* Re: [PATCH 1/1] usb: usleep_range is preferred over udelay where wakeup is flexible
2015-06-24 6:54 [PATCH 1/1] usb: usleep_range is preferred over udelay where wakeup is flexible Sunny Kumar
@ 2015-06-24 13:44 ` Alan Stern
2015-06-24 14:28 ` Sunny Kumar
` (2 subsequent siblings)
3 siblings, 0 replies; 6+ messages in thread
From: Alan Stern @ 2015-06-24 13:44 UTC (permalink / raw)
To: Sunny Kumar
Cc: mdharm-usb, gregkh, linux-usb, usb-storage, linux-kernel,
swt.sunny
On Wed, 24 Jun 2015, Sunny Kumar wrote:
> According to Documentation/timers/timers-howto.txt"
> udelay() is only called once from a place where sleeping is allowed.
> We can replace it with a call to usleep_range()
> with a reasonable upper limit.
>
> Signed-off-by: Sunny Kumar <sunny.kumar.roy@gmail.com>
> ---
> drivers/usb/storage/transport.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
> index 540add2..7cd45ac 100644
> --- a/drivers/usb/storage/transport.c
> +++ b/drivers/usb/storage/transport.c
> @@ -1111,7 +1111,7 @@ int usb_stor_Bulk_transport(struct scsi_cmnd *srb, struct us_data *us)
> * command phase and the data phase. Some devices need a little
> * more than that, probably because of clock rate inaccuracies. */
> if (unlikely(us->fflags & US_FL_GO_SLOW))
> - udelay(125);
> + usleep_range(100, 125);
You said you were going to use a reasonable upper limit. Instead, you
left the upper limit the same and decreased the lower limit, which
could cause errors.
Alan Stern
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH 1/1] usb: usleep_range is preferred over udelay where wakeup is flexible
2015-06-24 6:54 [PATCH 1/1] usb: usleep_range is preferred over udelay where wakeup is flexible Sunny Kumar
2015-06-24 13:44 ` Alan Stern
@ 2015-06-24 14:28 ` Sunny Kumar
2015-06-24 14:35 ` Alan Stern
2015-06-24 15:45 ` Sunny Kumar
2015-06-25 15:30 ` Sunny Kumar
3 siblings, 1 reply; 6+ messages in thread
From: Sunny Kumar @ 2015-06-24 14:28 UTC (permalink / raw)
To: mdharm-usb, gregkh, linux-usb, usb-storage, linux-kernel
[-- Attachment #1: Type: text/plain, Size: 1150 bytes --]
Ok so what about putting lower limit to 125 and increase upper limit 200
Sunny
On Wed, Jun 24, 2015 at 12:24:29PM +0530, Sunny Kumar wrote:
> According to Documentation/timers/timers-howto.txt"
> udelay() is only called once from a place where sleeping is allowed.
> We can replace it with a call to usleep_range()
> with a reasonable upper limit.
>
> Signed-off-by: Sunny Kumar <sunny.kumar.roy@gmail.com>
> ---
> drivers/usb/storage/transport.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
> index 540add2..7cd45ac 100644
> --- a/drivers/usb/storage/transport.c
> +++ b/drivers/usb/storage/transport.c
> @@ -1111,7 +1111,7 @@ int usb_stor_Bulk_transport(struct scsi_cmnd *srb, struct us_data *us)
> * command phase and the data phase. Some devices need a little
> * more than that, probably because of clock rate inaccuracies. */
> if (unlikely(us->fflags & US_FL_GO_SLOW))
> - udelay(125);
> + usleep_range(100, 125);
>
> if (transfer_length) {
> unsigned int pipe = srb->sc_data_direction == DMA_FROM_DEVICE ?
> --
> 2.1.4
>
[-- Attachment #2: 0001-usb-usleep_range-is-preferred-over-udelay-where-wake.patch --]
[-- Type: text/x-diff, Size: 1202 bytes --]
>From 89648c1532e860b986fc5286245c404ddbe45b35 Mon Sep 17 00:00:00 2001
From: Sunny Kumar <sunny.kumar.roy@gmail.com>
Date: Wed, 24 Jun 2015 19:50:30 +0530
Subject: [PATCH 1/1] usb: usleep_range is preferred over udelay where wakeup
is flexible
According to Documentation/timers/timers-howto.txt"
udelay() is only called once from a place where sleeping is allowed.
We can replace it with a call to usleep_range()
with a reasonable upper limit.
Signed-off-by: Sunny Kumar <sunny.kumar.roy@gmail.com>
---
drivers/usb/storage/transport.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
index 540add2..7e397f8 100644
--- a/drivers/usb/storage/transport.c
+++ b/drivers/usb/storage/transport.c
@@ -1111,7 +1111,7 @@ int usb_stor_Bulk_transport(struct scsi_cmnd *srb, struct us_data *us)
* command phase and the data phase. Some devices need a little
* more than that, probably because of clock rate inaccuracies. */
if (unlikely(us->fflags & US_FL_GO_SLOW))
- udelay(125);
+ usleep_range(125, 200);
if (transfer_length) {
unsigned int pipe = srb->sc_data_direction == DMA_FROM_DEVICE ?
--
2.1.4
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH 1/1] usb: usleep_range is preferred over udelay where wakeup is flexible
2015-06-24 6:54 [PATCH 1/1] usb: usleep_range is preferred over udelay where wakeup is flexible Sunny Kumar
2015-06-24 13:44 ` Alan Stern
2015-06-24 14:28 ` Sunny Kumar
@ 2015-06-24 15:45 ` Sunny Kumar
2015-06-25 15:30 ` Sunny Kumar
3 siblings, 0 replies; 6+ messages in thread
From: Sunny Kumar @ 2015-06-24 15:45 UTC (permalink / raw)
To: mdharm-usb, gregkh, linux-usb, usb-storage, linux-kernel
[-- Attachment #1: Type: text/plain, Size: 1144 bytes --]
Ok then,
This might be the final lower and upper limit value.
Sunny Kumar
On Wed, Jun 24, 2015 at 12:24:29PM +0530, Sunny Kumar wrote:
> According to Documentation/timers/timers-howto.txt"
> udelay() is only called once from a place where sleeping is allowed.
> We can replace it with a call to usleep_range()
> with a reasonable upper limit.
>
> Signed-off-by: Sunny Kumar <sunny.kumar.roy@gmail.com>
> ---
> drivers/usb/storage/transport.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
> index 540add2..7cd45ac 100644
> --- a/drivers/usb/storage/transport.c
> +++ b/drivers/usb/storage/transport.c
> @@ -1111,7 +1111,7 @@ int usb_stor_Bulk_transport(struct scsi_cmnd *srb, struct us_data *us)
> * command phase and the data phase. Some devices need a little
> * more than that, probably because of clock rate inaccuracies. */
> if (unlikely(us->fflags & US_FL_GO_SLOW))
> - udelay(125);
> + usleep_range(100, 125);
>
> if (transfer_length) {
> unsigned int pipe = srb->sc_data_direction == DMA_FROM_DEVICE ?
> --
> 2.1.4
>
[-- Attachment #2: 0001-usb-usleep_range-is-preferred-over-udelay-where-wake.patch --]
[-- Type: text/x-diff, Size: 1202 bytes --]
>From 62bc3514ec4da51e7e766199d00de0594b17d23b Mon Sep 17 00:00:00 2001
From: Sunny Kumar <sunny.kumar.roy@gmail.com>
Date: Wed, 24 Jun 2015 21:07:18 +0530
Subject: [PATCH 1/1] usb: usleep_range is preferred over udelay where wakeup
is flexible
According to Documentation/timers/timers-howto.txt"
udelay() is only called once from a place where sleeping is allowed.
We can replace it with a call to usleep_range()
with a reasonable upper limit.
Signed-off-by: Sunny Kumar <sunny.kumar.roy@gmail.com>
---
drivers/usb/storage/transport.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
index 540add2..5e67f63 100644
--- a/drivers/usb/storage/transport.c
+++ b/drivers/usb/storage/transport.c
@@ -1111,7 +1111,7 @@ int usb_stor_Bulk_transport(struct scsi_cmnd *srb, struct us_data *us)
* command phase and the data phase. Some devices need a little
* more than that, probably because of clock rate inaccuracies. */
if (unlikely(us->fflags & US_FL_GO_SLOW))
- udelay(125);
+ usleep_range(125, 150);
if (transfer_length) {
unsigned int pipe = srb->sc_data_direction == DMA_FROM_DEVICE ?
--
1.9.1
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH 1/1] usb: usleep_range is preferred over udelay where wakeup is flexible
2015-06-24 6:54 [PATCH 1/1] usb: usleep_range is preferred over udelay where wakeup is flexible Sunny Kumar
` (2 preceding siblings ...)
2015-06-24 15:45 ` Sunny Kumar
@ 2015-06-25 15:30 ` Sunny Kumar
3 siblings, 0 replies; 6+ messages in thread
From: Sunny Kumar @ 2015-06-25 15:30 UTC (permalink / raw)
To: mdharm-usb, gregkh, linux-usb, usb-storage, linux-kernel; +Cc: Sunny Kumar
According to Documentation/timers/timers-howto.txt"
udelay() is only called once from a place where sleeping is allowed.
We can replace it with a call to usleep_range()
with a reasonable upper limit.
Signed-off-by: Sunny Kumar <sunny.kumar.roy@gmail.com>
---
drivers/usb/storage/transport.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
index 540add2..5e67f63 100644
--- a/drivers/usb/storage/transport.c
+++ b/drivers/usb/storage/transport.c
@@ -1111,7 +1111,7 @@ int usb_stor_Bulk_transport(struct scsi_cmnd *srb, struct us_data *us)
* command phase and the data phase. Some devices need a little
* more than that, probably because of clock rate inaccuracies. */
if (unlikely(us->fflags & US_FL_GO_SLOW))
- udelay(125);
+ usleep_range(125, 150);
if (transfer_length) {
unsigned int pipe = srb->sc_data_direction == DMA_FROM_DEVICE ?
--
1.9.1
^ permalink raw reply related [flat|nested] 6+ messages in thread