public inbox for linux-ia64@vger.kernel.org
 help / color / mirror / Atom feed
* [Linux-ia64] O(1) scheduler fix for 2.5-ia64
@ 2002-11-04 18:53 Erich Focht
  0 siblings, 0 replies; only message in thread
From: Erich Focht @ 2002-11-04 18:53 UTC (permalink / raw)
  To: linux-ia64

[-- Attachment #1: Type: text/plain, Size: 423 bytes --]

Hi David,

this problem is wellknown and I actually thought that it was
solved. But it looks like the complex prepare_arch_switch()
etc. macros didn't make it into the 2.5 kernels for IA64. They are
needed on IA64 because we have to release the runqueue lock during
context switch. This is similar to sparc64. Thanks to John Hawks for
noticing this.

Please include this into your kernel tree.

Regards,
Erich

[-- Attachment #2: O1_switch_fix-2.5.45.patch --]
[-- Type: text/x-diff, Size: 720 bytes --]

diff -urNp linux-2.5.45-ia64/include/asm-ia64/system.h linux-2.5.45-ia64-o1fix/include/asm-ia64/system.h
--- linux-2.5.45-ia64/include/asm-ia64/system.h	Mon Nov  4 19:31:52 2002
+++ linux-2.5.45-ia64-o1fix/include/asm-ia64/system.h	Mon Nov  4 19:46:35 2002
@@ -432,6 +432,18 @@ extern void ia64_load_extra (struct task
 } while (0)
 #endif
 
+#define prepare_arch_switch(rq, next)		\
+do {	spin_lock(&(next)->switch_lock);	\
+	spin_unlock(&(rq)->lock);		\
+} while (0)
+
+#define finish_arch_switch(rq, prev)		\
+do {	spin_unlock_irq(&(prev)->switch_lock);	\
+} while (0)
+
+#define task_running(rq, p) \
+	((rq)->curr == (p) || spin_is_locked(&(p)->switch_lock))
+
 #endif /* __KERNEL__ */
 
 #endif /* __ASSEMBLY__ */

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2002-11-04 18:53 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2002-11-04 18:53 [Linux-ia64] O(1) scheduler fix for 2.5-ia64 Erich Focht

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox