From 38b4df5653ecbd9907a3d39b125640cd4fb7d012 Mon Sep 17 00:00:00 2001 From: JRPan <25518778+JRPan@users.noreply.github.com> Date: Tue, 20 Aug 2024 20:43:32 -0400 Subject: Stream stats (#71) * Temp commit for Justin and Cassie to sync on code changes for adding per-stream status. * Resolved compile errors. * Removed redundant parameter * Passed cuda_stream_id from accelsim to gpgpusim * Cleaned up unused changes * Changed vector to map, having operator problems. * StreamID defaults to zero * Implemented streams to inc_stats and so on * Fixed TOTAL_ACCESS counts * Implemented GLOBAL_TIMER. * Fixed m_shader->get_kernel SEGFAULT issue in shader.cc. * Use warp_init to track streamID instead of issue_warp * Removed temp debug print * Modified cache_stats to only print data from latest finished stream Added optional arg to cache_stats::print_stats, cache_stats::print_fail_stats and their upstream functions. When streamID is specified, print stats from that stream. When not specified, print all stats. NOTE: current implementation depending on streamid never equals -1 * Removed default arg values of streamID * modified constructor of mem_fetch to pass in streamID * changed get_streamid to get_streamID * Added TODO to gpgpusim_entrypoint.cc and power_stat.cc * Only collect power stats when enabled * print last finished stream in PTX mode using last_streamID * take out additional printf * Add a field to baseline cache to indicate cache level * save gpu object in cache * Print stream ID only once per kernel * rm test print * use -1 for default stream id * cleanup debug prints * remove GLOABL_TIMER * Automated clang-format * Should be correct to print everything in power model * addressing concerns & errors * Automated clang-format * add m_stats_pw in operator+ * Automated Format --------- Co-authored-by: Justin Qiao Co-authored-by: Justin Qiao <71228724+ShichenQiao@users.noreply.github.com> Co-authored-by: Tim Rogers Co-authored-by: JRPan Co-authored-by: purdue-jenkins --- src/gpgpu-sim/gpu-sim.h | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) (limited to 'src/gpgpu-sim/gpu-sim.h') diff --git a/src/gpgpu-sim/gpu-sim.h b/src/gpgpu-sim/gpu-sim.h index d43b399..8e81451 100644 --- a/src/gpgpu-sim/gpu-sim.h +++ b/src/gpgpu-sim/gpu-sim.h @@ -539,7 +539,7 @@ class gpgpu_sim : public gpgpu_t { (m_config.gpu_max_completed_cta_opt && (gpu_completed_cta >= m_config.gpu_max_completed_cta_opt)); } - void print_stats(); + void print_stats(unsigned long long streamID); void update_stats(); void deadlock_check(); void inc_completed_cta() { gpu_completed_cta++; } @@ -568,7 +568,7 @@ class gpgpu_sim : public gpgpu_t { void decrement_kernel_latency(); const gpgpu_sim_config &get_config() const { return m_config; } - void gpu_print_stat(); + void gpu_print_stat(unsigned long long streamID); void dump_pipeline(int mask, int s, int m) const; void perf_memcpy_to_gpu(size_t dst_start_addr, size_t count); @@ -685,6 +685,17 @@ class gpgpu_sim : public gpgpu_t { occupancy_stats gpu_occupancy; occupancy_stats gpu_tot_occupancy; + typedef struct { + unsigned long long start_cycle; + unsigned long long end_cycle; + } kernel_time_t; + std::map> + gpu_kernel_time; + unsigned long long last_streamID; + unsigned long long last_uid; + cache_stats aggregated_l1_stats; + cache_stats aggregated_l2_stats; + // performance counter for stalls due to congestion. unsigned int gpu_stall_dramfull; unsigned int gpu_stall_icnt2sh; @@ -712,6 +723,9 @@ class gpgpu_sim : public gpgpu_t { public: bool is_functional_sim() { return m_functional_sim; } kernel_info_t *get_functional_kernel() { return m_functional_sim_kernel; } + std::vector get_running_kernels() { + return m_running_kernels; + } void functional_launch(kernel_info_t *k) { m_functional_sim = true; m_functional_sim_kernel = k; -- cgit v1.3