From: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
To: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>,
Paul Mackerras <paulus@samba.org>,
Frederic Weisbecker <fweisbec@gmail.com>,
LKML <linux-kernel@vger.kernel.org>
Subject: [PATCH] perf tools: fix write_event()
Date: Wed, 20 Jan 2010 21:34:15 +0800 [thread overview]
Message-ID: <4B570657.5090105@cn.fujitsu.com> (raw)
We parse the event while it's read from mmap buffer in
write_event(), but sometimes the event will straddles the
mmap boundary, we should handle this case
And if we record events(such as perf kmem/sched) for long
times, Ctrl + C can't interrupt it just for this reason
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
---
tools/perf/builtin-record.c | 62 +++++++++++++++++++++++++++++++++---------
1 files changed, 48 insertions(+), 14 deletions(-)
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 7bb9ca1..97573b7 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -111,28 +111,62 @@ static void write_output(void *buf, size_t size)
}
}
-static void write_event(event_t *buf, size_t size)
+static void write_event(void *buf, size_t size)
{
- size_t processed_size = buf->header.size;
- event_t *ev = buf;
+ event_t *ev;
+ struct mmap_event mmap_ev;
+ static void *rec_buf;
+ static unsigned long rec_len;
- do {
- /*
- * Add it to the list of DSOs, so that when we finish this
- * record session we can pick the available build-ids.
- */
+ write_output(buf, size);
+
+ while (size > 0) {
+ if (rec_buf) {
+ int len;
+
+ memcpy(&mmap_ev, rec_buf, rec_len);
+ len = sizeof(mmap_ev.header) - rec_len;
+ if (len > 0) {
+ memcpy(&mmap_ev + rec_len, buf, len);
+ buf += len;
+ size -= len;
+ rec_len = sizeof((mmap_ev.header));
+ }
+
+ len = mmap_ev.header.size - rec_len;
+ if (mmap_ev.header.type == PERF_RECORD_MMAP)
+ memcpy(&mmap_ev + rec_len, buf, len);
+
+ buf += len;
+ size -= len;
+
+ rec_buf = NULL;
+ rec_len = 0;
+
+ ev = (event_t *)&mmap_ev;
+ goto handle_mmap;
+ }
+
+ ev = (event_t *)buf;
+ if (size < sizeof(struct perf_event_header) ||
+ ev->header.size > size) {
+ rec_buf = buf;
+ rec_len = size;
+ break;
+ }
+
+ buf += ev->header.size;
+ size -= ev->header.size;
+
+handle_mmap:
if (ev->header.type == PERF_RECORD_MMAP) {
struct list_head *head = &dsos__user;
if (ev->header.misc == 1)
head = &dsos__kernel;
__dsos__findnew(head, ev->mmap.filename);
}
+ }
- ev = ((void *)ev) + ev->header.size;
- processed_size += ev->header.size;
- } while (processed_size < size);
-
- write_output(buf, size);
}
static int process_synthesized_event(event_t *event,
--
1.6.1.2
next reply other threads:[~2010-01-20 13:36 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-20 13:34 Xiao Guangrong [this message]
2010-01-20 13:41 ` [PATCH] perf tools: fix write_event() Peter Zijlstra
2010-01-20 13:44 ` Peter Zijlstra
2010-01-20 14:09 ` Arnaldo Carvalho de Melo
2010-01-20 14:14 ` Peter Zijlstra
2010-01-20 14:26 ` Arnaldo Carvalho de Melo
2010-01-20 14:37 ` Peter Zijlstra
2010-01-20 15:43 ` Arnaldo Carvalho de Melo
2010-01-20 14:43 ` Peter Zijlstra
2010-01-20 15:30 ` Arnaldo Carvalho de Melo
2010-01-20 13:41 ` Frederic Weisbecker
2010-01-31 8:32 ` Ingo Molnar
2010-02-01 8:56 ` 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=4B570657.5090105@cn.fujitsu.com \
--to=xiaoguangrong@cn.fujitsu.com \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=paulus@samba.org \
--cc=peterz@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