Re: [PATCH 01/17] Fix "x86/alternatives: Lockdep-enforce text_mutex in text_poke*()"

On Jan 25, 2019, at 1:30 AM, Borislav Petkov [off-list ref] wrote:

On Wed, Jan 16, 2019 at 04:32:43PM -0800, Rick Edgecombe wrote:

quoted

From: Nadav Amit <redacted>

text_mutex is currently expected to be held before text_poke() is
called, but we kgdb does not take the mutex, and instead *supposedly*
ensures the lock is not taken and will not be acquired by any other core
while text_poke() is running.

The reason for the "supposedly" comment is that it is not entirely clear
that this would be the case if gdb_do_roundup is zero.

I guess that variable name is "kgdb_do_roundup” ?

Yes. Will fix.

quoted

This patch creates two wrapper functions, text_poke() and

Avoid having "This patch" or "This commit" in the commit message. It is
tautologically useless.

Also, do

$ git grep 'This patch' Documentation/process

for more details.

Ok.

quoted

+void *text_poke_kgdb(void *addr, const void *opcode, size_t len)

text_poke_unlocked() I guess. I don't think kgdb is that special that it
needs its own function flavor.

Tglx suggested this naming to prevent anyone from misusing text_poke_kdgb().
This is a very specific use-case that nobody else should need.

Regards,
Nadav