From: Adrian Hunter <adrian.hunter@intel.com>
To: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>, Namhyung Kim <namhyung@kernel.org>,
Ian Rogers <irogers@google.com>,
linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org
Subject: Re: [PATCH V3 0/1] perf tools: Allow config terms with breakpoints
Date: Tue, 6 Jun 2023 22:37:03 +0300 [thread overview]
Message-ID: <9a3776f9-9c9e-b23f-9f18-91b4b5b8364d@intel.com> (raw)
In-Reply-To: <ZH+Havh+xRIfwa4n@kernel.org>
On 6/06/23 22:22, Arnaldo Carvalho de Melo wrote:
> Em Tue, Jun 06, 2023 at 03:15:59PM -0300, Arnaldo Carvalho de Melo escreveu:
>> Em Tue, Jun 06, 2023 at 08:00:33AM +0300, Adrian Hunter escreveu:
>>> On 25/05/23 11:29, Adrian Hunter wrote:
>>>> Hi
>>>>
>>>> Here is a patch (V3) to the event parser for breakpoint events.
>>>> I am not that familiar with flex / bison, but it seemed to
>>>> need trailing context to stop the mem event colon and slash
>>>> delimiters from getting mixed up with delimiters for config
>>>> terms or event modifiers. Please look closely at that.
>>>>
>>>>
>>>> Changes in V3:
>>>>
>>>> Add Ian's Reviewed-by
>>>> Re-base
>>>
>>> Still applies. Any more comments?
>>
>> Tried it now, twice, once after removing the O= build dir:
>>
>> CC /tmp/build/perf-tools-next/tests/event-times.o
>> CC /tmp/build/perf-tools-next/tests/expr.o
>> BISON /tmp/build/perf-tools-next/util/parse-events-bison.c
>> util/parse-events.y:508.24-34: warning: unused value: $3 [-Wother]
>> 508 | PE_PREFIX_MEM PE_VALUE PE_BP_SLASH PE_VALUE PE_BP_COLON PE_MODIFIER_BP opt_event_config
>> | ^~~~~~~~~~~
>> util/parse-events.y:508.45-55: warning: unused value: $5 [-Wother]
>> 508 | PE_PREFIX_MEM PE_VALUE PE_BP_SLASH PE_VALUE PE_BP_COLON PE_MODIFIER_BP opt_event_config
>> | ^~~~~~~~~~~
>> util/parse-events.y:526.24-34: warning: unused value: $3 [-Wother]
>> 526 | PE_PREFIX_MEM PE_VALUE PE_BP_SLASH PE_VALUE opt_event_config
>> | ^~~~~~~~~~~
>> util/parse-events.y:543.24-34: warning: unused value: $3 [-Wother]
>> 543 | PE_PREFIX_MEM PE_VALUE PE_BP_COLON PE_MODIFIER_BP opt_event_config
>> | ^~~~~~~~~~~
>> CC /tmp/build/perf-tools-next/tests/backward-ring-buffer.o
>
> But it doesn't _break_ the build, just warns thiis when generating the
> .c file, the next build it notices it is already generated, no warnings.
The build script I was using has a flaw. It doesn't show warnings
unless the build fails, assuming warnings are errors. But these
are not.
The first hack that came to mind, and seems to work, is:
diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y
index bbfb8110947c..dd36be3832b6 100644
--- a/tools/perf/util/parse-events.y
+++ b/tools/perf/util/parse-events.y
@@ -510,6 +510,8 @@ PE_PREFIX_MEM PE_VALUE PE_BP_SLASH PE_VALUE PE_BP_COLON PE_MODIFIER_BP opt_event
struct list_head *list;
int err;
+ $3 = $3;
+ $5 = $5;
list = alloc_list();
ABORT_ON(!list);
err = parse_events_add_breakpoint(_parse_state, list,
@@ -528,6 +530,7 @@ PE_PREFIX_MEM PE_VALUE PE_BP_SLASH PE_VALUE opt_event_config
struct list_head *list;
int err;
+ $3 = $3;
list = alloc_list();
ABORT_ON(!list);
err = parse_events_add_breakpoint(_parse_state, list,
@@ -545,6 +548,7 @@ PE_PREFIX_MEM PE_VALUE PE_BP_COLON PE_MODIFIER_BP opt_event_config
struct list_head *list;
int err;
+ $3 = $3;
list = alloc_list();
ABORT_ON(!list);
err = parse_events_add_breakpoint(_parse_state, list,
>
> - Arnaldo
>
>>
>> ⬢[acme@toolbox perf-tools-next]$ cat /etc/redhat-release
>> Fedora release 36 (Thirty Six)
>>
>> ⬢[acme@toolbox perf-tools-next]$ rpm -q bison
>> bison-3.8.2-2.fc36.x86_64
>> ⬢[acme@toolbox perf-tools-next]$ rpm -q flex
>> flex-2.6.4-10.fc36.x86_64
>> ⬢[acme@toolbox perf-tools-next]$
>>
>>
>> - Arnaldo
>>
>>>>
>>>> Changes in V2:
>>>>
>>>> Add comments to tools/perf/util/parse-events.l
>>>> Add a test for 2 mem events back to back with config terms
>>>>
>>>>
>>>> Adrian Hunter (1):
>>>> perf tools: Allow config terms with breakpoints
>>>>
>>>> tools/perf/tests/parse-events.c | 157 ++++++++++++++++++++++++++++++++++++++++
>>>> tools/perf/util/parse-events.c | 23 +++++-
>>>> tools/perf/util/parse-events.h | 6 +-
>>>> tools/perf/util/parse-events.l | 23 +++++-
>>>> tools/perf/util/parse-events.y | 42 ++++++-----
>>>> 5 files changed, 224 insertions(+), 27 deletions(-)
>>>>
>>>>
>>>> Regards
>>>> Adrian
>>>
>>
>> --
>>
>> - Arnaldo
>
next prev parent reply other threads:[~2023-06-06 19:37 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-25 8:29 [PATCH V3 0/1] perf tools: Allow config terms with breakpoints Adrian Hunter
2023-05-25 8:29 ` [PATCH V3 1/1] " Adrian Hunter
2023-06-06 5:00 ` [PATCH V3 0/1] " Adrian Hunter
2023-06-06 18:15 ` Arnaldo Carvalho de Melo
2023-06-06 19:22 ` Arnaldo Carvalho de Melo
2023-06-06 19:37 ` Adrian Hunter [this message]
2023-06-07 5:33 ` Adrian Hunter
2023-06-07 13:48 ` Arnaldo Carvalho de Melo
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=9a3776f9-9c9e-b23f-9f18-91b4b5b8364d@intel.com \
--to=adrian.hunter@intel.com \
--cc=acme@kernel.org \
--cc=irogers@google.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=namhyung@kernel.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).