linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Corey Ashford <cjashfor@linux.vnet.ibm.com>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Jiri Olsa <jolsa@redhat.com>,
	acme@redhat.com, mingo@elte.hu, paulus@samba.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 5/9] perf: Adding sysfs group format attribute for pmu device
Date: Wed, 01 Feb 2012 16:33:08 -0800	[thread overview]
Message-ID: <4F29D9C4.6030401@linux.vnet.ibm.com> (raw)
In-Reply-To: <1328103384.2760.37.camel@laptop>

On 02/01/2012 05:36 AM, Peter Zijlstra wrote:
> On Tue, 2012-01-31 at 17:25 -0800, Corey Ashford wrote:
>>
>> One other comment that occurs to me is that it would be nice for perf to
>> know when a supplied value is out of range, or will have undefined
>> results.  For example, a field may be 8 bits wide, but not all 8-bit
>> values are legal.  For example, there may be 208 events, and the codes
>> may be somehwhat or even very sparsely encoded.  So, ideally, a config
>> field in sysfs might look like this:
>>
>> config1:0-7:0x0-0xd8,0xdb-0xe2,0xe4-0xe6
>>
>> This way perf could check for valid values before stuffing them into
>> registers, and give a good error message to the user.  If there is no
>> restriction field, it would be assumed all of the possible values are valid.
>>
>> I think the kernel code needs to check for bad values as well, because
>> people can bypass the restrictions exposed by sysfs and use the
>> perf_events API directly. 
> 
> Define bad? The only case where perf cares (from a kernel pov) is when
> it can make the hardware melt and similar things. Programming the PMU in
> a non-nonsensical but non-destructive way is fine, you get what you ask
> for.

The main definition of "bad" would be a known config that has no meaning
(or perhaps is improperly implemented on the chip).  This would just
give the user some extra help that he might have incorrectly configured
the PMU.  Perf could identify down to the specific field that may not be
set correctly.  You could add an override switch to perf (--force-config
etc.) that would allow the user to bypass the check if he really wanted
to use that config.

It's also conceivable that there would be known PMU modes, that let's
say, stores values into a memory buffer (like IBS, etc), that if you use
a particular config, it would overwhelm the memory bus and choke the
processor down.

That said, this feature would just be nice to have, and is only on my
wish list.

Thanks for considering comments,

- Corey


  reply	other threads:[~2012-02-02  0:33 UTC|newest]

Thread overview: 91+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-12-15 15:30 [RFC 0/3] perf tool: Add new event group management Jiri Olsa
2011-12-15 15:30 ` [PATCH 1/3] perf, tool: Add parser generator for events parsing Jiri Olsa
2011-12-16 14:02   ` Peter Zijlstra
2011-12-16 14:03     ` Peter Zijlstra
2011-12-20 10:31       ` Jiri Olsa
2011-12-20 10:47         ` Peter Zijlstra
2011-12-20 11:30           ` Peter Zijlstra
2011-12-20 11:39             ` Peter Zijlstra
2011-12-21 16:16               ` new syntax for perf event Jiri Olsa
2012-01-05  9:17                 ` Jiri Olsa
2012-01-05 14:10                 ` Peter Zijlstra
2012-01-09 15:28                   ` Jiri Olsa
2012-01-09 15:43                     ` Peter Zijlstra
2012-01-16 12:31                     ` [RFCv3 0/9] perf tool: parser generator for events parsing Jiri Olsa
2012-01-16 12:31                       ` [PATCH 1/9] perf, tool: Make perf_evlist__splice_list_tail global Jiri Olsa
2012-01-16 12:31                       ` [PATCH 2/9] perf, tool: Remove unused functions from debugfs object Jiri Olsa
2012-01-16 12:31                       ` [PATCH 3/9] perf, tool: Add sysfs mountpoint interface Jiri Olsa
2012-01-16 12:31                       ` [PATCH 4/9] perf, tool: Add bitmap_or function into bitmap object Jiri Olsa
2012-01-16 12:31                       ` [PATCH 5/9] perf: Add sysfs format attribute for pmu device Jiri Olsa
2012-01-23 15:13                         ` Eric W. Biederman
2012-01-23 15:33                           ` Jiri Olsa
2012-01-24 15:22                             ` Peter Zijlstra
2012-01-24 19:40                               ` Eric W. Biederman
2012-01-25  8:54                                 ` Jiri Olsa
2012-01-26 16:26                         ` Peter Zijlstra
2012-01-27 12:32                           ` Jiri Olsa
2012-01-16 12:31                       ` [PATCH 6/9] perf, tool: Add parser generator for events parsing Jiri Olsa
2012-01-24 16:02                         ` Peter Zijlstra
2012-01-25  8:42                           ` Jiri Olsa
2012-01-16 12:31                       ` [PATCH 7/9] perf, tool: Add config options support for event parsing Jiri Olsa
2012-01-16 12:31                       ` [PATCH 8/9] perf, tool: Add perf pmu object to access pmu format definition Jiri Olsa
2012-01-16 12:31                       ` [PATCH 9/9] perf, tool: Add support to specify pmu style event Jiri Olsa
2012-01-24 15:22                       ` [RFCv3 0/9] perf tool: parser generator for events parsing Peter Zijlstra
2012-01-24 16:26                       ` Peter Zijlstra
2012-01-25  0:53                         ` Greg KH
2012-01-25 10:49                           ` Peter Zijlstra
2012-01-25 14:37                             ` Jiri Olsa
2012-01-26 16:23                               ` Peter Zijlstra
2012-01-26 16:27                                 ` Greg KH
2012-01-25 17:01                             ` Greg KH
2012-01-27 14:34                       ` [PATCHv4 " Jiri Olsa
2012-01-27 14:34                         ` [PATCH 1/9] perf, tool: Make perf_evlist__splice_list_tail global Jiri Olsa
2012-02-07 19:31                           ` [tip:perf/core] perf evlist: Make splice_list_tail method public tip-bot for Jiri Olsa
2012-01-27 14:34                         ` [PATCH 2/9] perf, tool: Remove unused functions from debugfs object Jiri Olsa
2012-02-17  9:51                           ` [tip:perf/core] perf tools: " tip-bot for Jiri Olsa
2012-01-27 14:34                         ` [PATCH 3/9] perf, tool: Add sysfs mountpoint interface Jiri Olsa
2012-02-17  9:52                           ` [tip:perf/core] perf tools: " tip-bot for Jiri Olsa
2012-01-27 14:34                         ` [PATCH 4/9] perf, tool: Add bitmap_or function into bitmap object Jiri Olsa
2012-02-17  9:53                           ` [tip:perf/core] perf tools: " tip-bot for Jiri Olsa
2012-01-27 14:34                         ` [PATCH 5/9] perf: Adding sysfs group format attribute for pmu device Jiri Olsa
2012-01-27 21:08                           ` Corey Ashford
2012-01-27 21:19                             ` Peter Zijlstra
2012-02-01  0:47                               ` Corey Ashford
2012-01-30  9:52                             ` Jiri Olsa
2012-02-01  1:25                               ` Corey Ashford
2012-02-01 13:13                                 ` Jiri Olsa
2012-02-01 14:18                                   ` Peter Zijlstra
2012-02-01 14:31                                     ` Jiri Olsa
2012-02-01 14:40                                       ` Peter Zijlstra
2012-02-01 13:36                                 ` Peter Zijlstra
2012-02-02  0:33                                   ` Corey Ashford [this message]
2012-01-27 14:34                         ` [PATCH 6/9] perf, tool: Add parser generator for events parsing Jiri Olsa
2012-01-27 14:34                         ` [PATCH 7/9] perf, tool: Add config options support for event parsing Jiri Olsa
2012-01-27 14:34                         ` [PATCH 8/9] perf, tool: Add perf pmu object to access pmu format definition Jiri Olsa
2012-01-27 14:34                         ` [PATCH 9/9] perf, tool: Add support to specify pmu style event Jiri Olsa
2012-02-13 13:13                         ` [PATCHv4 0/9] perf tool: parser generator for events parsing Jiri Olsa
2012-02-14 16:28                         ` Peter Zijlstra
2012-02-14 16:43                           ` Peter Zijlstra
2012-02-14 20:20                             ` Peter Zijlstra
2012-02-14 20:57                               ` Peter Zijlstra
2012-02-14 21:03                                 ` Peter Zijlstra
2012-02-15  9:24                                   ` Jiri Olsa
2012-02-15 11:18                                     ` Peter Zijlstra
2012-02-15 13:32                                       ` Jiri Olsa
2012-02-15 13:39                                         ` Peter Zijlstra
2012-02-15  9:04                                 ` Jiri Olsa
2012-02-15 11:03                                   ` Peter Zijlstra
2011-12-22 19:32             ` [PATCH 1/3] perf, tool: Add " Vince Weaver
2011-12-19 14:37     ` Jiri Olsa
2011-12-20 10:29     ` [PATCHv2 0/2] perf tool: " Jiri Olsa
2011-12-20 10:29       ` [PATCHv2 1/2] perf, tool: Add " Jiri Olsa
2011-12-20 10:29       ` [PATCHv2 2/2] perf, tool: Add more automated tests for event parsing Jiri Olsa
2011-12-20 17:37   ` [PATCH 1/3] perf, tool: Add parser generator for events parsing Arnaldo Carvalho de Melo
2011-12-21  9:55     ` Jiri Olsa
2011-12-15 15:30 ` [PATCH 2/3] perf, tool: Add new event group management Jiri Olsa
2011-12-20 17:47   ` Arnaldo Carvalho de Melo
2011-12-20 21:20     ` Peter Zijlstra
2011-12-21 11:54       ` Arnaldo Carvalho de Melo
2011-12-15 15:30 ` [PATCH 3/3] perf, tool: Add more automated tests for event parsing Jiri Olsa
2011-12-20 17:38   ` Arnaldo Carvalho de Melo
2011-12-21  8:47   ` [tip:perf/core] perf test: " tip-bot for Jiri Olsa

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=4F29D9C4.6030401@linux.vnet.ibm.com \
    --to=cjashfor@linux.vnet.ibm.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=acme@redhat.com \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=paulus@samba.org \
    /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).