pali112/ollama
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge remote-tracking branch 'origin/main' into GraniteFour

Browse Source 
* origin/main:
ggml: Report ordinal IDs for AMD GPUs on Windows
doc: add MacOS docs (#11334)
Reduce default parallelism to 1 (#11330)
API/CLI context enhancements (#11331)
add `tool_name` to api.md (#11326)
template: add tool result compatibility (#11294)
ci: modularization (#11324)
Revert "ggml: Temporarily disable reporting UUIDs"
readme: update Ollama icon size
int: add performance integration tests (#11173)
doc: add NVIDIA blackwell to supported list (#11307)
Update base image to Ubuntu 24.04 LTS (#9681)
doc: Update link for mac install (#11288)
mimic logs for layers on new engine (#11278)
readme: add NativeMind to community integrations (#11242)
tools: fix parsing tool calls with empty arguments, missing required fields (#11233)
readme: add ollama-bash-toolshed to community integrations (#11224)
This commit is contained in:
Gabe Goodhart
2025-07-10 14:01:24 -06:00
parent 34ff84df43 35fda7b4af
commit e61826c180
28 changed files with 125358 additions and 147 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
llama/patches/0015-ggml-Export-GPU-UUIDs.patch
View File

@@ -7,19 +7,19 @@ This enables matching up devices and information reported by the backend
with tools (e.g. nvidia-smi) and system management libraries (e.g. nvml).
---
ggml/include/ggml-backend.h | 1 +
ggml/src/ggml-cuda/ggml-cuda.cu | 33 ++++++++++++++++++++++++++++++++
ggml/src/ggml-cuda/ggml-cuda.cu | 39 ++++++++++++++++++++++++++++++++
ggml/src/ggml-metal/ggml-metal.m | 1 +
3 files changed, 35 insertions(+)
3 files changed, 41 insertions(+)
diff --git a/ggml/include/ggml-backend.h b/ggml/include/ggml-backend.h
index 74e46716..a880df33 100644
index 74e46716..48839339 100644
--- a/ggml/include/ggml-backend.h
+++ b/ggml/include/ggml-backend.h
@@ -152,6 +152,7 @@ extern "C" {
struct ggml_backend_dev_props {
const char * name;
const char * description;
+ const char * uuid;
+ const char * id;
size_t memory_free;
size_t memory_total;
enum ggml_backend_dev_type type;
@@ -31,17 +31,17 @@ index b6cca93f..09ce299c 100644
int device;
std::string name;
std::string description;
+ std::string uuid;
+ std::string id;
};
static const char * ggml_backend_cuda_device_get_name(ggml_backend_dev_t dev) {
@@ -2951,6 +2952,11 @@ static const char * ggml_backend_cuda_device_get_description(ggml_backend_dev_t
return ctx->description.c_str();
}
+static const char * ggml_backend_cuda_device_get_uuid(ggml_backend_dev_t dev) {
+static const char * ggml_backend_cuda_device_get_id(ggml_backend_dev_t dev) {
+ ggml_backend_cuda_device_context * ctx = (ggml_backend_cuda_device_context *)dev->context;
+ return ctx->uuid.c_str();
+ return ctx->id.c_str();
+}
+
static void ggml_backend_cuda_device_get_memory(ggml_backend_dev_t dev, size_t * free, size_t * total) {
@@ -51,17 +51,17 @@ index b6cca93f..09ce299c 100644
static void ggml_backend_cuda_device_get_props(ggml_backend_dev_t dev, ggml_backend_dev_props * props) {
props->name = ggml_backend_cuda_device_get_name(dev);
props->description = ggml_backend_cuda_device_get_description(dev);
+ props->uuid = ggml_backend_cuda_device_get_uuid(dev);
+ props->id = ggml_backend_cuda_device_get_id(dev);
props->type = ggml_backend_cuda_device_get_type(dev);
ggml_backend_cuda_device_get_memory(dev, &props->memory_free, &props->memory_total);
@@ -3535,6 +3542,32 @@ ggml_backend_reg_t ggml_backend_cuda_reg() {
CUDA_CHECK(cudaGetDeviceProperties(&prop, i));
dev_ctx->description = prop.name;
+ #if !defined(GGML_USE_HIP)
+ char uuid[64];
+ snprintf(uuid, sizeof(uuid),
+ char id[64];
+ snprintf(id, sizeof(id),
+ "GPU-%02x%02x%02x%02x-%02x%02x-%02x%02x-%02x%02x-%02x%02x%02x%02x%02x%02x",
+ (unsigned char)prop.uuid.bytes[0],
+ (unsigned char)prop.uuid.bytes[1],
@@ -80,9 +80,15 @@ index b6cca93f..09ce299c 100644
+ (unsigned char)prop.uuid.bytes[14],
+ (unsigned char)prop.uuid.bytes[15]
+ );
+ dev_ctx->uuid = uuid;
+ dev_ctx->id = id;
+ #else
+ dev_ctx->uuid = "GPU-" + std::string(prop.uuid.bytes, 16);
+ #ifdef _WIN32
+ char id[16];
+ snprintf(id, sizeof(id), "%d", i);
+ dev_ctx->id = id;
+ #else
+ dev_ctx->id = "GPU-" + std::string(prop.uuid.bytes, 16);
+ #endif
+ #endif
+
ggml_backend_dev_t dev = new ggml_backend_device {
@@ -96,7 +102,7 @@ index 74fd6654..ea2d6218 100644
static void ggml_backend_metal_device_get_props(ggml_backend_dev_t dev, struct ggml_backend_dev_props * props) {
props->name = ggml_backend_metal_device_get_name(dev);
props->description = ggml_backend_metal_device_get_description(dev);
+ props->uuid = "0";
+ props->id = "0";
props->type = ggml_backend_metal_device_get_type(dev);
ggml_backend_metal_device_get_memory(dev, &props->memory_free, &props->memory_total);
props->caps = (struct ggml_backend_dev_caps) {