https://bugs.exim.org/show_bug.cgi?id=2509
Zoltan Herczeg <hzmester@???> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |hzmester@???
--- Comment #1 from Zoltan Herczeg <hzmester@???> ---
Thank you Petr for forwarding this issue.
After a quick glance, I would say this is not exactly sljit related, since this
is something high level. I mean "possible jump target addresses" is only known
by pcre-jit and not by sljit. They introduced many new instructions, and they
seem extremely complex, but I suspect we need to use only a few of them since
most of them are OS specific.
It would be great if an expert from Intel would explain what shall we exactly
do. For example, do we need to do anything with the "shadow stack" or it only
affects programs which use "longjmp" like techniques? Do we need to emit a
"ENDBR64" before every valid indirect jump target? Do we need to do anything
else?
--
You are receiving this mail because:
You are on the CC list for the bug.
