|
From: Christian B. <bor...@de...> - 2011-11-08 09:02:09
|
Another thing that helped on my system for interlocked atomic updates was this patch:
===================================================================
--- coregrind/m_scheduler/scheduler.c (revision 12257)
+++ coregrind/m_scheduler/scheduler.c (working copy)
@@ -1161,8 +1161,8 @@
before swapping to another. That means that short term
spins waiting for hardware to poke memory won't cause a
thread swap. */
- if (VG_(dispatch_ctr) > 2000)
- VG_(dispatch_ctr) = 2000;
+ if (VG_(dispatch_ctr) > 20)
+ VG_(dispatch_ctr) = 20;
break;
case VG_TRC_INNER_COUNTERZERO:
Julian, do you know if the reason for the high retry value of 2000 is still valid?
|