projects / p5sagit/p5-mst-13.2.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: c2f6176)
As best as I (and my minion, gcc -Os) can tell, PL_curcop and
authorNicholas Clark <nick@ccl4.org>
Fri, 22 Feb 2008 22:30:05 +0000 (22:30 +0000)
committerNicholas Clark <nick@ccl4.org>
Fri, 22 Feb 2008 22:30:05 +0000 (22:30 +0000)
commit910e06714ae57f4d3d804265c3d148726c7a8a2b
tree4b7a38ad29de464337a018aefc58c049c2040c9ctree | snapshot
parentc2f617693870df2d42e43b7a3ccc75d8f071f4bacommit | diff
As best as I (and my minion, gcc -Os) can tell, PL_curcop and
PL_in_eval do not need to be volatile. This improves the generated code
measurably - for example toke.o is 1.5% smaller. Every little helps.

p4raw-id: //depot/perl@33355
intrpvar.h diff | blob | blame | history
Unnamed repository; edit this file to name it for gitweb.