diff options
author | biao716.wang <biao716.wang@samsung.com> | 2023-08-28 14:40:25 +0900 |
---|---|---|
committer | biao716.wang <biao716.wang@samsung.com> | 2023-08-28 14:40:25 +0900 |
commit | d0ba767e7dc0fded6daae31c11d13b538ed8d53f (patch) | |
tree | eaaffe8ef043ee1803a7bbff50a9946e1613f58f /block/commit.c | |
parent | 0889ee8339e51dfdf78c39a8f15b6e5fe5ab49d5 (diff) | |
download | qemu-arm-static-d0ba767e7dc0fded6daae31c11d13b538ed8d53f.tar.gz qemu-arm-static-d0ba767e7dc0fded6daae31c11d13b538ed8d53f.tar.bz2 qemu-arm-static-d0ba767e7dc0fded6daae31c11d13b538ed8d53f.zip |
Upgrade version to 5.2upstream/5.2.0sandbox/wangbiao/qemu_5.2
Change-Id: I79199e4f7e9f27b498cd41113833a6158d4f07a2
Signed-off-by: biao716.wang <biao716.wang@samsung.com>
Diffstat (limited to 'block/commit.c')
-rw-r--r-- | block/commit.c | 597 |
1 files changed, 457 insertions, 140 deletions
diff --git a/block/commit.c b/block/commit.c index 2227fc2e6..71db7ba74 100644 --- a/block/commit.c +++ b/block/commit.c @@ -12,10 +12,15 @@ * */ +#include "qemu/osdep.h" +#include "qemu/cutils.h" #include "trace.h" #include "block/block_int.h" -#include "block/blockjob.h" +#include "block/blockjob_int.h" +#include "qapi/error.h" +#include "qapi/qmp/qerror.h" #include "qemu/ratelimit.h" +#include "sysemu/block-backend.h" enum { /* @@ -26,233 +31,545 @@ enum { COMMIT_BUFFER_SIZE = 512 * 1024, /* in bytes */ }; -#define SLICE_TIME 100000000ULL /* ns */ - typedef struct CommitBlockJob { BlockJob common; - RateLimit limit; - BlockDriverState *active; - BlockDriverState *top; - BlockDriverState *base; + BlockDriverState *commit_top_bs; + BlockBackend *top; + BlockBackend *base; + BlockDriverState *base_bs; + BlockDriverState *base_overlay; BlockdevOnError on_error; - int base_flags; - int orig_overlay_flags; + bool base_read_only; + bool chain_frozen; + char *backing_file_str; } CommitBlockJob; -static int coroutine_fn commit_populate(BlockDriverState *bs, - BlockDriverState *base, - int64_t sector_num, int nb_sectors, - void *buf) +static int commit_prepare(Job *job) { - int ret = 0; + CommitBlockJob *s = container_of(job, CommitBlockJob, common.job); + + bdrv_unfreeze_backing_chain(s->commit_top_bs, s->base_bs); + s->chain_frozen = false; - ret = bdrv_read(bs, sector_num, buf, nb_sectors); - if (ret) { - return ret; + /* Remove base node parent that still uses BLK_PERM_WRITE/RESIZE before + * the normal backing chain can be restored. */ + blk_unref(s->base); + s->base = NULL; + + /* FIXME: bdrv_drop_intermediate treats total failures and partial failures + * identically. Further work is needed to disambiguate these cases. */ + return bdrv_drop_intermediate(s->commit_top_bs, s->base_bs, + s->backing_file_str); +} + +static void commit_abort(Job *job) +{ + CommitBlockJob *s = container_of(job, CommitBlockJob, common.job); + BlockDriverState *top_bs = blk_bs(s->top); + + if (s->chain_frozen) { + bdrv_unfreeze_backing_chain(s->commit_top_bs, s->base_bs); } - ret = bdrv_write(base, sector_num, buf, nb_sectors); - if (ret) { - return ret; + /* Make sure commit_top_bs and top stay around until bdrv_replace_node() */ + bdrv_ref(top_bs); + bdrv_ref(s->commit_top_bs); + + if (s->base) { + blk_unref(s->base); } - return 0; + /* free the blockers on the intermediate nodes so that bdrv_replace_nodes + * can succeed */ + block_job_remove_all_bdrv(&s->common); + + /* If bdrv_drop_intermediate() failed (or was not invoked), remove the + * commit filter driver from the backing chain now. Do this as the final + * step so that the 'consistent read' permission can be granted. + * + * XXX Can (or should) we somehow keep 'consistent read' blocked even + * after the failed/cancelled commit job is gone? If we already wrote + * something to base, the intermediate images aren't valid any more. */ + bdrv_replace_node(s->commit_top_bs, s->commit_top_bs->backing->bs, + &error_abort); + + bdrv_unref(s->commit_top_bs); + bdrv_unref(top_bs); } -static void coroutine_fn commit_run(void *opaque) +static void commit_clean(Job *job) { - CommitBlockJob *s = opaque; - BlockDriverState *active = s->active; - BlockDriverState *top = s->top; - BlockDriverState *base = s->base; - BlockDriverState *overlay_bs; - int64_t sector_num, end; - int ret = 0; - int n = 0; - void *buf; - int bytes_written = 0; - int64_t base_len; + CommitBlockJob *s = container_of(job, CommitBlockJob, common.job); - ret = s->common.len = bdrv_getlength(top); + /* restore base open flags here if appropriate (e.g., change the base back + * to r/o). These reopens do not need to be atomic, since we won't abort + * even on failure here */ + if (s->base_read_only) { + bdrv_reopen_set_read_only(s->base_bs, true, NULL); + } + g_free(s->backing_file_str); + blk_unref(s->top); +} - if (s->common.len < 0) { - goto exit_restore_reopen; +static int coroutine_fn commit_run(Job *job, Error **errp) +{ + CommitBlockJob *s = container_of(job, CommitBlockJob, common.job); + int64_t offset; + uint64_t delay_ns = 0; + int ret = 0; + int64_t n = 0; /* bytes */ + void *buf = NULL; + int64_t len, base_len; + + ret = len = blk_getlength(s->top); + if (len < 0) { + goto out; } + job_progress_set_remaining(&s->common.job, len); - ret = base_len = bdrv_getlength(base); + ret = base_len = blk_getlength(s->base); if (base_len < 0) { - goto exit_restore_reopen; + goto out; } - if (base_len < s->common.len) { - ret = bdrv_truncate(base, s->common.len); + if (base_len < len) { + ret = blk_truncate(s->base, len, false, PREALLOC_MODE_OFF, 0, NULL); if (ret) { - goto exit_restore_reopen; + goto out; } } - end = s->common.len >> BDRV_SECTOR_BITS; - buf = qemu_blockalign(top, COMMIT_BUFFER_SIZE); + buf = blk_blockalign(s->top, COMMIT_BUFFER_SIZE); - for (sector_num = 0; sector_num < end; sector_num += n) { - uint64_t delay_ns = 0; + for (offset = 0; offset < len; offset += n) { bool copy; + bool error_in_source = true; -wait: /* Note that even when no rate limit is applied we need to yield * with no pending I/O here so that bdrv_drain_all() returns. */ - block_job_sleep_ns(&s->common, rt_clock, delay_ns); - if (block_job_is_cancelled(&s->common)) { + job_sleep_ns(&s->common.job, delay_ns); + if (job_is_cancelled(&s->common.job)) { break; } /* Copy if allocated above the base */ - ret = bdrv_co_is_allocated_above(top, base, sector_num, - COMMIT_BUFFER_SIZE / BDRV_SECTOR_SIZE, - &n); - copy = (ret == 1); - trace_commit_one_iteration(s, sector_num, n, ret); + ret = bdrv_is_allocated_above(blk_bs(s->top), s->base_overlay, true, + offset, COMMIT_BUFFER_SIZE, &n); + copy = (ret > 0); + trace_commit_one_iteration(s, offset, n, ret); if (copy) { - if (s->common.speed) { - delay_ns = ratelimit_calculate_delay(&s->limit, n); - if (delay_ns > 0) { - goto wait; + assert(n < SIZE_MAX); + + ret = blk_co_pread(s->top, offset, n, buf, 0); + if (ret >= 0) { + ret = blk_co_pwrite(s->base, offset, n, buf, 0); + if (ret < 0) { + error_in_source = false; } } - ret = commit_populate(top, base, sector_num, n, buf); - bytes_written += n * BDRV_SECTOR_SIZE; } if (ret < 0) { - if (s->on_error == BLOCKDEV_ON_ERROR_STOP || - s->on_error == BLOCKDEV_ON_ERROR_REPORT|| - (s->on_error == BLOCKDEV_ON_ERROR_ENOSPC && ret == -ENOSPC)) { - goto exit_free_buf; + BlockErrorAction action = + block_job_error_action(&s->common, s->on_error, + error_in_source, -ret); + if (action == BLOCK_ERROR_ACTION_REPORT) { + goto out; } else { n = 0; continue; } } /* Publish progress */ - s->common.offset += n * BDRV_SECTOR_SIZE; + job_progress_update(&s->common.job, n); + + if (copy) { + delay_ns = block_job_ratelimit_get_delay(&s->common, n); + } else { + delay_ns = 0; + } } ret = 0; - if (!block_job_is_cancelled(&s->common) && sector_num == end) { - /* success */ - ret = bdrv_drop_intermediate(active, top, base); - } - -exit_free_buf: +out: qemu_vfree(buf); -exit_restore_reopen: - /* restore base open flags here if appropriate (e.g., change the base back - * to r/o). These reopens do not need to be atomic, since we won't abort - * even on failure here */ - if (s->base_flags != bdrv_get_flags(base)) { - bdrv_reopen(base, s->base_flags, NULL); - } - overlay_bs = bdrv_find_overlay(active, top); - if (overlay_bs && s->orig_overlay_flags != bdrv_get_flags(overlay_bs)) { - bdrv_reopen(overlay_bs, s->orig_overlay_flags, NULL); - } + return ret; +} + +static const BlockJobDriver commit_job_driver = { + .job_driver = { + .instance_size = sizeof(CommitBlockJob), + .job_type = JOB_TYPE_COMMIT, + .free = block_job_free, + .user_resume = block_job_user_resume, + .run = commit_run, + .prepare = commit_prepare, + .abort = commit_abort, + .clean = commit_clean + }, +}; - block_job_completed(&s->common, ret); +static int coroutine_fn bdrv_commit_top_preadv(BlockDriverState *bs, + uint64_t offset, uint64_t bytes, QEMUIOVector *qiov, int flags) +{ + return bdrv_co_preadv(bs->backing, offset, bytes, qiov, flags); } -static void commit_set_speed(BlockJob *job, int64_t speed, Error **errp) +static void bdrv_commit_top_refresh_filename(BlockDriverState *bs) { - CommitBlockJob *s = container_of(job, CommitBlockJob, common); + pstrcpy(bs->exact_filename, sizeof(bs->exact_filename), + bs->backing->bs->filename); +} - if (speed < 0) { - error_set(errp, QERR_INVALID_PARAMETER, "speed"); - return; - } - ratelimit_set_speed(&s->limit, speed / BDRV_SECTOR_SIZE, SLICE_TIME); +static void bdrv_commit_top_child_perm(BlockDriverState *bs, BdrvChild *c, + BdrvChildRole role, + BlockReopenQueue *reopen_queue, + uint64_t perm, uint64_t shared, + uint64_t *nperm, uint64_t *nshared) +{ + *nperm = 0; + *nshared = BLK_PERM_ALL; } -static const BlockJobType commit_job_type = { - .instance_size = sizeof(CommitBlockJob), - .job_type = "commit", - .set_speed = commit_set_speed, +/* Dummy node that provides consistent read to its users without requiring it + * from its backing file and that allows writes on the backing file chain. */ +static BlockDriver bdrv_commit_top = { + .format_name = "commit_top", + .bdrv_co_preadv = bdrv_commit_top_preadv, + .bdrv_refresh_filename = bdrv_commit_top_refresh_filename, + .bdrv_child_perm = bdrv_commit_top_child_perm, + + .is_filter = true, }; -void commit_start(BlockDriverState *bs, BlockDriverState *base, - BlockDriverState *top, int64_t speed, - BlockdevOnError on_error, BlockDriverCompletionFunc *cb, - void *opaque, Error **errp) +void commit_start(const char *job_id, BlockDriverState *bs, + BlockDriverState *base, BlockDriverState *top, + int creation_flags, int64_t speed, + BlockdevOnError on_error, const char *backing_file_str, + const char *filter_node_name, Error **errp) { CommitBlockJob *s; - BlockReopenQueue *reopen_queue = NULL; - int orig_overlay_flags; - int orig_base_flags; - BlockDriverState *overlay_bs; + BlockDriverState *iter; + BlockDriverState *commit_top_bs = NULL; + BlockDriverState *filtered_base; Error *local_err = NULL; + int64_t base_size, top_size; + uint64_t base_perms, iter_shared_perms; + int ret; - if ((on_error == BLOCKDEV_ON_ERROR_STOP || - on_error == BLOCKDEV_ON_ERROR_ENOSPC) && - !bdrv_iostatus_is_enabled(bs)) { - error_set(errp, QERR_INVALID_PARAMETER_COMBINATION); + assert(top != bs); + if (bdrv_skip_filters(top) == bdrv_skip_filters(base)) { + error_setg(errp, "Invalid files for merge: top and base are the same"); return; } - /* Once we support top == active layer, remove this check */ - if (top == bs) { - error_setg(errp, - "Top image as the active layer is currently unsupported"); + base_size = bdrv_getlength(base); + if (base_size < 0) { + error_setg_errno(errp, -base_size, "Could not inquire base image size"); return; } - if (top == base) { - error_setg(errp, "Invalid files for merge: top and base are the same"); + top_size = bdrv_getlength(top); + if (top_size < 0) { + error_setg_errno(errp, -top_size, "Could not inquire top image size"); return; } - overlay_bs = bdrv_find_overlay(bs, top); + base_perms = BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE; + if (base_size < top_size) { + base_perms |= BLK_PERM_RESIZE; + } - if (overlay_bs == NULL) { - error_setg(errp, "Could not find overlay image for %s:", top->filename); + s = block_job_create(job_id, &commit_job_driver, NULL, bs, 0, BLK_PERM_ALL, + speed, creation_flags, NULL, NULL, errp); + if (!s) { return; } - orig_base_flags = bdrv_get_flags(base); - orig_overlay_flags = bdrv_get_flags(overlay_bs); + /* convert base to r/w, if necessary */ + s->base_read_only = bdrv_is_read_only(base); + if (s->base_read_only) { + if (bdrv_reopen_set_read_only(base, false, errp) != 0) { + goto fail; + } + } - /* convert base & overlay_bs to r/w, if necessary */ - if (!(orig_base_flags & BDRV_O_RDWR)) { - reopen_queue = bdrv_reopen_queue(reopen_queue, base, - orig_base_flags | BDRV_O_RDWR); + /* Insert commit_top block node above top, so we can block consistent read + * on the backing chain below it */ + commit_top_bs = bdrv_new_open_driver(&bdrv_commit_top, filter_node_name, 0, + errp); + if (commit_top_bs == NULL) { + goto fail; } - if (!(orig_overlay_flags & BDRV_O_RDWR)) { - reopen_queue = bdrv_reopen_queue(reopen_queue, overlay_bs, - orig_overlay_flags | BDRV_O_RDWR); + if (!filter_node_name) { + commit_top_bs->implicit = true; } - if (reopen_queue) { - bdrv_reopen_multiple(reopen_queue, &local_err); - if (local_err != NULL) { - error_propagate(errp, local_err); - return; - } + + /* So that we can always drop this node */ + commit_top_bs->never_freeze = true; + + commit_top_bs->total_sectors = top->total_sectors; + + bdrv_append(commit_top_bs, top, &local_err); + if (local_err) { + commit_top_bs = NULL; + error_propagate(errp, local_err); + goto fail; } + s->commit_top_bs = commit_top_bs; - s = block_job_create(&commit_job_type, bs, speed, cb, opaque, errp); - if (!s) { - return; + /* + * Block all nodes between top and base, because they will + * disappear from the chain after this operation. + * Note that this assumes that the user is fine with removing all + * nodes (including R/W filters) between top and base. Assuring + * this is the responsibility of the interface (i.e. whoever calls + * commit_start()). + */ + s->base_overlay = bdrv_find_overlay(top, base); + assert(s->base_overlay); + + /* + * The topmost node with + * bdrv_skip_filters(filtered_base) == bdrv_skip_filters(base) + */ + filtered_base = bdrv_cow_bs(s->base_overlay); + assert(bdrv_skip_filters(filtered_base) == bdrv_skip_filters(base)); + + /* + * XXX BLK_PERM_WRITE needs to be allowed so we don't block ourselves + * at s->base (if writes are blocked for a node, they are also blocked + * for its backing file). The other options would be a second filter + * driver above s->base. + */ + iter_shared_perms = BLK_PERM_WRITE_UNCHANGED | BLK_PERM_WRITE; + + for (iter = top; iter != base; iter = bdrv_filter_or_cow_bs(iter)) { + if (iter == filtered_base) { + /* + * From here on, all nodes are filters on the base. This + * allows us to share BLK_PERM_CONSISTENT_READ. + */ + iter_shared_perms |= BLK_PERM_CONSISTENT_READ; + } + + ret = block_job_add_bdrv(&s->common, "intermediate node", iter, 0, + iter_shared_perms, errp); + if (ret < 0) { + goto fail; + } } - s->base = base; - s->top = top; - s->active = bs; + if (bdrv_freeze_backing_chain(commit_top_bs, base, errp) < 0) { + goto fail; + } + s->chain_frozen = true; - s->base_flags = orig_base_flags; - s->orig_overlay_flags = orig_overlay_flags; + ret = block_job_add_bdrv(&s->common, "base", base, 0, BLK_PERM_ALL, errp); + if (ret < 0) { + goto fail; + } + s->base = blk_new(s->common.job.aio_context, + base_perms, + BLK_PERM_CONSISTENT_READ + | BLK_PERM_GRAPH_MOD + | BLK_PERM_WRITE_UNCHANGED); + ret = blk_insert_bs(s->base, base, errp); + if (ret < 0) { + goto fail; + } + blk_set_disable_request_queuing(s->base, true); + s->base_bs = base; + + /* Required permissions are already taken with block_job_add_bdrv() */ + s->top = blk_new(s->common.job.aio_context, 0, BLK_PERM_ALL); + ret = blk_insert_bs(s->top, top, errp); + if (ret < 0) { + goto fail; + } + blk_set_disable_request_queuing(s->top, true); + + s->backing_file_str = g_strdup(backing_file_str); s->on_error = on_error; - s->common.co = qemu_coroutine_create(commit_run); - trace_commit_start(bs, base, top, s, s->common.co, opaque); - qemu_coroutine_enter(s->common.co, s); + trace_commit_start(bs, base, top, s); + job_start(&s->common.job); + return; + +fail: + if (s->chain_frozen) { + bdrv_unfreeze_backing_chain(commit_top_bs, base); + } + if (s->base) { + blk_unref(s->base); + } + if (s->top) { + blk_unref(s->top); + } + if (s->base_read_only) { + bdrv_reopen_set_read_only(base, true, NULL); + } + job_early_fail(&s->common.job); + /* commit_top_bs has to be replaced after deleting the block job, + * otherwise this would fail because of lack of permissions. */ + if (commit_top_bs) { + bdrv_replace_node(commit_top_bs, top, &error_abort); + } +} + + +#define COMMIT_BUF_SIZE (2048 * BDRV_SECTOR_SIZE) + +/* commit COW file into the raw image */ +int bdrv_commit(BlockDriverState *bs) +{ + BlockBackend *src, *backing; + BlockDriverState *backing_file_bs = NULL; + BlockDriverState *commit_top_bs = NULL; + BlockDriver *drv = bs->drv; + AioContext *ctx; + int64_t offset, length, backing_length; + int ro; + int64_t n; + int ret = 0; + uint8_t *buf = NULL; + Error *local_err = NULL; + + if (!drv) + return -ENOMEDIUM; + + backing_file_bs = bdrv_cow_bs(bs); + + if (!backing_file_bs) { + return -ENOTSUP; + } + + if (bdrv_op_is_blocked(bs, BLOCK_OP_TYPE_COMMIT_SOURCE, NULL) || + bdrv_op_is_blocked(backing_file_bs, BLOCK_OP_TYPE_COMMIT_TARGET, NULL)) + { + return -EBUSY; + } + + ro = backing_file_bs->read_only; + + if (ro) { + if (bdrv_reopen_set_read_only(backing_file_bs, false, NULL)) { + return -EACCES; + } + } + + ctx = bdrv_get_aio_context(bs); + /* WRITE_UNCHANGED is required for bdrv_make_empty() */ + src = blk_new(ctx, BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE_UNCHANGED, + BLK_PERM_ALL); + backing = blk_new(ctx, BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL); + + ret = blk_insert_bs(src, bs, &local_err); + if (ret < 0) { + error_report_err(local_err); + goto ro_cleanup; + } + + /* Insert commit_top block node above backing, so we can write to it */ + commit_top_bs = bdrv_new_open_driver(&bdrv_commit_top, NULL, BDRV_O_RDWR, + &local_err); + if (commit_top_bs == NULL) { + error_report_err(local_err); + goto ro_cleanup; + } + + bdrv_set_backing_hd(commit_top_bs, backing_file_bs, &error_abort); + bdrv_set_backing_hd(bs, commit_top_bs, &error_abort); + + ret = blk_insert_bs(backing, backing_file_bs, &local_err); + if (ret < 0) { + error_report_err(local_err); + goto ro_cleanup; + } + + length = blk_getlength(src); + if (length < 0) { + ret = length; + goto ro_cleanup; + } + + backing_length = blk_getlength(backing); + if (backing_length < 0) { + ret = backing_length; + goto ro_cleanup; + } + + /* If our top snapshot is larger than the backing file image, + * grow the backing file image if possible. If not possible, + * we must return an error */ + if (length > backing_length) { + ret = blk_truncate(backing, length, false, PREALLOC_MODE_OFF, 0, + &local_err); + if (ret < 0) { + error_report_err(local_err); + goto ro_cleanup; + } + } + + /* blk_try_blockalign() for src will choose an alignment that works for + * backing as well, so no need to compare the alignment manually. */ + buf = blk_try_blockalign(src, COMMIT_BUF_SIZE); + if (buf == NULL) { + ret = -ENOMEM; + goto ro_cleanup; + } + + for (offset = 0; offset < length; offset += n) { + ret = bdrv_is_allocated(bs, offset, COMMIT_BUF_SIZE, &n); + if (ret < 0) { + goto ro_cleanup; + } + if (ret) { + ret = blk_pread(src, offset, buf, n); + if (ret < 0) { + goto ro_cleanup; + } + + ret = blk_pwrite(backing, offset, buf, n, 0); + if (ret < 0) { + goto ro_cleanup; + } + } + } + + ret = blk_make_empty(src, NULL); + /* Ignore -ENOTSUP */ + if (ret < 0 && ret != -ENOTSUP) { + goto ro_cleanup; + } + + blk_flush(src); + + /* + * Make sure all data we wrote to the backing device is actually + * stable on disk. + */ + blk_flush(backing); + + ret = 0; +ro_cleanup: + qemu_vfree(buf); + + blk_unref(backing); + if (bdrv_cow_bs(bs) != backing_file_bs) { + bdrv_set_backing_hd(bs, backing_file_bs, &error_abort); + } + bdrv_unref(commit_top_bs); + blk_unref(src); + + if (ro) { + /* ignoring error return here */ + bdrv_reopen_set_read_only(backing_file_bs, true, NULL); + } + + return ret; } |