vkCmdClearColorImage() and vkCmdClearDepthStencilImage() are currently
implemented using Blitter::clear(), which delegates to fastClear() for
certain formats, or falls back to using a generic Reactor routine.

This change benchmarks the performance for:
 - VK_FORMAT_R8G8B8A8_UNORM (fast path)
 - VK_FORMAT_R32_SFLOAT (slow path)
 - VK_FORMAT_D32_SFLOAT (slow path)

Bug: b/158231104
Change-Id: I5383b322e3ea5de6b21113843fb665a22a92b9bf
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45470
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>

Using https://github.com/KhronosGroup/Vulkan-Headers at tag v1.2.141.

Vulkan-Hpp is useful for writing unit tests and microbenchmarks, since
it includes support for loading a specific dynamically linked library:
https://github.com/KhronosGroup/Vulkan-Hpp#extensions--per-device-function-pointers

It also provides some abstraction to make tests more succinct and less
bug prone.

Bug: b/144696871
Change-Id: I4b22aefd205c01094e525b051b088ea9df14cca8
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45568
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>

In 34c59c9b ("Update Vulkan headers to version 1.1.111") the Vulkan
Headers were inadvertently converted to CRLF line endings, convert
them back.

Bug: b/157893908
Change-Id: I4ce4edd941e55cbfb7172c8d6041bf9e9a0254d3
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45430Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <nicolascapens@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

Bug: b/157926124
Change-Id: Iba985aa82cbda2d2d9778909d47abda2f0a7322b
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45489
Presubmit-Ready: Alexis Hétu <sugoi@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Reviewed-by: Nicolas Capens <nicolascapens@google.com>
Tested-by: Ari Suonpää <ari.suonpaa@siru.fi>

Bug: b/155664177
Change-Id: I9322be202434908dfd652980f50e1024cf2efa2d
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45368
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Reviewed-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Chris Forbes <chrisforbes@google.com>
Tested-by: Trevor Black <vantablack@google.com>
Presubmit-Ready: Trevor Black <vantablack@google.com>

`marl::Scheduler::setWorkerThreadCount()` and `marl::Scheduler::setThreadInitializer()` have been deprecated.
`marl::Scheduler::Config` is the new API.

See https://github.com/google/marl/issues/139 for context.

Bug: b/140546382
Change-Id: Ida768eea98a7ca5d22f55f5055ca288b41b353d0
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45493Tested-by: Ben Clayton <bclayton@google.com>
Presubmit-Ready: Ben Clayton <bclayton@google.com>
Reviewed-by: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

Bug: b/152339534
Change-Id: Icfbd4d3a00ffb322d083dccd7aee980ae1ee91d6
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/43109Reviewed-by: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Ben Clayton <bclayton@google.com>

* changes:
  Update Marl to 38c0c7a0f
  Squashed 'third_party/marl/' changes from 3c643dd4c..38c0c7a0f
  Fix LLVM_VERSION_MAJOR for LLVM 10 on macOS

Changes:
    38c0c7a0f Add Scheduler::Config and deprecate old [gs]etters
    17b12a448 Use CreateFiberEx instead of CreateFiber (#141)
    7236bd098 Test build / warning fixes for iOS
    c716869a7 Enable FreeBSD support
    00433d418 Add tests for using a marl::Scheduler without binding
    6b4f6472e Fix CMake warning when not built as subproject
    0591f0ef0 Kokoro Ubuntu: Switch to docker image
    bf9255db7 Optimization for marl::parallelize

Commands:
    ./third_party/update-marl.sh --squash

Bug: b/140546382
Change-Id: I5eaceebcbfef7636fd711d9e4dc7ae4a946e8a34

38c0c7a0f Add Scheduler::Config and deprecate old [gs]etters
17b12a448 Use CreateFiberEx instead of CreateFiber (#141)
7236bd098 Test build / warning fixes for iOS
c716869a7 Enable FreeBSD support
00433d418 Add tests for using a marl::Scheduler without binding
6b4f6472e Fix CMake warning when not built as subproject
0591f0ef0 Kokoro Ubuntu: Switch to docker image
bf9255db7 Optimization for marl::parallelize

git-subtree-dir: third_party/marl
git-subtree-split: 38c0c7a0fb1ceec42fb48d9c5380026a9eb9b076

Looks like I generated the configs using ToT which had already bumped version, insteaad of the snapshot held by SwiftShader `third_party/llvm-10.0`

Bug: b/152339534
Change-Id: I7d01a3c8d308916c6fa699ba73f0b3201fdf1658

Bug: b/157856449
Change-Id: I8363688a32df093244dd0b043ae2ce271e7387d6
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45428
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Reviewed-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>

Using headers from https://github.com/KhronosGroup/Vulkan-Headers at
tag v1.2.141.

* Modified vulkan.h to still include vulkan_fuchsia_extras.h like
  before.

* VK_NVX_device_generated_commands was replaced with a
  non-experimental version VK_NV_device_generated_commands.
  So replaced the symbols in VkStringify too.

* Preserved the line endings of the existing files.

* Note that VK_PIPELINE_BIND_POINT_RANGE_SIZE doesn't include the Ray
  Tracing pipeline bind point.  This is fine since it is not supported
  by SwiftShader.

Bug: b/157856449
Change-Id: I85f865e128dc034d39a24223621c105f9d1f5b03
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45429
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>

and move pointer typeinfo value construction from build time to runtime.
While this may seem counter-intuitive for performance, it's actually a
big win, as shader compilation can take a significant amount of time
with the amount of debug logic generated. This is also a first step
towards lazy/static `vk::dbg::Value` generation for array types.

Bug: b/124388146
Change-Id: I3448ffdabd553db17a22090081ac4809e0555b1d
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45488
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Ben Clayton <bclayton@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>

The 'builtins' scope was rather specific to the initial SPIR-V SSA debugging functionality, and doesn't really fit with the rich debug model.
Just replace it with a more sensible globals scope.

Fixes issues where these globals were only visible in the initial stack frame.

Bug: b/148401179
Change-Id: Id229e891edb8ece3694d989b48d3399d5d489d47
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45356Tested-by: Ben Clayton <bclayton@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

Bug: b/148401179
Change-Id: I0f9458c8e7c8bb04b3b4443f30bb1b7f6c98d686
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45354Tested-by: Ben Clayton <bclayton@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

The spec states that opaque types can have a None size instead of a constant integer size.

Bug: b/148401179
Change-Id: I52eeabe2336775643787599890857169825f3230
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45353Tested-by: Ben Clayton <bclayton@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

The type info is parsed into the corresponding structs, but nothing consumes these just yet.

Bug: b/148401179
Change-Id: I0568ea61fe22a5e8be5d6dc07e43868a6e5a35ec
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45352Tested-by: Ben Clayton <bclayton@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

Arrays are still WIP.
While I'm still working on these, don't fall through to the SSA handler. This can dramatically affect performance for large arrays.

Bug: b/148401179
Change-Id: I77e972f13cf9a1ce2b0d47a883194b6bceb68f4b
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45351
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Ben Clayton <bclayton@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>

If the shader contains `OpenCL.DebugInfo.100` rich debug info instructions, don't also include raw SSA values in the watch window.

Bug: b/148401179
Change-Id: Id70a702bbb24fb0bf79faa55cd22c0a87c973f62
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45350
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Ben Clayton <bclayton@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>

As part of chromium:944811, LLVM_ENABLE_THREADS was turned off, but
since https://swiftshader-review.googlesource.com/c/SwiftShader/+/33484
we may invoke Reactor JIT-compilation on multiple threads. Instead of
re-enabling LLVM_ENABLE_THREADS wholesale, just make ManagedStatic
thread-safe.

Since sys::Mutex usage is not thread-safe without LLVM_ENABLE_THREADS,
use standard C++ synchronization primitives instead.

Note this change only affects the Subzero JIT backend, not the LLVM JIT.

Bug: chromium:950343
Change-Id: Ie70be4adb2c395459cb9bbd4f6699e0b04969c74
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45328
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>

Previously only changes authored by Googlers would be tested, leaving
external contributions untested until the next 'daily' run.

This change enables testing of external contributions when a Googler
added a Code-Review+1, Code-Review+2, or Presubmit-Ready+1 vote.

Change-Id: I2f5c327848ad507d75967c9fd7f16085d0729a1a
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45508Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>

This header is now pre-built as part of third_party/SPIRV-Headers
It was first added by https://swiftshader-review.googlesource.com/c/SwiftShader/+/43035

Bug: b/150936967
Change-Id: I2931cb6475fd2d99cdc015d1a2e3bd759f9255d6
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45309Tested-by: Nicolas Capens <nicolascapens@google.com>
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Reviewed-by: Ben Clayton <bclayton@google.com>

When pCreateInfo->setLayoutCount is 0, descriptorSets[0].bindings would
not be assigned the initial bindingStorage value, which is the start of
the additional memory that is allocated for the PipelineLayout object.
It is used in PipelineLayout::destroy() for deallocating said memory,
which leads to attempting to free an initialized pointer.

This was a regression caused by https://swiftshader-review.googlesource.com/c/SwiftShader/+/44328

Bug: b/154522740
Change-Id: I98f63d58e9d64b36f15d966afd1bc70c1ec97f59
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45448
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Trevor Black <vantablack@google.com>

This is not a template class, so its methods don't have to be inlined.
Nor are they trivial enough to save us anything in performance or
binary size if they were inlined. Instead, this could reduce binary
size a bit because there are a large number of rr::Variable uses.

Bug: chromium:1086501
Change-Id: I1a6d078628ed7e634e0d90d6d8455e7d2938d134
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45388
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>

Changes:
    55af3902fc2 Fix function use (#3372)
    9cb2571a184 spirv-val: allow DebugInfoNone for DebugTypeComposite.Size (#3374)
    4386ef4234e Add validation support for ImageGatherBiasLodAMD (#3363)
    b0264b87ffb Fix validation failure on OpDecorationGroup (#3365)
    4410272bdda Remove deprecated interfaces from instrument passes (#3361)
    50b15578866 Preserve debug info in inline pass (#3349)
    4dbe18b0c86 Reject folding comparisons with unfoldable types. (#3370)
    55193b06e5e Improve build instructions for fuzzer (#3364)
    3c47dac2820 Add unrolling to performance passes (#3082)
    2b987c49a4e Handle OpConstantNull in ssa-rewrite (#3362)
    95df4c9643c Add in a bunch of missed files to the BUILD.gn (#3360)
    90930cb3115 Remove stale entries from BUILD.gn (#3358)
    18ba3d9a353 allow cross compiling for Windows Store, UWP, etc. (#3330)
    2f69ea849aa spirv-fuzz: Remove FuzzerPassAddUsefulConstructs (#3341)
    522561619a9 Add support for StorageBuffer (#3348)
    b75dbf82a69 Prevent Effcee install his things when build spirv-tools with testing enabled (#3256)
    85c7e7956bf Don't register edges twice in merge return (#3350)
    bd0a2da946c Revert "Revert "[spirv-opt] refactor inlining pass (#3328)" (#3342)" (#3345)
    31182763704 spirv-reduce: Remove unused struct members (#3329)
    a6b0e132ecc Add adjust branch weights transformation (#3336)
    d4fac3451b7 Revert "[spirv-opt] refactor inlining pass (#3328)" (#3342)
    233246bc9c5 [spirv-opt] refactor inlining pass (#3328)
    2992386ebea spirv-reduce: Remove unused uniforms and similar (#3321)
    a9f2a145e65 spirv-fuzz: Fix to fact manager (#3339)
    045a26e6e37 spirv-fuzz: Get rid of unnecessary template method (#3340)
    63fa9114a93 Do merge return if the return is not at the end of the function. (#3337)
    c8590c18bd0 Preserve debug info for wrap-opkill (#3331)
    d2b48621949 Validate ShaderCallKHR memory scope (#3332)
    2e1d208ed9d spirv-fuzz: Do not allow adding stores to read-only pointers (#3316)
    54fb17b2d30 reduce: increase default step limit (#3327)
    49842b88eec Generalize IsReadOnlyVariable() to apply to pointers (#3325)
    49ca250b44c Delete nullptr in function bb list immedietly (#3326)
    d0a87194f7b Set DebugScope for termination instructions (#3323)
    f278b467dfd spirv-fuzz: Do not outline regions that end with a loop header (#3312)
    23d68608b00 vscode: Handle '|' chains on BitEnum / ValueEnum (#3309)
    42268740c95 Add debug information analysis (#3305)
    eed48ae479d Add spvtools::opt::Operand::AsLiteralUint64 (#3320)
    94d6002dc53 spirv-fuzz: Pass on validator options during shrinking (#3317)
    88faf63ad3c spirv-fuzz: Clamp statically out-of-bounds accesses in code donation (#3315)
    b74199a22d4 spirv-fuzz: Fix memory management in the fact manager (#3313)
    d158ffe5405 spirv-fuzz: Do not replace the Sample argument in OpImageTexelPointer (#3311)
    5547553a0c7 Allow various validation options to be passed to spirv-opt (#3314)
    30ffe62e257 typo fix: in README.md exectuable->executable (#3306)
    67f4838659f spirv-fuzz: Make handling of synonym facts more efficient (#3301)
    61b7de3c39f Remove unreachable code. (#3304)
    ed96301c6c4 spirv-fuzz: Fix to outliner (#3302)
    c018fc6ae66 spirv-fuzz: Do not outline regions that produce pointer outputs (#3291)
    f460cca9dca spirv-fuzz: Handle OpRuntimeArray when replacing ids with synonyms (#3292)
    2f180468a71 spirv-fuzz: Handle image storage class in donation (#3290)
    f82d47003e7 spirv-fuzz: Respect rules for OpSampledImage (#3287)
    7ce2db1763b spirv-fuzz: Fix comment. (#3300)
    7d65bce0bbe Sampled images as read-only storage (#3295)
    2a2bdbd5d72 Remove implicit fallthrough (#3298)
    49566448944 Add tests for recently added command line option (#3297)
    ca5751590ed If SPIRV-Headers is in our tree, include it as subproject (#3299)
    e70d25f6fa5 Struct CFG analysus and single block loop (#3293)
    000040e707a Preserve debug info in eliminate-dead-functions (#3251)
    c531099eb34 Update acorn version (#3294)
    34be23373b9 Handle more cases in dead member elim (#3289)
    d0490ef080c Fix pch macro to ignore clang-cl (#3283)
    538512e8e89 spirv-fuzz: Improve the handling of equation facts (#3281)
    183e3242a36 spirv-fuzz: Handle more general SPIR-V in donation (#3280)
    4af38c49bfe spirv-fuzz: Improve support for compute shaders in donation (#3277)
    e95fbfb1f50 spirv-fuzz: Transformation to add OpConstantNull (#3273)
    5d491a7ed66 spirv-fuzz: Handle isomorphic types property in composite construction (#3262)
    bfd25ace084 spirv-fuzz: Limit adding of new variables to 'basic' types (#3257)
    f28cdeff16f spirv-fuzz: Only replace regular ids with synonyms (#3255)
    8d4261bc440 spirv-fuzz: Introduce TransformationContext (#3272)
    2fdea57d19d spirv-fuzz: Add validator options (#3254)
    af01d57b5e3 Update dominates to check for null nodes (#3271)
    f20c0d7971c Set wrapped kill basic block's parent (#3269)
    c37c94929bf Validate Buffer and BufferBlock apply only to struct types (#3259)

Commands:
    ./third_party/update-spirvtools.sh

Bug: b/123642959
Change-Id: Id033a5049da5a8b7c7fbf5f1896a985abfbd4ad8

55af3902fc2 Fix function use (#3372)
9cb2571a184 spirv-val: allow DebugInfoNone for DebugTypeComposite.Size (#3374)
4386ef4234e Add validation support for ImageGatherBiasLodAMD (#3363)
b0264b87ffb Fix validation failure on OpDecorationGroup (#3365)
4410272bdda Remove deprecated interfaces from instrument passes (#3361)
50b15578866 Preserve debug info in inline pass (#3349)
4dbe18b0c86 Reject folding comparisons with unfoldable types. (#3370)
55193b06e5e Improve build instructions for fuzzer (#3364)
3c47dac2820 Add unrolling to performance passes (#3082)
2b987c49a4e Handle OpConstantNull in ssa-rewrite (#3362)
95df4c9643c Add in a bunch of missed files to the BUILD.gn (#3360)
90930cb3115 Remove stale entries from BUILD.gn (#3358)
18ba3d9a353 allow cross compiling for Windows Store, UWP, etc. (#3330)
2f69ea849aa spirv-fuzz: Remove FuzzerPassAddUsefulConstructs (#3341)
522561619a9 Add support for StorageBuffer (#3348)
b75dbf82a69 Prevent Effcee install his things when build spirv-tools with testing enabled (#3256)
85c7e7956bf Don't register edges twice in merge return (#3350)
bd0a2da946c Revert "Revert "[spirv-opt] refactor inlining pass (#3328)" (#3342)" (#3345)
31182763704 spirv-reduce: Remove unused struct members (#3329)
a6b0e132ecc Add adjust branch weights transformation (#3336)
d4fac3451b7 Revert "[spirv-opt] refactor inlining pass (#3328)" (#3342)
233246bc9c5 [spirv-opt] refactor inlining pass (#3328)
2992386ebea spirv-reduce: Remove unused uniforms and similar (#3321)
a9f2a145e65 spirv-fuzz: Fix to fact manager (#3339)
045a26e6e37 spirv-fuzz: Get rid of unnecessary template method (#3340)
63fa9114a93 Do merge return if the return is not at the end of the function. (#3337)
c8590c18bd0 Preserve debug info for wrap-opkill (#3331)
d2b48621949 Validate ShaderCallKHR memory scope (#3332)
2e1d208ed9d spirv-fuzz: Do not allow adding stores to read-only pointers (#3316)
54fb17b2d30 reduce: increase default step limit (#3327)
49842b88eec Generalize IsReadOnlyVariable() to apply to pointers (#3325)
49ca250b44c Delete nullptr in function bb list immedietly (#3326)
d0a87194f7b Set DebugScope for termination instructions (#3323)
f278b467dfd spirv-fuzz: Do not outline regions that end with a loop header (#3312)
23d68608b00 vscode: Handle '|' chains on BitEnum / ValueEnum (#3309)
42268740c95 Add debug information analysis (#3305)
eed48ae479d Add spvtools::opt::Operand::AsLiteralUint64 (#3320)
94d6002dc53 spirv-fuzz: Pass on validator options during shrinking (#3317)
88faf63ad3c spirv-fuzz: Clamp statically out-of-bounds accesses in code donation (#3315)
b74199a22d4 spirv-fuzz: Fix memory management in the fact manager (#3313)
d158ffe5405 spirv-fuzz: Do not replace the Sample argument in OpImageTexelPointer (#3311)
5547553a0c7 Allow various validation options to be passed to spirv-opt (#3314)
30ffe62e257 typo fix: in README.md exectuable->executable (#3306)
67f4838659f spirv-fuzz: Make handling of synonym facts more efficient (#3301)
61b7de3c39f Remove unreachable code. (#3304)
ed96301c6c4 spirv-fuzz: Fix to outliner (#3302)
c018fc6ae66 spirv-fuzz: Do not outline regions that produce pointer outputs (#3291)
f460cca9dca spirv-fuzz: Handle OpRuntimeArray when replacing ids with synonyms (#3292)
2f180468a71 spirv-fuzz: Handle image storage class in donation (#3290)
f82d47003e7 spirv-fuzz: Respect rules for OpSampledImage (#3287)
7ce2db1763b spirv-fuzz: Fix comment. (#3300)
7d65bce0bbe Sampled images as read-only storage (#3295)
2a2bdbd5d72 Remove implicit fallthrough (#3298)
49566448944 Add tests for recently added command line option (#3297)
ca5751590ed If SPIRV-Headers is in our tree, include it as subproject (#3299)
e70d25f6fa5 Struct CFG analysus and single block loop (#3293)
000040e707a Preserve debug info in eliminate-dead-functions (#3251)
c531099eb34 Update acorn version (#3294)
34be23373b9 Handle more cases in dead member elim (#3289)
d0490ef080c Fix pch macro to ignore clang-cl (#3283)
538512e8e89 spirv-fuzz: Improve the handling of equation facts (#3281)
183e3242a36 spirv-fuzz: Handle more general SPIR-V in donation (#3280)
4af38c49bfe spirv-fuzz: Improve support for compute shaders in donation (#3277)
e95fbfb1f50 spirv-fuzz: Transformation to add OpConstantNull (#3273)
5d491a7ed66 spirv-fuzz: Handle isomorphic types property in composite construction (#3262)
bfd25ace084 spirv-fuzz: Limit adding of new variables to 'basic' types (#3257)
f28cdeff16f spirv-fuzz: Only replace regular ids with synonyms (#3255)
8d4261bc440 spirv-fuzz: Introduce TransformationContext (#3272)
2fdea57d19d spirv-fuzz: Add validator options (#3254)
af01d57b5e3 Update dominates to check for null nodes (#3271)
f20c0d7971c Set wrapped kill basic block's parent (#3269)
c37c94929bf Validate Buffer and BufferBlock apply only to struct types (#3259)

git-subtree-dir: third_party/SPIRV-Tools
git-subtree-split: 55af3902fc24db889b0ef8010a83efca04a6352c

See https://swiftshader-review.googlesource.com/c/SwiftShader/+/42088/7#message-18bed01232babd6cf5b5770b138d9034073fec95
for rationale.

Bug: b/151137030
Change-Id: Ieef8636734be1dc19ced8f572daea4990231106b
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45288
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>

Made 'Sample' come first again and thereby made it easier to use for
quick experiments and smoke testing.

Bug: b/156739423
Change-Id: Ic821aba66afbac97556b072750e24747af1010de
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45208
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

Previously we relied on the default copy constructor to be generated
when ENABLE_RR_DEBUG_INFO is disabled. This is not necessary as we can
just use the straightforward manual definition of it in either case.
Note RR_DEBUG_INFO_EMIT_VAR() is a no-op when ENABLE_RR_DEBUG_INFO is
disabled.

Bug: b/155302798
Change-Id: I04c020a795b4333e3c3d0a7ed375d21d9382e3d1
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45152
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

From https://gcc.gnu.org/onlinedocs/gcc-9.3.0/gcc/Name-lookup.html:
"The C++ standard prescribes that all names that are not dependent on
template parameters are bound to their present definitions when parsing
a template function or class.5 Only names that are dependent are looked
up at the point of instantiation.
...
To make the code valid either use this->f(), or Base<T>::f()"

It is best to use the this->f() solution because it could be a virtual
method.

Also use the type-safe load() and store() methods where possible.

Bug: b/155302798
Change-Id: Id636bfb0b1aae9529bf7f3f58213ec7823d9eddd
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45150
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>

We only need to know the array size during materialization of Array<>
types, so store it only for that class and make the allocation a
virtual method.

This change removes the REACTOR_MATERIALIZE_LVALUES_ON_DEFINITION
option, since it requires knowing the full size of Variable at
construction, which would require Array<> to still pass that into the
Variable constructor. If we need an debugging feature like this again,
we could also materialize variables on first use instead of at
construction.

Bug: b/155302798
Change-Id: I65b02c1258643dcbd3ae18f3f097cb815efa0830
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45148
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>

Bug: none
Change-Id: I1e9d97e95fea1e9c2863211c711feda6d794a825
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45268
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Antonio Maiorano <amaiorano@google.com>
Reviewed-by: Nicolas Capens <nicolascapens@google.com>

This will enable us to lazily create the underlying Value* in subsequent
changes.

Also make Argument<T> encapsulate its value, and make it type-safe.

Bug: b/155302798
Change-Id: I50eb16be06bc83e6c87b11f11effed33bee7107a
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/44671Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>

This methods returns an RValue<T> type so it's safer than 'loadValue()'.

Bug: b/155302798
Change-Id: Ib48819bd0a1b64ae2d22eda4910ad0b551b1eed7
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/44670
Kokoro-Result: kokoro <noreply+kokoro@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>

These methods operate on RValue<T> types so they're safer than
'loadValue()' and 'storeValue()'. They also enable overloading
the latter in subsequent changes.

Bug: b/155302798
Change-Id: I30903fc70e353dfd668873122ad7d8f74e857420
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/44669
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

This is just a refactoring which avoids storing the type as a Variable
fields, since the 'LValue<T>' subclass knows the type statically and we
only need it during materialization.

This gives us a vtable for overloading other methods in subsequent
changes, at no additional storage cost for Variable.

Bug: b/155302798
Change-Id: I36f6ef8f5f4d2b50ba04d539dc8b9735481c816a
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/44668
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

This is just a refactoring which will allow adding a virtual getType()
method in a subsequent change without overload naming conflicts.

'isVoid()' has been removed since it appears unused now.

Bug: b/155302798
Change-Id: I16f5cbd947da9581e0f20837497f6b61cfeb7a21
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45008
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

Previously any SPIR-V OpExecutionMode instruction would override the
execution mode parameters, so a shader module with multiple entry points
with different execution modes would not work as intended.

Bug: b/155550449
Change-Id: I86211e66a3cef5a3bbf9e535e9be8415b5d69d4b
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45249
Presubmit-Ready: Nicolas Capens <nicolascapens@google.com>
Tested-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Alexis Hétu <sugoi@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>

Doing a test run with LLVM after the one with Subzero results in various
tests timing out (including some 'UNSUPPORTED' ones). The Subzero test
run doesn't exhibit this issue. It's unclear at this point why this
happens, but reversing the order appears to stabilize it.

Bug: b/156728197
Change-Id: I6450d2d95d6043ef206d09f0020313fbf4b3f2e2
Reviewed-on: https://swiftshader-review.googlesource.com/c/SwiftShader/+/45149Tested-by: Nicolas Capens <nicolascapens@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro-Result: kokoro <noreply+kokoro@google.com>