ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/IO-AIO/README
(Generate patch)

Comparing IO-AIO/README (file contents):
Revision 1.24 by root, Sun Aug 5 16:45:54 2007 UTC vs.
Revision 1.27 by root, Sat Oct 6 14:05:37 2007 UTC

… …
258 } 258 }
259 }; 259 };
260 260
261 aio_close $fh, $callback->($status) 261 aio_close $fh, $callback->($status)
262 Asynchronously close a file and call the callback with the result 262 Asynchronously close a file and call the callback with the result
263 code. *WARNING:* although accepted, you should not pass in a perl 263 code.
264 filehandle here, as perl will likely close the file descriptor
265 another time when the filehandle is destroyed. Normally, you can
266 safely call perls "close" or just let filehandles go out of scope.
267 264
268 This is supposed to be a bug in the API, so that might change. It's 265 Unfortunately, you can't do this to perl. Perl *insists* very
269 therefore best to avoid this function. 266 strongly on closing the file descriptor associated with the
267 filehandle itself. Here is what aio_close will try:
268
269 1. dup()licate the fd
270 2. asynchronously close() the duplicated fd
271 3. dup()licate the fd once more
272 4. let perl close() the filehandle
273 5. asynchronously close the duplicated fd
274
275 The idea is that the first close() flushes stuff to disk that
276 closing an fd will flush, so when perl closes the fd, nothing much
277 will need to be flushed. The second async. close() will then flush
278 stuff to disk that closing the last fd to the file will flush.
279
280 Just FYI, SuSv3 has this to say on close:
281
282 All outstanding record locks owned by the process on the file
283 associated with the file descriptor shall be removed.
284
285 If fildes refers to a socket, close() shall cause the socket to be
286 destroyed. ... close() shall block for up to the current linger
287 interval until all data is transmitted.
288 [this actually sounds like a specification bug, but who knows]
289
290 And at least Linux additionally actually flushes stuff on every
291 close, even when the file itself is still open.
292
293 Sounds enourmously inefficient and complicated? Yes... please show
294 me how to nuke perl's fd out of existence...
270 295
271 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 296 aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
272 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval) 297 aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
273 Reads or writes $length bytes from the specified $fh and $offset 298 Reads or writes $length bytes from the specified $fh and $offset
274 into the scalar given by $data and offset $dataoffset and calls the 299 into the scalar given by $data and offset $dataoffset and calls the
275 callback without the actual number of bytes read (or -1 on error, 300 callback without the actual number of bytes read (or -1 on error,
276 just like the syscall). 301 just like the syscall).
277 302
278 If $offset is undefined, then the current file offset will be used 303 If $offset is undefined, then the current file descriptor offset
279 (and updated), otherwise the file offset will not be changed by 304 will be used (and updated), otherwise the file descriptor offset
280 these calls. 305 will not be changed by these calls.
281 306
282 If $length is undefined in "aio_write", use the remaining length of 307 If $length is undefined in "aio_write", use the remaining length of
283 $data. 308 $data.
284 309
285 If $dataoffset is less than zero, it will be counted from the end of 310 If $dataoffset is less than zero, it will be counted from the end of
… …
851 $oldmaxreqs = IO::AIO::max_outstanding $maxreqs 876 $oldmaxreqs = IO::AIO::max_outstanding $maxreqs
852 This is a very bad function to use in interactive programs because 877 This is a very bad function to use in interactive programs because
853 it blocks, and a bad way to reduce concurrency because it is 878 it blocks, and a bad way to reduce concurrency because it is
854 inexact: Better use an "aio_group" together with a feed callback. 879 inexact: Better use an "aio_group" together with a feed callback.
855 880
856 Sets the maximum number of outstanding requests to $nreqs. If you to 881 Sets the maximum number of outstanding requests to $nreqs. If you do
857 queue up more than this number of requests, the next call to the 882 queue up more than this number of requests, the next call to the
858 "poll_cb" (and "poll_some" and other functions calling "poll_cb") 883 "poll_cb" (and "poll_some" and other functions calling "poll_cb")
859 function will block until the limit is no longer exceeded. 884 function will block until the limit is no longer exceeded.
860 885
861 The default value is very large, so there is no practical limit on 886 The default value is very large, so there is no practical limit on
… …
907 bytes of memory. In addition, stat requests need a stat buffer (possibly 932 bytes of memory. In addition, stat requests need a stat buffer (possibly
908 a few hundred bytes), readdir requires a result buffer and so on. Perl 933 a few hundred bytes), readdir requires a result buffer and so on. Perl
909 scalars and other data passed into aio requests will also be locked and 934 scalars and other data passed into aio requests will also be locked and
910 will consume memory till the request has entered the done state. 935 will consume memory till the request has entered the done state.
911 936
912 This is now awfully much, so queuing lots of requests is not usually a 937 This is not awfully much, so queuing lots of requests is not usually a
913 problem. 938 problem.
914 939
915 Per-thread usage: 940 Per-thread usage:
916 941
917 In the execution phase, some aio requests require more memory for 942 In the execution phase, some aio requests require more memory for

Diff Legend

– Removed lines
+ Added lines
< Changed lines
> Changed lines