Peter Åstrand wrote:
> Regardless of which libraries we decide to use, we need to select a
> decent assembler. In my opinion, NASM is a great assembler. Before we
> decided to use it for SIMD/jpeg, we evaluated different options and
> NASM certainly looked like the most promising solution, and I think it
> still does.
Well, no, using intrinsics doesn't require a separate assembler.  It
uses the one already present in the compiler (GAS, in the case of GCC.)

> MSVC 2008 is certainly a problematic requirement for many people. As
> we have discussed before, one idea, which I still believe we should
> implement, is to get rid of all MS compiler dependencies. This means
> that the Autotools toolchain, GCC, and NASM could be used an all
> platforms.
I agree with getting rid of MS compiler dependencies and using MSYS or
Cygwin.  But I don't think MSVC 2008 is as problematic as you think. 
The compiler is available for free these days, and it's easier for most
people to install than a full-blown Cygwin distribution.  TurboVNC
requires it because TightVNC 1.3.x required it.  The TightVNC 1.3.x
server has some features that will not build with MinGW.

> It's true that NASM 2.0 isn't yet available on most distros such as
> RHEL5. But still, 2.0 was released more than one year ago. It's likely
> that it will be included in most future distributions, perhaps RHEL
> 5.4 and 6.0?
6.0 perhaps.  The 5.x series is "locked in" as to the versions they can use.

> Until then, installing NASM 2.0 seems to be very easy. I just tried it
> on our RHEL5 system:
>
> # wget
> http://www.nasm.us/pub/nasm/releasebuilds/2.05.01/nasm-2.05.01-1.src.rpm
> # rpmbuild --rebuild nasm-2.05.01-1.src.rpm
> # rpm -vhU /usr/src/redhat/RPMS/i386/nasm-2.05.01-1.i386.rpm
>
> Seems to work great. There are even binaries available for many
> platforms.
OK.  But even if I conceded this, that was one of only several arguments
in favor of using OML.  Maintainability is the biggest one.  But I'm
done arguing.


------------------------------------------------------------------------------
_______________________________________________
Tigervnc-devel mailing list
Tigervnc-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/tigervnc-devel

Reply via email to