This is the mail archive of the gcc-regression@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: 6 GCC HEAD regressions, 6 new, with your patch on 2004-11-20T00:01:18Z.



On Nov 20, 2004, at 11:20 AM, GCC regression checker wrote:


With your recent patch, GCC HEAD has some regression test failures,
which used to pass.  There are 6 new failures, and 0
failures that existed before and after that patch; 2 failures
have been fixed.

The new failures are:
native ./c-decl.o differs
native ./loop.o differs
native cp/pt.o differs
native java/parse.o differs
native g++.sum g++.old-deja/g++.abi/cxa_vec.C
native g++.sum g++.old-deja/g++.brendan/new3.C

The fixed failures are:
native g++.sum g++.old-deja/g++.abi/cxa_vec.C
native g++.sum g++.old-deja/g++.brendan/new3.C

This was caused by either: +2004-11-19 Jeff Law <law@redhat.com> + + * tree-ssa-threadupdate.c: Replace REDIRECTION_DATA varray with + a hash table. Extensive modifications throughout to support + that change. + (struct el): New. + (struct local_info): New. + (struct redirection_data): Add new INCOMING_EDGES and DO_NOT_DUPLICATE + fields. + (redirection_data): Now a hashtable. + (redirection_data_hash, redirection_data_eq): New. + (lookup_redirection_data, create_duplicates): New. + (create_edge_and_update_destionation_phis): New. + (fixup_template_block, redirect_edges): New. + (thread_block): Use hash table traversals instead of loops over + varray entries or incoming edge vectors.

2004-11-19 Kazu Hirata <kazu@cs.umass.edu>

+	* tree-cfg.c (reinstall_phi_args): New.
+	(tree_split_edge): Use it after redirecting an edge.  Don't
+	modify PHI_ARG_EDGE.
+

(Note that Jeff's changelog was in the wrong place and I fixed that)

Thanks,
Andrew Pinski


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]