Remove hppa lowlevellock.c

lowlevellock.c for hppa differs from the generic lowlevellock.c only in
insignificant ways, so can be removed. I don't have any hppa targets
to work with, so have not been able to test this patch.

The notable differences between the hppa and generic implementations are:

1) Some functions in hppa's lowlevellock.c set futex to 2 if it was 1.
The generic version always sets the futex to 2. As futex can only be
0, 1 or 2 on entry into these functions, the behaviour is equivalent.
(If the futex manages to be 0 on entry then we've just lost an
unlikely fast path out.)
2) hppa places most of the functions in this file in both libc and
libpthread. The generic implementation places only
__lll_lock_wait_private in libc. As these are internal functions I
think that, if this does cause a problem, it would show up at build
time - but I'm new to glibc so prepared to be corrected (and have not
been able to build).

I would be grateful if the maintainer could test/comment.

Regards,

Bernie

ports/ChangeLog.hppa
2014-04-24  Bernard Ogden  <bernie.ogden@linaro.org>

[BZ #15119]
* ports/sysdeps/unix/sysv/linux/hppa/nptl/lowlevellock.c: Remove file.


-

On 04/28/2014 11:51 AM, Bernie Ogden wrote:
> lowlevellock.c for hppa differs from the generic lowlevellock.c only in
> insignificant ways, so can be removed. I don't have any hppa targets
> to work with, so have not been able to test this patch.

Thanks for this.

I'll test this shortly.

> The notable differences between the hppa and generic implementations are:
> 
> 1) Some functions in hppa's lowlevellock.c set futex to 2 if it was 1.
> The generic version always sets the futex to 2. As futex can only be
> 0, 1 or 2 on entry into these functions, the behaviour is equivalent.
> (If the futex manages to be 0 on entry then we've just lost an
> unlikely fast path out.)

OK.

> 2) hppa places most of the functions in this file in both libc and
> libpthread. The generic implementation places only
> __lll_lock_wait_private in libc. As these are internal functions I
> think that, if this does cause a problem, it would show up at build
> time - but I'm new to glibc so prepared to be corrected (and have not
> been able to build).

I'll test this, but as they are internal functions it probably doesn't
matter but makes the final object size smaller.

> I would be grateful if the maintainer could test/comment.
> 
> Regards,
> 
> Bernie
> 
> ports/ChangeLog.hppa
> 2014-04-24  Bernard Ogden  <bernie.ogden@linaro.org>
> 
> [BZ #15119]
> * ports/sysdeps/unix/sysv/linux/hppa/nptl/lowlevellock.c: Remove file.

Slightly the wrong format.

Should be:

2014-04-24  Bernard Ogden  <bernie.ogden@linaro.org>
 
	[BZ #15119]
	* ports/sysdeps/unix/sysv/linux/hppa/nptl/lowlevellock.c: Remove file.

However, I haven't moved hppa out of ports, I'll do that at the end of
the day or if I get a break of time waiting on something. 
 
> diff --git a/ports/sysdeps/unix/sysv/linux/hppa/nptl/lowlevellock.c
> b/ports/sysdeps/unix/sysv/linux/hppa/nptl/lowlevellock.c
> deleted file mode 100644
> index d61c5d3..0000000
> --- a/ports/sysdeps/unix/sysv/linux/hppa/nptl/lowlevellock.c
> +++ /dev/null
> @@ -1,127 +0,0 @@
> -/* low level locking for pthread library.  Generic futex-using version.
> -   Copyright (C) 2003-2014 Free Software Foundation, Inc.
> -   This file is part of the GNU C Library.
> -   Contributed by Paul Mackerras <paulus@au.ibm.com>, 2003.
> -
> -   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/>.  */
> -
> -#include <errno.h>
> -#include <sysdep.h>
> -#include <lowlevellock.h>
> -#include <sys/time.h>
> -
> -void
> -__lll_lock_wait (lll_lock_t *futex, int private)
> -{
> -  do
> -    {
> -      int oldval = atomic_compare_and_exchange_val_acq (futex, 2, 1);
> -      if (oldval != 0)
> - lll_futex_wait (futex, 2, private);
> -    }
> -  while (atomic_compare_and_exchange_val_acq (futex, 2, 0) != 0);
> -}
> -
> -void
> -__lll_lock_wait_private (lll_lock_t *futex)
> -{
> -  do
> -    {
> -      int oldval = atomic_compare_and_exchange_val_acq (futex, 2, 1);
> -      if (oldval != 0)
> - lll_futex_wait (futex, 2, LLL_PRIVATE);
> -    }
> -  while (atomic_compare_and_exchange_val_acq (futex, 2, 0) != 0);
> -}
> -
> -int
> -__lll_timedlock_wait (lll_lock_t *futex, const struct timespec
> *abstime, int private)
> -{
> -  /* Reject invalid timeouts.  */
> -  if (abstime->tv_nsec < 0 || abstime->tv_nsec >= 1000000000)
> -    return EINVAL;
> -
> -  do
> -    {
> -      struct timeval tv;
> -      struct timespec rt;
> -
> -      /* Get the current time.  */
> -      (void) __gettimeofday (&tv, NULL);
> -
> -      /* Compute relative timeout.  */
> -      rt.tv_sec = abstime->tv_sec - tv.tv_sec;
> -      rt.tv_nsec = abstime->tv_nsec - tv.tv_usec * 1000;
> -      if (rt.tv_nsec < 0)
> - {
> -  rt.tv_nsec += 1000000000;
> -  --rt.tv_sec;
> - }
> -
> -      /* Already timed out?  */
> -      if (rt.tv_sec < 0)
> - return ETIMEDOUT;
> -
> -      /* Wait.  */
> -      int oldval = atomic_compare_and_exchange_val_acq (futex, 2, 1);
> -      if (oldval != 0)
> - lll_futex_timed_wait (futex, 2, &rt, private);
> -    }
> -  while (atomic_compare_and_exchange_bool_acq (futex, 2, 0) != 0);
> -  return 0;
> -}
> -
> -
> -/* These don't get included in libc.so  */
> -#ifdef IS_IN_libpthread
> -int
> -__lll_timedwait_tid (int *tidp, const struct timespec *abstime)
> -{
> -  int tid;
> -
> -  if (abstime->tv_nsec < 0 || abstime->tv_nsec >= 1000000000)
> -    return EINVAL;
> -
> -  /* Repeat until thread terminated.  */
> -  while ((tid = *tidp) != 0)
> -    {
> -      struct timeval tv;
> -      struct timespec rt;
> -
> -      /* Get the current time.  */
> -      (void) __gettimeofday (&tv, NULL);
> -
> -      /* Compute relative timeout.  */
> -      rt.tv_sec = abstime->tv_sec - tv.tv_sec;
> -      rt.tv_nsec = abstime->tv_nsec - tv.tv_usec * 1000;
> -      if (rt.tv_nsec < 0)
> - {
> -  rt.tv_nsec += 1000000000;
> -  --rt.tv_sec;
> - }
> -
> -      /* Already timed out?  */
> -      if (rt.tv_sec < 0)
> - return ETIMEDOUT;
> -
> -      /* Wait until thread terminates.  */
> -      if (lll_futex_timed_wait (tidp, tid, &rt, LLL_SHARED) == -ETIMEDOUT)
> - return ETIMEDOUT;
> -    }
> -
> -  return 0;
> -}
> -#endif
> -
> 

Cheers,
Carlos.

On 04/28/2014 11:51 AM, Bernie Ogden wrote:
> lowlevellock.c for hppa differs from the generic lowlevellock.c only in
> insignificant ways, so can be removed. I don't have any hppa targets
> to work with, so have not been able to test this patch.
> 
> The notable differences between the hppa and generic implementations are:
> 
> 1) Some functions in hppa's lowlevellock.c set futex to 2 if it was 1.
> The generic version always sets the futex to 2. As futex can only be
> 0, 1 or 2 on entry into these functions, the behaviour is equivalent.
> (If the futex manages to be 0 on entry then we've just lost an
> unlikely fast path out.)
> 2) hppa places most of the functions in this file in both libc and
> libpthread. The generic implementation places only
> __lll_lock_wait_private in libc. As these are internal functions I
> think that, if this does cause a problem, it would show up at build
> time - but I'm new to glibc so prepared to be corrected (and have not
> been able to build).
> 
> I would be grateful if the maintainer could test/comment.
> 
> Regards,
> 
> Bernie
> 
> ports/ChangeLog.hppa
> 2014-04-24  Bernard Ogden  <bernie.ogden@linaro.org>
> 
> [BZ #15119]
> * ports/sysdeps/unix/sysv/linux/hppa/nptl/lowlevellock.c: Remove file.

Looks good to me.

hppa's futex implementation should match the generic defaults at all times.

Cheers,
Carlos.

I've completed failed to ping this - sorry.

Would you mind committing?

Thanks,

Bernie

On 29 Apr 2014, at 10:14, Carlos O'Donell <carlos@redhat.com> wrote:

> On 04/28/2014 11:51 AM, Bernie Ogden wrote:
>> lowlevellock.c for hppa differs from the generic lowlevellock.c only in
>> insignificant ways, so can be removed. I don't have any hppa targets
>> to work with, so have not been able to test this patch.
>> 
>> The notable differences between the hppa and generic implementations are:
>> 
>> 1) Some functions in hppa's lowlevellock.c set futex to 2 if it was 1.
>> The generic version always sets the futex to 2. As futex can only be
>> 0, 1 or 2 on entry into these functions, the behaviour is equivalent.
>> (If the futex manages to be 0 on entry then we've just lost an
>> unlikely fast path out.)
>> 2) hppa places most of the functions in this file in both libc and
>> libpthread. The generic implementation places only
>> __lll_lock_wait_private in libc. As these are internal functions I
>> think that, if this does cause a problem, it would show up at build
>> time - but I'm new to glibc so prepared to be corrected (and have not
>> been able to build).
>> 
>> I would be grateful if the maintainer could test/comment.
>> 
>> Regards,
>> 
>> Bernie
>> 
>> ports/ChangeLog.hppa
>> 2014-04-24  Bernard Ogden  <bernie.ogden@linaro.org>
>> 
>> [BZ #15119]
>> * ports/sysdeps/unix/sysv/linux/hppa/nptl/lowlevellock.c: Remove file.
> 
> Looks good to me.
> 
> hppa's futex implementation should match the generic defaults at all times.
> 
> Cheers,
> Carlos.
>

On 06/09/2014 04:29 PM, Bernard Ogden wrote:
> I've completed failed to ping this - sorry.
> 
> Would you mind committing?

Pushed. All discussions now happen on libc-alpha.
This mailing list is legacy now.

Cheers,
Carlos.