1. Detect and account the number of secure RGB layers present.
2. Mark secure RGB layers for MDP comp in cache based stratergy.
3. If full MDP comp and partial MDP comp fails, Mark secure RGB
layers for MDP composition.
4. In partial MDP comp reject load based composition, if secure
layer is present to avoid GPU composition.
5. Disable idlefallback feature if secure RGB layer is present
to avoid GPU composition.
Change-Id: Idb4c4f09b081ec20240a2602aa3f20332fa49ad6
* Limit the number of rotator sessions to 4 both in
LayerRotMap and RotMgr structures
* Make use of MDPVersion getter fuctions to check
for maxMixerWidth and maxDownScale supported by MDP
Change-Id: I27dac877807fbb4ed6fcb3500a3a318a2ef5d8b2
Provides a way to enable/disable logs at runtime via binder.
The logs need to be enabled from a priviliged shell.
For example,
$ adb shell service call display.qservice 15 i32 0 i32 1
Here 15 is the integer code for DYNAMIC_DEBUG in IQService.h
The first parameter 0 is for DEBUG_ALL
The second parameter 1 is to enable.
Hence, this command enables all debug logs.
Another example - this command enables vsync logging
$ adb shell service call display.qservice 15 i32 2 i32 1
Change-Id: I13d5f397d9d03f0ac6a14d3a3c9098ed59fc62fd
- Consider two smallest layers for PTOR feature which uses MDP comp.
- These two layers are rendered on a buffer and are queued to MDP
to acheive full MDP Composition
- This helps to acheive performance enhancement
Change-Id: I43d9306ff19cd2d7a410c885316523965a44cbd4
1. Use MDP rotator to rotate RGB layers which are rendered by software
2. Restrict total number of rotator sessions to 4 for all displays
3. Disable RGB layer rotation for all MDP versions < MDP5
Change-Id: Ie93111082dff9a16b614e9768df5d8ce83ff9e25
MDP bandwidth limitations can be avoided, if the overlapping region
covered by the smallest layer at a higher z-order, can be composed
on a render buffer using Copybit, and then it can be queued to MDP
along with other layers. It helps to improve performace by avoiding
GPU Composition in use cases like Home Screen pan. To enable this
feature, set system prop "persist.hwc.ptor.enable" to "true" or '1'
and enable Copybit HAL compilation.
Change-Id: Ib719d3dd42477837519518cc427f1b95356c28f7
Provide support to simulate composition strategy tree to execute
and test multiple strategy combinations across test cases.
Change-Id: I5011fad32f8f45ecef47b6df513f9d72343697d2
Extends partial frame update support for split displays by calculating
changing region(ROI) independently for each half of the screen. In case of
only one half updating, display driver shuts down the DSI/MDP mixer control
of the non updating half.
Maintains two ROI's for each display. In case of Non-split
displays, only left ROI is updated.
Change-Id: I8a67fe40aac665a48b6f8a46beffb9f8027851b2
There is no need to re-program the H/W, if the frame geometry of all
active displays, has not changed. It will help to improve performance
by avoiding re-configuration of HAL and driver, if it is not needed.
Change-Id: Ie01dce33be89afc7308a8103128d27b4b9fb5c4b
Add support for returning pipes based on capabilities.
getPipe() API takes in pipe specifications like format class,
scaling, display, mixer, fb and decides what pipe to return.
The nextPipe() API is now just a private helper.
Change-Id: I0ab809662c0ad6297e19a44f39264ee929944d13
If using source split to stage 2 pipes on the same mixer stage,
the left pipe needs to be a higher priority than the right.
Add API in overlay to compare pipe priorities and use this in
source split config to stage pipes accordingly.
Add support for 1 pipe config for FB if updating rect is within 2048
pixels.
Change-Id: I8b33d5ebd0f8765ee842bab128d8abd67a110145
1. Set the GPU hint flag to high for MIXED/GPU composition only for
first frame after MDP to GPU/MIXED mode transition.
2. Set the GPU hint to default if the current composition type is GPU
due to idle fallback or MDP composition.
Change-Id: I208a778017435a5a4620142da9d9fb3c50e09155
When layer format is macro-tile, decimation cannot be supported.
Add necessary checks to bail out for such use-case.
Change-Id: I711f07a2963a76280a08887d04770cd418d8bd1e
Add support for SourceSplit. The mixers are abstracted from hwc.
There would be no translations to mixer understood destinations.
1) If a layer's crop and position are < 2048, only 1 pipe can be used
irrespective of position.
2) Else 2 pipes can be used with the layer equally split, without
regard to position.
Change-Id: I501a53838d147887c1e2299366663e05320d5096
While performing load based mixed composition, iterate from max
stages of mixer to at least 2.
Get rid of modes that preferred GPU or MDP. This new mode will
include both plus the missed out batches.
Change-Id: Ideb06a2332fe2f59968367d28df848be01f4aedc
Bandwidth checks will be done in the driver as part of resource check
which makes the existing checks in hwc redundant.
Change-Id: I39eb7de63618e8ea5981f9e1a1159224831f314a
While doing mixed mode composition, update only the region of the
framebuffer where layer contents are cached or updated. Avoids
MDP fetch for non-visible contents of FB.
Change-Id: I20997d79e02a6bac60d7333c28c6dde134263197
1. Use kernel timer interrupt to signal about the timeout to
Invalidator thread and fallback to GPU composition.
2. This implementation avoids waking up the Invalidator thread for
every idletimeout value during continuous update.
Change-Id: I4370f10a0ca06b5cb1e7bdcb047e9c8daec51afb
Pipe estimation was necessary when the unavailability from overlay
led to GPU fallback.
With the new design, hwc uses the next best strategy, so failure
to grab a pipe from overlay is ok.
Change-Id: Id20d2e6fd24aeb5af3ab84ec9dd8c52f6bbbef8b
1. Define classes for VDS and V4L2 based WFD solutions.
2. Overload the functionalities for VDS and V4L2 WFD solutions.
3. Instantiate VDS object or V4L2 object during hwc init based on the
property persist.hwc.enable_vds set by the user.
4. VDS solution is not verified as it requires some dependent changes.
Change-Id: I74bf944d532da138c850d4ec0c20b4a881489efa
Set default idle timeout to 70ms (~4 cycles on 60fps panel + some extra)
from the earlier 2sec.
The GPU timeout is around 80ms after which it turns off its rail,
clock and bus vote. So having a draw cycle before that is beneficial
Change-Id: I03291a4b8fd0707839567987febe1c088e21c92d
Currently, if programming MDP fails from a certain composition
strategy we fall back to GPU. This change, redesigns so that
composition types can have a chain of fall back strategies if
programming MDP fails. This happens currently only if resource
checks fail.
Change-Id: Ia7fb71788e595e59a91e95b1fb4eac3faea9451e
Combine the redundant code for FB programming and source split
adjustment into appropriate functions.
Change-Id: Ib99a2297a672e5ef40acb9f246021e57e41cd759
Add hw limitation checks for MDP composition once
we identify actual layers for MDP. This should be generic
way for all basic hw limitations of underlying MDP/MDSS.
Current limitations:
- a-family : MDP can not handle alpha scaling.
- 8226 & 8974 : Can not handle if multiple layers need
downscaling and blending
Change-Id: Ia944802d798f2d0e9cd4515d7e9854c7314cac78
CRs-fixed: 538803
The programYUV used for video-only cases is the same as programMDP
function. Remove the redundant function
Change-Id: I1d891404a77b2a47f90c236d4d6dcbbd02143d93
Use the low and high bandwidth values exported from driver per target
Use the low bandwidth value when the system has at least 1 video
mode panel, else use the high bandwidth value.
Newer targets could have bandwidth exceeding what an uint32 could
hold, so change the bytes read to be in gigs.
Do not use the framebuffer handle to get size, since the handle may
be null.
Change-Id: If3124ddc33334eba80d34ab7d8fc64d1b46697ce
To minimize usage of decimation when 4K x 2K video is played
on primary, split the yuv layer into 2 halves and configure
each layer to individual VG pipes. With this approach,
decimation can be avoided in some 4K x 2K video use cases,
thereby improving performance.
Change-Id: I7cddfab787c50f6ff132eaa7f5a3d9cfd636c282
GPU's perf is proportional to the number of pixels it processes, so
the load based strategy that prefers GPU, attempts to load it less
in terms of pixels, and at the same time makes sure the MDP has
enough pipes and mixer stages.
MDP's perf is real time and proportional to the instantaneous b/w
which is a function of overlaps. So the load based strategy that
prefers MDP would look at sending only as many full-screen layers
that MDP could support.
This mode is used when the GPU-preferred mode fails.
Change-Id: I3221a94bb01c43112370c248cd4ab217f2bd8ed1
Bytes claimed doesn't reflect the panel refresh rate.
So use the bandwidth claimed instead which factors this in.
This is important when 2 displays are running at different
refresh rates.
Change-Id: I3ab9573e43b8c25c31e01888fb2b181469b9e91b
Move resource checks, pipes, stages of mixer, bandwidth to a common
function and use it from all strategies.
Change-Id: I161fc957e2793d543b5943fbd23ae3607a4cfc87
When MDP cannot handle an updating layer and we marked it for GPU,
make sure framebuffer is refreshed for that frame. In the current design,
only video layers fall into these catches. But we may need
this change for future composition strategies.
Change-Id: I1d6a92da13feef4b353fa5e196d42f91148412e5
CRs-Fixed: 575340
Redraw the cache if current frame cache/drop data doesn't
match with cached frame data. Frame buffer z-order, fbcount
and mdp count are not reliable. With same fbz, fbcount and
mdpcount it is possible that layers in previous cached
frame are not part of current frame fb cache.
Change-Id: I858228dc1a3a8ab8673ab8d294e54fdbf43cc1ad
Add support for load based partial mdp comp.
This is used on geometry changes where a redraw is unavoidable.
We select a batch of layers, that has minimum pixels for FB comp,
the rest go to MDP.
Conflicts:
libhwcomposer/hwc_utils.cpp
libhwcomposer/hwc_utils.h
Change-Id: Ifc5eeb4785c75c37de97a2bb89ca81409d324691
While marking layers for cache, neglect the z-order when
possible, to improve the scope of mixed mode usage.
It is safe to neglect the z-order if a static layer
doesn't have any overlapping region with updating layers
in between static layer & contiguous static layers batch.
- if above layer doesn't have any overlapping with updating
layer in middle, push it to the batch.
- If above case fails, but layers below(existing batch) does not
have any overlapping region with the updating layer, bring the
batch to top and modify fb-zorder accordingly.
Change-Id: I4882a750d0a9a2e11272fcbb146202160673d04f
This change adds support in HWC for MDP partial update applicable
only for command mode panels. Presence of MDP FB (GRAM) in
command mode panels allows MDP clients to update only the updating region
of the frame as rest of the frame will be cached in the GRAM.
HWC calculates the updating region of a frame (ROI) by deducing
outer bounds of its updating layer destinations. Layers not lying
within the calculated ROI will be dropped from the composition
since its illegal to program MDP pipes for non participating layers.
Change-Id: I890d98ff7960fe888787981803cac62f68471201
When MDP cannot support a layer, we send the whole list of layers to
GPU. If such layers are not updating they can be cached on FB while
allowing the rest to take the MDP route.
Example # 1: Volume bar on video in 90 rotated mode. Since volume bar
is not h/w accelerated, there is no prerotation and we invoke the GPU
for redraws each round. With this change we won't because of caching.
Example # 2: If prerotation is applied to layers on primary, the
secondary layers would show a transform and the whole list is redrawn
with GPU. Now we could make use of caching.
This change also imposes a 1 layer limit on secondary, but now allows
full and partial MDP comp, thus benefiting secondary-only content.
Change-Id: I7eb43ba817f1081f317fefa99fa0c507344060e8
In video handling this patch makes sure:
1) Sufficient VG pipes are available, not just any pipes
2) If VG pipes are insufficient in multi-video scenarios
preference is given to secure videos
Change-Id: I170592463a1c28348108a1b12d60908cf3063d7d
