[mlir][linalg] Add quantized conv2d operator with FCHW,NCHW order (#107740)

This patch adds a quantized version of the `linalg.conv2d_nchw_fchw` Op.
This is the "channel-first" ordering typically used by PyTorch and
others.

Author: ubfx

mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py

@@ -876,6 +876,35 @@ def conv_2d_nhwc_fhwc_q(
     ) * (TypeFn.cast_signed(U, K[D.f, D.kh, D.kw, D.c]) - TypeFn.cast_signed(U, KZp))
 
 
+@linalg_structured_op
+def conv_2d_nchw_fchw_q(
+    I=TensorDef(T1, S.N, S.C, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW),
+    K=TensorDef(T2, S.F, S.C, S.KH, S.KW),
+    IZp=ScalarDef(I32),
+    KZp=ScalarDef(I32),
+    O=TensorDef(U, S.N, S.F, S.OH, S.OW, output=True),
+    strides=IndexAttrDef(S.SH, S.SW, default=[1, 1]),
+    dilations=IndexAttrDef(S.DH, S.DW, default=[1, 1]),
+):
+    """Performs 2-D convolution with zero point offsets.
+
+    Layout:
+      * Input: NCHW.
+      * Kernel: FCHW.
+
+    Numeric casting is performed on the operands to the inner multiply, promoting
+    them to the same data type as the accumulator/output. This includes the zero
+    point offsets common to quantized operations.
+    """
+    implements(ConvolutionOpInterface)
+    domain(D.n, D.f, D.oh, D.ow, D.c, D.kh, D.kw)
+    O[D.n, D.f, D.oh, D.ow] += (
+        TypeFn.cast_signed(
+            U, I[D.n, D.c, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW]
+        )
+        - TypeFn.cast_signed(U, IZp)
+    ) * (TypeFn.cast_signed(U, K[D.f, D.c, D.kh, D.kw]) - TypeFn.cast_signed(U, KZp))
+
 @linalg_structured_op
 def conv_2d_nchw_fchw(
     I=TensorDef(T1, S.N, S.C, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW),