Re: Need for arch pthread-offsets.h (was Re: [PATCH v2 05/15] ARC: Atomics and Locking primitives)

2019-02-04 Thread Andreas Schwab
On Jan 31 2019, Vineet Gupta wrote: > However what was not obvious is adjustment to > sysdeps//nptl/pthread-offsets.h to __PTHREAD_MUTEX_*_OFFSET. But these > don't > seem to be *actually* used anywhere. Sure there are assert checks in > pthread_mutext_init.c but what's the point: the struct is

Need for arch pthread-offsets.h (was Re: [PATCH v2 05/15] ARC: Atomics and Locking primitives)

2019-01-31 Thread Vineet Gupta
On 1/30/19 9:40 AM, Vineet Gupta wrote: > On 1/30/19 12:28 AM, Andreas Schwab wrote: >> On Jan 29 2019, Vineet Gupta wrote: >> >>> +#define __PTHREAD_MUTEX_NUSERS_AFTER_KIND 1 >>> +#define __PTHREAD_MUTEX_USE_UNION 1 >> New ports should use the preferred values for these macros. > >

Re: [PATCH v2 05/15] ARC: Atomics and Locking primitives

2019-01-30 Thread Joseph Myers
On Wed, 30 Jan 2019, Vineet Gupta wrote: > On 1/30/19 1:50 PM, Joseph Myers wrote: > >>> This seems like something that could be a common issue for new ports, so > >>> worth a mention at alongside > >>> such things as using init_array and

Re: [PATCH v2 05/15] ARC: Atomics and Locking primitives

2019-01-30 Thread Vineet Gupta
On 1/30/19 1:50 PM, Joseph Myers wrote: >>> This seems like something that could be a common issue for new ports, so >>> worth a mention at alongside >>> such things as using init_array and USE_ATOMIC_COMPILER_BUILTINS. >> Updated wiki page ! > Now

Re: [PATCH v2 05/15] ARC: Atomics and Locking primitives

2019-01-30 Thread Joseph Myers
On Wed, 30 Jan 2019, Vineet Gupta wrote: > On 1/30/19 1:04 PM, Joseph Myers wrote: > >>> +#define __PTHREAD_MUTEX_NUSERS_AFTER_KIND 1 > >>> +#define __PTHREAD_MUTEX_USE_UNION 1 > >> New ports should use the preferred values for these macros. > > This seems like something that could be a

Re: [PATCH v2 05/15] ARC: Atomics and Locking primitives

2019-01-30 Thread Vineet Gupta
On 1/30/19 1:04 PM, Joseph Myers wrote: >>> +#define __PTHREAD_MUTEX_NUSERS_AFTER_KIND 1 >>> +#define __PTHREAD_MUTEX_USE_UNION 1 >> New ports should use the preferred values for these macros. > This seems like something that could be a common issue for new ports, so > worth a mention

Re: [PATCH v2 05/15] ARC: Atomics and Locking primitives

2019-01-30 Thread Joseph Myers
On Wed, 30 Jan 2019, Andreas Schwab wrote: > On Jan 29 2019, Vineet Gupta wrote: > > > +#define __PTHREAD_MUTEX_NUSERS_AFTER_KIND 1 > > +#define __PTHREAD_MUTEX_USE_UNION 1 > > New ports should use the preferred values for these macros. This seems like something that could be a

Re: [PATCH v2 05/15] ARC: Atomics and Locking primitives

2019-01-30 Thread Vineet Gupta
On 1/30/19 12:28 AM, Andreas Schwab wrote: > On Jan 29 2019, Vineet Gupta wrote: > >> +#define __PTHREAD_MUTEX_NUSERS_AFTER_KIND 1 >> +#define __PTHREAD_MUTEX_USE_UNION 1 > New ports should use the preferred values for these macros. OK, changed to 0 and 0 per commit 06be6368da16104

Re: [PATCH v2 05/15] ARC: Atomics and Locking primitives

2019-01-30 Thread Andreas Schwab
On Jan 29 2019, Vineet Gupta wrote: > +#define __PTHREAD_MUTEX_NUSERS_AFTER_KIND 1 > +#define __PTHREAD_MUTEX_USE_UNION 1 New ports should use the preferred values for these macros. Andreas. -- Andreas Schwab, SUSE Labs, sch...@suse.de GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE

[PATCH v2 05/15] ARC: Atomics and Locking primitives

2019-01-29 Thread Vineet Gupta
Signed-off-by: Vineet Gupta --- ChangeLog | 3 ++ sysdeps/arc/atomic-machine.h | 73 +++ sysdeps/arc/nptl/bits/pthreadtypes-arch.h | 72 ++ sysdeps/arc/nptl/bits/semaphore.h | 32