happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

sampling : minor cleanup

This commit is contained in:
Georgi Gerganov 2025-12-03 15:39:44 +02:00
parent aad5a6afd7
commit cce3b2a8ad
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 8 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
src/llama-sampling.cpp
View File

@ -1047,8 +1047,8 @@ static void llama_sampler_top_k_free(struct llama_sampler * smpl) {
}
static void llama_sampler_top_k_backend_init(
struct llama_sampler * smpl,
ggml_backend_buffer_type_t buft) {
struct llama_sampler * smpl,
ggml_backend_buffer_type_t buft) {
auto * ctx_data = (llama_sampler_top_k *) smpl->ctx;
ctx_data->device = ggml_backend_buft_get_device(buft);
}
@ -1117,9 +1117,6 @@ struct llama_sampler_top_p {
const size_t min_keep;
std::vector<llama_token_data> buf_sort;
// Only required for checking operation support and can be removed later.
ggml_backend_dev_t device;
};
static const char * llama_sampler_top_p_name(const struct llama_sampler * /*smpl*/) {
@ -1193,8 +1190,8 @@ static void llama_sampler_top_p_free(struct llama_sampler * smpl) {
static void llama_sampler_top_p_backend_init(
struct llama_sampler * smpl,
ggml_backend_buffer_type_t buft) {
auto * sctx = (llama_sampler_top_p *) smpl->ctx;
sctx->device = ggml_backend_buft_get_device(buft);
GGML_UNUSED(smpl);
GGML_UNUSED(buft);
}
static void llama_sampler_top_p_backend_apply(
@ -1299,7 +1296,6 @@ struct llama_sampler * llama_sampler_init_top_p(float p, size_t min_keep) {
/* .p = */ p,
/* .min_keep = */ min_keep,
/* .buf_sort = */ {},
/* .device = */ nullptr,
}
);
}
@ -1309,9 +1305,6 @@ struct llama_sampler * llama_sampler_init_top_p(float p, size_t min_keep) {
struct llama_sampler_min_p {
const float p;
const size_t min_keep;
// Only required for checking operation support and can be removed later.
ggml_backend_dev_t device;
};
static const char * llama_sampler_min_p_name(const struct llama_sampler * /*smpl*/) {
@ -1384,8 +1377,8 @@ static void llama_sampler_min_p_free(struct llama_sampler * smpl) {
static void llama_sampler_min_p_backend_init(
struct llama_sampler * smpl,
ggml_backend_buffer_type_t buft) {
auto * sctx = (llama_sampler_min_p *) smpl->ctx;
sctx->device = ggml_backend_buft_get_device(buft);
GGML_UNUSED(smpl);
GGML_UNUSED(buft);
}
static void llama_sampler_min_p_backend_apply(
@ -1456,7 +1449,6 @@ struct llama_sampler * llama_sampler_init_min_p(float p, size_t min_keep) {
/* .ctx = */ new llama_sampler_min_p {
/* .p = */ p,
/* .min_keep = */ min_keep,
/* .device = */ nullptr,
}
);
}
@ -3105,8 +3097,8 @@ static void llama_sampler_logit_bias_backend_set_input(struct llama_sampler * sm
}
static void llama_sampler_logit_bias_backend_init(
struct llama_sampler * smpl,
ggml_backend_buffer_type_t buft) {
struct llama_sampler * smpl,
ggml_backend_buffer_type_t buft) {
auto * sctx = (llama_sampler_logit_bias *) smpl->ctx;
if (sctx->logit_bias.empty()) {