lean which should be set
to true whenever a fix is rather dangerous or too complicated using the
current refcount structures. Another example for this is refcount blocks
being referenced more than once.
Signed-off-by: Max Reitz
Reviewed-by: Benoît Canet
---
block/qcow2-refcou
On 20.10.2014 at 16:25, Kevin Wolf wrote:
Am 29.08.2014 um 23:40 hat Max Reitz geschrieben:
The size of a refblock entry is (in theory) variable; calculate
therefore the number of entries per refblock and the according bit shift
(1 << x == entry count) when opening an image.
Signed-off-b
refblocks beyond the image
end by resizing the image accordingly. Add several tests for this as
well.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
---
tests/qemu-iotests/108 | 141 +
tests/qemu-iotests/108.out | 110
x27;,
- 'rd_total_time_ns': 'int', 'wr_highest_offset': 'int' } }
+ 'rd_total_time_ns': 'int', 'wr_highest_offset': 'int',
+ 'rd_merged': 'int', 'wr_merged': 'int' } }
##
# @BlockStats:
With the hunk removing the empty line removed (I know that line should
not be there, but there's no point in removing it in this patch) and
with either 2.2 or 2.3 in block-core.json:
Reviewed-by: Max Reitz
;
-
bs->enable_write_cache = !!(flags & BDRV_O_CACHE_WB);
/* Open the image, either directly or using a protocol */
I know it's not even your patch originally, but I still don't like
spurious empty line removals, whoever might be responsible for them.
@@ -1010,6 +1035,8 @@ free_and_fail:
g_free(bs->opaque);
bs->opaque = NULL;
bs->drv = NULL;
+fail_opts:
+qemu_opts_del(opts);
return ret;
}
With that empty line removal hunk removed:
Reviewed-by: Max Reitz
-- a/qmp-commands.hx
+++ b/qmp-commands.hx
@@ -2104,6 +2104,7 @@ Each json-object contain the following:
- "iops_size": I/O size when limiting by iops (json-int)
- "detect_zeroes": detect and optimize zero writing (json-string)
- Possible values: "off", "on", "unmap"
+ - "write_merging": enable merging of write requests (json-bool)
- "image": the detail of the image, it is a json-object containing
the following:
- "filename": image file name (json-string)
@@ -2181,6 +2182,7 @@ Example:
"iops_wr_max": 0,
"iops_size": 0,
"detect_zeroes": "on",
+ "write_merging": "true",
"image":{
"filename":"disks/test.qcow2",
"format":"qcow2",
Reviewed-by: Max Reitz
simply MAX_MERGE_REQS or, more
complicated, MAX_MERGE_REQ_COUNT. But I'm me, and this mail's "From"
field says it's not from me.
Reviewed-by: Max Reitz
to disable multiwrite_merge
hw/virtio-blk: add a constant for max number of merged requests
In addition, I'd like a test for this (just the parameter would be
enough). But I don't object to this series without it.
Max
On 2014-10-20 at 22:48, Peter Lieven wrote:
Am 20.10.2014 um 17:56 schrieb Max Reitz :
On 20.10.2014 at 16:35, Peter Lieven wrote:
This adds some preparing patches for upcoming multiwrite modifications.
I will leave the dangerous patches for after 2.2 release.
Peter Lieven (4):
block: add
On 2014-10-20 at 18:44, Kevin Wolf wrote:
Am 20.10.2014 um 16:35 hat Max Reitz geschrieben:
If the qcow2 check function detects a refcount block located beyond the
image end, grow the image appropriately. This cannot break anything and
is the logical fix for such a case.
Signed-off-by: Max
mu to allocate and then write about 2G of data; yes, test 1 in 084
fails for me because qemu cannot allocate 4G for the bmap.
In fact, I can only test this once I'm home where I have more RAM
available (I made the mistake of activating swap space to test this only
once).
Max Reitz (2):
block/vdi:
There are macros for these operations, so make use of them.
Signed-off-by: Max Reitz
---
block/vdi.c | 9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/block/vdi.c b/block/vdi.c
index 9604721..19701ee 100644
--- a/block/vdi.c
+++ b/block/vdi.c
@@ -407,8 +407,7
()+bdrv_flush() instead of bdrv_pwrite_sync().
See: https://bugzilla.redhat.com/show_bug.cgi?id=1154940
Signed-off-by: Max Reitz
---
block/vdi.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/block/vdi.c b/block/vdi.c
index 19701ee..322efcd 100644
--- a/block/vdi.c
+++ b
On 2014-10-21 at 10:01, Peter Lieven wrote:
On 21.10.2014 09:06, Max Reitz wrote:
On 2014-10-20 at 22:48, Peter Lieven wrote:
Am 20.10.2014 um 17:56 schrieb Max Reitz :
On 20.10.2014 at 16:35, Peter Lieven wrote:
This adds some preparing patches for upcoming multiwrite
modifications.
I
on. g_free(l1_table); is enough.
return ret;
}
The change itself is good, it just needs some polishing.
Max
aio_write failed: Input/output error
Thanks for your report,
Max
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1353456
Title:
qemu-io: Failure on a qcow2 image with the fuzzed refcount table
Stat
On 2014-10-21 at 11:31, Kevin Wolf wrote:
Am 20.10.2014 um 16:35 hat Max Reitz geschrieben:
The previous commit introduced the "rebuild" variable to qcow2's
implementation of the image consistency check. Now make use of this by
adding a function which creates a completely new ref
On 2014-10-21 at 11:38, Kevin Wolf wrote:
Am 21.10.2014 um 11:07 hat Max Reitz geschrieben:
On 2014-10-21 at 10:01, Peter Lieven wrote:
On 21.10.2014 09:06, Max Reitz wrote:
On 2014-10-20 at 22:48, Peter Lieven wrote:
Am 20.10.2014 um 17:56 schrieb Max Reitz :
On 20.10.2014 at 16:35, Peter
On 2014-10-21 at 09:52, Kevin Wolf wrote:
Am 20.10.2014 um 16:35 hat Max Reitz geschrieben:
If a referenced cluster has a refcount of 0, increasing its refcount may
result in clusters being allocated for the refcount structures. This may
overwrite the referenced cluster, therefore we cannot
On 2014-10-21 at 11:59, Kevin Wolf wrote:
Am 20.10.2014 um 16:35 hat Max Reitz geschrieben:
Because the old refcount structure will be leaked after having rebuilt
it, we need to recalculate the refcounts and run a leak-fixing operation
afterwards (if leaks should be fixed at all).
Signed-off
;& l1_table ==
NULL)" conditional block.
Well, okay, I don't know about the callers of qcow2_process_discards(),
so they may have set s->cache_discards to true and then expect this
function to always call qcow2_process_discards() and set
s->cache_discards to false. Okay
** Changed in: qemu
Status: New => Fix Committed
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1354529
Title:
qemu-io: Assert failure on the fuzzed qcow2 image
Status in QEMU:
Fix Committ
** Changed in: qemu
Status: New => Fix Committed
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1353456
Title:
qemu-io: Failure on a qcow2 image with the fuzzed refcount table
Status in QEMU
n the repair function. But this
image is broken in a way that there's no real way to fix it. The best we
could do is ask the user to use qemu-img convert and then hope for the
best. I'll just mark this as fixed.
Max
** Changed in: qemu
Status: New => Fix Committed
--
You
e_clusters failed: Invalid argument
qcow2: Marking image as corrupt: Data cluster offset 0xfe00 unaligned
(guest offset: 0x2e1000); further corruption events will be suppressed
write failed: Input/output error
Thanks for your report (and your fuzzer),
Max
--
You received this bug notification b
Image end offset: 2883584
I'm still working towards the repair function actually doing its job.
Thank you for your report,
Max
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1355697
Title:
qem
s all we can do.
Therefore, I'm marking this fixed as well.
Max
** Changed in: qemu
Status: New => Fix Committed
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1355697
Title:
qemu-img: Seg
tefan Hajnoczi
---
blockjob.c | 45 +
include/block/blockjob.h | 19 +++
2 files changed, 64 insertions(+)
Reviewed-by: Max Reitz
/commit.c | 70 --
blockdev.c | 29
2 files changed, 64 insertions(+), 35 deletions(-)
Reviewed-by: Max Reitz
level driver that it needs
+ * to copy the buffer.
*/
-if ((bs->open_flags & BDRV_O_NOCACHE)) {
+if (s->needs_alignement) {
if (!bdrv_qiov_is_aligned(bs, qiov)) {
type |= QEMU_AIO_MISALIGNED;
#ifdef CONFIG_LINUX_AIO
With s/alignement/alignment/ and the
you are
modifying the current L1 table, you need to copy the result back to
s->l1_table.
Sorry for having missed these in my previous review.
Max
}
-if (l1_allocated)
-g_free(l1_table);
+g_free(l1_table);
return ret;
}
here.
OK, I will replace g_try_malloc0(ROUND_UP(l1_size2, BDRV_SECTOR_SIZE)) with
qemu_try_blockalign(bs, ROUND_UP(l1_size2, BDRV_SECTOR_SIZE)).
You can actually omit the ROUND_UP(). There are no accesses beyond
l1_table[l1_size - 1], if I'm not mistaken.
Max
(and we should be usin
On 2014-10-21 at 15:10, Max Reitz wrote:
On 2014-10-21 at 15:08, Zhang Haoyu wrote:
Use local variable to bdrv_pwrite_sync L1 table,
needless to make conversion of cached L1 table between
big-endian and host style.
Signed-off-by: Zhang Haoyu
---
v1 -> v2:
- remove the superflous assignm
On 2014-10-21 at 16:12, Kevin Wolf wrote:
Am 20.10.2014 um 16:35 hat Max Reitz geschrieben:
There are certain cases where repairing a qcow2 image might actually
damage it further (or rather, where repairing it has in fact damaged it
further with the old qcow2 check implementation). This should
On 2014-10-21 at 12:16, Max Reitz wrote:
On 2014-10-21 at 11:59, Kevin Wolf wrote:
Am 20.10.2014 um 16:35 hat Max Reitz geschrieben:
Because the old refcount structure will be leaked after having rebuilt
it, we need to recalculate the refcounts and run a leak-fixing
operation
afterwards (if
On 2014-10-21 at 17:11, Kevin Wolf wrote:
Am 21.10.2014 um 16:55 hat Max Reitz geschrieben:
On 2014-10-21 at 12:16, Max Reitz wrote:
On 2014-10-21 at 11:59, Kevin Wolf wrote:
Am 20.10.2014 um 16:35 hat Max Reitz geschrieben:
Because the old refcount structure will be leaked after having
On 2014-10-20 at 16:48, Kevin Wolf wrote:
Am 20.10.2014 um 16:39 hat Max Reitz geschrieben:
On 20.10.2014 at 16:25, Kevin Wolf wrote:
Am 29.08.2014 um 23:40 hat Max Reitz geschrieben:
The size of a refblock entry is (in theory) variable; calculate
therefore the number of entries per refblock
- fix file names that were changed by the commit
b707ab7 hw/xtensa: remove extraneous xtensa_ prefix from file names
- mark OpenCores 10/100 Mbit MAC model as maintained.
Signed-off-by: Max Filippov
---
MAINTAINERS | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a
least one distribution already having implemented this).
If you add a new driver, you either have to distribute its source so
people can build qemu themselves with the driver included; or you build
qemu yourself and then distribute the binary; or you find your driver
good enough to submit it for upstream qemu.
Max
008/13:[0007] [FC] 'qcow2: Fix refcount blocks beyond image end'
009/13:[0012] [FC] 'qcow2: Do not perform potentially damaging repairs'
010/13:[0082] [FC] 'qcow2: Rebuild refcount structure during check'
011/13:[0016] [FC] 'qcow2: Clean up after refcount rebuild
Put the code for calculating the reference counts and comparing them
during qemu-img check into own functions.
Signed-off-by: Max Reitz
Reviewed-by: Benoît Canet
Reviewed-by: Kevin Wolf
---
block/qcow2-refcount.c | 153 -
1 file changed, 102
These functions call their non-0-counterparts and then fill the
allocated buffer with 0 (if the allocation has been successful).
Signed-off-by: Max Reitz
---
block.c | 16
include/block/block.h | 2 ++
2 files changed, 18 insertions(+)
diff --git a/block.c b
If the qcow2 check function detects a refcount block located beyond the
image end, grow the image appropriately. This cannot break anything and
is the logical fix for such a case.
Signed-off-by: Max Reitz
---
block/qcow2-refcount.c | 63 ++
1 file
When opening dirty images, qcow2's repair function should not only
repair errors but leaks as well.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
Reviewed-by: Benoît Canet
Reviewed-by: Kevin Wolf
---
block/qcow2.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/
The size of a refblock entry is (in theory) variable; calculate
therefore the number of entries per refblock and the according bit shift
(1 << x == entry count) when opening an image.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
---
block/qcow2.c | 3 +++
block/qcow2.h | 2 ++
2
When implementing variable refcounts, we want to be able to easily find
all the places in qemu which are tied to a certain refcount order.
Replace sizeof(uint16_t) in the check code by sizeof(**refcount_table)
so we can later find it more easily.
Signed-off-by: Max Reitz
---
block/qcow2
Pull check_refblocks() before calculate_refcounts() so we can drop its
static declaration.
Signed-off-by: Max Reitz
Reviewed-by: Benoît Canet
Reviewed-by: Kevin Wolf
---
block/qcow2-refcount.c | 102 -
1 file changed, 49 insertions(+), 53
We will later call calculate_refcounts multiple times, so reuse the
refcount table if possible.
Signed-off-by: Max Reitz
Reviewed-by: Benoît Canet
Reviewed-by: Kevin Wolf
---
block/qcow2-refcount.c | 12 +++-
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/block/qcow2
ructure will be leaked, however. This leak will be
dealt with in a follow-up commit.
Signed-off-by: Max Reitz
---
block/qcow2-refcount.c | 310 -
1 file changed, 304 insertions(+), 6 deletions(-)
diff --git a/block/qcow2-refcount.c b/block/qcow2-r
lean which should be set
to true whenever a fix is rather dangerous or too complicated using the
current refcount structures. Another example for this is refcount blocks
being referenced more than once.
Signed-off-by: Max Reitz
---
block/qcow2-refcou
refblocks beyond the image
end by resizing the image accordingly. Add several tests for this as
well.
Signed-off-by: Max Reitz
---
tests/qemu-iotests/108 | 141 +
tests/qemu-iotests/108.out | 110 +++
tests/qemu-iotests
Because the old refcount structure will be leaked after having rebuilt
it, we need to recalculate the refcounts and run a leak-fixing operation
afterwards (if leaks should be fixed at all).
Signed-off-by: Max Reitz
---
block/qcow2-refcount.c | 45 +
1
rebuild the refcount structure as well.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
---
tests/qemu-iotests/039.out | 10 --
tests/qemu-iotests/060.out | 10 --
tests/qemu-iotests/061.out | 18 --
3 files changed, 28 insertions(+), 10 deletions(-)
diff --git a
On 2014-10-22 at 11:24, Kevin Wolf wrote:
Am 22.10.2014 um 10:08 hat Max Reitz geschrieben:
If the qcow2 check function detects a refcount block located beyond the
image end, grow the image appropriately. This cannot break anything and
is the logical fix for such a case.
Signed-off-by: Max
On 2014-10-22 at 11:14, Kevin Wolf wrote:
Am 22.10.2014 um 10:08 hat Max Reitz geschrieben:
The previous commit introduced the "rebuild" variable to qcow2's
implementation of the image consistency check. Now make use of this by
adding a function which creates a completely new ref
unconditionally.
Signed-off-by: Roger Pau Monné
Reviewed-by: Max Reitz
Cc: Kevin Wolf
Cc: Stefan Hajnoczi
---
Changes since v2:
- s/alignement/alignment/.
- Don't make needs_alignment a bit field.
- Add braces to single statement conditional block.
Changes since v1:
- Intead of appe
On 2014-10-21 at 13:03, Stefan Hajnoczi wrote:
Make sure that query-block-jobs acquires the BlockDriverState
AioContext so that the blockjob isn't running in another thread while we
access its state.
Signed-off-by: Stefan Hajnoczi
Reviewed-by: Max Reitz
---
blockdev.c | 9 +++--
1
On 2014-10-22 at 13:59, Gonglei wrote:
On 2014/10/22 19:45, Zhang Haoyu wrote:
Use local variable to bdrv_pwrite_sync L1 table,
needless to make conversion of cached L1 table between
big-endian and host style.
Signed-off-by: Zhang Haoyu
Reviewed-by: Max Reitz
---
v3 -> v4:
- convert lo
On 2014-10-22 at 14:01, Max Reitz wrote:
On 2014-10-22 at 13:59, Gonglei wrote:
On 2014/10/22 19:45, Zhang Haoyu wrote:
Use local variable to bdrv_pwrite_sync L1 table,
needless to make conversion of cached L1 table between
big-endian and host style.
Signed-off-by: Zhang Haoyu
Reviewed-by
rs'
014/17:[0001] [FC] 'qcow2: Rebuild refcount structure during check'
015/17:[0002] [FC] 'qcow2: Clean up after refcount rebuild'
016/17:[] [--] 'iotests: Fix test outputs'
017/17:[] [--] 'iotests: Add test for potentially damaging repairs'
Max Reitz (
When implementing variable refcounts, we want to be able to easily find
all the places in qemu which are tied to a certain refcount order.
Replace sizeof(uint16_t) in the check code by sizeof(**refcount_table)
so we can later find it more easily.
Signed-off-by: Max Reitz
---
block/qcow2
Put the code for calculating the reference counts and comparing them
during qemu-img check into own functions.
Signed-off-by: Max Reitz
Reviewed-by: Benoît Canet
Reviewed-by: Kevin Wolf
---
block/qcow2-refcount.c | 153 -
1 file changed, 102
The size of a refblock entry is (in theory) variable; calculate
therefore the number of entries per refblock and the according bit shift
(1 << x == entry count) when opening an image.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
---
block/qcow2.c | 3 +++
block/qcow2.h | 2 ++
2
As of a future patch, inc_refcounts() will have to throw errors which
are generally signaled by returning -errno. Therefore, let it return an
integer which is either 0 for success or -errno and handle the -errno
case in all callers.
Signed-off-by: Max Reitz
---
block/qcow2-refcount.c | 91
ructure will be leaked, however. This leak will be
dealt with in a follow-up commit.
Signed-off-by: Max Reitz
---
block/qcow2-refcount.c | 311 -
1 file changed, 305 insertions(+), 6 deletions(-)
diff --git a/block/qcow2-refcount.c b/block/qcow2-r
These functions call their non-0-counterparts and then fill the
allocated buffer with 0 (if the allocation has been successful).
Signed-off-by: Max Reitz
---
block.c | 16
include/block/block.h | 2 ++
2 files changed, 18 insertions(+)
diff --git a/block.c b
When opening dirty images, qcow2's repair function should not only
repair errors but leaks as well.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
Reviewed-by: Benoît Canet
Reviewed-by: Kevin Wolf
---
block/qcow2.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/
Pull check_refblocks() before calculate_refcounts() so we can drop its
static declaration.
Signed-off-by: Max Reitz
Reviewed-by: Benoît Canet
Reviewed-by: Kevin Wolf
---
block/qcow2-refcount.c | 102 -
1 file changed, 49 insertions(+), 53
If the qcow2 check function detects a refcount block located beyond the
image end, grow the image appropriately. This cannot break anything and
is the logical fix for such a case.
Signed-off-by: Max Reitz
---
block/qcow2-refcount.c | 67 +++---
1 file
Instead of printing out an error message, incrementing check_errors and
returning a fixed -errno, just do cleanups and return -ret, with ret set
by the code which threw the exception (jumped to the fail label).
Also, increment check_errors on error in check_refcounts_l2().
Signed-off-by: Max
Use int64_t for the entry count of the in-memory refcount table
throughout the check functions.
Signed-off-by: Max Reitz
---
block/qcow2-refcount.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c
index 24f297f..a3f4d47
Because the old refcount structure will be leaked after having rebuilt
it, we need to recalculate the refcounts and run a leak-fixing operation
afterwards (if leaks should be fixed at all).
Signed-off-by: Max Reitz
---
block/qcow2-refcount.c | 45 +
1
We will later call calculate_refcounts multiple times, so reuse the
refcount table if possible.
Signed-off-by: Max Reitz
Reviewed-by: Benoît Canet
Reviewed-by: Kevin Wolf
---
block/qcow2-refcount.c | 12 +++-
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/block/qcow2
refblocks beyond the image
end by resizing the image accordingly. Add several tests for this as
well.
Signed-off-by: Max Reitz
---
tests/qemu-iotests/108 | 141 +
tests/qemu-iotests/108.out | 110 +++
tests/qemu-iotests
Now that the refcount table can be passed around by reference, do that
for inc_refcounts() (and subsequently check_refcounts_l1() and
check_refcounts_l2()) and use it for resizing it when a cluster after
the image end is encountered.
Signed-off-by: Max Reitz
---
block/qcow2-refcount.c | 57
lean which should be set
to true whenever a fix is rather dangerous or too complicated using the
current refcount structures. Another example for this is refcount blocks
being referenced more than once.
Signed-off-by: Max Reitz
---
block/qcow2-refcou
rebuild the refcount structure as well.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
---
tests/qemu-iotests/039.out | 10 --
tests/qemu-iotests/060.out | 10 --
tests/qemu-iotests/061.out | 18 --
3 files changed, 28 insertions(+), 10 deletions(-)
diff --git a
On 2014-10-22 at 14:21, Gonglei wrote:
On 2014/10/22 20:02, Max Reitz wrote:
On 2014-10-22 at 14:01, Max Reitz wrote:
On 2014-10-22 at 13:59, Gonglei wrote:
On 2014/10/22 19:45, Zhang Haoyu wrote:
Use local variable to bdrv_pwrite_sync L1 table,
needless to make conversion of cached L1
On 2014-10-22 at 14:30, Gonglei wrote:
On 2014/10/22 20:24, Max Reitz wrote:
On 2014-10-22 at 14:21, Gonglei wrote:
On 2014/10/22 20:02, Max Reitz wrote:
On 2014-10-22 at 14:01, Max Reitz wrote:
On 2014-10-22 at 13:59, Gonglei wrote:
On 2014/10/22 19:45, Zhang Haoyu wrote:
Use local
Normally, discarded sectors should read back as zero. However, there are
cases in which a sector (or rather cluster) should be discarded as if
they were never written in the first place, that is, reading them should
fall through to the backing file again.
Signed-off-by: Max Reitz
Reviewed-by
the dirty flag set, which only works for
compat=1.1) and creating a trivial refcount structure.
If there are snapshots or for compat=0.10, fall back to the simple
implementation (discard all clusters).
Signed-off-by: Max Reitz
---
block/blkdebug.c | 2 +
block/qcow2.c | 143
As different image formats most probably map guest addresses to
different host addresses, add a filter to filter the host addresses out;
also, the image filename should be filtered.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
---
tests/qemu-iotests/common.filter | 7 +++
1 file
Implement this function by making all clusters in the image file fall
through to the backing file (by using the recently extended discard).
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
Reviewed-by: Kevin Wolf
---
block/qcow2.c | 27 +++
1 file changed, 27
it like QMP'
009/14:[] [-C] 'qemu-img: Empty image after commit'
010/14:[] [-C] 'qemu-img: Enable progress output for commit'
011/14:[] [-C] 'qemu-img: Specify backing file for commit'
012/14:[] [-C] 'iotests: Add _filter_qemu_img_map'
er in fact desirable.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
---
block/mirror.c | 34 ++
1 file changed, 22 insertions(+), 12 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index e8a43eb..2a1acfe 100644
--- a/block/mirror.c
+++ b/block/mirror.
function from iotests.py, the same
applies there as well which in turn affects tests 030, 055 and 056. On
the other hand, a block job's length does not have to be related to the
length of the image file in the first place, so that check was
questionable anyway.
Signed-off-by: Max Reitz
Reviewed-by:
Implement block_job_complete_sync() by doing the exact same thing as
block_job_cancel_sync() does, only with calling block_job_complete()
instead of block_job_cancel().
Signed-off-by: Max Reitz
Reviewed-by: Kevin Wolf
---
blockjob.c | 39
block-commit by using commit_active_start() and then
waiting for the block job to finish.
Signed-off-by: Max Reitz
---
block/Makefile.objs | 3 +-
qemu-img.c | 82 -
2 files changed, 64 insertions(+), 21 deletions(-)
diff --git a/
ck
job may be completed.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
---
blockjob.c | 3 +++
include/block/blockjob.h | 5 +
qapi/block-core.json | 4 +++-
3 files changed, 11 insertions(+), 1 deletion(-)
diff --git a/blockjob.c b/blockjob.c
index a7d57e3..448b9ce 10
Implement progress output for the commit command by querying the
progress of the block job.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
---
qemu-img-cmds.hx | 4 ++--
qemu-img.c | 24 ++--
qemu-img.texi| 2 +-
3 files changed, 25 insertions(+), 5 deletions
Introduce a new parameter for qemu-img commit which may be used to
explicitly specify the backing file into which an image should be
committed if the backing chain has more than a single layer.
Signed-off-by: Max Reitz
---
qemu-img-cmds.hx | 4 ++--
qemu-img.c | 24
After the top image has been committed, it should be emptied unless
specified otherwise.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
---
qemu-img-cmds.hx | 4 ++--
qemu-img.c | 34 +++---
qemu-img.texi| 6 +-
3 files changed, 38 insertions(+), 6
Add a test for qemu-img commit on backing chains with more than two
images. This test also checks whether the top image is emptied (unless
this is prevented by specifying either -d or -b) and does therefore not
work for qed and vmdk which requires it to be separate from 020.
Signed-off-by: Max
Add a test for qcow2's fast bdrv_make_empty implementation on images
without internal snapshots.
Signed-off-by: Max Reitz
---
tests/qemu-iotests/098 | 79 ++
tests/qemu-iotests/098.out | 45 ++
tests/qemu-iotests/
On 2014-10-22 at 15:24, Max Reitz wrote:
Patch 2:
The bdrv_is_allocated() functions may return a number of zero sectors
e.g. if a sector beyond the image end has been queried. Respect this
case in qemu-io's map implementation so it doesn't run into an infinite
loop (https://bugs.lau
. Therefore, this test does no longer require
_filter_qemu_img_map() (because there are no mappings shown) and this
series no longer depends on the qemu-img/QMP commit series.
Max Reitz (3):
block: Respect underlying file's EOF
qemu-io: Respect early image end for map
iotests:
: Max Reitz
---
block.c | 16 ++--
1 file changed, 14 insertions(+), 2 deletions(-)
diff --git a/block.c b/block.c
index 773e87e..68dc11d 100644
--- a/block.c
+++ b/block.c
@@ -3954,13 +3954,25 @@ static int64_t coroutine_fn
bdrv_co_get_block_status(BlockDriverState *bs,
if (bs
Add a test for qemu-img map and qemu-io -c map on truncated files.
Signed-off-by: Max Reitz
---
tests/qemu-iotests/102 | 64 ++
tests/qemu-iotests/102.out | 10
tests/qemu-iotests/group | 1 +
3 files changed, 75 insertions(+)
create
bdrv_is_allocated() may report zero clusters which most probably means
the image (file) is shorter than expected. Respect this case in order to
avoid an infinite loop.
Signed-off-by: Max Reitz
Reviewed-by: Eric Blake
Reviewed-by: Benoît Canet
---
qemu-io-cmds.c | 5 -
1 file changed, 4
On 2014-10-22 at 15:40, Kevin Wolf wrote:
Am 22.10.2014 um 15:24 hat Max Reitz geschrieben:
When falling through to the underlying file in
bdrv_co_get_block_status(), if it returns that the query offset is
beyond the file end (by setting *pnum to 0), return the range to be
zero and do not let
701 - 800 of 14987 matches
Mail list logo
