[PATCH v2] recordmcount: Correct st_shndx handling

[Mark-PK Tsai]

> On Tue, 15 Jun 2021 14:47:20 +0800
> Mark-PK Tsai <mark-pk.tsai at mediatek.com> wrote:
> 
> > One should only use st_shndx when >SHN_UNDEF and <SHN_LORESERVE. When
> > SHN_XINDEX, then use .symtab_shndx. Otherwise use 0.
> > 
> > This handles the case: st_shndx >= SHN_LORESERVE && st_shndx != SHN_XINDEX.
> > 
> > Reported-by: Mark-PK Tsai <mark-pk.tsai at mediatek.com>
> > Signed-off-by: Peter Zijlstra (Intel) <peterz at infradead.org>
> > Tested-by: Mark-PK Tsai <mark-pk.tsai at mediatek.com>
> > Signed-off-by: Mark-PK Tsai <mark-pk.tsai at mediatek.com>
> 
> Please explain the two signed-off-by's above. If you are just tweaking
> Peter's original patch, please add at the start:
> 
> From: Peter Zijlstra (Intel) <peterz at infradead.org>
> 
> And then just above your signed off by, add what you changed:
> 
> Tested-by: Mark-PK Tsai <mark-pk.tsai at mediatek.com>
> [ Changed something ]
> Signed-off-by: Mark-PK Tsai <mark-pk.tsai at mediatek.com>
> 
> But state what you changed.
> 
> Thanks!
> 
> -- Steve

Sorry for messing up.
I've fixed it in v3.

Thanks for your comment.