OSDN Git Service
Mike Frysinger [Thu, 28 Jul 2005 04:46:07 +0000 (04:46 -0000)]
the common sem.h is good enough for x86_64
Mike Frysinger [Thu, 28 Jul 2005 03:59:10 +0000 (03:59 -0000)]
check the return value of the clone call too
Mike Frysinger [Thu, 28 Jul 2005 03:58:43 +0000 (03:58 -0000)]
the errno settings was fixed but the return value was still being clobbered ... fix that too
Mike Frysinger [Thu, 28 Jul 2005 03:58:03 +0000 (03:58 -0000)]
tweak $TOPDIR usage to not have duplciate //
Manuel Novoa III [Thu, 28 Jul 2005 03:36:45 +0000 (03:36 -0000)]
Fix string.h. Change our _susv3_strerror_r to __xpg_strerror_r for glibc
compatibility. Also change _glibc_strerror_r to __glibc_strerror_r for
consistency.
Manuel Novoa III [Thu, 28 Jul 2005 03:34:44 +0000 (03:34 -0000)]
Add a config option for abort() to shutdown the stdio subsystem. This is
mainly to cut down on noise in the NIST/PCTS tests since older POSIX
behavior was to fclose() (and hence fflush()) all open streams.
Manuel Novoa III [Thu, 28 Jul 2005 03:30:57 +0000 (03:30 -0000)]
A while back, the sysvipc code was changed to use IPC_64 in the calls.
Unfortunatly, the semid_ds structs were not changed as needed. So
import the necessary (tweaked) files from glibc. Only tested on x86
today, but I'll be testing mips tomorrow.
Joakim Tjernlund [Wed, 27 Jul 2005 13:03:13 +0000 (13:03 -0000)]
Fix bug 358 by protecting _dl_symbol_tables as suggested.
Mike Frysinger [Wed, 27 Jul 2005 02:02:09 +0000 (02:02 -0000)]
add some asm magic to fix setting errno/ret values in the .S functions
Mike Frysinger [Wed, 27 Jul 2005 01:53:47 +0000 (01:53 -0000)]
dont run ls and make sure diff aborts if it fails
Mike Frysinger [Wed, 27 Jul 2005 01:44:05 +0000 (01:44 -0000)]
add a basic errno test based on one from ltp
Mike Frysinger [Tue, 26 Jul 2005 22:26:49 +0000 (22:26 -0000)]
update notes about alpha/x86_64
Eric Andersen [Tue, 26 Jul 2005 21:51:46 +0000 (21:51 -0000)]
Update the todo list for 0.9.28
Manuel Novoa III [Mon, 25 Jul 2005 03:50:55 +0000 (03:50 -0000)]
Fix 2 bugs in strftime related to glibc struct tm extensions.
1) Need to negate tm_gmtoff field value when used. (bug 336).
2) Deal with NULL ptr case for tm_zone field, which was causing
segfaults in both the NIST/PCTS tests and the Python 2.4.1
self-test suite.
NOTE: We set uninitialized timezone names to "???", and this
differs (intentionally) from glibc's behavior.
Manuel Novoa III [Sat, 23 Jul 2005 14:02:57 +0000 (14:02 -0000)]
Since __syscall_error is a C func, we need to store the syscall return value
in the appropriate register. Otherwise, errno is set to random garbage.
Mike Frysinger [Sun, 17 Jul 2005 00:54:46 +0000 (00:54 -0000)]
whoops, reversed naming of nano sec members ... kernel uses _, user space does not
Mike Frysinger [Sun, 17 Jul 2005 00:52:56 +0000 (00:52 -0000)]
syntax cleanups
Mike Frysinger [Sun, 17 Jul 2005 00:52:11 +0000 (00:52 -0000)]
tweak whitespace
Mike Frysinger [Sun, 17 Jul 2005 00:51:53 +0000 (00:51 -0000)]
rip out previous armeb old-stat abi compat code since arm linux opinion is that the old stat.h abi was for little endian only ... people should use this patch to fix their armeb kernels: arm.linux.org.uk/developer/patches/viewpatch.php?id=863/1
Mike Frysinger [Sun, 17 Jul 2005 00:49:38 +0000 (00:49 -0000)]
add support for nanoseconds
Mike Frysinger [Sat, 16 Jul 2005 04:01:17 +0000 (04:01 -0000)]
make sure the stat64 struct is packed to prevent further ABI headaches as done in mainline 2.6
Mike Frysinger [Sat, 16 Jul 2005 03:27:44 +0000 (03:27 -0000)]
with the help of John Bowler, track down the ugly ABI change between 2.4 / 2.6 on big endian arm kernels and work around it best we can at runtime
Mike Frysinger [Fri, 15 Jul 2005 22:58:11 +0000 (22:58 -0000)]
add support for passing nano seconds (if arch supports it)
Mike Frysinger [Fri, 15 Jul 2005 22:27:21 +0000 (22:27 -0000)]
allow people to build shared/ldso but disable threads
Mike Frysinger [Thu, 14 Jul 2005 04:17:11 +0000 (04:17 -0000)]
a [mostly working] elfinterp.c based on i386 / glibc
Mike Frysinger [Thu, 14 Jul 2005 04:16:21 +0000 (04:16 -0000)]
when running shared tests, use the toplevel ldso
Mike Frysinger [Thu, 14 Jul 2005 00:19:41 +0000 (00:19 -0000)]
remove unused instructions
Mike Frysinger [Thu, 14 Jul 2005 00:04:54 +0000 (00:04 -0000)]
import the rest of the glibc start.S so PIE works
Mike Frysinger [Wed, 13 Jul 2005 04:13:03 +0000 (04:13 -0000)]
eat pointless whitespace
Mike Frysinger [Wed, 13 Jul 2005 04:12:51 +0000 (04:12 -0000)]
dump the address we transfer control to
Mike Frysinger [Wed, 13 Jul 2005 03:10:28 +0000 (03:10 -0000)]
oops, fix typo in ELF_ST_BIND macros
Mike Frysinger [Wed, 13 Jul 2005 01:35:00 +0000 (01:35 -0000)]
use Elf32_Word instead of uint32_t since that is what the ELF spec refers to
Mike Frysinger [Tue, 12 Jul 2005 04:23:17 +0000 (04:23 -0000)]
use wrapper debug macros to improve readability
Mike Frysinger [Tue, 12 Jul 2005 01:43:06 +0000 (01:43 -0000)]
import fix from glibc to resolve segfault reported by wmq in Bug 335
Mike Frysinger [Mon, 11 Jul 2005 23:27:08 +0000 (23:27 -0000)]
dont use void* when we know the type of pointer we want
Mike Frysinger [Mon, 11 Jul 2005 23:25:00 +0000 (23:25 -0000)]
force hash buckets to work off of 32bit quantities
Mike Frysinger [Mon, 11 Jul 2005 22:31:34 +0000 (22:31 -0000)]
use the debug dprint macros to clean up readability
Mike Frysinger [Mon, 11 Jul 2005 22:29:29 +0000 (22:29 -0000)]
use the debug dprint macros to clean up readability
Mike Frysinger [Mon, 11 Jul 2005 22:28:40 +0000 (22:28 -0000)]
create a new wrapper macro for debug early output
Mike Frysinger [Mon, 11 Jul 2005 22:14:53 +0000 (22:14 -0000)]
minor touchups in debug output/comments
Mike Frysinger [Mon, 11 Jul 2005 22:13:30 +0000 (22:13 -0000)]
base __rtld_stack_end off of argv instead of args so we dont have to worry about how each arch treats args
Mike Frysinger [Mon, 11 Jul 2005 22:08:35 +0000 (22:08 -0000)]
declare reloc_addr as ElfW(Addr) instead of Word since it is an Addr, not a Word (and it will truncate addresses on 64bit hosts)
Mike Frysinger [Mon, 11 Jul 2005 22:04:57 +0000 (22:04 -0000)]
make sure the SEND_NUMBER_STDERR macro doesnt modify the number given to it
Mike Frysinger [Sun, 10 Jul 2005 07:21:07 +0000 (07:21 -0000)]
add some info on what needs to be done for specific arches
Mike Frysinger [Sun, 10 Jul 2005 07:15:35 +0000 (07:15 -0000)]
need syscall.h for __NR_clone definition
Mike Frysinger [Sun, 10 Jul 2005 06:59:41 +0000 (06:59 -0000)]
define an abort instruction for m68k
Mike Frysinger [Sun, 10 Jul 2005 06:57:25 +0000 (06:57 -0000)]
define an abort instruction for alpha
Mike Frysinger [Sun, 10 Jul 2005 06:54:30 +0000 (06:54 -0000)]
alphabetize and warn if an arch is undefined
Mike Frysinger [Sun, 10 Jul 2005 06:44:02 +0000 (06:44 -0000)]
merge updates from glibc and fix up to work again
Mike Frysinger [Sun, 10 Jul 2005 06:25:50 +0000 (06:25 -0000)]
update __UCLIBC_CTOR_DTOR__ check
Mike Frysinger [Sat, 9 Jul 2005 16:51:22 +0000 (16:51 -0000)]
punt unused __pthread_return_void func
Mike Frysinger [Sat, 9 Jul 2005 00:57:49 +0000 (00:57 -0000)]
import the __syscall_error trick for errno handling
Mike Frysinger [Sat, 9 Jul 2005 00:54:07 +0000 (00:54 -0000)]
this should be able to handle both 32 and 64 bit sparc
Mike Frysinger [Sat, 9 Jul 2005 00:52:25 +0000 (00:52 -0000)]
drop the envp code
Mike Frysinger [Sat, 9 Jul 2005 00:47:30 +0000 (00:47 -0000)]
style updates
Mike Frysinger [Sat, 9 Jul 2005 00:36:34 +0000 (00:36 -0000)]
rewrite crt in asm using glibc as a basis
Mike Frysinger [Sat, 9 Jul 2005 00:34:34 +0000 (00:34 -0000)]
update style
Mike Frysinger [Sat, 9 Jul 2005 00:32:45 +0000 (00:32 -0000)]
make sure we use long instead of int so we dont truncate 64bit values
Mike Frysinger [Sat, 9 Jul 2005 00:31:31 +0000 (00:31 -0000)]
we need the syscall6() version of mmap
Mike Frysinger [Sat, 9 Jul 2005 00:30:56 +0000 (00:30 -0000)]
the generic mmap.c is not good enough for us, we need the syscall6() version
Mike Frysinger [Sat, 9 Jul 2005 00:30:03 +0000 (00:30 -0000)]
might as well test mmap64() too
Mike Frysinger [Sat, 9 Jul 2005 00:29:15 +0000 (00:29 -0000)]
make sure the call to signal() worked
Mike Frysinger [Sat, 9 Jul 2005 00:28:32 +0000 (00:28 -0000)]
need unistd/string header files
Mike Frysinger [Sat, 9 Jul 2005 00:28:20 +0000 (00:28 -0000)]
need the ioctl header file
Mike Frysinger [Sat, 9 Jul 2005 00:27:42 +0000 (00:27 -0000)]
setting envp is pointless since no other arch does it
Mike Frysinger [Fri, 8 Jul 2005 23:37:31 +0000 (23:37 -0000)]
oops, we want to jmp to the syscall handler, not call it
Mike Frysinger [Fri, 8 Jul 2005 05:26:20 +0000 (05:26 -0000)]
add some more useful debug output
Mike Frysinger [Fri, 8 Jul 2005 05:18:41 +0000 (05:18 -0000)]
err, x86_64 is RELA, not REL
Mike Frysinger [Fri, 8 Jul 2005 04:39:36 +0000 (04:39 -0000)]
need _DEBUG macros for NUMBER too
Mike Frysinger [Fri, 8 Jul 2005 04:00:05 +0000 (04:00 -0000)]
CRT0_ -> CRT_
Mike Frysinger [Fri, 8 Jul 2005 03:13:40 +0000 (03:13 -0000)]
whitespace clean up
Mike Frysinger [Fri, 8 Jul 2005 02:38:05 +0000 (02:38 -0000)]
help if i named the var correctly ...
Mike Frysinger [Fri, 8 Jul 2005 02:37:53 +0000 (02:37 -0000)]
help if i named the var correctly ...
Mike Frysinger [Fri, 8 Jul 2005 02:37:19 +0000 (02:37 -0000)]
dl-debug.h for sparc
Mike Frysinger [Fri, 8 Jul 2005 02:36:24 +0000 (02:36 -0000)]
dl-debug.h for m68k
Mike Frysinger [Fri, 8 Jul 2005 02:01:24 +0000 (02:01 -0000)]
fix GET_ARGV() so it pulls out the argc/argv/envp properly
Mike Frysinger [Thu, 7 Jul 2005 23:45:58 +0000 (23:45 -0000)]
get rid of arch-specific defines in dl-startup.c and make each arch declare its requirements about initial bootstrapping
Mike Frysinger [Thu, 7 Jul 2005 23:41:49 +0000 (23:41 -0000)]
add debug code for showing argc/argv/envp and touch up some other debug output
Mike Frysinger [Thu, 7 Jul 2005 23:35:39 +0000 (23:35 -0000)]
clean up whitespace
Mike Frysinger [Thu, 7 Jul 2005 23:34:29 +0000 (23:34 -0000)]
remove broken asm code in elf_machine_dynamic() and just do it the glibc way (since it works)
Mike Frysinger [Thu, 7 Jul 2005 23:30:36 +0000 (23:30 -0000)]
clean up from psm ... merge glibc updates and punt errno handling in favor of jockes trick of using __syscall_error
Mike Frysinger [Thu, 7 Jul 2005 23:29:52 +0000 (23:29 -0000)]
dont need errno.h anymore
Mike Frysinger [Thu, 7 Jul 2005 23:29:25 +0000 (23:29 -0000)]
add update from psm to support 6 arguments to syscall
Mike Frysinger [Thu, 7 Jul 2005 05:19:33 +0000 (05:19 -0000)]
update syntax
Mike Frysinger [Thu, 7 Jul 2005 05:13:25 +0000 (05:13 -0000)]
Yoshinori Sato writes:
Dont forgive that length of munmap is "0" in current nommu.
This purpose cannot free memory area really.
It is a patch to work out a solution.
Mike Frysinger [Thu, 7 Jul 2005 04:46:29 +0000 (04:46 -0000)]
make RUNPATH/RPATH support configurable
Mike Frysinger [Thu, 7 Jul 2005 00:14:11 +0000 (00:14 -0000)]
rip out all the duplicated debug code and move into a shared file
Mike Frysinger [Wed, 6 Jul 2005 22:47:33 +0000 (22:47 -0000)]
need features.h for the misc attribute defines and uclibc features
Mike Frysinger [Wed, 6 Jul 2005 22:47:10 +0000 (22:47 -0000)]
setup ELF_R_TYPE() wrapper for ELF{32,64}_R_TYPE()
Mike Frysinger [Wed, 6 Jul 2005 22:43:22 +0000 (22:43 -0000)]
update comments/syntax and make sure we include features.h for the hidden define
Mike Frysinger [Wed, 6 Jul 2005 22:40:33 +0000 (22:40 -0000)]
Peter S. Mazinger writes: use the __syscall_error.c trick to handle setting errno
Mike Frysinger [Wed, 6 Jul 2005 22:26:19 +0000 (22:26 -0000)]
err, need to commit the rest of the changes from psm too
Mike Frysinger [Wed, 6 Jul 2005 22:24:34 +0000 (22:24 -0000)]
Peter S. Mazinger writes: we dont need features.h
Mike Frysinger [Wed, 6 Jul 2005 22:22:22 +0000 (22:22 -0000)]
syntax touchups as suggested by Peter S. Mazinger
Mike Frysinger [Wed, 6 Jul 2005 05:33:37 +0000 (05:33 -0000)]
import start of ldso based on glibc
Mike Frysinger [Wed, 6 Jul 2005 05:22:48 +0000 (05:22 -0000)]
make sure ldso.o depends on more files in subdirs
Mike Frysinger [Wed, 6 Jul 2005 01:24:17 +0000 (01:24 -0000)]
use % as the function marker instead of @
Mike Frysinger [Wed, 6 Jul 2005 01:23:08 +0000 (01:23 -0000)]
add support for __MALLOC_GLIBC_COMPAT__ (malloc(0) == NULL) so the malloc test wont fail
Mike Frysinger [Wed, 6 Jul 2005 01:22:24 +0000 (01:22 -0000)]
we need stdio.h for printf()
Mike Frysinger [Wed, 6 Jul 2005 01:22:13 +0000 (01:22 -0000)]
make test failures fatal