From 4739e7332cd70c18cdd783f37246d3cdcabff98b Mon Sep 17 00:00:00 2001 From: Stefan Behrens Date: Wed, 16 May 2012 18:51:28 +0200 Subject: Btrfs-progs: make scrub IO priority configurable The btrfs tool is changed in order to support command line parameters to configure the IO priority of the scrub tasks. Also the default is changed. The default IO priority for scrub is the idle class now. The behavior is the same as when one would type 'ionice ... btrfs scrub start ...' or 'ionice ... btrfs scrub resume ...' (without this patch applied). The only reason for adding this to the btrfs tool is that it was not documented and not obvious that it worked like this, that all internal scrub tasks inherited the IO priority values of the btrfs tool that is starting or resuming the scrub operation. Note that after applying the patch it is no longer possible to set the IO priority using ionice since the btrfs tool always configures the priority in order to run in the idle class by default. Some basic performance measurements have been done with the goal to measure which IO priority for scrub gives the best overall disk data throughput. The kernel was configured to use the CFQ IO scheduler with default configuration and without support for throttling. The summary is, that the more the disk head movements are avoided, the faster the overall disk transfer capacity is, which is not really a big surprise. Therefore it makes sense that the best data throughput was measured setting the scrub IO priority and the scrub readahead IO priority to the idle class priority. Running with idle class IO priority means that scrub and scrub readahead IO is paused while other tasks access the disk. Doing the tasks one after the other instead of concurrently avoids many disk head movements. The overall data throughput of rotating disks is improved this way. However, if it is desired to have the scrub task done within a reasonable time, and if at the same time the filesystem is heavily loaded, the idle IO priority should be avoided. Otherwise the scrub operation will never take place and thus never terminate. The best effort IO priority class with the subclass 7 (the lowest one in the best effort class) is recommended in the case of always heavily loaded hard disks. If the filesystem is not loaded all the time and leaves some idle slots for scrub, the idle class IO priority is recommended. The idle class now is the default if the scrub operation is started with the btrfs-progs tools. Note that the patch that sets the scrub readahead IO priority to the idle class is a seperate patch, this needs to be done in the kernel. Signed-off-by: Stefan Behrens --- cmds-scrub.c | 40 +++++++++++++++++++++++++++++++++++++--- 1 file changed, 37 insertions(+), 3 deletions(-) (limited to 'cmds-scrub.c') diff --git a/cmds-scrub.c b/cmds-scrub.c index f06eb20d..59223615 100644 --- a/cmds-scrub.c +++ b/cmds-scrub.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -60,6 +61,15 @@ struct scrub_stats { u64 canceled; }; +/* TBD: replace with #include "linux/ioprio.h" in some years */ +#if !defined (IOPRIO_H) +#define IOPRIO_WHO_PROCESS 1 +#define IOPRIO_CLASS_SHIFT 13 +#define IOPRIO_PRIO_VALUE(class, data) \ + (((class) << IOPRIO_CLASS_SHIFT) | (data)) +#define IOPRIO_CLASS_IDLE 3 +#endif + struct scrub_progress { struct btrfs_ioctl_scrub_args scrub_args; int fd; @@ -69,6 +79,8 @@ struct scrub_progress { struct scrub_file_record *resumed; int ioctl_errno; pthread_mutex_t progress_mutex; + int ioprio_class; + int ioprio_classdata; }; struct scrub_file_record { @@ -813,6 +825,14 @@ static void *scrub_one_dev(void *ctx) sp->stats.duration = 0; sp->stats.finished = 0; + ret = syscall(SYS_ioprio_set, IOPRIO_WHO_PROCESS, 0, + IOPRIO_PRIO_VALUE(sp->ioprio_class, + sp->ioprio_classdata)); + if (ret) + fprintf(stderr, + "WARNING: setting ioprio failed: %s (ignored).\n", + strerror(errno)); + ret = ioctl(sp->fd, BTRFS_IOC_SCRUB, &sp->scrub_args); gettimeofday(&tv, NULL); sp->ret = ret; @@ -1029,6 +1049,8 @@ static int scrub_start(int argc, char **argv, int resume) int do_record = 1; int readonly = 0; int do_stats_per_dev = 0; + int ioprio_class = IOPRIO_CLASS_IDLE; + int ioprio_classdata = 0; int n_start = 0; int n_skip = 0; int n_resume = 0; @@ -1054,7 +1076,7 @@ static int scrub_start(int argc, char **argv, int resume) u64 devid; optind = 1; - while ((c = getopt(argc, argv, "BdqrR")) != -1) { + while ((c = getopt(argc, argv, "BdqrRc:n:")) != -1) { switch (c) { case 'B': do_background = 0; @@ -1073,6 +1095,12 @@ static int scrub_start(int argc, char **argv, int resume) case 'R': print_raw = 1; break; + case 'c': + ioprio_class = (int)strtol(optarg, NULL, 10); + break; + case 'n': + ioprio_classdata = (int)strtol(optarg, NULL, 10); + break; case '?': default: usage(resume ? cmd_scrub_resume_usage : @@ -1182,6 +1210,8 @@ static int scrub_start(int argc, char **argv, int resume) sp[i].skip = 0; sp[i].scrub_args.end = (u64)-1ll; sp[i].scrub_args.flags = readonly ? BTRFS_SCRUB_READONLY : 0; + sp[i].ioprio_class = ioprio_class; + sp[i].ioprio_classdata = ioprio_classdata; } if (!n_start && !n_resume) { @@ -1435,13 +1465,15 @@ out: } static const char * const cmd_scrub_start_usage[] = { - "btrfs scrub start [-Bdqr] |", + "btrfs scrub start Bdqr] [-c ioprio_class -n ioprio_classdata] |\n", "Start a new scrub", "", "-B do not background", "-d stats per device (-B only)", "-q be quiet", "-r read only mode", + "-c set ioprio class (see ionice(1) manpage)", + "-n set ioprio classdata (see ionice(1) manpage)", NULL }; @@ -1494,13 +1526,15 @@ out: } static const char * const cmd_scrub_resume_usage[] = { - "btrfs scrub resume [-Bdqr] |", + "btrfs scrub resume [-Bdqr] [-c ioprio_class -n ioprio_classdata] |\n", "Resume previously canceled or interrupted scrub", "", "-B do not background", "-d stats per device (-B only)", "-q be quiet", "-r read only mode", + "-c set ioprio class (see ionice(1) manpage)", + "-n set ioprio classdata (see ionice(1) manpage)", NULL }; -- cgit v1.2.3