Re: [2.1.80] ftape fdc-io.c broken

Krzysztof G. Baranowski (kgb@manjak.knm.org.pl)
Fri, 23 Jan 1998 13:43:10 +0100 (CET)

On Thu, 22 Jan 1998, Fire Dragon wrote:
>building ftape's fdc-io.c as a module or into the kernel brings out the
>following errors. I would have attached a patch to fix it, but I do not
>have the time at the moment to track this down yet. If config information
>is needed from the .config file, let me know exactly what else is needed.
Hope this helps. Compiled but untested.

Kris

-- 
Krzysztof G. Baranowski - President of the Harmless Manyacs' Club
"Smith & Wesson - The original point and click interface..."
http://www.knm.org.pl/                 <prezes@manjak.knm.org.pl>


--- linux-2.1.80/drivers/char/ftape/lowlevel/fdc-io.c-old	Fri Jan 23 13:34:52 1998
+++ linux/drivers/char/ftape/lowlevel/fdc-io.c	Fri Jan 23 13:39:00 1998
@@ -385,7 +385,7 @@
 int fdc_interrupt_wait(unsigned int time)
 {
 	struct wait_queue wait = {current, NULL};
-	int current_blocked = current->blocked;
+	sigset_t old_sigmask;	
 	static int resetting = 0;
 	TRACE_FUN(ft_t_fdc_dma);
 
@@ -401,14 +401,23 @@
 	/* timeout time will be up to USPT microseconds too long ! */
 	current->timeout = jiffies + (1000 * time + FT_USPT - 1) / FT_USPT;
 	current->state = TASK_INTERRUPTIBLE;
-	current->blocked = _BLOCK_ALL; /* isn't this already set by the
-					* high level routines?
-					*/
+
+	spin_lock_irq(&current->sigmask_lock);
+        old_sigmask = current->blocked;
+        siginitset(&current->blocked, _BLOCK_ALL);
+        recalc_sigpending(current);
+        spin_unlock_irq(&current->sigmask_lock);
+
 	add_wait_queue(&ftape_wait_intr, &wait);
 	while (!ft_interrupt_seen && current->state != TASK_RUNNING) {
 		schedule();	/* sets TASK_RUNNING on timeout */
         }
-	current->blocked = current_blocked;	/* restore */
+
+	spin_lock_irq(&current->sigmask_lock);
+        current->blocked = old_sigmask;
+        recalc_sigpending(current);
+        spin_unlock_irq(&current->sigmask_lock);
+	
 	remove_wait_queue(&ftape_wait_intr, &wait);
 	/*  the following IS necessary. True: as well
 	 *  wake_up_interruptible() as the schedule() set TASK_RUNNING