I've been reading through LFS while following the instructions till I got to the point where I needed to compile glibc-2.25 for the actual system.
After running make check, have encountered the following failures:
FAIL: nptl/tst-cond17
FAIL: posix/tst-getaddrinfo4
FAIL: posix/tst-getaddrinfo5
Summary of test results:
3 FAIL
2640 PASS
26 UNSUPPORTED
43 XFAIL
2 XPASS
make[1]: *** [Makefile:355: tests] Error 1
make[1]: Leaving directory '/sources/glibc-2.25'
make: *** [Makefile:9: check] Error 2
Both posix/tst-getaddrinfo4 and posix/tst-getaddrinfo5 failures should pose no real threat as indicated by LFS, but I am not sure about the first failure nptl/tst-cond17.
I have checked the source file and found out that all it does is defining some sort of variable. Here's the code.
#define UNLOCK_AFTER_BROADCAST 1
#include "tst-cond16.c"
Is it not critical to the build process? or should I try to fix it somehow?
EDIT:
The files nptl/tst-cond17.o, nptl/tst-cond17.o.d and nptl/tst-cond17.out are empty, while the contents of the file nptl/tst-cond17.test-result are:
FAIL: nptl/tst-cond17
original exit status 127
I have checked our records, and tst-cond17 is not generally known to generate spurious failures (or as being affected by unfixed kernel bugs). I found a reference to a tst-cond17 failure in the glibc 2.20 release notes, but the submitter comments that, ‘The NPTL failures not mentioned as architecture-independent are thought to result from general unreliability of the board being used for testing.’, so I assume that this does not count.
I would say the tst-cond17 failure is worth investigating further, especially if you can reproduce it.