* [PATCH v3] x86/amd: don't set X86_BUG_SYSRET_SS_ATTRS when running under Xen
@ 2017-04-27 5:01 Juergen Gross
2017-05-05 5:09 ` Juergen Gross
2017-05-11 10:31 ` Juergen Gross
0 siblings, 2 replies; 5+ messages in thread
From: Juergen Gross @ 2017-04-27 5:01 UTC (permalink / raw)
To: linux-kernel, xen-devel, x86
Cc: boris.ostrovsky, hpa, tglx, mingo, Juergen Gross
When running as Xen pv guest X86_BUG_SYSRET_SS_ATTRS must not be set
on AMD cpus.
This bug/feature bit is kind of special as it will be used very early
when switching threads. Setting the bit and clearing it a little bit
later leaves a critical window where things can go wrong. This time
window has enlarged a little bit by using setup_clear_cpu_cap() instead
of the hypervisor's set_cpu_features callback. It seems this larger
window now makes it rather easy to hit the problem.
The proper solution is to never set the bit in case of Xen.
Signed-off-by: Juergen Gross <jgross@suse.com>
---
arch/x86/kernel/cpu/amd.c | 5 +++--
arch/x86/xen/enlighten_pv.c | 1 -
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index c36140d788fe..b6da6e75e3a8 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -799,8 +799,9 @@ static void init_amd(struct cpuinfo_x86 *c)
if (cpu_has(c, X86_FEATURE_3DNOW) || cpu_has(c, X86_FEATURE_LM))
set_cpu_cap(c, X86_FEATURE_3DNOWPREFETCH);
- /* AMD CPUs don't reset SS attributes on SYSRET */
- set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS);
+ /* AMD CPUs don't reset SS attributes on SYSRET, Xen does. */
+ if (!cpu_has(c, X86_FEATURE_XENPV))
+ set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS);
}
#ifdef CONFIG_X86_32
diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c
index a1af4f68278f..dcfd07faf1c3 100644
--- a/arch/x86/xen/enlighten_pv.c
+++ b/arch/x86/xen/enlighten_pv.c
@@ -290,7 +290,6 @@ static bool __init xen_check_xsave(void)
static void __init xen_init_capabilities(void)
{
- setup_clear_cpu_cap(X86_BUG_SYSRET_SS_ATTRS);
setup_force_cpu_cap(X86_FEATURE_XENPV);
setup_clear_cpu_cap(X86_FEATURE_DCA);
setup_clear_cpu_cap(X86_FEATURE_APERFMPERF);
--
2.12.0
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: [PATCH v3] x86/amd: don't set X86_BUG_SYSRET_SS_ATTRS when running under Xen
2017-04-27 5:01 [PATCH v3] x86/amd: don't set X86_BUG_SYSRET_SS_ATTRS when running under Xen Juergen Gross
@ 2017-05-05 5:09 ` Juergen Gross
2017-05-05 16:02 ` Boris Ostrovsky
2017-05-11 10:31 ` Juergen Gross
1 sibling, 1 reply; 5+ messages in thread
From: Juergen Gross @ 2017-05-05 5:09 UTC (permalink / raw)
To: linux-kernel, xen-devel, x86; +Cc: boris.ostrovsky, hpa, tglx, mingo
Any comments?
Juergen
On 27/04/17 07:01, Juergen Gross wrote:
> When running as Xen pv guest X86_BUG_SYSRET_SS_ATTRS must not be set
> on AMD cpus.
>
> This bug/feature bit is kind of special as it will be used very early
> when switching threads. Setting the bit and clearing it a little bit
> later leaves a critical window where things can go wrong. This time
> window has enlarged a little bit by using setup_clear_cpu_cap() instead
> of the hypervisor's set_cpu_features callback. It seems this larger
> window now makes it rather easy to hit the problem.
>
> The proper solution is to never set the bit in case of Xen.
>
> Signed-off-by: Juergen Gross <jgross@suse.com>
> ---
> arch/x86/kernel/cpu/amd.c | 5 +++--
> arch/x86/xen/enlighten_pv.c | 1 -
> 2 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
> index c36140d788fe..b6da6e75e3a8 100644
> --- a/arch/x86/kernel/cpu/amd.c
> +++ b/arch/x86/kernel/cpu/amd.c
> @@ -799,8 +799,9 @@ static void init_amd(struct cpuinfo_x86 *c)
> if (cpu_has(c, X86_FEATURE_3DNOW) || cpu_has(c, X86_FEATURE_LM))
> set_cpu_cap(c, X86_FEATURE_3DNOWPREFETCH);
>
> - /* AMD CPUs don't reset SS attributes on SYSRET */
> - set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS);
> + /* AMD CPUs don't reset SS attributes on SYSRET, Xen does. */
> + if (!cpu_has(c, X86_FEATURE_XENPV))
> + set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS);
> }
>
> #ifdef CONFIG_X86_32
> diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c
> index a1af4f68278f..dcfd07faf1c3 100644
> --- a/arch/x86/xen/enlighten_pv.c
> +++ b/arch/x86/xen/enlighten_pv.c
> @@ -290,7 +290,6 @@ static bool __init xen_check_xsave(void)
>
> static void __init xen_init_capabilities(void)
> {
> - setup_clear_cpu_cap(X86_BUG_SYSRET_SS_ATTRS);
> setup_force_cpu_cap(X86_FEATURE_XENPV);
> setup_clear_cpu_cap(X86_FEATURE_DCA);
> setup_clear_cpu_cap(X86_FEATURE_APERFMPERF);
>
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH v3] x86/amd: don't set X86_BUG_SYSRET_SS_ATTRS when running under Xen
2017-05-05 5:09 ` Juergen Gross
@ 2017-05-05 16:02 ` Boris Ostrovsky
0 siblings, 0 replies; 5+ messages in thread
From: Boris Ostrovsky @ 2017-05-05 16:02 UTC (permalink / raw)
To: Juergen Gross, linux-kernel, xen-devel, x86; +Cc: hpa, tglx, mingo
On 05/05/2017 01:09 AM, Juergen Gross wrote:
> Any comments?
>
>
> Juergen
>
> On 27/04/17 07:01, Juergen Gross wrote:
>> When running as Xen pv guest X86_BUG_SYSRET_SS_ATTRS must not be set
>> on AMD cpus.
>>
>> This bug/feature bit is kind of special as it will be used very early
>> when switching threads. Setting the bit and clearing it a little bit
>> later leaves a critical window where things can go wrong. This time
>> window has enlarged a little bit by using setup_clear_cpu_cap() instead
>> of the hypervisor's set_cpu_features callback. It seems this larger
>> window now makes it rather easy to hit the problem.
>>
>> The proper solution is to never set the bit in case of Xen.
FWIW,
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
>>
>> Signed-off-by: Juergen Gross <jgross@suse.com>
>> ---
>> arch/x86/kernel/cpu/amd.c | 5 +++--
>> arch/x86/xen/enlighten_pv.c | 1 -
>> 2 files changed, 3 insertions(+), 3 deletions(-)
>>
>> diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
>> index c36140d788fe..b6da6e75e3a8 100644
>> --- a/arch/x86/kernel/cpu/amd.c
>> +++ b/arch/x86/kernel/cpu/amd.c
>> @@ -799,8 +799,9 @@ static void init_amd(struct cpuinfo_x86 *c)
>> if (cpu_has(c, X86_FEATURE_3DNOW) || cpu_has(c, X86_FEATURE_LM))
>> set_cpu_cap(c, X86_FEATURE_3DNOWPREFETCH);
>>
>> - /* AMD CPUs don't reset SS attributes on SYSRET */
>> - set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS);
>> + /* AMD CPUs don't reset SS attributes on SYSRET, Xen does. */
>> + if (!cpu_has(c, X86_FEATURE_XENPV))
>> + set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS);
>> }
>>
>> #ifdef CONFIG_X86_32
>> diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c
>> index a1af4f68278f..dcfd07faf1c3 100644
>> --- a/arch/x86/xen/enlighten_pv.c
>> +++ b/arch/x86/xen/enlighten_pv.c
>> @@ -290,7 +290,6 @@ static bool __init xen_check_xsave(void)
>>
>> static void __init xen_init_capabilities(void)
>> {
>> - setup_clear_cpu_cap(X86_BUG_SYSRET_SS_ATTRS);
>> setup_force_cpu_cap(X86_FEATURE_XENPV);
>> setup_clear_cpu_cap(X86_FEATURE_DCA);
>> setup_clear_cpu_cap(X86_FEATURE_APERFMPERF);
>>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v3] x86/amd: don't set X86_BUG_SYSRET_SS_ATTRS when running under Xen
2017-04-27 5:01 [PATCH v3] x86/amd: don't set X86_BUG_SYSRET_SS_ATTRS when running under Xen Juergen Gross
2017-05-05 5:09 ` Juergen Gross
@ 2017-05-11 10:31 ` Juergen Gross
2017-05-11 12:46 ` Thomas Gleixner
1 sibling, 1 reply; 5+ messages in thread
From: Juergen Gross @ 2017-05-11 10:31 UTC (permalink / raw)
To: linux-kernel, xen-devel, x86; +Cc: boris.ostrovsky, hpa, tglx, mingo
On 27/04/17 07:01, Juergen Gross wrote:
> When running as Xen pv guest X86_BUG_SYSRET_SS_ATTRS must not be set
> on AMD cpus.
>
> This bug/feature bit is kind of special as it will be used very early
> when switching threads. Setting the bit and clearing it a little bit
> later leaves a critical window where things can go wrong. This time
> window has enlarged a little bit by using setup_clear_cpu_cap() instead
> of the hypervisor's set_cpu_features callback. It seems this larger
> window now makes it rather easy to hit the problem.
>
> The proper solution is to never set the bit in case of Xen.
>
> Signed-off-by: Juergen Gross <jgross@suse.com>
Any objections for carrying this through the Xen tree?
Juergen
> ---
> arch/x86/kernel/cpu/amd.c | 5 +++--
> arch/x86/xen/enlighten_pv.c | 1 -
> 2 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
> index c36140d788fe..b6da6e75e3a8 100644
> --- a/arch/x86/kernel/cpu/amd.c
> +++ b/arch/x86/kernel/cpu/amd.c
> @@ -799,8 +799,9 @@ static void init_amd(struct cpuinfo_x86 *c)
> if (cpu_has(c, X86_FEATURE_3DNOW) || cpu_has(c, X86_FEATURE_LM))
> set_cpu_cap(c, X86_FEATURE_3DNOWPREFETCH);
>
> - /* AMD CPUs don't reset SS attributes on SYSRET */
> - set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS);
> + /* AMD CPUs don't reset SS attributes on SYSRET, Xen does. */
> + if (!cpu_has(c, X86_FEATURE_XENPV))
> + set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS);
> }
>
> #ifdef CONFIG_X86_32
> diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c
> index a1af4f68278f..dcfd07faf1c3 100644
> --- a/arch/x86/xen/enlighten_pv.c
> +++ b/arch/x86/xen/enlighten_pv.c
> @@ -290,7 +290,6 @@ static bool __init xen_check_xsave(void)
>
> static void __init xen_init_capabilities(void)
> {
> - setup_clear_cpu_cap(X86_BUG_SYSRET_SS_ATTRS);
> setup_force_cpu_cap(X86_FEATURE_XENPV);
> setup_clear_cpu_cap(X86_FEATURE_DCA);
> setup_clear_cpu_cap(X86_FEATURE_APERFMPERF);
>
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH v3] x86/amd: don't set X86_BUG_SYSRET_SS_ATTRS when running under Xen
2017-05-11 10:31 ` Juergen Gross
@ 2017-05-11 12:46 ` Thomas Gleixner
0 siblings, 0 replies; 5+ messages in thread
From: Thomas Gleixner @ 2017-05-11 12:46 UTC (permalink / raw)
To: Juergen Gross; +Cc: linux-kernel, xen-devel, x86, boris.ostrovsky, hpa, mingo
On Thu, 11 May 2017, Juergen Gross wrote:
> On 27/04/17 07:01, Juergen Gross wrote:
> > When running as Xen pv guest X86_BUG_SYSRET_SS_ATTRS must not be set
> > on AMD cpus.
> >
> > This bug/feature bit is kind of special as it will be used very early
> > when switching threads. Setting the bit and clearing it a little bit
> > later leaves a critical window where things can go wrong. This time
> > window has enlarged a little bit by using setup_clear_cpu_cap() instead
> > of the hypervisor's set_cpu_features callback. It seems this larger
> > window now makes it rather easy to hit the problem.
> >
> > The proper solution is to never set the bit in case of Xen.
> >
> > Signed-off-by: Juergen Gross <jgross@suse.com>
>
> Any objections for carrying this through the Xen tree?
Acked-by: Thomas Gleixner <tglx@linutronix.de>
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2017-05-11 12:47 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-04-27 5:01 [PATCH v3] x86/amd: don't set X86_BUG_SYSRET_SS_ATTRS when running under Xen Juergen Gross
2017-05-05 5:09 ` Juergen Gross
2017-05-05 16:02 ` Boris Ostrovsky
2017-05-11 10:31 ` Juergen Gross
2017-05-11 12:46 ` Thomas Gleixner
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).