diff mbox

[U-Boot,2/2] tegra: enable lp0 on paz00

Message ID 644afaa87f727ab1558c1f1fa9e69faab10c8a95.1347302983.git.marvin24@gmx.de
State Changes Requested
Delegated to: Tom Warren
Headers show

Commit Message

Marc Dietrich Sept. 10, 2012, 6:51 p.m. UTC
This enables LP0 to support suspend / resume on PAZ00.

Signed-off-by: Marc Dietrich <marvin24@gmx.de>
---
 board/compal/paz00/Makefile |    6 ++++--
 include/configs/paz00.h     |    8 ++++++++
 2 files changed, 12 insertions(+), 2 deletions(-)

Comments

Stephen Warren Sept. 10, 2012, 7:07 p.m. UTC | #1
On 09/10/2012 12:51 PM, Marc Dietrich wrote:
> This enables LP0 to support suspend / resume on PAZ00.

Ooh. Did you test this out with the AC100 kernel, and have it work?
That'd be pretty cool...

> diff --git a/board/compal/paz00/Makefile b/board/compal/paz00/Makefile

> -COBJS	:= $(BOARD).o
> -COBJS	+= ../../nvidia/common/board.o
> +COBJS-y	:= $(BOARD).o
> +COBJS-y	+= ../../nvidia/common/board.o
> +COBJS-$(CONFIG_TEGRA_CLOCK_SCALING) += ../../nvidia/common/emc.o

Hmmm. That's odd. I'd expect that to be part of the core Tegra code,
rather than something boards have to pull in manually.

> diff --git a/include/configs/paz00.h b/include/configs/paz00.h

> +/* LP0 suspend / resume */
> +#define CONFIG_TEGRA20_LP0

That's been renamed CONFIG_TEGRA_LP0 in u-boot-tegra/next.
Marc Dietrich Sept. 10, 2012, 7:53 p.m. UTC | #2
On Monday 10 September 2012 13:07:43 Stephen Warren wrote:
> On 09/10/2012 12:51 PM, Marc Dietrich wrote:
> > This enables LP0 to support suspend / resume on PAZ00.
> 
> Ooh. Did you test this out with the AC100 kernel, and have it work?
> That'd be pretty cool...

yeah, it would.

Unfortunately, I failed to boot a nvidia 3.1 kernel with uboot (hangs 
somewhere in usb code), so I couldn't test it so far. But I'll retry soon.
 
> > diff --git a/board/compal/paz00/Makefile b/board/compal/paz00/Makefile
> > 
> > -COBJS	:= $(BOARD).o
> > -COBJS	+= ../../nvidia/common/board.o
> > +COBJS-y	:= $(BOARD).o
> > +COBJS-y	+= ../../nvidia/common/board.o
> > +COBJS-$(CONFIG_TEGRA_CLOCK_SCALING) += ../../nvidia/common/emc.o
> 
> Hmmm. That's odd. I'd expect that to be part of the core Tegra code,
> rather than something boards have to pull in manually.

Correct, will kill it in the next try.

> > diff --git a/include/configs/paz00.h b/include/configs/paz00.h
> > 
> > +/* LP0 suspend / resume */
> > +#define CONFIG_TEGRA20_LP0
> 
> That's been renamed CONFIG_TEGRA_LP0 in u-boot-tegra/next.

Thanks for the hint. At least I have a better chance to test it now ;-)

Marc
Marc Dietrich Sept. 16, 2012, 4:16 p.m. UTC | #3
Tom,

On Monday 10 September 2012 12:32:00 Tom Warren wrote:
> > -----Original Message-----
> > From: Stephen Warren [mailto:swarren@wwwdotorg.org]
> > Sent: Monday, September 10, 2012 12:08 PM
> > To: Marc Dietrich
> > Cc: u-boot@lists.denx.de; Tom Warren
> > Subject: Re: [PATCH 2/2] tegra: enable lp0 on paz00
> > 
> > On 09/10/2012 12:51 PM, Marc Dietrich wrote:
> > > This enables LP0 to support suspend / resume on PAZ00.
> > 
> > Ooh. Did you test this out with the AC100 kernel, and have it work?
> > That'd be pretty cool...
> > 
> > > diff --git a/board/compal/paz00/Makefile b/board/compal/paz00/Makefile
> > > 
> > > -COBJS	:= $(BOARD).o
> > > -COBJS	+= ../../nvidia/common/board.o
> > > +COBJS-y	:= $(BOARD).o
> > > +COBJS-y	+= ../../nvidia/common/board.o
> > > +COBJS-$(CONFIG_TEGRA_CLOCK_SCALING) += ../../nvidia/common/emc.o
> > 
> > Hmmm. That's odd. I'd expect that to be part of the core Tegra code,
> > rather
> > than something boards have to pull in manually.

I checked this again. The Makefile in nvidia/common is never executed on non 
nvidia boards (it is included from the topdir Makefile 
($vendor/common/Makefile). 

Therefore the explicit "COBJS	+= ../../nvidia/common/board.o" in the paz00 
Makefile is needed. So either we have to add ../../nvidia/common/foo.o to all 
non nvidia boards or we "source" the whole Makefile somehow else.

Marc
 
> Stephen's right - this is already done in ../cpu/tegra20-common/Makefile
> when CONFIG_TEGRA_CLOCK_SCALING is defined. So no need to change the PAZ00
> Makefile.
> > > diff --git a/include/configs/paz00.h b/include/configs/paz00.h
> > > 
> > > +/* LP0 suspend / resume */
> > > +#define CONFIG_TEGRA20_LP0
> > 
> > That's been renamed CONFIG_TEGRA_LP0 in u-boot-tegra/next.
> 
> As part of the pre-work for Tegra30 changes, I've changed generic Tegra
> defines, labels, etc. to be more non-specific, unless it does really refer
> to a Tegra20 feature, file, etc. As Stephen says, see u-boot-tegra/next's
> top commit.
> 
> Tom
diff mbox

Patch

diff --git a/board/compal/paz00/Makefile b/board/compal/paz00/Makefile
index 7f7287e..fbe28b3 100644
--- a/board/compal/paz00/Makefile
+++ b/board/compal/paz00/Makefile
@@ -20,9 +20,11 @@  $(shell mkdir -p $(obj)../../nvidia/common)
 
 LIB	= $(obj)lib$(BOARD).o
 
-COBJS	:= $(BOARD).o
-COBJS	+= ../../nvidia/common/board.o
+COBJS-y	:= $(BOARD).o
+COBJS-y	+= ../../nvidia/common/board.o
+COBJS-$(CONFIG_TEGRA_CLOCK_SCALING) += ../../nvidia/common/emc.o
 
+COBJS	:= $(COBJS-y)
 SRCS	:= $(COBJS:.o=.c)
 OBJS	:= $(addprefix $(obj),$(COBJS))
 
diff --git a/include/configs/paz00.h b/include/configs/paz00.h
index 7e21e66..1c96f1c 100644
--- a/include/configs/paz00.h
+++ b/include/configs/paz00.h
@@ -18,6 +18,14 @@ 
 #define __CONFIG_H
 
 #include <asm/sizes.h>
+
+/* LP0 suspend / resume */
+#define CONFIG_TEGRA20_LP0
+#define CONFIG_AES
+#define CONFIG_TEGRA_PMU
+#define CONFIG_TPS6586X_POWER
+#define CONFIG_TEGRA_CLOCK_SCALING
+
 #include "tegra20-common.h"
 
 /* Enable fdt support for Paz00. Flash the image in u-boot-dtb.bin */