- 2010.06.18 02:38 "Re: [Tiff] libtiff 4.0 compiler warning on x64 in tif_read.c", by Bob Friesenhahn
- 2010.06.28 13:22 "Re: [Tiff] libtiff 4.0 compiler warning on x64 in tif_read.c", by Edward Lam
- 2010.06.29 17:12 "Re: [Tiff] libtiff 4.0 compiler warning on x64 in tif_read.c", by Albert Cahalan
2010.06.18 13:18 "Re: [Tiff] libtiff 4.0 compiler warning on x64 in tif_read.c", by Edward Lam
tif_read.c(207): warning C4244: '=': conversion from 'tmsize_t' to 'int', possible loss of data
Please submit a well-tested patch via libtiff Bugzilla.
Except, that this is really recent code and I was hoping that the author of said code could give some indication of what was intended here. I just ran a cvs annotate and it looks like Frank added this with revision 1.31 just over 2 months ago. I've added Frank to the CC list.
I could go and just do blind type changes from int to tmsize_t until the compiler is satisfied but that's just a losing battle if we keep adding new code that has such warnings. Is there some gcc flag we can add to make it NOT compile such code?
Incidentally, what do we do for Windows? When new code is committed into libtiff, how do we even know that it builds under Windows? We can certainly make it so that at least the Windows compiler treats these warnings as errors.
-Edward
PS. The ChangeLog is wrong with the "CHUNK_STRIP_READ_SUPPORT" macro. It should read "CHUNKY_STRIP_READ_SUPPORT" (note the extra Y) instead.