* [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-18 2:28 ` Sharp.Xia
0 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-18 2:28 UTC (permalink / raw)
To: Ulf Hansson, Matthias Brugger, AngeloGioacchino Del Regno
Cc: wsd_upstream, Sharp Xia, linux-mmc, linux-kernel,
linux-arm-kernel, linux-mediatek
From: Sharp Xia <Sharp.Xia@mediatek.com>
MMC does not set readahead and uses the default VM_READAHEAD_PAGES
resulting in slower reading speed.
Use the max_req_size reported by host driver to set the optimal
I/O size to improve performance.
Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
---
drivers/mmc/core/queue.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
index b396e3900717..fc83c4917360 100644
--- a/drivers/mmc/core/queue.c
+++ b/drivers/mmc/core/queue.c
@@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card)
blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
blk_queue_max_hw_sectors(mq->queue,
min(host->max_blk_count, host->max_req_size / 512));
+ blk_queue_io_opt(mq->queue, host->max_req_size);
if (host->can_dma_map_merge)
WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
mmc_dev(host)),
--
2.18.0
^ permalink raw reply related [flat|nested] 26+ messages in thread
* [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-18 2:28 ` Sharp.Xia
0 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-18 2:28 UTC (permalink / raw)
To: Ulf Hansson, Matthias Brugger, AngeloGioacchino Del Regno
Cc: wsd_upstream, Sharp Xia, linux-mmc, linux-kernel,
linux-arm-kernel, linux-mediatek
From: Sharp Xia <Sharp.Xia@mediatek.com>
MMC does not set readahead and uses the default VM_READAHEAD_PAGES
resulting in slower reading speed.
Use the max_req_size reported by host driver to set the optimal
I/O size to improve performance.
Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
---
drivers/mmc/core/queue.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
index b396e3900717..fc83c4917360 100644
--- a/drivers/mmc/core/queue.c
+++ b/drivers/mmc/core/queue.c
@@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card)
blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
blk_queue_max_hw_sectors(mq->queue,
min(host->max_blk_count, host->max_req_size / 512));
+ blk_queue_io_opt(mq->queue, host->max_req_size);
if (host->can_dma_map_merge)
WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
mmc_dev(host)),
--
2.18.0
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-18 2:28 ` Sharp.Xia
@ 2023-08-24 10:55 ` Ulf Hansson
-1 siblings, 0 replies; 26+ messages in thread
From: Ulf Hansson @ 2023-08-24 10:55 UTC (permalink / raw)
To: Sharp.Xia
Cc: Matthias Brugger, AngeloGioacchino Del Regno, wsd_upstream,
linux-mmc, linux-kernel, linux-arm-kernel, linux-mediatek
On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
>
> From: Sharp Xia <Sharp.Xia@mediatek.com>
>
> MMC does not set readahead and uses the default VM_READAHEAD_PAGES
> resulting in slower reading speed.
> Use the max_req_size reported by host driver to set the optimal
> I/O size to improve performance.
This seems reasonable to me. However, it would be nice if you could
share some performance numbers too - comparing before and after
$subject patch.
Kind regards
Uffe
>
> Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> ---
> drivers/mmc/core/queue.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> index b396e3900717..fc83c4917360 100644
> --- a/drivers/mmc/core/queue.c
> +++ b/drivers/mmc/core/queue.c
> @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card)
> blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
> blk_queue_max_hw_sectors(mq->queue,
> min(host->max_blk_count, host->max_req_size / 512));
> + blk_queue_io_opt(mq->queue, host->max_req_size);
> if (host->can_dma_map_merge)
> WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
> mmc_dev(host)),
> --
> 2.18.0
>
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-24 10:55 ` Ulf Hansson
0 siblings, 0 replies; 26+ messages in thread
From: Ulf Hansson @ 2023-08-24 10:55 UTC (permalink / raw)
To: Sharp.Xia
Cc: Matthias Brugger, AngeloGioacchino Del Regno, wsd_upstream,
linux-mmc, linux-kernel, linux-arm-kernel, linux-mediatek
On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
>
> From: Sharp Xia <Sharp.Xia@mediatek.com>
>
> MMC does not set readahead and uses the default VM_READAHEAD_PAGES
> resulting in slower reading speed.
> Use the max_req_size reported by host driver to set the optimal
> I/O size to improve performance.
This seems reasonable to me. However, it would be nice if you could
share some performance numbers too - comparing before and after
$subject patch.
Kind regards
Uffe
>
> Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> ---
> drivers/mmc/core/queue.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> index b396e3900717..fc83c4917360 100644
> --- a/drivers/mmc/core/queue.c
> +++ b/drivers/mmc/core/queue.c
> @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card)
> blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
> blk_queue_max_hw_sectors(mq->queue,
> min(host->max_blk_count, host->max_req_size / 512));
> + blk_queue_io_opt(mq->queue, host->max_req_size);
> if (host->can_dma_map_merge)
> WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
> mmc_dev(host)),
> --
> 2.18.0
>
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-24 10:55 ` Ulf Hansson
@ 2023-08-25 7:10 ` Sharp Xia (夏宇彬)
-1 siblings, 0 replies; 26+ messages in thread
From: Sharp Xia (夏宇彬) @ 2023-08-25 7:10 UTC (permalink / raw)
To: ulf.hansson
Cc: linux-arm-kernel, matthias.bgg, wsd_upstream, linux-kernel,
angelogioacchino.delregno, linux-mediatek, linux-mmc
On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
>
> External email : Please do not click links or open attachments until
> you have verified the sender or the content.
> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
> >
> > From: Sharp Xia <Sharp.Xia@mediatek.com>
> >
> > MMC does not set readahead and uses the default VM_READAHEAD_PAGES
> > resulting in slower reading speed.
> > Use the max_req_size reported by host driver to set the optimal
> > I/O size to improve performance.
>
> This seems reasonable to me. However, it would be nice if you could
> share some performance numbers too - comparing before and after
> $subject patch.
>
> Kind regards
> Uffe
>
> >
> > Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> > ---
> > drivers/mmc/core/queue.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> > index b396e3900717..fc83c4917360 100644
> > --- a/drivers/mmc/core/queue.c
> > +++ b/drivers/mmc/core/queue.c
> > @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
> *mq, struct mmc_card *card)
> > blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
> > blk_queue_max_hw_sectors(mq->queue,
> > min(host->max_blk_count, host->max_req_size /
> 512));
> > + blk_queue_io_opt(mq->queue, host->max_req_size);
> > if (host->can_dma_map_merge)
> > WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
> > mmc_dev(hos
> t)),
> > --
> > 2.18.0
> >
I test this patch on internal platform(kernel-5.15).
Before:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
128
After:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
1024
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-25 7:10 ` Sharp Xia (夏宇彬)
0 siblings, 0 replies; 26+ messages in thread
From: Sharp Xia (夏宇彬) @ 2023-08-25 7:10 UTC (permalink / raw)
To: ulf.hansson
Cc: linux-arm-kernel, matthias.bgg, wsd_upstream, linux-kernel,
angelogioacchino.delregno, linux-mediatek, linux-mmc
On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
>
> External email : Please do not click links or open attachments until
> you have verified the sender or the content.
> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
> >
> > From: Sharp Xia <Sharp.Xia@mediatek.com>
> >
> > MMC does not set readahead and uses the default VM_READAHEAD_PAGES
> > resulting in slower reading speed.
> > Use the max_req_size reported by host driver to set the optimal
> > I/O size to improve performance.
>
> This seems reasonable to me. However, it would be nice if you could
> share some performance numbers too - comparing before and after
> $subject patch.
>
> Kind regards
> Uffe
>
> >
> > Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> > ---
> > drivers/mmc/core/queue.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> > index b396e3900717..fc83c4917360 100644
> > --- a/drivers/mmc/core/queue.c
> > +++ b/drivers/mmc/core/queue.c
> > @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
> *mq, struct mmc_card *card)
> > blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
> > blk_queue_max_hw_sectors(mq->queue,
> > min(host->max_blk_count, host->max_req_size /
> 512));
> > + blk_queue_io_opt(mq->queue, host->max_req_size);
> > if (host->can_dma_map_merge)
> > WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
> > mmc_dev(hos
> t)),
> > --
> > 2.18.0
> >
I test this patch on internal platform(kernel-5.15).
Before:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
128
After:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
1024
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-24 10:55 ` Ulf Hansson
@ 2023-08-25 7:25 ` Sharp.Xia
-1 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-25 7:25 UTC (permalink / raw)
To: ulf.hansson
Cc: Sharp.Xia, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
wsd_upstream
On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
>
> External email : Please do not click links or open attachments until
> you have verified the sender or the content.
> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
> >
> > From: Sharp Xia <Sharp.Xia@mediatek.com>
> >
> > MMC does not set readahead and uses the default VM_READAHEAD_PAGES
> > resulting in slower reading speed.
> > Use the max_req_size reported by host driver to set the optimal
> > I/O size to improve performance.
>
> This seems reasonable to me. However, it would be nice if you could
> share some performance numbers too - comparing before and after
> $subject patch.
>
> Kind regards
> Uffe
>
> >
> > Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> > ---
> > drivers/mmc/core/queue.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> > index b396e3900717..fc83c4917360 100644
> > --- a/drivers/mmc/core/queue.c
> > +++ b/drivers/mmc/core/queue.c
> > @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
> *mq, struct mmc_card *card)
> > blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
> > blk_queue_max_hw_sectors(mq->queue,
> > min(host->max_blk_count, host->max_req_size /
> 512));
> > + blk_queue_io_opt(mq->queue, host->max_req_size);
> > if (host->can_dma_map_merge)
> > WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
> > mmc_dev(hos
> t)),
> > --
> > 2.18.0
> >
I test this patch on internal platform(kernel-5.15).
Before:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
128
After:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
1024
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-25 7:25 ` Sharp.Xia
0 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-25 7:25 UTC (permalink / raw)
To: ulf.hansson
Cc: Sharp.Xia, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
wsd_upstream
On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
>
> External email : Please do not click links or open attachments until
> you have verified the sender or the content.
> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
> >
> > From: Sharp Xia <Sharp.Xia@mediatek.com>
> >
> > MMC does not set readahead and uses the default VM_READAHEAD_PAGES
> > resulting in slower reading speed.
> > Use the max_req_size reported by host driver to set the optimal
> > I/O size to improve performance.
>
> This seems reasonable to me. However, it would be nice if you could
> share some performance numbers too - comparing before and after
> $subject patch.
>
> Kind regards
> Uffe
>
> >
> > Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> > ---
> > drivers/mmc/core/queue.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> > index b396e3900717..fc83c4917360 100644
> > --- a/drivers/mmc/core/queue.c
> > +++ b/drivers/mmc/core/queue.c
> > @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
> *mq, struct mmc_card *card)
> > blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
> > blk_queue_max_hw_sectors(mq->queue,
> > min(host->max_blk_count, host->max_req_size /
> 512));
> > + blk_queue_io_opt(mq->queue, host->max_req_size);
> > if (host->can_dma_map_merge)
> > WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
> > mmc_dev(hos
> t)),
> > --
> > 2.18.0
> >
I test this patch on internal platform(kernel-5.15).
Before:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
128
After:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
1024
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-24 10:55 ` Ulf Hansson
@ 2023-08-25 7:26 ` Sharp.Xia
-1 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-25 7:26 UTC (permalink / raw)
To: ulf.hansson
Cc: Sharp.Xia, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
wsd_upstream
On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
>
> External email : Please do not click links or open attachments until
> you have verified the sender or the content.
> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
> >
> > From: Sharp Xia <Sharp.Xia@mediatek.com>
> >
> > MMC does not set readahead and uses the default VM_READAHEAD_PAGES
> > resulting in slower reading speed.
> > Use the max_req_size reported by host driver to set the optimal
> > I/O size to improve performance.
>
> This seems reasonable to me. However, it would be nice if you could
> share some performance numbers too - comparing before and after
> $subject patch.
>
> Kind regards
> Uffe
>
> >
> > Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> > ---
> > drivers/mmc/core/queue.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> > index b396e3900717..fc83c4917360 100644
> > --- a/drivers/mmc/core/queue.c
> > +++ b/drivers/mmc/core/queue.c
> > @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
> *mq, struct mmc_card *card)
> > blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
> > blk_queue_max_hw_sectors(mq->queue,
> > min(host->max_blk_count, host->max_req_size /
> 512));
> > + blk_queue_io_opt(mq->queue, host->max_req_size);
> > if (host->can_dma_map_merge)
> > WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
> > mmc_dev(hos
> t)),
> > --
> > 2.18.0
> >
I test this patch on internal platform(kernel-5.15).
Before:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
128
After:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
1024
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-25 7:26 ` Sharp.Xia
0 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-25 7:26 UTC (permalink / raw)
To: ulf.hansson
Cc: Sharp.Xia, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
wsd_upstream
On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
>
> External email : Please do not click links or open attachments until
> you have verified the sender or the content.
> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
> >
> > From: Sharp Xia <Sharp.Xia@mediatek.com>
> >
> > MMC does not set readahead and uses the default VM_READAHEAD_PAGES
> > resulting in slower reading speed.
> > Use the max_req_size reported by host driver to set the optimal
> > I/O size to improve performance.
>
> This seems reasonable to me. However, it would be nice if you could
> share some performance numbers too - comparing before and after
> $subject patch.
>
> Kind regards
> Uffe
>
> >
> > Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> > ---
> > drivers/mmc/core/queue.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> > index b396e3900717..fc83c4917360 100644
> > --- a/drivers/mmc/core/queue.c
> > +++ b/drivers/mmc/core/queue.c
> > @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
> *mq, struct mmc_card *card)
> > blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
> > blk_queue_max_hw_sectors(mq->queue,
> > min(host->max_blk_count, host->max_req_size /
> 512));
> > + blk_queue_io_opt(mq->queue, host->max_req_size);
> > if (host->can_dma_map_merge)
> > WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
> > mmc_dev(hos
> t)),
> > --
> > 2.18.0
> >
I test this patch on internal platform(kernel-5.15).
Before:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
128
After:
console:/ # echo 3 > /proc/sys/vm/drop_caches
console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
4485393+1 records in
4485393+1 records out
2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
1024
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-25 7:10 ` Sharp Xia (夏宇彬)
@ 2023-08-25 8:11 ` Shawn Lin
-1 siblings, 0 replies; 26+ messages in thread
From: Shawn Lin @ 2023-08-25 8:11 UTC (permalink / raw)
To: Sharp Xia (夏宇彬), ulf.hansson
Cc: shawn.lin, linux-arm-kernel, matthias.bgg, wsd_upstream,
linux-kernel, angelogioacchino.delregno, linux-mediatek,
linux-mmc
Hi Sharp,
On 2023/8/25 15:10, Sharp Xia (夏宇彬) wrote:
> On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
>>
>> External email : Please do not click links or open attachments until
>> you have verified the sender or the content.
>> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
>>>
>>> From: Sharp Xia <Sharp.Xia@mediatek.com>
>>>
>>> MMC does not set readahead and uses the default VM_READAHEAD_PAGES
>>> resulting in slower reading speed.
>>> Use the max_req_size reported by host driver to set the optimal
>>> I/O size to improve performance.
>>
>> This seems reasonable to me. However, it would be nice if you could
>> share some performance numbers too - comparing before and after
>> $subject patch.
>>
>> Kind regards
>> Uffe
>>
>>>
>>> Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
>>> ---
>>> drivers/mmc/core/queue.c | 1 +
>>> 1 file changed, 1 insertion(+)
>>>
>>> diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
>>> index b396e3900717..fc83c4917360 100644
>>> --- a/drivers/mmc/core/queue.c
>>> +++ b/drivers/mmc/core/queue.c
>>> @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
>> *mq, struct mmc_card *card)
>>> blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
>>> blk_queue_max_hw_sectors(mq->queue,
>>> min(host->max_blk_count, host->max_req_size /
>> 512));
>>> + blk_queue_io_opt(mq->queue, host->max_req_size);
>>> if (host->can_dma_map_merge)
>>> WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
>>> mmc_dev(hos
>> t)),
>>> --
>>> 2.18.0
>>>
>
> I test this patch on internal platform(kernel-5.15).
I patched this one and the test shows me a stable 11% performance drop.
Before:
echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
2048000+0 records in
2048000+0 records out
1048576000 bytes (0.9 G) copied, 3.912249 s, 256 M/s
After:
echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
2048000+0 records in
2048000+0 records out
1048576000 bytes (0.9 G) copied, 4.436271 s, 225 M/s
>
> Before:
> console:/ # echo 3 > /proc/sys/vm/drop_caches
> console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> 4485393+1 records in
> 4485393+1 records out
> 2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
> console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> 128
>
> After:
> console:/ # echo 3 > /proc/sys/vm/drop_caches
> console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> 4485393+1 records in
> 4485393+1 records out
> 2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
> console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> 1024
>
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-25 8:11 ` Shawn Lin
0 siblings, 0 replies; 26+ messages in thread
From: Shawn Lin @ 2023-08-25 8:11 UTC (permalink / raw)
To: Sharp Xia (夏宇彬), ulf.hansson
Cc: shawn.lin, linux-arm-kernel, matthias.bgg, wsd_upstream,
linux-kernel, angelogioacchino.delregno, linux-mediatek,
linux-mmc
Hi Sharp,
On 2023/8/25 15:10, Sharp Xia (夏宇彬) wrote:
> On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
>>
>> External email : Please do not click links or open attachments until
>> you have verified the sender or the content.
>> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
>>>
>>> From: Sharp Xia <Sharp.Xia@mediatek.com>
>>>
>>> MMC does not set readahead and uses the default VM_READAHEAD_PAGES
>>> resulting in slower reading speed.
>>> Use the max_req_size reported by host driver to set the optimal
>>> I/O size to improve performance.
>>
>> This seems reasonable to me. However, it would be nice if you could
>> share some performance numbers too - comparing before and after
>> $subject patch.
>>
>> Kind regards
>> Uffe
>>
>>>
>>> Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
>>> ---
>>> drivers/mmc/core/queue.c | 1 +
>>> 1 file changed, 1 insertion(+)
>>>
>>> diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
>>> index b396e3900717..fc83c4917360 100644
>>> --- a/drivers/mmc/core/queue.c
>>> +++ b/drivers/mmc/core/queue.c
>>> @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
>> *mq, struct mmc_card *card)
>>> blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_HIGH);
>>> blk_queue_max_hw_sectors(mq->queue,
>>> min(host->max_blk_count, host->max_req_size /
>> 512));
>>> + blk_queue_io_opt(mq->queue, host->max_req_size);
>>> if (host->can_dma_map_merge)
>>> WARN(!blk_queue_can_use_dma_map_merging(mq->queue,
>>> mmc_dev(hos
>> t)),
>>> --
>>> 2.18.0
>>>
>
> I test this patch on internal platform(kernel-5.15).
I patched this one and the test shows me a stable 11% performance drop.
Before:
echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
2048000+0 records in
2048000+0 records out
1048576000 bytes (0.9 G) copied, 3.912249 s, 256 M/s
After:
echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
2048000+0 records in
2048000+0 records out
1048576000 bytes (0.9 G) copied, 4.436271 s, 225 M/s
>
> Before:
> console:/ # echo 3 > /proc/sys/vm/drop_caches
> console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> 4485393+1 records in
> 4485393+1 records out
> 2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
> console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> 128
>
> After:
> console:/ # echo 3 > /proc/sys/vm/drop_caches
> console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> 4485393+1 records in
> 4485393+1 records out
> 2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
> console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> 1024
>
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-25 8:11 ` Shawn Lin
@ 2023-08-25 8:39 ` Sharp.Xia
-1 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-25 8:39 UTC (permalink / raw)
To: shawn.lin
Cc: Sharp.Xia, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
ulf.hansson, wsd_upstream
On Fri, 2023-08-25 at 16:11 +0800, Shawn Lin wrote:
>
> Hi Sharp,
>
> On 2023/8/25 15:10, Sharp Xia (夏宇彬) wrote:
> > On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
> >>
> >> External email : Please do not click links or open attachments
> until
> >> you have verified the sender or the content.
> >> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
> >>>
> >>> From: Sharp Xia <Sharp.Xia@mediatek.com>
> >>>
> >>> MMC does not set readahead and uses the default
> VM_READAHEAD_PAGES
> >>> resulting in slower reading speed.
> >>> Use the max_req_size reported by host driver to set the optimal
> >>> I/O size to improve performance.
> >>
> >> This seems reasonable to me. However, it would be nice if you
> could
> >> share some performance numbers too - comparing before and after
> >> $subject patch.
> >>
> >> Kind regards
> >> Uffe
> >>
> >>>
> >>> Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> >>> ---
> >>> drivers/mmc/core/queue.c | 1 +
> >>> 1 file changed, 1 insertion(+)
> >>>
> >>> diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> >>> index b396e3900717..fc83c4917360 100644
> >>> --- a/drivers/mmc/core/queue.c
> >>> +++ b/drivers/mmc/core/queue.c
> >>> @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
> >> *mq, struct mmc_card *card)
> >>> blk_queue_bounce_limit(mq->queue,
> BLK_BOUNCE_HIGH);
> >>> blk_queue_max_hw_sectors(mq->queue,
> >>> min(host->max_blk_count, host->max_req_size /
> >> 512));
> >>> + blk_queue_io_opt(mq->queue, host->max_req_size);
> >>> if (host->can_dma_map_merge)
> >>> WARN(!blk_queue_can_use_dma_map_merging(mq-
> >queue,
> >>> mmc_dev(
> hos
> >> t)),
> >>> --
> >>> 2.18.0
> >>>
> >
> > I test this patch on internal platform(kernel-5.15).
>
> I patched this one and the test shows me a stable 11% performance
> drop.
>
> Before:
> echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
>
> 2048000+0 records in
> 2048000+0 records out
> 1048576000 bytes (0.9 G) copied, 3.912249 s, 256 M/s
>
> After:
> echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
> 2048000+0 records in
> 2048000+0 records out
> 1048576000 bytes (0.9 G) copied, 4.436271 s, 225 M/s
>
> >
> > Before:
> > console:/ # echo 3 > /proc/sys/vm/drop_caches
> > console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> > 4485393+1 records in
> > 4485393+1 records out
> > 2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
> > console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> > 128
> >
> > After:
> > console:/ # echo 3 > /proc/sys/vm/drop_caches
> > console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> > 4485393+1 records in
> > 4485393+1 records out
> > 2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
> > console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> > 1024
> >
Hi Shawn,
What is your readahead value before and after applying this patch?
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-25 8:39 ` Sharp.Xia
0 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-25 8:39 UTC (permalink / raw)
To: shawn.lin
Cc: Sharp.Xia, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
ulf.hansson, wsd_upstream
On Fri, 2023-08-25 at 16:11 +0800, Shawn Lin wrote:
>
> Hi Sharp,
>
> On 2023/8/25 15:10, Sharp Xia (夏宇彬) wrote:
> > On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
> >>
> >> External email : Please do not click links or open attachments
> until
> >> you have verified the sender or the content.
> >> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
> >>>
> >>> From: Sharp Xia <Sharp.Xia@mediatek.com>
> >>>
> >>> MMC does not set readahead and uses the default
> VM_READAHEAD_PAGES
> >>> resulting in slower reading speed.
> >>> Use the max_req_size reported by host driver to set the optimal
> >>> I/O size to improve performance.
> >>
> >> This seems reasonable to me. However, it would be nice if you
> could
> >> share some performance numbers too - comparing before and after
> >> $subject patch.
> >>
> >> Kind regards
> >> Uffe
> >>
> >>>
> >>> Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> >>> ---
> >>> drivers/mmc/core/queue.c | 1 +
> >>> 1 file changed, 1 insertion(+)
> >>>
> >>> diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> >>> index b396e3900717..fc83c4917360 100644
> >>> --- a/drivers/mmc/core/queue.c
> >>> +++ b/drivers/mmc/core/queue.c
> >>> @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
> >> *mq, struct mmc_card *card)
> >>> blk_queue_bounce_limit(mq->queue,
> BLK_BOUNCE_HIGH);
> >>> blk_queue_max_hw_sectors(mq->queue,
> >>> min(host->max_blk_count, host->max_req_size /
> >> 512));
> >>> + blk_queue_io_opt(mq->queue, host->max_req_size);
> >>> if (host->can_dma_map_merge)
> >>> WARN(!blk_queue_can_use_dma_map_merging(mq-
> >queue,
> >>> mmc_dev(
> hos
> >> t)),
> >>> --
> >>> 2.18.0
> >>>
> >
> > I test this patch on internal platform(kernel-5.15).
>
> I patched this one and the test shows me a stable 11% performance
> drop.
>
> Before:
> echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
>
> 2048000+0 records in
> 2048000+0 records out
> 1048576000 bytes (0.9 G) copied, 3.912249 s, 256 M/s
>
> After:
> echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
> 2048000+0 records in
> 2048000+0 records out
> 1048576000 bytes (0.9 G) copied, 4.436271 s, 225 M/s
>
> >
> > Before:
> > console:/ # echo 3 > /proc/sys/vm/drop_caches
> > console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> > 4485393+1 records in
> > 4485393+1 records out
> > 2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
> > console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> > 128
> >
> > After:
> > console:/ # echo 3 > /proc/sys/vm/drop_caches
> > console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> > 4485393+1 records in
> > 4485393+1 records out
> > 2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
> > console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> > 1024
> >
Hi Shawn,
What is your readahead value before and after applying this patch?
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-25 8:39 ` Sharp.Xia
@ 2023-08-25 9:17 ` Shawn Lin
-1 siblings, 0 replies; 26+ messages in thread
From: Shawn Lin @ 2023-08-25 9:17 UTC (permalink / raw)
To: Sharp.Xia
Cc: shawn.lin, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
ulf.hansson, wsd_upstream
On 2023/8/25 16:39, Sharp.Xia@mediatek.com wrote:
> On Fri, 2023-08-25 at 16:11 +0800, Shawn Lin wrote:
>>
>> Hi Sharp,
...
>>> 1024
>>>
> Hi Shawn,
>
> What is your readahead value before and after applying this patch?
>
The original readahead is 128, and after applying the patch is 1024
cat /d/mmc0/ios
clock: 200000000 Hz
actual clock: 200000000 Hz
vdd: 18 (3.0 ~ 3.1 V)
bus mode: 2 (push-pull)
chip select: 0 (don't care)
power mode: 2 (on)
bus width: 3 (8 bits)
timing spec: 10 (mmc HS400 enhanced strobe)
signal voltage: 1 (1.80 V)
driver type: 0 (driver type B)
The driver I used is sdhci-of-dwcmshc.c with a KLMBG2JETDB041 eMMC chip.
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-25 9:17 ` Shawn Lin
0 siblings, 0 replies; 26+ messages in thread
From: Shawn Lin @ 2023-08-25 9:17 UTC (permalink / raw)
To: Sharp.Xia
Cc: shawn.lin, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
ulf.hansson, wsd_upstream
On 2023/8/25 16:39, Sharp.Xia@mediatek.com wrote:
> On Fri, 2023-08-25 at 16:11 +0800, Shawn Lin wrote:
>>
>> Hi Sharp,
...
>>> 1024
>>>
> Hi Shawn,
>
> What is your readahead value before and after applying this patch?
>
The original readahead is 128, and after applying the patch is 1024
cat /d/mmc0/ios
clock: 200000000 Hz
actual clock: 200000000 Hz
vdd: 18 (3.0 ~ 3.1 V)
bus mode: 2 (push-pull)
chip select: 0 (don't care)
power mode: 2 (on)
bus width: 3 (8 bits)
timing spec: 10 (mmc HS400 enhanced strobe)
signal voltage: 1 (1.80 V)
driver type: 0 (driver type B)
The driver I used is sdhci-of-dwcmshc.c with a KLMBG2JETDB041 eMMC chip.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-25 8:39 ` Sharp.Xia
@ 2023-08-25 12:23 ` Wenchao Chen
-1 siblings, 0 replies; 26+ messages in thread
From: Wenchao Chen @ 2023-08-25 12:23 UTC (permalink / raw)
To: Sharp.Xia
Cc: shawn.lin, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
ulf.hansson, wsd_upstream
On Fri, Aug 25, 2023 at 7:43 PM <Sharp.Xia@mediatek.com> wrote:
>
> On Fri, 2023-08-25 at 16:11 +0800, Shawn Lin wrote:
> >
> > Hi Sharp,
> >
> > On 2023/8/25 15:10, Sharp Xia (夏宇彬) wrote:
> > > On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
> > >>
> > >> External email : Please do not click links or open attachments
> > until
> > >> you have verified the sender or the content.
> > >> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
> > >>>
> > >>> From: Sharp Xia <Sharp.Xia@mediatek.com>
> > >>>
> > >>> MMC does not set readahead and uses the default
> > VM_READAHEAD_PAGES
> > >>> resulting in slower reading speed.
> > >>> Use the max_req_size reported by host driver to set the optimal
> > >>> I/O size to improve performance.
> > >>
> > >> This seems reasonable to me. However, it would be nice if you
> > could
> > >> share some performance numbers too - comparing before and after
> > >> $subject patch.
> > >>
> > >> Kind regards
> > >> Uffe
> > >>
> > >>>
> > >>> Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> > >>> ---
> > >>> drivers/mmc/core/queue.c | 1 +
> > >>> 1 file changed, 1 insertion(+)
> > >>>
> > >>> diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> > >>> index b396e3900717..fc83c4917360 100644
> > >>> --- a/drivers/mmc/core/queue.c
> > >>> +++ b/drivers/mmc/core/queue.c
> > >>> @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
> > >> *mq, struct mmc_card *card)
> > >>> blk_queue_bounce_limit(mq->queue,
> > BLK_BOUNCE_HIGH);
> > >>> blk_queue_max_hw_sectors(mq->queue,
> > >>> min(host->max_blk_count, host->max_req_size /
> > >> 512));
> > >>> + blk_queue_io_opt(mq->queue, host->max_req_size);
> > >>> if (host->can_dma_map_merge)
> > >>> WARN(!blk_queue_can_use_dma_map_merging(mq-
> > >queue,
> > >>> mmc_dev(
> > hos
> > >> t)),
> > >>> --
> > >>> 2.18.0
> > >>>
> > >
> > > I test this patch on internal platform(kernel-5.15).
> >
> > I patched this one and the test shows me a stable 11% performance
> > drop.
> >
> > Before:
> > echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
> >
> > 2048000+0 records in
> > 2048000+0 records out
> > 1048576000 bytes (0.9 G) copied, 3.912249 s, 256 M/s
> >
> > After:
> > echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
> > 2048000+0 records in
> > 2048000+0 records out
> > 1048576000 bytes (0.9 G) copied, 4.436271 s, 225 M/s
> >
> > >
> > > Before:
> > > console:/ # echo 3 > /proc/sys/vm/drop_caches
> > > console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> > > 4485393+1 records in
> > > 4485393+1 records out
> > > 2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
> > > console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> > > 128
> > >
> > > After:
> > > console:/ # echo 3 > /proc/sys/vm/drop_caches
> > > console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> > > 4485393+1 records in
> > > 4485393+1 records out
> > > 2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
> > > console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> > > 1024
> > >
> Hi Shawn,
>
> What is your readahead value before and after applying this patch?
>
Hi Sharp
Use "echo 1024 > sys/block/mmcblk0/queue/read_ahead_kb" instead of
"blk_queue_io_opt(mq->queue, host->max_req_size);"?
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-25 12:23 ` Wenchao Chen
0 siblings, 0 replies; 26+ messages in thread
From: Wenchao Chen @ 2023-08-25 12:23 UTC (permalink / raw)
To: Sharp.Xia
Cc: shawn.lin, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
ulf.hansson, wsd_upstream
On Fri, Aug 25, 2023 at 7:43 PM <Sharp.Xia@mediatek.com> wrote:
>
> On Fri, 2023-08-25 at 16:11 +0800, Shawn Lin wrote:
> >
> > Hi Sharp,
> >
> > On 2023/8/25 15:10, Sharp Xia (夏宇彬) wrote:
> > > On Thu, 2023-08-24 at 12:55 +0200, Ulf Hansson wrote:
> > >>
> > >> External email : Please do not click links or open attachments
> > until
> > >> you have verified the sender or the content.
> > >> On Fri, 18 Aug 2023 at 04:45, <Sharp.Xia@mediatek.com> wrote:
> > >>>
> > >>> From: Sharp Xia <Sharp.Xia@mediatek.com>
> > >>>
> > >>> MMC does not set readahead and uses the default
> > VM_READAHEAD_PAGES
> > >>> resulting in slower reading speed.
> > >>> Use the max_req_size reported by host driver to set the optimal
> > >>> I/O size to improve performance.
> > >>
> > >> This seems reasonable to me. However, it would be nice if you
> > could
> > >> share some performance numbers too - comparing before and after
> > >> $subject patch.
> > >>
> > >> Kind regards
> > >> Uffe
> > >>
> > >>>
> > >>> Signed-off-by: Sharp Xia <Sharp.Xia@mediatek.com>
> > >>> ---
> > >>> drivers/mmc/core/queue.c | 1 +
> > >>> 1 file changed, 1 insertion(+)
> > >>>
> > >>> diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
> > >>> index b396e3900717..fc83c4917360 100644
> > >>> --- a/drivers/mmc/core/queue.c
> > >>> +++ b/drivers/mmc/core/queue.c
> > >>> @@ -359,6 +359,7 @@ static void mmc_setup_queue(struct mmc_queue
> > >> *mq, struct mmc_card *card)
> > >>> blk_queue_bounce_limit(mq->queue,
> > BLK_BOUNCE_HIGH);
> > >>> blk_queue_max_hw_sectors(mq->queue,
> > >>> min(host->max_blk_count, host->max_req_size /
> > >> 512));
> > >>> + blk_queue_io_opt(mq->queue, host->max_req_size);
> > >>> if (host->can_dma_map_merge)
> > >>> WARN(!blk_queue_can_use_dma_map_merging(mq-
> > >queue,
> > >>> mmc_dev(
> > hos
> > >> t)),
> > >>> --
> > >>> 2.18.0
> > >>>
> > >
> > > I test this patch on internal platform(kernel-5.15).
> >
> > I patched this one and the test shows me a stable 11% performance
> > drop.
> >
> > Before:
> > echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
> >
> > 2048000+0 records in
> > 2048000+0 records out
> > 1048576000 bytes (0.9 G) copied, 3.912249 s, 256 M/s
> >
> > After:
> > echo 3 > proc/sys/vm/drop_caches && dd if=/data/1GB.img of=/dev/null
> > 2048000+0 records in
> > 2048000+0 records out
> > 1048576000 bytes (0.9 G) copied, 4.436271 s, 225 M/s
> >
> > >
> > > Before:
> > > console:/ # echo 3 > /proc/sys/vm/drop_caches
> > > console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> > > 4485393+1 records in
> > > 4485393+1 records out
> > > 2296521564 bytes (2.1 G) copied, 37.124446 s, 59 M/s
> > > console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> > > 128
> > >
> > > After:
> > > console:/ # echo 3 > /proc/sys/vm/drop_caches
> > > console:/ # dd if=/mnt/media_rw/8031-130D/super.img of=/dev/null
> > > 4485393+1 records in
> > > 4485393+1 records out
> > > 2296521564 bytes (2.1 G) copied, 28.956049 s, 76 M/s
> > > console:/ # cat /sys/block/mmcblk0/queue/read_ahead_kb
> > > 1024
> > >
> Hi Shawn,
>
> What is your readahead value before and after applying this patch?
>
Hi Sharp
Use "echo 1024 > sys/block/mmcblk0/queue/read_ahead_kb" instead of
"blk_queue_io_opt(mq->queue, host->max_req_size);"?
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-25 9:17 ` Shawn Lin
@ 2023-08-26 16:26 ` Sharp.Xia
-1 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-26 16:26 UTC (permalink / raw)
To: shawn.lin
Cc: Sharp.Xia, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
ulf.hansson, wsd_upstream
On Fri, 2023-08-25 at 17:17 +0800, Shawn Lin wrote:
>
>
> On 2023/8/25 16:39, Sharp.Xia@mediatek.com wrote:
> > On Fri, 2023-08-25 at 16:11 +0800, Shawn Lin wrote:
> >>
> >> Hi Sharp,
>
> ...
>
> >>> 1024
> >>>
> > Hi Shawn,
> >
> > What is your readahead value before and after applying this patch?
> >
>
> The original readahead is 128, and after applying the patch is 1024
>
>
> cat /d/mmc0/ios
> clock: 200000000 Hz
> actual clock: 200000000 Hz
> vdd: 18 (3.0 ~ 3.1 V)
> bus mode: 2 (push-pull)
> chip select: 0 (don't care)
> power mode: 2 (on)
> bus width: 3 (8 bits)
> timing spec: 10 (mmc HS400 enhanced strobe)
> signal voltage: 1 (1.80 V)
> driver type: 0 (driver type B)
>
> The driver I used is sdhci-of-dwcmshc.c with a KLMBG2JETDB041 eMMC
> chip.
I tested with RK3568 and sdhci-of-dwcmshc.c driver, the performance improved by 2~3%.
Before:
root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
2097152+0 records in
2097152+0 records out
real 0m 6.01s
user 0m 0.84s
sys 0m 2.89s
root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
128
After:
root@OpenWrt:/mnt/mmcblk0p3# echo 3 > /proc/sys/vm/drop_caches
root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
2097152+0 records in
2097152+0 records out
real 0m 5.86s
user 0m 1.04s
sys 0m 3.18s
root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
1024
root@OpenWrt:/sys/kernel/debug/mmc0# cat ios
clock: 200000000 Hz
actual clock: 200000000 Hz
vdd: 18 (3.0 ~ 3.1 V)
bus mode: 2 (push-pull)
chip select: 0 (don't care)
power mode: 2 (on)
bus width: 3 (8 bits)
timing spec: 9 (mmc HS200)
signal voltage: 1 (1.80 V)
driver type: 0 (driver type B)
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-26 16:26 ` Sharp.Xia
0 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-26 16:26 UTC (permalink / raw)
To: shawn.lin
Cc: Sharp.Xia, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
ulf.hansson, wsd_upstream
On Fri, 2023-08-25 at 17:17 +0800, Shawn Lin wrote:
>
>
> On 2023/8/25 16:39, Sharp.Xia@mediatek.com wrote:
> > On Fri, 2023-08-25 at 16:11 +0800, Shawn Lin wrote:
> >>
> >> Hi Sharp,
>
> ...
>
> >>> 1024
> >>>
> > Hi Shawn,
> >
> > What is your readahead value before and after applying this patch?
> >
>
> The original readahead is 128, and after applying the patch is 1024
>
>
> cat /d/mmc0/ios
> clock: 200000000 Hz
> actual clock: 200000000 Hz
> vdd: 18 (3.0 ~ 3.1 V)
> bus mode: 2 (push-pull)
> chip select: 0 (don't care)
> power mode: 2 (on)
> bus width: 3 (8 bits)
> timing spec: 10 (mmc HS400 enhanced strobe)
> signal voltage: 1 (1.80 V)
> driver type: 0 (driver type B)
>
> The driver I used is sdhci-of-dwcmshc.c with a KLMBG2JETDB041 eMMC
> chip.
I tested with RK3568 and sdhci-of-dwcmshc.c driver, the performance improved by 2~3%.
Before:
root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
2097152+0 records in
2097152+0 records out
real 0m 6.01s
user 0m 0.84s
sys 0m 2.89s
root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
128
After:
root@OpenWrt:/mnt/mmcblk0p3# echo 3 > /proc/sys/vm/drop_caches
root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
2097152+0 records in
2097152+0 records out
real 0m 5.86s
user 0m 1.04s
sys 0m 3.18s
root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
1024
root@OpenWrt:/sys/kernel/debug/mmc0# cat ios
clock: 200000000 Hz
actual clock: 200000000 Hz
vdd: 18 (3.0 ~ 3.1 V)
bus mode: 2 (push-pull)
chip select: 0 (don't care)
power mode: 2 (on)
bus width: 3 (8 bits)
timing spec: 9 (mmc HS200)
signal voltage: 1 (1.80 V)
driver type: 0 (driver type B)
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-25 12:23 ` Wenchao Chen
@ 2023-08-26 16:54 ` Sharp.Xia
-1 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-26 16:54 UTC (permalink / raw)
To: wenchao.chen666
Cc: Sharp.Xia, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg, shawn.lin,
ulf.hansson, wsd_upstream
On Fri, 2023-08-25 at 20:23 +0800, Wenchao Chen wrote:
>
>
> Hi Sharp
> Use "echo 1024 > sys/block/mmcblk0/queue/read_ahead_kb" instead of
> "blk_queue_io_opt(mq->queue, host->max_req_size);"?
Hi Wenchao,
User space does not know the max_req_size of each mmc host.
And when the SD card is hot inserted,
it is complicated for the user space to modify this value.
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-26 16:54 ` Sharp.Xia
0 siblings, 0 replies; 26+ messages in thread
From: Sharp.Xia @ 2023-08-26 16:54 UTC (permalink / raw)
To: wenchao.chen666
Cc: Sharp.Xia, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg, shawn.lin,
ulf.hansson, wsd_upstream
On Fri, 2023-08-25 at 20:23 +0800, Wenchao Chen wrote:
>
>
> Hi Sharp
> Use "echo 1024 > sys/block/mmcblk0/queue/read_ahead_kb" instead of
> "blk_queue_io_opt(mq->queue, host->max_req_size);"?
Hi Wenchao,
User space does not know the max_req_size of each mmc host.
And when the SD card is hot inserted,
it is complicated for the user space to modify this value.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-26 16:26 ` Sharp.Xia
@ 2023-08-28 2:27 ` Shawn Lin
-1 siblings, 0 replies; 26+ messages in thread
From: Shawn Lin @ 2023-08-28 2:27 UTC (permalink / raw)
To: Sharp.Xia
Cc: shawn.lin, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
ulf.hansson, wsd_upstream
Hi Sharp
On 2023/8/27 0:26, Sharp.Xia@mediatek.com wrote:
> On Fri, 2023-08-25 at 17:17 +0800, Shawn Lin wrote:
>>
>>
After more testing, most of my platforms which runs at HS400/HS200 mode
shows nearly no differences with the readahead ranging from 128 to 1024.
Yet just a board shows a performance drop now. Highly suspect it's eMMC
chip depends. I would recommand leave it to the BSP guys to decide which
readahead value is best for their usage.
>
> I tested with RK3568 and sdhci-of-dwcmshc.c driver, the performance improved by 2~3%.
>
> Before:
> root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
> 2097152+0 records in
> 2097152+0 records out
> real 0m 6.01s
> user 0m 0.84s
> sys 0m 2.89s
> root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
> 128
>
> After:
> root@OpenWrt:/mnt/mmcblk0p3# echo 3 > /proc/sys/vm/drop_caches
> root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
> 2097152+0 records in
> 2097152+0 records out
> real 0m 5.86s
> user 0m 1.04s
> sys 0m 3.18s
> root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
> 1024
>
> root@OpenWrt:/sys/kernel/debug/mmc0# cat ios
> clock: 200000000 Hz
> actual clock: 200000000 Hz
> vdd: 18 (3.0 ~ 3.1 V)
> bus mode: 2 (push-pull)
> chip select: 0 (don't care)
> power mode: 2 (on)
> bus width: 3 (8 bits)
> timing spec: 9 (mmc HS200)
> signal voltage: 1 (1.80 V)
> driver type: 0 (driver type B)
>
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-28 2:27 ` Shawn Lin
0 siblings, 0 replies; 26+ messages in thread
From: Shawn Lin @ 2023-08-28 2:27 UTC (permalink / raw)
To: Sharp.Xia
Cc: shawn.lin, angelogioacchino.delregno, linux-arm-kernel,
linux-kernel, linux-mediatek, linux-mmc, matthias.bgg,
ulf.hansson, wsd_upstream
Hi Sharp
On 2023/8/27 0:26, Sharp.Xia@mediatek.com wrote:
> On Fri, 2023-08-25 at 17:17 +0800, Shawn Lin wrote:
>>
>>
After more testing, most of my platforms which runs at HS400/HS200 mode
shows nearly no differences with the readahead ranging from 128 to 1024.
Yet just a board shows a performance drop now. Highly suspect it's eMMC
chip depends. I would recommand leave it to the BSP guys to decide which
readahead value is best for their usage.
>
> I tested with RK3568 and sdhci-of-dwcmshc.c driver, the performance improved by 2~3%.
>
> Before:
> root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
> 2097152+0 records in
> 2097152+0 records out
> real 0m 6.01s
> user 0m 0.84s
> sys 0m 2.89s
> root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
> 128
>
> After:
> root@OpenWrt:/mnt/mmcblk0p3# echo 3 > /proc/sys/vm/drop_caches
> root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
> 2097152+0 records in
> 2097152+0 records out
> real 0m 5.86s
> user 0m 1.04s
> sys 0m 3.18s
> root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
> 1024
>
> root@OpenWrt:/sys/kernel/debug/mmc0# cat ios
> clock: 200000000 Hz
> actual clock: 200000000 Hz
> vdd: 18 (3.0 ~ 3.1 V)
> bus mode: 2 (push-pull)
> chip select: 0 (don't care)
> power mode: 2 (on)
> bus width: 3 (8 bits)
> timing spec: 9 (mmc HS200)
> signal voltage: 1 (1.80 V)
> driver type: 0 (driver type B)
>
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
2023-08-28 2:27 ` Shawn Lin
@ 2023-08-28 9:04 ` Ulf Hansson
-1 siblings, 0 replies; 26+ messages in thread
From: Ulf Hansson @ 2023-08-28 9:04 UTC (permalink / raw)
To: sharp.xia, Shawn Lin
Cc: angelogioacchino.delregno, linux-arm-kernel, linux-kernel,
linux-mediatek, linux-mmc, matthias.bgg, wsd_upstream
On Mon, 28 Aug 2023 at 04:28, Shawn Lin <shawn.lin@rock-chips.com> wrote:
>
> Hi Sharp
>
> On 2023/8/27 0:26, Sharp.Xia@mediatek.com wrote:
> > On Fri, 2023-08-25 at 17:17 +0800, Shawn Lin wrote:
> >>
> >>
>
> After more testing, most of my platforms which runs at HS400/HS200 mode
> shows nearly no differences with the readahead ranging from 128 to 1024.
> Yet just a board shows a performance drop now. Highly suspect it's eMMC
> chip depends. I would recommand leave it to the BSP guys to decide which
> readahead value is best for their usage.
That's a very good point. The SD/eMMC card certainly behaves
differently, depending on the request-size.
Another thing we could consider doing, could be to combine the
information about the request-size from the mmc host, with some
relevant information from the registers in the card (not sure exactly
what though).
>
> >
> > I tested with RK3568 and sdhci-of-dwcmshc.c driver, the performance improved by 2~3%.
> >
> > Before:
> > root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
> > 2097152+0 records in
> > 2097152+0 records out
> > real 0m 6.01s
> > user 0m 0.84s
> > sys 0m 2.89s
> > root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
> > 128
> >
> > After:
> > root@OpenWrt:/mnt/mmcblk0p3# echo 3 > /proc/sys/vm/drop_caches
> > root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
> > 2097152+0 records in
> > 2097152+0 records out
> > real 0m 5.86s
> > user 0m 1.04s
> > sys 0m 3.18s
> > root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
> > 1024
> >
> > root@OpenWrt:/sys/kernel/debug/mmc0# cat ios
> > clock: 200000000 Hz
> > actual clock: 200000000 Hz
> > vdd: 18 (3.0 ~ 3.1 V)
> > bus mode: 2 (push-pull)
> > chip select: 0 (don't care)
> > power mode: 2 (on)
> > bus width: 3 (8 bits)
> > timing spec: 9 (mmc HS200)
> > signal voltage: 1 (1.80 V)
> > driver type: 0 (driver type B)
> >
Thanks for testing and sharing the data, both of you!
Kind regards
Uffe
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue
@ 2023-08-28 9:04 ` Ulf Hansson
0 siblings, 0 replies; 26+ messages in thread
From: Ulf Hansson @ 2023-08-28 9:04 UTC (permalink / raw)
To: sharp.xia, Shawn Lin
Cc: angelogioacchino.delregno, linux-arm-kernel, linux-kernel,
linux-mediatek, linux-mmc, matthias.bgg, wsd_upstream
On Mon, 28 Aug 2023 at 04:28, Shawn Lin <shawn.lin@rock-chips.com> wrote:
>
> Hi Sharp
>
> On 2023/8/27 0:26, Sharp.Xia@mediatek.com wrote:
> > On Fri, 2023-08-25 at 17:17 +0800, Shawn Lin wrote:
> >>
> >>
>
> After more testing, most of my platforms which runs at HS400/HS200 mode
> shows nearly no differences with the readahead ranging from 128 to 1024.
> Yet just a board shows a performance drop now. Highly suspect it's eMMC
> chip depends. I would recommand leave it to the BSP guys to decide which
> readahead value is best for their usage.
That's a very good point. The SD/eMMC card certainly behaves
differently, depending on the request-size.
Another thing we could consider doing, could be to combine the
information about the request-size from the mmc host, with some
relevant information from the registers in the card (not sure exactly
what though).
>
> >
> > I tested with RK3568 and sdhci-of-dwcmshc.c driver, the performance improved by 2~3%.
> >
> > Before:
> > root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
> > 2097152+0 records in
> > 2097152+0 records out
> > real 0m 6.01s
> > user 0m 0.84s
> > sys 0m 2.89s
> > root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
> > 128
> >
> > After:
> > root@OpenWrt:/mnt/mmcblk0p3# echo 3 > /proc/sys/vm/drop_caches
> > root@OpenWrt:/mnt/mmcblk0p3# time dd if=test.img of=/dev/null
> > 2097152+0 records in
> > 2097152+0 records out
> > real 0m 5.86s
> > user 0m 1.04s
> > sys 0m 3.18s
> > root@OpenWrt:/mnt/mmcblk0p3# cat /sys/block/mmcblk0/queue/read_ahead_kb
> > 1024
> >
> > root@OpenWrt:/sys/kernel/debug/mmc0# cat ios
> > clock: 200000000 Hz
> > actual clock: 200000000 Hz
> > vdd: 18 (3.0 ~ 3.1 V)
> > bus mode: 2 (push-pull)
> > chip select: 0 (don't care)
> > power mode: 2 (on)
> > bus width: 3 (8 bits)
> > timing spec: 9 (mmc HS200)
> > signal voltage: 1 (1.80 V)
> > driver type: 0 (driver type B)
> >
Thanks for testing and sharing the data, both of you!
Kind regards
Uffe
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
end of thread, other threads:[~2023-08-28 9:06 UTC | newest]
Thread overview: 26+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-08-18 2:28 [PATCH 1/1] mmc: Set optimal I/O size when mmc_setip_queue Sharp.Xia
2023-08-18 2:28 ` Sharp.Xia
2023-08-24 10:55 ` Ulf Hansson
2023-08-24 10:55 ` Ulf Hansson
2023-08-25 7:10 ` Sharp Xia (夏宇彬)
2023-08-25 7:10 ` Sharp Xia (夏宇彬)
2023-08-25 8:11 ` Shawn Lin
2023-08-25 8:11 ` Shawn Lin
2023-08-25 8:39 ` Sharp.Xia
2023-08-25 8:39 ` Sharp.Xia
2023-08-25 9:17 ` Shawn Lin
2023-08-25 9:17 ` Shawn Lin
2023-08-26 16:26 ` Sharp.Xia
2023-08-26 16:26 ` Sharp.Xia
2023-08-28 2:27 ` Shawn Lin
2023-08-28 2:27 ` Shawn Lin
2023-08-28 9:04 ` Ulf Hansson
2023-08-28 9:04 ` Ulf Hansson
2023-08-25 12:23 ` Wenchao Chen
2023-08-25 12:23 ` Wenchao Chen
2023-08-26 16:54 ` Sharp.Xia
2023-08-26 16:54 ` Sharp.Xia
2023-08-25 7:25 ` Sharp.Xia
2023-08-25 7:25 ` Sharp.Xia
2023-08-25 7:26 ` Sharp.Xia
2023-08-25 7:26 ` Sharp.Xia
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.