All of lore.kernel.org
 help / color / mirror / Atom feed
From: Steven Rostedt <rostedt@goodmis.org>
To: linux-kernel@vger.kernel.org
Cc: Ingo Molnar <mingo@kernel.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Ingo Molnar <mingo@redhat.com>, Baik Song An <bsahn@etri.re.kr>,
	Hong Yeon Kim <kimhy@etri.re.kr>,
	Taeung Song <taeung@reallinux.co.kr>,
	linuxgeek@linuxgeek.io, stable@vger.kernel.org,
	Wonhyuk Yang <vvghjk1234@gmail.com>
Subject: [for-next][PATCH 07/23] tracing: Fix return value of trace_pid_write()
Date: Fri, 27 May 2022 22:50:35 -0400	[thread overview]
Message-ID: <20220528025249.937888768@goodmis.org> (raw)
In-Reply-To: 20220528025028.850906216@goodmis.org

From: Wonhyuk Yang <vvghjk1234@gmail.com>

Setting set_event_pid with trailing whitespace lead to endless write
system calls like below.

    $ strace echo "123 " > /sys/kernel/debug/tracing/set_event_pid
    execve("/usr/bin/echo", ["echo", "123 "], ...) = 0
    ...
    write(1, "123 \n", 5)                   = 4
    write(1, "\n", 1)                       = 0
    write(1, "\n", 1)                       = 0
    write(1, "\n", 1)                       = 0
    write(1, "\n", 1)                       = 0
    write(1, "\n", 1)                       = 0
    ....

This is because, the result of trace_get_user's are not returned when it
read at least one pid. To fix it, update read variable even if
parser->idx == 0.

The result of applied patch is below.

    $ strace echo "123 " > /sys/kernel/debug/tracing/set_event_pid
    execve("/usr/bin/echo", ["echo", "123 "], ...) = 0
    ...
    write(1, "123 \n", 5)                   = 5
    close(1)                                = 0

Link: https://lkml.kernel.org/r/20220503050546.288911-1-vvghjk1234@gmail.com

Cc: Ingo Molnar <mingo@redhat.com>
Cc: Baik Song An <bsahn@etri.re.kr>
Cc: Hong Yeon Kim <kimhy@etri.re.kr>
Cc: Taeung Song <taeung@reallinux.co.kr>
Cc: linuxgeek@linuxgeek.io
Cc: stable@vger.kernel.org
Fixes: 4909010788640 ("tracing: Add set_event_pid directory for future use")
Signed-off-by: Wonhyuk Yang <vvghjk1234@gmail.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
---
 kernel/trace/trace.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 498ae22d4ffa..4825883b2ffd 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -721,13 +721,16 @@ int trace_pid_write(struct trace_pid_list *filtered_pids,
 		pos = 0;
 
 		ret = trace_get_user(&parser, ubuf, cnt, &pos);
-		if (ret < 0 || !trace_parser_loaded(&parser))
+		if (ret < 0)
 			break;
 
 		read += ret;
 		ubuf += ret;
 		cnt -= ret;
 
+		if (!trace_parser_loaded(&parser))
+			break;
+
 		ret = -EINVAL;
 		if (kstrtoul(parser.buffer, 0, &val))
 			break;
@@ -753,7 +756,6 @@ int trace_pid_write(struct trace_pid_list *filtered_pids,
 	if (!nr_pids) {
 		/* Cleared the list of pids */
 		trace_pid_list_free(pid_list);
-		read = ret;
 		pid_list = NULL;
 	}
 
-- 
2.35.1

  parent reply	other threads:[~2022-05-28  2:52 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-28  2:50 [for-next][PATCH 00/23] tracing: Last minute fixes and updates for 5.19 Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 01/23] tracing: Have event format check not flag %p* on __get_dynamic_array() Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 02/23] tracing: Reset the function filter after completing trampoline/graph selftest Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 03/23] tracing: Use trace_create_file() to simplify creation of tracefs entries Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 04/23] ftrace: Deal with error return code of the ftrace_process_locs() function Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 05/23] tracing: Use strim() to remove whitespace instead of doing it manually Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 06/23] tracing: Fix potential double free in create_var_ref() Steven Rostedt
2022-05-28  2:50 ` Steven Rostedt [this message]
2022-05-28  2:50 ` [for-next][PATCH 09/23] tracing/timerlat: Notify IRQ new max latency only if stop tracing is set Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 10/23] tracing/timerlat: Print stacktrace in the IRQ handler if needed Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 11/23] tracing/timerlat: Do not wakeup the thread if the trace stops at the IRQ Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 12/23] tracing: Change "char *" string form to "char []" Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 13/23] tracing: Cleanup code by removing init "char *name" Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 14/23] ftrace: Remove return value of ftrace_arch_modify_*() Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 15/23] ftrace: Fix typo in comment Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 16/23] tracing: Initialize integer variable to prevent garbage return value Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 17/23] tracing: Disable kcov on trace_preemptirq.c Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 18/23] tracing: Fix comments of create_filter() Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 19/23] ftrace: Clean up hash direct_functions on register failures Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 20/23] x86,tracing: Remove unused headers Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 21/23] x86/traceponit: Fix comment about irq vector tracepoints Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 22/23] tracing: Fix comments for event_trigger_separate_filter() Steven Rostedt
2022-05-28  2:50 ` [for-next][PATCH 23/23] ftrace: Add FTRACE_MCOUNT_MAX_OFFSET to avoid adding weak function Steven Rostedt
2022-05-28  3:17 ` [for-next][PATCH 00/23] tracing: Last minute fixes and updates for 5.19 Steven Rostedt

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=20220528025249.937888768@goodmis.org \
    --to=rostedt@goodmis.org \
    --cc=akpm@linux-foundation.org \
    --cc=bsahn@etri.re.kr \
    --cc=kimhy@etri.re.kr \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxgeek@linuxgeek.io \
    --cc=mingo@kernel.org \
    --cc=mingo@redhat.com \
    --cc=stable@vger.kernel.org \
    --cc=taeung@reallinux.co.kr \
    --cc=vvghjk1234@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.