[lkml]   [2023]   [Jan]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Patch in this message
Subject[Part 2 v2[cleanup] 4/4] x86/microcode: Do not call apply_microde() on sibling threads
Microcode updates are applied at the core, and both threads of HT siblings
will notice the update.

During late-load, after the primary has updated the microcode, it also
reflects that in the per-cpu structure (cpuinfo_x86) holding the current

Since the sibling hasn't had a chance to update the per-cpu revision,
the current code calls apply_microcode() just as a way to verify and also
update the per-cpu revision number.

But in the odd case when primary returned with an error, the secondary will
try to perform a patchload and the primary has already been released to the
system. This could be problematic.

Replace apply_microcode() with a call to collect_cpu_info() and let that
call also update the per-cpu structure instead of returning the previously
cached values.

Suggested-by: Thomas Gleixner <>
Signed-off-by: Ashok Raj <>
Cc: LKML <>
Cc: x86 <>
Cc: Ingo Molnar <>
Cc: Tony Luck <>
Cc: Dave Hansen <>
Cc: Alison Schofield <>
Cc: Reinette Chatre <>
Cc: Thomas Gleixner (Intel) <>
Cc: Tom Lendacky <>
Cc: Stefan Talpalaru <>
Cc: David Woodhouse <>
Cc: Benjamin Herrenschmidt <>
Cc: Jonathan Corbet <>
Cc: Rafael J. Wysocki <>
Cc: Peter Zilstra (Intel) <>
Cc: Andy Lutomirski <>
Cc: Andrew Cooper <>
Cc: Boris Ostrovsky <>
Cc: Martin Pohlack <>
arch/x86/kernel/cpu/microcode/core.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kernel/cpu/microcode/core.c b/arch/x86/kernel/cpu/microcode/core.c
index 6ade3d59c404..089636b1643f 100644
--- a/arch/x86/kernel/cpu/microcode/core.c
+++ b/arch/x86/kernel/cpu/microcode/core.c
@@ -386,6 +386,7 @@ static int __wait_for_cpus(atomic_t *t, long long timeout)
static int __reload_late(void *info)
+ struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
int cpu = smp_processor_id();
enum ucode_state err;
int ret = 0;
@@ -422,12 +423,11 @@ static int __reload_late(void *info)

* At least one thread has completed update on each core.
- * For others, simply call the update to make sure the
- * per-cpu cpuinfo can be updated with right microcode
- * revision.
+ * For siblings, collect the cpuinfo and update the
+ * per-cpu cpuinfo with the current microcode revision.
if (cpumask_first(topology_sibling_cpumask(cpu)) != cpu)
- err = microcode_ops->apply_microcode(cpu);
+ microcode_ops->collect_cpu_info(cpu, &uci->cpu_sig);

return ret;
 \ /
  Last update: 2023-03-26 23:51    [W:0.126 / U:0.224 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site