And I guess I should post here, too just to make sure it's not overlooked because it's kind of important if any other poor bastards are running hardware as ancient as the oldest 64-bit compute node around here:

http://d.puremagic.com/issues/show_bug.cgi?id=5597

Based on comparing cpuinfo for where this works vs. where it doesn't, I think it's related to the lahf instruction, which isn't present on some first-gen 64 CPUs.

On Wed, Feb 16, 2011 at 1:54 PM, Walter Bright <walter at digitalmars.com>wrote:

>
>
> David Simcha wrote:
>
>> Unfortunately I just found two new regressions:
>>
>> http://d.puremagic.com/issues/show_bug.cgi?id=5595 http://d.puremagic.com/issues/show_bug.cgi?id=5596
>>
>>
> Oh, damn. Just when I got to the last of your messages! _______________________________________________
>
> dmd-beta mailing list
> dmd-beta at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-beta
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.puremagic.com/pipermail/dmd-beta/attachments/20110216/c66723d8/attachment.html>

On 16 February 2011 19:54, Walter Bright <walter at digitalmars.com> wrote:
>
>
> David Simcha wrote:
>>
>> Unfortunately I just found two new regressions:
>>
>> http://d.puremagic.com/issues/show_bug.cgi?id=5595 http://d.puremagic.com/issues/show_bug.cgi?id=5596
>>
>
> Oh, damn. Just when I got to the last of your messages!

I'm working on cutting down 5595.

Another absolutely horrific bug has just been reported:
Bug 5591   EBX register not preserved when calling stdcall function pointer

David Simcha wrote:
> And I guess I should post here, too just to make sure it's not overlooked because it's kind of important if any other poor bastards are running hardware as ancient as the oldest 64-bit compute node around here:
>
> http://d.puremagic.com/issues/show_bug.cgi?id=5597
>
> Based on comparing cpuinfo for where this works vs. where it doesn't, I think it's related to the lahf instruction, which isn't present on some first-gen 64 CPUs.
>

I found the problem and the solution, but it'll take me a bit to fix it as I'm going to nwcpp tonight.

That's ok.  Rest assured, I would never use hardware that ancient and crappy for anything computationally intensive enough to require 64-bit. The fact that it took till now to find this bug says a lot about the hardware in question.

On 2/16/2011 8:01 PM, Walter Bright wrote:
>
>
> David Simcha wrote:
>> And I guess I should post here, too just to make sure it's not overlooked because it's kind of important if any other poor bastards are running hardware as ancient as the oldest 64-bit compute node around here:
>>
>> http://d.puremagic.com/issues/show_bug.cgi?id=5597
>>
>> Based on comparing cpuinfo for where this works vs. where it doesn't, I think it's related to the lahf instruction, which isn't present on some first-gen 64 CPUs.
>>
>
> I found the problem and the solution, but it'll take me a bit to fix
> it as I'm going to nwcpp tonight.
> _______________________________________________
> dmd-beta mailing list
> dmd-beta at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-beta
>

I'm happy to say that, after a few minor modifications to the code, SciD compiles and passes all* tests in both 32- and 64-bit mode.  Great work, guys! :)

-Lars


* I had to comment out one test with -m64, but I think that is a problem with the LAPACK bindings I use, and not DMD.




On Wed, 2011-02-16 at 08:37 -0500, David Simcha wrote:
> Great.  To everyone:  So that we have a successful release, I suggest putting this compiler though the paces on any code you can find today, before it's officially released.  That's my only secret to uncovering all these bugs:  I spent a whole bunch of time compiling dstats and std.parallelism and (to a lesser extent) plot2kill, isolating bugs and filing reports.  Anyone who has both the time and a codebase with good unit test coverage that they know works on 32-bit, please do the same.
> 
> On 2/16/2011 4:46 AM, Walter Bright wrote:
> > round up of more disastrous 64 bit bugs fixed
> >
> > http://ftp.digitalmars.com/dmd1beta.zip
> > http://ftp.digitalmars.com/dmd2beta.zip
> > _______________________________________________
> > dmd-beta mailing list
> > dmd-beta at puremagic.com
> > http://lists.puremagic.com/mailman/listinfo/dmd-beta
> >
> 
> _______________________________________________
> dmd-beta mailing list
> dmd-beta at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-beta

> David Simcha wrote:
>>
>> Unfortunately I just found two new regressions:
>>
>> http://d.puremagic.com/issues/show_bug.cgi?id=5595

I've posted a possible patch to this one, in Bugzilla.

>> http://d.puremagic.com/issues/show_bug.cgi?id=5596

I fear this might be another instance of bug 4269 (Regression(2.031):
invalid type accepted if evaluated while errors are gagged).