hardware_libhardware.git - hardware/libhardware

	Commit message (Collapse)	Author	Age	Files	Lines
...
*	fix [1965730] Device always give force close screen and completely unusable ↵	Mathias Agopian	2009-07-07	1	-0/+6
\| \| \| \|	in Éclair build ERC07,rebooted automatically
*	better fix for gralloc on sim	Mathias Agopian	2009-07-07	1	-10/+14
\|
*	Merge change 6375	Android (Google) Code Review	2009-07-07	1	-1/+5
\|\ \| \| \| \| \| \| \| \|	* changes: Fix build on some linux distros.
\| *	Fix build on some linux distros.	Andy McFadden	2009-07-07	1	-1/+5
\| \| \| \| \| \| \| \| \| \| \| \| \| \|	The place where the PAGE_SIZE value is defined varies, mostly because you're not supposed to be using it directly. sysconf(_SC_PAGE_SIZE) is the approved method, and in fact some Linux distros actually #define PAGE_SIZE to the library call.
* \|	fix [1965734] Screen flickers a lot [Sholes]	Mathias Agopian	2009-07-07	1	-10/+9
\|/
*	Simulator build fixes for the new GL stuff. The simulator still doesn't run, ↵	Marco Nelissen	2009-07-07	4	-3/+25
\| \| \| \|	but at least it builds.
*	workaroud [1964234] emulator renders funny with new gralloc.	Mathias Agopian	2009-07-06	1	-0/+10
\| \| \| \|	we detect we're running on the emu (using a very bad heuristic, see 1964251) and switch UPDATE_ON_DEMAND off.
*	fix [1964228] emulator crash with the new GL	Mathias Agopian	2009-07-06	1	-2/+2
\| \| \| \|	gralloc didn't keep the error state properly, so after the first pmem allocation failure, it would systematically fail without reverting to ashmem.
*	make sure to UNSLAP pmem buffers when they're destroyed/freed, as the client ↵	Mathias Agopian	2009-07-02	1	-21/+48
\| \| \| \|	could still have them mmapped
*	add support for YUV formats	Mathias Agopian	2009-06-25	1	-18/+38
\|
*	copybit now uses a native_handle_t* instead of a fd/offset	Mathias Agopian	2009-06-24	1	-0/+2
\|
*	fallback to ashmem when pmem is not availlable	Mathias Agopian	2009-06-19	1	-2/+17
\|
*	suppress debug log	Mathias Agopian	2009-06-16	1	-1/+1
\|
*	pmem allocator in gralloc. enabled for all surfaces. currently it uses a lot ↵	Mathias Agopian	2009-06-16	6	-41/+448
\| \| \| \|	more address space than needed.
*	add an offset field to gralloc handles so the framebuffer can work with copybit	Mathias Agopian	2009-06-10	2	-3/+5
\|
*	with the new lock/unlock API we don't really mean reference counting on ↵	Mathias Agopian	2009-06-09	3	-306/+78
\| \| \| \| \| \|	mmap/munmap because we're guaranteed to map the buffers only once within a process. no need to track all handles anymore, which simplifies the code a lot.
*	suppress spurious debug log	Mathias Agopian	2009-06-03	1	-3/+3
\|
*	better documentation for (*setUpdateRect)()	Mathias Agopian	2009-05-07	1	-0/+5
\|
*	better documentation and implementation for lock/unlock	Mathias Agopian	2009-05-05	2	-29/+77
\|
*	get rid of unused/needed flags	Mathias Agopian	2009-05-05	1	-1/+1
\|
*	lock will now return the vaddr of the buffer. map/umap are gone.	Mathias Agopian	2009-05-04	4	-64/+151
\| \| \| \| \| \|	- make sure to return an error if a buffer is locked twice or unlocked while not locked. - added registerBuffer() and unregisterBuffer() to the gralloc module so that we can do some cleanup when a buffer is no longer needed. this became necessary after we removed map/unmap so we have a place to unmap buffers without the need of a kernel module. - change the constants for GRALLOC_USAGE_SW_{READ\|WRITE}_NEVER to 0, so that NOT specifying them means "NEVER".
*	don't automatically unmap buffers when they're freed	Mathias Agopian	2009-04-14	1	-2/+2
\| \| \| \|	this is in preparation to an api change wrt gralloc map/unmap
*	Integrate from //sandbox/mathias/donut/...@145728	Mathias Agopian	2009-04-10	5	-0/+1143
	SurfaceFlinger rework for new EGL driver model support.