[Openmp-commits] [PATCH] Allow machine hierarchy expansion
Wilmarth, Terry L
terry.l.wilmarth at intel.com
Tue May 26 09:38:03 PDT 2015
So, yes, this allows the hierarchy to resize itself in the oversubscription case.
From: openmp-commits-bounces at cs.uiuc.edu [mailto:openmp-commits-bounces at cs.uiuc.edu] On Behalf Of Wilmarth, Terry L
Sent: Tuesday, May 26, 2015 11:37 AM
To: reviews+D9900+public+a0ab28293505a533 at reviews.llvm.org; Peyton, Jonathan L
Cc: llvm-commits at cs.uiuc.edu; openmp-commits at dcs-maillist2.engr.illinois.edu
Subject: Re: [Openmp-commits] [PATCH] Allow machine hierarchy expansion
Oversubscription. There's nothing to prevent the user from requesting an arbitrarily large number of threads.
From: hfinkel at anl.gov [mailto:hfinkel at anl.gov]
Sent: Saturday, May 23, 2015 7:57 PM
To: Peyton, Jonathan L; hfinkel at anl.gov
Cc: Wilmarth, Terry L; llvm-commits at cs.uiuc.edu; openmp-commits at dcs-maillist2.engr.illinois.edu
Subject: Re: [PATCH] Allow machine hierarchy expansion
I don't understand this patch; can you please explain more?
Is the idea that if, at runtime, someone adds more CPUs, the hierarchy will resize itself?
Comment at: runtime/src/kmp_affinity.cpp:320
@@ -319,3 +319,3 @@
etc. We don't want to get specific with nomenclature */
- static const kmp_uint32 maxLevels=7;
+ kmp_uint32 maxLevels;
Why are you making this dynamic? How could this end up needing more than 7?
Openmp-commits mailing list
Openmp-commits at dcs-maillist2.engr.illinois.edu
More information about the Openmp-commits