[graphite] Regression testing on 2011_01_20_14_51_01

1 view
Skip to first unread message

Sebastian Pop (ed)

unread,
Jan 20, 2011, 3:51:50 PM1/20/11
to gcc-grap...@googlegroups.com
14:51 => Start run 2011_01_20_14_51_01
14:51 => SVN update
14:51 => Updated to:
http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=169065
------------------------------------------------------------------------
r169065 | spop | 2011-01-20 14:26:06 -0600 (Thu, 20 Jan 2011) | 10 lines

Do not use a data ref with indices invalid in the loop of the close_phi.

2011-01-20 Sebastian Pop <sebast...@amd.com>

* graphite-sese-to-poly.c (dr_indices_valid_in_loop): New.
(close_phi_written_to_memory): Call for_each_index with
dr_indices_valid_in_loop.

* gfortran.dg/graphite/id-24.f: New.
* gfortran.dg/graphite/id-25.f: New.
------------------------------------------------------------------------
14:51 => Configure
14:51 => Build

Sebastian Pop (ed)

unread,
Jan 20, 2011, 4:00:15 PM1/20/11
to gcc-grap...@googlegroups.com

2011-01-20 Sebastian Pop <sebast...@amd.com>

15:00 => Install to /home/seb/daily_builds/install/2011_01_20_14_51_01
15:00 => Run extra tests

Sebastian Pop (ed)

unread,
Jan 21, 2011, 2:02:27 AM1/21/11
to gcc-grap...@googlegroups.com

2011-01-20 Sebastian Pop <sebast...@amd.com>

BUILD SUCCESSFUL
CPU2006 built with
# base: -O3 -fno-graphite-identity -fno-loop-interchange -fno-loop-strip-mine -fno-graphite
# peak: -O3 -fno-loop-strip-mine
Build errors: 459.GemsFDTD(base)
Error: 1x416.gamess 1x459.GemsFDTD

cpu2006
cpu2006.compile.gnu
cpu2006.compile.png
cpu2006.run.gnu
cpu2006.run.png
pb05.sum
2011_01_20_14_51_01
Reply all
Reply to author
Forward
0 new messages