From: "Chang S. Bae" <chang.seok.bae@intel.com>
To: <sashiko-reviews@lists.linux.dev>
Cc: <kvm@vger.kernel.org>
Subject: Re: [PATCH v5 19/20] KVM: x86: selftests: Add APX state and ABI test
Date: Wed, 22 Jul 2026 14:55:37 -0700 [thread overview]
Message-ID: <fd01143e-6c2c-415b-a5d8-8c0e16a61baa@intel.com> (raw)
In-Reply-To: <20260720181756.2190A1F000E9@smtp.kernel.org>
On 7/20/2026 11:17 AM, sashiko-bot@kernel.org wrote:
>>
>> +static inline unsigned long read_egpr(int reg)
>> +{
>> + unsigned long data = 0;
>> +
>> + /* mov %r16..%r31, %rax */
>> + switch (reg) {
>> + case 16:
>> + asm(".byte 0xd5, 0x48, 0x89, 0xc0" : "=a"(data));
>
> [Severity: Medium]
> Since this inline assembly has an output operand but lacks the volatile
> qualifier, can the compiler optimize away or cache the EGPR reads?
>
> When compiled with optimizations, GCC might treat this as a pure function and
> reuse the initially cached values across GUEST_SYNC barriers. If userspace
> updates the EGPRs in test_userspace_update(), will the guest compare against
> stale cached values instead of emitting instructions to re-read the hardware
> registers, causing the test to spuriously fail?
Without -mapxf, the compiler shouldn't include EGPRs in the optimization
options I think.
Thanks,
Chang
next prev parent reply other threads:[~2026-07-22 21:55 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-20 17:19 [PATCH v5 00/20] KVM: x86: Enable APX for guests Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 01/20] KVM: x86: Extend VCPU registers for EGPRs Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 02/20] KVM: VMX: Save guest EGPRs in VCPU cache Chang S. Bae
2026-07-20 18:20 ` sashiko-bot
2026-07-22 21:54 ` Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 03/20] KVM: x86: Support APX state for XSAVE ABI Chang S. Bae
2026-07-20 18:22 ` sashiko-bot
2026-07-22 21:54 ` Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 04/20] KVM: VMX: Refactor VMX instruction information access Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 05/20] KVM: VMX: Refactor instruction information decoding Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 06/20] KVM: VMX: Remove unused control-register access defines Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 07/20] KVM: VMX: Refactor register index retrieval from exit qualification Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 08/20] KVM: VMX: Support instruction information extension Chang S. Bae
2026-07-20 18:12 ` sashiko-bot
2026-07-22 21:54 ` Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 09/20] KVM: nVMX: Propagate extended instruction information Chang S. Bae
2026-07-20 18:10 ` sashiko-bot
2026-07-22 21:55 ` Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 10/20] KVM: x86: Support EGPR accessing and tracking for emulator Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 11/20] KVM: x86: Handle EGPR index and REX2-incompatible opcodes Chang S. Bae
2026-07-20 18:18 ` sashiko-bot
2026-07-22 21:55 ` Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 12/20] KVM: x86: Support REX2-prefixed opcode decode Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 13/20] KVM: x86: Reject EVEX-prefixed instructions Chang S. Bae
2026-07-20 18:11 ` sashiko-bot
2026-07-22 21:55 ` Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 14/20] KVM: x86: Move KVM_SUPPORTED_{XCR0,XSS} into kvm_x86_vendor_init() Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 15/20] KVM: x86: Guard valid XCR0.APX settings Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 16/20] KVM: x86: Add APX in supported XCR0 Chang S. Bae
2026-07-20 18:25 ` sashiko-bot
2026-07-22 21:55 ` Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 17/20] KVM: x86: Expose APX foundation feature to userspace Chang S. Bae
2026-07-20 18:12 ` sashiko-bot
2026-07-22 21:55 ` Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 18/20] KVM: x86: Expose APX sub-features " Chang S. Bae
2026-07-20 18:10 ` sashiko-bot
2026-07-22 21:55 ` Chang S. Bae
2026-07-20 17:19 ` [PATCH v5 19/20] KVM: x86: selftests: Add APX state and ABI test Chang S. Bae
2026-07-20 18:17 ` sashiko-bot
2026-07-22 21:55 ` Chang S. Bae [this message]
2026-07-20 17:19 ` [PATCH v5 20/20] KVM: x86: selftests: Add APX state handling and XCR0 sanity checks Chang S. Bae
2026-07-21 7:09 ` [PATCH v5 00/20] KVM: x86: Enable APX for guests Paolo Bonzini
2026-07-22 22:01 ` Chang S. Bae
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=fd01143e-6c2c-415b-a5d8-8c0e16a61baa@intel.com \
--to=chang.seok.bae@intel.com \
--cc=kvm@vger.kernel.org \
--cc=sashiko-reviews@lists.linux.dev \
/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