This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH 0/2] Port of lock elision to System/z v2
- From: Dominik Vogt <vogt at linux dot vnet dot ibm dot com>
- To: libc-alpha at sourceware dot org
- Date: Mon, 23 Sep 2013 09:39:45 +0200
- Subject: Re: [PATCH 0/2] Port of lock elision to System/z v2
- Authentication-results: sourceware.org; auth=none
- References: <20130920105925 dot GA374 at linux dot vnet dot ibm dot com> <523C2BC1 dot 2050707 at suse dot com> <20130920111704 dot GA3400 at linux dot vnet dot ibm dot com> <523C3693 dot 3000902 at suse dot com>
- Reply-to: libc-alpha at sourceware dot org
On Fri, Sep 20, 2013 at 01:50:43PM +0200, Andreas Jaeger wrote:
> I see:
>
> 2013-07-02 Andi Kleen <ak@linux.intel.com>
> Hongjiu Lu <hongjiu.lu@intel.com>
>
> * pthread_mutex_lock.c
> (__pthread_mutex_lock): Add lock elision support.
> * pthread_mutex_timedlock.c (pthread_mutex_timedlock): Likewise.
>
>
> So, it names not only the file but also the function changed - nothing
> like this is seen in your change.
Well, did you _look_ at my patch? Why do you claim that _my_
patch changes functions in these files like the Intel patch did,
when it doesn't?
> Also, the first entry contains a text, the rest "Likewise."
Which is wrong for the s390 patch. Elision support is not added
_individually_ in each file, but the complete set of _new_ files
adds elision support. Do I understand that correctly that I am
required to add additional comments that are "syntactically"
correct, even if they are semantically wrong?
--
2013-09-23 Dominik Vogt <vogt@linux.vnet.ibm.com>
* sysdeps/unix/sysv/linux/s390/elision-conf.c:
Add lock elision support for s390.
* sysdeps/unix/sysv/linux/s390/elision-conf.h: Likewise.
* sysdeps/unix/sysv/linux/s390/elision-lock.c: Likewise.
* sysdeps/unix/sysv/linux/s390/elision-timed.c: Likewise.
* sysdeps/unix/sysv/linux/s390/elision-trylock.c: Likewise.
* sysdeps/unix/sysv/linux/s390/elision-unlock.c: Likewise.
* sysdeps/unix/sysv/linux/s390/force-elision.h: Likewise.
* sysdeps/unix/sysv/linux/s390/pthread_mutex_cond_lock.c: Likewise.
* sysdeps/unix/sysv/linux/s390/pthread_mutex_lock.c: Likewise.
* sysdeps/unix/sysv/linux/s390/pthread_mutex_timedlock.c: Likewise.
* sysdeps/unix/sysv/linux/s390/pthread_mutex_trylock.c: Likewise.
* sysdeps/unix/sysv/linux/s390/pthread_mutex_unlock.c: Likewise.
* sysdeps/unix/sysv/linux/s390/bits/pthreadtypes.h (pthread_mutex_t):
Likewise.
* sysdeps/unix/sysv/linux/s390/Makefile: Likewise.
* sysdeps/unix/sysv/linux/s390/htm.h
(__builtin_tbegin, __builtin_tend, __builtin_tabort)
(__builtin_non_tx_store, __builtin_tx_nesting_depth)
(__builtin_tx_nesting_depth, TBEGIN, TEND, TABORT, NTSTG, ETND):
Add.
* sysdeps/unix/sysv/linux/s390/lowlevellock.h:
(__lll_timedlock_elision, __lll_lock_elision, __lll_unlock_elision)
(__lll_trylock_elision, lll_timedlock_elision, lll_lock_elision)
(lll_unlock_elision, lll_trylock_elision):
Add.
--
Ciao
Dominik ^_^ ^_^
--
Dominik Vogt
IBM Germany