lkml.org 
[lkml]   [2016]   [Aug]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] perf/x86/intel/uncore: correct uncore num_counters

* kan.liang@intel.com <kan.liang@intel.com> wrote:

> From: Kan Liang <kan.liang@intel.com>
>
> Some uncore boxes' num_counters for Haswell server and Broadwell server
> are not correct. This patch make them consistent with the uncore
> document.
>
> Reported-by: Lukasz Odzioba <lukasz.odzioba@intel.com>
> Signed-off-by: Kan Liang <kan.liang@intel.com>
> ---
> arch/x86/events/intel/uncore_snbep.c | 10 +++++-----
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/arch/x86/events/intel/uncore_snbep.c b/arch/x86/events/intel/uncore_snbep.c
> index 824e540..8aee83b 100644
> --- a/arch/x86/events/intel/uncore_snbep.c
> +++ b/arch/x86/events/intel/uncore_snbep.c
> @@ -2626,7 +2626,7 @@ void hswep_uncore_cpu_init(void)
>
> static struct intel_uncore_type hswep_uncore_ha = {
> .name = "ha",
> - .num_counters = 5,
> + .num_counters = 4,
> .num_boxes = 2,
> .perf_ctr_bits = 48,
> SNBEP_UNCORE_PCI_COMMON_INIT(),
> @@ -2645,7 +2645,7 @@ static struct uncore_event_desc hswep_uncore_imc_events[] = {
>
> static struct intel_uncore_type hswep_uncore_imc = {
> .name = "imc",
> - .num_counters = 5,
> + .num_counters = 4,
> .num_boxes = 8,
> .perf_ctr_bits = 48,
> .fixed_ctr_bits = 48,
> @@ -2691,7 +2691,7 @@ static struct intel_uncore_type hswep_uncore_irp = {
>
> static struct intel_uncore_type hswep_uncore_qpi = {
> .name = "qpi",
> - .num_counters = 5,
> + .num_counters = 4,
> .num_boxes = 3,
> .perf_ctr_bits = 48,
> .perf_ctr = SNBEP_PCI_PMON_CTR0,
> @@ -2773,7 +2773,7 @@ static struct event_constraint hswep_uncore_r3qpi_constraints[] = {
>
> static struct intel_uncore_type hswep_uncore_r3qpi = {
> .name = "r3qpi",
> - .num_counters = 4,
> + .num_counters = 3,
> .num_boxes = 3,
> .perf_ctr_bits = 44,
> .constraints = hswep_uncore_r3qpi_constraints,
> @@ -2972,7 +2972,7 @@ static struct intel_uncore_type bdx_uncore_ha = {
>
> static struct intel_uncore_type bdx_uncore_imc = {
> .name = "imc",
> - .num_counters = 5,
> + .num_counters = 4,
> .num_boxes = 8,
> .perf_ctr_bits = 48,
> .fixed_ctr_bits = 48,

So this changelog really sucks: what was the effect of the bug?

Did we report bogus (or zero) counts for those non-existent counters - or did the
code actually crash in a visible way?

Thanks,

Ingo

\
 
 \ /
  Last update: 2016-08-10 22:41    [W:0.041 / U:0.740 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site