qemu-e2k/include/block/block.h

798 lines
32 KiB
C
Raw Normal View History

#ifndef BLOCK_H
#define BLOCK_H
#include "block/aio.h"
#include "block/aio-wait.h"
#include "qemu/iov.h"
#include "qemu/coroutine.h"
#include "block/accounting.h"
#include "block/dirty-bitmap.h"
#include "block/blockjob.h"
#include "qemu/hbitmap.h"
/* block.c */
typedef struct BlockDriver BlockDriver;
typedef struct BdrvChild BdrvChild;
typedef struct BdrvChildClass BdrvChildClass;
typedef struct BlockDriverInfo {
/* in bytes, 0 if irrelevant */
int cluster_size;
/* offset at which the VM state can be saved (0 if not possible) */
int64_t vm_state_offset;
bool is_dirty;
/*
* True if unallocated blocks read back as zeroes. This is equivalent
* to the LBPRZ flag in the SCSI logical block provisioning page.
*/
bool unallocated_blocks_are_zero;
/*
* True if this block driver only supports compressed writes
*/
bool needs_compressed_writes;
} BlockDriverInfo;
typedef struct BlockFragInfo {
uint64_t allocated_clusters;
uint64_t total_clusters;
uint64_t fragmented_clusters;
uint64_t compressed_clusters;
} BlockFragInfo;
typedef enum {
BDRV_REQ_COPY_ON_READ = 0x1,
BDRV_REQ_ZERO_WRITE = 0x2,
/*
* The BDRV_REQ_MAY_UNMAP flag is used in write_zeroes requests to indicate
* that the block driver should unmap (discard) blocks if it is guaranteed
* that the result will read back as zeroes. The flag is only passed to the
* driver if the block device is opened with BDRV_O_UNMAP.
*/
BDRV_REQ_MAY_UNMAP = 0x4,
BDRV_REQ_FUA = 0x10,
BDRV_REQ_WRITE_COMPRESSED = 0x20,
/* Signifies that this write request will not change the visible disk
* content. */
BDRV_REQ_WRITE_UNCHANGED = 0x40,
/*
* BDRV_REQ_SERIALISING forces request serialisation for writes.
* It is used to ensure that writes to the backing file of a backup process
* target cannot race with a read of the backup target that defers to the
* backing file.
*
* Note, that BDRV_REQ_SERIALISING is _not_ opposite in meaning to
* BDRV_REQ_NO_SERIALISING. A more descriptive name for the latter might be
* _DO_NOT_WAIT_FOR_SERIALISING, except that is too long.
*/
BDRV_REQ_SERIALISING = 0x80,
/* Execute the request only if the operation can be offloaded or otherwise
* be executed efficiently, but return an error instead of using a slow
* fallback. */
BDRV_REQ_NO_FALLBACK = 0x100,
/*
* BDRV_REQ_PREFETCH may be used only together with BDRV_REQ_COPY_ON_READ
* on read request and means that caller doesn't really need data to be
* written to qiov parameter which may be NULL.
*/
BDRV_REQ_PREFETCH = 0x200,
/* Mask of valid flags */
BDRV_REQ_MASK = 0x3ff,
} BdrvRequestFlags;
typedef struct BlockSizes {
uint32_t phys;
uint32_t log;
} BlockSizes;
typedef struct HDGeometry {
uint32_t heads;
uint32_t sectors;
uint32_t cylinders;
} HDGeometry;
#define BDRV_O_RDWR 0x0002
#define BDRV_O_RESIZE 0x0004 /* request permission for resizing the node */
#define BDRV_O_SNAPSHOT 0x0008 /* open the file read only and save writes in a snapshot */
#define BDRV_O_TEMPORARY 0x0010 /* delete the file after use */
#define BDRV_O_NOCACHE 0x0020 /* do not use the host page cache */
#define BDRV_O_NATIVE_AIO 0x0080 /* use native AIO instead of the thread pool */
#define BDRV_O_NO_BACKING 0x0100 /* don't open the backing file */
#define BDRV_O_NO_FLUSH 0x0200 /* disable flushing on this disk */
#define BDRV_O_COPY_ON_READ 0x0400 /* copy read backing sectors into image */
#define BDRV_O_INACTIVE 0x0800 /* consistency hint for migration handoff */
#define BDRV_O_CHECK 0x1000 /* open solely for consistency check */
#define BDRV_O_ALLOW_RDWR 0x2000 /* allow reopen to change from r/o to r/w */
#define BDRV_O_UNMAP 0x4000 /* execute guest UNMAP/TRIM operations */
#define BDRV_O_PROTOCOL 0x8000 /* if no block driver is explicitly given:
select an appropriate protocol driver,
ignoring the format layer */
#define BDRV_O_NO_IO 0x10000 /* don't initialize for I/O */
block: Add auto-read-only option If a management application builds the block graph node by node, the protocol layer doesn't inherit its read-only option from the format layer any more, so it must be set explicitly. Backing files should work on read-only storage, but at the same time, a block job like commit should be able to reopen them read-write if they are on read-write storage. However, without option inheritance, reopen only changes the read-only option for the root node (typically the format layer), but not the protocol layer, so reopening fails (the format layer wants to get write permissions, but the protocol layer is still read-only). A simple workaround for the problem in the management tool would be to open the protocol layer always read-write and to make only the format layer read-only for backing files. However, sometimes the file is actually stored on read-only storage and we don't know whether the image can be opened read-write (for example, for NBD it depends on the server we're trying to connect to). This adds an option that makes QEMU try to open the image read-write, but allows it to degrade to a read-only mode without returning an error. The documentation for this option is consciously phrased in a way that allows QEMU to switch to a better model eventually: Instead of trying when the image is first opened, making the read-only flag dynamic and changing it automatically whenever the first BLK_PERM_WRITE user is attached or the last one is detached would be much more useful behaviour. Unfortunately, this more useful behaviour is also a lot harder to implement, and libvirt needs a solution now before it can switch to -blockdev, so let's start with this easier approach for now. Instead of adding a new auto-read-only option, turning the existing read-only into an enum (with a bool alternate for compatibility) was considered, but it complicated the implementation to the point that it didn't seem to be worth it. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
2018-10-05 18:57:40 +02:00
#define BDRV_O_AUTO_RDONLY 0x20000 /* degrade to read-only if opening read-write fails */
#define BDRV_O_IO_URING 0x40000 /* use io_uring instead of the thread pool */
#define BDRV_O_CACHE_MASK (BDRV_O_NOCACHE | BDRV_O_NO_FLUSH)
/* Option names of options parsed by the block layer */
#define BDRV_OPT_CACHE_WB "cache.writeback"
#define BDRV_OPT_CACHE_DIRECT "cache.direct"
#define BDRV_OPT_CACHE_NO_FLUSH "cache.no-flush"
#define BDRV_OPT_READ_ONLY "read-only"
block: Add auto-read-only option If a management application builds the block graph node by node, the protocol layer doesn't inherit its read-only option from the format layer any more, so it must be set explicitly. Backing files should work on read-only storage, but at the same time, a block job like commit should be able to reopen them read-write if they are on read-write storage. However, without option inheritance, reopen only changes the read-only option for the root node (typically the format layer), but not the protocol layer, so reopening fails (the format layer wants to get write permissions, but the protocol layer is still read-only). A simple workaround for the problem in the management tool would be to open the protocol layer always read-write and to make only the format layer read-only for backing files. However, sometimes the file is actually stored on read-only storage and we don't know whether the image can be opened read-write (for example, for NBD it depends on the server we're trying to connect to). This adds an option that makes QEMU try to open the image read-write, but allows it to degrade to a read-only mode without returning an error. The documentation for this option is consciously phrased in a way that allows QEMU to switch to a better model eventually: Instead of trying when the image is first opened, making the read-only flag dynamic and changing it automatically whenever the first BLK_PERM_WRITE user is attached or the last one is detached would be much more useful behaviour. Unfortunately, this more useful behaviour is also a lot harder to implement, and libvirt needs a solution now before it can switch to -blockdev, so let's start with this easier approach for now. Instead of adding a new auto-read-only option, turning the existing read-only into an enum (with a bool alternate for compatibility) was considered, but it complicated the implementation to the point that it didn't seem to be worth it. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
2018-10-05 18:57:40 +02:00
#define BDRV_OPT_AUTO_READ_ONLY "auto-read-only"
#define BDRV_OPT_DISCARD "discard"
#define BDRV_OPT_FORCE_SHARE "force-share"
#define BDRV_SECTOR_BITS 9
#define BDRV_SECTOR_SIZE (1ULL << BDRV_SECTOR_BITS)
#define BDRV_REQUEST_MAX_SECTORS MIN(SIZE_MAX >> BDRV_SECTOR_BITS, \
INT_MAX >> BDRV_SECTOR_BITS)
#define BDRV_REQUEST_MAX_BYTES (BDRV_REQUEST_MAX_SECTORS << BDRV_SECTOR_BITS)
/*
block: Convert bdrv_get_block_status() to bytes We are gradually moving away from sector-based interfaces, towards byte-based. In the common case, allocation is unlikely to ever use values that are not naturally sector-aligned, but it is possible that byte-based values will let us be more precise about allocation at the end of an unaligned file that can do byte-based access. Changing the name of the function from bdrv_get_block_status() to bdrv_block_status() ensures that the compiler enforces that all callers are updated. For now, the io.c layer still assert()s that all callers are sector-aligned, but that can be relaxed when a later patch implements byte-based block status in the drivers. There was an inherent limitation in returning the offset via the return value: we only have room for BDRV_BLOCK_OFFSET_MASK bits, which means an offset can only be mapped for sector-aligned queries (or, if we declare that non-aligned input is at the same relative position modulo 512 of the answer), so the new interface also changes things to return the offset via output through a parameter by reference rather than mashed into the return value. We'll have some glue code that munges between the two styles until we finish converting all uses. For the most part this patch is just the addition of scaling at the callers followed by inverse scaling at bdrv_block_status(), coupled with the tweak in calling convention. But some code, particularly bdrv_is_allocated(), gets a lot simpler because it no longer has to mess with sectors. For ease of review, bdrv_get_block_status_above() will be tackled separately. Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2017-10-12 05:47:03 +02:00
* Allocation status flags for bdrv_block_status() and friends.
*
* Public flags:
* BDRV_BLOCK_DATA: allocation for data at offset is tied to this layer
* BDRV_BLOCK_ZERO: offset reads as zero
* BDRV_BLOCK_OFFSET_VALID: an associated offset exists for accessing raw data
* BDRV_BLOCK_ALLOCATED: the content of the block is determined by this
block: Add .bdrv_co_block_status() callback We are gradually moving away from sector-based interfaces, towards byte-based. Now that the block layer exposes byte-based allocation, it's time to tackle the drivers. Add a new callback that operates on as small as byte boundaries. Subsequent patches will then update individual drivers, then finally remove .bdrv_co_get_block_status(). The new code also passes through the 'want_zero' hint, which will allow subsequent patches to further optimize callers that only care about how much of the image is allocated (want_zero is false), rather than full details about runs of zeroes and which offsets the allocation actually maps to (want_zero is true). As part of this effort, fix another part of the documentation: the claim in commit 4c41cb4 that BDRV_BLOCK_ALLOCATED is short for 'DATA || ZERO' is a lie at the block layer (see commit e88ae2264), even though it is how the bit is computed from the driver layer. After all, there are intentionally cases where we return ZERO but not ALLOCATED at the block layer, when we know that a read sees zero because the backing file is too short. Note that the driver interface is thus slightly different than the public interface with regards to which bits will be set, and what guarantees are provided on input. We also add an assertion that any driver using the new callback will make progress (the only time pnum will be 0 is if the block layer already handled an out-of-bounds request, or if there is an error); the old driver interface did not provide this guarantee, which could lead to some inf-loops in drastic corner-case failures. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2018-02-13 21:26:41 +01:00
* layer rather than any backing, set by block layer
* BDRV_BLOCK_EOF: the returned pnum covers through end of file for this
* layer, set by block layer
*
block: avoid recursive block_status call if possible drv_co_block_status digs bs->file for additional, more accurate search for hole inside region, reported as DATA by bs since 5daa74a6ebc. This accuracy is not free: assume we have qcow2 disk. Actually, qcow2 knows, where are holes and where is data. But every block_status request calls lseek additionally. Assume a big disk, full of data, in any iterative copying block job (or img convert) we'll call lseek(HOLE) on every iteration, and each of these lseeks will have to iterate through all metadata up to the end of file. It's obviously ineffective behavior. And for many scenarios we don't need this lseek at all. However, lseek is needed when we have metadata-preallocated image. So, let's detect metadata-preallocation case and don't dig qcow2's protocol file in other cases. The idea is to compare allocation size in POV of filesystem with allocations size in POV of Qcow2 (by refcounts). If allocation in fs is significantly lower, consider it as metadata-preallocation case. 102 iotest changed, as our detector can't detect shrinked file as metadata-preallocation, which don't seem to be wrong, as with metadata preallocation we always have valid file length. Two other iotests have a slight change in their QMP output sequence: Active 'block-commit' returns earlier because the job coroutine yields earlier on a blocking operation. This operation is loading the refcount blocks in qcow2_detect_metadata_preallocation(). Suggested-by: Denis V. Lunev <den@openvz.org> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2019-04-08 18:26:17 +02:00
* Internal flags:
* BDRV_BLOCK_RAW: for use by passthrough drivers, such as raw, to request
* that the block layer recompute the answer from the returned
* BDS; must be accompanied by just BDRV_BLOCK_OFFSET_VALID.
block: avoid recursive block_status call if possible drv_co_block_status digs bs->file for additional, more accurate search for hole inside region, reported as DATA by bs since 5daa74a6ebc. This accuracy is not free: assume we have qcow2 disk. Actually, qcow2 knows, where are holes and where is data. But every block_status request calls lseek additionally. Assume a big disk, full of data, in any iterative copying block job (or img convert) we'll call lseek(HOLE) on every iteration, and each of these lseeks will have to iterate through all metadata up to the end of file. It's obviously ineffective behavior. And for many scenarios we don't need this lseek at all. However, lseek is needed when we have metadata-preallocated image. So, let's detect metadata-preallocation case and don't dig qcow2's protocol file in other cases. The idea is to compare allocation size in POV of filesystem with allocations size in POV of Qcow2 (by refcounts). If allocation in fs is significantly lower, consider it as metadata-preallocation case. 102 iotest changed, as our detector can't detect shrinked file as metadata-preallocation, which don't seem to be wrong, as with metadata preallocation we always have valid file length. Two other iotests have a slight change in their QMP output sequence: Active 'block-commit' returns earlier because the job coroutine yields earlier on a blocking operation. This operation is loading the refcount blocks in qcow2_detect_metadata_preallocation(). Suggested-by: Denis V. Lunev <den@openvz.org> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2019-04-08 18:26:17 +02:00
* BDRV_BLOCK_RECURSE: request that the block layer will recursively search for
* zeroes in file child of current block node inside
* returned region. Only valid together with both
* BDRV_BLOCK_DATA and BDRV_BLOCK_OFFSET_VALID. Should not
* appear with BDRV_BLOCK_ZERO.
*
block: Add .bdrv_co_block_status() callback We are gradually moving away from sector-based interfaces, towards byte-based. Now that the block layer exposes byte-based allocation, it's time to tackle the drivers. Add a new callback that operates on as small as byte boundaries. Subsequent patches will then update individual drivers, then finally remove .bdrv_co_get_block_status(). The new code also passes through the 'want_zero' hint, which will allow subsequent patches to further optimize callers that only care about how much of the image is allocated (want_zero is false), rather than full details about runs of zeroes and which offsets the allocation actually maps to (want_zero is true). As part of this effort, fix another part of the documentation: the claim in commit 4c41cb4 that BDRV_BLOCK_ALLOCATED is short for 'DATA || ZERO' is a lie at the block layer (see commit e88ae2264), even though it is how the bit is computed from the driver layer. After all, there are intentionally cases where we return ZERO but not ALLOCATED at the block layer, when we know that a read sees zero because the backing file is too short. Note that the driver interface is thus slightly different than the public interface with regards to which bits will be set, and what guarantees are provided on input. We also add an assertion that any driver using the new callback will make progress (the only time pnum will be 0 is if the block layer already handled an out-of-bounds request, or if there is an error); the old driver interface did not provide this guarantee, which could lead to some inf-loops in drastic corner-case failures. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2018-02-13 21:26:41 +01:00
* If BDRV_BLOCK_OFFSET_VALID is set, the map parameter represents the
* host offset within the returned BDS that is allocated for the
* corresponding raw guest data. However, whether that offset
* actually contains data also depends on BDRV_BLOCK_DATA, as follows:
*
* DATA ZERO OFFSET_VALID
* t t t sectors read as zero, returned file is zero at offset
* t f t sectors read as valid from file at offset
* f t t sectors preallocated, read as zero, returned file not
* necessarily zero at offset
* f f t sectors preallocated but read from backing_hd,
* returned file contains garbage at offset
* t t f sectors preallocated, read as zero, unknown offset
* t f f sectors read from unknown file or offset
* f t f not allocated or unknown offset, read as zero
* f f f not allocated or unknown offset, read from backing_hd
*/
#define BDRV_BLOCK_DATA 0x01
#define BDRV_BLOCK_ZERO 0x02
#define BDRV_BLOCK_OFFSET_VALID 0x04
#define BDRV_BLOCK_RAW 0x08
#define BDRV_BLOCK_ALLOCATED 0x10
#define BDRV_BLOCK_EOF 0x20
block: avoid recursive block_status call if possible drv_co_block_status digs bs->file for additional, more accurate search for hole inside region, reported as DATA by bs since 5daa74a6ebc. This accuracy is not free: assume we have qcow2 disk. Actually, qcow2 knows, where are holes and where is data. But every block_status request calls lseek additionally. Assume a big disk, full of data, in any iterative copying block job (or img convert) we'll call lseek(HOLE) on every iteration, and each of these lseeks will have to iterate through all metadata up to the end of file. It's obviously ineffective behavior. And for many scenarios we don't need this lseek at all. However, lseek is needed when we have metadata-preallocated image. So, let's detect metadata-preallocation case and don't dig qcow2's protocol file in other cases. The idea is to compare allocation size in POV of filesystem with allocations size in POV of Qcow2 (by refcounts). If allocation in fs is significantly lower, consider it as metadata-preallocation case. 102 iotest changed, as our detector can't detect shrinked file as metadata-preallocation, which don't seem to be wrong, as with metadata preallocation we always have valid file length. Two other iotests have a slight change in their QMP output sequence: Active 'block-commit' returns earlier because the job coroutine yields earlier on a blocking operation. This operation is loading the refcount blocks in qcow2_detect_metadata_preallocation(). Suggested-by: Denis V. Lunev <den@openvz.org> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2019-04-08 18:26:17 +02:00
#define BDRV_BLOCK_RECURSE 0x40
typedef QTAILQ_HEAD(BlockReopenQueue, BlockReopenQueueEntry) BlockReopenQueue;
typedef struct BDRVReopenState {
BlockDriverState *bs;
int flags;
BlockdevDetectZeroesOptions detect_zeroes;
block: Handle child references in bdrv_reopen_queue() Children in QMP are specified with BlockdevRef / BlockdevRefOrNull, which can contain a set of child options, a child reference, or NULL. In optional attributes like "backing" it can also be missing. Only the first case (set of child options) is being handled properly by bdrv_reopen_queue(). This patch deals with all the others. Here's how these cases should be handled when bdrv_reopen_queue() is deciding what to do with each child of a BlockDriverState: 1) Set of child options: if the child was implicitly created (i.e inherits_from points to the parent) then the options are removed from the parent's options QDict and are passed to the child with a recursive bdrv_reopen_queue() call. This case was already working fine. 2) Child reference: there's two possibilites here. 2a) Reference to the current child: if the child was implicitly created then it is put in the reopen queue, keeping its current set of options (since this was a child reference there was no way to specify a different set of options). If the child is not implicit then it keeps its current set of options but it is not reopened (and therefore does not inherit any new option from the parent). 2b) Reference to a different BDS: the current child is not put in the reopen queue at all. Passing a reference to a different BDS can be used to replace a child, although at the moment no driver implements this, so it results in an error. In any case, the current child is not going to be reopened (and might in fact disappear if it's replaced) 3) NULL: This is similar to (2b). Although no driver allows this yet it can be used to detach the current child so it should not be put in the reopen queue. 4) Missing option: at the moment "backing" is the only case where this can happen. With "blockdev-add", leaving "backing" out means that the default backing file is opened. We don't want to open a new image during reopen, so we require that "backing" is always present. We'll relax this requirement a bit in the next patch. If keep_old_opts is true and "backing" is missing then this behaves like 2a (the current child is reopened). Signed-off-by: Alberto Garcia <berto@igalia.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2019-03-12 17:48:45 +01:00
bool backing_missing;
bool replace_backing_bs; /* new_backing_bs is ignored if this is false */
BlockDriverState *new_backing_bs; /* If NULL then detach the current bs */
uint64_t perm, shared_perm;
QDict *options;
QDict *explicit_options;
void *opaque;
} BDRVReopenState;
/*
* Block operation types
*/
typedef enum BlockOpType {
BLOCK_OP_TYPE_BACKUP_SOURCE,
BLOCK_OP_TYPE_BACKUP_TARGET,
BLOCK_OP_TYPE_CHANGE,
BLOCK_OP_TYPE_COMMIT_SOURCE,
BLOCK_OP_TYPE_COMMIT_TARGET,
BLOCK_OP_TYPE_DATAPLANE,
BLOCK_OP_TYPE_DRIVE_DEL,
BLOCK_OP_TYPE_EJECT,
BLOCK_OP_TYPE_EXTERNAL_SNAPSHOT,
BLOCK_OP_TYPE_INTERNAL_SNAPSHOT,
BLOCK_OP_TYPE_INTERNAL_SNAPSHOT_DELETE,
BLOCK_OP_TYPE_MIRROR_SOURCE,
BLOCK_OP_TYPE_MIRROR_TARGET,
BLOCK_OP_TYPE_RESIZE,
BLOCK_OP_TYPE_STREAM,
BLOCK_OP_TYPE_REPLACE,
BLOCK_OP_TYPE_MAX,
} BlockOpType;
/* Block node permission constants */
enum {
/**
* A user that has the "permission" of consistent reads is guaranteed that
* their view of the contents of the block device is complete and
* self-consistent, representing the contents of a disk at a specific
* point.
*
* For most block devices (including their backing files) this is true, but
* the property cannot be maintained in a few situations like for
* intermediate nodes of a commit block job.
*/
BLK_PERM_CONSISTENT_READ = 0x01,
/** This permission is required to change the visible disk contents. */
BLK_PERM_WRITE = 0x02,
/**
* This permission (which is weaker than BLK_PERM_WRITE) is both enough and
* required for writes to the block node when the caller promises that
* the visible disk content doesn't change.
*
* As the BLK_PERM_WRITE permission is strictly stronger, either is
* sufficient to perform an unchanging write.
*/
BLK_PERM_WRITE_UNCHANGED = 0x04,
/** This permission is required to change the size of a block node. */
BLK_PERM_RESIZE = 0x08,
/**
* This permission is required to change the node that this BdrvChild
* points to.
*/
BLK_PERM_GRAPH_MOD = 0x10,
BLK_PERM_ALL = 0x1f,
DEFAULT_PERM_PASSTHROUGH = BLK_PERM_CONSISTENT_READ
| BLK_PERM_WRITE
| BLK_PERM_WRITE_UNCHANGED
| BLK_PERM_RESIZE,
DEFAULT_PERM_UNCHANGED = BLK_PERM_ALL & ~DEFAULT_PERM_PASSTHROUGH,
};
/*
* Flags that parent nodes assign to child nodes to specify what kind of
* role(s) they take.
*
* At least one of DATA, METADATA, FILTERED, or COW must be set for
* every child.
*/
enum BdrvChildRoleBits {
/*
* This child stores data.
* Any node may have an arbitrary number of such children.
*/
BDRV_CHILD_DATA = (1 << 0),
/*
* This child stores metadata.
* Any node may have an arbitrary number of metadata-storing
* children.
*/
BDRV_CHILD_METADATA = (1 << 1),
/*
* A child that always presents exactly the same visible data as
* the parent, e.g. by virtue of the parent forwarding all reads
* and writes.
* This flag is mutually exclusive with DATA, METADATA, and COW.
* Any node may have at most one filtered child at a time.
*/
BDRV_CHILD_FILTERED = (1 << 2),
/*
* Child from which to read all data that isnt allocated in the
* parent (i.e., the backing child); such data is copied to the
* parent through COW (and optionally COR).
* This field is mutually exclusive with DATA, METADATA, and
* FILTERED.
* Any node may have at most one such backing child at a time.
*/
BDRV_CHILD_COW = (1 << 3),
/*
* The primary child. For most drivers, this is the child whose
* filename applies best to the parent node.
* Any node may have at most one primary child at a time.
*/
BDRV_CHILD_PRIMARY = (1 << 4),
/* Useful combination of flags */
BDRV_CHILD_IMAGE = BDRV_CHILD_DATA
| BDRV_CHILD_METADATA
| BDRV_CHILD_PRIMARY,
};
/* Mask of BdrvChildRoleBits values */
typedef unsigned int BdrvChildRole;
char *bdrv_perm_names(uint64_t perm);
uint64_t bdrv_qapi_perm_to_blk_perm(BlockPermission qapi_perm);
/* disk I/O throttling */
void bdrv_init(void);
void bdrv_init_with_whitelist(void);
2016-03-21 15:11:48 +01:00
bool bdrv_uses_whitelist(void);
int bdrv_is_whitelisted(BlockDriver *drv, bool read_only);
BlockDriver *bdrv_find_protocol(const char *filename,
bool allow_protocol_prefix,
Error **errp);
BlockDriver *bdrv_find_format(const char *format_name);
int bdrv_create(BlockDriver *drv, const char* filename,
QemuOpts *opts, Error **errp);
int bdrv_create_file(const char *filename, QemuOpts *opts, Error **errp);
BlockDriverState *bdrv_new(void);
void bdrv_append(BlockDriverState *bs_new, BlockDriverState *bs_top,
Error **errp);
void bdrv_replace_node(BlockDriverState *from, BlockDriverState *to,
Error **errp);
int bdrv_parse_aio(const char *mode, int *flags);
int bdrv_parse_cache_mode(const char *mode, int *flags, bool *writethrough);
int bdrv_parse_discard_flags(const char *mode, int *flags);
BdrvChild *bdrv_open_child(const char *filename,
QDict *options, const char *bdref_key,
BlockDriverState* parent,
const BdrvChildClass *child_class,
BdrvChildRole child_role,
bool allow_none, Error **errp);
BlockDriverState *bdrv_open_blockdev_ref(BlockdevRef *ref, Error **errp);
void bdrv_set_backing_hd(BlockDriverState *bs, BlockDriverState *backing_hd,
Error **errp);
int bdrv_open_backing_file(BlockDriverState *bs, QDict *parent_options,
const char *bdref_key, Error **errp);
BlockDriverState *bdrv_open(const char *filename, const char *reference,
QDict *options, int flags, Error **errp);
BlockDriverState *bdrv_new_open_driver(BlockDriver *drv, const char *node_name,
int flags, Error **errp);
BlockReopenQueue *bdrv_reopen_queue(BlockReopenQueue *bs_queue,
BlockDriverState *bs, QDict *options,
bool keep_old_opts);
int bdrv_reopen_multiple(BlockReopenQueue *bs_queue, Error **errp);
int bdrv_reopen_set_read_only(BlockDriverState *bs, bool read_only,
Error **errp);
int bdrv_reopen_prepare(BDRVReopenState *reopen_state,
BlockReopenQueue *queue, Error **errp);
void bdrv_reopen_commit(BDRVReopenState *reopen_state);
void bdrv_reopen_abort(BDRVReopenState *reopen_state);
int bdrv_pwrite_zeroes(BdrvChild *child, int64_t offset,
int bytes, BdrvRequestFlags flags);
int bdrv_make_zero(BdrvChild *child, BdrvRequestFlags flags);
int bdrv_pread(BdrvChild *child, int64_t offset, void *buf, int bytes);
int bdrv_preadv(BdrvChild *child, int64_t offset, QEMUIOVector *qiov);
int bdrv_pwrite(BdrvChild *child, int64_t offset, const void *buf, int bytes);
int bdrv_pwritev(BdrvChild *child, int64_t offset, QEMUIOVector *qiov);
int bdrv_pwrite_sync(BdrvChild *child, int64_t offset,
const void *buf, int count);
/*
* Efficiently zero a region of the disk image. Note that this is a regular
* I/O request like read or write and should have a reasonable size. This
* function is not suitable for zeroing the entire image in a single request
* because it may allocate memory for the entire region.
*/
int coroutine_fn bdrv_co_pwrite_zeroes(BdrvChild *child, int64_t offset,
int bytes, BdrvRequestFlags flags);
BlockDriverState *bdrv_find_backing_image(BlockDriverState *bs,
const char *backing_file);
void bdrv_refresh_filename(BlockDriverState *bs);
int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact,
PreallocMode prealloc, BdrvRequestFlags flags,
Error **errp);
int bdrv_truncate(BdrvChild *child, int64_t offset, bool exact,
PreallocMode prealloc, BdrvRequestFlags flags, Error **errp);
int64_t bdrv_nb_sectors(BlockDriverState *bs);
int64_t bdrv_getlength(BlockDriverState *bs);
int64_t bdrv_get_allocated_file_size(BlockDriverState *bs);
BlockMeasureInfo *bdrv_measure(BlockDriver *drv, QemuOpts *opts,
BlockDriverState *in_bs, Error **errp);
void bdrv_get_geometry(BlockDriverState *bs, uint64_t *nb_sectors_ptr);
void bdrv_refresh_limits(BlockDriverState *bs, Error **errp);
int bdrv_commit(BlockDriverState *bs);
int bdrv_make_empty(BdrvChild *c, Error **errp);
int bdrv_change_backing_file(BlockDriverState *bs,
const char *backing_file, const char *backing_fmt);
void bdrv_register(BlockDriver *bdrv);
int bdrv_drop_intermediate(BlockDriverState *top, BlockDriverState *base,
const char *backing_file_str);
BlockDriverState *bdrv_find_overlay(BlockDriverState *active,
BlockDriverState *bs);
BlockDriverState *bdrv_find_base(BlockDriverState *bs);
bool bdrv_is_backing_chain_frozen(BlockDriverState *bs, BlockDriverState *base,
Error **errp);
int bdrv_freeze_backing_chain(BlockDriverState *bs, BlockDriverState *base,
Error **errp);
void bdrv_unfreeze_backing_chain(BlockDriverState *bs, BlockDriverState *base);
int coroutine_fn bdrv_co_delete_file(BlockDriverState *bs, Error **errp);
typedef struct BdrvCheckResult {
int corruptions;
int leaks;
int check_errors;
int corruptions_fixed;
int leaks_fixed;
int64_t image_end_offset;
BlockFragInfo bfi;
} BdrvCheckResult;
typedef enum {
BDRV_FIX_LEAKS = 1,
BDRV_FIX_ERRORS = 2,
} BdrvCheckMode;
int bdrv_check(BlockDriverState *bs, BdrvCheckResult *res, BdrvCheckMode fix);
/* The units of offset and total_work_size may be chosen arbitrarily by the
* block driver; total_work_size may change during the course of the amendment
* operation */
typedef void BlockDriverAmendStatusCB(BlockDriverState *bs, int64_t offset,
int64_t total_work_size, void *opaque);
int bdrv_amend_options(BlockDriverState *bs_new, QemuOpts *opts,
BlockDriverAmendStatusCB *status_cb, void *cb_opaque,
Error **errp);
/* check if a named node can be replaced when doing drive-mirror */
BlockDriverState *check_to_replace_node(BlockDriverState *parent_bs,
const char *node_name, Error **errp);
/* async block I/O */
void bdrv_aio_cancel(BlockAIOCB *acb);
void bdrv_aio_cancel_async(BlockAIOCB *acb);
/* sg packet commands */
int bdrv_co_ioctl(BlockDriverState *bs, int req, void *buf);
/* Invalidate any cached metadata used by image formats */
void bdrv_invalidate_cache(BlockDriverState *bs, Error **errp);
void bdrv_invalidate_cache_all(Error **errp);
block: Inactivate BDS when migration completes So far, live migration with shared storage meant that the image is in a not-really-ready don't-touch-me state on the destination while the source is still actively using it, but after completing the migration, the image was fully opened on both sides. This is bad. This patch adds a block driver callback to inactivate images on the source before completing the migration. Inactivation means that it goes to a state as if it was just live migrated to the qemu instance on the source (i.e. BDRV_O_INACTIVE is set). You're then supposed to continue either on the source or on the destination, which takes ownership of the image. A typical migration looks like this now with respect to disk images: 1. Destination qemu is started, the image is opened with BDRV_O_INACTIVE. The image is fully opened on the source. 2. Migration is about to complete. The source flushes the image and inactivates it. Now both sides have the image opened with BDRV_O_INACTIVE and are expecting the other side to still modify it. 3. One side (the destination on success) continues and calls bdrv_invalidate_all() in order to take ownership of the image again. This removes BDRV_O_INACTIVE on the resuming side; the flag remains set on the other side. This ensures that the same image isn't written to by both instances (unless both are resumed, but then you get what you deserve). This is important because .bdrv_close for non-BDRV_O_INACTIVE images could write to the image file, which is definitely forbidden while another host is using the image. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com>
2015-12-22 14:07:08 +01:00
int bdrv_inactivate_all(void);
/* Ensure contents are flushed to disk. */
int bdrv_flush(BlockDriverState *bs);
int coroutine_fn bdrv_co_flush(BlockDriverState *bs);
int bdrv_flush_all(void);
void bdrv_close_all(void);
void bdrv_drain(BlockDriverState *bs);
block: Fix bdrv_drain in coroutine Using the nested aio_poll() in coroutine is a bad idea. This patch replaces the aio_poll loop in bdrv_drain with a BH, if called in coroutine. For example, the bdrv_drain() in mirror.c can hang when a guest issued request is pending on it in qemu_co_mutex_lock(). Mirror coroutine in this case has just finished a request, and the block job is about to complete. It calls bdrv_drain() which waits for the other coroutine to complete. The other coroutine is a scsi-disk request. The deadlock happens when the latter is in turn pending on the former to yield/terminate, in qemu_co_mutex_lock(). The state flow is as below (assuming a qcow2 image): mirror coroutine scsi-disk coroutine ------------------------------------------------------------- do last write qcow2:qemu_co_mutex_lock() ... scsi disk read tracked request begin qcow2:qemu_co_mutex_lock.enter qcow2:qemu_co_mutex_unlock() bdrv_drain while (has tracked request) aio_poll() In the scsi-disk coroutine, the qemu_co_mutex_lock() will never return because the mirror coroutine is blocked in the aio_poll(blocking=true). With this patch, the added qemu_coroutine_yield() allows the scsi-disk coroutine to make progress as expected: mirror coroutine scsi-disk coroutine ------------------------------------------------------------- do last write qcow2:qemu_co_mutex_lock() ... scsi disk read tracked request begin qcow2:qemu_co_mutex_lock.enter qcow2:qemu_co_mutex_unlock() bdrv_drain.enter > schedule BH > qemu_coroutine_yield() > qcow2:qemu_co_mutex_lock.return > ... tracked request end ... (resumed from BH callback) bdrv_drain.return ... Reported-by: Laurent Vivier <lvivier@redhat.com> Signed-off-by: Fam Zheng <famz@redhat.com> Message-id: 1459855253-5378-2-git-send-email-famz@redhat.com Suggested-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Fam Zheng <famz@redhat.com> Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2016-04-05 13:20:52 +02:00
void coroutine_fn bdrv_co_drain(BlockDriverState *bs);
void bdrv_drain_all_begin(void);
void bdrv_drain_all_end(void);
void bdrv_drain_all(void);
#define BDRV_POLL_WHILE(bs, cond) ({ \
BlockDriverState *bs_ = (bs); \
AIO_WAIT_WHILE(bdrv_get_aio_context(bs_), \
cond); })
int bdrv_pdiscard(BdrvChild *child, int64_t offset, int64_t bytes);
int bdrv_co_pdiscard(BdrvChild *child, int64_t offset, int64_t bytes);
int bdrv_has_zero_init_1(BlockDriverState *bs);
int bdrv_has_zero_init(BlockDriverState *bs);
bool bdrv_unallocated_blocks_are_zero(BlockDriverState *bs);
bool bdrv_can_write_zeroes_with_unmap(BlockDriverState *bs);
block: Convert bdrv_get_block_status() to bytes We are gradually moving away from sector-based interfaces, towards byte-based. In the common case, allocation is unlikely to ever use values that are not naturally sector-aligned, but it is possible that byte-based values will let us be more precise about allocation at the end of an unaligned file that can do byte-based access. Changing the name of the function from bdrv_get_block_status() to bdrv_block_status() ensures that the compiler enforces that all callers are updated. For now, the io.c layer still assert()s that all callers are sector-aligned, but that can be relaxed when a later patch implements byte-based block status in the drivers. There was an inherent limitation in returning the offset via the return value: we only have room for BDRV_BLOCK_OFFSET_MASK bits, which means an offset can only be mapped for sector-aligned queries (or, if we declare that non-aligned input is at the same relative position modulo 512 of the answer), so the new interface also changes things to return the offset via output through a parameter by reference rather than mashed into the return value. We'll have some glue code that munges between the two styles until we finish converting all uses. For the most part this patch is just the addition of scaling at the callers followed by inverse scaling at bdrv_block_status(), coupled with the tweak in calling convention. But some code, particularly bdrv_is_allocated(), gets a lot simpler because it no longer has to mess with sectors. For ease of review, bdrv_get_block_status_above() will be tackled separately. Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2017-10-12 05:47:03 +02:00
int bdrv_block_status(BlockDriverState *bs, int64_t offset,
int64_t bytes, int64_t *pnum, int64_t *map,
BlockDriverState **file);
block: Convert bdrv_get_block_status_above() to bytes We are gradually moving away from sector-based interfaces, towards byte-based. In the common case, allocation is unlikely to ever use values that are not naturally sector-aligned, but it is possible that byte-based values will let us be more precise about allocation at the end of an unaligned file that can do byte-based access. Changing the name of the function from bdrv_get_block_status_above() to bdrv_block_status_above() ensures that the compiler enforces that all callers are updated. Likewise, since it a byte interface allows an offset mapping that might not be sector aligned, split the mapping out of the return value and into a pass-by-reference parameter. For now, the io.c layer still assert()s that all uses are sector-aligned, but that can be relaxed when a later patch implements byte-based block status in the drivers. For the most part this patch is just the addition of scaling at the callers followed by inverse scaling at bdrv_block_status(), plus updates for the new split return interface. But some code, particularly bdrv_block_status(), gets a lot simpler because it no longer has to mess with sectors. Likewise, mirror code no longer computes s->granularity >> BDRV_SECTOR_BITS, and can therefore drop an assertion about alignment because the loop no longer depends on alignment (never mind that we don't really have a driver that reports sub-sector alignments, so it's not really possible to test the effect of sub-sector mirroring). Fix a neighboring assertion to use is_power_of_2 while there. For ease of review, bdrv_get_block_status() was tackled separately. Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2017-10-12 05:47:08 +02:00
int bdrv_block_status_above(BlockDriverState *bs, BlockDriverState *base,
int64_t offset, int64_t bytes, int64_t *pnum,
int64_t *map, BlockDriverState **file);
block: Make bdrv_is_allocated() byte-based We are gradually moving away from sector-based interfaces, towards byte-based. In the common case, allocation is unlikely to ever use values that are not naturally sector-aligned, but it is possible that byte-based values will let us be more precise about allocation at the end of an unaligned file that can do byte-based access. Changing the signature of the function to use int64_t *pnum ensures that the compiler enforces that all callers are updated. For now, the io.c layer still assert()s that all callers are sector-aligned on input and that *pnum is sector-aligned on return to the caller, but that can be relaxed when a later patch implements byte-based block status. Therefore, this code adds usages like DIV_ROUND_UP(,BDRV_SECTOR_SIZE) to callers that still want aligned values, where the call might reasonbly give non-aligned results in the future; on the other hand, no rounding is needed for callers that should just continue to work with byte alignment. For the most part this patch is just the addition of scaling at the callers followed by inverse scaling at bdrv_is_allocated(). But some code, particularly bdrv_commit(), gets a lot simpler because it no longer has to mess with sectors; also, it is now possible to pass NULL if the caller does not care how much of the image is allocated beyond the initial offset. Leave comments where we can further simplify once a later patch eliminates the need for sector-aligned requests through bdrv_is_allocated(). For ease of review, bdrv_is_allocated_above() will be tackled separately. Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2017-07-07 14:44:57 +02:00
int bdrv_is_allocated(BlockDriverState *bs, int64_t offset, int64_t bytes,
int64_t *pnum);
int bdrv_is_allocated_above(BlockDriverState *top, BlockDriverState *base,
bool include_base, int64_t offset, int64_t bytes,
int64_t *pnum);
bool bdrv_is_read_only(BlockDriverState *bs);
int bdrv_can_set_read_only(BlockDriverState *bs, bool read_only,
bool ignore_allow_rdw, Error **errp);
int bdrv_apply_auto_read_only(BlockDriverState *bs, const char *errmsg,
Error **errp);
bool bdrv_is_writable(BlockDriverState *bs);
bool bdrv_is_sg(BlockDriverState *bs);
bool bdrv_is_inserted(BlockDriverState *bs);
void bdrv_lock_medium(BlockDriverState *bs, bool locked);
void bdrv_eject(BlockDriverState *bs, bool eject_flag);
const char *bdrv_get_format_name(BlockDriverState *bs);
BlockDriverState *bdrv_find_node(const char *node_name);
BlockDeviceInfoList *bdrv_named_nodes_list(bool flat, Error **errp);
XDbgBlockGraph *bdrv_get_xdbg_block_graph(Error **errp);
qmp: Allow to change password on named block driver states. Signed-off-by: Benoit Canet <benoit@irqsave.net> Reviewed-by: Fam Zheng <famz@redhat.com> There was two candidate ways to implement named node manipulation: 1) { 'command': 'block_passwd', 'data': {'*device': 'str', '*node-name': 'str', 'password': 'str'} } 2) { 'command': 'block_passwd', 'data': {'device': 'str', '*device-is-node': 'bool', 'password': 'str'} } Luiz proposed 1 and says 2 was an abuse of the QMP interface and proposed to rewrite the QMP block interface for 2.0. Luiz does not like in 1 the fact that 2 fields are optional but one of them must be specified leading to an abuse of the QMP semantic. Kevin argumented that 2 what a clear abuse of the device field and would not be practical when reading fast some log file because the user would read "device" and think that a device is manipulated when it's in fact a node name. Documentation of 1 make it pretty clear what to do for the user. Kevin argued that all bs are node including devices ones so 2 does not make sense. Kevin also argued that rewriting the QMP block interface would not make disapear the current one. Kevin pushed the argument that making the QAPI generator compatible with the semantic of the operation would need a rewrite that no one has done yet. A vote has been done on the list to elect the version to use and 1 won. For reference the complete thread is: "[Qemu-devel] [PATCH V4 4/7] qmp: Allow to change password on names block driver states." Signed-off-by: Benoit Canet <benoit@irqsave.net> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2014-01-23 21:31:35 +01:00
BlockDriverState *bdrv_lookup_bs(const char *device,
const char *node_name,
Error **errp);
bool bdrv_chain_contains(BlockDriverState *top, BlockDriverState *base);
BlockDriverState *bdrv_next_node(BlockDriverState *bs);
BlockDriverState *bdrv_next_all_states(BlockDriverState *bs);
typedef struct BdrvNextIterator {
enum {
BDRV_NEXT_BACKEND_ROOTS,
BDRV_NEXT_MONITOR_OWNED,
} phase;
BlockBackend *blk;
BlockDriverState *bs;
} BdrvNextIterator;
BlockDriverState *bdrv_first(BdrvNextIterator *it);
BlockDriverState *bdrv_next(BdrvNextIterator *it);
void bdrv_next_cleanup(BdrvNextIterator *it);
BlockDriverState *bdrv_next_monitor_owned(BlockDriverState *bs);
bool bdrv_is_encrypted(BlockDriverState *bs);
void bdrv_iterate_format(void (*it)(void *opaque, const char *name),
void *opaque, bool read_only);
const char *bdrv_get_node_name(const BlockDriverState *bs);
const char *bdrv_get_device_name(const BlockDriverState *bs);
const char *bdrv_get_device_or_node_name(const BlockDriverState *bs);
int bdrv_get_flags(BlockDriverState *bs);
int bdrv_get_info(BlockDriverState *bs, BlockDriverInfo *bdi);
ImageInfoSpecific *bdrv_get_specific_info(BlockDriverState *bs,
Error **errp);
BlockStatsSpecific *bdrv_get_specific_stats(BlockDriverState *bs);
void bdrv_round_to_clusters(BlockDriverState *bs,
int64_t offset, int64_t bytes,
int64_t *cluster_offset,
int64_t *cluster_bytes);
void bdrv_get_backing_filename(BlockDriverState *bs,
char *filename, int filename_size);
char *bdrv_get_full_backing_filename(BlockDriverState *bs, Error **errp);
char *bdrv_get_full_backing_filename_from_filename(const char *backed,
const char *backing,
Error **errp);
char *bdrv_dirname(BlockDriverState *bs, Error **errp);
int path_has_protocol(const char *path);
int path_is_absolute(const char *path);
char *path_combine(const char *base_path, const char *filename);
int bdrv_readv_vmstate(BlockDriverState *bs, QEMUIOVector *qiov, int64_t pos);
int bdrv_writev_vmstate(BlockDriverState *bs, QEMUIOVector *qiov, int64_t pos);
int bdrv_save_vmstate(BlockDriverState *bs, const uint8_t *buf,
int64_t pos, int size);
int bdrv_load_vmstate(BlockDriverState *bs, uint8_t *buf,
int64_t pos, int size);
void bdrv_img_create(const char *filename, const char *fmt,
const char *base_filename, const char *base_fmt,
char *options, uint64_t img_size, int flags,
bool quiet, Error **errp);
/* Returns the alignment in bytes that is required so that no bounce buffer
* is required throughout the stack */
size_t bdrv_min_mem_align(BlockDriverState *bs);
/* Returns optimal alignment in bytes for bounce buffer */
size_t bdrv_opt_mem_align(BlockDriverState *bs);
void *qemu_blockalign(BlockDriverState *bs, size_t size);
void *qemu_blockalign0(BlockDriverState *bs, size_t size);
void *qemu_try_blockalign(BlockDriverState *bs, size_t size);
void *qemu_try_blockalign0(BlockDriverState *bs, size_t size);
bool bdrv_qiov_is_aligned(BlockDriverState *bs, QEMUIOVector *qiov);
void bdrv_enable_copy_on_read(BlockDriverState *bs);
void bdrv_disable_copy_on_read(BlockDriverState *bs);
void bdrv_ref(BlockDriverState *bs);
void bdrv_unref(BlockDriverState *bs);
void bdrv_unref_child(BlockDriverState *parent, BdrvChild *child);
BdrvChild *bdrv_attach_child(BlockDriverState *parent_bs,
BlockDriverState *child_bs,
const char *child_name,
const BdrvChildClass *child_class,
BdrvChildRole child_role,
Error **errp);
bool bdrv_op_is_blocked(BlockDriverState *bs, BlockOpType op, Error **errp);
void bdrv_op_block(BlockDriverState *bs, BlockOpType op, Error *reason);
void bdrv_op_unblock(BlockDriverState *bs, BlockOpType op, Error *reason);
void bdrv_op_block_all(BlockDriverState *bs, Error *reason);
void bdrv_op_unblock_all(BlockDriverState *bs, Error *reason);
bool bdrv_op_blocker_is_empty(BlockDriverState *bs);
#define BLKDBG_EVENT(child, evt) \
do { \
if (child) { \
bdrv_debug_event(child->bs, evt); \
} \
} while (0)
void bdrv_debug_event(BlockDriverState *bs, BlkdebugEvent event);
int bdrv_debug_breakpoint(BlockDriverState *bs, const char *event,
const char *tag);
int bdrv_debug_remove_breakpoint(BlockDriverState *bs, const char *tag);
int bdrv_debug_resume(BlockDriverState *bs, const char *tag);
bool bdrv_debug_is_suspended(BlockDriverState *bs, const char *tag);
/**
* bdrv_get_aio_context:
*
* Returns: the currently bound #AioContext
*/
AioContext *bdrv_get_aio_context(BlockDriverState *bs);
/**
* Transfer control to @co in the aio context of @bs
*/
void bdrv_coroutine_enter(BlockDriverState *bs, Coroutine *co);
void bdrv_set_aio_context_ignore(BlockDriverState *bs,
AioContext *new_context, GSList **ignore);
int bdrv_try_set_aio_context(BlockDriverState *bs, AioContext *ctx,
Error **errp);
int bdrv_child_try_set_aio_context(BlockDriverState *bs, AioContext *ctx,
BdrvChild *ignore_child, Error **errp);
bool bdrv_child_can_set_aio_context(BdrvChild *c, AioContext *ctx,
GSList **ignore, Error **errp);
bool bdrv_can_set_aio_context(BlockDriverState *bs, AioContext *ctx,
GSList **ignore, Error **errp);
int bdrv_probe_blocksizes(BlockDriverState *bs, BlockSizes *bsz);
int bdrv_probe_geometry(BlockDriverState *bs, HDGeometry *geo);
void bdrv_io_plug(BlockDriverState *bs);
void bdrv_io_unplug(BlockDriverState *bs);
/**
* bdrv_parent_drained_begin_single:
*
* Begin a quiesced section for the parent of @c. If @poll is true, wait for
* any pending activity to cease.
*/
void bdrv_parent_drained_begin_single(BdrvChild *c, bool poll);
block: Introduce BdrvChild.parent_quiesce_counter Commit 5cb2737e925042e6c7cd3fb0b01313950b03cddf laid out why bdrv_do_drained_end() must decrement the quiesce_counter after bdrv_drain_invoke(). It did not give a very good reason why it has to happen after bdrv_parent_drained_end(), instead only claiming symmetry to bdrv_do_drained_begin(). It turns out that delaying it for so long is wrong. Situation: We have an active commit job (i.e. a mirror job) from top to base for the following graph: filter | [file] | v top --[backing]--> base Now the VM is closed, which results in the job being cancelled and a bdrv_drain_all() happening pretty much simultaneously. Beginning the drain means the job is paused once whenever one of its nodes is quiesced. This is reversed when the drain ends. With how the code currently is, after base's drain ends (which means that it will have unpaused the job once), its quiesce_counter remains at 1 while it goes to undrain its parents (bdrv_parent_drained_end()). For some reason or another, undraining filter causes the job to be kicked and enter mirror_exit_common(), where it proceeds to invoke block_job_remove_all_bdrv(). Now base will be detached from the job. Because its quiesce_counter is still 1, it will unpause the job once more. So in total, undraining base will unpause the job twice. Eventually, this will lead to the job's pause_count going negative -- well, it would, were there not an assertion against this, which crashes qemu. The general problem is that if in bdrv_parent_drained_end() we undrain parent A, and then undrain parent B, which then leads to A detaching the child, bdrv_replace_child_noperm() will undrain A as if we had not done so yet; that is, one time too many. It follows that we cannot decrement the quiesce_counter after invoking bdrv_parent_drained_end(). Unfortunately, decrementing it before bdrv_parent_drained_end() would be wrong, too. Imagine the above situation in reverse: Undraining A leads to B detaching the child. If we had already decremented the quiesce_counter by that point, bdrv_replace_child_noperm() would undrain B one time too little; because it expects bdrv_parent_drained_end() to issue this undrain. But bdrv_parent_drained_end() won't do that, because B is no longer a parent. Therefore, we have to do something else. This patch opts for introducing a second quiesce_counter that counts how many times a child's parent has been quiesced (though c->role->drained_*). With that, bdrv_replace_child_noperm() just has to undrain the parent exactly that many times when removing a child, and it will always be right. Signed-off-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2019-07-19 11:26:09 +02:00
/**
* bdrv_parent_drained_end_single:
*
* End a quiesced section for the parent of @c.
block: Do not poll in bdrv_do_drained_end() We should never poll anywhere in bdrv_do_drained_end() (including its recursive callees like bdrv_drain_invoke()), because it does not cope well with graph changes. In fact, it has been written based on the postulation that no graph changes will happen in it. Instead, the callers that want to poll must poll, i.e. all currently globally available wrappers: bdrv_drained_end(), bdrv_subtree_drained_end(), bdrv_unapply_subtree_drain(), and bdrv_drain_all_end(). Graph changes there do not matter. They can poll simply by passing a pointer to a drained_end_counter and wait until it reaches 0. This patch also adds a non-polling global wrapper for bdrv_do_drained_end() that takes a drained_end_counter pointer. We need such a variant because now no function called anywhere from bdrv_do_drained_end() must poll. This includes BdrvChildRole.drained_end(), which already must not poll according to its interface documentation, but bdrv_child_cb_drained_end() just violates that by invoking bdrv_drained_end() (which does poll). Therefore, BdrvChildRole.drained_end() must take a *drained_end_counter parameter, which bdrv_child_cb_drained_end() can pass on to the new bdrv_drained_end_no_poll() function. Note that we now have a pattern of all drained_end-related functions either polling or receiving a *drained_end_counter to let the caller poll based on that. A problem with a single poll loop is that when the drained section in bdrv_set_aio_context_ignore() ends, some nodes in the subgraph may be in the old contexts, while others are in the new context already. To let the collective poll in bdrv_drained_end() work correctly, we must not hold a lock to the old context, so that the old context can make progress in case it is different from the current context. (In the process, remove the comment saying that the current context is always the old context, because it is wrong.) In all other places, all nodes in a subtree must be in the same context, so we can just poll that. The exception of course is bdrv_drain_all_end(), but that always runs in the main context, so we can just poll NULL (like bdrv_drain_all_begin() does). Signed-off-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2019-07-19 11:26:14 +02:00
*
* This polls @bs's AioContext until all scheduled sub-drained_ends
* have settled, which may result in graph changes.
block: Introduce BdrvChild.parent_quiesce_counter Commit 5cb2737e925042e6c7cd3fb0b01313950b03cddf laid out why bdrv_do_drained_end() must decrement the quiesce_counter after bdrv_drain_invoke(). It did not give a very good reason why it has to happen after bdrv_parent_drained_end(), instead only claiming symmetry to bdrv_do_drained_begin(). It turns out that delaying it for so long is wrong. Situation: We have an active commit job (i.e. a mirror job) from top to base for the following graph: filter | [file] | v top --[backing]--> base Now the VM is closed, which results in the job being cancelled and a bdrv_drain_all() happening pretty much simultaneously. Beginning the drain means the job is paused once whenever one of its nodes is quiesced. This is reversed when the drain ends. With how the code currently is, after base's drain ends (which means that it will have unpaused the job once), its quiesce_counter remains at 1 while it goes to undrain its parents (bdrv_parent_drained_end()). For some reason or another, undraining filter causes the job to be kicked and enter mirror_exit_common(), where it proceeds to invoke block_job_remove_all_bdrv(). Now base will be detached from the job. Because its quiesce_counter is still 1, it will unpause the job once more. So in total, undraining base will unpause the job twice. Eventually, this will lead to the job's pause_count going negative -- well, it would, were there not an assertion against this, which crashes qemu. The general problem is that if in bdrv_parent_drained_end() we undrain parent A, and then undrain parent B, which then leads to A detaching the child, bdrv_replace_child_noperm() will undrain A as if we had not done so yet; that is, one time too many. It follows that we cannot decrement the quiesce_counter after invoking bdrv_parent_drained_end(). Unfortunately, decrementing it before bdrv_parent_drained_end() would be wrong, too. Imagine the above situation in reverse: Undraining A leads to B detaching the child. If we had already decremented the quiesce_counter by that point, bdrv_replace_child_noperm() would undrain B one time too little; because it expects bdrv_parent_drained_end() to issue this undrain. But bdrv_parent_drained_end() won't do that, because B is no longer a parent. Therefore, we have to do something else. This patch opts for introducing a second quiesce_counter that counts how many times a child's parent has been quiesced (though c->role->drained_*). With that, bdrv_replace_child_noperm() just has to undrain the parent exactly that many times when removing a child, and it will always be right. Signed-off-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2019-07-19 11:26:09 +02:00
*/
void bdrv_parent_drained_end_single(BdrvChild *c);
/**
* bdrv_drain_poll:
*
* Poll for pending requests in @bs, its parents (except for @ignore_parent),
* and if @recursive is true its children as well (used for subtree drain).
*
* If @ignore_bds_parents is true, parents that are BlockDriverStates must
* ignore the drain request because they will be drained separately (used for
* drain_all).
*
* This is part of bdrv_drained_begin.
*/
bool bdrv_drain_poll(BlockDriverState *bs, bool recursive,
BdrvChild *ignore_parent, bool ignore_bds_parents);
/**
* bdrv_drained_begin:
*
* Begin a quiesced section for exclusive access to the BDS, by disabling
* external request sources including NBD server and device model. Note that
* this doesn't block timers or coroutines from submitting more requests, which
* means block_job_pause is still necessary.
*
* This function can be recursive.
*/
void bdrv_drained_begin(BlockDriverState *bs);
/**
* bdrv_do_drained_begin_quiesce:
*
* Quiesces a BDS like bdrv_drained_begin(), but does not wait for already
* running requests to complete.
*/
void bdrv_do_drained_begin_quiesce(BlockDriverState *bs,
BdrvChild *parent, bool ignore_bds_parents);
/**
* Like bdrv_drained_begin, but recursively begins a quiesced section for
* exclusive access to all child nodes as well.
*/
void bdrv_subtree_drained_begin(BlockDriverState *bs);
/**
* bdrv_drained_end:
*
* End a quiescent section started by bdrv_drained_begin().
block: Do not poll in bdrv_do_drained_end() We should never poll anywhere in bdrv_do_drained_end() (including its recursive callees like bdrv_drain_invoke()), because it does not cope well with graph changes. In fact, it has been written based on the postulation that no graph changes will happen in it. Instead, the callers that want to poll must poll, i.e. all currently globally available wrappers: bdrv_drained_end(), bdrv_subtree_drained_end(), bdrv_unapply_subtree_drain(), and bdrv_drain_all_end(). Graph changes there do not matter. They can poll simply by passing a pointer to a drained_end_counter and wait until it reaches 0. This patch also adds a non-polling global wrapper for bdrv_do_drained_end() that takes a drained_end_counter pointer. We need such a variant because now no function called anywhere from bdrv_do_drained_end() must poll. This includes BdrvChildRole.drained_end(), which already must not poll according to its interface documentation, but bdrv_child_cb_drained_end() just violates that by invoking bdrv_drained_end() (which does poll). Therefore, BdrvChildRole.drained_end() must take a *drained_end_counter parameter, which bdrv_child_cb_drained_end() can pass on to the new bdrv_drained_end_no_poll() function. Note that we now have a pattern of all drained_end-related functions either polling or receiving a *drained_end_counter to let the caller poll based on that. A problem with a single poll loop is that when the drained section in bdrv_set_aio_context_ignore() ends, some nodes in the subgraph may be in the old contexts, while others are in the new context already. To let the collective poll in bdrv_drained_end() work correctly, we must not hold a lock to the old context, so that the old context can make progress in case it is different from the current context. (In the process, remove the comment saying that the current context is always the old context, because it is wrong.) In all other places, all nodes in a subtree must be in the same context, so we can just poll that. The exception of course is bdrv_drain_all_end(), but that always runs in the main context, so we can just poll NULL (like bdrv_drain_all_begin() does). Signed-off-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2019-07-19 11:26:14 +02:00
*
* This polls @bs's AioContext until all scheduled sub-drained_ends
* have settled. On one hand, that may result in graph changes. On
* the other, this requires that the caller either runs in the main
* loop; or that all involved nodes (@bs and all of its parents) are
* in the caller's AioContext.
*/
void bdrv_drained_end(BlockDriverState *bs);
block: Do not poll in bdrv_do_drained_end() We should never poll anywhere in bdrv_do_drained_end() (including its recursive callees like bdrv_drain_invoke()), because it does not cope well with graph changes. In fact, it has been written based on the postulation that no graph changes will happen in it. Instead, the callers that want to poll must poll, i.e. all currently globally available wrappers: bdrv_drained_end(), bdrv_subtree_drained_end(), bdrv_unapply_subtree_drain(), and bdrv_drain_all_end(). Graph changes there do not matter. They can poll simply by passing a pointer to a drained_end_counter and wait until it reaches 0. This patch also adds a non-polling global wrapper for bdrv_do_drained_end() that takes a drained_end_counter pointer. We need such a variant because now no function called anywhere from bdrv_do_drained_end() must poll. This includes BdrvChildRole.drained_end(), which already must not poll according to its interface documentation, but bdrv_child_cb_drained_end() just violates that by invoking bdrv_drained_end() (which does poll). Therefore, BdrvChildRole.drained_end() must take a *drained_end_counter parameter, which bdrv_child_cb_drained_end() can pass on to the new bdrv_drained_end_no_poll() function. Note that we now have a pattern of all drained_end-related functions either polling or receiving a *drained_end_counter to let the caller poll based on that. A problem with a single poll loop is that when the drained section in bdrv_set_aio_context_ignore() ends, some nodes in the subgraph may be in the old contexts, while others are in the new context already. To let the collective poll in bdrv_drained_end() work correctly, we must not hold a lock to the old context, so that the old context can make progress in case it is different from the current context. (In the process, remove the comment saying that the current context is always the old context, because it is wrong.) In all other places, all nodes in a subtree must be in the same context, so we can just poll that. The exception of course is bdrv_drain_all_end(), but that always runs in the main context, so we can just poll NULL (like bdrv_drain_all_begin() does). Signed-off-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2019-07-19 11:26:14 +02:00
/**
* bdrv_drained_end_no_poll:
*
* Same as bdrv_drained_end(), but do not poll for the subgraph to
* actually become unquiesced. Therefore, no graph changes will occur
* with this function.
*
* *drained_end_counter is incremented for every background operation
* that is scheduled, and will be decremented for every operation once
* it settles. The caller must poll until it reaches 0. The counter
* should be accessed using atomic operations only.
*/
void bdrv_drained_end_no_poll(BlockDriverState *bs, int *drained_end_counter);
/**
* End a quiescent section started by bdrv_subtree_drained_begin().
*/
void bdrv_subtree_drained_end(BlockDriverState *bs);
void bdrv_add_child(BlockDriverState *parent, BlockDriverState *child,
Error **errp);
void bdrv_del_child(BlockDriverState *parent, BdrvChild *child, Error **errp);
bool bdrv_can_store_new_dirty_bitmap(BlockDriverState *bs, const char *name,
uint32_t granularity, Error **errp);
/**
*
* bdrv_register_buf/bdrv_unregister_buf:
*
* Register/unregister a buffer for I/O. For example, VFIO drivers are
* interested to know the memory areas that would later be used for I/O, so
* that they can prepare IOMMU mapping etc., to get better performance.
*/
void bdrv_register_buf(BlockDriverState *bs, void *host, size_t size);
void bdrv_unregister_buf(BlockDriverState *bs, void *host);
/**
*
* bdrv_co_copy_range:
*
* Do offloaded copy between two children. If the operation is not implemented
* by the driver, or if the backend storage doesn't support it, a negative
* error code will be returned.
*
* Note: block layer doesn't emulate or fallback to a bounce buffer approach
* because usually the caller shouldn't attempt offloaded copy any more (e.g.
* calling copy_file_range(2)) after the first error, thus it should fall back
* to a read+write path in the caller level.
*
* @src: Source child to copy data from
* @src_offset: offset in @src image to read data
* @dst: Destination child to copy data to
* @dst_offset: offset in @dst image to write data
* @bytes: number of bytes to copy
* @flags: request flags. Supported flags:
* BDRV_REQ_ZERO_WRITE - treat the @src range as zero data and do zero
* write on @dst as if bdrv_co_pwrite_zeroes is
* called. Used to simplify caller code, or
* during BlockDriver.bdrv_co_copy_range_from()
* recursion.
* BDRV_REQ_NO_SERIALISING - do not serialize with other overlapping
* requests currently in flight.
*
* Returns: 0 if succeeded; negative error code if failed.
**/
int coroutine_fn bdrv_co_copy_range(BdrvChild *src, uint64_t src_offset,
BdrvChild *dst, uint64_t dst_offset,
uint64_t bytes, BdrvRequestFlags read_flags,
BdrvRequestFlags write_flags);
#endif