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>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20191231085822.yxhph6wcguejb7al@frank-laptop>
Date:   Tue, 31 Dec 2019 03:58:22 -0500
From:   "Frank A. Cancio Bello" <frank@...eralsoftwareinc.com>
To:     Steven Rostedt <rostedt@...dmis.org>,
        Ingo Molnar <mingo@...hat.com>, linux-kernel@...r.kernel.org
Cc:     joel@...lfernandes.org, saiprakash.ranjan@...eaurora.org,
        nachukannan@...il.com
Subject: [PATCH] tracing: Resets the trace buffer after a snapshot

Currently, when a snapshot is taken the trace_buffer and the
max_buffer are swapped. After this swap, the "new" trace_buffer is
not reset. This produces an odd behavior: after a snapshot is taken
the previous snapshot entries become available to the next reader of
the trace_buffer as far as the reading occurs before the buffer is
refilled with new entries by a writer.

This patch resets the trace buffer after a snapshot is taken.

Signed-off-by: Frank A. Cancio Bello <frank@...eralsoftwareinc.com>
---

The following commands illustrate this odd behavior:

# cd /sys/kernel/debug/tracing
# echo nop > current_tracer
# echo 1 > tracing_on
# echo m1 > trace_marker
# echo 1 > snapshot
# echo m2 > trace_marker
# echo 1 > snapshot
# cat trace
# tracer: nop
#
# entries-in-buffer/entries-written: 1/1   #P:2
#
#                              _-----=> irqs-off
#                             / _----=> need-resched
#                            | / _---=> hardirq/softirq
#                            || / _--=> preempt-depth
#                            ||| /     delay
#           TASK-PID   CPU#  ||||    TIMESTAMP  FUNCTION
#              | |       |   ||||       |         |
            bash-550   [000] ....    50.479755: tracing_mark_write: m1


 kernel/trace/trace.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index ddb7e7f5fe8d..58373b5ae0cf 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -6867,10 +6867,13 @@ tracing_snapshot_write(struct file *filp, const char __user *ubuf, size_t cnt,
 			break;
 		local_irq_disable();
 		/* Now, we're going to swap */
-		if (iter->cpu_file == RING_BUFFER_ALL_CPUS)
+		if (iter->cpu_file == RING_BUFFER_ALL_CPUS) {
 			update_max_tr(tr, current, smp_processor_id(), NULL);
-		else
+			tracing_reset_online_cpus(&tr->trace_buffer);
+		} else {
 			update_max_tr_single(tr, current, iter->cpu_file);
+			tracing_reset_cpu(&tr->trace_buffer, iter->cpu_file);
+		}
 		local_irq_enable();
 		break;
 	default:
-- 
2.17.1

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ