Re: 2.6.18-rc6-mm1: GPF loop on early boot

[Ingo Molnar]

btw., i dont think it's the early-bootup behavior of lockdep that causes 
your oops. I think %gs somehow ended up being invalid upon pagefault 
entry, well after bootup. Not sure why. Andi's patch just papers over 
this i believe.

ah ... found it meanwhile: in the syscall exit path the PDA patches 
restore the userspace %gs too early, trace_hardirqs_on() is called 
_after_ the restoring of %gs. Will send a patch in a few minutes.

	Ingo
-
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/