2000-06-01 Jakub Jelinek <jakub@redhat.com>
* libc/sysdeps/alpha/fpu/bits/mathinline.h: Add __THROW to all
inlines to match prototypes in mathcalls.h.
* libc/sysdeps/i386/fpu/bits/mathinline.h: Likewise.
* libc/sysdeps/m68k/fpu/bits/mathinline.h: Likewise.
* libc/sysdeps/powerpc/fpu/bits/mathinline.h: Likewise.
* libc/sysdeps/sparc/fpu/bits/mathinline.h: Likewise.
* sysdeps/powerpc/dl-start.S: Close open comment.
2000-06-03 Andreas Jaeger <aj@suse.de>
* sysdeps/powerpc/dl-start.S: Close open comment.
2000-06-03 Jakub Jelinek <jakub@redhat.com>
* inet/Versions: Fix typo in rresvport_af export.
2000-06-01 Ulrich Drepper <drepper@redhat.com>
* sysdeps/i386/elf/setjmp.S: Work around change is recent
binutils. gas now emits a jump through the PLT instead of a
relative jump if the jump target is an exported symbol.
2000-05-31 Andreas Jaeger <aj@suse.de>
* sysdeps/unix/sysv/linux/mips/sys/ucontext.h: Use array for
gregset_t.
* sysdeps/unix/sysv/linux/mips/Versions: Export _flush_cache.
Patches by Maciej W. Rozycki <macro@ds2.pg.gda.pl>.
* sysdeps/unix/sysv/linux/mips/ftruncate64.c: Fix prototype of
syscall.
* sysdeps/unix/sysv/linux/mips/truncate64.c: Likewise.
2000-05-29 Jakub Jelinek <jakub@redhat.com>
* inet/rcmd.c (rresvport_af): Use correct port number.
2000-05-26 Andreas Schwab <schwab@suse.de>
* Makeconfig ($(common-objpfx)soversions.mk): Prepend `$lib=' to
entries in all-sonames.
($(common-objpfx)gnu/lib-names.h): Use it for the CPP symbol,
preserving the old name if different for compatibility.
2000-05-28 Ulrich Drepper <drepper@redhat.com>
* manual/arith.texi (Rounding Functions): Fix documentation of trunc.
Patch by Steven G. Johnson <stevenj@gil-galad.mit.edu>.
2000-05-27 Andreas Jaeger <aj@suse.de>
* sysdeps/unix/sysv/linux/i386/pread.c: Removed, we can use the
normal version now.
* sysdeps/unix/sysv/linux/i386/pwrite.c: Likewise.
* sysdeps/unix/sysv/linux/i386/pwrite64.c: Likewise.
* sysdeps/unix/sysv/linux/mips/pread.c: Likewise.
* sysdeps/unix/sysv/linux/mips/pwrite.c: Likewise.
* sysdeps/unix/sysv/linux/arm/pread.c: Likewise.
* sysdeps/unix/sysv/linux/arm/pwrite.c: Likewise.
* sysdeps/unix/sysv/linux/pread.c (__libc_pread): Check
endianness to decide how to pass arguments.
* sysdeps/unix/sysv/linux/pwrite.c (__libc_pwrite): Likewise.
* elf/vismain.c: Use test-skeleton interface.
(TEST_FUNCTION): New macro.
(do_test): Renamed from main.
2000-05-27 Ulrich Drepper <drepper@redhat.com>
* sysdeps/posix/getaddrinfo.c (gaih_inet): If req->ai_family ==
AF_UNSPEC don't allow the IPv6 lookup to return mapped IPv4 addresses.
* nss/nss_files/files-hosts.c (LINE_PARSER): Simplify. Correct
handling of AI_V4MAPPED flag.
* resolv/nss_dns/dns-host.c (_nss_dns_gethostbyname2_r): If lookup
of IPv6 address failed and RES_USE_INET6 bit in _res.options is set,
perform lookup for T_A type and tell getanswer_r to convert the
result with new argument.
(_nss_dns_gethostbyaddr_r): Pass zero as new argument to getanswer_r.
Don't convert addresses from IPv4 to IPv6 here.
(getanswer_r): Take new parameter indicating if mapping from IPv4 to
IPv6 is wanted. Allow qtype and type from reply to disagree if
IPv4 mapping has to be done. Perform mappings if required by the
caller and not if RES_USE_INET6 bit in _res.options is set.
Add several __builtin_expect.
* sysdeps/unix/sysv/linux/ftruncate64.c (ftruncate64): Make order
of arguments dependend on endianness.
* sysdeps/unix/sysv/linux/truncate64.c: Likewise.
2000-05-26 Ulrich Drepper <drepper@redhat.com>
* posix/tst-preadwrite.c: Alloc file being used also for testing
pread64 and pwrite64.
* posix/tst-preadwrite64.c: New file.
* sysdeps/unix/sysv/linux/i386/pread64.c: Removed, we can use the
normal version now.
* sysdeps/unix/sysv/linux/arm/pread64.c: Likewise.
* sysdeps/unix/sysv/linux/arm/pwrite64.c: Likewise.
* sysdeps/unix/sysv/linux/mips/pread64.c: Likewise.
* sysdeps/unix/sysv/linux/mips/pwrite64.c: Likewise.
* sysdeps/unix/sysv/linux/pwrite64.c (__libc_pwrite64): Check
endianness to check how to pass argument.
* sysdeps/unix/sysv/linux/pread64.c (__libc_pread64): Likewise.
2000-05-26 Andreas Jaeger <aj@suse.de>
* elf/ldconfig.h: Update parameter list for process_elf_file.
* sysdeps/unix/sysv/linux/sparc/readelflib.c (process_elf_file):
Add parameter file_length and pass it ot process_elf*file.
* sysdeps/generic/readelflib.c (check_ptr): New.
(process_elf_file): Use check_ptr to check all accesses to the
mmapped file. Add parameter file_length.
* elf/readlib.c (known_libs): Use <gnu/lib-names.h> to specify
library names.
* sunrpc/xdr_intXX_t.c (xdr_uint8_t): Fix conversion.
Closes PR libc/1573, reported by Bradley White
<bww@laurelnetworks.com>.
2000-05-22 Andreas Jaeger <aj@suse.de>
* math/Makefile: Add -D__FAST_MATH__ to CFLAGS-test-ifloat.c,
CFLAGS-test-idouble.c, CFLAGS-test-ildoubl.c.
* manual/math.texi (FP Function Optimizations): Document gcc
-ffast-math behaviour with mathinlines.
* sysdeps/i386/fpu/bits/mathinline.h: Only use save inline
functions unless -ffast-math is given to gcc.
2000-05-24 Ulrich Drepper <drepper@redhat.com>
* locale/programs/ld-collate.c (struct element_t): Add mbseqorder
and wcseqorder members.
(struct locale_collate_t): Likewise.
(collate_finish): Assign collation sequence value to each character.
Create tables for output.
(collate_output): Write out tables with collation sequence information.
* locale/C-collate.c: Provide C locale data for collation sequence
table.
* locale/langinfo.h: Add _NL_COLLATE_COLLSEQMB and
_NL_COLLATE_COLLSEQWC.
* locale/categories.def: Add entries for _NL_COLLATE_COLLSEQMB and
_NL_COLLATE_COLLSEQWC.
* posix/fnmatch.c: Define SUFFIX and WIDE_CHAR_VERSION before
include fnmatch_loop.c.
* posix/fnmatch_loop.c: Don't use strcoll while determining whether
character is matched by range expression. Use collation sequence
table. Outside glibc fall back on simple character value comparison.
