Re: [BUG] 2.5.73 Scheduling while atomic with taskfile IO and high

Mel Gorman (mel@csn.ul.ie)
Tue, 1 Jul 2003 01:02:57 +0100 (IST)


On Mon, 30 Jun 2003, Zwane Mwaikambo wrote:

> Could you try selecting your specific IDE chipset (or all), it doesn't
> look like PIO is getting along famously with various other bits. I also
> noticed TCQ, do you have any TCQ capable IDE devices?
>

Tried many things, in no particular order

o PCI chipsets disabled
o Disabled TCQ (device doesn't support it anyway)
o DMA disabled and ide_setup_dma() stub function added so it'll compile
o PCI Generic chipset support enabled
o Intel PIIX chipset support enabled
o All chipset under the sun supported

All come up with the same errors. The following workarounds let it
boot

o Removing inc_preempt_count() and dec_preempt_count() from kmap_atomic()
and kunmap_atomic()
o Disabling high memory
o Disabling taskfile IO

A quickie patch to sched.c shows that preempt_count() keeps incrementing
for each time the sleeping while atomic message is printed by the
cpu_idle() thread

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