This is the mail archive of the java-patches@gcc.gnu.org mailing list for the Java project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
Oyvind pointed out that environment variables currently can't be used to influence the GC on win32. This makes it significantly harder to diagnose and possibly work around problems such as http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10746 The reason getenv was disabled on win32 is given in http://gcc.gnu.org/ml/java-patches/2002-q1/msg00701.html (Based on some experiments on win32 and a quick look at the C89 standard, the Wine behavior described in that message is clearly wrong.) Does anyone know if the Wine bug that caused all this still exists? Assuming it does, the attached patch (only partially tested without Wine) should work around the problem while making environment variables usable on win32. The patch has the unpleasant side effect of treating environment variables with empty definitions as undefined environment variables under win32, but my impression is that is far more tolerable than the current situation. If someone could test this with mingw32 and ideally Wine, that would be appreciated. The attempt to generate this patch pointed out that I had some OSF1/Tru64 configury stuff in my tree that apparently was never checked into CVS. I left it in below. Is there a reason not to check this in? It is in the standalone version of the GC. Hans
Attachment:
0909gc.diff
Description: Binary data
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |