URL
https://opencores.org/ocsvn/neorv32/neorv32/trunk
Subversion Repositories neorv32
[/] [neorv32/] [trunk/] [sw/] [example/] [coremark/] [core_portme.c] - Rev 65
Go to most recent revision | Compare with Previous | Blame | View Log
/* Copyright 2018 Embedded Microprocessor Benchmark Consortium (EEMBC) Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. Original Author: Shay Gal-on */ /* Modified for the NEORV32 Processor - by Stephan Nolting */ #include "coremark.h" #include "core_portme.h" #if VALIDATION_RUN volatile ee_s32 seed1_volatile = 0x3415; volatile ee_s32 seed2_volatile = 0x3415; volatile ee_s32 seed3_volatile = 0x66; #endif #if PERFORMANCE_RUN volatile ee_s32 seed1_volatile = 0x0; volatile ee_s32 seed2_volatile = 0x0; volatile ee_s32 seed3_volatile = 0x66; #endif #if PROFILE_RUN volatile ee_s32 seed1_volatile = 0x8; volatile ee_s32 seed2_volatile = 0x8; volatile ee_s32 seed3_volatile = 0x8; #endif volatile ee_s32 seed4_volatile = ITERATIONS; volatile ee_s32 seed5_volatile = 0; /* Porting : Timing functions How to capture time and convert to seconds must be ported to whatever is supported by the platform. e.g. Read value from on board RTC, read value from cpu clock cycles performance counter etc. Sample implementation for standard time.h and windows.h definitions included. */ CORETIMETYPE barebones_clock() { /* #error \ "You must implement a method to measure time in barebones_clock()! This function should return current time.\n" */ return 1; } /* Define : TIMER_RES_DIVIDER Divider to trade off timer resolution and total time that can be measured. Use lower values to increase resolution, but make sure that overflow does not occur. If there are issues with the return value overflowing, increase this value. */ #define GETMYTIME(_t) (*_t = (CORETIMETYPE)neorv32_cpu_get_cycle()) #define MYTIMEDIFF(fin, ini) ((fin) - (ini)) #define TIMER_RES_DIVIDER 1 #define SAMPLE_TIME_IMPLEMENTATION 1 #define EE_TICKS_PER_SEC (CLOCKS_PER_SEC / TIMER_RES_DIVIDER) /** Define Host specific (POSIX), or target specific global time variables. */ static CORETIMETYPE start_time_val, stop_time_val; /* Function : start_time This function will be called right before starting the timed portion of the benchmark. Implementation may be capturing a system timer (as implemented in the example code) or zeroing some system parameters - e.g. setting the cpu clocks cycles to 0. */ void start_time(void) { neorv32_cpu_csr_write(CSR_MCOUNTINHIBIT, 0); // start all counters GETMYTIME(&start_time_val); } /* Function : stop_time This function will be called right after ending the timed portion of the benchmark. Implementation may be capturing a system timer (as implemented in the example code) or other system parameters - e.g. reading the current value of cpu cycles counter. */ void stop_time(void) { neorv32_cpu_csr_write(CSR_MCOUNTINHIBIT, -1); // stop all counters GETMYTIME(&stop_time_val); } /* Function : get_time Return an abstract "ticks" number that signifies time on the system. Actual value returned may be cpu cycles, milliseconds or any other value, as long as it can be converted to seconds by <time_in_secs>. This methodology is taken to accomodate any hardware or simulated platform. The sample implementation returns millisecs by default, and the resolution is controlled by <TIMER_RES_DIVIDER> */ CORE_TICKS get_time(void) { CORE_TICKS elapsed = (CORE_TICKS)(MYTIMEDIFF(stop_time_val, start_time_val)); return elapsed; } /* Function : time_in_secs Convert the value returned by get_time to seconds. The <secs_ret> type is used to accomodate systems with no support for floating point. Default implementation implemented by the EE_TICKS_PER_SEC macro above. */ secs_ret time_in_secs(CORE_TICKS ticks) { /* NEORV32-specific */ secs_ret retval = (secs_ret)(((CORE_TICKS)ticks) / ((CORE_TICKS)NEORV32_SYSINFO.CLK)); return retval; } ee_u32 default_num_contexts = 1; /* Number of available hardware performance monitors */ uint32_t num_hpm_cnts_global = 0; /* Function : portable_init Target specific initialization code Test for some common mistakes. */ #ifndef RUN_COREMARK void __attribute__((__noreturn__)) portable_init(core_portable *p, int *argc, char *argv[]) #else void portable_init(core_portable *p, int *argc, char *argv[]) #endif { /* NEORV32-specific */ neorv32_cpu_dint(); // no interrupt, thanks neorv32_rte_setup(); // capture all exceptions and give debug information, ho hw flow control neorv32_uart0_setup(BAUD_RATE, PARITY_NONE, FLOW_CONTROL_NONE); // Disable coremark compilation by default #ifndef RUN_COREMARK #warning COREMARK HAS NOT BEEN COMPILED! Use >>make USER_FLAGS+=-DRUN_COREMARK clean_all exe<< to compile it. // inform the user if you are actually executing this neorv32_uart0_printf("ERROR! CoreMark has not been compiled. Use >>make USER_FLAGS+=-DRUN_COREMARK clean_all exe<< to compile it.\n"); while(1); #endif // check available hardware extensions and compare with compiler flags neorv32_rte_check_isa(0); // silent = 0 -> show message if isa mismatch num_hpm_cnts_global = neorv32_cpu_hpm_get_counters(); // try to setup as many HPMs as possible neorv32_cpu_csr_write(CSR_MHPMCOUNTER3, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT3, 1 << HPMCNT_EVENT_CIR); neorv32_cpu_csr_write(CSR_MHPMCOUNTER4, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT4, 1 << HPMCNT_EVENT_WAIT_IF); neorv32_cpu_csr_write(CSR_MHPMCOUNTER5, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT5, 1 << HPMCNT_EVENT_WAIT_II); neorv32_cpu_csr_write(CSR_MHPMCOUNTER6, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT6, 1 << HPMCNT_EVENT_WAIT_MC); neorv32_cpu_csr_write(CSR_MHPMCOUNTER7, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT7, 1 << HPMCNT_EVENT_LOAD); neorv32_cpu_csr_write(CSR_MHPMCOUNTER8, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT8, 1 << HPMCNT_EVENT_STORE); neorv32_cpu_csr_write(CSR_MHPMCOUNTER9, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT9, 1 << HPMCNT_EVENT_WAIT_LS); neorv32_cpu_csr_write(CSR_MHPMCOUNTER10, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT10, 1 << HPMCNT_EVENT_JUMP); neorv32_cpu_csr_write(CSR_MHPMCOUNTER11, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT11, 1 << HPMCNT_EVENT_BRANCH); neorv32_cpu_csr_write(CSR_MHPMCOUNTER12, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT12, 1 << HPMCNT_EVENT_TBRANCH); neorv32_cpu_csr_write(CSR_MHPMCOUNTER13, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT13, 1 << HPMCNT_EVENT_TRAP); neorv32_cpu_csr_write(CSR_MHPMCOUNTER14, 0); neorv32_cpu_csr_write(CSR_MHPMEVENT14, 1 << HPMCNT_EVENT_ILLEGAL); neorv32_uart0_printf("NEORV32: Processor running at %u Hz\n", (uint32_t)NEORV32_SYSINFO.CLK); neorv32_uart0_printf("NEORV32: Executing coremark (%u iterations). This may take some time...\n\n", (uint32_t)ITERATIONS); // clear cycle counter neorv32_cpu_set_mcycle(0); neorv32_cpu_csr_write(CSR_MCOUNTEREN, -1); // enable access to all counters /* #error \ "Call board initialization routines in portable init (if needed), in particular initialize UART!\n" */ if (sizeof(ee_ptr_int) != sizeof(ee_u8 *)) { ee_printf( "ERROR! Please define ee_ptr_int to a type that holds a " "pointer!\n"); } if (sizeof(ee_u32) != 4) { ee_printf("ERROR! Please define ee_u32 to a 32b unsigned type!\n"); } p->portable_id = 1; #ifndef RUN_COREMARK while(1); #endif } /* Function : portable_fini Target specific final code */ void portable_fini(core_portable *p) { p->portable_id = 0; /* NEORV32-specific */ // show executed instructions, required cycles and resulting average CPI union { uint64_t uint64; uint32_t uint32[sizeof(uint64_t)/2]; } exe_instructions, exe_time; exe_time.uint64 = (uint64_t)get_time(); exe_instructions.uint64 = neorv32_cpu_get_instret(); neorv32_uart0_printf("\nNEORV32: All reported numbers only show the integer part.\n\n"); neorv32_uart0_printf("NEORV32: HPM results\n"); if (num_hpm_cnts_global == 0) {neorv32_uart0_printf("no HPMs available\n"); } if (num_hpm_cnts_global > 0) {neorv32_uart0_printf("# Retired compr. instructions: %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER3)); } if (num_hpm_cnts_global > 1) {neorv32_uart0_printf("# I-fetch wait cycles: %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER4)); } if (num_hpm_cnts_global > 2) {neorv32_uart0_printf("# I-issue wait cycles: %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER5)); } if (num_hpm_cnts_global > 3) {neorv32_uart0_printf("# Multi-cycle ALU wait cycles: %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER6)); } if (num_hpm_cnts_global > 4) {neorv32_uart0_printf("# Load operations: %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER7)); } if (num_hpm_cnts_global > 5) {neorv32_uart0_printf("# Store operations: %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER8)); } if (num_hpm_cnts_global > 6) {neorv32_uart0_printf("# Load/store wait cycles: %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER9)); } if (num_hpm_cnts_global > 7) {neorv32_uart0_printf("# Unconditional jumps: %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER10)); } if (num_hpm_cnts_global > 8) {neorv32_uart0_printf("# Conditional branches (all): %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER11)); } if (num_hpm_cnts_global > 9) {neorv32_uart0_printf("# Conditional branches (taken): %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER12)); } if (num_hpm_cnts_global > 10) {neorv32_uart0_printf("# Entered traps: %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER13)); } if (num_hpm_cnts_global > 11) {neorv32_uart0_printf("# Illegal operations: %u\n", (uint32_t)neorv32_cpu_csr_read(CSR_MHPMCOUNTER14)); } neorv32_uart0_printf("\n"); neorv32_uart0_printf("NEORV32: Executed instructions 0x%x_%x\n", (uint32_t)exe_instructions.uint32[1], (uint32_t)exe_instructions.uint32[0]); neorv32_uart0_printf("NEORV32: CoreMark core clock cycles 0x%x_%x\n", (uint32_t)exe_time.uint32[1], (uint32_t)exe_time.uint32[0]); uint64_t average_cpi_int = exe_time.uint64 / exe_instructions.uint64; neorv32_uart0_printf("NEORV32: Average CPI (integer part only): %u cycles/instruction\n", (uint32_t)average_cpi_int); }
Go to most recent revision | Compare with Previous | Blame | View Log