3998 Commits

Author SHA1 Message Date
Glenn Kasten
6b6e640323 am a8719ad9: Merge "Use the standard CC_LIKELY and CC_UNLIKELY macros"
* commit 'a8719ad9d53d3fe51e8031b2471e9558b8ef727f':
  Use the standard CC_LIKELY and CC_UNLIKELY macros
2012-01-05 07:40:36 -08:00
Glenn Kasten
a8719ad9d5 Merge "Use the standard CC_LIKELY and CC_UNLIKELY macros" 2012-01-05 07:38:29 -08:00
Glenn Kasten
e80a4ccd2b Use the standard CC_LIKELY and CC_UNLIKELY macros
Several source files privately defined macros LIKELY and UNLIKELY in terms
of __builtin_expect. But <cutils/compiler.h> already has CC_LIKELY and
CC_UNLIKELY which are intended for this purpose.  So rename the private
uses to use the standard names.

In addition, AudioFlinger was relying on the macro expanding to extra ( ).

Change-Id: I2494e087a0c0cac0ac998335f5e9c8ad02955873
2012-01-05 07:33:45 -08:00
Alex Sakhartchouk
6b908df7d9 Merge "Adding better error behavior to RS" into graphics-dev 2012-01-04 16:59:41 -08:00
Alex Sakhartchouk
afdf7be7ce Merge "Adding RS comments, cleaning up the gl implementation with new Element hal." into graphics-dev 2012-01-04 16:59:16 -08:00
Steve Block
c2daafc18d am 41891aea: Merge "Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE"
* commit '41891aeaf4e6e4006ae3087fa537786e9ec8caa4':
  Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF)  DO NOT MERGE
2012-01-04 10:16:14 -08:00
Romain Guy
0c16df2b20 am 173ab4d6: Merge "Properly restore the GL scissor after a GL draw functor Bug #5781254"
* commit '173ab4d61077c49f115b82eff34f97fda5a7273a':
  Properly restore the GL scissor after a GL draw functor Bug #5781254
2012-01-03 16:34:23 -08:00
Steve Block
5baa3a62a9 Rename (IF_)LOGD(_IF) to (IF_)ALOGD(_IF) DO NOT MERGE
See https://android-git.corp.google.com/g/156016

Bug: 5449033
Change-Id: I4c4e33bb9df3e39e11cd985e193e6fbab4635298
2012-01-03 22:38:27 +00:00
Romain Guy
82bc7a7727 Properly restore the GL scissor after a GL draw functor
Bug #5781254

Change-Id: I1dc4809563a793b6b579814951d4d73b4c34bf32
2012-01-03 14:14:55 -08:00
Alex Sakhartchouk
5b5664af2d Adding better error behavior to RS
Change-Id: Iea1cc76793f8978d60575404d7b1c5145e268d6e
2011-12-29 11:17:38 -08:00
Alex Sakhartchouk
36a7d6003d Adding RS comments, cleaning up the gl implementation with new Element hal.
Change-Id: I74de1e8248a73a45a39976e6027dac2626d580e9
2011-12-28 13:47:05 -08:00
Alex Sakhartchouk
fd79e02e0f Missing getter and additional tests.
Change-Id: I057dd0a29d2cad40d0a0627b4355faeb82bbc229
2011-12-22 14:30:55 -08:00
Alex Sakhartchouk
3aac0abe79 Adding tests for element/mesh getters.
Fixing bugs found by tests.

Change-Id: I6592a3b65f16b21255e7788fe8ee8aaafe268638
2011-12-22 13:11:48 -08:00
Alex Sakhartchouk
c9e7da6130 Adding simple hal struct to type.
Fixing off by 1 error in the element.

Change-Id: I0b142942c760cb861007af7099d35d0363f4e13b
2011-12-20 12:37:59 -08:00
Alex Sakhartchouk
8c998cf344 Merge "Adding comments to RS headers." into graphics-dev 2011-12-19 10:12:27 -08:00
Chet Haase
b6f3a279a8 am 0faa0556: Merge "De-allocate caches for large glyphs when trimming memory"
* commit '0faa05563992fd4b3632df50921a9c612378a049':
  De-allocate caches for large glyphs when trimming memory
2011-12-17 10:22:13 -08:00
Chet Haase
0faa055639 Merge "De-allocate caches for large glyphs when trimming memory" 2011-12-17 10:20:14 -08:00
Chet Haase
9a8245629d De-allocate caches for large glyphs when trimming memory
Currently, font renderers eliminate some texture caches when
memory is trimmed. This change makes it go further by eliminating the
large-glyph caches for all font renderers. These caches are
only allocated as needed, but continue to consume large amounts of
memory (CPU and GPU) after that allocation. De-allocating this memory
on a trim operation should prevent background apps from holding onto
this memory in the possible case that they have allocated it by drawing
large glyphs.

Change-Id: Id7a3ab49b244e036b442d87252fb40aeca8fdb26
2011-12-16 15:44:59 -08:00
Jeff Brown
f024eb2ebd am 1b678528: Merge "Fix invalid madvise() during concurrent alloc/dealloc of MemoryDealer"
* commit '1b678528237d4eea623840ce3ba5520c59ca4b2c':
  Fix invalid madvise() during concurrent alloc/dealloc of MemoryDealer
2011-12-16 14:37:07 -08:00
Jeff Brown
1b67852823 Merge "Fix invalid madvise() during concurrent alloc/dealloc of MemoryDealer" 2011-12-16 14:35:16 -08:00
Alex Sakhartchouk
11a2e341c5 Adding comments to RS headers.
Change-Id: Ib345fe9f651dd3e74fb384799d381034391de8a2
2011-12-16 14:15:21 -08:00
Alex Sakhartchouk
8d57f404fa Merge "Element getters." into graphics-dev 2011-12-16 09:51:53 -08:00
Alex Sakhartchouk
f5d8ac7cc3 Element getters.
Change-Id: Ibe570464557eeecad108b4d3cb8f02fad13943de
2011-12-16 09:44:26 -08:00
Chet Haase
ada81693a1 am 0904c203: Merge "Fix issues from recent glyph caching change"
* commit '0904c203c4c56e43f723522351b7a3bc304196ef':
  Fix issues from recent glyph caching change
2011-12-16 06:51:34 -08:00
Chet Haase
0904c203c4 Merge "Fix issues from recent glyph caching change" 2011-12-16 06:49:59 -08:00
Chet Haase
2a47c14e2a Fix issues from recent glyph caching change
There were 2 issues remaining after a recent change to support
glyph caching from multiple textures:
- memory in the GPU for all textures was being allocated automatically.
This is now lazy, being allocated only when those textures are first
needed.
- filtering (applied when a rendered object is transformed) was ignoring
the new multiple-texture structure. Filtering should be applied correctly
whenever we change textures.

Change-Id: I5c8eb8d46c73cd01782a353fc79b11cacc2146ab
2011-12-15 15:43:02 -08:00
Jason Sams
7d6bfab4bd Merge "Element HAL sturct, script side getters" into graphics-dev 2011-12-15 13:13:08 -08:00
Romain Guy
b4fb81773c am 18ca25be: Merge "Generate even fewer GL commands"
* commit '18ca25be58be75be78cdce7541b6b082a4b3864f':
  Generate even fewer GL commands
2011-12-15 10:10:48 -08:00
Romain Guy
374c12275a am 926d55eb: Merge "Reduce the number of active texture changes"
* commit '926d55eb40b9042c200f154fdf0c7c7969e0e359':
  Reduce the number of active texture changes
2011-12-15 10:08:20 -08:00
Romain Guy
58d0579eec am 9c986b7f: Merge "<Insert something improper about OpenGL>"
* commit '9c986b7f387adb8b65d9094ba931645141bc5253':
  <Insert something improper about OpenGL>
2011-12-15 10:07:52 -08:00
Romain Guy
27ecb1ab5f am 8a5f8ba4: Merge "Further reduce the number of GL commands sent to the driver"
* commit '8a5f8ba4f3e8352155c711adb649e6f536d194b5':
  Further reduce the number of GL commands sent to the driver
2011-12-15 10:07:33 -08:00
Alex Sakhartchouk
d24b1bb040 Element HAL sturct, script side getters
Change-Id: If98dd4c22bce58dca2c9739c8aee935a2dd0b493
2011-12-15 09:56:28 -08:00
Romain Guy
8f85e80b64 Generate even fewer GL commands
Change-Id: I0f4dcacb03ef5ee7f6ebd501df98bfead5f0a7f8
2011-12-14 19:23:32 -08:00
Ji-Hwan Lee
c17d9b75b6 Fix invalid madvise() during concurrent alloc/dealloc of MemoryDealer
Currently, madvise(MADV_REMOVE) is called after deallocation.
Another thread might allocate (and even write) the same region between
deallocation and madvise(), in which case the new thread will fail to read
what it have written.  So, call deallocate() after madvise(MADV_REMOVE).

Bug: 5654596
Change-Id: I26f36cd6013de499090768a0ddc68206a4a68219
2011-12-15 04:37:22 +09:00
Romain Guy
2d4fd36484 Reduce the number of active texture changes
Change-Id: I94046bdfe20740c26c8183822e3002d692fde7c4
2011-12-13 22:00:19 -08:00
Romain Guy
ec31f83bd3 <Insert something improper about OpenGL>
Change-Id: Ib645376093838156771588adc76a718da0ceb0db
2011-12-13 18:41:12 -08:00
Romain Guy
a1d3c91afb Further reduce the number of GL commands sent to the driver
Change-Id: Id922b2a166ea4573b767c27d3195e11c70320b23
2011-12-13 14:58:14 -08:00
Chet Haase
5bc5de87f3 am e829bc0f: Merge "Make glyph cache more flexible"
* commit 'e829bc0f0364e942bed01536d115a5c08d25d776':
  Make glyph cache more flexible
2011-12-13 14:29:06 -08:00
Romain Guy
07c04f7755 am 6f2de0a5: Merge "Reduce the number of GL commands generated by the UI"
* commit '6f2de0a5d6303ecff2795a94d5051ea0621be07f':
  Reduce the number of GL commands generated by the UI
2011-12-13 14:28:46 -08:00
Chet Haase
e829bc0f03 Merge "Make glyph cache more flexible" 2011-12-13 14:03:43 -08:00
Chet Haase
7de0cb12d0 Make glyph cache more flexible
Some GPU architectures could not handle the previous implementation
of our glyph cache. Frequent uploads would cause memory problems in the GPU
and eventually a crash due to these memory issues. The solution is to move to
a system of several, smaller caches instead of one monolythic cache for all
glyphs.

Change-Id: I0fc7a323360940d16d5a33eeb33abfab194c5920
2011-12-13 13:41:58 -08:00
Romain Guy
15bc6437f8 Reduce the number of GL commands generated by the UI
This optimization along with the previous one lets us render an
application like Gmail using only 30% of the number of GL commands
previously required

Change-Id: Ifee63edaf495e04490b5abd5433bb9a07bc327a8
2011-12-13 13:11:32 -08:00
Romain Guy
8c464b79bf am 5009f65c: Merge "Optimize state changes"
* commit '5009f65c9a676b64869e638e08ec0294e20d7e6e':
  Optimize state changes
2011-12-13 05:48:15 -08:00
Romain Guy
23d0deeb2b am da7642b3: Merge "Minimize the amount of data uploaded to draw text"
* commit 'da7642b31bb3490655016df979b0fad95425ac5f':
  Minimize the amount of data uploaded to draw text
2011-12-13 05:48:03 -08:00
Romain Guy
a72a3f077a am 51fbb009: Merge "Various OpenGL optimizations"
* commit '51fbb009661e8eeb41730fe107179425e5392b6d':
  Various OpenGL optimizations
2011-12-13 05:48:01 -08:00
Romain Guy
5009f65c9a Merge "Optimize state changes" 2011-12-12 20:36:22 -08:00
Romain Guy
f3a910b423 Optimize state changes
Change-Id: Iae59bc8dfd6427d0967472462cc1994987092827
2011-12-12 20:35:21 -08:00
Romain Guy
da7642b31b Merge "Minimize the amount of data uploaded to draw text" 2011-12-12 19:04:40 -08:00
Romain Guy
51fbb00966 Merge "Various OpenGL optimizations" 2011-12-12 19:04:32 -08:00
Romain Guy
d71dd367af Minimize the amount of data uploaded to draw text
Change-Id: I6313ac039291c9cd93aadafe3566ad9d60cab42d
2011-12-12 19:03:35 -08:00