linux-scsi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v1] ufs: core: remove dev cmd clock scaling busy
@ 2023-10-04  6:24 peter.wang
  2023-10-04  6:37 ` Stanley Chu
  2023-10-10  1:37 ` Martin K. Petersen
  0 siblings, 2 replies; 3+ messages in thread
From: peter.wang @ 2023-10-04  6:24 UTC (permalink / raw)
  To: stanley.chu, linux-scsi, martin.petersen, avri.altman,
	alim.akhtar, jejb
  Cc: wsd_upstream, linux-mediatek, peter.wang, chun-hung.wu,
	alice.chao, cc.chou, chaotian.jing, jiajie.hao, powen.kao,
	qilin.tan, lin.gui, tun-yu.yu, eddie.huang, naomi.chu

From: Peter Wang <peter.wang@mediatek.com>

If dev command timeout, clk_scaling.active_reqs is not decrease
and cause clock scaling framework abnormal. But it is complicated to
handle different dev command timeout case in legacy mode or mcq mode.
Besides, dev cmd is rare used and busy time is short.
So remove clock scaling busy window for dev cmd is properly.
Same as uic or tm cmd which doesn't update busy window too.

Signed-off-by: Peter Wang <peter.wang@mediatek.com>
---
 drivers/ufs/core/ufshcd.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
index c2df07545f96..474d5dded7ed 100644
--- a/drivers/ufs/core/ufshcd.c
+++ b/drivers/ufs/core/ufshcd.c
@@ -2165,7 +2165,8 @@ void ufshcd_send_command(struct ufs_hba *hba, unsigned int task_tag,
 	lrbp->compl_time_stamp = ktime_set(0, 0);
 	lrbp->compl_time_stamp_local_clock = 0;
 	ufshcd_add_command_trace(hba, task_tag, UFS_CMD_SEND);
-	ufshcd_clk_scaling_start_busy(hba);
+	if (lrbp->cmd)
+		ufshcd_clk_scaling_start_busy(hba);
 	if (unlikely(ufshcd_should_inform_monitor(hba, lrbp)))
 		ufshcd_start_monitor(hba, lrbp);
 
@@ -5405,7 +5406,6 @@ void ufshcd_compl_one_cqe(struct ufs_hba *hba, int task_tag,
 				lrbp->utr_descriptor_ptr->header.ocs = ocs;
 			}
 			complete(hba->dev_cmd.complete);
-			ufshcd_clk_scaling_update_busy(hba);
 		}
 	}
 }
-- 
2.18.0


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH v1] ufs: core: remove dev cmd clock scaling busy
  2023-10-04  6:24 [PATCH v1] ufs: core: remove dev cmd clock scaling busy peter.wang
@ 2023-10-04  6:37 ` Stanley Chu
  2023-10-10  1:37 ` Martin K. Petersen
  1 sibling, 0 replies; 3+ messages in thread
From: Stanley Chu @ 2023-10-04  6:37 UTC (permalink / raw)
  To: peter.wang
  Cc: stanley.chu, linux-scsi, martin.petersen, avri.altman,
	alim.akhtar, jejb, wsd_upstream, linux-mediatek, chun-hung.wu,
	alice.chao, cc.chou, chaotian.jing, jiajie.hao, powen.kao,
	qilin.tan, lin.gui, tun-yu.yu, eddie.huang, naomi.chu

On Wed, Oct 4, 2023 at 2:25 PM <peter.wang@mediatek.com> wrote:
>
> From: Peter Wang <peter.wang@mediatek.com>
>
> If dev command timeout, clk_scaling.active_reqs is not decrease
> and cause clock scaling framework abnormal. But it is complicated to
> handle different dev command timeout case in legacy mode or mcq mode.
> Besides, dev cmd is rare used and busy time is short.
> So remove clock scaling busy window for dev cmd is properly.
> Same as uic or tm cmd which doesn't update busy window too.
>
> Signed-off-by: Peter Wang <peter.wang@mediatek.com>

It makes sense to me since clock scaling is aimed at balancing
performance and power consumption for different workloads, which
should be data workloads rather than device commands.

Reviewed-by: Stanley Chu <stanley.chu@mediatek.com>

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH v1] ufs: core: remove dev cmd clock scaling busy
  2023-10-04  6:24 [PATCH v1] ufs: core: remove dev cmd clock scaling busy peter.wang
  2023-10-04  6:37 ` Stanley Chu
@ 2023-10-10  1:37 ` Martin K. Petersen
  1 sibling, 0 replies; 3+ messages in thread
From: Martin K. Petersen @ 2023-10-10  1:37 UTC (permalink / raw)
  To: peter.wang
  Cc: stanley.chu, linux-scsi, martin.petersen, avri.altman,
	alim.akhtar, jejb, wsd_upstream, linux-mediatek, chun-hung.wu,
	alice.chao, cc.chou, chaotian.jing, jiajie.hao, powen.kao,
	qilin.tan, lin.gui, tun-yu.yu, eddie.huang, naomi.chu


Peter,

> If dev command timeout, clk_scaling.active_reqs is not decrease and
> cause clock scaling framework abnormal. But it is complicated to
> handle different dev command timeout case in legacy mode or mcq mode.
> Besides, dev cmd is rare used and busy time is short. So remove clock
> scaling busy window for dev cmd is properly. Same as uic or tm cmd
> which doesn't update busy window too.

Applied to 6.7/scsi-staging, thanks!

-- 
Martin K. Petersen	Oracle Linux Engineering

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2023-10-10  1:38 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-10-04  6:24 [PATCH v1] ufs: core: remove dev cmd clock scaling busy peter.wang
2023-10-04  6:37 ` Stanley Chu
2023-10-10  1:37 ` Martin K. Petersen

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).