summaryrefslogtreecommitdiff
path: root/libc
AgeCommit message (Collapse)Author
2016-03-31bits/byteswap-common.h: import recent headers from glibcAnthony G. Basile
On systems where uClibc doesn't provide an arch specific byteswap.h, we fall back on bits/byteswap-common.h. However, there is a bug in this header in the __bswap_constant_64(x) macro. If, for example, a double is passed, we get 'invalid operands to binary &' in which we mismatch a 'double' and 'long long unsigned int'. The newer glibc headers fix this and so we import them. Also, since the inclusion of byteswap.h is not appropriate for assembly code, we guard a portion of endian.h which uses byteswap.h from inclusion in any assembly. This is needed, for example, for f2fs-tools 1.6.0 on 32-bit big endian PowerPC. Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
2016-03-31arm: simplify handling of Thumb related optionsThomas Petazzoni
Currently, the Thumb support on ARM has three related Config.in options, which are not trivial for users to understand, and are in fact not needed: - The USE_BX option is not needed: knowing whether BX is available or not is easy. If you have an ARM > v4 or ARMv4T, then BX is available, otherwise it's not. This is the logic used in glibc. - The USE_LDREXSTREX option is not needed: whenever Thumb2 is available, ldrex/strex are available, so we can simply rely on __thumb2__ to determine whether ldrex/strex should be used, without requiring a Config.in option. - Once USE_BX and USE_LDREXSTREX are removed, the only thing left that COMPILE_IN_THUMB does is to set -mthumb. This makes the option unnecessary, as on ARM at least, the user is already supposed to pass -march=<foo> or other compiler options tuning the library for a specific ARM variant. There is no reason to do otherwise for Thumb, which allows to get rid of the COMPILE_IN_THUMB option. Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
2016-03-31xtensa: fix __sigsetjmp call12 caseMax Filippov
Register window saving loop in __sigsetjmp incorrectly calculates offset in the jmp_buf in case it was called with call12. The bug was introduced in commit f4b1b7ade7971141 "xtensa: fix setjmp that didn't save all registers correctly". Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
2016-03-10getaddrinfo: correct AI_V4MAPPED handlingPeter Korsgaard
As recently reported on the Buildroot list: http://lists.busybox.net/pipermail/buildroot/2016-March/155325.html DNS lookups with Node.js currently fails on uClibc-ng. The reason for this is the way AI_V4MAPPED is handled. According to POSIX, AI_V4MAPPED should be ignored unless ai_family is AF_INET6: http://pubs.opengroup.org/onlinepubs/9699919799/functions/freeaddrinfo.html If the AI_V4MAPPED flag is specified along with an ai_family of AF_INET6, then getaddrinfo() shall return IPv4-mapped IPv6 addresses on finding no matching IPv6 addresses (ai_addrlen shall be 16). The AI_V4MAPPED flag shall be ignored unless ai_family equals AF_INET6. uClibc-ng was also handling AI_V4MAPPED for AF_UNSPEC, fix that. Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
2016-03-10DNS: Don't apply search domains to PTR lookupsTim Hockin
This lessens the load on upstream DNS servers (and it was just nonsensical). Signed-off-by: Tim Hockin <thockin@google.com>
2016-03-10DNS: don't count search-path miss as a retryTim Hockin
Currently a miss on a search-path entry is counted as a retry. This means that users with more than (num_nameservers * retries) entries in their search path list fail before trying all search paths. Concretely, a single nameserver with 4 search paths will never try the 4th search because the default retry is 3. The code doesn't currently retry a given nameserver in case of an error, so retries is sort of meaningless (though there are some comments indicating it might come). This change only treats total failure of a nameserver (try next server) as a retry. Signed-off-by: Tim Hockin <thockin@google.com>
2016-03-10ia64: sync with glibc headersWaldemar Brodkorb
2016-02-29arm: fix __JMP_BUF_SP definitionMax Filippov
__JMP_BUF_SP is the index of a stack pointer slot in the __jmp_buf. According to ARM __sigsetjmp code it does not depend on configuration parameter __UCLIBC_HAS_FPU__. Make its definition unconditional. __JMP_BUF_SP is used in the unwind_stop through the _JMPBUF_CFA_UNWINDS_ADJ macro, this change fixes cleanup routines call on thread cancellation in configurations with hard FP. Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
2016-02-24xtensa: fix stack unwinding over __default_sa_restorerMax Filippov
For some xtensa cores generated code gets arranged so that __default_sa_restorer immediately follows preceding function. E.g.: 40a1b: c03340 sub a3, a3, a4 40a1e: f01d retw.n 00040a20 <__default_sa_restorer>: 40a20: e1a022 movi a2, 225 40a23: 005000 syscall ... Contents of the .eh_frame section: ... 000007f4 00000014 000007f8 FDE cie=00000000 pc=0004076c..00040a20 DW_CFA_advance_loc4: 3 to 0004076f DW_CFA_def_cfa_offset: 48 Due to the way xtensa libgcc uw_frame_state_for calls _Unwind_Find_FDE for non-signal frames, FDE for the previous function is found during thread cancellation signal stack unwinding in that case. Signal stack frame is not recognized and is not unwound properly, breaking cleanup routines calling for cancelled thread. Insert padding before the __default_sa_restorer so that no FDE is found for it, MD_FALLBACK_FRAME_STATE_FOR is called by uw_frame_state_for and the frame is correctly recognized as signal frame. Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
2016-02-24Replace /etc/resolv.conf with _PATH_RESCONF to allow portability of the code ↵Ubaldo Porcheddu
on system where resolv.conf is not in /etc . Signed-off-by: Ubaldo Porcheddu <ubaldo@eja.it>
2016-02-24Replaced any occurence of /bin/sh with _PATH_BSHELL to allow easier ↵Ubaldo Porcheddu
portability on system with default shell on a different directory, like for instance on android. Signed-off-by: Ubaldo Porcheddu <ubaldo@eja.it>
2016-02-07frv: resurrect port somehow, totally untestedWaldemar Brodkorb
The FR-V port is really broken, and I have no emulator or hardware for this platform. I tried to get some hardware from RedHat, who made the FR-V port initially. Unfortunately Fujitsi didn't agreed to sent me some of their unused spare hardware lying @RedHat. As I invested some time to get stuff compiled, I decided to add the code and may be anytime later I can gain access to some emulator or hardware. GDB simulator for FR-V doesn't support booting Linux AFAIK.
2016-02-03mips: fix build if threads are disabledWaldemar Brodkorb
Reported-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
2016-01-31mips: fix clashing symbolsWaldemar Brodkorb
ISC bind9 uses ptrsize, better use something else. Reported-By: Gustavo Zacarias <gustavo@zacarias.com.ar> Signed-off-by: Waldemar Brodkorb <wbx@uclibc-ng.org>
2016-01-31mips64: fix memcpy, patch from glibcWaldemar Brodkorb
Signed-off-by: Waldemar Brodkorb <wbx@uclibc-ng.org>
2016-01-31Provide __adjtimex() alias, like glibc.Leonid Lisovskiy
Althought recent versions of sofware don't require it, old ntpd wants - see http://bugs.ntp.org/show_bug.cgi?id=769 Signed-off-by: Leonid Lisovskiy <lly.dev@gmail.com> Signed-off-by: Waldemar Brodkorb <wbx@uclibc-ng.org>
2016-01-31Make sure to always terminate decoded stringWaldemar Brodkorb
Write a terminating '\0' to dest when the first byte of the encoded data is 0. This corner case was previously missed. Signed-off-by: Daniel Fahlgren <daniel@fahlgren.se> Signed-off-by: Waldemar Brodkorb <wbx@uclibc-ng.org>
2016-01-31Do not follow compressed items forever.Waldemar Brodkorb
It is possible to get stuck in an infinite loop when receiving a specially crafted DNS reply. Exit the loop after a number of iteration and consider the packet invalid. Signed-off-by: Daniel Fahlgren <daniel@fahlgren.se> Signed-off-by: Waldemar Brodkorb <wbx@uclibc-ng.org>
2016-01-30add exp10() from glibcWaldemar Brodkorb
Some software make use of exp10, so add it. Basic testing with LTP. Signed-off-by: Waldemar Brodkorb <wbx@uclibc-ng.org>
2016-01-27nios2: allow to build toolchainWaldemar Brodkorb
At least allow to compile a toolchain targeting nios2 without MMU.
2016-01-25hppa: unbreak toolchain buildingWaldemar Brodkorb
At least allow to build a toolchain for hppa. Sync some headers with glibc.
2016-01-12Suppress warning "_GNU_SOURCE" redefinedWaldemar Brodkorb
libc/sysdeps/linux/common/euidaccess.c:1:1: warning: "_GNU_SOURCE" redefined In file included from <command-line>:0: ./include/libc-symbols.h:52:1: warning: this is the location of the previous definition Signed-Off-By: Leonid Lisovskiy <lly.dev@gmail.com>
2016-01-10h8300: O_DIRECT and O_DIRECTORY swapping.Yoshinori Sato
Different define of latest kernel. kernel define is #define O_DIRECT 00040000 /* direct disk access hint */ #define O_DIRECTORY 00200000 /* must be a directory */ Yoshinori Sato <ysato@users.sourceforge.jp>
2016-01-10order of special checks mattersWaldemar Brodkorb
The order of special checks seems critical for some applications. Xorg 1.18.0 fails to start with XNFreallocarray error. Took me some time to run with MALLOC_DEBUG=2 to find out. MALLOC_STANDARD is not affected.
2016-01-10Add implementation for copysignl for i386Waldemar Brodkorb
Patch from OpenWrt. Reported-by: Leonid Lisovskiy <lly.dev@gmail.com>
2016-01-10Provide XTI __t_scalar_t types, like glibcWaldemar Brodkorb
Reported-by: Leonid Lisovskiy <lly.dev@gmail.com>
2016-01-06Remove alias.Yoshinori Sato
open_cancel and open2_cancel have diffrent argument on open. So can't alias this functions. Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
2016-01-06openat argument fix.Yoshinori Sato
If argument passing to register case (ex. -mregparam=3). This case set all parameters set to register from caller. But callee refer to stack. So can't get parameter. Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
2016-01-06pthread_atfork handlers not removed during dlcloseWaldemar Brodkorb
Invoke pthread_atfork handler cleanup when removing the associated DSO... If a program loads a DSO (dlopen) that sets up a pthread_atfork handler(s), and then subsequently closes the DSO, the handler(s) are left in place. If fork() is subsequently called, the handlers are invoked even though the DSO has been removed causing crashes or unpredictable code execution. This is because the code in __cxa_finalize(atexit.c)to invoke the unregister_atfork() routine is ifdef'd out with the comment that it hasn't been "looked into this yet...". Refs.: http://bugs.busybox.net/show_bug.cgi?id=8211 http://sourceware.org/bugzilla/show_bug.cgi?id=13502 Add test-case, enable cleanup for NPTL only. Signed-off-by: John Ata <john.ata@baesystems.com> Signed-off-by: Leonid Lisovskiy <lly.dev@gmail.com>
2016-01-06__ARCH_HAS_DEPRECATED_SYSCALLS__ case fix. getdents is deprecated.Yoshinori Sato
Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
2016-01-06h8300: vfork flags update.Yoshinori Sato
Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
2016-01-06extra/Configs/Config.in: add syncfs() to Linux-specific functionsAnthony G. Basile
syncfs() was recently added (commit dfa593d4d). But man sync(2) specifies that syncfs() is Linux-specific. This was overlooked in the original commit so we add it to the set of Linux-specific functions supported by uClibc. Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
2016-01-03libc: getpass,getutent: allocate buffer dynamicallyWaldemar Brodkorb
Saves 0.6k bss with default buffer size(256). text data bss dec hex filename - 1172 8 408 1588 634 libc/misc/utmp/utent.os - 429 0 256 685 2ad libc/unistd/getpass.os + 1208 8 28 1244 4dc libc/misc/utmp/utent.os + 471 0 4 475 1db libc/unistd/getpass.os ================================================================ +78 -632 Signed-off-by: Leonid Lisovskiy <lly.dev@gmail.com>
2016-01-02libc/inet: Unbreak gethostent()Waldemar Brodkorb
Although gethostent() is obsoleted, there is no reason to keep it broken. Fix two problems: * commit f65e66078b "resolver: switch to config parser" leave an extra break statement in case of GETHOSTENT in __read_etc_hosts_r. In result, output buffer wasn't initialized at all. * gethostent static buffer has insufficient size to store aliases, so __read_etc_hosts_r always returns ERANGE. Restore ALIAS_DIM define. Add test-case. Signed-off-by: Leonid Lisovskiy <lly.dev@gmail.com>
2016-01-02Use dynamic buffers for gethostent/gethostbyname/gethostbyaddrWaldemar Brodkorb
Save ~1k static space (.bss) text data bss dec hex filename - 68 0 126 194 c2 libc/inet/gethostent.os - 79 0 460 539 21b libc/inet/gethostbyname2.os - 83 0 460 543 21f libc/inet/gethostbyaddr.os + 98 0 24 122 7a libc/inet/gethostent.os + 110 0 24 134 86 libc/inet/gethostbyname2.os + 113 0 24 137 89 libc/inet/gethostbyaddr.os ================================================================== +91 -974 Signed-off-by: Leonid Lisovskiy <lly.dev@gmail.com>
2016-01-02libc/pwd_grp: Allocate buffers for non-reentrant functions dynamicallyWaldemar Brodkorb
Saves 3k bss with default buffer size(256). text data bss dec hex filename - 61 0 272 333 14d fgetgrent.os - 61 0 284 345 159 fgetpwent.os - 61 0 292 353 161 fgetspent.os - 57 0 272 329 149 getgrent.os - 61 0 272 333 14d getgrgid.os - 61 0 272 333 14d getgrnam.os - 57 0 284 341 155 getpwent.os - 61 0 284 345 159 getpwnam.os - 61 0 284 345 159 getpwuid.os - 57 0 292 349 15d getspent.os - 61 0 292 353 161 getspnam.os - 61 0 292 353 161 sgetspent.os + 94 0 20 114 72 fgetgrent.os + 94 0 32 126 7e fgetpwent.os + 94 0 40 134 86 fgetspent.os + 87 0 20 107 6b getgrent.os + 94 0 20 114 72 getgrgid.os + 94 0 20 114 72 getgrnam.os + 87 0 32 119 77 getpwent.os + 94 0 32 126 7e getpwnam.os + 94 0 32 126 7e getpwuid.os + 87 0 40 127 7f getspent.os + 94 0 40 134 86 getspnam.os + 94 0 40 134 86 sgetspent.os ===================================================== +387 -3024 Signed-off-by: Leonid Lisovskiy <lly.dev@gmail.com>
2016-01-02lm32: fix testsuite compile problemsWaldemar Brodkorb
One test fails to compile with a gcc ICE, so disable it for now.
2016-01-02lm32: add missing typesWaldemar Brodkorb
2016-01-02remove attribute hidden, as the function is used in linuxthreads.oldWaldemar Brodkorb
On avr32/cris the build with MALLOC fails, when compiling linuxthreads.
2016-01-02remove deprecated architectures v850/i960Waldemar Brodkorb
Both architectures are more or less deprecated. No Linux upstream support, no gcc support for uClinux.
2016-01-02inet/getaddrinfo: simplest /etc/gai.conf to control IPv6/IPv4 addresses sort ↵Leonid Lisovskiy
order Implement simplest variant of /etc/gai.conf to control getaddrinfo IPv6/IPv4 addresses sorting. Keep the default sort order - IPv6 first, IPv4 second. To invert it, create /etc/gai.conf containing single line: precedence ::ffff:0:0/96 100 Example before: $ nslookup security.debian.org 8.8.8.8 Server: 8.8.8.8 Address 1: 8.8.8.8 google-public-dns-a.google.com Name: security.debian.org Address 1: 2001:a78:5:0:216:35ff:fe7f:be4f villa.debian.org Address 2: 2001:a78:5:1:216:35ff:fe7f:6ceb lobos.debian.org Address 3: 195.20.242.89 wieck.debian.org Address 4: 212.211.132.250 lobos.debian.org Address 5: 212.211.132.32 villa.debian.org After patch & precedence set in /etc/gai.conf: $ nslookup security.debian.org 8.8.8.8 Server: 8.8.8.8 Address 1: 8.8.8.8 google-public-dns-a.google.com Name: security.debian.org Address 1: 195.20.242.89 wieck.debian.org Address 2: 212.211.132.250 lobos.debian.org Address 3: 212.211.132.32 villa.debian.org Address 4: 2001:a78:5:0:216:35ff:fe7f:be4f villa.debian.org Address 5: 2001:a78:5:1:216:35ff:fe7f:6ceb lobos.debian.org bloat-o-meter report: function old new delta getaddrinfo 726 1138 +412 gaih_inet 2660 2692 +32 .rodata 16618 16643 +25 __gai_precedence - 1 +1 ------------------------------------------------------------------------------ (add/remove: 1/0 grow/shrink: 4/0 up/down: 882/0) Total: 470 bytes Signed-off-by: Leonid Lisovskiy <lly.dev@gmail.com>
2016-01-01Add argp implementationSalvatore Cro
Argp is an advanced support for parsing unix-style argument vectors. In addition to the common getopt interface, it provides automatic response to `--help' and `--version' options and use of custom parser in conjunction with argp native option parser, among others. Argp support is required by elfutils package and prelink. In uClibc argp functionalities has been moved from C library to libuargp.so Further the libc.so linker script contains an AS_NEEDED entry so that it doesn't need to link libuargp.so explicitely. Disable argp test if feature disabled. Signed-off-by: Salvatore Cro <salvatore.cro@st.com> Signed-off-by: Filippo Arcidiacono <filippo.arcidiacono@st.com> Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com> Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Waldemar Brodkorb <wbx@uclibc-ng.org>
2016-01-01Fix "pselect: Use linux pselect6 syscall when available"Leonid Lisovskiy
Commit e3c3bf2b58 introduce use of pselect6, but has following disadvantages: * Use of userspace types in args67 structure - it breaks, for example, configs when 32-bit uClibc-ng compiled against 64-bit kernel. Syscall will always return EINVAL. We must use __kernel_* types and __SYSCALL_SIGSET_T_SIZE. * It have excess checks for NSEC_PER_SEC. Original code from select() implementation has struct timeval => struct timespec conversion, kernel select() syscall implementation do the same. But none of libc versions (glibc, eglibc, musl) I know, perform similar checks for pselect() - there is no structure fields conversions, just struct timespec through all the calls. To have such checks in uClibc-ng we need one example, at least. * It is possible to avoid extra userspace reads from kernel code if sigmask == NULL. I suggest to do it, for a few bytes cost. * Commit didn't add test case to testsuite. Signed-off-by: Leonid Lisovskiy <lly.dev@gmail.com>
2015-12-29xtensa: fix vfork return address calculationsMax Filippov
vfork internally unwinds stack up one frame, saving the original return address in the a3 of the unwound frame. To do this in windowed ABI it needs to exchange two topmost bits of the original return address and of the helper entry point. Current code doesn't do it correctly for addresses above 0x40000000, resulting in illegal instruction exception for configurations that put code high, e.g. for noMMU. Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
2015-12-22Replace MIPS specific memcpy.S/memset.S with version from glibc/newlib.Steve Ellcey
These MIPS specific versions of memcpy.S and memset.S are faster than the current ones and match what is in newlib and glibc. They also have support for the mips32r6 and mips64r6 architectures. Signed-off-by: Steve Ellcey <sellcey@imgtec.com>
2015-12-22pselect: Use linux pselect syscall when availableNicolas S. Dade
Linux has a pselect syscall since 2.6.something. Using it rather than emulating it with sigprocmask+select+sigprocmask is smaller code, and works properly. (The emulation has race conditions when unblocked signals arrive before or after the select) The tv.nsec >= 1E9 handling comes from uclibc's linux select() implementation, which itself uses pselect() internally if the pselect syscall exists. I though it would be good to do the same here. Note that although the libc pselect() API has 6 arguments, the linux kernel syscall as 7 arguments. There is an extra, somewhat vestigial, sizeof the signal mask argument. Signed-off-by: Nicolas S. Dade <nic.dade@gmail.com>
2015-12-22libc/pwd_grp: Create template for non-reentrant functionsWaldemar Brodkorb
Avoid a lot of copy'n'paste code, no functionality change Signed-off-by: Leonid Lisovskiy <lly.dev@gmail.com>
2015-12-22Use gcc's __builtin_mempcpy() as __mempcpy(), if possibleWaldemar Brodkorb
Patch from OpenWrt. Reported-by: Leonid Lisovskiy <lly.dev@gmail.com>
2015-12-22Do define some IPv6 related symbols even with IPv6 disabledWaldemar Brodkorb
Fixes compilation of some faulty applictions. Patch from Freetz.org Reported-by: Leonid Lisovskiy <lly.dev@gmail.com>
2015-12-22resolv: fix gethostbyname2_r to match gethostbyname_r, fixing bugs with AAAA ↵Waldemar Brodkorb
lookups The latter half of gethostbyname2_r (doing AAAA queries) is rather dramatically different from the corresponding portion of gethostbyname_r (doing A queries). This leads to problems like calls to getaddrinfo only returning one IPv6 address, even when multiple exist. Seems to be entirely a case of divergent evolution -- a half-decade of fixes for the IPv4 code but no love for IPv6. Until now. ;) DNS behaviour for IPv6 is really no different than for IPv4 -- beyond the difference in address sizes, there's no need for the functions to be so different. Consequently, this patch really is almost just a cut-and-paste of gethostbyname_r, with the appropriate substitutions of in6_addr, AF_INET6, etc; while holding on to the few extra bits that actually belong in there (eg #ifdef __UCLIBC_HAS_IPV6__). Signed-off-by: Wes Campaigne <westacular@gmail.com>