[pypy-commit] stmgc default: Do we really need this?

arigo noreply at buildbot.pypy.org
Wed Feb 26 23:51:20 CET 2014


Author: Armin Rigo <arigo at tunes.org>
Branch: 
Changeset: r884:72584e5eb7df
Date: 2014-02-26 23:27 +0100
http://bitbucket.org/pypy/stmgc/changeset/72584e5eb7df/

Log:	Do we really need this?

diff --git a/c7/stm/contention.c b/c7/stm/contention.c
--- a/c7/stm/contention.c
+++ b/c7/stm/contention.c
@@ -77,6 +77,8 @@
 
         /* we will issue a safe point and wait: */
         STM_PSEGMENT->safe_point = SP_SAFE_POINT_CANNOT_COLLECT;
+        // XXX do we really need a safe_point here?  It seems we can
+        // kill it and the whole SP_SAFE_POINT_CANNOT_COLLECT
 
         /* wait, hopefully until the other thread broadcasts "I'm
            done aborting" (spurious wake-ups are ok). */


More information about the pypy-commit mailing list