linux-kernel - [for-linus][PATCH] ftrace: Document that multiple function

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <20250114105252.2cc71e08@gandalf.local.home>
Date: Tue, 14 Jan 2025 10:52:52 -0500
From: Steven Rostedt <rostedt@...dmis.org>
To: LKML <linux-kernel@...r.kernel.org>
Cc: Masami Hiramatsu <mhiramat@...nel.org>, Mark Rutland
 <mark.rutland@....com>, Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
Subject: [for-linus][PATCH] ftrace: Document that multiple function_graph
 tracing may have different times


Another update for v6.13:

- Document multiple function graph having slightly different timings

  Now that function graph tracer infrastructure is separate, this also
  allows the function graph tracer to run in multiple instances (it wasn't
  able to do so before). If two instances ran the function graph tracer and
  traced the same functions, the timings for them will be slightly
  different because each does their own timings and collects the timestamps
  differently. Document this to not have people be confused by it.

  git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace.git
trace/fixes

Head SHA1: 94d529a3255ce65496e932173d7f13e801170597


Steven Rostedt (1):
      ftrace: Document that multiple function_graph tracing may have different times

----
 Documentation/trace/ftrace.rst | 6 ++++++
 1 file changed, 6 insertions(+)
---------------------------
commit 94d529a3255ce65496e932173d7f13e801170597
Author: Steven Rostedt <rostedt@...dmis.org>
Date:   Tue Jan 14 10:12:02 2025 -0500

    ftrace: Document that multiple function_graph tracing may have different times
    
    The function graph tracer now calculates the calltime internally and for
    each instance. If there are two instances that are running function graph
    tracer and are tracing the same functions, the timings of the length of
    those functions may be slightly different:
    
     # trace-cmd record -B foo -p function_graph -B bar -p function_graph sleep 5
     # trace-cmd report
    [..]
    bar:            sleep-981   [000] ...1.  1101.109027: funcgraph_entry:        0.764 us   |          mutex_unlock(); (ret=0xffff8abcc256c300)
    foo:            sleep-981   [000] ...1.  1101.109028: funcgraph_entry:        0.748 us   |          mutex_unlock(); (ret=0xffff8abcc256c300)
    bar:            sleep-981   [000] .....  1101.109029: funcgraph_exit:         2.456 us   |        } (ret=0xffff8abcc256c300)
    foo:            sleep-981   [000] .....  1101.109029: funcgraph_exit:         2.403 us   |        } (ret=0xffff8abcc256c300)
    bar:            sleep-981   [000] d..1.  1101.109031: funcgraph_entry:        0.844 us   |  fpregs_assert_state_consistent(); (ret=0x0)
    foo:            sleep-981   [000] d..1.  1101.109032: funcgraph_entry:        0.803 us   |  fpregs_assert_state_consistent(); (ret=0x0)
    
    Link: https://lore.kernel.org/all/20250114101806.b2778cb01f34f5be9d23ad98@kernel.org/
    
    Cc: Mark Rutland <mark.rutland@....com>
    Cc: Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
    Suggested-by: Masami Hiramatsu <mhiramat@...nel.org>
    Link: https://lore.kernel.org/20250114101202.02e7bc68@gandalf.local.home
    Signed-off-by: Steven Rostedt (Google) <rostedt@...dmis.org>

diff --git a/Documentation/trace/ftrace.rst b/Documentation/trace/ftrace.rst
index 272464bb7c60..2b74f96d09d5 100644
--- a/Documentation/trace/ftrace.rst
+++ b/Documentation/trace/ftrace.rst
@@ -810,6 +810,12 @@ Here is the list of current tracers that may be configured.
 	to draw a graph of function calls similar to C code
 	source.
 
+	Note that the function graph calculates the timings of when the
+	function starts and returns internally and for each instance. If
+	there are two instances that run function graph tracer and traces
+	the same functions, the length of the timings may be slightly off as
+	each read the timestamp separately and not at the same time.
+
   "blk"
 
 	The block tracer. The tracer used by the blktrace user