Skip to content

optimize rope function to avoid call powf in the loop #807

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

Merged
merged 1 commit into from
Apr 14, 2023
Merged

optimize rope function to avoid call powf in the loop #807

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
22 changes: 12 additions & 10 deletions ggml.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7279,19 +7279,20 @@ static void ggml_compute_forward_rope_f32(
// row index used to determine which thread to use
int ir = 0;

const float theta_scale = powf(10000.0, ((float)-2)/n_dims);

for (int64_t i3 = 0; i3 < ne3; i3++) {
for (int64_t i2 = (mode == 0 ? 0 : n_past); i2 < ne2; i2++) {
const int p = (mode == 0 ? n_past + i2 : i2);
for (int64_t i1 = 0; i1 < ne1; i1++) {
if (ir++ < ir0) continue;
if (ir > ir1) break;

float theta = (float)p;
for (int i0 = 0; i0 < n_dims; i0 += 2) {
const float theta = powf(10000.0, ((float)-i0)/n_dims);

const float cos_theta = cosf(p*theta);
const float sin_theta = sinf(p*theta);
const float cos_theta = cosf(theta);
const float sin_theta = sinf(theta);

theta *= theta_scale;
const float * const src = (float *)((char *) src0->data + i3*nb3 + i2*nb2 + i1*nb1 + i0*nb0);
float * dst_data = (float *)((char *) dst->data + i3*nb3 + i2*nb2 + i1*nb1 + i0*nb0);

Expand Down Expand Up @@ -7352,19 +7353,20 @@ static void ggml_compute_forward_rope_f16(
// row index used to determine which thread to use
int ir = 0;

const float theta_scale = powf(10000.0, ((float)-2)/n_dims);

for (int64_t i3 = 0; i3 < ne3; i3++) {
for (int64_t i2 = (mode == 0 ? 0 : n_past); i2 < ne2; i2++) {
const int p = (mode == 0 ? n_past + i2 : i2);
for (int64_t i1 = 0; i1 < ne1; i1++) {
if (ir++ < ir0) continue;
if (ir > ir1) break;

float theta = (float)p;
for (int i0 = 0; i0 < n_dims; i0 += 2) {
const float theta = powf(10000.0, ((float)-i0)/n_dims);

const float cos_theta = cosf(p*theta);
const float sin_theta = sinf(p*theta);
const float cos_theta = cosf(theta);
const float sin_theta = sinf(theta);

theta *= theta_scale;
const ggml_fp16_t * const src = (ggml_fp16_t *)((char *) src0->data + i3*nb3 + i2*nb2 + i1*nb1 + i0*nb0);
ggml_fp16_t * dst_data = (ggml_fp16_t *)((char *) dst->data + i3*nb3 + i2*nb2 + i1*nb1 + i0*nb0);

Expand Down