Cgeist outputs something strange

[hm123450]

Hello,
I came here from the Polygeist website and it says I can ask questions here (although this doesn't seem like Polygeist's GitHub:)). I tried the example given on the website, but cgeist outputted something strange and I'm not sure what to do. My command was "./bin/cgeist gemm.c --function=matmul -S". The output is different from the one on the website and has a lot of weird things. I was wondering if this is normal? I would really appreciate it if someone could answer this question.

module attributes {dlti.dl_spec = #dlti.dl_spec<#dlti.dl_entry<f64, dense<64> : vector<2xi32>>, #dlti.dl_entry<f128, dense<128> : vector<2xi32>>, #dlti.dl_entry<f16, dense<16> : vector<2xi32>>, #dlti.dl_entry<i32, dense<32> : vector<2xi32>>, #dlti.dl_entry<i8, dense<8> : vector<2xi32>>, #dlti.dl_entry<i16, dense<16> : vector<2xi32>>, #dlti.dl_entry<i1, dense<8> : vector<2xi32>>, #dlti.dl_entry<!llvm.ptr, dense<64> : vector<4xi32>>, #dlti.dl_entry<i64, dense<64> : vector<2xi32>>, #dlti.dl_entry<f80, dense<128> : vector<2xi32>>, #dlti.dl_entry<!llvm.ptr<272>, dense<64> : vector<4xi32>>, #dlti.dl_entry<!llvm.ptr<270>, dense<32> : vector<4xi32>>, #dlti.dl_entry<!llvm.ptr<271>, dense<32> : vector<4xi32>>, #dlti.dl_entry<"dlti.endianness", "little">, #dlti.dl_entry<"dlti.stack_alignment", 128 : i32>>, llvm.data_layout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128", llvm.target_triple = "x86_64-unknown-linux-gnu", "polygeist.target-cpu" = "x86-64", "polygeist.target-features" = "+cmov,+cx8,+fxsr,+mmx,+sse,+sse2,+x87", "polygeist.tune-cpu" = "generic"} {
  func.func @matmul(%arg0: memref<?x400xf32>, %arg1: memref<?x300xf32>, %arg2: memref<?x300xf32>) attributes {llvm.linkage = #llvm.linkage<external>} {
    affine.for %arg3 = 0 to 200 {
      affine.for %arg4 = 0 to 300 {
        affine.for %arg5 = 0 to 400 {
          %0 = affine.load %arg0[%arg3, %arg5] : memref<?x400xf32>
          %1 = affine.load %arg1[%arg5, %arg4] : memref<?x300xf32>
          %2 = arith.mulf %0, %1 : f32
          %3 = affine.load %arg2[%arg3, %arg4] : memref<?x300xf32>
          %4 = arith.addf %3, %2 : f32
          affine.store %4, %arg2[%arg3, %arg4] : memref<?x300xf32>
        }
      }
    }
    return
  }
}

[hm123450]

My gemm.c is same as the example on the website(polygeist.llvm.org).

#define N 200
#define M 300
#define K 400
#define DATA_TYPE float

void matmul(DATA_TYPE A[N][K], DATA_TYPE B[K][M], DATA_TYPE C[N][M]) {
int i, j, k;
for (int i = 0; i < N; i++) {
for (int j = 0; j < M; j++) {
for (int k = 0; k < K; k++) {
C[i][j] += A[i][k] * B[k][j];
}
}
}
}

[Manuel Drehwald]

Github has a code tag, I updated your first post, so it does look nicer.
But I think you want to post this on the polygeist repo, here: https://github.com/llvm/Polygeist

[hm123450]

Github has a code tag, I updated your first post, so it does look nicer. But I think you want to post this on the polygeist repo, here: https://github.com/llvm/Polygeist
Thank you very much!!!