Cody Northrop [Thu, 7 Apr 2016 17:22:11 +0000 (11:22 -0600)]
android: Fix typo in Mac steps
guanghuafan [Thu, 31 Mar 2016 22:31:02 +0000 (15:31 -0700)]
android: minor indentation correction in Android.mk
guanghuafan [Thu, 31 Mar 2016 20:02:11 +0000 (13:02 -0700)]
android: Minor clean up for windows build template
guanghuafan [Thu, 31 Mar 2016 19:02:44 +0000 (12:02 -0700)]
android: fixed issues in windows build
tested: windows/mac
guanghuafan [Thu, 31 Mar 2016 14:49:00 +0000 (07:49 -0700)]
android: set hard float for arm7 correctly and minor cleanup
tested: Mac X
guanghuafan [Thu, 31 Mar 2016 04:16:13 +0000 (21:16 -0700)]
android: Changed android-generated.bat for windows (not tested )
guanghuafan [Tue, 29 Mar 2016 18:08:23 +0000 (11:08 -0700)]
android: Adding template for gradle wrapper build
Tobin Ehlis [Thu, 7 Apr 2016 17:35:46 +0000 (11:35 -0600)]
layers: GH241 vkSetEvent should affect queue stageMasks
vkSetEvent may occur asynchronously from cmd buffer execution
and the results of vkSetEvent should be immediately visible
across all queues of a device.
This fix will update any event that a queue has seen so that
its VK_PIPELINE_STAGE_HOST_BIT is set at vkSetEvent() time.
Jon Ashburn [Thu, 7 Apr 2016 17:57:03 +0000 (11:57 -0600)]
loader: Fix the std_validation version to really be loader's built version
Change-Id: I2ff3b6dab4209b152f487649db011cdfa7c8975b
David Pinedo [Thu, 7 Apr 2016 17:28:30 +0000 (11:28 -0600)]
winrtinstaller: re-sign powershell script
Karl Schultz [Thu, 7 Apr 2016 16:52:43 +0000 (10:52 -0600)]
layers: INVALID_ATTACH_COUNT -> INVALID_ATTACHMENT_COUNT
Fix error reported by vkvalidatelayerdoc test.
Change-Id: I7e1bb4e5cb8c113202d479ef95cc0d37334040b1
Chris Forbes [Wed, 6 Apr 2016 21:04:49 +0000 (09:04 +1200)]
loader: Preserve order during layer list expansion
As requested by Jon -- instead of appending the expanded layers, insert
them in place of the meta layer.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Wed, 6 Apr 2016 08:49:02 +0000 (20:49 +1200)]
loader: Don't scribble on caller memory in CreateInstance,CreateDevice
expand_... / unexpand_... scribbled on both the CreateInfo struct and
the list of layer strings, and then unscribbled them on the way back
out. This is a lousy thing to do, and just blows up if the memory isn't
writable (which it needn't be, given the API takes ptrs to const).
Instead, copy the *CreateInfo into a shadow struct on the stack, and be
careful in expand_layer_names never to scribble on the caller's layer
names array.
V2: slight tweak (missed initializer)
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Karl Schultz [Thu, 7 Apr 2016 15:40:30 +0000 (09:40 -0600)]
demos: Add missing spaces in error messages.
Change-Id: Ic9b2b2a2ea14bb087d74f6c0b697449b141899a4
Ian Elliott [Thu, 7 Apr 2016 15:05:45 +0000 (09:05 -0600)]
MR 289: layers: Changed other text/variables from "owned" to "acquired".
Ian Elliott [Wed, 6 Apr 2016 20:29:56 +0000 (14:29 -0600)]
MR 289: layers: Change the test about "too many images".
Given new spec language coming out in early April (clarifying various aspects
of acquiring presentable images), a change was needed to the WSI-swapchain
validation layer. The new validate-layer test will work for "BlitModel"
drivers that advertise a minImageCount of 1. Failing the test now results in
an error instead of a performance warning.
Jon Ashburn [Thu, 7 Apr 2016 13:17:22 +0000 (07:17 -0600)]
loader: temp workaround for advertising linux wsi surface extensions
Change-Id: I9a03c6ea16abe9e362ef81ee87ac217b8b3c6c92
Chris Forbes [Thu, 7 Apr 2016 01:21:07 +0000 (13:21 +1200)]
tests: Add helpers for verifying error monitor state
Open-coded verify was often not right:
- Actual error message would evolve and failure message would not get
updated.
- Dumping of other errors would be attempted /after/ FAIL(), which
throws, and so would never be reached.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Thu, 7 Apr 2016 01:20:10 +0000 (13:20 +1200)]
tests: Bump shader version to 450 in remaining cases
Allows us to not have 2 #extension lines in every shader.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Michael Lentine [Wed, 6 Apr 2016 22:40:22 +0000 (17:40 -0500)]
layers: Don't exceed max allowed color attachments.
Tony Barbour [Thu, 7 Apr 2016 00:17:57 +0000 (18:17 -0600)]
layers: Move call to validateCommandBuffersNotInUse inside lock
Fixes segv in CTS multithreaded command pool test
Change-Id: I80da12fd2bb6856e3c6824f17dd92f1597121a54
Jon Ashburn [Mon, 4 Apr 2016 22:37:37 +0000 (16:37 -0600)]
loader: Update specification doc for dispatchable object initialization
Change-Id: I80770b7ab14795316400279dc4e7c954fa5ad374
Dustin Graves [Wed, 6 Apr 2016 22:47:10 +0000 (16:47 -0600)]
layers: Fix 32-bit Windows build
Change reinterpret_cast<uint64_t> to reinterpret_cast<uint64_t&>.
Change-Id: I87b8479df6ba7f93b922bab11edbbf6c4fbfe7f6
Michael Lentine [Wed, 6 Apr 2016 18:15:46 +0000 (13:15 -0500)]
layers: Ensure bind point for subass is graphics.
Dustin Graves [Wed, 6 Apr 2016 17:06:19 +0000 (11:06 -0600)]
layers: Fix MSVS value truncation warning
Changed the type of the string validation constants to from char to
uint8_t to fix warning C4309: 'initializing' : truncation of constant
value, due to an implicit unsgined char to signed char conversion.
Change-Id: I86cca4b1a8c3f818a2e7a9a7e38f05ee8cd01150
Dustin Graves [Wed, 6 Apr 2016 16:16:05 +0000 (10:16 -0600)]
demos/tests: Fix MSVS type warnings
Fixes the following MSVS warnings:
- C4800 - forcing value to bool 'true' or 'false'
- C4244 - conversion from type1 to type2, possible loss of data
- C4267 - conversion from size_t to type, possible loss of data
Change-Id: I78576c48c5d42e4c9116a83c5f4a4f4a5035e0fc
Chris Forbes [Wed, 6 Apr 2016 03:16:26 +0000 (15:16 +1200)]
layers: Remove plumbing through of unused dispatchable object through MT
Much of this was eventually going to go away anyway, but getting rid of
this thread of it was trivial.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Tobin Ehlis [Tue, 5 Apr 2016 19:33:00 +0000 (13:33 -0600)]
layers: Merge binding maps into image and buffer maps
mem_tracker layer had specific maps to track memory binding for images
and buffers. This change merges the data from those maps into IMAGE_NODE
and BUFFER_NODE respectively and eliminates the binding maps.
Vinjn Zhang [Wed, 6 Apr 2016 07:38:24 +0000 (15:38 +0800)]
Update vk_layer.h
Correct comment for VkLayer*CreateInfo's sType
Piers Daniell [Tue, 5 Apr 2016 23:28:06 +0000 (17:28 -0600)]
loader: Fix regression with dEQP-VK.api.object_management.alloc_callback_fail.device
When the loader vkCreateDevice trampoline code calls down to
vkCreateDevice don't update the dev->device field with a bogus device
handle.
Change-Id: I703acb5283650ab0013483bda0f478da9b46e3ea
Chris Forbes [Tue, 5 Apr 2016 23:21:28 +0000 (11:21 +1200)]
layers: Insist on each shader's stage being unique in a pipeline
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Tue, 5 Apr 2016 23:18:47 +0000 (11:18 +1200)]
tests: Add test for multiple shaders for same stage
This is forbidden.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Michael Lentine [Tue, 5 Apr 2016 16:38:12 +0000 (11:38 -0500)]
Validate attachment index is in range.
Chris Forbes [Tue, 5 Apr 2016 05:51:35 +0000 (17:51 +1200)]
layers: Implement relaxed vec-vec and vec-scalar matching rules
- Track whether we've seen anything disqualifying -- non-stripped
arrays, matrices, structures, etc.
- Add relaxed vec/scalar and vec/vec matches if not disqualified.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Tue, 5 Apr 2016 05:50:35 +0000 (17:50 +1200)]
layers: Add test for relaxed type match
A positive test which matches:
vec3 -> float
ivec3 -> int
vec3 -> vec2
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Dustin Graves [Tue, 5 Apr 2016 21:15:40 +0000 (15:15 -0600)]
layers: Explicit VK_TRUE/VK_FALSE checks
Add explicit VK_FALSE/VK_TRUE conditional checks.
Change-Id: Idb5c82ce53d54c9f2c378dc557152efc695ae2b1
Dustin Graves [Tue, 5 Apr 2016 19:48:15 +0000 (13:48 -0600)]
layers: Fix unsafe bool mix warnings (part 2)
Fixes the MSVS warnings regarding unsafe mix of VkBool32 and bool for the
device_limits, image, parameter_validation, and threading layers.
Converts all Boolean usage internal to the layers to bool/true/false. The
VK_TRUE and VK_FALSE values continue to be used with the Vulkan API
components.
Change-Id: I612d3f26050fadcd77ffca4d8723a7c734613816
Dustin Graves [Tue, 5 Apr 2016 15:41:17 +0000 (09:41 -0600)]
layers: Fix unsafe bool mix warnings
Fixes the MSVS warnings regarding unsafe mix of VkBool32 and bool for the
core_validation layer. Converts all Boolean values internal to the layer
to bool/true/false. The VK_TRUE and VK_FALSE values continue to be used
with the Vulkan API components.
Change-Id: I9a298e2a9f5f92a388a2b7c158451d237fdb4fc0
Karl Schultz [Wed, 6 Apr 2016 13:18:42 +0000 (07:18 -0600)]
loader: remove unused instance_dispatch.png
Change-Id: I67ebce2e4e46a5a1439bb35fb966ed522c80ee52
David Pinedo [Mon, 4 Apr 2016 17:43:50 +0000 (11:43 -0600)]
winrtinstaller: sort in config script: <prerelease> is least significant
Also added execute permission to CreateInstallerRT.sh
David Pinedo [Tue, 5 Apr 2016 22:40:08 +0000 (16:40 -0600)]
winrtinstaller: fix install dir not being removed on uninstall.
This problem was introduced when logging was added.
Jeremy Hayes [Tue, 5 Apr 2016 21:46:17 +0000 (15:46 -0600)]
loader: bump linux version
Change-Id: I6c9e9b7650a7599797e93e60eb2191345c3e6ec8
Jon Ashburn [Mon, 4 Apr 2016 22:55:24 +0000 (16:55 -0600)]
loader: Temp workaround to add WSI surface extensions enumerated.
Change-Id: Ibb9ffa80bbb71d15296e39730f48f33c23bd331a
Jamie Madill [Mon, 4 Apr 2016 16:15:39 +0000 (12:15 -0400)]
layers: Fix return values of bool functions.
These functions were returning VK_ERROR_VALIDATION_FAILED_EXT,
which was giving a warning in Visual Studio. The app shouldn't be
checking for non-bool return values of these functions.
Change-Id: I9924a25b1c49ad6973c8915be1dd7aea4cc0edf4
Jamie Madill [Mon, 4 Apr 2016 15:54:43 +0000 (11:54 -0400)]
layers: Change include order of loader platform.
This fixes a bug where on Windows, windows.h would include
winsock.h before winsock2.h, causing numerous warnings.
Change-Id: I8eba1497385dcee6ebb201c74fe48268e6663b9c
Jamie Madill [Mon, 4 Apr 2016 19:09:51 +0000 (15:09 -0400)]
Fix 4 MSVS warnings in loader and layers.
warning C4267: conversion, possible loss of data
warning C4389: signed/unsigned mismatch
warning C4701: potentially uninitialized local variable used
warning C4702: unreachable code
Change-Id: Ibc2ce3825a986c3701f6e0e6244a0e30fbf7e87c
Tobin Ehlis [Mon, 4 Apr 2016 16:00:21 +0000 (10:00 -0600)]
layers: Remove incorrect check for colorblend & subpass attachment count match
GH256 has details on why this check is incorrect as-is. Removing the check here and a
fix will be addressed through GH156. Also disabling test for this case until the
correct check is added back in.
Tobin Ehlis [Fri, 1 Apr 2016 19:51:33 +0000 (13:51 -0600)]
layers: Clean up PIPELINE_NODE using safe_struct
PIPELINE_NODE was storing ptrs from the app and has some ugly initialization code
for the gfx pipeline create struct. This commit switches pipeline node to use
safe_struct wrappers for Gfx and Compute create info.
There were also some bugs found in the safe_struct wrappers making this migration
so those are cleaned up in this commit as well.
Jon Ashburn [Mon, 4 Apr 2016 19:52:53 +0000 (13:52 -0600)]
loader: ghlvl #204, Use __secure_getenv if secure_getenv is unavailable
Also add group id check to getenv suid check in loader.
Change-Id: Icbc08258498f893ee5fce144c043bdc6bd8e5423
Michael Mc Donnell [Mon, 4 Apr 2016 09:11:41 +0000 (02:11 -0700)]
tests: Initialize fields so test does not crash.
I had forgotten to define the LD_LIBRARY_PATH and VK_LAYER_PATH
environment variables in VS2015 before running the tests. The
test crashed when it tried to delete the un-initialized pointers.
I have added a default initialization to VK_NULL_HANDLE so you
get an error message instead of a crash.
Change-Id: Ia8e2770914d10dc67be744e418dafe82cc164d67
David Pinedo [Mon, 4 Apr 2016 21:32:49 +0000 (15:32 -0600)]
winrtinstaller: Re-sign powershell loader config script
Michael Mc Donnell [Sun, 3 Apr 2016 21:47:51 +0000 (14:47 -0700)]
layers: Fix 7 signed/unsigned comparison warnings on VS2015
Jamie Madill [Mon, 4 Apr 2016 18:39:53 +0000 (14:39 -0400)]
layers: Fix MSVS shadowing warning in core_validation.
warning 4457: declaration of 'fence' hides function parameter
Jamie Madill [Mon, 4 Apr 2016 18:42:21 +0000 (14:42 -0400)]
layers: Fix MSVS noexcept warning in core_validation.
warning 4577: 'noexcept' used with no exception handling mode specified
Dustin Graves [Mon, 4 Apr 2016 17:14:01 +0000 (11:14 -0600)]
layers: Fix paramcheck struct array NULL indexing
Fixes an issue where the generated parameter_validation code could
index a NULL pointer to an array of structs.
Change-Id: If78f1c97ea256911d857b0b8009b38e6ca6f2581
Tony Barbour [Mon, 4 Apr 2016 17:09:40 +0000 (11:09 -0600)]
Remove NumBlendAttachMismatch until check is in place
Change-Id: Ie4150e84ac4856f8a4e84acbd2c7e8b1ddad832c
Tony Barbour [Mon, 4 Apr 2016 16:59:56 +0000 (10:59 -0600)]
Revert "layers: LX265 Error if color blend attch count != subpass attch."
This reverts commit
dcce8143110568353a6217e5b885da407edd29be.
Change-Id: I0882b499b38e3cf7670846a9ffae48a0fbf015a3
Michael Lentine [Wed, 30 Mar 2016 20:57:52 +0000 (15:57 -0500)]
Fix find layout to search on each aspect bit.
Change-Id: I155ee469ef98d1ad9b1bfc8813a1cff43f30f8ca
Tobin Ehlis [Fri, 1 Apr 2016 00:02:51 +0000 (18:02 -0600)]
layers: GH226 Add check for descriptor slot to make sure it's updated
We have an existing check to make sure that a set has been updated, but there was
still a hole where individual slots may not have an update. This adds an additional
check to make sure that we flag any slots that are used by not updated at draw time.
Jon Ashburn [Mon, 4 Apr 2016 16:53:27 +0000 (10:53 -0600)]
layers: Update to version 1.0.8 JSON files
Change-Id: Ic35eb444c34004343e5fb19cc54f12518512c6be
Jon Ashburn [Mon, 4 Apr 2016 16:47:18 +0000 (10:47 -0600)]
header: update to vers 1.0.8 which had no changes
Change-Id: I809110e8483d1877f5106551d9f066ed6c25d8bc
Jamie Madill [Mon, 4 Apr 2016 14:58:33 +0000 (10:58 -0400)]
Fix def file generation to depend on vulkan.py.
Jamie Madill [Mon, 4 Apr 2016 15:20:24 +0000 (11:20 -0400)]
Allow vk-generate.py to work in Python 2.
There were two places where this script called super, one with
Python 2 and 3 compatible syntax and one with Python 3 only. Fix
it to use the compatible syntax in both locations.
Tobin Ehlis [Wed, 30 Mar 2016 15:32:19 +0000 (09:32 -0600)]
layers: Removed duplicate dynamicOffsets from core_validation
Missed this from a rebase conflict. Kill top-level cmd buffer node
dynamicOffsets vector and only use vector based off of lastBound data.
Chris Forbes [Mon, 4 Apr 2016 07:26:42 +0000 (19:26 +1200)]
tests: skip the tessellation shader tests if feature not available
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Mon, 4 Apr 2016 07:19:47 +0000 (19:19 +1200)]
tests: Add test case for patch decoration mismatch
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Mon, 4 Apr 2016 07:15:00 +0000 (19:15 +1200)]
tests: Add ability to override tessellation state, use it.
This fixes the remaining issue with the tess positive test.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Mon, 4 Apr 2016 07:05:33 +0000 (19:05 +1200)]
tests: Fix test framework to not pass a null vi
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Mon, 4 Apr 2016 06:52:54 +0000 (18:52 +1200)]
layers: Add simple VS/FS positive test, and a tess matching test
We didn't have a positive test which didn't exercise the VI path. This
hid from us that we were actually failing validation due to the test
framework passing a null VICI. This used to be a valid thing to do.
Tess matching test exercises interface matching fix.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Mon, 4 Apr 2016 06:50:38 +0000 (18:50 +1200)]
tests: fix multiple matching messages getting lost
When receiving a subsequent matching message, demote the previous one to
an 'other' message. Otherwise, in tests that should be positive, we lose
the real cause of the failure, due to paramcheck's emitting of an extra
message for "deeper layer returned validation failure".
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Mon, 4 Apr 2016 05:28:05 +0000 (17:28 +1200)]
tests: Bump version in SC tests for 450, drop exts
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Mon, 4 Apr 2016 05:22:42 +0000 (17:22 +1200)]
tests: Enable all supportable features on the device; expose to tests.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Tue, 29 Mar 2016 03:57:02 +0000 (16:57 +1300)]
layers: Flag difference in patch decoration correctly
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Tue, 29 Mar 2016 03:41:07 +0000 (16:41 +1300)]
layers: Strip array level from both sides of TCS->TES interface
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Tue, 29 Mar 2016 03:38:44 +0000 (16:38 +1300)]
layers: Rework array-of-verts stripping a bit to support patch
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Tue, 29 Mar 2016 03:14:02 +0000 (16:14 +1300)]
layers: Collect whether an interface_var has patch decoration
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Eric Engestrom [Sat, 2 Apr 2016 21:09:09 +0000 (22:09 +0100)]
docs: fix spelling mistakes
Eric Engestrom [Sat, 2 Apr 2016 21:08:10 +0000 (22:08 +0100)]
tests: fix spelling mistakes
Eric Engestrom [Sat, 2 Apr 2016 21:06:24 +0000 (22:06 +0100)]
loader: fix spelling mistakes
Eric Engestrom [Sat, 2 Apr 2016 21:06:13 +0000 (22:06 +0100)]
layers: fix spelling mistakes
Mark Lobodzinski [Fri, 1 Apr 2016 21:58:32 +0000 (15:58 -0600)]
layers: Enable safety check for invalid QueryObject
Typo allowed object access for invalid QueryToStateMap objects
Change-Id: I50845ca09564ea8b77173d2165a5318160dd8d7e
Dustin Graves [Fri, 1 Apr 2016 00:01:37 +0000 (18:01 -0600)]
layers: LX414: Fix device_limits alignment check
Fix buffer/descriptor offset alignment checks in device_limits:
- vkUpdateDescriptorSets offset check was not performed with
the correct physical device alignments.
- vkBindBufferMemory offset check was always performed with the
UBO alignment. This check has been moved to core_validation,
where the offset is checked against the alignment associated
with its usage type and the alignment retrieved from
vkGetBufferMemoryRequirements.
Change-Id: I4f64e06c44972a224d1930cb0798a9f472b768f2
Dustin Graves [Thu, 31 Mar 2016 15:50:42 +0000 (09:50 -0600)]
layers: Fix param checker struct array handling
Fix parameter validation issue with arrays of structs, where only
the first item in the array was processed. Validation messages
for arrays of structs will currently print the array variable name
with the letter 'i' as the array index instead of the actual index
number, producing a value like 'pCreateInfos[i]'. This will be
fixed by a future update.
Change-Id: I2cb773c23c6515a20ac8cb066e7cc61028e569f8
Dustin Graves [Fri, 25 Mar 2016 21:13:28 +0000 (15:13 -0600)]
layers: Update param_checker VkResult validation
VkResult validation now prints a description of the error
code being returned.
Change-Id: Ia6dc4e71ef6d1b27365e3c0499b24a3fc1d332ac
Dustin Graves [Thu, 24 Mar 2016 01:44:00 +0000 (19:44 -0600)]
layers: Codegen VkBool32/enum parameter validation
Now generating the VkBool32 and enum checks:
- Warns if a VkBool32 parameter is neither VK_TRUE nor VK_FALSE
- Warns if an enum token value falls outside of the enum's
begin/end range and the token was not added by an extension.
Only checkes enum's that have a begin/end range. Detection
of tokens added by extensions is based on the base_value defined
in appendic C.10 of the Vulkan specification.
Change-Id: Id49ace52fea8a35972f96d54247b98aba182c09e
Dustin Graves [Mon, 28 Mar 2016 22:17:38 +0000 (16:17 -0600)]
layers: 1.0.7 update for XML Registry files
Update local copies of generator.py, genvk.py, and vk.xml with the
latest versions from Vulkan-Docs.
Change-Id: Id2a3da34374fb1c39532cd3c124461eb953f6b89
Dustin Graves [Fri, 1 Apr 2016 21:00:09 +0000 (15:00 -0600)]
layers: Fix VS2013 build
Add macro to enable/disable noexcept based on compiler support.
Solution from:
http://stackoverflow.com/questions/
18387640/how-to-deal-with-noexcept-in-visual-studio
Change-Id: Ic2d22c9247b902d6e13120b17fc0b8647f079f7f
Mark Lobodzinski [Fri, 1 Apr 2016 19:34:08 +0000 (13:34 -0600)]
layers: LX459,GH#99,#100, Fix semaphore reference count
QueueSubmit waitSemaphore refcounts were getting incremented instead of
decremented, resulting in invalid 'semaphore still in use' errors.
Change-Id: I8ac224115b8ee43637b8de4b377750b277cfd22b
Jeremy Hayes [Fri, 1 Apr 2016 17:40:26 +0000 (11:40 -0600)]
loader: report implicit instance extensions
GL74: vkEnumerateInstanceExtensionProperties should report implicit
instance extensions when pLayerName is NULL or an implicit layer name.
Change-Id: Iba7664a168e146c4b862468b6255cdc752cc8c77
Mark Young [Wed, 30 Mar 2016 08:23:18 +0000 (02:23 -0600)]
layers: LX250 Verify color and depth/stencil are different.
Validate that the color and depth/stencil attachments in a
subpass are different.
Change-Id: I0010270e50334cd1dc5b275c29c76265f570b186
Dominik Witczak [Fri, 1 Apr 2016 11:19:49 +0000 (13:19 +0200)]
#238: Fixes the "num samples mismatch" error being shown incorrectly for subpasses with 0 attachments
Chris Forbes [Thu, 31 Mar 2016 05:11:28 +0000 (18:11 +1300)]
layers: Replace GLOBAL_CB_NODE::pMemObjList with unordered_set
One more std::list gone.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Thu, 31 Mar 2016 05:03:56 +0000 (18:03 +1300)]
layers: Get rid of a bunch of linked list copying when retiring CBs
There was no reason to copy this.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Thu, 31 Mar 2016 05:02:29 +0000 (18:02 +1300)]
layers: trivial: remove stray () from end of error message
This wasn't describing a function; the () on the end is just nonsense.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Thu, 31 Mar 2016 04:58:13 +0000 (17:58 +1300)]
layers: Replace DEVICE_MEM_INFO::pCommandBufferBindings with
unordered_set
More linked-list removal.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Thu, 31 Mar 2016 04:37:36 +0000 (17:37 +1300)]
layers: Replace DEVICE_MEM_INFO::pObjBindings with unordered_set
- Add operator== && std::hash specialization required for use in
unordered_set.
- Replace list by unordered_set
- Delete piles of stuff
V2: Fix accidentally inverted error condition
Signed-off-by: Chris Forbes <chrisforbes@google.com>
Chris Forbes [Thu, 31 Mar 2016 04:06:52 +0000 (17:06 +1300)]
layers: Get rid of DEVICE_MEM_INFO::refCount
This is always the sum of the two container sizes, which are guaranteed
accessible in constant time in C++11.
We can only screw it up by independently tracking it.
Signed-off-by: Chris Forbes <chrisforbes@google.com>
GregF [Fri, 1 Apr 2016 20:20:14 +0000 (14:20 -0600)]
SPIRV: Adjust to spirv-tools library move
Michael Mc Donnell [Fri, 18 Mar 2016 04:18:32 +0000 (21:18 -0700)]
layers: Fix signed/unsigned comparison warnings in thread_check.h
Change-Id: I3a5e8e64deed53b3817be1fe8e1e2b32c08691a7
Ian Elliott [Thu, 31 Mar 2016 16:48:19 +0000 (10:48 -0600)]
loader: vk{Create|Destroy}Instance can have multiple tmp callbacks
During code review of the Android version of the temporary debug_report
callbacks code, it was decided to allow an array of
VkDebugReportCallbackCreateInfoEXT structs to be passed to vkCreateInstance().
This code implements that, using some new utility functions in order to help
keep the code clean.