Re: [patch 1/2] oProfile: oops when profile_pc() return ~0LU

[Linus Torvalds <torvalds@linux-foundation.org>]

On Tue, 23 Oct 2007, Philippe Elie wrote:
> 
> For the signed-offs I thought the From: was an implicit Signed-offs.

No, there are no implicit sign-offs. The point of sign-offs is that it 
makes the copyright and flow of patches explicit, so an "implicit 
sign-off" would defeat the whole point.

> Test was done privately, Sami helped to narrow down the trouble, but
> he didn't test the last patch, nothing bad on Sami side, I was too
> confident the fix was obvious after narrowing it.

Well, I just wanted an ack that it was tested, since it seemed a bit 
subtle and (like the sign-offs) that explicit "yes, this was tested" was 
missing. Looks like I was right in asking for it:

> > The previous patch I tested by Philippe, oprof-fix-profile_pc-use.patch,
> > worked ok, but with this latest patch oprofiled aborts.
> > But kernel does not oops or print msgs.
> 
> argh, I just moved the wrong eip from kernel to user space where the same
> problem occur too, *sighs*, since I can't reproduce Sami problem, my own
> test obviously worked...
> 
> Sami, can you test this new patch. After testing can you report
> the contents of /dev/oprofile/stats/cpu*/sample_invalid_eip ?
> 
> Linus, there is two way to fix this problem, the attached patch fix it
> by sanitizing the sampled eip,

I'm perfectly happy with the attached patch, I just want it to be 
properly tested and have all the sign-offs (and explanations etc) in 
place, and I can apply it.

Of course, I cannot think of a single architecture where an EIP of ~0UL is 
valid anyway, so I'm also not opposed to just keeping ~0UL as the magic 
value.

		Linus