Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
gaoqiong
composable_kernel_ROCM
Commits
fcb9e1b4
Commit
fcb9e1b4
authored
Dec 17, 2024
by
M.Emin Ozturk
Browse files
clang all
parent
bd0a6f73
Changes
38
Show whitespace changes
Inline
Side-by-side
Showing
18 changed files
with
175 additions
and
177 deletions
+175
-177
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_km_nk_mn_mem_v2_kpadding_instance.cpp
...eamk_bf16_bf16_bf16_km_nk_mn_mem_v2_kpadding_instance.cpp
+11
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_km_nk_mn_mem_v2_mkpadding_instance.cpp
...amk_bf16_bf16_bf16_km_nk_mn_mem_v2_mkpadding_instance.cpp
+10
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_default_instance.cpp
...streamk_bf16_bf16_bf16_mk_kn_mn_comp_default_instance.cpp
+11
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_kpadding_instance.cpp
...treamk_bf16_bf16_bf16_mk_kn_mn_comp_kpadding_instance.cpp
+9
-9
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_mnkpadding_instance.cpp
...eamk_bf16_bf16_bf16_mk_kn_mn_comp_mnkpadding_instance.cpp
+9
-9
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_mnpadding_instance.cpp
...reamk_bf16_bf16_bf16_mk_kn_mn_comp_mnpadding_instance.cpp
+9
-9
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_default_instance.cpp
...reamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_default_instance.cpp
+11
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_kpadding_instance.cpp
...eamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_kpadding_instance.cpp
+11
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_mnkpadding_instance.cpp
...mk_bf16_bf16_bf16_mk_kn_mn_mem_v1_mnkpadding_instance.cpp
+10
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_default_instance.cpp
...reamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_default_instance.cpp
+11
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_kpadding_instance.cpp
...eamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_kpadding_instance.cpp
+5
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_mnkpadding_instance.cpp
...mk_bf16_bf16_bf16_mk_kn_mn_mem_v2_mnkpadding_instance.cpp
+4
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_comp_default_instance.cpp
...streamk_bf16_bf16_bf16_mk_nk_mn_comp_default_instance.cpp
+11
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_comp_kpadding_instance.cpp
...treamk_bf16_bf16_bf16_mk_nk_mn_comp_kpadding_instance.cpp
+9
-9
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v1_default_instance.cpp
...reamk_bf16_bf16_bf16_mk_nk_mn_mem_v1_default_instance.cpp
+11
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v1_kpadding_instance.cpp
...eamk_bf16_bf16_bf16_mk_nk_mn_mem_v1_kpadding_instance.cpp
+11
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_default_instance.cpp
...reamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_default_instance.cpp
+11
-10
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_kpadding_instance.cpp
...eamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_kpadding_instance.cpp
+11
-11
No files found.
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_km_nk_mn_mem_v2_kpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_km_nk_mn_mem_v2_kpaddi
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_km_nk_mn_mem_v2_kpaddi
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_km_nk_mn_mem_instances
<
Interwave
,
GemmKPadding
>
{});
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_km_nk_mn_mem_instances
<
Interwave
,
GemmKPadding
>
{});
}
}
}
// namespace instance
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_km_nk_mn_mem_v2_mkpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_km_nk_mn_mem_v2_mkpadd
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_km_nk_mn_mem_v2_mkpadd
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_default_instance.cpp
View file @
fcb9e1b4
...
@@ -17,10 +17,11 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_default_
...
@@ -17,10 +17,11 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_default_
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_instances
<
GemmDefault
>
{});
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_instances
<
GemmDefault
>
{});
}
}
}
// namespace instance
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_kpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_kpadding
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_kpadding
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_mnkpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_mnkpaddi
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_mnkpaddi
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_mnpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_mnpaddin
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_comp_mnpaddin
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_default_instance.cpp
View file @
fcb9e1b4
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_defaul
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_defaul
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_instances
<
Intrawave
,
GemmDefault
>
{});
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_instances
<
Intrawave
,
GemmDefault
>
{});
}
}
}
// namespace instance
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_kpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_kpaddi
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_kpaddi
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_instances
<
Intrawave
,
GemmKPadding
>
{});
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_instances
<
Intrawave
,
GemmKPadding
>
{});
}
}
}
// namespace instance
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_mnkpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_mnkpad
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v1_mnkpad
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_default_instance.cpp
View file @
fcb9e1b4
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_defaul
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_defaul
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_instances
<
Interwave
,
GemmDefault
>
{});
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_instances
<
Interwave
,
GemmDefault
>
{});
}
}
}
// namespace instance
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_kpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -9,19 +9,14 @@ namespace device {
...
@@ -9,19 +9,14 @@ namespace device {
namespace
instance
{
namespace
instance
{
void
add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_kpadding_instances
(
void
add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_kpadding_instances
(
std
::
vector
<
std
::
unique_ptr
<
DeviceGemmV2
<
Row
,
std
::
vector
<
std
::
unique_ptr
<
Row
,
DeviceGemmV2
<
Row
,
Row
,
Row
,
BF16
,
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
Row
,
instances
)
BF16
,
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_instances
<
Interwave
,
GemmKPadding
>
{});
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_instances
<
Interwave
,
GemmKPadding
>
{});
}
}
}
// namespace instance
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_mnkpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -9,15 +9,9 @@ namespace device {
...
@@ -9,15 +9,9 @@ namespace device {
namespace
instance
{
namespace
instance
{
void
add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_mnkpadding_instances
(
void
add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_kn_mn_mem_v2_mnkpadding_instances
(
std
::
vector
<
std
::
unique_ptr
<
DeviceGemmV2
<
Row
,
std
::
vector
<
std
::
unique_ptr
<
Row
,
DeviceGemmV2
<
Row
,
Row
,
Row
,
BF16
,
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
Row
,
instances
)
BF16
,
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_comp_default_instance.cpp
View file @
fcb9e1b4
...
@@ -17,10 +17,11 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_comp_default_
...
@@ -17,10 +17,11 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_comp_default_
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_comp_instances
<
GemmDefault
>
{});
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_comp_instances
<
GemmDefault
>
{});
}
}
}
// namespace instance
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_comp_kpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_comp_kpadding
...
@@ -17,7 +17,7 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_comp_kpadding
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v1_default_instance.cpp
View file @
fcb9e1b4
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v1_defaul
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v1_defaul
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_instances
<
Intrawave
,
GemmDefault
>
{});
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_instances
<
Intrawave
,
GemmDefault
>
{});
}
}
}
// namespace instance
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v1_kpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v1_kpaddi
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v1_kpaddi
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_instances
<
Intrawave
,
GemmKPadding
>
{});
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_instances
<
Intrawave
,
GemmKPadding
>
{});
}
}
}
// namespace instance
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_default_instance.cpp
View file @
fcb9e1b4
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_defaul
...
@@ -17,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_defaul
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_instances
<
Interwave
,
GemmDefault
>
{});
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_instances
<
Interwave
,
GemmDefault
>
{});
}
}
}
// namespace instance
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/gemm_universal_streamk/device_gemm_xdl_universal_streamk_bf16_bf16_bf16/device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_kpadding_instance.cpp
View file @
fcb9e1b4
...
@@ -9,8 +9,7 @@ namespace device {
...
@@ -9,8 +9,7 @@ namespace device {
namespace
instance
{
namespace
instance
{
void
add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_kpadding_instances
(
void
add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_kpadding_instances
(
std
::
vector
<
std
::
unique_ptr
<
std
::
vector
<
std
::
unique_ptr
<
DeviceGemm_Streamk_V2
<
Row
,
DeviceGemm_Streamk_V2
<
Row
,
Col
,
Col
,
Row
,
Row
,
BF16
,
BF16
,
...
@@ -18,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_kpaddi
...
@@ -18,11 +17,12 @@ void add_device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_v2_kpaddi
BF16
,
BF16
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
,
PassThrough
>>>&
instances
)
PassThrough
>>>&
instances
)
{
{
add_device_operation_instances
(
add_device_operation_instances
(
instances
,
instances
,
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_instances
<
Interwave
,
GemmKPadding
>
{});
device_gemm_xdl_universal_streamk_bf16_bf16_bf16_mk_nk_mn_mem_instances
<
Interwave
,
GemmKPadding
>
{});
}
}
}
// namespace instance
}
// namespace instance
...
...
Prev
1
2
Next
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment