From: jolsa@redhat.com (Jiri Olsa)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH V2 3/6] perf tools: add infrastructure for PMU specific configuration
Date: Tue, 26 Jul 2016 22:41:39 +0200 [thread overview]
Message-ID: <20160726204139.GA13377@krava> (raw)
In-Reply-To: <CANLsYkwx+iOE6uwOVPtsHFMmNaH0=NjeDP8z0GZhHja2oibt8g@mail.gmail.com>
On Fri, Jul 22, 2016 at 12:24:48PM -0600, Mathieu Poirier wrote:
SNIP
> > diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y
> > --- a/tools/perf/util/parse-events.y
> > +++ b/tools/perf/util/parse-events.y
> > @@ -602,12 +602,12 @@ PE_NAME array '=' PE_VALUE
> > $$ = term;
> > }
> > |
> > -PE_DRV_CFG_TERM
> > +'@' PE_DRV_CFG_TERM
> > {
> > struct parse_events_term *term;
> >
> > ABORT_ON(parse_events_term__str(&term, PARSE_EVENTS__TERM_TYPE_DRV_CFG,
> > - $1, $1, &@1, NULL));
> > + $2, $2, &@2, NULL));
> > $$ = term;
> > }
> >
>
> I've been experimenting with the above solution and it is not yielding
> the results one might think at first glance.
>
> If we use the example: -e event/@cfg1/ ...
>
> First if we leave things exactly the way they are suggested in the
> code snippet flex doesn't know what do to with the '@' character and
> returns an error. To deal with that a new clause
>
> "@" { return '@'; }
>
> can be inserted in the config state. But that doesn't link '@' with
> 'cfg1', and 'cfg1' gets interpreted as a PE_NAME. Introducing a new
> state upon hitting '@' would get us around that but we are moving away
> from our initial goal of keeping things simple.
hum, then how about keeping the flex atoms simple like for the
other terms and do something like below.. untested ;-)
thanks,
jirka
---
diff --git a/tools/perf/util/parse-events.l b/tools/perf/util/parse-events.l
index 1f7e11a6c5b3..8ba228e1c150 100644
--- a/tools/perf/util/parse-events.l
+++ b/tools/perf/util/parse-events.l
@@ -53,16 +53,6 @@ static int str(yyscan_t scanner, int token)
return token;
}
-static int drv_str(yyscan_t scanner, int token)
-{
- YYSTYPE *yylval = parse_events_get_lval(scanner);
- char *text = parse_events_get_text(scanner);
-
- /* Strip off the '@' */
- yylval->str = strdup(text + 1);
- return token;
-}
-
#define REWIND(__alloc) \
do { \
YYSTYPE *__yylval = parse_events_get_lval(yyscanner); \
@@ -134,7 +124,6 @@ num_hex 0x[a-fA-F0-9]+
num_raw_hex [a-fA-F0-9]+
name [a-zA-Z_*?][a-zA-Z0-9_*?.]*
name_minus [a-zA-Z_*?][a-zA-Z0-9\-_*?.:]*
-drv_cfg_term [a-zA-Z0-9_\.]+(=[a-zA-Z0-9_*?\.:]+)?
/* If you add a modifier you need to update check_modifier() */
modifier_event [ukhpPGHSDI]+
modifier_bp [rwx]{1,3}
@@ -216,11 +205,11 @@ no-inherit { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_NOINHERIT); }
overwrite { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_OVERWRITE); }
no-overwrite { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_NOOVERWRITE); }
, { return ','; }
+"@" { return '@'; }
"/" { BEGIN(INITIAL); return '/'; }
{name_minus} { return str(yyscanner, PE_NAME); }
\[all\] { return PE_ARRAY_ALL; }
"[" { BEGIN(array); return '['; }
-@{drv_cfg_term} { return drv_str(yyscanner, PE_DRV_CFG_TERM); }
}
<mem>{
diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y
index 879115f93edc..7e03e93dabca 100644
--- a/tools/perf/util/parse-events.y
+++ b/tools/perf/util/parse-events.y
@@ -602,12 +602,12 @@ PE_NAME array '=' PE_VALUE
$$ = term;
}
|
-PE_DRV_CFG_TERM
+'@' PE_NAME '=' PE_NAME
{
struct parse_events_term *term;
ABORT_ON(parse_events_term__str(&term, PARSE_EVENTS__TERM_TYPE_DRV_CFG,
- $1, $1, &@1, NULL));
+ $2, $4, &@2, &@4));
$$ = term;
}
next prev parent reply other threads:[~2016-07-26 20:41 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-20 20:38 [PATCH V2 0/6] perf: Driver specific configuration for PMU Mathieu Poirier
2016-07-20 20:38 ` [PATCH V2 1/6] perf/core: Adding PMU driver specific configuration Mathieu Poirier
2016-07-20 20:38 ` [PATCH V2 2/6] perf: Passing struct perf_event to function setup_aux() Mathieu Poirier
2016-07-20 20:38 ` [PATCH V2 3/6] perf tools: add infrastructure for PMU specific configuration Mathieu Poirier
2016-07-21 7:47 ` Jiri Olsa
2016-07-21 7:47 ` Jiri Olsa
2016-07-21 14:44 ` Mathieu Poirier
2016-07-21 14:54 ` Jiri Olsa
2016-07-21 15:47 ` Mathieu Poirier
2016-07-22 18:24 ` Mathieu Poirier
2016-07-26 20:41 ` Jiri Olsa [this message]
2016-07-27 17:59 ` Mathieu Poirier
2016-07-27 19:26 ` Jiri Olsa
2016-07-28 16:15 ` Mathieu Poirier
2016-07-28 16:52 ` Jiri Olsa
2016-07-21 7:47 ` Jiri Olsa
2016-07-20 20:38 ` [PATCH V2 4/6] perf tools: pushing driver configuration down to the kernel Mathieu Poirier
2016-07-21 7:47 ` Jiri Olsa
2016-07-22 19:57 ` Mathieu Poirier
2016-07-20 20:38 ` [PATCH V2 5/6] coresight: adding sink parameter to function coresight_build_path() Mathieu Poirier
2016-07-21 10:49 ` Suzuki K Poulose
2016-07-21 15:05 ` Mathieu Poirier
2016-07-20 20:38 ` [PATCH V2 6/6] coresight: etm-perf: incorporating sink definition from cmd line Mathieu Poirier
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=20160726204139.GA13377@krava \
--to=jolsa@redhat.com \
--cc=linux-arm-kernel@lists.infradead.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).