… | |
… | |
45 | Unlike the name component C<stamp> might indicate, it is also used for |
45 | Unlike the name component C<stamp> might indicate, it is also used for |
46 | time differences throughout libeio. |
46 | time differences throughout libeio. |
47 | |
47 | |
48 | =head2 FORK SUPPORT |
48 | =head2 FORK SUPPORT |
49 | |
49 | |
50 | Calling C<fork ()> is fully supported by this module. It is implemented in these steps: |
50 | Calling C<fork ()> is fully supported by this module - but you must not |
|
|
51 | rely on this. It is currently implemented in these steps: |
51 | |
52 | |
52 | 1. wait till all requests in "execute" state have been handled |
53 | 1. wait till all requests in "execute" state have been handled |
53 | (basically requests that are already handed over to the kernel). |
54 | (basically requests that are already handed over to the kernel). |
54 | 2. fork |
55 | 2. fork |
55 | 3. in the parent, continue business as usual, done |
56 | 3. in the parent, continue business as usual, done |
56 | 4. in the child, destroy all ready and pending requests and free the |
57 | 4. in the child, destroy all ready and pending requests and free the |
57 | memory used by the worker threads. This gives you a fully empty |
58 | memory used by the worker threads. This gives you a fully empty |
58 | libeio queue. |
59 | libeio queue. |
59 | |
60 | |
60 | Note, however, since libeio does use threads, thr above guarantee doesn't |
61 | Note, however, since libeio does use threads, the above guarantee doesn't |
61 | cover your libc, for example, malloc and other libc functions are not |
62 | cover your libc, for example, malloc and other libc functions are not |
62 | fork-safe, so there is very little you can do after a fork, and in fatc, |
63 | fork-safe, so there is very little you can do after a fork, and in fact, |
63 | the above might crash, and thus change. |
64 | the above might crash, and thus change. |
64 | |
65 | |
65 | =head1 INITIALISATION/INTEGRATION |
66 | =head1 INITIALISATION/INTEGRATION |
66 | |
67 | |
67 | Before you can call any eio functions you first have to initialise the |
68 | Before you can call any eio functions you first have to initialise the |
… | |
… | |
124 | =back |
125 | =back |
125 | |
126 | |
126 | For libev, you would typically use an C<ev_async> watcher: the |
127 | For libev, you would typically use an C<ev_async> watcher: the |
127 | C<want_poll> callback would invoke C<ev_async_send> to wake up the event |
128 | C<want_poll> callback would invoke C<ev_async_send> to wake up the event |
128 | loop. Inside the callback set for the watcher, one would call C<eio_poll |
129 | loop. Inside the callback set for the watcher, one would call C<eio_poll |
129 | ()> (followed by C<ev_async_send> again if C<eio_poll> indicates that not |
130 | ()>. |
130 | all requests have been handled yet). The race is taken care of because |
131 | |
131 | libev resets/rearms the async watcher before calling your callback, |
132 | If C<eio_poll ()> is configured to not handle all results in one go |
132 | and therefore, before calling C<eio_poll>. This might result in (some) |
133 | (i.e. it returns C<-1>) then you should start an idle watcher that calls |
133 | spurious wake-ups, but is generally harmless. |
134 | C<eio_poll> until it returns something C<!= -1>. |
|
|
135 | |
|
|
136 | A full-featured connector between libeio and libev would look as follows |
|
|
137 | (if C<eio_poll> is handling all requests, it can of course be simplified a |
|
|
138 | lot by removing the idle watcher logic): |
|
|
139 | |
|
|
140 | static struct ev_loop *loop; |
|
|
141 | static ev_idle repeat_watcher; |
|
|
142 | static ev_async ready_watcher; |
|
|
143 | |
|
|
144 | /* idle watcher callback, only used when eio_poll */ |
|
|
145 | /* didn't handle all results in one call */ |
|
|
146 | static void |
|
|
147 | repeat (EV_P_ ev_idle *w, int revents) |
|
|
148 | { |
|
|
149 | if (eio_poll () != -1) |
|
|
150 | ev_idle_stop (EV_A_ w); |
|
|
151 | } |
|
|
152 | |
|
|
153 | /* eio has some results, process them */ |
|
|
154 | static void |
|
|
155 | ready (EV_P_ ev_async *w, int revents) |
|
|
156 | { |
|
|
157 | if (eio_poll () == -1) |
|
|
158 | ev_idle_start (EV_A_ &repeat_watcher); |
|
|
159 | } |
|
|
160 | |
|
|
161 | /* wake up the event loop */ |
|
|
162 | static void |
|
|
163 | want_poll (void) |
|
|
164 | { |
|
|
165 | ev_async_send (loop, &ready_watcher) |
|
|
166 | } |
|
|
167 | |
|
|
168 | void |
|
|
169 | my_init_eio () |
|
|
170 | { |
|
|
171 | loop = EV_DEFAULT; |
|
|
172 | |
|
|
173 | ev_idle_init (&repeat_watcher, repeat); |
|
|
174 | ev_async_init (&ready_watcher, ready); |
|
|
175 | ev_async_start (loop &watcher); |
|
|
176 | |
|
|
177 | eio_init (want_poll, 0); |
|
|
178 | } |
134 | |
179 | |
135 | For most other event loops, you would typically use a pipe - the event |
180 | For most other event loops, you would typically use a pipe - the event |
136 | loop should be told to wait for read readiness on the read end. In |
181 | loop should be told to wait for read readiness on the read end. In |
137 | C<want_poll> you would write a single byte, in C<done_poll> you would try |
182 | C<want_poll> you would write a single byte, in C<done_poll> you would try |
138 | to read that byte, and in the callback for the read end, you would call |
183 | to read that byte, and in the callback for the read end, you would call |
139 | C<eio_poll>. The race is avoided here because the event loop should invoke |
184 | C<eio_poll>. |
140 | your callback again and again until the byte has been read (as the pipe |
185 | |
141 | read callback does not read it, only C<done_poll>). |
186 | You don't have to take special care in the case C<eio_poll> doesn't handle |
|
|
187 | all requests, as the done callback will not be invoked, so the event loop |
|
|
188 | will still signal readiness for the pipe until I<all> results have been |
|
|
189 | processed. |
142 | |
190 | |
143 | |
191 | |
144 | =head1 HIGH LEVEL REQUEST API |
192 | =head1 HIGH LEVEL REQUEST API |
145 | |
193 | |
146 | Libeio has both a high-level API, which consists of calling a request |
194 | Libeio has both a high-level API, which consists of calling a request |
… | |
… | |
213 | } |
261 | } |
214 | |
262 | |
215 | /* the first three arguments are passed to open(2) */ |
263 | /* the first three arguments are passed to open(2) */ |
216 | /* the remaining are priority, callback and data */ |
264 | /* the remaining are priority, callback and data */ |
217 | if (!eio_open ("/etc/passwd", O_RDONLY, 0, 0, file_open_done, 0)) |
265 | if (!eio_open ("/etc/passwd", O_RDONLY, 0, 0, file_open_done, 0)) |
218 | abort (); /* something ent wrong, we will all die!!! */ |
266 | abort (); /* something went wrong, we will all die!!! */ |
219 | |
267 | |
220 | Note that you additionally need to call C<eio_poll> when the C<want_cb> |
268 | Note that you additionally need to call C<eio_poll> when the C<want_cb> |
221 | indicates that requests are ready to be processed. |
269 | indicates that requests are ready to be processed. |
|
|
270 | |
|
|
271 | =head2 CANCELLING REQUESTS |
|
|
272 | |
|
|
273 | Sometimes the need for a request goes away before the request is |
|
|
274 | finished. In that case, one can cancel the request by a call to |
|
|
275 | C<eio_cancel>: |
|
|
276 | |
|
|
277 | =over 4 |
|
|
278 | |
|
|
279 | =item eio_cancel (eio_req *req) |
|
|
280 | |
|
|
281 | Cancel the request (and all its subrequests). If the request is currently |
|
|
282 | executing it might still continue to execute, and in other cases it might |
|
|
283 | still take a while till the request is cancelled. |
|
|
284 | |
|
|
285 | Even if cancelled, the finish callback will still be invoked - the |
|
|
286 | callbacks of all cancellable requests need to check whether the request |
|
|
287 | has been cancelled by calling C<EIO_CANCELLED (req)>: |
|
|
288 | |
|
|
289 | static int |
|
|
290 | my_eio_cb (eio_req *req) |
|
|
291 | { |
|
|
292 | if (EIO_CANCELLED (req)) |
|
|
293 | return 0; |
|
|
294 | } |
|
|
295 | |
|
|
296 | In addition, cancelled requests will I<either> have C<< req->result >> |
|
|
297 | set to C<-1> and C<errno> to C<ECANCELED>, or I<otherwise> they were |
|
|
298 | successfully executed, despite being cancelled (e.g. when they have |
|
|
299 | already been executed at the time they were cancelled). |
|
|
300 | |
|
|
301 | C<EIO_CANCELLED> is still true for requests that have successfully |
|
|
302 | executed, as long as C<eio_cancel> was called on them at some point. |
|
|
303 | |
|
|
304 | =back |
222 | |
305 | |
223 | =head2 AVAILABLE REQUESTS |
306 | =head2 AVAILABLE REQUESTS |
224 | |
307 | |
225 | The following request functions are available. I<All> of them return the |
308 | The following request functions are available. I<All> of them return the |
226 | C<eio_req *> on success and C<0> on failure, and I<all> of them have the |
309 | C<eio_req *> on success and C<0> on failure, and I<all> of them have the |
… | |
… | |
335 | =item eio_fstat (int fd, int pri, eio_cb cb, void *data) |
418 | =item eio_fstat (int fd, int pri, eio_cb cb, void *data) |
336 | |
419 | |
337 | Stats a file - if C<< req->result >> indicates success, then you can |
420 | Stats a file - if C<< req->result >> indicates success, then you can |
338 | access the C<struct stat>-like structure via C<< req->ptr2 >>: |
421 | access the C<struct stat>-like structure via C<< req->ptr2 >>: |
339 | |
422 | |
340 | EIO_STRUCT_STAT *statdata = (EIO_STRUCT_STAT *)req->ptr2; |
423 | EIO_STRUCT_STAT *statdata = (EIO_STRUCT_STAT *)req->ptr2; |
341 | |
424 | |
342 | =item eio_statvfs (const char *path, int pri, eio_cb cb, void *data) |
425 | =item eio_statvfs (const char *path, int pri, eio_cb cb, void *data) |
343 | |
426 | |
344 | =item eio_fstatvfs (int fd, int pri, eio_cb cb, void *data) |
427 | =item eio_fstatvfs (int fd, int pri, eio_cb cb, void *data) |
345 | |
428 | |
346 | Stats a filesystem - if C<< req->result >> indicates success, then you can |
429 | Stats a filesystem - if C<< req->result >> indicates success, then you can |
347 | access the C<struct statvfs>-like structure via C<< req->ptr2 >>: |
430 | access the C<struct statvfs>-like structure via C<< req->ptr2 >>: |
348 | |
431 | |
349 | EIO_STRUCT_STATVFS *statdata = (EIO_STRUCT_STATVFS *)req->ptr2; |
432 | EIO_STRUCT_STATVFS *statdata = (EIO_STRUCT_STATVFS *)req->ptr2; |
350 | |
433 | |
351 | =back |
434 | =back |
352 | |
435 | |
353 | =head3 READING DIRECTORIES |
436 | =head3 READING DIRECTORIES |
354 | |
437 | |
355 | Reading directories sounds simple, but can be rather demanding, especially |
438 | Reading directories sounds simple, but can be rather demanding, especially |
356 | if you want to do stuff such as traversing a diretcory hierarchy or |
439 | if you want to do stuff such as traversing a directory hierarchy or |
357 | processing all files in a directory. Libeio can assist thess complex tasks |
440 | processing all files in a directory. Libeio can assist these complex tasks |
358 | with it's C<eio_readdir> call. |
441 | with it's C<eio_readdir> call. |
359 | |
442 | |
360 | =over 4 |
443 | =over 4 |
361 | |
444 | |
362 | =item eio_readdir (const char *path, int flags, int pri, eio_cb cb, void *data) |
445 | =item eio_readdir (const char *path, int flags, int pri, eio_cb cb, void *data) |
… | |
… | |
394 | |
477 | |
395 | If this flag is specified, then, in addition to the names in C<ptr2>, |
478 | If this flag is specified, then, in addition to the names in C<ptr2>, |
396 | also an array of C<struct eio_dirent> is returned, in C<ptr1>. A C<struct |
479 | also an array of C<struct eio_dirent> is returned, in C<ptr1>. A C<struct |
397 | eio_dirent> looks like this: |
480 | eio_dirent> looks like this: |
398 | |
481 | |
399 | struct eio_dirent |
482 | struct eio_dirent |
400 | { |
483 | { |
401 | int nameofs; /* offset of null-terminated name string in (char *)req->ptr2 */ |
484 | int nameofs; /* offset of null-terminated name string in (char *)req->ptr2 */ |
402 | unsigned short namelen; /* size of filename without trailing 0 */ |
485 | unsigned short namelen; /* size of filename without trailing 0 */ |
403 | unsigned char type; /* one of EIO_DT_* */ |
486 | unsigned char type; /* one of EIO_DT_* */ |
404 | signed char score; /* internal use */ |
487 | signed char score; /* internal use */ |
405 | ino_t inode; /* the inode number, if available, otherwise unspecified */ |
488 | ino_t inode; /* the inode number, if available, otherwise unspecified */ |
406 | }; |
489 | }; |
407 | |
490 | |
408 | The only members you normally would access are C<nameofs>, which is the |
491 | The only members you normally would access are C<nameofs>, which is the |
409 | byte-offset from C<ptr2> to the start of the name, C<namelen> and C<type>. |
492 | byte-offset from C<ptr2> to the start of the name, C<namelen> and C<type>. |
410 | |
493 | |
411 | C<type> can be one of: |
494 | C<type> can be one of: |
… | |
… | |
454 | When this flag is specified, then the names will be returned in an order |
537 | When this flag is specified, then the names will be returned in an order |
455 | suitable for stat()'ing each one. That is, when you plan to stat() |
538 | suitable for stat()'ing each one. That is, when you plan to stat() |
456 | all files in the given directory, then the returned order will likely |
539 | all files in the given directory, then the returned order will likely |
457 | be fastest. |
540 | be fastest. |
458 | |
541 | |
459 | If both this flag and C<EIO_READDIR_DIRS_FIRST> are specified, then |
542 | If both this flag and C<EIO_READDIR_DIRS_FIRST> are specified, then the |
460 | the likely dirs come first, resulting in a less optimal stat order. |
543 | likely directories come first, resulting in a less optimal stat order. |
461 | |
544 | |
462 | =item EIO_READDIR_FOUND_UNKNOWN |
545 | =item EIO_READDIR_FOUND_UNKNOWN |
463 | |
546 | |
464 | This flag should not be specified when calling C<eio_readdir>. Instead, |
547 | This flag should not be specified when calling C<eio_readdir>. Instead, |
465 | it is being set by C<eio_readdir> (you can access the C<flags> via C<< |
548 | it is being set by C<eio_readdir> (you can access the C<flags> via C<< |
466 | req->int1 >>, when any of the C<type>'s found were C<EIO_DT_UNKNOWN>. The |
549 | req->int1 >>, when any of the C<type>'s found were C<EIO_DT_UNKNOWN>. The |
467 | absense of this flag therefore indicates that all C<type>'s are known, |
550 | absence of this flag therefore indicates that all C<type>'s are known, |
468 | which can be used to speed up some algorithms. |
551 | which can be used to speed up some algorithms. |
469 | |
552 | |
470 | A typical use case would be to identify all subdirectories within a |
553 | A typical use case would be to identify all subdirectories within a |
471 | directory - you would ask C<eio_readdir> for C<EIO_READDIR_DIRS_FIRST>. If |
554 | directory - you would ask C<eio_readdir> for C<EIO_READDIR_DIRS_FIRST>. If |
472 | then this flag is I<NOT> set, then all the entries at the beginning of the |
555 | then this flag is I<NOT> set, then all the entries at the beginning of the |
… | |
… | |
560 | |
643 | |
561 | eio_custom (my_open, 0, my_open_done, "/etc/passwd"); |
644 | eio_custom (my_open, 0, my_open_done, "/etc/passwd"); |
562 | |
645 | |
563 | =item eio_busy (eio_tstamp delay, int pri, eio_cb cb, void *data) |
646 | =item eio_busy (eio_tstamp delay, int pri, eio_cb cb, void *data) |
564 | |
647 | |
565 | This is a a request that takes C<delay> seconds to execute, but otherwise |
648 | This is a request that takes C<delay> seconds to execute, but otherwise |
566 | does nothing - it simply puts one of the worker threads to sleep for this |
649 | does nothing - it simply puts one of the worker threads to sleep for this |
567 | long. |
650 | long. |
568 | |
651 | |
569 | This request can be used to artificially increase load, e.g. for debugging |
652 | This request can be used to artificially increase load, e.g. for debugging |
570 | or benchmarking reasons. |
653 | or benchmarking reasons. |
… | |
… | |
586 | There are two primary use cases for this: a) bundle many requests into a |
669 | There are two primary use cases for this: a) bundle many requests into a |
587 | single, composite, request with a definite callback and the ability to |
670 | single, composite, request with a definite callback and the ability to |
588 | cancel the whole request with its subrequests and b) limiting the number |
671 | cancel the whole request with its subrequests and b) limiting the number |
589 | of "active" requests. |
672 | of "active" requests. |
590 | |
673 | |
591 | Further below you will find more dicussion of these topics - first follows |
674 | Further below you will find more discussion of these topics - first |
592 | the reference section detailing the request generator and other methods. |
675 | follows the reference section detailing the request generator and other |
|
|
676 | methods. |
593 | |
677 | |
594 | =over 4 |
678 | =over 4 |
595 | |
679 | |
596 | =item eio_grp (eio_cb cb, void *data) |
680 | =item eio_req *grp = eio_grp (eio_cb cb, void *data) |
597 | |
681 | |
598 | Creates and submits a group request. |
682 | Creates, submits and returns a group request. |
|
|
683 | |
|
|
684 | =item eio_grp_add (eio_req *grp, eio_req *req) |
|
|
685 | |
|
|
686 | Adds a request to the request group. |
|
|
687 | |
|
|
688 | =item eio_grp_cancel (eio_req *grp) |
|
|
689 | |
|
|
690 | Cancels all requests I<in> the group, but I<not> the group request |
|
|
691 | itself. You can cancel the group request via a normal C<eio_cancel> call. |
|
|
692 | |
|
|
693 | |
599 | |
694 | |
600 | =back |
695 | =back |
601 | |
696 | |
602 | |
697 | |
603 | |
698 | |
… | |
… | |
607 | /* groups */ |
702 | /* groups */ |
608 | |
703 | |
609 | eio_req *eio_grp (eio_cb cb, void *data); |
704 | eio_req *eio_grp (eio_cb cb, void *data); |
610 | void eio_grp_feed (eio_req *grp, void (*feed)(eio_req *req), int limit); |
705 | void eio_grp_feed (eio_req *grp, void (*feed)(eio_req *req), int limit); |
611 | void eio_grp_limit (eio_req *grp, int limit); |
706 | void eio_grp_limit (eio_req *grp, int limit); |
612 | void eio_grp_add (eio_req *grp, eio_req *req); |
|
|
613 | void eio_grp_cancel (eio_req *grp); /* cancels all sub requests but not the group */ |
707 | void eio_grp_cancel (eio_req *grp); /* cancels all sub requests but not the group */ |
614 | |
708 | |
615 | |
709 | |
616 | =back |
710 | =back |
617 | |
711 | |
… | |
… | |
624 | =head1 ANATOMY AND LIFETIME OF AN EIO REQUEST |
718 | =head1 ANATOMY AND LIFETIME OF AN EIO REQUEST |
625 | |
719 | |
626 | A request is represented by a structure of type C<eio_req>. To initialise |
720 | A request is represented by a structure of type C<eio_req>. To initialise |
627 | it, clear it to all zero bytes: |
721 | it, clear it to all zero bytes: |
628 | |
722 | |
629 | eio_req req; |
723 | eio_req req; |
630 | |
724 | |
631 | memset (&req, 0, sizeof (req)); |
725 | memset (&req, 0, sizeof (req)); |
632 | |
726 | |
633 | A more common way to initialise a new C<eio_req> is to use C<calloc>: |
727 | A more common way to initialise a new C<eio_req> is to use C<calloc>: |
634 | |
728 | |
635 | eio_req *req = calloc (1, sizeof (*req)); |
729 | eio_req *req = calloc (1, sizeof (*req)); |
636 | |
730 | |
637 | In either case, libeio neither allocates, initialises or frees the |
731 | In either case, libeio neither allocates, initialises or frees the |
638 | C<eio_req> structure for you - it merely uses it. |
732 | C<eio_req> structure for you - it merely uses it. |
639 | |
733 | |
640 | zero |
734 | zero |
… | |
… | |
658 | for example, in interactive programs, you might want to limit this time to |
752 | for example, in interactive programs, you might want to limit this time to |
659 | C<0.01> seconds or so. |
753 | C<0.01> seconds or so. |
660 | |
754 | |
661 | Note that: |
755 | Note that: |
662 | |
756 | |
|
|
757 | =over 4 |
|
|
758 | |
663 | a) libeio doesn't know how long your request callbacks take, so the time |
759 | =item a) libeio doesn't know how long your request callbacks take, so the |
664 | spent in C<eio_poll> is up to one callback invocation longer then this |
760 | time spent in C<eio_poll> is up to one callback invocation longer then |
665 | interval. |
761 | this interval. |
666 | |
762 | |
667 | b) this is implemented by calling C<gettimeofday> after each request, |
763 | =item b) this is implemented by calling C<gettimeofday> after each |
668 | which can be costly. |
764 | request, which can be costly. |
669 | |
765 | |
670 | c) at least one request will be handled. |
766 | =item c) at least one request will be handled. |
|
|
767 | |
|
|
768 | =back |
671 | |
769 | |
672 | =item eio_set_max_poll_reqs (unsigned int nreqs) |
770 | =item eio_set_max_poll_reqs (unsigned int nreqs) |
673 | |
771 | |
674 | When C<nreqs> is non-zero, then C<eio_poll> will not handle more than |
772 | When C<nreqs> is non-zero, then C<eio_poll> will not handle more than |
675 | C<nreqs> requests per invocation. This is a less costly way to limit the |
773 | C<nreqs> requests per invocation. This is a less costly way to limit the |