[RFC PATCH 0/2] ARM: update cpuinfo to print CPU model name

[RFC PATCH 0/2] ARM: update cpuinfo to print CPU model name

[Ruslan Bilovol]

Hi,

The following patches update cpuinfo to print CPU
model name for ARM. First patch exactly makes needed
changes for ARM architecture.
Second patch adds this ability to OMAP4 SoCs.

This adds a common approach to show SoC name.

Looks like there were few attempts to do similar
changes to cpuinfo without any luck.

So - comments are welcome

Ruslan Bilovol (2):
  ARM: kernel: update cpuinfo to print CPU model name
  ARM: OMAP4: setup CPU model name during ID initialisation

 arch/arm/include/asm/setup.h |    1 +
 arch/arm/kernel/setup.c      |   10 ++++++++++
 arch/arm/mach-omap2/id.c     |   18 ++++++++++++++++--
 3 files changed, 27 insertions(+), 2 deletions(-)

-- 
1.7.9.5

[RFC PATCH 1/2] ARM: kernel: update cpuinfo to print CPU model name

[Ruslan Bilovol]

Currently, reading /proc/cpuinfo provides userspace with CPU ID of
the CPU carrying out the read from the file.
Userspace using this information may decide what module
to load or how to configure some specific (and processor-depended)
settings or so.
However, since really different SoCs can share same ARM core,
this information currently is not so useful.
For example, TI OMAP4460 and OMAP4470 SoCs show the same
information in the /proc/cpuinfo whereas they are different.
The "cpuinfo" file looks like exactly that place, where
This information have to be displayed.

So added new line "CPU name" in the "cpuinfo" output for CPU model name

Example:

/ # cat proc/cpuinfo
processor       : 0
model name      : ARMv7 Processor rev 10 (v7l)
BogoMIPS        : 1590.23
Features        : swp half thumb fastmult vfp edsp thumbee neon vfpv3 tls
CPU implementer : 0x41
CPU name        : OMAP4470 ES1.0 HS
CPU architecture: 7
CPU variant     : 0x2
CPU part        : 0xc09
CPU revision    : 10
[...]

Signed-off-by: Ruslan Bilovol <ruslan.bilovol@ti.com>
---
 arch/arm/include/asm/setup.h |    1 +
 arch/arm/kernel/setup.c      |   10 ++++++++++
 2 files changed, 11 insertions(+)

diff --git a/arch/arm/include/asm/setup.h b/arch/arm/include/asm/setup.h
index c50f056..f49de38 100644
--- a/arch/arm/include/asm/setup.h
+++ b/arch/arm/include/asm/setup.h
@@ -52,5 +52,6 @@ extern struct meminfo meminfo;
 extern int arm_add_memory(phys_addr_t start, phys_addr_t size);
 extern void early_print(const char *str, ...);
 extern void dump_machine_table(void);
+extern void set_cpu_model_name(char *name);
 
 #endif
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 3f6cbb2..6fab2e2 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -134,6 +134,7 @@ char elf_platform[ELF_PLATFORM_SIZE];
 EXPORT_SYMBOL(elf_platform);
 
 static const char *cpu_name;
+static const char *cpu_model_name;
 static const char *machine_name;
 static char __initdata cmd_line[COMMAND_LINE_SIZE];
 struct machine_desc *machine_desc __initdata;
@@ -493,6 +494,11 @@ static void __init setup_processor(void)
 	cpu_init();
 }
 
+void set_cpu_model_name(char *name)
+{
+	cpu_model_name = name;
+}
+
 void __init dump_machine_table(void)
 {
 	struct machine_desc *p;
@@ -880,6 +886,10 @@ static int c_show(struct seq_file *m, void *v)
 				seq_printf(m, "%s ", hwcap_str[j]);
 
 		seq_printf(m, "\nCPU implementer\t: 0x%02x\n", cpuid >> 24);
+
+		if (cpu_model_name)
+			seq_printf(m, "CPU name\t: %s\n", cpu_model_name);
+
 		seq_printf(m, "CPU architecture: %s\n",
 			   proc_arch[cpu_architecture()]);
 
-- 
1.7.9.5

[RFC PATCH 1/2] ARM: kernel: update cpuinfo to print CPU model name

[Russell King - ARM Linux]

On Tue, Jan 29, 2013 at 05:54:24PM +0200, Ruslan Bilovol wrote:
> CPU implementer : 0x41
> CPU name        : OMAP4470 ES1.0 HS

Sigh.  No.  Look at what you're doing - look carefully at the above.

"CPU implementer" - 0x41.  That's A.  For ARM Ltd.  ARM Ltd implemented
this CPU.  Did ARM Ltd really implement OMAP4470 ?  I think TI would be
very upset if that were to be the case.

So no, OMAP4470 is _NOT_ a CPU.  It is a SoC.  The CPU inside the SoC is
a collection of ARM Ltd Cortex A9 _CPUs_.

See?  Please, learn what a CPU is as opposed to a SoC.

[RFC PATCH 1/2] ARM: kernel: update cpuinfo to print CPU model name

[Ruslan Bilovol]

Hi,

On Tue, Jan 29, 2013 at 6:08 PM, Russell King - ARM Linux
<linux@arm.linux.org.uk> wrote:
> On Tue, Jan 29, 2013 at 05:54:24PM +0200, Ruslan Bilovol wrote:
>> CPU implementer : 0x41
>> CPU name        : OMAP4470 ES1.0 HS
>
> Sigh.  No.  Look at what you're doing - look carefully at the above.
>
> "CPU implementer" - 0x41.  That's A.  For ARM Ltd.  ARM Ltd implemented
> this CPU.  Did ARM Ltd really implement OMAP4470 ?  I think TI would be
> very upset if that were to be the case.

Yes, it would be very surprisingly :)

>
> So no, OMAP4470 is _NOT_ a CPU.  It is a SoC.  The CPU inside the SoC is
> a collection of ARM Ltd Cortex A9 _CPUs_.
>
> See?  Please, learn what a CPU is as opposed to a SoC.

Completely agree with you. I will fix this

[RFC PATCH 1/2] ARM: kernel: update cpuinfo to print CPU model name

[Santosh Shilimkar]

On Wednesday 30 January 2013 04:42 AM, Ruslan Bilovol wrote:
> Hi,
>
> On Tue, Jan 29, 2013 at 6:08 PM, Russell King - ARM Linux
> <linux@arm.linux.org.uk> wrote:
>> On Tue, Jan 29, 2013 at 05:54:24PM +0200, Ruslan Bilovol wrote:
>>> CPU implementer : 0x41
>>> CPU name        : OMAP4470 ES1.0 HS
>>
>> Sigh.  No.  Look at what you're doing - look carefully at the above.
>>
>> "CPU implementer" - 0x41.  That's A.  For ARM Ltd.  ARM Ltd implemented
>> this CPU.  Did ARM Ltd really implement OMAP4470 ?  I think TI would be
>> very upset if that were to be the case.
>
> Yes, it would be very surprisingly :)
>
>>
>> So no, OMAP4470 is _NOT_ a CPU.  It is a SoC.  The CPU inside the SoC is
>> a collection of ARM Ltd Cortex A9 _CPUs_.
>>
>> See?  Please, learn what a CPU is as opposed to a SoC.
>
> Completely agree with you. I will fix this
>
Thank god you agreed to drop your current approach. Please elaborate
what you are going to fix and also state what user-space features
changes from OMAP4460 to OMAP4470.

Regards,
Santosh

[RFC PATCH 1/2] ARM: kernel: update cpuinfo to print CPU model name

[anish singh]

On Wed, Jan 30, 2013 at 11:22 AM, Santosh Shilimkar
<santosh.shilimkar@ti.com> wrote:
> On Wednesday 30 January 2013 04:42 AM, Ruslan Bilovol wrote:
>>
>> Hi,
>>
>> On Tue, Jan 29, 2013 at 6:08 PM, Russell King - ARM Linux
>> <linux@arm.linux.org.uk> wrote:
>>>
>>> On Tue, Jan 29, 2013 at 05:54:24PM +0200, Ruslan Bilovol wrote:
>>>>
>>>> CPU implementer : 0x41
>>>> CPU name        : OMAP4470 ES1.0 HS
>>>
>>>
>>> Sigh.  No.  Look at what you're doing - look carefully at the above.
>>>
>>> "CPU implementer" - 0x41.  That's A.  For ARM Ltd.  ARM Ltd implemented
Wouldn't it be nice to display the name rather than hex?I think this would have
definitely occurred to Russell but it is still done this way.I think
there would be
a valid reason for this and it would be nice to know that.

If not then I can cook up a patch.

>>> this CPU.  Did ARM Ltd really implement OMAP4470 ?  I think TI would be
>>> very upset if that were to be the case.
>>
>>
>> Yes, it would be very surprisingly :)
>>
>>>
>>> So no, OMAP4470 is _NOT_ a CPU.  It is a SoC.  The CPU inside the SoC is
>>> a collection of ARM Ltd Cortex A9 _CPUs_.
>>>
>>> See?  Please, learn what a CPU is as opposed to a SoC.
>>
>>
>> Completely agree with you. I will fix this
>>
> Thank god you agreed to drop your current approach. Please elaborate
> what you are going to fix and also state what user-space features
> changes from OMAP4460 to OMAP4470.
>
> Regards,
> Santosh
>
>
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo at vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

[RFC PATCH 2/2] ARM: OMAP4: setup CPU model name during ID initialisation

[Ruslan Bilovol]

Set up the CPU model name during OMAP ID initialisation
so it will be displayed in /proc/cpuinfo:

/ # cat proc/cpuinfo
processor       : 0
model name      : ARMv7 Processor rev 10 (v7l)
BogoMIPS        : 1590.23
Features        : swp half thumb fastmult vfp edsp thumbee neon vfpv3 tls
CPU implementer : 0x41
CPU name        : OMAP4470 ES1.0 HS
CPU architecture: 7
CPU variant     : 0x2
CPU part        : 0xc09
CPU revision    : 10
[...]

Signed-off-by: Ruslan Bilovol <ruslan.bilovol@ti.com>
---
 arch/arm/mach-omap2/id.c |   18 ++++++++++++++++--
 1 file changed, 16 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-omap2/id.c b/arch/arm/mach-omap2/id.c
index 45cc7ed4..d1416e9 100644
--- a/arch/arm/mach-omap2/id.c
+++ b/arch/arm/mach-omap2/id.c
@@ -20,6 +20,7 @@
 #include <linux/io.h>
 
 #include <asm/cputype.h>
+#include <asm/setup.h>
 
 #include "common.h"
 
@@ -33,8 +34,17 @@
 
 static unsigned int omap_revision;
 static const char *cpu_rev;
+static char omap_cpu_model_name[80];
 u32 omap_features;
 
+static const char const *omap_types[] = {
+	[OMAP2_DEVICE_TYPE_TEST]	= "TST",
+	[OMAP2_DEVICE_TYPE_EMU]		= "EMU",
+	[OMAP2_DEVICE_TYPE_SEC]		= "HS",
+	[OMAP2_DEVICE_TYPE_GP]		= "GP",
+	[OMAP2_DEVICE_TYPE_BAD]		= "BAD",
+};
+
 unsigned int omap_rev(void)
 {
 	return omap_revision;
@@ -502,8 +512,12 @@ void __init omap4xxx_check_revision(void)
 		omap_revision = OMAP4430_REV_ES2_3;
 	}
 
-	pr_info("OMAP%04x ES%d.%d\n", omap_rev() >> 16,
-		((omap_rev() >> 12) & 0xf), ((omap_rev() >> 8) & 0xf));
+	sprintf(omap_cpu_model_name, "OMAP%04x ES%d.%d %s", omap_rev() >> 16,
+			((omap_rev() >> 12) & 0xf), ((omap_rev() >> 8) & 0xf),
+			omap_types[omap_type()]);
+	set_cpu_model_name(omap_cpu_model_name);
+
+	pr_info("%s\n", omap_cpu_model_name);
 }
 
 void __init omap5xxx_check_revision(void)
-- 
1.7.9.5

[RFC PATCH 0/2] ARM: update cpuinfo to print CPU model name

[Nishanth Menon]

On 17:54-20130129, Ruslan Bilovol wrote:
> Hi,
> 
> The following patches update cpuinfo to print CPU
> model name for ARM. First patch exactly makes needed
> changes for ARM architecture.
> Second patch adds this ability to OMAP4 SoCs.
> 
> This adds a common approach to show SoC name.
> 
> Looks like there were few attempts to do similar
> changes to cpuinfo without any luck.
> 
> So - comments are welcome
> 
> Ruslan Bilovol (2):
>   ARM: kernel: update cpuinfo to print CPU model name
>   ARM: OMAP4: setup CPU model name during ID initialisation

We had an discussion on similar lines but a generic suggestion:
https://patchwork.kernel.org/patch/98720/
SoCinfo framework which was supposed to introduce this

Would'nt that be a better approach to take than an OMAP only solution?

Ccying eduardo who had driven this for some time.
-- 
Regards,
Nishanth Menon

[RFC PATCH 0/2] ARM: update cpuinfo to print CPU model name

[Ruslan Bilovol]

Hi,

On Tue, Jan 29, 2013 at 6:02 PM, Nishanth Menon <nm@ti.com> wrote:
>
> On 17:54-20130129, Ruslan Bilovol wrote:
> > Hi,
> >
> > The following patches update cpuinfo to print CPU
> > model name for ARM. First patch exactly makes needed
> > changes for ARM architecture.
> > Second patch adds this ability to OMAP4 SoCs.
> >
> > This adds a common approach to show SoC name.
> >
> > Looks like there were few attempts to do similar
> > changes to cpuinfo without any luck.
> >
> > So - comments are welcome
> >
> > Ruslan Bilovol (2):
> >   ARM: kernel: update cpuinfo to print CPU model name
> >   ARM: OMAP4: setup CPU model name during ID initialisation
>
> We had an discussion on similar lines but a generic suggestion:
> https://patchwork.kernel.org/patch/98720/
> SoCinfo framework which was supposed to introduce this
>
> Would'nt that be a better approach to take than an OMAP only solution?

My goal is only to say what is the SoC model name in the /proc/cpuinfo
(And it is not an OMAP-only solution - it is common. Support for OMAP
is added in second patch)
This is the only type of information that we can apply for any SoC.
My point is - any SoC-specific information should go through some other
way - like SoCinfo framework mentioned by you.
And additional point - in cpuinfo we already have CPU name and Machine name.
The SoC name (that is something between these two things) looks also relevant

>
> Ccying eduardo who had driven this for some time.

Yes, let's have additional comments

> --
> Regards,
> Nishanth Menon
> --
> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> the body of a message to majordomo at vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

[RFC PATCH 0/2] ARM: update cpuinfo to print CPU model name

[Nishanth Menon]

On 01:08-20130130, Ruslan Bilovol wrote:
> Hi,
> 
> On Tue, Jan 29, 2013 at 6:02 PM, Nishanth Menon <nm@ti.com> wrote:
> >
> > On 17:54-20130129, Ruslan Bilovol wrote:
> > > Hi,
> > >
> > > The following patches update cpuinfo to print CPU
> > > model name for ARM. First patch exactly makes needed
> > > changes for ARM architecture.
> > > Second patch adds this ability to OMAP4 SoCs.
> > >
> > > This adds a common approach to show SoC name.
> > >
> > > Looks like there were few attempts to do similar
> > > changes to cpuinfo without any luck.
> > >
> > > So - comments are welcome
> > >
> > > Ruslan Bilovol (2):
> > >   ARM: kernel: update cpuinfo to print CPU model name
> > >   ARM: OMAP4: setup CPU model name during ID initialisation
> >
> > We had an discussion on similar lines but a generic suggestion:
> > https://patchwork.kernel.org/patch/98720/
> > SoCinfo framework which was supposed to introduce this
> >
> > Would'nt that be a better approach to take than an OMAP only solution?
> 
> My goal is only to say what is the SoC model name in the /proc/cpuinfo
> (And it is not an OMAP-only solution - it is common. Support for OMAP
> is added in second patch)
> This is the only type of information that we can apply for any SoC.
> My point is - any SoC-specific information should go through some other
> way - like SoCinfo framework mentioned by you.
> And additional point - in cpuinfo we already have CPU name and Machine name.
> The SoC name (that is something between these two things) looks also relevant
Looking at the sample output in patch #1/2
OMAP4470 ES1.0 HS

OMAP4470 is SoC name
ES1.0 is SoC revision
HS is SoC type

We even have SoC features (e.g. at boot NEON etc..)

Is the intent to put all this inside /proc/cpuinfo?? I bet every SoC has
it's own interesting data it'd like to have (like some of the info
populated by DT even). I am hardpressed to think this fits inside
/proc/cpuinfo. I might even suspect that the list of interesting
information might even vary per SoC.

having machine name is something ARM specific? dumping it on my linux
machines (x86[1] and sparc[2]) dont seem to show that.

but in the interest of not breaking existing interfaces, new interfaces
should potentially belong elsewhere.. my 2 cents.

[1] http://pastebin.com/CF8HPDAC (Xubuntu 12.04 3.2.0-36)
[2] http://pastebin.com/qNwWHwiu (Debian 3.2.35-2)
-- 
Regards,
Nishanth Menon

[RFC PATCH 0/2] ARM: update cpuinfo to print CPU model name

[Ruslan Bilovol]

On Wed, Jan 30, 2013 at 1:38 AM, Nishanth Menon <nm@ti.com> wrote:
> On 01:08-20130130, Ruslan Bilovol wrote:
>> Hi,
>>
>> On Tue, Jan 29, 2013 at 6:02 PM, Nishanth Menon <nm@ti.com> wrote:
>> >
>> > On 17:54-20130129, Ruslan Bilovol wrote:
>> > > Hi,
>> > >
>> > > The following patches update cpuinfo to print CPU
>> > > model name for ARM. First patch exactly makes needed
>> > > changes for ARM architecture.
>> > > Second patch adds this ability to OMAP4 SoCs.
>> > >
>> > > This adds a common approach to show SoC name.
>> > >
>> > > Looks like there were few attempts to do similar
>> > > changes to cpuinfo without any luck.
>> > >
>> > > So - comments are welcome
>> > >
>> > > Ruslan Bilovol (2):
>> > >   ARM: kernel: update cpuinfo to print CPU model name
>> > >   ARM: OMAP4: setup CPU model name during ID initialisation
>> >
>> > We had an discussion on similar lines but a generic suggestion:
>> > https://patchwork.kernel.org/patch/98720/
>> > SoCinfo framework which was supposed to introduce this
>> >
>> > Would'nt that be a better approach to take than an OMAP only solution?
>>
>> My goal is only to say what is the SoC model name in the /proc/cpuinfo
>> (And it is not an OMAP-only solution - it is common. Support for OMAP
>> is added in second patch)
>> This is the only type of information that we can apply for any SoC.
>> My point is - any SoC-specific information should go through some other
>> way - like SoCinfo framework mentioned by you.
>> And additional point - in cpuinfo we already have CPU name and Machine name.
>> The SoC name (that is something between these two things) looks also relevant
> Looking at the sample output in patch #1/2
> OMAP4470 ES1.0 HS
>
> OMAP4470 is SoC name
> ES1.0 is SoC revision
> HS is SoC type

Will fix it in next patchseries. Let's leave only only SoC name

>
> We even have SoC features (e.g. at boot NEON etc..)
>
> Is the intent to put all this inside /proc/cpuinfo?? I bet every SoC has
> it's own interesting data it'd like to have (like some of the info
> populated by DT even). I am hardpressed to think this fits inside
> /proc/cpuinfo. I might even suspect that the list of interesting
> information might even vary per SoC.

No, I don't think we need all these information in cpuinfo - just because
it will be different for different SoC - so the file structure will be
very different..
Let's have only SoC name in cpuinfo.

>
> having machine name is something ARM specific? dumping it on my linux
> machines (x86[1] and sparc[2]) dont seem to show that.

Yes - it's ARM specific.

>
> but in the interest of not breaking existing interfaces, new interfaces
> should potentially belong elsewhere.. my 2 cents.
>
> [1] http://pastebin.com/CF8HPDAC (Xubuntu 12.04 3.2.0-36)
> [2] http://pastebin.com/qNwWHwiu (Debian 3.2.35-2)

Just look at changes in cpuinfo for ARM between kernels:
3.4 kernel: http://pastebin.com/TfmE7b2b
3.8-rc3: http://pastebin.com/zvvTGeNt

The file structure has huge changes

> --
> Regards,
> Nishanth Menon

[RFC PATCH 0/2] ARM: update cpuinfo to print CPU model name

[Rob Herring]

On 01/29/2013 09:54 AM, Ruslan Bilovol wrote:
> Hi,
> 
> The following patches update cpuinfo to print CPU
> model name for ARM. First patch exactly makes needed
> changes for ARM architecture.
> Second patch adds this ability to OMAP4 SoCs.
> 
> This adds a common approach to show SoC name.
> 
> Looks like there were few attempts to do similar
> changes to cpuinfo without any luck.

Care to point out how your solution is different from those or addresses
the issues with them.

Assuming we want to do this, I think this information should come from
DeviceTree. OMAP may have this in a register, but others don't.

Rob

> 
> So - comments are welcome
> 
> Ruslan Bilovol (2):
>   ARM: kernel: update cpuinfo to print CPU model name
>   ARM: OMAP4: setup CPU model name during ID initialisation
> 
>  arch/arm/include/asm/setup.h |    1 +
>  arch/arm/kernel/setup.c      |   10 ++++++++++
>  arch/arm/mach-omap2/id.c     |   18 ++++++++++++++++--
>  3 files changed, 27 insertions(+), 2 deletions(-)
>