would be just as simple and more
>> performant. It would also allow fancier things like emulating CPUID 0xD in
>> the guest if you want to go down that road.
>
> And before we do any of that "caching" or whatnot, I'd like to see
> numbers justifying its existence. Because if it is only a couple of
> CPUID invocations and the boot delay is immeasurable, then it's not
> worth the effort.
I added some rudimentary stats code to see how many times there was a
CPUID cache hit on a 64-vCPU guest during a kernel build (make clean
followed by make with -j 64):
SEV-ES CPUID cache statistics
0x00000000/0x00000000: 220,384
0x00000007/0x00000000: 213,306
0x80000000/0x00000000: 1,054,642
0x80000001/0x00000000: 213,306
0x80000005/0x00000000: 210,334
0x80000006/0x00000000: 420,668
0x80000007/0x00000000: 210,334
0x80000008/0x00000000: 420,684
2,963,658 cache hits
So it is significant in quantity, but I'm not sure what the overall
performance difference is. If I can find some more time I'll try to
compare the kernel builds with and without the caching to see if it is
noticeable.
Thanks,
Tom
>
Powered by blists - more mailing lists