0950b97393
If only HSA is enabled we dont need pciutils since AMD detection does not rely on it. Therefore we change CMakeLists.txt to build pciutils only if required. This commit has some side-effects: 1. We now don't build Intel backend by default. In other words, no backend is built by default, the user must specify which backend to use. 2. There were some issues with includes and wrongly used defines and variables. This commit fixes all that.
98 lines
1.8 KiB
C++
98 lines
1.8 KiB
C++
#ifndef __GPU__
|
|
#define __GPU__
|
|
|
|
#include <cstdint>
|
|
|
|
#define UNKNOWN_FREQ -1
|
|
|
|
enum {
|
|
GPU_VENDOR_NVIDIA,
|
|
GPU_VENDOR_AMD,
|
|
GPU_VENDOR_INTEL
|
|
};
|
|
|
|
enum {
|
|
MEMTYPE_UNKNOWN,
|
|
MEMTYPE_DDR3,
|
|
MEMTYPE_DDR4,
|
|
MEMTYPE_GDDR5,
|
|
MEMTYPE_GDDR5X,
|
|
MEMTYPE_GDDR6,
|
|
MEMTYPE_GDDR6X,
|
|
MEMTYPE_HBM2
|
|
};
|
|
|
|
typedef int32_t VENDOR;
|
|
typedef int32_t MEMTYPE;
|
|
|
|
struct cach {
|
|
int32_t size;
|
|
uint8_t num_caches;
|
|
bool exists;
|
|
};
|
|
|
|
struct cache {
|
|
struct cach* L2;
|
|
};
|
|
|
|
// CUDA topology
|
|
struct topology_c {
|
|
int32_t streaming_mp;
|
|
int32_t cores_per_mp;
|
|
int32_t cuda_cores;
|
|
int32_t tensor_cores;
|
|
};
|
|
|
|
// HSA topology
|
|
struct topology_h {
|
|
int32_t compute_units;
|
|
};
|
|
|
|
// Intel topology
|
|
struct topology_i {
|
|
int32_t slices;
|
|
int32_t subslices;
|
|
int32_t eu_subslice;
|
|
};
|
|
|
|
struct memory {
|
|
int64_t size_bytes;
|
|
MEMTYPE type;
|
|
int32_t bus_width;
|
|
int32_t freq;
|
|
int32_t clk_mul; // clock multiplier
|
|
};
|
|
|
|
struct gpu_info {
|
|
int32_t idx;
|
|
VENDOR vendor;
|
|
struct uarch* arch;
|
|
char* name;
|
|
int64_t freq;
|
|
struct pci* pci;
|
|
int64_t peak_performance;
|
|
// CUDA specific
|
|
int64_t peak_performance_tcu;
|
|
struct memory* mem;
|
|
struct cache* cach;
|
|
struct topology_c* topo_c;
|
|
// HSA specific
|
|
struct topology_h* topo_h;
|
|
// Intel specific
|
|
struct topology_i* topo_i;
|
|
};
|
|
|
|
VENDOR get_gpu_vendor(struct gpu_info* gpu);
|
|
char* get_str_gpu_name(struct gpu_info* gpu);
|
|
char* get_str_freq(struct gpu_info* gpu);
|
|
char* get_str_memory_size(struct gpu_info* gpu);
|
|
char* get_str_memory_type(struct gpu_info* gpu);
|
|
char* get_str_bus_width(struct gpu_info* gpu);
|
|
char* get_str_memory_clock(struct gpu_info* gpu);
|
|
char* get_str_l2(struct gpu_info* gpu);
|
|
char* get_str_peak_performance(struct gpu_info* gpu);
|
|
char* get_str_peak_performance_tensor(struct gpu_info* gpu);
|
|
char* get_str_generic(int32_t data);
|
|
|
|
#endif
|