On Thursday 29 June 2006 16:57 Matt Benson wrote: > Right... I was going to ask, are(n't) the modified > sources still compatible with ml.exe? If so, there > shouldn't be much reason -not- to adopt a > least-common-denominator approach (applying your > changes), should there? I'm not much on makefiles, > but it might be nice to be able to parameterize the > assembler in some way.
The reason I used NASM was that it uses Intel's instruction notations (compared to gas which uses AT&T) so port was easier. But no, the syntax of instructions is not the only difference. The main difference is compiler directives which every assermbler invents on its own. Also NASM has different syntax for memory operations. So if there is a common denominator I don't know about it. > I will have to go back and try to reassemble those > conversations; I intend to have a look at making all > the Ant build stuff as performant as possible, but I > suppose I will need to know what's on the horizon wrt > the structure of the native stuff. In any event, I > hope cygwin is supported soon (I doubt I personally > could get it working in anything like a reasonable > amount of time)... currently I open a VC window, run > the MS SDK SetEnv /XP32 /DEBUG, run another batch file > to make all the other env. changes I need, and use > that CMD prompt to run Ant. Then I do everything else > in a cygwin window. Annoying. :) I've written already how to change cygwin.bat to set up VS.NET environment for cygwin window. The cmd shell is so less usable than bash. You just have to remember to use windows paths when dealing with windows native progs and java. -- Gregory Shimansky, Intel Middleware Products Division --------------------------------------------------------------------- Terms of use : http://incubator.apache.org/harmony/mailing.html To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
