From: Jiri Olsa <jolsa@kernel.org>
To: Alexei Starovoitov <ast@kernel.org>,
Andrii Nakryiko <andrii@kernel.org>, Hao Luo <haoluo@google.com>,
Andrew Morton <akpm@linux-foundation.org>,
Alexander Viro <viro@zeniv.linux.org.uk>,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@redhat.com>,
Arnaldo Carvalho de Melo <acme@kernel.org>,
Matthew Wilcox <willy@infradead.org>
Cc: bpf@vger.kernel.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-perf-users@vger.kernel.org, Martin KaFai Lau <kafai@fb.com>,
Song Liu <songliubraving@fb.com>, Yonghong Song <yhs@fb.com>,
John Fastabend <john.fastabend@gmail.com>,
KP Singh <kpsingh@chromium.org>,
Stanislav Fomichev <sdf@google.com>,
Daniel Borkmann <daniel@iogearbox.net>,
Namhyung Kim <namhyung@gmail.com>,
Dave Chinner <david@fromorbit.com>
Subject: [PATCHv3 bpf-next 2/9] perf: Use file object build id in perf_event_mmap_event
Date: Thu, 16 Mar 2023 18:01:42 +0100 [thread overview]
Message-ID: <20230316170149.4106586-3-jolsa@kernel.org> (raw)
In-Reply-To: <20230316170149.4106586-1-jolsa@kernel.org>
Use build id from file object when available for perf's MMAP2 event
build id data.
The file's f_build_id is available (for CONFIG_FILE_BUILD_ID option)
when the file is mmap-ed and before the perf's callback is executed,
so we can use it, instead of reading it again.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
---
kernel/events/core.c | 43 +++++++++++++++++++++++++++++++++++++++----
1 file changed, 39 insertions(+), 4 deletions(-)
diff --git a/kernel/events/core.c b/kernel/events/core.c
index f79fd8b87f75..3a5398dda6f6 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -8527,6 +8527,9 @@ struct perf_mmap_event {
u32 prot, flags;
u8 build_id[BUILD_ID_SIZE_MAX];
u32 build_id_size;
+#ifdef CONFIG_FILE_BUILD_ID
+ struct build_id *f_build_id;
+#endif
struct {
struct perf_event_header header;
@@ -8539,6 +8542,38 @@ struct perf_mmap_event {
} event_id;
};
+#ifdef CONFIG_FILE_BUILD_ID
+static void build_id_read(struct perf_mmap_event *mmap_event)
+{
+ struct vm_area_struct *vma = mmap_event->vma;
+
+ mmap_event->f_build_id = vma->vm_file ? vma->vm_file->f_build_id : NULL;
+}
+
+static bool has_build_id(struct perf_mmap_event *mmap_event)
+{
+ return !IS_ERR_OR_NULL(mmap_event->f_build_id);
+}
+
+#define build_id_data mmap_event->f_build_id->data
+#define build_id_size mmap_event->f_build_id->sz
+#else
+static void build_id_read(struct perf_mmap_event *mmap_event)
+{
+ struct vm_area_struct *vma = mmap_event->vma;
+
+ build_id_parse(vma, mmap_event->build_id, &mmap_event->build_id_size);
+}
+
+static bool has_build_id(struct perf_mmap_event *mmap_event)
+{
+ return mmap_event->build_id_size;
+}
+
+#define build_id_data mmap_event->build_id
+#define build_id_size mmap_event->build_id_size
+#endif
+
static int perf_event_mmap_match(struct perf_event *event,
void *data)
{
@@ -8583,7 +8618,7 @@ static void perf_event_mmap_output(struct perf_event *event,
mmap_event->event_id.pid = perf_event_pid(event, current);
mmap_event->event_id.tid = perf_event_tid(event, current);
- use_build_id = event->attr.build_id && mmap_event->build_id_size;
+ use_build_id = event->attr.build_id && has_build_id(mmap_event);
if (event->attr.mmap2 && use_build_id)
mmap_event->event_id.header.misc |= PERF_RECORD_MISC_MMAP_BUILD_ID;
@@ -8592,10 +8627,10 @@ static void perf_event_mmap_output(struct perf_event *event,
if (event->attr.mmap2) {
if (use_build_id) {
- u8 size[4] = { (u8) mmap_event->build_id_size, 0, 0, 0 };
+ u8 size[4] = { (u8) build_id_size, 0, 0, 0 };
__output_copy(&handle, size, 4);
- __output_copy(&handle, mmap_event->build_id, BUILD_ID_SIZE_MAX);
+ __output_copy(&handle, build_id_data, BUILD_ID_SIZE_MAX);
} else {
perf_output_put(&handle, mmap_event->maj);
perf_output_put(&handle, mmap_event->min);
@@ -8727,7 +8762,7 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
mmap_event->event_id.header.size = sizeof(mmap_event->event_id) + size;
if (atomic_read(&nr_build_id_events))
- build_id_parse(vma, mmap_event->build_id, &mmap_event->build_id_size);
+ build_id_read(mmap_event);
perf_iterate_sb(perf_event_mmap_output,
mmap_event,
--
2.39.2
next prev parent reply other threads:[~2023-03-16 17:02 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-16 17:01 [PATCHv3 bpf-next 0/9] mm/bpf/perf: Store build id in file object Jiri Olsa
2023-03-16 17:01 ` [PATCHv3 bpf-next 1/9] mm: " Jiri Olsa
2023-03-16 22:07 ` Andrii Nakryiko
2023-03-16 17:01 ` Jiri Olsa [this message]
2023-03-16 17:01 ` [PATCHv3 bpf-next 3/9] bpf: Use file object build id in stackmap Jiri Olsa
2023-03-16 22:07 ` Andrii Nakryiko
2023-03-16 17:01 ` [PATCHv3 bpf-next 4/9] bpf: Switch BUILD_ID_SIZE_MAX to enum Jiri Olsa
2023-03-16 22:07 ` Andrii Nakryiko
2023-03-16 17:01 ` [PATCHv3 bpf-next 5/9] selftests/bpf: Add read_buildid function Jiri Olsa
2023-03-16 22:23 ` Andrii Nakryiko
2023-03-30 22:05 ` Jiri Olsa
2023-03-16 17:01 ` [PATCHv3 bpf-next 6/9] selftests/bpf: Add err.h header Jiri Olsa
2023-03-16 22:24 ` Andrii Nakryiko
2023-03-16 17:01 ` [PATCHv3 bpf-next 7/9] selftests/bpf: Replace extract_build_id with read_build_id Jiri Olsa
2023-03-16 17:01 ` [PATCHv3 bpf-next 8/9] selftests/bpf: Add iter_task_vma_buildid test Jiri Olsa
2023-03-16 22:31 ` Andrii Nakryiko
2023-03-16 17:01 ` [PATCHv3 bpf-next 9/9] selftests/bpf: Add file_build_id test Jiri Olsa
2023-03-16 19:59 ` Daniel Borkmann
2023-03-16 22:36 ` Andrii Nakryiko
2023-03-16 17:34 ` [PATCHv3 bpf-next 0/9] mm/bpf/perf: Store build id in file object Matthew Wilcox
2023-03-16 17:50 ` Ian Rogers
2023-03-16 21:51 ` Andrii Nakryiko
2023-03-17 3:51 ` Matthew Wilcox
2023-03-17 16:33 ` Andrii Nakryiko
2023-03-17 21:14 ` Al Viro
2023-03-17 21:21 ` Al Viro
2023-03-18 6:08 ` Andrii Nakryiko
2023-03-18 8:34 ` Jiri Olsa
2023-03-18 8:33 ` Jiri Olsa
2023-03-18 15:16 ` Matthew Wilcox
2023-03-18 17:40 ` Jiri Olsa
2023-03-22 15:45 ` Arnaldo Carvalho de Melo
2023-03-31 18:19 ` Andrii Nakryiko
2023-03-31 18:36 ` Matthew Wilcox
2023-03-31 20:27 ` Andrii Nakryiko
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=20230316170149.4106586-3-jolsa@kernel.org \
--to=jolsa@kernel.org \
--cc=acme@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=david@fromorbit.com \
--cc=haoluo@google.com \
--cc=john.fastabend@gmail.com \
--cc=kafai@fb.com \
--cc=kpsingh@chromium.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=namhyung@gmail.com \
--cc=peterz@infradead.org \
--cc=sdf@google.com \
--cc=songliubraving@fb.com \
--cc=viro@zeniv.linux.org.uk \
--cc=willy@infradead.org \
--cc=yhs@fb.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;
as well as URLs for NNTP newsgroup(s).