Message ID | Pine.LNX.4.64.1406251318060.27516@digraph.polyomino.org.uk |
---|---|
State | New |
Headers | show |
Ping (last ping for this patch for 2.20). This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is pending review.
Ping^2. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review.
On Wed 25 Jun 2014 13:19:48 Joseph S. Myers wrote: > This patch makes __ASSUME_UTIMES hppa-specific, removing mentions of > the macro from architecture-independent code and code for other > architectures. (All other architectures either have the utimes > syscall in all relevant kernel versions, or use the asm-generic > interface so only have utimensat and won't get the utimes syscall.) A > similar approach is used to that used for futimesat for MicroBlaze: if > the kernel is recent enough that the utimes syscall can be assumed to > be present, use the implementation in terms of the utimes syscall, and > otherwise use the linux/generic implementation in terms of utimensat. > > Tested x86_64 that the disassembly of installed shared libraries is > unchanged by the patch. Not tested for hppa. looks sane to me, but i'm guessing you're looking for Carlos to sign off -mike
Ping^3. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^4. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^5. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^6. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^7. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^8. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^9. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^10. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^11. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^12. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^13. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^14. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^15. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
On Mon, 1 Dec 2014, Joseph Myers wrote: Ping^16. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^17. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
Ping^18. This patch <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still pending review for hppa.
On Mon, Dec 22, 2014 at 01:08:37PM +0000, Joseph Myers wrote: > Ping^18. This patch > <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still > pending review for hppa. > for what good it is, your patch looks fine... though i question the utility of even bothering to wire up utimes() when asm-generic/unistd.h using architectures are never going to have it... it seems hppa'd be better off sticking to their guns and let arm64 deal with it, since that'll have more clout as a non-boat-anchor architecture. --kyle
On 12/22/2014 04:30 PM, Kyle McMartin wrote: > On Mon, Dec 22, 2014 at 01:08:37PM +0000, Joseph Myers wrote: >> Ping^18. This patch >> <https://sourceware.org/ml/libc-alpha/2014-06/msg00758.html> is still >> pending review for hppa. >> > > for what good it is, your patch looks fine... though i question the > utility of even bothering to wire up utimes() when asm-generic/unistd.h > using architectures are never going to have it... it seems hppa'd be > better off sticking to their guns and let arm64 deal with it, since > that'll have more clout as a non-boat-anchor architecture. Thanks for the review. I kept meaning to get to this, but to be honest I'm just barely getting through the -Werror fixes for hppa (and even then had to hack a set of new fans to boot my box). Cheers, Carlos.
diff --git a/sysdeps/unix/sysv/linux/aarch64/kernel-features.h b/sysdeps/unix/sysv/linux/aarch64/kernel-features.h index bd94fe1..b9963e3 100644 --- a/sysdeps/unix/sysv/linux/aarch64/kernel-features.h +++ b/sysdeps/unix/sysv/linux/aarch64/kernel-features.h @@ -29,6 +29,3 @@ #define __ASSUME_SENDMMSG_SYSCALL 1 #include_next <kernel-features.h> - -/* asm-generic architectures do not have the utimes syscall. */ -#undef __ASSUME_UTIMES diff --git a/sysdeps/unix/sysv/linux/hppa/kernel-features.h b/sysdeps/unix/sysv/linux/hppa/kernel-features.h index 25c3e36..cf0c574 100644 --- a/sysdeps/unix/sysv/linux/hppa/kernel-features.h +++ b/sysdeps/unix/sysv/linux/hppa/kernel-features.h @@ -32,9 +32,9 @@ # define __ASSUME_SENDMMSG_SYSCALL 1 #endif -#include_next <kernel-features.h> - -/* hppa did not get the utimes syscall until 3.14. */ -#if __LINUX_KERNEL_VERSION < 0x030e00 -# undef __ASSUME_UTIMES +/* Support for the utimes syscall was added in 3.14. */ +#if __LINUX_KERNEL_VERSION >= 0x030e00 +# define __ASSUME_UTIMES 1 #endif + +#include_next <kernel-features.h> diff --git a/sysdeps/unix/sysv/linux/hppa/utimes.c b/sysdeps/unix/sysv/linux/hppa/utimes.c new file mode 100644 index 0000000..f62dd05 --- /dev/null +++ b/sysdeps/unix/sysv/linux/hppa/utimes.c @@ -0,0 +1,30 @@ +/* Implement utimes for hppa. + Copyright (C) 2014 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <http://www.gnu.org/licenses/>. */ + +/* hppa has the utimensat syscall in all supported kernel versions but + gained the utimes syscall later, so use the linux-generic + implementation of utimes in terms of the utimensat syscall unless + the utimes syscall is known to be available. */ + +#include <kernel-features.h> + +#ifdef __ASSUME_UTIMES +# include <sysdeps/unix/sysv/linux/utimes.c> +#else +# include <sysdeps/unix/sysv/linux/generic/utimes.c> +#endif diff --git a/sysdeps/unix/sysv/linux/kernel-features.h b/sysdeps/unix/sysv/linux/kernel-features.h index 3c8ea50..2de5775 100644 --- a/sysdeps/unix/sysv/linux/kernel-features.h +++ b/sysdeps/unix/sysv/linux/kernel-features.h @@ -72,12 +72,6 @@ /* The statfs64 syscalls are available in 2.5.74 (but not for alpha). */ #define __ASSUME_STATFS64 1 -/* The utimes syscall has been available for some architectures - forever. For x86 it was introduced after 2.5.75, for x86-64, - ppc, and ppc64 it was introduced in 2.6.0-test3, for s390 it was - introduced in 2.6.21-rc5. */ -#define __ASSUME_UTIMES 1 - /* pselect/ppoll were introduced just after 2.6.16-rc1. On x86_64 and SH this appeared first in 2.6.19-rc1, on ia64 in 2.6.22-rc1. */ #define __ASSUME_PSELECT 1 diff --git a/sysdeps/unix/sysv/linux/tile/kernel-features.h b/sysdeps/unix/sysv/linux/tile/kernel-features.h index 78aacac..3dc0dfc 100644 --- a/sysdeps/unix/sysv/linux/tile/kernel-features.h +++ b/sysdeps/unix/sysv/linux/tile/kernel-features.h @@ -28,9 +28,6 @@ #include_next <kernel-features.h> -/* asm-generic architectures do not have the utimes syscall. */ -#undef __ASSUME_UTIMES - /* Define this if your 32-bit syscall API requires 64-bit register pairs to start with an even-number register. */ #define __ASSUME_ALIGNED_REGISTER_PAIRS 1 diff --git a/sysdeps/unix/sysv/linux/utimes.c b/sysdeps/unix/sysv/linux/utimes.c index 2bcebeb..5bea069 100644 --- a/sysdeps/unix/sysv/linux/utimes.c +++ b/sysdeps/unix/sysv/linux/utimes.c @@ -20,40 +20,16 @@ #include <utime.h> #include <sys/time.h> #include <sysdep.h> -#include <kernel-features.h> +/* Consider moving to syscalls.list. */ + /* Change the access time of FILE to TVP[0] and the modification time of FILE to TVP[1]. */ int __utimes (const char *file, const struct timeval tvp[2]) { -#ifdef __NR_utimes - int result = INLINE_SYSCALL (utimes, 2, file, tvp); -# ifndef __ASSUME_UTIMES - if (result != -1 || errno != ENOSYS) -# endif - return result; -#endif - - /* The utimes() syscall does not exist or is not available in the - used kernel. Use utime(). For this we have to convert to the - data format utime() expects. */ -#ifndef __ASSUME_UTIMES - struct utimbuf buf; - struct utimbuf *times; - - if (tvp != NULL) - { - times = &buf; - buf.actime = tvp[0].tv_sec + tvp[0].tv_usec / 1000000; - buf.modtime = tvp[1].tv_sec + tvp[1].tv_usec / 1000000; - } - else - times = NULL; - - return INLINE_SYSCALL (utime, 2, file, times); -#endif + return INLINE_SYSCALL (utimes, 2, file, tvp); } weak_alias (__utimes, utimes)