[PATCH] oprofile: add support for Intel processor model 30

[PATCH] oprofile: add support for Intel processor model 30

[Josh Hunt]

Newer Intel processors identifying themselves as model 30 are not recognized by
oprofile.

<cpuinfo snippet>
model           : 30
model name      : Intel(R) Xeon(R) CPU           X3470  @ 2.93GHz
</cpuinfo snippet>

Running oprofile on these machines gives the following:
+ opcontrol --init
+ opcontrol --list-events
oprofile: available events for CPU type "Intel Architectural Perfmon"

See Intel 64 and IA-32 Architectures Software Developer's Manual
Volume 3B (Document 253669) Chapter 18 for architectural perfmon events
This is a limited set of fallback events because oprofile doesn't know your CPU
CPU_CLK_UNHALTED: (counter: all)
        Clock cycles when not halted (min count: 6000)
INST_RETIRED: (counter: all)
        number of instructions retired (min count: 6000)
LLC_MISSES: (counter: all)
        Last level cache demand requests from this core that missed the LLC
(min count: 6000)
        Unit masks (default 0x41)
        ----------
        0x41: No unit mask
LLC_REFS: (counter: all)
        Last level cache demand requests from this core (min count: 6000)
        Unit masks (default 0x4f)
        ----------
        0x4f: No unit mask
BR_MISS_PRED_RETIRED: (counter: all)
        number of mispredicted branches retired (precise) (min count: 500)
+ opcontrol --shutdown

Tested using oprofile 0.9.6.

Signed-off-by: Josh Hunt <johunt@akamai.com>
---
 arch/x86/oprofile/nmi_int.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index b28d2f1..6f95251 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -657,6 +657,7 @@ static int __init ppro_init(char **cpu_type)
 		break;
 	case 0x2e:
 	case 26:
+	case 30:
 		spec = &op_arch_perfmon_spec;
 		*cpu_type = "i386/core_i7";
 		break;
-- 
1.7.0.4

Re: [PATCH] oprofile: add support for Intel processor model 30

[Andi Kleen]

> Tested using oprofile 0.9.6.
> 
> Signed-off-by: Josh Hunt <johunt@akamai.com>

Reviewed-by: Andi Kleen <ak@linux.intel.com>

This change is ok since it's a 45nm Nehalem Xeon.

-Andi

Re: [PATCH] oprofile: add support for Intel processor model 30

[Robert Richter]

On 04.08.10 20:27:05, Josh Hunt wrote:
> Newer Intel processors identifying themselves as model 30 are not recognized by
> oprofile.

> Tested using oprofile 0.9.6.
> 
> Signed-off-by: Josh Hunt <johunt@akamai.com>

Applied to:

 git://git.kernel.org/pub/scm/linux/kernel/git/rric/oprofile.git core

(I converted the model number to hex.)

Thanks Josh.

Ingo,

please (ff-only) pull this again for tip/perf/core. You already send
the pull request for Linus, but I think this patch will make it for
v2.6.36 too?

Thanks.

-Robert

> ---
>  arch/x86/oprofile/nmi_int.c |    1 +
>  1 files changed, 1 insertions(+), 0 deletions(-)
> 
> diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
> index b28d2f1..6f95251 100644
> --- a/arch/x86/oprofile/nmi_int.c
> +++ b/arch/x86/oprofile/nmi_int.c
> @@ -657,6 +657,7 @@ static int __init ppro_init(char **cpu_type)
>  		break;
>  	case 0x2e:
>  	case 26:
> +	case 30:
>  		spec = &op_arch_perfmon_spec;
>  		*cpu_type = "i386/core_i7";
>  		break;
> -- 
> 1.7.0.4
> 
> 

-- 
Advanced Micro Devices, Inc.
Operating System Research Center

Re: [PATCH] oprofile: add support for Intel processor model 30

[Ingo Molnar]

* Robert Richter <robert.richter@amd.com> wrote:

> On 04.08.10 20:27:05, Josh Hunt wrote:
> > Newer Intel processors identifying themselves as model 30 are not recognized by
> > oprofile.
> 
> > Tested using oprofile 0.9.6.
> > 
> > Signed-off-by: Josh Hunt <johunt@akamai.com>
> 
> Applied to:
> 
>  git://git.kernel.org/pub/scm/linux/kernel/git/rric/oprofile.git core
> 
> (I converted the model number to hex.)
> 
> Thanks Josh.
> 
> Ingo,
> 
> please (ff-only) pull this again for tip/perf/core. You already send
> the pull request for Linus, but I think this patch will make it for
> v2.6.36 too?

Yeah, we are still early in the merge window, and it's a very small patch in 
any case - i pulled it.

Thanks,

	Ingo