Re: Race condition in module load causing undefined symbols

[Pozsár Balázs]

On Sat, Jun 11, 2005 at 03:09:08PM -0500, Steve Lord wrote:
> Well, the bizarre part is that I think this has been around for a while, 
> but it
> does not exhibit itself in redhat kernels which postdate the earliest
> recolloction of me seeing it. 2.6.11-rc1 is the earliest I remember,
> but I am not religious about updating the kernel on this box so
> my samples are spotty.
> 
> The difference between the two may be that I recompile for a P4
> while redhat uses a lowest common denominator cpu type.
> 
> If I get a chance this weekend I will try some other kernels and
> report back. Maybe just start out by dumbing down my cpu
> type.

I always used CONFIG_M586=y.

Did you turn preempt on? I did with the 2.6.12-rc, but not with the 
(working) 2.6.9, so this migth be the difference.
Sorry, I do not have currently for testing.


-- 
pozsy

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/