Since we may not have /cache partition on A/B devices, let recovery
handle /cache related operations gracefully if /cache doesn't exist.
(1) Disable the wipe for /cache partition.
(2) Skip wiping /cache while wiping /data (i.e. factory reset).
(3) Disable logging-related features, until we figure out better
ways / places to store recovery logs (mainly for factory resets on A/B
devices).
Bug: 27176738
Change-Id: I7b14e53ce18960fe801ddfc15380dac6ceef1198
(cherry picked from commit 26112e5870)
If two libraries both use LOCAL_WHOLE_STATIC_LIBRARIES and include a same
library, there would be linking errors when generating a shared library
(or executable) that depends on the two libraries both.
Also clean up Android.mk files.
Remove the "LOCAL_MODULE_TAGS := eng" line for the updater module. The
module will then default to "optional" which won't be built until needed.
Change-Id: I3ec227109b8aa744b7568e7f82f575aae3fe0e6f
Their parents will take care of them.
Good parents.
Fixes double restoration of subpartitions when they are
also set to backup in the recovery fstab.
Change-Id: I876c179135e0cb00754e9a8cfc8eac164c4b7fd4
We used to rely on files (e.g. /cache/recovery/command and
/cache/recovery/uncrypt_status) to communicate between uncrypt and its
caller (i.e. system_server). Since A/B devices may not have /cache
partitions anymore, we switch to socket communication instead.
We will keep the use of /cache/recovery/uncrypt_file to indicate the OTA
package to be uncrypt'd though. Because there is existing logic in
ShutdownThread.java that depends on the existence of the file to
detect pending uncrypt works. This part won't affect A/B devices without
/cache partitions, because such devices won't need uncrypt service (i.e
the real de-encrypt work) anyway.
Bug: 27176738
Change-Id: I481406e09e3ffc7b80f2c9e39003b9fca028742e
When I/O error happens, reboot and retry installation two times
before we abort this OTA update.
Bug: 25633753
Change-Id: Iba6d4203a343a725aa625a41d237606980d62f69
Since we may not have /cache partition on A/B devices, let recovery
handle /cache related operations gracefully if /cache doesn't exist.
(1) Disable the wipe for /cache partition.
(2) Skip wiping /cache while wiping /data (i.e. factory reset).
(3) Disable logging-related features, until we figure out better
ways / places to store recovery logs (mainly for factory resets on A/B
devices).
Bug: 27176738
Change-Id: I7b14e53ce18960fe801ddfc15380dac6ceef1198
This patch mirrors what was done in the main init.rc to relabel
/postinstall.
Bug: 27178350
Bug: 27177071
Change-Id: I39cd03f3c55a42c03367957e8c259c9a3155203c
All updates and fixes applied from upstream libtar as of
March 1, 2016.
Debug flag is disabled, however non-debug output now
provides 1 line of useful output per object extracted.
I've also merged some fixes from CyanogenMod's
fork of libtar:
From: Tom Marshall <tdm@cyngn.com>
Date: Thu, 11 Feb 2016 16:24:40 -0800
Subject: libtar: Cleanup, secure, and extend numeric fields
Commit: e18b457ea1cbf6be1adc3b75450ed1c737cd82ea
From: Tom Marshall <tdm@cyngn.com>
Date: Thu, 11 Feb 2016 12:49:30 -0800
Subject: libtar: Make file sizes 64-bit clean
Commit: e628c2025549a24018bc568351465130a05daafb
From: Tom Marshall <tdm@cyngn.com>
Date: Thu, 17 Apr 2014 09:39:25 -0700
Subject: libtar: Add methods for in-memory files
Commit: 8ec5627a8ff0a91724c6d5b344f0e887da922527
From: Tom Marshall <tdm@cyngn.com>
Date: Wed, 2 Jul 2014 09:34:40 -0700
Subject: libtar: Fix hardlink extract
Commit: 166d83a51e0c51abcea37694dbd7df92d03c1f56
From: philz-cwm6 <phytowardt@gmail.com>
Date: Sat, 26 Apr 2014 01:11:35 +0200
Subject: libtar: Various bug fixes and enhancements
Commit: a271d763e94235ccee9ecaabdb52bf4b9b2f8c06
(Some of this was not merged in, as better solutions were
available from upstream libtar)
From: Tom Marshall <tdm@cyngn.com>
Date: Wed, 9 Apr 2014 09:35:54 -0700
Subject: libtar: Add const qualifiers to reduce compile warnings
Commit: 0600afa19fe827d06d3fcf24a7aabd52dbf487b4
Change-Id: I6d008cb6fdf950f835bbed63aeb8727cc5c86083
On some devices TWRP tries to access the framebuffer before all
device nodes have been created. Retry opening instead of crashing later.
Change-Id: I189a8fe80a8906b46fb6cece53c0bf83c00c0e09
Some people insist on running make without specifying any target
like recoveryimage or otapackage, which triggers a build of
verifier_test, so this fix will keep people from asking about
this build error anymore.
Change-Id: Ic955b7a3fb2599894e712fd8b418227da240d288
bionic process initialization calls personality
(specifically personality-8)
personality wants to load a kernel module
loading a kernel module calls /sbin/modprobe
loading /sbin/modprobe is a bionic process initialization
bionic process initialization calls personality
personality wants to load a kernel module
loading a kernel module calls /sbin/modprobe . . .
Before you know it, it takes 0.5 seconds to do anything.
Note: modprobe is still technically available, but the symlink
has been removed, so you can still call it directly by running
busybox modprobe if you like.
From what I can tell, this issue only affects 32 bit devices
with CONFIG_MODULES=y in the defconfig. The problem can be also
patched out of the kernel by commenting or otherwise removing
the block of code in kernel/exec_domain.c inside the CONFIG_MODULES
ifdef block
It is also possible to patch the problem in bionic libc by commenting
out or otherwise removing the __initialize_personality in bionic/libc/
bionic/libc_init_common.cpp file.
Change-Id: Iebac314616080ac18320d73b087980ac1b98b951
If a path is specified for Find_Next_Storage() and datamedia is
excluded, the function currently runs to the end of available
partitions and reports no other storage is available. Fix this by
restoring the original mount_point == search_path check to break the
loop. The subsequent partitions loop will skip data-media storage as
needed.
Change-Id: I92baed866cd36f13d851bd58ba5346e47edb142e
Tao Bao
Alex Deymo
James Christopher Adduono
Tianjie Xu
Mattias Nissler
The Android Automerger
Prashant Malani
Matt Mower
that
Ethan Yonker
Paul Lawrence