[1/1] s390: Cleanup removed IOSCHED_DEADLINE

Commit Message

Stefan Haberland April 6, 2020, 7:41 a.m. UTC

From: Krzysztof Kozlowski <[e-mail]>

CONFIG_IOSCHED_DEADLINE is gone since commit f382fb0bcef4 ("block:
remove legacy IO schedulers").

Signed-off-by: Krzysztof Kozlowski <[e-mail]>
Signed-off-by: Stefan Haberland <[e-mail]>
---
 drivers/s390/block/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Jens Axboe April 6, 2020, 3:32 p.m. UTC | #1

On 4/6/20 1:41 AM, Stefan Haberland wrote:
> From: Krzysztof Kozlowski <[e-mail]>
> 
> CONFIG_IOSCHED_DEADLINE is gone since commit f382fb0bcef4 ("block:
> remove legacy IO schedulers").

Isn't this a leftover thing from when dasd selected deadline
internally? I don't think we need this anymore, just kill the
select completely.

Stefan Haberland April 7, 2020, 10:01 a.m. UTC | #2

Am 06.04.20 um 17:32 schrieb Jens Axboe:
> On 4/6/20 1:41 AM, Stefan Haberland wrote:
>> From: Krzysztof Kozlowski <[e-mail]>
>>
>> CONFIG_IOSCHED_DEADLINE is gone since commit f382fb0bcef4 ("block:
>> remove legacy IO schedulers").
> Isn't this a leftover thing from when dasd selected deadline
> internally? I don't think we need this anymore, just kill the
> select completely.
>

Basically yes.
We still have the recommendation to use deadline and there are some udev
rules in place which select it.
So I thought it might be a good idea to keep the line. But I am also
fine to drop it completely if you prefer this.

Regards,
Stefan

Jens Axboe April 7, 2020, 7:24 p.m. UTC | #3

On 4/7/20 3:01 AM, Stefan Haberland wrote:
> Am 06.04.20 um 17:32 schrieb Jens Axboe:
>> On 4/6/20 1:41 AM, Stefan Haberland wrote:
>>> From: Krzysztof Kozlowski <[e-mail]>
>>>
>>> CONFIG_IOSCHED_DEADLINE is gone since commit f382fb0bcef4 ("block:
>>> remove legacy IO schedulers").
>> Isn't this a leftover thing from when dasd selected deadline
>> internally? I don't think we need this anymore, just kill the
>> select completely.
>>
> 
> Basically yes.
> We still have the recommendation to use deadline and there are some udev
> rules in place which select it.
> So I thought it might be a good idea to keep the line. But I am also
> fine to drop it completely if you prefer this.

I don't feel super strongly about it, but at the same time I don't
think there's a reason for it. Besides, I think everyone has deadline
enabled. So I guess we should just kill it.

Stefan Haberland April 15, 2020, 9:18 a.m. UTC | #4

Am 07.04.20 um 21:24 schrieb Jens Axboe:
> On 4/7/20 3:01 AM, Stefan Haberland wrote:
>> Am 06.04.20 um 17:32 schrieb Jens Axboe:
>>> On 4/6/20 1:41 AM, Stefan Haberland wrote:
>>>> From: Krzysztof Kozlowski <[e-mail]>
>>>>
>>>> CONFIG_IOSCHED_DEADLINE is gone since commit f382fb0bcef4 ("block:
>>>> remove legacy IO schedulers").
>>> Isn't this a leftover thing from when dasd selected deadline
>>> internally? I don't think we need this anymore, just kill the
>>> select completely.
>>>
>> Basically yes.
>> We still have the recommendation to use deadline and there are some udev
>> rules in place which select it.
>> So I thought it might be a good idea to keep the line. But I am also
>> fine to drop it completely if you prefer this.
> I don't feel super strongly about it, but at the same time I don't
> think there's a reason for it. Besides, I think everyone has deadline
> enabled. So I guess we should just kill it.
>

OK, thanks for your opinion. I will remove the line and come up with a
different patch.

Patch

diff --git a/drivers/s390/block/Kconfig b/drivers/s390/block/Kconfig
index a8682f69effc..1f06b19cb290 100644
--- a/drivers/s390/block/Kconfig
+++ b/drivers/s390/block/Kconfig
@@ -26,7 +26,7 @@  config DASD
 	def_tristate y
 	prompt "Support for DASD devices"
 	depends on CCW && BLOCK
-	select IOSCHED_DEADLINE
+	select MQ_IOSCHED_DEADLINE
 	help
 	  Enable this option if you want to access DASDs directly utilizing
 	  S/390s channel subsystem commands. This is necessary for running