| Age | Commit message (Collapse) | Author |
|---|
|
(#161670)
Use the `Base` type alias from
https://github.com/llvm/llvm-project/pull/158433.
|
|
See https://github.com/llvm/llvm-project/pull/147168 for more info.
|
|
This patch is part of a project to move the Presburger library into
LLVM.
|
|
This patch implements the lowering for vector
deinterleave for vector of n-dimensions. Process
involves unrolling the n-d vector to a series
of one-dimensional vectors. The deinterleave
operation is then used on these vectors.
From:
```
%0, %1 = vector.deinterleave %a : vector<2x8xi8> -> vector<2x4xi8>
```
To:
```
%cst = arith.constant dense<0> : vector<2x4xi32>
%0 = vector.extract %arg0[0] : vector<8xi32> from vector<2x8xi32>
%res1, %res2 = vector.deinterleave %0 : vector<8xi32> -> vector<4xi32>
%1 = vector.insert %res1, %cst [0] : vector<4xi32> into vector<2x4xi32>
%2 = vector.insert %res2, %cst [0] : vector<4xi32> into vector<2x4xi32>
%3 = vector.extract %arg0[1] : vector<8xi32> from vector<2x8xi32>
%res1_0, %res2_1 = vector.deinterleave %3 : vector<8xi32> -> vector<4xi32>
%4 = vector.insert %res1_0, %1 [1] : vector<4xi32> into vector<2x4xi32>
%5 = vector.insert %res2_1, %2 [1] : vector<4xi32> into vector<2x4xi32>
...etc.
```
|
|
This is to make it more obvious for what the result type is, especially
with some less trivial cases like 0-d inputs resulting in 1-d inputs or
interaction with scalable vector types. Note that `vector.deinterleave`
uses the same format with explicit result type.
Also improve examples and clean up surrounding code.
|
|
VectorToSPIRV (#92012)
This is the second attempt at merging #91800, which bounced due to a
linker error apparently caused by an undeclared dependency.
`MLIRVectorToSPIRV` needed to depend on `MLIRVectorTransforms`. In fact
that was a preexisting issue already flagged by the tool in
https://discourse.llvm.org/t/ninja-can-now-check-for-missing-cmake-dependencies-on-generated-files/74344.
Context: https://github.com/iree-org/iree/issues/17346.
Test IREE integrate showing it's fixing the problem it's intended to
fix, i.e. it allows IREE to drop its local revert of
https://github.com/llvm/llvm-project/pull/89131:
https://github.com/iree-org/iree/pull/17359
This is added to VectorToSPIRV because SPIRV doesn't currently handle
`vector.interleave` (see motivating context above).
This is limited to 1D, non-scalable vectors.
|
|
enable in VectorToSPIRV" (#92006)
Reverts llvm/llvm-project#91800
Reason: https://lab.llvm.org/buildbot/#/builders/268/builds/13935
|
|
VectorToSPIRV (#91800)
Context: https://github.com/iree-org/iree/issues/17346.
Test IREE integrate showing it's fixing the problem it's intended to
fix, i.e. it allows IREE to drop its local revert of
https://github.com/llvm/llvm-project/pull/89131:
https://github.com/iree-org/iree/pull/17359
This is added to VectorToSPIRV because SPIRV doesn't currently handle
`vector.interleave` (see motivating context above).
This is limited to 1D, non-scalable vectors.
|
|
This unrolls n-D vector.interleave ops like:
```mlir
vector.interleave %i, %j : vector<6x3xf32>
```
To a sequence of 1-D operations:
```mlir
%i_0 = vector.extract %i[0]
%j_0 = vector.extract %j[0]
%res_0 = vector.interleave %i_0, %j_0 : vector<3xf32>
vector.insert %res_0, %result[0] :
// ... repeated x6
```
The 1-D operations can then be directly lowered to LLVM.
Depends on: #80966
|
