14.08.2012 06:23, Bart Oldeman wrote: > Hi Stas, > > it looks good to me, if it works well and is the cleanest solution by > all means go for it; I still haven't made up my mind on whether it works well or whether its a cleanest solution, but it already allowed me to find 2 absolutely critical bugs in the main code: http://dosemu.git.sourceforge.net/git/gitweb.cgi?p=dosemu/dosemu;a=commit;h=fb607434d0c8650c7c0c22c45ec75f2bc212a910 http://dosemu.git.sourceforge.net/git/gitweb.cgi?p=dosemu/dosemu;a=commit;h=818d4dabcfd9f1a1258930ac90c6505590d19fd1 I am posting the links here because for some reasons SF decided to not show these commits in a notification e-mails.
Please review the above commits: I am not entirely sure I set the stack the right way (but it seems bochs does this so) As for coopthreads, I am still trying to get something stable, but the deeper I integrate it, the more unstable it becomes so far. It uncovers bugs and assumptions in dosemu code and I don't know how far will it go fixing them all. > as essentially the DPMI code and DOSEMU code are two coroutines I think the sense of coroutines can be completely eliminated if you don't save dosemu context on DPMI switch. Since you only have a fixed entry point, you can use some dummy context, or some saved-once context. There is no need to always save it as it does not change. If this is done, then obviously this won't even remotely resemble coroutines. ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Dosemu-devel mailing list Dosemu-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dosemu-devel
