From: Robert Richter <robert.richter@amd.com>
To: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: Ingo Molnar <mingo@elte.hu>, Lin Ming <ming.m.lin@intel.com>,
"H. Peter Anvin" <hpa@zytor.com>,
Thomas Gleixner <tglx@linutronix.de>,
Peter Zijlstra <peterz@infradead.org>,
Arnaldo Carvalho de Melo <acme@redhat.com>,
Stephane Eranian <eranian@google.com>,
Frederic Weisbecker <fweisbec@gmail.com>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [RFC] x86,perf: Implement minimal P4 PMU driver v14
Date: Wed, 10 Mar 2010 20:29:28 +0100 [thread overview]
Message-ID: <20100310192928.GD7219@erda.amd.com> (raw)
In-Reply-To: <20100310183102.GC8070@lenovo>
On 10.03.10 21:31:02, Cyrill Gorcunov wrote:
> arch/x86/include/asm/perf_event.h | 2
> arch/x86/include/asm/perf_p4.h | 707 +++++++++++++++++++++++++++++++++
If so, it should be perf_event_p4.h.
> arch/x86/kernel/cpu/perf_event.c | 46 +-
> arch/x86/kernel/cpu/perf_event_amd.c | 2
> arch/x86/kernel/cpu/perf_event_intel.c | 15
> arch/x86/kernel/cpu/perf_event_p4.c | 612 ++++++++++++++++++++++++++++
> arch/x86/kernel/cpu/perf_event_p6.c | 2
> 7 files changed, 1363 insertions(+), 23 deletions(-)
> Index: linux-2.6.git/arch/x86/kernel/cpu/perf_event.c
> =====================================================================
> --- linux-2.6.git.orig/arch/x86/kernel/cpu/perf_event.c
> +++ linux-2.6.git/arch/x86/kernel/cpu/perf_event.c
> @@ -190,6 +190,8 @@ struct x86_pmu {
> void (*enable_all)(void);
> void (*enable)(struct perf_event *);
> void (*disable)(struct perf_event *);
> + int (*hw_config)(struct perf_event_attr *attr, struct hw_perf_event *hwc);
> + int (*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign);
I don't like this extension since it widened the interface without
additional use.
(*hw_config) could be instead implemented in (*event_map).
(*schedule_events) could be implemented by a special p4 handler for
(*enable) in struct pmu. Maybe there are other solutions for both
cases, but it should be possible by adoption of existing functions.
The current implementation of model specific functions is
sufficient. We have already the following:
* event initialization: x86_pmu.raw_event(), x86_pmu.event_map()
* event enable: event->pmu->enable(), x86_pmu.enable()
* event disable: event->pmu->disable(), x86_pmu.disable()
Maybe I miss something in the list above. The introduction of more
function pointers should be reduced to a minimum.
If the pmu differs heavily you even could return a different pmu for
such an event.
-Robert
> unsigned eventsel;
> unsigned perfctr;
> u64 (*event_map)(int);
> @@ -415,6 +417,25 @@ set_ext_hw_attr(struct hw_perf_event *hw
> return 0;
> }
--
Advanced Micro Devices, Inc.
Operating System Research Center
email: robert.richter@amd.com
next prev parent reply other threads:[~2010-03-10 19:29 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-10 18:31 [RFC] x86,perf: Implement minimal P4 PMU driver v14 Cyrill Gorcunov
2010-03-10 19:29 ` Robert Richter [this message]
2010-03-10 19:43 ` Cyrill Gorcunov
2010-03-11 2:32 ` Lin Ming
2010-03-11 4:12 ` Cyrill Gorcunov
2010-03-11 16:54 ` Cyrill Gorcunov
2010-03-11 18:16 ` Ingo Molnar
2010-03-11 18:29 ` Cyrill Gorcunov
2010-03-11 18:39 ` Ingo Molnar
2010-03-11 21:15 ` Cyrill Gorcunov
2010-03-11 21:24 ` Peter Zijlstra
2010-03-11 21:31 ` Cyrill Gorcunov
2010-03-11 21:38 ` Peter Zijlstra
2010-03-11 21:41 ` Cyrill Gorcunov
2010-03-11 21:50 ` Cyrill Gorcunov
2010-03-12 9:54 ` [tip:perf/x86] x86, perf: Fix NULL deref on not assigned x86_pmu tip-bot for Cyrill Gorcunov
2010-03-11 18:33 ` [tip:perf/x86] perf, x86: Implement initial P4 PMU driver tip-bot for Cyrill Gorcunov
2010-03-16 16:07 ` Robert Richter
2010-03-16 16:23 ` Cyrill Gorcunov
2010-03-17 1:05 ` Lin Ming
2010-03-17 9:48 ` [tip:perf/core] perf, x86: Report error code that returned from x86_pmu.hw_config() tip-bot for Robert Richter
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20100310192928.GD7219@erda.amd.com \
--to=robert.richter@amd.com \
--cc=acme@redhat.com \
--cc=eranian@google.com \
--cc=fweisbec@gmail.com \
--cc=gorcunov@openvz.org \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=ming.m.lin@intel.com \
--cc=mingo@elte.hu \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox