Skip to content

onednn/fc_kernel.cc add use_onednn [fluid_ops] #74325

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Aug 1, 2025
Merged

onednn/fc_kernel.cc add use_onednn [fluid_ops] #74325

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
fbf6e8f
Fix
co63oc Jul 31, 2025
589a883
Fix
co63oc Jul 31, 2025
File filter

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
41 changes: 25 additions & 16 deletions paddle/phi/kernels/fusion/onednn/fc_kernel.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -574,6 +574,10 @@ void FCKernel(const Context& dev_ctx,
dev_ctx.HasDnnAttr("use_mkldnn")
? PADDLE_GET_CONST(bool, dev_ctx.GetDnnAttr("use_mkldnn"))
: false;
const bool use_onednn =
(!use_mkldnn && dev_ctx.HasDnnAttr("use_onednn"))
? PADDLE_GET_CONST(bool, dev_ctx.GetDnnAttr("use_onednn"))
: use_mkldnn;
const bool use_quantizer =
dev_ctx.HasDnnAttr("use_quantizer")
? PADDLE_GET_CONST(bool, dev_ctx.GetDnnAttr("use_quantizer"))
Expand All @@ -583,6 +587,11 @@ void FCKernel(const Context& dev_ctx,
? PADDLE_GET_CONST(std::string,
dev_ctx.GetDnnAttr("mkldnn_data_type"))
: "float32";
const std::string onednn_data_type =
(use_onednn && dev_ctx.HasDnnAttr("onednn_data_type"))
? PADDLE_GET_CONST(std::string,
dev_ctx.GetDnnAttr("onednn_data_type"))
: mkldnn_data_type;
const float scale_in =
dev_ctx.HasDnnAttr("Scale_in")
? PADDLE_GET_CONST(float, dev_ctx.GetDnnAttr("Scale_in"))
Expand All @@ -601,24 +610,24 @@ void FCKernel(const Context& dev_ctx,
dev_ctx.HasDnnAttr("force_fp32_output")
? PADDLE_GET_CONST(bool, dev_ctx.GetDnnAttr("force_fp32_output"))
: false;
std::vector<std::string> mkldnn_data_type_list = {
std::vector<std::string> onednn_data_type_list = {
"float32", "int8", "bfloat16"};
PADDLE_ENFORCE_EQ(std::find(mkldnn_data_type_list.begin(),
mkldnn_data_type_list.end(),
mkldnn_data_type) != mkldnn_data_type_list.end(),
PADDLE_ENFORCE_EQ(std::find(onednn_data_type_list.begin(),
onednn_data_type_list.end(),
onednn_data_type) != onednn_data_type_list.end(),
true,
common::errors::InvalidArgument(
"The mkldnn_data_type should be [float32, "
"The onednn_data_type should be [float32, "
"int8, bfloat16], but found %s.",
mkldnn_data_type.c_str()));
onednn_data_type.c_str()));
auto in_dims = input.dims();
if (use_mkldnn) {
if (use_onednn) {
PADDLE_ENFORCE_EQ(
in_dims.size() >= 2 && in_dims.size() <= 4,
true,
common::errors::Unimplemented(
"The Input of fc is expected to be a 2-D, 3-D or 4-D tensor when "
"use_mkldnn is set. But received the number of Input's "
"use_onednn is set. But received the number of Input's "
"dimensions is %d, Input's shape is %s.",
in_dims.size(),
in_dims));
Expand All @@ -632,10 +641,10 @@ void FCKernel(const Context& dev_ctx,
bias,
in_num_col_dims,
activation_type,
use_mkldnn,
use_onednn,
padding_weights,
use_quantizer,
mkldnn_data_type,
onednn_data_type,
scale_in,
scale_weights,
scale_out,
Expand All @@ -649,10 +658,10 @@ void FCKernel(const Context& dev_ctx,
bias,
in_num_col_dims,
activation_type,
use_mkldnn,
use_onednn,
padding_weights,
use_quantizer,
mkldnn_data_type,
onednn_data_type,
scale_in,
scale_weights,
scale_out,
Expand All @@ -665,10 +674,10 @@ void FCKernel(const Context& dev_ctx,
bias,
in_num_col_dims,
activation_type,
use_mkldnn,
use_onednn,
padding_weights,
use_quantizer,
mkldnn_data_type,
onednn_data_type,
scale_in,
scale_weights,
scale_out,
Expand All @@ -682,10 +691,10 @@ void FCKernel(const Context& dev_ctx,
bias,
in_num_col_dims,
activation_type,
use_mkldnn,
use_onednn,
padding_weights,
use_quantizer,
mkldnn_data_type,
onednn_data_type,
scale_in,
scale_weights,
scale_out,
Expand Down