Hi.
I have noticed that Cygwin's spinlock goes into heavy sleeping code
for each spin. It seems it would be a good idea to actually try to
spin a bit first. There is this 'pause' instruction which let's the
CPU make such busy loops be less busy. Here is a patch to do this.
--
VH
diff --git a/winsup/cygwin/thread.cc b/winsup/cygwin/thread.cc
index 01cfd5b..56e66f1 100644
--- a/winsup/cygwin/thread.cc
+++ b/winsup/cygwin/thread.cc
@@ -1917,6 +1917,19 @@ pthread_spinlock::lock ()
{
pthread_t self = ::pthread_self ();
int result = -1;
+ unsigned spins = 0;
+
+ /*
+ We want to spin using 'pause' instruction on multi-core system but we
+ want to avoid this on single-core systems.
+
+ The limit of 1000 spins is semi-arbitrary. Microsoft suggests (in their
+ InitializeCriticalSectionAndSpinCount documentation on MSDN) they are
+ using spin count limit 4000 for their heap manager critical
+ sections. Other source suggest spin count as small as 200 for fast path
+ of mutex locking.
+ */
+ unsigned const FAST_SPINS_LIMIT = wincap.cpu_count () != 1 ? 1000 : 0;
do
{
@@ -1925,8 +1938,13 @@ pthread_spinlock::lock ()
set_owner (self);
result = 0;
}
- else if (pthread::equal (owner, self))
+ else if (unlikely(pthread::equal (owner, self)))
result = EDEADLK;
+ else if (spins < FAST_SPINS_LIMIT)
+ {
+ ++spins;
+ __asm__ volatile ("pause":::);
+ }
else
{
/* Minimal timeout to minimize CPU usage while still spinning. */
