This is the mail archive of the gcc@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]
Other format: [Raw text]

Re: Test result email failure


Bill Northcott wrote:

Below are the summaries. The compiler can't be that broken. After all it managed to bootstrap and build the libraries. Although I have to admit that there seem to be a lot of failures. Maybe some of the tests are broken on the platform?

This is about what you see when things build and link, but nothing executes successfully. (I'm recently familiar with all this, from first attempts at 64-bit. :-) ) Did you try building and running just a plain hello world?

Stan


Since this, I have rebuilt the compiler with --enable-nls and adding f77 and java to langauges. Currently I am stuck on java, but it appears to be a header issue on the host dues to wreckage of previous gcc compilers. I have asked the admins to fix it.


Bill Northcott

=== g++ Summary ===

# of expected passes 8227
# of unexpected failures 1262
# of unexpected successes 3
# of expected failures 67
# of unresolved testcases 25
# of untested testcases 30
# of unsupported tests 83
/home/bank/s9573347/src/GNU/gcc-build-7Feb/gcc/testsuite/../g++ version 3.4.3


=== gcc Summary ===

# of expected passes        11826
# of unexpected failures    7001
# of expected failures        61
# of unresolved testcases    5127
# of untested testcases        322
# of unsupported tests        353
/home/bank/s9573347/src/GNU/gcc-build-7Feb/gcc/xgcc version 3.4.3


=== objc Summary ===


# of expected passes        308
# of unexpected failures    536
# of unresolved testcases    508
# of unsupported tests        8
/home/bank/s9573347/src/GNU/gcc-build-7Feb/gcc/xgcc version 3.4.3


Compiler version: 3.4.3
Platform: alphaev56-dec-osf5.1
configure flags: --prefix=/home/bank/s9573347 --disable-nls --with-gc=simple --with-stabs --disable-threads --enable-languages=c,c++,objc --enable-shared --program-suffix=-3.4




On 09/02/2005, at 10:40 AM, Joe Buck wrote:

Maybe you could give us a summary of it.

How many lines in it begin with FAIL: ?

....


To get such a huge output, probably almost all of your tests are failing,
due to some major breakage like a completely nonfunctional compiler.






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