On 13 November 2015 at 16:31, Fernando Seiti Furusato <ferse...@br.ibm.com> wrote: > Hi Felipe. Thanks for the quick action. > > Unfortunately, the package did not build successfully on buildd, but it is > a problem with a binary file. > Specifically the binary file called "no-rpath", within "tests" directory. > So the test suite failed. > > It is shipped with the source pkg, but it differs from upstream, which > actually works. > > Unlike upstream, there is not a no-rpath.c to compile either. > Is there anything you can do to fix this?
Hmm, I just checked upstream master branch on the ppc64el porterbox and it fails on 5 tests: FAIL: no-rpath-armel.sh FAIL: no-rpath-armhf.sh FAIL: no-rpath-hurd-i386.sh FAIL: no-rpath-i386.sh FAIL: no-rpath-kfreebsd-i386.sh Interestingly, all the tests that fail are 32bit archs (but some 32bit archs survive, like mipsel or powerpc). The test shows the following output: first reserved offset/addr is 0x26e/0x804826e And then complains that those values % pageSize are not equal. It dawns on me that one should not expect a binary for other architecture to have offsets consistent with our own pageSize, but rather the other kernel's pagesize, right? -- Saludos, Felipe Sateler
