This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

are the egcs patches useless?


Dear egcs maintainers,

I prefer to download patches of the egcs source tree instead of the
whole huge tarballs. However, some of these patches appear to be buggy
(from the technical point of view): they don't apply cleanly as they
evidently were not generated using the distributed source trees, but
some changes to the source trees were made "silently", and were not
reflected in the patches. :-(

Isn't the patch-generation procedure automated, and such errors are
common? If so, why?

I patched the egcs source tree from egcs-970828 to egcs-980321; there
were some problems with unclean patching; most of them were in 1997,
and I resolved manually, except 3 files:

gcc/testsuite/g++.old-deja/g++.jason/2371.C
gcc/testsuite/g++.old-deja/g++.mike/p784.C
gcc/testsuite/g++.old-deja/g++.mike/p785.C

I checked the source tree using the .brik file, and _all_ was Ok
except those 3 files.  After that, all patches were clean, --- good.
But when I tried to apply the egcs-980321-980328.diff.gz and got
rejects again! These were with those 3 files (which is probably
acceptable, since my files were broken due to incorrect old patches),
and also:

texinfo/util/texi2dvi
.brik

The analysis of the rejects shows that someone silently updated the
texinfo directory, but these changes did not get into the patch!

Thank you, and I'm sorry for bothering you. Please Cc: me your answers
as I'm not subscribed to egcs-bugs.

	Best regards, -- Vladimir.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]