On Wed, 11 Jul 2018 17:35:10 +0200 Hans de Goede hdegoede@redhat.com wrote:
OK, so if we don't remove it, we should probably make it so that it can be used without triggering any WARN_ONs, which would require changing the existing WARN_CONSOLE_UNLOCKED() so that the calls from drivers/tty/vt/vt.c also do not trigger it ?
I guess one can just ignore the oopses when debugging, but debugging surely would be easier if there are just no oopses ?
What about adding this patch (untested, not even compiled), and then set it from the fb module.
-- Steve
diff --git a/include/linux/console.h b/include/linux/console.h index dfd6b0e97855..11cd4956a57f 100644 --- a/include/linux/console.h +++ b/include/linux/console.h @@ -200,8 +200,10 @@ void vcs_make_sysfs(int index); void vcs_remove_sysfs(int index);
/* Some debug stub to catch some of the obvious races in the VT code */ +extern bool ignore_console_lock_warning; #if 1 -#define WARN_CONSOLE_UNLOCKED() WARN_ON(!is_console_locked() && !oops_in_progress) +#define WARN_CONSOLE_UNLOCKED() \ + WARN_ON(!ignore_console_lock_warning && !is_console_locked() && !oops_in_progress) #else #define WARN_CONSOLE_UNLOCKED() #endif diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index 247808333ba4..fa15d7ddf0c4 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -66,6 +66,9 @@ int console_printk[4] = { CONSOLE_LOGLEVEL_DEFAULT, /* default_console_loglevel */ };
+bool ignore_console_lock_warning; +EXPORT_SYMBOL(ignore_console_lock_warning); + /* * Low level drivers may need that to know if they can schedule in * their unblank() callback or not. So let's export it.