ARM: at91: remove useless init_time for DT-only SoCs

PATCH Superseded Boris Brezillon 3 participants 7 messages Nov 10, 2014 – Nov 20, 2014

Subsystems (3)

Arm (32-bit) Arm sub-architectures Arm/Microchip (AT91) SoC support

lore

Share & tools

lore

Boris Brezillon

author

2014-11-10 07:31 UTC #0 lore raw

PATCH 1/1 ARM: at91: remove useless init_time for DT-only SoCs

Commits bcf8c7e7703bb2bbde66bec26a81fed4be9cd1d4 and 4bf7753b8ab7c704ca864b1653367ea9cfe68a25 introduced compilation errors ("error: 'NR_IRQS_LEGACY' undeclared (first use in this function)") because they remove the asm/irq.h inclusion while the init_time function needs it for the NR_IRQS_LEGACY definition.

In the other hand, the point of these commits is to remove board file support, and init_time is only needed when booting non-DT boards, we can thus safely remove init_time functions.

Signed-off-by Boris Brezillon <[email protected]>

---
Hi Olof,

You recently pulled some changes on Nicolas' at91-cleanup tag and two of those commits (see above for commit hashes) are actually introducing compilation errors.
This patch fixes those compilation errors.

Best Regards,

Boris

 arch/arm/mach-at91/at91sam9g45.c | 6 ------
 arch/arm/mach-at91/at91sam9rl.c  | 6 ------
 2 files changed, 12 deletions(-)

arch/arm/mach-at91/at91sam9g45.c +0 -6

index dd03f65..2ad0db4 100644
@@ -34,13 +34,7 @@ static void __init at91sam9g45_initialize(void)
 	at91_sysirq_mask_rtt(AT91SAM9G45_BASE_RTT);
 }
 
-static void __init at91sam9g45_init_time(void)
-{
-	at91sam926x_pit_init(NR_IRQS_LEGACY + AT91_ID_SYS);
-}
-
 AT91_SOC_START(at91sam9g45)
 	.map_io = at91sam9g45_map_io,
 	.init = at91sam9g45_initialize,
-	.init_time = at91sam9g45_init_time,
 AT91_SOC_END

arch/arm/mach-at91/at91sam9rl.c +0 -6

index 6ca7fc5..bd7e568 100644
@@ -46,13 +46,7 @@ static void __init at91sam9rl_initialize(void)
 	at91_sysirq_mask_rtt(AT91SAM9RL_BASE_RTT);
 }
 
-static void __init at91sam9rl_init_time(void)
-{
-	at91sam926x_pit_init(NR_IRQS_LEGACY + AT91_ID_SYS);
-}
-
 AT91_SOC_START(at91sam9rl)
 	.map_io = at91sam9rl_map_io,
 	.init = at91sam9rl_initialize,
-	.init_time = at91sam9rl_init_time,
 AT91_SOC_END

Signature · 1 line

1.9.1

Arnd Bergmann

2014-11-19 22:23 UTC #1 lore raw

PATCH 1/1 ARM: at91: remove useless init_time for DT-only SoCs

Quoted 7 lines · Boris Brezillon

Hi Olof,
You recently pulled some changes on Nicolas' at91-cleanup tag and two of those commits (see above for commit hashes) are actually introducing compilation errors.
This patch fixes those compilation errors.

Applied on next/cleanup, thanks!

Arnd

Boris Brezillon

author

2014-11-20 12:46 UTC #2 lore raw

PATCH 1/1 ARM: at91: remove useless init_time for DT-only SoCs

Hi Arnd,

Quoted 10 lines · On Wed, 19 Nov 2014 23:23:34 +0100 Arnd Bergmann <[email protected]> wrote:

Quoted 7 lines · Boris Brezillon
Hi Olof,
You recently pulled some changes on Nicolas' at91-cleanup tag and two of those commits (see above for commit hashes) are actually introducing compilation errors.
This patch fixes those compilation errors.
Applied on next/cleanup, thanks!

Actually, I've made a v2 of this patch (see [1]) after realizing Olof had already fixed the initial compilation error.
This is pretty much the same patch, except it removes asm/irq.h inclusion.

[1]http://thread.gmane.org/gmane.linux.ports.arm.kernel/370398

Best Regards,

Boris

Signature · 3 lines

Boris Brezillon, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com

Arnd Bergmann

2014-11-20 13:17 UTC #3 lore raw

PATCH 1/1 ARM: at91: remove useless init_time for DT-only SoCs

Quoted 7 lines · On Thursday 20 November 2014 13:46:15 Boris Brezillon wrote:

Actually, I've made a v2 of this patch (see [1]) after realizing Olof had already fixed the initial compilation error.
This is pretty much the same patch, except it removes asm/irq.h inclusion.
[1]http://thread.gmane.org/gmane.linux.ports.arm.kernel/370398

Sorry about this. Do you want to send an add-on patch to change that one line, or should we just keep the current version?

Arnd

Nicolas Ferre

2014-11-20 13:32 UTC #4 lore raw

PATCH 1/1 ARM: at91: remove useless init_time for DT-only SoCs

In reply to Boris Brezillon

On 20/11/2014 13:46, Boris Brezillon :

Quoted 14 lines

Hi Arnd,
Quoted 9 lines · On Wed, 19 Nov 2014 23:23:34 +0100 Arnd Bergmann <[email protected]> wrote:
Quoted 7 lines · Boris Brezillon
Hi Olof,
You recently pulled some changes on Nicolas' at91-cleanup tag and two of those commits (see above for commit hashes) are actually introducing compilation errors.
This patch fixes those compilation errors.

Quoted 8 lines

Quoted 1 line
Applied on next/cleanup, thanks!
Actually, I've made a v2 of this patch (see [1]) after realizing Olof had already fixed the initial compilation error.
This is pretty much the same patch, except it removes asm/irq.h inclusion.
[1]http://thread.gmane.org/gmane.linux.ports.arm.kernel/370398

And this v2 was included in the pull-request I made and that Arnd have just taken (at91-cleanup2 tag).

It seems that the files still have the "#include <asm/irq.h>" in the result branch.

So, Arnd can whether remove the following commit (in next/cleanup):
440ae45119e9 (ARM: at91: remove useless init_time for DT-only SoCs) or we can add another patch just to remove these additional lines.

Arnd, what do you prefer?

Bye,

Signature · 1 line

Nicolas Ferre

Arnd Bergmann

2014-11-20 15:26 UTC #5 lore raw

PATCH 1/1 ARM: at91: remove useless init_time for DT-only SoCs

Quoted 27 lines · On Thursday 20 November 2014 14:32:29 Nicolas Ferre wrote:

On 20/11/2014 13:46, Boris Brezillon :
Quoted 14 lines
Hi Arnd,
Quoted 9 lines · On Wed, 19 Nov 2014 23:23:34 +0100 Arnd Bergmann <[email protected]> wrote:
Quoted 7 lines · Boris Brezillon
Hi Olof,
You recently pulled some changes on Nicolas' at91-cleanup tag and two of those commits (see above for commit hashes) are actually introducing compilation errors.
This patch fixes those compilation errors.
Quoted 8 lines
Quoted 1 line
Applied on next/cleanup, thanks!
Actually, I've made a v2 of this patch (see [1]) after realizing Olof had already fixed the initial compilation error.
This is pretty much the same patch, except it removes asm/irq.h inclusion.
[1]http://thread.gmane.org/gmane.linux.ports.arm.kernel/370398
And this v2 was included in the pull-request I made and that Arnd have just taken (at91-cleanup2 tag).

Quoted 8 lines

It seems that the files still have the "#include <asm/irq.h>" in the result branch.
So, Arnd can whether remove the following commit (in next/cleanup):
440ae45119e9 (ARM: at91: remove useless init_time for DT-only SoCs) or we can add another patch just to remove these additional lines.
Arnd, what do you prefer?

I don't want to change the merge I already did, so just send a patch that I can apply on top.

Arnd

Nicolas Ferre

2014-11-20 16:20 UTC #6 lore raw

PATCH 1/1 ARM: at91: remove useless init_time for DT-only SoCs

On 20/11/2014 16:26, Arnd Bergmann :

Quoted 31 lines · On Thursday 20 November 2014 14:32:29 Nicolas Ferre wrote:

Quoted 30 lines
On 20/11/2014 13:46, Boris Brezillon :
Quoted 14 lines
Hi Arnd,
Quoted 9 lines · On Wed, 19 Nov 2014 23:23:34 +0100 Arnd Bergmann <[email protected]> wrote:
Quoted 7 lines · Boris Brezillon
Hi Olof,
You recently pulled some changes on Nicolas' at91-cleanup tag and two of those commits (see above for commit hashes) are actually introducing compilation errors.
This patch fixes those compilation errors.
Quoted 8 lines
Quoted 1 line
Applied on next/cleanup, thanks!
Actually, I've made a v2 of this patch (see [1]) after realizing Olof had already fixed the initial compilation error.
This is pretty much the same patch, except it removes asm/irq.h inclusion.
[1]http://thread.gmane.org/gmane.linux.ports.arm.kernel/370398
And this v2 was included in the pull-request I made and that Arnd have just taken (at91-cleanup2 tag).
It seems that the files still have the "#include <asm/irq.h>" in the result branch.

Quoted 8 lines

Quoted 5 lines
So, Arnd can whether remove the following commit (in next/cleanup):
440ae45119e9 (ARM: at91: remove useless init_time for DT-only SoCs) or we can add another patch just to remove these additional lines.
Arnd, what do you prefer?
I don't want to change the merge I already did, so just send a patch that I can apply on top.

Sure I will.

I will then add it to my next "cleanup" pull-request to you.

Thanks for your help, bye.

Signature · 1 line

Nicolas Ferre