http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25145
Richard Guenther changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|
--- Comment #10 from rguenth at gcc dot gnu dot org 2008-01-12 13:05
---
The first testcase in the original report is fixed by comparison
canonicalization,
even if a temporary is used (forwprop re-instantiates the canonicalized
comparison).
The testcase in comment #1 is fixed by FRE si
--- Comment #9 from rguenth at gcc dot gnu dot org 2006-12-28 12:14 ---
The testcase in comment #1 is fixed by comparison canonicalization of i <= j -
1
to i < j. Of course it fails again if we use a temporary for j - 1 like in the
testcases in other comments.
--
http://gcc.gnu.org
--- Comment #8 from rguenth at gcc dot gnu dot org 2006-04-26 16:09 ---
Now the patch for 25148 fixes the wrong answer for the testcase in comment #2
if the patch in comment #5 is applied. It needs
Index: tree-ssa-propagate.c
--- Comment #7 from pinskia at gcc dot gnu dot org 2005-12-12 21:01 ---
Confirmed.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCON
--- Comment #6 from pinskia at gcc dot gnu dot org 2005-11-29 17:34 ---
(In reply to comment #5)
> Note this was the simple fix which exposes those latent bugs as far as I can
> see that should work, we get the correct range but the rest of VRP goes
> bonkers:
I should also note it does
--- Comment #5 from pinskia at gcc dot gnu dot org 2005-11-29 00:38 ---
Note this was the simple fix which exposes those latent bugs as far as I can
see that should work, we get the correct range but the rest of VRP goes
bonkers:
Index: tree-vrp.c
--- Comment #4 from pinskia at gcc dot gnu dot org 2005-11-29 00:34 ---
I run into a different regression if we try to compile the first example in
comment #0.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25145
--- Comment #3 from pinskia at gcc dot gnu dot org 2005-11-29 00:21 ---
This needs PR 25148 fixed to do the correct thing for comment #1 as we get the
wrong answer for a + -1 < a, we get false when we should get true.
--
pinskia at gcc dot gnu dot org changed:
What|Re
--- Comment #2 from pinskia at gcc dot gnu dot org 2005-11-28 23:43 ---
extract_range_from_binary_expr should do symbolic ranges for stuff like this
(or at least it seems like it should).
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25145
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-11-28 22:56 ---
Even the simple code like:
int f(int i, int j )
{
int k;
k = i+ - 1;
return k < i;
}
Does not get VRP to optimize it which means we don't do that much symbolic
ranges as we should. From looking at things,
11 matches
Mail list logo
