Merge tag 'for-upstream' of https://gitlab.com/bonzini/qemu into staging

* rust: miscellaneous changes
* target/i386: small code generation improvements
* target/i386: various cleanups and fixes
* cpu: remove env->nr_cores

# -----BEGIN PGP SIGNATURE-----
#
# iQFIBAABCAAyFiEE8TM4V0tmI4mGbHaCv/vSX3jHroMFAmeBoIgUHHBib256aW5p
# QHJlZGhhdC5jb20ACgkQv/vSX3jHroOD2gf+NK7U1EhNIrsbBsbtu2i7+tnbRKIB
# MTu+Mxb2wz4C7//pxq+vva4bgT3iOuL9RF19PRe/63CMD65xMiwyyNrEWX2HbRIJ
# 5dytLLLdef3yMhHh2x1uZfm54g12Ppvn9kulMCbPawrlqWgg1sZbkUBrRtFzS45c
# NeYjGWWSpBDe7LtsrgSRYLPnz6wWEiy3tDpu2VoDtjrE86UVDXwyzpbtBk9Y8jPi
# CKdvLyQeO9xDE5OoXMjJMlJeQq3D9iwYEprXUqy+RUZtpW7YmqMCf2JQ4dAjVCad
# 07v/kITF4brGCVnzDcDA6W7LqHpBu1w+Hn23yLw3HEDDBt11o9JjQCl9qA==
# =xIQ4
# -----END PGP SIGNATURE-----
# gpg: Signature made Fri 10 Jan 2025 17:34:48 EST
# gpg:                using RSA key F13338574B662389866C7682BFFBD25F78C7AE83
# gpg:                issuer "pbonzini@redhat.com"
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" [full]
# gpg:                 aka "Paolo Bonzini <pbonzini@redhat.com>" [full]
# Primary key fingerprint: 46F5 9FBD 57D6 12E7 BFD4  E2F7 7E15 100C CD36 69B1
#      Subkey fingerprint: F133 3857 4B66 2389 866C  7682 BFFB D25F 78C7 AE83

* tag 'for-upstream' of https://gitlab.com/bonzini/qemu: (38 commits)
  i386/cpu: Set and track CPUID_EXT3_CMP_LEG in env->features[FEAT_8000_0001_ECX]
  i386/cpu: Set up CPUID_HT in x86_cpu_expand_features() instead of cpu_x86_cpuid()
  cpu: Remove nr_cores from struct CPUState
  i386/cpu: Hoist check of CPUID_EXT3_TOPOEXT against threads_per_core
  i386/cpu: Track a X86CPUTopoInfo directly in CPUX86State
  i386/topology: Introduce helpers for various topology info of different level
  i386/topology: Update the comment of x86_apicid_from_topo_ids()
  i386/cpu: Drop cores_per_pkg in cpu_x86_cpuid()
  i386/cpu: Drop the variable smp_cores and smp_threads in x86_cpu_pre_plug()
  i386/cpu: Extract a common fucntion to setup value of MSR_CORE_THREAD_COUNT
  target/i386/kvm: Replace ARRAY_SIZE(msr_handlers) with KVM_MSR_FILTER_MAX_RANGES
  target/i386/kvm: Clean up error handling in kvm_arch_init()
  target/i386/kvm: Return -1 when kvm_msr_energy_thread_init() fails
  target/i386/kvm: Clean up return values of MSR filter related functions
  target/i386/confidential-guest: Fix comment of x86_confidential_guest_kvm_type()
  target/i386/kvm: Drop workaround for KVM_X86_DISABLE_EXITS_HTL typo
  target/i386/kvm: Only save/load kvmclock MSRs when kvmclock enabled
  target/i386/kvm: Remove local MSR_KVM_WALL_CLOCK and MSR_KVM_SYSTEM_TIME definitions
  target/i386/kvm: Add feature bit definitions for KVM CPUID
  i386/cpu: Mark avx10_version filtered when prefix is NULL
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

1 files changed, 54 insertions, 65 deletions

diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 0b639848cd..1b9c11022c 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -312,13 +312,11 @@ static uint32_t num_threads_by_topo_level(X86CPUTopoInfo *topo_info,
     case CPU_TOPOLOGY_LEVEL_CORE:
         return topo_info->threads_per_core;
     case CPU_TOPOLOGY_LEVEL_MODULE:
-        return topo_info->threads_per_core * topo_info->cores_per_module;
+        return x86_threads_per_module(topo_info);
     case CPU_TOPOLOGY_LEVEL_DIE:
-        return topo_info->threads_per_core * topo_info->cores_per_module *
-               topo_info->modules_per_die;
+        return x86_threads_per_die(topo_info);
     case CPU_TOPOLOGY_LEVEL_SOCKET:
-        return topo_info->threads_per_core * topo_info->cores_per_module *
-               topo_info->modules_per_die * topo_info->dies_per_pkg;
+        return x86_threads_per_pkg(topo_info);
     default:
         g_assert_not_reached();
     }
@@ -6498,18 +6496,10 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
     CPUState *cs = env_cpu(env);
     uint32_t limit;
     uint32_t signature[3];
-    X86CPUTopoInfo topo_info;
-    uint32_t cores_per_pkg;
+    X86CPUTopoInfo *topo_info = &env->topo_info;
     uint32_t threads_per_pkg;
 
-    topo_info.dies_per_pkg = env->nr_dies;
-    topo_info.modules_per_die = env->nr_modules;
-    topo_info.cores_per_module = cs->nr_cores / env->nr_dies / env->nr_modules;
-    topo_info.threads_per_core = cs->nr_threads;
-
-    cores_per_pkg = topo_info.cores_per_module * topo_info.modules_per_die *
-                    topo_info.dies_per_pkg;
-    threads_per_pkg = cores_per_pkg * topo_info.threads_per_core;
+    threads_per_pkg = x86_threads_per_pkg(topo_info);
 
     /* Calculate & apply limits for different index ranges */
     if (index >= 0xC0000000) {
@@ -6548,7 +6538,6 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
         *edx = env->features[FEAT_1_EDX];
         if (threads_per_pkg > 1) {
             *ebx |= threads_per_pkg << 16;
-            *edx |= CPUID_HT;
         }
         if (!cpu->enable_pmu) {
             *ecx &= ~CPUID_EXT_PDCM;
@@ -6586,12 +6575,12 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
                 int host_vcpus_per_cache = 1 + ((*eax & 0x3FFC000) >> 14);
 
                 *eax &= ~0xFC000000;
-                *eax |= max_core_ids_in_package(&topo_info) << 26;
+                *eax |= max_core_ids_in_package(topo_info) << 26;
                 if (host_vcpus_per_cache > threads_per_pkg) {
                     *eax &= ~0x3FFC000;
 
                     /* Share the cache at package level. */
-                    *eax |= max_thread_ids_for_cache(&topo_info,
+                    *eax |= max_thread_ids_for_cache(topo_info,
                                 CPU_TOPOLOGY_LEVEL_SOCKET) << 14;
                 }
             }
@@ -6603,7 +6592,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
             switch (count) {
             case 0: /* L1 dcache info */
                 encode_cache_cpuid4(env->cache_info_cpuid4.l1d_cache,
-                                    &topo_info,
+                                    topo_info,
                                     eax, ebx, ecx, edx);
                 if (!cpu->l1_cache_per_core) {
                     *eax &= ~MAKE_64BIT_MASK(14, 12);
@@ -6611,7 +6600,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
                 break;
             case 1: /* L1 icache info */
                 encode_cache_cpuid4(env->cache_info_cpuid4.l1i_cache,
-                                    &topo_info,
+                                    topo_info,
                                     eax, ebx, ecx, edx);
                 if (!cpu->l1_cache_per_core) {
                     *eax &= ~MAKE_64BIT_MASK(14, 12);
@@ -6619,13 +6608,13 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
                 break;
             case 2: /* L2 cache info */
                 encode_cache_cpuid4(env->cache_info_cpuid4.l2_cache,
-                                    &topo_info,
+                                    topo_info,
                                     eax, ebx, ecx, edx);
                 break;
             case 3: /* L3 cache info */
                 if (cpu->enable_l3_cache) {
                     encode_cache_cpuid4(env->cache_info_cpuid4.l3_cache,
-                                        &topo_info,
+                                        topo_info,
                                         eax, ebx, ecx, edx);
                     break;
                 }
@@ -6708,12 +6697,12 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
 
         switch (count) {
         case 0:
-            *eax = apicid_core_offset(&topo_info);
-            *ebx = topo_info.threads_per_core;
+            *eax = apicid_core_offset(topo_info);
+            *ebx = topo_info->threads_per_core;
             *ecx |= CPUID_B_ECX_TOPO_LEVEL_SMT << 8;
             break;
         case 1:
-            *eax = apicid_pkg_offset(&topo_info);
+            *eax = apicid_pkg_offset(topo_info);
             *ebx = threads_per_pkg;
             *ecx |= CPUID_B_ECX_TOPO_LEVEL_CORE << 8;
             break;
@@ -6739,7 +6728,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
             break;
         }
 
-        encode_topo_cpuid1f(env, count, &topo_info, eax, ebx, ecx, edx);
+        encode_topo_cpuid1f(env, count, topo_info, eax, ebx, ecx, edx);
         break;
     case 0xD: {
         /* Processor Extended State */
@@ -6964,17 +6953,6 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
         *ecx = env->features[FEAT_8000_0001_ECX];
         *edx = env->features[FEAT_8000_0001_EDX];
 
-        /* The Linux kernel checks for the CMPLegacy bit and
-         * discards multiple thread information if it is set.
-         * So don't set it here for Intel to make Linux guests happy.
-         */
-        if (threads_per_pkg > 1) {
-            if (env->cpuid_vendor1 != CPUID_VENDOR_INTEL_1 ||
-                env->cpuid_vendor2 != CPUID_VENDOR_INTEL_2 ||
-                env->cpuid_vendor3 != CPUID_VENDOR_INTEL_3) {
-                *ecx |= 1 << 1;    /* CmpLegacy bit */
-            }
-        }
         if (tcg_enabled() && env->cpuid_vendor1 == CPUID_VENDOR_INTEL_1 &&
             !(env->hflags & HF_LMA_MASK)) {
             *edx &= ~CPUID_EXT2_SYSCALL;
@@ -7042,7 +7020,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
              * thread ID within a package".
              * Bits 7:0 is "The number of threads in the package is NC+1"
              */
-            *ecx = (apicid_pkg_offset(&topo_info) << 12) |
+            *ecx = (apicid_pkg_offset(topo_info) << 12) |
                    (threads_per_pkg - 1);
         } else {
             *ecx = 0;
@@ -7071,19 +7049,19 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
         switch (count) {
         case 0: /* L1 dcache info */
             encode_cache_cpuid8000001d(env->cache_info_amd.l1d_cache,
-                                       &topo_info, eax, ebx, ecx, edx);
+                                       topo_info, eax, ebx, ecx, edx);
             break;
         case 1: /* L1 icache info */
             encode_cache_cpuid8000001d(env->cache_info_amd.l1i_cache,
-                                       &topo_info, eax, ebx, ecx, edx);
+                                       topo_info, eax, ebx, ecx, edx);
             break;
         case 2: /* L2 cache info */
             encode_cache_cpuid8000001d(env->cache_info_amd.l2_cache,
-                                       &topo_info, eax, ebx, ecx, edx);
+                                       topo_info, eax, ebx, ecx, edx);
             break;
         case 3: /* L3 cache info */
             encode_cache_cpuid8000001d(env->cache_info_amd.l3_cache,
-                                       &topo_info, eax, ebx, ecx, edx);
+                                       topo_info, eax, ebx, ecx, edx);
             break;
         default: /* end of info */
             *eax = *ebx = *ecx = *edx = 0;
@@ -7095,7 +7073,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
         break;
     case 0x8000001E:
         if (cpu->core_id <= 255) {
-            encode_topo_cpuid8000001e(cpu, &topo_info, eax, ebx, ecx, edx);
+            encode_topo_cpuid8000001e(cpu, topo_info, eax, ebx, ecx, edx);
         } else {
             *eax = 0;
             *ebx = 0;
@@ -7539,6 +7517,19 @@ void x86_cpu_expand_features(X86CPU *cpu, Error **errp)
         }
     }
 
+    if (x86_threads_per_pkg(&env->topo_info) > 1) {
+        env->features[FEAT_1_EDX] |= CPUID_HT;
+
+        /*
+         * The Linux kernel checks for the CMPLegacy bit and
+         * discards multiple thread information if it is set.
+         * So don't set it here for Intel to make Linux guests happy.
+         */
+        if (!IS_INTEL_CPU(env)) {
+            env->features[FEAT_8000_0001_ECX] |= CPUID_EXT3_CMP_LEG;
+        }
+    }
+
     for (i = 0; i < ARRAY_SIZE(feature_dependencies); i++) {
         FeatureDep *d = &feature_dependencies[i];
         if (!(env->features[d->from.index] & d->from.mask)) {
@@ -7719,8 +7710,10 @@ static bool x86_cpu_filter_features(X86CPU *cpu, bool verbose)
             env->avx10_version = version;
             have_filtered_features = true;
         }
-    } else if (env->avx10_version && prefix) {
-        warn_report("%s: avx10.%d.", prefix, env->avx10_version);
+    } else if (env->avx10_version) {
+        if (prefix) {
+            warn_report("%s: avx10.%d.", prefix, env->avx10_version);
+        }
         have_filtered_features = true;
     }
 
@@ -7891,6 +7884,21 @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
      */
     cpu->mwait.ecx |= CPUID_MWAIT_EMX | CPUID_MWAIT_IBE;
 
+    /*
+     * Most Intel and certain AMD CPUs support hyperthreading. Even though QEMU
+     * fixes this issue by adjusting CPUID_0000_0001_EBX and CPUID_8000_0008_ECX
+     * based on inputs (sockets,cores,threads), it is still better to give
+     * users a warning.
+     */
+    if (IS_AMD_CPU(env) &&
+        !(env->features[FEAT_8000_0001_ECX] & CPUID_EXT3_TOPOEXT) &&
+        env->topo_info.threads_per_core > 1) {
+            warn_report_once("This family of AMD CPU doesn't support "
+                             "hyperthreading(%d). Please configure -smp "
+                             "options properly or try enabling topoext "
+                             "feature.", env->topo_info.threads_per_core);
+    }
+
     /* For 64bit systems think about the number of physical bits to present.
      * ideally this should be the same as the host; anything other than matching
      * the host can cause incorrect guest behaviour.
@@ -7995,24 +8003,6 @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
     x86_cpu_gdb_init(cs);
     qemu_init_vcpu(cs);
 
-    /*
-     * Most Intel and certain AMD CPUs support hyperthreading. Even though QEMU
-     * fixes this issue by adjusting CPUID_0000_0001_EBX and CPUID_8000_0008_ECX
-     * based on inputs (sockets,cores,threads), it is still better to give
-     * users a warning.
-     *
-     * NOTE: the following code has to follow qemu_init_vcpu(). Otherwise
-     * cs->nr_threads hasn't be populated yet and the checking is incorrect.
-     */
-    if (IS_AMD_CPU(env) &&
-        !(env->features[FEAT_8000_0001_ECX] & CPUID_EXT3_TOPOEXT) &&
-        cs->nr_threads > 1) {
-            warn_report_once("This family of AMD CPU doesn't support "
-                             "hyperthreading(%d). Please configure -smp "
-                             "options properly or try enabling topoext "
-                             "feature.", cs->nr_threads);
-    }
-
 #ifndef CONFIG_USER_ONLY
     x86_cpu_apic_realize(cpu, &local_err);
     if (local_err != NULL) {
@@ -8171,8 +8161,7 @@ static void x86_cpu_init_default_topo(X86CPU *cpu)
 {
     CPUX86State *env = &cpu->env;
 
-    env->nr_modules = 1;
-    env->nr_dies = 1;
+    env->topo_info = (X86CPUTopoInfo) {1, 1, 1, 1};
 
     /* thread, core and socket levels are set by default. */
     set_bit(CPU_TOPOLOGY_LEVEL_THREAD, env->avail_cpu_topo);