[PATCH v2 0/2] Improve mvebu_get_soc_id

[PATCH v2 0/2] Improve mvebu_get_soc_id

[Gregory CLEMENT]

Hi,

The first use -ENODEV as return value instead of -1 as suggested by
Arnd for the second patch.

The second patch allows to get the SoC Id and the revision without using
the PCI register on Armada 38x.

Thanks,

Gregory

Changelog:

v1 -> v2:

- Fix typo in mvebu_system_controller_get_soc_id.
- Use -ENODEV as return value instead of -1


Gregory CLEMENT (2):
  ARM: mvebu: Use the a standard errno in mvebu_get_soc_id
  ARM: mvebu: Use system controller to get the soc id when possible

 arch/arm/mach-mvebu/common.h            |  1 +
 arch/arm/mach-mvebu/mvebu-soc-id.c      | 22 ++++++++++++++++++++--
 arch/arm/mach-mvebu/system-controller.c | 19 +++++++++++++++++++
 3 files changed, 40 insertions(+), 2 deletions(-)

-- 
1.8.1.2

[PATCH v2 1/2] ARM: mvebu: Use the a standard errno in mvebu_get_soc_id

[Gregory CLEMENT]

Instead of using -1 as error value, use a standard errno.

Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
---
 arch/arm/mach-mvebu/mvebu-soc-id.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/mach-mvebu/mvebu-soc-id.c b/arch/arm/mach-mvebu/mvebu-soc-id.c
index d0f35b4d4a23..12c66cac967d 100644
--- a/arch/arm/mach-mvebu/mvebu-soc-id.c
+++ b/arch/arm/mach-mvebu/mvebu-soc-id.c
@@ -51,7 +51,7 @@ int mvebu_get_soc_id(u32 *dev, u32 *rev)
 		*rev = soc_rev;
 		return 0;
 	} else
-		return -1;
+		return -ENODEV;
 }
 
 static int __init mvebu_soc_id_init(void)
-- 
1.8.1.2

[PATCH v2 1/2] ARM: mvebu: Use the a standard errno in mvebu_get_soc_id

[Thomas Petazzoni]

Dear Gregory CLEMENT,

On Fri, 20 Jun 2014 16:35:52 +0200, Gregory CLEMENT wrote:
> Instead of using -1 as error value, use a standard errno.
> 
> Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
> ---
>  arch/arm/mach-mvebu/mvebu-soc-id.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Acked-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

[PATCH v2 1/2] ARM: mvebu: Use the a standard errno in mvebu_get_soc_id

[Jason Cooper]

On Fri, Jun 20, 2014 at 04:35:52PM +0200, Gregory CLEMENT wrote:
> Instead of using -1 as error value, use a standard errno.
> 
> Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
> ---
>  arch/arm/mach-mvebu/mvebu-soc-id.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Applied to mvebu/soc

thx,

Jason.

[PATCH v2 2/2] ARM: mvebu: Use system controller to get the soc id when possible

[Gregory CLEMENT]

On Armada 38x it is possible to get the SoC Id and the revision
without using the PCI register. Accessing the PCI registers implies
enabling its clock and, because of the initialization issue, not
keeping them enable. So if possible it is better to avoid it.

Armada 370 and Armada XP provides the SoC ID values from the system
controller but not the revision.

Armada 375 provides both but the SoC ID value looks buggy (0x6660
instead of 0x6720).

Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
---
 arch/arm/mach-mvebu/common.h            |  1 +
 arch/arm/mach-mvebu/mvebu-soc-id.c      | 20 +++++++++++++++++++-
 arch/arm/mach-mvebu/system-controller.c | 19 +++++++++++++++++++
 3 files changed, 39 insertions(+), 1 deletion(-)

diff --git a/arch/arm/mach-mvebu/common.h b/arch/arm/mach-mvebu/common.h
index b67fb7a10d8b..927d88627a6e 100644
--- a/arch/arm/mach-mvebu/common.h
+++ b/arch/arm/mach-mvebu/common.h
@@ -21,6 +21,7 @@ void mvebu_restart(enum reboot_mode mode, const char *cmd);
 int mvebu_cpu_reset_deassert(int cpu);
 void mvebu_pmsu_set_cpu_boot_addr(int hw_cpu, void *boot_addr);
 void mvebu_system_controller_set_cpu_boot_addr(void *boot_addr);
+int mvebu_system_controller_get_soc_id(u32 *dev, u32 *rev);
 
 void armada_xp_cpu_die(unsigned int cpu);
 
diff --git a/arch/arm/mach-mvebu/mvebu-soc-id.c b/arch/arm/mach-mvebu/mvebu-soc-id.c
index 12c66cac967d..922b1380b0be 100644
--- a/arch/arm/mach-mvebu/mvebu-soc-id.c
+++ b/arch/arm/mach-mvebu/mvebu-soc-id.c
@@ -25,8 +25,10 @@
 #include <linux/of_address.h>
 #include <linux/slab.h>
 #include <linux/sys_soc.h>
+#include "common.h"
 #include "mvebu-soc-id.h"
 
+
 #define PCIE_DEV_ID_OFF		0x0
 #define PCIE_DEV_REV_OFF	0x8
 
@@ -54,7 +56,7 @@ int mvebu_get_soc_id(u32 *dev, u32 *rev)
 		return -ENODEV;
 }
 
-static int __init mvebu_soc_id_init(void)
+static int __init get_soc_id_by_pci(void)
 {
 	struct device_node *np;
 	int ret = 0;
@@ -129,6 +131,22 @@ clk_err:
 
 	return ret;
 }
+
+static int __init mvebu_soc_id_init(void)
+{
+
+	/*
+	 * First try to get the ID and the revision by the system
+	 * regsiter and use PCI registers only if it is not possible
+	 */
+	if (!mvebu_system_controller_get_soc_id(&soc_dev_id, &soc_rev)) {
+		is_id_valid = true;
+		pr_info("MVEBU SoC ID=0x%X, Rev=0x%X\n", soc_dev_id, soc_rev);
+		return 0;
+	}
+
+	return get_soc_id_by_pci();
+}
 early_initcall(mvebu_soc_id_init);
 
 static int __init mvebu_soc_device(void)
diff --git a/arch/arm/mach-mvebu/system-controller.c b/arch/arm/mach-mvebu/system-controller.c
index 0c5524ac75b7..de90821129da 100644
--- a/arch/arm/mach-mvebu/system-controller.c
+++ b/arch/arm/mach-mvebu/system-controller.c
@@ -39,6 +39,9 @@ struct mvebu_system_controller {
 	u32 system_soft_reset;
 
 	u32 resume_boot_addr;
+
+	u32 dev_id;
+	u32 rev_id;
 };
 static struct mvebu_system_controller *mvebu_sc;
 
@@ -47,6 +50,8 @@ static const struct mvebu_system_controller armada_370_xp_system_controller = {
 	.system_soft_reset_offset = 0x64,
 	.rstoutn_mask_reset_out_en = 0x1,
 	.system_soft_reset = 0x1,
+	.dev_id = 0x38,
+	.rev_id = 0x3c,
 };
 
 static const struct mvebu_system_controller armada_375_system_controller = {
@@ -55,6 +60,8 @@ static const struct mvebu_system_controller armada_375_system_controller = {
 	.rstoutn_mask_reset_out_en = 0x1,
 	.system_soft_reset = 0x1,
 	.resume_boot_addr = 0xd4,
+	.dev_id = 0x38,
+	.rev_id = 0x3c,
 };
 
 static const struct mvebu_system_controller orion_system_controller = {
@@ -101,6 +108,18 @@ void mvebu_restart(enum reboot_mode mode, const char *cmd)
 		;
 }
 
+int mvebu_system_controller_get_soc_id(u32 *dev, u32 *rev)
+{
+	if (of_machine_is_compatible("marvell,armada38x") &&
+		system_controller_base) {
+		*dev = readl(system_controller_base + mvebu_sc->dev_id) >> 16;
+		*rev = (readl(system_controller_base + mvebu_sc->rev_id) >> 8)
+			& 0xF;
+		return 0;
+	} else
+		return -ENODEV;
+}
+
 #ifdef CONFIG_SMP
 void mvebu_system_controller_set_cpu_boot_addr(void *boot_addr)
 {
-- 
1.8.1.2

[PATCH v2 2/2] ARM: mvebu: Use system controller to get the soc id when possible

[Ezequiel Garcia]

On 20 Jun 04:35 PM, Gregory CLEMENT wrote:
> diff --git a/arch/arm/mach-mvebu/mvebu-soc-id.c b/arch/arm/mach-mvebu/mvebu-soc-id.c
> index 12c66cac967d..922b1380b0be 100644
> --- a/arch/arm/mach-mvebu/mvebu-soc-id.c
> +++ b/arch/arm/mach-mvebu/mvebu-soc-id.c
> @@ -25,8 +25,10 @@
>  #include <linux/of_address.h>
>  #include <linux/slab.h>
>  #include <linux/sys_soc.h>
> +#include "common.h"
>  #include "mvebu-soc-id.h"
>  
> +

Spurious whitespace?

>  #define PCIE_DEV_ID_OFF		0x0
>  #define PCIE_DEV_REV_OFF	0x8
>  
-- 
Ezequiel Garc?a, Free Electrons
Embedded Linux, Kernel and Android Engineering
http://free-electrons.com

[PATCH v2 2/2] ARM: mvebu: Use system controller to get the soc id when possible

[Jason Cooper]

On Fri, Jun 20, 2014 at 04:35:53PM +0200, Gregory CLEMENT wrote:
> On Armada 38x it is possible to get the SoC Id and the revision
> without using the PCI register. Accessing the PCI registers implies
> enabling its clock and, because of the initialization issue, not
> keeping them enable. So if possible it is better to avoid it.
> 
> Armada 370 and Armada XP provides the SoC ID values from the system
> controller but not the revision.
> 
> Armada 375 provides both but the SoC ID value looks buggy (0x6660
> instead of 0x6720).
> 
> Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
> ---
>  arch/arm/mach-mvebu/common.h            |  1 +
>  arch/arm/mach-mvebu/mvebu-soc-id.c      | 20 +++++++++++++++++++-
>  arch/arm/mach-mvebu/system-controller.c | 19 +++++++++++++++++++
>  3 files changed, 39 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/arm/mach-mvebu/common.h b/arch/arm/mach-mvebu/common.h
> index b67fb7a10d8b..927d88627a6e 100644
> --- a/arch/arm/mach-mvebu/common.h
> +++ b/arch/arm/mach-mvebu/common.h
> @@ -21,6 +21,7 @@ void mvebu_restart(enum reboot_mode mode, const char *cmd);
>  int mvebu_cpu_reset_deassert(int cpu);
>  void mvebu_pmsu_set_cpu_boot_addr(int hw_cpu, void *boot_addr);
>  void mvebu_system_controller_set_cpu_boot_addr(void *boot_addr);
> +int mvebu_system_controller_get_soc_id(u32 *dev, u32 *rev);
>  
>  void armada_xp_cpu_die(unsigned int cpu);
>  
> diff --git a/arch/arm/mach-mvebu/mvebu-soc-id.c b/arch/arm/mach-mvebu/mvebu-soc-id.c
> index 12c66cac967d..922b1380b0be 100644
> --- a/arch/arm/mach-mvebu/mvebu-soc-id.c
> +++ b/arch/arm/mach-mvebu/mvebu-soc-id.c
> @@ -25,8 +25,10 @@
>  #include <linux/of_address.h>
>  #include <linux/slab.h>
>  #include <linux/sys_soc.h>
> +#include "common.h"
>  #include "mvebu-soc-id.h"
>  
> +
>  #define PCIE_DEV_ID_OFF		0x0

whitespace?

>  #define PCIE_DEV_REV_OFF	0x8
>  
> @@ -54,7 +56,7 @@ int mvebu_get_soc_id(u32 *dev, u32 *rev)
>  		return -ENODEV;
>  }
>  
> -static int __init mvebu_soc_id_init(void)
> +static int __init get_soc_id_by_pci(void)
>  {
>  	struct device_node *np;
>  	int ret = 0;
> @@ -129,6 +131,22 @@ clk_err:
>  
>  	return ret;
>  }
> +
> +static int __init mvebu_soc_id_init(void)
> +{
> +
> +	/*
> +	 * First try to get the ID and the revision by the system
> +	 * regsiter and use PCI registers only if it is not possible

s/regsiter/register/

> +	 */
> +	if (!mvebu_system_controller_get_soc_id(&soc_dev_id, &soc_rev)) {
> +		is_id_valid = true;
> +		pr_info("MVEBU SoC ID=0x%X, Rev=0x%X\n", soc_dev_id, soc_rev);
> +		return 0;
> +	}
> +
> +	return get_soc_id_by_pci();
> +}
>  early_initcall(mvebu_soc_id_init);
>  
>  static int __init mvebu_soc_device(void)
> diff --git a/arch/arm/mach-mvebu/system-controller.c b/arch/arm/mach-mvebu/system-controller.c
> index 0c5524ac75b7..de90821129da 100644
> --- a/arch/arm/mach-mvebu/system-controller.c
> +++ b/arch/arm/mach-mvebu/system-controller.c
> @@ -39,6 +39,9 @@ struct mvebu_system_controller {
>  	u32 system_soft_reset;
>  
>  	u32 resume_boot_addr;
> +
> +	u32 dev_id;
> +	u32 rev_id;
>  };
>  static struct mvebu_system_controller *mvebu_sc;
>  
> @@ -47,6 +50,8 @@ static const struct mvebu_system_controller armada_370_xp_system_controller = {
>  	.system_soft_reset_offset = 0x64,
>  	.rstoutn_mask_reset_out_en = 0x1,
>  	.system_soft_reset = 0x1,
> +	.dev_id = 0x38,
> +	.rev_id = 0x3c,
>  };
>  
>  static const struct mvebu_system_controller armada_375_system_controller = {
> @@ -55,6 +60,8 @@ static const struct mvebu_system_controller armada_375_system_controller = {
>  	.rstoutn_mask_reset_out_en = 0x1,
>  	.system_soft_reset = 0x1,
>  	.resume_boot_addr = 0xd4,
> +	.dev_id = 0x38,
> +	.rev_id = 0x3c,
>  };
>  
>  static const struct mvebu_system_controller orion_system_controller = {
> @@ -101,6 +108,18 @@ void mvebu_restart(enum reboot_mode mode, const char *cmd)
>  		;
>  }
>  
> +int mvebu_system_controller_get_soc_id(u32 *dev, u32 *rev)
> +{
> +	if (of_machine_is_compatible("marvell,armada38x") &&

As Sergei commented on your documentation patch, we try to avoid
wildcards in compatible strings.  Perhaps a match list would be more
appropriate?

> +		system_controller_base) {
> +		*dev = readl(system_controller_base + mvebu_sc->dev_id) >> 16;
> +		*rev = (readl(system_controller_base + mvebu_sc->rev_id) >> 8)
> +			& 0xF;
> +		return 0;
> +	} else
> +		return -ENODEV;
> +}
> +
>  #ifdef CONFIG_SMP
>  void mvebu_system_controller_set_cpu_boot_addr(void *boot_addr)
>  {
> -- 
> 1.8.1.2
> 

thx,

Jason.

[PATCH v2 2/2] ARM: mvebu: Use system controller to get the soc id when possible

[Andrew Lunn]

> > +int mvebu_system_controller_get_soc_id(u32 *dev, u32 *rev)
> > +{
> > +	if (of_machine_is_compatible("marvell,armada38x") &&
> 
> As Sergei commented on your documentation patch, we try to avoid
> wildcards in compatible strings.  Perhaps a match list would be more
> appropriate?

I'm being pedantic, but in this invocation, it is not a wildcard. It will
match one and only one compatible string.

However, we have the issue that 38x is not introduced in this patch,
it has been there for a while. So would you like another patch series
removing it from the DT files? A quick grep suggests nothing is using
"marvell,armada38x" except this patch. So it should not be a backwards
compatible issue.

	   Andrew

[PATCH v2 2/2] ARM: mvebu: Use system controller to get the soc id when possible

[Jason Cooper]

On Fri, Jun 20, 2014 at 09:12:04PM +0200, Andrew Lunn wrote:
> > > +int mvebu_system_controller_get_soc_id(u32 *dev, u32 *rev)
> > > +{
> > > +	if (of_machine_is_compatible("marvell,armada38x") &&
> > 
> > As Sergei commented on your documentation patch, we try to avoid
> > wildcards in compatible strings.  Perhaps a match list would be more
> > appropriate?
> 
> I'm being pedantic, but in this invocation, it is not a wildcard. It will
> match one and only one compatible string.
> 
> However, we have the issue that 38x is not introduced in this patch,
> it has been there for a while. So would you like another patch series
> removing it from the DT files? 

Please do.  Sorry I missed it the first time around.  Also, may as well
put the documentation patch into the same series.

> A quick grep suggests nothing is using "marvell,armada38x" except this
> patch. So it should not be a backwards compatible issue.

Agreed.

thx,

Jason.

[PATCH v2 0/2] Improve mvebu_get_soc_id

[Andrew Lunn]

On Fri, Jun 20, 2014 at 04:35:51PM +0200, Gregory CLEMENT wrote:
> Hi,
> 
> The first use -ENODEV as return value instead of -1 as suggested by
> Arnd for the second patch.
> 
> The second patch allows to get the SoC Id and the revision without using
> the PCI register on Armada 38x.

Hi Gregory

I just tested on Kirkwood. No regression.

Once we get the compatible string worked out, i will post a Tested-by:

     Andrew