[PATCH] x86/S3: don't lose previously collected CPU data

[PATCH] x86/S3: don't lose previously collected CPU data

[Jan Beulich]

smp_store_cpu_info() doesn't call identify_cpu() again during S3 resume,
hence it is wrong to overwrite APs' data using boot_cpu_data. With
7126b7f806d54 ("x86/CPU: re-work populating of cpu_data[]") this now
results in a crash, as the cpu_to_socket() use in smp_store_cpu_info()
then obtains a bad socket number, much like was already observed in [1].

[1] https://lists.xen.org/archives/html/xen-devel/2025-06/msg01879.html

Fixes: bb502a8ca5925 ("x86: check feature flags after resume")
Signed-off-by: Jan Beulich <jbeulich@suse.com>

--- a/xen/arch/x86/smpboot.c
+++ b/xen/arch/x86/smpboot.c
@@ -370,7 +370,9 @@ void asmlinkage start_secondary(void)
 
     cpu_init();
 
-    initialize_cpu_data(cpu);
+    /* During resume, must not clear previously collected data. */
+    if ( system_state != SYS_STATE_resume )
+        initialize_cpu_data(cpu);
 
     microcode_update_one();
 

Re: [PATCH] x86/S3: don't lose previously collected CPU data

[Roger Pau Monné]

On Thu, Jul 17, 2025 at 12:14:54PM +0200, Jan Beulich wrote:
> smp_store_cpu_info() doesn't call identify_cpu() again during S3 resume,
> hence it is wrong to overwrite APs' data using boot_cpu_data. With
> 7126b7f806d54 ("x86/CPU: re-work populating of cpu_data[]") this now
> results in a crash, as the cpu_to_socket() use in smp_store_cpu_info()
> then obtains a bad socket number, much like was already observed in [1].
> 
> [1] https://lists.xen.org/archives/html/xen-devel/2025-06/msg01879.html
> 
> Fixes: bb502a8ca5925 ("x86: check feature flags after resume")
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

Acked-by: Roger Pau Monné <roger.pau@citrix.com>

Thanks, Roger.