Make __ASSUME_UTIMES hppa-specific
Commit Message
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.
2014-06-25 Joseph Myers <joseph@codesourcery.com>
* sysdeps/unix/sysv/linux/kernel-features.h (__ASSUME_UTIMES): Do
not define.
* sysdeps/unix/sysv/linux/utimes.c: Do not include
<kernel-features.h>.
(__utimes) [__NR_utimes]: Make code unconditional.
(__utimes) [!__ASSUME_UTIMES]: Remove conditional code.
* sysdeps/unix/sysv/linux/aarch64/kernel-features.h
(__ASSUME_UTIMES): Do not undefine.
* sysdeps/unix/sysv/linux/tile/kernel-features.h
(__ASSUME_UTIMES): Likewise.
* sysdeps/unix/sysv/linux/hppa/kernel-features.h
(__ASSUME_UTIMES): Define for [__LINUX_KERNEL_VERSION >= 0x030e00]
instead of undefining for [__LINUX_KERNEL_VERSION < 0x030e00].
* sysdeps/unix/sysv/linux/hppa/utimes.c: New file.
Comments
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.
@@ -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
@@ -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>
new file mode 100644
@@ -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
@@ -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
@@ -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
@@ -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)