From e5f532a3b65e17f49991ed08a275f87ac2d68d0a Mon Sep 17 00:00:00 2001 From: aamir Date: Fri, 1 Jun 2018 09:52:12 -0700 Subject: wmma load working --- cuda-kernels/gpgpu_inst_stats.txt | 19 + cuda-kernels/log | 6328 +++++++++++++++++++++++++++++++++++++ cuda-kernels/log1 | 512 +++ cuda-kernels/tensor_core | Bin 2750968 -> 2750968 bytes src/cuda-sim/Makefile | 2 +- src/cuda-sim/cuda-math.h | 1 + src/cuda-sim/cuda-sim.cc | 2 +- src/cuda-sim/half.hpp | 3067 ++++++++++++++++++ src/cuda-sim/instructions.cc | 96 +- src/cuda-sim/ptx_sim.h | 6 +- 10 files changed, 9988 insertions(+), 45 deletions(-) create mode 100644 cuda-kernels/log create mode 100644 cuda-kernels/log1 create mode 100644 src/cuda-sim/half.hpp diff --git a/cuda-kernels/gpgpu_inst_stats.txt b/cuda-kernels/gpgpu_inst_stats.txt index acb1839..41f06a4 100755 --- a/cuda-kernels/gpgpu_inst_stats.txt +++ b/cuda-kernels/gpgpu_inst_stats.txt @@ -1 +1,20 @@ kernel line : count latency dram_traffic smem_bk_conflicts smem_warp gmem_access_generated gmem_warp exposed_latency warp_divergence +_1.ptx 164 : 512 2560 1024 0 0 16 16 0 0 +_1.ptx 163 : 512 5696 0 0 0 0 0 0 0 +_1.ptx 162 : 512 5664 0 0 0 0 0 0 0 +_1.ptx 161 : 512 3616 0 0 0 0 0 0 0 +_1.ptx 158 : 512 3616 0 0 0 0 0 0 0 +_1.ptx 156 : 512 134048 2048 0 0 16 16 0 0 +_1.ptx 155 : 512 5632 0 0 0 0 0 0 0 +_1.ptx 154 : 512 5376 0 0 0 0 0 0 0 +_1.ptx 143 : 512 100864 128 0 0 0 0 0 0 +_1.ptx 167 : 512 3072 0 0 0 0 0 0 0 +_1.ptx 144 : 512 100864 0 0 0 0 0 0 0 +_1.ptx 145 : 512 1536 0 0 0 0 0 0 0 +_1.ptx 146 : 512 3072 0 0 0 0 0 0 0 +_1.ptx 147 : 512 3072 0 0 0 0 0 0 0 +_1.ptx 148 : 512 3072 0 0 0 0 0 0 0 +_1.ptx 149 : 512 8384 0 0 0 0 0 0 0 +_1.ptx 150 : 512 4096 0 0 0 0 0 0 0 +_1.ptx 151 : 512 0 0 0 0 0 0 0 0 +_1.ptx 153 : 512 3968 0 0 0 0 0 0 0 diff --git a/cuda-kernels/log b/cuda-kernels/log new file mode 100644 index 0000000..98df26a --- /dev/null +++ b/cuda-kernels/log @@ -0,0 +1,6328 @@ + + + *** GPGPU-Sim Simulator Version 3.2.2 [build gpgpu-sim_git-commit-03de5ae03420ba5666d669c6f76faccf2704fa58_modified_17] *** + + +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: Configuration options: + +-network_mode 1 # Interconnection network mode +-inter_config_file config_fermi_islip.icnt # Interconnection network config file +-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] +-gpgpu_ptx_convert_to_ptxplus 0 # Convert SASS (native ISA) to ptxplus and run ptxplus +-gpgpu_ptx_force_max_capability 70 # 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 +-gpgpu_shader_core_pipeline 2048:32 # shader core pipeline config, i.e., {:} +-gpgpu_tex_cache:l1 16:128:24,L:R:m:N:L,F:128:4,128:2 # per-shader L1 texture cache (READ-ONLY) config {::,:::,::,:} +-gpgpu_const_cache:l1 128:64:2,L:R:f:N:L,A:2:64,4 # per-shader L1 constant memory cache (READ-ONLY) config {::,:::,::,} +-gpgpu_cache:il1 8:128:4,L:R:f:N:L,A:2:48,4 # shader L1 instruction cache config {::,:::,::,} +-gpgpu_cache:dl1 64:128:6,L:L:m:N:H,A:128:8,8 # per-shader L1 data cache config {::,:::,::, | none} +-gpgpu_cache:dl1PrefL1 none # per-shader L1 data cache config {::,:::,::, | none} +-gpgpu_cache:dl1PreShared none # per-shader L1 data cache config {::,:::,::, | none} +-gmem_skip_L1D 1 # global memory access skip L1D cache (implements -Xptxas -dlcm=cg, default=no skip) +-gpgpu_perfect_mem 0 # enable perfect memory mode (no cache miss) +-n_regfile_gating_group 4 # group of lanes that should be read/written together) +-gpgpu_clock_gated_reg_file 0 # enable clock gated reg file for power calculations +-gpgpu_clock_gated_lanes 0 # enable clock gated lanes for power calculations +-gpgpu_shader_registers 65536 # Number of registers per shader core. Limits number of concurrent CTAs. (default 8192) +-gpgpu_shader_cta 32 # Maximum number of concurrent CTAs in shader (default 8) +-gpgpu_num_cta_barriers 16 # Maximum number of named barriers per CTA (default 16) +-gpgpu_n_clusters 40 # number of processing clusters +-gpgpu_n_cores_per_cluster 1 # number of simd cores per cluster +-gpgpu_n_cluster_ejection_buffer_size 8 # number of packets in ejection buffer +-gpgpu_n_ldst_response_buffer_size 2 # number of response packets in ld/st unit ejection buffer +-gpgpu_shmem_size 16384 # Size of shared memory per shader core (default 16kB) +-gpgpu_shmem_size 98304 # Size of shared memory per shader core (default 16kB) +-gpgpu_shmem_size_PrefL1 16384 # Size of shared memory per shader core (default 16kB) +-gpgpu_shmem_size_PrefShared 16384 # Size of shared memory per shader core (default 16kB) +-gpgpu_shmem_num_banks 32 # Number of banks in the shared memory in each shader core (default 16) +-gpgpu_shmem_limited_broadcast 0 # Limit shared memory to do one broadcast per cycle (default on) +-gpgpu_shmem_warp_parts 1 # Number of portions a warp is divided into for shared memory bank conflict check +-gpgpu_warpdistro_shader -1 # Specify which shader core to collect the warp size distribution from +-gpgpu_warp_issue_shader 0 # Specify which shader core to collect the warp issue distribution from +-gpgpu_local_mem_map 1 # Mapping from local memory space address to simulated GPU physical address space (default = enabled) +-gpgpu_num_reg_banks 32 # Number of register banks (default = 8) +-gpgpu_reg_bank_use_warp_id 0 # Use warp ID in mapping registers to banks (default = off) +-gpgpu_operand_collector_num_units_sp 20 # number of collector units (default = 4) +-gpgpu_operand_collector_num_units_sfu 4 # number of collector units (default = 4) +-gpgpu_operand_collector_num_units_mem 8 # number of collector units (default = 2) +-gpgpu_operand_collector_num_units_gen 0 # number of collector units (default = 0) +-gpgpu_operand_collector_num_in_ports_sp 4 # number of collector unit in ports (default = 1) +-gpgpu_operand_collector_num_in_ports_sfu 1 # number of collector unit in ports (default = 1) +-gpgpu_operand_collector_num_in_ports_mem 1 # number of collector unit in ports (default = 1) +-gpgpu_operand_collector_num_in_ports_gen 0 # number of collector unit in ports (default = 0) +-gpgpu_operand_collector_num_out_ports_sp 4 # number of collector unit in ports (default = 1) +-gpgpu_operand_collector_num_out_ports_sfu 1 # number of collector unit in ports (default = 1) +-gpgpu_operand_collector_num_out_ports_mem 1 # number of collector unit in ports (default = 1) +-gpgpu_operand_collector_num_out_ports_gen 0 # number of collector unit in ports (default = 0) +-gpgpu_coalesce_arch 13 # Coalescing arch (default = 13, anything else is off for now) +-gpgpu_num_sched_per_core 2 # Number of warp schedulers per core +-gpgpu_max_insn_issue_per_warp 2 # Max number of instructions that can be issued per warp in one cycle by scheduler +-gpgpu_simt_core_sim_order 1 # Select the simulation order of cores in a cluster (0=Fix, 1=Round-Robin) +-gpgpu_pipeline_widths 4,1,1,4,1,1,6 # Pipeline widths ID_OC_SP,ID_OC_SFU,ID_OC_MEM,OC_EX_SP,OC_EX_SFU,OC_EX_MEM,EX_WB +-gpgpu_num_sp_units 4 # Number of SP units (default=1) +-gpgpu_num_sfu_units 1 # Number of SF units (default=1) +-gpgpu_num_mem_units 1 # Number if ldst units (default=1) WARNING: not hooked up to anything +-gpgpu_scheduler gto # Scheduler configuration: < lrr | gto | two_level_active > If two_level_active:::For complete list of prioritization values see shader.h enum scheduler_prioritization_typeDefault: gto +-gpgpu_concurrent_kernel_sm 0 # Support concurrent kernels on a SM (default = disabled) +-gpgpu_dram_scheduler 1 # 0 = fifo, 1 = FR-FCFS (defaul) +-gpgpu_dram_partition_queues 8:8:8:8 # i2$:$2d:d2$:$2i +-l2_ideal 0 # Use a ideal L2 cache that always hit +-gpgpu_cache:dl2 64:128:16,L:B:m:W:L,A:1024:1024,4:0,32 # unified banked L2 data cache config {::,:::,::,} +-gpgpu_cache:dl2_texture_only 0 # L2 cache used for texture only +-gpgpu_n_mem 11 # number of memory modules (e.g. memory controllers) in gpu +-gpgpu_n_sub_partition_per_mchannel 2 # number of memory subpartition in each memory module +-gpgpu_n_mem_per_ctrlr 1 # number of memory chips per memory controller +-gpgpu_memlatency_stat 14 # track and display latency statistics 0x2 enables MC, 0x4 enables queue logs +-gpgpu_frfcfs_dram_sched_queue_size 64 # 0 = unlimited (default); # entries per chip +-gpgpu_dram_return_queue_size 116 # 0 = unlimited (default); # entries per chip +-gpgpu_dram_buswidth 4 # default = 4 bytes (8 bytes per cycle at DDR) +-gpgpu_dram_burst_length 8 # Burst length of each DRAM request (default = 4 data bus cycle) +-dram_data_command_freq_ratio 4 # Frequency ratio between DRAM data bus and command bus (default = 2 times, i.e. DDR) +-gpgpu_dram_timing_opt nbk=16:CCD=2:RRD=6:RCD=12:RAS=28:RP=12:RC=40: CL=12:WL=4:CDLR=5:WR=12:nbkgrp=1:CCDL=0:RTPL=0 # DRAM timing parameters = {nbk:tCCD:tRRD:tRCD:tRAS:tRP:tRC:CL:WL:tCDLR:tWR:nbkgrp:tCCDL:tRTPL} +-rop_latency 120 # ROP queue latency (default 85) +-dram_latency 100 # DRAM latency (default 30) +-gpgpu_mem_addr_mapping dramid@8;00000000.00000000.00000000.00000000.0000RRRR.RRRRRRRR.RBBBCCCC.BCCSSSSS # mapping memory address to dram model {dramid@;} +-gpgpu_mem_addr_test 0 # run sweep test to check address mapping for aliased address +-gpgpu_mem_address_mask 1 # 0 = old addressing mask, 1 = new addressing mask, 2 = new add. mask + flipped bank sel and chip sel bits +-gpuwattch_xml_file gpuwattch_gtx1080Ti.xml # GPUWattch XML file +-power_simulation_enabled 1 # Turn on power simulator (1=On, 0=Off) +-power_per_cycle_dump 0 # Dump detailed power output each cycle +-power_trace_enabled 0 # produce a file for the power trace (1=On, 0=Off) +-power_trace_zlevel 6 # Compression level of the power trace output log (0=no comp, 9=highest) +-steady_power_levels_enabled 0 # produce a file for the steady power levels (1=On, 0=Off) +-steady_state_definition 8:4 # allowed deviation:number of samples +-gpgpu_max_cycle 0 # terminates gpu simulation early (0 = no limit) +-gpgpu_max_insn 0 # terminates gpu simulation early (0 = no limit) +-gpgpu_max_cta 0 # terminates gpu simulation early (0 = no limit) +-gpgpu_runtime_stat 500 # display runtime statistics such as dram utilization {:} +-liveness_message_freq 1 # Minimum number of seconds between simulation liveness messages (0 = always print) +-gpgpu_flush_l1_cache 0 # Flush L1 cache at the end of each kernel call +-gpgpu_flush_l2_cache 0 # Flush L2 cache at the end of each kernel call +-gpgpu_deadlock_detect 1 # Stop the simulation at deadlock (1=on (default), 0=off) +-gpgpu_ptx_instruction_classification 0 # if enabled will classify ptx instruction types per kernel (Max 255 kernels now) +-gpgpu_ptx_sim_mode 0 # Select between Performance (default) or Functional simulation (1) +-gpgpu_clock_domains 1481.0:2962.0:1481.0:2750.0 # Clock Domain Frequencies in MhZ {:::} +-gpgpu_max_concurrent_kernel 8 # maximum kernels that can run concurrently on GPU +-gpgpu_cflog_interval 0 # Interval between each snapshot in control flow logger +-visualizer_enabled 0 # Turn on visualizer output (1=On, 0=Off) +-visualizer_outputfile NULL # Specifies the output log file for visualizer +-visualizer_zlevel 6 # Compression level of the visualizer output log (0=no comp, 9=highest) +-trace_enabled 0 # Turn on traces +-trace_components none # comma seperated list of traces to enable. Complete list found in trace_streams.tup. Default none +-trace_sampling_core 0 # The core which is printed using CORE_DPRINTF. Default 0 +-trace_sampling_memory_partition -1 # The memory partition which is printed using MEMPART_DPRINTF. Default -1 (i.e. all) +-enable_ptx_file_line_stats 1 # Turn on PTX source line statistic profiling. (1 = On) +-ptx_line_stats_filename gpgpu_inst_stats.txt # Output file for PTX source line statistics. +-gpgpu_kernel_launch_latency 0 # Kernel launch latency in cycles. Default: 0 +-gpgpu_cdp_enabled 0 # Turn on CDP +-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 +-ptx_opcode_latency_int 4,13,4,5,145,16,4 # Opcode latencies for integers Default 1,1,19,25,145,1,4 +-ptx_opcode_latency_fp 4,13,4,5,39 # Opcode latencies for single precision floating points Default 1,1,1,1,30 +-ptx_opcode_latency_dp 8,19,8,8,330 # Opcode latencies for double precision floating points Default 8,8,8,8,335 +-ptx_opcode_initiation_int 1,2,2,2,8,16,4 # Opcode initiation intervals for integers Default 1,1,4,4,32,1,1 +-ptx_opcode_initiation_fp 1,2,1,1,4 # Opcode initiation intervals for single precision floating points Default 1,1,1,1,5 +-ptx_opcode_initiation_dp 1,2,1,1,130 # Opcode initiation intervals for double precision floating points Default 8,8,8,8,130 +-cdp_latency 7200,8000,100,12000,1600 # CDP API latency Default 7200,8000,100,12000,1600 +DRAM Timing Options: +nbk 16 # number of banks +CCD 2 # column to column delay +RRD 6 # minimal delay between activation of rows in different banks +RCD 12 # row to column delay +RAS 28 # time needed to activate row +RP 12 # time needed to precharge (deactivate) row +RC 40 # row cycle time +CDLR 5 # switching from write to read (changes tWTR) +WR 12 # last data-in to row precharge +CL 12 # CAS latency +WL 4 # Write latency +nbkgrp 1 # number of bank groups +CCDL 0 # column to column delay between accesses to different bank groups +RTPL 0 # read to precharge delay between accesses to different bank groups +Total number of memory sub partition = 22 +addr_dec_mask[CHIP] = 0000000000000000 high:64 low:0 +addr_dec_mask[BK] = 0000000000007080 high:15 low:7 +addr_dec_mask[ROW] = 000000000fff8000 high:28 low:15 +addr_dec_mask[COL] = 0000000000000f7f high:12 low:0 +addr_dec_mask[BURST] = 000000000000001f high:5 low:0 +sub_partition_id_mask = 0000000000000080 +GPGPU-Sim uArch: clock freqs: 1481000000.000000:2962000000.000000:1481000000.000000:2750000000.000000 +GPGPU-Sim uArch: clock periods: 0.00000000067521944632:0.00000000033760972316:0.00000000067521944632:0.00000000036363636364 +*** Initializing Memory Statistics *** +GPGPU-Sim uArch: interconnect node map (shaderID+MemID to icntID) +GPGPU-Sim uArch: Memory nodes ID start from index: 40 +GPGPU-Sim uArch: 0 1 2 3 4 5 6 +GPGPU-Sim uArch: 7 8 9 10 11 12 13 +GPGPU-Sim uArch: 14 15 16 17 18 19 20 +GPGPU-Sim uArch: 21 22 23 24 25 26 27 +GPGPU-Sim uArch: 28 29 30 31 32 33 34 +GPGPU-Sim uArch: 35 36 37 38 39 40 41 +GPGPU-Sim uArch: 42 43 44 45 46 47 48 +GPGPU-Sim uArch: 49 50 51 52 53 54 55 +GPGPU-Sim uArch: 56 57 58 59 60 61 +GPGPU-Sim uArch: interconnect node reverse map (icntID to shaderID+MemID) +GPGPU-Sim uArch: Memory nodes start from ID: 40 +GPGPU-Sim uArch: 0 1 2 3 4 5 6 +GPGPU-Sim uArch: 7 8 9 10 11 12 13 +GPGPU-Sim uArch: 14 15 16 17 18 19 20 +GPGPU-Sim uArch: 21 22 23 24 25 26 27 +GPGPU-Sim uArch: 28 29 30 31 32 33 34 +GPGPU-Sim uArch: 35 36 37 38 39 40 41 +GPGPU-Sim uArch: 42 43 44 45 46 47 48 +GPGPU-Sim uArch: 49 50 51 52 53 54 55 +GPGPU-Sim uArch: 56 57 58 59 60 61 +a9478053306cbb4803bedf0d6ea12100 /home/araihan/gpgpusim-tensorcore/cuda-kernels/tensor_core +GPGPU-Sim uArch: performance model initialization complete. +GPGPU-Sim PTX: __cudaRegisterFatBinary, fat_cubin_handle = 1, filename=default +self exe links to: /home/araihan/gpgpusim-tensorcore/cuda-kernels/tensor_core +Running md5sum using "md5sum /home/araihan/gpgpusim-tensorcore/cuda-kernels/tensor_core " +Parsing file _cuobjdump_complete_output_c7ZC8M +######### cuobjdump parser ######## +## Adding new section PTX +Adding ptx filename: _cuobjdump_1.ptx +Adding arch: sm_70 +Adding identifier: default +Done parsing!!! +GPGPU-Sim PTX: __cudaRegisterFunction _Z17convertFp32ToFp16P6__halfPfi : hostFun 0x0x401dd7, fat_cubin_handle = 1 +WARNING: No guarantee that PTX will be parsed for SM version 70 + _1.ptx:13 => (ptx_parser.cc:175) start_function + _1.ptx:13 => (ptx_parser.cc:144) init_directive_state + _1.ptx:13 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:13 => (ptx_parser.cc:144) init_directive_state + _1.ptx:13 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:13 => (ptx_parser.cc:631) add_scalar_type_spec "B32_TYPE" + _1.ptx:13 => (ptx_parser.cc:331) set_variable_type space_spec=param_space_local scalar_type_spec=B32_TYPE + _1.ptx:14 => (ptx_parser.cc:189) add_function_name vprintf (extern) + _1.ptx:14 => (ptx_parser.cc:381) add_identifier "func_retval0" (0) +GPGPU-Sim PTX: allocating stack frame region for .param "func_retval0" from 0x0 to 0x4 + _1.ptx:14 => (ptx_parser.cc:144) init_directive_state + _1.ptx:15 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:15 => (ptx_parser.cc:631) add_scalar_type_spec "B64_TYPE" + _1.ptx:15 => (ptx_parser.cc:331) set_variable_type space_spec=param_space_local scalar_type_spec=B64_TYPE + _1.ptx:15 => (ptx_parser.cc:381) add_identifier "vprintf_param_0" (1) +GPGPU-Sim PTX: allocating stack frame region for .param "vprintf_param_0" from 0x4 to 0xc + _1.ptx:15 => (ptx_parser.cc:577) add_function_arg "vprintf_param_0" + _1.ptx:15 => (ptx_parser.cc:219) add_directive + _1.ptx:15 => (ptx_parser.cc:144) init_directive_state + _1.ptx:16 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:16 => (ptx_parser.cc:631) add_scalar_type_spec "B64_TYPE" + _1.ptx:16 => (ptx_parser.cc:331) set_variable_type space_spec=param_space_local scalar_type_spec=B64_TYPE + _1.ptx:17 => (ptx_parser.cc:381) add_identifier "vprintf_param_1" (2) +GPGPU-Sim PTX: allocating stack frame region for .param "vprintf_param_1" from 0xc to 0x14 + _1.ptx:17 => (ptx_parser.cc:577) add_function_arg "vprintf_param_1" + _1.ptx:17 => (ptx_parser.cc:219) add_directive + _1.ptx:17 => (ptx_parser.cc:144) init_directive_state + _1.ptx:19 => (ptx_parser.cc:605) add_space_spec "global_space" + _1.ptx:19 => (ptx_parser.cc:590) add_alignment_spec + _1.ptx:19 => (ptx_parser.cc:631) add_scalar_type_spec "B8_TYPE" + _1.ptx:19 => (ptx_parser.cc:331) set_variable_type space_spec=global_space scalar_type_spec=B8_TYPE + _1.ptx:19 => (ptx_parser.cc:381) add_identifier "$str" (3) +GPGPU-Sim PTX: allocating global region for "$str" from 0x100 to 0x109 (global memory space) + _1.ptx:19 => (ptx_parser.cc:883) add_literal_int + _1.ptx:19 => (ptx_parser.cc:883) add_literal_int + _1.ptx:19 => (ptx_parser.cc:883) add_literal_int + _1.ptx:19 => (ptx_parser.cc:883) add_literal_int + _1.ptx:19 => (ptx_parser.cc:883) add_literal_int + _1.ptx:19 => (ptx_parser.cc:883) add_literal_int + _1.ptx:19 => (ptx_parser.cc:883) add_literal_int + _1.ptx:19 => (ptx_parser.cc:883) add_literal_int + _1.ptx:19 => (ptx_parser.cc:883) add_literal_int + _1.ptx:19 => (ptx_parser.cc:319) add_variables + _1.ptx:19 => (ptx_parser.cc:144) init_directive_state + _1.ptx:21 => (ptx_parser.cc:175) start_function + _1.ptx:21 => (ptx_parser.cc:144) init_directive_state + _1.ptx:21 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:21 => (ptx_parser.cc:144) init_directive_state + _1.ptx:21 => (ptx_parser.cc:189) add_function_name _Z12wmma_exampleP6__halfS0_Pfiiiff (entrypoint) + _1.ptx:22 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:22 => (ptx_parser.cc:631) add_scalar_type_spec "U64_TYPE" + _1.ptx:22 => (ptx_parser.cc:331) set_variable_type space_spec=param_space_kernel scalar_type_spec=U64_TYPE + _1.ptx:22 => (ptx_parser.cc:381) add_identifier "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_0" (4) + _1.ptx:22 => (ptx_parser.cc:577) add_function_arg "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_0" + _1.ptx:22 => (ptx_parser.cc:219) add_directive + _1.ptx:22 => (ptx_parser.cc:144) init_directive_state + _1.ptx:23 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:23 => (ptx_parser.cc:631) add_scalar_type_spec "U64_TYPE" + _1.ptx:23 => (ptx_parser.cc:331) set_variable_type space_spec=param_space_kernel scalar_type_spec=U64_TYPE + _1.ptx:23 => (ptx_parser.cc:381) add_identifier "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_1" (5) + _1.ptx:23 => (ptx_parser.cc:577) add_function_arg "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_1" + _1.ptx:23 => (ptx_parser.cc:219) add_directive + _1.ptx:23 => (ptx_parser.cc:144) init_directive_state + _1.ptx:24 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:24 => (ptx_parser.cc:631) add_scalar_type_spec "U64_TYPE" + _1.ptx:24 => (ptx_parser.cc:331) set_variable_type space_spec=param_space_kernel scalar_type_spec=U64_TYPE + _1.ptx:24 => (ptx_parser.cc:381) add_identifier "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_2" (6) + _1.ptx:24 => (ptx_parser.cc:577) add_function_arg "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_2" + _1.ptx:24 => (ptx_parser.cc:219) add_directive + _1.ptx:24 => (ptx_parser.cc:144) init_directive_state + _1.ptx:25 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:25 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:25 => (ptx_parser.cc:331) set_variable_type space_spec=param_space_kernel scalar_type_spec=U32_TYPE + _1.ptx:25 => (ptx_parser.cc:381) add_identifier "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_3" (7) + _1.ptx:25 => (ptx_parser.cc:577) add_function_arg "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_3" + _1.ptx:25 => (ptx_parser.cc:219) add_directive + _1.ptx:25 => (ptx_parser.cc:144) init_directive_state + _1.ptx:26 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:26 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:26 => (ptx_parser.cc:331) set_variable_type space_spec=param_space_kernel scalar_type_spec=U32_TYPE + _1.ptx:26 => (ptx_parser.cc:381) add_identifier "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_4" (8) + _1.ptx:26 => (ptx_parser.cc:577) add_function_arg "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_4" + _1.ptx:26 => (ptx_parser.cc:219) add_directive + _1.ptx:26 => (ptx_parser.cc:144) init_directive_state + _1.ptx:27 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:27 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:27 => (ptx_parser.cc:331) set_variable_type space_spec=param_space_kernel scalar_type_spec=U32_TYPE + _1.ptx:27 => (ptx_parser.cc:381) add_identifier "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_5" (9) + _1.ptx:27 => (ptx_parser.cc:577) add_function_arg "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_5" + _1.ptx:27 => (ptx_parser.cc:219) add_directive + _1.ptx:27 => (ptx_parser.cc:144) init_directive_state + _1.ptx:28 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:28 => (ptx_parser.cc:631) add_scalar_type_spec "F32_TYPE" + _1.ptx:28 => (ptx_parser.cc:331) set_variable_type space_spec=param_space_kernel scalar_type_spec=F32_TYPE + _1.ptx:28 => (ptx_parser.cc:381) add_identifier "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_6" (10) + _1.ptx:28 => (ptx_parser.cc:577) add_function_arg "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_6" + _1.ptx:28 => (ptx_parser.cc:219) add_directive + _1.ptx:28 => (ptx_parser.cc:144) init_directive_state + _1.ptx:29 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:29 => (ptx_parser.cc:631) add_scalar_type_spec "F32_TYPE" + _1.ptx:29 => (ptx_parser.cc:331) set_variable_type space_spec=param_space_kernel scalar_type_spec=F32_TYPE + _1.ptx:30 => (ptx_parser.cc:381) add_identifier "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_7" (11) + _1.ptx:30 => (ptx_parser.cc:577) add_function_arg "_Z12wmma_exampleP6__halfS0_Pfiiiff_param_7" + _1.ptx:30 => (ptx_parser.cc:219) add_directive + _1.ptx:30 => (ptx_parser.cc:144) init_directive_state + _1.ptx:32 => (ptx_parser.cc:605) add_space_spec "local_space" + _1.ptx:32 => (ptx_parser.cc:590) add_alignment_spec + _1.ptx:32 => (ptx_parser.cc:631) add_scalar_type_spec "B8_TYPE" + _1.ptx:32 => (ptx_parser.cc:331) set_variable_type space_spec=local_space scalar_type_spec=B8_TYPE + _1.ptx:32 => (ptx_parser.cc:381) add_identifier "__local_depot0" (12) +GPGPU-Sim PTX: allocating stack frame region for .local "__local_depot0" from 0x0 to 0x8 + _1.ptx:32 => (ptx_parser.cc:319) add_variables + _1.ptx:32 => (ptx_parser.cc:144) init_directive_state + _1.ptx:32 => (ptx_parser.cc:219) add_directive + _1.ptx:32 => (ptx_parser.cc:144) init_directive_state + _1.ptx:33 => (ptx_parser.cc:605) add_space_spec "reg_space" + _1.ptx:33 => (ptx_parser.cc:631) add_scalar_type_spec "B64_TYPE" + _1.ptx:33 => (ptx_parser.cc:331) set_variable_type space_spec=reg_space scalar_type_spec=B64_TYPE + _1.ptx:33 => (ptx_parser.cc:381) add_identifier "%SP" (13) + _1.ptx:33 => (ptx_parser.cc:319) add_variables + _1.ptx:33 => (ptx_parser.cc:144) init_directive_state + _1.ptx:33 => (ptx_parser.cc:219) add_directive + _1.ptx:33 => (ptx_parser.cc:144) init_directive_state + _1.ptx:34 => (ptx_parser.cc:605) add_space_spec "reg_space" + _1.ptx:34 => (ptx_parser.cc:631) add_scalar_type_spec "B64_TYPE" + _1.ptx:34 => (ptx_parser.cc:331) set_variable_type space_spec=reg_space scalar_type_spec=B64_TYPE + _1.ptx:34 => (ptx_parser.cc:381) add_identifier "%SPL" (14) + _1.ptx:34 => (ptx_parser.cc:319) add_variables + _1.ptx:34 => (ptx_parser.cc:144) init_directive_state + _1.ptx:34 => (ptx_parser.cc:219) add_directive + _1.ptx:34 => (ptx_parser.cc:144) init_directive_state + _1.ptx:35 => (ptx_parser.cc:605) add_space_spec "reg_space" + _1.ptx:35 => (ptx_parser.cc:631) add_scalar_type_spec "PRED_TYPE" + _1.ptx:35 => (ptx_parser.cc:331) set_variable_type space_spec=reg_space scalar_type_spec=PRED_TYPE + _1.ptx:35 => (ptx_parser.cc:381) add_identifier "%p0" (15) + _1.ptx:35 => (ptx_parser.cc:381) add_identifier "%p1" (16) + _1.ptx:35 => (ptx_parser.cc:381) add_identifier "%p2" (17) + _1.ptx:35 => (ptx_parser.cc:381) add_identifier "%p3" (18) + _1.ptx:35 => (ptx_parser.cc:381) add_identifier "%p4" (19) + _1.ptx:35 => (ptx_parser.cc:381) add_identifier "%p5" (20) + _1.ptx:35 => (ptx_parser.cc:319) add_variables + _1.ptx:35 => (ptx_parser.cc:144) init_directive_state + _1.ptx:35 => (ptx_parser.cc:219) add_directive + _1.ptx:35 => (ptx_parser.cc:144) init_directive_state + _1.ptx:36 => (ptx_parser.cc:605) add_space_spec "reg_space" + _1.ptx:36 => (ptx_parser.cc:631) add_scalar_type_spec "F32_TYPE" + _1.ptx:36 => (ptx_parser.cc:331) set_variable_type space_spec=reg_space scalar_type_spec=F32_TYPE + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f0" (21) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f1" (22) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f2" (23) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f3" (24) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f4" (25) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f5" (26) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f6" (27) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f7" (28) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f8" (29) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f9" (30) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f10" (31) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f11" (32) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f12" (33) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f13" (34) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f14" (35) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f15" (36) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f16" (37) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f17" (38) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f18" (39) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f19" (40) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f20" (41) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f21" (42) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f22" (43) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f23" (44) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f24" (45) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f25" (46) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f26" (47) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f27" (48) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f28" (49) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f29" (50) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f30" (51) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f31" (52) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f32" (53) + _1.ptx:36 => (ptx_parser.cc:381) add_identifier "%f33" (54) + _1.ptx:36 => (ptx_parser.cc:319) add_variables + _1.ptx:36 => (ptx_parser.cc:144) init_directive_state + _1.ptx:36 => (ptx_parser.cc:219) add_directive + _1.ptx:36 => (ptx_parser.cc:144) init_directive_state + _1.ptx:37 => (ptx_parser.cc:605) add_space_spec "reg_space" + _1.ptx:37 => (ptx_parser.cc:631) add_scalar_type_spec "B32_TYPE" + _1.ptx:37 => (ptx_parser.cc:331) set_variable_type space_spec=reg_space scalar_type_spec=B32_TYPE + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r0" (55) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r1" (56) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r2" (57) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r3" (58) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r4" (59) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r5" (60) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r6" (61) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r7" (62) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r8" (63) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r9" (64) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r10" (65) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r11" (66) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r12" (67) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r13" (68) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r14" (69) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r15" (70) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r16" (71) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r17" (72) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r18" (73) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r19" (74) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r20" (75) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r21" (76) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r22" (77) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r23" (78) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r24" (79) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r25" (80) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r26" (81) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r27" (82) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r28" (83) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r29" (84) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r30" (85) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r31" (86) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r32" (87) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r33" (88) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r34" (89) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r35" (90) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r36" (91) + _1.ptx:37 => (ptx_parser.cc:381) add_identifier "%r37" (92) + _1.ptx:37 => (ptx_parser.cc:319) add_variables + _1.ptx:37 => (ptx_parser.cc:144) init_directive_state + _1.ptx:37 => (ptx_parser.cc:219) add_directive + _1.ptx:37 => (ptx_parser.cc:144) init_directive_state + _1.ptx:38 => (ptx_parser.cc:605) add_space_spec "reg_space" + _1.ptx:38 => (ptx_parser.cc:631) add_scalar_type_spec "B64_TYPE" + _1.ptx:38 => (ptx_parser.cc:331) set_variable_type space_spec=reg_space scalar_type_spec=B64_TYPE + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd0" (93) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd1" (94) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd2" (95) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd3" (96) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd4" (97) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd5" (98) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd6" (99) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd7" (100) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd8" (101) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd9" (102) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd10" (103) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd11" (104) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd12" (105) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd13" (106) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd14" (107) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd15" (108) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd16" (109) + _1.ptx:38 => (ptx_parser.cc:381) add_identifier "%rd17" (110) + _1.ptx:38 => (ptx_parser.cc:319) add_variables + _1.ptx:38 => (ptx_parser.cc:144) init_directive_state + _1.ptx:38 => (ptx_parser.cc:219) add_directive + _1.ptx:38 => (ptx_parser.cc:144) init_directive_state + _1.ptx:41 => (ptx_parser.cc:631) add_scalar_type_spec "U64_TYPE" + _1.ptx:41 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:41 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:41 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:41 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:41 => (ptx_parser.cc:144) init_directive_state + _1.ptx:42 => (ptx_parser.cc:605) add_space_spec "local_space" + _1.ptx:42 => (ptx_parser.cc:631) add_scalar_type_spec "U64_TYPE" + _1.ptx:42 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:42 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:42 => (ptx_parser.cc:295) add_instruction: cvta + _1.ptx:42 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:42 => (ptx_parser.cc:144) init_directive_state + _1.ptx:43 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:43 => (ptx_parser.cc:631) add_scalar_type_spec "U64_TYPE" + _1.ptx:43 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:43 => (ptx_parser.cc:929) add_address_operand + _1.ptx:43 => (ptx_parser.cc:766) add_memory_operand + _1.ptx:43 => (ptx_parser.cc:295) add_instruction: ld + _1.ptx:43 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:43 => (ptx_parser.cc:144) init_directive_state + _1.ptx:44 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:44 => (ptx_parser.cc:631) add_scalar_type_spec "U64_TYPE" + _1.ptx:44 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:44 => (ptx_parser.cc:929) add_address_operand + _1.ptx:44 => (ptx_parser.cc:766) add_memory_operand + _1.ptx:44 => (ptx_parser.cc:295) add_instruction: ld + _1.ptx:44 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:44 => (ptx_parser.cc:144) init_directive_state + _1.ptx:45 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:45 => (ptx_parser.cc:631) add_scalar_type_spec "U64_TYPE" + _1.ptx:45 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:45 => (ptx_parser.cc:929) add_address_operand + _1.ptx:45 => (ptx_parser.cc:766) add_memory_operand + _1.ptx:45 => (ptx_parser.cc:295) add_instruction: ld + _1.ptx:45 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:45 => (ptx_parser.cc:144) init_directive_state + _1.ptx:46 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:46 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:46 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:46 => (ptx_parser.cc:929) add_address_operand + _1.ptx:46 => (ptx_parser.cc:766) add_memory_operand + _1.ptx:46 => (ptx_parser.cc:295) add_instruction: ld + _1.ptx:46 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:46 => (ptx_parser.cc:144) init_directive_state + _1.ptx:47 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:47 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:47 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:47 => (ptx_parser.cc:929) add_address_operand + _1.ptx:47 => (ptx_parser.cc:766) add_memory_operand + _1.ptx:47 => (ptx_parser.cc:295) add_instruction: ld + _1.ptx:47 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:47 => (ptx_parser.cc:144) init_directive_state + _1.ptx:48 => (ptx_parser.cc:605) add_space_spec "param_space_unclassified" + _1.ptx:48 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:48 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:48 => (ptx_parser.cc:929) add_address_operand + _1.ptx:48 => (ptx_parser.cc:766) add_memory_operand + _1.ptx:48 => (ptx_parser.cc:295) add_instruction: ld + _1.ptx:48 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:48 => (ptx_parser.cc:144) init_directive_state + _1.ptx:50 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:50 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:50 => (ptx_parser.cc:760) add_builtin_operand + _1.ptx:50 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:50 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:50 => (ptx_parser.cc:144) init_directive_state + _1.ptx:52 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:52 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:52 => (ptx_parser.cc:760) add_builtin_operand + _1.ptx:52 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:52 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:52 => (ptx_parser.cc:144) init_directive_state + _1.ptx:53 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:53 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:53 => (ptx_parser.cc:760) add_builtin_operand + _1.ptx:53 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:53 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:53 => (ptx_parser.cc:144) init_directive_state + _1.ptx:54 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:54 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:54 => (ptx_parser.cc:760) add_builtin_operand + _1.ptx:54 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:54 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:54 => (ptx_parser.cc:144) init_directive_state + _1.ptx:55 => (ptx_parser.cc:672) add_option + _1.ptx:55 => (ptx_parser.cc:631) add_scalar_type_spec "S32_TYPE" + _1.ptx:55 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:55 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:55 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:55 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:55 => (ptx_parser.cc:295) add_instruction: mad + _1.ptx:55 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:55 => (ptx_parser.cc:144) init_directive_state + _1.ptx:56 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:56 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:56 => (ptx_parser.cc:760) add_builtin_operand + _1.ptx:56 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:56 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:56 => (ptx_parser.cc:144) init_directive_state + _1.ptx:57 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:57 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:57 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:57 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:57 => (ptx_parser.cc:295) add_instruction: div + _1.ptx:57 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:57 => (ptx_parser.cc:144) init_directive_state + _1.ptx:58 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:58 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:58 => (ptx_parser.cc:760) add_builtin_operand + _1.ptx:58 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:58 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:58 => (ptx_parser.cc:144) init_directive_state + _1.ptx:59 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:59 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:59 => (ptx_parser.cc:760) add_builtin_operand + _1.ptx:59 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:59 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:59 => (ptx_parser.cc:144) init_directive_state + _1.ptx:60 => (ptx_parser.cc:631) add_scalar_type_spec "U32_TYPE" + _1.ptx:60 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:60 => (ptx_parser.cc:760) add_builtin_operand + _1.ptx:60 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:60 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:60 => (ptx_parser.cc:144) init_directive_state + _1.ptx:61 => (ptx_parser.cc:672) add_option + _1.ptx:61 => (ptx_parser.cc:631) add_scalar_type_spec "S32_TYPE" + _1.ptx:61 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:61 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:61 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:61 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:61 => (ptx_parser.cc:295) add_instruction: mad + _1.ptx:61 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:61 => (ptx_parser.cc:144) init_directive_state + _1.ptx:62 => (ptx_parser.cc:631) add_scalar_type_spec "B32_TYPE" + _1.ptx:62 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:62 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:62 => (ptx_parser.cc:883) add_literal_int + _1.ptx:62 => (ptx_parser.cc:295) add_instruction: shl + _1.ptx:62 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:62 => (ptx_parser.cc:144) init_directive_state + _1.ptx:63 => (ptx_parser.cc:631) add_scalar_type_spec "B32_TYPE" + _1.ptx:63 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:63 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:63 => (ptx_parser.cc:883) add_literal_int + _1.ptx:63 => (ptx_parser.cc:295) add_instruction: shl + _1.ptx:63 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:63 => (ptx_parser.cc:144) init_directive_state + _1.ptx:64 => (ptx_parser.cc:672) add_option + _1.ptx:64 => (ptx_parser.cc:631) add_scalar_type_spec "S32_TYPE" + _1.ptx:64 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:64 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:64 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:64 => (ptx_parser.cc:295) add_instruction: setp + _1.ptx:64 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:64 => (ptx_parser.cc:144) init_directive_state + _1.ptx:65 => (ptx_parser.cc:672) add_option + _1.ptx:65 => (ptx_parser.cc:631) add_scalar_type_spec "S32_TYPE" + _1.ptx:65 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:65 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:65 => (ptx_parser.cc:883) add_literal_int + _1.ptx:65 => (ptx_parser.cc:295) add_instruction: setp + _1.ptx:65 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:65 => (ptx_parser.cc:144) init_directive_state + _1.ptx:66 => (ptx_parser.cc:631) add_scalar_type_spec "PRED_TYPE" + _1.ptx:66 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:66 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:66 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:66 => (ptx_parser.cc:295) add_instruction: and + _1.ptx:66 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:66 => (ptx_parser.cc:144) init_directive_state + _1.ptx:67 => (ptx_parser.cc:672) add_option + _1.ptx:67 => (ptx_parser.cc:631) add_scalar_type_spec "S32_TYPE" + _1.ptx:67 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:67 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:67 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:67 => (ptx_parser.cc:295) add_instruction: setp + _1.ptx:67 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:67 => (ptx_parser.cc:144) init_directive_state + _1.ptx:68 => (ptx_parser.cc:631) add_scalar_type_spec "PRED_TYPE" + _1.ptx:68 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:68 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:68 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:68 => (ptx_parser.cc:295) add_instruction: and + _1.ptx:68 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:68 => (ptx_parser.cc:144) init_directive_state + _1.ptx:69 => (ptx_parser.cc:631) add_scalar_type_spec "F32_TYPE" + _1.ptx:69 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:69 => (ptx_parser.cc:889) add_literal_float + _1.ptx:69 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:69 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:69 => (ptx_parser.cc:144) init_directive_state + _1.ptx:70 => (ptx_parser.cc:631) add_scalar_type_spec "F32_TYPE" + _1.ptx:70 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:70 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:70 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:70 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:70 => (ptx_parser.cc:144) init_directive_state + _1.ptx:71 => (ptx_parser.cc:631) add_scalar_type_spec "F32_TYPE" + _1.ptx:71 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:71 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:71 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:71 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:71 => (ptx_parser.cc:144) init_directive_state + _1.ptx:72 => (ptx_parser.cc:631) add_scalar_type_spec "F32_TYPE" + _1.ptx:72 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:72 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:72 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:72 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:72 => (ptx_parser.cc:144) init_directive_state + _1.ptx:73 => (ptx_parser.cc:631) add_scalar_type_spec "F32_TYPE" + _1.ptx:73 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:73 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:73 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:73 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:73 => (ptx_parser.cc:144) init_directive_state + _1.ptx:74 => (ptx_parser.cc:631) add_scalar_type_spec "F32_TYPE" + _1.ptx:74 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:74 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:74 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:74 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:74 => (ptx_parser.cc:144) init_directive_state + _1.ptx:75 => (ptx_parser.cc:631) add_scalar_type_spec "F32_TYPE" + _1.ptx:75 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:75 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:75 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:75 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:75 => (ptx_parser.cc:144) init_directive_state + _1.ptx:76 => (ptx_parser.cc:631) add_scalar_type_spec "F32_TYPE" + _1.ptx:76 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:76 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:76 => (ptx_parser.cc:295) add_instruction: mov + _1.ptx:76 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:76 => (ptx_parser.cc:144) init_directive_state + _1.ptx:77 => (ptx_parser.cc:659) add_pred + _1.ptx:77 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:77 => (ptx_parser.cc:295) add_instruction: bra + _1.ptx:77 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:77 => (ptx_parser.cc:144) init_directive_state + _1.ptx:78 => (ptx_parser.cc:672) add_option + _1.ptx:78 => (ptx_parser.cc:901) add_scalar_operand + _1.ptx:78 => (ptx_parser.cc:295) add_instruction: bra + _1.ptx:78 => (ptx_parser.cc:159) init_instruction_state + _1.ptx:78 => (ptx_parser.cc:144) init_directive_state + _1.ptx:80 => (ptx_parser.cc:643) add_label + _1.ptx:80 => (ptx_parser.cc:295) add_instruction: