[cherry-pick] updating mul and matmul with set_mem_desc and fix squee…

…ze_transpose for MKLDNN (#47951) * Fix slice bugs in MKLDNN when input dims are zeros (#46671) * fix slice bugs * fix * update code * fix * update code * updating mul and matmul with set_mem_desc (#45624) * - mul & matmul changes - fix - bs16 correction of strides * - cosmetic fixes * - lint * - fix * - fix * - format -> mem_desc * - fix * - fix * - fix * - fix * - fix * fix squueze_transpose (#47911) Co-authored-by: Jacek Czaja <jacek.czaja@intel.com>
PaddlePaddle · Nov 29, 2022 · 9e2ba9b · 9e2ba9b
1 parent 7a0b862
commit 9e2ba9b
Show file tree

Hide file tree

Showing 5 changed files with 37 additions and 32 deletions.
diff --git a/paddle/fluid/framework/ir/graph_pattern_detector.cc b/paddle/fluid/framework/ir/graph_pattern_detector.cc
@@ -1045,6 +1045,7 @@ PDNode *patterns::SeqConvEltAddRelu::operator()(
 PDNode *patterns::Squeeze2Transpose2::operator()() {
   auto *squeeze2_op_in = pattern->NewNode(squeeze2_op_in_repr())
                              ->AsInput()
+                             ->assert_has_n_outputs(1)
                              ->assert_is_op_input("squeeze2", "X");
   auto *squeeze2_op = pattern->NewNode(squeeze2_op_repr())
                           ->assert_is_op("squeeze2")

diff --git a/paddle/fluid/operators/mkldnn/matmul_v2_mkldnn_op.cc b/paddle/fluid/operators/mkldnn/matmul_v2_mkldnn_op.cc
@@ -214,10 +214,7 @@ class MatMulMKLDNNHandler
     }
     astream.wait();
 
-    auto format =
-        MKLDNNFormatForSize(out->dims().size(), dnnl::memory::format_tag::nchw);
-    out->set_format(format);
-    out->set_layout(DataLayout::kMKLDNN);
+    out->set_mem_desc(dst_memory_p->get_desc().reshape(out->dims()));
   }
 
   std::shared_ptr<dnnl::memory> AcquireDstMemory(
@@ -651,10 +648,18 @@ void ExecuteMatMulV2(const ExecutionContext &ctx,
   auto &astream = MKLDNNDeviceContext::tls().get_stream();
   matmul_p->execute(astream, matmul_args);
   astream.wait();
-  auto format =
-      MKLDNNFormatForSize(out->dims().size(), dnnl::memory::format_tag::nchw);
-  out->set_format(format);
-  out->set_layout(DataLayout::kMKLDNN);
+
+  // TODO(jczaja): Explain why int8 format of dst is ABCD and do not need
+  // permute
+  if (IsOutputFused(ctx) && !IsInt8<T_out>()) {
+    auto axis = ctx.Attr<std::vector<int>>("fused_transpose_Out");
+    auto permuted_md = dst_memory_p->get_desc().permute_axes(axis);
+    out->set_mem_desc(
+        permuted_md.reshape(phi::vectorize<int64_t>(out->dims())));
+  } else {
+    out->set_mem_desc(
+        dst_memory_p->get_desc().reshape(phi::vectorize<int64_t>(out->dims())));
+  }
 }
 
 template <typename T>
@@ -836,8 +841,7 @@ class MatMulV2GradMKLDNNKernel : public paddle::framework::OpKernel<T> {
     reduction_p->execute(astream, reduction_args);
     astream.wait();
 
-    dx->set_format(paddle::platform::GetMKLDNNFormat(
-        dst_memory_p->get_desc().reshape(squeezed_dims)));
+    dx->set_mem_desc(dst_memory_p->get_desc().reshape(squeezed_dims));
   }
 
   std::vector<int64_t> ExtendDimsWithOnes(const std::vector<int64_t> &dims,
@@ -1119,9 +1123,8 @@ void MatMulGradMKLDNNKernel<T>::ExecuteMatMulGrad(
   matmul_p->execute(astream, matmul_args);
   astream.wait();
 
-  out->set_layout(framework::DataLayout::kMKLDNN);
-  out->set_format(platform::GetMKLDNNFormat(
-      dst_memory_p->get_desc().reshape(vectorize<int64_t>(out->dims()))));
+  out->set_mem_desc(
+      dst_memory_p->get_desc().reshape(vectorize<int64_t>(out->dims())));
 }
 
 template <typename T>
@@ -1184,13 +1187,13 @@ void MatMulGradMKLDNNKernel<T>::RunKernel(const ExecutionContext &ctx) const {
   if (dx) {
     if (dx_dims != x.dims()) {
       dx->Resize(dx_dims);
-      dx->set_format(x.format());
+      dx->set_mem_desc(x.mem_desc());
     }
   }
   if (dy) {
     if (dy_dims != y.dims()) {
       dy->Resize(dy_dims);
-      dy->set_format(y.format());
+      dy->set_mem_desc(y.mem_desc());
     }
   }
 }

diff --git a/paddle/fluid/operators/mkldnn/mul_mkldnn_op.cc b/paddle/fluid/operators/mkldnn/mul_mkldnn_op.cc
@@ -221,7 +221,7 @@ class MulPrimitiveFactory {
               to_void_cast<T>(x_tmp.data<T>()));
 
       x_tmp.Resize(data->dims());
-      x_tmp.set_format(platform::GetMKLDNNFormat(dst_mdesc));
+      x_tmp.set_mem_desc(dst_mdesc);
       data_matrix = framework::ReshapeToMatrix(x_tmp, num_col_dims);
     } else {
       data_matrix = framework::ReshapeToMatrix(*data, num_col_dims);
@@ -235,11 +235,7 @@ class MulPrimitiveFactory {
                           const Tensor *in) {
     x_input_->set_data_handle(to_void_cast<XT>(in->data<XT>()));
     output_->set_data_handle(out->mutable_data<OT>(ctx.GetPlace()));
-
-    if (out->format() == MKLDNNMemoryFormat::undef) {
-      auto output_format = platform::GetMKLDNNFormat(*output_);
-      out->set_format((MKLDNNMemoryFormat)output_format);
-    }
+    out->set_mem_desc(output_->get_desc());
   }
 
   template <typename T>
@@ -272,7 +268,7 @@ class MulPrimitiveFactory {
     auto buffer_size = dst_desc.get_size();
 
     OT *output_data = output->mutable_data<OT>(ctx.GetPlace(), buffer_size);
-    output->set_format(paddle::platform::GetMKLDNNFormat(dst_desc));
+    output->set_mem_desc(dst_desc);
     return memory(dst_desc, engine_, to_void_cast<OT>(output_data));
   }
 
@@ -392,9 +388,10 @@ class MulMKLDNNINT8Kernel : public framework::OpKernel<XT> {
     if (out_dims.size() != 2) {
       out->Resize(out_dims);
     }
-    out->set_layout(DataLayout::kMKLDNN);
-    out->set_format(platform::MKLDNNFormatForSize(out_dims.size(),
-                                                  MKLDNNMemoryFormat::nchw));
+
+    auto in_md = dnnl::memory::desc(*dnnl_primitive_desc_query_md(
+        mul.get_primitive_desc(), dnnl_query_dst_md, 0));
+    out->set_mem_desc(in_md.reshape(phi::vectorize<int64_t>(out->dims())));
   }
 };
 
@@ -442,10 +439,11 @@ class MulMKLDNNKernel : public framework::OpKernel<XT> {
     matmul_p->execute(astream, matmul_args);
     astream.wait();
 
-    out->set_layout(framework::DataLayout::kMKLDNN);
-    // plain output formats are enforced inside handler
-    out->set_format(platform::MKLDNNFormatForSize(
-        out->dims().size(), dnnl::memory::format_tag::nchw));
+    // This kernel is flattening dims so then we need to unflattened version
+    // that should be set in out reshape require plain layout, but
+    // MatmulV2MKLDNNHanlder enforces one so it should work
+    out->set_mem_desc(
+        dst_memory_p->get_desc().reshape(phi::vectorize<int64_t>(out->dims())));
   }
 
  private:

diff --git a/paddle/fluid/platform/mkldnn_reuse.h b/paddle/fluid/platform/mkldnn_reuse.h
@@ -301,7 +301,8 @@ class MatMulV2MKLDNNHandler
       out_strides[i] = out_ddims[i + 1] * out_strides[i + 1];
     }
 
-    if (!IsInt8<OT>() && !IsBfloat16<OT>() && is_output_fused) {
+    // TODO(jczaja): Why not for int8??
+    if (!IsInt8<OT>() && is_output_fused) {
       out_strides = FakeTransposeStrides(out_ddims);
     }
 

diff --git a/paddle/phi/kernels/transfer_layout_kernel.cc b/paddle/phi/kernels/transfer_layout_kernel.cc
@@ -121,8 +121,10 @@ void TransferLayoutMKLDNN(const Context& dev_ctx,
       OneDNNContext::tls().set_cur_paddle_data_layout(src_layout);
     }
 
-    out->set_layout(DataLayout::ONEDNN);
-    out->set_format(out_format);
+    dnnl::memory::desc out_mem_desc(vectorize<int64_t>(out->dims()),
+                                    funcs::ToOneDNNDataType(x.dtype()),
+                                    out_format);
+    out->set_mem_desc(out_mem_desc);
   } else if (src_layout == DataLayout::ONEDNN &&
              dst_layout != DataLayout::ONEDNN) {
     // Case2 - transfrom from MKLDNN OPKernel to Non-MKLDNN OPKernel