Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
|
|
Also fixing the string length of the affected strncmp.
Credit to Mällon Lobjakas
|
|
Starting with iOS 10.3 the ipsw files don't contain any 'manifest' file
anymore which we used to determine the files required for the reply to the
'NorData' request. Since this info is also included in the BuildManifest
we us that instead, unless a 'manifest' file is present.
|
|
memory leaks
|
|
conditions when tempnam generated name is not random enough.
|
|
firmware
|
|
The iPhone 7/7+ models with Mav16 baseband failed to restore as the baseband
firmware files are now called RestoreSBL1-HastTableDigest and
SBL1-HashTableDigest. The code only stripped off the '-Blob' part from the
baseband tss response so it couldn't match the firmware filenames properly.
This commit fixes the issue by only keeping the first part of the string
(everything before the first occurance of '-').
|
|
While not tested, the new Mav16 baseband firmware files should
also work with this update.
|
|
|
|
the restore
|
|
|
|
|
|
it for iOS 9+
|
|
|
|
|
|
|
|
|
|
|
|
In very rare cases the changes introduced by commit can actually cause
an invalid free. Reverted those changes. Also, plist_free() will check
for NULL before doing anything, so an extra check for that is not needed.
|
|
|
|
libimobiledevice version
|
|
identity to make sure we use correct values.
|
|
|
|
versions.
|
|
|
|
|
|
Only send an empty reply but this seems to be enough to restore
devices that already have valid data.
|
|
|
|
|
|
|
|
|
|
|
|
request
|
|
|
|
|
|
|
|
Apparently restored is very picky about the order and expects the iBoot
component as the first item in the array. This fixes restore of new models.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|