Skip to content

UPSTREAM PR #17725: CANN: In the ROPE operator, yarn_ramp uses cache #427

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

Open
loci-dev wants to merge 1 commit into
base: main
Choose a base branch
from
Open

UPSTREAM PR #17725: CANN: In the ROPE operator, yarn_ramp uses cache #427

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
CANN: In the ROPE operator, yarn_ramp uses cache
  • Loading branch information
@TianHao324
TianHao324 committed Dec 4, 2025
commit e3f00322de144c19f9d7cdfedfef47f7d26b73e2
14 changes: 8 additions & 6 deletions ggml/src/ggml-cann/aclnn_ops.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2338,20 +2338,20 @@ static void aclnn_rope_cache_init(ggml_backend_cann_context & ctx,

// Step1.2: prepare rope_yarn_ramp, if this part updated, should update theta_scale_tensor.
bool yarn_ramp_tensor_updated = false;
ggml_cann_pool_alloc yarn_ramp_allocator(ctx.pool());
acl_tensor_ptr acl_yarn_ramp_tensor;
if (ext_factor != 0 &&
// TODO: check more parameter.
(ctx.rope_cache.theta_scale_length != theta_scale_length || ctx.rope_cache.freq_scale != freq_scale)) {
yarn_ramp_tensor_updated = true;

if (ctx.rope_cache.yarn_ramp_cache != nullptr) {
ACL_CHECK(aclrtFree(ctx.rope_cache.yarn_ramp_cache));
}
ACL_CHECK(aclrtMalloc(&ctx.rope_cache.yarn_ramp_cache, theta_scale_length * sizeof(float), ACL_MEM_MALLOC_HUGE_FIRST));
// -rope_yarn_ramp
// const float y = (i0 / 2 - low) / MAX(0.001f, high - low);
// return MIN(1, MAX(0, y)) - 1;
yarn_ramp_allocator.alloc(theta_scale_length * sizeof(float));
void * yarn_ramp_buffer = yarn_ramp_allocator.get();
acl_yarn_ramp_tensor =
ggml_cann_create_tensor(yarn_ramp_buffer, ACL_FLOAT, sizeof(float), theta_scale_ne, theta_scale_nb, 1);
ggml_cann_create_tensor(ctx.rope_cache.yarn_ramp_cache, ACL_FLOAT, sizeof(float), theta_scale_ne, theta_scale_nb, 1);
float zero_value = 0, one_value = 1;
float denom_safe_value = MAX(0.001f, corr_dims[1] - corr_dims[0]);
acl_scalar_ptr low = ggml_cann_create_scalar(&corr_dims[0], aclDataType::ACL_FLOAT);
Expand Down Expand Up @@ -2381,8 +2381,10 @@ static void aclnn_rope_cache_init(ggml_backend_cann_context & ctx,
acl_scalar_ptr freq_scale_1_sc = ggml_cann_create_scalar(&freq_scale_1, aclDataType::ACL_FLOAT);
GGML_CANN_CALL_ACLNN_OP(ctx, InplaceMuls, acl_yarn_ramp_tensor.get(), freq_scale_1_sc.get());
GGML_CANN_CALL_ACLNN_OP(ctx, InplaceAdds, acl_yarn_ramp_tensor.get(), freq_scale_sc.get(), one.get());
} else {
acl_yarn_ramp_tensor =
ggml_cann_create_tensor(ctx.rope_cache.yarn_ramp_cache, ACL_FLOAT, sizeof(float), theta_scale_ne, theta_scale_nb, 1);
}

// Step 1.3: update theta_scale_tensor according to ext_factor or freq_scale.
if (ext_factor != 0) {
if (theta_scale_updated || yarn_ramp_tensor_updated) {
Expand Down
8 changes: 6 additions & 2 deletions ggml/src/ggml-cann/common.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -315,9 +315,12 @@ struct ggml_cann_rope_cache {
if (theta_scale_exp_host) {
free(theta_scale_exp_host);
}
if(position_select_index_host) {
if (position_select_index_host) {
free(position_select_index_host);
}
if (yarn_ramp_cache) {
ACL_CHECK(aclrtFree(yarn_ramp_cache));
}
}

bool equal(int64_t theta_scale_length,
Expand All @@ -340,7 +343,7 @@ struct ggml_cann_rope_cache {

void set(int64_t theta_scale_length,
int64_t position_length,
float ext_factor,
float ext_factor,
float theta_scale,
float freq_scale,
float attn_factor,
Expand Down Expand Up @@ -370,6 +373,7 @@ struct ggml_cann_rope_cache {
float * theta_scale_exp_host = nullptr;
int * position_select_index_host = nullptr;
void * position_select_index = nullptr;
void * yarn_ramp_cache = nullptr;
// sin/cos cache, used only to accelerate first layer on each device
void * sin_cache = nullptr;
void * cos_cache = nullptr;
Expand Down