[ warning: getting a bit off topic ]

> > > Another point, though I do not know if this is maybe the "industry
> > > standard" is that I find the Xilinx tools, at least ISE/Xst just not
> > > suitable for serious use.
> 
> Welcome to the world of hardware synthesis.
> I've used the Xilinx tools at work and Synplify for my master
> thesis, and i can tell you, even though i'm used to unintuitive
> and broken software like gcc or sun cc, both were more horrible
> than any broken shit i've used in my life. But that's my personal
> opinion.

It has been 15 or 20 years since I've found a bug in gcc.
(not counting desiring improvements to the error/warning messages,
and the arch support problem)

> > I understand that it's frustrating at times -- it's definitely not GCC 
> > :) -- but as long as one approaches it as a new tool with a learning 
> > curve, I believe it's perfectly manageable.
> 
> Unless you are a few weeks before a dead line and fight with the
> bug of the week every time you change something small.
> 
> And using gcc for comparison is like using a worm-eaten apple
> to tell people that a rotten apple isn't that bad.
> gcc is considered by too many people as 1) broken 2) too complex to fix
> 3) having undecipherable error messages and 4) producing bad code
> more often than not.

You left out the biggie: gcc is dropping arches.  :-(

> It's only being used because there is nothing better.

Depending on what arch you are targeting there may be a proprietary
compiler than generates better code, at least for some cases.

Due to gcc dropping arch support, some people are looking into
alternatives, like pcc.  I would think that forking an earlier
version of gcc would be a possibility, but I don't know if
anyone is considering that.

Back when gcc came out, twenty-something years ago, it was an improvement.
When your CPU is only 0.6 MIPS, anything that made the code run faster
was very welcome indeed.  I haven't followed it closely, but apparently
there were personell changes that have lead to the current situation.

> And i'm always astonished when i talk with VLSI/FPGA guys, that
> they can cope with the brokenness of their tools which are by far
> worse than gcc, especially considering what Synplicity and Mentor
> Graphics ask for their tools[1].

What choice do they have?  You do the best you can with the tools you have.

Someone should write a FLOSS alternative.  Unfortunately that is more than
an easy Saturday afternoon project.

There is lots of software that is buggier than gcc:  firefox, ffmpeg, linux,
xpdf, ...
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to