Mark Lobodzinski [Mon, 8 Aug 2016 16:18:49 +0000 (10:18 -0600)]
layers: Remove swapchain INVALID_COUNT macro
Change-Id: I44d19882af69541cfe9876a22fecc0a09056ba0b
Mark Lobodzinski [Mon, 8 Aug 2016 15:57:09 +0000 (09:57 -0600)]
layers: Remove swapchain NULL_POINTER checks
These are all overed in parameter_validation.
Change-Id: I23a7fc281414b6b8bcf30c5a9660560f2b51177f
Mark Lobodzinski [Mon, 8 Aug 2016 15:43:22 +0000 (09:43 -0600)]
layers: Removed LOG_ERROR_NON_VALID_OBJ macro
Change-Id: I0609694a3743ca764bcda6de66c0431fb63c1ad1
Mark Lobodzinski [Mon, 8 Aug 2016 15:38:42 +0000 (09:38 -0600)]
layers: Replace swapchain LOG_ERROR macros
Fixed some missing return codes, incorrect parameter order, bad object
handles, and instance-device mismatches. Inlined log-error calls.
Change-Id: I6d272cafcd8554f068896312a2b0f41a42d108be
Mark Lobodzinski [Tue, 16 Aug 2016 14:03:31 +0000 (08:03 -0600)]
layers: Fix typo in layer validation details doc
DynamicBlendConstantsNotBound was misspelled.
Change-Id: I33f4d53e336dc39b5f23583ee24df88441a339ac
Mark Lobodzinski [Wed, 10 Aug 2016 21:12:32 +0000 (15:12 -0600)]
layers: Update details document for new PV test
Added test to use a non-solid fillmode when this feature is disabled.
Change-Id: If43d8d2f30fa892b96cb90585eb0513649c535f7
Mark Lobodzinski [Tue, 9 Aug 2016 19:10:51 +0000 (13:10 -0600)]
layers: GH785, Validate rasterinfo->polygonMode
If fillModeNonSolid is not enabled, polygonMode cannot be set to
POINT or LINE.
Change-Id: I6cfdfac8f5c9ff7d27b94bf0aa80ae1249b0ce25
Mark Lobodzinski [Mon, 15 Aug 2016 22:22:50 +0000 (16:22 -0600)]
layers: Fix missing PV WSI registration
Change-Id: I1c924977a420bc7a417710b82f0a42f4ae5dd4c3
Lenny Komow [Mon, 15 Aug 2016 20:41:26 +0000 (14:41 -0600)]
winsdk: Move uninstall log to temp
Tobin Ehlis [Fri, 12 Aug 2016 20:26:31 +0000 (14:26 -0600)]
layers: Style fixup, don't use pass-by-reference params
Change pass-by-reference to a ptr to comply with Google style guide.
This way it's more clear that function is modifying param, especially when calling.
Tobin Ehlis [Fri, 12 Aug 2016 20:12:44 +0000 (14:12 -0600)]
layers: Naming corrections
Some minor fix-ups to variable names and comments for memory aliasing code.
Tobin Ehlis [Thu, 11 Aug 2016 01:11:54 +0000 (19:11 -0600)]
layers: Refactor set/validate memory valid functions
Create custom image and buffer versions of functions to minimize casting.
Tobin Ehlis [Thu, 11 Aug 2016 00:01:27 +0000 (18:01 -0600)]
layers: Push some casts into helper functions
Tobin Ehlis [Wed, 10 Aug 2016 23:00:51 +0000 (17:00 -0600)]
layers: Handle memory mapped prior to object binding
This adds initial support for memory that is updated via a MapMemory() call
prior to an object being bound. Initially any mapped allocation will just have
global_valid set for whole allocation. This value is then picked up by any
ranges made off of that allocation.
Really, we need to create new mapped regions and mark them valid, then make
sure that any overlapped objects pick up valid from mapped regions. This
requires more thought and design, though, so commiting this changes as a
simpler and incomplete initial fix.
Tobin Ehlis [Wed, 10 Aug 2016 22:08:00 +0000 (16:08 -0600)]
layers : Update memory valid to be per-range
We previously only tracked a single valid bool for an entire allocation.
That can easily get out of synch when multiple objects are bound to the
same allocation. This update moves the valid tracking to be per-range so
that it can be tracked on a per-object basis.
This does handle aliasing cases where various objects completely alias each
other, but there are still various aliased resource cases that aren't handled.
Tobin Ehlis [Wed, 10 Aug 2016 15:55:59 +0000 (09:55 -0600)]
layers: Refactor InsertMemoryRange
Work with local MEMORY_RANGE struct and insert to map at the end.
This avoids a loop pass and the awkward loop check of a range against itself.
Adds the slight complexity of storing aliases during loop and updating them
after the loop, but I prefer that to the alternative.
Tobin Ehlis [Mon, 8 Aug 2016 18:33:11 +0000 (12:33 -0600)]
layers: Update bound memory tracking and alias checking
Store all ranges bound to a single memory allocation in a single map indexed by object id.
Add two separate unordered_sets for independent image and vector processing.
Added a set of aliases to MEMORY_RANGE struct to hold any aliased ranges.
Insert aliased ranges at create time and remove the aliases when a range
is destroyed.
Have a single function, rangesInterset, to track if regions bound to a memory
allocation overlap, and if linear/non-linear overlap in violation of the spec.
Rene Lindsay [Fri, 12 Aug 2016 23:56:09 +0000 (17:56 -0600)]
demos : SAM6 Flip projection matrix, instead of vertex-shader
Change-Id: I0000000000decaf15bad0000000000000000006b
Lenny Komow [Fri, 12 Aug 2016 19:26:20 +0000 (13:26 -0600)]
winsdk: Clean up confiure_runtime.c and docs
Lenny Komow [Wed, 10 Aug 2016 16:31:07 +0000 (10:31 -0600)]
winsdk: Make RT installer use C program
Lenny Komow [Tue, 9 Aug 2016 21:56:15 +0000 (15:56 -0600)]
winsdk: Add C program for runtime configuration
Tony Barbour [Wed, 10 Aug 2016 16:32:42 +0000 (10:32 -0600)]
docs: Add renderpass tests from GH427
Change-Id: I3804836836af714fd9ba3138518d5c78832fc4d5
Mark Young [Thu, 11 Aug 2016 14:46:55 +0000 (08:46 -0600)]
spirv-headers : Add file to restrict to commit
Restrict spirv-headers to a specific commit (just as we do to
spirv-tools and glslang) by creating a new file
(spirv-headers_revision) to track the commit we want to
syncrhonize to. Also update the update_external_sources scripts
to use it. This will make sure that we don't run into the
scenario where spirv-tools and spirv-headers are incompatible.
Change-Id: Ia4fdd921a5c1c500d7f751db08b6998b3d704f2d
Mark Lobodzinski [Thu, 11 Aug 2016 19:25:38 +0000 (13:25 -0600)]
layers: Update PV variable names, clang-format
Change-Id: I1c1da95bbfa1dce31b2b109ec7dea0c770af44c9
Mark Lobodzinski [Tue, 9 Aug 2016 16:44:38 +0000 (10:44 -0600)]
layers: Add WSI support to parameter_validation
Change-Id: Ieb52e121632c6297eee5e37baeade5f0c8f8f3e5
Mark Lobodzinski [Tue, 9 Aug 2016 19:41:09 +0000 (13:41 -0600)]
layers: Validate Enabled Features in PV
ParameterValidation was using the hardware supported feature set for
validation -- these are now replaced with the application-enabled
features at CreateDevice time.
Change-Id: If52d3ae3e55c69611e02437d33a651d88ba7043d
Lenny Komow [Thu, 11 Aug 2016 17:23:15 +0000 (11:23 -0600)]
misc: Add vulkan C++ header
Lenny Komow [Thu, 11 Aug 2016 16:57:38 +0000 (10:57 -0600)]
demos: Fix useless NULL check in cube and tri
Tobin Ehlis [Thu, 11 Aug 2016 13:43:03 +0000 (07:43 -0600)]
layers: Refactor function names
Capitalize first letter of Validate[Image|Buffer]UsageFlags() functions
Tobin Ehlis [Thu, 11 Aug 2016 13:16:53 +0000 (07:16 -0600)]
layers: Fixes GH846 Use IMAGE_USAGE bits for images
Cut-and-paste bug. Various checks for images usage flags were being made
against BUFFER_USAGE enums instead of IMAGE_USAGE enums
Mark Lobodzinski [Wed, 10 Aug 2016 19:12:23 +0000 (13:12 -0600)]
layers: Update details doc w/new queue family test
Added MismatchedQueueFamiliesOnSubmit() test for validating
DRAWSTATE_INVALID_QUEUE_FAMILY errors.
Change-Id: I86ff4da21840a679e755790ae6c7810a0f752028
Mark Lobodzinski [Tue, 9 Aug 2016 22:42:24 +0000 (16:42 -0600)]
layers: GH790, Validate queue family at submit-time
Command buffers must come from a command pool created with the same
queue family index as the queue they are submitted on.
Change-Id: Ifd1d73a09643c11852d807cb7edfc881827005dd
Chris Forbes [Tue, 9 Aug 2016 22:41:27 +0000 (10:41 +1200)]
layers: Adjust color/input attachment layout checks style
For consistency with depth
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Tony Barbour [Wed, 10 Aug 2016 19:39:36 +0000 (13:39 -0600)]
demos: Fix operator precedence problem with window width
Change-Id: Ief0b260528a6eb38788f88ec8a463307522849e8
Cody Northrop [Fri, 5 Aug 2016 20:21:21 +0000 (14:21 -0600)]
toolchain: Fix SPIRV-Headers update script
Without specifying a branch, git pull fails.
Tobin Ehlis [Thu, 4 Aug 2016 16:17:19 +0000 (10:17 -0600)]
layers: Fix MapMem image layout validation
When mapping memory, check all of the bound image ranges and for any
that overlap with the map range, make sure that the image layout is
either PREINITIALIZED or GENERAL.
Tobin Ehlis [Thu, 4 Aug 2016 15:33:02 +0000 (09:33 -0600)]
layers: Rename validateMemRange()
This function is only used in MapMemory case so renamed to
ValidateMapMemRange() and added a header comment to make use
more clear.
Tobin Ehlis [Thu, 4 Aug 2016 14:03:32 +0000 (08:03 -0600)]
layers: Rename DEVICE_MEM_INFO struct members
Update names from camelcase to lowercase w/ "_" as we're slowly migrating
to that uniform style
Tobin Ehlis [Thu, 4 Aug 2016 13:53:46 +0000 (07:53 -0600)]
layers: Remove image member from DEVICE_MEM_INFO
The image member is broken and was superceded by objBindings which track
all objects bound to a single memory allocation.
This is also temporarily disabling a check for image layout when mapping
memory, but that check was broken and will be fixed in subsequent CL.
Chris Forbes [Mon, 8 Aug 2016 20:17:14 +0000 (08:17 +1200)]
layers: Account for read-only depthstencil attachments in ValidateLayout
The depth stencil attachment can be DEPTH_STENCIL_READ_ONLY_OPTIMAL.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Mark Lobodzinski [Fri, 5 Aug 2016 18:05:16 +0000 (12:05 -0600)]
layers: Update validation layer details doc
Added new image test for undefined format in CreateRenderPass
attachment description and updated notes for missing tests.
Change-Id: I6a928213da8dc86d59a33a704cbbec8bd3cfbd45
Mark Lobodzinski [Fri, 5 Aug 2016 17:30:15 +0000 (11:30 -0600)]
layers: LX578, Add CreateRenderPass format check
Validate that AttachmentDescription.Format is not UNDEFINED.
Change-Id: Id348905aadfdc552c3b674a5752cdbc350022d8e
Mark Lobodzinski [Fri, 5 Aug 2016 17:20:21 +0000 (11:20 -0600)]
layers: Add minor Image::CreateRenderPass updates
Combined some repeated for-loops, code standards fixups, etc., in
the image layer.
Change-Id: I3952691d392898ce3c8fa7cc0a6762a40d06220d
Tony Barbour [Mon, 8 Aug 2016 17:04:17 +0000 (11:04 -0600)]
demos: Fix ifdefs to allow compilation without XLIB
Change-Id: I3fbed70e842bf327437d77f56cd0fa7848ab5d19
Tobin Ehlis [Tue, 19 Jul 2016 19:25:12 +0000 (13:25 -0600)]
tests: Add ColorBlendLogicOpTests
This is a dual-purpose test. If the underlying device supports logicOp,
then attempt to use an invalid logicOp and verify that an error is flagged.
If the device doesn't support logicOp, attempt to use it and verify that
the correct error is flagged.
Also update documentation.
Tobin Ehlis [Tue, 19 Jul 2016 17:22:29 +0000 (11:22 -0600)]
tests: Add IndexBufferNotBound test
Attempt and indexed draw with no index buffer bound.
Tobin Ehlis [Tue, 19 Jul 2016 17:20:24 +0000 (11:20 -0600)]
layers: Exclude index buffer from CBSTATUS_ALL mask
The CBSTATUS_ALL mask is set to default all dynamic state to static in the
PSO. The index buffer doesn't fall under the other dynamic state umbrella
so excluding it from the CBSTATUS_ALL mask.
Tobin Ehlis [Tue, 19 Jul 2016 16:50:30 +0000 (10:50 -0600)]
tests: Add FramebufferIncompatible test
Test where secondary command buffer has different framebuffer than
primary command buffer, which is a violation.
Update documentation.
Tobin Ehlis [Tue, 19 Jul 2016 16:45:24 +0000 (10:45 -0600)]
layers: Remove duplicate framebuffer incompatible check
The existing DRAWSTATE_FRAMEBUFFER_INCOMPATIBLE check was a duplicate
of a check performed a few lines earlier in the validateFramebuffer()
function.
Update the check in that function to use the DRAWSTATE_FRAMEBUFFER_INCOMPATIBLE
enum and remove the duplicate check.
Tony Barbour [Fri, 5 Aug 2016 20:55:20 +0000 (14:55 -0600)]
demos: Fix cube to handle >1 event per frame
Change-Id: I6532f75f285b7428993da9314e128bfe5e476829
Cody Northrop [Fri, 5 Aug 2016 18:30:44 +0000 (12:30 -0600)]
demos: Fix Android build on OSX and Windows
Mike Stroyan [Thu, 4 Aug 2016 17:20:14 +0000 (11:20 -0600)]
demos: Flip cube texture on android
Flip the cube texture on android to track commit cb7f374.
Recreate lunarg.ppm.h from lunarg.ppm using xxd.
xxd -g 1 -c 12 -i lunarg.ppm > android/include/lunarg.ppm.h
Mark Lobodzinski [Fri, 5 Aug 2016 15:38:18 +0000 (09:38 -0600)]
layers: Reenable fix for DS layout transitions
Fix for GH650, plus added fix for cube/tri.
Change-Id: I66c2332c30eb34a0f022c033efd9b1c5b8284c32
Ian Elliott [Fri, 5 Aug 2016 15:08:30 +0000 (09:08 -0600)]
GH 818: Improve swapchain error for vkCreateSwapchainKHR()
When the the application calls vkCreateSwapchainKHR(), the
pCreateInfo->surface used must have been previously checked as supported,
by calling vkGetPhysicalDeviceSurfaceSupportKHR(). The error message needed
to be more clear about this.
davidhubbard [Thu, 21 Jul 2016 22:11:59 +0000 (15:11 -0700)]
build: pr787 - CMAKE_INSTALL_PREFIX default
CMAKE_INSTALL_PREFIX default is /usr/local on unix
In order to build with a custom CMAKE_INSTALL_PREFIX, the line
set (CMAKE_INSTALL_PREFIX "") must be removed.
CMake set will overwrite anything specified on the command
line with cmake -DCMAKE_INSTALL_PREFIX.
Change-Id: I3bc905b4ad8fb08917e500e4987eb0277ba74a40
Lenny Komow [Thu, 4 Aug 2016 16:32:27 +0000 (10:32 -0600)]
winsdk: Make installer version programmable
Tobin Ehlis [Wed, 3 Aug 2016 22:51:43 +0000 (16:51 -0600)]
layers: Update string concatenation in struct_string_helper_cpp
For structs with many fields, the generation of the final string representation
created by vk_struct_string_helper_cpp.h was getting obscene. The final string
concatentation for VkPhysicalDeviceFeatures exceeded 3000 cols and was crashing
clang.
This change builds the final string in a number of steps rather than trying to
do it all on a single line.
John Drinkwater [Mon, 1 Aug 2016 16:00:00 +0000 (17:00 +0100)]
loader: support manifests in XDG locations
Some users set the environment variable XDG_DATA_HOME to relocate
~/.local/share/ to a more friendly location and so it is not wise to
look solely at that location.
This change adds basic support for that env var without changing the
behaviour for typical users. If XDG_DATA_HOME is defined, the loader
checks XDG_DATA_HOME/vulkan/..., if HOME is defined the loader will
check HOME/.local/share/vulkan/..., and with neither it skips looking
for user manifests.
loader_get_manifest_files() now takes a relative location inside the
config path eg `/vulkan/implicit_layer.d` and appends that onto the
preferred location.
vk_loader_platform.h Linux defines got minor adjustments to support this
change, with an added bonus of being more readable.
Change-Id: I44e8a99f2a984f94dc33cdda122e6c417d3a3653
Chris Forbes [Thu, 4 Aug 2016 03:59:38 +0000 (15:59 +1200)]
layers: Fix function interception list for param validation
A bunch of things were missing here.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Tobin Ehlis [Thu, 4 Aug 2016 02:30:07 +0000 (20:30 -0600)]
layers: Disable previous fix for DS layout transitions
Fix to always transition both depth and stencil aspects for DS images
is causing errors on cube and tri. Disabling the fix and test for now
to avoid unwanted false positives.
Mark Lobodzinski [Mon, 1 Aug 2016 21:51:39 +0000 (15:51 -0600)]
layers: GH650, Fix D/S image layout validation
When an imageView of a depth/stencil image is used as a depth/stencil
framebuffer attachment, the aspectMask is ignored and both depth and
stencil image subresources are used. Validation was throwing errors
as only the specified aspectMask transition was tracked.
Change-Id: I4f45eec74040448ba3ed52b07268fab3b3cb9318
Tobin Ehlis [Wed, 3 Aug 2016 15:59:17 +0000 (09:59 -0600)]
layers: Check vtx buffer binding
Need to check vtx buffer binding based on binding values set in
pVertexBindingDescriptions array if present.
Ian Elliott [Wed, 3 Aug 2016 20:57:11 +0000 (14:57 -0600)]
Fix Android build for mips64: NULL -> VK_NULL_HANDLE
This should affect other architectures as well. A handle can't be assigned to
NULL (may get 32-64 type issues). Must set a Vulkan handle to VK_NULL_HANDLE.
Tony Barbour [Thu, 28 Jul 2016 18:01:45 +0000 (12:01 -0600)]
demos: Remove the need for QueueWaitIdle in cube
Change-Id: Iddccf0d672632ad7f0821fb98c685c944c2b8b77
Tony Barbour [Thu, 28 Jul 2016 16:50:13 +0000 (10:50 -0600)]
demos: Use renderpass finalLayout to get to PRESENT_SRC_KHR
instead of an explicit memory barrier
Change-Id: I9c9c19ee6640fc696576d67ffc608b9205d84166
Tobin Ehlis [Fri, 22 Jul 2016 22:55:19 +0000 (16:55 -0600)]
layers: Fix bug in cmd buffer framebuffer binding
If cmd buffers were freed via DestroyCommandPool, then the framebuffer links to the
individual cb_nodes were not getting cleared. Fix that issue.
This is initial, simple solution to avoid crash, need smarter solution long-term.
Cody Northrop [Mon, 1 Aug 2016 16:27:25 +0000 (10:27 -0600)]
toolchain: Update revisions for Android
Also add spirv-headers, a new dependency for shaderc.
Tobin Ehlis [Tue, 2 Aug 2016 16:50:29 +0000 (10:50 -0600)]
layers: Fix object_tracker build
This moves a static array of maps to be a vector of maps.
Building android from Windows crashes clang without this change.
clang in the NDK doesn't like arrays of unordered_maps.
Mark Lobodzinski [Tue, 2 Aug 2016 15:30:28 +0000 (09:30 -0600)]
layers: Remove debug comments from OT.h
Change-Id: I1cf485ddb54223b873746ff218f0f34085a87450
Chris Forbes [Mon, 1 Aug 2016 20:08:16 +0000 (08:08 +1200)]
layers: Fix #808 broken mapping of image types to view types
This was trying to be too clever, and was foiled by bad assumptions
about the enum order.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Mark Young [Mon, 1 Aug 2016 17:06:53 +0000 (11:06 -0600)]
layers: Generate full vk_dispatch_table_helper
Generate the full vk_dispatch_table-helper file for all platforms
as part of the build process now. This includes all WSI
extension entry-points for all build targets we support increasing
portablility of the header.
Change-Id: Ie83423e7ea169f7251c5369992a6144a97db1ddc
Johannes van Waveren [Fri, 22 Jul 2016 21:10:11 +0000 (16:10 -0500)]
layers: gh793 gen vk_dispatch_table_helper.h
Generate one vk_dispatch_table_helper.h for all platforms.
Change-Id: I17fa435bc69cacfdd75dd93fbb0a5a37222aaa2b
Mark Young [Mon, 1 Aug 2016 17:34:36 +0000 (11:34 -0600)]
loader : Fix segfault in vulkaninfo
Last commit introduced a segfault in VulkanInfo. This corrects it.
Change-Id: Ie4115bbe24191bdad9f539e5d9b1658982417f51
davidhubbard [Thu, 28 Jul 2016 00:59:58 +0000 (17:59 -0700)]
loader: gh571 vkEnumerateInstanceLayerProperties()
This fixed a memory leak.
Steps to reproduce:
Run valgrind --leak_check=full on a this test file:
```
int main() {
uint32_t layerCount = 0;
vkEnumerateInstanceLayerProperties(&layerCount, nullptr);
printf("layerCount=%u\n", layerCount);
return 0;
}
```
valgrind output will look like:
```
==PID==
==PID== 49,920 bytes in 6 blocks are definitely lost in loss record 57 of 58
==PID== at 0x40307C4: malloc (vg_replace_malloc.c:270)
==PID== by 0x4A2B5F3: loader_instance_heap_alloc (loader/loader.c:178)
==PID== by 0x4A2CB0D: loader_init_generic_list (loader/loader.c:858)
==PID== by 0x4A2CBDE: loader_add_to_ext_list (loader/loader.c:889)
==PID== by 0x4A30A32: loader_read_json_layer (loader/loader.c:2311)
==PID== by 0x4A31373: loader_add_layer_properties (loader/loader.c:2509)
==PID== by 0x4A3286F: loader_layer_scan (loader/loader.c:3032)
==PID== by 0x4A38DFC: vkEnumerateInstanceLayerProperties (loader/trampoline.c:239)
==PID==
```
The fix is simple: vkEnumerateInstanceLayerProperties() in loader/trampoline.c should call loader_delete_layer_properties(), not loader_destroy_layer_list() in both places it frees resources associated with the instance_layer_list.
Change-Id: I3275195dd27703c2747971288327d546efe5ed25
Chris Forbes [Fri, 24 Jun 2016 05:04:54 +0000 (17:04 +1200)]
layers: Add some TODOs for weird code that should go away
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Thu, 23 Jun 2016 05:03:55 +0000 (17:03 +1200)]
layers: Drop QUEUE_NODE::device
This was unused.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Thu, 28 Jul 2016 02:15:38 +0000 (14:15 +1200)]
layers: Track valid dynamic scissor and viewport indices with masks
CmdSetViewport/CmdSetScissor were broken; they assumed the update
replaced all the viewports/scissors; further, the drawtime validation
was too aggressive -- nowhere does the spec require the number of valid
dynamic viewports/ dynamic scissors to be equal to the number used by
the PSO; just that all viewports/scissors used by the PSO must be
provided.
V2: Add more parens to quiet warnings in android build
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Tue, 26 Jul 2016 05:06:10 +0000 (17:06 +1200)]
layers: Validate image view type requirement for descriptor sets
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Mon, 25 Jul 2016 06:12:05 +0000 (18:12 +1200)]
layers: Generate appropriate sets of requirement flags for OpTypeImage
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Mon, 25 Jul 2016 06:10:41 +0000 (18:10 +1200)]
layers: Add plumbing for descriptor requirements
We want to be able to impose requirements on images & views bound to
descriptor sets based on their usage in the shader. This adds the
plumbing from pipelines into descriptor sets to enable that.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Mon, 25 Jul 2016 06:11:50 +0000 (18:11 +1200)]
layers: Add enum for descriptor requirements
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Rene Lindsay [Fri, 29 Jul 2016 17:49:11 +0000 (10:49 -0700)]
demos: Flip cube texture and TC's, so image is no longer inverted
Rene Lindsay [Fri, 29 Jul 2016 17:43:47 +0000 (10:43 -0700)]
demos: Remove unused function parameter.
Mark Young [Fri, 22 Jul 2016 14:51:05 +0000 (08:51 -0600)]
docs: Clean up LoaderAndLayerInterface markdown
Perform some wordsmithing and general clean-up of the
LoaderAndLayerInterface markdown file.
Change-Id: I286d0f1d01241b376ba2f4dfa97d2329b04d515d
Michael Lentine [Thu, 21 Jul 2016 22:24:56 +0000 (17:24 -0500)]
layers: GH722 Add query function calls from secondary buffers.
Cody Northrop [Wed, 13 Jul 2016 23:24:59 +0000 (17:24 -0600)]
tests: Create APK version of layer validation tests
Tony Barbour [Mon, 18 Jul 2016 19:21:06 +0000 (13:21 -0600)]
demos: Fix cube to handle seperate gfx / present queues
Change-Id: I92497bb966aa6d35746184249e921155ed26d67e
Jeremy Hayes [Wed, 6 Jul 2016 18:02:03 +0000 (12:02 -0600)]
loader: return incomplete when count < available
The list was being destroyed before the check.
Change-Id: I500a22c8e2fffa8c852235f688026cd80f84088d
Cody Northrop [Tue, 26 Jul 2016 20:25:45 +0000 (14:25 -0600)]
android: Lift global offset table size limit (-mxgot)
Currently the mips64 build for Android is broken due to size
increases of core_validation. If you disabled optimizations (-O0),
you'll get the following error:
relocation truncated to fit: R_MIPS_CALL16
The linker flag to lift the limit is -mxgot, documented here:
https://gcc.gnu.org/onlinedocs/gcc-4.0.0/gcc/MIPS-Options.html
It is a slight de-optimization for global symbol lookups (3 inst
instead of 1). Hopefully we aren't using global lookups in
performance critical code.
Tony Barbour [Tue, 26 Jul 2016 16:17:28 +0000 (10:17 -0600)]
Tools: Bump spirv-tools revision to include fix
Covers fix for SPIRV-Tools issue #270 - access violation in
Vulkan conformance test
Change-Id: Idc3470f9ba68d9189d3536fe776459ce2ccba388
Mark Lobodzinski [Fri, 22 Jul 2016 21:30:27 +0000 (15:30 -0600)]
layers: Restore OT WSI GPA prefixes
Change-Id: Ia83e40c8c5c92a8ac50dc2d5bf5bf9b2b255594b
Tobin Ehlis [Thu, 21 Jul 2016 20:48:21 +0000 (14:48 -0600)]
layers: Fix copy-paste "DEPTH" instead of "STENCIL"
Tobin Ehlis [Tue, 19 Jul 2016 23:39:34 +0000 (17:39 -0600)]
layers: Fix image aspect checks
Previously validation required DS images to have both DEPTH and STENCIL
aspect bits set, but spec only requires that at least one is set.
Also, we weren't checking case of depth or stencil only formats, or
color format to make sure that correct aspect mask was set.
Mark Lobodzinski [Thu, 21 Jul 2016 15:13:18 +0000 (09:13 -0600)]
tests: Update doc validation script
Changed object_tracker location, status.
Change-Id: I2b9834002cdeb8a3b48e83d8da6f727edfd2ffdd
Mark Lobodzinski [Thu, 21 Jul 2016 15:09:35 +0000 (09:09 -0600)]
layers: Update codegen for OT removal
Removed object_tracker code from vk_layer_generate.py.
Change-Id: I0ebd1c379deabfb0712d6692751145f34b430459
Mark Lobodzinski [Mon, 18 Jul 2016 23:06:52 +0000 (17:06 -0600)]
layers: Fixup OT error messages
Print object type explicitly; also fixup a couple of format
specifiers.
Change-Id: I6f5036f70cae05bd5f7ee9320e24b2e16bdc811b
Mark Lobodzinski [Wed, 13 Jul 2016 17:29:00 +0000 (11:29 -0600)]
layers: Fix thread-safety fails in OT FreeXxx APIs
Tracking data was cleaned up after calling down the chain in
FreeCommandBuffers and FreeDescriptorSets -- this left a hole that
another thread could walk through and corrupt the tracking data.
Change-Id: I7cc9c9bfae9f1ea179122125dd09952650f14d81
Mark Lobodzinski [Fri, 1 Jul 2016 16:53:31 +0000 (10:53 -0600)]
layers: Update object_tracker layer
Removed dead code, added per-instance and per-device object tracking,
(this was broken in codegen'd version with no cross-device
validation), updated for new layer architectures and coding standards,
removed OT-related codegen -- it is now a standalone cpp file.
Change-Id: I64464b855e1b4841c8e3a581387e0e9065b006f7
Mark Young [Tue, 5 Jul 2016 22:41:50 +0000 (16:41 -0600)]
layers : gh682 move gipa dispatch table init
Move the GIPA setup of the dispatch tables for layers into
the layer_init_device_dispatch_table call since we're already
doing a majority of it there. This removes the need to separately
setup the WSI extension entry-points.
Additionally, memset the table to 0, just to make sure anything
that gets added in the future is obvoiusly not set.
Change-Id: I63c7c107cd9f7957e2766fdbeb8a06bc1ae0eda6
Mark Young [Thu, 21 Jul 2016 15:45:32 +0000 (09:45 -0600)]
docs: Remove extra line in doc headers
GitHub markdown adds a line after H1 and H2 entries. Normal
markdown doesn't do this. So my WSYIWYG editor incorrectly showed
the results of my previous edits.
Change-Id: I3c197c1fb1f364bf4b1050d1eaa1141affefe9d1