[git patch] atang tree: fix 2.6.32 SSD performance regression

[git patch] atang tree: fix 2.6.32 SSD performance regression

[Bartlomiej Zolnierkiewicz]

DISCLAIMER: the fact of getting patches merged into atang tree means
that from now on they will be getting updates for changes happening
in atang tree and it should not be treated as an indication regarding
decisions taken by 'upstream' kernel trees


Fix 2.6.32 regression bringing back the desired SSD performance.

[ 'upstream' may want to wait for a proper resolution, for atang tree
  performance of big non-ATA boxes owned by a very few people is not
  of such a big concern as a major regression on a commodity hardware ]


The following changes since commit ded92e0d01fbb7f27bec7a35606847a2547e95e0:
  Bartlomiej Zolnierkiewicz (1):
        libata: enhanced SSD detection

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/bart/misc.git atang-v4.8

Bartlomiej Zolnierkiewicz (1):
      block: fix SSD performance regression

 block/blk-core.c       |   11 ++---------
 include/linux/blkdev.h |    2 --
 2 files changed, 2 insertions(+), 11 deletions(-)

commit cad2f21db052f21beafd44d3396a565d9a485073
Author: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date:   Mon Feb 22 22:19:14 2010 +0100

    block: fix SSD performance regression
    
    Revert "block: improve queue_should_plug() by looking at IO depths".
    
    This reverts commit fb1e75389bd06fd5987e9cda1b4e0305c782f854 which was
    reported to negatively affect performance for SSD devices in 2.6.32:
    
    	http://article.gmane.org/gmane.linux.ide/45053
    
    "Without that patch my SSD (Super Talent Ultradrive GX MLC 64GB)
     reaches about 200MB/s sequentiell read. After applying the patch it
     reaches only 70MB/s."
    
    Reported-by: "Benjamin S." <sbenni@gmx.de>
    Cc: Jens Axboe <jens.axboe@oracle.com>
    Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

diff --git a/block/blk-core.c b/block/blk-core.c
index 71da511..f8f01f2 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1147,7 +1147,7 @@ void init_request_from_bio(struct request *req, struct bio *bio)
  */
 static inline bool queue_should_plug(struct request_queue *q)
 {
-	return !(blk_queue_nonrot(q) && blk_queue_queuing(q));
+	return !(blk_queue_nonrot(q) && blk_queue_tagged(q));
 }
 
 static int __make_request(struct request_queue *q, struct bio *bio)
@@ -1859,15 +1859,8 @@ void blk_dequeue_request(struct request *rq)
 	 * and to it is freed is accounted as io that is in progress at
 	 * the driver side.
 	 */
-	if (blk_account_rq(rq)) {
+	if (blk_account_rq(rq))
 		q->in_flight[rq_is_sync(rq)]++;
-		/*
-		 * Mark this device as supporting hardware queuing, if
-		 * we have more IOs in flight than 4.
-		 */
-		if (!blk_queue_queuing(q) && queue_in_flight(q) > 4)
-			set_bit(QUEUE_FLAG_CQ, &q->queue_flags);
-	}
 }
 
 /**
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 221cecd..7db4f99 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -457,7 +457,6 @@ struct request_queue
 #define QUEUE_FLAG_NONROT      14	/* non-rotational device (SSD) */
 #define QUEUE_FLAG_VIRT        QUEUE_FLAG_NONROT /* paravirt device */
 #define QUEUE_FLAG_IO_STAT     15	/* do IO stats */
-#define QUEUE_FLAG_CQ	       16	/* hardware does queuing */
 #define QUEUE_FLAG_DISCARD     17	/* supports DISCARD */
 
 #define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
@@ -582,7 +581,6 @@ enum {
 
 #define blk_queue_plugged(q)	test_bit(QUEUE_FLAG_PLUGGED, &(q)->queue_flags)
 #define blk_queue_tagged(q)	test_bit(QUEUE_FLAG_QUEUED, &(q)->queue_flags)
-#define blk_queue_queuing(q)	test_bit(QUEUE_FLAG_CQ, &(q)->queue_flags)
 #define blk_queue_stopped(q)	test_bit(QUEUE_FLAG_STOPPED, &(q)->queue_flags)
 #define blk_queue_nomerges(q)	test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags)
 #define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)

Re: [git patch] atang tree: fix 2.6.32 SSD performance regression

[Jens Axboe]

On Mon, Feb 22 2010, Bartlomiej Zolnierkiewicz wrote:
> DISCLAIMER: the fact of getting patches merged into atang tree means
> that from now on they will be getting updates for changes happening
> in atang tree and it should not be treated as an indication regarding
> decisions taken by 'upstream' kernel trees
> 
> 
> Fix 2.6.32 regression bringing back the desired SSD performance.
> 
> [ 'upstream' may want to wait for a proper resolution, for atang tree
>   performance of big non-ATA boxes owned by a very few people is not
>   of such a big concern as a major regression on a commodity hardware ]

What is the point of this, other than continuing your past history of
political tirades?

-- 
Jens Axboe

Re: [git patch] atang tree: fix 2.6.32 SSD performance regression

[Bartlomiej Zolnierkiewicz]

On Tuesday 23 February 2010 07:23:24 am Jens Axboe wrote:
> On Mon, Feb 22 2010, Bartlomiej Zolnierkiewicz wrote:
> > DISCLAIMER: the fact of getting patches merged into atang tree means
> > that from now on they will be getting updates for changes happening
> > in atang tree and it should not be treated as an indication regarding
> > decisions taken by 'upstream' kernel trees
> > 
> > 
> > Fix 2.6.32 regression bringing back the desired SSD performance.
> > 
> > [ 'upstream' may want to wait for a proper resolution, for atang tree
> >   performance of big non-ATA boxes owned by a very few people is not
> >   of such a big concern as a major regression on a commodity hardware ]
> 
> What is the point of this, other than continuing your past history of
> political tirades?

I worry that you may be reading too much into it -- I don't use Linus'
tree or linux-next tree any longer so the next kernel I would see fix
for the issue is most likely 2.6.34 and there is a little point from my
personal perspective in waiting for such long time..

--
Bartlomiej Zolnierkiewicz

Re: [git patch] atang tree: fix 2.6.32 SSD performance regression

[Jens Axboe]

On Tue, Feb 23 2010, Bartlomiej Zolnierkiewicz wrote:
> On Tuesday 23 February 2010 07:23:24 am Jens Axboe wrote:
> > On Mon, Feb 22 2010, Bartlomiej Zolnierkiewicz wrote:
> > > DISCLAIMER: the fact of getting patches merged into atang tree means
> > > that from now on they will be getting updates for changes happening
> > > in atang tree and it should not be treated as an indication regarding
> > > decisions taken by 'upstream' kernel trees
> > > 
> > > 
> > > Fix 2.6.32 regression bringing back the desired SSD performance.
> > > 
> > > [ 'upstream' may want to wait for a proper resolution, for atang tree
> > >   performance of big non-ATA boxes owned by a very few people is not
> > >   of such a big concern as a major regression on a commodity hardware ]
> > 
> > What is the point of this, other than continuing your past history of
> > political tirades?
> 
> I worry that you may be reading too much into it -- I don't use Linus'
> tree or linux-next tree any longer so the next kernel I would see fix
> for the issue is most likely 2.6.34 and there is a little point from my
> personal perspective in waiting for such long time..

It'll go into 2.6.33, I already prepared an updated for-linus branch
this morning and will send it out later today.

-- 
Jens Axboe

Re: [git patch] atang tree: fix 2.6.32 SSD performance regression

[Bartlomiej Zolnierkiewicz]

On Tuesday 23 February 2010 01:23:20 pm Jens Axboe wrote:
> On Tue, Feb 23 2010, Bartlomiej Zolnierkiewicz wrote:
> > On Tuesday 23 February 2010 07:23:24 am Jens Axboe wrote:
> > > On Mon, Feb 22 2010, Bartlomiej Zolnierkiewicz wrote:
> > > > DISCLAIMER: the fact of getting patches merged into atang tree means
> > > > that from now on they will be getting updates for changes happening
> > > > in atang tree and it should not be treated as an indication regarding
> > > > decisions taken by 'upstream' kernel trees
> > > > 
> > > > 
> > > > Fix 2.6.32 regression bringing back the desired SSD performance.
> > > > 
> > > > [ 'upstream' may want to wait for a proper resolution, for atang tree
> > > >   performance of big non-ATA boxes owned by a very few people is not
> > > >   of such a big concern as a major regression on a commodity hardware ]
> > > 
> > > What is the point of this, other than continuing your past history of
> > > political tirades?
> > 
> > I worry that you may be reading too much into it -- I don't use Linus'
> > tree or linux-next tree any longer so the next kernel I would see fix
> > for the issue is most likely 2.6.34 and there is a little point from my
> > personal perspective in waiting for such long time..
> 
> It'll go into 2.6.33, I already prepared an updated for-linus branch
> this morning and will send it out later today.

Great!  I'll just drop my patch during re-base to 2.6.33 then..

--
Bartlomiej Zolnierkiewicz