[PATCH 1/2] printk: export pr_flush()

[PATCH 1/2] printk: export pr_flush()

[Martin Hundebøll]

Printk users might want to assure whatever printed has reached its
destination before continuing. E.g. during the shutdown-procedure, where
printk-buffers aren't necessarily emptied before the system goes down.

Signed-off-by: Martin Hundebøll <martin@geanix.com>
---
 include/linux/printk.h | 5 +++++
 kernel/printk/printk.c | 4 +---
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/include/linux/printk.h b/include/linux/printk.h
index 8ef499ab3c1e..a2a33494c222 100644
--- a/include/linux/printk.h
+++ b/include/linux/printk.h
@@ -192,6 +192,7 @@ void show_regs_print_info(const char *log_lvl);
 extern asmlinkage void dump_stack_lvl(const char *log_lvl) __cold;
 extern asmlinkage void dump_stack(void) __cold;
 void printk_trigger_flush(void);
+bool pr_flush(int timeout_ms, bool reset_on_progress);
 #else
 static inline __printf(1, 0)
 int vprintk(const char *s, va_list args)
@@ -271,6 +272,10 @@ static inline void dump_stack(void)
 static inline void printk_trigger_flush(void)
 {
 }
+static inline bool pr_flush(int timeout_ms, bool reset_on_progress)
+{
+	return true;
+}
 #endif
 
 #ifdef CONFIG_SMP
diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
index 0b3af1529778..dc1d2c880eb0 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -2336,7 +2336,6 @@ asmlinkage __visible int _printk(const char *fmt, ...)
 }
 EXPORT_SYMBOL(_printk);
 
-static bool pr_flush(int timeout_ms, bool reset_on_progress);
 static bool __pr_flush(struct console *con, int timeout_ms, bool reset_on_progress);
 
 #else /* CONFIG_PRINTK */
@@ -2365,7 +2364,6 @@ static ssize_t msg_print_ext_body(char *buf, size_t size,
 static void console_lock_spinning_enable(void) { }
 static int console_lock_spinning_disable_and_check(int cookie) { return 0; }
 static bool suppress_message_printing(int level) { return false; }
-static bool pr_flush(int timeout_ms, bool reset_on_progress) { return true; }
 static bool __pr_flush(struct console *con, int timeout_ms, bool reset_on_progress) { return true; }
 
 #endif /* CONFIG_PRINTK */
@@ -3813,7 +3811,7 @@ static bool __pr_flush(struct console *con, int timeout_ms, bool reset_on_progre
  * Context: Process context. May sleep while acquiring console lock.
  * Return: true if all usable printers are caught up.
  */
-static bool pr_flush(int timeout_ms, bool reset_on_progress)
+bool pr_flush(int timeout_ms, bool reset_on_progress)
 {
 	return __pr_flush(NULL, timeout_ms, reset_on_progress);
 }
-- 
2.42.0

[PATCH 2/2] reboot: flush printk buffers before final shutdown

[Martin Hundebøll]

Make sure printed messages are in fact printed before putting the system
down.

Signed-off-by: Martin Hundebøll <martin@geanix.com>
---
 kernel/reboot.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/kernel/reboot.c b/kernel/reboot.c
index 3bba88c7ffc6..bab8350d5dae 100644
--- a/kernel/reboot.c
+++ b/kernel/reboot.c
@@ -273,6 +273,7 @@ void kernel_restart(char *cmd)
 	else
 		pr_emerg("Restarting system with command '%s'\n", cmd);
 	kmsg_dump(KMSG_DUMP_SHUTDOWN);
+	pr_flush(1000, false);
 	machine_restart(cmd);
 }
 EXPORT_SYMBOL_GPL(kernel_restart);
-- 
2.42.0

Re: [PATCH 2/2] reboot: flush printk buffers before final shutdown

[John Ogness]

On 2023-10-30, Martin Hundebøll <martin@geanix.com> wrote:
> Make sure printed messages are in fact printed before putting the system
> down.
>
> Signed-off-by: Martin Hundebøll <martin@geanix.com>
> ---
>  kernel/reboot.c | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/kernel/reboot.c b/kernel/reboot.c
> index 3bba88c7ffc6..bab8350d5dae 100644
> --- a/kernel/reboot.c
> +++ b/kernel/reboot.c
> @@ -273,6 +273,7 @@ void kernel_restart(char *cmd)
>  	else
>  		pr_emerg("Restarting system with command '%s'\n", cmd);
>  	kmsg_dump(KMSG_DUMP_SHUTDOWN);
> +	pr_flush(1000, false);

printk() tries to print directly from the calling context. Are you
experiencing problems where you do not see the restarting message?

John Ogness

Re: [PATCH 2/2] reboot: flush printk buffers before final shutdown

[John Ogness]

On 2023-10-31, Martin Hundebøll <martin@geanix.com> wrote:
>> printk() tries to print directly from the calling context. Are you
>> experiencing problems where you do not see the restarting message?
>
> Yes, but only with rt-patches. Sorry I forgot to mention that in the
> commit message. I considered sending the patches to the "rt-tree", but
> figured they don't need more patches to maintain, when the same
> changes could live in mainline just as well.

Thanks for clarifying. These patches really are only appropriate for the
rt-tree.

John Ogness

Re: [PATCH 1/2] printk: export pr_flush()

[John Ogness]

On 2023-10-30, Martin Hundebøll <martin@geanix.com> wrote:
> Printk users might want to assure whatever printed has reached its
> destination before continuing. E.g. during the shutdown-procedure, where
> printk-buffers aren't necessarily emptied before the system goes down.

This is reverting:

commit c60ba2d34608 ("printk: Make pr_flush() static")

I agree that it should be exported at some point, but we need to have
outside users.

John Ogness