perf, x86: Implement initial P4 PMU driver
in [Kernel]
|
From: Robert Richter on 16 Mar 2010 12:10 Please see my patch below that fixes ther reporting error code that returns from x86_pmu.hw_config. -Robert On 11.03.10 18:33:55, tip-bot for Cyrill Gorcunov wrote: > Commit-ID: a072738e04f0eb26370e39ec679e9a0d65e49aea > Gitweb: http://git.kernel.org/tip/a072738e04f0eb26370e39ec679e9a0d65e49aea > Author: Cyrill Gorcunov <gorcunov(a)openvz.org> > AuthorDate: Thu, 11 Mar 2010 19:54:39 +0300 > Committer: Ingo Molnar <mingo(a)elte.hu> > CommitDate: Thu, 11 Mar 2010 18:51:08 +0100 > > perf, x86: Implement initial P4 PMU driver > > The netburst PMU is way different from the "architectural > perfomance monitoring" specification that current CPUs use. > P4 uses a tuple of ESCR+CCCR+COUNTER MSR registers to handle > perfomance monitoring events. -- From: Robert Richter <robert.richter(a)amd.com> Date: Tue, 16 Mar 2010 16:38:19 +0100 Subject: [PATCH] perf, x86: reporting error code that returns from x86_pmu.hw_config() If x86_pmu.hw_config() fails a fixed error code (-EOPNOTSUPP) is return even if a different error was reported. This patch fixes this. Signed-off-by: Robert Richter <robert.richter(a)amd.com> --- arch/x86/kernel/cpu/perf_event.c | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 4e2480f..8982d92 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -467,8 +467,9 @@ static int __hw_perf_event_init(struct perf_event *event) hwc->last_tag = ~0ULL; /* Processor specifics */ - if (x86_pmu.hw_config(attr, hwc)) - return -EOPNOTSUPP; + err = x86_pmu.hw_config(attr, hwc); + if (err) + return err; if (!hwc->sample_period) { hwc->sample_period = x86_pmu.max_period; -- 1.7.0 -- Advanced Micro Devices, Inc. Operating System Research Center email: robert.richter(a)amd.com -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Cyrill Gorcunov on 16 Mar 2010 12:30 On Tue, Mar 16, 2010 at 05:07:33PM +0100, Robert Richter wrote: [...] > -- > > From: Robert Richter <robert.richter(a)amd.com> > Date: Tue, 16 Mar 2010 16:38:19 +0100 > Subject: [PATCH] perf, x86: reporting error code that returns from x86_pmu.hw_config() > > If x86_pmu.hw_config() fails a fixed error code (-EOPNOTSUPP) is > return even if a different error was reported. This patch fixes this. > > Signed-off-by: Robert Richter <robert.richter(a)amd.com> > --- > arch/x86/kernel/cpu/perf_event.c | 5 +++-- > 1 files changed, 3 insertions(+), 2 deletions(-) > > diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c > index 4e2480f..8982d92 100644 > --- a/arch/x86/kernel/cpu/perf_event.c > +++ b/arch/x86/kernel/cpu/perf_event.c > @@ -467,8 +467,9 @@ static int __hw_perf_event_init(struct perf_event *event) > hwc->last_tag = ~0ULL; > > /* Processor specifics */ > - if (x86_pmu.hw_config(attr, hwc)) > - return -EOPNOTSUPP; > + err = x86_pmu.hw_config(attr, hwc); > + if (err) > + return err; > > if (!hwc->sample_period) { > hwc->sample_period = x86_pmu.max_period; > -- > 1.7.0 > Though at moment all hw_config callees return 0, it's better to be ready if one day we may start returning some particular errors. Looks good to me. Objections? -- Cyrill -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Lin Ming on 16 Mar 2010 21:30 On Wed, 2010-03-17 at 00:23 +0800, Cyrill Gorcunov wrote: > On Tue, Mar 16, 2010 at 05:07:33PM +0100, Robert Richter wrote: > [...] > > -- > > > > From: Robert Richter <robert.richter(a)amd.com> > > Date: Tue, 16 Mar 2010 16:38:19 +0100 > > Subject: [PATCH] perf, x86: reporting error code that returns from x86_pmu.hw_config() > > > > If x86_pmu.hw_config() fails a fixed error code (-EOPNOTSUPP) is > > return even if a different error was reported. This patch fixes this. > > > > Signed-off-by: Robert Richter <robert.richter(a)amd.com> > > --- > > arch/x86/kernel/cpu/perf_event.c | 5 +++-- > > 1 files changed, 3 insertions(+), 2 deletions(-) > > > > diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c > > index 4e2480f..8982d92 100644 > > --- a/arch/x86/kernel/cpu/perf_event.c > > +++ b/arch/x86/kernel/cpu/perf_event.c > > @@ -467,8 +467,9 @@ static int __hw_perf_event_init(struct perf_event *event) > > hwc->last_tag = ~0ULL; > > > > /* Processor specifics */ > > - if (x86_pmu.hw_config(attr, hwc)) > > - return -EOPNOTSUPP; > > + err = x86_pmu.hw_config(attr, hwc); > > + if (err) > > + return err; > > > > if (!hwc->sample_period) { > > hwc->sample_period = x86_pmu.max_period; > > -- > > 1.7.0 > > > > Though at moment all hw_config callees return 0, it's better > to be ready if one day we may start returning some particular > errors. Looks good to me. Objections? Looks good. Lin Ming > > -- Cyrill -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
|
Pages: 1 Prev: Failure with the 2.6.34-rc1 kernel Next: 2.6.34-rc1 sky2 mtu on second port defaults to 576 |