maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/mlir/test/Quantizer/matmul.mlir

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

52 lines
5.0 KiB
MLIR
Raw Normal View History

Convert the Canonicalize and CSE passes to generic Operation Passes. This allows for them to be used on other non-function, or even other function-like, operations. The algorithms are already generic, so this is simply changing the derived pass type. The majority of this change is just ensuring that the nesting of these passes remains the same, as the pass manager won't auto-nest them anymore. PiperOrigin-RevId: 276573038
2019-10-25 06:00:36 +08:00
// RUN: mlir-opt %s -quantizer-infer-quantized-types -quant-convert-const -quantizer-remove-instrumentation -pass-pipeline='func(canonicalize)' -split-input-file | FileCheck %s
Upstream the Quantizer tool (part 4). This adds the basic passes needed and ties them into mlir-opt. Also adds two specific unit tests that exercise them. Next step is a standalone quantizer tool and additional cleanup. Tested: ninja check-mlir -- PiperOrigin-RevId: 249167690
2019-05-21 09:27:38 +08:00
// ----
// A matmul without fused clamp or bias.
// CHECK-LABEL: @matmul
Modify the syntax of the the ElementsAttrs to print the type as a colon type. This is the standard syntax for types on operations, and is also already used by IntegerAttr and FloatAttr. Example: dense<5> : tensor<i32> dense<[3]> : tensor<1xi32> PiperOrigin-RevId: 255069157
2019-06-26 07:06:13 +08:00
// CHECK: %cst = constant dense{{.*}}tensor<3x5xi8>
Upstream the Quantizer tool (part 4). This adds the basic passes needed and ties them into mlir-opt. Also adds two specific unit tests that exercise them. Next step is a standalone quantizer tool and additional cleanup. Tested: ninja check-mlir -- PiperOrigin-RevId: 249167690
2019-05-21 09:27:38 +08:00
// CHECK-NEXT: %0 = "quant.qcast"(%arg0) : (tensor<300x3xf32>) -> tensor<300x3x!quant.uniform<i8:f32, 0.037564418067230126:35>>
// CHECK-NEXT: %1 = "quant.scast"(%cst) : (tensor<3x5xi8>) -> tensor<3x5x!quant.uniform<i8:f32, 0.0062823070315864236:-1>>
// CHECK-NEXT: %2 = "fxpmath.real_matmul"(%0, %1) : (tensor<300x3x!quant.uniform<i8:f32, 0.037564418067230126:35>>, tensor<3x5x!quant.uniform<i8:f32, 0.0062823070315864236:-1>>) -> tensor<300x5x!quant.uniform<i8:f32, 0.0629921259842528:-1>>
// CHECK-NEXT: %3 = "quant.dcast"(%2) : (tensor<300x5x!quant.uniform<i8:f32, 0.0629921259842528:-1>>) -> tensor<300x5xf32>
func @matmul(%arg0: tensor<300x3xf32>) -> tensor<300x5xf32> {
Change the attribute dictionary syntax to separate name and value with '='. The current syntax separates the name and value with ':', but ':' is already overloaded by several other things(e.g. trailing types). This makes the syntax difficult to parse in some situtations: Old: "foo: 10 : i32" New: "foo = 10 : i32" PiperOrigin-RevId: 255097928
2019-06-26 10:06:06 +08:00
%0 = "quant.stats"(%arg0) {layerStats = dense<[-6.123e+00, 3.45e+00]> : tensor<2xf32>} : (tensor<300x3xf32>) -> tensor<300x3xf32>
%cst = constant {name = "constant.35"} dense<[[-1.060230e-01, 1.215050e-01, 8.002390e-01, -7.688850e-01, 0.0966112986], [6.890140e-01, -4.070560e-01, -0.797852993, 3.789250e-03, -2.088810e-01], [-6.085290e-01, 2.766170e-02, 2.685570e-01, 5.774010e-01, -4.284370e-01]]> : tensor<3x5xf32>
Upstream the Quantizer tool (part 4). This adds the basic passes needed and ties them into mlir-opt. Also adds two specific unit tests that exercise them. Next step is a standalone quantizer tool and additional cleanup. Tested: ninja check-mlir -- PiperOrigin-RevId: 249167690
2019-05-21 09:27:38 +08:00
%1 = "fxpmath.real_matmul"(%0, %cst) : (tensor<300x3xf32>, tensor<3x5xf32>) -> tensor<300x5xf32>
Change the attribute dictionary syntax to separate name and value with '='. The current syntax separates the name and value with ':', but ':' is already overloaded by several other things(e.g. trailing types). This makes the syntax difficult to parse in some situtations: Old: "foo: 10 : i32" New: "foo = 10 : i32" PiperOrigin-RevId: 255097928
2019-06-26 10:06:06 +08:00
%2 = "quant.stats"(%1) {layerStats = dense<[-8.000000e+00, 8.000000e+00]> : tensor<2xf32>} : (tensor<300x5xf32>) -> tensor<300x5xf32>
Upstream the Quantizer tool (part 4). This adds the basic passes needed and ties them into mlir-opt. Also adds two specific unit tests that exercise them. Next step is a standalone quantizer tool and additional cleanup. Tested: ninja check-mlir -- PiperOrigin-RevId: 249167690
2019-05-21 09:27:38 +08:00
return %2 : tensor<300x5xf32>
}
// ----
// A matmul with fused clamp which serves as statistics for the result.
// CHECK-LABEL: @matmul_clamp
Modify the syntax of the the ElementsAttrs to print the type as a colon type. This is the standard syntax for types on operations, and is also already used by IntegerAttr and FloatAttr. Example: dense<5> : tensor<i32> dense<[3]> : tensor<1xi32> PiperOrigin-RevId: 255069157
2019-06-26 07:06:13 +08:00
// CHECK: %cst = constant dense{{.*}}tensor<3x5xi8>
Upstream the Quantizer tool (part 4). This adds the basic passes needed and ties them into mlir-opt. Also adds two specific unit tests that exercise them. Next step is a standalone quantizer tool and additional cleanup. Tested: ninja check-mlir -- PiperOrigin-RevId: 249167690
2019-05-21 09:27:38 +08:00
// CHECK-NEXT: %0 = "quant.qcast"(%arg0) : (tensor<300x3xf32>) -> tensor<300x3x!quant.uniform<i8:f32, 0.037564418067230126:35>>
// CHECK-NEXT: %1 = "quant.scast"(%cst) : (tensor<3x5xi8>) -> tensor<3x5x!quant.uniform<i8:f32, 0.0062823070315864236:-1>>
Change the attribute dictionary syntax to separate name and value with '='. The current syntax separates the name and value with ':', but ':' is already overloaded by several other things(e.g. trailing types). This makes the syntax difficult to parse in some situtations: Old: "foo: 10 : i32" New: "foo = 10 : i32" PiperOrigin-RevId: 255097928
2019-06-26 10:06:06 +08:00
// CHECK-NEXT: %2 = "fxpmath.real_matmul"(%0, %1) {clamp_max = 6.100000e+00 : f64, clamp_min = -1.225000e+01 : f64} : (tensor<300x3x!quant.uniform<i8:f32, 0.037564418067230126:35>>, tensor<3x5x!quant.uniform<i8:f32, 0.0062823070315864236:-1>>) -> tensor<300x5x!quant.uniform<i8:f32, 0.072058823529412216:42>>
Upstream the Quantizer tool (part 4). This adds the basic passes needed and ties them into mlir-opt. Also adds two specific unit tests that exercise them. Next step is a standalone quantizer tool and additional cleanup. Tested: ninja check-mlir -- PiperOrigin-RevId: 249167690
2019-05-21 09:27:38 +08:00
// CHECK-NEXT: %3 = "quant.dcast"(%2) : (tensor<300x5x!quant.uniform<i8:f32, 0.072058823529412216:42>>) -> tensor<300x5xf32>
func @matmul_clamp(%arg0: tensor<300x3xf32>) -> tensor<300x5xf32> {
Change the attribute dictionary syntax to separate name and value with '='. The current syntax separates the name and value with ':', but ':' is already overloaded by several other things(e.g. trailing types). This makes the syntax difficult to parse in some situtations: Old: "foo: 10 : i32" New: "foo = 10 : i32" PiperOrigin-RevId: 255097928
2019-06-26 10:06:06 +08:00
%0 = "quant.stats"(%arg0) {layerStats = dense<[-6.123e+00, 3.45e+00]> : tensor<2xf32>} : (tensor<300x3xf32>) -> tensor<300x3xf32>
%cst = constant {name = "constant.35"} dense<[[-1.060230e-01, 1.215050e-01, 8.002390e-01, -7.688850e-01, 0.0966112986], [6.890140e-01, -4.070560e-01, -0.797852993, 3.789250e-03, -2.088810e-01], [-6.085290e-01, 2.766170e-02, 2.685570e-01, 5.774010e-01, -4.284370e-01]]> : tensor<3x5xf32>
%1 = "fxpmath.real_matmul"(%0, %cst) {clamp_max = 6.10, clamp_min = -12.25} : (tensor<300x3xf32>, tensor<3x5xf32>) -> tensor<300x5xf32>
Upstream the Quantizer tool (part 4). This adds the basic passes needed and ties them into mlir-opt. Also adds two specific unit tests that exercise them. Next step is a standalone quantizer tool and additional cleanup. Tested: ninja check-mlir -- PiperOrigin-RevId: 249167690
2019-05-21 09:27:38 +08:00
return %1 : tensor<300x5xf32>
}
// ----
// A matmul with bias and clamp.
// CHECK-LABEL: @matmul_add_clamp
Modify the syntax of the the ElementsAttrs to print the type as a colon type. This is the standard syntax for types on operations, and is also already used by IntegerAttr and FloatAttr. Example: dense<5> : tensor<i32> dense<[3]> : tensor<1xi32> PiperOrigin-RevId: 255069157
2019-06-26 07:06:13 +08:00
// CHECK: %cst = constant dense{{.*}}tensor<3x5xi8>
// CHECK-NEXT: %cst_0 = constant dense<[14, 28, 42, 56, 69]> : tensor<5xi32>
Upstream the Quantizer tool (part 4). This adds the basic passes needed and ties them into mlir-opt. Also adds two specific unit tests that exercise them. Next step is a standalone quantizer tool and additional cleanup. Tested: ninja check-mlir -- PiperOrigin-RevId: 249167690
2019-05-21 09:27:38 +08:00
// CHECK-NEXT: %0 = "quant.qcast"(%arg0) : (tensor<300x3xf32>) -> tensor<300x3x!quant.uniform<i8:f32, 0.037564418067230126:35>>
// CHECK-NEXT: %1 = "quant.scast"(%cst) : (tensor<3x5xi8>) -> tensor<3x5x!quant.uniform<i8:f32, 0.0062823070315864236:-1>>
// CHECK-NEXT: %2 = "quant.scast"(%cst_0) : (tensor<5xi32>) -> tensor<5x!quant.uniform<i32:f32, 0.072058823529412216>>
Change the attribute dictionary syntax to separate name and value with '='. The current syntax separates the name and value with ':', but ':' is already overloaded by several other things(e.g. trailing types). This makes the syntax difficult to parse in some situtations: Old: "foo: 10 : i32" New: "foo = 10 : i32" PiperOrigin-RevId: 255097928
2019-06-26 10:06:06 +08:00
// CHECK-NEXT: %3 = "fxpmath.real_matmul_bias"(%0, %1, %2) {clamp_max = 6.100000e+00 : f64, clamp_min = -1.225000e+01 : f64} : (tensor<300x3x!quant.uniform<i8:f32, 0.037564418067230126:35>>, tensor<3x5x!quant.uniform<i8:f32, 0.0062823070315864236:-1>>, tensor<5x!quant.uniform<i32:f32, 0.072058823529412216>>) -> tensor<300x5x!quant.uniform<i8:f32, 0.072058823529412216:42>>
Upstream the Quantizer tool (part 4). This adds the basic passes needed and ties them into mlir-opt. Also adds two specific unit tests that exercise them. Next step is a standalone quantizer tool and additional cleanup. Tested: ninja check-mlir -- PiperOrigin-RevId: 249167690
2019-05-21 09:27:38 +08:00
// CHECK-NEXT: %4 = "quant.dcast"(%3) : (tensor<300x5x!quant.uniform<i8:f32, 0.072058823529412216:42>>) -> tensor<300x5xf32>
func @matmul_add_clamp(%arg0: tensor<300x3xf32>) -> tensor<300x5xf32> {
Change the attribute dictionary syntax to separate name and value with '='. The current syntax separates the name and value with ':', but ':' is already overloaded by several other things(e.g. trailing types). This makes the syntax difficult to parse in some situtations: Old: "foo: 10 : i32" New: "foo = 10 : i32" PiperOrigin-RevId: 255097928
2019-06-26 10:06:06 +08:00
%0 = "quant.stats"(%arg0) {layerStats = dense<[-6.123e+00, 3.45e+00]> : tensor<2xf32>} : (tensor<300x3xf32>) -> tensor<300x3xf32>
%cst = constant {name = "constant.35"} dense<[[-1.060230e-01, 1.215050e-01, 8.002390e-01, -7.688850e-01, 0.0966112986], [6.890140e-01, -4.070560e-01, -0.797852993, 3.789250e-03, -2.088810e-01], [-6.085290e-01, 2.766170e-02, 2.685570e-01, 5.774010e-01, -4.284370e-01]]> : tensor<3x5xf32>
%cst_0 = constant {name = "constant.37"} dense<[1.000000e+00, 2.000000e+00, 3.000000e+00, 4.000000e+00, 5.000000e+00]> : tensor<5xf32>
%1 = "fxpmath.real_matmul_bias"(%0, %cst, %cst_0) {clamp_max = 6.10, clamp_min = -12.25} : (tensor<300x3xf32>, tensor<3x5xf32>, tensor<5xf32>) -> tensor<300x5xf32>
Upstream the Quantizer tool (part 4). This adds the basic passes needed and ties them into mlir-opt. Also adds two specific unit tests that exercise them. Next step is a standalone quantizer tool and additional cleanup. Tested: ninja check-mlir -- PiperOrigin-RevId: 249167690
2019-05-21 09:27:38 +08:00
return %1 : tensor<300x5xf32>
}