- 24 Oct, 2016 1 commit
-
-
Mark Salyzyn authored
(cherry pick from commit fa07f9dc) mLastWorstPidOfSystem is filled with iterator references that are not from AID_SYSTEM to aid the performance. But we only clear entries from the list during erase if they are from AID_SYSTEM. Remove the filter check in erase so the stale references will be removed. The conditions that caused this failure are difficult to reproduce and are rare. Test: gTests logd-unit-tests, liblog-unit-tests and logcat-unit-tests Bug: 32247044 Bug: 31237377 Change-Id: Ie405dd643203b816cac15eef5c97600551cee450
-
- 21 Oct, 2016 2 commits
-
-
Tim Murray authored
-
Tim Murray authored
The binder fd doesn't need to use EPOLLWAKEUP and shouldn't; if it does, it will prevent suspend unnecessarily. bug 32180327 Change-Id: I292f1ada0c698290f6bd5518b237867ed69ec391
-
- 18 Oct, 2016 2 commits
-
-
Tim Murray authored
-
Tim Murray authored
This reverts commit fb896944. bug 32180327 Change-Id: Ic26bdffac6c56b102348f5393e6eb5d895960d7e
-
- 12 Oct, 2016 1 commit
-
-
Lingfeng Yang authored
bug: 31950237 There are two lists of active ADB transports (devices), and with the emulator, they can go out of sync. This CL more conservatively checks if there are no transports in either list before commiting to register a new transport for the emulator. Change-Id: Id1201dc59c70825881dad80925c2e5bcc13dbd5e (cherry picked from commit edaedfd5da41b2f5aa14b4d52742a6d8caa49214)
-
- 11 Oct, 2016 1 commit
-
-
Tim Murray authored
If for some reason the system can't suspend (usually a driver bug), libsuspend will currently attempt to retry suspend after 100ms. Because entering suspend takes a significant amount of CPU time, this can be extremely expensive and be a major contributor to rapid battery drain. Move autosuspend to use exponential backoff if the previous suspend attempt failed. bug 32092914 Change-Id: I3e9e944f290de5f1853a02e3f61721ba9159df46
-
- 10 Oct, 2016 1 commit
-
-
Glenn Kasten authored
-
- 09 Oct, 2016 1 commit
-
-
Glenn Kasten authored
Bug: 31996208 Test: will need a CTS, not yet done Change-Id: I5ecc7f0519d42a83065b7b97a31cdb5b33549cda
-
- 07 Oct, 2016 9 commits
-
-
Tianjie Xu authored
am: 167562aa Change-Id: Icb066cdaf63d87324c6eca8b5b45dbc23235aaa2
-
Tianjie Xu authored
am: 87885912 Change-Id: I2f0ca98c74f23b1125f29b1eaf810143a83754b4
-
Tianjie Xu authored
am: c24dd976 Change-Id: Ifddcfb7e658156c22539e07915115f4b657925b1
-
Tianjie Xu authored
am: 1ee4892e Change-Id: I26b48df5a9d39933b90ac063172b4b4ae8428bfd
- 26 Sep, 2016 1 commit
-
-
Robb Glasser authored
-
- 23 Sep, 2016 1 commit
-
-
Tianjie Xu authored
The boundary check of an invalid EOCD record may succeed due to the overflow of uint32_t. Fix the check. Test: Open the crash.apk and libziparchive reports the offset error as expected. Bug: 31251826 Change-Id: I1d8092a19b73886a671bc9d291cfc27d65e3d236 (cherry picked from commit ae8180c06dee228cd1378c56afa6020ae98d8a24)
-
- 21 Sep, 2016 13 commits
-
-
Narayan Kamath authored
am: c681771b Change-Id: I0fb6263aa98124acb03e79b9c102c7ee6d3e94ba
-
Narayan Kamath authored
am: af43cd03 -s ours Change-Id: Ifd07ba06cecf61ba53af64bae08f0dac679c9e06
-
Narayan Kamath authored
am: b5f75c45 Change-Id: Iacea86ad3c142d6f14b5fbdfb5b615f4705b1bd1
-
Narayan Kamath authored
am: 61f27551 Change-Id: I418fb945e685e2983187c2ba43ad66ec99b468ec
-
Narayan Kamath authored
Test: test_zipfile with known bad files. Change-Id: I25dde73c9b94d253c8017e441e89cb7339907c7f
-
Narayan Kamath authored
am: c307830f Change-Id: I5d856cec4528541ff5416456cbe1338c78a09699
-
Narayan Kamath authored
am: a1ec2363 Change-Id: I290e611776bd58b47266a2cb9371a832bdc0744e
- 20 Sep, 2016 3 commits
-
-
Dan Willemsen authored
am: 15366542 Change-Id: Idb1bea870de3251d7fa9682526e9634fe965424d
-
Dan Willemsen authored
When building without vendor/... in the tree, we won't be creating a vendorimage, so BOARD_VENDORIMAGE_FILE_SYSTEM_TYPE won't be defined. But we still need to use the vendor image that will be present. Bug: 31545062 Bug: 30040825 Change-Id: I756adc615aea11da60da4d3a4568caca9fed93ab
-
Mark Salyzyn authored
This reverts commit f894b96a. Kernel binaries have incorporated the performance-enhanced pmsg driver that has removed the bounce buffer. Bug: 31057326 Bug: 30375418 Change-Id: Id6cec3ac144f57e8e5465aa48bbdc829cf29e283
-
- 13 Sep, 2016 1 commit
-
-
Narayan Kamath authored
- Make sure the start and end of the data for a given entry are within the allocated buffer. - Make sure all central directory entries start and end within the central directory buffer. - Reject zip file entries that have no filenames. bug: 30916186 test: test_zipfile with known bad zip files. Change-Id: Ibf3f6469e60c85ec1608f5ce613d40867d2d09b7
-
- 08 Sep, 2016 3 commits
-
-
Mark Salyzyn authored
am: 1b1b6a48 -s ours Change-Id: I21849cf64861536e911f94d95181822ffb06805c