[Bug tree-optimization/38207] Union in structs are not well optimized

2009-04-03 Thread rguenth at gcc dot gnu dot org
--- Comment #8 from rguenth at gcc dot gnu dot org 2009-04-03 10:34 --- Fixed. -- rguenth at gcc dot gnu dot org changed: What|Removed |Added Status|ASSIGNED

[Bug tree-optimization/38207] Union in structs are not well optimized

2009-04-03 Thread rguenth at gcc dot gnu dot org
--- Comment #7 from rguenth at gcc dot gnu dot org 2009-04-03 10:26 --- Subject: Bug 38207 Author: rguenth Date: Fri Apr 3 10:24:28 2009 New Revision: 145494 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=145494 Log: 2009-04-03 Richard Guenther PR middle-end/13146

[Bug tree-optimization/38207] Union in structs are not well optimized

2008-12-10 Thread rguenth at gcc dot gnu dot org
--- Comment #6 from rguenth at gcc dot gnu dot org 2008-12-10 16:26 --- Subject: Bug 38207 Author: rguenth Date: Wed Dec 10 16:25:27 2008 New Revision: 142649 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=142649 Log: 2008-12-10 Richard Guenther <[EMAIL PROTECTED]> PR

[Bug tree-optimization/38207] Union in structs are not well optimized

2008-11-22 Thread rguenth at gcc dot gnu dot org
--- Comment #5 from rguenth at gcc dot gnu dot org 2008-11-22 17:47 --- "The problem is a general defect in FRE that causes us to never CSE a load with the default definition of the virtual SSA name." after walking, that is. I have a patch that also makes PR38212 to be visible on the t

[Bug tree-optimization/38207] Union in structs are not well optimized

2008-11-21 Thread rguenth at gcc dot gnu dot org
--- Comment #4 from rguenth at gcc dot gnu dot org 2008-11-21 10:52 --- Which means that we correctly optimize int f(struct a *c) { int d; c->c = 1; d = c->a; c->c = 1; return c->a + d; } -- rguenth at gcc dot gnu dot org changed: What|Removed

[Bug tree-optimization/38207] Union in structs are not well optimized

2008-11-21 Thread rguenth at gcc dot gnu dot org
--- Comment #3 from rguenth at gcc dot gnu dot org 2008-11-21 10:51 --- The oracle correctly figures that c_1(D)->D.1593.a and c_1(D)->D.1597.c do not alias. The problem is a general defect in FRE that causes us to never CSE a load with the default definition of the virtual SSA name.

[Bug tree-optimization/38207] Union in structs are not well optimized

2008-11-20 Thread pinskia at gcc dot gnu dot org
--- Comment #2 from pinskia at gcc dot gnu dot org 2008-11-20 23:47 --- This works correctly on the RTL level which means the aliasing oracle does not say c->a and c->c cannot alias. -- pinskia at gcc dot gnu dot org changed: What|Removed |Added --

[Bug tree-optimization/38207] Union in structs are not well optimized

2008-11-20 Thread pinskia at gcc dot gnu dot org
--- Comment #1 from pinskia at gcc dot gnu dot org 2008-11-20 23:31 --- Note this was found in clang: http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20081117/009546.html -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38207

[Bug tree-optimization/38207] Union in structs are not well optimized

2008-11-20 Thread pinskia at gcc dot gnu dot org
-- pinskia at gcc dot gnu dot org changed: What|Removed |Added Severity|normal |enhancement http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38207