It has been more than one year since we tried to change the default
from legacy to multi queue in SCSI. Back then we had to retract the
change because of performance issues with rotating disks.
In the meantime there have been a substantial amount of performance
improvements, thus we can re-enable scsi-mq without a significant
performance penalty.
Signed-off-by: Johannes Thumshirn <[email protected]>
---
drivers/scsi/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index 35c909bbf8ba..bd115bab162e 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -49,6 +49,7 @@ config SCSI_NETLINK
config SCSI_MQ_DEFAULT
bool "SCSI: use blk-mq I/O path by default"
+ default y
depends on SCSI
---help---
This option enables the new blk-mq based I/O path for SCSI
--
2.16.4
On 07/03/2018 08:54 AM, Johannes Thumshirn wrote:
> It has been more than one year since we tried to change the default
> from legacy to multi queue in SCSI. Back then we had to retract the
> change because of performance issues with rotating disks.
>
> In the meantime there have been a substantial amount of performance
> improvements, thus we can re-enable scsi-mq without a significant
> performance penalty.
>
> Signed-off-by: Johannes Thumshirn <[email protected]>
> ---
> drivers/scsi/Kconfig | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
> index 35c909bbf8ba..bd115bab162e 100644
> --- a/drivers/scsi/Kconfig
> +++ b/drivers/scsi/Kconfig
> @@ -49,6 +49,7 @@ config SCSI_NETLINK
>
> config SCSI_MQ_DEFAULT
> bool "SCSI: use blk-mq I/O path by default"
> + default y
> depends on SCSI
> ---help---
> This option enables the new blk-mq based I/O path for SCSI
>
Yeah, let's try it.
Reviewed-by: Hannes Reinecke <[email protected]>
Cheers,
Hannes
--
Dr. Hannes Reinecke Teamlead Storage & Networking
[email protected] +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)
On Tue, Jul 3, 2018 at 2:54 PM, Johannes Thumshirn <[email protected]> wrote:
> It has been more than one year since we tried to change the default
> from legacy to multi queue in SCSI. Back then we had to retract the
> change because of performance issues with rotating disks.
>
> In the meantime there have been a substantial amount of performance
> improvements, thus we can re-enable scsi-mq without a significant
> performance penalty.
>
> Signed-off-by: Johannes Thumshirn <[email protected]>
> ---
> drivers/scsi/Kconfig | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
> index 35c909bbf8ba..bd115bab162e 100644
> --- a/drivers/scsi/Kconfig
> +++ b/drivers/scsi/Kconfig
> @@ -49,6 +49,7 @@ config SCSI_NETLINK
>
> config SCSI_MQ_DEFAULT
> bool "SCSI: use blk-mq I/O path by default"
> + default y
> depends on SCSI
> ---help---
> This option enables the new blk-mq based I/O path for SCSI
> --
> 2.16.4
>
Reviewed-by: Ming Lei <[email protected]>
thanks,
Ming Lei
On Tue, 2018-07-03 at 08:54 +0200, Johannes Thumshirn wrote:
> It has been more than one year since we tried to change the default
> from legacy to multi queue in SCSI. Back then we had to retract the
> change because of performance issues with rotating disks.
>
> In the meantime there have been a substantial amount of performance
> improvements, thus we can re-enable scsi-mq without a significant
> performance penalty.
Hello Johannes,
Please include references to commits cbe7dfa26eee ("Revert "scsi:
default to scsi-mq"") and 5c279bd9e406 ("scsi: default to scsi-mq").
I think it's a good idea to mention that suspend/resume now works
with scsi-mq.
Bart.
On 04/07/2018 00:21, Bart Van Assche wrote:
> On Tue, 2018-07-03 at 08:54 +0200, Johannes Thumshirn wrote:
>> It has been more than one year since we tried to change the default
>> from legacy to multi queue in SCSI. Back then we had to retract the
>> change because of performance issues with rotating disks.
>>
>> In the meantime there have been a substantial amount of performance
>> improvements, thus we can re-enable scsi-mq without a significant
>> performance penalty.
>
> Hello Johannes,
>
> Please include references to commits cbe7dfa26eee ("Revert "scsi:
> default to scsi-mq"") and 5c279bd9e406 ("scsi: default to scsi-mq").
> I think it's a good idea to mention that suspend/resume now works
> with scsi-mq.
>
> Bart.
We were seeing performance drops previously for hisi_sas when originally
enabled by default, but results on 4.18 seem almost identical with and
without MQ default.
FWIW,
Acked-by: John Garry <[email protected]>
>
>
>
>
>
>
>
>
On Wed, Jul 04, 2018 at 09:18:23AM +0100, John Garry wrote:
> We were seeing performance drops previously for hisi_sas when originally
> enabled by default, but results on 4.18 seem almost identical with and
> without MQ default.
>
> FWIW,
>
> Acked-by: John Garry <[email protected]>
Thanks for letting me know.
--
Johannes Thumshirn Storage
[email protected] +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 N?rnberg
GF: Felix Imend?rffer, Jane Smithard, Graham Norton
HRB 21284 (AG N?rnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850
On Tue, Jul 03, 2018 at 11:21:18PM +0000, Bart Van Assche wrote:
> Please include references to commits cbe7dfa26eee ("Revert "scsi:
> default to scsi-mq"") and 5c279bd9e406 ("scsi: default to scsi-mq").
> I think it's a good idea to mention that suspend/resume now works
> with scsi-mq.
That's actually a good idea. Will be sending out a non-rfc version
with these changelog changes and Reviews/Acks applied.
--
Johannes Thumshirn Storage
[email protected] +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 N?rnberg
GF: Felix Imend?rffer, Jane Smithard, Graham Norton
HRB 21284 (AG N?rnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850