Commit 465bee1d authored by Peter Lieven's avatar Peter Lieven Committed by Kevin Wolf
Browse files

block: optimize zero writes with bdrv_write_zeroes 



this patch tries to optimize zero write requests
by automatically using bdrv_write_zeroes if it is
supported by the format.

This significantly speeds up file system initialization and
should speed zero write test used to test backend storage
performance.

I ran the following 2 tests on my internal SSD with a
50G QCOW2 container and on an attached iSCSI storage.

a) mkfs.ext4 -E lazy_itable_init=0,lazy_journal_init=0 /dev/vdX

QCOW2         [off]     [on]     [unmap]
-----
runtime:       14secs    1.1secs  1.1secs
filesize:      937M      18M      18M

iSCSI         [off]     [on]     [unmap]
----
runtime:       9.3s      0.9s     0.9s

b) dd if=/dev/zero of=/dev/vdX bs=1M oflag=direct

QCOW2         [off]     [on]     [unmap]
-----
runtime:       246secs   18secs   18secs
filesize:      51G       192K     192K
throughput:    203M/s    2.3G/s   2.3G/s

iSCSI*        [off]     [on]     [unmap]
----
runtime:       8mins     45secs   33secs
throughput:    106M/s    1.2G/s   1.6G/s
allocated:     100%      100%     0%

* The storage was connected via an 1Gbit interface.
  It seems to internally handle writing zeroes
  via WRITESAME16 very fast.

Signed-off-by: default avatarPeter Lieven <pl@kamp.de>
Signed-off-by: default avatarKevin Wolf <kwolf@redhat.com>
parent 82a402e9

block.c

+9 −0
Original line number Diff line number Diff line
@@ -3289,6 +3289,15 @@ static int coroutine_fn bdrv_aligned_pwritev(BlockDriverState *bs, 


    ret = notifier_with_return_list_notify(&bs->before_write_notifiers, req);



    if (!ret && bs->detect_zeroes != BLOCKDEV_DETECT_ZEROES_OPTIONS_OFF &&

        !(flags & BDRV_REQ_ZERO_WRITE) && drv->bdrv_co_write_zeroes &&

        qemu_iovec_is_zero(qiov)) {

        flags |= BDRV_REQ_ZERO_WRITE;

        if (bs->detect_zeroes == BLOCKDEV_DETECT_ZEROES_OPTIONS_UNMAP) {

            flags |= BDRV_REQ_MAY_UNMAP;

        }

    }



    if (ret < 0) {

        /* Do nothing, write notifier decided to fail this request */

    } else if (flags & BDRV_REQ_ZERO_WRITE) {

block/qapi.c

+1 −0
Original line number Diff line number Diff line
@@ -50,6 +50,7 @@ BlockDeviceInfo *bdrv_block_device_info(BlockDriverState *bs) 
    }



    info->backing_file_depth = bdrv_get_backing_file_depth(bs);

    info->detect_zeroes = bs->detect_zeroes;



    if (bs->io_limits_enabled) {

        ThrottleConfig cfg;

blockdev.c

+24 −0
Original line number Diff line number Diff line
@@ -343,6 +343,7 @@ static DriveInfo *blockdev_init(const char *file, QDict *bs_opts, 
    QemuOpts *opts;

    const char *id;

    bool has_driver_specific_opts;

    BlockdevDetectZeroesOptions detect_zeroes;

    BlockDriver *drv = NULL;



    /* Check common options by copying from bs_opts to opts, all other options
@@ -471,6 +472,24 @@ static DriveInfo *blockdev_init(const char *file, QDict *bs_opts, 
        }

    }



    detect_zeroes =

        parse_enum_option(BlockdevDetectZeroesOptions_lookup,

                          qemu_opt_get(opts, "detect-zeroes"),

                          BLOCKDEV_DETECT_ZEROES_OPTIONS_MAX,

                          BLOCKDEV_DETECT_ZEROES_OPTIONS_OFF,

                          &error);

    if (error) {

        error_propagate(errp, error);

        goto early_err;

    }



    if (detect_zeroes == BLOCKDEV_DETECT_ZEROES_OPTIONS_UNMAP &&

        !(bdrv_flags & BDRV_O_UNMAP)) {

        error_setg(errp, "setting detect-zeroes to unmap is not allowed "

                         "without setting discard operation to unmap");

        goto early_err;

    }



    /* init */

    dinfo = g_malloc0(sizeof(*dinfo));

    dinfo->id = g_strdup(qemu_opts_id(opts));
@@ -481,6 +500,7 @@ static DriveInfo *blockdev_init(const char *file, QDict *bs_opts, 
    }

    dinfo->bdrv->open_flags = snapshot ? BDRV_O_SNAPSHOT : 0;

    dinfo->bdrv->read_only = ro;

    dinfo->bdrv->detect_zeroes = detect_zeroes;

    dinfo->refcount = 1;

    if (serial != NULL) {

        dinfo->serial = g_strdup(serial);
@@ -2474,6 +2494,10 @@ QemuOptsList qemu_common_drive_opts = { 
            .name = "copy-on-read",

            .type = QEMU_OPT_BOOL,

            .help = "copy read data from backing file into image file",

        },{

            .name = "detect-zeroes",

            .type = QEMU_OPT_STRING,

            .help = "try to optimize zero writes (off, on, unmap)",

        },

        { /* end of list */ }

    },

hmp.c

+5 −0
Original line number Diff line number Diff line
@@ -341,6 +341,11 @@ void hmp_info_block(Monitor *mon, const QDict *qdict) 
                           info->value->inserted->backing_file_depth);

        }



        if (info->value->inserted->detect_zeroes != BLOCKDEV_DETECT_ZEROES_OPTIONS_OFF) {

            monitor_printf(mon, "    Detect zeroes:    %s\n",

                           BlockdevDetectZeroesOptions_lookup[info->value->inserted->detect_zeroes]);

        }



        if (info->value->inserted->bps

            || info->value->inserted->bps_rd

            || info->value->inserted->bps_wr

include/block/block_int.h

+1 −0
Original line number Diff line number Diff line
@@ -364,6 +364,7 @@ struct BlockDriverState { 
    BlockJob *job;



    QDict *options;

    BlockdevDetectZeroesOptions detect_zeroes;

};



int get_tmp_filename(char *filename, int size);