suport multiple cards: split-mode - layer|row

README-sycl.md

@@ -1,6 +1,7 @@
 # llama.cpp for SYCL
 
 - [Background](#background)
+- [News](#news)
 - [OS](#os)
 - [Intel GPU](#intel-gpu)
 - [Docker](#docker)
@@ -25,6 +26,16 @@ The llama.cpp for SYCL is used to support Intel GPUs.
 
 For Intel CPU, recommend to use llama.cpp for X86 (Intel MKL building).
 
+## News
+
+- 2024.3
+  - Support multiple cards: **--split-mode**: none|layer|row
+  - Support to assign main GPU by **--main-gpu**
+
+- 2024.1
+  - Create SYCL backend for Intel GPU.
+  - Support Windows build
+
 ## OS
 
 |OS|Status|Verified|
@@ -449,6 +460,7 @@ Using device **0** (Intel(R) Arc(TM) A770 Graphics) as main device
 |-|-|-|
 |GGML_SYCL_DEVICE|0 (default) or 1|Set the device id used. Check the device ids by default running output|
 |GGML_SYCL_DEBUG|0 (default) or 1|Enable log function by macro: GGML_SYCL_DEBUG|
+|ZES_ENABLE_SYSMAN| 0 (default) or 1|Support to get free memory of GPU by sycl::aspect::ext_intel_free_memory.<br>Recommended to use when --split-mode = layer|
 
 ## Known Issue
 
@@ -458,6 +470,10 @@ Using device **0** (Intel(R) Arc(TM) A770 Graphics) as main device
 
   Solution: add **--no-mmap** or **--mmap 0**.
 
+- Run with split-mode: row is slow
+
+  It's limited by the missed function: memory copy from device to device. We use dirty solution. It will be fixed in the future.
+
 ## Q&A
 
 - Error:  `error while loading shared libraries: libsycl.so.7: cannot open shared object file: No such file or directory`.

examples/llama-bench/llama-bench.cpp

@@ -125,18 +125,15 @@ static std::string get_gpu_info() {
 #ifdef GGML_USE_SYCL
     int device_list[GGML_SYCL_MAX_DEVICES];
     ggml_sycl_get_gpu_list(device_list, GGML_SYCL_MAX_DEVICES);
-
-    for (int i = 0; i < GGML_SYCL_MAX_DEVICES; i++) {
-        if (device_list[i] >0 ){
+    int count = ggml_backend_sycl_get_device_count();
+    for (int i = 0; i < count; i++) {
         char buf[128];
         ggml_sycl_get_device_description(i, buf, sizeof(buf));
         id += buf;
+        if (i < count - 1) {
             id += "/";
         }
     }
-    if (id.length() >2 ) {
-        id.pop_back();
-    }
 #endif
     // TODO: other backends
     return id;

examples/sycl/run-llama2.sh

@@ -8,12 +8,19 @@ INPUT2="Building a website can be done in 10 simple steps:\nStep 1:"
 source /opt/intel/oneapi/setvars.sh
 
 if [ $# -gt 0 ]; then
-    export GGML_SYCL_DEVICE=$1
+    GGML_SYCL_DEVICE=$1
 else
-    export GGML_SYCL_DEVICE=0
+    GGML_SYCL_DEVICE=0
 fi
-echo GGML_SYCL_DEVICE=$GGML_SYCL_DEVICE
+echo "use $GGML_SYCL_DEVICE as main GPU"
 #export GGML_SYCL_DEBUG=1
-./build/bin/main -m models/llama-2-7b.Q4_0.gguf -p "${INPUT2}" -n 400 -e -ngl 33 -s 0
-#./build/bin/main -m models/llama-2-7b.Q4_0.gguf -p "${INPUT2}" -n 5 -e -ngl 33 -t 1 -s 0
+
+
+#ZES_ENABLE_SYSMAN=1, Support to get free memory of GPU by sycl::aspect::ext_intel_free_memory. Recommended to use when --split-mode = layer.
+
+#use all GPUs with same max compute units
+ZES_ENABLE_SYSMAN=1 ./build/bin/main -m models/llama-2-7b.Q4_0.gguf -p "${INPUT2}" -n 400 -e -ngl 33 -s 0 -mg $GGML_SYCL_DEVICE
+
+#use main GPU only
+#ZES_ENABLE_SYSMAN=1 ./build/bin/main -m models/llama-2-7b.Q4_0.gguf -p "${INPUT2}" -n 400 -e -ngl 33 -s 0 -mg $GGML_SYCL_DEVICE -sm none
 

ggml-sycl.h

@@ -24,6 +24,11 @@ GGML_API ggml_backend_buffer_type_t ggml_backend_sycl_host_buffer_type(void);
 GGML_API void   ggml_backend_sycl_print_sycl_devices(void);
 GGML_API GGML_CALL void   ggml_sycl_get_gpu_list(int *id_list, int max_len);
 GGML_API GGML_CALL void   ggml_sycl_get_device_description(int device, char *description, size_t description_size);
+GGML_API GGML_CALL int   ggml_backend_sycl_get_device_count();
+GGML_API GGML_CALL ggml_backend_buffer_type_t ggml_backend_sycl_split_buffer_type(const float * tensor_split);
+GGML_API GGML_CALL void ggml_backend_sycl_get_device_memory(int device, size_t *free, size_t *total);
+GGML_API GGML_CALL int ggml_backend_sycl_get_device_index(int device_id);
+
 #ifdef  __cplusplus
 }
 #endif

llama.cpp

@@ -102,6 +102,10 @@
 #define LLAMA_MAX_NODES   8192
 #define LLAMA_MAX_EXPERTS 8
 
+#if (defined(GGML_USE_CUBLAS) || defined(GGML_USE_SYCL))
+#define GGML_USE_CUBLAS_SYCL
+#endif
+
 //
 // logging
 //
@@ -1461,6 +1465,12 @@ static ggml_backend_buffer_type_t llama_default_buffer_type_split(int fallback_g
     }
 #endif
 
+#ifdef GGML_USE_SYCL
+    if (ggml_backend_sycl_get_device_count() > 1) {
+        buft = ggml_backend_sycl_split_buffer_type(tensor_split);
+    }
+#endif
+
     if (buft == nullptr) {
         buft = llama_default_buffer_type_offload(fallback_gpu);
     }
@@ -1472,6 +1482,8 @@ static ggml_backend_buffer_type_t llama_default_buffer_type_split(int fallback_g
 static size_t llama_get_device_count() {
 #if defined(GGML_USE_CUBLAS)
     return ggml_backend_cuda_get_device_count();
+#elif defined(GGML_USE_SYCL)
+    return ggml_backend_sycl_get_device_count();
 #elif defined(GGML_USE_VULKAN)
     return ggml_backend_vk_get_device_count();
 #else
@@ -1485,6 +1497,11 @@ static size_t llama_get_device_memory(int device) {
     size_t free;
     ggml_backend_cuda_get_device_memory(device, &total, &free);
     return free;
+#elif defined(GGML_USE_SYCL)
+    size_t total;
+    size_t free;
+    ggml_backend_sycl_get_device_memory(device, &total, &free);
+    return free;
 #elif defined(GGML_USE_VULKAN)
     size_t total;
     size_t free;
@@ -11974,13 +11991,31 @@ struct llama_context * llama_new_context_with_model(
         }
 #elif defined(GGML_USE_SYCL)
         if (model->n_gpu_layers > 0) {
-            ggml_backend_t backend = ggml_backend_sycl_init(model->main_gpu);
+            // with split_mode LLAMA_SPLIT_MODE_NONE or LLAMA_SPLIT_MODE_ROW, only the main GPU backend is used
+            if (model->split_mode == LLAMA_SPLIT_MODE_NONE || model->split_mode == LLAMA_SPLIT_MODE_ROW) {
+                int main_gpu_index = ggml_backend_sycl_get_device_index(model->main_gpu);
+                ggml_backend_t backend = ggml_backend_sycl_init(main_gpu_index);
                 if (backend == nullptr) {
-                LLAMA_LOG_ERROR("%s: failed to initialize SYCL%d backend\n", __func__, model->main_gpu);
+                    LLAMA_LOG_ERROR("%s: failed to initialize SYCL%d (index %d)backend\n", __func__, model->main_gpu, main_gpu_index);
                     llama_free(ctx);
                     return nullptr;
                 }
                 ctx->backends.push_back(backend);
+            } else {
+                // LLAMA_SPLIT_LAYER requires a backend for each GPU
+                int id_list[GGML_SYCL_MAX_DEVICES];
+                ggml_sycl_get_gpu_list(id_list, GGML_SYCL_MAX_DEVICES);
+                for (int i = 0; i < ggml_backend_sycl_get_device_count(); ++i) {
+                    int device_id = id_list[i];
+                    ggml_backend_t backend = ggml_backend_sycl_init(i);
+                    if (backend == nullptr) {
+                        LLAMA_LOG_ERROR("%s: failed to initialize SYCL%d (index %d)backend\n", __func__, device_id, i);
+                        llama_free(ctx);
+                        return nullptr;
+                    }
+                    ctx->backends.push_back(backend);
+                }
+            }
         }
 #elif defined(GGML_USE_KOMPUTE)
         if (model->n_gpu_layers > 0) {