diff options
Diffstat (limited to 'src/gpgpu-sim/gpu-sim.cc')
| -rw-r--r-- | src/gpgpu-sim/gpu-sim.cc | 18 |
1 files changed, 10 insertions, 8 deletions
diff --git a/src/gpgpu-sim/gpu-sim.cc b/src/gpgpu-sim/gpu-sim.cc index e44551e..5af244b 100644 --- a/src/gpgpu-sim/gpu-sim.cc +++ b/src/gpgpu-sim/gpu-sim.cc @@ -1,6 +1,7 @@ // Copyright (c) 2009-2021, Tor M. Aamodt, Wilson W.L. Fung, George L. Yuan, -// Ali Bakhoda, Andrew Turner, Ivan Sham, Vijay Kandiah, Nikos Hardavellas -// The University of British Columbia, Northwestern University +// Ali Bakhoda, Andrew Turner, Ivan Sham, Vijay Kandiah, Nikos Hardavellas, +// Mahmoud Khairy, Junrui Pan, Timothy G. Rogers +// The University of British Columbia, Northwestern University, Purdue University // All rights reserved. // // Redistribution and use in source and binary forms, with or without @@ -394,7 +395,7 @@ void shader_core_config::reg_options(class OptionParser *opp) { "gpgpu_ignore_resources_limitation (default 0)", "0"); option_parser_register( opp, "-gpgpu_shader_cta", OPT_UINT32, &max_cta_per_core, - "Maximum number of concurrent CTAs in shader (default 8)", "8"); + "Maximum number of concurrent CTAs in shader (default 32)", "32"); option_parser_register( opp, "-gpgpu_num_cta_barriers", OPT_UINT32, &max_barriers_per_cta, "Maximum number of named barriers per CTA (default 16)", "16"); @@ -700,7 +701,8 @@ void gpgpu_sim_config::reg_options(option_parser_t opp) { "500.0:2000.0:2000.0:2000.0"); option_parser_register( opp, "-gpgpu_max_concurrent_kernel", OPT_INT32, &max_concurrent_kernel, - "maximum kernels that can run concurrently on GPU", "8"); + "maximum kernels that can run concurrently on GPU, set this value " + "according to max resident grids for your compute capability", "32"); option_parser_register( opp, "-gpgpu_cflog_interval", OPT_INT32, &gpgpu_cflog_interval, "Interval between each snapshot in control flow logger", "0"); @@ -1638,9 +1640,9 @@ bool shader_core_ctx::occupy_shader_resource_1block(kernel_info_t &k, SHADER_DPRINTF(LIVENESS, "GPGPU-Sim uArch: Occupied %u threads, %u shared mem, %u " - "registers, %u ctas\n", + "registers, %u ctas, on shader %d\n", m_occupied_n_threads, m_occupied_shmem, m_occupied_regs, - m_occupied_ctas); + m_occupied_ctas, m_sid); } return true; @@ -1806,9 +1808,9 @@ void shader_core_ctx::issue_block2core(kernel_info_t &kernel) { shader_CTA_count_log(m_sid, 1); SHADER_DPRINTF(LIVENESS, "GPGPU-Sim uArch: cta:%2u, start_tid:%4u, end_tid:%4u, " - "initialized @(%lld,%lld)\n", + "initialized @(%lld,%lld), kernel_uid:%u, kernel_name:%s\n", free_cta_hw_id, start_thread, end_thread, m_gpu->gpu_sim_cycle, - m_gpu->gpu_tot_sim_cycle); + m_gpu->gpu_tot_sim_cycle, kernel.get_uid(), kernel.get_name().c_str()); } /////////////////////////////////////////////////////////////////////////////////////////// |