client: support --status-interval option in client/server mode
[fio.git] / init.c
diff --git a/init.c b/init.c
index ab7e399752dbad7e763a0e727ad4ec4daa1ba7ef..06f69719328f575c12c0a2886ef64e9b256d4236 100644 (file)
--- a/init.c
+++ b/init.c
@@ -258,7 +258,7 @@ static struct option l_opts[FIO_NR_OPTIONS] = {
        {
                .name           = (char *) "status-interval",
                .has_arg        = required_argument,
-               .val            = 'L',
+               .val            = 'L' | FIO_CLIENT_FLAG,
        },
        {
                .name           = (char *) "trigger-file",
@@ -531,7 +531,7 @@ static void put_job(struct thread_data *td)
 
 static int __setup_rate(struct thread_data *td, enum fio_ddir ddir)
 {
-       unsigned int bs = td->o.min_bs[ddir];
+       unsigned long long bs = td->o.min_bs[ddir];
 
        assert(ddir_rw(ddir));
 
@@ -574,23 +574,6 @@ static int fixed_block_size(struct thread_options *o)
                o->min_bs[DDIR_READ] == o->min_bs[DDIR_TRIM];
 }
 
-
-static unsigned long long get_rand_start_delay(struct thread_data *td)
-{
-       unsigned long long delayrange;
-       uint64_t frand_max;
-       unsigned long r;
-
-       delayrange = td->o.start_delay_high - td->o.start_delay;
-
-       frand_max = rand_max(&td->delay_state);
-       r = __rand(&td->delay_state);
-       delayrange = (unsigned long long) ((double) delayrange * (r / (frand_max + 1.0)));
-
-       delayrange += td->o.start_delay;
-       return delayrange;
-}
-
 /*
  * <3 Johannes
  */
@@ -611,13 +594,19 @@ static int fixup_options(struct thread_data *td)
        struct thread_options *o = &td->o;
        int ret = 0;
 
+       if (read_only && (td_write(td) || td_trim(td))) {
+               log_err("fio: trim and write operations are not allowed"
+                        " with the --readonly parameter.\n");
+               ret |= 1;
+       }
+
 #ifndef CONFIG_PSHARED
        if (!o->use_thread) {
                log_info("fio: this platform does not support process shared"
                         " mutexes, forcing use of threads. Use the 'thread'"
                         " option to get rid of this warning.\n");
                o->use_thread = 1;
-               ret = warnings_fatal;
+               ret |= warnings_fatal;
        }
 #endif
 
@@ -625,7 +614,7 @@ static int fixup_options(struct thread_data *td)
                log_err("fio: read iolog overrides write_iolog\n");
                free(o->write_iolog_file);
                o->write_iolog_file = NULL;
-               ret = warnings_fatal;
+               ret |= warnings_fatal;
        }
 
        /*
@@ -679,20 +668,25 @@ static int fixup_options(struct thread_data *td)
            !o->norandommap) {
                log_err("fio: Any use of blockalign= turns off randommap\n");
                o->norandommap = 1;
-               ret = warnings_fatal;
+               ret |= warnings_fatal;
        }
 
        if (!o->file_size_high)
                o->file_size_high = o->file_size_low;
 
-       if (o->start_delay_high)
-               o->start_delay = get_rand_start_delay(td);
+       if (o->start_delay_high) {
+               if (!o->start_delay_orig)
+                       o->start_delay_orig = o->start_delay;
+               o->start_delay = rand_between(&td->delay_state,
+                                               o->start_delay_orig,
+                                               o->start_delay_high);
+       }
 
        if (o->norandommap && o->verify != VERIFY_NONE
            && !fixed_block_size(o))  {
                log_err("fio: norandommap given for variable block sizes, "
                        "verify limited\n");
-               ret = warnings_fatal;
+               ret |= warnings_fatal;
        }
        if (o->bs_unaligned && (o->odirect || td_ioengine_flagged(td, FIO_RAWIO)))
                log_err("fio: bs_unaligned may not work with raw io\n");
@@ -736,7 +730,7 @@ static int fixup_options(struct thread_data *td)
                log_err("fio: checking for in-flight overlaps when the "
                        "io_submit_mode is offload is not supported\n");
                o->serialize_overlap = 0;
-               ret = warnings_fatal;
+               ret |= warnings_fatal;
        }
 
        if (o->nr_files > td->files_index)
@@ -750,7 +744,7 @@ static int fixup_options(struct thread_data *td)
            ((o->ratemin[DDIR_READ] + o->ratemin[DDIR_WRITE] + o->ratemin[DDIR_TRIM]) &&
            (o->rate_iops_min[DDIR_READ] + o->rate_iops_min[DDIR_WRITE] + o->rate_iops_min[DDIR_TRIM]))) {
                log_err("fio: rate and rate_iops are mutually exclusive\n");
-               ret = 1;
+               ret |= 1;
        }
        if ((o->rate[DDIR_READ] && (o->rate[DDIR_READ] < o->ratemin[DDIR_READ])) ||
            (o->rate[DDIR_WRITE] && (o->rate[DDIR_WRITE] < o->ratemin[DDIR_WRITE])) ||
@@ -759,13 +753,13 @@ static int fixup_options(struct thread_data *td)
            (o->rate_iops[DDIR_WRITE] && (o->rate_iops[DDIR_WRITE] < o->rate_iops_min[DDIR_WRITE])) ||
            (o->rate_iops[DDIR_TRIM] && (o->rate_iops[DDIR_TRIM] < o->rate_iops_min[DDIR_TRIM]))) {
                log_err("fio: minimum rate exceeds rate\n");
-               ret = 1;
+               ret |= 1;
        }
 
        if (!o->timeout && o->time_based) {
                log_err("fio: time_based requires a runtime/timeout setting\n");
                o->time_based = 0;
-               ret = warnings_fatal;
+               ret |= warnings_fatal;
        }
 
        if (o->fill_device && !o->size)
@@ -781,7 +775,7 @@ static int fixup_options(struct thread_data *td)
                        log_info("fio: multiple writers may overwrite blocks "
                                "that belong to other jobs. This can cause "
                                "verification failures.\n");
-                       ret = warnings_fatal;
+                       ret |= warnings_fatal;
                }
 
                /*
@@ -793,7 +787,7 @@ static int fixup_options(struct thread_data *td)
                        log_info("fio: verification read phase will never "
                                 "start because write phase uses all of "
                                 "runtime\n");
-                       ret = warnings_fatal;
+                       ret |= warnings_fatal;
                }
 
                if (!fio_option_is_set(o, refill_buffers))
@@ -829,15 +823,15 @@ static int fixup_options(struct thread_data *td)
                if (td_ioengine_flagged(td, FIO_PIPEIO)) {
                        log_info("fio: cannot pre-read files with an IO engine"
                                 " that isn't seekable. Pre-read disabled.\n");
-                       ret = warnings_fatal;
+                       ret |= warnings_fatal;
                }
        }
 
-       if (!o->unit_base) {
+       if (o->unit_base == N2S_NONE) {
                if (td_ioengine_flagged(td, FIO_BIT_BASED))
-                       o->unit_base = 1;
+                       o->unit_base = N2S_BITPERSEC;
                else
-                       o->unit_base = 8;
+                       o->unit_base = N2S_BYTEPERSEC;
        }
 
 #ifndef FIO_HAVE_ANY_FALLOCATE
@@ -853,7 +847,7 @@ static int fixup_options(struct thread_data *td)
                         " this warning\n");
                o->fsync_blocks = o->fdatasync_blocks;
                o->fdatasync_blocks = 0;
-               ret = warnings_fatal;
+               ret |= warnings_fatal;
        }
 #endif
 
@@ -866,7 +860,7 @@ static int fixup_options(struct thread_data *td)
                log_err("fio: Windows does not support direct or non-buffered io with"
                                " the synchronous ioengines. Use the 'windowsaio' ioengine"
                                " with 'direct=1' and 'iodepth=1' instead.\n");
-               ret = 1;
+               ret |= 1;
        }
 #endif
 
@@ -897,9 +891,9 @@ static int fixup_options(struct thread_data *td)
         * If size is set but less than the min block size, complain
         */
        if (o->size && o->size < td_min_bs(td)) {
-               log_err("fio: size too small, must not be less than minimum block size: %llu < %u\n",
+               log_err("fio: size too small, must not be less than minimum block size: %llu < %llu\n",
                        (unsigned long long) o->size, td_min_bs(td));
-               ret = 1;
+               ret |= 1;
        }
 
        /*
@@ -916,7 +910,7 @@ static int fixup_options(struct thread_data *td)
 
        if (td_ioengine_flagged(td, FIO_NOEXTEND) && o->file_append) {
                log_err("fio: can't append/extent with IO engine %s\n", td->io_ops->name);
-               ret = 1;
+               ret |= 1;
        }
 
        if (fio_option_is_set(o, gtod_cpu)) {
@@ -933,7 +927,7 @@ static int fixup_options(struct thread_data *td)
                log_err("fio: block error histogram only available "
                        "with a single file per job, but %d files "
                        "provided\n", o->nr_files);
-               ret = 1;
+               ret |= 1;
        }
 
        if (fio_option_is_set(o, clat_percentiles) &&
@@ -947,7 +941,7 @@ static int fixup_options(struct thread_data *td)
                   o->lat_percentiles && o->clat_percentiles) {
                log_err("fio: lat_percentiles and clat_percentiles are "
                        "mutually exclusive\n");
-               ret = 1;
+               ret |= 1;
        }
 
        if (o->disable_lat)
@@ -997,23 +991,26 @@ void td_fill_verify_state_seed(struct thread_data *td)
 
 static void td_fill_rand_seeds_internal(struct thread_data *td, bool use64)
 {
+       unsigned int read_seed = td->rand_seeds[FIO_RAND_BS_OFF];
+       unsigned int write_seed = td->rand_seeds[FIO_RAND_BS1_OFF];
+       unsigned int trim_seed = td->rand_seeds[FIO_RAND_BS2_OFF];
        int i;
 
        /*
         * trimwrite is special in that we need to generate the same
         * offsets to get the "write after trim" effect. If we are
         * using bssplit to set buffer length distributions, ensure that
-        * we seed the trim and write generators identically.
+        * we seed the trim and write generators identically. Ditto for
+        * verify, read and writes must have the same seed, if we are doing
+        * read verify.
         */
-       if (td_trimwrite(td)) {
-               init_rand_seed(&td->bsrange_state[DDIR_READ], td->rand_seeds[FIO_RAND_BS_OFF], use64);
-               init_rand_seed(&td->bsrange_state[DDIR_WRITE], td->rand_seeds[FIO_RAND_BS1_OFF], use64);
-               init_rand_seed(&td->bsrange_state[DDIR_TRIM], td->rand_seeds[FIO_RAND_BS1_OFF], use64);
-       } else {
-               init_rand_seed(&td->bsrange_state[DDIR_READ], td->rand_seeds[FIO_RAND_BS_OFF], use64);
-               init_rand_seed(&td->bsrange_state[DDIR_WRITE], td->rand_seeds[FIO_RAND_BS1_OFF], use64);
-               init_rand_seed(&td->bsrange_state[DDIR_TRIM], td->rand_seeds[FIO_RAND_BS2_OFF], use64);
-       }
+       if (td->o.verify != VERIFY_NONE)
+               write_seed = read_seed;
+       if (td_trimwrite(td))
+               trim_seed = write_seed;
+       init_rand_seed(&td->bsrange_state[DDIR_READ], read_seed, use64);
+       init_rand_seed(&td->bsrange_state[DDIR_WRITE], write_seed, use64);
+       init_rand_seed(&td->bsrange_state[DDIR_TRIM], trim_seed, use64);
 
        td_fill_verify_state_seed(td);
        init_rand_seed(&td->rwmix_state, td->rand_seeds[FIO_RAND_MIX_OFF], false);
@@ -1184,7 +1181,7 @@ static void init_flags(struct thread_data *td)
            fio_option_is_set(o, zero_buffers)))
                td->flags |= TD_F_SCRAMBLE_BUFFERS;
        if (o->verify != VERIFY_NONE)
-               td->flags |= TD_F_VER_NONE;
+               td->flags |= TD_F_DO_VERIFY;
 
        if (o->verify_async || o->io_submit_mode == IO_MODE_OFFLOAD)
                td->flags |= TD_F_NEED_LOCK;
@@ -1453,6 +1450,11 @@ static int add_job(struct thread_data *td, const char *jobname, int job_add_num,
                }
        }
 
+       if (setup_random_seeds(td)) {
+               td_verror(td, errno, "setup_random_seeds");
+               goto err;
+       }
+
        if (fixup_options(td))
                goto err;
 
@@ -1508,11 +1510,6 @@ static int add_job(struct thread_data *td, const char *jobname, int job_add_num,
        td->groupid = groupid;
        prev_group_jobs++;
 
-       if (setup_random_seeds(td)) {
-               td_verror(td, errno, "setup_random_seeds");
-               goto err;
-       }
-
        if (setup_rate(td))
                goto err;
 
@@ -1970,7 +1967,8 @@ static int __parse_jobs_ini(struct thread_data *td,
                        if (p[0] == '[') {
                                if (nested) {
                                        log_err("No new sections in included files\n");
-                                       return 1;
+                                       ret = 1;
+                                       goto out;
                                }
 
                                skip_fgets = 1;
@@ -2160,7 +2158,7 @@ static void usage(const char *name)
        printf("  --showcmd\t\tTurn a job file into command line options\n");
        printf("  --eta=when\t\tWhen ETA estimate should be printed\n");
        printf("            \t\tMay be \"always\", \"never\" or \"auto\"\n");
-       printf("  --eta-newline=time\tForce a new line for every 'time'");
+       printf("  --eta-newline=t\tForce a new line for every 't'");
        printf(" period passed\n");
        printf("  --status-interval=t\tForce full status dump every");
        printf(" 't' period passed\n");