vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

ggml_sycl_compute_forward: fixup function calling names and remove comments

Browse source
This commit is contained in:
Akarshan Biswas 2025-02-01 09:40:44 +05:30
parent 5288bd5896
commit a153f1972d
No known key found for this signature in database
GPG key ID: 52A578A14B32134D
1 changed files with 50 additions and 50 deletions
Download patch file Download diff file Expand all files Collapse all files

100
ggml/src/ggml-sycl/ggml-sycl.cpp
View file

@ -3665,138 +3665,138 @@ bool ggml_sycl_compute_forward(ggml_backend_sycl_context & ctx, struct ggml_tens
switch (dst->op) { switch (dst->op) {
case GGML_OP_ARGMAX: case GGML_OP_ARGMAX:
ggml_sycl_op_argmax(ctx, dst); // done ggml_sycl_argmax(ctx, dst);
break; break;
case GGML_OP_CONV_TRANSPOSE_1D: case GGML_OP_CONV_TRANSPOSE_1D:
ggml_sycl_op_conv_transpose_1d(ctx, dst); // already good ggml_sycl_op_conv_transpose_1d(ctx, dst);
break; break;
case GGML_OP_REPEAT: case GGML_OP_REPEAT:
ggml_sycl_repeat(ctx, dst); // partially done ggml_sycl_repeat(ctx, dst);
break; break;
case GGML_OP_GET_ROWS: case GGML_OP_GET_ROWS:
ggml_sycl_op_get_rows(ctx, dst); // done ggml_sycl_op_get_rows(ctx, dst);
break; break;
case GGML_OP_DUP: case GGML_OP_DUP:
ggml_sycl_dup(ctx, dst); // done ggml_sycl_dup(ctx, dst);
break; break;
case GGML_OP_ADD: case GGML_OP_ADD:
case GGML_OP_ADD1: // TODO: more efficient implementation case GGML_OP_ADD1: // TODO: more efficient implementation
ggml_sycl_add(ctx, dst); // partially done ggml_sycl_add(ctx, dst);
break; break;
case GGML_OP_SUB: case GGML_OP_SUB:
ggml_sycl_sub(ctx, dst); // partially done ggml_sycl_sub(ctx, dst);
break; break;
case GGML_OP_ACC: case GGML_OP_ACC:
ggml_sycl_acc(ctx, dst); // fully done ggml_sycl_acc(ctx, dst);
break; break;
case GGML_OP_MUL: case GGML_OP_MUL:
ggml_sycl_mul(ctx, dst); // partially done ggml_sycl_mul(ctx, dst);
break; break;
case GGML_OP_LOG: case GGML_OP_LOG:
ggml_sycl_log(ctx, dst); // fully done ggml_sycl_log(ctx, dst);
break; break;
case GGML_OP_DIV: case GGML_OP_DIV:
ggml_sycl_div(ctx, dst); // partially done ggml_sycl_div(ctx, dst);
break; break;
case GGML_OP_UNARY: case GGML_OP_UNARY:
switch (ggml_get_unary_op(dst)) { switch (ggml_get_unary_op(dst)) {
case GGML_UNARY_OP_NEG: case GGML_UNARY_OP_NEG:
ggml_sycl_neg(ctx, dst); // done ggml_sycl_neg(ctx, dst);
break; break;
case GGML_UNARY_OP_STEP: case GGML_UNARY_OP_STEP:
ggml_sycl_step(ctx, dst); // done ggml_sycl_step(ctx, dst);
break; break;
case GGML_UNARY_OP_GELU: case GGML_UNARY_OP_GELU:
ggml_sycl_gelu(ctx, dst); // done ggml_sycl_gelu(ctx, dst);
break; break;
case GGML_UNARY_OP_SILU: case GGML_UNARY_OP_SILU:
ggml_sycl_silu(ctx, dst); // done ggml_sycl_silu(ctx, dst);
break; break;
case GGML_UNARY_OP_GELU_QUICK: case GGML_UNARY_OP_GELU_QUICK:
ggml_sycl_gelu_quick(ctx, dst); // done ggml_sycl_gelu_quick(ctx, dst);
break; break;
case GGML_UNARY_OP_TANH: case GGML_UNARY_OP_TANH:
ggml_sycl_tanh(ctx, dst); // done ggml_sycl_tanh(ctx, dst);
break; break;
case GGML_UNARY_OP_RELU: case GGML_UNARY_OP_RELU:
ggml_sycl_relu(ctx, dst); // done ggml_sycl_relu(ctx, dst);
break; break;
case GGML_UNARY_OP_SIGMOID: case GGML_UNARY_OP_SIGMOID:
ggml_sycl_sigmoid(ctx, dst); // done ggml_sycl_sigmoid(ctx, dst);
break; break;
case GGML_UNARY_OP_HARDSIGMOID: case GGML_UNARY_OP_HARDSIGMOID:
ggml_sycl_hardsigmoid(ctx, dst); // done ggml_sycl_hardsigmoid(ctx, dst);
break; break;
case GGML_UNARY_OP_HARDSWISH: case GGML_UNARY_OP_HARDSWISH:
ggml_sycl_hardswish(ctx, dst); // done ggml_sycl_hardswish(ctx, dst);
break; break;
case GGML_UNARY_OP_EXP: case GGML_UNARY_OP_EXP:
ggml_sycl_exp(ctx, dst); // done ggml_sycl_exp(ctx, dst);
break; break;
default: default:
return false; return false;
} }
break; break;
case GGML_OP_NORM: case GGML_OP_NORM:
ggml_sycl_norm(ctx, dst); // done ggml_sycl_norm(ctx, dst);
break; break;
case GGML_OP_GROUP_NORM: case GGML_OP_GROUP_NORM:
ggml_sycl_group_norm(ctx, dst); // done ggml_sycl_group_norm(ctx, dst);
break; break;
case GGML_OP_CONCAT: case GGML_OP_CONCAT:
ggml_sycl_op_concat(ctx, dst); // already good ggml_sycl_op_concat(ctx, dst);
break; break;
case GGML_OP_UPSCALE: case GGML_OP_UPSCALE:
ggml_sycl_upscale(ctx, dst); // done ggml_sycl_upscale(ctx, dst);
break; break;
case GGML_OP_PAD: case GGML_OP_PAD:
ggml_sycl_pad(ctx, dst); // done ggml_sycl_pad(ctx, dst);
break; break;
case GGML_OP_LEAKY_RELU: case GGML_OP_LEAKY_RELU:
ggml_sycl_leaky_relu(ctx, dst); // done ggml_sycl_leaky_relu(ctx, dst);
break; break;
case GGML_OP_RMS_NORM: case GGML_OP_RMS_NORM:
ggml_sycl_rms_norm(ctx, dst); // done ggml_sycl_rms_norm(ctx, dst);
break; break;
case GGML_OP_MUL_MAT: case GGML_OP_MUL_MAT:
if (dst->src[0]->ne[3] != dst->src[1]->ne[3]) { if (dst->src[0]->ne[3] != dst->src[1]->ne[3]) {
return false; return false;
} }
/* ggml_sycl_mul_mat_id is dependent on ggml_sycl_mul_mat */ /* ggml_sycl_mul_mat_id is dependent on ggml_sycl_mul_mat */
ggml_sycl_mul_mat(ctx, dst->src[0], dst->src[1], dst); // good ggml_sycl_mul_mat(ctx, dst->src[0], dst->src[1], dst);
break; break;
case GGML_OP_MUL_MAT_ID: case GGML_OP_MUL_MAT_ID:
if (dst->src[0]->ne[3] != dst->src[1]->ne[3]) { if (dst->src[0]->ne[3] != dst->src[1]->ne[3]) {
return false; return false;
} }
ggml_sycl_mul_mat_id(ctx, dst); // good ggml_sycl_mul_mat_id(ctx, dst);
break; break;
case GGML_OP_OUT_PROD: case GGML_OP_OUT_PROD:
ggml_sycl_op_out_prod(ctx, dst); // good ggml_sycl_op_out_prod(ctx, dst);
break; break;
case GGML_OP_SCALE: case GGML_OP_SCALE:
ggml_sycl_scale(ctx, dst); // done ggml_sycl_scale(ctx, dst);
break; break;
case GGML_OP_SQR: case GGML_OP_SQR:
ggml_sycl_sqr(ctx, dst); // done ggml_sycl_sqr(ctx, dst);
break; break;
case GGML_OP_SQRT: case GGML_OP_SQRT:
ggml_sycl_sqrt(ctx, dst); // done ggml_sycl_sqrt(ctx, dst);
break; break;
case GGML_OP_SIN: case GGML_OP_SIN:
ggml_sycl_sin(ctx, dst); //done ggml_sycl_sin(ctx, dst);
break; break;
case GGML_OP_COS: case GGML_OP_COS:
ggml_sycl_cos(ctx, dst); // done ggml_sycl_cos(ctx, dst);
break; break;
case GGML_OP_CLAMP: case GGML_OP_CLAMP:
ggml_sycl_clamp(ctx, dst); // done ggml_sycl_clamp(ctx, dst);
break; break;
case GGML_OP_CPY: case GGML_OP_CPY:
ggml_sycl_cpy(ctx, dst->src[0], dst->src[1]); // okayish, need check ggml_sycl_cpy(ctx, dst->src[0], dst->src[1]);
break; break;
case GGML_OP_CONT: case GGML_OP_CONT:
ggml_sycl_dup(ctx, dst); // done ggml_sycl_dup(ctx, dst);
break; break;
case GGML_OP_NONE: case GGML_OP_NONE:
case GGML_OP_RESHAPE: case GGML_OP_RESHAPE:
@ -3806,34 +3806,34 @@ bool ggml_sycl_compute_forward(ggml_backend_sycl_context & ctx, struct ggml_tens
GGML_SYCL_DEBUG("%s: Tensor NO-OP\n", __func__); GGML_SYCL_DEBUG("%s: Tensor NO-OP\n", __func__);
break; break;
case GGML_OP_DIAG_MASK_INF: case GGML_OP_DIAG_MASK_INF:
ggml_sycl_diag_mask_inf(ctx, dst); // done ggml_sycl_diag_mask_inf(ctx, dst);
break; break;
case GGML_OP_SOFT_MAX: case GGML_OP_SOFT_MAX:
ggml_sycl_op_soft_max(ctx, dst); // already good ggml_sycl_op_soft_max(ctx, dst);
break; break;
case GGML_OP_ROPE: case GGML_OP_ROPE:
ggml_sycl_rope(ctx, dst); // done ggml_sycl_rope(ctx, dst);
break; break;
case GGML_OP_IM2COL: case GGML_OP_IM2COL:
ggml_sycl_im2col(ctx, dst); // done ggml_sycl_im2col(ctx, dst);
break; break;
case GGML_OP_POOL_2D: case GGML_OP_POOL_2D:
ggml_sycl_pool2d(ctx, dst); // done ggml_sycl_pool2d(ctx, dst);
break; break;
case GGML_OP_SUM: case GGML_OP_SUM:
ggml_sycl_sum(ctx, dst); // done ggml_sycl_sum(ctx, dst);
break; break;
case GGML_OP_SUM_ROWS: case GGML_OP_SUM_ROWS:
ggml_sycl_sum_rows(ctx, dst); // done ggml_sycl_sum_rows(ctx, dst);
break; break;
case GGML_OP_ARGSORT: case GGML_OP_ARGSORT:
ggml_sycl_argsort(ctx, dst); // done ggml_sycl_argsort(ctx, dst);
break; break;
case GGML_OP_TIMESTEP_EMBEDDING: case GGML_OP_TIMESTEP_EMBEDDING:
ggml_sycl_op_timestep_embedding(ctx, dst); // already pretty good ggml_sycl_op_timestep_embedding(ctx, dst);
break; break;
case GGML_OP_RWKV_WKV6: case GGML_OP_RWKV_WKV6:
ggml_sycl_op_rwkv_wkv6(ctx, dst); // good ggml_sycl_op_rwkv_wkv6(ctx, dst);
break; break;
case GGML_OP_GATED_LINEAR_ATTN: case GGML_OP_GATED_LINEAR_ATTN:
ggml_sycl_op_gated_linear_attn(ctx, dst); ggml_sycl_op_gated_linear_attn(ctx, dst);