ARM: pxa: fix regulator related build fail in magician_defconfig

PATCH Superseded Paul Gortmaker 5 participants 16 messages Mar 6, 2012 – Mar 28, 2012

Subsystems (4)

Arm (32-bit) Arm sub-architectures Linux-next tree PXA2xx/PXA3xx support

lore

Share & tools

lore

Paul Gortmaker

author

2012-03-06 19:46 UTC #0 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

commit 737f360d5bef5e01c6cfa755dca0b449a154c1e0 (linux-next)

"regulator: Remove support for supplies specified by struct device"

deletes the field from the struct, but this platform was still trying to set those fields. Delete the assignments.

arch/arm/mach-pxa/magician.c:581: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:581: warning: initialization from incompatible pointer type arch/arm/mach-pxa/magician.c:585: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:585: warning: initialization from incompatible pointer type

arch/arm/mach-pxa/hx4700.c:683: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:683: warning: initialization from incompatible pointer type arch/arm/mach-pxa/hx4700.c:687: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:687: warning: initialization from incompatible pointer type

Signed-off-by Paul Gortmaker <[email protected]>

arch/arm/mach-pxa/hx4700.c +0 -2

index f2c23ea..76de30c 100644
@@ -680,11 +680,9 @@ static struct platform_device power_supply = {
 
 static struct regulator_consumer_supply bq24022_consumers[] = {
 	{
-		.dev = &gpio_vbus.dev,
 		.supply = "vbus_draw",
 	},
 	{
-		.dev = &power_supply.dev,
 		.supply = "ac_draw",
 	},
 };

arch/arm/mach-pxa/magician.c +0 -2

index 3d6baf9..bd0f86a 100644
@@ -578,11 +578,9 @@ static struct platform_device power_supply = {
 
 static struct regulator_consumer_supply bq24022_consumers[] = {
 	{
-		.dev = &gpio_vbus.dev,
 		.supply = "vbus_draw",
 	},
 	{
-		.dev = &power_supply.dev,
 		.supply = "ac_draw",
 	},
 };

Signature · 1 line

1.7.9.1

Mark Brown

2012-03-06 20:01 UTC #1 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

arch/arm/mach-pxa/hx4700.c static struct platform_device power_supply = { · line 683

Quoted 4 lines · Paul Gortmaker

static struct regulator_consumer_supply bq24022_consumers[] = {
arch/arm/mach-pxa/hx4700.c +0 -1
 	{
-		.dev = &gpio_vbus.dev,
 		.supply = "vbus_draw",

No, this isn't a good fix - it will break the user. You need to replace
dev with the equivalent dev_name.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20120306/b2636232/attachment.sig>

Paul Gortmaker

author

2012-03-06 20:49 UTC #2 lore raw

PATCH v2 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

commit 737f360d5bef5e01c6cfa755dca0b449a154c1e0 (linux-next)

"regulator: Remove support for supplies specified by struct device"

deletes the field from the struct, but this platform was still trying to set those fields. Convert them to use the REGULATOR_SUPPLY macro so as to fix this failure:

Signed-off-by Paul Gortmaker <[email protected]>

---

[v2: convert to using REGULATOR_SUPPLY instead of just deleting the
instances of setting the .dev field ]

arch/arm/mach-pxa/hx4700.c +2 -8

index f2c23ea..53c3366 100644
@@ -679,14 +679,8 @@ static struct platform_device power_supply = {
  */
 
 static struct regulator_consumer_supply bq24022_consumers[] = {
-	{
-		.dev = &gpio_vbus.dev,
-		.supply = "vbus_draw",
-	},
-	{
-		.dev = &power_supply.dev,
-		.supply = "ac_draw",
-	},
+	REGULATOR_SUPPLY("vbus_draw", "gpio_vbus"),
+	REGULATOR_SUPPLY("ac_draw", "power_supply"),
 };
 
 static struct regulator_init_data bq24022_init_data = {

arch/arm/mach-pxa/magician.c +2 -8

index 3d6baf9..71a3a3d 100644
@@ -577,14 +577,8 @@ static struct platform_device power_supply = {
  */
 
 static struct regulator_consumer_supply bq24022_consumers[] = {
-	{
-		.dev = &gpio_vbus.dev,
-		.supply = "vbus_draw",
-	},
-	{
-		.dev = &power_supply.dev,
-		.supply = "ac_draw",
-	},
+	REGULATOR_SUPPLY("vbus_draw", "gpio_vbus"),
+	REGULATOR_SUPPLY("ac_draw", "power_supply"),
 };
 
 static struct regulator_init_data bq24022_init_data = {

Signature · 1 line

1.7.9.1

Paul Parsons

2012-03-07 01:46 UTC #3 lore raw

PATCH v2 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

Quoted 56 lines · Paul Gortmaker

commit 737f360d5bef5e01c6cfa755dca0b449a154c1e0 (linux-next)
? "regulator: Remove support for supplies specified by struct device"
deletes the field from the struct, but this platform was still trying
to set those fields.? Convert them to use the REGULATOR_SUPPLY macro
so as to fix this failure:
arch/arm/mach-pxa/magician.c:581: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:581: warning: initialization from incompatible pointer type arch/arm/mach-pxa/magician.c:585: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:585: warning: initialization from incompatible pointer type
arch/arm/mach-pxa/hx4700.c:683: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:683: warning: initialization from incompatible pointer type arch/arm/mach-pxa/hx4700.c:687: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:687: warning: initialization from incompatible pointer type
Signed-off-by Paul Gortmaker <[email protected]>
---
[v2: convert to using REGULATOR_SUPPLY instead of just deleting the
instances of setting the .dev field ]
arch/arm/mach-pxa/hx4700.c +0 -0
index f2c23ea..53c3366 100644
@@ -679,14 +679,8 @@ static struct platform_device
 power_supply = {
? */
static struct regulator_consumer_supply bq24022_consumers[]
= {
-??? {
-??? ??? .dev =
&gpio_vbus.dev,
-??? ??? .supply =
"vbus_draw",
-??? },
-??? {
-??? ??? .dev =

Quoted 8 lines

&power_supply.dev,
-??? ??? .supply = "ac_draw",
-??? },
+??? REGULATOR_SUPPLY("vbus_draw",
"gpio_vbus"),
+??? REGULATOR_SUPPLY("ac_draw",
"power_supply"),
};

The above hx4700 patch yields the following warnings:

_regulator_get: gpio-vbus supply vbus_draw not found, using dummy regulator _regulator_get: pda-power supply ac_draw not found, using dummy regulator

I suggest the following lines instead:

	REGULATOR_SUPPLY("vbus_draw", "gpio-vbus"),
	REGULATOR_SUPPLY("ac_draw", "pda-power"),

Paul Gortmaker

author

2012-03-07 14:46 UTC #4 lore raw

PATCH v2 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

Quoted 68 lines · Paul Parsons

Quoted 60 lines · Paul Gortmaker
commit 737f360d5bef5e01c6cfa755dca0b449a154c1e0 (linux-next)
"regulator: Remove support for supplies specified by
struct device"
deletes the field from the struct, but this platform was still trying
to set those fields. Convert them to use the REGULATOR_SUPPLY macro
so as to fix this failure:
arch/arm/mach-pxa/magician.c:581: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:581: warning: initialization from incompatible pointer type arch/arm/mach-pxa/magician.c:585: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:585: warning: initialization from incompatible pointer type
arch/arm/mach-pxa/hx4700.c:683: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:683: warning: initialization from incompatible pointer type arch/arm/mach-pxa/hx4700.c:687: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:687: warning: initialization from incompatible pointer type
Signed-off-by Paul Gortmaker <[email protected]>
---
[v2: convert to using REGULATOR_SUPPLY instead of just deleting the
instances of setting the .dev field ]
arch/arm/mach-pxa/hx4700.c +1 -8
index f2c23ea..53c3366 100644
@@ -679,14 +679,8 @@ static struct platform_device
 power_supply = {
  */
 
 static struct regulator_consumer_supply bq24022_consumers[]
 = {
-    {
-        .dev =
 &gpio_vbus.dev,
-        .supply =
 "vbus_draw",
-    },
-    {
-        .dev =
 &power_supply.dev,
-        .supply = "ac_draw",
-    },
+    REGULATOR_SUPPLY("vbus_draw",
Quoted 4 lines
"gpio_vbus"),
+ REGULATOR_SUPPLY("ac_draw",
"power_supply"),
};
The above hx4700 patch yields the following warnings:

Quoted 7 lines

_regulator_get: gpio-vbus supply vbus_draw not found, using dummy regulator _regulator_get: pda-power supply ac_draw not found, using dummy regulator
I suggest the following lines instead:
	REGULATOR_SUPPLY("vbus_draw", "gpio-vbus"),
	REGULATOR_SUPPLY("ac_draw", "pda-power"),

Thanks for the report. The regulator code is rather foreign to me, but I think I figured out the expectations on the names last night when looking at a similar patch here:

http://www.spinics.net/lists/arm-kernel/msg163418.html

I'll fix these two patches up, but next time I'll just report the regulator regressions and let the sub-system maintainer or the platform maintainer sort them out.

Paul.

Quoted 1 line

Mark Brown

2012-03-07 14:57 UTC #5 lore raw

PATCH v2 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

Quoted 3 lines · Paul Gortmaker

Thanks for the report. The regulator code is rather foreign to me, but I think I figured out the expectations on the names last night when looking at a similar patch here:

All it's looking for is the output of dev_name() for the device.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20120307/54328836/attachment.sig>

Paul Gortmaker

author

2012-03-08 21:34 UTC #6 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

In reply to Paul Parsons

commit 737f360d5bef5e01c6cfa755dca0b449a154c1e0 (linux-next)

"regulator: Remove support for supplies specified by struct device"

deletes the field from the struct, but this platform was still trying to set those fields. Convert them to use the REGULATOR_SUPPLY macro.

Signed-off-by Paul Gortmaker <[email protected]>

---

[v3: fix arg2 in REGULATOR_SUPPLY to have correct names ]

arch/arm/mach-pxa/hx4700.c +2 -8

index f2c23ea..b6a338b 100644
@@ -679,14 +679,8 @@ static struct platform_device power_supply = {
  */
 
 static struct regulator_consumer_supply bq24022_consumers[] = {
-	{
-		.dev = &gpio_vbus.dev,
-		.supply = "vbus_draw",
-	},
-	{
-		.dev = &power_supply.dev,
-		.supply = "ac_draw",
-	},
+	REGULATOR_SUPPLY("vbus_draw", "gpio-vbus"),
+	REGULATOR_SUPPLY("ac_draw", "pda-power"),
 };
 
 static struct regulator_init_data bq24022_init_data = {

arch/arm/mach-pxa/magician.c +2 -8

index 3d6baf9..1a18e63 100644
@@ -577,14 +577,8 @@ static struct platform_device power_supply = {
  */
 
 static struct regulator_consumer_supply bq24022_consumers[] = {
-	{
-		.dev = &gpio_vbus.dev,
-		.supply = "vbus_draw",
-	},
-	{
-		.dev = &power_supply.dev,
-		.supply = "ac_draw",
-	},
+	REGULATOR_SUPPLY("vbus_draw", "gpio-vbus"),
+	REGULATOR_SUPPLY("ac_draw", "pda-power"),
 };
 
 static struct regulator_init_data bq24022_init_data = {

Signature · 1 line

1.7.9.1

Paul Parsons

2012-03-09 02:47 UTC #7 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

Thanks Paul.

The hx4700 patch works for me on linux-3.3.0-rc6.
I'm not able to test the magician patch.

Tested-by Paul Parsons <[email protected]>

Quoted 96 lines · Paul Gortmaker

commit 737f360d5bef5e01c6cfa755dca0b449a154c1e0 (linux-next)
? "regulator: Remove support for supplies specified by struct device"
deletes the field from the struct, but this platform was still trying
to set those fields.? Convert them to use the REGULATOR_SUPPLY macro.
arch/arm/mach-pxa/magician.c:581: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:581: warning: initialization from incompatible pointer type arch/arm/mach-pxa/magician.c:585: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:585: warning: initialization from incompatible pointer type
arch/arm/mach-pxa/hx4700.c:683: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:683: warning: initialization from incompatible pointer type arch/arm/mach-pxa/hx4700.c:687: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:687: warning: initialization from incompatible pointer type
Signed-off-by Paul Gortmaker <[email protected]>
---
[v3: fix arg2 in REGULATOR_SUPPLY to have correct names ]
arch/arm/mach-pxa/hx4700.c +0 -0
index f2c23ea..b6a338b 100644
@@ -679,14 +679,8 @@ static struct platform_device
 power_supply = {
? */
static struct regulator_consumer_supply bq24022_consumers[]
= {
-??? {
-??? ??? .dev =
&gpio_vbus.dev,
-??? ??? .supply =
"vbus_draw",
-??? },
-??? {
-??? ??? .dev =
&power_supply.dev,
-??? ??? .supply = "ac_draw",
-??? },
+??? REGULATOR_SUPPLY("vbus_draw",
"gpio-vbus"),
+??? REGULATOR_SUPPLY("ac_draw",
"pda-power"),
};
static struct regulator_init_data bq24022_init_data = {
arch/arm/mach-pxa/magician.c +0 -0
index 3d6baf9..1a18e63 100644
@@ -577,14 +577,8 @@ static struct platform_device
 power_supply = {
? */
static struct regulator_consumer_supply bq24022_consumers[]
= {
-??? {
-??? ??? .dev =
&gpio_vbus.dev,
-??? ??? .supply =
"vbus_draw",
-??? },
-??? {
-??? ??? .dev =
&power_supply.dev,
-??? ??? .supply = "ac_draw",
-??? },
+??? REGULATOR_SUPPLY("vbus_draw",
"gpio-vbus"),
+??? REGULATOR_SUPPLY("ac_draw",
"pda-power"),
};
static struct regulator_init_data bq24022_init_data = {
Signature · 1 line
1.7.9.1

Haojian Zhuang

maintainer

2012-03-14 08:38 UTC #8 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

In reply to Paul Gortmaker

Quoted 29 lines · Paul Gortmaker

commit 737f360d5bef5e01c6cfa755dca0b449a154c1e0 (linux-next)
?"regulator: Remove support for supplies specified by struct device"
deletes the field from the struct, but this platform was still trying to set those fields. ?Convert them to use the REGULATOR_SUPPLY macro.
arch/arm/mach-pxa/magician.c:581: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:581: warning: initialization from incompatible pointer type arch/arm/mach-pxa/magician.c:585: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:585: warning: initialization from incompatible pointer type
arch/arm/mach-pxa/hx4700.c:683: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:683: warning: initialization from incompatible pointer type arch/arm/mach-pxa/hx4700.c:687: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:687: warning: initialization from incompatible pointer type
Signed-off-by Paul Gortmaker <[email protected]>
---
[v3: fix arg2 in REGULATOR_SUPPLY to have correct names ]
arch/arm/mach-pxa/hx4700.c +0 -0
index f2c23ea..b6a338b 100644
@@ -679,14 +679,8 @@ static struct platform_device power_supply = {
?*/

Quoted 11 lines

?static struct regulator_consumer_supply bq24022_consumers[] = {
- ? ? ? {
- ? ? ? ? ? ? ? .dev = &gpio_vbus.dev,
- ? ? ? ? ? ? ? .supply = "vbus_draw",
- ? ? ? },
- ? ? ? {
- ? ? ? ? ? ? ? .dev = &power_supply.dev,
- ? ? ? ? ? ? ? .supply = "ac_draw",
- ? ? ? },
+ ? ? ? REGULATOR_SUPPLY("vbus_draw", "gpio-vbus"),
+ ? ? ? REGULATOR_SUPPLY("ac_draw", "pda-power"),

How do you think about Mark's comments? Use dev_name() at here.

Quoted 27 lines

?};
?static struct regulator_init_data bq24022_init_data = {
arch/arm/mach-pxa/magician.c +0 -0
index 3d6baf9..1a18e63 100644
@@ -577,14 +577,8 @@ static struct platform_device power_supply = {
?*/
?static struct regulator_consumer_supply bq24022_consumers[] = {
- ? ? ? {
- ? ? ? ? ? ? ? .dev = &gpio_vbus.dev,
- ? ? ? ? ? ? ? .supply = "vbus_draw",
- ? ? ? },
- ? ? ? {
- ? ? ? ? ? ? ? .dev = &power_supply.dev,
- ? ? ? ? ? ? ? .supply = "ac_draw",
- ? ? ? },
+ ? ? ? REGULATOR_SUPPLY("vbus_draw", "gpio-vbus"),
+ ? ? ? REGULATOR_SUPPLY("ac_draw", "pda-power"),
?};
?static struct regulator_init_data bq24022_init_data = { --
1.7.9.1

Paul Gortmaker

author

2012-03-14 12:30 UTC #9 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

Quoted 31 lines · Haojian Zhuang

Quoted 29 lines · Paul Gortmaker
commit 737f360d5bef5e01c6cfa755dca0b449a154c1e0 (linux-next)
"regulator: Remove support for supplies specified by struct device"
deletes the field from the struct, but this platform was still trying to set those fields. Convert them to use the REGULATOR_SUPPLY macro.
arch/arm/mach-pxa/magician.c:581: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:581: warning: initialization from incompatible pointer type arch/arm/mach-pxa/magician.c:585: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/magician.c:585: warning: initialization from incompatible pointer type
arch/arm/mach-pxa/hx4700.c:683: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:683: warning: initialization from incompatible pointer type arch/arm/mach-pxa/hx4700.c:687: error: unknown field 'dev' specified in initializer arch/arm/mach-pxa/hx4700.c:687: warning: initialization from incompatible pointer type
Signed-off-by Paul Gortmaker <[email protected]>
---
[v3: fix arg2 in REGULATOR_SUPPLY to have correct names ]
arch/arm/mach-pxa/hx4700.c +0 -0
index f2c23ea..b6a338b 100644
@@ -679,14 +679,8 @@ static struct platform_device power_supply = {
 */
 

Quoted 13 lines

arch/arm/mach-pxa/hx4700.c static struct platform_device power_supply = { · line 683
Quoted 11 lines
arch/arm/mach-pxa/hx4700.c +2 -8
 static struct regulator_consumer_supply bq24022_consumers[] = {
-       {
-               .dev = &gpio_vbus.dev,
-               .supply = "vbus_draw",
-       },
-       {
-               .dev = &power_supply.dev,
-               .supply = "ac_draw",
-       },
+       REGULATOR_SUPPLY("vbus_draw", "gpio-vbus"),
+       REGULATOR_SUPPLY("ac_draw", "pda-power"),
How do you think about Mark's comments? Use dev_name() at here.

Hi Haojian,

I'm not sure I quite understand you. I'd looked up the structs and got their respective value that would be returned by dev_name(), which is what I thought was requested of me.

Thanks,
Paul.

Quoted 27 lines

arch/arm/mach-pxa/hx4700.c +0 -0

 };
 
 static struct regulator_init_data bq24022_init_data = {

arch/arm/mach-pxa/magician.c +2 -9

index 3d6baf9..1a18e63 100644
@@ -577,14 +577,8 @@ static struct platform_device power_supply = {
 */
 
 static struct regulator_consumer_supply bq24022_consumers[] = {
-       {
-               .dev = &gpio_vbus.dev,
-               .supply = "vbus_draw",
-       },
-       {
-               .dev = &power_supply.dev,
-               .supply = "ac_draw",
-       },
+       REGULATOR_SUPPLY("vbus_draw", "gpio-vbus"),
+       REGULATOR_SUPPLY("ac_draw", "pda-power"),
 };
 
 static struct regulator_init_data bq24022_init_data = {
--

1.7.9.1

Mark Brown

2012-03-14 12:32 UTC #10 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

Quoted 8 lines · Paul Gortmaker

Quoted 1 line · Haojian Zhuang
How do you think about Mark's comments? Use dev_name() at here.
I'm not sure I quite understand you. I'd looked up the structs and got their respective value that would be returned by dev_name(), which is what I thought was requested of me.

Yes, that's what you should do. Calling dev_name() is only possible if
you have the struct device and removing the need for the struct device
is exactly what we're trying to avoid here.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20120314/557ec99d/attachment.sig>

Russell King - ARM Linux

2012-03-28 09:43 UTC #11 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

Quoted 5 lines · Mark Brown

Quoted 3 lines · Paul Gortmaker
Quoted 1 line · Haojian Zhuang
How do you think about Mark's comments? Use dev_name() at here.

Quoted 8 lines

Quoted 4 lines
I'm not sure I quite understand you. I'd looked up the structs and got their respective value that would be returned by dev_name(), which is what I thought was requested of me.
Yes, that's what you should do. Calling dev_name() is only possible if you have the struct device and removing the need for the struct device is exactly what we're trying to avoid here.

So what the hell is happening here to fix the breakage which the regulator tree caused in mainline? From this thread, it looks like it's been known about since 8th March which is now twenty days ago, and it's still broken.

Mark Brown

2012-03-28 10:18 UTC #12 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

Quoted 10 lines · Russell King - ARM Linux

Quoted 3 lines · Mark Brown
Yes, that's what you should do. Calling dev_name() is only possible if you have the struct device and removing the need for the struct device is exactly what we're trying to avoid here.
So what the hell is happening here to fix the breakage which the regulator tree caused in mainline? From this thread, it looks like it's been known about since 8th March which is now twenty days ago, and it's still broken.

Yes, please at least apply Paul's patches - honestly, given how old
these boards are and the general lack of activity I'd rather expect that
the only thing anyone's doing with them these days is build tests.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20120328/e6126460/attachment.sig>

Haojian Zhuang

maintainer Applied

2012-03-28 13:36 UTC #13 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

Quoted 7 lines · Mark Brown

Quoted 5 lines · Russell King - ARM Linux
Quoted 3 lines · Mark Brown
Yes, that's what you should do. ?Calling dev_name() is only possible if you have the struct device and removing the need for the struct device is exactly what we're trying to avoid here.

Quoted 8 lines

Quoted 4 lines
So what the hell is happening here to fix the breakage which the regulator tree caused in mainline? ?From this thread, it looks like it's been known about since 8th March which is now twenty days ago, and it's still broken.
Yes, please at least apply Paul's patches - honestly, given how old these boards are and the general lack of activity I'd rather expect that the only thing anyone's doing with them these days is build tests.

Applied. I'll submit this fix to Arnd and Olof.

Best Regards
Haojian

Russell King - ARM Linux

2012-03-28 17:56 UTC #14 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

In reply to Mark Brown

Quoted 7 lines · Mark Brown

Quoted 5 lines · Russell King - ARM Linux
Quoted 3 lines · Mark Brown
Yes, that's what you should do. Calling dev_name() is only possible if you have the struct device and removing the need for the struct device is exactly what we're trying to avoid here.

Quoted 8 lines

Quoted 4 lines
So what the hell is happening here to fix the breakage which the regulator tree caused in mainline? From this thread, it looks like it's been known about since 8th March which is now twenty days ago, and it's still broken.
Yes, please at least apply Paul's patches - honestly, given how old these boards are and the general lack of activity I'd rather expect that the only thing anyone's doing with them these days is build tests.

Is that a reason to push breakage to them upstream? No. Please take Paul's patches to fix the breakage which was introduced through your tree yourself. I've got enough problems this merge window to deal with without having to add further patches to my tree.

Thanks.

Mark Brown

2012-03-28 18:10 UTC #15 lore raw

PATCH 1/1 ARM: pxa: fix regulator related build fail in magician_defconfig

Quoted 10 lines · Russell King - ARM Linux

Quoted 3 lines · Mark Brown
Yes, please at least apply Paul's patches - honestly, given how old these boards are and the general lack of activity I'd rather expect that the only thing anyone's doing with them these days is build tests.
Is that a reason to push breakage to them upstream? No. Please take Paul's patches to fix the breakage which was introduced through your tree yourself. I've got enough problems this merge window to deal with without having to add further patches to my tree.

Well, I'd expect this stuff to be going via the pxa tree to arm-soc rather than via your tree anyway (which seems to be what's happening at this point) - half the point with arm-soc was to ensure you didn't have to deal with things like this.

Quite frankly I thought the fixes had already been applied, they were
posted and reviewed in good time for the merge window and the issues
weren't being reported in -next any more; had someone asked I'd have
applied them as I did with the equivalent stuff for u300. If they were
being totally ignored I'd probably have applied them too, though perhaps
not.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20120328/6f4abb48/attachment.sig>