[GIT PULL] Third Round of Renesas ARM Based SoC Fixes for v3.19

[GIT PULL] Third Round of Renesas ARM Based SoC Fixes for v3.19

[Simon Horman]

Hi Olof, Hi Kevin, Hi Arnd,

Please consider these third round of Renesas ARM based SoC fixes for v3.19.

This pull request is based on the previous round of
such requests, tagged as renesas-soc-fixes2-for-v3.19,
which I have already sent a pull-request for.


This fixes a regression in the r8a7790 and r8a73a4 SoCs due to
9a1091ef0017c40a ("irqchip: gic: Support hierarchy irq domain.")
which was included in v3.19-rc1.

This pull-request is targeted at v3.19. r8a7790 patch in this pull-request
conflicts with the patch "ARM: shmobile: lager: Remove legacy board
support" which was included in the pull request "Renesas ARM Based SoC
Lager Board Removal for v3.20" which you have pulled for v3.20. The
conflict occurs because board-lager.c is removed in that patch whereas
this series updates that file.

In other words, this series fixes board code for the lager/r8a7790 for v3.19
which is scheduled for removal in v3.20.


The following changes since commit f469cde20a7ee342070b4a459b6fce469a3186db:

  ARM: shmobile: r8a7779: Instantiate GIC from C board code in legacy builds (2015-01-17 09:28:41 +0900)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git tags/renesas-soc-fixes3-for-v3.19

for you to fetch changes up to 77cf5166f21c3b73b04571311dc89a327424f594:

  ARM: shmobile: r8a7790: Instantiate GIC from C board code in legacy builds (2015-01-29 17:52:38 +0900)

----------------------------------------------------------------
Third Round of Renesas ARM Based SoC Fixes for v3.19

* Instantiate GIC from C board code in legacy builds on r8a7790 and r8a73a4

----------------------------------------------------------------
Magnus Damm (2):
      ARM: shmobile: r8a73a4: Instantiate GIC from C board code in legacy builds
      ARM: shmobile: r8a7790: Instantiate GIC from C board code in legacy builds

 arch/arm/mach-shmobile/board-ape6evm.c   | 20 ++++++++++++++++++++
 arch/arm/mach-shmobile/board-lager.c     | 13 +++++++++++++
 arch/arm/mach-shmobile/setup-rcar-gen2.c |  2 ++
 arch/arm/mach-shmobile/timer.c           | 12 ++++++++++++
 4 files changed, 47 insertions(+)

[PATCH 1/2] ARM: shmobile: r8a73a4: Instantiate GIC from C board code in legacy builds

[Simon Horman]

From: Magnus Damm <damm+renesas@opensource.se>

As of commit 9a1091ef0017c40a ("irqchip: gic: Support hierarchy irq
domain."), the APE6EVM legacy board support is known to be broken.

The IRQ numbers of the GIC are now virtual, and no longer match the
hardcoded hardware IRQ numbers in the legacy platform board code.

To fix this issue specific to non-muliplatform r8a73a4 and APE6EVM:
 1) Instantiate the GIC from platform board code and also
 2) Skip over the DT arch timer as well as
 3) Force delay setup based on DT CPU frequency

With these 3 fixes in place interrupts on APE6EVM are now unbroken.

Partially based on legacy GIC fix by Geert Uytterhoeven, thanks to
him for the initial work.

Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
---
 arch/arm/mach-shmobile/board-ape6evm.c | 20 ++++++++++++++++++++
 arch/arm/mach-shmobile/timer.c         |  7 +++++++
 2 files changed, 27 insertions(+)

diff --git a/arch/arm/mach-shmobile/board-ape6evm.c b/arch/arm/mach-shmobile/board-ape6evm.c
index b222f68..2cca5ef 100644
--- a/arch/arm/mach-shmobile/board-ape6evm.c
+++ b/arch/arm/mach-shmobile/board-ape6evm.c
@@ -22,6 +22,8 @@
 #include <linux/gpio_keys.h>
 #include <linux/input.h>
 #include <linux/interrupt.h>
+#include <linux/irqchip.h>
+#include <linux/irqchip/arm-gic.h>
 #include <linux/kernel.h>
 #include <linux/mfd/tmio.h>
 #include <linux/mmc/host.h>
@@ -277,6 +279,22 @@ static void __init ape6evm_add_standard_devices(void)
 				      sizeof(ape6evm_leds_pdata));
 }
 
+static void __init ape6evm_legacy_init_time(void)
+{
+	/* Do not invoke DT-based timers via clocksource_of_init() */
+}
+
+static void __init ape6evm_legacy_init_irq(void)
+{
+	void __iomem *gic_dist_base = ioremap_nocache(0xf1001000, 0x1000);
+	void __iomem *gic_cpu_base = ioremap_nocache(0xf1002000, 0x1000);
+
+	gic_init(0, 29, gic_dist_base, gic_cpu_base);
+
+	/* Do not invoke DT-based interrupt code via irqchip_init() */
+}
+
+
 static const char *ape6evm_boards_compat_dt[] __initdata = {
 	"renesas,ape6evm",
 	NULL,
@@ -284,7 +302,9 @@ static const char *ape6evm_boards_compat_dt[] __initdata = {
 
 DT_MACHINE_START(APE6EVM_DT, "ape6evm")
 	.init_early	= shmobile_init_delay,
+	.init_irq       = ape6evm_legacy_init_irq,
 	.init_machine	= ape6evm_add_standard_devices,
 	.init_late	= shmobile_init_late,
 	.dt_compat	= ape6evm_boards_compat_dt,
+	.init_time	= ape6evm_legacy_init_time,
 MACHINE_END
diff --git a/arch/arm/mach-shmobile/timer.c b/arch/arm/mach-shmobile/timer.c
index 32ee335..44d32a8 100644
--- a/arch/arm/mach-shmobile/timer.c
+++ b/arch/arm/mach-shmobile/timer.c
@@ -75,6 +75,13 @@ void __init shmobile_init_delay(void)
 	if (!max_freq)
 		return;
 
+#ifdef CONFIG_ARCH_SHMOBILE_LEGACY
+	/* Non-multiplatform r8a73a4 SoC cannot use arch timer due
+	 * to GIC being initialized from C and arch timer via DT */
+	if (of_machine_is_compatible("renesas,r8a73a4"))
+		has_arch_timer = false;
+#endif
+
 	if (!has_arch_timer || !IS_ENABLED(CONFIG_ARM_ARCH_TIMER)) {
 		if (is_a7_a8_a9)
 			shmobile_setup_delay_hz(max_freq, 1, 3);
-- 
2.1.4

[PATCH 2/2] ARM: shmobile: r8a7790: Instantiate GIC from C board code in legacy builds

[Simon Horman]

From: Magnus Damm <damm+renesas@opensource.se>

As of commit 9a1091ef0017c40a ("irqchip: gic: Support hierarchy irq
domain."), the Lager legacy board support is known to be broken.

The IRQ numbers of the GIC are now virtual, and no longer match the
hardcoded hardware IRQ numbers in the legacy platform board code.

To fix this issue specific to non-multiplatform r8a7790 and Lager:
 1) Instantiate the GIC from platform board code and also
 2) Skip over the DT arch timer as well as
 3) Force delay setup based on DT CPU frequency

With these 3 fixes in place interrupts on Lager are now unbroken.

Partially based on legacy GIC fix by Geert Uytterhoeven, thanks to
him for the initial work.

Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
---
 arch/arm/mach-shmobile/board-lager.c     | 13 +++++++++++++
 arch/arm/mach-shmobile/setup-rcar-gen2.c |  2 ++
 arch/arm/mach-shmobile/timer.c           |  5 +++++
 3 files changed, 20 insertions(+)

diff --git a/arch/arm/mach-shmobile/board-lager.c b/arch/arm/mach-shmobile/board-lager.c
index 571327b..3f51327 100644
--- a/arch/arm/mach-shmobile/board-lager.c
+++ b/arch/arm/mach-shmobile/board-lager.c
@@ -25,6 +25,8 @@
 #include <linux/input.h>
 #include <linux/interrupt.h>
 #include <linux/irq.h>
+#include <linux/irqchip.h>
+#include <linux/irqchip/arm-gic.h>
 #include <linux/kernel.h>
 #include <linux/leds.h>
 #include <linux/mfd/tmio.h>
@@ -873,6 +875,16 @@ static void __init lager_init(void)
 					  lager_ksz8041_fixup);
 }
 
+static void __init lager_legacy_init_irq(void)
+{
+	void __iomem *gic_dist_base = ioremap_nocache(0xf1001000, 0x1000);
+	void __iomem *gic_cpu_base = ioremap_nocache(0xf1002000, 0x1000);
+
+	gic_init(0, 29, gic_dist_base, gic_cpu_base);
+
+	/* Do not invoke DT-based interrupt code via irqchip_init() */
+}
+
 static const char * const lager_boards_compat_dt[] __initconst = {
 	"renesas,lager",
 	NULL,
@@ -881,6 +893,7 @@ static const char * const lager_boards_compat_dt[] __initconst = {
 DT_MACHINE_START(LAGER_DT, "lager")
 	.smp		= smp_ops(r8a7790_smp_ops),
 	.init_early	= shmobile_init_delay,
+	.init_irq	= lager_legacy_init_irq,
 	.init_time	= rcar_gen2_timer_init,
 	.init_machine	= lager_init,
 	.init_late	= shmobile_init_late,
diff --git a/arch/arm/mach-shmobile/setup-rcar-gen2.c b/arch/arm/mach-shmobile/setup-rcar-gen2.c
index 7ed9279..101b3e4 100644
--- a/arch/arm/mach-shmobile/setup-rcar-gen2.c
+++ b/arch/arm/mach-shmobile/setup-rcar-gen2.c
@@ -137,7 +137,9 @@ void __init rcar_gen2_timer_init(void)
 #ifdef CONFIG_COMMON_CLK
 	rcar_gen2_clocks_init(mode);
 #endif
+#ifdef CONFIG_ARCH_SHMOBILE_MULTI
 	clocksource_of_init();
+#endif
 }
 
 struct memory_reserve_config {
diff --git a/arch/arm/mach-shmobile/timer.c b/arch/arm/mach-shmobile/timer.c
index 44d32a8..88f067b 100644
--- a/arch/arm/mach-shmobile/timer.c
+++ b/arch/arm/mach-shmobile/timer.c
@@ -80,6 +80,11 @@ void __init shmobile_init_delay(void)
 	 * to GIC being initialized from C and arch timer via DT */
 	if (of_machine_is_compatible("renesas,r8a73a4"))
 		has_arch_timer = false;
+
+	/* Non-multiplatform r8a7790 SoC cannot use arch timer due
+	 * to GIC being initialized from C and arch timer via DT */
+	if (of_machine_is_compatible("renesas,r8a7790"))
+		has_arch_timer = false;
 #endif
 
 	if (!has_arch_timer || !IS_ENABLED(CONFIG_ARM_ARCH_TIMER)) {
-- 
2.1.4

[GIT PULL] Third Round of Renesas ARM Based SoC Fixes for v3.19

[Olof Johansson]

Simon,


On Thu, Jan 29, 2015 at 4:45 PM, Simon Horman
<horms+renesas@verge.net.au> wrote:
> Hi Olof, Hi Kevin, Hi Arnd,
>
> Please consider these third round of Renesas ARM based SoC fixes for v3.19.
>
> This pull request is based on the previous round of
> such requests, tagged as renesas-soc-fixes2-for-v3.19,
> which I have already sent a pull-request for.
>
>
> This fixes a regression in the r8a7790 and r8a73a4 SoCs due to
> 9a1091ef0017c40a ("irqchip: gic: Support hierarchy irq domain.")
> which was included in v3.19-rc1.
>
> This pull-request is targeted at v3.19. r8a7790 patch in this pull-request
> conflicts with the patch "ARM: shmobile: lager: Remove legacy board
> support" which was included in the pull request "Renesas ARM Based SoC
> Lager Board Removal for v3.20" which you have pulled for v3.20. The
> conflict occurs because board-lager.c is removed in that patch whereas
> this series updates that file.
>
> In other words, this series fixes board code for the lager/r8a7790 for v3.19
> which is scheduled for removal in v3.20.
>
>
> The following changes since commit f469cde20a7ee342070b4a459b6fce469a3186db:
>
>   ARM: shmobile: r8a7779: Instantiate GIC from C board code in legacy builds (2015-01-17 09:28:41 +0900)
>
> are available in the git repository at:
>
>   git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git tags/renesas-soc-fixes3-for-v3.19
>
> for you to fetch changes up to 77cf5166f21c3b73b04571311dc89a327424f594:
>
>   ARM: shmobile: r8a7790: Instantiate GIC from C board code in legacy builds (2015-01-29 17:52:38 +0900)
>
> ----------------------------------------------------------------
> Third Round of Renesas ARM Based SoC Fixes for v3.19
>
> * Instantiate GIC from C board code in legacy builds on r8a7790 and r8a73a4
>
> ----------------------------------------------------------------
> Magnus Damm (2):
>       ARM: shmobile: r8a73a4: Instantiate GIC from C board code in legacy builds
>       ARM: shmobile: r8a7790: Instantiate GIC from C board code in legacy builds

This is the third branch you've sent us with the same type of fixes.
Was there a reason for why you couldn't have prepared all of them at
the same time once it was known to be a systematic problem?


-Olof

[GIT PULL] Third Round of Renesas ARM Based SoC Fixes for v3.19

[Simon Horman]

On Sat, Jan 31, 2015 at 05:30:32PM -0800, Olof Johansson wrote:
> Simon,
> 
> 
> On Thu, Jan 29, 2015 at 4:45 PM, Simon Horman
> <horms+renesas@verge.net.au> wrote:
> > Hi Olof, Hi Kevin, Hi Arnd,
> >
> > Please consider these third round of Renesas ARM based SoC fixes for v3.19.
> >
> > This pull request is based on the previous round of
> > such requests, tagged as renesas-soc-fixes2-for-v3.19,
> > which I have already sent a pull-request for.
> >
> >
> > This fixes a regression in the r8a7790 and r8a73a4 SoCs due to
> > 9a1091ef0017c40a ("irqchip: gic: Support hierarchy irq domain.")
> > which was included in v3.19-rc1.
> >
> > This pull-request is targeted at v3.19. r8a7790 patch in this pull-request
> > conflicts with the patch "ARM: shmobile: lager: Remove legacy board
> > support" which was included in the pull request "Renesas ARM Based SoC
> > Lager Board Removal for v3.20" which you have pulled for v3.20. The
> > conflict occurs because board-lager.c is removed in that patch whereas
> > this series updates that file.
> >
> > In other words, this series fixes board code for the lager/r8a7790 for v3.19
> > which is scheduled for removal in v3.20.
> >
> >
> > The following changes since commit f469cde20a7ee342070b4a459b6fce469a3186db:
> >
> >   ARM: shmobile: r8a7779: Instantiate GIC from C board code in legacy builds (2015-01-17 09:28:41 +0900)
> >
> > are available in the git repository at:
> >
> >   git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git tags/renesas-soc-fixes3-for-v3.19
> >
> > for you to fetch changes up to 77cf5166f21c3b73b04571311dc89a327424f594:
> >
> >   ARM: shmobile: r8a7790: Instantiate GIC from C board code in legacy builds (2015-01-29 17:52:38 +0900)
> >
> > ----------------------------------------------------------------
> > Third Round of Renesas ARM Based SoC Fixes for v3.19
> >
> > * Instantiate GIC from C board code in legacy builds on r8a7790 and r8a73a4
> >
> > ----------------------------------------------------------------
> > Magnus Damm (2):
> >       ARM: shmobile: r8a73a4: Instantiate GIC from C board code in legacy builds
> >       ARM: shmobile: r8a7790: Instantiate GIC from C board code in legacy builds
> 
> This is the third branch you've sent us with the same type of fixes.
> Was there a reason for why you couldn't have prepared all of them at
> the same time once it was known to be a systematic problem?

Sorry about that, I agree this could have been batched up a bit better.

The main reason that things occurred in this manner was that it took
us a while to fix the problem on all the effected SoCs and as a result
some of the fixes were ready quite some time ago while others only
became available quite recently. It seemed reasonable to me to push out the
fixes as they became available. Rather than risk all of the SoCs being
broken in v3.19.

However, as this seems to be a bit of a pain point for you, I'll work
on batching things up more if we encounter such a wide-spread problem
again. We are also working on being able to catch regressions earlier
to try to avoid having to patching things up in this manner.

[GIT PULL] Third Round of Renesas ARM Based SoC Fixes for v3.19

[Olof Johansson]

On Sun, Feb 1, 2015 at 4:24 PM, Simon Horman <horms@verge.net.au> wrote:
> On Sat, Jan 31, 2015 at 05:30:32PM -0800, Olof Johansson wrote:
>> Simon,
>>
>>
>> On Thu, Jan 29, 2015 at 4:45 PM, Simon Horman
>> <horms+renesas@verge.net.au> wrote:
>> > Hi Olof, Hi Kevin, Hi Arnd,
>> >
>> > Please consider these third round of Renesas ARM based SoC fixes for v3.19.
>> >
>> > This pull request is based on the previous round of
>> > such requests, tagged as renesas-soc-fixes2-for-v3.19,
>> > which I have already sent a pull-request for.
>> >
>> >
>> > This fixes a regression in the r8a7790 and r8a73a4 SoCs due to
>> > 9a1091ef0017c40a ("irqchip: gic: Support hierarchy irq domain.")
>> > which was included in v3.19-rc1.
>> >
>> > This pull-request is targeted at v3.19. r8a7790 patch in this pull-request
>> > conflicts with the patch "ARM: shmobile: lager: Remove legacy board
>> > support" which was included in the pull request "Renesas ARM Based SoC
>> > Lager Board Removal for v3.20" which you have pulled for v3.20. The
>> > conflict occurs because board-lager.c is removed in that patch whereas
>> > this series updates that file.
>> >
>> > In other words, this series fixes board code for the lager/r8a7790 for v3.19
>> > which is scheduled for removal in v3.20.
>> >
>> >
>> > The following changes since commit f469cde20a7ee342070b4a459b6fce469a3186db:
>> >
>> >   ARM: shmobile: r8a7779: Instantiate GIC from C board code in legacy builds (2015-01-17 09:28:41 +0900)
>> >
>> > are available in the git repository at:
>> >
>> >   git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git tags/renesas-soc-fixes3-for-v3.19
>> >
>> > for you to fetch changes up to 77cf5166f21c3b73b04571311dc89a327424f594:
>> >
>> >   ARM: shmobile: r8a7790: Instantiate GIC from C board code in legacy builds (2015-01-29 17:52:38 +0900)
>> >
>> > ----------------------------------------------------------------
>> > Third Round of Renesas ARM Based SoC Fixes for v3.19
>> >
>> > * Instantiate GIC from C board code in legacy builds on r8a7790 and r8a73a4
>> >
>> > ----------------------------------------------------------------
>> > Magnus Damm (2):
>> >       ARM: shmobile: r8a73a4: Instantiate GIC from C board code in legacy builds
>> >       ARM: shmobile: r8a7790: Instantiate GIC from C board code in legacy builds
>>
>> This is the third branch you've sent us with the same type of fixes.
>> Was there a reason for why you couldn't have prepared all of them at
>> the same time once it was known to be a systematic problem?
>
> Sorry about that, I agree this could have been batched up a bit better.
>
> The main reason that things occurred in this manner was that it took
> us a while to fix the problem on all the effected SoCs and as a result
> some of the fixes were ready quite some time ago while others only
> became available quite recently. It seemed reasonable to me to push out the
> fixes as they became available. Rather than risk all of the SoCs being
> broken in v3.19.
>
> However, as this seems to be a bit of a pain point for you, I'll work
> on batching things up more if we encounter such a wide-spread problem
> again. We are also working on being able to catch regressions earlier
> to try to avoid having to patching things up in this manner.

It's not a pain point, and you can send them when ready, it just
seemed odd that they kept trickling in over time. Which is why I
brought it up.


-Olof

[GIT PULL] Third Round of Renesas ARM Based SoC Fixes for v3.19

[Simon Horman]

On Sun, Feb 01, 2015 at 04:27:50PM -0800, Olof Johansson wrote:
> On Sun, Feb 1, 2015 at 4:24 PM, Simon Horman <horms@verge.net.au> wrote:
> > On Sat, Jan 31, 2015 at 05:30:32PM -0800, Olof Johansson wrote:
> >> Simon,
> >>
> >>
> >> On Thu, Jan 29, 2015 at 4:45 PM, Simon Horman
> >> <horms+renesas@verge.net.au> wrote:
> >> > Hi Olof, Hi Kevin, Hi Arnd,
> >> >
> >> > Please consider these third round of Renesas ARM based SoC fixes for v3.19.
> >> >
> >> > This pull request is based on the previous round of
> >> > such requests, tagged as renesas-soc-fixes2-for-v3.19,
> >> > which I have already sent a pull-request for.
> >> >
> >> >
> >> > This fixes a regression in the r8a7790 and r8a73a4 SoCs due to
> >> > 9a1091ef0017c40a ("irqchip: gic: Support hierarchy irq domain.")
> >> > which was included in v3.19-rc1.
> >> >
> >> > This pull-request is targeted at v3.19. r8a7790 patch in this pull-request
> >> > conflicts with the patch "ARM: shmobile: lager: Remove legacy board
> >> > support" which was included in the pull request "Renesas ARM Based SoC
> >> > Lager Board Removal for v3.20" which you have pulled for v3.20. The
> >> > conflict occurs because board-lager.c is removed in that patch whereas
> >> > this series updates that file.
> >> >
> >> > In other words, this series fixes board code for the lager/r8a7790 for v3.19
> >> > which is scheduled for removal in v3.20.
> >> >
> >> >
> >> > The following changes since commit f469cde20a7ee342070b4a459b6fce469a3186db:
> >> >
> >> >   ARM: shmobile: r8a7779: Instantiate GIC from C board code in legacy builds (2015-01-17 09:28:41 +0900)
> >> >
> >> > are available in the git repository at:
> >> >
> >> >   git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git tags/renesas-soc-fixes3-for-v3.19
> >> >
> >> > for you to fetch changes up to 77cf5166f21c3b73b04571311dc89a327424f594:
> >> >
> >> >   ARM: shmobile: r8a7790: Instantiate GIC from C board code in legacy builds (2015-01-29 17:52:38 +0900)
> >> >
> >> > ----------------------------------------------------------------
> >> > Third Round of Renesas ARM Based SoC Fixes for v3.19
> >> >
> >> > * Instantiate GIC from C board code in legacy builds on r8a7790 and r8a73a4
> >> >
> >> > ----------------------------------------------------------------
> >> > Magnus Damm (2):
> >> >       ARM: shmobile: r8a73a4: Instantiate GIC from C board code in legacy builds
> >> >       ARM: shmobile: r8a7790: Instantiate GIC from C board code in legacy builds
> >>
> >> This is the third branch you've sent us with the same type of fixes.
> >> Was there a reason for why you couldn't have prepared all of them at
> >> the same time once it was known to be a systematic problem?
> >
> > Sorry about that, I agree this could have been batched up a bit better.
> >
> > The main reason that things occurred in this manner was that it took
> > us a while to fix the problem on all the effected SoCs and as a result
> > some of the fixes were ready quite some time ago while others only
> > became available quite recently. It seemed reasonable to me to push out the
> > fixes as they became available. Rather than risk all of the SoCs being
> > broken in v3.19.
> >
> > However, as this seems to be a bit of a pain point for you, I'll work
> > on batching things up more if we encounter such a wide-spread problem
> > again. We are also working on being able to catch regressions earlier
> > to try to avoid having to patching things up in this manner.
> 
> It's not a pain point, and you can send them when ready, it just
> seemed odd that they kept trickling in over time. Which is why I
> brought it up.

Thanks, understood.