OSDN Git Service
Yann E. MORIN [Sun, 9 Jan 2011 00:45:05 +0000 (01:45 +0100)]
ARM: introduce blind options to select & force THUMB mode
Add three new blind options to set use of Thumb mode:
- COMPILE_IN_THUMB_MODE
- if set, CFLAGS will contain -mthumb
- if unset, the compiler's default is used
- HAS_THUMB
- CPUS with Thumb instruction set can select this
- use of BX depends on this
- FORCE_THUMB
- CPUs that are Thumb-only must select this
- this selects: HAS_THUMB, COMPILE_IN_THUMB_MODE and USE_BX
Also, remove leading space in Rules.mak.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Cc: Khem Raj <raj.khem@gmail.com>
Cc: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Cc: Carmelo AMOROSO <carmelo.amoroso@st.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Yann E. MORIN [Sun, 9 Jan 2011 00:45:04 +0000 (01:45 +0100)]
ARM: reorder "Use BX" option
"Use BX" is not available on all CPUs, so the option depends on
a correct CPU to be chosen . It is weird that e BX" then appears
_above_ the CPU selection, not below.
Move the "Use BX" after the CPU selection.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Cc: Khem Raj <raj.khem@gmail.com>
Cc: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Cc: Carmelo AMOROSO <carmelo.amoroso@st.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Phil Blundell [Sat, 11 Jun 2011 05:10:46 +0000 (01:10 -0400)]
Fix __libc_epoll_pwait compile failure on x86
This prevents "memory input 7 is not directly addressable" errors.
| libc/sysdeps/linux/common/epoll.c: In function '__libc_epoll_pwait':
| libc/sysdeps/linux/common/epoll.c:71:80: error: memory input 7 is not directly addressable
| libc/sysdeps/linux/common/epoll.c:75:86: error: memory input 7 is not directly addressable
| make: *** [libc/sysdeps/linux/common/epoll.o] Error 1
| make: *** Waiting for unfinished jobs....
Signed-off-by: Phil Blundell <pb@pbcl.net>
Signed-off-by: Jason Woodward <jason.woodward@timesys.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Khem Raj [Fri, 10 Jun 2011 20:05:34 +0000 (13:05 -0700)]
nptl/Makefile.in: Fix the build break caused by UCLIBC_CTOR_DTOR enabling
Extra / somehow does not match the target and complains that
$(top_builddir)libpthread/nptl/sysdeps/pthread/crti.o has no rules to
build
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Carmelo Amoroso [Mon, 30 May 2011 12:22:57 +0000 (14:22 +0200)]
buildsys: fix pregen target (!NPTL with LOCALE)
When NPTL is not enabled, and LOCALE support is used, compilation fails
with the following error:
-------------------------------------------------------------------------
MKDIR include/bits
GEN include/bits/sysnum.h
make[1]: *** No rule to make target `../..//include/bits/uClibc_ctype.h', needed by `../../extra/locale/gen_wc8bit.c'. Stop.
make: *** [pregen] Error 2
-------------------------------------------------------------------------
The problem is that "headers" is not listed as prerequisite of pregen.
These have been changed multiple times recently:
o Austin: order headers before pregen-headers
1f5c73c9f8a98b1d2a35cea868a585c97ab0e436
o Khem: pregen: Fix the parallel build problem in pregen target
046035ecde92262d96eff2192ba3cda716f04909
o Khem: pregen-headers: Add new target which depends on headers.
ef18cfe8ebab25f5ef92e81956f50e2dc57df602
o Carmelo: build: Fix infinite loop when no threading support is enabled
ff5e4de7088fe8d34812c2a1e604bf04be713606
In all of these commits headers was actually as prerequisite
In commit
9381d622e2411a35a5fd73a5a573eb269e2dd9c9 [nptl: fix buildsys] by Bernhard, headers was removed,
causing the problem reported above.
Signed-off-by: Filippo Arcidiacono <filippo.arcidiacono@st.com>
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Bernhard Reutner-Fischer [Tue, 31 May 2011 07:47:12 +0000 (09:47 +0200)]
libubacktrace: Fix ASNEEDED emission
In master the variable is called SHARED_LIBNAME and not
SHARED_MAJORNAME.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Tue, 31 May 2011 07:45:44 +0000 (09:45 +0200)]
NPTL: Fix CTOR_DTOR
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Filippo Arcidiacono [Fri, 27 May 2011 15:36:20 +0000 (17:36 +0200)]
libc: make common longjmp usable with NPTL
Call _longjmp_unwind conditionally under NPTL config option,
making longjmp usable with NPTL.
Signed-off-by: Filippo Arcidiacono <filippo.arcidiacono@st.com>
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Filippo Arcidiacono [Fri, 27 May 2011 15:17:51 +0000 (17:17 +0200)]
sh: use the common implementation for longjmp
Signed-off-by: Filippo Arcidiacono <filippo.arcidiacono@st.com>
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Filippo Arcidiacono [Fri, 27 May 2011 15:08:36 +0000 (17:08 +0200)]
sh: fix build in !NPTL case
Fix build issue due to missing symbols in !NPTL case:
- vfork
- _longjmp_unwind
Signed-off-by: Filippo Arcidiacono <filippo.arcidiacono@st.com>
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Alan Davis [Thu, 26 May 2011 16:15:06 +0000 (18:15 +0200)]
Correct a bug when remapping textrel segments on nommu
From: Alan Davis <adavis@ti.com>
On C6X, when trying to execute a program that has a textrel DSO, it
fails to load. The telltale line in the LD_DEBUG output is:
_dl_get_ready_to_run:779: file=''; needed by './a.out'
The corresponding DT_NEEDED entry has 'libc.so.0', but here the
filename is empty. This is what is happening in
_dl_elf_shared_library():
First, map all segments according to their permissions. Text gets
initially mapped read-only.
Then, parse the dynamic information. The dynamic table is in RW but
some of the tags may point to RO. For example, DT_NEEDED points to a
string in .dynstr which is in RO. These pointers get computed
according to the loadmap from the original mapping.
Then, in response to a DT_TEXTREL tag, the RO segment gets remapped,
thereby invaliding anything that points to it, in particular certain
dynamic tags such as DT_NEEDED.
The following patch re-parses the dynamic info after the remapping so
as to re-compute any invalid pointers.
Signed-off-by: Alan Davis <adavis@ti.com>
Signed-off-by: Bernd Schmidt <bernds@codesourcery.com>
Khem Raj [Fri, 13 May 2011 22:31:06 +0000 (15:31 -0700)]
unwind.h: Move to libc/sysdeps from nptl sysdeps
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Bernhard Reutner-Fischer [Fri, 13 May 2011 10:16:29 +0000 (12:16 +0200)]
epoll: redo cancellation to match master
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Peter S. Mazinger [Thu, 12 May 2011 23:56:08 +0000 (01:56 +0200)]
epoll.c: add cancellation to epoll_[p]wait()
While there, fix epoll_pwait syscall, it takes 6 arguments
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Thu, 12 May 2011 10:26:41 +0000 (12:26 +0200)]
libm: fix building out-of-tree
Thanks to Thierry Reding for noticing!
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Thu, 12 May 2011 10:11:14 +0000 (12:11 +0200)]
wordexp: fix mishandled return value of pipe call
and use pipe2 if available while at it.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Thu, 12 May 2011 08:37:36 +0000 (10:37 +0200)]
linux_specific: handle accept4 and pipe2
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Thu, 12 May 2011 08:30:48 +0000 (10:30 +0200)]
NPTL: needs madvise internally
This fixes
8d09a50a044638fde2ed3e1a1c4d3c7c5a3cce5c since NPTL also uses
madvise internally.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Thu, 12 May 2011 08:07:25 +0000 (10:07 +0200)]
socket: fix stubs
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 11 May 2011 20:22:51 +0000 (22:22 +0200)]
commentary typo fix
s/decriptor/descriptor/
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 11 May 2011 20:17:23 +0000 (22:17 +0200)]
epoll: unify epoll.h
Bernhard Reutner-Fischer [Wed, 11 May 2011 20:03:33 +0000 (22:03 +0200)]
commentary typo fix
use cancellation (with two 'l') uniformly.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Thierry Reding [Wed, 11 May 2011 07:28:13 +0000 (09:28 +0200)]
Implement epoll_create1 and epoll_pwait system calls.
Note: TODO: This lacks cancellation support.
Signed-off-by: Thierry Reding <thierry.reding@avionic-design.de>
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 11 May 2011 19:16:23 +0000 (21:16 +0200)]
buildsys: document UCLIBC_EXTRA_CPPFLAGS
Rename EXTRA_CPPFLAGS to UCLIBC_EXTRA_CPPFLAGS and make them override
previous flags (by appending - not prepending - them to the other flags).
Bernhard Reutner-Fischer [Wed, 11 May 2011 18:42:18 +0000 (20:42 +0200)]
lll: avoid shadow warnings
use __ret uniformly to avoid shadow warnings as seen on x86_64.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 11 May 2011 18:29:59 +0000 (20:29 +0200)]
x86_64: add TODO for R_X86_64_RELATIVE
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 11 May 2011 17:52:16 +0000 (19:52 +0200)]
libubacktrace: mv unwind.h to include
Fixes compilation if NPTL is off.
I hope this doesn't break ARM EABI but cannot check ATM.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 11 May 2011 17:16:17 +0000 (19:16 +0200)]
Revert "libubacktrace: fix config-symbol handling"
This reverts commit
7a080cd149c7b25d415d76506510d55b34819fc2.
This symbol is in fact different (as it is set no 'n').
Bernhard Reutner-Fischer [Wed, 11 May 2011 11:12:14 +0000 (13:12 +0200)]
add pipe2()
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 11 May 2011 10:05:20 +0000 (12:05 +0200)]
socketcalls: add stubs
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 11 May 2011 09:31:45 +0000 (11:31 +0200)]
accept4: Implement cancellation
.. and add proper prototype, move it into it's own obj and other such
cleanups.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Thierry Reding [Wed, 11 May 2011 06:40:58 +0000 (08:40 +0200)]
Implement accept4 system call.
Signed-off-by: Thierry Reding <thierry.reding@avionic-design.de>
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Fri, 6 May 2011 17:07:12 +0000 (19:07 +0200)]
libubacktrace: fix config-symbol handling
Config symbals that are not set are empty, not 'n'.
Bernhard Reutner-Fischer [Wed, 11 May 2011 07:37:35 +0000 (09:37 +0200)]
buildsys: Turn on strict aliasing for everyone
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Tue, 10 May 2011 19:22:25 +0000 (21:22 +0200)]
ldso: commentary typo fix
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Tue, 10 May 2011 18:19:46 +0000 (20:19 +0200)]
getcwd: Fix memory-leak in error path
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Tue, 10 May 2011 18:09:36 +0000 (20:09 +0200)]
__uc_malloc: Fix memory-leak in error path
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Filippo Arcidiacono [Fri, 6 May 2011 14:49:28 +0000 (16:49 +0200)]
libdl: fix local symbol's address handling in dladdr
Fix dladdr to correctly handle local function's address so backtrace_symbols
print only the function address for these function, instead of showing the name
of nearest one.
Indeed the dladdr walk through the hash table to find the nearest symbol, that
doesn't contain local symbols.
Signed-off-by: Filippo Arcidiacono <filippo.arcidiacono@st.com>
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Bernhard Reutner-Fischer [Tue, 10 May 2011 08:03:58 +0000 (10:03 +0200)]
*/crtn.S: Remove .size directive for _init and _fini
These are split across objects so setting size does not (and never did)
work since the expression cannot be computed at assembly time.
This avoids errors from recent (> 2.21) gas.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Timo Teräs [Wed, 20 Apr 2011 10:59:15 +0000 (13:59 +0300)]
posix_fadvise64: fix x86 implementation
Commit
73d59554144f429b1cf0d4d7fa7de42bdf59ad92 completely broke
the x86 implementation of posix_fadvise64. It moved the first
the assembly code retn instruction gets missing depending on the
Technically the file has two implementaions for posix_fadvise64,
one when __NR_fadvise64_64 is available, and second one if only
__NR_fadvise64 is there. Fix the #ifdef's to be proper for that.
Signed-off-by: Timo Teräs <timo.teras@iki.fi>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Bernhard Reutner-Fischer [Thu, 5 May 2011 07:33:14 +0000 (09:33 +0200)]
i386: fix .size of _init/_fini
binutils-2.21 barf on .size that do not evaluate to const, so use
the section size and not a function that is not visible here.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Thu, 5 May 2011 07:27:15 +0000 (09:27 +0200)]
x86_64: fix .size of _init/_fini
binutils-2.21 barf on .size that do not evaluate to const, so use the
section size and not a function that is not visible here.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Thu, 5 May 2011 07:19:16 +0000 (09:19 +0200)]
Revert "*/crt?.S: remove .size directives"
Removing them generally was not a good idea
This reverts commit
233c504cd940d9802226b6a3a092368b86978f5e.
Bernhard Reutner-Fischer [Wed, 4 May 2011 08:01:24 +0000 (10:01 +0200)]
*/crt?.S: remove .size directives
sed -i -e '/\.size[[:space:]]/d' $(grep -l "\.size" libc/sysdeps/linux/*/crt*.[sSc])
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Henning Heinold [Fri, 29 Apr 2011 11:58:02 +0000 (13:58 +0200)]
libubacktrace: fix backtrace support on arm-eabi, which needs libgcc_eh linked too
Signed-off-by: Henning Heinold <heinold@inf.fu-berlin.de>
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Denys Vlasenko [Tue, 3 May 2011 15:29:35 +0000 (17:29 +0200)]
getaddrinfo.c: fix incorrect check for ERANGE from gethostbyaddr_r
Also, freeaddrinfo(NULL) is ok, no need to check parameted for NULL
before calling it.
Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
Denys Vlasenko [Tue, 3 May 2011 15:28:19 +0000 (17:28 +0200)]
getaddrinfo.c: improve code readability. No functional changes
Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
Bernhard Reutner-Fischer [Tue, 3 May 2011 14:40:31 +0000 (16:40 +0200)]
string: remove unused variable
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Tue, 3 May 2011 14:36:15 +0000 (16:36 +0200)]
x86_64: silence warning if !TLS
TODO: fix all other arches
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Tue, 3 May 2011 14:24:08 +0000 (16:24 +0200)]
buildsys: prettify ssp.c handling
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Tue, 3 May 2011 14:15:30 +0000 (16:15 +0200)]
madvise is LINUX_SPECIFIC
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Filippo Arcidiacono [Tue, 3 May 2011 07:32:53 +0000 (09:32 +0200)]
test_nptl: fix expected result for tst-cputimer[123]
Signed-off-by: Filippo Arcidiacono <filippo.arcidiacono@st.com>
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Filippo Arcidiacono [Mon, 2 May 2011 10:59:18 +0000 (12:59 +0200)]
test_nptl: fix expected result for tst-clock2 test
tst-clock2 should return 0 when _SC_THREAD_CPUTIME option isn't
available, instead of treating it as an error.
Further set the expected ret value as 0 avoiding to hide any real failures
in case of THREAD_CPUTIME feature available.
Signed-off-by: Filippo Arcidiacono <filippo.arcidiacono@st.com>
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Bernhard Reutner-Fischer [Mon, 2 May 2011 08:15:41 +0000 (10:15 +0200)]
buildsys: make $(LOCAL_INSTALL_PATH) phony
Could depend on the individual files too but this makes sure that
nothing is missed.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Mon, 2 May 2011 08:10:47 +0000 (10:10 +0200)]
ether_aton: reject invalid input
fixes PR2227
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Sun, 1 May 2011 16:00:04 +0000 (18:00 +0200)]
tests: disable ether tests if !HAS_SOCKET
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
David A Ramos [Sun, 1 May 2011 15:28:31 +0000 (17:28 +0200)]
inet: add ether_aton testcase
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Fri, 29 Apr 2011 12:14:27 +0000 (14:14 +0200)]
sysconf: clock_getres depends on HAS_REALTIME
Bug was introduced in revision
a202cf6f.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Fri, 29 Apr 2011 12:05:54 +0000 (14:05 +0200)]
__rt_sigwaitinfo: depends on HAS_REALTIME
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Carmelo Amoroso [Wed, 27 Apr 2011 07:29:44 +0000 (09:29 +0200)]
buildsys: minor fixes in Makefile.arch for C6X
Just a tidy-up by removing commented-out lines.
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Carmelo Amoroso [Wed, 27 Apr 2011 07:20:48 +0000 (09:20 +0200)]
buildsys: minor fixes in Makefile.arch for microblaze
Do not include Makefile.commonarch directly from within arch specific
Makefile, as it is already done in parent Maefile.in.
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Carmelo Amoroso [Tue, 26 Apr 2011 11:54:13 +0000 (13:54 +0200)]
libubacktrace: enabled for all archs indeed.
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Austin Foxley [Fri, 22 Apr 2011 23:21:30 +0000 (16:21 -0700)]
sparc: don't access fp registers when configured for no fpu
Signed-off-by: Austin Foxley <austinf@cetoncorp.com>
Carmelo Amoroso [Fri, 22 Apr 2011 10:55:43 +0000 (12:55 +0200)]
libubacktrace: generic implementation based dwarf
Use the initial implementation for SH4 based on dwarf for all archs.
Indeed there are not obvious reason for which it should not work in general.
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Salvatore Cro [Wed, 20 Apr 2011 10:49:25 +0000 (12:49 +0200)]
Added fts support for traversing UNIX file hierarchies.
It is required by libdwfl in elfutils package.
Signed-off-by: Salvatore Cro <salvatore.cro at st.com>
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Salvatore Cro [Fri, 8 Apr 2011 15:56:22 +0000 (17:56 +0200)]
test: fix tst-tls13 expected result and timeout
tst-tls13 is expected to return 0 when passing.
Set higher timeout to avoid failure on slow archs.
Signed-off-by: Salvatore Cro <salvatore.cro@st.com>
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Kevin Cernekee [Sat, 16 Apr 2011 17:02:00 +0000 (10:02 -0700)]
dl-startup: fix typos in block comment
Signed-off-by: Kevin Cernekee <cernekee@gmail.com>
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Maksim Rayskiy [Fri, 15 Apr 2011 17:25:40 +0000 (10:25 -0700)]
MIPS LDSO: pass sym_ref parameter to _dl_find_hash() to support PROTECTED symbols
_dl_find_hash() relies on sym_ref parameter to check if the looked-up symbol
is protected. The code fixes a case when _dl_perform_mips_global_got_relocations()
was calling _dl_find_hash() without providing sym_ref parameter.
The bug was causing hangs if a library exporting non-protected symbol was earlier in
link order than library with the same symbol declared as protected.
Signed-off-by: Maksim Rayskiy <mrayskiy@broadcom.com>
Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
Bernhard Reutner-Fischer [Thu, 14 Apr 2011 07:59:26 +0000 (09:59 +0200)]
buildsys: do_rm ARCH_HEADERS
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 13 Apr 2011 17:38:40 +0000 (19:38 +0200)]
resolv: really fix res_close not to hang with ipv6
Fix goof in previous commit.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 13 Apr 2011 08:26:38 +0000 (10:26 +0200)]
buildsys: do not use $(and)
make-3.80 does not have $(and) and $(or), so workaround for now.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Kevin Cernekee [Sun, 10 Apr 2011 02:55:07 +0000 (19:55 -0700)]
buildsys: fix GNU make v3.80 compatibility again
GNU make 3.80 cannot handle "$(and)" or "$(or)" from commit
18e7136e
(buildsys: use kbuild style). Replace them with ifeq/ifneq.
Signed-off-by: Kevin Cernekee <cernekee@gmail.com>
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Tue, 12 Apr 2011 07:39:20 +0000 (09:39 +0200)]
resolv: fix res_close not to hang with ipv6
Timo Teräs writes:
The memory release loop is missing an obvious counter increment.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernd Schmidt [Mon, 11 Apr 2011 11:21:23 +0000 (13:21 +0200)]
Support dynamic assignment of DSBT_INDEX
For DSBT targets (C6X only at this point), we'd like to support the case
where the user did not specify --dsbt-index at link time when building a
shared library. The dynamic linker can still assign an index at runtime
and fix up the DSBT_INDEX relocs, at the cost of startup time and memory
space.
Signed-off-by: Bernd Schmidt <bernds@codesourcery.com>
Bernd Schmidt [Mon, 11 Apr 2011 11:19:05 +0000 (13:19 +0200)]
Fix malloc alignment
In commit
3e0a1f388, Richard tried to fix malloc alignments by using
alignof (double __attribute_aligned__(sizeof (size_t))).
This doesn't work, since attribute_aligned overrides the alignment
rather than providing a minimum. On C6X, malloc returns four-byte
aligned values rather than the necessary eight-byte alignment.
It's simpler to use a comparison and pick the bigger of the two values,
so that's what I've done.
Signed-off-by: Bernd Schmidt <bernds@codesourcery.com>
Bernd Schmidt [Mon, 11 Apr 2011 11:13:18 +0000 (13:13 +0200)]
Fix nommu handling of DT_TEXTREL
We have a problem with DT_TEXTREL shared libraries on nommu machines.
The dynamic linker's strategy is to map the text segment read-only
first, then look for DT_TEXTREL, and use mprotect to change protections
if necessary. This fails on nommu, since a nommu kernel can decide to
share the memory for private read-only file mappings, and mprotect
doesn't (can't) do anything about this sharing. Existing nommu targets
apparently have no need for this, but on C6X, we may need to assign
library indices at run-time if no --dsbt-index option was passed to the
linker at build time.
Hence, the following patch, which instead of using mprotect, redoes the
mapping with PF_W set.
Signed-off-by: Bernd Schmidt <bernds@codesourcery.com>
Kevin Day [Wed, 30 Mar 2011 11:51:22 +0000 (04:51 -0700)]
utils/ldd: Check for returned pointer from strrchr not the value it holds
Signed-off-by: Kevin Day <thekevinday@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Thorsten Glaser [Sat, 26 Mar 2011 15:11:51 +0000 (15:11 +0000)]
cris: add provide arch-specific vfork implementation
vfork must be called with nothing at all on the stack,
so implementing it via syscall() does not work.
Signed-off-by: Thorsten Glaser <tg@mirbsd.org>
Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Peter S. Mazinger [Fri, 18 Mar 2011 14:37:35 +0000 (15:37 +0100)]
lutimes.c, stubs.c: fix compiling lutimes, if __NR_utimensat is not defined
while there, provide stubs for functions depending on utimensat syscall.
Reported-by: Sedat Dilek <sedat.dilek@googlemail.com>
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Bernhard Reutner-Fischer [Wed, 16 Mar 2011 19:20:32 +0000 (20:20 +0100)]
bump version to 0.9.32-rc3-git
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 16 Mar 2011 19:11:13 +0000 (20:11 +0100)]
release 0.9.32-rc3
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Bernhard Reutner-Fischer [Wed, 16 Mar 2011 19:09:43 +0000 (20:09 +0100)]
memalign: include sys/param.h for MAX
libc/stdlib/malloc/memalign.c:22:1: warning: "MAX" redefined In file included from
./libpthread/nptl/sysdeps/unix/sysv/linux/i386/lowlevellock.h:25,
from ./include/bits/libc-lock.h:36,
from ./include/bits/stdio-lock.h:23,
from ./include/bits/uClibc_mutex.h:71,
from libc/stdlib/malloc/malloc.h:135,
from libc/stdlib/malloc/memalign.c:18:
./include/sys/param.h:75:1: warning: this is the location of the previous definition
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Khem Raj [Tue, 15 Mar 2011 05:45:33 +0000 (22:45 -0700)]
arm/bits/atomic.h: Include common/bit/atomic.h for thumb1
This restores the behavior for thumb1 builds and yet uses
the new atomic.h for arm and thumb2 modes.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Peter S. Mazinger [Thu, 10 Mar 2011 12:15:10 +0000 (13:15 +0100)]
wctype.h: fix libc_hidden_proto for iswupper and add it for iswspace
Old typo, became visible due to other changes.
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Thu, 10 Mar 2011 12:13:26 +0000 (13:13 +0100)]
add libc_hidden_proto for wcs[n]casecmp_l
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Thu, 10 Mar 2011 12:02:19 +0000 (13:02 +0100)]
really fix missing __libc_drand48_data
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Thu, 10 Mar 2011 11:59:15 +0000 (12:59 +0100)]
Revert "missing prototype of __libc_drand48_data fixed"
This reverts commit
e2cea9b9edabe885b9bf2586f9541ffb8fe93f3f.
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Thu, 10 Mar 2011 10:43:07 +0000 (11:43 +0100)]
missing prototype of __libc_drand48_data fixed
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Thu, 10 Mar 2011 10:34:38 +0000 (11:34 +0100)]
time.c, time.h: remove unused hidden strftime/strptime
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Thu, 10 Mar 2011 10:33:51 +0000 (11:33 +0100)]
nanosleep.c: remove duplicated libc_hidden_proto
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Thu, 10 Mar 2011 10:06:25 +0000 (11:06 +0100)]
ctype.c, ctype.h: remove commented parts that were banned for removal after 0.9.31
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Thu, 10 Mar 2011 09:50:01 +0000 (10:50 +0100)]
_wctype.c, wctype.h: remove unused isw* and wctype_l hidden functions
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Wed, 9 Mar 2011 23:49:01 +0000 (00:49 +0100)]
time.c, wchar.h: remove unused hidden wcsftime
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Wed, 9 Mar 2011 23:01:10 +0000 (00:01 +0100)]
str[n]casecmp.c: fix hidden usage
Provide visible str[n]casecmp[_l], wcs[n]casecmp[_l].
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Wed, 9 Mar 2011 22:20:16 +0000 (23:20 +0100)]
remove unused hidden functions
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Wed, 9 Mar 2011 21:37:05 +0000 (22:37 +0100)]
frv/memset.S: add missing libc_hidden_def
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Wed, 9 Mar 2011 20:55:39 +0000 (21:55 +0100)]
memmem.c, string.h: remove unused hidden memmem
It is used only by gen_collate, not included into libc.
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Wed, 9 Mar 2011 20:40:01 +0000 (21:40 +0100)]
_collate.c: remove duplicated libc_hidden_proto() for strcoll*/wcscoll*
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Wed, 9 Mar 2011 20:30:53 +0000 (21:30 +0100)]
wchar.h, _collate.c, strlcpy.c: use a common prototype of __wcslcpy
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>
Peter S. Mazinger [Wed, 9 Mar 2011 20:11:47 +0000 (21:11 +0100)]
ctype.c, _collate.c, str[n]casecmp.c, strlcpy.c: remove unused hidden functions
Signed-off-by: Peter S. Mazinger <ps.m@gmx.net>