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-prev] [thread-next>] [day] [month] [year] [list]
Date:   Sat, 28 May 2022 13:54:11 +0000
From:   Junwen Wu <wudaemon@....com>
To:     rostedt@...dmis.org, Junwen Wu <wudaemon@....com>
Cc:     frederic@...nel.org, linux-kernel@...r.kernel.org,
        mingo@...hat.com, tannerlove@...gle.com, tglx@...utronix.de
Subject: Re: [PATCH v1] softirq: Add tracepoint for tasklet_entry/exit

From: Steven Rostedt <rostedt@...dmis.org>

>> +
>> +	TP_printk("function=%ps", __entry->func)

>This needs an acked-by from Thomas, but regardless, the above two trace
>events are identical. You need to have a DECLARE_EVENT_CLASS() followed by
>two DEFINE_EVENT()s, otherwise you are wasting memory.

I have use DECLARE_EVENT_CLASS() define use tracepoint,Steve.
Thomas, can you give me acked-by,here is patch v2.


>From 68d2298372018ec90c8587d5421a0e1a3c9f879f Mon Sep 17 00:00:00 2001
From: Junwen Wu <wudaemon@....com>
Date: Sun, 8 May 2022 15:29:56 +0000
Subject: [PATCH v2] softirq: Add tracepoint for tasklet_entry/exit

Usually softirq handler is pre-defined,only tasklet can be register by
driver.We expand tracepoint for tasklet_entry/exit to trace
tasklet handler.

Signed-off-by: Junwen Wu <wudaemon@....com>
---
 include/trace/events/irq.h | 41 ++++++++++++++++++++++++++++++++++++++
 kernel/softirq.c           |  4 ++++
 2 files changed, 45 insertions(+)

diff --git a/include/trace/events/irq.h b/include/trace/events/irq.h
index eeceafaaea4c..62b7e53121da 100644
--- a/include/trace/events/irq.h
+++ b/include/trace/events/irq.h
@@ -160,6 +160,47 @@ DEFINE_EVENT(softirq, softirq_raise,
        TP_ARGS(vec_nr)
 );

+DECLARE_EVENT_CLASS(tasklet,
+
+       TP_PROTO(void *func),
+
+       TP_ARGS(func),
+
+       TP_STRUCT__entry(
+               __field(        void*,     func )
+       ),
+
+       TP_fast_assign(
+               __entry->func = func;
+       ),
+
+       TP_printk("function=%ps", __entry->func)
+);
+
+/**
+ * tasklet_entry - called immediately when a tasklet exec
+ * @func: the callback function when a tasklet exec
+ *
+ * When used in combination with the tasklet_entry tracepoint
+ * we can determine the callback function when a tasklet exec.
+ */
+DEFINE_EVENT(tasklet, tasklet_entry,
+
+       TP_PROTO(void *func),
+
+       TP_ARGS(func)
+);
+
+/**
+ * tasklet_exit - called immediately after the tasklet handler returns
+ * @func: the callback function when a tasklet exec
+ */
+DEFINE_EVENT(tasklet, tasklet_exit,
+
+       TP_PROTO(void *func),
+
+       TP_ARGS(func)
+);
 #endif /*  _TRACE_IRQ_H */

 /* This part must be outside protection */
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 41f470929e99..79b3814ad166 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -780,10 +780,14 @@ static void tasklet_action_common(struct softirq_action *a,
                if (tasklet_trylock(t)) {
                        if (!atomic_read(&t->count)) {
                                if (tasklet_clear_sched(t)) {
+                                       trace_tasklet_entry(t->use_callback ? (void *)t->callback
+                                                       : (void *)t->func);
                                        if (t->use_callback)
                                                t->callback(t);
                                        else
                                                t->func(t->data);
+                                       trace_tasklet_exit(t->use_callback ? (void *)t->callback
+                                                       : (void *)t->func);
                                }
                                tasklet_unlock(t);
                                continue;
--
2.25.1

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ