Messages in this thread | | | Subject | Re: [PATCH v7 05/10] x86/tdx: Handle port I/O | From | Sathyanarayanan Kuppuswamy <> | Date | Mon, 18 Oct 2021 11:42:07 -0700 |
| |
On 10/18/21 6:52 AM, Tom Lendacky wrote: > On 10/17/21 3:35 PM, Sathyanarayanan Kuppuswamy wrote: >> >> On 10/17/21 12:58 PM, Thomas Gleixner wrote: >>> On Tue, Oct 05 2021 at 13:41, Kuppuswamy Sathyanarayanan wrote: >>>> \ >>>> static inline void outs##bwl(int port, const void *addr, unsigned >>>> long count) \ >>>> { \ >>>> - if (sev_key_active()) { \ >>>> + if (sev_key_active() || \ >>>> + cc_platform_has(CC_ATTR_GUEST_UNROLL_STRING_IO)) { \ >>> Instead of adding an extra check, can you please replace that >>> sev_key_active() with cc_platform_has() completely? >> >> Yes. sev_key_active() can be removed and replaced with >> cc_platform_has(). >> >> Thomas Lendacky also proposed to introduce as common >> static key which can be set by both AMD SEV and TDX code. >> >> @Thomas Lendacky, any comments? > > Either way works for me. > > For the cc_platform_has() path, you will need to update the > amd_cc_platform_has() to return true for > CC_ATTR_GUEST_UNROLL_STRING_IO for SEV only, i.e.: > > case CC_ATTR_GUEST_UNROLL_STRING_IO: > return (sev_status & MSR_AMD64_SEV_ENABLED) && > !(sev_status & MSR_AMD64_SEV_ES_ENABLED);
I will submit a separate cleanup patch to make SEV code use CC_ATTR_GUEST_UNROLL_STRING_IO first. That way, in this patch I can just set CC_ATTR_GUEST_UNROLL_STRING_IO.
> > Thanks, > Tom
-- Sathyanarayanan Kuppuswamy Linux Kernel Developer
| |