From: tip-bot for Arnaldo Carvalho de Melo <tipbot@zytor.com>
To: linux-tip-commits@vger.kernel.org
Cc: tglx@linutronix.de, hpa@zytor.com, hekuang@huawei.com,
linux-kernel@vger.kernel.org, lizefan@huawei.com,
wangnan0@huawei.com, namhyung@kernel.org, jolsa@kernel.org,
mhiramat@kernel.org, mingo@kernel.org, acme@kernel.org,
acme@redhat.com
Subject: [tip:perf/core] perf evsel: Fix write_backwards fallback
Date: Wed, 22 Jun 2016 00:43:45 -0700 [thread overview]
Message-ID: <tip-7da36e94e7fad768ca8640b61ed1f49b284e1dc5@git.kernel.org> (raw)
In-Reply-To: <20160616214724.GI13337@kernel.org>
Commit-ID: 7da36e94e7fad768ca8640b61ed1f49b284e1dc5
Gitweb: http://git.kernel.org/tip/7da36e94e7fad768ca8640b61ed1f49b284e1dc5
Author: Arnaldo Carvalho de Melo <acme@kernel.org>
AuthorDate: Mon, 20 Jun 2016 10:47:18 +0000
Committer: Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Tue, 21 Jun 2016 13:18:35 -0300
perf evsel: Fix write_backwards fallback
Commit b90dc17a5d14 "perf evsel: Add overwrite attribute and check
write_backward" misunderstood the 'order' should be obeyed in
__perf_evsel__open.
But the way this was done for attr.write_backwards was buggy, as we need
to check features in the inverse order of their introduction to the
kernel, so that a newer tool checks first the newest perf_event_attr
fields, detecting that the older kernel doesn't have support for them.
Also, we can avoid calling sys_perf_event_open() if we have already
detected the missing of write_backward.
Cc: He Kuang <hekuang@huawei.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Zefan Li <lizefan@huawei.com>
Cc: pi3orama@163.com
Fixes: b90dc17a5d14 ("perf evsel: Add overwrite attribute and check write_backward")
Link: http://lkml.kernel.org/r/1466419645-75551-2-git-send-email-wangnan0@huawei.com
Link: http://lkml.kernel.org/r/20160616214724.GI13337@kernel.org
Signed-off-by: Wang Nan <wangnan0@huawei.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/evsel.c | 23 ++++++++++-------------
1 file changed, 10 insertions(+), 13 deletions(-)
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 9b2e3e6..1d8f2bb 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -1389,8 +1389,11 @@ fallback_missing_features:
if (perf_missing_features.lbr_flags)
evsel->attr.branch_sample_type &= ~(PERF_SAMPLE_BRANCH_NO_FLAGS |
PERF_SAMPLE_BRANCH_NO_CYCLES);
- if (perf_missing_features.write_backward)
+ if (perf_missing_features.write_backward) {
+ if (evsel->overwrite)
+ return -EINVAL;
evsel->attr.write_backward = false;
+ }
retry_sample_id:
if (perf_missing_features.sample_id_all)
evsel->attr.sample_id_all = 0;
@@ -1453,12 +1456,6 @@ retry_open:
err = -EINVAL;
goto out_close;
}
-
- if (evsel->overwrite &&
- perf_missing_features.write_backward) {
- err = -EINVAL;
- goto out_close;
- }
}
}
@@ -1496,7 +1493,10 @@ try_fallback:
* Must probe features in the order they were added to the
* perf_event_attr interface.
*/
- if (!perf_missing_features.clockid_wrong && evsel->attr.use_clockid) {
+ if (!perf_missing_features.write_backward && evsel->attr.write_backward) {
+ perf_missing_features.write_backward = true;
+ goto fallback_missing_features;
+ } else if (!perf_missing_features.clockid_wrong && evsel->attr.use_clockid) {
perf_missing_features.clockid_wrong = true;
goto fallback_missing_features;
} else if (!perf_missing_features.clockid && evsel->attr.use_clockid) {
@@ -1521,12 +1521,7 @@ try_fallback:
PERF_SAMPLE_BRANCH_NO_FLAGS))) {
perf_missing_features.lbr_flags = true;
goto fallback_missing_features;
- } else if (!perf_missing_features.write_backward &&
- evsel->attr.write_backward) {
- perf_missing_features.write_backward = true;
- goto fallback_missing_features;
}
-
out_close:
do {
while (--thread >= 0) {
@@ -2409,6 +2404,8 @@ int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
"We found oprofile daemon running, please stop it and try again.");
break;
case EINVAL:
+ if (evsel->overwrite && perf_missing_features.write_backward)
+ return scnprintf(msg, size, "Reading from overwrite event is not supported by this kernel.");
if (perf_missing_features.clockid)
return scnprintf(msg, size, "clockid feature not supported.");
if (perf_missing_features.clockid_wrong)
next prev parent reply other threads:[~2016-06-22 7:44 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-15 2:23 [PATCH v7 0/8] perf tools: Support overwritable ring buffer Wang Nan
2016-06-15 2:23 ` [PATCH v7 1/8] perf evlist: Introduce aux evlist Wang Nan
2016-06-15 2:23 ` [PATCH v7 2/8] perf tests: Add testcase for auxiliary evlist Wang Nan
2016-06-15 2:23 ` [PATCH v7 3/8] perf record: Introduce rec->overwrite_evlist for overwritable events Wang Nan
2016-06-15 2:23 ` [PATCH v7 4/8] perf record: Toggle overwrite ring buffer for reading Wang Nan
2016-06-15 2:23 ` [PATCH v7 5/8] perf tools: Enable overwrite settings Wang Nan
2016-06-15 2:23 ` [PATCH v7 6/8] perf tools: Don't warn about out of order event if write_backward is used Wang Nan
2016-06-15 2:23 ` [PATCH v7 7/8] perf tools: Check write_backward during evlist config Wang Nan
2016-06-16 21:47 ` Arnaldo Carvalho de Melo
2016-06-20 4:09 ` Wangnan (F)
2016-06-22 7:43 ` tip-bot for Arnaldo Carvalho de Melo [this message]
2016-06-15 2:23 ` [PATCH v7 8/8] perf record: Unmap overwrite evlist when event terminate Wang Nan
2016-06-16 20:59 ` Arnaldo Carvalho de Melo
2016-06-20 8:04 ` Wangnan (F)
2016-06-20 10:47 [PATCH v8 0/8] perf tools: Support overwritable ring buffer Wang Nan
2016-06-20 10:47 ` [PATCH v8 1/8] perf tools: Fix write_backwards fallback Wang Nan
2016-06-20 10:47 ` [PATCH v8 2/8] perf evlist: Introduce aux evlist Wang Nan
2016-06-20 20:36 ` Arnaldo Carvalho de Melo
2016-06-21 1:31 ` Wangnan (F)
2016-06-20 10:47 ` [PATCH v8 3/8] perf tests: Add testcase for auxiliary evlist Wang Nan
2016-06-21 21:05 ` Arnaldo Carvalho de Melo
2016-06-22 4:10 ` Wangnan (F)
2016-06-20 10:47 ` [PATCH v8 4/8] perf record: Introduce rec->overwrite_evlist for overwritable events Wang Nan
2016-06-21 21:30 ` Arnaldo Carvalho de Melo
2016-06-20 10:47 ` [PATCH v8 5/8] perf record: Toggle overwrite ring buffer for reading Wang Nan
2016-06-20 10:47 ` [PATCH v8 6/8] perf tools: Enable overwrite settings Wang Nan
2016-06-21 21:49 ` Arnaldo Carvalho de Melo
2016-06-20 10:47 ` [PATCH v8 7/8] perf tools: Don't warn about out of order event if write_backward is used Wang Nan
2016-06-20 10:47 ` [PATCH v8 8/8] perf tools: Add --tail-synthesize option Wang Nan
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=tip-7da36e94e7fad768ca8640b61ed1f49b284e1dc5@git.kernel.org \
--to=tipbot@zytor.com \
--cc=acme@kernel.org \
--cc=acme@redhat.com \
--cc=hekuang@huawei.com \
--cc=hpa@zytor.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tip-commits@vger.kernel.org \
--cc=lizefan@huawei.com \
--cc=mhiramat@kernel.org \
--cc=mingo@kernel.org \
--cc=namhyung@kernel.org \
--cc=tglx@linutronix.de \
--cc=wangnan0@huawei.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
Powered by JetHome