Re: ohci1394_dma=early crash since 2.6.32 (was Re: [Bug #14487] PANIC:early exception 08 rip 246:10 error ffffffff810251b5 cr2 0)

From: Justin P. Mattock
Date: Tue Feb 02 2010 - 02:02:26 EST


On 02/01/10 22:55, Stefan Richter wrote:
Justin P. Mattock wrote:
As for anything changed in the kernel
(2.6.31 - present), tough to say
from what I remember I had created a new fresh
lfs system using these CFLAGS:

CFLAGS="-mtune=core2 -march=core2 -O2 -pipe -fomit-frame-pointer"
CXXFLAGS="${CFLAGS}" MAKEOPTS="{-j3}"
(without -m option gcc defaults(I think)to -m32).

which booted with ohci1394_dma=early just fine.

then decided to build another lfs system with the same CFLAGS except
added -m64 (pure64) to the build process.
(then this showed up).

What I can try is do a git revert to 2.6.29/27 to see if this thing
fires off(before going any further). if the system boots then do a bisect.

Do I understand correctly that at this moment it is only known that the
bug could be
- *either* a 2.6.31 -> 2.6.32 regression
- *or* an x86-64 specific bug that does not occur on x86-32,
right?


at first I was under the impression this was an arch thing because of building an x86_32, and then building x86_64(and hitting this). but now after reverting to 2.6.27 I'm thinking other wise.(my bad, should of done this at first but didn't even think too);

I have an Core 2 Duo based PC with x86-32 kernel and userland and an AMD
based x86-64 PC and could give ohci1394_dma=early a try on both (never
tested it myself before). I could furthermore attempt to build and
install an x86-64 kernel on the Core 2 Duo PC but I am afraid I am far
too short of spare time for that.

no..
I need to do a bisect from 2.6.27 to present to see
(just need to crash for a few hrs, then can start);
then I'll go from there.

Justin P. Mattock
--
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/