Re: inconsistent kallsyms data [2.6.11-mm2]

From: Paulo Marques
Date: Mon Mar 14 2005 - 08:37:05 EST

Sam Ravnborg wrote:
On Thu, Mar 10, 2005 at 12:12:22PM +0000, Paulo Marques wrote:

Paulo Marques wrote:

A simple and robust way is to do the sampling on a list of symbols sorted by symbol name. This way, even if the symbol positions that are given to scripts/kallsyms change, the symbols sampled will be the same.

I'll do the patch to do this and send it ASAP.

Ok, here it is.

Dominik can you try the attached patch and see if it solves the problem?

Hi Paulo.

Hi Sam :)

Alexander Stohr had similar problems with down and __sched_text_start.

I figured out that what was causing the troubles was the fact that the
linker generated symbol __sched_text_start changed value from pass 1 to
pass 2. The reason for this was the alingment used within that section.

Damn, you're right. Looking more carefully at Dominik's files I can see that on the first pass we have:

T __sched_text_start PTR 0xc0420482
t __down PTR 0xc0420484

and on the second pass:

t __down PTR 0xc0420484
T __sched_text_start PTR 0xc0420484

I only looked at the addresses on the second pass and noticed they were aliased symbols and that the symbol order changed from the first pass :P

I never came around submitting this since I do not know what the correct
number for function alignment is on different paltforms.

If this will just align the beginning of a section, I don't think it will be a problem to always align at 8 bytes even on platforms that need only a 4 byte alignment.

So I think that your patch should definitely go in, as it solves a real problem.

As for my patch it could potentially solve problems that we don't currently have(*), so it is probably better to wait for them to appear before trying to solve an non-existent problem :)

Paulo Marques -

All that is necessary for the triumph of evil is that good men do nothing.
Edmund Burke (1729 - 1797)

(*) order of aliased symbols changing, or 'nm' returning non sorted addresses.
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at