trace/hwlat: Remove printk from sampling loop
authorDaniel Bristot de Oliveira <bristot@redhat.com>
Tue, 22 Jun 2021 14:42:25 +0000 (16:42 +0200)
committerSteven Rostedt (VMware) <rostedt@goodmis.org>
Fri, 25 Jun 2021 22:26:12 +0000 (18:26 -0400)
hwlat has some time operation checks on the sample loop, and it is
currently using pr_err (printk) to report them. The problem is that
this can lead the system to an unresponsible state due to an overflow of
printk messages. This problem can be mitigated by writing the error
message to the trace buffer.

Remove the printk messages from the sampling loop, switching the to
messages in the trace buffer.

No functional change.

Link: https://lkml.kernel.org/r/9d77c34869748aa105e965c769d24642914eea3a.1624372313.git.bristot@redhat.com
Cc: Phil Auld <pauld@redhat.com>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Kate Carcia <kcarcia@redhat.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Alexandre Chartre <alexandre.chartre@oracle.com>
Cc: Clark Willaims <williams@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: linux-doc@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Daniel Bristot de Oliveira <bristot@redhat.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
kernel/trace/trace_hwlat.c

index 44f46bc..a625bfd 100644 (file)
@@ -182,6 +182,15 @@ void trace_hwlat_callback(bool enter)
                kdata->nmi_count++;
 }
 
+/*
+ * hwlat_err - report a hwlat error.
+ */
+#define hwlat_err(msg) ({                                                      \
+       struct trace_array *tr = hwlat_trace;                                   \
+                                                                               \
+       trace_array_printk_buf(tr->array_buffer.buffer, _THIS_IP_, msg);        \
+})
+
 /**
  * get_sample - sample the CPU TSC and look for likely hardware latencies
  *
@@ -225,7 +234,7 @@ static int get_sample(void)
                        outer_diff = time_to_us(time_sub(t1, last_t2));
                        /* This shouldn't happen */
                        if (outer_diff < 0) {
-                               pr_err(BANNER "time running backwards\n");
+                               hwlat_err(BANNER "time running backwards\n");
                                goto out;
                        }
                        if (outer_diff > outer_sample)
@@ -237,7 +246,7 @@ static int get_sample(void)
 
                /* Check for possible overflows */
                if (total < last_total) {
-                       pr_err("Time total overflowed\n");
+                       hwlat_err("Time total overflowed\n");
                        break;
                }
                last_total = total;
@@ -253,7 +262,7 @@ static int get_sample(void)
 
                /* This shouldn't happen */
                if (diff < 0) {
-                       pr_err(BANNER "time running backwards\n");
+                       hwlat_err(BANNER "time running backwards\n");
                        goto out;
                }