mattsinc
commented
4 years ago Those specific prints only happen once at the beginning of your run. But, if you want to turn them off while leaving the prints in the code, you could use the g_debug_execution flag (which is set based on the PTX_SIM_DEBUG environment variable). You can look around in the simulator and see other instances of this -- basically, the higher you set the debug level, the more prints you'll see. For example, you could make the prints you want to remove be at debug level 1, and then set PTX_SIM_DEBUG to 0 and they wouldn't print.
Matt
brad-mengchi
KMFtcy
While I'm running a cuda application on gpgpu-sim, there will be lots of gpu status pumped out. ` GPGPU-Sim Simulator Version 4.0.0 [build gpgpu-sim_git-commit-e7fbfaa347c0acf8a6702c1e684a8e2ad8d3f733_modified_0]
GPGPU-Sim PTX: simulation mode 0 (can change with PTX_SIM_MODE_FUNC environment variable: 1=functional simulation only, 0=detailed performance simulator) GPGPU-Sim PTX: overriding embedded ptx with ptx file (PTX_SIM_USE_PTX_FILE is set) GPGPU-Sim: Configuration options:
-save_embedded_ptx 0 # saves ptx files embedded in binary as.ptx
-keep 0 # keep intermediate files created by GPGPU-Sim when interfacing with external programs
-gpgpu_ptx_save_converted_ptxplus 0 # Saved converted ptxplus to a file
-gpgpu_occupancy_sm_number 20 # The SM number to pass to ptxas when getting register usage for computing GPU occupancy. This parameter is required in the config.
-ptx_opcode_latency_int 4,13,4,5,145,32 # Opcode latencies for integers <ADD,MAX,MUL,MAD,DIV,SHFL>Default 1,1,19,25,145,32
-ptx_opcode_latency_fp 4,13,4,5,39 # Opcode latencies for single precision floating points <ADD,MAX,MUL,MAD,DIV>Default 1,1,1,1,30
-ptx_opcode_latency_dp 8,19,8,8,330 # Opcode latencies for double precision floating points <ADD,MAX,MUL,MAD,DIV>Default 8,8,8,8,335
-ptx_opcode_latency_sfu 8 # Opcode latencies for SFU instructionsDefault 8
-ptx_opcode_latency_tesnor 64 # Opcode latencies for Tensor instructionsDefault 64
-ptx_opcode_initiation_int 1,2,2,1,8,4 # Opcode initiation intervals for integers <ADD,MAX,MUL,MAD,DIV,SHFL>Default 1,1,4,4,32,4
-ptx_opcode_initiation_fp 1,2,1,1,4 # Opcode initiation intervals for single precision floating points <ADD,MAX,MUL,MAD,DIV>Default 1,1,1,1,5
-ptx_opcode_initiation_dp 8,16,8,8,130 # Opcode initiation intervals for double precision floating points <ADD,MAX,MUL,MAD,DIV>Default 8,8,8,8,130
-ptx_opcode_initiation_sfu 8 # Opcode initiation intervals for sfu instructionsDefault 8
-ptx_opcode_initiation_tensor 64 # Opcode initiation intervals for tensor instructionsDefault 64
-cdp_latency 7200,8000,100,12000,1600 # CDP API latency <cudaStreamCreateWithFlags, cudaGetParameterBufferV2_init_perWarp, cudaGetParameterBufferV2_perKernel, cudaLaunchDeviceV2_init_perWarp, cudaLaunchDevicV2_perKernel>Default 7200,8000,100,12000,1600
-network_mode 1 # Interconnection network mode
-inter_config_file config_fermi_islip.icnt # Interconnection network config file
-inct_in_buffer_limit 64 # in_buffer_limit
-inct_out_buffer_limit 64 # out_buffer_limit
-inct_subnets 2 # subnets
-gpgpu_ptx_use_cuobjdump 1 # Use cuobjdump to extract ptx and sass from binaries
-gpgpu_experimental_lib_support 0 # Try to extract code from cuda libraries [Broken because of unknown cudaGetExportTable]
-checkpoint_option 0 # checkpointing flag (0 = no checkpoint)
-checkpoint_kernel 1 # checkpointing during execution of which kernel (1- 1st kernel)
-checkpoint_CTA 0 # checkpointing after # of CTA (< less than total CTA)
-resume_option 0 # resume flag (0 = no resume)
-resume_kernel 0 # Resume from which kernel (1= 1st kernel)
-resume_CTA 0 # resume from which CTA
-checkpoint_CTA_t 0 # resume from which CTA
-checkpoint_insn_Y 0 # resume from which CTA
-gpgpu_ptx_convert_to_ptxplus 0 # Convert SASS (native ISA) to ptxplus and run ptxplus
-gpgpu_ptx_force_max_capability 20 # Force maximum compute capability
-gpgpu_ptx_inst_debug_to_file 0 # Dump executed instructions' debug information to file
-gpgpu_ptx_inst_debug_file inst_debug.txt # Executed instructions' debug output file
-gpgpu_ptx_inst_debug_thread_uid 1 # Thread UID for executed instructions' debug output
-gpgpu_simd_model 1 # 1 = post-dominator
...
`
Sometimes I want to focus on application debuging. And I have log file to trace hardware data, these ouput is not necessary. Can I turn off it if it is not needed?