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
Commits
1ebc21d4
Commit
1ebc21d4
authored
Sep 14, 2022
by
wangshaojie6
Browse files
fix n2 compute error
parent
336a7065
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
1 addition
and
1 deletion
+1
-1
include/ck/tensor_operation/gpu/grid/gridwise_batched_gemm_softmax_gemm_xdl_cshuffle_v1.hpp
...id/gridwise_batched_gemm_softmax_gemm_xdl_cshuffle_v1.hpp
+1
-1
No files found.
include/ck/tensor_operation/gpu/grid/gridwise_batched_gemm_softmax_gemm_xdl_cshuffle_v1.hpp
View file @
1ebc21d4
...
@@ -790,7 +790,7 @@ struct GridwiseBatchedGemmSoftmaxGemm_Xdl_CShuffle
...
@@ -790,7 +790,7 @@ struct GridwiseBatchedGemmSoftmaxGemm_Xdl_CShuffle
const
index_t
nstartxdl
=
nstart
+
n0_i
*
NPerRepeat
;
const
index_t
nstartxdl
=
nstart
+
n0_i
*
NPerRepeat
;
const
index_t
acc_idx_n0
=
acc_idx_m0
+
n0_i
*
n2
*
n4
;
const
index_t
acc_idx_n0
=
acc_idx_m0
+
n0_i
*
n2
*
n4
;
static_for
<
0
,
n2
,
1
>
{}([
&
](
auto
n2_i
)
{
static_for
<
0
,
n2
,
1
>
{}([
&
](
auto
n2_i
)
{
const
index_t
nstartgroup
=
nstartxdl
+
thread_n_cluster_id
*
n4
+
n2_i
*
n3
*
n4
;
const
index_t
nstartgroup
=
nstartxdl
+
thread_n_cluster_id
*
n4
+
n2_i
*
(
warpSize
/
MPerXdl
)
*
n4
;
const
index_t
acc_idx_n2
=
acc_idx_n0
+
n2_i
*
n4
;
const
index_t
acc_idx_n2
=
acc_idx_n0
+
n2_i
*
n4
;
static_for
<
0
,
n4
,
1
>
{}([
&
](
auto
n4_i
)
{
static_for
<
0
,
n4
,
1
>
{}([
&
](
auto
n4_i
)
{
const
index_t
n_global
=
nstartgroup
+
n4_i
;
const
index_t
n_global
=
nstartgroup
+
n4_i
;
...
...
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