Skip to contentSkip to conversation

Renesas ARM based SoC fixes for v3.10

GIT PULL New Simon Horman 6 participants 17 messages May 29, 2013 – Jun 7, 2013

Subsystems (5)
lore
Share & tools
lore
#0 lore raw

The following changes since commit c7788792a5e7b0d5d7f96d0766b4cb6112d47d75:

Linux 3.10-rc2 (2013-05-20 14:37:38 -0700)

are available in the git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git

for you to fetch changes up to c7f413548dce3d25609c12efe963b3e6c8e678f3:

ARM: shmobile: sh73a0: Update CMT clockevent rating to 80 (2013-05-28 14:46:38 +0900)

----------------------------------------------------------------
Renesas ARM based SoC fixes for v3.10

* Correction to USB OVC and PENC pin groupings on r8a7779 SoC. This avoids conflicts when the USB_OVCn pins are used by another function. This has been observed to be a problem in v3.10-rc1.

* Update CMT clock rating for sh73a0 SoC to resolve boot failure on kzm9g-reference. This resolves a regression between v3.9 and v3.10-rc1.

----------------------------------------------------------------
Laurent Pinchart (1):
sh-pfc: r8a7779: Don't group USB OVC and PENC pins

Simon Horman (1):
ARM: shmobile: sh73a0: Update CMT clockevent rating to 80

 arch/arm/mach-shmobile/setup-sh73a0.c |    2 +-
 drivers/pinctrl/sh-pfc/pfc-r8a7779.c  |   45 ++++++++++++++++++++++++++-------
 2 files changed, 37 insertions(+), 10 deletions(-)
#1 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

From: Laurent Pinchart <[email protected]>

The USB_OVCn pins are alternate options for USB over-current detection when using a 3.3V USB interface. As they're not mandatory they can be used independently of the USB PENC pins. Don't group the USB_OVCn and PENC pins to avoid conflicts when the USB_OVCn pins are used by another function.

---
 drivers/pinctrl/sh-pfc/pfc-r8a7779.c |   45 +++++++++++++++++++++++++++-------
 1 file changed, 36 insertions(+), 9 deletions(-)
drivers/pinctrl/sh-pfc/pfc-r8a7779.c +36 -9
index 791a671..8cd90e7 100644
@@ -2357,27 +2357,48 @@ static const unsigned int sdhi3_wp_mux[] = {
 };
 /* - USB0 ------------------------------------------------------------------- */
 static const unsigned int usb0_pins[] = {
-	/* OVC */
-	150, 154,
+	/* PENC */
+	154,
 };
 static const unsigned int usb0_mux[] = {
-	USB_OVC0_MARK, USB_PENC0_MARK,
+	USB_PENC0_MARK,
+};
+static const unsigned int usb0_ovc_pins[] = {
+	/* USB_OVC */
+	150
+};
+static const unsigned int usb0_ovc_mux[] = {
+	USB_OVC0_MARK,
 };
 /* - USB1 ------------------------------------------------------------------- */
 static const unsigned int usb1_pins[] = {
-	/* OVC */
-	152, 155,
+	/* PENC */
+	155,
 };
 static const unsigned int usb1_mux[] = {
-	USB_OVC1_MARK, USB_PENC1_MARK,
+	USB_PENC1_MARK,
+};
+static const unsigned int usb1_ovc_pins[] = {
+	/* USB_OVC */
+	152,
+};
+static const unsigned int usb1_ovc_mux[] = {
+	USB_OVC1_MARK,
 };
 /* - USB2 ------------------------------------------------------------------- */
 static const unsigned int usb2_pins[] = {
-	/* OVC, PENC */
-	125, 156,
+	/* PENC */
+	156,
 };
 static const unsigned int usb2_mux[] = {
-	USB_OVC2_MARK, USB_PENC2_MARK,
+	USB_PENC2_MARK,
+};
+static const unsigned int usb2_ovc_pins[] = {
+	/* USB_OVC */
+	125,
+};
+static const unsigned int usb2_ovc_mux[] = {
+	USB_OVC2_MARK,
 };
 
 static const struct sh_pfc_pin_group pinmux_groups[] = {
@@ -2501,8 +2522,11 @@ static const struct sh_pfc_pin_group pinmux_groups[] = {
 	SH_PFC_PIN_GROUP(sdhi3_cd),
 	SH_PFC_PIN_GROUP(sdhi3_wp),
 	SH_PFC_PIN_GROUP(usb0),
+	SH_PFC_PIN_GROUP(usb0_ovc),
 	SH_PFC_PIN_GROUP(usb1),
+	SH_PFC_PIN_GROUP(usb1_ovc),
 	SH_PFC_PIN_GROUP(usb2),
+	SH_PFC_PIN_GROUP(usb2_ovc),
 };
 
 static const char * const du0_groups[] = {
@@ -2683,14 +2707,17 @@ static const char * const sdhi3_groups[] = {
 
 static const char * const usb0_groups[] = {
 	"usb0",
+	"usb0_ovc",
 };
 
 static const char * const usb1_groups[] = {
 	"usb1",
+	"usb1_ovc",
 };
 
 static const char * const usb2_groups[] = {
 	"usb2",
+	"usb2_ovc",
 };
 
 static const struct sh_pfc_function pinmux_functions[] = {
Signature · 1 line
1.7.10.4
#2 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

Hello.

Quoted 11 lines · Simon Horman

From: Laurent Pinchart <[email protected]>

The USB_OVCn pins are alternate options for USB over-current detection when using a 3.3V USB interface. As they're not mandatory they can be used independently of the USB PENC pins. Don't group the USB_OVCn and PENC pins to avoid conflicts when the USB_OVCn pins are used by another function.

I suspect this is needed in -stable. The bug is older than 3.10-rc1.

WBR, Sergei

#3 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

Hi Sergei,

Quoted 9 lines · On Thursday 30 May 2013 02:35:51 Sergei Shtylyov wrote:
Quoted 8 lines · Simon Horman

From: Laurent Pinchart <[email protected]>

The USB_OVCn pins are alternate options for USB over-current detection when using a 3.3V USB interface. As they're not mandatory they can be used independently of the USB PENC pins. Don't group the USB_OVCn and PENC pins to avoid conflicts when the USB_OVCn pins are used by another function.

Quoted 6 lines
Quoted 4 lines
Signed-off-by Laurent Pinchart

<[email protected]>

I suspect this is needed in -stable. The bug is older than 3.10-rc1.

Is it ? The offending commit is

commit 97d40c4224172451f666febdd865c24b1c3c3fe5

Author: Laurent Pinchart <[email protected]>
Date:   Thu Mar 7 13:38:51 2013 +0100

sh-pfc: r8a7779: Add USB pin groups and functions

Signed-off-by: Laurent Pinchart
<[email protected]>

and got applied to mainline in v3.10-rc1.

Signature · 3 lines
Regards,

Laurent Pinchart
#4 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

Hello.

Quoted 21 lines · Laurent Pinchart
Quoted 8 lines
Quoted 7 lines

From: Laurent Pinchart <[email protected]>

The USB_OVCn pins are alternate options for USB over-current detection when using a 3.3V USB interface. As they're not mandatory they can be used independently of the USB PENC pins. Don't group the USB_OVCn and PENC pins to avoid conflicts when the USB_OVCn pins are used by another function.

Quoted 6 lines
Quoted 4 lines
Signed-off-by Laurent Pinchart

<[email protected]>

I suspect this is needed in -stable. The bug is older than 3.10-rc1.

Is it ? The offending commit is

commit 97d40c4224172451f666febdd865c24b1c3c3fe5

Author: Laurent Pinchart <[email protected]>
Date:   Thu Mar 7 13:38:51 2013 +0100
Quoted 7 lines

sh-pfc: r8a7779: Add USB pin groups and functions

Signed-off-by: Laurent Pinchart
<[email protected]>

and got applied to mainline in v3.10-rc1.

Ah, then sorry for the false alarm. Consequency of working in the
'next' branch all the time...

WBR, Sergei

#5 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

In reply to Simon Horman
Quoted 11 lines · Simon Horman

From: Laurent Pinchart <[email protected]>

The USB_OVCn pins are alternate options for USB over-current detection when using a 3.3V USB interface. As they're not mandatory they can be used independently of the USB PENC pins. Don't group the USB_OVCn and PENC pins to avoid conflicts when the USB_OVCn pins are used by another function.

OK, Acked-by, do you want me to take this as a fix in the pinctrl tree for v3.10?

Just asking because Simon usually takes everything else relating to the SH. ...

Yours,
Linus Walleij

#6 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

Quoted 15 lines · Linus Walleij
Quoted 11 lines · Simon Horman

From: Laurent Pinchart <[email protected]>

The USB_OVCn pins are alternate options for USB over-current detection when using a 3.3V USB interface. As they're not mandatory they can be used independently of the USB PENC pins. Don't group the USB_OVCn and PENC pins to avoid conflicts when the USB_OVCn pins are used by another function.

Quoted 5 lines

OK, Acked-by, do you want me to take this as a fix in the pinctrl tree for v3.10?

Just asking because Simon usually takes everything else relating to the SH. ...

This was a pull request from my tree to arm-soc, so thats they way I was planning to handle things.

Olof, Arnd,

should I re-spin the pull request to include Linus's ack?

#7 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

Hi,

Quoted 22 lines · Simon Horman
Quoted 15 lines · Linus Walleij
Quoted 11 lines · Simon Horman

From: Laurent Pinchart <[email protected]>

The USB_OVCn pins are alternate options for USB over-current detection when using a 3.3V USB interface. As they're not mandatory they can be used independently of the USB PENC pins. Don't group the USB_OVCn and PENC pins to avoid conflicts when the USB_OVCn pins are used by another function.

Quoted 5 lines

OK, Acked-by, do you want me to take this as a fix in the pinctrl tree for v3.10?

Just asking because Simon usually takes everything else relating to the SH. ...

Quoted 6 lines

This was a pull request from my tree to arm-soc, so thats they way I was planning to handle things.

Olof, Arnd,

should I re-spin the pull request to include Linus's ack?

You can if you want, but for an odd and specific fix like this we'll all know he's ok with the code going in.

-Olof

#8 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

Quoted 28 lines · Olof Johansson

Hi,

Quoted 25 lines · Simon Horman
Quoted 15 lines · Linus Walleij
Quoted 11 lines · Simon Horman

From: Laurent Pinchart <[email protected]>

The USB_OVCn pins are alternate options for USB over-current detection when using a 3.3V USB interface. As they're not mandatory they can be used independently of the USB PENC pins. Don't group the USB_OVCn and PENC pins to avoid conflicts when the USB_OVCn pins are used by another function.

Quoted 5 lines

OK, Acked-by, do you want me to take this as a fix in the pinctrl tree for v3.10?

Just asking because Simon usually takes everything else relating to the SH. ...

This was a pull request from my tree to arm-soc, so thats they way I was planning to handle things.

Quoted 6 lines
Quoted 3 lines

Olof, Arnd,

should I re-spin the pull request to include Linus's ack?

You can if you want, but for an odd and specific fix like this we'll all know he's ok with the code going in.

In that case my preference is to leave the pull request as is. Could you consider merging it?

As an aside, I do expect to send some more v3.10 fixes for shmobile. There seem to be more than usual.

#9 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

Quoted 31 lines · Simon Horman
Quoted 30 lines · Olof Johansson

Hi,

Quoted 27 lines · Simon Horman
Quoted 15 lines · Linus Walleij
Quoted 11 lines · Simon Horman

From: Laurent Pinchart <[email protected]>

The USB_OVCn pins are alternate options for USB over-current detection when using a 3.3V USB interface. As they're not mandatory they can be used independently of the USB PENC pins. Don't group the USB_OVCn and PENC pins to avoid conflicts when the USB_OVCn pins are used by another function.

Quoted 5 lines

OK, Acked-by, do you want me to take this as a fix in the pinctrl tree for v3.10?

Just asking because Simon usually takes everything else relating to the SH. ...

This was a pull request from my tree to arm-soc, so thats they way I was planning to handle things.

Olof, Arnd,

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

should I re-spin the pull request to include Linus's ack?

You can if you want, but for an odd and specific fix like this we'll all know he's ok with the code going in.

In that case my preference is to leave the pull request as is. Could you consider merging it?

Yes, I will do a pass tomorrow and handle the outstanding merges.

Quoted 2 lines

As an aside, I do expect to send some more v3.10 fixes for shmobile. There seem to be more than usual.

A little unfortunate but it's hard to do much about.

Please be conservative with what you pick up for 3.10 vs 3.11 -- Linus has said he really wants to see things slow down.

And please make our case as easy as possible by making sure it's pretty clear from the patch descriptions what kind of bugs they fix (and if they are regressions).

-Olof

#10 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

Quoted 8 lines · Olof Johansson

[Simon]

Quoted 2 lines

As an aside, I do expect to send some more v3.10 fixes for shmobile. There seem to be more than usual.

A little unfortunate but it's hard to do much about.

Please be conservative with what you pick up for 3.10 vs 3.11 -- Linus has said he really wants to see things slow down.

What we might need to think about is the evolutional pace of the ARM kernel - compared to a few years back we have increased the change rate by orders of a magnitude (at least that is my feeling...)

There may be some point where we actually need to tell people to hold changes back because they are just pushing too large volumes through one merge window, resulting in a corresponding amount of fixes.

This merge window both Samsung Exynos and SH mobile seem to have been a little bit trigger-happy :-/

Yours,
Linus Walleij

#11 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

Quoted 14 lines · Linus Walleij
Quoted 8 lines · Olof Johansson

[Simon]

Quoted 2 lines

As an aside, I do expect to send some more v3.10 fixes for shmobile. There seem to be more than usual.

A little unfortunate but it's hard to do much about.

Please be conservative with what you pick up for 3.10 vs 3.11 -- Linus has said he really wants to see things slow down.

What we might need to think about is the evolutional pace of the ARM kernel - compared to a few years back we have increased the change rate by orders of a magnitude (at least that is my feeling...)

Quoted 7 lines

There may be some point where we actually need to tell people to hold changes back because they are just pushing too large volumes through one merge window, resulting in a corresponding amount of fixes.

This merge window both Samsung Exynos and SH mobile seem to have been a little bit trigger-happy :-/

To quantify things, for shmobile I am currently aware of 4 problems that seem to be worthy of fixing for v3.10. A patche for one has been accepted into arm-soc, Patches for two are this pull request. And one more has a patch in the pipeline.

These known problems seem to fall into two categories:

1. Problems that have been in linux-next for a while but only recently appeared in the renesas tree. These were observed when boot testing kernels during the course of rebasing the renesas tree on v3.10-rcX.

I believe that these problems could and should have been spotted earlier by boot testing linux-next more often, allowing them to be resolved in the merge-window rather than at -rc time.

I intend to make this happen.

2. Problems that have been present in the renesas tree and thus linux-next for a while but were not observed as early as they might have been.

Statistically the rate of these problems could have been reduced by accepting changes more slowly. But at this point we are talking about one or two bugs out of many many changes. Bugs that only affect a small subset of shmobile hardware support.

#12 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

In reply to Linus Walleij
Quoted 5 lines · Linus Walleij
Quoted 4 lines · Olof Johansson

[Simon]

Quoted 2 lines

As an aside, I do expect to send some more v3.10 fixes for shmobile. There seem to be more than usual.

Quoted 8 lines
Quoted 4 lines

A little unfortunate but it's hard to do much about.

Please be conservative with what you pick up for 3.10 vs 3.11 -- Linus has said he really wants to see things slow down.

What we might need to think about is the evolutional pace of the ARM kernel - compared to a few years back we have increased the change rate by orders of a magnitude (at least that is my feeling...)

Oh, very much so. And there's a lot of additions of new code.

Linus clarified just this week at LC Japan what ticks him off about -rc patches too; it's an indication that people aren't sending code that's baked and tested enough during the merge window, causing all these fixes on top.

Given the rate of change of arm platforms, some of this is definitely expected (i.e. proportional to amount of merge window changes). Still, we have empirical evidence that far from all maintainers keep a close eye on linux-next status for their platforms. :-)

I have a small board farm that I make sure boot more or less daily with linux-next now, but my space is limited. It also shouldn't be my responsibility to make sure that various platforms keep working; it's up to each submaintainer to do so.

That doesn't mean there won't be corner cases still, hardware that maintainers don't have themselves affected, etc. But by keeping an eye on linux-next, a handful of -rc fixes can be avoided.

Quoted 4 lines

There may be some point where we actually need to tell people to hold changes back because they are just pushing too large volumes through one merge window, resulting in a corresponding amount of fixes.

That's essentially what we did with the initial multiplatform changes for 3.10 for Exynos, where they seemed to need more bake time and got held off.

Quoted 2 lines

This merge window both Samsung Exynos and SH mobile seem to have been a little bit trigger-happy :-/

Both of them saw a lot of changes. Exynos saw the onslaught of Linaro engineers starting to work more on Arndale, and shmobile saw a huge amount of changes due to their cutting over subsystems. The latter is hard to avoid since it can be complicated to keep both old and new infrastructure around, and the former we should just make sure to catch with linux-next testing.

But you do bring up valid points, and it's something to keep in mind going forward.

-Olof

#13 lore raw

PATCH 1/2 sh-pfc: r8a7779: Don't group USB OVC and PENC pins

In reply to Olof Johansson
Quoted 45 lines · Olof Johansson
Quoted 31 lines · Simon Horman
Quoted 30 lines · Olof Johansson

Hi,

Quoted 27 lines · Simon Horman
Quoted 15 lines · Linus Walleij
Quoted 11 lines · Simon Horman

From: Laurent Pinchart <[email protected]>

The USB_OVCn pins are alternate options for USB over-current detection when using a 3.3V USB interface. As they're not mandatory they can be used independently of the USB PENC pins. Don't group the USB_OVCn and PENC pins to avoid conflicts when the USB_OVCn pins are used by another function.

Quoted 5 lines

OK, Acked-by, do you want me to take this as a fix in the pinctrl tree for v3.10?

Just asking because Simon usually takes everything else relating to the SH. ...

This was a pull request from my tree to arm-soc, so thats they way I was planning to handle things.

Olof, Arnd,

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

should I re-spin the pull request to include Linus's ack?

You can if you want, but for an odd and specific fix like this we'll all know he's ok with the code going in.

In that case my preference is to leave the pull request as is. Could you consider merging it?

Yes, I will do a pass tomorrow and handle the outstanding merges.

Quoted 2 lines

As an aside, I do expect to send some more v3.10 fixes for shmobile. There seem to be more than usual.

Quoted 8 lines

A little unfortunate but it's hard to do much about.

Please be conservative with what you pick up for 3.10 vs 3.11 -- Linus has said he really wants to see things slow down.

And please make our case as easy as possible by making sure it's pretty clear from the patch descriptions what kind of bugs they fix (and if they are regressions).

Understood. I will be careful to only send the bare essentials for v3.10.

#14 lore raw

PATCH 2/2 ARM: shmobile: sh73a0: Update CMT clockevent rating to 80

In reply to Simon Horman

Update the CMT clockevent rating from 125 to 80.

This resolves a boot-failure regression for kzm9g-reference in v3.10-rc1 introduced by f7db706b132f11c79ae1d74b2382e0926cf31644 ("ARM: 7674/1: smp:
Avoid dummy clockevent being preferred over real").

The patch noted above reduces the rating of dummy clockevent from 400 to 100. This patch reduces the rating of CMT so that it is once again less than that of the dummy clockevent.

---
 arch/arm/mach-shmobile/setup-sh73a0.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/mach-shmobile/setup-sh73a0.c +1 -1
index fdf3894..9696f36 100644
@@ -252,7 +252,7 @@ static struct sh_timer_config cmt10_platform_data = {
 	.name = "CMT10",
 	.channel_offset = 0x10,
 	.timer_bit = 0,
-	.clockevent_rating = 125,
+	.clockevent_rating = 80,
 	.clocksource_rating = 125,
 };
 
Signature · 1 line
1.7.10.4
#15 lore raw
In reply to Simon Horman
Quoted 7 lines · Simon Horman

The following changes since commit c7788792a5e7b0d5d7f96d0766b4cb6112d47d75:

Linux 3.10-rc2 (2013-05-20 14:37:38 -0700)

are available in the git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git

Hi, no tag or branch? Looks like you forgot to push.

I don't see any branch or tag at c7f413548dce3d25609c12efe963b3e6c8e678f3 in that tree, so it's not a case of pushing branch but forgetting tag.

-Olof

#16 lore raw
Quoted 5 lines · Olof Johansson
Quoted 4 lines · Simon Horman

The following changes since commit c7788792a5e7b0d5d7f96d0766b4cb6112d47d75:

Linux 3.10-rc2 (2013-05-20 14:37:38 -0700)

Quoted 8 lines
Quoted 3 lines

are available in the git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git

Hi, no tag or branch? Looks like you forgot to push.

I don't see any branch or tag at c7f413548dce3d25609c12efe963b3e6c8e678f3 in that tree, so it's not a case of pushing branch but forgetting tag.

Sorry about that. I have re-pushed and sent a fresh pull request.

For reference it is revision f5dbcc6e917f12e306acfc9575d437b5d5e24134 tagged as renesas-fixes-for-v3.10.