qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Igor Mammedov <imammedo@redhat.com>
To: Eduardo Habkost <ehabkost@redhat.com>
Cc: qemu-devel@nongnu.org, dahi@linux.vnet.ibm.com,
	Paolo Bonzini <pbonzini@redhat.com>,
	Jiri Denemark <jdenemar@redhat.com>,
	Markus Armbruster <armbru@redhat.com>,
	Richard Henderson <rth@twiddle.net>,
	libvir-list@redhat.com
Subject: Re: [Qemu-devel] [PATCH v6 2/3] target-i386: x86_cpu_load_features() function
Date: Mon, 10 Oct 2016 14:25:32 +0200	[thread overview]
Message-ID: <20161010142532.6ee864b8@nial.brq.redhat.com> (raw)
In-Reply-To: <1475872142-3986-3-git-send-email-ehabkost@redhat.com>

On Fri,  7 Oct 2016 17:29:01 -0300
Eduardo Habkost <ehabkost@redhat.com> wrote:

> When probing for CPU model information, we need to reuse the code
> that initializes CPUID fields, but not the remaining side-effects
> of x86_cpu_realizefn(). Move that code to a separate function
> that can be reused later.
> 
> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
> ---
> Changes v5 -> v6:
> * Move x86_cpu_filter_features() outside x86_cpu_load_features(),
>   as the CPU model querying API won't run
>   x86_cpu_filter_features() on most cases

> 
> Changes v4 -> v5:
> * Fix typo on x86_cpu_load_features() comment
>   Reported-by: Paolo Bonzini <pbonzini@redhat.com>
> 
> Changes series v3 -> v4:
> * New patch added to series
> ---
>  target-i386/cpu.c | 67 +++++++++++++++++++++++++++++++++++--------------------
>  1 file changed, 43 insertions(+), 24 deletions(-)
> 
> diff --git a/target-i386/cpu.c b/target-i386/cpu.c
> index 1e8127b..23cc19b 100644
> --- a/target-i386/cpu.c
> +++ b/target-i386/cpu.c
> @@ -2993,34 +2993,13 @@ static void x86_cpu_enable_xsave_components(X86CPU *cpu)
>      env->features[FEAT_XSAVE_COMP_HI] = mask >> 32;
>  }
>  
> -#define IS_INTEL_CPU(env) ((env)->cpuid_vendor1 == CPUID_VENDOR_INTEL_1 && \
> -                           (env)->cpuid_vendor2 == CPUID_VENDOR_INTEL_2 && \
> -                           (env)->cpuid_vendor3 == CPUID_VENDOR_INTEL_3)
> -#define IS_AMD_CPU(env) ((env)->cpuid_vendor1 == CPUID_VENDOR_AMD_1 && \
> -                         (env)->cpuid_vendor2 == CPUID_VENDOR_AMD_2 && \
> -                         (env)->cpuid_vendor3 == CPUID_VENDOR_AMD_3)
> -static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
> +/* Load CPUID data based on configured features */
> +static void x86_cpu_load_features(X86CPU *cpu, Error **errp)
>  {
> -    CPUState *cs = CPU(dev);
> -    X86CPU *cpu = X86_CPU(dev);
> -    X86CPUClass *xcc = X86_CPU_GET_CLASS(dev);
>      CPUX86State *env = &cpu->env;
> -    Error *local_err = NULL;
> -    static bool ht_warned;
>      FeatureWord w;
>      GList *l;
> -
> -    if (xcc->kvm_required && !kvm_enabled()) {
> -        char *name = x86_cpu_class_get_model_name(xcc);
> -        error_setg(&local_err, "CPU model '%s' requires KVM", name);
> -        g_free(name);
> -        goto out;
> -    }
> -
> -    if (cpu->apic_id == UNASSIGNED_APIC_ID) {
> -        error_setg(errp, "apic-id property was not initialized properly");
> -        return;
> -    }
> +    Error *local_err = NULL;
>  
>      /*TODO: cpu->host_features incorrectly overwrites features
>       * set using "feat=on|off". Once we fix this, we can convert
> @@ -3086,6 +3065,46 @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
>          env->cpuid_xlevel2 = env->cpuid_min_xlevel2;
>      }
>  
> +out:
> +    if (local_err != NULL) {
> +        error_propagate(errp, local_err);
> +    }
> +}
> +
> +#define IS_INTEL_CPU(env) ((env)->cpuid_vendor1 == CPUID_VENDOR_INTEL_1 && \
> +                           (env)->cpuid_vendor2 == CPUID_VENDOR_INTEL_2 && \
> +                           (env)->cpuid_vendor3 == CPUID_VENDOR_INTEL_3)
> +#define IS_AMD_CPU(env) ((env)->cpuid_vendor1 == CPUID_VENDOR_AMD_1 && \
> +                         (env)->cpuid_vendor2 == CPUID_VENDOR_AMD_2 && \
> +                         (env)->cpuid_vendor3 == CPUID_VENDOR_AMD_3)
> +static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
> +{
> +    CPUState *cs = CPU(dev);
> +    X86CPU *cpu = X86_CPU(dev);
> +    X86CPUClass *xcc = X86_CPU_GET_CLASS(dev);
> +    CPUX86State *env = &cpu->env;
> +    Error *local_err = NULL;
> +    static bool ht_warned;
> +
> +    if (xcc->kvm_required && !kvm_enabled()) {
> +        char *name = x86_cpu_class_get_model_name(xcc);
> +        error_setg(&local_err, "CPU model '%s' requires KVM", name);
> +        g_free(name);
> +        goto out;
> +    }
> +
> +    if (cpu->apic_id == UNASSIGNED_APIC_ID) {
> +        error_setg(errp, "apic-id property was not initialized properly");
> +        return;
> +    }
> +
> +    x86_cpu_load_features(cpu, &local_err);
> +    if (local_err) {
> +        goto out;
> +    }
> +
> +    x86_cpu_filter_features(cpu);
that makes 2 invocations of ^^ inside realize,
see followup line vvvv
 
[...]
>      if (x86_cpu_filter_features(cpu) &&
>          (cpu->check_cpuid || cpu->enforce_cpuid)) {
>          x86_cpu_report_filtered_features(cpu);

  reply	other threads:[~2016-10-10 12:25 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-10-07 20:28 [Qemu-devel] [PATCH v6 0/3] Add runnability info to query-cpu-definitions Eduardo Habkost
2016-10-07 20:29 ` [Qemu-devel] [PATCH v6 1/3] target-i386: Move warning code outside x86_cpu_filter_features() Eduardo Habkost
2016-10-10 11:57   ` Igor Mammedov
2016-10-10 17:04     ` Eduardo Habkost
2016-10-07 20:29 ` [Qemu-devel] [PATCH v6 2/3] target-i386: x86_cpu_load_features() function Eduardo Habkost
2016-10-10 12:25   ` Igor Mammedov [this message]
2016-10-10 16:58     ` [Qemu-devel] [PATCH] fixup! " Eduardo Habkost
2016-10-11 11:41       ` Igor Mammedov
2016-10-14 14:59         ` Eduardo Habkost
2016-10-07 20:29 ` [Qemu-devel] [PATCH v6 3/3] target-i386: Return runnability information on query-cpu-definitions Eduardo Habkost
2016-10-10 12:27   ` Igor Mammedov
2016-10-10 17:01     ` Eduardo Habkost
2016-10-11 11:45       ` Igor Mammedov
2016-10-11 11:58         ` Eduardo Habkost
2016-10-11 13:21           ` Igor Mammedov
2016-10-11 13:24             ` Eduardo Habkost
2016-10-11 13:51               ` Igor Mammedov
2016-10-11 14:13   ` Igor Mammedov
2016-10-14 14:59     ` Eduardo Habkost

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20161010142532.6ee864b8@nial.brq.redhat.com \
    --to=imammedo@redhat.com \
    --cc=armbru@redhat.com \
    --cc=dahi@linux.vnet.ibm.com \
    --cc=ehabkost@redhat.com \
    --cc=jdenemar@redhat.com \
    --cc=libvir-list@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).