public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] Fix x86-64 properly with MSI & Suresh's change
@ 2004-09-27 10:41 Andi Kleen
  0 siblings, 0 replies; only message in thread
From: Andi Kleen @ 2004-09-27 10:41 UTC (permalink / raw)
  To: Mallick, Asit K, Siddha, Suresh B, tom.l.nguyen, linux-kernel


Together with Suresh's recent LH workaround: this patch makes x86-64 
compile again with MSI on.  i386 uses an CPU number, x86-64
an CPU mask for MSI_TARGET_CPUS and that didn't work very well.

I must admit I don't fully understand how MSI irq affinity
is supposed to work (why do you always redirect to the current CPU?),
but this matches i386 which is presumably the best tested MSI
platform.

Signed-off-by: Andi Kleen <ak@suse.de>

diff -u linux/include/asm-x86_64/msi.h-o linux/include/asm-x86_64/msi.h
--- linux/include/asm-x86_64/msi.h-o	2004-09-24 13:04:06.000000000 +0200
+++ linux/include/asm-x86_64/msi.h	2004-09-27 12:19:56.000000000 +0200
@@ -7,10 +7,11 @@
 #define ASM_MSI_H
 
 #include <asm/desc.h>
+#include <asm/smp.h>
 
 #define LAST_DEVICE_VECTOR		232
 #define MSI_DEST_MODE			MSI_LOGICAL_MODE
 #define MSI_TARGET_CPU_SHIFT		12
-#define MSI_TARGET_CPU			TARGET_CPUS
+#define MSI_TARGET_CPU			hard_smp_processor_id()
 
 #endif /* ASM_MSI_H */



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

only message in thread, other threads:[~2004-09-27 10:42 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-09-27 10:41 [PATCH] Fix x86-64 properly with MSI & Suresh's change Andi Kleen

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