/* * self_view.c - example of self-monitoring with PMD access via remapping * * Copyright (c) 2005-2006 Hewlett-Packard Development Company, L.P. * Contributed by Stephane Eranian * * Permission is hereby granted, free of charge, to any person obtaining a copy * of this software and associated documentation files (the "Software"), to deal * in the Software without restriction, including without limitation the rights * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies * of the Software, and to permit persons to whom the Software is furnished to do so, * subject to the following conditions: * * The above copyright notice and this permission notice shall be included in all * copies or substantial portions of the Software. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, * INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A * PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF * CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE * OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #include #include #include #include #include #include #include #include #include #include #include #include #include "detect_pmcs.h" #define NUM_PMCS PFMLIB_MAX_PMCS #define NUM_PMDS PFMLIB_MAX_PMDS #define MAX_EVT_NAME_LEN 128 typedef struct { uint64_t val; unsigned int reg_num; char name[MAX_EVT_NAME_LEN]; } pmd_val_t; static pfm_set_view_t *view0; static pmd_val_t pmdv[NUM_PMDS]; static unsigned int num_pmdv; static uint64_t ovfl_mask; static void fatal_error(char *fmt,...) __attribute__((noreturn)); static void fatal_error(char *fmt, ...) { va_list ap; va_start(ap, fmt); vfprintf(stderr, fmt, ap); va_end(ap); exit(1); } #ifdef __ia64__ #define get_pmd(n) ia64_get_pmd(n) #elif defined(__x86_64__) || defined(__i386__) /* * IMPORTANT: * there is an implicit assumption here that the perfmon-2 * PMD mapping and the rdpmc mapping correspond. As such * pmd0 -> rdpmc(0) * pmd1 -> rdpmc(1) * ... */ #define rdpmc(counter,low,high) \ __asm__ __volatile__("rdpmc" \ : "=a" (low), "=d" (high) \ : "c" (counter)) uint64_t get_pmd(unsigned int n) { uint32_t high, low; uint64_t tmp; rdpmc(n, low, high); tmp = (uint64_t)high <<32 | low; return tmp; } #elif defined(__mips__) || defined(__powerpc__) /* * XXX: MIPS does not have an instruction to read a counter at the user level */ uint64_t get_pmd(unsigned int n) { return(0ULL); } #else #error "you need to define get_pmd for your architecture" #endif static void show_view_self(int is_loaded) { unsigned int i, is_active; uint64_t val, hw_val; uint64_t mask = ovfl_mask; unsigned long retries = 0; unsigned long start_seq, end_seq; retries = -1; /* * print the results */ retry: retries++; /* * get initial sequence number, if the number changes * while we are scanning, it means the view was updated * and we need to retry */ start_seq = view0->set_seq; /* * active is true if the set is the active set */ is_active = view0->set_status & PFM_SETVFL_ACTIVE; printf("retries=%lu active=%d view_seq=%lu set_runs=%"PRIu64"\n", retries, view0->set_status, view0->set_seq, view0->set_runs); /* * extract information directly from view */ for (i=0; i < num_pmdv; i++) { val = view0->set_pmds[pmdv[i].reg_num]; /* * is context is attached and set is active then * we need to complement the software value with * the current hardware value. For self-monitoring * with simply need to read the PMD */ if (is_loaded && is_active) { hw_val = get_pmd(pmdv[i].reg_num); /* * update the lower portion of the 64-bit * value with the hardware value */ val = (val & ~mask) | (hw_val & mask); } pmdv[i].val = val; } /* * check if sequence number is still the same */ end_seq = view0->set_seq; if (end_seq != start_seq) goto retry; /* * print results */ for (i=0; i < num_pmdv; i++) { printf("%20"PRIu64" %s\n", pmdv[i].val, pmdv[i].name); } } uint64_t noploop(uint64_t loop) { while (loop--) { if ((loop % 10000) == 0) show_view_self(1); } return loop; } int main(int argc, char **argv) { char **p; unsigned int i; int ret, ctx_fd; pfmlib_input_param_t inp; pfmlib_output_param_t outp; pfarg_pmd_t pd[NUM_PMDS]; pfarg_pmc_t pc[NUM_PMCS]; pfarg_ctx_t ctx[1]; pfarg_setinfo_t setinfo; pfarg_load_t load_args; pfmlib_options_t pfmlib_options; unsigned int num_counters, width; #ifdef __mips__ printf("<>\n"); #endif /* * pass options to library (optional) */ memset(&pfmlib_options, 0, sizeof(pfmlib_options)); pfmlib_options.pfm_debug = 0; /* set to 1 for debug */ pfmlib_options.pfm_verbose = 1; /* set to 1 for debug */ pfm_set_options(&pfmlib_options); /* * Initialize pfm library (required before we can use it) */ ret = pfm_initialize(); if (ret != PFMLIB_SUCCESS) fatal_error("Cannot initialize library: %s\n", pfm_strerror(ret)); pfm_get_num_counters(&num_counters); pfm_get_hw_counter_width(&width); ovfl_mask = (1ULL << width)-1; printf("width=%u ovfl_mask=0x%"PRIx64"\n", width, ovfl_mask); memset(pd, 0, sizeof(pd)); memset(pc, 0, sizeof(pc)); memset(ctx, 0, sizeof(ctx)); memset(&load_args, 0, sizeof(load_args)); memset(&setinfo, 0, sizeof(setinfo)); /* * prepare parameters to library. we don't use any Itanium * specific features here. so the pfp_model is NULL. */ memset(&inp,0, sizeof(inp)); memset(&outp,0, sizeof(outp)); /* * be nice to user! */ if (argc > 1) { p = argv+1; for (i=0; *p ; i++, p++) { if (pfm_find_full_event(*p, &inp.pfp_events[i]) != PFMLIB_SUCCESS) { fatal_error("Cannot find %s event\n", *p); } } } else { if (pfm_get_cycle_event(&inp.pfp_events[0]) != PFMLIB_SUCCESS) { fatal_error("cannot find cycle event\n"); } if (pfm_get_inst_retired_event(&inp.pfp_events[1]) != PFMLIB_SUCCESS) { fatal_error("cannot find inst retired event\n"); } i = 2; } /* * set the default privilege mode for all counters: * PFM_PLM3 : user level only */ inp.pfp_dfl_plm = PFM_PLM3; if (i > num_counters) { i = num_counters; printf("too many events provided (max=%d events), using first %d event(s)\n", num_counters, i); } /* * how many counters we use */ inp.pfp_event_count = i; ctx[0].ctx_flags = PFM_FL_MAP_SETS; /* * now create a new context, per process context. * This just creates a new context with some initial state, it is not * active nor attached to any process. */ ctx_fd = pfm_create_context(ctx, NULL, NULL, 0); if (ctx_fd == -1) { if (errno == ENOSYS) { fatal_error("Your kernel does not have performance monitoring support!\n"); } fatal_error("Can't create PFM context %s\n", strerror(errno)); } /* * build the pfp_unavail_pmcs bitmask by looking * at what perfmon has available. It is not always * the case that all PMU registers are actually available * to applications. For instance, on IA-32 platforms, some * registers may be reserved for the NMI watchdog timer. * * With this bitmap, the library knows which registers NOT to * use. Of source, it is possible that no valid assignement may * be possible if certina PMU registers are not available. */ detect_unavail_pmcs(ctx_fd, &inp.pfp_unavail_pmcs); /* * let the library figure out the values for the PMCS */ if ((ret=pfm_dispatch_events(&inp, NULL, &outp, NULL)) != PFMLIB_SUCCESS) { fatal_error("cannot configure events: %s\n", pfm_strerror(ret)); } /* * Now prepare the argument to initialize the PMDs and PMCS. * We must pfp_pmc_count to determine the number of PMC to intialize. * We must use pfp_event_count to determine the number of PMD to initialize. * Some events causes extra PMCs to be used, so pfp_pmc_count may be >= pfp_event_count. * * This step is new compared to libpfm-2.x. It is necessary because the library no * longer knows about the kernel data structures. */ for (i=0; i < outp.pfp_pmc_count; i++) { pc[i].reg_num = outp.pfp_pmcs[i].reg_num; pc[i].reg_value = outp.pfp_pmcs[i].reg_value; } for(i=0; i < outp.pfp_pmd_count; i++) pd[i].reg_num = outp.pfp_pmds[i].reg_num; /* * Now program the registers * * We don't use the same variable to indicate the number of elements passed to * the kernel because, as we said earlier, pc may contain more elements than * the number of events (pmd) we specified, i.e., contains more than counting * monitors. */ if (pfm_write_pmcs(ctx_fd, pc, outp.pfp_pmc_count)) fatal_error("pfm_write_pmcs error errno %d\n",errno); /* * To be read, each PMD must be either written or declared * as being part of a sample (reg_smpl_pmds) */ if (pfm_write_pmds(ctx_fd, pd, outp.pfp_pmd_count)) fatal_error("pfm)write_pmds error errno %d\n",errno); /* * now we load (i.e., attach) the context to ourself */ load_args.load_pid = getpid(); if (pfm_load_context(ctx_fd, &load_args)) fatal_error("pfm_load_context error errno %d\n",errno); /* * retrieve set0 information */ if (pfm_getinfo_evtsets(ctx_fd, &setinfo, 1)) fatal_error("pfm_getinfo_evtsets error errno %d\n",errno); printf("set_id=0 mmap_offset=%llu\n", (unsigned long long)setinfo.set_mmap_offset); view0 = (pfm_set_view_t *)mmap(NULL, getpagesize(), PROT_READ, MAP_PRIVATE, ctx_fd, setinfo.set_mmap_offset); if (view0 == MAP_FAILED) { fatal_error("cannot mmap set view errno %d\n", errno); } printf("view=%p set_id=0 set_status=%u\n", view0, view0->set_status); /* * Let's roll now */ pfm_self_start(ctx_fd); noploop(10000000); pfm_self_stop(ctx_fd); /* * unload context (stop + unload) */ if (pfm_unload_context(ctx_fd)) fatal_error( "pfm_unload_context error errno %d\n",errno); /* * now read the results */ if (pfm_read_pmds(ctx_fd, pd, outp.pfp_pmd_count)) fatal_error( "pfm_read_pmds error errno %d\n",errno); printf("results using pfm_read_pmds:\n"); for (i=0; i < num_pmdv; i++) { printf("%20"PRIu64" %s\n", pd[i].reg_value, pmdv[i].name); } show_view_self(0); /* * and destroy our context */ munmap(view0, getpagesize()); close(ctx_fd); return 0; }