vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-05-22 03:10:03 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

do not use vmm for ggml v3

Browse source
This commit is contained in:
Concedo 2024-01-20 11:30:22 +08:00
parent 9ab904562e
commit 21f0ce2502
1 changed files with 2 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

12
otherarch/ggml_v3-cuda.cu
View file

@ -7434,19 +7434,11 @@ static void ggml_v3_cuda_pool_free_vmm(int device, void * ptr, size_t size) {
}
static void * ggml_v3_cuda_pool_malloc(int device, size_t size, size_t * actual_size) {
if (g_device_caps[device].vmm) {
return ggml_v3_cuda_pool_malloc_vmm(device, size, actual_size);
} else {
return ggml_v3_cuda_pool_malloc_leg(device, size, actual_size);
}
return ggml_v3_cuda_pool_malloc_leg(device, size, actual_size);
}
static void ggml_v3_cuda_pool_free(int device, void * ptr, size_t size) {
if (g_device_caps[device].vmm) {
ggml_v3_cuda_pool_free_vmm(device, ptr, size);
} else {
ggml_v3_cuda_pool_free_leg(device, ptr, size);
}
ggml_v3_cuda_pool_free_leg(device, ptr, size);
}
#else
#define ggml_v3_cuda_pool_malloc ggml_v3_cuda_pool_malloc_leg