free_irq

From: Russell King (rmk@arm.linux.org.uk)
Date: Sun Sep 29 2002 - 08:15:50 EST


Hi,

The comments surrounding free_irq() include these line:

 * On a shared IRQ the caller must ensure the interrupt is disabled
 * on the card it drives before calling this function. The function
 * does not return until any executing interrupts for this IRQ
 * have completed.
 *
 * This function may be called from interrupt context.

This last line is quite bogus.

- if we call this function to free IRQ "n" from an interrupt handler
  running on IRQ "n", then we will deadlock; we call synchronize_irq(n)
  which will wait for us to exit.

- if we call this function to free IRQ "n" from an interrupt handler
  running on IRQ "b", and a handler for IRQ "n" is already running and
  IRQ "n" interrupted IRQ "n", then we will deadlock.

In light of the above, I'd suggest changing the last line to:

 * This function must not be called from interrupt context.

This patch is against 2.5.34:

--- orig/arch/i386/kernel/irq.c Fri Aug 30 14:52:51 2002
+++ linux/arch/i386/kernel/irq.c Sun Sep 29 14:13:39 2002
@@ -483,10 +483,7 @@
  * does not return until any executing interrupts for this IRQ
  * have completed.
  *
- * This function may be called from interrupt context.
- *
- * Bugs: Attempting to free an irq in a handler for the same irq hangs
- * the machine.
+ * This function must not be called from interrupt context.
  */
  
 void free_irq(unsigned int irq, void *dev_id)

-- 
Russell King (rmk@arm.linux.org.uk)                The developer of ARM Linux
             http://www.arm.linux.org.uk/personal/aboutme.html

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Mon Sep 30 2002 - 22:00:40 EST