On 2019-05-07 14:14 -0500, Bruce Dubbs via lfs-dev wrote: > On 5/7/19 12:44 PM, Xi Ruoyao via lfs-dev wrote: > > On 2019-05-07 11:46 -0500, Bruce Dubbs via lfs-dev wrote: > > > On 5/7/19 6:07 AM, Xi Ruoyao via lfs-dev wrote: > > > > Hi folks, > > > > > > > > I just updated the status of GCC test suite in r11593. I noticed two > > > > failures > > > > because the lack of /etc/hosts and iana-etc. They can be resolved by > > > > using > > > > the > > > > minimal hosts file from Perl page, and installing iana-etc before GCC. > > > > > > > > The edited book is at > > > > http://www.linuxfromscratch.org/~xry111/LFS-BOOK-rfc/. > > > > > > I saw those updates. Thanks. > > > > > > If you want to dig into gcc, take a look at the 100 gcc errors when > > > valgrind and gdb are installed. Most seem to be related, but I didn't > > > want to try to figure them out. Upstream knows about them as I found > > > them in the test results page. > > > > I just ignore all failures of "guality" tests. > > > > > If test failures are due to the test environment or bugs in the tests > > > themselves, I generally don't think is is worth our time to try to > > > figure out fixes. > > > > Only because I was using some CFLAGS/CXXFLAGS, I spent some time to make > > sure > > these flags didn't cause the failures. I think it's OK to document these > > failures in the book and let the readers to ignore them. So let's keep the > > book > > like now. > > I agree. > > > But, just curious, why did we move shadow for several gcc tests? > > From the change log last August: > > "Move shadow to before gcc so the gcc tests can > use su to run as a non-privileged user."
I mean we moved shadow to run gcc testsuite as a non-privileged user, only in order to fix several failures in libstdc++fs tests... -- Xi Ruoyao <xry...@mengyan1223.wang> School of Aerospace Science and Technology, Xidian University -- http://lists.linuxfromscratch.org/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
