Re: [BUG] 2.6.0-test3 and cciss driver (or blk_queue_hardsect_size)

[Francois Romieu <romieu@fr.zoreil.com>]

Greetings,

HABBINGA,ERIK (HP-Loveland,ex1) <erik.habbinga@hp.com> :
> I'm wondering if anyone else is having problems with 2.6.0-test3 and the
> cciss driver, or with the function blk_queue_hardsect_size.  I was able to
> successfully boot 2.6.0-test2 in previous weeks, but trying 2.6.0-test3
> today gave me:


Jens, does the following patch make sense ?

hba[i]->queue went from 'struct request_queue queue' to
'struct request_queue *queue' and it now needs to be explicitly set.


 drivers/block/cciss.c |    1 +
 1 files changed, 1 insertion(+)

diff -puN drivers/block/cciss.c~oops-cciss drivers/block/cciss.c
--- linux-2.6.0-test3/drivers/block/cciss.c~oops-cciss	Mon Aug 11 21:53:11 2003
+++ linux-2.6.0-test3-fr/drivers/block/cciss.c	Mon Aug 11 22:00:55 2003
@@ -2537,6 +2537,7 @@ err_all:
 	cciss_procinit(i);
 
         q->queuedata = hba[i];
+	hba[i]->queue = q;
 	blk_queue_bounce_limit(q, hba[i]->pdev->dma_mask);
 
 	/* This is a hardware imposed limit. */

_