Re: [GIT PULL] ftrace: Fixes for v6.13

[Steven Rostedt]

[
  I'm back from a lovely day with my wife and friends visiting the
  Finger Lake wineries which I would love to share with the Linux
  Plumbers attendees if we ever get to have Plumbers at the Ithaca
  Conference center. ;-)
]

On Sun, 15 Dec 2024 16:40:34 -0500
"Theodore Ts'o" <tytso@xxxxxxx> wrote:

> I'm not convinced that it's worth it in this particular case, so I
> think I side with Linus here; maybe all of this hackery isn't worth
> it?  Steven, what am I missing?  Why did we go down this particular
> path in the first place?  I assume there must have been something that
> seemed like a good reason at the time?

It has nothing to do with performance. It has only to do with debugging
and catching the issue where developers use "%s" in the TP_printk() of
TRACE_EVENT() for a string that can be freed after the trace event was
executed but before it is read from the ring buffer.

I stated in another email that I didn't like this solution when I wrote
it, but it was the only solution I could think of to catch this common
bug. But I also stated in another email that I think there's another
solution that doesn't deal with va_list implementations and should be
acceptable by Linus.

I'm not against ripping out the code. I agree with Linus that it's a
hack, but the hack that caught the bugs early seemed to me worth it at
the time.

-- Steve