[PATCH] ARM: ftrace: Ensure code modifications are synchronised across all cpus

[Russell King - ARM Linux]

On Mon, Dec 10, 2012 at 09:06:05AM -0500, Steven Rostedt wrote:
> On Mon, 2012-12-10 at 13:57 +0000, Russell King - ARM Linux wrote:
> > On Mon, Dec 10, 2012 at 08:02:17AM -0500, Steven Rostedt wrote:
> > > On Mon, 2012-12-10 at 10:04 +0000, Will Deacon wrote:
> > > > Yes, and I think if you do use two 16-bit nops, you can even get rid of all
> > > > the intermediate `sync' operations (I guess you might want one at the end if
> > > > you want the call to become visible at a particular point).
> > > 
> > > Wont work. We are replacing a 32bit call with a nop. That nop must also
> > > be 32bits, because we could eventually replace the nop(s) with a 32bit
> > > call.
> > 
> > ... which, if it's misaligned to a 32-bit boundary, which can happen with
> > Thumb-2 code, will require the replacement to be done atomically; you will
> > need to use stop_machine() to ensure that other CPUs don't try to execute
> > the instruction mid-way through modification... as I have already
> > explained in my previous mails.
> 
> If there's no way to modify a 32bit operation without stop_machine(),
> ever with a breakpoint, than we can stop the discussion here. ARM will
> forever require stop_machine() for use with tracepoints and ftrace. Too
> bad, as ARM was the x86 competitor. Here's something that x86 has a one
> up on ARM.

You think that kind of blackmail makes a difference?  Look closely at what
I've written - I didn't say that there's no way to modify any 32-bit
operation without stop_machine().