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
flash-attention
Commits
6b4a4821
Commit
6b4a4821
authored
Jan 25, 2023
by
Tri Dao
Browse files
[FA] Remove unused variable rng_engine_inputs
parent
78b7a1dc
Changes
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
0 additions
and
2 deletions
+0
-2
csrc/flash_attn/fmha_api.cpp
csrc/flash_attn/fmha_api.cpp
+0
-2
No files found.
csrc/flash_attn/fmha_api.cpp
View file @
6b4a4821
...
...
@@ -309,7 +309,6 @@ mha_fwd(const at::Tensor &q, // total_q x num_heads x head_size, total_q
// state
// We use a custom RNG that increases the offset by batch_size * nheads * 32.
int64_t
counter_offset
=
launch_params
.
params
.
b
*
launch_params
.
params
.
h
*
32
;
at
::
PhiloxCudaState
rng_engine_inputs
;
if
(
is_dropout
)
{
// See Note [Acquire lock when using random generators]
...
...
@@ -612,7 +611,6 @@ mha_fwd_block(const at::Tensor &q, // total_q x num_heads x head_size, t
// number of times random will be generated per thread, to offset philox counter in thc random
// state
int64_t
counter_offset
=
launch_params
.
elts_per_thread
;
at
::
PhiloxCudaState
rng_engine_inputs
;
if
(
is_dropout
)
{
// See Note [Acquire lock when using random generators]
...
...
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