Currently, kexec_image_info() is called during load time, and
right before kernel is being kexec'ed. There is no need to do both.
So, call it only once when segments are loaded and the physical
location of page with copy of arm64_relocate_new_kernel is known.
Signed-off-by: Pavel Tatashin <[email protected]>
---
arch/arm64/kernel/machine_kexec.c | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/arch/arm64/kernel/machine_kexec.c b/arch/arm64/kernel/machine_kexec.c
index ec71a153cc2d..cee3be586384 100644
--- a/arch/arm64/kernel/machine_kexec.c
+++ b/arch/arm64/kernel/machine_kexec.c
@@ -66,6 +66,7 @@ int machine_kexec_post_load(struct kimage *kimage)
memcpy(reloc_code, arm64_relocate_new_kernel,
arm64_relocate_new_kernel_size);
kimage->arch.kern_reloc = __pa(reloc_code);
+ kexec_image_info(kimage);
return 0;
}
@@ -79,8 +80,6 @@ int machine_kexec_post_load(struct kimage *kimage)
*/
int machine_kexec_prepare(struct kimage *kimage)
{
- kexec_image_info(kimage);
-
if (kimage->type != KEXEC_TYPE_CRASH && cpus_are_stuck_in_kernel()) {
pr_err("Can't kexec: CPUs are stuck in the kernel.\n");
return -EBUSY;
@@ -166,8 +165,6 @@ void machine_kexec(struct kimage *kimage)
WARN(in_kexec_crash && (stuck_cpus || smp_crash_stop_failed()),
"Some CPUs may be stale, kdump will be unreliable.\n");
- kexec_image_info(kimage);
-
/* Flush the reboot_code_buffer in preparation for its execution. */
__flush_dcache_area(reboot_code_buffer, arm64_relocate_new_kernel_size);
--
2.17.1
Hi Pavel,
On 26/03/2020 03:24, Pavel Tatashin wrote:
> Currently, kexec_image_info() is called during load time, and
> right before kernel is being kexec'ed. There is no need to do both.
I think the original logic was if debugging, you'd see the load-time value in dmesg, and
the kexec-time values when the machine panic()d and you hadn't made a note of the previous
set. But I'm not sure why you'd have these debug messages enabled unless you were
debugging kexec.
> So, call it only once when segments are loaded and the physical
> location of page with copy of arm64_relocate_new_kernel is known.
Sure, keep the earlier version that is more likely to be seen.
Acked-by: James Morse <[email protected]>
Thanks,
James