[xen staging-4.20] x86/S3: don't lose previously collected CPU data (AMD extra)

[patchbot@xxxxxxx]

commit 69568ab30622ba09a398a65c940f1101b927a099
Author:     Jan Beulich <jbeulich@xxxxxxxx>
AuthorDate: Tue Jul 22 12:42:32 2025 +0200
Commit:     Jan Beulich <jbeulich@xxxxxxxx>
CommitDate: Tue Jul 22 12:42:32 2025 +0200

    x86/S3: don't lose previously collected CPU data (AMD extra)
    
    smp_store_cpu_info() doesn't call identify_cpu() again during S3 resume,
    hence it is wrong to zap APs' data while bringing them down (and we're
    not just parking them there) during suspend.
    
    Fixes: bb502a8ca5925 ("x86: check feature flags after resume")
    Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
    Reviewed-by: Juergen Gross <jgross@xxxxxxxx>
    master commit: 55719030b0bb0069fc8b57cd808dc98dc9d39add
    master date: 2025-07-17 17:55:54 +0200
---
 xen/arch/x86/smpboot.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/xen/arch/x86/smpboot.c b/xen/arch/x86/smpboot.c
index 10d57dfb82..348e0b9581 100644
--- a/xen/arch/x86/smpboot.c
+++ b/xen/arch/x86/smpboot.c
@@ -978,9 +978,12 @@ static void cpu_smpboot_free(unsigned int cpu, bool remove)
 
     if ( remove )
     {
-        c[cpu].phys_proc_id = XEN_INVALID_SOCKET_ID;
-        c[cpu].cpu_core_id = XEN_INVALID_CORE_ID;
-        c[cpu].compute_unit_id = INVALID_CUID;
+        if ( system_state != SYS_STATE_suspend )
+        {
+            c[cpu].phys_proc_id = XEN_INVALID_SOCKET_ID;
+            c[cpu].cpu_core_id = XEN_INVALID_CORE_ID;
+            c[cpu].compute_unit_id = INVALID_CUID;
+        }
 
         FREE_CPUMASK_VAR(per_cpu(cpu_sibling_mask, cpu));
         FREE_CPUMASK_VAR(per_cpu(cpu_core_mask, cpu));
--
generated by git-patchbot for /home/xen/git/xen.git#staging-4.20