Closed
Description
openedon Aug 28, 2024
What happened?
repro.mlir:72:11: error: 'linalg.conv_2d_nchw_fchw' op inferred input/output operand #1 has shape's dimension #1 to be 4, but found 3
%24 = linalg.conv_2d_nchw_fchw {dilations = dense<1> : vector<2xi64>, strides = dense<2> : vector<2xi64>} ins(%padded, %arg1 : tensor<?x?x?x?xf32>, tensor<32x3x3x3xf32>) outs(%broadcasted : tensor<?x32x?x?xf32>) -> tensor<?x32x?x?xf32>
^
repro.mlir:7:3: note: called from
func.func @torch_jit(%arg0: tensor<1x3x224x224xf32>, %arg1: tensor<32x3x3x3xf32>) -> tensor<?x32x?x?xf32> {
^
repro.mlir:72:11: note: see current operation:
%12 = "linalg.conv_2d_nchw_fchw"(%8, %5, %11) <{dilations = dense<1> : vector<2xi64>, operandSegmentSizes = array<i32: 2, 1>, strides = dense<2> : vector<2xi64>}> ({
^bb0(%arg5: f32, %arg6: f32, %arg7: f32):
%17 = "arith.mulf"(%arg5, %arg6) <{fastmath = #arith.fastmath<none>}> : (f32, f32) -> f32
%18 = "arith.addf"(%arg7, %17) <{fastmath = #arith.fastmath<none>}> : (f32, f32) -> f32
"linalg.yield"(%18) : (f32) -> ()
}) {linalg.memoized_indexing_maps = [affine_map<(d0, d1, d2, d3, d4, d5, d6) -> (d0, d4, d2 * 2 + d5, d3 * 2 + d6)>, affine_map<(d0, d1, d2, d3, d4, d5, d6) -> (d1, d4, d5, d6)>, affine_map<(d0, d1, d2, d3, d4, d5, d6) -> (d0, d1, d2, d3)>]} : (tensor<2x4x224x224xf32>, tensor<32x3x3x3xf32>, tensor<2x32x111x111xf32>) -> tensor<2x32x111x111xf32>
%24 = linalg.conv_2d_nchw_fchw {dilations = dense<1> : vector<2xi64>, strides = dense<2> : vector<2xi64>} ins(%padded, %arg1 : tensor<?x?x?x?xf32>, tensor<32x3x3x3xf32>) outs(%broadcasted : tensor<?x32x?x?xf32>) -> tensor<?x32x?x?xf32>
^
Steps to reproduce your issue
- repro: https://gist.github.com/jinchen62/57c051d783a05c99512a82f8cd80855f
iree-compile --iree-input-demote-i64-to-i32 --iree-hal-target-backends=llvm-cpu repro.mlir -o test.vmfb
What component(s) does this issue relate to?
No response
Version information
Build TOM locally
Additional context
No response
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Metadata
Assignees
Labels
Type
Projects
Status
Done