[Qemu-devel] [PATCH 0/5] [PULL] qemu-kvm.git uq/master queue

[Qemu-devel] [PATCH 0/5] [PULL] qemu-kvm.git uq/master queue

[Marcelo Tosatti]

The following changes since commit 03ecd2c80a64d030a22fe67cc7a60f24e17ff211:

  virtio-serial-bus: Ports are expected to implement 'have_data' callback (2011-12-21 15:00:29 -0600)

are available in the git repository at:
  git://git.kernel.org/pub/scm/virt/kvm/qemu-kvm.git uq/master

Gleb Natapov (1):
      enable architectural PMU cpuid leaf for kvm

Jan Kiszka (3):
      kvm: x86: Use symbols for all xsave field
      kvm: x86: Avoid runtime allocation of xsave buffer
      kvm: x86: Drop redundant apic base and tpr update from kvm_get_sregs

Vasilis Liaskovitis (1):
      Set numa topology for max_cpus

 hw/pc.c             |    8 ++++----
 target-i386/cpu.h   |    3 ++-
 target-i386/cpuid.c |   17 +++++++++++++----
 target-i386/kvm.c   |   34 +++++++++++++++++-----------------
 vl.c                |    2 +-
 5 files changed, 37 insertions(+), 27 deletions(-)

[Qemu-devel] [PATCH 1/5] kvm: x86: Use symbols for all xsave field

[Marcelo Tosatti]

From: Jan Kiszka <jan.kiszka@siemens.com>

Field 0 (FCW+FSW) and 1 (FTW+FOP) were hard-coded so far.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
---
 target-i386/kvm.c |   14 ++++++++------
 1 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/target-i386/kvm.c b/target-i386/kvm.c
index 5bfc21f..d2f70f9 100644
--- a/target-i386/kvm.c
+++ b/target-i386/kvm.c
@@ -759,6 +759,8 @@ static int kvm_put_fpu(CPUState *env)
     return kvm_vcpu_ioctl(env, KVM_SET_FPU, &fpu);
 }
 
+#define XSAVE_FCW_FSW     0
+#define XSAVE_FTW_FOP     1
 #define XSAVE_CWD_RIP     2
 #define XSAVE_CWD_RDP     4
 #define XSAVE_MXCSR       6
@@ -786,8 +788,8 @@ static int kvm_put_xsave(CPUState *env)
     for (i = 0; i < 8; ++i) {
         twd |= (!env->fptags[i]) << i;
     }
-    xsave->region[0] = (uint32_t)(swd << 16) + cwd;
-    xsave->region[1] = (uint32_t)(env->fpop << 16) + twd;
+    xsave->region[XSAVE_FCW_FSW] = (uint32_t)(swd << 16) + cwd;
+    xsave->region[XSAVE_FTW_FOP] = (uint32_t)(env->fpop << 16) + twd;
     memcpy(&xsave->region[XSAVE_CWD_RIP], &env->fpip, sizeof(env->fpip));
     memcpy(&xsave->region[XSAVE_CWD_RDP], &env->fpdp, sizeof(env->fpdp));
     memcpy(&xsave->region[XSAVE_ST_SPACE], env->fpregs,
@@ -991,10 +993,10 @@ static int kvm_get_xsave(CPUState *env)
         return ret;
     }
 
-    cwd = (uint16_t)xsave->region[0];
-    swd = (uint16_t)(xsave->region[0] >> 16);
-    twd = (uint16_t)xsave->region[1];
-    env->fpop = (uint16_t)(xsave->region[1] >> 16);
+    cwd = (uint16_t)xsave->region[XSAVE_FCW_FSW];
+    swd = (uint16_t)(xsave->region[XSAVE_FCW_FSW] >> 16);
+    twd = (uint16_t)xsave->region[XSAVE_FTW_FOP];
+    env->fpop = (uint16_t)(xsave->region[XSAVE_FTW_FOP] >> 16);
     env->fpstt = (swd >> 11) & 7;
     env->fpus = swd;
     env->fpuc = cwd;
-- 
1.7.6.4

[Qemu-devel] [PATCH 2/5] kvm: x86: Avoid runtime allocation of xsave buffer

[Marcelo Tosatti]

From: Jan Kiszka <jan.kiszka@siemens.com>

Keep a per-VCPU xsave buffer for kvm_put/get_xsave instead of
continuously allocating and freeing it on state sync.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
---
 target-i386/cpu.h |    3 ++-
 target-i386/kvm.c |   15 +++++++--------
 2 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/target-i386/cpu.h b/target-i386/cpu.h
index a08ce9d..37dde79 100644
--- a/target-i386/cpu.h
+++ b/target-i386/cpu.h
@@ -751,7 +751,8 @@ typedef struct CPUX86State {
     uint32_t cpuid_svm_features;
     bool tsc_valid;
     int tsc_khz;
-    
+    void *kvm_xsave_buf;
+
     /* in order to simplify APIC support, we leave this pointer to the
        user */
     struct DeviceState *apic_state;
diff --git a/target-i386/kvm.c b/target-i386/kvm.c
index d2f70f9..06f4401 100644
--- a/target-i386/kvm.c
+++ b/target-i386/kvm.c
@@ -516,6 +516,10 @@ int kvm_arch_init_vcpu(CPUState *env)
         }
     }
 
+    if (kvm_has_xsave()) {
+        env->kvm_xsave_buf = qemu_memalign(4096, sizeof(struct kvm_xsave));
+    }
+
     return 0;
 }
 
@@ -771,15 +775,14 @@ static int kvm_put_fpu(CPUState *env)
 
 static int kvm_put_xsave(CPUState *env)
 {
-    int i, r;
-    struct kvm_xsave* xsave;
+    struct kvm_xsave* xsave = env->kvm_xsave_buf;
     uint16_t cwd, swd, twd;
+    int i, r;
 
     if (!kvm_has_xsave()) {
         return kvm_put_fpu(env);
     }
 
-    xsave = qemu_memalign(4096, sizeof(struct kvm_xsave));
     memset(xsave, 0, sizeof(struct kvm_xsave));
     twd = 0;
     swd = env->fpus & ~(7 << 11);
@@ -801,7 +804,6 @@ static int kvm_put_xsave(CPUState *env)
     memcpy(&xsave->region[XSAVE_YMMH_SPACE], env->ymmh_regs,
             sizeof env->ymmh_regs);
     r = kvm_vcpu_ioctl(env, KVM_SET_XSAVE, xsave);
-    g_free(xsave);
     return r;
 }
 
@@ -978,7 +980,7 @@ static int kvm_get_fpu(CPUState *env)
 
 static int kvm_get_xsave(CPUState *env)
 {
-    struct kvm_xsave* xsave;
+    struct kvm_xsave* xsave = env->kvm_xsave_buf;
     int ret, i;
     uint16_t cwd, swd, twd;
 
@@ -986,10 +988,8 @@ static int kvm_get_xsave(CPUState *env)
         return kvm_get_fpu(env);
     }
 
-    xsave = qemu_memalign(4096, sizeof(struct kvm_xsave));
     ret = kvm_vcpu_ioctl(env, KVM_GET_XSAVE, xsave);
     if (ret < 0) {
-        g_free(xsave);
         return ret;
     }
 
@@ -1013,7 +1013,6 @@ static int kvm_get_xsave(CPUState *env)
     env->xstate_bv = *(uint64_t *)&xsave->region[XSAVE_XSTATE_BV];
     memcpy(env->ymmh_regs, &xsave->region[XSAVE_YMMH_SPACE],
             sizeof env->ymmh_regs);
-    g_free(xsave);
     return 0;
 }
 
-- 
1.7.6.4

[Qemu-devel] [PATCH 3/5] kvm: x86: Drop redundant apic base and tpr update from kvm_get_sregs

[Marcelo Tosatti]

From: Jan Kiszka <jan.kiszka@siemens.com>

The latter was already commented out, the former is redundant as well.
We always get the latest changes after return from the guest via
kvm_arch_post_run.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
---
 target-i386/kvm.c |    5 ++---
 1 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/target-i386/kvm.c b/target-i386/kvm.c
index 06f4401..d206852 100644
--- a/target-i386/kvm.c
+++ b/target-i386/kvm.c
@@ -1082,10 +1082,9 @@ static int kvm_get_sregs(CPUState *env)
     env->cr[3] = sregs.cr3;
     env->cr[4] = sregs.cr4;
 
-    cpu_set_apic_base(env->apic_state, sregs.apic_base);
-
     env->efer = sregs.efer;
-    //cpu_set_apic_tpr(env->apic_state, sregs.cr8);
+
+    /* changes to apic base and cr8/tpr are read back via kvm_arch_post_run */
 
 #define HFLAG_COPY_MASK \
     ~( HF_CPL_MASK | HF_PE_MASK | HF_MP_MASK | HF_EM_MASK | \
-- 
1.7.6.4

[Qemu-devel] [PATCH 4/5] Set numa topology for max_cpus

[Marcelo Tosatti]

From: Vasilis Liaskovitis <vasilis.liaskovitis@profitbricks.com>

qemu-kvm passes numa/SRAT topology information for smp_cpus to SeaBIOS. However
SeaBIOS always expects to setup max_cpus number of SRAT cpu entries
(MaxCountCPUs variable in build_srat function of Seabios). When qemu-kvm runs
with smp_cpus != max_cpus (e.g. -smp 2,maxcpus=4), Seabios will mistakenly use
memory SRAT info for setting up CPU SRAT entries for the offline CPUs. Wrong
SRAT memory entries are also created. This breaks NUMA in a guest.
Fix by setting up SRAT info for max_cpus in qemu-kvm.

Signed-off-by: Vasilis Liaskovitis <vasilis.liaskovitis@profitbricks.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
---
 hw/pc.c |    8 ++++----
 vl.c    |    2 +-
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/hw/pc.c b/hw/pc.c
index 3a71992..f51afa8 100644
--- a/hw/pc.c
+++ b/hw/pc.c
@@ -624,9 +624,9 @@ static void *bochs_bios_init(void)
      * of nodes, one word for each VCPU->node and one word for each node to
      * hold the amount of memory.
      */
-    numa_fw_cfg = g_malloc0((1 + smp_cpus + nb_numa_nodes) * 8);
+    numa_fw_cfg = g_malloc0((1 + max_cpus + nb_numa_nodes) * 8);
     numa_fw_cfg[0] = cpu_to_le64(nb_numa_nodes);
-    for (i = 0; i < smp_cpus; i++) {
+    for (i = 0; i < max_cpus; i++) {
         for (j = 0; j < nb_numa_nodes; j++) {
             if (node_cpumask[j] & (1 << i)) {
                 numa_fw_cfg[i + 1] = cpu_to_le64(j);
@@ -635,10 +635,10 @@ static void *bochs_bios_init(void)
         }
     }
     for (i = 0; i < nb_numa_nodes; i++) {
-        numa_fw_cfg[smp_cpus + 1 + i] = cpu_to_le64(node_mem[i]);
+        numa_fw_cfg[max_cpus + 1 + i] = cpu_to_le64(node_mem[i]);
     }
     fw_cfg_add_bytes(fw_cfg, FW_CFG_NUMA, (uint8_t *)numa_fw_cfg,
-                     (1 + smp_cpus + nb_numa_nodes) * 8);
+                     (1 + max_cpus + nb_numa_nodes) * 8);
 
     return fw_cfg;
 }
diff --git a/vl.c b/vl.c
index c03abb6..d925424 100644
--- a/vl.c
+++ b/vl.c
@@ -3305,7 +3305,7 @@ int main(int argc, char **argv, char **envp)
          * real machines which also use this scheme.
          */
         if (i == nb_numa_nodes) {
-            for (i = 0; i < smp_cpus; i++) {
+            for (i = 0; i < max_cpus; i++) {
                 node_cpumask[i % nb_numa_nodes] |= 1 << i;
             }
         }
-- 
1.7.6.4

[Qemu-devel] [PATCH 5/5] enable architectural PMU cpuid leaf for kvm

[Marcelo Tosatti]

From: Gleb Natapov <gleb@redhat.com>

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
---
 target-i386/cpuid.c |   17 +++++++++++++----
 1 files changed, 13 insertions(+), 4 deletions(-)

diff --git a/target-i386/cpuid.c b/target-i386/cpuid.c
index 0b3af90..91a104b 100644
--- a/target-i386/cpuid.c
+++ b/target-i386/cpuid.c
@@ -1180,10 +1180,19 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
         break;
     case 0xA:
         /* Architectural Performance Monitoring Leaf */
-        *eax = 0;
-        *ebx = 0;
-        *ecx = 0;
-        *edx = 0;
+        if (kvm_enabled()) {
+            KVMState *s = env->kvm_state;
+
+            *eax = kvm_arch_get_supported_cpuid(s, 0xA, count, R_EAX);
+            *ebx = kvm_arch_get_supported_cpuid(s, 0xA, count, R_EBX);
+            *ecx = kvm_arch_get_supported_cpuid(s, 0xA, count, R_ECX);
+            *edx = kvm_arch_get_supported_cpuid(s, 0xA, count, R_EDX);
+        } else {
+            *eax = 0;
+            *ebx = 0;
+            *ecx = 0;
+            *edx = 0;
+        }
         break;
     case 0xD:
         /* Processor Extended State */
-- 
1.7.6.4

Re: [Qemu-devel] [PATCH 0/5] [PULL] qemu-kvm.git uq/master queue

[Anthony Liguori]

On 12/22/2011 02:13 PM, Marcelo Tosatti wrote:
> The following changes since commit 03ecd2c80a64d030a22fe67cc7a60f24e17ff211:
>
>    virtio-serial-bus: Ports are expected to implement 'have_data' callback (2011-12-21 15:00:29 -0600)
>
> are available in the git repository at:
>    git://git.kernel.org/pub/scm/virt/kvm/qemu-kvm.git uq/master

Pulled.  Thanks.

Regards,

Anthony Liguori

>
> Gleb Natapov (1):
>        enable architectural PMU cpuid leaf for kvm
>
> Jan Kiszka (3):
>        kvm: x86: Use symbols for all xsave field
>        kvm: x86: Avoid runtime allocation of xsave buffer
>        kvm: x86: Drop redundant apic base and tpr update from kvm_get_sregs
>
> Vasilis Liaskovitis (1):
>        Set numa topology for max_cpus
>
>   hw/pc.c             |    8 ++++----
>   target-i386/cpu.h   |    3 ++-
>   target-i386/cpuid.c |   17 +++++++++++++----
>   target-i386/kvm.c   |   34 +++++++++++++++++-----------------
>   vl.c                |    2 +-
>   5 files changed, 37 insertions(+), 27 deletions(-)
>
>