linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jiri Olsa <jolsa@redhat.com>
To: "Yan, Zheng" <zheng.z.yan@intel.com>
Cc: a.p.zijlstra@chello.nl, mingo@elte.hu, andi@firstfloor.org,
	eranian@google.com, ming.m.lin@intel.com,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 9/9] perf tool: Add pmu event alias support
Date: Thu, 3 May 2012 12:56:55 +0200	[thread overview]
Message-ID: <20120503105655.GB1671@m.brq.redhat.com> (raw)
In-Reply-To: <1335924440-11242-10-git-send-email-zheng.z.yan@intel.com>

On Wed, May 02, 2012 at 10:07:20AM +0800, Yan, Zheng wrote:
> From: "Yan, Zheng" <zheng.z.yan@intel.com>
> 
> The definition of pmu event alias is located at:
>   ${sysfs_mount}/bus/event_source/devices/${pmu}/events/
> 
> Each file in the 'events' directory defines a event alias. Its contents
> is like:
>   config=1,config1=2
> 
> Using pmu event alias, event could be now specified like:
>   uncore/CLOCKTICKS/
> 
> Signed-off-by: Zheng Yan <zheng.z.yan@intel.com>
> ---
>  tools/perf/util/parse-events.c |   24 ++++++++-
>  tools/perf/util/parse-events.y |    2 +-
>  tools/perf/util/pmu.c          |  117 ++++++++++++++++++++++++++++++++++++++++
>  tools/perf/util/pmu.h          |   10 +++-
>  4 files changed, 149 insertions(+), 4 deletions(-)
> 
> diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
> index c587ae8..764b2c31 100644
> --- a/tools/perf/util/parse-events.c
> +++ b/tools/perf/util/parse-events.c
> @@ -653,8 +653,12 @@ int parse_events_add_numeric(struct list_head *list, int *idx,
>  int parse_events_add_pmu(struct list_head *list, int *idx,
>  			 char *name, struct list_head *head_config)
>  {
> +	LIST_HEAD(event);
>  	struct perf_event_attr attr;
>  	struct perf_pmu *pmu;
> +	const char *config;
> +	char *str;
> +	int ret;
>  
>  	pmu = perf_pmu__find(name);
>  	if (!pmu)
> @@ -668,10 +672,26 @@ int parse_events_add_pmu(struct list_head *list, int *idx,
>  	 */
>  	config_attr(&attr, head_config, 0);
>  
> -	if (perf_pmu__config(pmu, &attr, head_config))
> +	ret = perf_pmu__config(pmu, &attr, head_config);
> +	if (!ret)
> +		return add_event(list, idx, &attr, (char *) "pmu");
> +
> +	config = perf_pmu__alias(pmu, head_config);
> +	if (!config)
>  		return -EINVAL;
hi,
could we have the interface with string only:
	config = perf_pmu__alias(pmu, alias);

and AFAICS check if there's only single term and it's string,
then check for alias


I've got an idea for another approach, that would not need reentrant
parser and might be more gentle to the sysfs file rule

- in sysfs you would have directory with aliases (now called 'events')
- each alias is sysfs dir, with file attrs:
	file name = term name, file value = term value
  eg.:
	events/
		CAS_COUNT_RD/
			# files:
			config  - value 1
			config1 - value 2
			mask	- value ...

- on init you read all aliases and load its terms
  so each alias is defined by list of terms
- in parse_events_add_pmu before you run perf_pmu__config,
  you check if any term matches any defined alias
  and replace that term with all the terms defined for the alias
- run perf_pmu__config with new set of terms..

this way it's also possible to add extra terms to existing alias
in command line if needed... might be handy

thoughts?
jirka

  reply	other threads:[~2012-05-03 10:57 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-05-02  2:07 [PATCH V3 0/9] perf: Intel uncore pmu counting support Yan, Zheng
2012-05-02  2:07 ` [PATCH 1/9] perf: Export perf_assign_events Yan, Zheng
2012-05-02  2:07 ` [PATCH 2/9] perf: Allow pmu to choose cpu on which to install event Yan, Zheng
2012-05-09  6:38   ` Anshuman Khandual
2012-05-10  1:09     ` Yan, Zheng
2012-05-10  3:41       ` Anshuman Khandual
2012-05-10 10:56         ` Peter Zijlstra
2012-05-02  2:07 ` [PATCH 3/9] perf: Introduce perf_pmu_migrate_context Yan, Zheng
2012-05-02  2:07 ` [PATCH 4/9] perf: Generic intel uncore support Yan, Zheng
2012-05-03 17:12   ` Peter Zijlstra
2012-05-04  7:33     ` Yan, Zheng
2012-05-04 17:57       ` Peter Zijlstra
2012-05-10  7:34     ` Yan, Zheng
2012-05-10 10:05       ` Peter Zijlstra
2012-05-11  1:54         ` Yan, Zheng
2012-05-03 21:49   ` Peter Zijlstra
2012-05-11  6:31   ` Anshuman Khandual
2012-05-11  6:41     ` Yan, Zheng
2012-05-02  2:07 ` [PATCH 5/9] perf: Add Nehalem and Sandy Bridge " Yan, Zheng
2012-05-03 21:04   ` Peter Zijlstra
2012-05-04  5:47     ` Yan, Zheng
2012-05-03 21:04   ` Peter Zijlstra
2012-05-02  2:07 ` [PATCH 6/9] perf: Generic pci uncore device support Yan, Zheng
2012-05-03 21:37   ` Peter Zijlstra
2012-05-03 21:39   ` Peter Zijlstra
2012-05-03 21:46   ` Peter Zijlstra
2012-05-04  6:07     ` Yan, Zheng
2012-05-02  2:07 ` [PATCH 7/9] perf: Add Sandy Bridge-EP uncore support Yan, Zheng
2012-05-03 21:12   ` Peter Zijlstra
2012-05-02  2:07 ` [PATCH 8/9] perf tool: Make the event parser reentrantable Yan, Zheng
2012-05-02  2:07 ` [PATCH 9/9] perf tool: Add pmu event alias support Yan, Zheng
2012-05-03 10:56   ` Jiri Olsa [this message]
2012-05-03 11:24     ` Peter Zijlstra
2012-05-03 20:05       ` Jiri Olsa
2012-05-04 12:32         ` Yan, Zheng
2012-05-07  8:34         ` Yan, Zheng
2012-05-10  9:52           ` Jiri Olsa
2012-05-07 17:14         ` Peter Zijlstra

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=20120503105655.GB1671@m.brq.redhat.com \
    --to=jolsa@redhat.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=andi@firstfloor.org \
    --cc=eranian@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ming.m.lin@intel.com \
    --cc=mingo@elte.hu \
    --cc=zheng.z.yan@intel.com \
    /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;
as well as URLs for NNTP newsgroup(s).