diff mbox series

[RFC,v2,07/11] scsi: ufs: core: Perform read back after writing UTP_TASK_REQ_LIST_BASE_H

Message ID 20231221-ufs-reset-ensure-effect-before-delay-v2-7-75af2a9bae51@redhat.com (mailing list archive)
State Superseded
Headers show
Series scsi: ufs: Remove overzealous memory barriers | expand

Commit Message

Andrew Halaney Dec. 21, 2023, 6:25 p.m. UTC
Currently, the UTP_TASK_REQ_LIST_BASE_L/UTP_TASK_REQ_LIST_BASE_H regs
are written to and then completed with an mb().

mb() ensure that the write completes, but completion doesn't mean that
it isn't stored in a buffer somewhere. The recommendation for
ensuring these bits have taken effect on the device is to perform a read
back to force it to make it all the way to the device. This is
documented in device-io.rst and a talk by Will Deacon on this can
be seen over here:

    https://youtu.be/i6DayghhA8Q?si=MiyxB5cKJXSaoc01&t=1678

Let's do that to ensure the bits hit the device. Because the mb()'s
purpose wasn't to add extra ordering (on top of the ordering guaranteed
by writel()/readl()), it can safely be removed.

Fixes: 88441a8d355d ("scsi: ufs: core: Add hibernation callbacks")
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
---
 drivers/ufs/core/ufshcd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Manivannan Sadhasivam Dec. 27, 2023, 6:10 a.m. UTC | #1
On Thu, Dec 21, 2023 at 12:25:24PM -0600, Andrew Halaney wrote:
> Currently, the UTP_TASK_REQ_LIST_BASE_L/UTP_TASK_REQ_LIST_BASE_H regs
> are written to and then completed with an mb().
> 
> mb() ensure that the write completes, but completion doesn't mean that
> it isn't stored in a buffer somewhere. The recommendation for
> ensuring these bits have taken effect on the device is to perform a read
> back to force it to make it all the way to the device. This is
> documented in device-io.rst and a talk by Will Deacon on this can
> be seen over here:
> 
>     https://youtu.be/i6DayghhA8Q?si=MiyxB5cKJXSaoc01&t=1678
> 
> Let's do that to ensure the bits hit the device. Because the mb()'s
> purpose wasn't to add extra ordering (on top of the ordering guaranteed
> by writel()/readl()), it can safely be removed.
> 
> Fixes: 88441a8d355d ("scsi: ufs: core: Add hibernation callbacks")
> Signed-off-by: Andrew Halaney <ahalaney@redhat.com>

This also I'm not sure whether we can safely remove readback. So,

Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>

- Mani

> ---
>  drivers/ufs/core/ufshcd.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
> index d1e33328ff3f..7bfb556e5b8e 100644
> --- a/drivers/ufs/core/ufshcd.c
> +++ b/drivers/ufs/core/ufshcd.c
> @@ -10351,7 +10351,7 @@ int ufshcd_system_restore(struct device *dev)
>  	 * are updated with the latest queue addresses. Only after
>  	 * updating these addresses, we can queue the new commands.
>  	 */
> -	mb();
> +	ufshcd_readl(hba, REG_UTP_TASK_REQ_LIST_BASE_H);
>  
>  	/* Resuming from hibernate, assume that link was OFF */
>  	ufshcd_set_link_off(hba);
> 
> -- 
> 2.43.0
>
diff mbox series

Patch

diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
index d1e33328ff3f..7bfb556e5b8e 100644
--- a/drivers/ufs/core/ufshcd.c
+++ b/drivers/ufs/core/ufshcd.c
@@ -10351,7 +10351,7 @@  int ufshcd_system_restore(struct device *dev)
 	 * are updated with the latest queue addresses. Only after
 	 * updating these addresses, we can queue the new commands.
 	 */
-	mb();
+	ufshcd_readl(hba, REG_UTP_TASK_REQ_LIST_BASE_H);
 
 	/* Resuming from hibernate, assume that link was OFF */
 	ufshcd_set_link_off(hba);