summaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
...
* Fix compatibility with latest changes in libimobiledeviceNikias Bassen2019-11-111-3/+6
|
* Define _BSD_SOURCE again for older platforms.Nikias Bassen2019-11-022-0/+2
| | | | | While defining _BSD_SOURCE prints deprecation warnings, defining both _DEFAULT_SOURCE and _BSD_SOURCE will not.
* preflight: Prevent possible UaF if usb device is removed while preflight is ↵Nikias Bassen2019-10-071-0/+5
| | | | | | | | | in progress The device serial number is only used by reference, however since the preflight helper runs in a separate thread the usb device might be invalidated before the preflight operation is complete, leading to a use-after-free when passing on the device info, followed by accessing the device serial number. By copying the serial number this can be avoided.
* device: Make sure to operate on the copy of the device list in device_get_list()Nikias Bassen2019-10-071-1/+1
|
* Replace obsolete _BSD_SOURCE define with _DEFAULT_SOURCENikias Bassen2019-10-072-2/+2
|
* preflight: Let insecure notification proxy terminate when pairing is doneNikias Bassen2019-10-041-2/+6
|
* device: Plug memory leak in error condition when TH_SYN sending failsNikias Bassen2019-09-131-4/+4
|
* Avoid using configuration 5mrmacete2019-07-171-4/+8
| | | | - this fixes setting configuration for iOS 11 devices inside virtual machines which caused timeout and subsequent reboot of the device when unplugged from USB
* client: Plug memory leaks in update_client_info()Nikias Bassen2019-03-051-13/+4
| | | | | | plist_get_string_val() was used to get the string values of the respective nodes, however those strings were never freed. Improved by simply copying the nodes instead.
* allow preflight error to pass so simple mode will still workAdam Laurie2019-02-121-1/+3
|
* Allow logging to a dedicated logfileNikias Bassen2019-02-071-16/+35
|
* utils: Fix free on invalid pointerNikias Bassen2018-11-291-1/+1
| | | | Credit to tjps see: #100 https://github.com/libimobiledevice/usbmuxd/pull/100
* client: Fix Use-after-Free in handling of SavePairRecord message when ↵Eric Chen2018-11-291-1/+1
| | | | reading device id
* usb: Fix adding hyphen to new style UDIDs by not counting '\0' towards ↵Nikias Bassen2018-10-121-0/+2
| | | | string length
* usb: Add hyphen to new style UDIDs introduced with iPhone XS/XRNikias Bassen2018-10-051-1/+8
|
* preflight: Fix compatibility with latest internal changes to libimobiledeviceNikias Bassen2018-10-021-1/+5
|
* client: Send 'Paired' message when a device's pairing record has been stored ↵Nikias Bassen2018-09-153-1/+60
| | | | | | | | successfully This requires the SavePairRecord message sent from the client to contain a usbmux device id so the device can be matched accordingly. For the record: This is the original behavior.
* client: Implement ListListeners commandNikias Bassen2018-07-231-4/+117
|
* device: Flush buffer to client when remote side unexpectedly terminates ↵Nikias Bassen2018-05-152-2/+16
| | | | connection
* client: Set socket options for client connections to improve performanceNikias Bassen2018-05-151-0/+12
|
* Remove whitespace errorsNikias Bassen2018-05-091-6/+6
|
* log: Fix timestamps being printed incorrectly when running in foregroundNikias Bassen2018-04-091-5/+11
| | | | | | | | | | Due to usage of wrong function (get_tick_count) the timestamps have been printed incorrectly based on clock_gettime. This commit fixes it by using gettimeofday correctly and also makes sure that this is thread-safe by using localtime_r if available. Furthermore, this commit will also have the effect that when logging through syslog we don't determine the current time anymore because the timestamp is not even used.
* usb: Remove leftover debug log outputNikias Bassen2018-01-311-1/+0
|
* Fix typo in error message.Carlos Maddela2018-01-081-1/+1
|
* usb: Fix calculation of RX loop countNikias Bassen2018-01-081-0/+1
|
* usb: Use heap buffer instead of making the usb_device struct largerNikias Bassen2018-01-081-6/+18
| | | | | Since this buffer is only used during device initialization we don't want the usb_device struct to be unecessary big.
* usb: Use libusb asynchronous I/O for getting initial device information.Alexis Ballier2018-01-081-35/+111
| | | | | | | | | usb_device_add may now be called from libusb main loop via the hotplug callbacks. No blocking call must occur there and libusb 1.0.21 now returns an error when trying to perform blocking I/O in this callback. Should fix the error when hotpluging a device reported in #81
* usb: Store the whole device descriptor in struct usb_deviceAlexis Ballier2018-01-081-4/+3
|
* client: Plug memory leakRudolf Tammekivi2017-12-281-0/+1
|
* conf: Report an error if writing to config file failsNikias Bassen2017-10-291-3/+8
|
* Better error handling when saving pair recordsFrederik Carlier2017-10-292-12/+34
| | | | plist_write_to_filename and buffer_write_to_filename now return 0 if the actual write operation failed (e.g. because access is denied to the file), and set errno if required.
* utils: Use autoconf to check for availability of clock_gettime()Nikias Bassen2017-07-171-1/+1
|
* Try to autodetect ppoll(2) instead of hardcoding the list of platforms ↵Jan Beich2016-01-281-1/+1
| | | | lacking it thus using fallback.
* device: Plug small memory leakNikias Bassen2016-01-281-0/+1
|
* Fix debug message printing the wrong timeout valueNikias Bassen2016-01-281-1/+1
|
* Add a static clock_gettime() substitute for OS X.Aaron Burghardt2016-01-281-0/+33
|
* internal_get_value: fix over-free of node returned by plist_dict_get_item.Aaron Burghardt2015-07-311-1/+0
|
* send_system_buid: fix leak of buid string returned by config_get_system_buid.Aaron Burghardt2015-07-311-0/+1
|
* preflight: Remove obsoleted extern function declarationsNikias Bassen2014-12-021-3/+0
|
* usb: Implement device discovery using libusb hotplug eventsNikias Bassen2014-12-021-202/+268
|
* Rename usb-linux.c to more suitable usb.cNikias Bassen2014-12-022-2/+2
|
* Use non-blocking sockets for client communicationNikias Bassen2014-11-122-1/+31
| | | | | | | This approach is better than using blocking sockets and select() since there's no guarantee that send() doesn't block. Plus we're using poll() anyway so send() and recv() will only be called if the socket is actually ready for writing/reading.
* Revert "client: Make sure fd is writable before calling send() to avoid ↵Nikias Bassen2014-11-121-18/+1
| | | | | | blocking" This reverts commit 23ecea077d8f22d9da5cae50df3e2ff3406fee90.
* client: Make sure fd is writable before calling send() to avoid blockingNikias Bassen2014-11-111-1/+18
|
* Get USB speed for device and use it for device attached client messageMartin Szulecki2014-11-115-2/+34
|
* client: Remove invalid line feed from log messageNikias Bassen2014-11-111-1/+1
|
* client: Log pid of connecting clients (if supported)Nikias Bassen2014-11-111-0/+16
|
* Use new get_tick_count() to avoid timing issues on packetsMartin Szulecki2014-11-114-9/+24
|
* Make sure usbmuxd-proto.h file is added to distributionMartin Szulecki2014-10-171-1/+1
|
* Only conditionally compile bits of udev and systemd supportMartin Szulecki2014-10-161-1/+21
|
View Lorry Controller Status