On Fri, 2021-08-20 at 01:26 +0000, HAGIO KAZUHITO(萩尾 一仁) wrote:
> -----Original Message-----
> > Dear Kazu,
> >
> > Sorry for late reply
> > Thanks for your suggestion and I have prepared a V2 patch, please
> > help
> > to review.
>
> ok, I've modified your v2 patch to fix the following compilation
> warning
> and rewrite the commit log, and attached it.
>
> arm64.c: In function ‘arm64_init’:
> arm64.c:3728:35: warning: ‘note’ may be used uninitialized in this
> function [-Wmaybe-uninitialized]
This warning can be eliminated as follows:
--- a/arm64.c
+++ b/arm64.c
@@ -3698,14 +3698,48 @@ arm64_get_crash_notes(void)
{
struct machine_specific *ms = machdep->machspec;
ulong crash_notes;
- Elf64_Nhdr *note;
+ Elf64_Nhdr *note = NULL;
ulong offset;
+++ b/arm64.c
@@ -3698,14 +3698,48 @@ arm64_get_crash_notes(void)
{
struct machine_specific *ms = machdep->machspec;
ulong crash_notes;
- Elf64_Nhdr *note;
+ Elf64_Nhdr *note = NULL;
ulong offset;
But I didn't see this correction in V2. Can you help to add the initialization of the note variable Kazu. Otherwise:
Acked-by: Lianbo Jiang <lijiang@xxxxxxxxxx>
BTW: the patch format is still unfriendly to me.
Thanks
Lianbo
> offset = roundup(offset + note->n_namesz, 4);
> ^~
> Kazu, thanks for your kind help.
> Lianbo, could you review the attached patch?
>
> Thanks,
> Kazu
-- Crash-utility mailing list Crash-utility@xxxxxxxxxx https://listman.redhat.com/mailman/listinfo/crash-utility