[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[PULL 06/23] i386/cpu: Fix i/d-cache topology to core level for Intel CP
From: |
Paolo Bonzini |
Subject: |
[PULL 06/23] i386/cpu: Fix i/d-cache topology to core level for Intel CPU |
Date: |
Thu, 23 May 2024 17:00:19 +0200 |
From: Zhao Liu <zhao1.liu@intel.com>
For i-cache and d-cache, current QEMU hardcodes the maximum IDs for CPUs
sharing cache (CPUID.04H.00H:EAX[bits 25:14] and CPUID.04H.01H:EAX[bits
25:14]) to 0, and this means i-cache and d-cache are shared in the SMT
level.
This is correct if there's single thread per core, but is wrong for the
hyper threading case (one core contains multiple threads) since the
i-cache and d-cache are shared in the core level other than SMT level.
For AMD CPU, commit 8f4202fb1080 ("i386: Populate AMD Processor Cache
Information for cpuid 0x8000001D") has already introduced i/d cache
topology as core level by default.
Therefore, in order to be compatible with both multi-threaded and
single-threaded situations, we should set i-cache and d-cache be shared
at the core level by default.
This fix changes the default i/d cache topology from per-thread to
per-core. Potentially, this change in L1 cache topology may affect the
performance of the VM if the user does not specifically specify the
topology or bind the vCPU. However, the way to achieve optimal
performance should be to create a reasonable topology and set the
appropriate vCPU affinity without relying on QEMU's default topology
structure.
Fixes: 7e3482f82480 ("i386: Helpers to encode cache information consistently")
Suggested-by: Robert Hoo <robert.hu@linux.intel.com>
Signed-off-by: Zhao Liu <zhao1.liu@intel.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Tested-by: Yongwei Ma <yongwei.ma@intel.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Message-ID: <20240424154929.1487382-6-zhao1.liu@intel.com>
[Add compat property. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
hw/i386/pc.c | 1 +
target/i386/cpu.c | 6 ++++--
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index 4a2d6f5a97f..6126bfdd2a7 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -79,6 +79,7 @@
{ "athlon-" TYPE_X86_CPU, "model-id", "QEMU Virtual CPU version " v, },
GlobalProperty pc_compat_9_0[] = {
+ { TYPE_X86_CPU, "x-l1-cache-per-thread", "false" },
{ TYPE_X86_CPU, "guest-phys-bits", "0" },
{ "sev-guest", "legacy-vm-type", "true" },
{ TYPE_X86_CPU, "legacy-multi-node", "on" },
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index de1ad7270cf..3c66242f6d3 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -6258,15 +6258,16 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index,
uint32_t count,
*eax = *ebx = *ecx = *edx = 0;
} else {
*eax = 0;
+ int apic_ids_sharing_l1 = cpu->l1_cache_per_core ? cs->nr_threads
: 1;
switch (count) {
case 0: /* L1 dcache info */
encode_cache_cpuid4(env->cache_info_cpuid4.l1d_cache,
- 1, cs->nr_cores,
+ apic_ids_sharing_l1, cs->nr_cores,
eax, ebx, ecx, edx);
break;
case 1: /* L1 icache info */
encode_cache_cpuid4(env->cache_info_cpuid4.l1i_cache,
- 1, cs->nr_cores,
+ apic_ids_sharing_l1, cs->nr_cores,
eax, ebx, ecx, edx);
break;
case 2: /* L2 cache info */
@@ -8105,6 +8106,7 @@ static Property x86_cpu_properties[] = {
false),
DEFINE_PROP_BOOL("x-intel-pt-auto-level", X86CPU, intel_pt_auto_level,
true),
+ DEFINE_PROP_BOOL("x-l1-cache-per-thread", X86CPU, l1_cache_per_core, true),
DEFINE_PROP_END_OF_LIST()
};
--
2.45.1
- [PULL 00/23] i386 changes for 2024-05-23, Paolo Bonzini, 2024/05/23
- [PULL 01/23] target/i386: generate simpler code for ROL/ROR with immediate count, Paolo Bonzini, 2024/05/23
- [PULL 02/23] target/i386: clean up AAM/AAD, Paolo Bonzini, 2024/05/23
- [PULL 03/23] hw/i386/pc_sysfw: Alias rather than copy isa-bios region, Paolo Bonzini, 2024/05/23
- [PULL 04/23] target/i386: add support for LAM in CPUID enumeration, Paolo Bonzini, 2024/05/23
- [PULL 05/23] target/i386: add control bits support for LAM, Paolo Bonzini, 2024/05/23
- [PULL 06/23] i386/cpu: Fix i/d-cache topology to core level for Intel CPU,
Paolo Bonzini <=
- [PULL 07/23] i386/cpu: Use APIC ID info to encode cache topo in CPUID[4], Paolo Bonzini, 2024/05/23
- [PULL 08/23] i386/cpu: Use APIC ID info get NumSharingCache for CPUID[0x8000001D].EAX[bits 25:14], Paolo Bonzini, 2024/05/23
- [PULL 09/23] i386/cpu: Consolidate the use of topo_info in cpu_x86_cpuid(), Paolo Bonzini, 2024/05/23
- [PULL 10/23] i386/cpu: Introduce bitmap to cache available CPU topology levels, Paolo Bonzini, 2024/05/23
- [PULL 11/23] i386: Split topology types of CPUID[0x1F] from the definitions of CPUID[0xB], Paolo Bonzini, 2024/05/23
- [PULL 17/23] i386/cpu: Introduce module-id to X86CPU, Paolo Bonzini, 2024/05/23
- [PULL 14/23] i386: Support modules_per_die in X86CPUTopoInfo, Paolo Bonzini, 2024/05/23
- [PULL 16/23] i386: Support module_id in X86CPUTopoIDs, Paolo Bonzini, 2024/05/23
- [PULL 12/23] i386/cpu: Decouple CPUID[0x1F] subleaf with specific topology level, Paolo Bonzini, 2024/05/23
- [PULL 13/23] i386: Introduce module level cpu topology to CPUX86State, Paolo Bonzini, 2024/05/23