OSDN Git Service

uclinux-h8/uClibc.git
18 years agoRevert earlier patch to install_dev and runtime targets. Need a better solution.
Peter S. Mazinger [Mon, 26 Sep 2005 19:43:49 +0000 (19:43 -0000)]
Revert earlier patch to install_dev and runtime targets. Need a better solution.

18 years agoremove unneeded ';' and add utils to .PHONY
Peter S. Mazinger [Mon, 26 Sep 2005 17:36:29 +0000 (17:36 -0000)]
remove unneeded ';' and add utils to .PHONY

18 years agoAdd __cxa_atexit and __cxa_finalize, thanks to Stephen Warren. This patch breaks...
Peter S. Mazinger [Mon, 26 Sep 2005 17:31:47 +0000 (17:31 -0000)]
Add __cxa_atexit and __cxa_finalize, thanks to Stephen Warren. This patch breaks compatibility with existing binaries, unless the new COMPAT_ATEXIT option is enabled.

18 years agoCorrect targets install_dev and install_runtime to depend on shared
Peter S. Mazinger [Mon, 26 Sep 2005 08:07:11 +0000 (08:07 -0000)]
Correct targets install_dev and install_runtime to depend on shared

18 years agomake sure our utils link against the currently built libc rather than the toolchain one
Mike Frysinger [Sat, 24 Sep 2005 23:57:33 +0000 (23:57 -0000)]
make sure our utils link against the currently built libc rather than the toolchain one

18 years agoRemove unneeded if clause. We won't be in this directory unless we
"Steven J. Hill" [Sat, 24 Sep 2005 16:01:53 +0000 (16:01 -0000)]
Remove unneeded if clause. We won't be in this directory unless we
have enabled threads.

18 years agogenerate headers after running menuconfig
Mike Frysinger [Sat, 24 Sep 2005 04:58:20 +0000 (04:58 -0000)]
generate headers after running menuconfig

18 years agoDo not build __uClibc_main.o with ssp enabled
Peter S. Mazinger [Fri, 23 Sep 2005 20:17:39 +0000 (20:17 -0000)]
Do not build __uClibc_main.o with ssp enabled

18 years agoChanged ssp.c license to LGPL and added vapier as contributor
Peter S. Mazinger [Fri, 23 Sep 2005 20:01:57 +0000 (20:01 -0000)]
Changed ssp.c license to LGPL and added vapier as contributor

18 years agoNew ssp code using syscalls where possible.
Peter S. Mazinger [Fri, 23 Sep 2005 19:52:33 +0000 (19:52 -0000)]
New ssp code using syscalls where possible.
Moved guard_setup to __uClibc_main.c, the only place where it is called.
Removed SIGKILL option, not usable with sigaction.

18 years agoDon't remove libc-internal.h on install, else gcc does not build
Peter S. Mazinger [Fri, 23 Sep 2005 18:44:53 +0000 (18:44 -0000)]
Don't remove libc-internal.h on install, else gcc does not build

18 years agoI really really HATE gcc-4.1 and its whimsical inline rules.
"Steven J. Hill" [Fri, 23 Sep 2005 03:15:09 +0000 (03:15 -0000)]
I really really HATE gcc-4.1 and its whimsical inline rules.

18 years agono check needed in libm/libpthread, we do not recurse anymore if the related option...
Peter S. Mazinger [Thu, 22 Sep 2005 09:45:02 +0000 (09:45 -0000)]
no check needed in libm/libpthread, we do not recurse anymore if the related option is not set, more Makefile cleanups

18 years agodont screw around with lib/ if it doesnt exist
Mike Frysinger [Thu, 22 Sep 2005 03:04:19 +0000 (03:04 -0000)]
dont screw around with lib/ if it doesnt exist

18 years agosplit out nearbyint, round, trunc from libm/powerpc/s_modf.c
Peter S. Mazinger [Thu, 22 Sep 2005 00:51:06 +0000 (00:51 -0000)]
split out nearbyint, round, trunc from libm/powerpc/s_modf.c

18 years agoweaks moved after the related function so gcc4 won't warn
Peter S. Mazinger [Thu, 22 Sep 2005 00:43:18 +0000 (00:43 -0000)]
weaks moved after the related function so gcc4 won't warn

18 years agoclean 32bit-isms and make code 64/32 bit clean
Mike Frysinger [Thu, 22 Sep 2005 00:20:12 +0000 (00:20 -0000)]
clean 32bit-isms and make code 64/32 bit clean

18 years agoMakefile cleanups
Peter S. Mazinger [Thu, 22 Sep 2005 00:04:48 +0000 (00:04 -0000)]
Makefile cleanups

18 years agoadd attribute_relro, should be unconditional, if gcc/ld supports it
Peter S. Mazinger [Wed, 21 Sep 2005 23:51:53 +0000 (23:51 -0000)]
add attribute_relro, should be unconditional, if gcc/ld supports it

18 years agosync copyright with glibc
Mike Frysinger [Wed, 21 Sep 2005 23:49:33 +0000 (23:49 -0000)]
sync copyright with glibc

18 years agoadd some more defines from binutils since our readelf uses it
Mike Frysinger [Wed, 21 Sep 2005 23:48:24 +0000 (23:48 -0000)]
add some more defines from binutils since our readelf uses it

18 years agotouchup CFLAGS and also link the link.h header file like we do with elf.h
Mike Frysinger [Wed, 21 Sep 2005 23:48:05 +0000 (23:48 -0000)]
touchup CFLAGS and also link the link.h header file like we do with elf.h

18 years agoalso symlink thread_db.h instead of duplicating it, fix the clean target, and touchup...
Mike Frysinger [Wed, 21 Sep 2005 23:47:10 +0000 (23:47 -0000)]
also symlink thread_db.h instead of duplicating it, fix the clean target, and touchup the headers target so we dont create bogus dir symlinks

18 years agodo not recurse into subdirs if related options are not set
Peter S. Mazinger [Wed, 21 Sep 2005 23:44:10 +0000 (23:44 -0000)]
do not recurse into subdirs if related options are not set

18 years agoutils should be built all the time (iconv), disable building readelf
Peter S. Mazinger [Wed, 21 Sep 2005 23:40:27 +0000 (23:40 -0000)]
utils should be built all the time (iconv), disable building readelf

18 years agoinclude headers only if the related option is enabled
Peter S. Mazinger [Wed, 21 Sep 2005 23:33:29 +0000 (23:33 -0000)]
include headers only if the related option is enabled

18 years agodisable obstack_printf in header, not supported
Peter S. Mazinger [Wed, 21 Sep 2005 23:18:07 +0000 (23:18 -0000)]
disable obstack_printf in header, not supported

18 years agoremove unused headers on install and clean up all thread related ones
Peter S. Mazinger [Wed, 21 Sep 2005 23:12:18 +0000 (23:12 -0000)]
remove unused headers on install and clean up all thread related ones

18 years agoenable flockfile and friends if threads are not enabled
Peter S. Mazinger [Wed, 21 Sep 2005 23:00:49 +0000 (23:00 -0000)]
enable flockfile and friends if threads are not enabled

18 years agocorrect some typos
Peter S. Mazinger [Wed, 21 Sep 2005 22:53:33 +0000 (22:53 -0000)]
correct some typos

18 years ago__set_errno must be redefined after #include "sys/syscall.h".
Joakim Tjernlund [Wed, 21 Sep 2005 21:18:04 +0000 (21:18 -0000)]
__set_errno must be redefined after #include "sys/syscall.h".
syscall.h should not have it's own version of __set_errno, instead syscall.h should
include errno.h(see powerpc) since errno handling depends on threads/no threads.

I leave the the fixing of syscall.h since there might be some smaller compile problems.

18 years ago- typo in documentation
Bernhard Reutner-Fischer [Wed, 21 Sep 2005 17:48:32 +0000 (17:48 -0000)]
- typo in documentation

18 years agoremove redundant paragraph
Eric Andersen [Wed, 21 Sep 2005 17:45:12 +0000 (17:45 -0000)]
remove redundant paragraph

18 years ago'pthread.h' is not the only file that needs to be symlinked. When NPTL gets merged...
"Steven J. Hill" [Wed, 21 Sep 2005 11:31:54 +0000 (11:31 -0000)]
'pthread.h' is not the only file that needs to be symlinked. When NPTL gets merged, you will need to do 'semaphore.h' and 'bits/pthreadtypes.h'. I think I have been a good little boy and keeping my NPTL changes out of mainline, so I'm going to go ahead and cross the line this once to get this change in.

18 years ago'pthread.h' is not the only file that needs to be symlinked. When NPTL gets merged...
"Steven J. Hill" [Wed, 21 Sep 2005 05:14:02 +0000 (05:14 -0000)]
'pthread.h' is not the only file that needs to be symlinked. When NPTL gets merged, you will need to do 'semaphore.h' and 'bits/pthreadtypes.h'. I think I have been a good little boy and keeping my NPTL changes out of mainline, so I'm going to go ahead and cross the line this once to get this change in.

18 years agomerge some misc stuff from glibc
Mike Frysinger [Wed, 21 Sep 2005 03:55:21 +0000 (03:55 -0000)]
merge some misc stuff from glibc

18 years agomove misc internal definitions to a sep file to minimize differences with features...
Mike Frysinger [Wed, 21 Sep 2005 02:28:48 +0000 (02:28 -0000)]
move misc internal definitions to a sep file to minimize differences with features.h from glibc

18 years agomerge x86_64 optimized string support
Mike Frysinger [Wed, 21 Sep 2005 02:18:29 +0000 (02:18 -0000)]
merge x86_64 optimized string support

18 years agomerge some updates from glibc to fix 32/64 bit support
Mike Frysinger [Wed, 21 Sep 2005 02:17:00 +0000 (02:17 -0000)]
merge some updates from glibc to fix 32/64 bit support

18 years agodont duplicate pthread.h in include and libpthread subdirs, so symlink it
Mike Frysinger [Wed, 21 Sep 2005 02:05:48 +0000 (02:05 -0000)]
dont duplicate pthread.h in include and libpthread subdirs, so symlink it

18 years agoforgot MMAP_HAS_6_ARGS
Joakim Tjernlund [Tue, 20 Sep 2005 21:25:39 +0000 (21:25 -0000)]
forgot MMAP_HAS_6_ARGS

18 years agoMake powerpc use sys/syscall.h, instead of defining its own syscalls.
Joakim Tjernlund [Tue, 20 Sep 2005 21:19:51 +0000 (21:19 -0000)]
Make powerpc use sys/syscall.h, instead of defining its own syscalls.

18 years agoFix bug 370.
Joakim Tjernlund [Tue, 20 Sep 2005 21:16:25 +0000 (21:16 -0000)]
Fix bug 370.

18 years agoRemove unused function.
Joakim Tjernlund [Sun, 18 Sep 2005 14:14:56 +0000 (14:14 -0000)]
Remove unused function.

18 years agodeclare _init/_fini weak if UCLIBC_CTOR_DTOR is not defined
Mike Frysinger [Sun, 18 Sep 2005 08:07:11 +0000 (08:07 -0000)]
declare _init/_fini weak if UCLIBC_CTOR_DTOR is not defined

18 years agoFix bug 430 and reduce ldso size with 1KB :). Not sure if I need to use fstat64,...
Joakim Tjernlund [Sat, 17 Sep 2005 17:06:59 +0000 (17:06 -0000)]
Fix bug 430 and reduce ldso size with 1KB :). Not sure if I need to use fstat64, but I am sure someone will
tell me.

18 years agoAdd atomic support for PPC.
Joakim Tjernlund [Sat, 17 Sep 2005 12:00:06 +0000 (12:00 -0000)]
Add atomic support for PPC.
Merge glibc bits/atomic.h  powerpc32/bits/atomic.h powerpc64/bits/atomic.h into one file

18 years agoChange to glibc way of defining syscalls. This should make PPC
Joakim Tjernlund [Sat, 17 Sep 2005 11:23:39 +0000 (11:23 -0000)]
Change to glibc way of defining syscalls. This should make PPC
compile with gcc 4.0/4.1

18 years agomake sure we treat count as size_t, not ssize_t ... fixes with help from khem
Mike Frysinger [Fri, 16 Sep 2005 23:51:31 +0000 (23:51 -0000)]
make sure we treat count as size_t, not ssize_t ... fixes with help from khem

18 years agosyntax/whitespace touchups
Mike Frysinger [Fri, 16 Sep 2005 23:42:25 +0000 (23:42 -0000)]
syntax/whitespace touchups

18 years agoRobin Getz from blackfin.uclinux.org writes:
Mike Frysinger [Fri, 16 Sep 2005 04:58:13 +0000 (04:58 -0000)]
Robin Getz from blackfin.uclinux.org writes:
Bernd Schmidt found/fixed this problem in uClibc:

http://blackfin.uclinux.org/tracker/index.php?func=detail&aid=882&group_id=17&atid=141

Basically, the uClinux dist includes a few thread demos - thdm and bcdm -
running them simultaneously sometimes causes bus errors.

Bernd wrote:
>They occur in pthread_handle_sigrestart; thread_self returns a bogus value.
>I managed to capture a debugging log, it's attached as bad-log.  If you
>look at it you'll notice that the initial stack bounds are bogus: bottom
>of stack is higher than top of stack.  This appears to be because of a bug
>in NOMMU_INITIAL_THREAD_BOUNDS(tos,bos): if a new thread has a BOS equal
>to the initial thread's current TOS, it'll munge the initial thread's
>stack bounds.  Fixed with the attached patch, which I've committed.
>This should fix the crashes people have seen, but since the bug was always
>hard to reproduce, I can't be 100% certain they are gone.  Please retest,
>everyone...

This patch was made against our cvs, which has an older version of uClibc
in it - this code was recently moved from internals.h to descr.h in the
uclibc svn, but it still has the same problem

18 years agoremove duplicated p_canceled setting and extra whitespace
Mike Frysinger [Fri, 9 Sep 2005 03:05:37 +0000 (03:05 -0000)]
remove duplicated p_canceled setting and extra whitespace

18 years agosync style changes and misc fixes from glibc
Mike Frysinger [Fri, 9 Sep 2005 02:58:06 +0000 (02:58 -0000)]
sync style changes and misc fixes from glibc

18 years agosync some style changes and stack fixes with glibc
Mike Frysinger [Fri, 9 Sep 2005 02:54:46 +0000 (02:54 -0000)]
sync some style changes and stack fixes with glibc

18 years agosync with glibc and macro out the strong_alias/versioned-symbols to cut down on the...
Mike Frysinger [Fri, 9 Sep 2005 02:46:25 +0000 (02:46 -0000)]
sync with glibc and macro out the strong_alias/versioned-symbols to cut down on the diffs

18 years agonewer linuxthreads renamed the function invalid_handle() to nonexisting_handle()
Mike Frysinger [Fri, 9 Sep 2005 02:45:42 +0000 (02:45 -0000)]
newer linuxthreads renamed the function invalid_handle() to nonexisting_handle()

18 years agosync with glibc by moving the thread_desc stuff out of internals.h and into a new...
Mike Frysinger [Fri, 9 Sep 2005 02:39:12 +0000 (02:39 -0000)]
sync with glibc by moving the thread_desc stuff out of internals.h and into a new file descr.h ... also redo thread_desc init by specifying members in the struct by name rather than position

18 years agorevert pthread_exit change until all files using new prototype is ready
Mike Frysinger [Fri, 9 Sep 2005 00:06:28 +0000 (00:06 -0000)]
revert pthread_exit change until all files using new prototype is ready

18 years agosync all the strong_alias and version compat stuff but make them no-ops with #if...
Mike Frysinger [Thu, 8 Sep 2005 23:57:11 +0000 (23:57 -0000)]
sync all the strong_alias and version compat stuff but make them no-ops with #if magic to cut down on diffs

18 years agoadd hidden_def() macros from glibc but stub them out to cut down on diffs
Mike Frysinger [Thu, 8 Sep 2005 23:50:45 +0000 (23:50 -0000)]
add hidden_def() macros from glibc but stub them out to cut down on diffs

18 years agosync changelog with glibc
Mike Frysinger [Thu, 8 Sep 2005 23:48:03 +0000 (23:48 -0000)]
sync changelog with glibc

18 years agosync prototypes with glibc
Mike Frysinger [Thu, 8 Sep 2005 23:47:35 +0000 (23:47 -0000)]
sync prototypes with glibc

18 years agopunt wrapsyscall.c as glibc has
Mike Frysinger [Thu, 8 Sep 2005 23:46:56 +0000 (23:46 -0000)]
punt wrapsyscall.c as glibc has

18 years agomake sure we use optimized version when building for newer intel/amd cpus
Mike Frysinger [Thu, 8 Sep 2005 04:25:06 +0000 (04:25 -0000)]
make sure we use optimized version when building for newer intel/amd cpus

18 years agoimport some fixes from glibc
Mike Frysinger [Thu, 8 Sep 2005 04:24:25 +0000 (04:24 -0000)]
import some fixes from glibc

18 years agosync with glibc and import pthread_rwlock_timed{rd,wr}lock in the process
Mike Frysinger [Thu, 8 Sep 2005 04:07:49 +0000 (04:07 -0000)]
sync with glibc and import pthread_rwlock_timed{rd,wr}lock in the process

18 years agouClibc has pthread_attr_{get,set}stack now
Mike Frysinger [Thu, 8 Sep 2005 04:04:02 +0000 (04:04 -0000)]
uClibc has pthread_attr_{get,set}stack now

18 years agosync misc trivial touchups from glibc
Mike Frysinger [Thu, 8 Sep 2005 03:56:33 +0000 (03:56 -0000)]
sync misc trivial touchups from glibc

18 years agosync with glibc coding style to make further updates easier
Mike Frysinger [Thu, 8 Sep 2005 03:41:40 +0000 (03:41 -0000)]
sync with glibc coding style to make further updates easier

18 years agosync license with glibc
Mike Frysinger [Thu, 8 Sep 2005 03:28:24 +0000 (03:28 -0000)]
sync license with glibc

18 years agosync license with glibc
Mike Frysinger [Thu, 8 Sep 2005 03:26:31 +0000 (03:26 -0000)]
sync license with glibc

18 years agosync with glibc
Mike Frysinger [Thu, 8 Sep 2005 03:25:37 +0000 (03:25 -0000)]
sync with glibc

18 years agoimport some fixes from glibc
Mike Frysinger [Thu, 8 Sep 2005 03:21:02 +0000 (03:21 -0000)]
import some fixes from glibc

18 years agoimport pthread_attr_{get,set}stack functions from glibc along with a few attr fixes
Mike Frysinger [Thu, 8 Sep 2005 03:19:01 +0000 (03:19 -0000)]
import pthread_attr_{get,set}stack functions from glibc along with a few attr fixes

18 years agoimport the helper function pthread_for_each_thread from glibc
Mike Frysinger [Thu, 8 Sep 2005 03:17:05 +0000 (03:17 -0000)]
import the helper function pthread_for_each_thread from glibc

18 years agoimport misc defines from glibc
Mike Frysinger [Thu, 8 Sep 2005 03:14:23 +0000 (03:14 -0000)]
import misc defines from glibc

18 years agomake sure function is always inlined
Mike Frysinger [Thu, 8 Sep 2005 03:00:39 +0000 (03:00 -0000)]
make sure function is always inlined

18 years agosync with glibc and remove duplicated SEM_VALUE_MAX define
Mike Frysinger [Thu, 8 Sep 2005 03:00:11 +0000 (03:00 -0000)]
sync with glibc and remove duplicated SEM_VALUE_MAX define

18 years agosync with glibc
Mike Frysinger [Thu, 8 Sep 2005 02:59:43 +0000 (02:59 -0000)]
sync with glibc

18 years agosync with glibc
Mike Frysinger [Thu, 8 Sep 2005 02:59:01 +0000 (02:59 -0000)]
sync with glibc

18 years agoslight tweak to minimize diffs with glibc
Mike Frysinger [Thu, 8 Sep 2005 02:58:50 +0000 (02:58 -0000)]
slight tweak to minimize diffs with glibc

18 years agosync code style with glibc
Mike Frysinger [Thu, 8 Sep 2005 02:58:12 +0000 (02:58 -0000)]
sync code style with glibc

18 years agoimport license updates from glibc
Mike Frysinger [Thu, 8 Sep 2005 02:36:16 +0000 (02:36 -0000)]
import license updates from glibc

18 years agopunt _DL_FINI_CRT_COMPAT option now that 0.9.28 has been released
Mike Frysinger [Thu, 8 Sep 2005 02:35:30 +0000 (02:35 -0000)]
punt _DL_FINI_CRT_COMPAT option now that 0.9.28 has been released

18 years agoFix by Martin Schlemmer:
Mike Frysinger [Thu, 8 Sep 2005 02:29:37 +0000 (02:29 -0000)]
Fix by Martin Schlemmer:
If _DL_FINI_CRT_COMPAT is defined, _dl_fini is setup to run at exit via
atexit(), but this makes it run _before_ the fini (__app_fini()) of the
app, causing stuff like sandbox that frees structs, etc via its fini to
segfault.
http://bugs.gentoo.org/98187

18 years agounused
Mike Frysinger [Thu, 8 Sep 2005 02:25:22 +0000 (02:25 -0000)]
unused

18 years agodlvsym stub
Mike Frysinger [Thu, 8 Sep 2005 02:25:01 +0000 (02:25 -0000)]
dlvsym stub

18 years agograb some updates from glibc
Mike Frysinger [Thu, 8 Sep 2005 02:24:40 +0000 (02:24 -0000)]
grab some updates from glibc

18 years agoclean up code and make it easy to extend
Mike Frysinger [Wed, 7 Sep 2005 22:18:47 +0000 (22:18 -0000)]
clean up code and make it easy to extend

18 years agodont fail if target doesnt support mmap64
Mike Frysinger [Wed, 7 Sep 2005 22:18:16 +0000 (22:18 -0000)]
dont fail if target doesnt support mmap64

18 years agograb a tiny mmap implementation from glibc
Mike Frysinger [Wed, 7 Sep 2005 22:16:43 +0000 (22:16 -0000)]
grab a tiny mmap implementation from glibc

18 years agomake sure the syscall mmap func isnt exported
Mike Frysinger [Wed, 7 Sep 2005 22:16:12 +0000 (22:16 -0000)]
make sure the syscall mmap func isnt exported

18 years agofix string functions to not treat the size_t as ssize_t
Mike Frysinger [Mon, 5 Sep 2005 07:29:47 +0000 (07:29 -0000)]
fix string functions to not treat the size_t as ssize_t

18 years agoupdate developer verbage
Eric Andersen [Fri, 2 Sep 2005 14:16:56 +0000 (14:16 -0000)]
update developer verbage

18 years agoadd (size_t)-1 border case tests for strncat and strncmp, and add a bunch of tests...
Mike Frysinger [Wed, 31 Aug 2005 22:27:45 +0000 (22:27 -0000)]
add (size_t)-1 border case tests for strncat and strncmp, and add a bunch of tests for strnlen based on strlen

18 years agoadd M_PERTURB define from glibc and Frank Mehnert writes: the following patch fixes...
Mike Frysinger [Tue, 30 Aug 2005 23:02:27 +0000 (23:02 -0000)]
add M_PERTURB define from glibc and Frank Mehnert writes: the following patch fixes a compile problem where malloc.h was included into a C++ file

18 years agoanemo writes in Bug 9: use ElfW(auxv_t) instead of Elf32_auxv_t
Mike Frysinger [Mon, 29 Aug 2005 22:45:52 +0000 (22:45 -0000)]
anemo writes in Bug 9: use ElfW(auxv_t) instead of Elf32_auxv_t

18 years agosome fixes by anemo in Bug 9 to play nicely with 32 or 64 bit hosts
Mike Frysinger [Mon, 29 Aug 2005 22:45:11 +0000 (22:45 -0000)]
some fixes by anemo in Bug 9 to play nicely with 32 or 64 bit hosts

18 years agopoint to uclibc.org
Eric Andersen [Sun, 28 Aug 2005 12:50:09 +0000 (12:50 -0000)]
point to uclibc.org

18 years agoasierllano writes in bug 370:
Joakim Tjernlund [Sat, 27 Aug 2005 11:41:56 +0000 (11:41 -0000)]
asierllano writes in bug 370:
The gcc compiler has deprecated the type __uint128 for 32-bits platforms. Due to the fact that this
type was introduced in gcc-3.1 it is better to drop its use.
(glibc has the same problem and uses this same solution).