1=pod
2
3=head1 NAME
4
5ASYNC_get_wait_ctx,
6ASYNC_init_thread, ASYNC_cleanup_thread, ASYNC_start_job, ASYNC_pause_job,
7ASYNC_get_current_job, ASYNC_block_pause, ASYNC_unblock_pause, ASYNC_is_capable
8- asynchronous job management functions
9
10=head1 SYNOPSIS
11
12 #include <openssl/async.h>
13
14 int ASYNC_init_thread(size_t max_size, size_t init_size);
15 void ASYNC_cleanup_thread(void);
16
17 int ASYNC_start_job(ASYNC_JOB **job, ASYNC_WAIT_CTX *ctx, int *ret,
18                     int (*func)(void *), void *args, size_t size);
19 int ASYNC_pause_job(void);
20
21 ASYNC_JOB *ASYNC_get_current_job(void);
22 ASYNC_WAIT_CTX *ASYNC_get_wait_ctx(ASYNC_JOB *job);
23 void ASYNC_block_pause(void);
24 void ASYNC_unblock_pause(void);
25
26 int ASYNC_is_capable(void);
27
28=head1 DESCRIPTION
29
30OpenSSL implements asynchronous capabilities through an ASYNC_JOB. This
31represents code that can be started and executes until some event occurs. At
32that point the code can be paused and control returns to user code until some
33subsequent event indicates that the job can be resumed.
34
35The creation of an ASYNC_JOB is a relatively expensive operation. Therefore, for
36efficiency reasons, jobs can be created up front and reused many times. They are
37held in a pool until they are needed, at which point they are removed from the
38pool, used, and then returned to the pool when the job completes. If the user
39application is multi-threaded, then ASYNC_init_thread() may be called for each
40thread that will initiate asynchronous jobs. Before
41user code exits per-thread resources need to be cleaned up. This will normally
42occur automatically (see L<OPENSSL_init_crypto(3)>) but may be explicitly
43initiated by using ASYNC_cleanup_thread(). No asynchronous jobs must be
44outstanding for the thread when ASYNC_cleanup_thread() is called. Failing to
45ensure this will result in memory leaks.
46
47The B<max_size> argument limits the number of ASYNC_JOBs that will be held in
48the pool. If B<max_size> is set to 0 then no upper limit is set. When an
49ASYNC_JOB is needed but there are none available in the pool already then one
50will be automatically created, as long as the total of ASYNC_JOBs managed by the
51pool does not exceed B<max_size>. When the pool is first initialised
52B<init_size> ASYNC_JOBs will be created immediately. If ASYNC_init_thread() is
53not called before the pool is first used then it will be called automatically
54with a B<max_size> of 0 (no upper limit) and an B<init_size> of 0 (no ASYNC_JOBs
55created up front).
56
57An asynchronous job is started by calling the ASYNC_start_job() function.
58Initially B<*job> should be NULL. B<ctx> should point to an ASYNC_WAIT_CTX
59object created through the L<ASYNC_WAIT_CTX_new(3)> function. B<ret> should
60point to a location where the return value of the asynchronous function should
61be stored on completion of the job. B<func> represents the function that should
62be started asynchronously. The data pointed to by B<args> and of size B<size>
63will be copied and then passed as an argument to B<func> when the job starts.
64ASYNC_start_job will return one of the following values:
65
66=over 4
67
68=item B<ASYNC_ERR>
69
70An error occurred trying to start the job. Check the OpenSSL error queue (e.g.
71see L<ERR_print_errors(3)>) for more details.
72
73=item B<ASYNC_NO_JOBS>
74
75There are no jobs currently available in the pool. This call can be retried
76again at a later time.
77
78=item B<ASYNC_PAUSE>
79
80The job was successfully started but was "paused" before it completed (see
81ASYNC_pause_job() below). A handle to the job is placed in B<*job>. Other work
82can be performed (if desired) and the job restarted at a later time. To restart
83a job call ASYNC_start_job() again passing the job handle in B<*job>. The
84B<func>, B<args> and B<size> parameters will be ignored when restarting a job.
85When restarting a job ASYNC_start_job() B<must> be called from the same thread
86that the job was originally started from.
87
88=item B<ASYNC_FINISH>
89
90The job completed. B<*job> will be NULL and the return value from B<func> will
91be placed in B<*ret>.
92
93=back
94
95At any one time there can be a maximum of one job actively running per thread
96(you can have many that are paused). ASYNC_get_current_job() can be used to get
97a pointer to the currently executing ASYNC_JOB. If no job is currently executing
98then this will return NULL.
99
100If executing within the context of a job (i.e. having been called directly or
101indirectly by the function "func" passed as an argument to ASYNC_start_job())
102then ASYNC_pause_job() will immediately return control to the calling
103application with ASYNC_PAUSE returned from the ASYNC_start_job() call. A
104subsequent call to ASYNC_start_job passing in the relevant ASYNC_JOB in the
105B<*job> parameter will resume execution from the ASYNC_pause_job() call. If
106ASYNC_pause_job() is called whilst not within the context of a job then no
107action is taken and ASYNC_pause_job() returns immediately.
108
109ASYNC_get_wait_ctx() can be used to get a pointer to the ASYNC_WAIT_CTX
110for the B<job>. ASYNC_WAIT_CTXs can have a "wait" file descriptor associated
111with them. Applications can wait for the file descriptor to be ready for "read"
112using a system function call such as select or poll (being ready for "read"
113indicates that the job should be resumed). If no file descriptor is made
114available then an application will have to periodically "poll" the job by
115attempting to restart it to see if it is ready to continue.
116
117An example of typical usage might be an async capable engine. User code would
118initiate cryptographic operations. The engine would initiate those operations
119asynchronously and then call L<ASYNC_WAIT_CTX_set_wait_fd(3)> followed by
120ASYNC_pause_job() to return control to the user code. The user code can then
121perform other tasks or wait for the job to be ready by calling "select" or other
122similar function on the wait file descriptor. The engine can signal to the user
123code that the job should be resumed by making the wait file descriptor
124"readable". Once resumed the engine should clear the wake signal on the wait
125file descriptor.
126
127The ASYNC_block_pause() function will prevent the currently active job from
128pausing. The block will remain in place until a subsequent call to
129ASYNC_unblock_pause(). These functions can be nested, e.g. if you call
130ASYNC_block_pause() twice then you must call ASYNC_unblock_pause() twice in
131order to re-enable pausing. If these functions are called while there is no
132currently active job then they have no effect. This functionality can be useful
133to avoid deadlock scenarios. For example during the execution of an ASYNC_JOB an
134application acquires a lock. It then calls some cryptographic function which
135invokes ASYNC_pause_job(). This returns control back to the code that created
136the ASYNC_JOB. If that code then attempts to acquire the same lock before
137resuming the original job then a deadlock can occur. By calling
138ASYNC_block_pause() immediately after acquiring the lock and
139ASYNC_unblock_pause() immediately before releasing it then this situation cannot
140occur.
141
142Some platforms cannot support async operations. The ASYNC_is_capable() function
143can be used to detect whether the current platform is async capable or not.
144
145=head1 RETURN VALUES
146
147ASYNC_init_thread returns 1 on success or 0 otherwise.
148
149ASYNC_start_job returns one of ASYNC_ERR, ASYNC_NO_JOBS, ASYNC_PAUSE or
150ASYNC_FINISH as described above.
151
152ASYNC_pause_job returns 0 if an error occurred or 1 on success. If called when
153not within the context of an ASYNC_JOB then this is counted as success so 1 is
154returned.
155
156ASYNC_get_current_job returns a pointer to the currently executing ASYNC_JOB or
157NULL if not within the context of a job.
158
159ASYNC_get_wait_ctx() returns a pointer to the ASYNC_WAIT_CTX for the job.
160
161ASYNC_is_capable() returns 1 if the current platform is async capable or 0
162otherwise.
163
164=head1 NOTES
165
166On Windows platforms the openssl/async.h header is dependent on some
167of the types customarily made available by including windows.h. The
168application developer is likely to require control over when the latter
169is included, commonly as one of the first included headers. Therefore
170it is defined as an application developer's responsibility to include
171windows.h prior to async.h.
172
173=head1 EXAMPLES
174
175The following example demonstrates how to use most of the core async APIs:
176
177 #ifdef _WIN32
178 # include <windows.h>
179 #endif
180 #include <stdio.h>
181 #include <unistd.h>
182 #include <openssl/async.h>
183 #include <openssl/crypto.h>
184
185 int unique = 0;
186
187 void cleanup(ASYNC_WAIT_CTX *ctx, const void *key, OSSL_ASYNC_FD r, void *vw)
188 {
189     OSSL_ASYNC_FD *w = (OSSL_ASYNC_FD *)vw;
190
191     close(r);
192     close(*w);
193     OPENSSL_free(w);
194 }
195
196 int jobfunc(void *arg)
197 {
198     ASYNC_JOB *currjob;
199     unsigned char *msg;
200     int pipefds[2] = {0, 0};
201     OSSL_ASYNC_FD *wptr;
202     char buf = 'X';
203
204     currjob = ASYNC_get_current_job();
205     if (currjob != NULL) {
206         printf("Executing within a job\n");
207     } else {
208         printf("Not executing within a job - should not happen\n");
209         return 0;
210     }
211
212     msg = (unsigned char *)arg;
213     printf("Passed in message is: %s\n", msg);
214
215     if (pipe(pipefds) != 0) {
216         printf("Failed to create pipe\n");
217         return 0;
218     }
219     wptr = OPENSSL_malloc(sizeof(OSSL_ASYNC_FD));
220     if (wptr == NULL) {
221         printf("Failed to malloc\n");
222         return 0;
223     }
224     *wptr = pipefds[1];
225     ASYNC_WAIT_CTX_set_wait_fd(ASYNC_get_wait_ctx(currjob), &unique,
226                                pipefds[0], wptr, cleanup);
227
228     /*
229      * Normally some external event would cause this to happen at some
230      * later point - but we do it here for demo purposes, i.e.
231      * immediately signalling that the job is ready to be woken up after
232      * we return to main via ASYNC_pause_job().
233      */
234     write(pipefds[1], &buf, 1);
235
236     /* Return control back to main */
237     ASYNC_pause_job();
238
239     /* Clear the wake signal */
240     read(pipefds[0], &buf, 1);
241
242     printf ("Resumed the job after a pause\n");
243
244     return 1;
245 }
246
247 int main(void)
248 {
249     ASYNC_JOB *job = NULL;
250     ASYNC_WAIT_CTX *ctx = NULL;
251     int ret;
252     OSSL_ASYNC_FD waitfd;
253     fd_set waitfdset;
254     size_t numfds;
255     unsigned char msg[13] = "Hello world!";
256
257     printf("Starting...\n");
258
259     ctx = ASYNC_WAIT_CTX_new();
260     if (ctx == NULL) {
261         printf("Failed to create ASYNC_WAIT_CTX\n");
262         abort();
263     }
264
265     for (;;) {
266         switch (ASYNC_start_job(&job, ctx, &ret, jobfunc, msg, sizeof(msg))) {
267         case ASYNC_ERR:
268         case ASYNC_NO_JOBS:
269             printf("An error occurred\n");
270             goto end;
271         case ASYNC_PAUSE:
272             printf("Job was paused\n");
273             break;
274         case ASYNC_FINISH:
275             printf("Job finished with return value %d\n", ret);
276             goto end;
277         }
278
279         /* Wait for the job to be woken */
280         printf("Waiting for the job to be woken up\n");
281
282         if (!ASYNC_WAIT_CTX_get_all_fds(ctx, NULL, &numfds)
283                 || numfds > 1) {
284             printf("Unexpected number of fds\n");
285             abort();
286         }
287         ASYNC_WAIT_CTX_get_all_fds(ctx, &waitfd, &numfds);
288         FD_ZERO(&waitfdset);
289         FD_SET(waitfd, &waitfdset);
290         select(waitfd + 1, &waitfdset, NULL, NULL, NULL);
291     }
292
293 end:
294     ASYNC_WAIT_CTX_free(ctx);
295     printf("Finishing\n");
296
297     return 0;
298 }
299
300The expected output from executing the above example program is:
301
302 Starting...
303 Executing within a job
304 Passed in message is: Hello world!
305 Job was paused
306 Waiting for the job to be woken up
307 Resumed the job after a pause
308 Job finished with return value 1
309 Finishing
310
311=head1 SEE ALSO
312
313L<crypto(7)>, L<ERR_print_errors(3)>
314
315=head1 HISTORY
316
317ASYNC_init_thread, ASYNC_cleanup_thread,
318ASYNC_start_job, ASYNC_pause_job, ASYNC_get_current_job, ASYNC_get_wait_ctx(),
319ASYNC_block_pause(), ASYNC_unblock_pause() and ASYNC_is_capable() were first
320added in OpenSSL 1.1.0.
321
322=head1 COPYRIGHT
323
324Copyright 2015-2019 The OpenSSL Project Authors. All Rights Reserved.
325
326Licensed under the OpenSSL license (the "License").  You may not use
327this file except in compliance with the License.  You can obtain a copy
328in the file LICENSE in the source distribution or at
329L<https://www.openssl.org/source/license.html>.
330
331=cut
332