Am 10.10.2014 um 14:29 schrieb Benoît Canet:
On Fri, Aug 29, 2014 at 11:40:53PM +0200, Max Reitz wrote:
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 <mre...@redhat.com>
---
block/qcow2.c | 2 ++
block/qcow2.h | 2 ++
2 files changed, 4 insertions(+)
diff --git a/block/qcow2.c b/block/qcow2.c
index f9e045f..172ad00 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -689,6 +689,8 @@ static int qcow2_open(BlockDriverState *bs, QDict *options,
int flags,
s->l2_bits = s->cluster_bits - 3; /* L2 is always one cluster */
s->l2_size = 1 << s->l2_bits;
+ s->refcount_block_bits = s->cluster_bits - (s->refcount_order - 3);
After carefull examination (s->refcount_order - 3) == REFCOUNT_SHIFT.
Why not also creating s->refcount_shift and make use of it in order to avoid
torturing the mind of the next reader.
I'm sorry. *g*
Well, I'm against creating s->refcount_shift, because that name implies
you could use it for shifts. But you shouldn't do that, because it may
be both negative and positive.
Or simply recall in a comment that there is 2^3 bits in a byte.
I like this more. :-)
Max
Best regards
Benoît
+ s->refcount_block_size = 1 << s->refcount_block_bits;
bs->total_sectors = header.size / 512;
s->csize_shift = (62 - (s->cluster_bits - 8));
s->csize_mask = (1 << (s->cluster_bits - 8)) - 1;
diff --git a/block/qcow2.h b/block/qcow2.h
index 6aeb7ea..7c01fb7 100644
--- a/block/qcow2.h
+++ b/block/qcow2.h
@@ -222,6 +222,8 @@ typedef struct BDRVQcowState {
int l2_size;
int l1_size;
int l1_vm_state_index;
+ int refcount_block_bits;
+ int refcount_block_size;
int csize_shift;
int csize_mask;
uint64_t cluster_offset_mask;
--
2.1.0