Skip to content

CUDA: add a prop in ggml_cuda_device_infor for distinguish iGPU or dGPU in cuda (#13856) #13895

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 10 commits into from
May 31, 2025
Merged

CUDA: add a prop in ggml_cuda_device_infor for distinguish iGPU or dGPU in cuda (#13856) #13895

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
db0444f
1. add "integrated" in ggml_cuda_device_info for distinguish whether…
Yangxiaoz May 29, 2025
99cbf62
Update ggml/src/ggml-cuda/ggml-cuda.cu
Yangxiaoz May 30, 2025
eea49ab
Update ggml/src/ggml-cuda/ggml-cuda.cu
Yangxiaoz May 30, 2025
075fae3
Update ggml/src/ggml-cuda/ggml-cuda.cu
Yangxiaoz May 30, 2025
bd21613
add a host_buft assert in case of integrated_cuda_device with func:'e…
Yangxiaoz May 30, 2025
1959c24
Update ggml/src/ggml-cuda/ggml-cuda.cu
Yangxiaoz May 30, 2025
308886a
Update ggml/src/ggml-cuda/ggml-cuda.cu
Yangxiaoz May 30, 2025
63db683
revoke the suggest commit changes due to it's not applicable in jetso…
Yangxiaoz May 30, 2025
73472d1
Update ggml/src/ggml-cuda/ggml-cuda.cu
Yangxiaoz May 30, 2025
2ab5def
Update ggml/src/ggml-cuda/ggml-cuda.cu
Yangxiaoz May 30, 2025
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
1 change: 1 addition & 0 deletions ggml/src/ggml-cuda/common.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -635,6 +635,7 @@ struct ggml_cuda_device_info {
int nsm; // number of streaming multiprocessors
size_t smpb; // max. shared memory per block
size_t smpbo; // max. shared memory per block (with opt-in)
bool integrated; // Device is integrated as opposed to discrete
bool vmm; // virtual memory support
size_t vmm_granularity; // granularity of virtual memory
size_t total_vram;
Expand Down
14 changes: 12 additions & 2 deletions ggml/src/ggml-cuda/ggml-cuda.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -243,7 +243,7 @@ static ggml_cuda_device_info ggml_cuda_init() {

info.default_tensor_split[id] = total_vram;
total_vram += prop.totalGlobalMem;

info.devices[id].integrated= prop.integrated;
info.devices[id].nsm = prop.multiProcessorCount;
info.devices[id].smpb = prop.sharedMemPerBlock;
info.devices[id].warp_size = prop.warpSize;
Expand Down Expand Up @@ -1065,6 +1065,10 @@ static const char * ggml_backend_cuda_host_buffer_type_name(ggml_backend_buffer_
GGML_UNUSED(buft);
}

static bool ggml_backend_buft_is_cuda_host(ggml_backend_buffer_type_t buft) {
return buft->iface.get_name == ggml_backend_cuda_host_buffer_type_name;
}

static void ggml_backend_cuda_host_buffer_free_buffer(ggml_backend_buffer_t buffer) {
CUDA_CHECK(cudaFreeHost(buffer->context));
}
Expand Down Expand Up @@ -3263,7 +3267,13 @@ static bool ggml_backend_cuda_device_supports_op(ggml_backend_dev_t dev, const g
}

static bool ggml_backend_cuda_device_supports_buft(ggml_backend_dev_t dev, ggml_backend_buffer_type_t buft) {
return (ggml_backend_buft_is_cuda(buft) || ggml_backend_buft_is_cuda_split(buft)) && buft->device == dev;
ggml_backend_cuda_device_context * dev_ctx = (ggml_backend_cuda_device_context *) dev->context;
const int integrated = ggml_cuda_info().devices[dev_ctx->device].integrated;
if(integrated){
return (ggml_backend_buft_is_cuda(buft) || ggml_backend_buft_is_cuda_split(buft) ||ggml_backend_buft_is_cuda_host(buft)) && buft->device == dev;
}else{
return (ggml_backend_buft_is_cuda(buft) || ggml_backend_buft_is_cuda_split(buft)) && buft->device == dev;
}
}

static int64_t get_op_batch_size(const ggml_tensor * op) {
Expand Down