[PATCH] x86/microcode/amd: fix the return value when microcode has no update

[PATCH] x86/microcode/amd: fix the return value when microcode has no update

[Annie Li]

In commit 6f059e634dcd("x86/microcode: Clarify the late load logic"), the
return value is UCODE_OK if the load is up-to-date in amd platform, which
leads to load_late_locked() returning -EBADFD.

This is different from the intel platform, which will return UCODE_NFOUND
and is more reasonable.

Fix the return to UCODE_NFOUND to avoid -EBADFD error.

Fixes: 6f059e634dcd ("x86/microcode: Clarify the late load logic")
Signed-off-by: Annie Li <jiayanli@google.com>
---
 arch/x86/kernel/cpu/microcode/amd.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/x86/kernel/cpu/microcode/amd.c b/arch/x86/kernel/cpu/microcode/amd.c
index b61028cf5c8a3..f90779dc1cac5 100644
--- a/arch/x86/kernel/cpu/microcode/amd.c
+++ b/arch/x86/kernel/cpu/microcode/amd.c
@@ -1074,6 +1074,8 @@ static enum ucode_state load_microcode_amd(u8 family, const u8 *data, size_t siz
 	if (ret != UCODE_OK)
 		return ret;
 
+	ret = UCODE_NFOUND;
+
 	for_each_node_with_cpus(nid) {
 		cpu = cpumask_first(cpumask_of_node(nid));
 		c = &cpu_data(cpu);
-- 
2.49.0.901.g37484f566f-goog

Re: [PATCH] x86/microcode/amd: fix the return value when microcode has no update

[Borislav Petkov]

On Tue, Apr 29, 2025 at 08:43:16AM +0000, Annie Li wrote:
> In commit 6f059e634dcd("x86/microcode: Clarify the late load logic"), the
> return value is UCODE_OK if the load is up-to-date in amd platform, which
> leads to load_late_locked() returning -EBADFD.
> 
> This is different from the intel platform, which will return UCODE_NFOUND
> and is more reasonable.
> 
> Fix the return to UCODE_NFOUND to avoid -EBADFD error.

... and return -ENOENT instead?

Is that better?

Or would it be even better for the switch-case to handle UCODE_OK too?

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette