| Age | Commit message (Collapse) | Author |
|---|
|
|
|
The recvmmsg and sendmmsg is very important for UDP stream application.
If we only use recvmsg for UDP stream, it will only copy one mtu size
of data in a syscall. And recvmmsg copy as many as you want in a syscall.
So recvmmsg is more efficient,and some applications will depends on the
recvmmsg and sendmmsg, eg: UDP media stream player.
Signed-off-by: Guo Ren <ren_guo@c-sky.com>
|
|
In the test-case of statfs from ltp-testsuite, it pass -1 to 2th
argument. eg: fstatfs(fd , -1)
When uclibc-ng parse the buf32 to buf will cause illegal address
access, the kernel will signal the process with SIGSEGV.
If we pass the -1 directly to the syscall of statfs/fstatfs, kernel
use copy_to_user() to prevent the signal of SIGSEGV and just return
EINVAL.
This is the ltp-testsuite expect.
Signed-off-by: Guo Ren <ren_guo@c-sky.com>
|
|
The current uclibc-ng use 4 arguments, and this will cause
ltp-testsuite's preadv/pwritev case failed.
The syscall of preadv/pwritev in current linux-kernel is 5 arguments:
linux/fs/read_write.c:
SYSCALL_DEFINE5(preadv, unsigned long, fd, const struct iovec __user *, vec,
unsigned long, vlen, unsigned long, pos_l, unsigned long, pos_h)
SYSCALL_DEFINE5(pwritev, unsigned long, fd, const struct iovec __user *, vec,
unsigned long, vlen, unsigned long, pos_l, unsigned long, pos_h)
So just update to 5-args-syscall, and off_t could be 32bit or 64bit.
Signed-off-by: Guo Ren <ren_guo@c-sky.com>
|
|
Reported-by: Lauri Kasanen <cand@gmx.com>
|
|
An issue has been found with current implementation of signal restorer
function in uClibc and how GDB handles it. When debugger information is not
present, everything worked fine, because GDB would use a built-in logic to
determine if function is a signal restorer. However when debugging information
is present, debugger would rely solely on it and wouldn't use ARC-specific
functions to detect signal handler frames. Because debug information for signal restorer is generated
completely by the compiler, it lacks a marker, that identifies this as a
signal frame that requires special handling. While it is possible to insert
that marker via inline assembly, that still doesn't solve the whole problem,
because some other expectations are not met by the debug information - there
is no "nop" in front of the function, needed to fool debugger into thinking
that this was a function call, and references to previous frame information
need to be described manually. The simplest way to fix the problem is just
to make sure that signal restorer function will not have any debug function
at all, which can be done by writing it in assembly.
Alternative, more complex, solution, where debug information for signal
frame is manually defined can be found in
glibc/sysdeps/unix/sysv/linux/x86_64/sigaction.c [1].
[1] https://sourceware.org/git/?p=glibc.git;a=blob;f=sysdeps/unix/sysv/linux/x86_64/sigaction.c;hb=HEAD
Signed-off-by: Anton Kolesov <Anton.Kolesov@synopsys.com>
|
|
In cde74b83f9b2 "ARC: remove special CFLAGS/LDFLAGS handling" we
got rid of CONFIG_ARC_CPU_HS which was used to select ARCv2-specific
implementation of optimized string routines. So now ARCv2-tuned
memset/memcpy/strcmp are not used, instead those for ARC700 used for
both ARC700 and ARCHS.
Without uClibc config option we may only tell which CPU type we're
targeting by built-in defines of GCC. I.e. no more conditional file
inclusion in Makefiles. That leaves us only one option - merge both
implementations in 1 file and use ifdefs.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
|
|
Ensure to always allow a build and fallback to C.UTF-8
and even C.
Signed-off-by: Eugene Yudin <e.yudin@ndmsystems.com>
|
|
Pthread pid caching was removed recently but an assert is still
present which checks pthread->pid, and this breaks the build when
debugging is enabled.
Reported-By: Bogdan Harjoc <harjoc@gmail.com>
|
|
Remove enum __ptrace_flags along with the only constant it contains,
PTRACE_SEIZE_DEVEL, from Linux's sys/ptrace.h files.
Following GNU C library commit:
60e2846e2633a990bdf474004a373bde54c0bc5f
|
|
Latest binutils 2.29 release emits a R_MICROBLAZE_NONE
relocation, which breaks shared library loader bootstrap
relocation.
|
|
GDB 8.0 is compiled and linked with g++, but the
linking of static targets (f.e. coldfire) fails,
without declaring the functions in thread_db.h
extern C.
The compilation of gdb errors out with:
thread-db.o: In function `thread_db_init()':
thread-db.c:(.text+0x5b6): undefined reference to `td_ta_new(ps_prochandle*, td_thragent**)'
thread-db.c:(.text+0x61e): undefined reference to `td_thr_get_info(td_thrhandle const*, td_thrinfo*)'
thread-db.c:(.text+0x632): undefined reference to `td_symbol_list()'
..
|
|
|
|
|
|
The function towlower doesn't work with locales different from C.
Issue was introduced in commit: 8cde3a9bf2856dcb9a759dec7ecb04a68e712254
Call to setlocale is needed for correct generation of the table uplow_diff.
Otherwise you receive compile time error "range assumption error" after
uncommenting the call.
Similar problem described here:
http://lists.uclibc.org/pipermail/uclibc/2015-March/048852.html
This commit fix the problem by using int32_t values.
|
|
|
|
|
|
Similar to a changeset planned in GNU C library remove
any assembly code from sysdep-cancel.h. This allows
to port to new architectures in a simpler way.
|
|
|
|
|
|
This reverts commit 9b42da7d0558884e2a3cc9a8674ccfc752369610.
This fixes a long standing bug in dlclose().
Can be seen with apache and mod_php or just by executing
php -m to list all modules.
Reported-by: Bernd Kuhls <bernd.kuhls@t-online.de>
|
|
Fix compile of latest iproute2 package.
Reported-By: Baruch Siach <baruch@tkos.co.il>
|
|
|
|
Following glibc commit a358c805300e358e30d4788a6f19c69988623a5c
|
|
|
|
No NPTL, no LDSO support.
Bootup with Busybox Ash in Qemu working.
Testuite shows only two failures, but mksh continue/break
support doesn't work.
|
|
Remove __FAVOR_BSD and sync with GNU C library
Some issues compiling knock application fixed.
Reported-By: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
|
|
|
|
|
|
|
|
|
|
Sync with GNU C library. Found while trying to compile
linux-rdma to uClibc-ng.
Reported-by: Alexey Brodkin <Alexey.Brodkin@synopsys.com>
|
|
All flavors of mkstemp create files with mode S_IRUSR | S_IWUSR, as
per POSIX.1-2008. Make mkostemp64 follow that too instead of creating
files with mode S_IRUSR | S_IWUSR | S_IXUSR.
Signed-off-by: Ignacy Gawędzki <ignacy.gawedzki@green-communications.fr>
|
|
Fix compile issues with uchar.h users, f.e. systemd.
|
|
In commit 4b7634a5e03b0da6f8875de9d3f74c1cf6f2a6e8 non-null
check was removed. Fixes a compiler warning when trying
to compile systemd.
|
|
Fixed in glibc commit f9cfa295ae3f2556bd8808f0ff693cfe44f4ac25.
Found via Buildroot autobuilder and gdbm package cross-compile.
|
|
|
|
|
|
|
|
man feature_test_macros(7) specifies that _BSD_SOURCE and _SVID_SOURCE have
been deprecated in favor of _DEFAULT_SOURCE since libc 2.20. Specifying
either of the former is now equivalent to specifying just the latter. We add
this macro to conform to this standard, but do not add the compiler warning
to maintain full backwards compatibility with earlier version of glibc and
uclibc.
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
|
|
Remove a lot of unused 64 Bit header stuff.
|
|
This syncs up <sys/reboot.h> with the GNU C library, see their
commit 1a09dc56. These constants are needed by init systems like
openrc.
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
|
|
Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
|
|
Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
|
|
As reported by Bering-uClibc project uClibc-ng for x86
generates a segfault on SMP machines when trying to run ntpd.
A small test case is here: https://gist.github.com/ddrown
Use a similar nop instruction as for x86_64 in the code.
Signed-off-by: KP.Kirchdoerfer <kapeka@bering-uclibc.de>
|
|
Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
|
|
|
|
For the linux kernel (since 2.6.12) MAKE_THREAD_CPUCLOCK
macro-like computation should be used to get clockid.
|
|
|
|
|
