[PATCH] drivers/char/console.c and reentering

[Petr V. <van...@vc...>]

Hi Alan
  as nobody answered to my mail from Monday, please apply
this patch to 2.4.0-test1. It fixes one codepath which was 
proven that there is missing lock. Patch was generated 
against 2.4.0-test1-ac7.

  I tried to verify all other callers of set_cursor and
I believe that others are safe - or at least immediate callers
of set_cursor are at least sometime called with console_lock
held. But there are some callers (set_selection, 
complete_change_console) for which I was not able to verify
whether they are called with lock held or not.

  I'm using this change since monday on my SMP system and
I do not see any bad effect - only one good that fbdev is
no more reentered.
				Thanks,
					Petr Vandrovec
					van...@vc...




diff -urdN linux/drivers/char/console.c linux/drivers/char/console.c
--- linux/drivers/char/console.c	Sun Apr  2 22:20:27 2000
+++ linux/drivers/char/console.c	Wed May 31 15:52:52 2000
@@ -2290,10 +2290,13 @@
 
 static void con_flush_chars(struct tty_struct *tty)
 {
+	unsigned long flags;
 	struct vt_struct *vt = (struct vt_struct *)tty->driver_data;
 
 	pm_access(pm_con);
+	spin_lock_irqsave(&console_lock, flags);
 	set_cursor(vt->vc_num);
+	spin_unlock_irqrestore(&console_lock, flags);
 }
 
 /*