On Wed, 2013-11-27 at 15:54 +1300, David Koontz wrote: > On 27 Nov 2013, at 4:46 am, Brian Drummond <br...@shapes.demon.co.uk> wrote: > > > > > I have managed to get GHDL to build with gcc4.8.2 as both the compiler and > > the sources. > > > > Instructions (README), build script and patch attached. > > > > Testing so far shows no problems, including the simple OSVVM demo. > > > > Apologies for formatting of previous message. > > I had no idea Evolution would simply "cat" some of the attachments (but > > not others!) instead of attaching them properly. > > > > Anyway a better home for the patch and instructions is at: > > https://gna.org/bugs/index.php?21305 > > > Any chance of a change log (even by groups of files)? There seems to be > three > groups of changes in this patch file. You're OS-VVM patches, the GCC 4.8.2 > patches and other changes.
Sure; I'll append an informal one to the "bug report" today. Are you willing (and authorised) to change the status of Support and Bug requests as and when they have solutions? The patch fixes Support #3028 (Number exceeds 32-bit range) and probably Support #2903 (Floating point exception). The first is now a more friendly error message and the second I can't reproduce; I think it went away alongside other changes. So these can be updated now. Plus another bug whose number I haven't traced yet (may have been on "Discussion" in which case I'll report it on "bugs" to give it a number). I also have here (not yet uploaded) patches for Support #2940 (function name is a design unit) and Bug #19195 (failure aliasing a record) which are in the bugtracker but never made it into my trunk. I want to clear up a few more issues and create another patch : your opinion on the best way would be appreciated. I am leaning towards a separate patch to apply on top of yesterdays to clean up several issues. That way it can remain separate until it meets with e.g. your approval or something more formal like (see below) Several of the issues are genuinely bad VHDL causing ghdl to bugbox, (and Support #3060 and #2686 are the same "raise") so "better" error messages are the right fix : please suggest improvements to the messages where appropriate! (e.g. #2940 now reports GCD.vhd:73:25: function name is a design unit because someone is calling an entity instead of a function! Would "cannot call a design unit as a function" be clearer?) > Funny, the weather today pretty much matches that of Dùn Omhain here in the > Antipodes (where it's supposed to be summer). +1 for the correct Gaelic spelling! > Also Nick Gasson has hosted the Philip Wilsey's VHDL tESTs (VESTs) on > github) and reported -r150 ghdl passed 3558 of the 3759 no error (analyze) > tests with 201 failures. > There are tests that are supposed to fail during elaboration or simulation > not included in those numbers. > I think I saw something that would knock at least one of those failures out > in your patch. > See https://github.com/nickg/vests (and this is a case of having to know the > path). GPLv2 which is why they're in a separate repository. I'll look at these pages later today. Are the Failures available as a separate webpage or document? If so I suggest creating a "bug" for it, or linking to it if Nick permits. > I'm not interested in getting stuck chasing gcc releases but could be > persuaded to > work on VHDL language compliance issues and testing. While, having climbed to gcc4.7 took a month, gcc4.8 took less than a week (back in May but paid work intervened) so - at least for now - I can cover that base. And it transpired I had grabbed "trunk" by mistake so we may have a head start on gcc4.9. And I can patch simple failures at the moment but your oversight would be VERY welcome to confirm that the LRM agrees with me(!) ... especially if/when I move on to more involved stuff. > I spent a week or so screening > ghdl bug reports and trimming them down, added a few more. (VHDL Lawyering > when done > properly is labor intensive, too bad we don't get paid like lawyers). Saw that : let me get a few simpler ones under my belt and I'll look at them. > After suitable testing I could probably manage to enter the current spate of > patches into the gna ghdl archive, > although I think it may be time to host a branch somewhere else. Hence the > interest in change logs. I've been meaning to put my trunk under Mercurial for revision control. How about I start a branch on sourceforge? I already have a compiler there. http://sourceforge.net/projects/msp430ada/ We can call it a "testing" branch. I don't want to take anything away from Tristan; perhaps gna.org hosts the "stable" (too stable?) branch and he can pull changes from "testing" and approve them (or otherwise) as time permits. Thoughts? - Brian _______________________________________________ Ghdl-discuss mailing list Ghdl-discuss@gna.org https://mail.gna.org/listinfo/ghdl-discuss
