Kiyoshi Ueda wrote:
> This patch converts xsysace to use blk_end_request interfaces.
> Related 'uptodate' arguments are converted to 'error'.
>
> xsysace is a little bit different from "normal" drivers.
> xsysace driver has a state machine in it.
> It calls end_that_request_first() and end_that_request_last()
> from different states. (ACE_FSM_STATE_REQ_TRANSFER and
> ACE_FSM_STATE_REQ_COMPLETE, respectively.)
>
> However, those states are consecutive and without any interruption
> inbetween.
> So we can just follow the standard conversion rule (b) mentioned in
> the patch subject "[PATCH 01/30] blk_end_request: add new request
> completion interface".
>
> Cc: Grant Likely <grant.likely@secretlab.ca>
> Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
> Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
> ---
> drivers/block/xsysace.c | 5 +----
> 1 files changed, 1 insertion(+), 4 deletions(-)
>
> Index: 2.6.24-rc4/drivers/block/xsysace.c
> ================================================== =================
> --- 2.6.24-rc4.orig/drivers/block/xsysace.c
> +++ 2.6.24-rc4/drivers/block/xsysace.c
> @@ -703,7 +703,7 @@ static void ace_fsm_dostate(struct ace_d
>
> /* bio finished; is there another one? */
> i = ace->req->current_nr_sectors;
> - if (end_that_request_first(ace->req, 1, i)) {
> + if (__blk_end_request(ace->req, 0, i)) {
end_that_request_first() took sectors __blk_end_request() now takes
bytes
> /* dev_dbg(ace->dev, "next block; h=%li c=%i
",
> * ace->req->hard_nr_sectors,
> * ace->req->current_nr_sectors);
> @@ -718,9 +718,6 @@ static void ace_fsm_dostate(struct ace_d
> break;
>
> case ACE_FSM_STATE_REQ_COMPLETE:
> - /* Complete the block request */
> - blkdev_dequeue_request(ace->req);
> - end_that_request_last(ace->req, 1);
> ace->req = NULL;
>
> /* Finished request; go to idle state */
> -
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
Boaz
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
12-12-2007, 04:06 PM
Kiyoshi Ueda
blk_end_request: changing xsysace (take 4)
Hi,
On Wed, 12 Dec 2007 11:09:12 +0200, Boaz Harrosh <bharrosh@panasas.com> wrote:
> > Index: 2.6.24-rc4/drivers/block/xsysace.c
> > ================================================== =================
> > --- 2.6.24-rc4.orig/drivers/block/xsysace.c
> > +++ 2.6.24-rc4/drivers/block/xsysace.c
> > @@ -703,7 +703,7 @@ static void ace_fsm_dostate(struct ace_d
> >
> > /* bio finished; is there another one? */
> > i = ace->req->current_nr_sectors;
> > - if (end_that_request_first(ace->req, 1, i)) {
> > + if (__blk_end_request(ace->req, 0, i)) {
>
> end_that_request_first() took sectors __blk_end_request() now takes
> bytes
Thank you for pointing it out! And I'm very sorry for the bug.
I have checked all conversions between sectors and bytes through
all patches again, and I found no other miss conversions.
This patch converts xsysace to use blk_end_request interfaces.
Related 'uptodate' arguments are converted to 'error'.
xsysace is a little bit different from "normal" drivers.
xsysace driver has a state machine in it.
It calls end_that_request_first() and end_that_request_last()
from different states. (ACE_FSM_STATE_REQ_TRANSFER and
ACE_FSM_STATE_REQ_COMPLETE, respectively.)
However, those states are consecutive and without any interruption
inbetween.
So we can just follow the standard conversion rule (b) mentioned in
the patch subject "[PATCH 01/30] blk_end_request: add new request
completion interface".
In ace_fsm_dostate(), the variable 'i' was used only for passing
sector size of the request to end_that_request_first().
So I removed it and changed the code to pass the size in bytes
directly to __blk_end_request().
/* Transfer the next buffer */
- i = 16;
if (ace->fsm_task == ACE_TASK_WRITE)
ace->reg_ops->dataout(ace);
else
@@ -702,8 +700,8 @@ static void ace_fsm_dostate(struct ace_d
}
/* bio finished; is there another one? */
- i = ace->req->current_nr_sectors;
- if (end_that_request_first(ace->req, 1, i)) {
+ if (__blk_end_request(ace->req, 0,
+ blk_rq_cur_bytes(ace->req))) {
/* dev_dbg(ace->dev, "next block; h=%li c=%i
",
* ace->req->hard_nr_sectors,
* ace->req->current_nr_sectors);
@@ -718,9 +716,6 @@ static void ace_fsm_dostate(struct ace_d
break;
case ACE_FSM_STATE_REQ_COMPLETE:
- /* Complete the block request */
- blkdev_dequeue_request(ace->req);
- end_that_request_last(ace->req, 1);
ace->req = NULL;
/* Finished request; go to idle state */
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
12-12-2007, 05:12 PM
Boaz Harrosh
blk_end_request: changing xsysace (take 4)
On Wed, Dec 12 2007 at 19:06 +0200, Kiyoshi Ueda <k-ueda@ct.jp.nec.com> wrote:
> Hi,
>
> On Wed, 12 Dec 2007 11:09:12 +0200, Boaz Harrosh <bharrosh@panasas.com> wrote:
>>> Index: 2.6.24-rc4/drivers/block/xsysace.c
>>> ================================================== =================
>>> --- 2.6.24-rc4.orig/drivers/block/xsysace.c
>>> +++ 2.6.24-rc4/drivers/block/xsysace.c
>>> @@ -703,7 +703,7 @@ static void ace_fsm_dostate(struct ace_d
>>>
>>> /* bio finished; is there another one? */
>>> i = ace->req->current_nr_sectors;
>>> - if (end_that_request_first(ace->req, 1, i)) {
>>> + if (__blk_end_request(ace->req, 0, i)) {
>> end_that_request_first() took sectors __blk_end_request() now takes
>> bytes
>
> Thank you for pointing it out! And I'm very sorry for the bug.
> I have checked all conversions between sectors and bytes through
> all patches again, and I found no other miss conversions.
>
> Below is the revised patch for xsysace.
>
> Thanks,
> Kiyoshi Ueda
>
>
NP, I know how it is, after you rebased your work for the
"who's counting" times, you become blind. You need fresh
eyes to look at it. Thanks for doing all this.
Boaz
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
blk_end_request: changing xsysace (take 4)
