From: Oleg Nesterov <oleg@redhat.com>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>,
Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
Anton Arapov <anton@redhat.com>,
Frederic Weisbecker <fweisbec@gmail.com>,
Ingo Molnar <mingo@redhat.com>,
linux-kernel@vger.kernel.org
Subject: [PATCH v2 4/7] uprobes/tracing: Make uprobe_{trace,perf}_print() uretprobe-friendly
Date: Tue, 9 Apr 2013 21:32:35 +0200 [thread overview]
Message-ID: <20130409193235.GA27244@redhat.com> (raw)
In-Reply-To: <20130409193200.GA26731@redhat.com>
Change uprobe_trace_print() and uprobe_perf_print() to check
is_ret_probe() and fill ring_buffer_event accordingly.
Also change uprobe_trace_func() and uprobe_perf_func() to not
_print() if is_ret_probe() is true. Note that we keep ->handler()
nontrivial even for uretprobe, we need this for filtering and for
other potential extensions.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Tested-by: Anton Arapov <anton@redhat.com>
---
kernel/trace/trace_uprobe.c | 34 +++++++++++++++++++++++++---------
1 files changed, 25 insertions(+), 9 deletions(-)
diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c
index 72aa45e..0ed99a2 100644
--- a/kernel/trace/trace_uprobe.c
+++ b/kernel/trace/trace_uprobe.c
@@ -516,15 +516,22 @@ static void uprobe_trace_print(struct trace_uprobe *tu,
int size, i;
struct ftrace_event_call *call = &tu->call;
- size = SIZEOF_TRACE_ENTRY(false) + tu->size;
+ size = SIZEOF_TRACE_ENTRY(is_ret_probe(tu));
event = trace_current_buffer_lock_reserve(&buffer, call->event.type,
- size, 0, 0);
+ size + tu->size, 0, 0);
if (!event)
return;
entry = ring_buffer_event_data(event);
- entry->vaddr[0] = instruction_pointer(regs);
- data = DATAOF_TRACE_ENTRY(entry, false);
+ if (is_ret_probe(tu)) {
+ entry->vaddr[0] = func;
+ entry->vaddr[1] = instruction_pointer(regs);
+ data = DATAOF_TRACE_ENTRY(entry, true);
+ } else {
+ entry->vaddr[0] = instruction_pointer(regs);
+ data = DATAOF_TRACE_ENTRY(entry, false);
+ }
+
for (i = 0; i < tu->nr_args; i++)
call_fetch(&tu->args[i].fetch, regs, data + tu->args[i].offset);
@@ -535,7 +542,8 @@ static void uprobe_trace_print(struct trace_uprobe *tu,
/* uprobe handler */
static int uprobe_trace_func(struct trace_uprobe *tu, struct pt_regs *regs)
{
- uprobe_trace_print(tu, 0, regs);
+ if (!is_ret_probe(tu))
+ uprobe_trace_print(tu, 0, regs);
return 0;
}
@@ -784,7 +792,7 @@ static void uprobe_perf_print(struct trace_uprobe *tu,
void *data;
int size, rctx, i;
- size = SIZEOF_TRACE_ENTRY(false);
+ size = SIZEOF_TRACE_ENTRY(is_ret_probe(tu));
size = ALIGN(size + tu->size + sizeof(u32), sizeof(u64)) - sizeof(u32);
if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, "profile buffer not large enough"))
return;
@@ -795,8 +803,15 @@ static void uprobe_perf_print(struct trace_uprobe *tu,
goto out;
ip = instruction_pointer(regs);
- entry->vaddr[0] = ip;
- data = DATAOF_TRACE_ENTRY(entry, false);
+ if (is_ret_probe(tu)) {
+ entry->vaddr[0] = func;
+ entry->vaddr[1] = ip;
+ data = DATAOF_TRACE_ENTRY(entry, true);
+ } else {
+ entry->vaddr[0] = ip;
+ data = DATAOF_TRACE_ENTRY(entry, false);
+ }
+
for (i = 0; i < tu->nr_args; i++)
call_fetch(&tu->args[i].fetch, regs, data + tu->args[i].offset);
@@ -812,7 +827,8 @@ static int uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs)
if (!uprobe_perf_filter(&tu->consumer, 0, current->mm))
return UPROBE_HANDLER_REMOVE;
- uprobe_perf_print(tu, 0, regs);
+ if (!is_ret_probe(tu))
+ uprobe_perf_print(tu, 0, regs);
return 0;
}
--
1.5.5.1
next prev parent reply other threads:[~2013-04-09 19:35 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-29 18:15 [PATCH 0/4] uprobes/tracing: uretprobes, initial preparations Oleg Nesterov
2013-03-29 18:15 ` [PATCH 1/4] uprobes/tracing: Kill the pointless task_pt_regs() calls Oleg Nesterov
2013-04-02 8:57 ` Anton Arapov
2013-04-04 14:24 ` Srikar Dronamraju
2013-03-29 18:15 ` [PATCH 2/4] uprobes/tracing: Kill the pointless seq_print_ip_sym() call Oleg Nesterov
2013-04-02 8:57 ` Anton Arapov
2013-04-04 14:24 ` Srikar Dronamraju
2013-03-29 18:15 ` [PATCH 3/4] uprobes/tracing: Kill the pointless local_save_flags/preempt_count calls Oleg Nesterov
2013-04-02 8:58 ` Anton Arapov
2013-04-04 14:25 ` Srikar Dronamraju
2013-04-05 3:47 ` Masami Hiramatsu
2013-04-05 15:01 ` Oleg Nesterov
2013-04-08 9:29 ` Masami Hiramatsu
2013-04-10 14:58 ` [PATCH 0/1] uprobes/tracing: Don't pass addr=ip to perf_trace_buf_submit() Oleg Nesterov
2013-04-10 14:58 ` [PATCH 1/1] " Oleg Nesterov
2013-04-11 10:19 ` Masami Hiramatsu
2013-04-13 9:28 ` Srikar Dronamraju
2013-04-11 10:38 ` [PATCH 0/1] " Masami Hiramatsu
2013-04-11 11:59 ` Oleg Nesterov
2013-04-12 18:01 ` Steven Rostedt
2013-04-12 21:19 ` Steven Rostedt
2013-04-13 14:02 ` [PATCH 0/1] uprobes/perf: Avoid perf_trace_buf_prepare/submit if ->perf_events is empty Oleg Nesterov
2013-04-13 14:02 ` [PATCH 1/1] " Oleg Nesterov
2013-04-13 18:22 ` [PATCH 0/1] " Oleg Nesterov
2013-04-08 15:58 ` [PATCH 3/4] uprobes/tracing: Kill the pointless local_save_flags/preempt_count calls Steven Rostedt
2013-04-09 14:58 ` Oleg Nesterov
2013-04-09 15:12 ` Steven Rostedt
2013-03-29 18:15 ` [PATCH 4/4] uprobes/tracing: generalize struct uprobe_trace_entry_head Oleg Nesterov
2013-04-02 8:59 ` Anton Arapov
2013-04-04 14:25 ` Srikar Dronamraju
2013-04-08 15:55 ` Steven Rostedt
2013-04-09 14:50 ` Oleg Nesterov
2013-04-09 15:07 ` Steven Rostedt
2013-04-09 19:32 ` [PATCH v2 0/7] uprobes/tracing: uretprobes Oleg Nesterov
2013-04-09 19:32 ` [PATCH v2 1/7] uprobes/tracing: Generalize struct uprobe_trace_entry_head Oleg Nesterov
2013-04-09 19:32 ` [PATCH v2 2/7] uprobes/tracing: Introduce uprobe_{trace,perf}_print() helpers Oleg Nesterov
2013-04-09 19:32 ` [PATCH v2 3/7] uprobes/tracing: Introduce is_ret_probe() and uretprobe_dispatcher() Oleg Nesterov
2013-04-09 19:32 ` Oleg Nesterov [this message]
2013-04-13 9:33 ` [PATCH v2 4/7] uprobes/tracing: Make uprobe_{trace,perf}_print() uretprobe-friendly Srikar Dronamraju
2013-04-09 19:32 ` [PATCH v2 5/7] uprobes/tracing: Make register_uprobe_event() paths uretprobe-friendly Oleg Nesterov
2013-04-09 19:32 ` [PATCH v2 6/7] uprobes/tracing: Make seq_printf() code uretprobe-friendly Oleg Nesterov
2013-04-09 19:32 ` [PATCH v2 7/7] uprobes/tracing: Change create_trace_uprobe() to support uretprobes Oleg Nesterov
2013-04-01 16:08 ` [PATCH 0/6] uprobes/tracing: uretprobes Oleg Nesterov
2013-04-01 16:08 ` [PATCH 1/6] uprobes/tracing: Introduce uprobe_{trace,perf}_print() helpers Oleg Nesterov
2013-04-07 13:58 ` Srikar Dronamraju
2013-04-01 16:08 ` [PATCH 2/6] uprobes/tracing: Introduce is_ret_probe() and uretprobe_dispatcher() Oleg Nesterov
2013-04-07 14:12 ` Srikar Dronamraju
2013-04-01 16:08 ` [PATCH 3/6] uprobes/tracing: Make uprobe_{trace,perf}_print() uretprobe-friendly Oleg Nesterov
2013-04-07 10:31 ` Srikar Dronamraju
2013-04-09 13:33 ` Oleg Nesterov
2013-04-13 9:31 ` Srikar Dronamraju
2013-04-08 17:08 ` Steven Rostedt
2013-04-01 16:08 ` [PATCH 4/6] uprobes/tracing: Make register_uprobe_event() paths uretprobe-friendly Oleg Nesterov
2013-04-07 14:14 ` Srikar Dronamraju
2013-04-01 16:08 ` [PATCH 5/6] uprobes/tracing: Make seq_printf() code uretprobe-friendly Oleg Nesterov
2013-04-07 14:15 ` Srikar Dronamraju
2013-04-01 16:09 ` [PATCH 6/6] uprobes/tracing: Change create_trace_uprobe() to support uretprobes Oleg Nesterov
2013-04-07 14:17 ` Srikar Dronamraju
2013-04-02 13:25 ` [PATCH 0/6] uprobes/tracing: uretprobes Anton Arapov
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=20130409193235.GA27244@redhat.com \
--to=oleg@redhat.com \
--cc=ananth@in.ibm.com \
--cc=anton@redhat.com \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=rostedt@goodmis.org \
--cc=srikar@linux.vnet.ibm.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).