[Bug tree-optimization/68964] [6 regression] Internal compiler error for test case gcc.dg/tm/20100610.c since r231674

2016-01-13 Thread rth at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68964 --- Comment #5 from Richard Henderson --- Author: rth Date: Wed Jan 13 17:03:42 2016 New Revision: 232330 URL: https://gcc.gnu.org/viewcvs?rev=232330=gcc=rev Log: PR 68964 gcc/ PR tree-opt/68964 * target.def (builtin_tm_load,

[Bug tree-optimization/68964] [6 regression] Internal compiler error for test case gcc.dg/tm/20100610.c since r231674

2015-12-22 Thread rth at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68964 --- Comment #3 from Richard Henderson --- The path on which this test goes off the rails is supposed to only be used for structure assignments, where memmove must be used, and thus taking the address of both sides of the assignment will work.

[Bug tree-optimization/68964] [6 regression] Internal compiler error for test case gcc.dg/tm/20100610.c since r231674

2015-12-22 Thread rth at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68964 --- Comment #4 from Richard Henderson --- Oh, yes, duh. These vector types are (at least partially) target-specific, and so the function definitions are left for the target to fill in at startup time. See config/i386/i386.c bdesc_tm for

[Bug tree-optimization/68964] [6 regression] Internal compiler error for test case gcc.dg/tm/20100610.c since r231674

2015-12-18 Thread rguenth at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68964 Richard Biener changed: What|Removed |Added CC||rth at gcc dot gnu.org Target