OSDN Git Service
Aurimas Liutikas [Tue, 16 Feb 2016 20:46:45 +0000 (20:46 +0000)]
Merge "Fix -Wgnu-designator warnigns in cmds/flatland/Main.cpp" am:
f4475023e7
am:
8b8d7741e6
* commit '
8b8d7741e68ea10648369e1b80f8697413c22bfc':
Fix -Wgnu-designator warnigns in cmds/flatland/Main.cpp
Aurimas Liutikas [Tue, 16 Feb 2016 19:37:10 +0000 (19:37 +0000)]
Merge "Fix -Wgnu-designator warnigns in cmds/flatland/Main.cpp"
am:
f4475023e7
* commit '
f4475023e728db1456a72b0e9a1292e11fb38245':
Fix -Wgnu-designator warnigns in cmds/flatland/Main.cpp
Aurimas Liutikas [Tue, 16 Feb 2016 19:30:38 +0000 (19:30 +0000)]
Merge "Fix -Wgnu-designator warnigns in cmds/flatland/Main.cpp"
Pablo Ceballos [Tue, 16 Feb 2016 18:57:05 +0000 (18:57 +0000)]
Merge "Fix Surface slot caching" into nyc-dev
am:
3130422a1d
* commit '
3130422a1d7a7d2a5d528cd161e860fb5f66d353':
Fix Surface slot caching
Pablo Ceballos [Tue, 16 Feb 2016 18:46:40 +0000 (18:46 +0000)]
Merge "Fix Surface slot caching" into nyc-dev
Aurimas Liutikas [Tue, 16 Feb 2016 18:42:52 +0000 (10:42 -0800)]
Fix -Wgnu-designator warnigns in cmds/flatland/Main.cpp
Bug: http://b/
27068438
Change-Id: I3064950f761411fe5271c6259867ad437c71e66d
Christopher Wiley [Tue, 16 Feb 2016 17:04:22 +0000 (17:04 +0000)]
Merge "Replace AppOpsManager locking with std::lock_guard" am:
d3a9228d1d
am:
3f40743c4f
* commit '
3f40743c4f2cc86c505238cf5f597efca41277de':
Replace AppOpsManager locking with std::lock_guard
Christopher Wiley [Tue, 16 Feb 2016 16:57:44 +0000 (16:57 +0000)]
Merge "Replace AppOpsManager locking with std::lock_guard"
am:
d3a9228d1d
* commit '
d3a9228d1d2388f21b4e8013cf2fad7b4c91ace7':
Replace AppOpsManager locking with std::lock_guard
Christopher Wiley [Tue, 16 Feb 2016 16:50:01 +0000 (16:50 +0000)]
Merge "Replace AppOpsManager locking with std::lock_guard"
Xia Yang [Tue, 16 Feb 2016 15:31:30 +0000 (15:31 +0000)]
dumpstate: remove length modifier on assignment suppressed scanf item
am:
60292e5c83
* commit '
60292e5c8324475c5afbd0a6703b6b3ae2c94b1a':
dumpstate: remove length modifier on assignment suppressed scanf item
Xia Yang [Tue, 16 Feb 2016 15:31:03 +0000 (15:31 +0000)]
dumpstate: remove -std=gnu99
am:
861dd225fc
* commit '
861dd225fcd53f3e8d10469600648226c6e887cf':
dumpstate: remove -std=gnu99
Xia Yang [Tue, 16 Feb 2016 11:05:18 +0000 (03:05 -0800)]
dumpstate: remove length modifier on assignment suppressed scanf item
gcc throws warning for using the two together as there is no point specifying
storage size if the item is not to be stored. With -Werror enabled in this
module, such warning breaks the build under gcc compiling.
Change-Id: I6478c7cdb39f4709a63089191cc4ff541564cea4
Xia Yang [Tue, 16 Feb 2016 10:26:09 +0000 (02:26 -0800)]
dumpstate: remove -std=gnu99
-std=gnu99 no longer has effect now that dumpstate is C++.
The flag also break build if target platform is configured to default
gcc compiler since gcc throws warning for having -std=gnu99 when compiling
C++ and we have -Werror for the module.
Reproduce:
mm LOCAL_CLANG=false
...
error: command line option '-std=gnu99' is valid for C/ObjC but not for C++ [-Werror]
Change-Id: If1c272868ec4626873e7d91342b1fea51a15c52e
Courtney Goeltzenleuchter [Mon, 15 Feb 2016 16:41:12 +0000 (16:41 +0000)]
Merge changes Id64bb14c,I47fb1475,Iefd17b62,Ib0e963e8,I236183d5 into nyc-dev
am:
deee0d4d19
* commit '
deee0d4d1983f495bd57e2506d0f96993c5255f6':
loader: Add extension before CreateInstance chain
vkinfo: Add debug command line option
vkinfo: Add validation layers
loader: Fix enumerate device extension properties
loader: Fix enumerate device layer properties
Courtney Goeltzenleuchter [Mon, 15 Feb 2016 16:33:30 +0000 (16:33 +0000)]
Merge changes Id64bb14c,I47fb1475,Iefd17b62,Ib0e963e8,I236183d5 into nyc-dev
* changes:
loader: Add extension before CreateInstance chain
vkinfo: Add debug command line option
vkinfo: Add validation layers
loader: Fix enumerate device extension properties
loader: Fix enumerate device layer properties
Dimitry Ivanov [Sat, 13 Feb 2016 01:56:41 +0000 (01:56 +0000)]
Merge "Add missing liblog dependency" am:
39f1abac3c
am:
3b3e298583
* commit '
3b3e2985832ffb5d3fb3b6af4dc8be66f8fd01b0':
Add missing liblog dependency
Dimitry Ivanov [Sat, 13 Feb 2016 01:50:29 +0000 (01:50 +0000)]
Merge "Add missing liblog dependency"
am:
39f1abac3c
* commit '
39f1abac3c21e2f5a4a450b21d1c52ba5f099ddb':
Add missing liblog dependency
Dimitry Ivanov [Sat, 13 Feb 2016 01:43:29 +0000 (01:43 +0000)]
Merge "Add missing liblog dependency"
Christopher Wiley [Sat, 13 Feb 2016 01:32:31 +0000 (01:32 +0000)]
Merge "Don\'t rely on AppOpsManager in systems without applications" am:
46c32a1b1c
am:
c61c5c05ef
* commit '
c61c5c05ef2c64d6a0dca6be8eab7842143529b6':
Don't rely on AppOpsManager in systems without applications
Christopher Wiley [Sat, 13 Feb 2016 01:32:23 +0000 (01:32 +0000)]
Merge "Fix race in AppOpsManager client library" am:
712fe62ab2
am:
f2936d7bec
* commit '
f2936d7bec9ce8e56f1bd4df052376dea192c809':
Fix race in AppOpsManager client library
Dimitry Ivanov [Sat, 13 Feb 2016 01:13:10 +0000 (17:13 -0800)]
Add missing liblog dependency
Bug: http://b/
27171986
Change-Id: Ifbe01c6b325e7198a95ea865f557b0c9acb65809
Christopher Wiley [Sat, 13 Feb 2016 00:07:49 +0000 (00:07 +0000)]
Merge "Don\'t rely on AppOpsManager in systems without applications"
am:
46c32a1b1c
* commit '
46c32a1b1c895ec7630686aa8228785479f8fe92':
Don't rely on AppOpsManager in systems without applications
Christopher Wiley [Sat, 13 Feb 2016 00:07:40 +0000 (00:07 +0000)]
Merge "Fix race in AppOpsManager client library"
am:
712fe62ab2
* commit '
712fe62ab282b3897fc1ee98e35b12d95663be52':
Fix race in AppOpsManager client library
Christopher Wiley [Fri, 12 Feb 2016 23:41:08 +0000 (15:41 -0800)]
Replace AppOpsManager locking with std::lock_guard
Change-Id: Ice4c101659837cea594194cfb335f9f3f6837f38
Christopher Wiley [Fri, 12 Feb 2016 23:43:07 +0000 (23:43 +0000)]
Merge "Don't rely on AppOpsManager in systems without applications"
Christopher Wiley [Fri, 12 Feb 2016 23:43:00 +0000 (23:43 +0000)]
Merge "Fix race in AppOpsManager client library"
Courtney Goeltzenleuchter [Fri, 12 Feb 2016 14:53:12 +0000 (07:53 -0700)]
loader: Add extension before CreateInstance chain
A developer can cause the loader to enable VK_EXT_debug_report
extension by setting a specific property. However, that check
was happening after the instance chain was called and thus
no layers saw the extension enabled.
Also fix use of AddExtensionToCreateInfo to always allocate
and insert VK_EXT_debug_report. Otherwise cannot distinguish
between the failure case and the extension is already present
case.
Change-Id: Id64bb14cb2086e94190625b4364329d634afda70
Courtney Goeltzenleuchter [Tue, 9 Feb 2016 03:31:46 +0000 (20:31 -0700)]
vkinfo: Add debug command line option
Really handy for loader debug to switch between
waiting and not waiting for a debugger to attach.
Change-Id: I47fb1475d8aa62b8f4b2b7e6a156793f108a83ed
Felipe Leme [Fri, 12 Feb 2016 18:06:21 +0000 (18:06 +0000)]
Changed permission of temporary bugreport and dumpstate log files to Shell.
am:
6fe9db67f6
* commit '
6fe9db67f6c92d5fbf87d371da5cca412f672630':
Changed permission of temporary bugreport and dumpstate log files to Shell.
Felipe Leme [Fri, 12 Feb 2016 17:04:16 +0000 (09:04 -0800)]
Changed permission of temporary bugreport and dumpstate log files to Shell.
BUG:
27164100
Change-Id: I569cbe4854fad44c95c9bf3e2efa9c3c93317003
Felipe Leme [Fri, 12 Feb 2016 00:49:57 +0000 (00:49 +0000)]
Merge "Split bugreport identifier into id and pid." into nyc-dev
am:
421c1c07e7
* commit '
421c1c07e73ef91a8fb887ad0045f37335722b2e':
Split bugreport identifier into id and pid.
Felipe Leme [Fri, 12 Feb 2016 00:41:10 +0000 (00:41 +0000)]
Merge "Split bugreport identifier into id and pid." into nyc-dev
Courtney Goeltzenleuchter [Tue, 9 Feb 2016 03:31:25 +0000 (20:31 -0700)]
vkinfo: Add validation layers
Change-Id: Iefd17b622f0d91219b81bd1df09ffeb3397f9553
Courtney Goeltzenleuchter [Sun, 7 Feb 2016 17:32:27 +0000 (10:32 -0700)]
loader: Fix enumerate device extension properties
Same as device layer property issue.
Loader needs to provide aggregate data rather
than send to instance chain.
Change-Id: Ib0e963e8d1780e4870110c556051da921f94353b
Courtney Goeltzenleuchter [Sat, 6 Feb 2016 00:10:59 +0000 (17:10 -0700)]
loader: Fix enumerate device layer properties
Loader was calling down the chain when is needs to
respond with aggregate data. This patch fixes
https://buganizer.corp.google.com/u/0/issues/
26982782
Change-Id: I236183d5d781a6efc3d84db4a51376b8cc29664f
Casey Dahlin [Thu, 11 Feb 2016 23:35:44 +0000 (23:35 +0000)]
Merge "libbinder: Allow readByteVector to take uint8_t" am:
dda39f2a22
am:
f226203232
* commit '
f2262032320c31a87659df81b0b1db76ea3310dd':
libbinder: Allow readByteVector to take uint8_t
Casey Dahlin [Thu, 11 Feb 2016 23:31:07 +0000 (23:31 +0000)]
Merge "libbinder: Allow readByteVector to take uint8_t"
am:
dda39f2a22
* commit '
dda39f2a2229ea74cc4738daa1c6a1c758284d7d':
libbinder: Allow readByteVector to take uint8_t
Casey Dahlin [Thu, 11 Feb 2016 23:24:41 +0000 (23:24 +0000)]
Merge "libbinder: Allow readByteVector to take uint8_t"
Dan Stoza [Thu, 11 Feb 2016 21:29:36 +0000 (21:29 +0000)]
Merge "SF: Prevent non-sync transactions from syncing" into nyc-dev
am:
da8ca93b43
* commit '
da8ca93b430cdb9f1b559e065e194d6701d30291':
SF: Prevent non-sync transactions from syncing
Dan Stoza [Thu, 11 Feb 2016 21:24:35 +0000 (21:24 +0000)]
Merge "SF: Prevent non-sync transactions from syncing" into nyc-dev
Dan Stoza [Thu, 11 Feb 2016 19:43:58 +0000 (11:43 -0800)]
SF: Prevent non-sync transactions from syncing
Clears the sync handle and frame number after registering a deferred
transaction to prevent other non-synchronized transactions from
erroneously getting marked as synchronized.
Also changes the application of pending states such that all valid
pending transactions are applied instead of breaking after the first
one.
Bug:
27061473
Change-Id: I98f0f26e9d9bb0d266563f209e94dfd776d8f8ea
Felipe Leme [Thu, 11 Feb 2016 19:05:45 +0000 (19:05 +0000)]
Merge "Send broadcasts as Shell." into nyc-dev
am:
c30e3e7c09
* commit '
c30e3e7c095603c3c48b3c2691db80fec570c731':
Send broadcasts as Shell.
Felipe Leme [Thu, 11 Feb 2016 18:53:46 +0000 (18:53 +0000)]
Merge "Send broadcasts as Shell." into nyc-dev
Felipe Leme [Thu, 11 Feb 2016 17:12:39 +0000 (09:12 -0800)]
Send broadcasts as Shell.
The BUGREPORT_STARTED was being sent as root, which would cause it to
bypass Shell restrictions when DISALLOW_DEBUGGING_FEATURES is set.
Also added the missing --receiver-background and unhardcoded shell's uid
from previous changes.
BUG:
26814173
Change-Id: I867361efafb96c7e709f83806a9301afc637a27e
Pablo Ceballos [Thu, 11 Feb 2016 17:56:59 +0000 (17:56 +0000)]
Merge "BQ: Fix iterator use" into nyc-dev
am:
5d9d711a16
* commit '
5d9d711a16d7560f4be62cad2413dbcb5feddfde':
BQ: Fix iterator use
Pablo Ceballos [Thu, 11 Feb 2016 17:54:03 +0000 (17:54 +0000)]
Merge "BQ: Fix iterator use" into nyc-dev
Srinath Sridharan [Thu, 11 Feb 2016 17:18:12 +0000 (17:18 +0000)]
Split dumpsys into critical and non-critical sections.
am:
fdf52d3697
* commit '
fdf52d3697aa0396bd9d8883892937b99a6772a0':
Split dumpsys into critical and non-critical sections.
Srinath Sridharan [Mon, 1 Feb 2016 23:50:22 +0000 (15:50 -0800)]
Split dumpsys into critical and non-critical sections.
Order in which we report stats in bugreport can disrupt cpu and memory
stats. This CL sorts the bugreport process
from least impactful to most impactful to try and keep the
system stats as close to its initial state as possible throughout the BR (eg.
things like dumpsys meminfo should run before dumping thread stacks)
BUG:
26639621
Change-Id: I29dccfd423d83a6a69b7e025d45cb9cebfc41cc4
Courtney Goeltzenleuchter [Thu, 11 Feb 2016 00:06:33 +0000 (00:06 +0000)]
loader: Can\'t dlclose until all done with the layers
am:
e6e6968345
* commit '
e6e6968345e12f633bed0add4fa4600e7ccb7542':
loader: Can't dlclose until all done with the layers
Courtney Goeltzenleuchter [Thu, 11 Feb 2016 00:06:20 +0000 (00:06 +0000)]
loader: Implement new layer initialization
am:
a90ce61c1a
* commit '
a90ce61c1a4c0eae79b117f2137e9fe7ed41aa6d':
loader: Implement new layer initialization
Courtney Goeltzenleuchter [Thu, 11 Feb 2016 00:06:13 +0000 (00:06 +0000)]
loader: Add GetName and GetDescription for logs
am:
eff631189b
* commit '
eff631189b135a9579a7f8510f9b1ccadd6fb4f9':
loader: Add GetName and GetDescription for logs
Courtney Goeltzenleuchter [Thu, 11 Feb 2016 00:06:01 +0000 (00:06 +0000)]
loader: Layer init structures
am:
1dd712aefd
* commit '
1dd712aefdc771297823b24f26d768cd8bd14d49':
loader: Layer init structures
Courtney Goeltzenleuchter [Thu, 11 Feb 2016 00:05:48 +0000 (00:05 +0000)]
loader: Only use driver\'s DebugReport if it has one
am:
6fecdd563f
* commit '
6fecdd563f3bf94dadedc78512a1b28c08c67e07':
loader: Only use driver's DebugReport if it has one
Courtney Goeltzenleuchter [Thu, 11 Feb 2016 00:05:35 +0000 (00:05 +0000)]
tools: Need to specify pQueuePriorities
am:
ca472ab8e4
* commit '
ca472ab8e43efe92db30660db9997117e4b00ab9':
tools: Need to specify pQueuePriorities
Pablo Ceballos [Wed, 10 Feb 2016 23:43:22 +0000 (15:43 -0800)]
BQ: Fix iterator use
Don't use an iterator after it's been removed from the container.
Change-Id: I374b454bba95d10dcec6a600bf8f16e5636e426b
Courtney Goeltzenleuchter [Fri, 29 Jan 2016 00:26:17 +0000 (17:26 -0700)]
loader: Can't dlclose until all done with the layers
The loader was freeing the layers (dlclose) at the
bottom of the DestroyInstance chain. Would get a
segfault when returning from DestroyInstance_Bottom
as layers would try to run that are no longer in memory.
Change-Id: I25cc611fed9d8b3a90fd2652e53cd450daefa08c
(cherry picked from commit
3160835e44fcb0c5367d2f50b132e41c789998d5)
Courtney Goeltzenleuchter [Tue, 9 Feb 2016 03:48:05 +0000 (20:48 -0700)]
loader: Implement new layer initialization
We now call down the instance and device chains
as part of the creation process. The loader puts
together "LinkInfo" structures that each layer
uses to reference the next layer down in the stack.
These info structures are attached to the CreateInfo
pNext structure. Also have an info structure that
the loader terminator functions (CreateInstance_Bottom
and CreateDevice_Bottom) that includes loader state
they need.
As part of this the CreateDevice function now lives
on the device chain and it has been split into
Top and Bottom functions. Top builds the LinkInfo
structures and Bottom makes the actual call to the ICD
to create the device object.
Change-Id: I5f5a6d579c34942511cce288ba8fa51f9f89f67f
(cherry picked from commit
9091095245975597a0b2031c3b4847f780c1721e)
Courtney Goeltzenleuchter [Tue, 9 Feb 2016 03:12:59 +0000 (20:12 -0700)]
loader: Add GetName and GetDescription for logs
Handy to have more layer detail for logging.
Change-Id: I6933a8fcb90b9cfb70fc85a358244863aac9d702
(cherry picked from commit
cf275f37b4a903b5dcc95b5100db667c906aa33f)
Courtney Goeltzenleuchter [Tue, 2 Feb 2016 18:51:32 +0000 (11:51 -0700)]
loader: Layer init structures
Update to match LoaderAndTools vk_layer.h
which includes the new layer initialization structures.
Change-Id: Ie6bf2a1705373bad3c77dd0b04be1263f1c4fc46
(cherry picked from commit
f7491b28bfcc47ce09b32168a5fd25b58c6aceab)
Courtney Goeltzenleuchter [Wed, 3 Feb 2016 22:14:46 +0000 (15:14 -0700)]
loader: Only use driver's DebugReport if it has one
Change-Id: Ic020b9d5a95c9ddd20dd4c94fd6e7de050b83f2b
(cherry picked from commit
b776ba1cff149bb724a1958cc3f470ba5c113b63)
Courtney Goeltzenleuchter [Tue, 2 Feb 2016 03:09:00 +0000 (20:09 -0700)]
tools: Need to specify pQueuePriorities
Change-Id: I689c3e8da9e8cf00705f2f90d472c1268a978f1b
(cherry picked from commit
363ac5d35ca076f419c14f14bce603d5d82553fa)
Jesse Hall [Wed, 10 Feb 2016 18:02:35 +0000 (18:02 +0000)]
Merge "libvulkan: Support R5G6B5_UNORM_PACK16 as a swapchain format" into nyc-dev
am:
3d92c53d99
* commit '
3d92c53d99866ddd28c9ff20f610074c82f70d43':
libvulkan: Support R5G6B5_UNORM_PACK16 as a swapchain format
Jesse Hall [Wed, 10 Feb 2016 17:49:49 +0000 (17:49 +0000)]
Merge "libvulkan: Support R5G6B5_UNORM_PACK16 as a swapchain format" into nyc-dev
Mark Salyzyn [Wed, 10 Feb 2016 16:15:53 +0000 (16:15 +0000)]
Merge "dumpstate: Continued build breakage" am:
16072e7e17
am:
c01e8dc4a6
* commit '
c01e8dc4a687ae3445f68b8d005e2b29a84f6e7f':
dumpstate: Continued build breakage
Mark Salyzyn [Wed, 10 Feb 2016 16:13:54 +0000 (16:13 +0000)]
Merge "dumpstate: Continued build breakage"
am:
16072e7e17
* commit '
16072e7e1717bcf1ea8be99eb7af90974dbbeeb7':
dumpstate: Continued build breakage
Mark Salyzyn [Wed, 10 Feb 2016 16:12:50 +0000 (16:12 +0000)]
Merge "dumpstate: Continued build breakage"
Jesse Hall [Wed, 10 Feb 2016 16:11:49 +0000 (16:11 +0000)]
Merge "libvulkan: Un-revert and fix VK_PRESENT_MODE_MAILBOX_KHR" into nyc-dev
am:
148a1b45ff
* commit '
148a1b45ff2c205f6aa9c46c7f039d29d877a35c':
libvulkan: Un-revert and fix VK_PRESENT_MODE_MAILBOX_KHR
Jesse Hall [Wed, 10 Feb 2016 16:11:34 +0000 (16:11 +0000)]
Merge "vknulldrv: Report physical device limits" into nyc-dev
am:
45ba96e44c
* commit '
45ba96e44c0d7d5dac6e86f0bfbfc44b4f2618ac':
vknulldrv: Report physical device limits
Mark Salyzyn [Wed, 10 Feb 2016 15:41:12 +0000 (07:41 -0800)]
dumpstate: Continued build breakage
- drop assignment length in scanf
- drop -std=gnu99 as invalid for C++
Change-Id: I924d2561069b3c289cb9deb475f732c46bb5a67e
Jesse Hall [Wed, 10 Feb 2016 16:08:26 +0000 (16:08 +0000)]
Merge "libvulkan: Initialize VkSurfacePropertiesKHR::supportedCompositeAlpha" into nyc-dev
am:
c9456b532d
* commit '
c9456b532d83edafc997d60b2c31d67bb77ac3f9':
libvulkan: Initialize VkSurfacePropertiesKHR::supportedCompositeAlpha
Jesse Hall [Wed, 10 Feb 2016 16:07:53 +0000 (16:07 +0000)]
Merge "libvulkan: Un-revert and fix VK_PRESENT_MODE_MAILBOX_KHR" into nyc-dev
Jesse Hall [Wed, 10 Feb 2016 16:06:54 +0000 (16:06 +0000)]
Merge "vknulldrv: Report physical device limits" into nyc-dev
Jesse Hall [Wed, 10 Feb 2016 16:02:55 +0000 (16:02 +0000)]
Merge "libvulkan: Initialize VkSurfacePropertiesKHR::supportedCompositeAlpha" into nyc-dev
Jesse Hall [Wed, 10 Feb 2016 08:07:18 +0000 (00:07 -0800)]
libvulkan: Support R5G6B5_UNORM_PACK16 as a swapchain format
Change-Id: I489cf12965372d5281d200102a4632e6e8d75ede
Jesse Hall [Wed, 10 Feb 2016 06:49:12 +0000 (06:49 +0000)]
Merge "vkinfo: Fix bytes -> megabytes conversion" into nyc-dev
am:
e59624db83
* commit '
e59624db8323445569b2a9d04aea694298f9a2eb':
vkinfo: Fix bytes -> megabytes conversion
Jesse Hall [Wed, 10 Feb 2016 06:46:40 +0000 (06:46 +0000)]
Merge "vkinfo: Fix bytes -> megabytes conversion" into nyc-dev
Jesse Hall [Tue, 9 Feb 2016 21:26:59 +0000 (13:26 -0800)]
libvulkan: Initialize VkSurfacePropertiesKHR::supportedCompositeAlpha
Also rearrange the code to set the fields of this struct in the same
order they appear in the struct, to make it easier to tell when
something is missing.
Bug:
26991609
Change-Id: I681ed42c96ac75bb1589d30ab75f760f660eacdd
Jesse Hall [Wed, 10 Feb 2016 06:35:04 +0000 (22:35 -0800)]
vknulldrv: Report physical device limits
Limits are set to the minimum values allowed by the spec.
Change-Id: Ib3ca99ad6d1537e2ed3fa29e15368419f2331ba7
Jesse Hall [Wed, 10 Feb 2016 06:13:34 +0000 (22:13 -0800)]
libvulkan: Un-revert and fix VK_PRESENT_MODE_MAILBOX_KHR
This reverts commit
9ec497c, which was a revert of
f333922, and fixes
the problem that caused the initial revert.
MODE_MAILBOX corresponds to BufferQueue "async" mode; in that mode we
need one extra un-dequeued buffer at all times. We weren't accounting
for that, and weren't allocating enough bufers.
Bug:
26918467
Change-Id: Ice67e6d16828e50413ae453ecf3c67b8fb288f1a
Pablo Ceballos [Fri, 5 Feb 2016 21:39:27 +0000 (13:39 -0800)]
Fix Surface slot caching
- Now that it's possible to resize a BufferQueue while buffers are
dequeued/acquired, it's no longer correct for Surface to clear its
cache when the BufferQueue is resized since it must keep at least
the currently dequeued buffers.
- Add an onSlotsFreed callback to IProducerListener so that producers
that wish to be notified about buffers being freed can do so. Note
that this isn't currently used in Surface.
- Review and fixup all the places where the producer/consumer
listeners for freed buffers should be called.
Change-Id: I4ab0c92bc69b75a3c072ddf5d74d78f941dba4c8
Felipe Leme [Wed, 10 Feb 2016 02:13:52 +0000 (02:13 +0000)]
Felipe Leme [Wed, 10 Feb 2016 00:35:14 +0000 (16:35 -0800)]
Fixed format.
Change-Id: If0322e3dcf56005541aa08e0ab91dae10d2d385d
Felipe Leme [Wed, 10 Feb 2016 00:18:27 +0000 (00:18 +0000)]
Fixed C++ syntax.
am:
60869c9e1e
* commit '
60869c9e1e91170d487715fc30b795d786630c54':
Fixed C++ syntax.
Felipe Leme [Wed, 10 Feb 2016 00:07:20 +0000 (16:07 -0800)]
Fixed C++ syntax.
Change-Id: I9702fdede589dd37cb498bbf63c0773816a7e3b8
Mark Salyzyn [Tue, 9 Feb 2016 23:39:46 +0000 (23:39 +0000)]
dumpstate: report process times
am:
a297c3258f
* commit '
a297c3258f6f7ea52cc2dcc42d62f85fda12a163':
dumpstate: report process times
Casey Dahlin [Tue, 9 Feb 2016 19:08:35 +0000 (11:08 -0800)]
libbinder: Allow readByteVector to take uint8_t
Change-Id: I2f0d1f3b4c04f04e7a866e9a4cf76c899b4b4980
Test: AIDL integration tests pass
Bug:
27078230
Mark Salyzyn [Fri, 5 Feb 2016 23:33:17 +0000 (15:33 -0800)]
dumpstate: report process times
(cherry picked from commit
0751efa64cddfab69749236193e776058c9db595)
- Fix per-pid helper to pick up comm if cmdline is empty
and assume kernel thread.
- Add per pid report command line user time if not kernel thread,
system time and I/O time, plus percentage if non zero.
Bug:
26966375
Change-Id: I6b499f535cc3687eb468b32ae39e7d151dffb11f
Felipe Leme [Tue, 9 Feb 2016 18:40:07 +0000 (10:40 -0800)]
Split bugreport identifier into id and pid.
Although 'pid' is more useful when diagnosing problems with the
bugreport workflow, it could be confusing to the end user. Hence, a
sequential id (started at 1 after a reboot) would be more useful.
BUG:
27076108
Change-Id: I6fd7ea8c703e8040034af183ef5cd35adfbcb4ef
Mark Salyzyn [Tue, 9 Feb 2016 17:47:49 +0000 (09:47 -0800)]
resolve merge conflicts of
c1c11d20de to nyc-dev-plus-aosp.
Change-Id: Ia9520e7ac0d13b440ab5a83991585c03eccf00a0
Mark Salyzyn [Tue, 9 Feb 2016 17:25:01 +0000 (17:25 +0000)]
Merge changes I6b499f53,If821ed21
am:
c658aee7d6
* commit '
c658aee7d6dc56b1b3b1722412029830e966ba71':
dumpstate: report process times
Add option to #define _DUMPSTATE_DRY_RUN_ to skip the actual dumps.
Mark Salyzyn [Tue, 9 Feb 2016 17:22:50 +0000 (17:22 +0000)]
Merge changes I6b499f53,If821ed21
* changes:
dumpstate: report process times
Add option to #define _DUMPSTATE_DRY_RUN_ to skip the actual dumps.
Mark Salyzyn [Fri, 5 Feb 2016 23:33:17 +0000 (15:33 -0800)]
dumpstate: report process times
- Fix per-pid helper to pick up comm if cmdline is empty
and assume kernel thread.
- Add per pid report command line user time if not kernel thread,
system time and I/O time, plus percentage if non zero.
Bug:
26966375
Change-Id: I6b499f535cc3687eb468b32ae39e7d151dffb11f
Felipe Leme [Wed, 11 Nov 2015 04:10:25 +0000 (20:10 -0800)]
Add option to #define _DUMPSTATE_DRY_RUN_ to skip the actual dumps.
(cherry picked from commit
93d705b6d742e8f00167ff2384e35c2205a6f9da)
Such option is useful when debugging dumpstate itself, since it will
finish much sooner and its output will be much smaller.
Change-Id: If821ed21715461bf82eea0b2be4b926239ad69da
Jesse Hall [Tue, 9 Feb 2016 05:20:20 +0000 (21:20 -0800)]
vkinfo: Fix bytes -> megabytes conversion
Also only report 256 MiB (chosen arbitrarily) as the null driver's
memory heap size. The previous value wasn't reasonable on 64-bit
systems (2^63 + 1).
Change-Id: Id57a4895752af45d046072ae8fee386ca219e82d
Jesse Hall [Mon, 18 Jan 2016 12:01:10 +0000 (04:01 -0800)]
vkinfo: Print VkPhysicalDeviceFeatures
Change-Id: Ifa045de906ce2612382f470168438f80c7188b0a
Jesse Hall [Tue, 9 Feb 2016 00:47:53 +0000 (00:47 +0000)]
Merge "vkinfo: improve indentation, add verbosity parameters" into nyc-dev
Jesse Hall [Mon, 18 Jan 2016 12:00:57 +0000 (04:00 -0800)]
vknulldrv: Report valid VkPhysicalDeviceFeatures
Change-Id: I616f8e40d42abb812cc150d31b3a07d41d86b43c
Jesse Hall [Mon, 18 Jan 2016 11:41:35 +0000 (03:41 -0800)]
vkinfo: improve indentation, add verbosity parameters
Change-Id: I23b4a08a24d6fb1463db55dbbb342eca866fc228
Jesse Hall [Mon, 1 Feb 2016 22:43:47 +0000 (14:43 -0800)]
vulkan: Document codegen instructions and code style
Change-Id: I7565506a866b72f4ce31ac6bdd3d97e357497ab0
(cherry picked from commit
e0d27bb1f3671401c2f6d84a60111f5c695c97d3)
Felipe Leme [Mon, 8 Feb 2016 17:53:18 +0000 (09:53 -0800)]
Improved dumpstate logging.
dumpstate generates 3 types of output:
- stdout: used for the bugreport itself and is redirected to a file
if requested.
- stderr: used for errors, is only useful when dumpstate is called from
adb (otherwise is ignored).
- ALOG: logged into logcat and available in the bugreport.
This current approach has 2 problems:
- stderr output is ignored when dumpstatet is called from bugreport
services.
- bugs on dumpstate are often hard to diagnose
(AKA 'Who watches the watchmen?').
This change mitigates these problems by redirecting stderr into a log
file (which will also be included in the bugreport notification) and by
keeping the .tmp file around (for the extreme cases where the .zip file
could not be generated or got corrupted).
BUG:
26906985
Change-Id: I73e7b1dd10ad9f83b7aa1043131ff3b74c426fdb