From: Beau Belgrave <beaub@linux.microsoft.com>
To: rostedt@goodmis.org, mhiramat@kernel.org
Cc: linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org,
beaub@linux.microsoft.com
Subject: [PATCH v7 05/13] user_events: Add basic perf and eBPF support
Date: Thu, 9 Dec 2021 14:32:02 -0800 [thread overview]
Message-ID: <20211209223210.1818-6-beaub@linux.microsoft.com> (raw)
In-Reply-To: <20211209223210.1818-1-beaub@linux.microsoft.com>
Adds support to write out user_event data to perf_probe/perf files as
well as to any attached eBPF program.
Signed-off-by: Beau Belgrave <beaub@linux.microsoft.com>
---
kernel/trace/trace_events_user.c | 63 ++++++++++++++++++++++++++++++++
1 file changed, 63 insertions(+)
diff --git a/kernel/trace/trace_events_user.c b/kernel/trace/trace_events_user.c
index ae3ceaaa9098..807db0af74fb 100644
--- a/kernel/trace/trace_events_user.c
+++ b/kernel/trace/trace_events_user.c
@@ -541,6 +541,50 @@ static void user_event_ftrace(struct user_event *user, void *data, u32 datalen,
trace_event_buffer_commit(&event_buffer);
}
+#ifdef CONFIG_PERF_EVENTS
+/*
+ * Writes the user supplied payload out to perf ring buffer or eBPF program.
+ */
+static void user_event_perf(struct user_event *user, void *data, u32 datalen,
+ void *tpdata)
+{
+ struct hlist_head *perf_head;
+
+ if (bpf_prog_array_valid(&user->call)) {
+ struct user_bpf_context context = {0};
+
+ context.data_len = datalen;
+ context.data_type = USER_BPF_DATA_KERNEL;
+ context.kdata = data;
+
+ trace_call_bpf(&user->call, &context);
+ }
+
+ perf_head = this_cpu_ptr(user->call.perf_events);
+
+ if (perf_head && !hlist_empty(perf_head)) {
+ struct trace_entry *perf_entry;
+ struct pt_regs *regs;
+ size_t size = sizeof(*perf_entry) + datalen;
+ int context;
+
+ perf_entry = perf_trace_buf_alloc(ALIGN(size, 8),
+ ®s, &context);
+
+ if (unlikely(!perf_entry))
+ return;
+
+ perf_fetch_caller_regs(regs);
+
+ memcpy(perf_entry + 1, data, datalen);
+
+ perf_trace_buf_submit(perf_entry, size, context,
+ user->call.event.type, 1, regs,
+ perf_head, NULL);
+ }
+}
+#endif
+
/*
* Update the register page that is shared between user processes.
*/
@@ -563,6 +607,10 @@ static void update_reg_page_for(struct user_event *user)
if (probe_func == user_event_ftrace)
status |= EVENT_STATUS_FTRACE;
+#ifdef CONFIG_PERF_EVENTS
+ else if (probe_func == user_event_perf)
+ status |= EVENT_STATUS_PERF;
+#endif
else
status |= EVENT_STATUS_OTHER;
} while ((++probe_func_ptr)->func);
@@ -604,7 +652,19 @@ static int user_event_reg(struct trace_event_call *call,
#ifdef CONFIG_PERF_EVENTS
case TRACE_REG_PERF_REGISTER:
+ ret = tracepoint_probe_register(call->tp,
+ call->class->perf_probe,
+ data);
+ if (!ret)
+ goto inc;
+ break;
+
case TRACE_REG_PERF_UNREGISTER:
+ tracepoint_probe_unregister(call->tp,
+ call->class->perf_probe,
+ data);
+ goto dec;
+
case TRACE_REG_PERF_OPEN:
case TRACE_REG_PERF_CLOSE:
case TRACE_REG_PERF_ADD:
@@ -862,6 +922,9 @@ static int user_event_parse(char *name, char *args, char *flags,
user->class.get_fields = user_event_get_fields;
user->class.reg = user_event_reg;
user->class.probe = user_event_ftrace;
+#ifdef CONFIG_PERF_EVENTS
+ user->class.perf_probe = user_event_perf;
+#endif
mutex_lock(&event_mutex);
ret = user_event_trace_register(user);
--
2.17.1
next prev parent reply other threads:[~2021-12-09 22:32 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-12-09 22:31 [PATCH v7 00/13] user_events: Enable user processes to create and write to trace events Beau Belgrave
2021-12-09 22:31 ` [PATCH v7 01/13] user_events: Add UABI header for user access to user_events Beau Belgrave
2021-12-10 13:30 ` Masami Hiramatsu
2021-12-10 17:29 ` Beau Belgrave
2021-12-09 22:31 ` [PATCH v7 02/13] user_events: Add minimal support for trace_event into ftrace Beau Belgrave
2021-12-10 10:43 ` Masami Hiramatsu
2021-12-10 17:43 ` Steven Rostedt
2021-12-13 0:09 ` Masami Hiramatsu
2021-12-13 16:48 ` Steven Rostedt
2021-12-10 18:03 ` Beau Belgrave
2021-12-13 4:24 ` Masami Hiramatsu
2021-12-13 17:58 ` Beau Belgrave
2021-12-09 22:32 ` [PATCH v7 03/13] user_events: Add print_fmt generation support for basic types Beau Belgrave
2021-12-10 2:50 ` Masami Hiramatsu
2021-12-09 22:32 ` [PATCH v7 04/13] user_events: Handle matching arguments from dyn_events Beau Belgrave
2021-12-09 22:32 ` Beau Belgrave [this message]
2021-12-09 22:32 ` [PATCH v7 06/13] user_events: Add self-test for ftrace integration Beau Belgrave
2021-12-09 22:32 ` [PATCH v7 07/13] user_events: Add self-test for dynamic_events integration Beau Belgrave
2021-12-09 22:32 ` [PATCH v7 08/13] user_events: Add self-test for perf_event integration Beau Belgrave
2021-12-09 22:32 ` [PATCH v7 09/13] user_events: Optimize writing events by only copying data once Beau Belgrave
2021-12-10 14:51 ` Masami Hiramatsu
2021-12-10 18:36 ` Beau Belgrave
2021-12-09 22:32 ` [PATCH v7 10/13] user_events: Add documentation file Beau Belgrave
2021-12-09 22:32 ` [PATCH v7 11/13] user_events: Add sample code for typical usage Beau Belgrave
2021-12-09 22:32 ` [PATCH v7 12/13] user_events: Validate user payloads for size and null termination Beau Belgrave
2021-12-10 14:46 ` Masami Hiramatsu
2021-12-09 22:32 ` [PATCH v7 13/13] user_events: Use __get_rel_str for relative string fields Beau Belgrave
2021-12-10 1:23 ` Masami Hiramatsu
2021-12-10 18:45 ` Beau Belgrave
2021-12-12 15:12 ` Masami Hiramatsu
2021-12-13 18:47 ` Beau Belgrave
2021-12-14 6:30 ` Masami Hiramatsu
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=20211209223210.1818-6-beaub@linux.microsoft.com \
--to=beaub@linux.microsoft.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-trace-devel@vger.kernel.org \
--cc=mhiramat@kernel.org \
--cc=rostedt@goodmis.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).