Place to attach GCC 4.7 pending patches and PRs
Created attachment 22447 [details] Improve reload inheritance's code generation and predictability http://gcc.gnu.org/ml/gcc-patches/2010-11/msg01917.html
Created attachment 22603 [details] Improve reload inheritance's code generation and predictability Updated version taking into account comments from various individuals. Compile time cost is down to less than .25% according to my testing.
Created attachment 22749 [details] Patch to combine several distinct arrays into a single VEC Patch to move various reg_equiv_xxx arrays into a single structure to ease debugging and growing the structures in response to creating new pseudos. About .1% compile-time slowdown, possibly due to the checking built into the VEC implementation.
http://gcc.gnu.org/ml/gcc-patches/2011-03/msg00609.html Patch to improve XOR optimizations.
http://gcc.gnu.org/ml/gcc-patches/2011-02/msg01443.html Vlad approved this on the main gcc list in Mar 2011.
http://gcc.gnu.org/ml/gcc/2011-03/msg00098.html Minor IRA patch from Vlad.
http://gcc.gnu.org/ml/gcc-patches/2011-01/msg00176.html which was reviewed in http://gcc.gnu.org/ml/gcc-patches/2011-01/msg00206.html?
Created attachment 23693 [details] Minor improvement to register allocation Was showing good results for x86; however, was regressing for mn103 and put on indefinite hold pending resolution of the regressions. While it's on hold, I don't want it to get lost completely.
Created attachment 23694 [details] Andy's frame-pointerless mcount patch This is Andy's frame-pointerless mcount patch. There was a thread on one of the lists about this patch, but I couldn't find it in my quick search. Again, moving it here so it doesn't get lost.
Closing as fixed as 4.7 is stage 4 now.
(In reply to comment #10) > Closing as fixed as 4.7 is stage 4 now. Before closing this, all pending patches from 4.7 should be moved to the 4.8 meta-bug. Otherwise, they will get lost. It wouldn't be the first time that patches are lost.
(In reply to comment #11) > (In reply to comment #10) > > Closing as fixed as 4.7 is stage 4 now. > > Before closing this, all pending patches from 4.7 should be moved to the 4.8 > meta-bug. Otherwise, they will get lost. It wouldn't be the first time that > patches are lost. All of those patches have already been lost as far as I can see. Most of them are over a year old and some have been in the pending patches meta-bug for 2 or 3 major releases already.