Commit graph

30 commits

Author SHA1 Message Date
Masashi Honma
f0356ec85c eloop: Add epoll option for better performance
This patch adds epoll option for the eloop implementation. This can be
selected with the CONFIG_ELOOP_EPOLL=y build option.

[merit]
See Table1.

Table1. comparison table
+--------+--------+-----------+------------+-------------+
|        | add fd | remove fd | prepare fd | dispatch fd |
+--------+--------+-----------+------------+-------------+
| select | O(1)   | O(1)      | O(N)       | O(N)        |
+--------+--------+-----------+------------+-------------+
| poll   | O(1)   | O(1)      | O(N)       | O(N)        |
+--------+--------+-----------+------------+-------------+
| epoll  | O(1)   | O(1)      | 0          | O(M)        |
+--------+--------+-----------+------------+-------------+
"add fd" is addition of fd by eloop_sock_table_add_sock().
"remove fd" is removal of fd by eloop_sock_table_remove_sock().
"prepare fd" is preparation of fds before wait in eloop_run().
"dispatch fd" is dispatchment of fds by eloop_sock_table_dispatch().
"N" is all watching fds.
"M" is fds which could be dispatched after waiting.

As shown in Table1, epoll option has better performance on "prepare fd" column.
Because select/poll option requires setting fds before every select()/poll().
But epoll_wait() doesn't need it.

And epoll option has also better performance on "dispatch fd" column.
Because select/poll option needs to check all registered fds to find out
dispatchable fds. But epoll option doesn't require checking all registered fds.
Because epoll_wait() returns dispatchable fd set.

So epoll option is effective for GO/AP functionality.

[demerit]
The epoll option requires additional heap memory. In case of P2P GO, it is
about 8K bytes.

Signed-off-by: Masashi Honma <masashi.honma@gmail.com>
2014-05-16 18:25:51 +03:00
Masashi Honma
da96a6f793 eloop: Separate event loop select/poll implementation
This allows yet another eloop.c option to be added.

Signed-off-by: Masashi Honma <masashi.honma@gmail.com>
2014-05-16 18:23:09 +03:00
Maxime Bizon
762c41ae99 eloop: Add assert() on negative fd when using select() code path
Signed-off-by: Maxime Bizon <mbizon@freebox.fr>
2014-03-26 00:17:07 +02:00
Jouni Malinen
a09ffd5f2f Fix req_scan-deplete-timeout and update eloop API for this
Commit e2f5a9889a was supposed to prevent
new scan request from pushing out the old one. However, it did not
really do that since eloop_deplete_timeout() returned 0 both for the
case where the old timeout existed (and was sooner) and if the old
timeout did not exist. It returned 1 only for the case where an old
timeout did exist and was larger than the new requested value. That case
used to result in wpa_supplicant_req_scan() rescheduling the timeout,
but hew code in eloop_deplete_timeout() did the exact same thing and as
such, did not really change anything apart from the debug log message.

Extend the eloop_deplete_timeout() (and eloop_replenish_timeout() for
that matter since it is very similar) to return three different values
based on whether the timeout existed or not and if yes, whether it was
modified. This allows wpa_supplicant_req_scan() to schedule a new
timeout only in the case there was no old timeout.

Signed-hostap: Jouni Malinen <j@w1.fi>
2014-01-07 10:45:10 +02:00
Johannes Berg
594516b4c2 Use monotonic clock for relative time for eloop if available
Relative time shouldn't be calculated based on gettimeofday
because that clock can jump (e.g., when the time is adjusted
by the system administrator.)

On systems where that is available, use CLOCK_BOOTTIME (on
fairly recent Linux systems, this clock takes into account
the time spend suspended) or CLOCK_MONOTONIC (on Linux and
some POSIX systems, this clock is just freely running with
no adjustments.)

Reported-by: Holger Schurig <holgerschurig@gmail.com>
Signed-hostap: Johannes Berg <johannes.berg@intel.com>
2013-11-20 23:52:56 +02:00
Dmitry Shmidt
e2f5a9889a Deplete scan request if new time is less than remaining
This avoids pushing out previous scheduled scan requests based on new
events.

Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
2013-11-17 16:46:58 +02:00
Kyeyoon Park
e5e74e5500 eloop: Add support for replenishing a registered timeout
eloop_replenish_timeout() finds a registered matching
<handler,eloop_data,user_data> timeout. If found, replenishes
the timeout if remaining time is less than the requested time.

Signed-hostap: Jouni Malinen <jouni@qca.qualcomm.com>
2013-11-08 00:22:32 +02:00
Jouni Malinen
61323e70e1 Convert perror/printf calls to wpa_printf
This makes debug and error logging more consistent and allows them to be
directed to a file more easily.

Signed-hostap: Jouni Malinen <j@w1.fi>
2013-11-02 12:58:33 +02:00
Simon Wunderlich
a7505b1775 eloop: Allow to run event loop multiple times in a row
DFS implementation requires to run an eventloop while monitoring
the Channel Availability Check (CAC). After that, the "real" event
loop is started, and should not fail doing so.

Signed-hostap: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
2013-05-09 16:42:14 +03:00
Pontus Fuchs
c869536ce9 eloop: Add a timer cancel that returns the remaining time
This new cancel timer will give back the remaining time if it was
pending.

Signed-hostap: Pontus Fuchs <pontus.fuchs@gmail.com>
2013-02-03 17:17:08 +02:00
Jouni Malinen
067ffa2696 Convert os_realloc() for an array to use os_realloc_array()
Signed-hostap: Jouni Malinen <j@w1.fi>
2012-08-13 21:21:23 +03:00
Jouni Malinen
4be921ac60 eloop: Fix EVENT_TYPE_WRITE with poll()-based eloop
This needs to use POLLOUT instead of POLLIN to get the correct event.

Signed-hostap: Jouni Malinen <j@w1.fi>
2012-07-04 21:36:38 +03:00
Ben Greear
f115560c21 eloop/poll: Handle POLLER | POLLHUP in read logic
Without this, we can get into a tight loop because the
code in general doesn't add eloop exception handlers,
so socket reporting the POLLERR would never be read.

With this change, any socket with POLLERR or POLLHUP
asserted will be handled by the read logic.

Signed-hostap: Ben Greear <greearb@candelatech.com>
2012-06-24 13:19:34 +03:00
Jouni Malinen
2624ed4311 eloop: Fix allocation failure handling in poll() version
eloop_sock_table_add_sock() needs to fail if pollfd array allocation
fails instead of returning success and leaving behind no buffer.

Signed-hostap: Jouni Malinen <j@w1.fi>
2012-02-12 21:33:42 +02:00
Ben Greear
2df4c4ef2f eloop: Support poll() in eloop
When using more than around 200 virtual stations, we start hitting the
max number of file descriptors supported by select(). This patch adds
support for poll(), which has no hard upper limit.

Signed-hostap: Ben Greear <greearb@candelatech.com>
2012-02-12 21:12:22 +02:00
Jouni Malinen
0f3d578efc Remove the GPL notification from files contributed by Jouni Malinen
Remove the GPL notification text from the files that were
initially contributed by myself.

Signed-hostap: Jouni Malinen <j@w1.fi>
2012-02-11 19:39:36 +02:00
Jouni Malinen
0fa0ad4e17 eloop: Fix integer overflow in long timeouts
If the os_time_t variable used for the expiration time (seconds)
overflows when the registered timeout value is being added,
assume that the event would happen after an infinite time, i.e.,
would not really happen in practice. This fixes issues with
long key timeouts getting converted to immediate expiration due
to the overflow.
2011-03-06 14:31:20 +02:00
Jouni Malinen
7f5957abcf eloop: Fix crash on signal handler cancelling next timeout
It is possible that the timeout pointer becomes invalid in one of the
signal handlers, so we need to reload the pointer after those.
2010-05-28 22:18:28 +03:00
Jouni Malinen
2e320d8db5 eloop: Clear timeout data during allocation
Better make sure the eloop_timeout data gets fully initialized. The
current code is filling in all the fields, but it is clearer to just
zero the buffer to make sure any new field added to the structure gets
initialized.
2010-01-10 21:48:27 +02:00
Jouni Malinen
459489c99d eloop: Fix timeout handler to use local copy of func pointer
We need to copy not only the context pointers, but also the function
pointer before the timeout gets freed.
2009-12-26 14:30:50 +02:00
Jouni Malinen
9aca440199 Drop WPA_TRACE reference before eloop timeout handler call
This avoids bogus error reports for cases where the timeout handler
frees the memory that was pointed to by the eloop timeout context.
2009-12-24 12:41:20 +02:00
Jouni Malinen
5e5223bf29 trace: Show eloop unregistered handler function name/file/line 2009-12-22 01:52:48 +02:00
Jouni Malinen
a6ff0e0810 trace: Add active reference tracking
This WPA_TRACE=y additions allows components to register active references
to memory that has been provided to them as a pointer. If such an actively
referenced memory area is freed, tracer will report this as an error and
backtraces of both the invalid free and the location where this pointer
was marked referenced are shown.
2009-12-22 01:11:15 +02:00
Jouni Malinen
f45fb672cc eloop: Do not use printf() or fprintf() directly 2009-12-19 20:27:55 +02:00
Jouni Malinen
eaa3f04b97 eloop: Use struct dl_list for timeouts 2009-12-19 20:26:22 +02:00
Jouni Malinen
0456ea16d8 eloop: Remove global user data pointer
This is not really needed since all signal handlers can use a context
pointer provided during signal handler registration.
2009-12-19 19:22:16 +02:00
Jouni Malinen
930f704aac Add backtrace support for debugging
WPA_TRACE=y can now be used to enable internal backtrace support that
will provide more details about implementation errors, e.g., when some
resources are not released correctly. In addition, this will print out
a backtrace automatically if SIGSEGV is received.
2009-12-19 18:40:54 +02:00
Dan Williams
7e1488494e Do not continually reschedule specific scans to help finding hidden SSIDs
In situations where the driver does background scanning and sends a
steady stream of scan results, wpa_supplicant would continually
reschedule the scan.  This resulted in specific SSID scans never
happening for a hidden AP, and the supplicant never connecting to the AP
because it never got found.  Instead, if there's an already scheduled
scan, and a request comes in to reschedule it, and there are enabled
scan_ssid=1 network blocks, let the scan happen anyway so the hidden
SSID has a chance to be found.
2008-06-03 11:37:48 +03:00
Jouni Malinen
a6ee047fcb Verify that os_get_time() does not fail before using the time value when
registering an eloop timeout.
2008-02-27 17:50:36 -08:00
Jouni Malinen
6fc6879bd5 Re-initialize hostapd/wpa_supplicant git repository based on 0.6.3 release 2008-02-27 17:34:43 -08:00