2007-11-11 03:51:17 +01:00
|
|
|
#ifndef BLOCK_H
|
|
|
|
#define BLOCK_H
|
|
|
|
|
2012-12-17 18:19:44 +01:00
|
|
|
#include "block/aio.h"
|
2018-02-11 10:36:01 +01:00
|
|
|
#include "qapi/qapi-types-block-core.h"
|
2018-02-16 17:50:12 +01:00
|
|
|
#include "block/aio-wait.h"
|
2016-03-09 10:52:44 +01:00
|
|
|
#include "qemu/iov.h"
|
2015-09-01 15:48:02 +02:00
|
|
|
#include "qemu/coroutine.h"
|
2014-09-05 15:46:18 +02:00
|
|
|
#include "block/accounting.h"
|
2016-03-08 05:44:55 +01:00
|
|
|
#include "block/dirty-bitmap.h"
|
2016-10-27 18:07:00 +02:00
|
|
|
#include "block/blockjob.h"
|
2016-03-08 05:44:53 +01:00
|
|
|
#include "qemu/hbitmap.h"
|
2008-09-22 21:17:18 +02:00
|
|
|
|
2007-11-11 03:51:17 +01:00
|
|
|
/* block.c */
|
|
|
|
typedef struct BlockDriver BlockDriver;
|
2015-06-15 13:24:19 +02:00
|
|
|
typedef struct BdrvChild BdrvChild;
|
2015-04-08 13:43:47 +02:00
|
|
|
typedef struct BdrvChildRole BdrvChildRole;
|
2007-11-11 03:51:17 +01:00
|
|
|
|
|
|
|
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;
|
2012-03-15 13:13:33 +01:00
|
|
|
bool is_dirty;
|
2013-10-24 12:06:53 +02:00
|
|
|
/*
|
|
|
|
* True if unallocated blocks read back as zeroes. This is equivalent
|
2015-08-26 13:17:13 +02:00
|
|
|
* to the LBPRZ flag in the SCSI logical block provisioning page.
|
2013-10-24 12:06:53 +02:00
|
|
|
*/
|
|
|
|
bool unallocated_blocks_are_zero;
|
2014-05-06 15:08:43 +02:00
|
|
|
/*
|
|
|
|
* True if this block driver only supports compressed writes
|
|
|
|
*/
|
|
|
|
bool needs_compressed_writes;
|
2007-11-11 03:51:17 +01:00
|
|
|
} BlockDriverInfo;
|
|
|
|
|
2012-03-15 13:13:31 +01:00
|
|
|
typedef struct BlockFragInfo {
|
|
|
|
uint64_t allocated_clusters;
|
|
|
|
uint64_t total_clusters;
|
|
|
|
uint64_t fragmented_clusters;
|
2013-02-07 17:15:04 +01:00
|
|
|
uint64_t compressed_clusters;
|
2012-03-15 13:13:31 +01:00
|
|
|
} BlockFragInfo;
|
|
|
|
|
2013-10-24 12:06:50 +02:00
|
|
|
typedef enum {
|
2015-09-08 05:28:32 +02:00
|
|
|
BDRV_REQ_COPY_ON_READ = 0x1,
|
|
|
|
BDRV_REQ_ZERO_WRITE = 0x2,
|
2018-07-25 13:20:32 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* 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.
|
2013-10-24 12:06:52 +02:00
|
|
|
*/
|
2015-09-08 05:28:32 +02:00
|
|
|
BDRV_REQ_MAY_UNMAP = 0x4,
|
2018-07-09 18:37:16 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* The BDRV_REQ_NO_SERIALISING flag is only valid for reads and means that
|
|
|
|
* we don't want wait_serialising_requests() during the read operation.
|
|
|
|
*
|
|
|
|
* This flag is used for backup copy-on-write operations, when we need to
|
|
|
|
* read old data before write (write notifier triggered). It is okay since
|
|
|
|
* we already waited for other serializing requests in the initiating write
|
|
|
|
* (see bdrv_aligned_pwritev), and it is necessary if the initiating write
|
|
|
|
* is already serializing (without the flag, the read would deadlock
|
|
|
|
* waiting for the serialising write to complete).
|
|
|
|
*/
|
2015-12-01 10:36:28 +01:00
|
|
|
BDRV_REQ_NO_SERIALISING = 0x8,
|
2016-03-04 14:28:01 +01:00
|
|
|
BDRV_REQ_FUA = 0x10,
|
2016-07-22 10:17:42 +02:00
|
|
|
BDRV_REQ_WRITE_COMPRESSED = 0x20,
|
2016-06-13 20:56:35 +02:00
|
|
|
|
2018-04-21 15:29:23 +02:00
|
|
|
/* Signifies that this write request will not change the visible disk
|
|
|
|
* content. */
|
|
|
|
BDRV_REQ_WRITE_UNCHANGED = 0x40,
|
|
|
|
|
2018-07-09 18:37:18 +02:00
|
|
|
/*
|
|
|
|
* 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,
|
|
|
|
|
2016-06-13 20:56:35 +02:00
|
|
|
/* Mask of valid flags */
|
2018-07-09 18:37:18 +02:00
|
|
|
BDRV_REQ_MASK = 0xff,
|
2013-10-24 12:06:50 +02:00
|
|
|
} BdrvRequestFlags;
|
|
|
|
|
2015-02-16 12:47:54 +01:00
|
|
|
typedef struct BlockSizes {
|
|
|
|
uint32_t phys;
|
|
|
|
uint32_t log;
|
|
|
|
} BlockSizes;
|
|
|
|
|
|
|
|
typedef struct HDGeometry {
|
|
|
|
uint32_t heads;
|
|
|
|
uint32_t sectors;
|
|
|
|
uint32_t cylinders;
|
|
|
|
} HDGeometry;
|
|
|
|
|
2007-11-11 03:51:17 +01:00
|
|
|
#define BDRV_O_RDWR 0x0002
|
2017-02-17 15:07:38 +01:00
|
|
|
#define BDRV_O_RESIZE 0x0004 /* request permission for resizing the node */
|
2007-11-11 03:51:17 +01:00
|
|
|
#define BDRV_O_SNAPSHOT 0x0008 /* open the file read only and save writes in a snapshot */
|
2014-04-11 19:16:36 +02:00
|
|
|
#define BDRV_O_TEMPORARY 0x0010 /* delete the file after use */
|
2008-10-14 16:42:54 +02:00
|
|
|
#define BDRV_O_NOCACHE 0x0020 /* do not use the host page cache */
|
2009-08-20 16:58:35 +02:00
|
|
|
#define BDRV_O_NATIVE_AIO 0x0080 /* use native AIO instead of the thread pool */
|
2010-01-12 12:55:16 +01:00
|
|
|
#define BDRV_O_NO_BACKING 0x0100 /* don't open the backing file */
|
2010-05-26 17:51:49 +02:00
|
|
|
#define BDRV_O_NO_FLUSH 0x0200 /* disable flushing on this disk */
|
2011-11-28 17:08:47 +01:00
|
|
|
#define BDRV_O_COPY_ON_READ 0x0400 /* copy read backing sectors into image */
|
2016-01-13 15:56:06 +01:00
|
|
|
#define BDRV_O_INACTIVE 0x0800 /* consistency hint for migration handoff */
|
2012-08-09 14:05:56 +02:00
|
|
|
#define BDRV_O_CHECK 0x1000 /* open solely for consistency check */
|
block: correctly set the keep_read_only flag
I believe the bs->keep_read_only flag is supposed to reflect
the initial open state of the device. If the device is initially
opened R/O, then commit operations, or reopen operations changing
to R/W, are prohibited.
Currently, the keep_read_only flag is only accurate for the active
layer, and its backing file. Subsequent images end up always having
the keep_read_only flag set.
For instance, what happens now:
[ base ] kro = 1, ro = 1
|
v
[ snap-1 ] kro = 1, ro = 1
|
v
[ snap-2 ] kro = 0, ro = 1
|
v
[ active ] kro = 0, ro = 0
What we want:
[ base ] kro = 0, ro = 1
|
v
[ snap-1 ] kro = 0, ro = 1
|
v
[ snap-2 ] kro = 0, ro = 1
|
v
[ active ] kro = 0, ro = 0
Signed-off-by: Jeff Cody <jcody@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2012-09-20 21:13:17 +02:00
|
|
|
#define BDRV_O_ALLOW_RDWR 0x2000 /* allow reopen to change from r/o to r/w */
|
2013-02-08 14:06:11 +01:00
|
|
|
#define BDRV_O_UNMAP 0x4000 /* execute guest UNMAP/TRIM operations */
|
2014-02-18 18:33:07 +01:00
|
|
|
#define BDRV_O_PROTOCOL 0x8000 /* if no block driver is explicitly given:
|
|
|
|
select an appropriate protocol driver,
|
|
|
|
ignoring the format layer */
|
2016-03-21 15:11:42 +01:00
|
|
|
#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 */
|
2008-10-14 16:42:54 +02:00
|
|
|
|
2016-03-18 17:46:45 +01:00
|
|
|
#define BDRV_O_CACHE_MASK (BDRV_O_NOCACHE | BDRV_O_NO_FLUSH)
|
2007-11-11 03:51:17 +01:00
|
|
|
|
2015-04-07 16:55:00 +02:00
|
|
|
|
|
|
|
/* 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"
|
2016-09-15 16:53:02 +02:00
|
|
|
#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"
|
2016-09-12 18:03:18 +02:00
|
|
|
#define BDRV_OPT_DISCARD "discard"
|
2017-05-02 18:35:37 +02:00
|
|
|
#define BDRV_OPT_FORCE_SHARE "force-share"
|
2015-04-07 16:55:00 +02:00
|
|
|
|
|
|
|
|
2009-11-30 18:21:19 +01:00
|
|
|
#define BDRV_SECTOR_BITS 9
|
2010-05-27 15:46:55 +02:00
|
|
|
#define BDRV_SECTOR_SIZE (1ULL << BDRV_SECTOR_BITS)
|
2010-05-01 08:23:32 +02:00
|
|
|
#define BDRV_SECTOR_MASK ~(BDRV_SECTOR_SIZE - 1)
|
2009-11-30 18:21:19 +01:00
|
|
|
|
2015-02-06 11:54:11 +01:00
|
|
|
#define BDRV_REQUEST_MAX_SECTORS MIN(SIZE_MAX >> BDRV_SECTOR_BITS, \
|
|
|
|
INT_MAX >> BDRV_SECTOR_BITS)
|
2017-01-20 17:25:26 +01:00
|
|
|
#define BDRV_REQUEST_MAX_BYTES (BDRV_REQUEST_MAX_SECTORS << BDRV_SECTOR_BITS)
|
2015-02-06 11:54:11 +01:00
|
|
|
|
2014-11-10 10:10:38 +01:00
|
|
|
/*
|
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.
|
2017-05-07 02:05:43 +02:00
|
|
|
*
|
|
|
|
* 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
|
2014-05-06 15:25:36 +02:00
|
|
|
* 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
|
2013-09-04 19:00:29 +02:00
|
|
|
*
|
2017-05-07 02:05:43 +02:00
|
|
|
* Internal flag:
|
2017-06-05 22:38:44 +02:00
|
|
|
* 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.
|
2013-09-04 19:00:29 +02:00
|
|
|
*
|
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:
|
2013-09-04 19:00:29 +02:00
|
|
|
*
|
|
|
|
* DATA ZERO OFFSET_VALID
|
2017-05-07 02:05:43 +02:00
|
|
|
* 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
|
2013-09-04 19:00:29 +02:00
|
|
|
* necessarily zero at offset
|
|
|
|
* f f t sectors preallocated but read from backing_hd,
|
2017-05-07 02:05:43 +02:00
|
|
|
* returned file contains garbage at offset
|
2013-09-04 19:00:29 +02:00
|
|
|
* 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
|
|
|
|
*/
|
2014-05-06 15:25:36 +02:00
|
|
|
#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
|
2017-05-05 04:14:59 +02:00
|
|
|
#define BDRV_BLOCK_EOF 0x20
|
2013-09-04 19:00:29 +02:00
|
|
|
#define BDRV_BLOCK_OFFSET_MASK BDRV_SECTOR_MASK
|
|
|
|
|
2012-09-20 21:13:19 +02:00
|
|
|
typedef QSIMPLEQ_HEAD(BlockReopenQueue, BlockReopenQueueEntry) BlockReopenQueue;
|
|
|
|
|
|
|
|
typedef struct BDRVReopenState {
|
|
|
|
BlockDriverState *bs;
|
|
|
|
int flags;
|
2018-09-06 11:37:09 +02:00
|
|
|
BlockdevDetectZeroesOptions detect_zeroes;
|
2019-03-12 17:48:45 +01:00
|
|
|
bool backing_missing;
|
2017-07-03 17:07:35 +02:00
|
|
|
uint64_t perm, shared_perm;
|
2015-04-10 17:50:50 +02:00
|
|
|
QDict *options;
|
2015-05-08 16:15:03 +02:00
|
|
|
QDict *explicit_options;
|
2012-09-20 21:13:19 +02:00
|
|
|
void *opaque;
|
|
|
|
} BDRVReopenState;
|
|
|
|
|
2014-05-23 15:29:41 +02:00
|
|
|
/*
|
|
|
|
* Block operation types
|
|
|
|
*/
|
|
|
|
typedef enum BlockOpType {
|
|
|
|
BLOCK_OP_TYPE_BACKUP_SOURCE,
|
|
|
|
BLOCK_OP_TYPE_BACKUP_TARGET,
|
|
|
|
BLOCK_OP_TYPE_CHANGE,
|
2014-09-11 07:14:00 +02:00
|
|
|
BLOCK_OP_TYPE_COMMIT_SOURCE,
|
|
|
|
BLOCK_OP_TYPE_COMMIT_TARGET,
|
2014-05-23 15:29:41 +02:00
|
|
|
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,
|
2015-12-24 05:45:02 +01:00
|
|
|
BLOCK_OP_TYPE_MIRROR_SOURCE,
|
2015-12-24 05:45:04 +01:00
|
|
|
BLOCK_OP_TYPE_MIRROR_TARGET,
|
2014-05-23 15:29:41 +02:00
|
|
|
BLOCK_OP_TYPE_RESIZE,
|
|
|
|
BLOCK_OP_TYPE_STREAM,
|
2014-06-27 18:25:25 +02:00
|
|
|
BLOCK_OP_TYPE_REPLACE,
|
2014-05-23 15:29:41 +02:00
|
|
|
BLOCK_OP_TYPE_MAX,
|
|
|
|
} BlockOpType;
|
2012-09-20 21:13:19 +02:00
|
|
|
|
2016-12-20 16:52:41 +01:00
|
|
|
/* 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.
|
2018-04-21 15:29:22 +02:00
|
|
|
*
|
|
|
|
* As the BLK_PERM_WRITE permission is strictly stronger, either is
|
|
|
|
* sufficient to perform an unchanging write.
|
2016-12-20 16:52:41 +01:00
|
|
|
*/
|
|
|
|
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,
|
2018-07-03 16:48:47 +02:00
|
|
|
|
|
|
|
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,
|
2016-12-20 16:52:41 +01:00
|
|
|
};
|
|
|
|
|
2017-05-02 18:35:36 +02:00
|
|
|
char *bdrv_perm_names(uint64_t perm);
|
|
|
|
|
2011-11-03 09:57:25 +01:00
|
|
|
/* disk I/O throttling */
|
2007-11-11 03:51:17 +01:00
|
|
|
void bdrv_init(void);
|
2009-10-27 18:41:44 +01:00
|
|
|
void bdrv_init_with_whitelist(void);
|
2016-03-21 15:11:48 +01:00
|
|
|
bool bdrv_uses_whitelist(void);
|
2018-02-21 11:47:43 +01:00
|
|
|
int bdrv_is_whitelisted(BlockDriver *drv, bool read_only);
|
2013-07-10 15:47:39 +02:00
|
|
|
BlockDriver *bdrv_find_protocol(const char *filename,
|
2015-02-05 19:58:12 +01:00
|
|
|
bool allow_protocol_prefix,
|
|
|
|
Error **errp);
|
2007-11-11 03:51:17 +01:00
|
|
|
BlockDriver *bdrv_find_format(const char *format_name);
|
2009-05-18 16:42:10 +02:00
|
|
|
int bdrv_create(BlockDriver *drv, const char* filename,
|
2014-06-05 11:21:11 +02:00
|
|
|
QemuOpts *opts, Error **errp);
|
|
|
|
int bdrv_create_file(const char *filename, QemuOpts *opts, Error **errp);
|
2014-10-07 13:59:03 +02:00
|
|
|
BlockDriverState *bdrv_new(void);
|
2017-02-20 12:46:42 +01:00
|
|
|
void bdrv_append(BlockDriverState *bs_new, BlockDriverState *bs_top,
|
|
|
|
Error **errp);
|
2017-03-06 16:20:51 +01:00
|
|
|
void bdrv_replace_node(BlockDriverState *from, BlockDriverState *to,
|
|
|
|
Error **errp);
|
2015-09-15 11:58:23 +02:00
|
|
|
|
2016-03-14 11:40:23 +01:00
|
|
|
int bdrv_parse_cache_mode(const char *mode, int *flags, bool *writethrough);
|
2013-02-08 14:06:11 +01:00
|
|
|
int bdrv_parse_discard_flags(const char *mode, int *flags);
|
2015-06-15 13:24:19 +02:00
|
|
|
BdrvChild *bdrv_open_child(const char *filename,
|
|
|
|
QDict *options, const char *bdref_key,
|
|
|
|
BlockDriverState* parent,
|
|
|
|
const BdrvChildRole *child_role,
|
|
|
|
bool allow_none, Error **errp);
|
2018-01-10 15:52:33 +01:00
|
|
|
BlockDriverState *bdrv_open_blockdev_ref(BlockdevRef *ref, Error **errp);
|
2017-02-17 20:42:32 +01:00
|
|
|
void bdrv_set_backing_hd(BlockDriverState *bs, BlockDriverState *backing_hd,
|
|
|
|
Error **errp);
|
2015-01-16 18:23:41 +01:00
|
|
|
int bdrv_open_backing_file(BlockDriverState *bs, QDict *parent_options,
|
|
|
|
const char *bdref_key, Error **errp);
|
2016-05-17 16:41:31 +02:00
|
|
|
BlockDriverState *bdrv_open(const char *filename, const char *reference,
|
|
|
|
QDict *options, int flags, Error **errp);
|
2017-01-18 17:16:41 +01:00
|
|
|
BlockDriverState *bdrv_new_open_driver(BlockDriver *drv, const char *node_name,
|
|
|
|
int flags, Error **errp);
|
2012-09-20 21:13:19 +02:00
|
|
|
BlockReopenQueue *bdrv_reopen_queue(BlockReopenQueue *bs_queue,
|
2019-03-12 17:48:44 +01:00
|
|
|
BlockDriverState *bs, QDict *options,
|
|
|
|
bool keep_old_opts);
|
2016-10-27 12:49:02 +02:00
|
|
|
int bdrv_reopen_multiple(AioContext *ctx, BlockReopenQueue *bs_queue, Error **errp);
|
2018-11-12 15:00:33 +01:00
|
|
|
int bdrv_reopen_set_read_only(BlockDriverState *bs, bool read_only,
|
|
|
|
Error **errp);
|
2012-09-20 21:13:19 +02:00
|
|
|
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);
|
2016-05-30 16:48:35 +02:00
|
|
|
int bdrv_read(BdrvChild *child, int64_t sector_num,
|
2007-11-11 03:51:17 +01:00
|
|
|
uint8_t *buf, int nb_sectors);
|
2016-05-31 14:42:08 +02:00
|
|
|
int bdrv_write(BdrvChild *child, int64_t sector_num,
|
2007-11-11 03:51:17 +01:00
|
|
|
const uint8_t *buf, int nb_sectors);
|
2016-06-16 15:13:15 +02:00
|
|
|
int bdrv_pwrite_zeroes(BdrvChild *child, int64_t offset,
|
2017-06-09 12:18:08 +02:00
|
|
|
int bytes, BdrvRequestFlags flags);
|
2016-06-16 15:13:15 +02:00
|
|
|
int bdrv_make_zero(BdrvChild *child, BdrvRequestFlags flags);
|
2016-06-20 18:24:02 +02:00
|
|
|
int bdrv_pread(BdrvChild *child, int64_t offset, void *buf, int bytes);
|
|
|
|
int bdrv_preadv(BdrvChild *child, int64_t offset, QEMUIOVector *qiov);
|
2016-06-20 20:09:15 +02:00
|
|
|
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);
|
2012-02-07 14:27:25 +01:00
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*/
|
2016-06-20 21:31:46 +02:00
|
|
|
int coroutine_fn bdrv_co_pwrite_zeroes(BdrvChild *child, int64_t offset,
|
2017-06-09 12:18:08 +02:00
|
|
|
int bytes, BdrvRequestFlags flags);
|
2012-01-18 15:40:51 +01:00
|
|
|
BlockDriverState *bdrv_find_backing_image(BlockDriverState *bs,
|
|
|
|
const char *backing_file);
|
2014-07-18 20:24:56 +02:00
|
|
|
void bdrv_refresh_filename(BlockDriverState *bs);
|
block: Convert .bdrv_truncate callback to coroutine_fn
bdrv_truncate() is an operation that can block (even for a quite long
time, depending on the PreallocMode) in I/O paths that shouldn't block.
Convert it to a coroutine_fn so that we have the infrastructure for
drivers to make their .bdrv_co_truncate implementation asynchronous.
This change could potentially introduce new race conditions because
bdrv_truncate() isn't necessarily executed atomically any more. Whether
this is a problem needs to be evaluated for each block driver that
supports truncate:
* file-posix/win32, gluster, iscsi, nfs, rbd, ssh, sheepdog: The
protocol drivers are trivially safe because they don't actually yield
yet, so there is no change in behaviour.
* copy-on-read, crypto, raw-format: Essentially just filter drivers that
pass the request to a child node, no problem.
* qcow2: The implementation modifies metadata, so it needs to hold
s->lock to be safe with concurrent I/O requests. In order to avoid
double locking, this requires pulling the locking out into
preallocate_co() and using qcow2_write_caches() instead of
bdrv_flush().
* qed: Does a single header update, this is fine without locking.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
2018-06-21 17:54:35 +02:00
|
|
|
|
|
|
|
int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset,
|
|
|
|
PreallocMode prealloc, Error **errp);
|
2017-06-13 22:20:53 +02:00
|
|
|
int bdrv_truncate(BdrvChild *child, int64_t offset, PreallocMode prealloc,
|
|
|
|
Error **errp);
|
block: Convert .bdrv_truncate callback to coroutine_fn
bdrv_truncate() is an operation that can block (even for a quite long
time, depending on the PreallocMode) in I/O paths that shouldn't block.
Convert it to a coroutine_fn so that we have the infrastructure for
drivers to make their .bdrv_co_truncate implementation asynchronous.
This change could potentially introduce new race conditions because
bdrv_truncate() isn't necessarily executed atomically any more. Whether
this is a problem needs to be evaluated for each block driver that
supports truncate:
* file-posix/win32, gluster, iscsi, nfs, rbd, ssh, sheepdog: The
protocol drivers are trivially safe because they don't actually yield
yet, so there is no change in behaviour.
* copy-on-read, crypto, raw-format: Essentially just filter drivers that
pass the request to a child node, no problem.
* qcow2: The implementation modifies metadata, so it needs to hold
s->lock to be safe with concurrent I/O requests. In order to avoid
double locking, this requires pulling the locking out into
preallocate_co() and using qcow2_write_caches() instead of
bdrv_flush().
* qed: Does a single header update, this is fine without locking.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
2018-06-21 17:54:35 +02:00
|
|
|
|
2014-06-26 13:23:17 +02:00
|
|
|
int64_t bdrv_nb_sectors(BlockDriverState *bs);
|
2007-11-11 03:51:17 +01:00
|
|
|
int64_t bdrv_getlength(BlockDriverState *bs);
|
2011-07-12 13:56:39 +02:00
|
|
|
int64_t bdrv_get_allocated_file_size(BlockDriverState *bs);
|
2017-07-05 14:57:30 +02:00
|
|
|
BlockMeasureInfo *bdrv_measure(BlockDriver *drv, QemuOpts *opts,
|
|
|
|
BlockDriverState *in_bs, Error **errp);
|
2007-12-17 02:35:20 +01:00
|
|
|
void bdrv_get_geometry(BlockDriverState *bs, uint64_t *nb_sectors_ptr);
|
2014-07-16 17:48:16 +02:00
|
|
|
void bdrv_refresh_limits(BlockDriverState *bs, Error **errp);
|
2007-11-11 03:51:17 +01:00
|
|
|
int bdrv_commit(BlockDriverState *bs);
|
2010-01-12 12:55:17 +01:00
|
|
|
int bdrv_change_backing_file(BlockDriverState *bs,
|
|
|
|
const char *backing_file, const char *backing_fmt);
|
2009-05-10 00:03:42 +02:00
|
|
|
void bdrv_register(BlockDriver *bdrv);
|
2017-06-27 20:36:18 +02:00
|
|
|
int bdrv_drop_intermediate(BlockDriverState *top, BlockDriverState *base,
|
block: extend block-commit to accept a string for the backing file
On some image chains, QEMU may not always be able to resolve the
filenames properly, when updating the backing file of an image
after a block commit.
For instance, certain relative pathnames may fail, or drives may
have been specified originally by file descriptor (e.g. /dev/fd/???),
or a relative protocol pathname may have been used.
In these instances, QEMU may lack the information to be able to make
the correct choice, but the user or management layer most likely does
have that knowledge.
With this extension to the block-commit api, the user is able to change
the backing file of the overlay image as part of the block-commit
operation.
This allows the change to be 'safe', in the sense that if the attempt
to write the overlay image metadata fails, then the block-commit
operation returns failure, without disrupting the guest.
If the commit top is the active layer, then specifying the backing
file string will be treated as an error (there is no overlay image
to modify in that case).
If a backing file string is not specified in the command, the backing
file string to use is determined in the same manner as it was
previously.
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Jeff Cody <jcody@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2014-06-25 21:40:10 +02:00
|
|
|
const char *backing_file_str);
|
2012-09-27 19:29:12 +02:00
|
|
|
BlockDriverState *bdrv_find_overlay(BlockDriverState *active,
|
|
|
|
BlockDriverState *bs);
|
2012-09-27 19:29:15 +02:00
|
|
|
BlockDriverState *bdrv_find_base(BlockDriverState *bs);
|
2019-03-12 17:48:40 +01:00
|
|
|
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);
|
2009-05-10 00:03:42 +02:00
|
|
|
|
2010-06-29 11:43:13 +02:00
|
|
|
|
|
|
|
typedef struct BdrvCheckResult {
|
|
|
|
int corruptions;
|
|
|
|
int leaks;
|
|
|
|
int check_errors;
|
2012-05-11 18:16:54 +02:00
|
|
|
int corruptions_fixed;
|
|
|
|
int leaks_fixed;
|
2013-01-28 12:59:46 +01:00
|
|
|
int64_t image_end_offset;
|
2012-03-15 13:13:31 +01:00
|
|
|
BlockFragInfo bfi;
|
2010-06-29 11:43:13 +02:00
|
|
|
} BdrvCheckResult;
|
|
|
|
|
2012-05-11 16:07:02 +02:00
|
|
|
typedef enum {
|
|
|
|
BDRV_FIX_LEAKS = 1,
|
|
|
|
BDRV_FIX_ERRORS = 2,
|
|
|
|
} BdrvCheckMode;
|
|
|
|
|
|
|
|
int bdrv_check(BlockDriverState *bs, BdrvCheckResult *res, BdrvCheckMode fix);
|
2010-06-29 11:43:13 +02:00
|
|
|
|
2014-10-27 11:12:50 +01:00
|
|
|
/* 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,
|
2015-07-27 17:51:32 +02:00
|
|
|
int64_t total_work_size, void *opaque);
|
2014-10-27 11:12:50 +01:00
|
|
|
int bdrv_amend_options(BlockDriverState *bs_new, QemuOpts *opts,
|
2018-05-09 23:00:18 +02:00
|
|
|
BlockDriverAmendStatusCB *status_cb, void *cb_opaque,
|
|
|
|
Error **errp);
|
2013-09-03 10:09:50 +02:00
|
|
|
|
2013-10-02 14:33:48 +02:00
|
|
|
/* external snapshots */
|
2014-01-23 21:31:36 +01:00
|
|
|
bool bdrv_recurse_is_first_non_filter(BlockDriverState *bs,
|
|
|
|
BlockDriverState *candidate);
|
|
|
|
bool bdrv_is_first_non_filter(BlockDriverState *candidate);
|
2013-10-02 14:33:48 +02:00
|
|
|
|
2014-06-27 18:25:25 +02:00
|
|
|
/* check if a named node can be replaced when doing drive-mirror */
|
2015-07-17 04:12:22 +02:00
|
|
|
BlockDriverState *check_to_replace_node(BlockDriverState *parent_bs,
|
|
|
|
const char *node_name, Error **errp);
|
2014-06-27 18:25:25 +02:00
|
|
|
|
2007-11-11 03:51:17 +01:00
|
|
|
/* async block I/O */
|
2014-10-07 13:59:14 +02:00
|
|
|
void bdrv_aio_cancel(BlockAIOCB *acb);
|
|
|
|
void bdrv_aio_cancel_async(BlockAIOCB *acb);
|
2007-11-11 03:51:17 +01:00
|
|
|
|
2009-03-12 20:57:08 +01:00
|
|
|
/* sg packet commands */
|
2016-10-20 12:56:14 +02:00
|
|
|
int bdrv_co_ioctl(BlockDriverState *bs, int req, void *buf);
|
2009-03-12 20:57:08 +01:00
|
|
|
|
2011-11-14 22:09:45 +01:00
|
|
|
/* Invalidate any cached metadata used by image formats */
|
2014-03-12 15:59:16 +01:00
|
|
|
void bdrv_invalidate_cache(BlockDriverState *bs, Error **errp);
|
|
|
|
void bdrv_invalidate_cache_all(Error **errp);
|
2015-12-22 14:07:08 +01:00
|
|
|
int bdrv_inactivate_all(void);
|
2011-11-14 22:09:45 +01:00
|
|
|
|
2007-11-11 03:51:17 +01:00
|
|
|
/* Ensure contents are flushed to disk. */
|
2010-10-21 16:43:43 +02:00
|
|
|
int bdrv_flush(BlockDriverState *bs);
|
2011-10-17 12:32:12 +02:00
|
|
|
int coroutine_fn bdrv_co_flush(BlockDriverState *bs);
|
2016-09-23 03:45:50 +02:00
|
|
|
int bdrv_flush_all(void);
|
2010-05-28 04:44:57 +02:00
|
|
|
void bdrv_close_all(void);
|
2014-10-21 13:03:55 +02:00
|
|
|
void bdrv_drain(BlockDriverState *bs);
|
2016-04-05 13:20:52 +02:00
|
|
|
void coroutine_fn bdrv_co_drain(BlockDriverState *bs);
|
2016-10-28 09:08:02 +02:00
|
|
|
void bdrv_drain_all_begin(void);
|
|
|
|
void bdrv_drain_all_end(void);
|
2011-11-30 13:23:43 +01:00
|
|
|
void bdrv_drain_all(void);
|
2008-10-06 15:55:43 +02:00
|
|
|
|
2016-10-27 12:48:55 +02:00
|
|
|
#define BDRV_POLL_WHILE(bs, cond) ({ \
|
|
|
|
BlockDriverState *bs_ = (bs); \
|
2018-09-18 17:09:16 +02:00
|
|
|
AIO_WAIT_WHILE(bdrv_get_aio_context(bs_), \
|
2018-02-16 17:50:12 +01:00
|
|
|
cond); })
|
2016-10-27 12:48:55 +02:00
|
|
|
|
2018-07-10 08:31:17 +02:00
|
|
|
int bdrv_pdiscard(BdrvChild *child, int64_t offset, int bytes);
|
|
|
|
int bdrv_co_pdiscard(BdrvChild *child, int64_t offset, int bytes);
|
2013-06-28 12:47:42 +02:00
|
|
|
int bdrv_has_zero_init_1(BlockDriverState *bs);
|
2010-04-14 17:30:35 +02:00
|
|
|
int bdrv_has_zero_init(BlockDriverState *bs);
|
2013-10-24 12:06:54 +02:00
|
|
|
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);
|
2013-02-13 09:09:39 +01:00
|
|
|
int bdrv_is_allocated_above(BlockDriverState *top, BlockDriverState *base,
|
block: Make bdrv_is_allocated_above() 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,
but that can be relaxed when a later patch implements byte-based
block status. Therefore, 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 stream_run(),
gets a lot simpler because it no longer has to mess with sectors.
Leave comments where we can further simplify by switching to
byte-based iterations, once later patches eliminate the need for
sector-aligned operations.
For ease of review, bdrv_is_allocated() was tackled separately.
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2017-07-07 14:44:59 +02:00
|
|
|
int64_t offset, int64_t bytes, int64_t *pnum);
|
2007-11-11 03:51:17 +01:00
|
|
|
|
2016-06-24 00:37:26 +02:00
|
|
|
bool bdrv_is_read_only(BlockDriverState *bs);
|
2017-08-03 17:02:58 +02:00
|
|
|
int bdrv_can_set_read_only(BlockDriverState *bs, bool read_only,
|
|
|
|
bool ignore_allow_rdw, Error **errp);
|
2018-10-12 11:27:41 +02:00
|
|
|
int bdrv_apply_auto_read_only(BlockDriverState *bs, const char *errmsg,
|
|
|
|
Error **errp);
|
2018-06-06 21:37:00 +02:00
|
|
|
bool bdrv_is_writable(BlockDriverState *bs);
|
2016-06-24 00:37:26 +02:00
|
|
|
bool bdrv_is_sg(BlockDriverState *bs);
|
2015-10-19 17:53:11 +02:00
|
|
|
bool bdrv_is_inserted(BlockDriverState *bs);
|
2011-09-06 18:58:47 +02:00
|
|
|
void bdrv_lock_medium(BlockDriverState *bs, bool locked);
|
2012-02-03 19:24:53 +01:00
|
|
|
void bdrv_eject(BlockDriverState *bs, bool eject_flag);
|
2012-06-13 10:11:48 +02:00
|
|
|
const char *bdrv_get_format_name(BlockDriverState *bs);
|
2014-01-23 21:31:32 +01:00
|
|
|
BlockDriverState *bdrv_find_node(const char *node_name);
|
2015-04-17 13:52:43 +02:00
|
|
|
BlockDeviceInfoList *bdrv_named_nodes_list(Error **errp);
|
2018-12-21 18:09:07 +01:00
|
|
|
XDbgBlockGraph *bdrv_get_xdbg_block_graph(Error **errp);
|
2014-01-23 21:31:35 +01:00
|
|
|
BlockDriverState *bdrv_lookup_bs(const char *device,
|
|
|
|
const char *node_name,
|
|
|
|
Error **errp);
|
2014-06-25 21:40:09 +02:00
|
|
|
bool bdrv_chain_contains(BlockDriverState *top, BlockDriverState *base);
|
2014-10-31 04:32:54 +01:00
|
|
|
BlockDriverState *bdrv_next_node(BlockDriverState *bs);
|
2018-03-28 18:29:18 +02:00
|
|
|
BlockDriverState *bdrv_next_all_states(BlockDriverState *bs);
|
2016-05-20 18:49:07 +02:00
|
|
|
|
|
|
|
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);
|
2017-11-10 18:25:45 +01:00
|
|
|
void bdrv_next_cleanup(BdrvNextIterator *it);
|
2016-05-20 18:49:07 +02:00
|
|
|
|
2016-03-16 19:54:41 +01:00
|
|
|
BlockDriverState *bdrv_next_monitor_owned(BlockDriverState *bs);
|
2016-06-24 00:37:26 +02:00
|
|
|
bool bdrv_is_encrypted(BlockDriverState *bs);
|
2007-11-11 03:51:17 +01:00
|
|
|
void bdrv_iterate_format(void (*it)(void *opaque, const char *name),
|
2019-03-07 14:33:58 +01:00
|
|
|
void *opaque, bool read_only);
|
2014-10-31 04:32:55 +01:00
|
|
|
const char *bdrv_get_node_name(const BlockDriverState *bs);
|
2014-10-07 13:59:11 +02:00
|
|
|
const char *bdrv_get_device_name(const BlockDriverState *bs);
|
2015-04-08 11:29:18 +02:00
|
|
|
const char *bdrv_get_device_or_node_name(const BlockDriverState *bs);
|
2012-06-05 16:49:24 +02:00
|
|
|
int bdrv_get_flags(BlockDriverState *bs);
|
2007-11-11 03:51:17 +01:00
|
|
|
int bdrv_get_info(BlockDriverState *bs, BlockDriverInfo *bdi);
|
2019-02-08 16:06:06 +01:00
|
|
|
ImageInfoSpecific *bdrv_get_specific_info(BlockDriverState *bs,
|
|
|
|
Error **errp);
|
2013-01-21 17:09:42 +01:00
|
|
|
void bdrv_round_to_clusters(BlockDriverState *bs,
|
2017-10-12 05:46:59 +02:00
|
|
|
int64_t offset, int64_t bytes,
|
2016-06-02 11:41:52 +02:00
|
|
|
int64_t *cluster_offset,
|
2017-10-12 05:46:59 +02:00
|
|
|
int64_t *cluster_bytes);
|
2007-11-11 03:51:17 +01:00
|
|
|
|
|
|
|
void bdrv_get_backing_filename(BlockDriverState *bs,
|
|
|
|
char *filename, int filename_size);
|
2019-02-01 20:29:15 +01:00
|
|
|
char *bdrv_get_full_backing_filename(BlockDriverState *bs, Error **errp);
|
2019-02-01 20:29:14 +01:00
|
|
|
char *bdrv_get_full_backing_filename_from_filename(const char *backed,
|
|
|
|
const char *backing,
|
|
|
|
Error **errp);
|
2019-02-01 20:29:18 +01:00
|
|
|
char *bdrv_dirname(BlockDriverState *bs, Error **errp);
|
2007-11-11 03:51:17 +01:00
|
|
|
|
2014-12-03 14:57:22 +01:00
|
|
|
int path_has_protocol(const char *path);
|
2007-11-11 03:51:17 +01:00
|
|
|
int path_is_absolute(const char *path);
|
2019-02-01 20:29:13 +01:00
|
|
|
char *path_combine(const char *base_path, const char *filename);
|
2007-11-11 03:51:17 +01:00
|
|
|
|
2016-06-09 16:50:16 +02:00
|
|
|
int bdrv_readv_vmstate(BlockDriverState *bs, QEMUIOVector *qiov, int64_t pos);
|
2013-04-05 21:27:53 +02:00
|
|
|
int bdrv_writev_vmstate(BlockDriverState *bs, QEMUIOVector *qiov, int64_t pos);
|
2009-07-10 23:11:57 +02:00
|
|
|
int bdrv_save_vmstate(BlockDriverState *bs, const uint8_t *buf,
|
|
|
|
int64_t pos, int size);
|
2009-04-05 21:10:55 +02:00
|
|
|
|
2009-07-10 23:11:57 +02:00
|
|
|
int bdrv_load_vmstate(BlockDriverState *bs, uint8_t *buf,
|
|
|
|
int64_t pos, int size);
|
2009-04-05 21:10:55 +02:00
|
|
|
|
2012-11-30 13:52:09 +01:00
|
|
|
void bdrv_img_create(const char *filename, const char *fmt,
|
|
|
|
const char *base_filename, const char *base_fmt,
|
2013-02-13 09:09:40 +01:00
|
|
|
char *options, uint64_t img_size, int flags,
|
2017-04-21 14:27:01 +02:00
|
|
|
bool quiet, Error **errp);
|
2010-12-16 13:52:15 +01:00
|
|
|
|
2013-11-28 10:23:32 +01:00
|
|
|
/* Returns the alignment in bytes that is required so that no bounce buffer
|
|
|
|
* is required throughout the stack */
|
2015-05-12 16:30:55 +02:00
|
|
|
size_t bdrv_min_mem_align(BlockDriverState *bs);
|
|
|
|
/* Returns optimal alignment in bytes for bounce buffer */
|
2013-11-28 10:23:32 +01:00
|
|
|
size_t bdrv_opt_mem_align(BlockDriverState *bs);
|
2011-08-03 15:08:19 +02:00
|
|
|
void *qemu_blockalign(BlockDriverState *bs, size_t size);
|
2014-10-22 14:09:27 +02:00
|
|
|
void *qemu_blockalign0(BlockDriverState *bs, size_t size);
|
2014-05-20 12:24:05 +02:00
|
|
|
void *qemu_try_blockalign(BlockDriverState *bs, size_t size);
|
2014-10-22 14:09:27 +02:00
|
|
|
void *qemu_try_blockalign0(BlockDriverState *bs, size_t size);
|
2013-01-11 16:41:27 +01:00
|
|
|
bool bdrv_qiov_is_aligned(BlockDriverState *bs, QEMUIOVector *qiov);
|
2011-08-03 15:08:19 +02:00
|
|
|
|
2011-11-28 17:08:47 +01:00
|
|
|
void bdrv_enable_copy_on_read(BlockDriverState *bs);
|
|
|
|
void bdrv_disable_copy_on_read(BlockDriverState *bs);
|
|
|
|
|
2013-08-23 03:14:46 +02:00
|
|
|
void bdrv_ref(BlockDriverState *bs);
|
|
|
|
void bdrv_unref(BlockDriverState *bs);
|
2015-06-15 13:51:04 +02:00
|
|
|
void bdrv_unref_child(BlockDriverState *parent, BdrvChild *child);
|
2016-05-10 09:36:38 +02:00
|
|
|
BdrvChild *bdrv_attach_child(BlockDriverState *parent_bs,
|
|
|
|
BlockDriverState *child_bs,
|
|
|
|
const char *child_name,
|
2016-12-20 22:21:17 +01:00
|
|
|
const BdrvChildRole *child_role,
|
|
|
|
Error **errp);
|
2010-03-15 17:27:00 +01:00
|
|
|
|
2014-05-23 15:29:42 +02:00
|
|
|
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);
|
|
|
|
|
2015-06-16 14:19:22 +02:00
|
|
|
#define BLKDBG_EVENT(child, evt) \
|
|
|
|
do { \
|
|
|
|
if (child) { \
|
|
|
|
bdrv_debug_event(child->bs, evt); \
|
|
|
|
} \
|
|
|
|
} while (0)
|
|
|
|
|
2015-11-18 09:52:54 +01:00
|
|
|
void bdrv_debug_event(BlockDriverState *bs, BlkdebugEvent event);
|
2010-03-15 17:27:00 +01:00
|
|
|
|
2012-12-06 14:32:58 +01:00
|
|
|
int bdrv_debug_breakpoint(BlockDriverState *bs, const char *event,
|
|
|
|
const char *tag);
|
2013-11-20 03:01:54 +01:00
|
|
|
int bdrv_debug_remove_breakpoint(BlockDriverState *bs, const char *tag);
|
2012-12-06 14:32:58 +01:00
|
|
|
int bdrv_debug_resume(BlockDriverState *bs, const char *tag);
|
|
|
|
bool bdrv_debug_is_suspended(BlockDriverState *bs, const char *tag);
|
|
|
|
|
2014-05-15 13:22:05 +02:00
|
|
|
/**
|
|
|
|
* bdrv_get_aio_context:
|
|
|
|
*
|
|
|
|
* Returns: the currently bound #AioContext
|
|
|
|
*/
|
|
|
|
AioContext *bdrv_get_aio_context(BlockDriverState *bs);
|
|
|
|
|
2017-04-10 14:09:25 +02:00
|
|
|
/**
|
|
|
|
* Transfer control to @co in the aio context of @bs
|
|
|
|
*/
|
|
|
|
void bdrv_coroutine_enter(BlockDriverState *bs, Coroutine *co);
|
|
|
|
|
2014-05-08 16:34:37 +02:00
|
|
|
/**
|
|
|
|
* bdrv_set_aio_context:
|
|
|
|
*
|
|
|
|
* Changes the #AioContext used for fd handlers, timers, and BHs by this
|
|
|
|
* BlockDriverState and all its children.
|
|
|
|
*
|
block: Forbid bdrv_set_aio_context outside BQL
Even if the caller has both the old and the new AioContext's, there can
be a deadlock, due to the leading bdrv_drain_all.
Suppose there are four io threads (A, B, A0, B0) with A and B owning a
BDS for each (bs_a, bs_b); Now A wants to move bs_a to iothread A0, and
B wants to move bs_b to B0, at the same time:
iothread A iothread B
--------------------------------------------------------------------------
aio_context_acquire(A0) /* OK */ aio_context_acquire(B0) /* OK */
bdrv_set_aio_context(bs_a, A0) bdrv_set_aio_context(bs_b, B0)
-> bdrv_drain_all() -> bdrv_drain_all()
-> acquire A /* OK */ -> acquire A /* blocked */
-> acquire B /* blocked */ -> acquire B
... ...
Deadlock happens because A is waiting for B, and B is waiting for A.
Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <1423969591-23646-2-git-send-email-famz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2015-02-15 04:06:30 +01:00
|
|
|
* This function must be called with iothread lock held.
|
2014-05-08 16:34:37 +02:00
|
|
|
*/
|
|
|
|
void bdrv_set_aio_context(BlockDriverState *bs, AioContext *new_context);
|
2015-02-16 12:47:54 +01:00
|
|
|
int bdrv_probe_blocksizes(BlockDriverState *bs, BlockSizes *bsz);
|
|
|
|
int bdrv_probe_geometry(BlockDriverState *bs, HDGeometry *geo);
|
2014-05-08 16:34:37 +02:00
|
|
|
|
2014-07-04 12:04:33 +02:00
|
|
|
void bdrv_io_plug(BlockDriverState *bs);
|
|
|
|
void bdrv_io_unplug(BlockDriverState *bs);
|
|
|
|
|
2017-04-08 05:34:45 +02:00
|
|
|
/**
|
|
|
|
* bdrv_parent_drained_begin:
|
|
|
|
*
|
|
|
|
* Begin a quiesced section of all users of @bs. This is part of
|
|
|
|
* bdrv_drained_begin.
|
|
|
|
*/
|
2018-05-29 17:17:45 +02:00
|
|
|
void bdrv_parent_drained_begin(BlockDriverState *bs, BdrvChild *ignore,
|
|
|
|
bool ignore_bds_parents);
|
2017-04-08 05:34:45 +02:00
|
|
|
|
2018-06-29 18:01:31 +02:00
|
|
|
/**
|
|
|
|
* 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);
|
|
|
|
|
2017-04-08 05:34:45 +02:00
|
|
|
/**
|
|
|
|
* bdrv_parent_drained_end:
|
|
|
|
*
|
|
|
|
* End a quiesced section of all users of @bs. This is part of
|
|
|
|
* bdrv_drained_end.
|
|
|
|
*/
|
2018-05-29 17:17:45 +02:00
|
|
|
void bdrv_parent_drained_end(BlockDriverState *bs, BdrvChild *ignore,
|
|
|
|
bool ignore_bds_parents);
|
2017-04-08 05:34:45 +02:00
|
|
|
|
2018-03-22 14:11:20 +01:00
|
|
|
/**
|
|
|
|
* bdrv_drain_poll:
|
|
|
|
*
|
2018-03-23 12:40:41 +01:00
|
|
|
* Poll for pending requests in @bs, its parents (except for @ignore_parent),
|
2018-05-29 17:17:45 +02:00
|
|
|
* 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).
|
2018-03-23 12:40:41 +01:00
|
|
|
*
|
|
|
|
* This is part of bdrv_drained_begin.
|
2018-03-22 14:11:20 +01:00
|
|
|
*/
|
2018-03-23 12:40:41 +01:00
|
|
|
bool bdrv_drain_poll(BlockDriverState *bs, bool recursive,
|
2018-05-29 17:17:45 +02:00
|
|
|
BdrvChild *ignore_parent, bool ignore_bds_parents);
|
2018-03-22 14:11:20 +01:00
|
|
|
|
2015-10-23 05:08:09 +02:00
|
|
|
/**
|
|
|
|
* 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);
|
|
|
|
|
2018-03-23 15:57:20 +01:00
|
|
|
/**
|
|
|
|
* 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,
|
2018-05-29 17:17:45 +02:00
|
|
|
BdrvChild *parent, bool ignore_bds_parents);
|
2018-03-23 15:57:20 +01:00
|
|
|
|
2017-12-06 17:05:44 +01:00
|
|
|
/**
|
|
|
|
* 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);
|
|
|
|
|
2015-10-23 05:08:09 +02:00
|
|
|
/**
|
|
|
|
* bdrv_drained_end:
|
|
|
|
*
|
|
|
|
* End a quiescent section started by bdrv_drained_begin().
|
|
|
|
*/
|
|
|
|
void bdrv_drained_end(BlockDriverState *bs);
|
|
|
|
|
2017-12-06 17:05:44 +01:00
|
|
|
/**
|
|
|
|
* End a quiescent section started by bdrv_subtree_drained_begin().
|
|
|
|
*/
|
|
|
|
void bdrv_subtree_drained_end(BlockDriverState *bs);
|
|
|
|
|
2016-05-10 09:36:37 +02:00
|
|
|
void bdrv_add_child(BlockDriverState *parent, BlockDriverState *child,
|
|
|
|
Error **errp);
|
|
|
|
void bdrv_del_child(BlockDriverState *parent, BdrvChild *child, Error **errp);
|
|
|
|
|
2017-06-28 14:05:21 +02:00
|
|
|
bool bdrv_can_store_new_dirty_bitmap(BlockDriverState *bs, const char *name,
|
|
|
|
uint32_t granularity, Error **errp);
|
2018-01-16 07:08:56 +01:00
|
|
|
/**
|
|
|
|
*
|
|
|
|
* 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);
|
2018-06-01 11:26:39 +02:00
|
|
|
|
|
|
|
/**
|
|
|
|
*
|
|
|
|
* 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
|
2018-07-03 04:37:57 +02:00
|
|
|
* @flags: request flags. Supported flags:
|
2018-06-01 11:26:39 +02:00
|
|
|
* 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.
|
2018-07-03 04:37:57 +02:00
|
|
|
* BDRV_REQ_NO_SERIALISING - do not serialize with other overlapping
|
|
|
|
* requests currently in flight.
|
2018-06-01 11:26:39 +02:00
|
|
|
*
|
|
|
|
* 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,
|
2018-07-09 18:37:17 +02:00
|
|
|
uint64_t bytes, BdrvRequestFlags read_flags,
|
|
|
|
BdrvRequestFlags write_flags);
|
2012-07-10 11:12:40 +02:00
|
|
|
#endif
|