On 2013-07-01 20:08-0700 Austin English wrote:
In other words, while fixing Cygwin issues is a valuable effort, it takes a lot of time and effort that is better spent elsewhere.
Hi Austin:
I agree that fixing Cygwin on Wine issues will take some considerable initial effort learning about Cygwin, but after some Wine developer has made that effort (or if they already have that Cygwin expertise), then in addition they have complete access to the Cygwin source code for the Windows calls that are not properly working on Wine, and that is a big advantage in knocking off Wine problems one by one that are turned up this way. And I want to emphasize again that fixing those Wine problems should be a big advantage for _all_ Windows applications being run on Wine, not just Cygwin.
[...] [A]sking others to spend their time to debug issues that fix Cygwin isn't likely to find many volunteers.
Well, there is at least one volunteer working on this, and he has made some good progress. Andrey Turkin's small patch for Wine (see his attachment at http://bugs.winehq.org/show_bug.cgi?id=24018) combined with the Cygwin fork fix now works for him and me in the sense that the postinstall scripts now all run without catastrophic runtime errors or hangs. I have added all the details about my experience with that breakthrough to that bug report. I have included an attachment there of a script to allow convenient updating of cygwin1.dll with the fork-fixed version.
Note, the first postinstall script that is run creates an empty /etc/fstab rather than the desired result and also fails to create /dev. These errors very likely create problems for the other postinstall scripts and might be some/all of the reason why four of those scripts generate error messages and have non-zero return codes.
So the status right now is that the postinstall scripts run by setup.exe on Wine still do not give clean results, but the results are much better than before the Cygwin fork fix and Andrey's patch. There are probably a couple of Wine bugs that still need to be fixed before clean postinstall script results can be achieved. Of course, a clean result from the postinstall stage of setup.exe would only be the first step, but that step should give solid footing for finding and fixing the rest of the Cygwin on Wine bugs.
Alan __________________________ Alan W. Irwin
Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca).
Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________
Linux-powered Science __________________________