atari_scsi: Fix atari_scsi deadlocks on Falcon
authorFinn Thain <fthain@telegraphics.com.au>
Wed, 12 Nov 2014 05:12:08 +0000 (16:12 +1100)
committerChristoph Hellwig <hch@lst.de>
Thu, 20 Nov 2014 08:11:13 +0000 (09:11 +0100)
commit16b29e75a78ae03250233468b68f7ae467d3dc7a
treea5ba9e016d5557d2a6f5fb60dc83ae906ae771af
parentcbad48deb38d8e442db9760ca1f950cd24429707
atari_scsi: Fix atari_scsi deadlocks on Falcon

Don't disable irqs when waiting for the ST DMA "lock"; its release may
require an interrupt.

Introduce stdma_try_lock() for use in soft irq context. atari_scsi now tells
the SCSI mid-layer to defer queueing a command if the ST DMA lock is not
available, as per Michael's patch:
http://marc.info/?l=linux-m68k&m=139095335824863&w=2

The falcon_got_lock variable is race prone: we can't disable IRQs while
waiting to acquire the lock, so after acquiring it there must be some
interval during which falcon_got_lock remains false. Introduce
stdma_is_locked_by() to replace falcon_got_lock.

The falcon_got_lock tests in the EH handlers are incorrect these days. It
can happen that an EH handler is called after a command completes normally.
Remove these checks along with falcon_got_lock.

Also remove the complicated and racy fairness wait queues. If fairness is an
issue (when SCSI competes with IDE for the ST DMA interrupt), the solution
is likely to be a lower value for host->can_queue.

Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Tested-by: Michael Schmitz <schmitzmic@gmail.com>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
arch/m68k/atari/stdma.c
arch/m68k/include/asm/atari_stdma.h
drivers/scsi/atari_NCR5380.c
drivers/scsi/atari_scsi.c