OSDN Git Service
Mike Frysinger [Wed, 8 Feb 2006 01:32:09 +0000 (01:32 -0000)]
all my other 64bit changes seems to have "magically" fixed pthreads tests
Mike Frysinger [Wed, 8 Feb 2006 01:31:24 +0000 (01:31 -0000)]
fix warnings on 64bit hosts
Mike Frysinger [Wed, 8 Feb 2006 01:31:06 +0000 (01:31 -0000)]
janlana writes in Bug 700: use standard uint* types rather than u_int
Mike Frysinger [Wed, 8 Feb 2006 01:30:31 +0000 (01:30 -0000)]
sync with glibc
Mike Frysinger [Tue, 7 Feb 2006 05:10:21 +0000 (05:10 -0000)]
include stdint.h and check __intptr_t_defined to see if we need to setup uintptr_t
Mike Frysinger [Tue, 7 Feb 2006 04:47:42 +0000 (04:47 -0000)]
remove outdated file
Mike Frysinger [Tue, 7 Feb 2006 03:08:53 +0000 (03:08 -0000)]
grab syscall() from glibc
Mike Frysinger [Mon, 6 Feb 2006 05:06:37 +0000 (05:06 -0000)]
thanks to Thomas, nios should be happy again!
Mike Frysinger [Mon, 6 Feb 2006 03:15:24 +0000 (03:15 -0000)]
kill _dummy symbol
Mike Frysinger [Sat, 4 Feb 2006 03:24:24 +0000 (03:24 -0000)]
rfelker reports in Bug 683 that we leak file descriptors if the fcntl() fails ... so fix that. add back in fstat() ofter we open() the directory as we need the blocksize further down in the code. unify the memory error handling to shrink the code there a little.
Mike Frysinger [Sat, 4 Feb 2006 03:10:45 +0000 (03:10 -0000)]
fix building
Mike Frysinger [Sat, 4 Feb 2006 02:48:47 +0000 (02:48 -0000)]
fix segfault do to sizeof(NULL) usage
Mike Frysinger [Sat, 4 Feb 2006 02:20:46 +0000 (02:20 -0000)]
add note about racecondition
Mike Frysinger [Sat, 4 Feb 2006 02:19:23 +0000 (02:19 -0000)]
use O_DIRECTORY when possible, saves us from having to use stat() thus cutting codesize/race condition
Mike Frysinger [Sat, 4 Feb 2006 02:18:41 +0000 (02:18 -0000)]
add some copyright/license info
Mike Frysinger [Sat, 4 Feb 2006 02:17:00 +0000 (02:17 -0000)]
use just __getdents64 rather than either __getdents and __getdents64 when possible (saves space and gives us access to d_type most of the time)
Mike Frysinger [Sat, 4 Feb 2006 02:04:00 +0000 (02:04 -0000)]
we get d_type from the kernel so use it
Mike Frysinger [Sat, 4 Feb 2006 01:28:22 +0000 (01:28 -0000)]
make build output a little less noisy
Mike Frysinger [Sat, 4 Feb 2006 01:28:11 +0000 (01:28 -0000)]
make build output a little less noisy
Mike Frysinger [Sat, 4 Feb 2006 01:27:57 +0000 (01:27 -0000)]
fix warning when building on non-mmu hosts
Mike Frysinger [Sat, 4 Feb 2006 01:27:08 +0000 (01:27 -0000)]
glibc sync
Mike Frysinger [Sat, 4 Feb 2006 01:26:24 +0000 (01:26 -0000)]
touchup test to have some more helpful output
Mike Frysinger [Sat, 4 Feb 2006 01:20:54 +0000 (01:20 -0000)]
import files from glibc for an ia64/static port
Peter S. Mazinger [Sat, 4 Feb 2006 01:11:18 +0000 (01:11 -0000)]
Dont link and remove link.h
Mike Frysinger [Sat, 4 Feb 2006 00:51:49 +0000 (00:51 -0000)]
integrate frv link.h
Mike Frysinger [Fri, 3 Feb 2006 23:29:34 +0000 (23:29 -0000)]
unify duplicated code cause i get tired of updating this stuff
Peter S. Mazinger [Fri, 3 Feb 2006 20:02:23 +0000 (20:02 -0000)]
Weaken libc's hidden fork versions, ltp test failed
Peter S. Mazinger [Fri, 3 Feb 2006 16:19:05 +0000 (16:19 -0000)]
attribute_hidden is enough in the prototype, sync getopt-susv3 w/ gnu
Peter S. Mazinger [Fri, 3 Feb 2006 16:12:58 +0000 (16:12 -0000)]
disable hidden_proto/def for __BCC__ generally
Peter S. Mazinger [Fri, 3 Feb 2006 16:06:06 +0000 (16:06 -0000)]
add -DNOT_IN_libc
Peter S. Mazinger [Fri, 3 Feb 2006 14:37:09 +0000 (14:37 -0000)]
hide __td_agent_list
Peter S. Mazinger [Fri, 3 Feb 2006 14:33:12 +0000 (14:33 -0000)]
Use write instead of __libc_write for debug, hide __td_debug
Peter S. Mazinger [Fri, 3 Feb 2006 14:19:24 +0000 (14:19 -0000)]
Hide internal wrapper
Peter S. Mazinger [Fri, 3 Feb 2006 11:43:16 +0000 (11:43 -0000)]
err() is not used internally
Peter S. Mazinger [Fri, 3 Feb 2006 11:29:32 +0000 (11:29 -0000)]
getopt_long[_only] visible only if GNU getopt used
Peter S. Mazinger [Fri, 3 Feb 2006 11:26:19 +0000 (11:26 -0000)]
Use internal versions of strchr and stderr for non-BCC
Peter S. Mazinger [Thu, 2 Feb 2006 13:47:57 +0000 (13:47 -0000)]
even the hidden version of the weaks in libc (as strong in libpthread) have to be weaks
Peter S. Mazinger [Thu, 2 Feb 2006 12:12:58 +0000 (12:12 -0000)]
prototype was wrong, change not tested
Peter S. Mazinger [Thu, 2 Feb 2006 11:37:06 +0000 (11:37 -0000)]
fix typo s/__initbrk/__init_brk/
Peter S. Mazinger [Wed, 1 Feb 2006 23:02:24 +0000 (23:02 -0000)]
missed another global data marking
Peter S. Mazinger [Wed, 1 Feb 2006 22:11:03 +0000 (22:11 -0000)]
Make it build w/ HAVE_SHARED disabled and SSP enabled
Peter S. Mazinger [Wed, 1 Feb 2006 22:10:08 +0000 (22:10 -0000)]
Use -DSTATIC and remove stub_warning, not accepted
Peter S. Mazinger [Wed, 1 Feb 2006 22:08:43 +0000 (22:08 -0000)]
everything uses now CFLAGS, no need to add anything to CPU_FLAGS, add -DSTATIC for those not needing shared libs, needed to be seen in libc-symbols.h
Peter S. Mazinger [Wed, 1 Feb 2006 11:58:12 +0000 (11:58 -0000)]
missed two __curbrk
Peter S. Mazinger [Wed, 1 Feb 2006 11:55:56 +0000 (11:55 -0000)]
I will risk this one: __curbrk
Peter S. Mazinger [Wed, 1 Feb 2006 11:19:52 +0000 (11:19 -0000)]
Go __malloc_lock, go ...
Peter S. Mazinger [Wed, 1 Feb 2006 11:19:10 +0000 (11:19 -0000)]
weaks for __pthread_internal_tsd* are added if not built w/ -DSHARED, header issue
Peter S. Mazinger [Wed, 1 Feb 2006 09:34:11 +0000 (09:34 -0000)]
gcc41 does not like this at all
Peter S. Mazinger [Wed, 1 Feb 2006 09:25:33 +0000 (09:25 -0000)]
change asm/volatile to __x__
Peter S. Mazinger [Wed, 1 Feb 2006 09:22:45 +0000 (09:22 -0000)]
global data uses libc_hidden_data_def, convert all -I hope- and add some new
Peter S. Mazinger [Wed, 1 Feb 2006 02:07:53 +0000 (02:07 -0000)]
typo
Peter S. Mazinger [Wed, 1 Feb 2006 02:06:12 +0000 (02:06 -0000)]
remove hidden versions of bzero, unused in libc.so
Peter S. Mazinger [Wed, 1 Feb 2006 01:36:33 +0000 (01:36 -0000)]
same in old
Peter S. Mazinger [Wed, 1 Feb 2006 01:34:36 +0000 (01:34 -0000)]
__pthread_exit is noreturn only within libpthread
Peter S. Mazinger [Wed, 1 Feb 2006 01:29:20 +0000 (01:29 -0000)]
correct __uselocale
Mike Frysinger [Wed, 1 Feb 2006 00:14:52 +0000 (00:14 -0000)]
add syscall6
Mike Frysinger [Wed, 1 Feb 2006 00:14:43 +0000 (00:14 -0000)]
add prototype to get rid of warning
Mike Frysinger [Wed, 1 Feb 2006 00:14:33 +0000 (00:14 -0000)]
cast buf to get rid of warning
Mike Frysinger [Wed, 1 Feb 2006 00:14:11 +0000 (00:14 -0000)]
disable lazy code for now
Peter S. Mazinger [Tue, 31 Jan 2006 23:31:32 +0000 (23:31 -0000)]
s:THUMB_INTERWORK:USE_BX:, thx Khem Raj
Peter S. Mazinger [Tue, 31 Jan 2006 22:59:12 +0000 (22:59 -0000)]
add comment to know why it's weak
Peter S. Mazinger [Tue, 31 Jan 2006 22:46:03 +0000 (22:46 -0000)]
change strong to weak_alias, arm fails, thx Khem Raj
Peter S. Mazinger [Tue, 31 Jan 2006 22:32:17 +0000 (22:32 -0000)]
pt-allocrtsig.c is irrelevant for now, remove related too, add libc_hidden_proto from old version
Peter S. Mazinger [Tue, 31 Jan 2006 20:54:59 +0000 (20:54 -0000)]
New linuxthreads builds on x86, vapier: pt-allocrtsig.c is missing from sources
Peter S. Mazinger [Tue, 31 Jan 2006 14:41:37 +0000 (14:41 -0000)]
s/IS_IN_linuxthreads/IS_IN_libpthread/, some prototypes, correct Makefile.in, disable __res_state, unavailable in resolv.h
Peter S. Mazinger [Tue, 31 Jan 2006 09:28:52 +0000 (09:28 -0000)]
include math.h to get the prototype
Mike Frysinger [Tue, 31 Jan 2006 05:05:15 +0000 (05:05 -0000)]
initial import of latest linuxthreads
Mike Frysinger [Tue, 31 Jan 2006 02:28:11 +0000 (02:28 -0000)]
make the extra warnings optional
Mike Frysinger [Tue, 31 Jan 2006 01:50:28 +0000 (01:50 -0000)]
John Bowler writes in Bug 385:
The patch fixes up the .align directives to '2' (i.e. a multiple
of 4) not '4' (a multiple of 16 - apparently an error since it seems
to be unnecessary, there is no advantage here in cache line alignment).
this is an arm "feature" ... the value given to .align is not in bytes:
.align 4 in arm means .align 2 ^ 4
Peter S. Mazinger [Tue, 31 Jan 2006 01:42:44 +0000 (01:42 -0000)]
don't include wchar.h if WCHAR is disabled
Mike Frysinger [Tue, 31 Jan 2006 01:41:16 +0000 (01:41 -0000)]
Khem Raj says:
add support for using BX instruction for THUMB aware architectures
Mike Frysinger [Tue, 31 Jan 2006 01:34:21 +0000 (01:34 -0000)]
rename local var brk to _brk
Mike Frysinger [Tue, 31 Jan 2006 01:32:33 +0000 (01:32 -0000)]
John Bowler says: use (!thumb) instead of (arm) since thumb defines arm
Mike Frysinger [Tue, 31 Jan 2006 00:33:26 +0000 (00:33 -0000)]
John Bowler writes in Bug 385:
This is a work round for a fairly serious GCC compiler bug - when
the syscall assembler overwrites r7 (required on thumb) the
compiler fails to protect the register when it is using it as a
frame pointer.
Mike Frysinger [Tue, 31 Jan 2006 00:29:30 +0000 (00:29 -0000)]
John Bowler writes in Bug 385:
This patch changes all cases where the ARM assembler mov pc,rx
instructions are used to ensure that the thumb/arm interwork change of
process more works - in essence mov pc,rx needs to become bx rc.
The ldr pc or ldm rx, {pc} instructions are not changed - this is
fine on ARM >=v5 but will fail to restore thumb mode on ARM v4T,
i.e. this code will not provide support for thumb on ARM v4T.
One mov pc is left in resolve.S, this is fixed in a different patch -
thumb-resolve.patch
The changes are protected by __THUMB_INTERWORK__ - the original
mov instruction will work on newer architectures and is required on
arch v4 (not v4t) and earlier - those which did not support thumb -
so this is safe. See gcc lib1asmfuncs for a more exact test.
Mike Frysinger [Tue, 31 Jan 2006 00:26:22 +0000 (00:26 -0000)]
John Bowler writes in Bug 385:
in various places defined(__arm__) is used to protect/select code which
is ARM specific, that code must also be selected for __thumb__ because
__thumb__ is an ARM but __arm__ is not set...
Peter S. Mazinger [Mon, 30 Jan 2006 22:48:02 +0000 (22:48 -0000)]
strlen is -cheaper- then strnlen
Peter S. Mazinger [Mon, 30 Jan 2006 21:35:42 +0000 (21:35 -0000)]
Add IMA compiling
"Jan-Benedict Glaw" [Mon, 30 Jan 2006 21:23:23 +0000 (21:23 -0000)]
Assebler sources ment to be preprocessed shouldn't contain assembler
comments, but C comments instead...
Peter S. Mazinger [Mon, 30 Jan 2006 21:01:49 +0000 (21:01 -0000)]
Undo -std=c99 and related -D__ASSEMBLER__
Peter S. Mazinger [Mon, 30 Jan 2006 18:09:59 +0000 (18:09 -0000)]
back to inline, mips fails
Peter S. Mazinger [Mon, 30 Jan 2006 16:06:33 +0000 (16:06 -0000)]
some more prototypes
Peter S. Mazinger [Mon, 30 Jan 2006 16:05:33 +0000 (16:05 -0000)]
make regex a little bit smaller
Peter S. Mazinger [Mon, 30 Jan 2006 15:18:45 +0000 (15:18 -0000)]
another attempt to fix sendfile64
Peter S. Mazinger [Mon, 30 Jan 2006 15:04:09 +0000 (15:04 -0000)]
Sync w/ glibc
Peter S. Mazinger [Mon, 30 Jan 2006 13:03:19 +0000 (13:03 -0000)]
Enable -std=c99 if supported, this will define __STDC_VERSION__, have to -D__ASSEMBLER__, bug in all gcc
Peter S. Mazinger [Mon, 30 Jan 2006 09:39:51 +0000 (09:39 -0000)]
make it build if WCHAR is disabled
Peter S. Mazinger [Mon, 30 Jan 2006 07:51:24 +0000 (07:51 -0000)]
Too many sparc versions ;-(, another hidden bcopy removed
Peter S. Mazinger [Mon, 30 Jan 2006 07:45:03 +0000 (07:45 -0000)]
Remove arch specific hidden bcopy
Peter S. Mazinger [Mon, 30 Jan 2006 07:35:34 +0000 (07:35 -0000)]
Change the only bcopy user to memmove and remove internal version of bcopy
Peter S. Mazinger [Mon, 30 Jan 2006 07:33:45 +0000 (07:33 -0000)]
Why this?
Peter S. Mazinger [Mon, 30 Jan 2006 07:21:58 +0000 (07:21 -0000)]
Using hidden_proto solves this internally, my proposal was wrong
Peter S. Mazinger [Mon, 30 Jan 2006 07:09:18 +0000 (07:09 -0000)]
Use memset instead of bzero
Peter S. Mazinger [Mon, 30 Jan 2006 06:44:51 +0000 (06:44 -0000)]
hidden moved to prototype
Mike Frysinger [Sun, 29 Jan 2006 21:02:08 +0000 (21:02 -0000)]
Patch by Joseph S. Myers to add support for ARM EABI
Mike Frysinger [Sun, 29 Jan 2006 20:29:37 +0000 (20:29 -0000)]
use typeof to create the prototype
Peter S. Mazinger [Sun, 29 Jan 2006 18:08:05 +0000 (18:08 -0000)]
Add back the current glibc copyright notice, although the file is far different
Mike Frysinger [Sun, 29 Jan 2006 16:45:42 +0000 (16:45 -0000)]
rename local i var to idx to get rid of shadow warning
Mike Frysinger [Sun, 29 Jan 2006 16:44:25 +0000 (16:44 -0000)]
remove unneeded double reloc_type definition
Mike Frysinger [Sun, 29 Jan 2006 16:43:10 +0000 (16:43 -0000)]
make sure clone/__clone are created properly