On (07/09/19 16:29), Petr Mladek wrote:
>
> Anyway, if nobody vetoes the patch, I would schedule it for 5.4.
> We are already in the merge window and it deserves some testing
> in linux-next.
>
I don't really want to veto it, but I'm not comfortable with fixing
off-by-one error by introducing another off-by-one error. The former
one leads to a missing message, the latter one to memory corruption.
Given that Vincent did find the code which does `buf[len] = 0x00'
[kudos to Vincent] let's have a better solution here. Otherwise, the
patch can make it to -stable via AUTOSEL machinery and then "we have
a problem".
Objections?
-ss
