Re: [Qemu-devel] [PATCHv3 1/2] qemu-img: find the highest offset in use during check

[Kevin Wolf]

Am 21.01.2013 17:27, schrieb Eric Blake:
> On 01/21/2013 07:25 AM, Federico Simoncelli wrote:
>> This patch adds the support for reporting the highest offset in use by
>> an image. This is particularly useful after a conversion (or a rebase)
>> where the destination is a block device in order to find the actual
>> amount of space in use.
>>
>> Signed-off-by: Federico Simoncelli <address@hidden>
>> ---
>>  block/qcow2-refcount.c       |   10 ++++++++--
>>  include/block/block.h        |    1 +
>>  qemu-img.c                   |    4 ++++
>>  tests/qemu-iotests/026       |    6 +++---
>>  tests/qemu-iotests/036       |    3 ++-
>>  tests/qemu-iotests/039       |    2 +-
>>  tests/qemu-iotests/044.out   |    1 +
>>  tests/qemu-iotests/common.rc |    5 +++--
>>  8 files changed, 23 insertions(+), 9 deletions(-)
>>
> 
>> @@ -1231,6 +1236,7 @@ int qcow2_check_refcounts(BlockDriverState *bs, 
>> BdrvCheckResult *res,
>>          }
>>      }
>>  
>> +    res->highest_offset = (highest_cluster + 1) * s->cluster_size;
> 
> That's actually one byte larger than the actual highest offset in use.
> That is, if I have a file with exactly 4096 bytes, this computation
> would return 4096 (the file size) rather than 4095 (the last offset in use).
> 
>> +++ b/qemu-img.c
>> @@ -475,6 +475,10 @@ static int img_check(int argc, char **argv)
>>          result.bfi.fragmented_clusters * 100.0 / 
>> result.bfi.allocated_clusters);
>>      }
>>  
>> +    if (result.highest_offset > 0) {
>> +        printf("Highest offset in use: %" PRId64 "\n", 
>> result.highest_offset);
> 
> This output message feels off by one.  Either you need to subtract 1
> from res->highest_offset to get the address of the last used byte, or
> you need to document it as the first unused byte, or instead of calling
> it 'highest offset', you should call it 'used bytes' (except that with
> sparse files, it's hard to argue that all earlier bytes were 'used').

Good point. I think the number is what we wanted, and what users are
interested in is probably "used bytes" rather than "first unused byte".
Maybe we can find a better word for "used" (it has the same problem in
all three contexts), but I can't think of one off the top of my head.

Kevin