summaryrefslogtreecommitdiffstats
path: root/src
AgeCommit message (Collapse)AuthorFilesLines
2013-10-24Return ‘DeviceTree’ instead of ‘RestoreDeviceTree’ from ↵Gravatar Aaron Burghardt1-1/+1
get_component_name(). Returning ‘RestoreDeviceTree’ is inconsistent with other component names, and the caller apparently handles the need for it. Also, the API doc for plist_dictionary_insert_item states that it asserts if the key to be inserted already exists. if plist_dictionary_insert_item actually does assert as described, the caller of get_component_name triggers that assertion if ‘RestoreDeviceTree’ is returned. Signed-off-by: Martin Szulecki <m.szulecki@libimobiledevice.org>
2013-10-24Fall back to ‘RestoreKernelCaches’ if ‘KernelCachesByTarget’ not in ↵Gravatar Aaron Burghardt1-12/+16
Restore.plist. Signed-off-by: Martin Szulecki <m.szulecki@libimobiledevice.org>
2013-10-24Fixed misuse of free() instead of plist_free().Gravatar Aaron Burghardt1-1/+1
Signed-off-by: Martin Szulecki <m.szulecki@libimobiledevice.org>
2013-10-24Fixed pass-by-reference inconsistency in asr_send().Gravatar Aaron Burghardt2-2/+2
Signed-off-by: Martin Szulecki <m.szulecki@libimobiledevice.org>
2013-10-03AppleTV: set auto-boot to true again after restoreGravatar Nikias Bassen3-1/+16
Apparently AppleTV units don't boot up in normal mode after restore but switch into recovery mode so we set auto-boot to true to make them boot up in normal mode.
2013-10-02ipsw: change buffers to use type unsigned char*Gravatar Nikias Bassen5-23/+26
2013-10-02dfu: make sure buffers are of type unsigned char*Gravatar Nikias Bassen2-7/+7
2013-10-02tss: make sure buffers are of type unsigned char*Gravatar Nikias Bassen3-21/+21
2013-10-02img3: make buffers and sizes unsignedGravatar Nikias Bassen3-16/+16
2013-10-02use dynamically allocated buffer when reading cached SHSH filesGravatar Nikias Bassen1-3/+24
2013-10-02silence implicit declaration compiler warningsGravatar Nikias Bassen9-2/+17
2013-10-02common: supress more compiler warningsGravatar Nikias Bassen2-1/+3
2013-10-02asr: supress compiler warnings with proper castGravatar Nikias Bassen1-2/+3
2013-10-02custom fw: fall back to User ram disk if Update ramdisk is not foundGravatar Nikias Bassen1-7/+14
This is a fix for AppleTV firmwares since they don't have an Update ram disk but only a User ram disk. When restoring a custom firmware idevicerestore fails if -e parameter is not given. This fix will automatically set erase mode enabled if no update ram disk is found.
2013-09-27Require libirecovery >= 0.2.0 and port code to it's new APIGravatar Martin Szulecki11-89/+121
2013-09-27ipsw: also check product type in ipsw_get_latest_fw to get largest major versionGravatar Nikias Bassen1-3/+6
2013-09-27ipsw: remove obsolete major version increase in ipsw_get_latest_fwGravatar Nikias Bassen1-8/+2
2013-09-27ipsw: fix version data parsingGravatar Nikias Bassen1-41/+9
2013-09-24recovery: fix possible segfault and plug some small memory leaksGravatar Nikias Bassen1-4/+20
2013-09-23idevicerestore: Update usage output to use a more common standardGravatar Martin Szulecki1-21/+21
2013-09-23normal: Fix pair record removal due to changed libimobiledevice APIGravatar Martin Szulecki1-1/+1
2013-09-23ipsw: Add a temporary workaround to support iOS 7 firmware lookupsGravatar Martin Szulecki1-0/+5
It appears that iTunes does no longer parse the version XML structure as currently implemented in idevicerestore. We'll need to modify the lookup for the latest firmware in the future, this will do for a while.
2013-09-23Update AUTHORS and file copyrights accordinglyGravatar Martin Szulecki26-24/+110
2013-09-23restore: Handle "Failed to mount filesystems" status messageGravatar Martin Szulecki1-0/+3
2013-07-04common: do not recurse mkdir_with_parents if dirname(dir) == dirGravatar Nikias Bassen1-1/+1
2013-06-15img3: don't fail if SHSH is appended (instead of replaced)Gravatar Nikias Bassen1-4/+0
2013-04-24limera1n: use chip_id constants from libirecovery to detect device modelGravatar Nikias Bassen1-6/+11
2013-04-24limera1n: verify device mode after sending exploit to check for failureGravatar Nikias Bassen3-6/+12
2013-04-24main: updated version URLGravatar Nikias Bassen1-1/+1
2013-03-06Disable SSL verification to allow downloads from untrusted https locationsGravatar Martin Szulecki2-0/+8
2013-01-11idevicerestore: removed unused cpid/bdid variablesGravatar Nikias Bassen1-2/+0
2013-01-11idevicrestore: don't use info as variable name (collides with info() function)Gravatar Nikias Bassen1-24/+23
2013-01-11restore: remove unused variableGravatar Nikias Bassen1-3/+0
2013-01-11restore: don't use info as variable name (collides with info() function)Gravatar Nikias Bassen1-6/+6
2013-01-11fix possible memory corruption due to bad cast from int to uint64_tGravatar Nikias Bassen1-1/+3
2013-01-09idevicerestore: Improve maintenance of recovery/dfu mode device compatibilityGravatar Martin Szulecki1-193/+5
Instead of maintaining a large switch() monster which replicates the exact same logic that was already used in libirecovery, using the dfu_check_device() helper to quickly determine the device automatically is a better approach.
2013-01-09dfu: Implement dfu_check_device() helperGravatar Martin Szulecki2-0/+22
2013-01-09dfu: Fix minor code style issueGravatar Martin Szulecki1-1/+1
2012-12-07dfu: Fix codepath after uploading iBSS due to typoGravatar Martin Szulecki1-1/+1
2012-12-07restore: Fix "hanging" and not exiting on failed restore due to variable typoGravatar Martin Szulecki1-1/+1
2012-11-08idevicerestore: Fix compilation due to typoGravatar Martin Szulecki1-3/+3
2012-11-08Add basic support for iPad3,4 and iPad2,5 (untested)Gravatar Martin Szulecki1-0/+21
2012-11-07idevicerestore: Show build major version number, tooGravatar Martin Szulecki1-1/+1
2012-11-07normal: Fix compiler warningsGravatar Martin Szulecki1-1/+2
2012-11-07idevicerestore: Make first nonce request not mandatory to support "old" iOS ↵Gravatar Martin Szulecki1-2/+2
builds Devices running iOS 4.x and older do not use or supply a nonce before an actual iBEC of the target firmware version is loaded which does so. This caused older devices to fail to restore.
2012-11-07idevicerestore: Move nonce reporting into get_nonce() to avoid code duplicationGravatar Martin Szulecki1-15/+14
2012-11-07Added support for iPhone5,1 and iPhone5,2 (including baseband update)Gravatar Nikias Bassen2-0/+21
2012-11-07restore: Change wording to "personalized" instead of "signed"Gravatar Martin Szulecki1-2/+2
2012-11-07restore: Plug possible memory leakGravatar Martin Szulecki1-2/+1
2012-11-07idevicerestore: Only sign components if we have a blob and don't treat it as ↵Gravatar Martin Szulecki1-27/+38
error Since iOS 6 some blobs are not provided in the TSS response. However, that is not fatal as the missing blobs are presigned already on iOS 6.