CARVIEW |
Navigation Menu
-
Notifications
You must be signed in to change notification settings - Fork 5.8k
[XPU] support weight_only_linear for xpu && fix l3 bug #64185
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.
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
Conversation
你的PR提交成功,感谢你对开源项目的贡献! |
LGTM |
@@ -331,6 +331,8 @@ if(WITH_XPU) | |||
${XPU_PASS_DEPS}) | |||
pass_library(spatial_transformer_resblock_xpu_fuse_pass inference DIR xpu | |||
DEPS ${XPU_PASS_DEPS}) | |||
pass_library(permute_int8_weight_only_pass inference DIR xpu DEPS |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
weight_only_linear_xpu_pass
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done
458b9b3
to
0b20eb7
Compare
@@ -0,0 +1,166 @@ | |||
/* Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
2024,新增文件使用当前日期
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
ok
@@ -0,0 +1,135 @@ | |||
// Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
2024
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这个文件放到paddle/phi/kernels/fusion/xpu/weight_only_linear_xpu_kernel.cc
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
ok
@@ -137,6 +137,7 @@ void MemoryOptimizePass::CollectVarMemorySize( | |||
"lod_reset", | |||
"fetch", | |||
"share_data"}; | |||
std::set<std::string> inplace_op = {"silu", "group_norm"}; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这两个为什么需要特殊处理?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
已删除
0b20eb7
to
ed30dc2
Compare
ed30dc2
to
874038b
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
PR Category
Custom Device
PR Types
Others
Description