OSDN Git Service

uclinux-h8/uclibc-ng.git
17 years agoEnable all of the remaining baud rates. All of the other architectures have them...
"Steven J. Hill" [Wed, 18 Jul 2007 03:16:05 +0000 (03:16 -0000)]
Enable all of the remaining baud rates. All of the other architectures have them enabled, so let's go all the way. There was a quick discussion on IRC about making the max baud rate configurable. That requires a bunch of #ifdef salting throughout all of the 'termios.h' files for the pre-processor to chew on. I'm going to pass on that for now.

17 years agoApply patch as submitted by Blue Smurfy Swirly <blauwirbel@gmail.com> for Sparc32.
"Steven J. Hill" [Wed, 18 Jul 2007 03:08:36 +0000 (03:08 -0000)]
Apply patch as submitted by Blue Smurfy Swirly <blauwirbel@gmail.com> for Sparc32.

17 years agosyslog.h: allow prioritynames[]/facilitynames[] to be defined const
Denis Vlasenko [Sat, 14 Jul 2007 14:52:24 +0000 (14:52 -0000)]
syslog.h: allow prioritynames[]/facilitynames[] to be defined const
(selectable by #defining SYSLOG_NAMES_CONST)

17 years agoFix broken syscall() function since the original patch did not work.
"Steven J. Hill" [Thu, 12 Jul 2007 00:58:41 +0000 (00:58 -0000)]
Fix broken syscall() function since the original patch did not work.

17 years agotweak msync prototype to match header
Mike Frysinger [Fri, 29 Jun 2007 08:19:22 +0000 (08:19 -0000)]
tweak msync prototype to match header

17 years ago- make sure to define _LIBC in libc-symbols.h before including anything
Bernhard Reutner-Fischer [Sun, 24 Jun 2007 17:23:54 +0000 (17:23 -0000)]
- make sure to define _LIBC in libc-symbols.h before including anything
  else.
- internal_function is only used inside libc, so do not define it in
  uClibc_arch_features.h on i386 if we are outside of libc.

17 years agoas noted by psm, make sure we accumulate PICFLAG-y back into PICFLAG
Mike Frysinger [Fri, 22 Jun 2007 16:00:45 +0000 (16:00 -0000)]
as noted by psm, make sure we accumulate PICFLAG-y back into PICFLAG

17 years agopull from Blackfin repo: add -mid-shared-library for UCLIBC_FORMAT_SHARED_FLAT and...
Mike Frysinger [Fri, 22 Jun 2007 07:11:53 +0000 (07:11 -0000)]
pull from Blackfin repo: add -mid-shared-library for UCLIBC_FORMAT_SHARED_FLAT and -msep-data for UCLIBC_FORMAT_FLAT_SEP_DATA

17 years agoCarmelo AMOROSO <carmelo.amoroso@st.com> writes:
Joakim Tjernlund [Wed, 13 Jun 2007 16:23:59 +0000 (16:23 -0000)]
Carmelo AMOROSO <carmelo.amoroso@st.com> writes:
Hi All,
this patch fixes a mismatch in ucontext struct between uClibc and kernel.
In this way a signal handler installed by sigaction can access
to the whole machine state correctly.
This can be applied both to trunk and nptl branch

17 years agoCorrect return value. Pointed out by psm.
Joakim Tjernlund [Mon, 4 Jun 2007 22:16:54 +0000 (22:16 -0000)]
Correct return value. Pointed out by psm.

17 years agochange the auxvt code from being dependent on MMU to !LDSO as auxvt things are ELF...
Mike Frysinger [Fri, 1 Jun 2007 19:17:36 +0000 (19:17 -0000)]
change the auxvt code from being dependent on MMU to !LDSO as auxvt things are ELF-specific, not MMU-specific

17 years agoBernd Schmidt writes: make sure custom stacks work properly for no-mmu
Mike Frysinger [Thu, 31 May 2007 08:26:29 +0000 (08:26 -0000)]
Bernd Schmidt writes: make sure custom stacks work properly for no-mmu

17 years agoAtsushi Nemoto writes:
Mike Frysinger [Thu, 31 May 2007 08:22:35 +0000 (08:22 -0000)]
Atsushi Nemoto writes:
Current MIPS readahead(), posix_fadvise(), posix_fadvise64() do not
match with kernel on all ABIs.

On O32 ABI, a padding is needed before a long long argument.
On N32/N64, a long long argument should be passed via a single register.

17 years agoAtsushi Nemoto writes:
Mike Frysinger [Thu, 31 May 2007 08:21:38 +0000 (08:21 -0000)]
Atsushi Nemoto writes:
http://www.opengroup.org/onlinepubs/009695399/functions/posix_fadvise.html
states it returns error code instead of setting errno.

17 years agoJean-Christian de Rivaz writes:
Mike Frysinger [Fri, 25 May 2007 18:34:46 +0000 (18:34 -0000)]
Jean-Christian de Rivaz writes:
I actually suspect this code into the file uClibc/libc/sysdeps/linux/common/poll.c:
tval.tv_nsec = (timeout % 1000) *1000;  <==== make only usec!
From milisecond this really needs a * 1000000 to make nanosecond. Without this
a 1100 milisecond timeout is converted into a 1 seconde and 100 microsecond
timeout! This can explain the weird result of the test code.

17 years agoRemove the arch specific mmap impl. for Powerpc. The common one will do.
Joakim Tjernlund [Fri, 25 May 2007 18:06:02 +0000 (18:06 -0000)]
Remove the arch specific mmap impl. for Powerpc. The common one will do.

17 years agouse POSIX -eq rather than bash == #1360
Mike Frysinger [Thu, 24 May 2007 04:37:20 +0000 (04:37 -0000)]
use POSIX -eq rather than bash == #1360

17 years agoFix MIPS syscall() and pipe functions to set errno correctly as reported by Daniel...
"Steven J. Hill" [Thu, 24 May 2007 03:00:08 +0000 (03:00 -0000)]
Fix MIPS syscall() and pipe functions to set errno correctly as reported by Daniel Jacobowitz on the mailing list. More information available at <uclibc.org/lists/uclibc/2007-May/017968.html> .

17 years agoFix ARM EABI signal unwinding to accomodate signal frame layout between Linux kernel...
"Steven J. Hill" [Thu, 24 May 2007 02:56:59 +0000 (02:56 -0000)]
Fix ARM EABI signal unwinding to accomodate signal frame layout between Linux kernel versions as reported by Joseph S. Myers on the mailing list. More information available at <uclibc.org/lists/uclibc/2007-May/017971.html> .

17 years agoRestored indentation.
Peter Kjellerstedt [Tue, 15 May 2007 12:05:06 +0000 (12:05 -0000)]
Restored indentation.

17 years agoMade it compile again.
Peter Kjellerstedt [Tue, 15 May 2007 11:58:37 +0000 (11:58 -0000)]
Made it compile again.

17 years agoFix arm mmap when using mmap2 syscall. Fixes bug #1303
Khem Raj [Tue, 15 May 2007 00:37:02 +0000 (00:37 -0000)]
Fix arm mmap when using mmap2 syscall. Fixes bug #1303

17 years agoRemove preprocessor warning due to missing definition of USE_TLS
Khem Raj [Tue, 15 May 2007 00:35:00 +0000 (00:35 -0000)]
Remove preprocessor warning due to missing definition of USE_TLS

17 years agoNickolai Zeldovich writes: if the temp name already exists, then the retry code does...
Mike Frysinger [Sun, 6 May 2007 09:18:39 +0000 (09:18 -0000)]
Nickolai Zeldovich writes: if the temp name already exists, then the retry code does not create a new temp name as the code to do so is outside of the retry loop

17 years ago- commentary typo fix
Bernhard Reutner-Fischer [Fri, 4 May 2007 14:13:17 +0000 (14:13 -0000)]
- commentary typo fix

17 years agogive proper prototypes
Mike Frysinger [Thu, 3 May 2007 23:13:04 +0000 (23:13 -0000)]
give proper prototypes

17 years agocancellation-points needs librt
Mike Frysinger [Wed, 2 May 2007 18:20:58 +0000 (18:20 -0000)]
cancellation-points needs librt

17 years agoexercise all required POSIX pthread cancellation points
Mike Frysinger [Wed, 2 May 2007 08:10:22 +0000 (08:10 -0000)]
exercise all required POSIX pthread cancellation points

17 years agoNickolai Zeldovich writes:
Mike Frysinger [Wed, 2 May 2007 08:05:09 +0000 (08:05 -0000)]
Nickolai Zeldovich writes:
Currently, tcgetpgrp() in uClibc uses an int to store a PID (fetched
via ioctl TIOCGPGRP).  This causes problems on platforms where pid_t
is defined to be larger (e.g., uint64_t).  Other functions in termios,
such as tcgetsid() and tcsetpgrp(), already pass a pid_t to ioctl(),
so the following patch does the same in tcgetpgrp() as well.

17 years agoDaniel Jacobowitz: sleep()/usleep() relies on nanosleep() being a cancellation
Mike Frysinger [Wed, 2 May 2007 08:03:22 +0000 (08:03 -0000)]
Daniel Jacobowitz: sleep()/usleep() relies on nanosleep() being a cancellation
point but the files have "libc_hidden_proto(nanosleep)" which means it always
calls the libc.so version, never the wrapped version in libpthread.so that's a
cancellation point.

17 years agoCarmelo AMOROSO writes:
Mike Frysinger [Mon, 30 Apr 2007 21:23:22 +0000 (21:23 -0000)]
Carmelo AMOROSO writes:
running LTP test suite on uClibc-nptl for sh4 I found a bug into pread
and pwrite functions.  When the offset is negative it is not correctly
handled due to a missing shift operation, so it is passed to the
syscall as the highest unsigned positive value.

17 years agofix from Bernd Schmidt for realloc shrinkage bug
Mike Frysinger [Tue, 24 Apr 2007 05:24:08 +0000 (05:24 -0000)]
fix from Bernd Schmidt for realloc shrinkage bug

17 years ago- include stdbool.h
Bernhard Reutner-Fischer [Mon, 23 Apr 2007 17:01:05 +0000 (17:01 -0000)]
- include stdbool.h

17 years agoThis test needs XLOCALE to work.
"Steven J. Hill" [Sat, 21 Apr 2007 19:42:21 +0000 (19:42 -0000)]
This test needs XLOCALE to work.

17 years agoBugfix from Atsushi Nemoto: return nonzero error code on failure
Rob Landley [Wed, 18 Apr 2007 16:26:16 +0000 (16:26 -0000)]
Bugfix from Atsushi Nemoto: return nonzero error code on failure
(without which busybox ping segfaults on error).  Broken during locking
changes in in svn 16801.

17 years agoif chroot() failed, then reset "chroot_dir" to "." so that after we do our fallback...
Mike Frysinger [Tue, 17 Apr 2007 23:08:20 +0000 (23:08 -0000)]
if chroot() failed, then reset "chroot_dir" to "." so that after we do our fallback chdir(), we dont continue to prefix the chroot path ... fixes `ldconfig.host -r some/dir/`

17 years agocleanup style; no functional changes
Mike Frysinger [Tue, 17 Apr 2007 22:31:01 +0000 (22:31 -0000)]
cleanup style; no functional changes

17 years agofix ld.so.cache handling on no-mmu setups
Mike Frysinger [Tue, 17 Apr 2007 21:54:11 +0000 (21:54 -0000)]
fix ld.so.cache handling on no-mmu setups

17 years ago- shrink resolver code
Bernhard Reutner-Fischer [Tue, 17 Apr 2007 17:24:29 +0000 (17:24 -0000)]
- shrink resolver code

17 years agoforce ourselves to write proper code and develop testscases
Mike Frysinger [Tue, 17 Apr 2007 13:34:11 +0000 (13:34 -0000)]
force ourselves to write proper code and develop testscases

17 years agomove uclibc.org out of actual source tree
Mike Frysinger [Tue, 17 Apr 2007 13:11:21 +0000 (13:11 -0000)]
move uclibc.org out of actual source tree

17 years agodont force a . before EXTRAVERSION so we can be a little more flexible with versioning
Mike Frysinger [Tue, 17 Apr 2007 13:05:20 +0000 (13:05 -0000)]
dont force a . before EXTRAVERSION so we can be a little more flexible with versioning

17 years agocatch a small overflow in malloc-standard
Mike Frysinger [Tue, 17 Apr 2007 11:58:35 +0000 (11:58 -0000)]
catch a small overflow in malloc-standard

17 years agodisable for now
Mike Frysinger [Tue, 17 Apr 2007 11:45:20 +0000 (11:45 -0000)]
disable for now

17 years agoadd a basic .depend target to better catch updates in source files
Mike Frysinger [Tue, 17 Apr 2007 11:38:21 +0000 (11:38 -0000)]
add a basic .depend target to better catch updates in source files

17 years ago- doesn't make sense to impose a limit that is too narrow. 12 ought to be fine..
Bernhard Reutner-Fischer [Sat, 14 Apr 2007 17:34:47 +0000 (17:34 -0000)]
- doesn't make sense to impose a limit that is too narrow. 12 ought to be fine..

17 years ago- dtrt
Bernhard Reutner-Fischer [Sat, 14 Apr 2007 17:03:18 +0000 (17:03 -0000)]
- dtrt

17 years agofix up malloc debug building
Mike Frysinger [Fri, 13 Apr 2007 18:01:21 +0000 (18:01 -0000)]
fix up malloc debug building

17 years agoPatch by Ricard Wanderlof <ricardw at axis dot com>:
Peter Kjellerstedt [Fri, 13 Apr 2007 08:32:18 +0000 (08:32 -0000)]
Patch by Ricard Wanderlof <ricardw at axis dot com>:
* Add configurable buffer sizes for getpwnam() and getgrnam().
  The default buffer size is, as before, 256 (glibc seems to use
  1024 by default).

17 years agoremove __* symbols and enable llrintf()
Mike Frysinger [Thu, 12 Apr 2007 03:41:16 +0000 (03:41 -0000)]
remove __* symbols and enable llrintf()

17 years agotest case to make sure realloc() can shrink buffers properly
Mike Frysinger [Wed, 11 Apr 2007 22:53:54 +0000 (22:53 -0000)]
test case to make sure realloc() can shrink buffers properly

17 years agoPOSIX says you can use realloc() to shrink buffers ... make sure we dont trigger...
Mike Frysinger [Wed, 11 Apr 2007 22:52:20 +0000 (22:52 -0000)]
POSIX says you can use realloc() to shrink buffers ... make sure we dont trigger a buffer overflow in that case

17 years agoimport llrint() from glibc
Mike Frysinger [Tue, 3 Apr 2007 01:03:42 +0000 (01:03 -0000)]
import llrint() from glibc

17 years agoimplement carg
Mike Frysinger [Mon, 2 Apr 2007 20:52:18 +0000 (20:52 -0000)]
implement carg

17 years agofixes from Carmelo AMOROSO to depend less on kernel headers
Mike Frysinger [Mon, 2 Apr 2007 19:30:58 +0000 (19:30 -0000)]
fixes from Carmelo AMOROSO to depend less on kernel headers

17 years agoPOSIX requires that errno be set whenever 0 is returned by malloc()
Mike Frysinger [Mon, 2 Apr 2007 12:06:00 +0000 (12:06 -0000)]
POSIX requires that errno be set whenever 0 is returned by malloc()

17 years agoFix fallout from the e500 math integration.
Joakim Tjernlund [Mon, 2 Apr 2007 11:10:22 +0000 (11:10 -0000)]
Fix fallout from the e500 math integration.
from Steve Papacharalambous.

17 years agoBernd Schmidt writes:
Joakim Tjernlund [Sat, 31 Mar 2007 13:42:48 +0000 (13:42 -0000)]
Bernd Schmidt writes:
Currently a static libdl.a doesn't run all the constructors or
destructors of the libraries it loads.  I can't see a good reason for
that, and it does cause aborts in the destructors it does run for things
like libgcc.so on the Blackfin.

Fixed with the patch below - untested in mainline, but the equivalent
has been in our Blackfin tree for a while now.

17 years agoUpdate comment regarding passing dl_fini.
Joakim Tjernlund [Sat, 31 Mar 2007 13:36:07 +0000 (13:36 -0000)]
Update comment regarding passing dl_fini.

17 years agoFrom Steve Papacharalambous:
Joakim Tjernlund [Sat, 31 Mar 2007 13:28:15 +0000 (13:28 -0000)]
From Steve Papacharalambous:
Add math support for PowerPC e500.

17 years agoRevert last change until we figure out the correct fix.
Rob Landley [Thu, 29 Mar 2007 17:18:09 +0000 (17:18 -0000)]
Revert last change until we figure out the correct fix.

17 years agoMake the code match the comments. (I narrowed down the bug, Khem Raj
Rob Landley [Wed, 28 Mar 2007 22:50:26 +0000 (22:50 -0000)]
Make the code match the comments.  (I narrowed down the bug, Khem Raj
pointed out the solution.  This fixes a segfault for me on exit from a
statically linked "hello world".)

17 years agoWarning fix for uninitialized variable, by Al Stone.
Rob Landley [Wed, 28 Mar 2007 19:13:06 +0000 (19:13 -0000)]
Warning fix for uninitialized variable, by Al Stone.

17 years agoPaul Brook writes: filter .svn dirs to greatly speed up the file grep test
Mike Frysinger [Sun, 25 Mar 2007 13:08:31 +0000 (13:08 -0000)]
Paul Brook writes: filter .svn dirs to greatly speed up the file grep test

17 years ago- add arm1176 options
Bernhard Reutner-Fischer [Tue, 20 Mar 2007 14:19:37 +0000 (14:19 -0000)]
- add arm1176 options

17 years agoJoseph S. Myers writes:
Joakim Tjernlund [Mon, 19 Mar 2007 18:02:16 +0000 (18:02 -0000)]
Joseph S. Myers writes:
On PowerPC, r13 is used for a small data pointer and needs to be set up
from _SDA_BASE_ (defined by the linker) at startup.

This is needed for the GCC testcase gcc.target/powerpc/980827-1.c to work.
This patch fixes that testcase (verified for both static and dynamic
linking).

17 years agodont want to actually bump EXTRAVERSION
Mike Frysinger [Mon, 19 Mar 2007 09:59:50 +0000 (09:59 -0000)]
dont want to actually bump EXTRAVERSION

17 years ago#1273 if EXTRAVERSION is set, make sure we respect it
Mike Frysinger [Mon, 19 Mar 2007 09:49:04 +0000 (09:49 -0000)]
#1273 if EXTRAVERSION is set, make sure we respect it

17 years agoremove dead link as pointed out by Denis Vlasenko
Mike Frysinger [Sun, 18 Mar 2007 21:50:02 +0000 (21:50 -0000)]
remove dead link as pointed out by Denis Vlasenko

17 years agofix one more to use html entities rather than the ISO 8859 version
Mike Frysinger [Sun, 18 Mar 2007 21:42:48 +0000 (21:42 -0000)]
fix one more to use html entities rather than the ISO 8859 version

17 years agouse html entities rather than the ISO 8859 version
Mike Frysinger [Sun, 18 Mar 2007 21:41:49 +0000 (21:41 -0000)]
use html entities rather than the ISO 8859 version

17 years agoBernd Schmidt writes: The following patch is needed for libc to be able to emulate...
Mike Frysinger [Sun, 18 Mar 2007 21:37:44 +0000 (21:37 -0000)]
Bernd Schmidt writes: The following patch is needed for libc to be able to emulate poll with ppoll.

17 years agofix typo pointed out by Al Stone
Mike Frysinger [Sun, 18 Mar 2007 21:33:32 +0000 (21:33 -0000)]
fix typo pointed out by Al Stone

17 years agoHave Makefile.in call conf-header.sh instead of having kconfig call it
Rob Landley [Sat, 17 Mar 2007 02:45:59 +0000 (02:45 -0000)]
Have Makefile.in call conf-header.sh instead of having kconfig call it
via system().  Also move it to extra/scripts since it's a uClibc
build script, not part of upstream kconfig.  Adjust the "grabbing a new
kconfig snapshot" procedure accordingly.

17 years agoNo configs actually set LIBGCC_CFLAGS to anything, and it's not even
Rob Landley [Fri, 16 Mar 2007 19:38:14 +0000 (19:38 -0000)]
No configs actually set LIBGCC_CFLAGS to anything, and it's not even
written into the resulting .config, so remove it.

17 years agomore ifdef tweaks from holin (#1247) for retarded hosts (like OS X)
Mike Frysinger [Wed, 14 Mar 2007 03:45:56 +0000 (03:45 -0000)]
more ifdef tweaks from holin (#1247) for retarded hosts (like OS X)

17 years agodont include asm/user.h
Mike Frysinger [Wed, 14 Mar 2007 03:10:29 +0000 (03:10 -0000)]
dont include asm/user.h

17 years agoBe consistent about spelling. LGPL says "License" not "Licence", so go
Rob Landley [Tue, 13 Mar 2007 18:34:52 +0000 (18:34 -0000)]
Be consistent about spelling.  LGPL says "License" not "Licence", so go
with that.  (Spotted by Xride on irc.)

17 years agoCarmelo Amoroso writes:
Joakim Tjernlund [Tue, 13 Mar 2007 11:02:33 +0000 (11:02 -0000)]
Carmelo Amoroso writes:
I experiencied some problems while trying to debug the ld.so (sh4)
in debug mode. I found that the _start code was not placed into any
PT_LOAD segment causing a SIGSEV.
I solved this issue explicitely placing the _start code into
the .text section (see attached patch).

17 years agoCarmelo Amoroso writes:
Joakim Tjernlund [Tue, 13 Mar 2007 10:13:35 +0000 (10:13 -0000)]
Carmelo Amoroso writes:
Hi All,
the attached patch completes my previous patch on statically linked
application calling dlopen enabling the support for LD_DEBUG.
Further it fixes the missing initialization of the _dl_library_path
variable.
This patch has been fully tested with the uClibc-nptl branch (sh4).

17 years agoUsing local variable 'result' caused problems with lseek64 syscall failing with gcc...
Khem Raj [Tue, 13 Mar 2007 00:02:10 +0000 (00:02 -0000)]
Using local variable 'result' caused problems with lseek64 syscall failing with gcc 4.2. Renaming to make it a bit uncommon name.

17 years agomake sure undefine references in dlopen()-ed libs are handled graciously
Mike Frysinger [Mon, 12 Mar 2007 19:44:58 +0000 (19:44 -0000)]
make sure undefine references in dlopen()-ed libs are handled graciously

17 years agomake sure dlsym() properly returns NULL with undefined symbols
Mike Frysinger [Mon, 12 Mar 2007 19:22:52 +0000 (19:22 -0000)]
make sure dlsym() properly returns NULL with undefined symbols

17 years agofix from upstream blackfin repo for sigsetjmp
Mike Frysinger [Mon, 12 Mar 2007 18:48:13 +0000 (18:48 -0000)]
fix from upstream blackfin repo for sigsetjmp

17 years agoPatch from Al Stone to fix ia64:
Rob Landley [Sun, 11 Mar 2007 23:06:39 +0000 (23:06 -0000)]
Patch from Al Stone to fix ia64:

The attached patch works around some compilation failures on
ia64 caused by the use of the C preprocessor ## operator, and
allows the ia64 default configuration to build again -- or, at
least get as far as an x86 build does (the current source tree
has some problems compiling because of conflicting type
declarations for __kernel_dev_t).

17 years agoSync kernel_stat64 with kernel. Thanks to Daniel Jacobowitz for hints.
Khem Raj [Thu, 8 Mar 2007 18:11:34 +0000 (18:11 -0000)]
Sync kernel_stat64 with kernel. Thanks to Daniel Jacobowitz for hints.

17 years agoPatch from Khem Raj that fixes running gcc on arm for me. (Otherwise it
Rob Landley [Thu, 8 Mar 2007 18:00:42 +0000 (18:00 -0000)]
Patch from Khem Raj that fixes running gcc on arm for me.  (Otherwise it
says "virtual memory exhausted" trying to build hello world.)

17 years agoNow Test.mak includes toplevel Rules.mak so if DO_ASSERT is not set then -DNDEBUG...
Khem Raj [Thu, 8 Mar 2007 08:55:52 +0000 (08:55 -0000)]
Now Test.mak includes toplevel Rules.mak so if DO_ASSERT is not set then -DNDEBUG is added
to CFLAGS so we need to get the seeko() calls out of the assert otherwise they get
cancelled when NDEBUG is defined.

17 years agosvn 17941/17943 broke the build. 18032 made the threaded case compile
Rob Landley [Thu, 8 Mar 2007 01:26:49 +0000 (01:26 -0000)]
svn 17941/17943 broke the build.  18032 made the threaded case compile
again, this makes the non-threaded case compile again.  I have no idea
if this is the proper fix, I'm just tired if it not compiling for two
weeks.

17 years agoFix a build break due to errno and h_errno being unresolved symbols.
Rob Landley [Wed, 7 Mar 2007 23:39:28 +0000 (23:39 -0000)]
Fix a build break due to errno and h_errno being unresolved symbols.
(No idea if this is the right fix but it's something for now.)

17 years agoSync sh's procfs.h with glibc, fixes gdbserver.
Paul Mundt [Wed, 7 Mar 2007 09:47:39 +0000 (09:47 -0000)]
Sync sh's procfs.h with glibc, fixes gdbserver.
Patch from Carmelo Amoroso <carmelo73@gmail.com>.

17 years agoFix PLTJMP() token pasting build error, patch from
Paul Mundt [Wed, 7 Mar 2007 09:46:22 +0000 (09:46 -0000)]
Fix PLTJMP() token pasting build error, patch from
Carmelo Amoroso <carmelo73@gmail.com>.

17 years agoI always miss svn add.
Khem Raj [Tue, 6 Mar 2007 19:13:31 +0000 (19:13 -0000)]
I always miss svn add.

17 years agoRemove debian directory at the request of Simon Richter. (The debian
Rob Landley [Tue, 6 Mar 2007 18:37:15 +0000 (18:37 -0000)]
Remove debian directory at the request of Simon Richter.  (The debian
developers prefer to maintain their own debian subdir as a patch.)

17 years agoAdd posix_fadvise posix_fadvise64 for arm.
Khem Raj [Tue, 6 Mar 2007 08:06:26 +0000 (08:06 -0000)]
Add posix_fadvise posix_fadvise64 for arm.
Import INTERNAL_SYSCALL macro for i386 from glibc.
Use above macro in posix_fadvise implementation if an arch defines it.

17 years agoNew test targets 'compile' and 'run'
Khem Raj [Mon, 5 Mar 2007 20:49:57 +0000 (20:49 -0000)]
New test targets 'compile' and 'run'

17 years agoinclude stddef.h to get NULL definition.
Khem Raj [Thu, 1 Mar 2007 22:26:51 +0000 (22:26 -0000)]
include stddef.h to get NULL definition.

17 years agoFix compiler warning about shadowed declaration.
Khem Raj [Thu, 1 Mar 2007 22:02:09 +0000 (22:02 -0000)]
Fix compiler warning about shadowed declaration.

17 years agoShould be __sigset_t instead of sigset_t.
Khem Raj [Thu, 1 Mar 2007 21:59:59 +0000 (21:59 -0000)]
Should be __sigset_t instead of sigset_t.

17 years agoadd support for ppoll() and emulate poll() with it when __NR_poll does not exist
Mike Frysinger [Wed, 28 Feb 2007 22:25:41 +0000 (22:25 -0000)]
add support for ppoll() and emulate poll() with it when __NR_poll does not exist