The 2.16 release is tagged with tag glibc-2.16.0. A mistake was made tagging glibc-2.16 and a second tag glibc-2.16.0 had to be created.
The 2.16 release branch is release/2.16/master.
Linux-based configurations unless indicated.
- x86, Builds, No testsuite failures. (By AJ: Linux 3.4, GCC 4.7 using -m32, binutils-2.22)
- x86 GNU Hurd, doesn't build without patches
- x86-64, Builds, No testsuite failures. (By AJ: Linux 3.4, GCC 4.7, binutils-2.22)
- x32, Builds, No testsuite failures (By H.J. Lu: Linux 3.4, GCC 4.7 using -mx32, binutils-18.104.22.168.4)
- ARM, Builds, No new testsuite failures.
- Power, Builds, No new testsuite failures.
- Power64, Builds, No new testsuite failures.
- s390 (32-bit), Builds, No new testsuite failures (By AJ: Linux 2.6.32, GCC 4.7, binutils 2.22) (By AK: Linux 2.6.35, GCC 4.5.3, binutils 22.214.171.12420502)
- s390x (64-bit), Builds, No new testsuite failures (By AJ: Linux 2.6.32, GCC 4.7, binutils 2.22) (By AK: Linux 3.3.4, GCC 4.7, binutils 126.96.36.19920622)
- MIPS, Builds, No new testsuite failures.
- sparc, Builds, No testsuite failures.
- sparc64, Builds, No testsuite failures.
- Alpha, Builds, Some testsuite failures under qemu; no native testing yet.
- IA64, Builds, No testsuite regressions. (Linux: 3.1.6, GCC 4.5, binutils-2.21.1)
- HP-PARISC, Does not build, Some testsuite failures.
- SH, Builds, No testsuite regressions. (Linux: 2.6.30, GCC 4.6.3, binutils-2.22)
- M68K, builds, some math testsuite failures; no native testing
- tilepro/tilegx, Builds, Some testsuite failures listed below. (Linux: 2.6.38, GCC 4.4.6, binutils-2.21)
- Some math failures
- makecontext3 doesn't scale like other ports
- tst-getcpu fails due to kernel bug
On systems with VFPv3 or VFPv4 (not VFPv2, VFPv3U or VFPv4U), math/test-fenv fails; see libc-ports message.
- On systems not using VFP, math/test-fenv, math/test-misc, math/test-fpucw, math/bug-nextafter, math/bug-nexttoward, math/test-float, math/test-double, math/test-ifloat and math/test-idouble fail (the lack of support for floating point exceptions and rounding modes is determined at run time, and while tests generally check for errors from fesetround they do not handle a run-time lack of exceptions being raised).
- On MIPS64 (n32 and n64), long double is implemented using fp-bit without support for exceptions and rounding modes, which causes the following tests to fail: math/test-misc, math/bug-nextafter, math/bug-nexttoward, math/test-ldouble, math/test-ildoubl.
- On soft-float MIPS systems, there is no support for floating-point exceptions and rounding modes, causing the following tests to fail: math/test-fenv, math/test-misc, math/bug-nextafter, math/bug-nexttoward, math/test-float, math/test-double, math/test-ifloat, math/test-idouble.
- MIPS GCC does not use PT_GNU_STACK markers (this is a GCC issue), causing elf/check-execstack to fail (the relevant kernel support went in around February 2010).
Peculiarities of MIPS ELF cause elf/tst-audit1, elf/tst-audit2 and elf/check-localplt to fail. For the first two, see bug 2981.
tile has issues with TLS on dlopen()'ed shared objects; see bug 14237.
tile doesn't handle -fexceptions correctly for pthread_cancel(); see bug 14266.
_mcount symbol missing at version GLIBC_2.0; see bug 14042
- Intermittent/known test-suite failure with rt/tst-cpuclock2 due to timing/granularity.
- Known test-suite failures in stdlib/tst-setcontext, stdlib/tst-makecontext3, and tst-setcontext-fpscr due to kernel bug.
S/390 System z
- nptl/tst-eintr1 fails from time to time with: "Expected signal 'Alarm clock' from child, got none"
- Known test-suite failures in rt/tst-cputimer1, rt/tst-cpuclock2, due to kernel bug.
On systems with limited memory, stdio-common/test-vfprintf and stdio-common/bug22 may fail (bug 14231).
On some systems, a race condition may cause nptl/tst-cancel7 and nptl/tst-cancelx7 to fail (bug 14232).
nptl/tst-getpid2 is an incorrect test (bug 11214).
Race condition on timezone/tst-timezone (bug 14096).
Gnulib Testsuite failures
As part of the 2.16 pre-release we ran the gnulib testsuite for x86_64 to look for problems.
Four issues were detected in the gnulib testsuite run, three were testsuite issues, and one is a known issue.
- remove.o - gnulib was unconditionally using it's own copy of remove. Gnulib needs fixing.
logf.o - gnulib didn't use -lm during build. Gnulib testsuite bug.
log10f.o - gnulib didn't use -lm during build. Gnulib testsuite bug.
strstr.o - Known issue: http://sourceware.org/bugzilla/show_bug.cgi?id=12100
In summary the gnulib testing did not detect any new problems, but did alert the release manager to known issue BZ#12100 which we hope to fix in 2.17.