[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20220106032118.34459-1-likexu@tencent.com>
Date: Thu, 6 Jan 2022 11:21:18 +0800
From: Like Xu <like.xu.linux@...il.com>
To: Paolo Bonzini <pbonzini@...hat.com>
Cc: Sean Christopherson <seanjc@...gle.com>,
Vitaly Kuznetsov <vkuznets@...hat.com>,
Wanpeng Li <wanpengli@...cent.com>,
Jim Mattson <jmattson@...gle.com>,
Joerg Roedel <joro@...tes.org>, kvm@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: [PATCH v2] KVM: x86/pmu: Make top-down.slots event unavailable in supported leaf
From: Like Xu <likexu@...cent.com>
When we choose to disable the fourth fixed counter TOPDOWN.SLOTS,
we also need to comply with the specification and set 0AH.EBX.[bit 7]
to 1 if the guest (e.g. on the ICX) has a value of 0AH.EAX[31:24] > 7.
Fixes: 2e8cd7a3b8287 ("kvm: x86: limit the maximum number of vPMU fixed counters to 3")
Signed-off-by: Like Xu <likexu@...cent.com>
---
v1 -> v2 Changelog:
- Make it simpler to keep forward compatibility; (Sean)
- Wrap related comment at ~80 chars; (Sean)
Previous:
https://lore.kernel.org/kvm/20220105050711.67280-1-likexu@tencent.com/
arch/x86/kvm/cpuid.c | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index 0b920e12bb6d..4fe17a537084 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -782,6 +782,18 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
eax.split.mask_length = cap.events_mask_len;
edx.split.num_counters_fixed = min(cap.num_counters_fixed, MAX_FIXED_COUNTERS);
+
+ /*
+ * The 8th Intel architectural event (Topdown Slots) will be supported
+ * if the 4th fixed counter exists && EAX[31:24] > 7 && EBX[7] = 0.
+ *
+ * Currently, KVM needs to set EAX[31:24] < 8 or EBX[7] == 1
+ * to make this event unavailable in a consistent way.
+ */
+ if (edx.split.num_counters_fixed < 4 &&
+ eax.split.mask_length > 7)
+ cap.events_mask |= BIT_ULL(7);
+
edx.split.bit_width_fixed = cap.bit_width_fixed;
if (cap.version)
edx.split.anythread_deprecated = 1;
--
2.33.1
Powered by blists - more mailing lists