summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2014-12-28update INSTALLWaldemar Brodkorb
2014-12-28revert getopt changes to testsWaldemar Brodkorb
The upstream changes are not complete and let the tests fail. Revert the changes for now, may be upstream will make a better fix later.
2014-12-28for mips64 with N32 ABI we need to use newfstatatWaldemar Brodkorb
Fixes compile errors with latest util-linux.
2014-12-28relocation fixesWaldemar Brodkorb
From OpenWrt: https://dev.openwrt.org/browser/trunk/toolchain/uClibc/patches-0.9.33.2/613-mips64_more_relocation_fixes.patch
2014-12-28sparc64 support was removed recentlyWaldemar Brodkorb
Remove leftover code.
2014-12-26don't optimize for special mips cpuWaldemar Brodkorb
When building optimized code for Lemote Yeelong system, a conflict occurs. Better use optimized flags in your buildsystem, not in uClibc-ng.
2014-12-25merge upstream changesWaldemar Brodkorb
2014-12-25sync tcp header with glibcWaldemar Brodkorb
Nginx tries to find TCP_FASTOPEN, which is missing here.
2014-12-15mkostemp: fix implementationAnthony G. Basile
mkostemp(char *template, int flags) generates a unique temporary filename from a template. The flags parameter accepts three of the same flags as open(2): O_APPEND, O_CLOEXEC, and O_SYNC. The current implementation of mkostemp(3) does not respect the flags and in fact confuses the flags with the file mode which should always be S_IRUSR | S_IWUSR. This patch corrects this issue. Signed-off-by: Anthony G. Basile <blueness@gentoo.org> Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-12-15test: Some more tests under conditionalsBernhard Reutner-Fischer
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-12-15stdio: Fix printing 0.0Bernhard Reutner-Fischer
We were relying on FE_DIVBYZERO being turned off when printing "%f", +-.0 Avoid the whole issue by looking at the sign-bit (in a rough approximation). Note that we do not handle gracefully: printf ("\n%llf\n", -0.0); printf ("\n%llf\n", 0.0); nor %Lf for both when NOT cast to long double. Avoiding an FPE due to broken numbers like these does not make sense to me. Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-12-12unistd: allow to turn off getopt_longBernhard Reutner-Fischer
The GNU variant of getopt() previously had no way to turn off getopt_long() support.
2014-12-12test: disable ptytest unless HAS_PTYBernhard Reutner-Fischer
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-12-12config: move STDIO_FUTEXES up a bitBernhard Reutner-Fischer
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-12-10disable test for noMMUWaldemar Brodkorb
2014-12-10mkostemp: fix implementationAnthony G. Basile
mkostemp(char *template, int flags) generates a unique temporary filename from a template. The flags parameter accepts three of the same flags as open(2): O_APPEND, O_CLOEXEC, and O_SYNC. The current implementation of mkostemp(3) does not respect the flags and in fact confuses the flags with the file mode which should always be S_IRUSR | S_IWUSR. This patch corrects this issue. Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
2014-12-10add argument check in mknodwangyufen
mknod() in glibc/eglibc will check the argument, like this, ... if (k_dev != dev) { __set_errno (EINVAL); return -1; } ... So add argument check in uclibc's mknod() too. Signed-off-by: Wang Yufen <wangyufen@huawei.com>
2014-12-09Merge remote-tracking branch 'origin/upstream'Waldemar Brodkorb
2014-12-09sigaction: Fix !HAS_REALTIME compilationBernhard Reutner-Fischer
In file included from ./include/signal.h:279:0, from libc/sysdeps/linux/x86_64/sigaction.c:22: ./include/bits/sigaction.h:32:29: error: unknown type name 'siginfo_t' void (*sa_sigaction)(int, siginfo_t *, void *); Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-12-08remove text files not useful for uClibc-ngWaldemar Brodkorb
2014-12-06arm: add RESET_PID in the clone implWang Yufen
Called getpid() When creating a new process with clone(), getpid() returns the father_process's value. It should be child_process's value. The reason is missing a RESET_PID in the arm clone impl. Signed-off-by: Wang Yufen <wangyufen@huawei.com> Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
2014-12-04Merge remote-tracking branch 'origin/upstream'Waldemar Brodkorb
2014-12-03mips: rename siginfo _timer membersBernhard Reutner-Fischer
Rename _timer[12] to si_tid and si_overrun to fix compilation of strace-4.9+ Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-12-03buildsys: remove some obsolete optimization CFLAGSBernhard Reutner-Fischer
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-12-01Merge remote-tracking branch 'origin/upstream'Waldemar Brodkorb
2014-11-19locale: Add wcsftime()Bernhard Reutner-Fischer
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-11-13Merge remote-tracking branch 'origin/upstream'Waldemar Brodkorb
2014-11-13nptl: compile pt-vfork in ARM-modeBernhard Reutner-Fischer
could need a thumb implementation.. Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-11-11Merge remote-tracking branch 'origin/upstream'Waldemar Brodkorb
2014-11-10Config.in.arch: enable long double math for xtensaMax Filippov
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com> Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-11-10statfs: fix compile error when UCLIBC_LINUX_SPECIFIC is not setAnthony G. Basile
statfs() is a Linux-specific function. When building without UCLIBC_LINUX_SPECIFIC set, libc_hidden_proto(statfs) in include/sys/statfs.h is unmatched with libc_hidden_def(statfs) in libc/sysdeps/linux/common/statfs.c, leading to a compile error. This patch fixes this, as well as adds statfs() to the list of Linux-specific functions in extra/Configs/Config.in. Signed-off-by: Anthony G. Basile <blueness@gentoo.org> Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
2014-10-14reduce the list of supported architecturesWaldemar Brodkorb
As most of these architectures either does not compile or embedded hardware is difficult to get, I disable them for now. I do not remove the existing code and hope to enable many of them in the future, but not for 1.0.
2014-10-11m68k: remove reg.hWaldemar Brodkorb
If reg.h exist, gdbserver fails to compile. I am not sure if this header file is needed for anything.
2014-10-10c6x: disable it at the momentWaldemar Brodkorb
There is a NPTL port of uCLibc for c6x, but it doesn't work with uClibc master. May be someone with real hardware could sync it. Unfortunately I got no respone from Aurelien and no response on the mailing list.
2014-10-08test: disable for noMMU caseWaldemar Brodkorb
Pthread tests are failing for coldfire with linuxthreads enabled.
2014-10-05disable a test for avr32, compile failure.Waldemar Brodkorb
2014-10-05disable some test for avr32, compile failures.Waldemar Brodkorb
2014-10-03linuxthreads: fix compile error for non-MMUWaldemar Brodkorb
Otherwise you get following linking error, because of missing functions in libc.a: undefined reference to `__libc_enable_asynccancel' Fix by Thorsten Glaser while hacking session.
2014-10-03m68k: fix linuxthreads compile for coldfireWaldemar Brodkorb
The tas instruction is not available for most coldfire CPU's. Use bset instead in this case as already used in linuxthreads.old.
2014-10-03disable linkage of gcc_ehWaldemar Brodkorb
As recently discussed on the uClibc mailing list here: http://lists.uclibc.org/pipermail/uclibc/2014-September/048659.html I think it is not required for gcc 4.8.3, which is default in f.e. OpenADK. Tested with a DODEBUG build for x86.
2014-10-01remove forced gcc optimizationWaldemar Brodkorb
It was added in 6d6bd8ba78434ecb09395582b5f3e41febd4d4ee, but it is unclear for me, why this is needed or if it is required. I don't think we should depend on some gcc optimization, which might change in the future. Anyway, this breaks c6x toolchain building (ICE), so I remove it. Testsuite run for supported architectures didn't add any new errors.
2014-09-26libpthread: disable unimplemented configurationsWaldemar Brodkorb
When NPTL is available for an architecture just allow the use of it. If NPTL is not available for an architecture, allow to choose between LT and LT.old. This minimizes misconfiguration of threading library.
2014-09-26enable this symbol by defaultWaldemar Brodkorb
disabling this results in non-working toolchain building, so better do not allow anyone to disable it.
2014-09-26Add eventfd_read() and eventfd_write()Waldemar Brodkorb
Signed-off-by: Khem Raj <raj.khem@gmail.com>
2014-09-26Do not define unimplemented functionsCristian Morales Vega
e.g. fminf() is not implemented, only fmin(), but both are defined. Signed-off-by: Cristian Morales Vega <cristian@samknows.com>
2014-09-26cris: remove call to HIDDEN_JUMPTARGETWaldemar Brodkorb
errno_location is no longer hidden, so remove the call to the macro, like done for sparc recently.
2014-09-26cris: do not include asm/elf.hWaldemar Brodkorb
elf.h is not exported by the Linux kernel, do not include it here. Done like in other architectures.
2014-09-23testsuite: fix xtensa macrosWaldemar Brodkorb
2014-09-23add tls test macros for xtensaWaldemar Brodkorb
http://lists.linux-xtensa.org/pipermail/linux-xtensa/Week-of-Mon-20130819/001153.html
2014-09-22xtensa: add support for NPTLWaldemar Brodkorb
Changes from: https://github.com/foss-xtensa/uClibc/commits/xtensa_nptl Author: Chris Zankel <chris@zankel.net> Author: Baruch Siach <baruch@tkos.co.il>