Skip to contentSkip to conversation

scsi: ufs: Allow async suspend/resume callbacks

PATCH Applied Vincent Palomares 4 participants 6 messages Jul 28, 2021 – Jul 31, 2021 SCSI

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 scsi: ufs: Allow async suspend/resume callbacks

Allow UFS suspend/resume callbacks to run in parallel with other suspend/resume callbacks. This can recoup dozens of milliseconds on the resume path if UFS hardware needs to be powered back on.

Suspending and resuming asynchronously is safe to do so long as the driver callbacks only depend on resources made available by either a) parent devices or b) devices explicitly marked as suppliers with device_link_add.

Cc Bjorn Helgaas <[email protected]>
Cc Jaegeuk Kim <[email protected]>
Cc Bart Van Assche <[email protected]>
Cc Adrian Hunter <[email protected]>
Cc Stanley Chu <[email protected]>
Cc Can Guo <[email protected]>
Cc Asutosh Das <[email protected]>
Cc Avri Altman <[email protected]>
Cc Martin K. Petersen <[email protected]>

---

Are there any suspend/resume dependencies for UFS drivers not tracked by the device parent relationship?

drivers/scsi/ufs/ufshcd.c | 1 +

 1 file changed, 1 insertion(+)
drivers/scsi/ufs/ufshcd.c +1 -0
index b87ff68aa9aa..9ec5c308a0ea 100644
@@ -9625,6 +9625,7 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq)
 	async_schedule(ufshcd_async_scan, hba);
 	ufs_sysfs_add_nodes(hba->dev);
 
+	device_enable_async_suspend(dev);
 	return 0;
 
 free_tmf_queue:
Signature · 1 line
2.32.0.432.gabb21c7263-goog
Reviewed-by on v1 1/1
#1 lore raw
Quoted 2 lines · Vincent Palomares

Are there any suspend/resume dependencies for UFS drivers not tracked by the device parent relationship?

Not that I know of. Since I like this patch:

#2 lore raw
In reply to Vincent Palomares
Quoted 32 lines

Allow UFS suspend/resume callbacks to run in parallel with other suspend/resume callbacks. This can recoup dozens of milliseconds on the resume path if UFS hardware needs to be powered back on.

Suspending and resuming asynchronously is safe to do so long as the driver callbacks only depend on resources made available by either a) parent devices or b) devices explicitly marked as suppliers with device_link_add.

Cc Bjorn Helgaas <[email protected]>
Cc Jaegeuk Kim <[email protected]>
Cc Bart Van Assche <[email protected]>
Cc Adrian Hunter <[email protected]>
Cc Stanley Chu <[email protected]>
Cc Can Guo <[email protected]>
Cc Asutosh Das <[email protected]>
Cc Avri Altman <[email protected]>
Cc Martin K. Petersen <[email protected]>

---

Are there any suspend/resume dependencies for UFS drivers not tracked by the device parent relationship?

drivers/scsi/ufs/ufshcd.c | 1 +

 1 file changed, 1 insertion(+)
drivers/scsi/ufs/ufshcd.c +0 -0
index b87ff68aa9aa..9ec5c308a0ea 100644

drivers/scsi/ufs/ufshcd.c int ufshcd_init(struct ufs_hba *hba, void __iomem · line 9630

Quoted 6 lines
drivers/scsi/ufs/ufshcd.c +1 -0
@@ -9625,6 +9625,7 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem
 *mmio_base, unsigned int irq)
        async_schedule(ufshcd_async_scan, hba);
        ufs_sysfs_add_nodes(hba->dev);
 
+       device_enable_async_suspend(dev);
        return 0;

Isn't device_enable_async_suspend is being called for each lun in scsi_sysfs_add_sdev Anyway?

Thanks,
Avri

Quoted 4 lines
drivers/scsi/ufs/ufshcd.c +0 -1
 free_tmf_queue:
--

2.32.0.432.gabb21c7263-goog

#3 lore raw
Quoted 10 lines · Avri Altman
Quoted 9 lines · Vincent
drivers/scsi/ufs/ufshcd.c +0 -0
index b87ff68aa9aa..9ec5c308a0ea 100644
@@ -9625,6 +9625,7 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem
 *mmio_base, unsigned int irq)
         async_schedule(ufshcd_async_scan, hba);
         ufs_sysfs_add_nodes(hba->dev);
 
Quoted 3 lines

drivers/scsi/ufs/ufshcd.c int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq) · line 9629

Quoted 2 lines
drivers/scsi/ufs/ufshcd.c +1 -0
+       device_enable_async_suspend(dev);
         return 0;

Isn't device_enable_async_suspend is being called for each lun in scsi_sysfs_add_sdev Anyway?

Hi Avri,

Our measurements have shown that resume takes longer than it should with encryption enabled. While suspending we change the power mode of the UFS device to a mode in which it loses crypto keys. Restoring crypto keys during resume (blk_ksm_reprogram_all_keys()) takes about 31 ms. This is the long pole and takes much more time than resuming LUNs. This patch makes UFS resume happen concurrently with resuming other devices in the system instead of serializing it. Measurements have shown that this patch significantly improves the time needed to resume an Android device.

Bart.

#4 lore raw
Quoted 18 lines · Avri Altman
Quoted 10 lines
Quoted 9 lines · Vincent
drivers/scsi/ufs/ufshcd.c +0 -0
index b87ff68aa9aa..9ec5c308a0ea 100644
@@ -9625,6 +9625,7 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem
 *mmio_base, unsigned int irq)
         async_schedule(ufshcd_async_scan, hba);
         ufs_sysfs_add_nodes(hba->dev);
 
Quoted 3 lines

drivers/scsi/ufs/ufshcd.c int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq) · line 9629

Quoted 2 lines
drivers/scsi/ufs/ufshcd.c +1 -0
+       device_enable_async_suspend(dev);
         return 0;

Isn't device_enable_async_suspend is being called for each lun in

scsi_sysfs_add_sdev Anyway?

Hi Avri,

Quoted 8 lines

Our measurements have shown that resume takes longer than it should with encryption enabled. While suspending we change the power mode of the UFS device to a mode in which it loses crypto keys. Restoring crypto keys during resume (blk_ksm_reprogram_all_keys()) takes about 31 ms. This is the long pole and takes much more time than resuming LUNs. This patch makes UFS resume happen concurrently with resuming other devices in the system instead of serializing it. Measurements have shown that this patch significantly improves the time needed to resume an Android device.

OK.
Thanks for the extra info.

Thanks,
Avri

Quoted 2 lines

Bart.

#5 lore raw
In reply to Vincent Palomares

Vincent,

Quoted 3 lines

Allow UFS suspend/resume callbacks to run in parallel with other suspend/resume callbacks. This can recoup dozens of milliseconds on the resume path if UFS hardware needs to be powered back on.

Applied to 5.15/scsi-staging, thanks!

Signature · 1 line
Martin K. Petersen	Oracle Linux Engineering