Re: Oops on 2.5.23 and IDE

[Jens Axboe]

On Wed, Jun 19 2002, Zwane Mwaikambo wrote:
> Hi Martin,
> 	IDE seems to be doing odd things.
> 
> >>EIP; c0117318 <schedule+28/4f0>   <=====
> Trace; c0117c6a <wait_for_completion+11a/1d0>
> Trace; c0117820 <default_wake_function+0/40>
> Trace; c0117820 <default_wake_function+0/40>
> Trace; c022e1ae <ide_do_drive_cmd+18e/1b0>
> Trace; c01bc082 <vsnprintf+2a2/420>
> Trace; c022e439 <ide_raw_taskfile+59/60>
> Trace; c02302f4 <do_recalibrate+54/70>
> Trace; c022e1d0 <special_intr+0/210>
> Trace; c0234e75 <ide_dma_intr+115/120>
> Trace; c0231138 <ata_irq_request+148/230>
> Trace; c0234d60 <ide_dma_intr+0/120>
> 
> 
> int ide_do_drive_cmd()
> {
> [...]
> 	if (action == ide_wait) {
>                 wait_for_completion(&wait);     /* wait for it to be serviced */
>                 return rq->errors ? -EIO : 0;   /* return -EIO if errors */
>         }
> [...]
> }
> 
> Was that function really designed to be called from interrupt context?

With action other than ide_wait sure, using ide_wait of course not.

-- 
Jens Axboe