Well, I wasn't going to consider the assembler yet because I didn't want to stir things up too much, but now that you ask ...
I have a lot of experience with a lot of assemblers on a lot of machines. I was never too fond of masm, although I did write a lot of code using it. I preferred (note past tense) tasm even though it was just as quirky in its syntax. I currently use nasm and really like it. Now, I will have to disclose that I have not tried jwasm. There's no reason behind that except for laziness. Having learned nasm and finding it met my needs, I just didn't try jwasm because I didn't want to learn it. So, I cannot compare nasm to it. Given my druthers, I would like to see nasm as the standard for the project. Again, this is only for what is currently on SourceForge.net and maybe the base in the future, but that is a personal choice and not one scientifically arrived at. I could be wrong. Pat Villani Project Coordinator On Sat, Aug 21, 2010 at 1:15 PM, Christian Masloch <[email protected]>wrote: > > I'm thinking it would be a good idea to standardize on one C/C++ compiler > > for the project. [...] > > I would be interested in your opinions about standardizing one or several > assemblers for the project. I think that the kernel and FreeCOM assembler > source files currently use NASM, a 2-clause BSD licensed assembler. > > NASM uses a slightly different syntax than the one of MASM that most DOS > programmers were used to. A free assembler with nearly the same syntax as > MASM is Japheth's recent JWASM, an extended fork of Open Watcom's > assembler. It is distributed under the same license as Open Watcom. > > I propose that any assembler source code added to the project should be > compilable with either NASM or JWASM. (It might be useful to some that > NASM also provides limited TASM compatibility; refer to its documentation.) > > > I have another concern regarding the use of inline assembler code in C > source code, as mentioned in this thread. While reworking current code > isn't required, please consider to implement functions you might write > with inline assembler code either in pure C or pure assembler. > > In the later case, that source code should be moved to a proper file on > its own. That file would be compiled into an object format using one of > the stand-alone assembler. This might also make such functions easier to > port to other compilers or platforms since the porter won't have to look > through C source files for inline assembler code. > > > An off topic question: Is there any source code in the full distribution > that uses C++ and OOP? > > Regards, > Christian > > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by > > Make an app they can't live without > Enter the BlackBerry Developer Challenge > http://p.sf.net/sfu/RIM-dev2dev > _______________________________________________ > Freedos-devel mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/freedos-devel >
------------------------------------------------------------------------------ This SF.net email is sponsored by Make an app they can't live without Enter the BlackBerry Developer Challenge http://p.sf.net/sfu/RIM-dev2dev
_______________________________________________ Freedos-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freedos-devel
