2020-02-14 02:29:13 +08:00
|
|
|
// RUN: mlir-opt %s -launch-func-to-vulkan | FileCheck %s
|
|
|
|
|
|
|
|
|
|
// CHECK: llvm.mlir.global internal constant @kernel_spv_entry_point_name
|
|
|
|
|
// CHECK: llvm.mlir.global internal constant @SPIRV_BIN
|
2020-08-04 17:37:50 +08:00
|
|
|
// CHECK: %[[Vulkan_Runtime_ptr:.*]] = llvm.call @initVulkan() : () -> !llvm.ptr<i8>
|
2020-02-14 02:29:13 +08:00
|
|
|
// CHECK: %[[addressof_SPIRV_BIN:.*]] = llvm.mlir.addressof @SPIRV_BIN
|
|
|
|
|
// CHECK: %[[SPIRV_BIN_ptr:.*]] = llvm.getelementptr %[[addressof_SPIRV_BIN]]
|
|
|
|
|
// CHECK: %[[SPIRV_BIN_size:.*]] = llvm.mlir.constant
|
[mlir] replace LLVM dialect float types with built-ins
Continue the convergence between LLVM dialect and built-in types by replacing
the bfloat, half, float and double LLVM dialect types with their built-in
counterparts. At the API level, this is a direct replacement. At the syntax
level, we change the keywords to `bf16`, `f16`, `f32` and `f64`, respectively,
to be compatible with the built-in type syntax. The old keywords can still be
parsed but produce a deprecation warning and will be eventually removed.
Depends On D94178
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94179
2021-01-06 23:21:08 +08:00
|
|
|
// CHECK: llvm.call @bindMemRef1DFloat(%[[Vulkan_Runtime_ptr]], %{{.*}}, %{{.*}}, %{{.*}}) : (!llvm.ptr<i8>, i32, i32, !llvm.ptr<struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>>) -> !llvm.void
|
[mlir] replace LLVMIntegerType with built-in integer type
The LLVM dialect type system has been closed until now, i.e. did not support
types from other dialects inside containers. While this has had obvious
benefits of deriving from a common base class, it has led to some simple types
being almost identical with the built-in types, namely integer and floating
point types. This in turn has led to a lot of larger-scale complexity: simple
types must still be converted, numerous operations that correspond to LLVM IR
intrinsics are replicated to produce versions operating on either LLVM dialect
or built-in types leading to quasi-duplicate dialects, lowering to the LLVM
dialect is essentially required to be one-shot because of type conversion, etc.
In this light, it is reasonable to trade off some local complexity in the
internal implementation of LLVM dialect types for removing larger-scale system
complexity. Previous commits to the LLVM dialect type system have adapted the
API to support types from other dialects.
Replace LLVMIntegerType with the built-in IntegerType plus additional checks
that such types are signless (these are isolated in a utility function that
replaced `isa<LLVMType>` and in the parser). Temporarily keep the possibility
to parse `!llvm.i32` as a synonym for `i32`, but add a deprecation notice.
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94178
2021-01-06 23:19:04 +08:00
|
|
|
// CHECK: llvm.call @setBinaryShader(%[[Vulkan_Runtime_ptr]], %[[SPIRV_BIN_ptr]], %[[SPIRV_BIN_size]]) : (!llvm.ptr<i8>, !llvm.ptr<i8>, i32) -> !llvm.void
|
2020-02-14 02:29:13 +08:00
|
|
|
// CHECK: %[[addressof_entry_point:.*]] = llvm.mlir.addressof @kernel_spv_entry_point_name
|
|
|
|
|
// CHECK: %[[entry_point_ptr:.*]] = llvm.getelementptr %[[addressof_entry_point]]
|
2020-08-04 17:37:50 +08:00
|
|
|
// CHECK: llvm.call @setEntryPoint(%[[Vulkan_Runtime_ptr]], %[[entry_point_ptr]]) : (!llvm.ptr<i8>, !llvm.ptr<i8>) -> !llvm.void
|
[mlir] replace LLVMIntegerType with built-in integer type
The LLVM dialect type system has been closed until now, i.e. did not support
types from other dialects inside containers. While this has had obvious
benefits of deriving from a common base class, it has led to some simple types
being almost identical with the built-in types, namely integer and floating
point types. This in turn has led to a lot of larger-scale complexity: simple
types must still be converted, numerous operations that correspond to LLVM IR
intrinsics are replicated to produce versions operating on either LLVM dialect
or built-in types leading to quasi-duplicate dialects, lowering to the LLVM
dialect is essentially required to be one-shot because of type conversion, etc.
In this light, it is reasonable to trade off some local complexity in the
internal implementation of LLVM dialect types for removing larger-scale system
complexity. Previous commits to the LLVM dialect type system have adapted the
API to support types from other dialects.
Replace LLVMIntegerType with the built-in IntegerType plus additional checks
that such types are signless (these are isolated in a utility function that
replaced `isa<LLVMType>` and in the parser). Temporarily keep the possibility
to parse `!llvm.i32` as a synonym for `i32`, but add a deprecation notice.
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94178
2021-01-06 23:19:04 +08:00
|
|
|
// CHECK: llvm.call @setNumWorkGroups(%[[Vulkan_Runtime_ptr]], %{{.*}}, %{{.*}}, %{{.*}}) : (!llvm.ptr<i8>, i64, i64, i64) -> !llvm.void
|
2020-08-04 17:37:50 +08:00
|
|
|
// CHECK: llvm.call @runOnVulkan(%[[Vulkan_Runtime_ptr]]) : (!llvm.ptr<i8>) -> !llvm.void
|
|
|
|
|
// CHECK: llvm.call @deinitVulkan(%[[Vulkan_Runtime_ptr]]) : (!llvm.ptr<i8>) -> !llvm.void
|
2020-02-14 02:29:13 +08:00
|
|
|
|
[mlir] replace LLVMIntegerType with built-in integer type
The LLVM dialect type system has been closed until now, i.e. did not support
types from other dialects inside containers. While this has had obvious
benefits of deriving from a common base class, it has led to some simple types
being almost identical with the built-in types, namely integer and floating
point types. This in turn has led to a lot of larger-scale complexity: simple
types must still be converted, numerous operations that correspond to LLVM IR
intrinsics are replicated to produce versions operating on either LLVM dialect
or built-in types leading to quasi-duplicate dialects, lowering to the LLVM
dialect is essentially required to be one-shot because of type conversion, etc.
In this light, it is reasonable to trade off some local complexity in the
internal implementation of LLVM dialect types for removing larger-scale system
complexity. Previous commits to the LLVM dialect type system have adapted the
API to support types from other dialects.
Replace LLVMIntegerType with the built-in IntegerType plus additional checks
that such types are signless (these are isolated in a utility function that
replaced `isa<LLVMType>` and in the parser). Temporarily keep the possibility
to parse `!llvm.i32` as a synonym for `i32`, but add a deprecation notice.
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94178
2021-01-06 23:19:04 +08:00
|
|
|
// CHECK: llvm.func @bindMemRef1DHalf(!llvm.ptr<i8>, i32, i32, !llvm.ptr<struct<(ptr<i16>, ptr<i16>, i64, array<1 x i64>, array<1 x i64>)>>)
|
2020-08-22 01:34:12 +08:00
|
|
|
|
2020-02-14 02:29:13 +08:00
|
|
|
module attributes {gpu.container_module} {
|
[mlir] replace LLVMIntegerType with built-in integer type
The LLVM dialect type system has been closed until now, i.e. did not support
types from other dialects inside containers. While this has had obvious
benefits of deriving from a common base class, it has led to some simple types
being almost identical with the built-in types, namely integer and floating
point types. This in turn has led to a lot of larger-scale complexity: simple
types must still be converted, numerous operations that correspond to LLVM IR
intrinsics are replicated to produce versions operating on either LLVM dialect
or built-in types leading to quasi-duplicate dialects, lowering to the LLVM
dialect is essentially required to be one-shot because of type conversion, etc.
In this light, it is reasonable to trade off some local complexity in the
internal implementation of LLVM dialect types for removing larger-scale system
complexity. Previous commits to the LLVM dialect type system have adapted the
API to support types from other dialects.
Replace LLVMIntegerType with the built-in IntegerType plus additional checks
that such types are signless (these are isolated in a utility function that
replaced `isa<LLVMType>` and in the parser). Temporarily keep the possibility
to parse `!llvm.i32` as a synonym for `i32`, but add a deprecation notice.
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94178
2021-01-06 23:19:04 +08:00
|
|
|
llvm.func @malloc(i64) -> !llvm.ptr<i8>
|
2020-03-11 03:17:07 +08:00
|
|
|
llvm.func @foo() {
|
[mlir] replace LLVMIntegerType with built-in integer type
The LLVM dialect type system has been closed until now, i.e. did not support
types from other dialects inside containers. While this has had obvious
benefits of deriving from a common base class, it has led to some simple types
being almost identical with the built-in types, namely integer and floating
point types. This in turn has led to a lot of larger-scale complexity: simple
types must still be converted, numerous operations that correspond to LLVM IR
intrinsics are replicated to produce versions operating on either LLVM dialect
or built-in types leading to quasi-duplicate dialects, lowering to the LLVM
dialect is essentially required to be one-shot because of type conversion, etc.
In this light, it is reasonable to trade off some local complexity in the
internal implementation of LLVM dialect types for removing larger-scale system
complexity. Previous commits to the LLVM dialect type system have adapted the
API to support types from other dialects.
Replace LLVMIntegerType with the built-in IntegerType plus additional checks
that such types are signless (these are isolated in a utility function that
replaced `isa<LLVMType>` and in the parser). Temporarily keep the possibility
to parse `!llvm.i32` as a synonym for `i32`, but add a deprecation notice.
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94178
2021-01-06 23:19:04 +08:00
|
|
|
%0 = llvm.mlir.constant(12 : index) : i64
|
[mlir] replace LLVM dialect float types with built-ins
Continue the convergence between LLVM dialect and built-in types by replacing
the bfloat, half, float and double LLVM dialect types with their built-in
counterparts. At the API level, this is a direct replacement. At the syntax
level, we change the keywords to `bf16`, `f16`, `f32` and `f64`, respectively,
to be compatible with the built-in type syntax. The old keywords can still be
parsed but produce a deprecation warning and will be eventually removed.
Depends On D94178
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94179
2021-01-06 23:21:08 +08:00
|
|
|
%1 = llvm.mlir.null : !llvm.ptr<f32>
|
[mlir] replace LLVMIntegerType with built-in integer type
The LLVM dialect type system has been closed until now, i.e. did not support
types from other dialects inside containers. While this has had obvious
benefits of deriving from a common base class, it has led to some simple types
being almost identical with the built-in types, namely integer and floating
point types. This in turn has led to a lot of larger-scale complexity: simple
types must still be converted, numerous operations that correspond to LLVM IR
intrinsics are replicated to produce versions operating on either LLVM dialect
or built-in types leading to quasi-duplicate dialects, lowering to the LLVM
dialect is essentially required to be one-shot because of type conversion, etc.
In this light, it is reasonable to trade off some local complexity in the
internal implementation of LLVM dialect types for removing larger-scale system
complexity. Previous commits to the LLVM dialect type system have adapted the
API to support types from other dialects.
Replace LLVMIntegerType with the built-in IntegerType plus additional checks
that such types are signless (these are isolated in a utility function that
replaced `isa<LLVMType>` and in the parser). Temporarily keep the possibility
to parse `!llvm.i32` as a synonym for `i32`, but add a deprecation notice.
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94178
2021-01-06 23:19:04 +08:00
|
|
|
%2 = llvm.mlir.constant(1 : index) : i64
|
[mlir] replace LLVM dialect float types with built-ins
Continue the convergence between LLVM dialect and built-in types by replacing
the bfloat, half, float and double LLVM dialect types with their built-in
counterparts. At the API level, this is a direct replacement. At the syntax
level, we change the keywords to `bf16`, `f16`, `f32` and `f64`, respectively,
to be compatible with the built-in type syntax. The old keywords can still be
parsed but produce a deprecation warning and will be eventually removed.
Depends On D94178
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94179
2021-01-06 23:21:08 +08:00
|
|
|
%3 = llvm.getelementptr %1[%2] : (!llvm.ptr<f32>, i64) -> !llvm.ptr<f32>
|
|
|
|
|
%4 = llvm.ptrtoint %3 : !llvm.ptr<f32> to i64
|
[mlir] replace LLVMIntegerType with built-in integer type
The LLVM dialect type system has been closed until now, i.e. did not support
types from other dialects inside containers. While this has had obvious
benefits of deriving from a common base class, it has led to some simple types
being almost identical with the built-in types, namely integer and floating
point types. This in turn has led to a lot of larger-scale complexity: simple
types must still be converted, numerous operations that correspond to LLVM IR
intrinsics are replicated to produce versions operating on either LLVM dialect
or built-in types leading to quasi-duplicate dialects, lowering to the LLVM
dialect is essentially required to be one-shot because of type conversion, etc.
In this light, it is reasonable to trade off some local complexity in the
internal implementation of LLVM dialect types for removing larger-scale system
complexity. Previous commits to the LLVM dialect type system have adapted the
API to support types from other dialects.
Replace LLVMIntegerType with the built-in IntegerType plus additional checks
that such types are signless (these are isolated in a utility function that
replaced `isa<LLVMType>` and in the parser). Temporarily keep the possibility
to parse `!llvm.i32` as a synonym for `i32`, but add a deprecation notice.
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94178
2021-01-06 23:19:04 +08:00
|
|
|
%5 = llvm.mul %0, %4 : i64
|
|
|
|
|
%6 = llvm.call @malloc(%5) : (i64) -> !llvm.ptr<i8>
|
[mlir] replace LLVM dialect float types with built-ins
Continue the convergence between LLVM dialect and built-in types by replacing
the bfloat, half, float and double LLVM dialect types with their built-in
counterparts. At the API level, this is a direct replacement. At the syntax
level, we change the keywords to `bf16`, `f16`, `f32` and `f64`, respectively,
to be compatible with the built-in type syntax. The old keywords can still be
parsed but produce a deprecation warning and will be eventually removed.
Depends On D94178
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94179
2021-01-06 23:21:08 +08:00
|
|
|
%7 = llvm.bitcast %6 : !llvm.ptr<i8> to !llvm.ptr<f32>
|
|
|
|
|
%8 = llvm.mlir.undef : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%9 = llvm.insertvalue %7, %8[0] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%10 = llvm.insertvalue %7, %9[1] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
[mlir] replace LLVMIntegerType with built-in integer type
The LLVM dialect type system has been closed until now, i.e. did not support
types from other dialects inside containers. While this has had obvious
benefits of deriving from a common base class, it has led to some simple types
being almost identical with the built-in types, namely integer and floating
point types. This in turn has led to a lot of larger-scale complexity: simple
types must still be converted, numerous operations that correspond to LLVM IR
intrinsics are replicated to produce versions operating on either LLVM dialect
or built-in types leading to quasi-duplicate dialects, lowering to the LLVM
dialect is essentially required to be one-shot because of type conversion, etc.
In this light, it is reasonable to trade off some local complexity in the
internal implementation of LLVM dialect types for removing larger-scale system
complexity. Previous commits to the LLVM dialect type system have adapted the
API to support types from other dialects.
Replace LLVMIntegerType with the built-in IntegerType plus additional checks
that such types are signless (these are isolated in a utility function that
replaced `isa<LLVMType>` and in the parser). Temporarily keep the possibility
to parse `!llvm.i32` as a synonym for `i32`, but add a deprecation notice.
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94178
2021-01-06 23:19:04 +08:00
|
|
|
%11 = llvm.mlir.constant(0 : index) : i64
|
[mlir] replace LLVM dialect float types with built-ins
Continue the convergence between LLVM dialect and built-in types by replacing
the bfloat, half, float and double LLVM dialect types with their built-in
counterparts. At the API level, this is a direct replacement. At the syntax
level, we change the keywords to `bf16`, `f16`, `f32` and `f64`, respectively,
to be compatible with the built-in type syntax. The old keywords can still be
parsed but produce a deprecation warning and will be eventually removed.
Depends On D94178
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94179
2021-01-06 23:21:08 +08:00
|
|
|
%12 = llvm.insertvalue %11, %10[2] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
[mlir] replace LLVMIntegerType with built-in integer type
The LLVM dialect type system has been closed until now, i.e. did not support
types from other dialects inside containers. While this has had obvious
benefits of deriving from a common base class, it has led to some simple types
being almost identical with the built-in types, namely integer and floating
point types. This in turn has led to a lot of larger-scale complexity: simple
types must still be converted, numerous operations that correspond to LLVM IR
intrinsics are replicated to produce versions operating on either LLVM dialect
or built-in types leading to quasi-duplicate dialects, lowering to the LLVM
dialect is essentially required to be one-shot because of type conversion, etc.
In this light, it is reasonable to trade off some local complexity in the
internal implementation of LLVM dialect types for removing larger-scale system
complexity. Previous commits to the LLVM dialect type system have adapted the
API to support types from other dialects.
Replace LLVMIntegerType with the built-in IntegerType plus additional checks
that such types are signless (these are isolated in a utility function that
replaced `isa<LLVMType>` and in the parser). Temporarily keep the possibility
to parse `!llvm.i32` as a synonym for `i32`, but add a deprecation notice.
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94178
2021-01-06 23:19:04 +08:00
|
|
|
%13 = llvm.mlir.constant(1 : index) : i64
|
[mlir] replace LLVM dialect float types with built-ins
Continue the convergence between LLVM dialect and built-in types by replacing
the bfloat, half, float and double LLVM dialect types with their built-in
counterparts. At the API level, this is a direct replacement. At the syntax
level, we change the keywords to `bf16`, `f16`, `f32` and `f64`, respectively,
to be compatible with the built-in type syntax. The old keywords can still be
parsed but produce a deprecation warning and will be eventually removed.
Depends On D94178
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94179
2021-01-06 23:21:08 +08:00
|
|
|
%14 = llvm.insertvalue %0, %12[3, 0] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%15 = llvm.insertvalue %13, %14[4, 0] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
[mlir] replace LLVMIntegerType with built-in integer type
The LLVM dialect type system has been closed until now, i.e. did not support
types from other dialects inside containers. While this has had obvious
benefits of deriving from a common base class, it has led to some simple types
being almost identical with the built-in types, namely integer and floating
point types. This in turn has led to a lot of larger-scale complexity: simple
types must still be converted, numerous operations that correspond to LLVM IR
intrinsics are replicated to produce versions operating on either LLVM dialect
or built-in types leading to quasi-duplicate dialects, lowering to the LLVM
dialect is essentially required to be one-shot because of type conversion, etc.
In this light, it is reasonable to trade off some local complexity in the
internal implementation of LLVM dialect types for removing larger-scale system
complexity. Previous commits to the LLVM dialect type system have adapted the
API to support types from other dialects.
Replace LLVMIntegerType with the built-in IntegerType plus additional checks
that such types are signless (these are isolated in a utility function that
replaced `isa<LLVMType>` and in the parser). Temporarily keep the possibility
to parse `!llvm.i32` as a synonym for `i32`, but add a deprecation notice.
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94178
2021-01-06 23:19:04 +08:00
|
|
|
%16 = llvm.mlir.constant(1 : index) : i64
|
[mlir] replace LLVM dialect float types with built-ins
Continue the convergence between LLVM dialect and built-in types by replacing
the bfloat, half, float and double LLVM dialect types with their built-in
counterparts. At the API level, this is a direct replacement. At the syntax
level, we change the keywords to `bf16`, `f16`, `f32` and `f64`, respectively,
to be compatible with the built-in type syntax. The old keywords can still be
parsed but produce a deprecation warning and will be eventually removed.
Depends On D94178
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94179
2021-01-06 23:21:08 +08:00
|
|
|
%17 = llvm.extractvalue %15[0] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%18 = llvm.extractvalue %15[1] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%19 = llvm.extractvalue %15[2] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%20 = llvm.extractvalue %15[3, 0] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%21 = llvm.extractvalue %15[4, 0] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
2020-04-16 03:02:41 +08:00
|
|
|
llvm.call @vulkanLaunch(%16, %16, %16, %17, %18, %19, %20, %21) {spirv_blob = "\03\02#\07\00", spirv_entry_point = "kernel"}
|
[mlir] replace LLVM dialect float types with built-ins
Continue the convergence between LLVM dialect and built-in types by replacing
the bfloat, half, float and double LLVM dialect types with their built-in
counterparts. At the API level, this is a direct replacement. At the syntax
level, we change the keywords to `bf16`, `f16`, `f32` and `f64`, respectively,
to be compatible with the built-in type syntax. The old keywords can still be
parsed but produce a deprecation warning and will be eventually removed.
Depends On D94178
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94179
2021-01-06 23:21:08 +08:00
|
|
|
: (i64, i64, i64, !llvm.ptr<f32>, !llvm.ptr<f32>, i64, i64, i64) -> ()
|
2020-03-11 03:17:07 +08:00
|
|
|
llvm.return
|
2020-02-14 02:29:13 +08:00
|
|
|
}
|
[mlir] replace LLVM dialect float types with built-ins
Continue the convergence between LLVM dialect and built-in types by replacing
the bfloat, half, float and double LLVM dialect types with their built-in
counterparts. At the API level, this is a direct replacement. At the syntax
level, we change the keywords to `bf16`, `f16`, `f32` and `f64`, respectively,
to be compatible with the built-in type syntax. The old keywords can still be
parsed but produce a deprecation warning and will be eventually removed.
Depends On D94178
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94179
2021-01-06 23:21:08 +08:00
|
|
|
llvm.func @vulkanLaunch(%arg0: i64, %arg1: i64, %arg2: i64, %arg6: !llvm.ptr<f32>, %arg7: !llvm.ptr<f32>, %arg8: i64, %arg9: i64, %arg10: i64) {
|
|
|
|
|
%0 = llvm.mlir.undef : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%1 = llvm.insertvalue %arg6, %0[0] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%2 = llvm.insertvalue %arg7, %1[1] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%3 = llvm.insertvalue %arg8, %2[2] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%4 = llvm.insertvalue %arg9, %3[3, 0] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
|
|
|
|
%5 = llvm.insertvalue %arg10, %4[4, 0] : !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>
|
[mlir] replace LLVMIntegerType with built-in integer type
The LLVM dialect type system has been closed until now, i.e. did not support
types from other dialects inside containers. While this has had obvious
benefits of deriving from a common base class, it has led to some simple types
being almost identical with the built-in types, namely integer and floating
point types. This in turn has led to a lot of larger-scale complexity: simple
types must still be converted, numerous operations that correspond to LLVM IR
intrinsics are replicated to produce versions operating on either LLVM dialect
or built-in types leading to quasi-duplicate dialects, lowering to the LLVM
dialect is essentially required to be one-shot because of type conversion, etc.
In this light, it is reasonable to trade off some local complexity in the
internal implementation of LLVM dialect types for removing larger-scale system
complexity. Previous commits to the LLVM dialect type system have adapted the
API to support types from other dialects.
Replace LLVMIntegerType with the built-in IntegerType plus additional checks
that such types are signless (these are isolated in a utility function that
replaced `isa<LLVMType>` and in the parser). Temporarily keep the possibility
to parse `!llvm.i32` as a synonym for `i32`, but add a deprecation notice.
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94178
2021-01-06 23:19:04 +08:00
|
|
|
%6 = llvm.mlir.constant(1 : index) : i64
|
[mlir] replace LLVM dialect float types with built-ins
Continue the convergence between LLVM dialect and built-in types by replacing
the bfloat, half, float and double LLVM dialect types with their built-in
counterparts. At the API level, this is a direct replacement. At the syntax
level, we change the keywords to `bf16`, `f16`, `f32` and `f64`, respectively,
to be compatible with the built-in type syntax. The old keywords can still be
parsed but produce a deprecation warning and will be eventually removed.
Depends On D94178
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94179
2021-01-06 23:21:08 +08:00
|
|
|
%7 = llvm.alloca %6 x !llvm.struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)> : (i64) -> !llvm.ptr<struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>>
|
|
|
|
|
llvm.store %5, %7 : !llvm.ptr<struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>>
|
|
|
|
|
llvm.call @_mlir_ciface_vulkanLaunch(%arg0, %arg1, %arg2, %7) : (i64, i64, i64, !llvm.ptr<struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>>) -> ()
|
2020-03-17 19:49:00 +08:00
|
|
|
llvm.return
|
|
|
|
|
}
|
[mlir] replace LLVM dialect float types with built-ins
Continue the convergence between LLVM dialect and built-in types by replacing
the bfloat, half, float and double LLVM dialect types with their built-in
counterparts. At the API level, this is a direct replacement. At the syntax
level, we change the keywords to `bf16`, `f16`, `f32` and `f64`, respectively,
to be compatible with the built-in type syntax. The old keywords can still be
parsed but produce a deprecation warning and will be eventually removed.
Depends On D94178
Reviewed By: mehdi_amini, silvas, antiagainst
Differential Revision: https://reviews.llvm.org/D94179
2021-01-06 23:21:08 +08:00
|
|
|
llvm.func @_mlir_ciface_vulkanLaunch(i64, i64, i64, !llvm.ptr<struct<(ptr<f32>, ptr<f32>, i64, array<1 x i64>, array<1 x i64>)>>)
|
2020-02-14 02:29:13 +08:00
|
|
|
}
|