2020-10-21 17:58:46 +03:00
|
|
|
/*
|
|
|
|
* preallocate filter driver
|
|
|
|
*
|
|
|
|
* The driver performs preallocate operation: it is injected above
|
|
|
|
* some node, and before each write over EOF it does additional preallocating
|
|
|
|
* write-zeroes request.
|
|
|
|
*
|
|
|
|
* Copyright (c) 2020 Virtuozzo International GmbH.
|
|
|
|
*
|
|
|
|
* Author:
|
|
|
|
* Sementsov-Ogievskiy Vladimir <vsementsov@virtuozzo.com>
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or modify
|
|
|
|
* it under the terms of the GNU General Public License as published by
|
|
|
|
* the Free Software Foundation; either version 2 of the License, or
|
|
|
|
* (at your option) any later version.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
* GNU General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License
|
|
|
|
* along with this program. If not, see <http://www.gnu.org/licenses/>.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include "qemu/osdep.h"
|
|
|
|
|
|
|
|
#include "qapi/error.h"
|
|
|
|
#include "qemu/module.h"
|
|
|
|
#include "qemu/option.h"
|
|
|
|
#include "qemu/units.h"
|
|
|
|
#include "block/block_int.h"
|
|
|
|
|
|
|
|
|
|
|
|
typedef struct PreallocateOpts {
|
|
|
|
int64_t prealloc_size;
|
|
|
|
int64_t prealloc_align;
|
|
|
|
} PreallocateOpts;
|
|
|
|
|
|
|
|
typedef struct BDRVPreallocateState {
|
|
|
|
PreallocateOpts opts;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Track real data end, to crop preallocation on close. If < 0 the status is
|
|
|
|
* unknown.
|
|
|
|
*
|
|
|
|
* @data_end is a maximum of file size on open (or when we get write/resize
|
|
|
|
* permissions) and all write request ends after it. So it's safe to
|
|
|
|
* truncate to data_end if it is valid.
|
|
|
|
*/
|
|
|
|
int64_t data_end;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Start of trailing preallocated area which reads as zero. May be smaller
|
|
|
|
* than data_end, if user does over-EOF write zero operation. If < 0 the
|
|
|
|
* status is unknown.
|
|
|
|
*
|
|
|
|
* If both @zero_start and @file_end are valid, the region
|
|
|
|
* [@zero_start, @file_end) is known to be preallocated zeroes. If @file_end
|
|
|
|
* is not valid, @zero_start doesn't make much sense.
|
|
|
|
*/
|
|
|
|
int64_t zero_start;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Real end of file. Actually the cache for bdrv_getlength(bs->file->bs),
|
|
|
|
* to avoid extra lseek() calls on each write operation. If < 0 the status
|
|
|
|
* is unknown.
|
|
|
|
*/
|
|
|
|
int64_t file_end;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* All three states @data_end, @zero_start and @file_end are guaranteed to
|
|
|
|
* be invalid (< 0) when we don't have both exclusive BLK_PERM_RESIZE and
|
|
|
|
* BLK_PERM_WRITE permissions on file child.
|
|
|
|
*/
|
|
|
|
} BDRVPreallocateState;
|
|
|
|
|
|
|
|
#define PREALLOCATE_OPT_PREALLOC_ALIGN "prealloc-align"
|
|
|
|
#define PREALLOCATE_OPT_PREALLOC_SIZE "prealloc-size"
|
|
|
|
static QemuOptsList runtime_opts = {
|
|
|
|
.name = "preallocate",
|
|
|
|
.head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
|
|
|
|
.desc = {
|
|
|
|
{
|
|
|
|
.name = PREALLOCATE_OPT_PREALLOC_ALIGN,
|
|
|
|
.type = QEMU_OPT_SIZE,
|
|
|
|
.help = "on preallocation, align file length to this number, "
|
|
|
|
"default 1M",
|
|
|
|
},
|
|
|
|
{
|
|
|
|
.name = PREALLOCATE_OPT_PREALLOC_SIZE,
|
|
|
|
.type = QEMU_OPT_SIZE,
|
|
|
|
.help = "how much to preallocate, default 128M",
|
|
|
|
},
|
|
|
|
{ /* end of list */ }
|
|
|
|
},
|
|
|
|
};
|
|
|
|
|
|
|
|
static bool preallocate_absorb_opts(PreallocateOpts *dest, QDict *options,
|
|
|
|
BlockDriverState *child_bs, Error **errp)
|
|
|
|
{
|
|
|
|
QemuOpts *opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
|
|
|
|
|
|
|
|
if (!qemu_opts_absorb_qdict(opts, options, errp)) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
dest->prealloc_align =
|
|
|
|
qemu_opt_get_size(opts, PREALLOCATE_OPT_PREALLOC_ALIGN, 1 * MiB);
|
|
|
|
dest->prealloc_size =
|
|
|
|
qemu_opt_get_size(opts, PREALLOCATE_OPT_PREALLOC_SIZE, 128 * MiB);
|
|
|
|
|
|
|
|
qemu_opts_del(opts);
|
|
|
|
|
|
|
|
if (!QEMU_IS_ALIGNED(dest->prealloc_align, BDRV_SECTOR_SIZE)) {
|
|
|
|
error_setg(errp, "prealloc-align parameter of preallocate filter "
|
|
|
|
"is not aligned to %llu", BDRV_SECTOR_SIZE);
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!QEMU_IS_ALIGNED(dest->prealloc_align,
|
|
|
|
child_bs->bl.request_alignment)) {
|
|
|
|
error_setg(errp, "prealloc-align parameter of preallocate filter "
|
|
|
|
"is not aligned to underlying node request alignment "
|
|
|
|
"(%" PRIi32 ")", child_bs->bl.request_alignment);
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int preallocate_open(BlockDriverState *bs, QDict *options, int flags,
|
|
|
|
Error **errp)
|
|
|
|
{
|
|
|
|
BDRVPreallocateState *s = bs->opaque;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* s->data_end and friends should be initialized on permission update.
|
|
|
|
* For this to work, mark them invalid.
|
|
|
|
*/
|
|
|
|
s->file_end = s->zero_start = s->data_end = -EINVAL;
|
|
|
|
|
|
|
|
bs->file = bdrv_open_child(NULL, options, "file", bs, &child_of_bds,
|
|
|
|
BDRV_CHILD_FILTERED | BDRV_CHILD_PRIMARY,
|
|
|
|
false, errp);
|
|
|
|
if (!bs->file) {
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!preallocate_absorb_opts(&s->opts, options, bs->file->bs, errp)) {
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
|
|
|
bs->supported_write_flags = BDRV_REQ_WRITE_UNCHANGED |
|
|
|
|
(BDRV_REQ_FUA & bs->file->bs->supported_write_flags);
|
|
|
|
|
|
|
|
bs->supported_zero_flags = BDRV_REQ_WRITE_UNCHANGED |
|
|
|
|
((BDRV_REQ_FUA | BDRV_REQ_MAY_UNMAP | BDRV_REQ_NO_FALLBACK) &
|
|
|
|
bs->file->bs->supported_zero_flags);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void preallocate_close(BlockDriverState *bs)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
BDRVPreallocateState *s = bs->opaque;
|
|
|
|
|
|
|
|
if (s->data_end < 0) {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (s->file_end < 0) {
|
|
|
|
s->file_end = bdrv_getlength(bs->file->bs);
|
|
|
|
if (s->file_end < 0) {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (s->data_end < s->file_end) {
|
|
|
|
ret = bdrv_truncate(bs->file, s->data_end, true, PREALLOC_MODE_OFF, 0,
|
|
|
|
NULL);
|
|
|
|
s->file_end = ret < 0 ? ret : s->data_end;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Handle reopen.
|
|
|
|
*
|
|
|
|
* We must implement reopen handlers, otherwise reopen just don't work. Handle
|
|
|
|
* new options and don't care about preallocation state, as it is handled in
|
|
|
|
* set/check permission handlers.
|
|
|
|
*/
|
|
|
|
|
|
|
|
static int preallocate_reopen_prepare(BDRVReopenState *reopen_state,
|
|
|
|
BlockReopenQueue *queue, Error **errp)
|
|
|
|
{
|
|
|
|
PreallocateOpts *opts = g_new0(PreallocateOpts, 1);
|
|
|
|
|
|
|
|
if (!preallocate_absorb_opts(opts, reopen_state->options,
|
|
|
|
reopen_state->bs->file->bs, errp)) {
|
|
|
|
g_free(opts);
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
|
|
|
reopen_state->opaque = opts;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void preallocate_reopen_commit(BDRVReopenState *state)
|
|
|
|
{
|
|
|
|
BDRVPreallocateState *s = state->bs->opaque;
|
|
|
|
|
|
|
|
s->opts = *(PreallocateOpts *)state->opaque;
|
|
|
|
|
|
|
|
g_free(state->opaque);
|
|
|
|
state->opaque = NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void preallocate_reopen_abort(BDRVReopenState *state)
|
|
|
|
{
|
|
|
|
g_free(state->opaque);
|
|
|
|
state->opaque = NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static coroutine_fn int preallocate_co_preadv_part(
|
block: use int64_t instead of uint64_t in driver read handlers
We are generally moving to int64_t for both offset and bytes parameters
on all io paths.
Main motivation is realization of 64-bit write_zeroes operation for
fast zeroing large disk chunks, up to the whole disk.
We chose signed type, to be consistent with off_t (which is signed) and
with possibility for signed return type (where negative value means
error).
So, convert driver read handlers parameters which are already 64bit to
signed type.
While being here, convert also flags parameter to be BdrvRequestFlags.
Now let's consider all callers. Simple
git grep '\->bdrv_\(aio\|co\)_preadv\(_part\)\?'
shows that's there three callers of driver function:
bdrv_driver_preadv() in block/io.c, passes int64_t, checked by
bdrv_check_qiov_request() to be non-negative.
qcow2_load_vmstate() does bdrv_check_qiov_request().
do_perform_cow_read() has uint64_t argument. And a lot of things in
qcow2 driver are uint64_t, so converting it is big job. But we must
not work with requests that don't satisfy bdrv_check_qiov_request(),
so let's just assert it here.
Still, the functions may be called directly, not only by drv->...
Let's check:
git grep '\.bdrv_\(aio\|co\)_preadv\(_part\)\?\s*=' | \
awk '{print $4}' | sed 's/,//' | sed 's/&//' | sort | uniq | \
while read func; do git grep "$func(" | \
grep -v "$func(BlockDriverState"; done
The only one such caller:
QEMUIOVector qiov = QEMU_IOVEC_INIT_BUF(qiov, &data, 1);
...
ret = bdrv_replace_test_co_preadv(bs, 0, 1, &qiov, 0);
in tests/unit/test-bdrv-drain.c, and it's OK obviously.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210903102807.27127-4-vsementsov@virtuozzo.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
[eblake: fix typos]
Signed-off-by: Eric Blake <eblake@redhat.com>
2021-09-03 13:27:59 +03:00
|
|
|
BlockDriverState *bs, int64_t offset, int64_t bytes,
|
|
|
|
QEMUIOVector *qiov, size_t qiov_offset, BdrvRequestFlags flags)
|
2020-10-21 17:58:46 +03:00
|
|
|
{
|
|
|
|
return bdrv_co_preadv_part(bs->file, offset, bytes, qiov, qiov_offset,
|
|
|
|
flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int coroutine_fn preallocate_co_pdiscard(BlockDriverState *bs,
|
block: use int64_t instead of int in driver discard handlers
We are generally moving to int64_t for both offset and bytes parameters
on all io paths.
Main motivation is realization of 64-bit write_zeroes operation for
fast zeroing large disk chunks, up to the whole disk.
We chose signed type, to be consistent with off_t (which is signed) and
with possibility for signed return type (where negative value means
error).
So, convert driver discard handlers bytes parameter to int64_t.
The only caller of all updated function is bdrv_co_pdiscard in
block/io.c. It is already prepared to work with 64bit requests, but
pass at most max(bs->bl.max_pdiscard, INT_MAX) to the driver.
Let's look at all updated functions:
blkdebug: all calculations are still OK, thanks to
bdrv_check_qiov_request().
both rule_check and bdrv_co_pdiscard are 64bit
blklogwrites: pass to blk_loc_writes_co_log which is 64bit
blkreplay, copy-on-read, filter-compress: pass to bdrv_co_pdiscard, OK
copy-before-write: pass to bdrv_co_pdiscard which is 64bit and to
cbw_do_copy_before_write which is 64bit
file-posix: one handler calls raw_account_discard() is 64bit and both
handlers calls raw_do_pdiscard(). Update raw_do_pdiscard, which pass
to RawPosixAIOData::aio_nbytes, which is 64bit (and calls
raw_account_discard())
gluster: somehow, third argument of glfs_discard_async is size_t.
Let's set max_pdiscard accordingly.
iscsi: iscsi_allocmap_set_invalid is 64bit,
!is_byte_request_lun_aligned is 64bit.
list.num is uint32_t. Let's clarify max_pdiscard and
pdiscard_alignment.
mirror_top: pass to bdrv_mirror_top_do_write() which is
64bit
nbd: protocol limitation. max_pdiscard is alredy set strict enough,
keep it as is for now.
nvme: buf.nlb is uint32_t and we do shift. So, add corresponding limits
to nvme_refresh_limits().
preallocate: pass to bdrv_co_pdiscard() which is 64bit.
rbd: pass to qemu_rbd_start_co() which is 64bit.
qcow2: calculations are still OK, thanks to bdrv_check_qiov_request(),
qcow2_cluster_discard() is 64bit.
raw-format: raw_adjust_offset() is 64bit, bdrv_co_pdiscard too.
throttle: pass to bdrv_co_pdiscard() which is 64bit and to
throttle_group_co_io_limits_intercept() which is 64bit as well.
test-block-iothread: bytes argument is unused
Great! Now all drivers are prepared to handle 64bit discard requests,
or else have explicit max_pdiscard limits.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210903102807.27127-11-vsementsov@virtuozzo.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
2021-09-03 13:28:06 +03:00
|
|
|
int64_t offset, int64_t bytes)
|
2020-10-21 17:58:46 +03:00
|
|
|
{
|
|
|
|
return bdrv_co_pdiscard(bs->file, offset, bytes);
|
|
|
|
}
|
|
|
|
|
|
|
|
static bool can_write_resize(uint64_t perm)
|
|
|
|
{
|
|
|
|
return (perm & BLK_PERM_WRITE) && (perm & BLK_PERM_RESIZE);
|
|
|
|
}
|
|
|
|
|
|
|
|
static bool has_prealloc_perms(BlockDriverState *bs)
|
|
|
|
{
|
|
|
|
BDRVPreallocateState *s = bs->opaque;
|
|
|
|
|
|
|
|
if (can_write_resize(bs->file->perm)) {
|
|
|
|
assert(!(bs->file->shared_perm & BLK_PERM_WRITE));
|
|
|
|
assert(!(bs->file->shared_perm & BLK_PERM_RESIZE));
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
assert(s->data_end < 0);
|
|
|
|
assert(s->zero_start < 0);
|
|
|
|
assert(s->file_end < 0);
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Call on each write. Returns true if @want_merge_zero is true and the region
|
|
|
|
* [offset, offset + bytes) is zeroed (as a result of this call or earlier
|
|
|
|
* preallocation).
|
|
|
|
*
|
|
|
|
* want_merge_zero is used to merge write-zero request with preallocation in
|
|
|
|
* one bdrv_co_pwrite_zeroes() call.
|
|
|
|
*/
|
|
|
|
static bool coroutine_fn handle_write(BlockDriverState *bs, int64_t offset,
|
|
|
|
int64_t bytes, bool want_merge_zero)
|
|
|
|
{
|
|
|
|
BDRVPreallocateState *s = bs->opaque;
|
|
|
|
int64_t end = offset + bytes;
|
|
|
|
int64_t prealloc_start, prealloc_end;
|
|
|
|
int ret;
|
2022-02-15 15:16:09 +03:00
|
|
|
uint32_t file_align = bs->file->bs->bl.request_alignment;
|
|
|
|
uint32_t prealloc_align = MAX(s->opts.prealloc_align, file_align);
|
|
|
|
|
|
|
|
assert(QEMU_IS_ALIGNED(prealloc_align, file_align));
|
2020-10-21 17:58:46 +03:00
|
|
|
|
|
|
|
if (!has_prealloc_perms(bs)) {
|
|
|
|
/* We don't have state neither should try to recover it */
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (s->data_end < 0) {
|
|
|
|
s->data_end = bdrv_getlength(bs->file->bs);
|
|
|
|
if (s->data_end < 0) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (s->file_end < 0) {
|
|
|
|
s->file_end = s->data_end;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (end <= s->data_end) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We have valid s->data_end, and request writes beyond it. */
|
|
|
|
|
|
|
|
s->data_end = end;
|
|
|
|
if (s->zero_start < 0 || !want_merge_zero) {
|
|
|
|
s->zero_start = end;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (s->file_end < 0) {
|
|
|
|
s->file_end = bdrv_getlength(bs->file->bs);
|
|
|
|
if (s->file_end < 0) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Now s->data_end, s->zero_start and s->file_end are valid. */
|
|
|
|
|
|
|
|
if (end <= s->file_end) {
|
|
|
|
/* No preallocation needed. */
|
|
|
|
return want_merge_zero && offset >= s->zero_start;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Now we want new preallocation, as request writes beyond s->file_end. */
|
|
|
|
|
2022-02-15 15:16:09 +03:00
|
|
|
prealloc_start = QEMU_ALIGN_UP(
|
|
|
|
want_merge_zero ? MIN(offset, s->file_end) : s->file_end,
|
|
|
|
file_align);
|
|
|
|
prealloc_end = QEMU_ALIGN_UP(
|
|
|
|
MAX(prealloc_start, end) + s->opts.prealloc_size,
|
|
|
|
prealloc_align);
|
|
|
|
|
|
|
|
want_merge_zero = want_merge_zero && (prealloc_start <= offset);
|
2020-10-21 17:58:46 +03:00
|
|
|
|
|
|
|
ret = bdrv_co_pwrite_zeroes(
|
|
|
|
bs->file, prealloc_start, prealloc_end - prealloc_start,
|
|
|
|
BDRV_REQ_NO_FALLBACK | BDRV_REQ_SERIALISING | BDRV_REQ_NO_WAIT);
|
|
|
|
if (ret < 0) {
|
|
|
|
s->file_end = ret;
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
s->file_end = prealloc_end;
|
|
|
|
return want_merge_zero;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int coroutine_fn preallocate_co_pwrite_zeroes(BlockDriverState *bs,
|
block: use int64_t instead of int in driver write_zeroes handlers
We are generally moving to int64_t for both offset and bytes parameters
on all io paths.
Main motivation is realization of 64-bit write_zeroes operation for
fast zeroing large disk chunks, up to the whole disk.
We chose signed type, to be consistent with off_t (which is signed) and
with possibility for signed return type (where negative value means
error).
So, convert driver write_zeroes handlers bytes parameter to int64_t.
The only caller of all updated function is bdrv_co_do_pwrite_zeroes().
bdrv_co_do_pwrite_zeroes() itself is of course OK with widening of
callee parameter type. Also, bdrv_co_do_pwrite_zeroes()'s
max_write_zeroes is limited to INT_MAX. So, updated functions all are
safe, they will not get "bytes" larger than before.
Still, let's look through all updated functions, and add assertions to
the ones which are actually unprepared to values larger than INT_MAX.
For these drivers also set explicit max_pwrite_zeroes limit.
Let's go:
blkdebug: calculations can't overflow, thanks to
bdrv_check_qiov_request() in generic layer. rule_check() and
bdrv_co_pwrite_zeroes() both have 64bit argument.
blklogwrites: pass to blk_log_writes_co_log() with 64bit argument.
blkreplay, copy-on-read, filter-compress: pass to
bdrv_co_pwrite_zeroes() which is OK
copy-before-write: Calls cbw_do_copy_before_write() and
bdrv_co_pwrite_zeroes, both have 64bit argument.
file-posix: both handler calls raw_do_pwrite_zeroes, which is updated.
In raw_do_pwrite_zeroes() calculations are OK due to
bdrv_check_qiov_request(), bytes go to RawPosixAIOData::aio_nbytes
which is uint64_t.
Check also where that uint64_t gets handed:
handle_aiocb_write_zeroes_block() passes a uint64_t[2] to
ioctl(BLKZEROOUT), handle_aiocb_write_zeroes() calls do_fallocate()
which takes off_t (and we compile to always have 64-bit off_t), as
does handle_aiocb_write_zeroes_unmap. All look safe.
gluster: bytes go to GlusterAIOCB::size which is int64_t and to
glfs_zerofill_async works with off_t.
iscsi: Aha, here we deal with iscsi_writesame16_task() that has
uint32_t num_blocks argument and iscsi_writesame16_task() has
uint16_t argument. Make comments, add assertions and clarify
max_pwrite_zeroes calculation.
iscsi_allocmap_() functions already has int64_t argument
is_byte_request_lun_aligned is simple to update, do it.
mirror_top: pass to bdrv_mirror_top_do_write which has uint64_t
argument
nbd: Aha, here we have protocol limitation, and NBDRequest::len is
uint32_t. max_pwrite_zeroes is cleanly set to 32bit value, so we are
OK for now.
nvme: Again, protocol limitation. And no inherent limit for
write-zeroes at all. But from code that calculates cdw12 it's obvious
that we do have limit and alignment. Let's clarify it. Also,
obviously the code is not prepared to handle bytes=0. Let's handle
this case too.
trace events already 64bit
preallocate: pass to handle_write() and bdrv_co_pwrite_zeroes(), both
64bit.
rbd: pass to qemu_rbd_start_co() which is 64bit.
qcow2: offset + bytes and alignment still works good (thanks to
bdrv_check_qiov_request()), so tail calculation is OK
qcow2_subcluster_zeroize() has 64bit argument, should be OK
trace events updated
qed: qed_co_request wants int nb_sectors. Also in code we have size_t
used for request length which may be 32bit. So, let's just keep
INT_MAX as a limit (aligning it down to pwrite_zeroes_alignment) and
don't care.
raw-format: Is OK. raw_adjust_offset and bdrv_co_pwrite_zeroes are both
64bit.
throttle: Both throttle_group_co_io_limits_intercept() and
bdrv_co_pwrite_zeroes() are 64bit.
vmdk: pass to vmdk_pwritev which is 64bit
quorum: pass to quorum_co_pwritev() which is 64bit
Hooray!
At this point all block drivers are prepared to support 64bit
write-zero requests, or have explicitly set max_pwrite_zeroes.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210903102807.27127-8-vsementsov@virtuozzo.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
[eblake: use <= rather than < in assertions relying on max_pwrite_zeroes]
Signed-off-by: Eric Blake <eblake@redhat.com>
2021-09-03 13:28:03 +03:00
|
|
|
int64_t offset, int64_t bytes, BdrvRequestFlags flags)
|
2020-10-21 17:58:46 +03:00
|
|
|
{
|
|
|
|
bool want_merge_zero =
|
|
|
|
!(flags & ~(BDRV_REQ_ZERO_WRITE | BDRV_REQ_NO_FALLBACK));
|
|
|
|
if (handle_write(bs, offset, bytes, want_merge_zero)) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
return bdrv_co_pwrite_zeroes(bs->file, offset, bytes, flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
static coroutine_fn int preallocate_co_pwritev_part(BlockDriverState *bs,
|
block: use int64_t instead of uint64_t in driver write handlers
We are generally moving to int64_t for both offset and bytes parameters
on all io paths.
Main motivation is realization of 64-bit write_zeroes operation for
fast zeroing large disk chunks, up to the whole disk.
We chose signed type, to be consistent with off_t (which is signed) and
with possibility for signed return type (where negative value means
error).
So, convert driver write handlers parameters which are already 64bit to
signed type.
While being here, convert also flags parameter to be BdrvRequestFlags.
Now let's consider all callers. Simple
git grep '\->bdrv_\(aio\|co\)_pwritev\(_part\)\?'
shows that's there three callers of driver function:
bdrv_driver_pwritev() and bdrv_driver_pwritev_compressed() in
block/io.c, both pass int64_t, checked by bdrv_check_qiov_request() to
be non-negative.
qcow2_save_vmstate() does bdrv_check_qiov_request().
Still, the functions may be called directly, not only by drv->...
Let's check:
git grep '\.bdrv_\(aio\|co\)_pwritev\(_part\)\?\s*=' | \
awk '{print $4}' | sed 's/,//' | sed 's/&//' | sort | uniq | \
while read func; do git grep "$func(" | \
grep -v "$func(BlockDriverState"; done
shows several callers:
qcow2:
qcow2_co_truncate() write at most up to @offset, which is checked in
generic qcow2_co_truncate() by bdrv_check_request().
qcow2_co_pwritev_compressed_task() pass the request (or part of the
request) that already went through normal write path, so it should
be OK
qcow:
qcow_co_pwritev_compressed() pass int64_t, it's updated by this patch
quorum:
quorum_co_pwrite_zeroes() pass int64_t and int - OK
throttle:
throttle_co_pwritev_compressed() pass int64_t, it's updated by this
patch
vmdk:
vmdk_co_pwritev_compressed() pass int64_t, it's updated by this
patch
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210903102807.27127-5-vsementsov@virtuozzo.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
2021-09-03 13:28:00 +03:00
|
|
|
int64_t offset,
|
|
|
|
int64_t bytes,
|
2020-10-21 17:58:46 +03:00
|
|
|
QEMUIOVector *qiov,
|
|
|
|
size_t qiov_offset,
|
block: use int64_t instead of uint64_t in driver write handlers
We are generally moving to int64_t for both offset and bytes parameters
on all io paths.
Main motivation is realization of 64-bit write_zeroes operation for
fast zeroing large disk chunks, up to the whole disk.
We chose signed type, to be consistent with off_t (which is signed) and
with possibility for signed return type (where negative value means
error).
So, convert driver write handlers parameters which are already 64bit to
signed type.
While being here, convert also flags parameter to be BdrvRequestFlags.
Now let's consider all callers. Simple
git grep '\->bdrv_\(aio\|co\)_pwritev\(_part\)\?'
shows that's there three callers of driver function:
bdrv_driver_pwritev() and bdrv_driver_pwritev_compressed() in
block/io.c, both pass int64_t, checked by bdrv_check_qiov_request() to
be non-negative.
qcow2_save_vmstate() does bdrv_check_qiov_request().
Still, the functions may be called directly, not only by drv->...
Let's check:
git grep '\.bdrv_\(aio\|co\)_pwritev\(_part\)\?\s*=' | \
awk '{print $4}' | sed 's/,//' | sed 's/&//' | sort | uniq | \
while read func; do git grep "$func(" | \
grep -v "$func(BlockDriverState"; done
shows several callers:
qcow2:
qcow2_co_truncate() write at most up to @offset, which is checked in
generic qcow2_co_truncate() by bdrv_check_request().
qcow2_co_pwritev_compressed_task() pass the request (or part of the
request) that already went through normal write path, so it should
be OK
qcow:
qcow_co_pwritev_compressed() pass int64_t, it's updated by this patch
quorum:
quorum_co_pwrite_zeroes() pass int64_t and int - OK
throttle:
throttle_co_pwritev_compressed() pass int64_t, it's updated by this
patch
vmdk:
vmdk_co_pwritev_compressed() pass int64_t, it's updated by this
patch
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210903102807.27127-5-vsementsov@virtuozzo.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
2021-09-03 13:28:00 +03:00
|
|
|
BdrvRequestFlags flags)
|
2020-10-21 17:58:46 +03:00
|
|
|
{
|
|
|
|
handle_write(bs, offset, bytes, false);
|
|
|
|
|
|
|
|
return bdrv_co_pwritev_part(bs->file, offset, bytes, qiov, qiov_offset,
|
|
|
|
flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int coroutine_fn
|
|
|
|
preallocate_co_truncate(BlockDriverState *bs, int64_t offset,
|
|
|
|
bool exact, PreallocMode prealloc,
|
|
|
|
BdrvRequestFlags flags, Error **errp)
|
|
|
|
{
|
|
|
|
ERRP_GUARD();
|
|
|
|
BDRVPreallocateState *s = bs->opaque;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (s->data_end >= 0 && offset > s->data_end) {
|
|
|
|
if (s->file_end < 0) {
|
|
|
|
s->file_end = bdrv_getlength(bs->file->bs);
|
|
|
|
if (s->file_end < 0) {
|
|
|
|
error_setg(errp, "failed to get file length");
|
|
|
|
return s->file_end;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (prealloc == PREALLOC_MODE_FALLOC) {
|
|
|
|
/*
|
|
|
|
* If offset <= s->file_end, the task is already done, just
|
|
|
|
* update s->data_end, to move part of "filter preallocation"
|
|
|
|
* to "preallocation requested by user".
|
|
|
|
* Otherwise just proceed to preallocate missing part.
|
|
|
|
*/
|
|
|
|
if (offset <= s->file_end) {
|
|
|
|
s->data_end = offset;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* We have to drop our preallocation, to
|
|
|
|
* - avoid "Cannot use preallocation for shrinking files" in
|
|
|
|
* case of offset < file_end
|
|
|
|
* - give PREALLOC_MODE_OFF a chance to keep small disk
|
|
|
|
* usage
|
|
|
|
* - give PREALLOC_MODE_FULL a chance to actually write the
|
|
|
|
* whole region as user expects
|
|
|
|
*/
|
|
|
|
if (s->file_end > s->data_end) {
|
|
|
|
ret = bdrv_co_truncate(bs->file, s->data_end, true,
|
|
|
|
PREALLOC_MODE_OFF, 0, errp);
|
|
|
|
if (ret < 0) {
|
|
|
|
s->file_end = ret;
|
|
|
|
error_prepend(errp, "preallocate-filter: failed to drop "
|
|
|
|
"write-zero preallocation: ");
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
s->file_end = s->data_end;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
s->data_end = offset;
|
|
|
|
}
|
|
|
|
|
|
|
|
ret = bdrv_co_truncate(bs->file, offset, exact, prealloc, flags, errp);
|
|
|
|
if (ret < 0) {
|
|
|
|
s->file_end = s->zero_start = s->data_end = ret;
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (has_prealloc_perms(bs)) {
|
|
|
|
s->file_end = s->zero_start = s->data_end = offset;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int coroutine_fn preallocate_co_flush(BlockDriverState *bs)
|
|
|
|
{
|
|
|
|
return bdrv_co_flush(bs->file->bs);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int64_t preallocate_getlength(BlockDriverState *bs)
|
|
|
|
{
|
|
|
|
int64_t ret;
|
|
|
|
BDRVPreallocateState *s = bs->opaque;
|
|
|
|
|
|
|
|
if (s->data_end >= 0) {
|
|
|
|
return s->data_end;
|
|
|
|
}
|
|
|
|
|
|
|
|
ret = bdrv_getlength(bs->file->bs);
|
|
|
|
|
|
|
|
if (has_prealloc_perms(bs)) {
|
|
|
|
s->file_end = s->zero_start = s->data_end = ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int preallocate_check_perm(BlockDriverState *bs,
|
|
|
|
uint64_t perm, uint64_t shared, Error **errp)
|
|
|
|
{
|
|
|
|
BDRVPreallocateState *s = bs->opaque;
|
|
|
|
|
|
|
|
if (s->data_end >= 0 && !can_write_resize(perm)) {
|
|
|
|
/*
|
|
|
|
* Lose permissions.
|
|
|
|
* We should truncate in check_perm, as in set_perm bs->file->perm will
|
|
|
|
* be already changed, and we should not violate it.
|
|
|
|
*/
|
|
|
|
if (s->file_end < 0) {
|
|
|
|
s->file_end = bdrv_getlength(bs->file->bs);
|
|
|
|
if (s->file_end < 0) {
|
|
|
|
error_setg(errp, "Failed to get file length");
|
|
|
|
return s->file_end;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (s->data_end < s->file_end) {
|
|
|
|
int ret = bdrv_truncate(bs->file, s->data_end, true,
|
|
|
|
PREALLOC_MODE_OFF, 0, NULL);
|
|
|
|
if (ret < 0) {
|
|
|
|
error_setg(errp, "Failed to drop preallocation");
|
|
|
|
s->file_end = ret;
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
s->file_end = s->data_end;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void preallocate_set_perm(BlockDriverState *bs,
|
|
|
|
uint64_t perm, uint64_t shared)
|
|
|
|
{
|
|
|
|
BDRVPreallocateState *s = bs->opaque;
|
|
|
|
|
|
|
|
if (can_write_resize(perm)) {
|
|
|
|
if (s->data_end < 0) {
|
|
|
|
s->data_end = s->file_end = s->zero_start =
|
|
|
|
bdrv_getlength(bs->file->bs);
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* We drop our permissions, as well as allow shared
|
|
|
|
* permissions (see preallocate_child_perm), anyone will be able to
|
|
|
|
* change the child, so mark all states invalid. We'll regain control if
|
|
|
|
* get good permissions back.
|
|
|
|
*/
|
|
|
|
s->data_end = s->file_end = s->zero_start = -EINVAL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void preallocate_child_perm(BlockDriverState *bs, BdrvChild *c,
|
|
|
|
BdrvChildRole role, BlockReopenQueue *reopen_queue,
|
|
|
|
uint64_t perm, uint64_t shared, uint64_t *nperm, uint64_t *nshared)
|
|
|
|
{
|
|
|
|
bdrv_default_perms(bs, c, role, reopen_queue, perm, shared, nperm, nshared);
|
|
|
|
|
|
|
|
if (can_write_resize(perm)) {
|
|
|
|
/* This should come by default, but let's enforce: */
|
|
|
|
*nperm |= BLK_PERM_WRITE | BLK_PERM_RESIZE;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Don't share, to keep our states s->file_end, s->data_end and
|
|
|
|
* s->zero_start valid.
|
|
|
|
*/
|
|
|
|
*nshared &= ~(BLK_PERM_WRITE | BLK_PERM_RESIZE);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
BlockDriver bdrv_preallocate_filter = {
|
|
|
|
.format_name = "preallocate",
|
|
|
|
.instance_size = sizeof(BDRVPreallocateState),
|
|
|
|
|
|
|
|
.bdrv_getlength = preallocate_getlength,
|
|
|
|
.bdrv_open = preallocate_open,
|
|
|
|
.bdrv_close = preallocate_close,
|
|
|
|
|
|
|
|
.bdrv_reopen_prepare = preallocate_reopen_prepare,
|
|
|
|
.bdrv_reopen_commit = preallocate_reopen_commit,
|
|
|
|
.bdrv_reopen_abort = preallocate_reopen_abort,
|
|
|
|
|
|
|
|
.bdrv_co_preadv_part = preallocate_co_preadv_part,
|
|
|
|
.bdrv_co_pwritev_part = preallocate_co_pwritev_part,
|
|
|
|
.bdrv_co_pwrite_zeroes = preallocate_co_pwrite_zeroes,
|
|
|
|
.bdrv_co_pdiscard = preallocate_co_pdiscard,
|
|
|
|
.bdrv_co_flush = preallocate_co_flush,
|
|
|
|
.bdrv_co_truncate = preallocate_co_truncate,
|
|
|
|
|
|
|
|
.bdrv_check_perm = preallocate_check_perm,
|
|
|
|
.bdrv_set_perm = preallocate_set_perm,
|
|
|
|
.bdrv_child_perm = preallocate_child_perm,
|
|
|
|
|
|
|
|
.has_variable_length = true,
|
|
|
|
.is_filter = true,
|
|
|
|
};
|
|
|
|
|
|
|
|
static void bdrv_preallocate_init(void)
|
|
|
|
{
|
|
|
|
bdrv_register(&bdrv_preallocate_filter);
|
|
|
|
}
|
|
|
|
|
|
|
|
block_init(bdrv_preallocate_init);
|