From: "Carlos O'Donell" <carlos@redhat.com>
To: Bernie Ogden <bernie.ogden@linaro.org>, libc-ports@sourceware.org
Subject: Re: [PATCH] Remove hppa lowlevellock.c
Date: Mon, 28 Apr 2014 17:02:00 -0000 [thread overview]
Message-ID: <535E899C.30509@redhat.com> (raw)
In-Reply-To: <CALE0ps19HNFvdrD0j0BPhx6tyBW+DuyKbYEMXJ513FqUeYPDEg@mail.gmail.com>
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.
next prev parent reply other threads:[~2014-04-28 17:02 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-04-28 15:51 Bernie Ogden
2014-04-28 17:02 ` Carlos O'Donell [this message]
2014-04-29 9:14 ` Carlos O'Donell
2014-06-09 20:29 ` Bernard Ogden
2014-06-10 3:25 ` Carlos O'Donell
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=535E899C.30509@redhat.com \
--to=carlos@redhat.com \
--cc=bernie.ogden@linaro.org \
--cc=libc-ports@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).