[Qemu-devel] [PULL 0/3] QOM CPUState patch queue 2014-09-04

[Qemu-devel] [PULL 0/3] QOM CPUState patch queue 2014-09-04

[Andreas Färber]

Hello Peter,

This is my QOM CPU patch queue. Please pull.

Regards,
Andreas

Cc: Peter Maydell <peter.maydell@linaro.org>

Cc: Eduardo Habkost <ehabkost@redhat.com>

The following changes since commit 01eb313907dda97313b8fea62e5632fca64f069c:

  Merge remote-tracking branch 'remotes/mjt/tags/trivial-patches-2014-09-03' into staging (2014-09-04 13:33:53 +0100)

are available in the git repository at:


  git://github.com/afaerber/qemu-cpu.git tags/qom-cpu-for-peter

for you to fetch changes up to 295efacb5eaae69f0b69d91bb7e9f7b3ead9ca3e:

  target-i386: Reject invalid CPU feature names on the command-line (2014-09-04 19:06:19 +0200)

----------------------------------------------------------------
X86CPU

* Fix -cpu *,migratable=foo
* Error out on unknown -cpu *,+foo,-bar

----------------------------------------------------------------
Eduardo Habkost (2):
      target-i386: Support migratable=no properly
      target-i386: Reject invalid CPU feature names on the command-line

Pavel Dovgaluk (1):
      exec: Save CPUState::exception_index field

 exec.c                | 35 +++++++++++++++++++++++++++++++++++
 target-i386/cpu-qom.h |  1 +
 target-i386/cpu.c     | 32 ++++++++++++++++++++------------
 3 files changed, 56 insertions(+), 12 deletions(-)

[Qemu-devel] [PULL v2 1/3] exec: Save CPUState::exception_index field

[Andreas Färber]

From: Pavel Dovgaluk <Pavel.Dovgaluk@ispras.ru>

This patch adds a subsection with exception_index field to the VMState for
correct saving the CPU state.
Without this patch, simulator could miss the pending exception in the saved
virtual machine state.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
Signed-off-by: Andreas Färber <afaerber@suse.de>
---
 exec.c | 35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/exec.c b/exec.c
index 5122a33..7dddcc8 100644
--- a/exec.c
+++ b/exec.c
@@ -430,15 +430,50 @@ static int cpu_common_post_load(void *opaque, int version_id)
     return 0;
 }
 
+static int cpu_common_pre_load(void *opaque)
+{
+    CPUState *cpu = opaque;
+
+    cpu->exception_index = 0;
+
+    return 0;
+}
+
+static bool cpu_common_exception_index_needed(void *opaque)
+{
+    CPUState *cpu = opaque;
+
+    return cpu->exception_index != 0;
+}
+
+static const VMStateDescription vmstate_cpu_common_exception_index = {
+    .name = "cpu_common/exception_index",
+    .version_id = 1,
+    .minimum_version_id = 1,
+    .fields = (VMStateField[]) {
+        VMSTATE_INT32(exception_index, CPUState),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
 const VMStateDescription vmstate_cpu_common = {
     .name = "cpu_common",
     .version_id = 1,
     .minimum_version_id = 1,
+    .pre_load = cpu_common_pre_load,
     .post_load = cpu_common_post_load,
     .fields = (VMStateField[]) {
         VMSTATE_UINT32(halted, CPUState),
         VMSTATE_UINT32(interrupt_request, CPUState),
         VMSTATE_END_OF_LIST()
+    },
+    .subsections = (VMStateSubsection[]) {
+        {
+            .vmsd = &vmstate_cpu_common_exception_index,
+            .needed = cpu_common_exception_index_needed,
+        } , {
+            /* empty */
+        }
     }
 };
 
-- 
1.8.4.5

[Qemu-devel] [PULL v2 2/3] target-i386: Support migratable=no properly

[Andreas Färber]

From: Eduardo Habkost <ehabkost@redhat.com>

When the "migratable" property was implemented, the behavior was tested
by changing the default on the code, but actually using the option on
the command-line (e.g. "-cpu host,migratable=false") doesn't work as
expected. This is a regression for a common use case of "-cpu host",
which is to enable features that are supported by the host CPU + kernel
before feature-specific code is added to QEMU.

Fix this by initializing the feature words for "-cpu host" on
x86_cpu_parse_featurestr(), right after parsing the CPU options.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Andreas Färber <afaerber@suse.de>
---
 target-i386/cpu-qom.h |  1 +
 target-i386/cpu.c     | 17 ++++++++++++-----
 2 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/target-i386/cpu-qom.h b/target-i386/cpu-qom.h
index 71a1b97..7755466 100644
--- a/target-i386/cpu-qom.h
+++ b/target-i386/cpu-qom.h
@@ -92,6 +92,7 @@ typedef struct X86CPU {
     bool enforce_cpuid;
     bool expose_kvm;
     bool migratable;
+    bool host_features;
 
     /* if true the CPUID code directly forward host cache leaves to the guest */
     bool cache_info_passthrough;
diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index fa811a0..60d0dd5 100644
--- a/target-i386/cpu.c
+++ b/target-i386/cpu.c
@@ -1318,18 +1318,18 @@ static void host_x86_cpu_initfn(Object *obj)
     X86CPU *cpu = X86_CPU(obj);
     CPUX86State *env = &cpu->env;
     KVMState *s = kvm_state;
-    FeatureWord w;
 
     assert(kvm_enabled());
 
+    /* We can't fill the features array here because we don't know yet if
+     * "migratable" is true or false.
+     */
+    cpu->host_features = true;
+
     env->cpuid_level = kvm_arch_get_supported_cpuid(s, 0x0, 0, R_EAX);
     env->cpuid_xlevel = kvm_arch_get_supported_cpuid(s, 0x80000000, 0, R_EAX);
     env->cpuid_xlevel2 = kvm_arch_get_supported_cpuid(s, 0xC0000000, 0, R_EAX);
 
-    for (w = 0; w < FEATURE_WORDS; w++) {
-        env->features[w] =
-            x86_cpu_get_supported_feature_word(w, cpu->migratable);
-    }
     object_property_set_bool(OBJECT(cpu), true, "pmu", &error_abort);
 }
 
@@ -1828,6 +1828,13 @@ static void x86_cpu_parse_featurestr(CPUState *cs, char *features,
         featurestr = strtok(NULL, ",");
     }
 
+    if (cpu->host_features) {
+        for (w = 0; w < FEATURE_WORDS; w++) {
+            env->features[w] =
+                x86_cpu_get_supported_feature_word(w, cpu->migratable);
+        }
+    }
+
     for (w = 0; w < FEATURE_WORDS; w++) {
         env->features[w] |= plus_features[w];
         env->features[w] &= ~minus_features[w];
-- 
1.8.4.5

[Qemu-devel] [PULL v2 3/3] target-i386: Reject invalid CPU feature names on the command-line

[Andreas Färber]

From: Eduardo Habkost <ehabkost@redhat.com>

Instead of simply printing a warning, report an error when invalid CPU
options are provided on the CPU model string.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Andreas Färber <afaerber@suse.de>
---
 target-i386/cpu.c | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index 60d0dd5..88b64d8 100644
--- a/target-i386/cpu.c
+++ b/target-i386/cpu.c
@@ -592,7 +592,8 @@ static bool lookup_feature(uint32_t *pval, const char *s, const char *e,
 }
 
 static void add_flagname_to_bitmaps(const char *flagname,
-                                    FeatureWordArray words)
+                                    FeatureWordArray words,
+                                    Error **errp)
 {
     FeatureWord w;
     for (w = 0; w < FEATURE_WORDS; w++) {
@@ -603,7 +604,7 @@ static void add_flagname_to_bitmaps(const char *flagname,
         }
     }
     if (w == FEATURE_WORDS) {
-        fprintf(stderr, "CPU feature %s not found\n", flagname);
+        error_setg(errp, "CPU feature %s not found", flagname);
     }
 }
 
@@ -1254,6 +1255,9 @@ void x86_cpu_compat_set_features(const char *cpu_model, FeatureWord w,
     }
 }
 
+static uint32_t x86_cpu_get_supported_feature_word(FeatureWord w,
+                                                   bool migratable_only);
+
 #ifdef CONFIG_KVM
 
 static int cpu_x86_fill_model_id(char *str)
@@ -1310,9 +1314,6 @@ static void host_x86_cpu_class_init(ObjectClass *oc, void *data)
     dc->props = host_x86_cpu_properties;
 }
 
-static uint32_t x86_cpu_get_supported_feature_word(FeatureWord w,
-                                                   bool migratable_only);
-
 static void host_x86_cpu_initfn(Object *obj)
 {
     X86CPU *cpu = X86_CPU(obj);
@@ -1761,9 +1762,9 @@ static void x86_cpu_parse_featurestr(CPUState *cs, char *features,
     while (featurestr) {
         char *val;
         if (featurestr[0] == '+') {
-            add_flagname_to_bitmaps(featurestr + 1, plus_features);
+            add_flagname_to_bitmaps(featurestr + 1, plus_features, &local_err);
         } else if (featurestr[0] == '-') {
-            add_flagname_to_bitmaps(featurestr + 1, minus_features);
+            add_flagname_to_bitmaps(featurestr + 1, minus_features, &local_err);
         } else if ((val = strchr(featurestr, '='))) {
             *val = 0; val++;
             feat2prop(featurestr);
-- 
1.8.4.5

Re: [Qemu-devel] [PULL v2 3/3] target-i386: Reject invalid CPU feature names on the command-line

[Eduardo Habkost]

On Thu, Sep 04, 2014 at 07:20:13PM +0200, Andreas Färber wrote:
[...]
> +static uint32_t x86_cpu_get_supported_feature_word(FeatureWord w,
> +                                                   bool migratable_only);
> +
>  #ifdef CONFIG_KVM
>  
>  static int cpu_x86_fill_model_id(char *str)
> @@ -1310,9 +1314,6 @@ static void host_x86_cpu_class_init(ObjectClass *oc, void *data)
>      dc->props = host_x86_cpu_properties;
>  }
>  
> -static uint32_t x86_cpu_get_supported_feature_word(FeatureWord w,
> -                                                   bool migratable_only);
> -

This was supposed to be squashed into patch 1/2, not this one.

(That was my fault, as I mentioned the wrong patch when I submitted the
build error fix.)

-- 
Eduardo

Re: [Qemu-devel] [PULL 0/3] QOM CPUState patch queue 2014-09-04

[Peter Maydell]

On 4 September 2014 18:20, Andreas Färber <afaerber@suse.de> wrote:
> Hello Peter,
>
> This is my QOM CPU patch queue. Please pull.
>
> Regards,
> Andreas
>
> Cc: Peter Maydell <peter.maydell@linaro.org>
>
> Cc: Eduardo Habkost <ehabkost@redhat.com>
>
> The following changes since commit 01eb313907dda97313b8fea62e5632fca64f069c:
>
>   Merge remote-tracking branch 'remotes/mjt/tags/trivial-patches-2014-09-03' into staging (2014-09-04 13:33:53 +0100)
>
> are available in the git repository at:
>
>
>   git://github.com/afaerber/qemu-cpu.git tags/qom-cpu-for-peter
>
> for you to fetch changes up to 295efacb5eaae69f0b69d91bb7e9f7b3ead9ca3e:
>
>   target-i386: Reject invalid CPU feature names on the command-line (2014-09-04 19:06:19 +0200)

Hi; I'm dropping this from my to-process queue since
Eduardo says a fixup got squashed into the wrong patch
by mistake.

thanks
-- PMM

Re: [Qemu-devel] [PULL v2 1/3] exec: Save CPUState::exception_index field

[Peter Maydell]

On 4 September 2014 18:20, Andreas Färber <afaerber@suse.de> wrote:
> From: Pavel Dovgaluk <Pavel.Dovgaluk@ispras.ru>
>
> This patch adds a subsection with exception_index field to the VMState for
> correct saving the CPU state.
> Without this patch, simulator could miss the pending exception in the saved
> virtual machine state.
>
> Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
> Signed-off-by: Andreas Färber <afaerber@suse.de>

Is this patch a worthwhile candidate for stable?

-- PMM

Re: [Qemu-devel] [PULL v2 1/3] exec: Save CPUState::exception_index field

[Andreas Färber]

Am 05.09.2014 00:50, schrieb Peter Maydell:
> On 4 September 2014 18:20, Andreas Färber <afaerber@suse.de> wrote:
>> From: Pavel Dovgaluk <Pavel.Dovgaluk@ispras.ru>
>>
>> This patch adds a subsection with exception_index field to the VMState for
>> correct saving the CPU state.
>> Without this patch, simulator could miss the pending exception in the saved
>> virtual machine state.
>>
>> Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
>> Signed-off-by: Andreas Färber <afaerber@suse.de>
> 
> Is this patch a worthwhile candidate for stable?

Well, it came up in the context of reverse execution, but it wouldn't
hurt to have it backported, I guess. It's difficult to test though...

Andreas

-- 
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg

Re: [Qemu-devel] [PULL 0/3] QOM CPUState patch queue 2014-09-04

[Andreas Färber]

Am 04.09.2014 20:40, schrieb Peter Maydell:
> On 4 September 2014 18:20, Andreas Färber <afaerber@suse.de> wrote:
>> Hello Peter,
>>
>> This is my QOM CPU patch queue. Please pull.
>>
>> Regards,
>> Andreas
>>
>> Cc: Peter Maydell <peter.maydell@linaro.org>
>>
>> Cc: Eduardo Habkost <ehabkost@redhat.com>
>>
>> The following changes since commit 01eb313907dda97313b8fea62e5632fca64f069c:
>>
>>   Merge remote-tracking branch 'remotes/mjt/tags/trivial-patches-2014-09-03' into staging (2014-09-04 13:33:53 +0100)
>>
>> are available in the git repository at:
>>
>>
>>   git://github.com/afaerber/qemu-cpu.git tags/qom-cpu-for-peter
>>
>> for you to fetch changes up to 295efacb5eaae69f0b69d91bb7e9f7b3ead9ca3e:
>>
>>   target-i386: Reject invalid CPU feature names on the command-line (2014-09-04 19:06:19 +0200)
> 
> Hi; I'm dropping this from my to-process queue since
> Eduardo says a fixup got squashed into the wrong patch
> by mistake.

Yeah, v3 is out now. Thanks to both of you for catching this. I must've
screwed up my testing as I only noticed after the final patch...

Andreas

-- 
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg