http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #1 from Stephan 2013-01-24 08:06:00 PST
---
During debugging, I actually looked at the value of every single local
variable, and you can actually see how the value of some variables (for example
"a") changes from one iteration to t
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #2 from Stephan 2013-01-24 10:00:41 PST
---
I just checked: The bug definitely was introduced with version 2.061!
With dmd version 2.060, everything works fine, with and without the "-O"
switch.
--
Configure issuemail: http://d.p
http://d.puremagic.com/issues/show_bug.cgi?id=9387
Walter Bright changed:
What|Removed |Added
CC||bugzi...@digitalmars.com
Seve
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #3 from Walter Bright 2013-01-25
00:30:21 PST ---
I can't reproduce this with the latest dmd. I'll upload a new beta tomorrow you
can try.
--
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
--- You
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #4 from Stephan 2013-01-25 05:01:07 PST
---
(In reply to comment #3)
> I can't reproduce this with the latest dmd. I'll upload a new beta tomorrow
> you
> can try.
What actually seems to be corrupted are the precompiled executabl
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #5 from Stephan 2013-01-25 06:53:50 PST
---
(In reply to comment #3)
> I can't reproduce this with the latest dmd. I'll upload a new beta tomorrow
> you
> can try.
Sorry to jump back and forth here. I have to again correct my pre
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #6 from Walter Bright 2013-01-25
10:37:02 PST ---
When I compile and run separately, it works fine.
You should also clarify whether you are using -m64 or not.
--
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?t
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #7 from Stephan 2013-01-25 10:43:26 PST
---
Right, I use the 64bit model.
And I tested this on OSX and on linux, with same outcomes on both platforms.
It's frustating that you can't reproduce. Thanks for responding quickly on this
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #8 from Don 2013-01-28 01:01:42 PST ---
BTW you might be interested in std.numeric.findRoot, which is the
root-finding-by-bracketing algorithm (in contrast to "Brent's algorithm" which
is minima-finding-by-bracketing). In terms of n
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #9 from Don 2013-01-28 01:03:34 PST ---
...and I can reproduce your bug.
--
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #10 from Don 2013-01-28 01:13:35 PST ---
I think there is an uninitialized variable in there. When I compile with -O, if
I run the same executable multiple times, sometimes it passes, sometimes it
fails.
--
Configure issuemail: ht
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #11 from Stephan 2013-01-28 06:57:56
PST ---
Hi Don,
glad to hear that you can reproduce the bug! I tested initializing all
variables by hand, and the bug still occurs.
Thanks for the suggestion to use std.numeric. Looks very usef
http://d.puremagic.com/issues/show_bug.cgi?id=9387
Don changed:
What|Removed |Added
Keywords||wrong-code
--- Comment #12 from Don 2013-01-29
http://d.puremagic.com/issues/show_bug.cgi?id=9387
Don changed:
What|Removed |Added
Keywords||ice
OS/Version|Mac OS X
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #14 from Don 2013-01-30 00:20:15 PST ---
ICE, further reduced:
--
void bug9387a(double x) { }
void ice9387()
{
double x = 0.3;
double r = x*0.1;
double q = x*0.1 + r;
double p = x*0.1 + r*0.2;
if ( q
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #15 from Don 2013-01-30 03:58:21 PST ---
And a reduction for the wrong-code case. This sometimes segfaults but usually
hangs. Looks like the saved RBX register gets trampled:
double brent(double x) { return x; }
void wrong9387()
{
/bfa5d0f0ba80c7ff6e0d67806714763584666fb2
fix Issue 9387 - Compiler switch -O changes behavior of correct code
--
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #17 from Walter Bright 2013-01-30
14:42:56 PST ---
https://github.com/D-Programming-Language/dmd/pull/1584
Thanks, Don, for the minimizations which made it easy for me to find the
problem. It was not a regression, although it look
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #18 from Stephan 2013-01-30 15:29:45
PST ---
Don and Walter, thanks for reducing the code and fixing the bug, all on a very
short timescale! This is going to be a very important fix for me. Using the
optimization switch is critical
/06d991f039eab23561398aea4ea764ea49a6dea4
fix Issue 9387 - Compiler switch -O changes behavior of correct code
https://github.com/D-Programming-Language/dmd/commit/9f3ab3f0b4713bd12a3ada71ca783bab1edae663
Merge pull request #1584 from WalterBright/b45
fix Issue 9387 - Compiler switch -O changes behavior of correct code
--
Configure
http://d.puremagic.com/issues/show_bug.cgi?id=9387
Walter Bright changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|
http://d.puremagic.com/issues/show_bug.cgi?id=9387
--- Comment #20 from Don 2013-01-31 00:52:53 PST ---
> Don and Walter, thanks for reducing the code and fixing the bug, all on a very
short timescale!
Thanks. Optimizer bugs get top priority, and this was the one of the worst bugs
of all time.
22 matches
Mail list logo
