#include "options.h"
#include "profile.h"
#include "time.h"
+#include "gettime.h"
#include "lib/getopt.h"
#include "lib/rand.h"
#include "server.h"
+#include "stat.h"
+#include "flow.h"
#ifdef FIO_HAVE_GUASI
#include <guasi.h>
#include <sys/asynch.h>
#endif
-struct group_run_stats {
- unsigned long long max_run[2], min_run[2];
- unsigned long long max_bw[2], min_bw[2];
- unsigned long long io_kb[2];
- unsigned long long agg[2];
- unsigned int kb_base;
-};
+#ifdef FIO_HAVE_LIBNUMA
+#include <linux/mempolicy.h>
+#include <numa.h>
+
+/*
+ * "local" is pseudo-policy
+ */
+#define MPOL_LOCAL MPOL_MAX
+#endif
/*
* What type of allocation to use for io buffers
};
/*
- * How many depth levels to log
+ * What type of errors to continue on when continue_on_error is used
*/
-#define FIO_IO_U_MAP_NR 7
-#define FIO_IO_U_LAT_U_NR 10
-#define FIO_IO_U_LAT_M_NR 12
-
-/*
- * Aggregate clat samples to report percentile(s) of them.
- *
- * EXECUTIVE SUMMARY
- *
- * FIO_IO_U_PLAT_BITS determines the maximum statistical error on the
- * value of resulting percentiles. The error will be approximately
- * 1/2^(FIO_IO_U_PLAT_BITS+1) of the value.
- *
- * FIO_IO_U_PLAT_GROUP_NR and FIO_IO_U_PLAT_BITS determine the maximum
- * range being tracked for latency samples. The maximum value tracked
- * accurately will be 2^(GROUP_NR + PLAT_BITS -1) microseconds.
- *
- * FIO_IO_U_PLAT_GROUP_NR and FIO_IO_U_PLAT_BITS determine the memory
- * requirement of storing those aggregate counts. The memory used will
- * be (FIO_IO_U_PLAT_GROUP_NR * 2^FIO_IO_U_PLAT_BITS) * sizeof(int)
- * bytes.
- *
- * FIO_IO_U_PLAT_NR is the total number of buckets.
- *
- * DETAILS
- *
- * Suppose the clat varies from 0 to 999 (usec), the straightforward
- * method is to keep an array of (999 + 1) buckets, in which a counter
- * keeps the count of samples which fall in the bucket, e.g.,
- * {[0],[1],...,[999]}. However this consumes a huge amount of space,
- * and can be avoided if an approximation is acceptable.
- *
- * One such method is to let the range of the bucket to be greater
- * than one. This method has low accuracy when the value is small. For
- * example, let the buckets be {[0,99],[100,199],...,[900,999]}, and
- * the represented value of each bucket be the mean of the range. Then
- * a value 0 has an round-off error of 49.5. To improve on this, we
- * use buckets with non-uniform ranges, while bounding the error of
- * each bucket within a ratio of the sample value. A simple example
- * would be when error_bound = 0.005, buckets are {
- * {[0],[1],...,[99]}, {[100,101],[102,103],...,[198,199]},..,
- * {[900,909],[910,919]...} }. The total range is partitioned into
- * groups with different ranges, then buckets with uniform ranges. An
- * upper bound of the error is (range_of_bucket/2)/value_of_bucket
- *
- * For better efficiency, we implement this using base two. We group
- * samples by their Most Significant Bit (MSB), extract the next M bit
- * of them as an index within the group, and discard the rest of the
- * bits.
- *
- * E.g., assume a sample 'x' whose MSB is bit n (starting from bit 0),
- * and use M bit for indexing
- *
- * | n | M bits | bit (n-M-1) ... bit 0 |
- *
- * Because x is at least 2^n, and bit 0 to bit (n-M-1) is at most
- * (2^(n-M) - 1), discarding bit 0 to (n-M-1) makes the round-off
- * error
- *
- * 2^(n-M)-1 2^(n-M) 1
- * e <= --------- <= ------- = ---
- * 2^n 2^n 2^M
- *
- * Furthermore, we use "mean" of the range to represent the bucket,
- * the error e can be lowered by half to 1 / 2^(M+1). By using M bits
- * as the index, each group must contains 2^M buckets.
- *
- * E.g. Let M (FIO_IO_U_PLAT_BITS) be 6
- * Error bound is 1/2^(6+1) = 0.0078125 (< 1%)
- *
- * Group MSB #discarded range of #buckets
- * error_bits value
- * ----------------------------------------------------------------
- * 0* 0~5 0 [0,63] 64
- * 1* 6 0 [64,127] 64
- * 2 7 1 [128,255] 64
- * 3 8 2 [256,511] 64
- * 4 9 3 [512,1023] 64
- * ... ... ... [...,...] ...
- * 18 23 17 [8838608,+inf]** 64
- *
- * * Special cases: when n < (M-1) or when n == (M-1), in both cases,
- * the value cannot be rounded off. Use all bits of the sample as
- * index.
- *
- * ** If a sample's MSB is greater than 23, it will be counted as 23.
- */
-
-#define FIO_IO_U_PLAT_BITS 6
-#define FIO_IO_U_PLAT_VAL (1 << FIO_IO_U_PLAT_BITS)
-#define FIO_IO_U_PLAT_GROUP_NR 19
-#define FIO_IO_U_PLAT_NR (FIO_IO_U_PLAT_GROUP_NR * FIO_IO_U_PLAT_VAL)
-#define FIO_IO_U_LIST_MAX_LEN 20 /* The size of the default and user-specified
- list of percentiles */
-
-#define MAX_PATTERN_SIZE 512
-
-struct thread_stat {
- char *name;
- char *verror;
- int error;
- int groupid;
- pid_t pid;
- char *description;
- int members;
-
- struct io_log *slat_log;
- struct io_log *clat_log;
- struct io_log *lat_log;
- struct io_log *bw_log;
-
- /*
- * bandwidth and latency stats
- */
- struct io_stat clat_stat[2]; /* completion latency */
- struct io_stat slat_stat[2]; /* submission latency */
- struct io_stat lat_stat[2]; /* total latency */
- struct io_stat bw_stat[2]; /* bandwidth stats */
-
- unsigned long long stat_io_bytes[2];
- struct timeval stat_sample_time[2];
-
- /*
- * fio system usage accounting
- */
- struct rusage ru_start;
- struct rusage ru_end;
- unsigned long usr_time;
- unsigned long sys_time;
- unsigned long ctx;
- unsigned long minf, majf;
-
- /*
- * IO depth and latency stats
- */
- unsigned int clat_percentiles;
- double *percentile_list;
-
- unsigned int io_u_map[FIO_IO_U_MAP_NR];
- unsigned int io_u_submit[FIO_IO_U_MAP_NR];
- unsigned int io_u_complete[FIO_IO_U_MAP_NR];
- unsigned int io_u_lat_u[FIO_IO_U_LAT_U_NR];
- unsigned int io_u_lat_m[FIO_IO_U_LAT_M_NR];
- unsigned int io_u_plat[2][FIO_IO_U_PLAT_NR];
- unsigned long total_io_u[3];
- unsigned long short_io_u[3];
- unsigned long total_submit;
- unsigned long total_complete;
-
- unsigned long long io_bytes[2];
- unsigned long long runtime[2];
- unsigned long total_run_time;
-
- /*
- * IO Error related stats
- */
- unsigned continue_on_error;
- unsigned long total_err_count;
- int first_error;
+enum error_type_bit {
+ ERROR_TYPE_READ_BIT = 0,
+ ERROR_TYPE_WRITE_BIT = 1,
+ ERROR_TYPE_VERIFY_BIT = 2,
+ ERROR_TYPE_CNT = 3,
+};
- unsigned int kb_base;
+enum error_type {
+ ERROR_TYPE_NONE = 0,
+ ERROR_TYPE_READ = 1 << ERROR_TYPE_READ_BIT,
+ ERROR_TYPE_WRITE = 1 << ERROR_TYPE_WRITE_BIT,
+ ERROR_TYPE_VERIFY = 1 << ERROR_TYPE_VERIFY_BIT,
+ ERROR_TYPE_ANY = 0xffff,
};
struct bssplit {
unsigned long long file_size_high;
unsigned long long start_offset;
- unsigned int bs[2];
- unsigned int ba[2];
- unsigned int min_bs[2];
- unsigned int max_bs[2];
- struct bssplit *bssplit[2];
- unsigned int bssplit_nr[2];
+ unsigned int bs[DDIR_RWDIR_CNT];
+ unsigned int ba[DDIR_RWDIR_CNT];
+ unsigned int min_bs[DDIR_RWDIR_CNT];
+ unsigned int max_bs[DDIR_RWDIR_CNT];
+ struct bssplit *bssplit[DDIR_RWDIR_CNT];
+ unsigned int bssplit_nr[DDIR_RWDIR_CNT];
+
+ int *ignore_error[ERROR_TYPE_CNT];
+ unsigned int ignore_error_nr[ERROR_TYPE_CNT];
+ unsigned int error_dump;
unsigned int nr_files;
unsigned int open_files;
unsigned int create_serialize;
unsigned int create_fsync;
unsigned int create_on_open;
+ unsigned int create_only;
unsigned int end_fsync;
unsigned int pre_read;
unsigned int sync_io;
unsigned int use_os_rand;
unsigned int write_lat_log;
unsigned int write_bw_log;
+ unsigned int write_iops_log;
+ unsigned int log_avg_msec;
unsigned int norandommap;
unsigned int softrandommap;
unsigned int bs_unaligned;
unsigned int fsync_on_close;
+ unsigned int random_distribution;
+ double zipf_theta;
+ double pareto_h;
+
unsigned int hugepage_size;
unsigned int rw_min_bs;
unsigned int thinktime;
unsigned long long ramp_time;
unsigned int overwrite;
unsigned int bw_avg_time;
+ unsigned int iops_avg_time;
unsigned int loops;
+ unsigned long long zone_range;
unsigned long long zone_size;
unsigned long long zone_skip;
enum fio_memtype mem_type;
unsigned int mem_align;
+ unsigned int max_latency;
+
unsigned int stonewall;
unsigned int new_group;
unsigned int numjobs;
unsigned int cpumask_set;
os_cpu_mask_t verify_cpumask;
unsigned int verify_cpumask_set;
+#ifdef FIO_HAVE_LIBNUMA
+ struct bitmask *numa_cpunodesmask;
+ unsigned int numa_cpumask_set;
+ unsigned short numa_mem_mode;
+ unsigned int numa_mem_prefer_node;
+ struct bitmask *numa_memnodesmask;
+ unsigned int numa_memmask_set;
+#endif
unsigned int iolog;
unsigned int rwmixcycle;
unsigned int rwmix[2];
unsigned int zero_buffers;
unsigned int refill_buffers;
unsigned int scramble_buffers;
+ unsigned int compress_percentage;
+ unsigned int compress_chunk;
unsigned int time_based;
unsigned int disable_lat;
unsigned int disable_clat;
unsigned long long trim_backlog;
unsigned int clat_percentiles;
unsigned int overwrite_plist;
- double percentile_list[FIO_IO_U_LIST_MAX_LEN];
+ fio_fp64_t percentile_list[FIO_IO_U_LIST_MAX_LEN];
char *read_iolog_file;
char *write_iolog_file;
char *bw_log_file;
char *lat_log_file;
+ char *iops_log_file;
char *replay_redirect;
/*
char *exec_prerun;
char *exec_postrun;
- unsigned int rate[2];
- unsigned int ratemin[2];
+ unsigned int rate[DDIR_RWDIR_CNT];
+ unsigned int ratemin[DDIR_RWDIR_CNT];
unsigned int ratecycle;
- unsigned int rate_iops[2];
- unsigned int rate_iops_min[2];
+ unsigned int rate_iops[DDIR_RWDIR_CNT];
+ unsigned int rate_iops_min[DDIR_RWDIR_CNT];
char *ioscheduler;
/*
* I/O Error handling
*/
- unsigned int continue_on_error;
+ enum error_type continue_on_error;
/*
* Benchmark profile type
unsigned int uid;
unsigned int gid;
- unsigned int sync_file_range;
+ int flow_id;
+ int flow;
+ int flow_watermark;
+ unsigned int flow_sleep;
- unsigned int userspace_libaio_reap;
-};
+ unsigned long long offset_increment;
-#define FIO_VERROR_SIZE 128
+ unsigned int sync_file_range;
+};
/*
* This describes a single thread/process executing a fio job.
*/
struct thread_data {
struct thread_options o;
+ void *eo;
char verror[FIO_VERROR_SIZE];
pthread_t thread;
int thread_number;
int groupid;
struct thread_stat ts;
+
+ struct io_log *slat_log;
+ struct io_log *clat_log;
+ struct io_log *lat_log;
+ struct io_log *bw_log;
+ struct io_log *iops_log;
+
+ uint64_t stat_io_bytes[DDIR_RWDIR_CNT];
+ struct timeval bw_sample_time;
+
+ uint64_t stat_io_blocks[DDIR_RWDIR_CNT];
+ struct timeval iops_sample_time;
+
+ struct rusage ru_start;
+ struct rusage ru_end;
+
struct fio_file **files;
unsigned int files_size;
unsigned int files_index;
struct frand_state __next_file_state;
};
int error;
+ int sig;
int done;
pid_t pid;
char *orig_buffer;
struct ioengine_ops *io_ops;
/*
- * Current IO depth and list of free and busy io_u's.
+ * Queue depth of io_u's that fio MIGHT do
*/
unsigned int cur_depth;
+
+ /*
+ * io_u's about to be committed
+ */
unsigned int io_u_queued;
+
+ /*
+ * io_u's submitted but not completed yet
+ */
+ unsigned int io_u_in_flight;
+
+ /*
+ * List of free and busy io_u's
+ */
struct flist_head io_u_freelist;
struct flist_head io_u_busylist;
struct flist_head io_u_requeues;
/*
* Rate state
*/
- unsigned long rate_nsec_cycle[2];
- long rate_pending_usleep[2];
- unsigned long rate_bytes[2];
- unsigned long rate_blocks[2];
- struct timeval lastrate[2];
+ unsigned long long rate_bps[DDIR_RWDIR_CNT];
+ long rate_pending_usleep[DDIR_RWDIR_CNT];
+ unsigned long rate_bytes[DDIR_RWDIR_CNT];
+ unsigned long rate_blocks[DDIR_RWDIR_CNT];
+ struct timeval lastrate[DDIR_RWDIR_CNT];
unsigned long long total_io_size;
unsigned long long fill_device_size;
- unsigned long io_issues[2];
- unsigned long long io_blocks[2];
- unsigned long long io_bytes[2];
+ unsigned long io_issues[DDIR_RWDIR_CNT];
+ unsigned long long io_blocks[DDIR_RWDIR_CNT];
+ unsigned long long this_io_blocks[DDIR_RWDIR_CNT];
+ unsigned long long io_bytes[DDIR_RWDIR_CNT];
unsigned long long io_skip_bytes;
- unsigned long long this_io_bytes[2];
+ unsigned long long this_io_bytes[DDIR_RWDIR_CNT];
unsigned long long zone_bytes;
struct fio_mutex *mutex;
unsigned int total_err_count;
int first_error;
+ struct fio_flow *flow;
+
/*
* Can be overloaded by profiles
*/
#define td_vmsg(td, err, msg, func) \
__td_verror((td), (err), (msg), (func))
+#define __fio_stringify_1(x) #x
+#define __fio_stringify(x) __fio_stringify_1(x)
+
extern int exitall_on_terminate;
-extern int thread_number;
-extern int nr_process, nr_thread;
+extern unsigned int thread_number;
+extern unsigned int nr_process, nr_thread;
extern int shm_id;
extern int groupid;
-extern int terse_output;
+extern int output_format;
extern int temp_stall_ts;
extern unsigned long long mlock_size;
-extern unsigned long page_mask, page_size;
+extern uintptr_t page_mask, page_size;
extern int read_only;
extern int eta_print;
extern unsigned long done_secs;
extern int terse_version;
extern int is_backend;
extern int nr_clients;
+extern int log_syslog;
+extern const char fio_version_string[];
+extern const fio_fp64_t def_percentile_list[FIO_IO_U_LIST_MAX_LEN];
extern struct thread_data *threads;
#define REAL_MAX_JOBS 2048
-#define td_non_fatal_error(e) ((e) == EIO || (e) == EILSEQ)
+static inline enum error_type td_error_type(enum fio_ddir ddir, int err)
+{
+ if (err == EILSEQ)
+ return ERROR_TYPE_VERIFY_BIT;
+ if (ddir == DDIR_READ)
+ return ERROR_TYPE_READ_BIT;
+ return ERROR_TYPE_WRITE_BIT;
+}
+
+static int __NON_FATAL_ERR[] = {EIO, EILSEQ};
+static inline int td_non_fatal_error(struct thread_data *td,
+ enum error_type_bit etype, int err)
+{
+ int i;
+ if (!td->o.ignore_error[etype]) {
+ td->o.ignore_error[etype] = __NON_FATAL_ERR;
+ td->o.ignore_error_nr[etype] = sizeof(__NON_FATAL_ERR)
+ / sizeof(int);
+ }
+
+ if (!(td->o.continue_on_error & (1 << etype)))
+ return 0;
+ for (i = 0; i < td->o.ignore_error_nr[etype]; i++)
+ if (td->o.ignore_error[etype][i] == err)
+ return 1;
+ return 0;
+}
static inline void update_error_count(struct thread_data *td, int err)
{
*/
extern int __must_check parse_options(int, char **);
extern int parse_jobs_ini(char *, int, int);
-extern int exec_run(void);
+extern int parse_cmd_line(int, char **);
+extern int fio_backend(void);
extern void reset_fio_state(void);
+extern void clear_io_state(struct thread_data *);
extern int fio_options_parse(struct thread_data *, char **, int);
extern void fio_keywords_init(void);
extern int fio_cmd_option_parse(struct thread_data *, const char *, char *);
+extern int fio_cmd_ioengine_option_parse(struct thread_data *, const char *, char *);
extern void fio_fill_default_options(struct thread_data *);
extern int fio_show_option_help(const char *);
+extern void fio_options_set_ioengine_opts(struct option *long_options, struct thread_data *td);
extern void fio_options_dup_and_init(struct option *);
-extern void options_mem_dupe(struct thread_data *);
-extern void options_mem_free(struct thread_data *);
+extern void fio_options_mem_dupe(struct thread_data *);
+extern void options_mem_dupe(void *data, struct fio_option *options);
extern void td_fill_rand_seeds(struct thread_data *);
extern void add_job_opts(const char **);
extern char *num2str(unsigned long, int, int, int);
+extern int ioengine_load(struct thread_data *);
-#define FIO_GETOPT_JOB 0x89988998
+#define FIO_GETOPT_JOB 0x89000000
+#define FIO_GETOPT_IOENGINE 0x98000000
#define FIO_NR_OPTIONS (FIO_MAX_OPTS + 128)
/*
TD_NOT_CREATED = 0,
TD_CREATED,
TD_INITIALIZED,
+ TD_SETTING_UP,
TD_RAMP,
TD_RUNNING,
TD_PRE_READING,
};
extern void td_set_runstate(struct thread_data *, int);
+#define TERMINATE_ALL (-1)
+extern void fio_terminate_threads(int);
/*
* Memory helpers
{
int ret = 0;
- if (bytes_done[0])
- ret |= __should_check_rate(td, 0);
- if (bytes_done[1])
- ret |= __should_check_rate(td, 1);
+ if (bytes_done[DDIR_READ])
+ ret |= __should_check_rate(td, DDIR_READ);
+ if (bytes_done[DDIR_WRITE])
+ ret |= __should_check_rate(td, DDIR_WRITE);
+ if (bytes_done[DDIR_TRIM])
+ ret |= __should_check_rate(td, DDIR_TRIM);
return ret;
}
pthread_cond_signal(&td->free_cond);
}
+extern const char *fio_get_arch_string(int);
+extern const char *fio_get_os_string(int);
+
+enum {
+ FIO_OUTPUT_TERSE = 0,
+ FIO_OUTPUT_JSON,
+ FIO_OUTPUT_NORMAL,
+};
+
+enum {
+ FIO_RAND_DIST_RANDOM = 0,
+ FIO_RAND_DIST_ZIPF,
+ FIO_RAND_DIST_PARETO,
+};
+
#endif