The gralloc API now provides a way for using lock/unlock with the Android
explicit synchronisation concept. This changes updates the GraphicBuffer class
to also expose this functionality, and updates the Surface class to make use of
in line with the dequeueBuffer/queueBuffer mechanism.
This new behaviour is dependent on GRALLOC_MODULE_API_VERSION_0_3. If the local
gralloc module does not support this then the existing synchronous lock/unlock
mechanism will be used.
Change-Id: I77daa1beb197b63b1c2f281b8414ac4ae4b5b03c
The instruction-set is used to construct the dalvik cache
path and is also passed down to dex2oat.
(cherry picked from commit 791781bfb8)
Change-Id: I43396b16f6eaecacf0fb7d052526fc5a239167ac
It can help to detect some kind of error, such as why GraphicBuffer::flatten
will crash when handle is null.
Change-Id: I703cd035b96edb7afb324cf24d8230d4e55f4f52
Signed-off-by: Jun Jiang <jun.a.jiang@intel.com>
It's wrong to just concatenate the apk_path and .odex.
The bug prevents the prebuilt odex being used since Kitkat.
The patch is copied from the code of JellyBean.
Change-Id: I0ce8a877e3df8ae1ab9a0e3aeeef2d5253efc223
The Fence object was writing a size_t into the binder buffer
in flatten, which changes size if the producer and consumer
are running in a 32-bit and a 64-bit process. Use a uint32_t
instead.
Change-Id: Ifed526513800ce27f9d605101cddd922292cca37
This is required so that it will be assigned the correct SELinux
security context on first creation by installd.
Bug: 13927667
Change-Id: I4857d031f9e7e60d48b8c72fcb22a81b3a2ebaaa
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
These weren't really being used and they make it
very hard to reason about who looks at command line
arguments.
Processes started via app_process (this includes all
zygote forks and the system_server) can get information
about command line arguments from the AndroidRuntime
class, which is available via a call to
AndroidRuntime::getRuntime.
bug: 13647418
Change-Id: I6f92680c3619a68c6d4b0995db4cdc9adc788e36
The types of OMX_U32 and OMX_S32 were appropriate to a 32-bit system
but were not appropriate to a 64-bit system. They have been changed to
make them portable between the two systems. With this change and a
change to frameworks/av/media/libstagefright/foundation/AString.cpp,
the android.media.cts.DecoderTest#testFlush CTS test passes.
Change-Id: I8e16c7169f54599c23fea71585b531e48541c78a
Signed-off-by: Marcus Oakland <marcus.oakland@arm.com>
Changes above with PMS and below with libselinux
have resulted in a few changes to the restorecon data
api. This change is needed in order to support the new
way to issue a recursive restorecon of certain
/data/data directories.
The restorecondata function has also been modified to
find all users on the device for a given package name
(argument to the function) and to issue a separate
recursive restorecon call for each.
Change-Id: Ie440cba2c96f0907458086348197e1506d31c1b6
Signed-off-by: rpcraig <rpcraig@tycho.ncsc.mil>
GraphicProducerWrapper(GPW) changed how the methods of
BpGraphicBufferProducer(BpGBP) are executed.
First, "fake" BpGBP is created. Its remote is GPW. The GPW has
wrapped the real BpGBP.
All the method calls to the fake BpGPB will be intercepted by
the GPW inside it when the methods run into remote()->transact().
Then the GPW will invoke the transact() of the real BpGBP. And
Everything runs well except that the GPW forgets to store the
transact status and always return NO_ERROR to the fake BpGBP.
It would be disastrous if the binder call of the IGBP failed and
the out parameter "reply" of transact() was in unkown state.
E.g. the queueBuffer() in the fake BpGBP will try to operate on
the "reply". This will crash the SurfaceFlinger.
Change-Id: I01b31f64e1fc92804da3f16c1fb1420dcfb3b855
Signed-off-by: bdeng3X <bingx.deng@intel.com>
Signed-off-by: Guobin Zhang <guobin.zhang@intel.com>
Otherwise we can't readily see log messages from libselinux calls
made by installd.
Change-Id: I319b30c181470468fe19dd5fbe9251ef03f1163b
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
To make struct Event have consistent memory layout between x86
and x86_64, force nsecs_t timestamp align to 64bit because it's
not true for x86.
Change-Id: I49eaddd89e10d595cf6a5130f9762b5388c96625
Signed-off-by: Fengwei Yin <fengwei.yin@intel.com>
Implement new restorecondata function which will allow
the relabeling of /data/data and /data/user directories.
This is needed in the case of certain OTAs. Not every
boot will apply this relabeling however. Consult change
I77bf2a0c4c34b1feef6fdf4d6c3bd92dbf32f4a1 for
clarification on this issue.
Change-Id: I05e8b438950ddb908e46c9168ea6ee601e6d674f
Signed-off-by: rpcraig <rpcraig@tycho.ncsc.mil>
