Re: [2.6] kernel BUG at arch/i386/mm/highmem.c:14!

[Jens Axboe]

On Tue, Aug 26 2003, Roger Luethi wrote:
> This BUG immediately followed my turning off DMA on hda (with a mounted
> partition and dirty buffers). Register and stack info are available upon
> request (screenshot). 2.6.0-test4.
> 
> Roger
> 
> kernel BUG at arch/i386/mm/highmem.c:14!
> invalid operand: 0000 [#1]
> CPU:    0
> EIP:    0060:[<c011bc36>]   Not tainted
> EFLAGS: 00010206
> EIP is at kunmap+0x16/0x30
> [...]
> Process pdflush [...]
> 
>  [<c0158f69>] __blk_queue_bounce+0x1e9/0x230
>  [<c0158fe7>] blk_queue_bounce+0x37/0x60

Funky, it is indeed strictly forbidden to enter this path with
interrupts disabled.

>  [<c022be20>] __make_request+0x50/0x670
>  [<c023ccb0>] task_mulout_intr+0x0/0x290
>  [<c022c584>] generic_make_request+0x144/0x1e0

Possibly some artifact of using read/write multiple on IDE, does it work
if you disable that with -m0 on the device? I'm guessing the error is
100% reproducible?

-- 
Jens Axboe

-
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/