| Commit message (Collapse) | Author | Age |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This PR implements `Access.Immutable` to allow pointers to immutable
data.
The new type `ImmutablePtr<T>` is defined as an alias of `Ptr<T,
Address.Immutable>`.
By forming a immutable pointer, the programmer is conveying to the
compiler that the data at the pointer address will never change during
the execution of the current program. Therefore loads from immutable
pointers can be deduplicated by the compiler, and will translate to
`__ldg` when generating code for CUDA.
The SPIRV backend is not changed in this PR, since the current SPIRV
spec makes it very difficult to specify loads from immutable address
without generating tons of wrappers and boilerplate type declarations.
We would like to see the spec evolved a bit to around its support of
`NonWritable` physical storage pointers or immutable loads before we
attempt to express such immutability in SPIRV. For now we simply emit
ordinary pointers and loads when generating spirv.
---------
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fixes: https://github.com/shader-slang/slang/issues/7634
Duplicate of PR https://github.com/shader-slang/slang/pull/8052
Primary Changes:
* Added `storeCoherent` and `loadCoherent` for coherent load/store via
pointers. This is backed by `IRMemoryScopeAttr` which is an `IRAttr`
attached to `IRLoad` and `IRStore`
* Logic in `source\slang\slang-emit-spirv.cpp` for load/store emitting
has been reworked to be less messy and more maintainable
* Add to `hlsl.meta.slang` coop vector and coop matrix coherent
load/store operations
Secondary Changes:
* Added a missing load/store test for coop matrix:
`tests\cooperative-matrix\load-store-pointer.slang`
---------
Co-authored-by: ArielG-NV <aglasroth@nvidia.com>
Co-authored-by: ArielG-NV <159081215+ArielG-NV@users.noreply.github.com>
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
Co-authored-by: Nathan V. Morrical <natemorrical@gmail.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Note that while this change touched a large numer of files, there are no
changes to functionality being made here. The only things being done are
renaming various symbols and, in a few cases, updating or adding
comments for consistency with the new names.
The core of the naming changes are:
* Most things named to refer to `OutType` (e.g., `IROutType`,
`IRBuilder::getOutType()`, etc.) have been consistently renamed to refer
to `OutParamType`, to emphasize that the relevant AST/IR node types are
only intended for use to represent `out` parameters.
* The same change as described above for `OutType` is also made for
`RefType`, which becomes `RefParamType` in most cases. One mess that
this exposes is the way that the `ExplicitRef<T>` type in the core
module currently lowers to `IRRefParamType`. This change sticks to the
rule of not making functional changes, so that mess is left as-is for
now.
* Names referring to `InOutType` have been changed to instead refer to
`BorrowInOutType`. The intention with this naming change is to emphasize
that the Slang rules for `inout` are semantically those of a borrow (or
at least our interpretation of what a borrow means).
* Names referring to `ConstRefType` have been changed to instead refer
to `BorrowInType`. This change starts work on clarifying that the
existing `__constref` modifier was never intended to be a read-only
analogue of `__ref`, and instead is the input-only analogue of `inout`.
* The `ParameterDirection` enum type has been changed to
`ParamPassingMode`, to reflect the fact that the concept of "direction"
fails to capture what is actually being encoded, particularly once we
have modes beyond simple `in`/`out`/`inout`.
While this change does not alter behavior in any case (the user-exposed
Slang language is unchanged), it is intended to set up subsequence
changes that will work to make the handling of these types in the
compiler more nuanced and correct. Breaking this part of the change out
separately is primarily motivated by a desire to minimize the effort for
reviewers.
---------
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
packing/unpacking. (#8526)
Part of the effort to improve the performance of generated SPIRV code.
The existing lower-buffer-element-type pass works by loading the entire
buffer element content from memory, and translate it to logical type
stored in a local variable at the earliest reference of a buffer handle.
This means that is can generate inefficient code that reads more than
necessary.
Consider this example:
```
struct BigStruct { bool values[1024]; }
ConstantBuffer<BigStruct> cb;
void test(BigStruct v)
{
if (v.values[0]) { printf("ok"); }
}
[numthreads(1,1,1)]
void computeMain()
{
test(cb);
}
```
In IR, the `computeMain` function before lower-buffer-element-type pass
is something like following:
```
func test:
%v = param : BigStruct
%barr = fieldExtract(%v, "values")
%element = elementExtract(%barr, 0)
... // uses %element
func computeMain:
%v = load(cb)
call %test %v
```
The existing lower-buffer-element-type pass will rewrite the bool array
in `BigStruct` into `int` array so it is legal in SPIRV. However, it
does so by inserting the translation on the first `load` of the constant
buffer:
```
struct BigStruct_std430 {
int values[1024];
}
var cb : ConstantBuffer<BigStruct_std430>;
func computeMain:
%tmpVar : var<BigStruct>
call %unpackStorage(%tmpVar, cb)
%v : BigStruct = load %tmpVar
call %test %v
```
This means that the entire array will be loaded and translated to int,
before calling `test`, which only uses one element. It turns out that
the downstream compiler isn't always able to optimize out this
inefficient translation/copy.
This PR completely rewrites the way buffer-element-type lowering is
handled to avoid producing this inefficient code. It works in two parts:
first we turn on the `transformParamsToConstRef` pass for SPIRV target
as well, so we will translate the `test` function to take the `v`
parameter as `constref`. The second part is a redesigned
buffer-element-type pass that defers the storage-type to logical-type
translation until a value is actually used by a `load` instruction.
In this example, after `transformParamsToConstRef`, the IR is:
```
func test:
%v = param : ConstRef<BigStruct>
%barr = fieldAddr(%v, "values")
%elementPtr = elementAddr(%barr, 0)
%element = load(%elementPtr)
... // uses %element
func computeMain:
call %test %cb
```
The new `buffer-element-type-lowering` pass will take this IR, and
insert translation at latest possible time across the entire call graph,
and translate the IR into:
```
func test:
%v = param : ConstRef<BigStruct_std430>
%barr = fieldAddr(%v, "values")
%elementPtr : ptr<int> = elementAddr(%barr, 0)
%element_int = load(%elementPtr)
%element = cast(%element_int) : %bool
... // uses %element
func computeMain:
call %test %cb
```
In this new IR, there is no longer a load and conversion of the entire
array.
See new comment in `slang-ir-lower-buffer-element-type.cpp` for more
details of how the pass works.
This PR also address many other issues surfaced by turning on
`transformParamsToConstRef` pass on SPIRV backend.
---------
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit removes unnecessary Load and Store pairs in IR.
When the IR is like
```
let %1 = var
let %2 = load(%ptr)
store(%1 %2)
```
This PR will replace all uses of %1 with %ptr.
And the load and store instructions will be removed.
But I found that there can be cases where %2 might be still used later
in other IRs.
For these cases, the removal of load instruction relies on DCE.
---------
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
when we have `GetElementPtr -> load -> GetElement` in our use-chain, the
final `GetElement` may use the `load` as a `Index`, not a base.
This is a non-issue with `getFieldExtract` since a field is a StructKey.
We will still add this check to ensure no bugs down the line.
---------
Co-authored-by: Harsh Aggarwal <haaggarwal@nvidia.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fixes #7574
Changes:
* Add an initial (fairly simple) optimization pass which is able to
eliminate redundant copies.
* Our current existing optimizer passes remove redundant load/store very
robustly, this pass will focus on other cases of copy elimination
* Primary approach is to make all functions which are `in T` and `T` is
trivial to copy into a `__constref T`. We then (depending on scenario)
manually insert a variable+load if a pass-by-reference is not possible;
otherwise we pass by `constref`.
* Added optimizations to eliminate redundant code which causes
`constref` to fail to compile
---------
Co-authored-by: Harsh Aggarwal <haaggarwal@nvidia.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>
|
| |
|
|
|
|
|
|
|
| |
This fixes issue #6654
Only hoist instructions that are optimized by prepareFuncForForwardDiff.
Add flag hoistLoopInvariantInsts to IRSimplificationOptions and set this
to true only if called from prepareFuncForForwardDiff, then only hoist
if the flag is set. Additionally, do not hoist loops if they only have a
single trivial iteration.
|
| |
|
|
|
| |
* Fix fast single iteration loop test in redundancy pass.
* Fix.
|
| |
|
|
|
|
|
|
|
| |
* Move switch statement bodies to their own lines
* format
---------
Co-authored-by: Yong He <yonghe@outlook.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Precompiled SPIR-V import support
Adds appropriate linkage and function declaration syntax
for SPIR-V functions that are declared, to be imported
from another SPIR-V module.
Unlike DXIL, stripping the Slang IR for a function down
to a declaration requires retaining a block of parameters,
as the function declaration must be emitted to SPIR-V
with the same parameters as a definition. Because that
thwarts the logic in Slang to tell the difference between
a declaration and definition, and explicit decoration is
introduced to explicitly mark functions which need to be
treated as declarations during emit phase.
Fixes #4992
Co-authored-by: Yong He <yonghe@outlook.com>
|
| |
|
|
|
|
|
| |
* format
* Minor test fixes
* enable checking cpp format in ci
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Initial -embed-spirv support
Add support for SPIR-V precompilation using the framework
established for DXIL.
Work on #4883
* SLANG_UNUSED
* Add linkage attributes to exported spirv functions
* Combine DXIL and SPIRV paths
* Whitespace fix
* Merge remaining precompiled spirv/dxil paths
* Change inst accessors to return codegentarget
* Add unit test for precompiled spirv
---------
Co-authored-by: Yong He <yonghe@outlook.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Support mixture of precompiled and non-precompiled modules
This changes the implementation of precompile DXIL modules to
accept combinations of modules with precompiled DXIL, ones without,
and ones with a mixture of precompiled DXIL and Slang IR.
During precompilation, module IR is analyzed to find public functions
which appear to be capable of being compiled as HLSL, and those
functions are given a HLSLExport decoration, ensuring they are emitted
as HLSL and preserved in the precompiled DXIL blob. The IR for those
functions is then tagged with a new decoration AvailableInDXIL, which
marks that their implementation is present in the embedded DXIL blob.
The DXIL blob is attached to the IR as before, inside a EmbeddedDXIL
BlobLit instruction.
The logic that determines whether or not functions should be
precompiled to DXIL is a placeholder at this point, returning true
always. A subsequent change will add selection criteria.
During module linking, the full module IR is available, as well
as the optional EmbeddedDXIL blob. The IR for functions implemented
by the blob are tagged with AvailableInDXIL in the module IR.
After linking the IR for all modules to program level IR, the IR for
the functions marked AvailableInDXIL are deleted from the linked IR,
prior to emitting HLSL and compiling linking the result.
This change also changes the point of time when the module IR is
checked for EmbeddedDXIL blobs. Instead of happening at load time
as before, it happens during immediately before final linking, meaning
that the blob does not need to be independently stored with the module
separate from the IR as was done previously.
Work on #4792
* Clean up debug prints
* Call isSimpleHLSLDataType stub
* Address feedback on precompiled dxil support
Allow for IR filtering both before and after linking.
Only mark AvailableInDXIL those functions which pass
both filtering stages. Functions are corrlated using
mangled function names.
Rather than delete functions entirely when linking with
libraries that include precompiled DXIL, instead convert
the IR function definitions to declarations by gutting
them, removing child blocks.
* Use artifact metadata and name list instead of linkedir hack
* Use String instead of UnownedStringSlice
* Update tests
* Renaming
* Minor edits
* Don't fully remove functions post-link
* Unexport before collecting metadata
|
| |
|
|
|
|
|
|
|
|
|
| |
* Tuple swizzling and element access.
* Update proposal status.
* Cleanup.
* Fix merrge error.
* Address review.
|
| |
|
|
| |
code (#4250)
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* SPIRV compiler performance fixes.
* Cleanup.
* update project files
* Cleanup debug code.
* Make redundancy removal non-recursive.
---------
Co-authored-by: Yong He <yhe@nvidia.com>
|
| |
|
|
|
|
|
|
|
| |
* Do not move movable insts in fuse-satcoop
* Add case for IRCall in isMovableInst
---------
Co-authored-by: Yong He <yonghe@outlook.com>
|
| |
|
|
|
|
|
| |
* Disable code motion for expensive insts (call & div)
The current redundancy removal pass does not consider control-flow within loops and as a result can sometimes move dynamic dispatch code outside their switch blocks, if they are nested in a single-iter-loop.
* Update liveness.slang.expected
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Simplify lookup.
* Various bug fixes.
* Report type dictionary size in perf benchmark.
* Remove type duplication.
* increase initial dict size.
* Bug fix.
* Fix bugs.
* Fixup.
* Revert type legalization looping.
* Fix specialization pass.
---------
Co-authored-by: Yong He <yhe@nvidia.com>
|
| |
|
| |
Co-authored-by: Yong He <yhe@nvidia.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Use scratchData on `IRInst` to replace HashSets.
* Update test results.
* Initialize scratchData.
* Update autodiff documentation.
* Use enum instead of bool.
---------
Co-authored-by: Yong He <yhe@nvidia.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
* SSA Register Allocation improvements.
* Fix.
* Rename `Use`->`UseOrPseudoUse`.
---------
Co-authored-by: Yong He <yhe@nvidia.com>
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* #include an absolute path didn't work - because paths were taken to always be relative.
* WIP lowerCamel Dictionary.
* WIP more lowerCamel fixes for Dictionary.
* Add/Remove/Clear
* GetValue/Contains
* Fix tabs in dictionary.
Count -> getCount
* Fix fields with caps.
* Key -> key
Value -> value
Use m_ for members where appropriate.
Use lowerCamel in linked list.
* Some small fixes/improvements to Dictionary.
* Kick CI.
|
| |
|
| |
Co-authored-by: Yong He <yhe@nvidia.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
* Fix optimization pass not converging.
* Fix.
* Fix tests.
---------
Co-authored-by: Yong He <yhe@nvidia.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
* Detect and deduplicate read-only resource access.
* Fix tests.
* Fix tests.
---------
Co-authored-by: Yong He <yhe@nvidia.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* More control flow and Phi param simplifications.
* Fix.
* Fix gcc error.
* Fix.
* More IR cleanup.
* Fix bug in phi param dce + ifelse simplify.
* Propagate and DCE side-effect-free functions.
* Enhance CFG simplifcation to remove loops with no side effects.
* Fix.
* Fixes.
* Fix tests. Add [__AlwaysFoldIntoUseSite] for rayPayloadLocation.
* More cleanup.
* Fixes.
* Fix.
---------
Co-authored-by: Yong He <yhe@nvidia.com>
|
| |
|
|
|
|
|
|
|
| |
* Overhaul global inst deduplication and cpp/cuda backend.
* Update IR documentation.
---------
Co-authored-by: Yong He <yhe@nvidia.com>
|
| | |
|
| |
|
|
|
|
|
| |
* Unify UpdateField and UpdateElement with access chain.
* Fix warnings.
Co-authored-by: Yong He <yhe@nvidia.com>
|
| |
|
|
|
|
|
|
|
| |
* Reimplement address elimination pass.
* Fix error.
* Update test references.
Co-authored-by: Yong He <yhe@nvidia.com>
|
|
|
Co-authored-by: Yong He <yhe@nvidia.com>
|
