This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ 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] | |
My initial patch to fix this a week ago was wrong. I have higher hopes
for this patch, which I've tested by constructing various incorrect
libstdc++.so binaries, all of which were caught as expected.
So, baseline symbol files with *_LDBL_* version names won't be
marked missing if the test symbol file doesn't contain any *_LDBL_*
versions.
In addition, this adds a check for __float128 mangled names not in
*_LDBL_* version names.
I believe these are the last abi_check FAILs that involve
testsuite/util/* sources. The rest are configure/build issues.
In particular, this doesn't fix the issue that Jakub pointed out here:
http://gcc.gnu.org/ml/gcc-patches/2008-03/msg01262.html
where
baseline_dir="$glibcxx_srcdir/config/abi/post/${abi_baseline_pair}\$(MULTISUBDIR)"
isn't correct when --with-cpu=default32 on some platforms.
I would like to fix this too, but haven't investigated this yet.
tested x86/linux
-benjaminAttachment:
p.20080326-3
Description: Text document
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |