Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(235)

Unified Diff: third_party/tcmalloc/vendor/src/base/spinlock_linux-inl.h

Issue 9701040: Revert 126715 - Update the tcmalloc vendor branch to r144 (gperftools 2.0). (Closed) Base URL: svn://svn.chromium.org/chrome/trunk/src/
Patch Set: Created 8 years, 9 months ago
Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.
Jump to:
View side-by-side diff with in-line comments
Download patch
Index: third_party/tcmalloc/vendor/src/base/spinlock_linux-inl.h
===================================================================
--- third_party/tcmalloc/vendor/src/base/spinlock_linux-inl.h (revision 126727)
+++ third_party/tcmalloc/vendor/src/base/spinlock_linux-inl.h (working copy)
@@ -31,7 +31,6 @@
* This file is a Linux-specific part of spinlock_internal.cc
*/
-#include <errno.h>
#include <sched.h>
#include <time.h>
#include <limits.h>
@@ -76,12 +75,12 @@
struct timespec tm;
tm.tv_sec = 0;
if (have_futex) {
- tm.tv_nsec = base::internal::SuggestedDelayNS(loop);
+ tm.tv_nsec = 1000000; // 1ms; really we're trying to sleep for one
+ // kernel clock tick
} else {
tm.tv_nsec = 2000001; // above 2ms so linux 2.4 doesn't spin
}
if (have_futex) {
- tm.tv_nsec *= 16; // increase the delay; we expect explicit wakeups
sys_futex(reinterpret_cast<int *>(const_cast<Atomic32 *>(w)),
FUTEX_WAIT | futex_private_flag,
value, reinterpret_cast<struct kernel_timespec *>(&tm));
« no previous file with comments | « third_party/tcmalloc/vendor/src/base/spinlock_internal.cc ('k') | third_party/tcmalloc/vendor/src/base/spinlock_posix-inl.h » ('j') | no next file with comments »

Powered by Google App Engine
This is Rietveld 408576698