x86/cpu: populate CPUID 0x1.edx features early for self-snoop detection

[PATCH for-4.21] x86/cpu: populate CPUID 0x1.edx features early for self-snoop detection

Posted by Roger Pau Monne 1 week, 3 days ago

Otherwise the check for the SS feature in
check_memory_type_self_snoop_errata() fails unconditionally, which leads to
X86_FEATURE_XEN_SELFSNOOP never being set.

We could also avoid this by not doing the reset_cpuinfo() for the BSP in
identify_cpu(), because SS detection uses boot_cpu_data.  However that
creates an imbalance on the state of the BSP versus the APs in the
identify_cpu() code.

I've opted for the less controversial solution of populating FEATURESET_1d
in generic_identify(), as the value is already there.  The same is done for
the AMD faulting probe code.

Fixes: f2663ca2e520 ("x86/cpu/intel: Clear cache self-snoop capability in CPUs with known errata")
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
---
 xen/arch/x86/cpu/common.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/xen/arch/x86/cpu/common.c b/xen/arch/x86/cpu/common.c
index 530b9eb39abc..35dcdf0c8801 100644
--- a/xen/arch/x86/cpu/common.c
+++ b/xen/arch/x86/cpu/common.c
@@ -490,6 +490,9 @@ static void generic_identify(struct cpuinfo_x86 *c)
 	c->apicid = phys_pkg_id((ebx >> 24) & 0xFF, 0);
 	c->phys_proc_id = c->apicid;
 
+	/* Early init of Self Snoop support requires 0x1.edx. */
+	c->x86_capability[FEATURESET_1d] = edx;
+
 	eax = cpuid_eax(0x80000000);
 	if ((eax >> 16) == 0x8000)
 		c->extended_cpuid_level = eax;
-- 
2.51.0

Re: [PATCH for-4.21] x86/cpu: populate CPUID 0x1.edx features early for self-snoop detection

Posted by Andrew Cooper 1 week, 3 days ago

On 24/09/2025 4:00 am, Roger Pau Monne wrote:
> Otherwise the check for the SS feature in
> check_memory_type_self_snoop_errata() fails unconditionally, which leads to
> X86_FEATURE_XEN_SELFSNOOP never being set.
>
> We could also avoid this by not doing the reset_cpuinfo() for the BSP in
> identify_cpu(), because SS detection uses boot_cpu_data.

Doesn't this, mean ...

>   However that
> creates an imbalance on the state of the BSP versus the APs in the
> identify_cpu() code.
>
> I've opted for the less controversial solution of populating FEATURESET_1d
> in generic_identify(), as the value is already there.  The same is done for
> the AMD faulting probe code.
>
> Fixes: f2663ca2e520 ("x86/cpu/intel: Clear cache self-snoop capability in CPUs with known errata")
> Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>

... this Fixes tag is incorrect?

~Andrew