[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230712183136.85561-1-itazur@amazon.com>
Date: Wed, 12 Jul 2023 19:31:36 +0100
From: Takahiro Itazuri <itazur@...zon.com>
To: Sean Christopherson <seanjc@...gle.com>,
Paolo Bonzini <pbonzini@...hat.com>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
Dave Hansen <dave.hansen@...ux.intel.com>, <x86@...nel.org>,
"H . Peter Anvin" <hpa@...or.com>, <kvm@...r.kernel.org>,
<linux-kernel@...r.kernel.org>
CC: Takahiro Itazuri <zulinx86@...il.com>,
Takahiro Itazuri <itazur@...zon.com>
Subject: [PATCH] KVM: pass through CPUID 0x80000005
Pass CPUID 0x80000005 (L1 cache and TLB info).
CPUID 0x80000006 (L2 cache and TLB and L3 cache info) has been returned
since commit 43d05de2bee7 ("KVM: pass through CPUID(0x80000006)").
Enumerating both 0x80000005 and 0x80000006 with KVM_GET_SUPPORTED_CPUID
would be better than reporting either, and 0x80000005 could be helpful
for VMM to pass it to KVM_SET_CPUID{,2} for the same reason with
0x80000006..
Signed-off-by: Takahiro Itazuri <itazur@...zon.com>
---
Discussion was made a bit on
https://lore.kernel.org/all/20230712170258.75355-1-itazur@amazon.com/.
If there is any reason that leaf 0x80000005 should not be enumerated or
dropping leaf 0x80000006 is preferred, please feel free to share your
thoughts.
---
arch/x86/kvm/cpuid.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index 0c9660a07b23..54a5b256c484 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -1152,6 +1152,9 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
cpuid_entry_override(entry, CPUID_8000_0001_EDX);
cpuid_entry_override(entry, CPUID_8000_0001_ECX);
break;
+ case 0x80000005:
+ /* Pass host L1 cache and TLB info. */
+ break;
case 0x80000006:
/* Drop reserved bits, pass host L2 cache and TLB info. */
entry->edx &= ~GENMASK(17, 16);
--
2.38.0
Powered by blists - more mailing lists