[PATCH] HID: intel_ish-hid: use %pUL for uuid formatting

[PATCH] HID: intel_ish-hid: use %pUL for uuid formatting

[Rasmus Villemoes]

We have the %pU printf extension for doing exactly this. Saves some
.text, and is likely also a little faster.

Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
---
 drivers/hid/intel-ish-hid/ishtp/bus.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/drivers/hid/intel-ish-hid/ishtp/bus.c b/drivers/hid/intel-ish-hid/ishtp/bus.c
index 256521509d20..f4cbc744e657 100644
--- a/drivers/hid/intel-ish-hid/ishtp/bus.c
+++ b/drivers/hid/intel-ish-hid/ishtp/bus.c
@@ -585,14 +585,7 @@ int ishtp_bus_new_client(struct ishtp_device *dev)
 	 */
 	i = dev->fw_client_presentation_num - 1;
 	device_uuid = dev->fw_clients[i].props.protocol_name;
-	dev_name = kasprintf(GFP_KERNEL,
-		"{%02X%02X%02X%02X-%02X%02X-%02X%02X-%02X%02X-%02X%02X%02X%02X%02X%02X}",
-		device_uuid.b[3], device_uuid.b[2], device_uuid.b[1],
-		device_uuid.b[0], device_uuid.b[5], device_uuid.b[4],
-		device_uuid.b[7], device_uuid.b[6], device_uuid.b[8],
-		device_uuid.b[9], device_uuid.b[10], device_uuid.b[11],
-		device_uuid.b[12], device_uuid.b[13], device_uuid.b[14],
-		device_uuid.b[15]);
+	dev_name = kasprintf(GFP_KERNEL, "{%pUL}", device_uuid.b);
 	if (!dev_name)
 		return	-ENOMEM;
 
-- 
2.1.4

Re: [PATCH] HID: intel_ish-hid: use %pUL for uuid formatting

[Benjamin Tissoires]

On Nov 30 2016 or thereabouts, Rasmus Villemoes wrote:
> We have the %pU printf extension for doing exactly this. Saves some
> .text, and is likely also a little faster.
> 
> Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>

Reviewed-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>

> ---
>  drivers/hid/intel-ish-hid/ishtp/bus.c | 9 +--------
>  1 file changed, 1 insertion(+), 8 deletions(-)
> 
> diff --git a/drivers/hid/intel-ish-hid/ishtp/bus.c b/drivers/hid/intel-ish-hid/ishtp/bus.c
> index 256521509d20..f4cbc744e657 100644
> --- a/drivers/hid/intel-ish-hid/ishtp/bus.c
> +++ b/drivers/hid/intel-ish-hid/ishtp/bus.c
> @@ -585,14 +585,7 @@ int ishtp_bus_new_client(struct ishtp_device *dev)
>  	 */
>  	i = dev->fw_client_presentation_num - 1;
>  	device_uuid = dev->fw_clients[i].props.protocol_name;
> -	dev_name = kasprintf(GFP_KERNEL,
> -		"{%02X%02X%02X%02X-%02X%02X-%02X%02X-%02X%02X-%02X%02X%02X%02X%02X%02X}",
> -		device_uuid.b[3], device_uuid.b[2], device_uuid.b[1],
> -		device_uuid.b[0], device_uuid.b[5], device_uuid.b[4],
> -		device_uuid.b[7], device_uuid.b[6], device_uuid.b[8],
> -		device_uuid.b[9], device_uuid.b[10], device_uuid.b[11],
> -		device_uuid.b[12], device_uuid.b[13], device_uuid.b[14],
> -		device_uuid.b[15]);
> +	dev_name = kasprintf(GFP_KERNEL, "{%pUL}", device_uuid.b);
>  	if (!dev_name)
>  		return	-ENOMEM;
>  
> -- 
> 2.1.4
> 

Re: [PATCH] HID: intel_ish-hid: use %pUL for uuid formatting

[Srinivas Pandruvada]

On Thu, 2016-12-01 at 15:21 +0100, Benjamin Tissoires wrote:
> On Nov 30 2016 or thereabouts, Rasmus Villemoes wrote:
> > 
> > We have the %pU printf extension for doing exactly this. Saves some
> > .text, and is likely also a little faster.
> > 
> > Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
> 
> Reviewed-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Acked-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>

> 
> > 
> > ---
> >  drivers/hid/intel-ish-hid/ishtp/bus.c | 9 +--------
> >  1 file changed, 1 insertion(+), 8 deletions(-)
> > 
> > diff --git a/drivers/hid/intel-ish-hid/ishtp/bus.c
> > b/drivers/hid/intel-ish-hid/ishtp/bus.c
> > index 256521509d20..f4cbc744e657 100644
> > --- a/drivers/hid/intel-ish-hid/ishtp/bus.c
> > +++ b/drivers/hid/intel-ish-hid/ishtp/bus.c
> > @@ -585,14 +585,7 @@ int ishtp_bus_new_client(struct ishtp_device
> > *dev)
> >  	 */
> >  	i = dev->fw_client_presentation_num - 1;
> >  	device_uuid = dev->fw_clients[i].props.protocol_name;
> > -	dev_name = kasprintf(GFP_KERNEL,
> > -		"{%02X%02X%02X%02X-%02X%02X-%02X%02X-%02X%02X-
> > %02X%02X%02X%02X%02X%02X}",
> > -		device_uuid.b[3], device_uuid.b[2],
> > device_uuid.b[1],
> > -		device_uuid.b[0], device_uuid.b[5],
> > device_uuid.b[4],
> > -		device_uuid.b[7], device_uuid.b[6],
> > device_uuid.b[8],
> > -		device_uuid.b[9], device_uuid.b[10],
> > device_uuid.b[11],
> > -		device_uuid.b[12], device_uuid.b[13],
> > device_uuid.b[14],
> > -		device_uuid.b[15]);
> > +	dev_name = kasprintf(GFP_KERNEL, "{%pUL}", device_uuid.b);
> >  	if (!dev_name)
> >  		return	-ENOMEM;
> >  
> > -- 
> > 2.1.4
> > 

Re: [PATCH] HID: intel_ish-hid: use %pUL for uuid formatting

[Jiri Kosina]

On Wed, 30 Nov 2016, Rasmus Villemoes wrote:

> We have the %pU printf extension for doing exactly this. Saves some
> .text, and is likely also a little faster.
> 
> Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>

Applied, thanks.

-- 
Jiri Kosina
SUSE Labs

Issue in applying patch [hid-sensor-hub: clear memory to avoid random data]

[Srinivas Pandruvada]

Hi Jiri,

As per the comment here, it seems that you applied v4 version of patch.

https://patchwork.kernel.org/patch/9428715/

But when I look at the upstream kernel, the very first version of the
patch was applied, which was wrong.

Basically the change should be in sensor_hub_get_feature() not in
sensor_hub_set_feature().
Otherwise it breaks sensor functionality.

How should we handle this?


Thanks,
Srinivas

Re: Issue in applying patch [hid-sensor-hub: clear memory to avoid random data]

[Jiri Kosina]

On Wed, 14 Dec 2016, Srinivas Pandruvada wrote:

> As per the comment here, it seems that you applied v4 version of patch.
> 
> https://patchwork.kernel.org/patch/9428715/
> 
> But when I look at the upstream kernel, the very first version of the
> patch was applied, which was wrong.
> 
> Basically the change should be in sensor_hub_get_feature() not in
> sensor_hub_set_feature().
> Otherwise it breaks sensor functionality.
> 
> How should we handle this?

Hi Srinivas,

my whole queue is now merged by Linus, so any fixups should be submitted 
as standalone patches on top of current Linus' tree; I'll push the fix to 
Linus then for 4.10 still.

Ideally we also put proper "Fixes:" tag, so that it gets picked up by 
4.9-stable once/if there is one.

Thanks, and sorry for my brainfart; I'll check my archives to see what 
went wrong.

-- 
Jiri Kosina
SUSE Labs

Re: Issue in applying patch [hid-sensor-hub: clear memory to avoid random data]

[Srinivas Pandruvada]

On Wed, 2016-12-14 at 21:33 +0100, Jiri Kosina wrote:
> On Wed, 14 Dec 2016, Srinivas Pandruvada wrote:
> 
> > As per the comment here, it seems that you applied v4 version of
> patch.
> > 
> > https://patchwork.kernel.org/patch/9428715/
> > 
> > But when I look at the upstream kernel, the very first version of
> the
> > patch was applied, which was wrong.
> > 
> > Basically the change should be in sensor_hub_get_feature() not in
> > sensor_hub_set_feature().
> > Otherwise it breaks sensor functionality.
> > 
> > How should we handle this?
> 
> Hi Srinivas,
> 
> my whole queue is now merged by Linus, so any fixups should be
> submitted 
> as standalone patches on top of current Linus' tree; I'll push the
> fix to 
> Linus then for 4.10 still.
> 
> Ideally we also put proper "Fixes:" tag, so that it gets picked up
> by 
> 4.9-stable once/if there is one.
> 
> Thanks, and sorry for my brainfart; I'll check my archives to see
> what 
> went wrong.
No problem at all.

Thanks,
Srinivas