OSDN Git Service
Chih-Wei Huang [Thu, 14 Feb 2013 14:01:11 +0000 (22:01 +0800)]
ueventd: load modules in a forked process
Loading some modules (e.g., ath3k) will be blocked to wait for
firmware. The forking is necessary, otherwise ueventd has no
chance to load the firmware.
Chih-Wei Huang [Wed, 6 Feb 2013 12:52:34 +0000 (20:52 +0800)]
allow to load modules from standard module path
Chih-Wei Huang [Thu, 25 Jul 2013 08:01:59 +0000 (16:01 +0800)]
Merge branch 'android-ia3' into jb-x86
Conflicts:
adb/file_sync_service.c
fs_mgr/fs_mgr_priv.h
include/system/graphics.h
init/devices.c
init/init.c
init/init.h
init/readme.txt
libsuspend/autosuspend.c
toolbox/Android.mk
Ken Lierman [Thu, 20 Jun 2013 16:27:13 +0000 (09:27 -0700)]
Bound the ADB connect time with a non-blocking connect
After a disconnect, the initial blocking connect takes
a long time to return, while subsequent calls return
quicks. Switch to a non-blocking connect to make the
re-connect time more consistent and faster overall.
Issue: AXIA-2990
Change-Id: Ide95f75c7d763e1c2071408659e7a2aa7a13d9a0
Robert Chiras [Tue, 18 Jun 2013 14:14:03 +0000 (17:14 +0300)]
Fix adb connect when trying to reconnect
If adb tries to connect to a disconnected device while a reconnect
thread is already trying to connect, adb just printed an error but
didn't abort the current connect sequence.
So, if we have a reconnect thread handling a connection also exit after
printing the error.
Change-Id: If09ea2486f181c620033e1a755360c7913e8eb48
For: AXIA-3129
Signed-off-by: Robert Chiras <robert.chiras@intel.com>
Chih-Wei Huang [Tue, 20 Nov 2012 17:50:51 +0000 (01:50 +0800)]
init: handle DEBUG and HWACCEL variables
On DEBUG=1, trigger actions in class debug.
On HWACCEL=0, set debug.egl.hw=0 which disable hardware acceleration.
Chih-Wei Huang [Sun, 29 Apr 2012 07:04:55 +0000 (15:04 +0800)]
Set correct permission for pppd
The pppd needs setuid root to setup ppp interface.
Chih-Wei Huang [Sun, 1 Jan 2012 19:16:41 +0000 (03:16 +0800)]
init: avoid virtual console be set blanked
Chih-Wei Huang [Thu, 14 Apr 2011 01:48:52 +0000 (09:48 +0800)]
init: builtins: Add 'setkeycode' command to init.rc parser
The command is used to set an entry into the kernel's scancode-to-keycode map.
Change-Id: If772b3590c08d3696ff9d015f1faf0e75405e262
Chih-Wei Huang [Tue, 12 Apr 2011 08:46:45 +0000 (16:46 +0800)]
init: make sure the last parameter to execve is NULL
Extend the buffer to hold one more pointer, and set it is NULL.
Change-Id: I3018d62a1b0d61c0a3f5f2f7ffa12c7ce9ee32c5
Chih-Wei Huang [Tue, 1 Feb 2011 07:47:03 +0000 (15:47 +0800)]
init: change FIRMWARE_DIR1 for Android-x86
Change-Id: I0f54d1dcfcd7b1fe4b5f1dbe4ee3aa7289a5ecea
Chih-Wei Huang [Sat, 15 Jan 2011 22:44:37 +0000 (06:44 +0800)]
ueventd.rc: remove redundant definitions
Change-Id: Icb4f051abf941052aca9752ca2949386738d94e5
Chih-Wei Huang [Sat, 15 Jan 2011 22:25:37 +0000 (06:25 +0800)]
init.rc: use relative path for symlink
Usually it's better to use relative path for symlink. For example,
to use Android in a chroot environment, the absolute symlink will
fail outside the chroot environment. Using relative path guarantees
the link will work both inside and outside chroot.
Change-Id: I2f654a05189b9a666aed1fc7bdfb5a50357aff78
Magnus Malmborn [Fri, 14 Jan 2011 08:42:18 +0000 (09:42 +0100)]
Adding an 'exec' implementation to init
This is useful for running programs synchronously
during startup.
Change-Id: I0212fad22613124dd13b4f20b86d0dcc6cb59b6e
Weichuan Yan [Tue, 28 Dec 2010 07:45:02 +0000 (15:45 +0800)]
Fix a crash when there is no command action in some trigger
Change-Id: I5c002c9f86afceca83c39715576392cf88dd8614
Signed-off-by: Weichuan Yan <wchyan@marvell.com>
Jim Huang [Wed, 9 Sep 2009 03:47:40 +0000 (11:47 +0800)]
Let console applications have a controlling tty.
Since the logic in open_console() is changed, terminating console
application within a shell is now possible via Ctrl-C.
Ajay Dudani [Sun, 21 Jul 2013 16:03:41 +0000 (09:03 -0700)]
init: Increase maximum number of Android properties to 372
Bug:
9941046 -- encryption failed
Change-Id: Ib9c2921a2b574542cce98d2eadbd9204a1ad0bf0
Signed-off-by: Iliyan Malchev <malchev@google.com>
Robert Greenwalt [Tue, 16 Jul 2013 16:46:17 +0000 (09:46 -0700)]
Add writable data space for radio.
Storing carrier provisioning urls updates in /data/misc/radio.
bug:
9623159
Change-Id: I8d62d4638229733dea0f11f1729c4d22ae2295d1
I36697ed341353b7a3dbec5afe20241102e76f6f1
The Android Automerger [Sun, 16 Jun 2013 12:55:49 +0000 (05:55 -0700)]
merge in jb-mr2-zeroday-release history after reset to jb-mr2-dev
Andy Ross [Tue, 28 May 2013 18:02:25 +0000 (11:02 -0700)]
Allow for overriding environment variables in export
Don't silently append duplicates, delete the older version. Also
increase the size limit, as the set of shell variables on my test
system is already at 28.
Change-Id: Id2373cc197662ae14fabad4de5e4516fe30d44cc
For: AXIA-2870
Signed-off-by: Andy Ross <andy.ross@windriver.com>
Ken Lierman [Mon, 6 May 2013 23:45:48 +0000 (16:45 -0700)]
Have TCP connections attempt to auto-reconnect
2 major chunks of changes.
First, TCP connections are setup with keep-alive turned on, so when the device
end goes away, the server finds out in a reasonable time.
Second, when a link goes down, spin off a reconnect thread to periodically
attempt to reconnect. React appropriately to adb devices, connect, and
disconnect commands.
Issue: AXIA-2515
Change-Id: I17d8031ae3238260a3c5d584e5002388ac96b182
Ken Lierman [Wed, 1 May 2013 17:36:45 +0000 (10:36 -0700)]
cleanup ADB's debug output
The server was using stderr to talk back to the client to ACK it
starting correctly, which made the debug output from the server get
lost (before it later gets redirected to a file). Change it to stdout
and add a fflush() to force the output.
Add the pid to the output so you can tell which process the output is
coming from.
Add basename to the __FILE__ part of the output to remove the unnecessary
and redundant system/core/adb path info from every output line.
Remove the duplicated fdevents debug output macro so it gets redirected
into the server log file like it should.
Change-Id: I8eaf3c8ccbca62f907b0ee7b52a0e179db5ff82d
Sean V Kelley [Wed, 1 May 2013 21:35:07 +0000 (14:35 -0700)]
system: Add hw specific pixel format
Relocate hw pixel format to graphics header for convenience
from drm_gralloc codebase.
Change-Id: I405398b172fab19949fef33c89a60132b1bd0ea9
Signed-off-by: Sean V Kelley <sean.v.kelley@intel.com>
Billy [Thu, 14 Mar 2013 08:00:23 +0000 (16:00 +0800)]
Init: Setting system time back to utc
Fix:
When rtc_local_time is enabled (ro.rtc_local_time = 1),
system time loaded from rtc should be local time too during boot.
Setting system time back to utc as soon as fast, which avoid programs
get confused about it.
For convert local time to utc, we need to get timezone information.
Android store timezone in property variable "persist.sys.timezone", and
its raw file stored on /data partition. So, the reasonalbe time when we
trigger sysclk should be after /data partition loaded and persist_props
loaded. We must add a new property variable for triggering sysclk if we
want to trigger it after load_persist_props commmand, which will make code
more complicated. Thinking about above, I add sysclk command on post-fs-data
section in init.bigcore.rc. At this point, partitions loaded, persist_props
not loaded yet. I get the timezone infromation through reading raw file
directly instead of call propert_get function
Issue: AXIA-2051
Change-Id: I0ff1d5932c5b1aaa8695129de44c06dba31cb443
Signed-off-by: Xiaobing Feng <xiaobing.feng@windriver.com>
ravindranath [Tue, 2 Apr 2013 14:53:11 +0000 (10:53 -0400)]
MUST_REVERT ueventd: Add USB VID/PID rule type
Add a new type of rule in ueventd.rc files to intercept 'Add' uevents
for specific USB devices, as identified by vendor ID and product ID,
and set permissions on the device nodes as requested by the user.
This is a MUST_REVERT patch at Andrew's request because the implementation
is not as general as we'd eventually like. See AXIA-2364 for tracking
the reversion of this patch.
Issue: AXIA-2229
Change-Id: Ie8e4feaee87175f3c97899dd94f7cc1432dd2de3
Signed-off-by: Ravindranath Doddi <ravindranathx.doddi@wipro.com>
Signed-off-by: Krishnan V <krishnanx.vaidyanathan.venkitakrishnan@intel.com>
Signed-off-by: Matt Gumbel <matthew.k.gumbel@intel.com>
Thomas Daniel [Mon, 7 Jan 2013 17:02:31 +0000 (17:02 +0000)]
adbd: set sent file ownership per android fs config table
do_send() uses fs_config() to retrieve mode, user and group for a sent file but
handle_send_file() only gets passed the mode. User and group are discarded.
This causes problems as it means any files updated with adb sync will have their
owner set to root.
For example, the file "/system/etc/dbus.conf" should be owned by "bluetooth" as
specified in android_filesystem_config.h, and its permissions should be read for
user and group but nothing for other. Using adb sync to send this file means
that the owner is set to root, and no access permissions for others. This means
that when a process running as "bluetooth" tries to access the file it fails,
resulting in a boot failure.
This change adds a new wrapper function adb_chown() which calls chown() to match
the existing pattern for wrapping file I/O functions. This adb_chown() is now
called in do_send() to correctly set the owner and group as specified in the
fs_config table.
Change-Id: I37661273780346a6a868e58d9069171052e81e34
Andrew Boie [Tue, 19 Mar 2013 18:30:48 +0000 (11:30 -0700)]
always try to create by-name links based on install id
Previously, if a block device was detected as a platform device,
the by-name logic which uses install id was skipped in favor of
the older logic which includes the bus in the path name. This
resulted in the installation of Android failing on these devices
since the expected devices nodes don't show up after the disk is
partitioned.
Now always try to use the install id logic, and fall back to the
older mechanism only if it doesn't produce any symlinks.
Change-Id: Idc096eb6425af478ddca7fe4ba9c0fe24b922d6e
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Matt Gumbel [Wed, 14 Nov 2012 18:16:17 +0000 (10:16 -0800)]
Support adb client connect to remote server
ADB client: allow user to specify hostname and port number of remote
adb server.
ADB server: bind server to all network interfaces instead of just
localhost when user gives -a flag.
Primary use-case for this change is to support remote testing of USB
devices. HostA is running some test automation software which invokes adb
client. HostB has USB-only device attached and is running adb server. adb
client on HostA makes connection to adb server on HostB to talk to the
USB device.
Change-Id: I845cc8c00350b400317f8c18f813e6fd79bd5470
Signed-off-by: Dean Kwon <daex.i.kwon@intel.com>
Signed-off-by: Jim Bride <jim.bride@intel.com>
Signed-off-by: Matt Gumbel <matthew.k.gumbel@intel.com>
Matt Gumbel [Fri, 4 Jan 2013 01:15:47 +0000 (17:15 -0800)]
Fix dhcpd startup when net.hostname is not set
Commit
62d6f74 swapped p2p_interface and DHCP_CONFIG_PATH
erroneously. This reverts them to their correct ordering and makes dhcpcd
startup work again.
Change-Id: Iea033a7e0dad98bb2a63fb39755330675cfbb0ab
Signed-off-by: Matt Gumbel <matthew.k.gumbel@linux.intel.com>
Andrew Boie [Fri, 21 Dec 2012 19:06:57 +0000 (11:06 -0800)]
charger: fix build
Change-Id: I796c71dca32cd879405f10be9fa17edba8aa100e
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Daniel Leung [Tue, 11 Dec 2012 16:39:57 +0000 (08:39 -0800)]
libsparse: Use BIONIC implementation of mmap64()
Change-Id: Ifad484c8b1a22f7b1a0ef5dfd52d7fe12fd892c6
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Andrew Boie [Tue, 4 Dec 2012 23:47:20 +0000 (15:47 -0800)]
ueventd: Increase uevent buffer size
As it turns out, no, 64K is not enough. A device provisioning
tool was generating many uevents as it manipulated the GPT,
causing events near the end of the process (including events
as a result of the final BLKRRPART ioctl()) to get lost.
It's not clear what the best value for this should be, but
increasing to 256K fixed the problem in this case.
Change-Id: I4883b34e96c89e8a6fa581bc9cd121bb021b5694
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Andrew Boie [Thu, 25 Oct 2012 21:10:07 +0000 (14:10 -0700)]
init: expand usage of property lookup and allow file lookup
Currently, the init.rc syntax ${prop.name} to substitute a value
looked up from the property service is only supported for the
setprop, write, and import commands. Here we generalize it so that
the arguments for all commands and services allow for references like
this.
We also introduce the syntax $[/path/to/file] to substitute the
contents of a file (typically a sysfs node), pruning any trailing
newlines from the file first.
Substitutions are always done at the last possible moment; right
when the command is executed or the service is started. If a service
is started/stopped multiple times, the substitutions will always be
re-done each time it starts.
If any substitition fails, the command will not run/service will not
start.
The deprecated syntax $prop.name is no longer supported.
Change-Id: I8560b70f48c0396832d9a7fadc546441e8e81ff9
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Andrew Boie [Fri, 16 Nov 2012 22:51:19 +0000 (14:51 -0800)]
init: create by-name device links for marked GPT partitions
Create device node symlinks by name if the uevent for a block
device has a name in it, and the name passed in matches the
prefix for the installation ID. This prevents collisions with
multiple Android installations on the same system. The created
symlinks do not have the prefix in them.
If the installation ID isn't set (through ro.boot.install_id)
this mechanism doesn't do anything. There is a possibility that
specially crafted removable storage could alias internal storage
through this mechanism, so locked-down devices won't want to use
this.
Change-Id: I04782691d4e5ff0db6f68d8d9345a8585cfe5f86
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Andrew Boie [Mon, 6 Aug 2012 17:49:08 +0000 (10:49 -0700)]
init: remove logo code
This code is legacy and should never need to be used. The bootloader
should put an image on the screen and the kernel shouldn't reconfigure
the display controller until userland is ready to draw something,
such as the boot animation, charger, recovery console, etc.
Instead of hard-coding tty0, the "A N D R O I D" text is written out
to whatever the active console is.
Change-Id: I5fd21413f721d2ba6966d34d632d8fc3913ca073
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Daniel Leung [Mon, 22 Oct 2012 23:32:52 +0000 (16:32 -0700)]
libsuspend: Use autosleep first if available
Change the order to look for sysfs nodes to do autosleep first,
then early suspend.
Change-Id: Iff34b9354dd8e5104befd05e8097962d7f7eee7f
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Andrew Boie [Tue, 30 Oct 2012 22:16:04 +0000 (15:16 -0700)]
init: Allow for configurable init.rc path
If androidboot.initrc is passed in the kernel command line,
the path it refers to will be used for the initial init.rc
and not /init.rc.
This is to support specialized boot modes using the same
ramdisk.
Change-Id: Icfa179c422bf1d548c4ddef33c41cc89718d762d
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Andrew Boie [Fri, 12 Oct 2012 22:23:40 +0000 (15:23 -0700)]
init: Add 'readprops' command
This reads the contents of a text file and sets system properties
based on its contents.
Change-Id: Ic03de7ee9afb8a12439d0dff343560011e152beb
For: AXIA-102
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Bruce E. Robertson [Mon, 3 Dec 2012 21:01:28 +0000 (13:01 -0800)]
ADBD: set sent file permissions per android fs config table.
This fix allows the standard CTS installation to succeed while
preserving the previous fix's intent to allow init to operate
properly.
The first CTS APK installation failed when adbd, the adb daemon, on
the target invoked pm (package manager). "pm" runs as another user
than adbd and could not open the APK because adbd had cleared the
"other" read permission.
For: AXIA-813
Change-Id: If27a628072f51ffac78db7b5448670ea4666575f
Signed-off-by: Bruce E. Robertson <bruce.e.robertson@intel.com>
Conflicts:
adb/file_sync_service.c
Octavian Purdila [Thu, 27 Sep 2012 16:28:43 +0000 (19:28 +0300)]
devices: allow sending on device-add notifications for major/minor-less devices
This is usefull for getting device plug/unplug notifications for
things like USB ethernet adapters which do not have major/minor
numbers.
Change-Id: I8f032ef983aeb36dd013fafc4a5df5238decbb17
Signed-off-by: Octavian Purdila <octavian.purdila@intel.com>
Andrew Boie [Tue, 27 Nov 2012 20:58:19 +0000 (12:58 -0800)]
init: HACK: Try multiple times to open console
If the console is set to be a USB-to-serial adapter, there is
a bootup race for the creation of its associated device nodes
and init's one-time console detection.
If the console can't be opened, try polling up to 5 seconds. Log
an error if it ultimately fails.
A better solution would be to reimplement the console functionality
to be hot-pluggable, but this would require much more invasive changes
to make it robust.
Change-Id: I20b19095a2f22efffafe2a7ac22d2be753a62d4b
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Daniel Leung [Tue, 7 Aug 2012 22:03:34 +0000 (15:03 -0700)]
libcutils: wait for properties and value comparison
Add a property_try_get() function to wait for property to appear.
This can be used to block operations until some properties are set.
Add property_cmp() and property_try_cmp() to compare property
to specified value.
Also add a trygetprop, cmpprop, trycmpprop command in toolbox
to use these.
Issue: AXIA-539
Change-Id: I9d531e57c84140cebe6440efe3d3195c505932fa
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Andrew Boie [Tue, 27 Nov 2012 21:21:38 +0000 (13:21 -0800)]
init: wildcard matching for chmod/chown in init.rc
Implement wildcard matching for the builtin command
chmod and chown for use in init.rc.
For security reason, the number of slashes has to match.
For example, to match /dir_a/dir_b/dir_c/dir_d/file,
the path needs to be /dir_a/*/*/*/file. Simply doing
/dir_a/*/file does not work.
Issue: AXIA-576
Change-Id: I21b5aa58b3d3142725bc73ff31afb65d4fd4bc99
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Andrew Boie [Wed, 25 Jul 2012 21:33:39 +0000 (14:33 -0700)]
fs_mgr_priv: Extend fs wait timeout to 20 seconds
In practice 5 seconds is too short to wait for a disk device node to
show up if the disk is USB; 20 seconds is a much more comfortable
window.
Change-Id: Iaf2c1f46b41a44fc1240d52d8498ca9cb639ea80
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Andrew Boie [Tue, 29 Nov 2011 20:45:58 +0000 (12:45 -0800)]
util.c: Fix file_read() with sysfs nodes
Sysfs nodes always report a file size of 4096 regardless of how much
data can be read. Hence don't fail if read() can't get the full size;
specifically look for errors or 0 (which is EOF).
Change-Id: Ief1ffbe08ac5e4daa80805713efa4f2af0a1a33e
Daniel Leung [Thu, 24 May 2012 23:54:49 +0000 (16:54 -0700)]
Use SSE2 enhanced memset for capable x86 processors
Originally, if TARGET_ARCH_VARIANT is x86-atom, the SSE2 enhanced
memset is used. This patch extends this to all x86 processors which
support SSE2 (i.e. ARCH_X86_HAVE_SSE2 is true).
For: AIA-502
Change-Id: Ibb7ff6ff1ce48da4262e4284618c26f2b516c8bf
Depends-Change-Id: I61c66496df66f6d8adc45cddd48f8750fc95f83c
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Alin Dobre [Thu, 1 Mar 2012 13:11:48 +0000 (15:11 +0200)]
Fix build failure when switching between build variants
When switching from the eng to the userdebug build variant, the init
binary compilation returns error ar linking stage:
.../obj/EXECUTABLES/init_intermediates/init.o: In function `bootchart_init_action':
system/core/init/init.c:729: undefined reference to `bootchart_init'
.../obj/EXECUTABLES/init_intermediates/init.o: In function `main':
system/core/init/init.c:881: undefined reference to `bootchart_step'
system/core/init/init.c:882: undefined reference to `bootchart_finish'
collect2: ld returned 1 exit status
make: *** [.../obj/EXECUTABLES/init_intermediates/LINKED/init] Error 1
Issue: ANDROID-1757
Change-Id: Ide50286e190ffeb75ec56bdfa2f5c612e0ae2392
Signed-off-by: Alin Dobre <alin.m.dobre@intel.com>
Octavian Purdila [Mon, 20 Feb 2012 07:42:15 +0000 (09:42 +0200)]
bootchart: add threads support
Grab information about threads not only processes. This gives us
additional important information about the boot processes especially
for heavy processes such as system server and zygote.
Issue: ANDROID-609
Change-Id: Ia941da116de1589c46101f262db64b0bc730e643
Signed-off-by: Octavian Purdila <octavian.purdila@intel.com>
Pauli Nieminen [Tue, 15 Nov 2011 18:14:51 +0000 (20:14 +0200)]
Fix warnings about unused parameters
Change-Id: I29139ffa688fb9215956b4d585dd4c47f70612b8
Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Andrew Boie [Thu, 12 Jul 2012 22:39:25 +0000 (15:39 -0700)]
adb: reserve a wakelock when clients are connected over socket
USB OTG connections already have a wakelock reserved, but if a network
socket connection is made, the device can unexpectedly suspend
in the middle of a session. The lock is released when the client
disconnects.
Change-Id: Id5ac01699fab9350955c9814042259542fc6a694
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Jianxun Zhang [Mon, 2 Jul 2012 18:32:30 +0000 (11:32 -0700)]
add modprobe-like capability and automatic ueventd loading
Author: Jianxun Zhang <jianxun.zhang@intel.com>
Author: Daniel Leung <daniel.leung@intel.com>
Author: Andrew Boie <andrew.p.boie@intel.com>
- insmod_by_dep() added to libcutils; loads a module into kernel.
Modules the target module depends on will be loaded first. Loading
will be stopped when an error occurs.
- rmmod_by_dep() added to libcutils; removes a module from kernel.
It also tries to remove other modules the target module depends
on until an error occurs.
- Implement wildcard matching for ueventd rules.
The PCI and USB addresses for devices can change from devices
from devices for a particular class of peripheral, for example,
bluetooth. The ueventd rules created with these addresses are
then device-specific.
This changes the way ueventd rules with wildcard are handled.
Instead of matching just the prefix with a trailing wildcard,
now these rules can have wildcard anywhere in the rule.
The wildcard matching is implemented using fnmatch(), where
its matching is simliar to shell pathname expansion. It suits
this particular usage model well.
For example, instead of creating a rule to match:
/sys/devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.3/2-1.3:1.0/bluetooth/hci0/rfkill*
, this would suffice:
/sys/devices/*/bluetooth/hci0/rfkill*
- Let ueventd auto-load kernel modules. Implements the functionality
for ueventd to auto-load kernel modules when uevents are triggered.
Since /system may not be mounted when uevents are fired,
a deferred loading mechanism is implemented. Once mapping of
module and alias is available, these modules are then loaded.
Modules can also be blacklisted so they will not be loaded
automatically. One example would be the Wifi driver, as
Android's has to control its loading and unloading.
- add 'probemod' builtin command. This command accepts the name of a
kernel module plus a set of command line arguments. The module will
be loaded, along with all its dependencies, using the libcutils
insmod_by_dep() API.
- Drivers in kernel can request modules by launching a program in
user space, the program's path by default is "/sbin/modprobe".
Because Android system has no modprobe and ueventd is the only
program handling the module aliases so far, This patch provides a
cheap approach to handle kernel's requests in ueventd executable.
- Add new builtin init command "coldboot". The main purpose is to
provide an approach in init.*.rc files to fire uevents for devices
under the path which is passed as the the argument. This should be
called after /system is mounted so any queued events that need to
load a module can be fired.
Ajay Dudani [Wed, 12 Jun 2013 22:55:26 +0000 (15:55 -0700)]
Free file handle for /firmware/image after use
Change-Id: Id59e25486b45dffc87a89a961970a6e8128a5237
The Android Automerger [Thu, 23 May 2013 12:55:39 +0000 (05:55 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
Matthew Xie [Wed, 22 May 2013 17:36:53 +0000 (17:36 +0000)]
Merge "Add permission for bluetooth app in non-primary users" into jb-mr2-dev
Matthew Xie [Mon, 20 May 2013 21:22:01 +0000 (14:22 -0700)]
Add permission for bluetooth app in non-primary users
bug
8954114
Change-Id: Id6a42f5a5beefa22c3cc8eecfb39a2222017a3b4
The Android Automerger [Tue, 21 May 2013 14:47:06 +0000 (07:47 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
repo sync [Fri, 17 May 2013 19:48:34 +0000 (12:48 -0700)]
Revert "Add logic to fixup file contexts after a policy update."
This reverts commit
ebcf93e3bf92bf46c54c8e0f2ec8936bb2db6aea.
repo sync [Fri, 17 May 2013 19:48:18 +0000 (12:48 -0700)]
Revert "Add persistent property for SELinux enforcment status."
This reverts commit
b89bdfde422eff751b3f939f10ab57e22af83eaa.
repo sync [Fri, 17 May 2013 19:47:04 +0000 (12:47 -0700)]
Revert "Reload policy after setting up the data partition."
This reverts commit
fee250d27a9c03af1ba439047b976d89563b1887.
repo sync [Fri, 17 May 2013 19:46:31 +0000 (12:46 -0700)]
Revert "Remove /system from the dirs to relabel."
This reverts commit
3d4d07b2fa3b2447646d3c87a0b401e993ad8363.
repo sync [Fri, 17 May 2013 19:46:00 +0000 (12:46 -0700)]
Revert "Add a version check for SELinux policy on device."
This reverts commit
921be8b6568df0057c4eacbac2e1022b71e09620.
The Android Automerger [Wed, 15 May 2013 12:55:51 +0000 (05:55 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
repo sync [Fri, 10 May 2013 06:04:07 +0000 (23:04 -0700)]
Add a version check for SELinux policy on device.
This helps to ensure that when a new system image is installed,
old userdata policy isn't applied over the top of it.
Bug:
8841348
Change-Id: I135af32250aa62979763e775842ce0af3c8b6f9f
The Android Automerger [Wed, 8 May 2013 12:55:54 +0000 (05:55 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
repo sync [Wed, 8 May 2013 02:18:20 +0000 (02:18 +0000)]
Merge "Remove /system from the dirs to relabel." into jb-mr2-dev
repo sync [Wed, 8 May 2013 01:50:39 +0000 (18:50 -0700)]
Remove /system from the dirs to relabel.
Bug:
8841348
Change-Id: Ia1545288e8a8184feffaaa6158faad62f95cbcce
Jamie Gennis [Wed, 8 May 2013 01:12:18 +0000 (01:12 +0000)]
Merge "libcutils: add a trace tag for Dalvik" into jb-mr2-dev
Jamie Gennis [Tue, 7 May 2013 22:09:17 +0000 (15:09 -0700)]
libcutils: add a trace tag for Dalvik
This change adds a trace tag bit for Dalvik. It also allows the ATRACE_*
macros to be called from code compiled for the host with no effect.
Bug:
8856374
Change-Id: I9c17129d6cc170cae00b6500485b5b5fa4b5e647
The Android Automerger [Wed, 1 May 2013 12:56:12 +0000 (05:56 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
repo sync [Tue, 30 Apr 2013 05:21:06 +0000 (22:21 -0700)]
Reload policy after setting up the data partition.
This forces a policy reload + fixcon to deal with dynamically
delivered policy changing labels on device nodes.
It's implemented as a new keyword in init.
Bug:
8702843
Change-Id: I803cf1ecf6ff8318ce25dcc5cda4f292adc9738c
The Android Automerger [Mon, 29 Apr 2013 12:56:19 +0000 (05:56 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
repo sync [Fri, 26 Apr 2013 20:31:21 +0000 (20:31 +0000)]
Merge "Add persistent property for SELinux enforcment status." into jb-mr2-dev
repo sync [Wed, 24 Apr 2013 18:18:29 +0000 (11:18 -0700)]
Add persistent property for SELinux enforcment status.
Also adds triggers for going between enforcing and
permissive status.
Bug:
8702843
Change-Id: I841d80999b96acf6b6fd0a5408fb074952d6ae72
The Android Automerger [Thu, 25 Apr 2013 12:55:48 +0000 (05:55 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
Jeff Tinker [Thu, 25 Apr 2013 04:54:02 +0000 (04:54 +0000)]
Merge "Add a new group for MediaDrm engine plugins" into jb-mr2-dev
Jeff Tinker [Wed, 24 Apr 2013 02:54:17 +0000 (19:54 -0700)]
Add a new group for MediaDrm engine plugins
bug:
8702754
Change-Id: I3b7988b64b1dcf4685624e4c1af938e132b82696
Matthew Xie [Tue, 23 Apr 2013 19:32:49 +0000 (12:32 -0700)]
Remove DBUS from make and header files
DBUS had been needed by bluetooth bluz stack. It is not needed after
we replaced bluez stack with bluedroid stack.
bug
6872904
Change-Id: I3fa41c1dd4ac80bc679d5950b3b20c7f6d12265f
The Android Automerger [Mon, 22 Apr 2013 12:56:17 +0000 (05:56 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
Christopher Tate [Thu, 18 Apr 2013 21:44:37 +0000 (21:44 +0000)]
Merge "Fix regression: log mem maps around native fault address" into jb-mr2-dev
Christopher Tate [Thu, 18 Apr 2013 21:05:52 +0000 (14:05 -0700)]
Fix regression: log mem maps around native fault address
Restore the logging of the mem maps around the fault address along
with the rest of the faulting thread's information. (It was still
being written to the tombstone file, but the logging got dropped on
the floor in the refactoring around AM report integration).
Bug
8654694
Change-Id: Id8851fa765dfe6b6ce41ccfc39e85eaac0acc629
Alex Ray [Thu, 18 Apr 2013 20:39:32 +0000 (20:39 +0000)]
Merge "graphics: add Android flexible YCbCr format" into jb-mr2-dev
The Android Automerger [Wed, 17 Apr 2013 12:55:59 +0000 (05:55 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
Jamie Gennis [Tue, 16 Apr 2013 22:12:47 +0000 (22:12 +0000)]
Merge "cutils: add a way to disable tracing for a process" into jb-mr2-dev
The Android Automerger [Tue, 16 Apr 2013 12:55:51 +0000 (05:55 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
Jamie Gennis [Tue, 16 Apr 2013 01:50:22 +0000 (18:50 -0700)]
cutils: add a way to disable tracing for a process
This change adds the atrace_set_tracing_enabled call to libcutils. The call
can be used to disable all tracing (of the atrace variety) for the current
process. This is to be used to disable tracing in the Zygote process, as there
is no way for Zygote to be notified of changes to the enabled trace tags.
Change-Id: I0b691cc0dcfc65b16e3d17e1db2866a4deb253a7
Geremy Condra [Thu, 11 Apr 2013 00:51:53 +0000 (17:51 -0700)]
Add logic to fixup file contexts after a policy update.
Bug:
8116902
(cherry picked from commit
47677a506febfe386d186f8f854d967f165a342f)
Change-Id: I2a50f68a53aee2c94d03d3e5ced1ea36337fde33
Ken Sumrall [Mon, 15 Apr 2013 19:33:17 +0000 (12:33 -0700)]
Include liblog in fs_mgr to fix the build.
A recent change to how libcutils is built requires liblog
to be explicitly included in the link list if it's needed.
Change-Id: I8547f5e65c488c8f6e314ccd4eb96606742272be
Ken Sumrall [Mon, 15 Apr 2013 17:34:15 +0000 (17:34 +0000)]
Merge "fs_mgr: Capture the output of e2fsck and add to the kernel log" into jb-mr2-dev
Ken Sumrall [Mon, 15 Apr 2013 17:33:44 +0000 (17:33 +0000)]
Merge "logwrapper: Add ability to log to kernel log" into jb-mr2-dev
Ken Sumrall [Mon, 15 Apr 2013 17:33:27 +0000 (17:33 +0000)]
Merge "klog: Have klog_write() call klog_init() if needed" into jb-mr2-dev
The Android Automerger [Mon, 15 Apr 2013 12:56:35 +0000 (05:56 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
Ken Sumrall [Wed, 20 Mar 2013 02:38:44 +0000 (19:38 -0700)]
fs_mgr: Capture the output of e2fsck and add to the kernel log
Currently, the output of e2fsck is not saved, and we have no insight
into how many errors e2fsck is finding and fixing. Using the new
abbreviated logging feature in liblogwrap, up to the first 100 lines,
and last 4K bytes of the output of e2fsck is captured by fs_mgr, and
added to the kernel log.
Usually, the filesystem will be clean, and this will only add a few
lines to the kernel log on boot, but when things go wrong, it should
save enough to indicate what the problem is, without potentially
filling the kernel log with only e2fsck output if the filesystem is
really corrupted.
Change-Id: I9c264798e6fe721c8f818b5ce15d0975027ddbdd
Ken Sumrall [Wed, 3 Apr 2013 20:45:10 +0000 (13:45 -0700)]
logwrapper: Add ability to log to kernel log
Also add ability to do abbreviated logging where only the first
4K bytes and last 4K bytes of output are added to the desginated log.
Also update standalog logwrapper command to support the new options.
Change-Id: Ia49cbe58479b9f9ed077498d6852e20b21287bad
Dianne Hackborn [Fri, 12 Apr 2013 21:51:43 +0000 (14:51 -0700)]
Add new resources trace.
Change-Id: I63398f282fe7bf04f1c4cefa8f602100f403fa7c
Ken Sumrall [Wed, 3 Apr 2013 20:43:11 +0000 (13:43 -0700)]
klog: Have klog_write() call klog_init() if needed
Also change klog_init() to do nothing if already initialized.
Change-Id: Ia2dfe914c9d9fd119fb8939508d57b15c9884663
Ying Wang [Thu, 11 Apr 2013 00:27:35 +0000 (17:27 -0700)]
We need libcutils to include liblog
For some prebuilts were built against libcutils with liblog in it.
Unfortunately that means in the future if someone checks in prebuilts
that are built against libcutils and they don't explicitly link against
liblog, they will again prevent us from removing liblog from libcutils.
Bug:
8587800
Bug:
8580410
Change-Id: I2a5ed6233df7e7d542d1e4167bec405750c0e671
The Android Automerger [Wed, 10 Apr 2013 12:56:59 +0000 (05:56 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev
Ying Wang [Wed, 10 Apr 2013 05:03:45 +0000 (22:03 -0700)]
Add liblog
Bug:
8580410
Change-Id: Iab3a9b4307f207c14a04a922cc7350c54e60e9ad
Alex Ray [Tue, 19 Mar 2013 08:41:32 +0000 (01:41 -0700)]
graphics: add Android flexible YCbCr format
Change-Id: I3f6676873febc793f7a4a5e06892cd9fadb35665
The Android Automerger [Mon, 8 Apr 2013 12:57:16 +0000 (05:57 -0700)]
merge in jb-mr2-release history after reset to jb-mr2-dev