This patchset includes the following:

 * the modified GC that's capable of efficiently dealing with multiple stacks/thread
 * the MmFile/win32 fix for broken anonymous MmFiles
 * A workaround for the broken AA[] resize issue

As of a week or so, scrapple.tools requires this patchset to work. I'm sorry, but with the degree of neglect Phobos is experiencing, I cannot wait for the standard library to finally get fixed/updated in a year or so.

If you know of, or experience, any problems with this patch, please let me know.
If you know how to patch GDC so that it includes std/stack.d in the build process, please let me know.
If you have some experience with D's array handling, kindly try to work out why GDC/DMD consistently calls the wrong library function for arrays of associative arrays (resizing and newing).

Thanks for reading. :)

 --downs, faithful though disillusioned Phobos user.

downs Wrote:

> This patchset includes the following:
> 
>  * the modified GC that's capable of efficiently dealing with multiple stacks/thread
>  * the MmFile/win32 fix for broken anonymous MmFiles
>  * A workaround for the broken AA[] resize issue
> 
> As of a week or so, scrapple.tools requires this patchset to work. I'm sorry, but with the degree of neglect Phobos is experiencing, I cannot wait for the standard library to finally get fixed/updated in a year or so.
> 
> If you know of, or experience, any problems with this patch, please let me know.
> If you know how to patch GDC so that it includes std/stack.d in the build process, please let me know.
> If you have some experience with D's array handling, kindly try to work out why GDC/DMD consistently calls the wrong library function for arrays of associative arrays (resizing and newing).
> 
> Thanks for reading. :)

This is probably the wrong place to post this kind of stuff.  GDC has its own bug/feature request tracking system (I don't have a URL though).  Additionally, there's a D.bugs mailing list that is probably more appropriate.

downs wrote:
>
> If you have some experience with D's array handling, kindly try to work out why GDC/DMD consistently calls the wrong library function for arrays of associative arrays (resizing and newing).

I believe there's an outstanding bug report regarding the wrong function being called for an array operation, though I can't remember what it is offhand.  But this is something that has to be fixed in the compiler.


Sean