[PATCH 0/4] Tegra114 Add Minimal SDHCI support

[PATCH 0/4] Tegra114 Add Minimal SDHCI support

[Rhyland Klein]

This patchset adds the minimal support for Tegra114 SDHCI support
and enables it for tegra114-dalmore.

This patchset is dependent on:
 - Tegra114 Clock framework (v7) from Peter De Schrijver
   ( http://patchwork.ozlabs.org/patch/220727/ ...)
 - Tegra114 Pinctrl patches (v2) I sent 2/11
   ( http://patchwork.ozlabs.org/patch/219686/
     http://patchwork.ozlabs.org/patch/219687/ )
 - mmc: tegra: assume CONFIG_OF, remove platform data
   ( https://patchwork.kernel.org/patch/2150351/)

I tested merging this with and without the mmc patches from Joseph Lo
and it merged cleanly in either case, so there shouldn't be a nasty
dependency between them.
   ( http://patchwork.ozlabs.org/patch/221969/
     http://patchwork.ozlabs.org/patch/221970/ )

Pritesh Raithatha (1):
  ARM tegra: add SDHCI nodes with common properties

Rhyland Klein (3):
  ARM: tegra: add SDHCI support for Dalmore
  mmc: host: sdhci-tegra: cleanup ifdefs
  mmc: host: sdhci-tegra: add basic support for Tegra114

 arch/arm/boot/dts/tegra114-dalmore.dts |   11 +++++++++++
 arch/arm/boot/dts/tegra114.dtsi        |   32 ++++++++++++++++++++++++++++++++
 drivers/mmc/host/sdhci-tegra.c         |   22 ++++++++++++++--------
 3 files changed, 57 insertions(+), 8 deletions(-)

-- 
1.7.9.5

[PATCH 2/4] ARM: tegra: add SDHCI support for Dalmore

[Rhyland Klein]

Based on changes by: Pritesh Raithatha <praithatha@nvidia.com>
Signed-off-by: Rhyland Klein <rklein@nvidia.com>
---
 arch/arm/boot/dts/tegra114-dalmore.dts |   11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/arch/arm/boot/dts/tegra114-dalmore.dts b/arch/arm/boot/dts/tegra114-dalmore.dts
index a30aca6..7be8bd2 100644
--- a/arch/arm/boot/dts/tegra114-dalmore.dts
+++ b/arch/arm/boot/dts/tegra114-dalmore.dts
@@ -15,6 +15,17 @@
 		clock-frequency = <408000000>;
 	};
 
+	sdhci@78000400 {
+		cd-gpios = <&gpio 170 1>; /* gpio PV2 */
+		bus-width = <4>;
+		status = "okay";
+	};
+
+	sdhci@78000600 {
+		bus-width = <8>;
+		status = "okay";
+	};
+
 	pmc {
 		nvidia,invert-interrupt;
 	};
-- 
1.7.9.5

Re: [PATCH 2/4] ARM: tegra: add SDHCI support for Dalmore

[Stephen Warren]

On 02/20/2013 11:35 AM, Rhyland Klein wrote:
> Based on changes by: Pritesh Raithatha <praithatha@nvidia.com>
> Signed-off-by: Rhyland Klein <rklein@nvidia.com>

A patch description would be useful here. Something simple like:

Dalmore has a built-in eMMC device and a user-accessible SD card slot.
Add device tree nodes to enable these.

The same rebasing comment will apply here too.

Re: [PATCH 2/4] ARM: tegra: add SDHCI support for Dalmore

[Rhyland Klein]

On 2/20/2013 2:07 PM, Stephen Warren wrote:
> On 02/20/2013 11:35 AM, Rhyland Klein wrote:
>> Based on changes by: Pritesh Raithatha <praithatha@nvidia.com>
>> Signed-off-by: Rhyland Klein <rklein@nvidia.com>
> A patch description would be useful here. Something simple like:
>
> Dalmore has a built-in eMMC device and a user-accessible SD card slot.
> Add device tree nodes to enable these.
>
> The same rebasing comment will apply here too.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-tegra" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
Will update patch description for next series.

I can rebase on the cpp usage, but for some stuff like the #define 
replacements, it sounded like things were still being hashed out (like 
.h location, include paths, etc) so I figured its safer not to base on 
those initially. If they are settled by the time i send this out again, 
i can use them as a base.

-rhyland

-- 
nvpublic

Re: [PATCH 2/4] ARM: tegra: add SDHCI support for Dalmore

[Stephen Warren]

On 02/20/2013 12:14 PM, Rhyland Klein wrote:
> On 2/20/2013 2:07 PM, Stephen Warren wrote:
>> On 02/20/2013 11:35 AM, Rhyland Klein wrote:
>>> Based on changes by: Pritesh Raithatha <praithatha@nvidia.com>
>>> Signed-off-by: Rhyland Klein <rklein@nvidia.com>
>> A patch description would be useful here. Something simple like:
>>
>> Dalmore has a built-in eMMC device and a user-accessible SD card slot.
>> Add device tree nodes to enable these.
>>
>> The same rebasing comment will apply here too.
>> -- 
>> To unsubscribe from this list: send the line "unsubscribe linux-tegra" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Will update patch description for next series.
> 
> I can rebase on the cpp usage, but for some stuff like the #define
> replacements, it sounded like things were still being hashed out (like
> .h location, include paths, etc) so I figured its safer not to base on
> those initially. If they are settled by the time i send this out again,
> i can use them as a base.

Yes, it's not all 100% settled yet; I pointed it out more as a heads-up
that I might ask you to rebase later. There's no need to actively rebase
just yet.

Re: [PATCH 2/4] ARM: tegra: add SDHCI support for Dalmore

[Stephen Warren]

On 02/20/2013 12:07 PM, Stephen Warren wrote:
> On 02/20/2013 11:35 AM, Rhyland Klein wrote:
>> Based on changes by: Pritesh Raithatha <praithatha@nvidia.com>
>> Signed-off-by: Rhyland Klein <rklein@nvidia.com>
> 
> A patch description would be useful here. Something simple like:
> 
> Dalmore has a built-in eMMC device and a user-accessible SD card slot.
> Add device tree nodes to enable these.
> 
> The same rebasing comment will apply here too.

I went ahead and applied patches 1/2 and 2/2 to Tegra's for-3.10/dt
branch. I added the commit description I mentioned above.

I think what I'll do for the dtc+cpp usage is to apply patches that come
in before all the dependencies for dtc+cpp are in place, then rebase and
repost my series that converts to use dtc+cpp. That'll save blocking any
other patches on dtc+cpp availability, and save everyone rebasing.

[PATCH 1/4] ARM tegra: add SDHCI nodes with common properties

[Rhyland Klein]

From: Pritesh Raithatha <praithatha-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>

This patch adds in the SDHCI nodes for the busses supported on Tegra114
boards.

Signed-off-by: Pritesh Raithatha <praithatha-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
[Rhyland added clk refs to & reordered sdhci nodes and removed spaces]
Signed-off-by: Rhyland Klein <rklein-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
---
 arch/arm/boot/dts/tegra114.dtsi |   32 ++++++++++++++++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/arch/arm/boot/dts/tegra114.dtsi b/arch/arm/boot/dts/tegra114.dtsi
index 1dfaf28..92d4c7d 100644
--- a/arch/arm/boot/dts/tegra114.dtsi
+++ b/arch/arm/boot/dts/tegra114.dtsi
@@ -92,6 +92,38 @@
 		status = "disabled";
 	};
 
+	sdhci@78000000 {
+		compatible = "nvidia,tegra114-sdhci", "nvidia,tegra30-sdhci";
+		reg = <0x78000000 0x200>;
+		interrupts = <0 14 0x04>;
+		clocks = <&tegra_car 14>;
+		status = "disable";
+	};
+
+	sdhci@78000200 {
+		compatible = "nvidia,tegra114-sdhci", "nvidia,tegra30-sdhci";
+		reg = <0x78000200 0x200>;
+		interrupts = <0 15 0x04>;
+		clocks = <&tegra_car 9>;
+		status = "disable";
+	};
+
+	sdhci@78000400 {
+		compatible = "nvidia,tegra114-sdhci", "nvidia,tegra30-sdhci";
+		reg = <0x78000400 0x200>;
+		interrupts = <0 19 0x04>;
+		clocks = <&tegra_car 69>;
+		status = "disable";
+	};
+
+	sdhci@78000600 {
+		compatible = "nvidia,tegra114-sdhci", "nvidia,tegra30-sdhci";
+		reg = <0x78000600 0x200>;
+		interrupts = <0 31 0x04>;
+		clocks = <&tegra_car 15>;
+		status = "disable";
+	};
+
 	rtc {
 		compatible = "nvidia,tegra114-rtc", "nvidia,tegra20-rtc";
 		reg = <0x7000e000 0x100>;
-- 
1.7.9.5

Re: [PATCH 1/4] ARM tegra: add SDHCI nodes with common properties

[Stephen Warren]

On 02/20/2013 11:35 AM, Rhyland Klein wrote:
> From: Pritesh Raithatha <praithatha@nvidia.com>
> 
> This patch adds in the SDHCI nodes for the busses supported on Tegra114
> boards.

This patch looks fine, although it'd need some modifications to rebase
it on top of the cpp usage in *.dts, which I believe I'll check in early
for 3.10.

[PATCH 3/4] mmc: host: sdhci-tegra: cleanup ifdefs

[Rhyland Klein]

The structs wrapped with the SOC ifdefs are small enough where having
them always there shouldn't be a big overhead. Removing the ifdefs
also makes the code a little cleaner.

Signed-off-by: Rhyland Klein <rklein-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
---
 drivers/mmc/host/sdhci-tegra.c |    8 --------
 1 file changed, 8 deletions(-)

diff --git a/drivers/mmc/host/sdhci-tegra.c b/drivers/mmc/host/sdhci-tegra.c
index 1276131..5e126bf 100644
--- a/drivers/mmc/host/sdhci-tegra.c
+++ b/drivers/mmc/host/sdhci-tegra.c
@@ -173,7 +173,6 @@ static struct sdhci_ops tegra_sdhci_ops = {
 	.platform_reset_exit = tegra_sdhci_reset_exit,
 };
 
-#ifdef CONFIG_ARCH_TEGRA_2x_SOC
 static struct sdhci_pltfm_data sdhci_tegra20_pdata = {
 	.quirks = SDHCI_QUIRK_BROKEN_TIMEOUT_VAL |
 		  SDHCI_QUIRK_SINGLE_POWER_WRITE |
@@ -187,9 +186,7 @@ static struct sdhci_tegra_soc_data soc_data_tegra20 = {
 	.nvquirks = NVQUIRK_FORCE_SDHCI_SPEC_200 |
 		    NVQUIRK_ENABLE_BLOCK_GAP_DET,
 };
-#endif
 
-#ifdef CONFIG_ARCH_TEGRA_3x_SOC
 static struct sdhci_pltfm_data sdhci_tegra30_pdata = {
 	.quirks = SDHCI_QUIRK_BROKEN_TIMEOUT_VAL |
 		  SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK |
@@ -203,15 +200,10 @@ static struct sdhci_tegra_soc_data soc_data_tegra30 = {
 	.pdata = &sdhci_tegra30_pdata,
 	.nvquirks = NVQUIRK_ENABLE_SDHCI_SPEC_300,
 };
-#endif
 
 static const struct of_device_id sdhci_tegra_dt_match[] = {
-#ifdef CONFIG_ARCH_TEGRA_3x_SOC
 	{ .compatible = "nvidia,tegra30-sdhci", .data = &soc_data_tegra30 },
-#endif
-#ifdef CONFIG_ARCH_TEGRA_2x_SOC
 	{ .compatible = "nvidia,tegra20-sdhci", .data = &soc_data_tegra20 },
-#endif
 	{}
 };
 MODULE_DEVICE_TABLE(of, sdhci_dt_ids);
-- 
1.7.9.5

Re: [PATCH 3/4] mmc: host: sdhci-tegra: cleanup ifdefs

[Stephen Warren]

On 02/20/2013 11:35 AM, Rhyland Klein wrote:
> The structs wrapped with the SOC ifdefs are small enough where having
> them always there shouldn't be a big overhead. Removing the ifdefs
> also makes the code a little cleaner.

Patch 3 and 4 to the MMC driver look fine. But I'd ask Chris to hold off
on applying them for 3.10 until I've sorted out the dependencies of
Joseph Lo's MMC patches; if I can get the device tree fix for the CD
GPIO polarity into 3.9/stable as a bug-fix, there will be no dependency
issue I think. Otherwise, I'll probably want to take all the Tegra SDHCI
changes through the Tegra tree for 3.10 to resolve the dependencies.

Re: [PATCH 3/4] mmc: host: sdhci-tegra: cleanup ifdefs

[Stephen Warren]

On 02/20/2013 11:35 AM, Rhyland Klein wrote:
> The structs wrapped with the SOC ifdefs are small enough where having
> them always there shouldn't be a big overhead. Removing the ifdefs
> also makes the code a little cleaner.

Chris,

I think patches 3 and 4 of this series can go through either the MMC
tree (I checked there are no conflicts with the MMC patch I have to take
through the Tegra tree), or the Tegra tree (since I'm taking another
Tegra MMC patch though there), both without any dependency issues.

It's your call. Let me know if you want me to take them through the
Tegra tree with an ack. Or, if you want to take them through the MMC
tree that's fine too. If you need them reposted to see the content, let
me know, or they're at:

http://patchwork.ozlabs.org/patch/222136/
http://patchwork.ozlabs.org/patch/222138/

Thanks.

Re: [PATCH 3/4] mmc: host: sdhci-tegra: cleanup ifdefs

[Chris Ball]

Hi,

On Wed, Mar 06 2013, Stephen Warren wrote:
> On 02/20/2013 11:35 AM, Rhyland Klein wrote:
>> The structs wrapped with the SOC ifdefs are small enough where having
>> them always there shouldn't be a big overhead. Removing the ifdefs
>> also makes the code a little cleaner.
>
> I think patches 3 and 4 of this series can go through either the MMC
> tree (I checked there are no conflicts with the MMC patch I have to take
> through the Tegra tree), or the Tegra tree (since I'm taking another
> Tegra MMC patch though there), both without any dependency issues.
>
> It's your call. Let me know if you want me to take them through the
> Tegra tree with an ack. Or, if you want to take them through the MMC
> tree that's fine too. If you need them reposted to see the content, let
> me know, or they're at:

Via the MMC tree with an ack sounds good, just for the mundane reason
that I like my "highlights for 3.10" pull request to Linus to mention
when support for new MMC hardware is added, and it won't do that if
the tegra110 patch is coming through your tree.

Thanks,

- Chris.
-- 
Chris Ball   <cjb-2X9k7bc8m7Mdnm+yROfE0A@public.gmane.org>   <http://printf.net/>
One Laptop Per Child

Re: [PATCH 3/4] mmc: host: sdhci-tegra: cleanup ifdefs

[Stephen Warren]

On 03/06/2013 03:54 PM, Chris Ball wrote:
> Hi,
> 
> On Wed, Mar 06 2013, Stephen Warren wrote:
>> On 02/20/2013 11:35 AM, Rhyland Klein wrote:
>>> The structs wrapped with the SOC ifdefs are small enough where having
>>> them always there shouldn't be a big overhead. Removing the ifdefs
>>> also makes the code a little cleaner.
>>
>> I think patches 3 and 4 of this series can go through either the MMC
>> tree (I checked there are no conflicts with the MMC patch I have to take
>> through the Tegra tree), or the Tegra tree (since I'm taking another
>> Tegra MMC patch though there), both without any dependency issues.
>>
>> It's your call. Let me know if you want me to take them through the
>> Tegra tree with an ack. Or, if you want to take them through the MMC
>> tree that's fine too. If you need them reposted to see the content, let
>> me know, or they're at:
> 
> Via the MMC tree with an ack sounds good

OK, thanks. Patches 3/4 and 4/4,
Acked-by: Stephen Warren <swarren-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>

Re: [PATCH 3/4] mmc: host: sdhci-tegra: cleanup ifdefs

[Chris Ball]

Hi,

On Wed, Mar 06 2013, Stephen Warren wrote:
>> Via the MMC tree with an ack sounds good
>
> OK, thanks. Patches 3/4 and 4/4,
> Acked-by: Stephen Warren <swarren-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>

Thanks, both pushed out to mmc-next for 3.10 now.

- Chris.
-- 
Chris Ball   <cjb-2X9k7bc8m7Mdnm+yROfE0A@public.gmane.org>   <http://printf.net/>
One Laptop Per Child

[PATCH 4/4] mmc: host: sdhci-tegra: add basic support for Tegra114

[Rhyland Klein]

The mmc controller on Tegra114 platforms is basically compatible with
the settings used for Tegra30. However there is a difference where we
don't need the extra ENABLE_SDHCI_SPEC_300 quirk as Tegra114 hardware
advertises v3.0 support already.

Signed-off-by: Rhyland Klein <rklein-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
---
 drivers/mmc/host/sdhci-tegra.c |   14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/drivers/mmc/host/sdhci-tegra.c b/drivers/mmc/host/sdhci-tegra.c
index 5e126bf..931d9d7 100644
--- a/drivers/mmc/host/sdhci-tegra.c
+++ b/drivers/mmc/host/sdhci-tegra.c
@@ -201,7 +201,21 @@ static struct sdhci_tegra_soc_data soc_data_tegra30 = {
 	.nvquirks = NVQUIRK_ENABLE_SDHCI_SPEC_300,
 };
 
+static struct sdhci_pltfm_data sdhci_tegra114_pdata = {
+	.quirks = SDHCI_QUIRK_BROKEN_TIMEOUT_VAL |
+		  SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK |
+		  SDHCI_QUIRK_SINGLE_POWER_WRITE |
+		  SDHCI_QUIRK_NO_HISPD_BIT |
+		  SDHCI_QUIRK_BROKEN_ADMA_ZEROLEN_DESC,
+	.ops  = &tegra_sdhci_ops,
+};
+
+static struct sdhci_tegra_soc_data soc_data_tegra114 = {
+	.pdata = &sdhci_tegra114_pdata,
+};
+
 static const struct of_device_id sdhci_tegra_dt_match[] = {
+	{ .compatible = "nvidia,tegra114-sdhci", .data = &soc_data_tegra114 },
 	{ .compatible = "nvidia,tegra30-sdhci", .data = &soc_data_tegra30 },
 	{ .compatible = "nvidia,tegra20-sdhci", .data = &soc_data_tegra20 },
 	{}
-- 
1.7.9.5