X-Git-Url: https://git.kernel.dk/?p=fio.git;a=blobdiff_plain;f=t%2Fdedupe.c;h=12ede8f0494566017674c97f1295b8c0e6efd372;hp=16c1748d86f19131f7985daa9b6eecda168fbbf0;hb=9ca144fec3077984e37397d0c1d1f6d098841e2e;hpb=1956a9e5ce9947a81a3c6de619957b6d1313b3f9;ds=sidebyside diff --git a/t/dedupe.c b/t/dedupe.c index 16c1748d..12ede8f0 100644 --- a/t/dedupe.c +++ b/t/dedupe.c @@ -27,13 +27,8 @@ #include "../gettime.h" #include "../fio_time.h" -FILE *f_err; -struct timeval *fio_tv = NULL; -unsigned int fio_debug = 0; - -void __dprint(int type, const char *str, ...) -{ -} +#include "../lib/bloom.h" +#include "debug.h" struct worker_thread { pthread_t thread; @@ -45,6 +40,7 @@ struct worker_thread { uint64_t size; unsigned long items; + unsigned long dupes; int err; }; @@ -66,6 +62,7 @@ struct item { }; static struct rb_root rb_root; +static struct bloom *bloom; static struct fio_mutex *rb_lock; static unsigned int blocksize = 4096; @@ -75,6 +72,7 @@ static unsigned int dump_output; static unsigned int odirect; static unsigned int collision_check; static unsigned int print_progress = 1; +static unsigned int use_bloom = 1; static uint64_t total_size; static uint64_t cur_offset; @@ -116,17 +114,17 @@ static int get_work(uint64_t *offset, uint64_t *size) return ret; } -static int read_block(int fd, void *buf, off_t offset) +static int __read_block(int fd, void *buf, off_t offset, size_t count) { ssize_t ret; - ret = pread(fd, buf, blocksize, offset); + ret = pread(fd, buf, count, offset); if (ret < 0) { perror("pread"); return 1; } else if (!ret) return 1; - else if (ret != blocksize) { + else if (ret != count) { log_err("dedupe: short read on block\n"); return 1; } @@ -134,6 +132,11 @@ static int read_block(int fd, void *buf, off_t offset) return 0; } +static int read_block(int fd, void *buf, off_t offset) +{ + return __read_block(fd, buf, offset, blocksize); +} + static void add_item(struct chunk *c, struct item *i) { /* @@ -231,14 +234,24 @@ add: add_item(c, i); } -static void insert_chunks(struct item *items, unsigned int nitems) +static void insert_chunks(struct item *items, unsigned int nitems, + uint64_t *ndupes) { int i; fio_mutex_down(rb_lock); - for (i = 0; i < nitems; i++) - insert_chunk(&items[i]); + for (i = 0; i < nitems; i++) { + if (bloom) { + unsigned int s; + int r; + + s = sizeof(items[i].hash) / sizeof(uint32_t); + r = bloom_set(bloom, items[i].hash, s); + *ndupes += r; + } else + insert_chunk(&items[i]); + } fio_mutex_up(rb_lock); } @@ -252,30 +265,46 @@ static void crc_buf(void *buf, uint32_t *hash) fio_md5_final(&ctx); } +static unsigned int read_blocks(int fd, void *buf, off_t offset, size_t size) +{ + if (__read_block(fd, buf, offset, size)) + return 0; + + return size / blocksize; +} + static int do_work(struct worker_thread *thread, void *buf) { unsigned int nblocks, i; off_t offset; - int err = 0, nitems = 0; + int nitems = 0; + uint64_t ndupes = 0; struct item *items; - nblocks = thread->size / blocksize; offset = thread->cur_offset; + + nblocks = read_blocks(thread->fd, buf, offset, min(thread->size, (uint64_t)chunk_size)); + if (!nblocks) + return 1; + items = malloc(sizeof(*items) * nblocks); for (i = 0; i < nblocks; i++) { - if (read_block(thread->fd, buf, offset)) - break; - items[i].offset = offset; - crc_buf(buf, items[i].hash); + void *thisptr = buf + (i * blocksize); + + if (items) + items[i].offset = offset; + crc_buf(thisptr, items[i].hash); offset += blocksize; nitems++; } - insert_chunks(items, nitems); - thread->items += nitems; + insert_chunks(items, nitems, &ndupes); + free(items); - return err; + thread->items += nitems; + thread->dupes += ndupes; + return 0; } static void *thread_fn(void *data) @@ -283,7 +312,7 @@ static void *thread_fn(void *data) struct worker_thread *thread = data; void *buf; - buf = fio_memalign(blocksize, blocksize); + buf = fio_memalign(blocksize, chunk_size); do { if (get_work(&thread->cur_offset, &thread->size)) { @@ -297,7 +326,7 @@ static void *thread_fn(void *data) } while (1); thread->done = 1; - fio_memfree(buf, blocksize); + fio_memfree(buf, chunk_size); return NULL; } @@ -343,7 +372,8 @@ static void show_progress(struct worker_thread *threads, unsigned long total) }; } -static int run_dedupe_threads(int fd, uint64_t dev_size) +static int run_dedupe_threads(int fd, uint64_t dev_size, uint64_t *nextents, + uint64_t *nchunks) { struct worker_thread *threads; unsigned long nitems, total_items; @@ -371,20 +401,27 @@ static int run_dedupe_threads(int fd, uint64_t dev_size) show_progress(threads, total_items); nitems = 0; + *nextents = 0; + *nchunks = 1; for (i = 0; i < num_threads; i++) { void *ret; pthread_join(threads[i].thread, &ret); nitems += threads[i].items; + *nchunks += threads[i].dupes; } printf("Threads(%u): %lu items processed\n", num_threads, nitems); + *nextents = nitems; + *nchunks = nitems - *nchunks; + fio_mutex_remove(size_lock); free(threads); return err; } -static int dedupe_check(const char *filename) +static int dedupe_check(const char *filename, uint64_t *nextents, + uint64_t *nchunks) { uint64_t dev_size; struct stat sb; @@ -412,9 +449,16 @@ static int dedupe_check(const char *filename) return 1; } - printf("Will check <%s>, size <%llu>\n", filename, (unsigned long long) dev_size); + if (use_bloom) { + uint64_t bloom_entries; + + bloom_entries = (3 * dev_size ) / (blocksize * 2); + bloom = bloom_new(bloom_entries); + } + + printf("Will check <%s>, size <%llu>, using %u threads\n", filename, (unsigned long long) dev_size, num_threads); - return run_dedupe_threads(dev_fd, dev_size); + return run_dedupe_threads(dev_fd, dev_size, nextents, nchunks); } static void show_chunk(struct chunk *c) @@ -429,14 +473,25 @@ static void show_chunk(struct chunk *c) } } -static void iter_rb_tree(void) +static void show_stat(uint64_t nextents, uint64_t nchunks) +{ + double perc, ratio; + + printf("Extents=%lu, Unique extents=%lu\n", (unsigned long) nextents, (unsigned long) nchunks); + ratio = (double) nextents / (double) nchunks; + printf("De-dupe ratio: 1:%3.2f\n", ratio - 1.0); + + perc = 1.00 - ((double) nchunks / (double) nextents); + perc *= 100.0; + printf("Fio setting: dedupe_percentage=%u\n", (int) (perc + 0.50)); + +} + +static void iter_rb_tree(uint64_t *nextents, uint64_t *nchunks) { struct rb_node *n; - uint64_t nchunks; - uint64_t nextents; - double perc; - nchunks = nextents = 0; + *nchunks = *nextents = 0; n = rb_first(&rb_root); if (!n) @@ -446,20 +501,13 @@ static void iter_rb_tree(void) struct chunk *c; c = rb_entry(n, struct chunk, rb_node); - nchunks++; - nextents += c->count; + (*nchunks)++; + *nextents += c->count; if (dump_output) show_chunk(c); } while ((n = rb_next(n)) != NULL); - - printf("Extents=%lu, Unique extents=%lu\n", (unsigned long) nextents, (unsigned long) nchunks); - printf("De-dupe factor: %3.2f\n", (double) nextents / (double) nchunks); - - perc = 1.00 - ((double) nchunks / (double) nextents); - perc *= 100.0; - printf("Fio setting: dedupe_percentage=%u\n", (int) (perc + 0.50)); } static int usage(char *argv[]) @@ -471,15 +519,19 @@ static int usage(char *argv[]) log_err("\t-d\tFull extent/chunk debug output\n"); log_err("\t-o\tUse O_DIRECT\n"); log_err("\t-c\tFull collision check\n"); + log_err("\t-B\tUse probabilistic bloom filter\n"); log_err("\t-p\tPrint progress indicator\n"); return 1; } int main(int argc, char *argv[]) { + uint64_t nextents = 0, nchunks = 0; int c, ret; - while ((c = getopt(argc, argv, "b:t:d:o:c:p:")) != -1) { + debug_init(); + + while ((c = getopt(argc, argv, "b:t:d:o:c:p:B:")) != -1) { switch (c) { case 'b': blocksize = atoi(optarg); @@ -499,12 +551,18 @@ int main(int argc, char *argv[]) case 'p': print_progress = atoi(optarg); break; + case 'B': + use_bloom = atoi(optarg); + break; case '?': default: return usage(argv); } } + if (collision_check || dump_output) + use_bloom = 0; + if (!num_threads) num_threads = cpus_online(); @@ -516,11 +574,18 @@ int main(int argc, char *argv[]) rb_root = RB_ROOT; rb_lock = fio_mutex_init(FIO_MUTEX_UNLOCKED); - ret = dedupe_check(argv[optind]); + ret = dedupe_check(argv[optind], &nextents, &nchunks); + + if (!ret) { + if (!bloom) + iter_rb_tree(&nextents, &nchunks); - iter_rb_tree(); + show_stat(nextents, nchunks); + } fio_mutex_remove(rb_lock); + if (bloom) + bloom_free(bloom); scleanup(); return ret; }