33dee81933
The getaddrinfo(3), getipnodebyname(3) and resolver(3) can coincide now with what should be totally reentrant, and h_errno values will now be preserved correctly, but this does not affect interfaces such as gethostbyname(3) which are still mostly non-reentrant. In all of these relevant functions, the thread-safety has been pushed down as far as it seems possible right now. This means that operations that are selected via nsdispatch(3) (i.e. files, yp, dns) are protected still under global locks that getaddrinfo(3) defines, but where possible the locking is greatly reduced. The most noticeable improvement is that multiple DNS lookups can now be run at the same time, and this shows major improvement in performance of DNS-lookup threaded programs, and solves the "Mozilla tab serialization" problem. No single-threaded applications need to be recompiled. Multi-threaded applications that reference "_res" to change resolver(3) options will need to be recompiled, and ones which reference "h_errno" will also if they desire the correct h_errno values. If the applications already understood that _res and h_errno were not thread-safe and had their own locking, they will see no performance improvement but will not actually break in any way. Please note that when NSS modules are used, or when nsdispatch(3) defaults to adding any lookups of its own to the individual libc _nsdispatch() calls, those MUST be reentrant as well.
136 lines
5.4 KiB
C
136 lines
5.4 KiB
C
/*-
|
|
* Copyright (c) 1997,98 The NetBSD Foundation, Inc.
|
|
* All rights reserved.
|
|
*
|
|
* This code is derived from software contributed to The NetBSD Foundation
|
|
* by J.T. Conklin.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
* 3. All advertising materials mentioning features or use of this software
|
|
* must display the following acknowledgement:
|
|
* This product includes software developed by the NetBSD
|
|
* Foundation, Inc. and its contributors.
|
|
* 4. Neither the name of The NetBSD Foundation nor the names of its
|
|
* contributors may be used to endorse or promote products derived
|
|
* from this software without specific prior written permission.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
|
|
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
|
|
* TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
|
|
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
|
|
* BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
|
|
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
|
|
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
|
|
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
|
|
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
|
|
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
|
|
* POSSIBILITY OF SUCH DAMAGE.
|
|
*
|
|
* $FreeBSD$
|
|
*/
|
|
|
|
/*
|
|
* Requirements:
|
|
*
|
|
* 1. The thread safe mechanism should be lightweight so the library can
|
|
* be used by non-threaded applications without unreasonable overhead.
|
|
*
|
|
* 2. There should be no dependency on a thread engine for non-threaded
|
|
* applications.
|
|
*
|
|
* 3. There should be no dependency on any particular thread engine.
|
|
*
|
|
* 4. The library should be able to be compiled without support for thread
|
|
* safety.
|
|
*
|
|
*
|
|
* Rationale:
|
|
*
|
|
* One approach for thread safety is to provide discrete versions of the
|
|
* library: one thread safe, the other not. The disadvantage of this is
|
|
* that libc is rather large, and two copies of a library which are 99%+
|
|
* identical is not an efficent use of resources.
|
|
*
|
|
* Another approach is to provide a single thread safe library. However,
|
|
* it should not add significant run time or code size overhead to non-
|
|
* threaded applications.
|
|
*
|
|
* Since the NetBSD C library is used in other projects, it should be
|
|
* easy to replace the mutual exclusion primitives with ones provided by
|
|
* another system. Similarly, it should also be easy to remove all
|
|
* support for thread safety completely if the target environment does
|
|
* not support threads.
|
|
*
|
|
*
|
|
* Implementation Details:
|
|
*
|
|
* The mutex primitives used by the library (mutex_t, mutex_lock, etc.)
|
|
* are macros which expand to the cooresponding primitives provided by
|
|
* the thread engine or to nothing. The latter is used so that code is
|
|
* not unreasonably cluttered with #ifdefs when all thread safe support
|
|
* is removed.
|
|
*
|
|
* The mutex macros can be directly mapped to the mutex primitives from
|
|
* pthreads, however it should be reasonably easy to wrap another mutex
|
|
* implementation so it presents a similar interface.
|
|
*
|
|
* Stub implementations of the mutex functions are provided with *weak*
|
|
* linkage. These functions simply return success. When linked with a
|
|
* thread library (i.e. -lpthread), the functions will override the
|
|
* stubs.
|
|
*/
|
|
|
|
#include <pthread.h>
|
|
#include <pthread_np.h>
|
|
#include "libc_private.h"
|
|
|
|
#define mutex_t pthread_mutex_t
|
|
#define cond_t pthread_cond_t
|
|
#define rwlock_t pthread_rwlock_t
|
|
#define once_t pthread_once_t
|
|
|
|
#define thread_key_t pthread_key_t
|
|
#define MUTEX_INITIALIZER PTHREAD_MUTEX_INITIALIZER
|
|
#define RWLOCK_INITIALIZER PTHREAD_RWLOCK_INITIALIZER
|
|
#define ONCE_INITIALIZER PTHREAD_ONCE_INIT
|
|
|
|
#define mutex_init(m, a) _pthread_mutex_init(m, a)
|
|
#define mutex_lock(m) if (__isthreaded) \
|
|
_pthread_mutex_lock(m)
|
|
#define mutex_unlock(m) if (__isthreaded) \
|
|
_pthread_mutex_unlock(m)
|
|
#define mutex_trylock(m) (__isthreaded ? 0 : _pthread_mutex_trylock(m))
|
|
|
|
#define cond_init(c, a, p) _pthread_cond_init(c, a)
|
|
#define cond_signal(m) if (__isthreaded) \
|
|
_pthread_cond_signal(m)
|
|
#define cond_broadcast(m) if (__isthreaded) \
|
|
_pthread_cond_broadcast(m)
|
|
#define cond_wait(c, m) if (__isthreaded) \
|
|
_pthread_cond_wait(c, m)
|
|
|
|
#define rwlock_init(l, a) _pthread_rwlock_init(l, a)
|
|
#define rwlock_rdlock(l) if (__isthreaded) \
|
|
_pthread_rwlock_rdlock(l)
|
|
#define rwlock_wrlock(l) if (__isthreaded) \
|
|
_pthread_rwlock_wrlock(l)
|
|
#define rwlock_unlock(l) if (__isthreaded) \
|
|
_pthread_rwlock_unlock(l)
|
|
|
|
#define thr_keycreate(k, d) _pthread_key_create(k, d)
|
|
#define thr_setspecific(k, p) _pthread_setspecific(k, p)
|
|
#define thr_getspecific(k) _pthread_getspecific(k)
|
|
#define thr_sigsetmask(f, n, o) _pthread_sigmask(f, n, o)
|
|
|
|
#define thr_once(o, i) _pthread_once(o, i)
|
|
#define thr_self() _pthread_self()
|
|
#define thr_exit(x) _pthread_exit(x)
|
|
#define thr_main() _pthread_main_np()
|