Commit Graph

61 Commits

Author SHA1 Message Date
Mathias Agopian
0a757814f3 fix [3259708] Graphic Buffer Mapper does not support YV12
remove a bunch of a code that was there only to support broken gralloc implementations

Change-Id: I3c1a9172224cbcc283601abfbbd695a20815451f
2010-12-08 16:48:28 -08:00
Jamie Gennis
54cc83e8a4 Implement reducing the buffer count of a Surface.
Change-Id: I7f979c60c06d654aa8265002836277434bc1a64f
Bug: 3095167
2010-11-11 14:06:38 -08:00
Mathias Agopian
19957553cb fix [2835280] Add support for cancelling buffers to ANativeWindow
There is a new ANativeWindow::cancelBuffer() API that can be used to
cancel any dequeued buffer, BEFORE it's been enqueued. The buffer is
returned to the list of availlable buffers. dequeue and cancel are not
mutually thread safe, they must be called from the same thread or
external synchronization must be used.

Change-Id: I86cc7985bace8b6a93ad2c75d2bef5c3c2cb4d61
2010-10-04 20:10:11 -07:00
Mathias Agopian
0e7f4292c7 fix a race in SF buffer management
also remove some unused code.

Change-Id: Iae2c3309b7a08055f3e13a5b866c5c084993e352
2010-08-26 17:42:27 -07:00
Mathias Agopian
b661d66013 fix [2931513] Add support for setting the orientation of an ANativeWindow
Also implement support for cropping.

Change-Id: Iba5888dd242bf2feaac9e9ce26e404c1f404c280
2010-08-24 15:40:50 -07:00
Mathias Agopian
208cb07724 fix [2873058] Surface::dequeueBuffer blocks on last buffer, i.e. cannot dequeue all allocated buffers at once.
this situation happened when the last buffer needed to be resized
(or allocated, the first time). the assumption was that the buffer
was in use by SF itself as the current buffer (obviously, this
assumption made no sense when the buffer had never been allocated, btw).

the system would wait until some other buffer became the "front" buffer.

we fix this problem by entirely removing the requirement that the
buffer being resized cannot be the front buffer. instead, we just
allocate a new buffer and replace the front buffer by the new one.

the downside is that this uses more memory (an extra buffer) for a
brief amount of time while the old buffer is being reallocated and
before it has actually been replaced.

Change-Id: I022e4621209474ceb1c671b23deb4188eaaa7285
2010-07-27 20:11:35 -07:00
Mathias Agopian
579b3f88d0 allow re-targetting of surfaces
Surfaces can now be parcelized and sent to remote
processes. When a surface crosses a process
boundary, it looses its connection with the
current process and gets attached to the new one.

Change-Id: I39c7b055bcd3ea1162ef2718d3d4b866bf7c81c0
2010-06-08 20:10:02 -07:00
Mathias Agopian
b7e930db17 split surface management from surface's buffers management
Change-Id: If3c5655d1231f8f0c49ba68f972b1b20c93b3f87
2010-06-04 18:57:41 -07:00
Mathias Agopian
7e27f05739 more clean-up in preparation of bigger changes
the most important change here is the renaming of
ISurfaceFlingerClient to ISurfaceComposerClient

Change-Id: I94e18b0417f50e06f21377446639c61f65f959b3
2010-05-28 17:19:16 -07:00
Mathias Agopian
631f358d34 fix [2712278] The preview buffer left some black borders in left and bottom edges
we were incorrectly flagging push_buffer surfaces as invalid

Change-Id: I4dfd4ffbbe8a71f7e23e835db8d71966416c29bb
2010-05-25 17:51:34 -07:00
Mathias Agopian
a138f89c5e added the notion of fixed-size buffers
the new native_window_set_buffers_geometry allows
to specify a size and format for all buffers to be
dequeued. the buffer will be scalled to the window's
size.

Change-Id: I2c378b85c88d29cdd827a5f319d5c704d79ba381
2010-05-24 18:26:01 -07:00
Mathias Agopian
a0b3f1d2eb fix a bug when reallocating a window's buffers is needed
we need to mark the buffers that need to be reallocated,
NOT the buffer's indices.

Change-Id: I809e2e1b03b56c4d2ab983c25523dae99aa1da74
2010-05-24 13:47:24 -07:00
Mathias Agopian
f10d7fde03 added native_window_set_buffer_count()
this method can be used to change the number of buffers
associated to a native window. the default is two.

Change-Id: I608b959e6b29d77f95edb23c31dc9b099a758f2f
2010-05-21 14:34:32 -07:00
Mathias Agopian
bb641244d7 fix the threading issue for setBuffercount()
this change introduces R/W locks in the right places.
on the server-side, it guarantees that setBufferCount()
is synchronized with "retire" and "resize".
on the client-side, it guarantees that setBufferCount()
is synchronized with "dequeue", "lockbuffer" and "queue"
2010-05-20 18:00:42 -07:00
Mathias Agopian
d6297f7d09 fix some bugs in SharedBufferStack::resize
added buffers should now be labeled properly.

Change-Id: I28aa753fbe89ab89134e7753575319478934c7fa
2010-05-17 17:27:26 -07:00
Mathias Agopian
b5b7f260da SharedBufferStack now can grow up to 16 buffers.
there is a new resize() api, which currently only allows growing.

Change-Id: Ia37b81b73be466d2491ffed7f3a23cd8e113c6fe
2010-05-12 17:28:20 -07:00
Mathias Agopian
c0a9164e9e Add support for enqueuing buffers in arbitrary order
Also added a very simple SharedBufferStack unit test.

Change-Id: I253dbbe98a53c966b78d22d4d6dd59f8aefc8c40
2010-04-28 16:12:54 -07:00
Mathias Agopian
b296533607 cleanup. waitForCondition() now uses polymorphsim instead of templtes
the reason for the above change is that waitForCondition() had become
large over time, mainly to handle error cases, using inlines to
evaluate the condition doesn't buys us much anymore while it increases
code size.

Change-Id: I2595d850832628954b900ab8bb1796c863447bc7
2010-04-28 16:12:54 -07:00
Mathias Agopian
0a8cd0689a fix a race condition in undoDequeue(), where 'tail' could be computed incorrectly.
in the undoDequeue() case, 'tail' was recalculated from 'available' and 'head'
however there was a race between this and retireAndLock(), which could cause
'tail' to be recalculated wrongly.

the interesting thing though is that retireAndLock() shouldn't have any impact
on the value of 'tail', which is client-side only attribute.
we fix the race by saving the value of 'tail' before dequeue() and restore it
in the case of undoDequeue(), since we know it doesn't depend on retireAndLock().

Change-Id: I4bcc4d16b6bc4dd93717ee739c603040b18295a0
2010-04-28 16:12:54 -07:00
Mathias Agopian
cc08e688f5 added setCrop() to android_native_window_t
hooked up the new method up to Surface.cpp
the actual crop is not implemented in SF yet

Change-Id: Ic6e313c98fd880f127a051a0ccc71808bd689751
2010-04-20 13:36:29 -07:00
Mathias Agopian
1100c8b901 add support for up to 16 buffers per surface
also increase the dirtyregion size from 1 to 6 rectangles.
Overall we now need 27KiB process instead of 4KiB

Change-Id: Iebda5565015158f49d9ca8dbcf55e6ad04855be3
2010-04-20 13:36:29 -07:00
Mathias Agopian
9cce325fae split libsurfaceflinger_client and libcamera_client out of libui 2010-02-11 13:16:22 -08:00
Mathias Agopian
04a3f57d19 fix [2133133] Software OpenGL ES Lighting is buggy (GL Gears washed out bug)
A typo caused GL_AMBIENT_AND_DIFFUSE to only set the the ambient color.

Fix another typo which caused the viewer position to be wrong for
specular highlights.

Switch back to eye-space lighting, since there are still some issues
with some demos (San Angeles in particular).
2010-02-03 12:13:15 -08:00
Mathias Agopian
18b915a102 implement [2396050] Add ETC1 texture support to AGL 2010-02-02 20:19:54 -08:00
Mathias Agopian
6fee064809 fix[2228133] pixelflinger ignores the "vertical stride" leading to artifacts when playing back video
we lost the concept of vertical stride when moving video playback to EGLImage.
Here we bring it back in a somewhat hacky-way that will work only for the
softgl/mdp backend.
2009-11-02 17:48:33 -08:00
Mathias Agopian
38a7fa2ae3 fix [2182249] [MR1] valgrind error in surface flinger 2009-10-15 18:08:15 -07:00
Mathias Agopian
e700501d0e fix [2170319] gmail bulk operation checkbox latency on passion
This also fixes [2152536] ANR in browser

When SF is enqueuing buffers faster than SF dequeues them.
The update flag in SF is not counted and under some situations SF will only
dequeue the first buffer. The state at this point is not technically
corrupted, it's valid, but just delayed by one buffer.

In the case of the Browser ANR, because the last enqueued buffer was delayed
the resizing of the current buffer couldn't happen.

The system would always fall back onto its feet if anything -else- in
tried to draw, because the "late" buffer would be picked up then.
2009-10-07 16:44:10 -07:00
Mathias Agopian
9ec430adae fix [2152536] ANR in browser
A window is created and the browser is about to render into it the
very first time, at that point it does an IPC to SF to request a new
buffer. Meanwhile, the window manager removes that window from the
list and the shared memory block it uses is marked as invalid.
However, at that point, another window is created and is given the
same index (that just go freed), but a different identity and resets
the "invalid" bit in the shared block. When we go back to the buffer
allocation code, we're stuck because the surface we're allocating for
is gone and we don't detect it's invalid because the invalid bit has
been reset.

It is not sufficient to check for the invalid bit, I should
also check that identities match.
2009-10-06 19:00:57 -07:00
Mathias Agopian
b26af23744 fix [2168531] have software-only gralloc buffer side-step the HAL 2009-10-06 17:00:25 -07:00
Mathias Agopian
3330b20303 fix [2167050] glTexImage2D code path buggy in SurfaceFlinger
When EGLImage extension is not available, SurfaceFlinger will fallback to using
glTexImage2D and glTexSubImage2D instead, which requires 50% more memory and an
extra copy. However this code path has never been exercised and had some bugs
which this patch fix.

Mainly the scale factor wasn't computed right when falling back on glDrawElements.
We also fallback to this mode of operation if a buffer doesn't have the adequate
usage bits for EGLImage usage.

This changes only code that is currently not executed. Some refactoring was needed to
keep the change clean. This doesn't change anything functionaly.
2009-10-06 17:00:25 -07:00
Mathias Agopian
0b3ad46a26 Attempt to fix [2152536] ANR in browser
The ANR is caused by SurfaceFlinger waiting for buffers of a removed surface to become availlable.
When it is removed from the current list, a Surface is marked as NO_INIT, which causes SF to return
immediately in the above case. For some reason, the surface here wasn't marked as NO_INIT.

This change makes the code more robust by always (irregadless or errors) setting the NO_INIT status
in all code paths where a surface is removed from the list.

Additionaly added more information in the logs, should this happen again.
2009-10-02 18:12:30 -07:00
Mathias Agopian
86f7329080 add basic time stats for surfaces lock time 2009-09-17 01:35:28 -07:00
Mathias Agopian
c7d5601081 make sure to update the tail pointer when undoing a dequeue 2009-09-14 15:48:42 -07:00
Mathias Agopian
48d819a131 fix [2112575] stuck on DequeueCondition for a surface that doesn't exist anymore
this also fixes part of [2111536] Device is soft rebooted after ending the call through voice dialer
2009-09-10 19:41:18 -07:00
Mathias Agopian
b58b5d72d6 make sure conditions will return when the status of a surface is not NO_ERROR 2009-09-10 16:55:13 -07:00
Mathias Agopian
cbb288bfe8 fix [2068105] implement queueBuffer/lockBuffer/dequeueBuffer properly
Rewrote SurfaceFlinger's buffer management from the ground-up.
The design now support an arbitrary number of buffers per surface, however the current implementation is limited to four. Currently only 2 buffers are used in practice.

The main new feature is to be able to dequeue all buffers at once (very important when there are only two). 

A client can dequeue all buffers until there are none available, it can lock all buffers except the last one that is used for composition. The client will block then, until a new buffer is enqueued.

The current implementation requires that buffers are locked in the same order they are dequeued and enqueued in the same order they are locked. Only one buffer can be locked at a time.

eg. Allowed sequence:   DQ, DQ, LOCK, Q, LOCK, Q
eg. Forbidden sequence: DQ, DQ, LOCK, LOCK, Q, Q
2009-09-07 16:32:45 -07:00
Martin Storsjo
e03eaa1551 Calculate specular lighting correctly
Since the lighting calculations are done in object space, the vector
from the object to the viewer also needs to be transformed to object
space.
2009-08-25 18:35:54 -07:00
Mathias Agopian
238a66e67b be more friendly with C 2009-08-13 18:27:10 -07:00
Mathias Agopian
eb9fd685f8 add a ctor to Mutex to specify the type, which can be shared. This is used by sf and af an soon will allow some optimization in the kernel for non shared mutexes 2009-07-13 22:06:36 -07:00
Mathias Agopian
ec0f1f6720 implement Mutex and Condition with pthread instead of calling futex directly.
internally pthread uses futex. the implementation consists of simple inlines
there are no implementation files anymore.
2009-07-13 15:14:23 -07:00
Mathias Agopian
e7c4c28d0f add a virtual destructor to region_rasterizer, to be on the safe side 2009-07-07 12:29:17 -07:00
Mathias Agopian
4c4b387afb merge master in master_gl 2009-06-24 18:31:21 -07:00
Android (Google) Code Review
952663ef57 am 5277103d: Merge change 5035 into donut
Merge commit '5277103db9033226814edc8fcba6f4ba4e846678'

* commit '5277103db9033226814edc8fcba6f4ba4e846678':
  add glTexParameteri() and fix glTexParameteriv()
2009-06-22 22:04:12 -07:00
Mathias Agopian
b12f99b315 add glTexParameteri() and fix glTexParameteriv() 2009-06-22 18:09:09 -07:00
Mathias Agopian
0a3139a21e fix a bug where copybit only renders in the first buffer when used with s/w GL 2009-06-10 16:15:32 -07:00
Mathias Agopian
7e2a937c4f revive the copybit test. 2009-06-09 21:38:08 -07:00
Mathias Agopian
0c97ed3a2e cleanup Debug.h a bit 2009-06-04 23:29:29 -07:00
Mathias Agopian
310f8da0c3 merge master to master_gl 2009-05-22 02:16:08 -07:00
Mathias Agopian
c5b2c0bf80 move libbinder's header files under includes/binder 2009-05-20 12:55:03 -07:00
Mathias Agopian
208059f67e checkpoint: split libutils into libutils + libbinder 2009-05-20 12:55:02 -07:00