We noticed a drop in n/w performance due to the irq_desc being cacheline
aligned rather than internode aligned. We see 50% of expected performance
when two e1000 nics local to two different nodes have consecutive irq
descriptors allocated, due to false sharing.
Note that this patch does away with cacheline padding for the UP case, as it
does not seem useful for UP configurations.
Signed-off-by: Ravikiran Thirumalai <[email protected]>
Signed-off-by: Shai Fultheim <[email protected]>
Index: linux-2.6.21-rc5/include/linux/irq.h
===================================================================
--- linux-2.6.21-rc5.orig/include/linux/irq.h 2007-04-09 10:16:23.560848473 -0700
+++ linux-2.6.21-rc5/include/linux/irq.h 2007-04-09 10:16:45.401177929 -0700
@@ -175,7 +175,7 @@ struct irq_desc {
struct proc_dir_entry *dir;
#endif
const char *name;
-} ____cacheline_aligned;
+} ____cacheline_internodealigned_in_smp;
extern struct irq_desc irq_desc[NR_IRQS];
Index: linux-2.6.21-rc5/kernel/irq/handle.c
===================================================================
--- linux-2.6.21-rc5.orig/kernel/irq/handle.c 2007-02-04 10:44:54.000000000 -0800
+++ linux-2.6.21-rc5/kernel/irq/handle.c 2007-04-09 12:26:40.473326023 -0700
@@ -48,7 +48,7 @@ handle_bad_irq(unsigned int irq, struct
*
* Controller mappings for all interrupt sources:
*/
-struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned = {
+struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = {
[0 ... NR_IRQS-1] = {
.status = IRQ_DISABLED,
.chip = &no_irq_chip,
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]
