This patch includes optimized 64bit memcpy/memmove for Atom, Core 2 and
Core i7. It improves memcpy by up to 3X on Atom, up to 4X on Core 2 and
up to 1X on Core i7. It also improves memmove by up to 3X on Atom, up to
4X on Core 2 and up to 2X on Core i7.
This patch adds SSSE3 strcpy/stpcpy. I got up to 4X speed up on Core 2
and Core i7. I disabled it on Atom since SSSE3 version is slower for
shorter (<64byte) data.
* sysdeps/x86_64/dl-trampoline.S (_dl_runtime_profile): Fix
alignement of La_x86_64_regs. Store xmm parameters.
Patch mostly by Jiri Olsa <olsajiri@gmail.com>.
* string/strverscmp.c (__strverscmp): Fix case of different digits
in fractional part of string.
Patch by Jingyu Liu <jyliu@fortinet.com>.
* string/Makefile (tests): Add tst-svc2.
* string/tst-svc2.c: New file.
* string/strverscmp.c (__strverscmp): Optimize size of tables.
* wcsmbs/wchar.h: mbsnrtowcs, open_wmemstream, wcpcpy, wcpncpy,
wcscasecmp, wcsdup, wcsncasecmp, wcsnlen, wcsnrtombs,
wcscasecmp_l, wcsncasecmp_l, wcscoll_l, and wcsxfrm_l.
* sysdeps/mach/hurd/bits/posix_opt.h: Reset value of macros from
200112L to 200809L.
* sysdeps/unix/sysv/linux/bits/posix_opt.h: Likewise.
* posix/getconf.c (vars): Add _SC_THREAD_ROBUST_PRIO_INHERIT and
_SC_THREAD_ROBUST_PRIO_PROTECT entries.
* bits/confname.h: Add _SC_THREAD_ROBUST_PRIO_INHERIT and
_SC_THREAD_ROBUST_PRIO_PROTECT.
* posix/unistd.h: fexecve is in POSIX 2008.
* time/time.h: strftime_l is in POSIX 2008.
* io/sys/stat.h: futimens is in POSIX 2008.
* string/strings.h: strcasecmp_l and strncasecmp_l are in POSIX 2008.
* string/string.h: stpcpy, stpncpy, strndup, strnlen, strsignal,
strcoll_l, strerror_l, and strxfrm_l are in POSIX 2008.
* stdlib/stdlib.h: mkdtemp is in POSIX 2008.
* string/test-memmem.c: New file.
* string/test-string.h (BUF1PAGES): Define to 1 if undefined.
(test_init): Size buf1 according to BUF1PAGES.
* stdio-common/scanf16.c: New file.
* stdio-common/scanf17.c: New file.
* scripts/data/localplt-sparc64-linux-gnu.data: New file.
2008-03-29 Eric Blake <ebb9@byu.net>
Rewrite string searches to O(n) rather than O(n^2).
* string/str-two-way.h: New file. For linear fixed-allocation
string searching.
* string/memmem.c: New implementation.
* string/strstr.c: New implementation.
* string/strcasestr.c: New implementation.
* sysdeps/posix/getaddrinfo.c (getaddrinfo): Call _res_hconf_init
* string/bits/string2.h (__strdup): Cast parameters to calloc to
avoid warning with -Wconversion.
(__strndup): Likewise.
Patch to 50% by Christian Iseli <christian.iseli@licr.org>.
* wcsmbs/wcsnlen.c: Don't reference before checking the length.
* string/stratcliff.c: Make usable to test wide char functions.
* wcsmbs/wcsatcliff.c: New file.
* wcsmbs/Makefiel (tests): Add wcsatcliff.
* misc/bits/syslog.h (syslog): When __va_arg_pack is defined,
implement as __extern_always_inline function.
(vsyslog): Define as __extern_always_inline function unconditionally.
* libio/bits/stdio2.h (sprintf, snprintf, printf, fprintf):
When __va_arg_pack is defined, implement as __extern_always_inline
functions.
(vsprintf, vsnprintf, vprintf, vfprintf): Define as
__extern_always_inline functions unconditionally.
* libio/bits/stdio.h (vprintf): Ifdef out the inline when
bits/stdio2.h will be included.
* wcsmbs/bits/wchar2.h (__swprintf_alias): New redirect.
(swprintf, wprintf, fwprintf): When __va_arg_pack is defined,
implement as __extern_always_inline functions.
(vswprintf, vwprintf, vfwprintf): Define as
__extern_always_inline functions unconditionally.
* debug/tst-chk1.c (do_test): Enable remaining tests for C++.
2007-09-03 Jakub Jelinek <jakub@redhat.com>
* misc/sys/cdefs.h (__extern_inline, __extern_always_inline): Only
define in C++ for GCC 4.3+, in C++ always use __gnu_inline__
attribute.
* include/features.h (__USE_EXTERN_INLINES): Define only when
__extern_inline is defined.
* stdlib/stdlib.h: Include bits/stdlib.h when __extern_always_inline
is defined instead of when not __cplusplus.
* misc/sys/syslog.h: Include bits/syslog.h when __extern_always_inline
is defined instead of when not __cplusplus.
* socket/sys/socket.h: Include bits/socket2.h when
__extern_always_inline is defined instead of when not __cplusplus.
* libio/stdio.h: Include bits/stdio2.h when __extern_always_inline
is defined instead of when not __cplusplus.
* posix/unistd.h: Include bits/unistd.h when __extern_always_inline
is defined instead of when not __cplusplus.
* string/string.h: Include bits/string3.h when __extern_always_inline
is defined instead of when not __cplusplus.
* wcsmbs/wchar.h: Include bits/wchar2.h when __extern_always_inline
is defined instead of when not __cplusplus.
(btowc, wctob): Don't guard the inlines with ifndef __cplusplus.
* io/fcntl.h: Don't include bits/fcntl2.h if __extern_always_inline
is not defined.
* misc/bits/syslog-ldbl.h: Guard *_chk stuff with
defined __extern_always_inline instead of !defined __cplusplus.
* libio/bits/stdio-ldbl.h: Likewise.
* wcsmbs/bits/wchar-ldbl.h: Likewise.
* misc/bits/syslog.h (syslog): Don't define for C++.
(vsyslog): Use __extern_always_inline function for C++ instead of
a macro.
* libio/bits/stdio.h (__STDIO_INLINE): Define to __extern_inline
whenever that macro is defined.
(vprintf): Don't provide the inline for C++.
(fread_unlocked, fwrite_unlocked): Don't define the macros for C++.
* libio/bits/stdio2.h (sprintf, snprintf, printf, fprintf): Don't
define the macros for C++.
(vsprintf, vsnprintf, vprintf, vfprintf): Define as
__extern_always_inline functions for C++.
* io/sys/stat.h (stat, lstat, fstat, fstatat, mknod, mknodat,
stat64, lstat64, fstat64, fstatat64): Don't define if not
__USE_EXTERN_INLINES.
* wcsmbs/bits/wchar2.h: Fix #error message.
(swprintf, wprintf, fwprintf): Don't define the macros for C++.
(vswprintf, vwprintf, vfwprintf): Define using
__extern_always_inline functions for C++.
* string/bits/string3.h: Don't #undef macros if __cplusplus.
(memcpy, memmove, mempcpy, memset, bcopy, bzero, strcpy, stpcpy,
strncpy, strcat, strncat): Define as __extern_always_inline
functions instead of macros for C++.
* math/bits/cmathcalls.h: Guard __extern_inline routines with
defined __extern_inline.
* sysdeps/alpha/fpu/bits/mathinline.h (__MATH_INLINE): Define
to __extern_inline whenever that macro is defined.
* sysdeps/ia64/fpu/bits/mathinline.h (__MATH_INLINE): Likewise.
* sysdeps/i386/fpu/bits/mathinline.h (__MATH_INLINE): Likewise.
* sysdeps/i386/i486/bits/string.h (__STRING_INLINE): Likewise.
* sysdeps/s390/bits/string.h (__STRING_INLINE): Likewise.
* sysdeps/s390/fpu/bits/mathinline.h (__MATH_INLINE): Likewise.
* sysdeps/powerpc/fpu/bits/mathinline.h (__MATH_INLINE): Likewise.
* sysdeps/x86_64/fpu/bits/mathinline.h (__MATH_INLINE): Likewise.
* sysdeps/sparc/fpu/bits/mathinline.h (__MATH_INLINE): Likewise.
* sysdeps/unix/sysv/linux/sys/sysmacros.h (gnu_dev_major,
gnu_dev_minor, gnu_dev_makedev): Remove __extern_inline from
prototypes. Only provide __extern_inline routines if
__USE_EXTERN_INLINES.
* debug/Makefile: Add rules to build and run tst-{,lfs}chk{4,5,6}
tests.
* debug/tst-chk1.c (do_prepare, do_test): Allow compilation as C++.
For now avoid some *printf tests in C++. Skip all testing
if __USE_FORTIFY_LEVEL is defined, but __extern_always_inline macro
is not.
* debug/tst-chk4.cc: New file.
* debug/tst-chk5.cc: New file.
* debug/tst-chk6.cc: New file.
* debug/tst-lfschk4.cc: New file.
* debug/tst-lfschk5.cc: New file.
* debug/tst-lfschk6.cc: New file.
* include/wchar.h (__vfwprintf_chk, __vswprintf_chk): Avoid
prototypes in C++.
* include/stdio.h (__sprintf_chk, __snprintf_chk, __vsprintf_chk,
__vsnprintf_chk, __printf_chk, __fprintf_chk, __vprintf_chk,
__vfprintf_chk, __fgets_unlocked_chk, __fgets_chk): Likewise.
smaller scopes.
* elf/dl-dst.h (DL_DST_REQ_STATIC): Add l as macro argument.
(DL_DST_REQUIRED): Adjust user.
* include/dlfcn.h (struct link_map): New forward decl.
* inet/getnameinfo.c: Include stddef.h.
(getnameinfo): Use offsetof.
* time/tst-mktime2.c (do_test): Don't rely on signed wrap.
* stdio-common/vfprintf.c (_itoa): Undef before redefining.
* string/strerror_l.c: Include stdlib.h.
returning.
(PTR_DEMANGLE): Real definition now that it's not the same as
PRT_MANGLE anymore.
* sysdeps/unix/sysv/linux/x86_64/sysdep.h: Likewise.
* string/strerror_l.c: New file.
* string/Makefile (routines): Add strerror_l.
* string/string.h: Declare strerror_l.
* string/Versions: Export strerror_l for GLIBC_2.6.
* nscd/connections.c (nscd_run): Don't define MAXKEYLEN here.
if N is one bigger than return value.
* string/tst-strxfrm2.c (do_test): Also test strxfrm with l1 + 1
and l1 last arguments, if buf is defined, verify the return value
equals to strlen (buf) and verify no byte beyond passed length
is modified.
look further, return original strings.
(_nl_find_msg): Do not return found translation if the conversion
failed. Either signal the string is unusable or that something went
wrong and the original should be used.
2006-06-21 Ulrich Drepper <drepper@redhat.com>
* string/_strerror.c (__strerror_r): Add __builtin_expect.
2006-06-02 Ryan S. Arnold <rsa@us.ibm.com>
[BZ #2703]
* string/envz.c (envz_strip): Correct erroneously reversed src
and dest parameters to memmove() invocation.
[Coverity CID 229, 230]
* argp/argp-help.c (hol_entry_help): Handle STATE==NULL in ARG and
DGETTEXT calls.
(hol_help): Likewise. [Coverity CID 226, 227]
* string/argz-replace.c (__argz_replace): Unconditionally call
free on SRC. [Coverity CID 225]
* nis/nis_creategroup.c (nis_creategroup): No need to duplicate
the return value of __nis_default_owner and __nis_default_group,
it has been especially allocated. [Coverity CID 224]
* include/bits/wchar2.h: New file.
* wcsmbs/wchar.h: Include <bits/wchar2.h> if fortification is
requested.
* wcsmbs/wcsncpy.c: Add __wcsncpy alias.
* string/bits/string3.h: Add fortified stpncpy definitions.
* sysdeps/generic/stpncpy_chk.c: New file.
* libio/vswprintf.c: Move _IO_wstrnfile definition to strfile.h.
Export _IO_wstrn_jumps.
* libio/strfile.h: Define _IO_wstrnfile and declare _IO_wstrn_jumps.
* include/wchar.h: Declare __wcsncpy and __vswprintf_chk.
* debug/fgetws_chk.c: New file.
* debug/fgetws_u_chk.c: New file.
* debug/fwprintf_chk.c: New file.
* debug/swprintf_chk.c: New file.
* debug/vfwprintf_chk.c: New file.
* debug/vswprintf_chk.c: New file.
* debug/vwprintf_chk.c: New file.
* debug/wcpcpy_chk.c: New file.
* debug/wcpncpy_chk.c: New file.
* debug/wcscat_chk.c: New file.
* debug/wcscpy_chk.c: New file.
* debug/wcsncat_chk.c: New file.
* debug/wcsncpy_chk.c: New file.
* debug/wmemcpy_chk.c: New file.
* debug/wmemmove_chk.c: New file.
* debug/wmempcpy_chk.c: New file.
* debug/wmemset_chk.c: New file.
* debug/wprintf_chk.c: New file.
* debug/tst-chk1.c: Add tests for new functions.
* debug/Versions: Export new functions.
* debug/Makefile (routines): Add new functions.
of internal.
(__hidden_def1, __hidden_dot_def1): Remove.
(__hidden_def2, __hidden_def3): Remove.
(__hidden_ver1): New.
(hidden_ver, hidden_def, hidden_weak): Use it.
(hidden_data_ver, hidden_data_ver, hidden_data_weak): Use non-data
version of the macro.
* include/wchar.h (__wcscoll): Remove.
* wcsmbs/wcscoll.c: Define wcscoll directly instead of via __wcscoll.
* string/strcoll.c: Don't issue libc_hidden_def STRCOLL redefined.
determine the call will never trigger a failure.
* sysdeps/i386/i686/memset_chk.S: Remove alias and warning.
* sysdeps/x86_64/memset_chk.S: Likewise.
2005-02-24 Roland McGrath <roland@redhat.com>
* debug/Versions (libc: GLIBC_2.4): Remove
__memset_zero_constant_len_parameter.
* sysdeps/generic/memset_chk.c: Remove alias and warning.
* misc/sys/cdefs.h (__warndecl): New macro.
* debug/warning-nop.c: New file.
* string/bits/string3.h (memset): Call __warn_memset_zero_len with no
arguments, instead of calling __memset_zero_constant_len_parameter.
Use __warndecl for __warn_memset_zero_len.
* debug/Makefile (routines): Add $(static-only-routines).
(static-only-routines): New variable.
map if requested.
* debug/chk_fail.c: Request backtrace and memory map dump.
* Versions.def: Add GLIBC_2.4 for libc.
* debug/fgets_chk.c: New file.
* debug/fgets_u_chk.c: New file.
* debug/getcwd_chk.c: New file.
* debug/getwd_chk.c: New file.
* debug/readlink_chk.c: New file.
* debug/read_chk.c: New file.
* debug/pread_chk.c: New file.
* debug/pread64_chk.c: New file.
* debug/recv_chk.c: New file.
* debug/recvfrom_chk.c: New file.
* debug/Versions: Add all new functions with version GLIBC_2.4.
* debug/Makefile (routines): Add fgets_chk, fgets_u_chk, read_chk,
pread_chk, pread64_chk, recv_chk, recvfrom_chk, readlink_chk,
getwd_chk, and getcwd_chk. Plus appropriate CFLAGS definitions.
* debug/tst-chk1.c: Add more tests.
* libio/bits/stdio2.h: Add macros for fgets and fgets_unlocked.
* include/stdio.h: Declare __fgets_chk and __fgets_unlocked_chk.
* posix/unistd.h: Include <bits/unistd.h> for fortification.
* posix/bits/unistd.h: New file.
* posix/Makefile (headers): Add bits/unistd.h.
* socket/sys/socket.h: Include <bits/socket2.h> for fortification.
* socket/bits/socket2.h: New file.
* socket/Makefile (headers): Add bits/socket2.h.
* string/bits/string3.h: Extend memset macro to check for zero 3rd
parameter and use __memset_zero_constant_len_parameter in that case.
* sysdeps/generic/memset_chk.c: Add
__memset_zero_constant_len_parameter alias and linker warning.
* debug/Versions: Add __memset_zero_constant_len_parameter to libc
with version GLIBC_2.4.
* sysdeps/generic/bits/types.h: Don't unnecessarily use __extension__
in __STD_TYPE definition.
2005-02-21 Jakub Jelinek <jakub@redhat.com>
* malloc/malloc.c (malloc_printerr): If MALLOC_CHECK_={5,7}, print
the error message rather than program name.
2005-02-21 Ulrich Drepper <drepper@redhat.com>
2004-05-26 Jakub Jelinek <jakub@redhat.com>
* include/string.h (mempcpy, stpcpy): Add libc_hidden_builtin_proto.
* string/bits/string2.h (memset): Disable macro for GCC 3.0+.
(__mempcpy): Use __builtin_mempcpy for GCC 3.4+.
(strchr): For GCC 3.2+, only use __rawmemchr if second argument is
constant '\0' and first argument is not constant.
(__stpcpy): Use __builtin_stpcpy for GCC 3.4+.
(strncpy): Remove #ifdef _USE_STRING_ARCH_mempcpy variant.
For GCC 3.2+ use __builtin_strncpy.
(strncat): For GCC 3.2+ use __builtin_strncat.
(strcmp): For GCC 3.2+ use __builtin_strcmp if both arguments are
constant.
(strcspn, strspn, strpbrk): For GCC 3.2+, use builtin function
if both arguments are constant.
2004-05-26 Ulrich Drepper <drepper@redhat.com>
* nss/nss_files/files-hosts.c: Fix condition for looking up IPv4
mapped addresses in gethostbyaddr.
2004-03-22 Jakub Jelinek <jakub@redhat.com>
* sysdeps/unix/sysv/linux/sparc/sparc32/getpagesize.c
(__getpagesize): Avoid warning about writing into read-only memory.
* string/Makefile (routines): Add xpg-strerror.
* string/string.h (strerror_r): If __USE_XOPEN2K but not __USE_GNU,
redirect strerror_r to __xpg_strerror_r.
* string/Versions (libc): Add __xpg_strerror_r@@GLIBC_2.3.4.
* sysdeps/generic/xpg-strerror.c: New file.
* sysdeps/mach/xpg-strerror.c: New file.
2004-03-14 Ulrich Drepper <drepper@redhat.com>
Make the non-_l functions wrappers around the _l functions.
* include/monetary.h: Declare __vstrmon_l.
* include/string.h: Add libc_hidden_proto for __strcoll_l and
__strxfrm_l.
* include/time.h: Define ptime_locale_status. Declare
__strptime_internal.
* include/wchar.h: Add libc_hidden_proto for __wcscoll_l and
__wcsxfrm_l.
* stdlib/strfmon.c: Move the code to strfmon_l.c. Add little
wrapper around __vstrfmon_l.
* stdlib/strfmon_l.c: Add real implementation. Split into new
function __vstrfmon_l to allow calling it from strfmon.
* stdlib/strtod.c: Move real code to strtod_l.c and add wrapper.
* stdlib/strtod_l.c: Add real implementation.
* stdlib/strtof.c: Adjust to changed strtod.c.
* stdlib/strtof_l.c: Include strtod_l.c now.
* stdlib/strtold.c: New file.
* stdlib/strtold_l.c: Removed.
* string/strcoll.c: Move real code to strcoll_l.c: Add wrapper.
* string/strcoll_l.c: Add real implementation.
* string/strxfrm.c: Move real code to strxfrm_l.c: Add wrapper.
* string/strxfrm_l.c: Add real implementation.
* sysdeps/generic/strtol.c: Move real implementation to strtol_l.c.
Add wrappers.
* sysdeps/generic/strtol_l.c: Add real implementation.
* sysdeps/generic/strtold.c: Removed.
* sysdeps/generic/strtold_l.c: New file.
* sysdeps/generic/strtoll_l.c: Include strtol_l.c now. Adjust
#defines.
* sysdeps/generic/strtoul_l.c: Likewise.
* sysdeps/generic/strtoull_l.c: Likewise.
* sysdeps/generic/wcstol_l.c: Likewise.
* sysdeps/generic/wcstoll_l.c: Likewise.
* sysdeps/generic/wcstoul_l.c: Likewise.
* sysdeps/generic/wcstoull_l.c: Likewise.
* sysdeps/ieee754/ldbl-128/strtold.c: Removed.
* sysdeps/ieee754/ldbl-128/strtold_l.c: New file.
* sysdeps/ieee754/ldbl-96/strtold.c: Removed.
* sysdeps/ieee754/ldbl-96/strtold_l.c: New file.
* sysdeps/m68k/strtold.c: Removed.
* sysdeps/m68k/strtold_l.c: New file.
* time/strftime.c: Move real code to strftime_l.c. Add wrapper.
* time/strftime_l.c: Add real implementation.
* time/strptime.c: Move real code to strptime_l.c. Add wrapper.
* time/strptime_l.c: Add real implementation.
* time/wcsftime.c: Simplify since only wrappers are defined in
strftime.c.
* time/wcsftime_l.c: Include strftime_l.c.
* wcsmbs/wcscoll.c: Simplify since the file is not used by wcscoll_l.c
anymore.
* wcsmbs/wcscoll_l.c: Include strcoll_l.c.
* wcsmbs/wcsxfrm.c: Simplify since the file is not used by wcsxfrm_l.c
anymore.
* wcsmbs/wcsxfrm_l.c: Include strxfrm_l.c.
* wcsmbs/wcstod.c: Prepare to include new strtod.c.
* wcsmbs/wcstod_l.c: Include strtod_l.c.
* wcsmbs/wcstof.c: Prepare to include new strtof.c.
* wcsmbs/wcstof_l.c: Include strtof_l.c.
* wcsmbs/wcstold.c: Prepare to include new strtold.c.
* wcsmbs/wcstold_l.c: Include strtold_l.c.
* locale/uselocale.c: Use _NL_CURRENT_LOCALE instead of __libc_tsd_get.
* sysdeps/generic/strcasecmp.c: Optimize a bit. It's better to get
a reference to the current locale and then use the _l functions.
* sysdeps/generic/strncase.c: Likewise.
More cancellation handling fixups.
* sysdeps/unix/sysv/linux/not-cancel.h: Add waitpid_not_cancel.
* sysdeps/generic/not-cancel.h: Likewise.
* catgets/open_catalog.c: Use not-cancelable syscalls.
* time/Makefile (CFLAGS-getdate.c): Add -fexceptions.
* sysdeps/unix/sysv/linux/llseek.c: Must not be cancelable.
* sysdeps/unix/syscalls.list: Don't mark lseek as cancelable.
* dlfcn/dlfcn.h: Mark dlopen with __THROW again.
* io/fcntl.h: Don't mark posix_fallocate with __THROW.
* libio/fileops.c: Use not-cancelable syscalls for fclose.
* libio/iopopen.c: Use no-cancelable syscalls.
* libio/stdio.h: Mark popen and pclose with __THROW again.
* misc/Makefile (CFLAGS-syslog.c): Add -fexceptions.
* misc/syslog.c: Fix locking and cancellation cleanup handling.
* posix/unistd.h: Mark ttyname and ttyname_r again with __THROW.
* stdio-common/Makefile (CFLAGS-tmpfile.c, CFLAGS-tmpfile64.c,
CFLAGS-tempname.c): Add -fexceptions.
* stdlib/Makefile (CFLAGS-mkstemp.c): Add -fexceptions.
* string/string.h: Mark strerror and strerror_r with _THROW again.
* sysdeps/generic/unwind.inc: New file. Copied from gcc.
* sysdeps/generic/unwind-dw2.c: Update from gcc version. Remove
#ifs since we now need all the code compiled.
* sysdeps/posix/spawni.c: Use close_not_cancel instead of close.
* sysdeps/unix/closedir.c: Use not-cancelable syscalls.
* sysdeps/unix/opendir.c: Likewise.
has been promoted to wordsize if the ABI requires caller to do so.
* string/test-strncmp.c (do_random_tests): Likewise.
* string/test-memcmp.c (do_random_tests): Likewise.
* sysdeps/powerpc/powerpc64/strcmp.S (strcmp): Sign extend rRTN
before returning.
