Re: [PATCH 11/14] iotests: bash tests: filter compression type - Vladimir Sementsov-Ogievskiy

From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
To: Max Reitz <mreitz@redhat.com>, qemu-block@nongnu.org
Cc: qemu-devel@nongnu.org, kwolf@redhat.com, den@openvz.org,
	jsnow@redhat.com
Subject: Re: [PATCH 11/14] iotests: bash tests: filter compression type
Date: Fri, 16 Jul 2021 17:35:21 +0300	[thread overview]
Message-ID: <4cf48c9b-8d6d-1a87-4712-a405a8fbdbcc@virtuozzo.com> (raw)
In-Reply-To: <9646362b-c01f-3ed8-b5c4-4fdefd9786eb@redhat.com>

16.07.2021 16:17, Max Reitz wrote:
> On 05.07.21 11:15, Vladimir Sementsov-Ogievskiy wrote:
>> We want iotests pass with both the default zlib compression and with
>> IMGOPTS='compression_type=zstd'.
>>
>> Actually the only test that is interested in real compression type in
>> test output is 287 (test for qcow2 compression type), so implement
>> specific option for it.
>>
>> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
>> ---
>>   tests/qemu-iotests/060.out       |  2 +-
>>   tests/qemu-iotests/061.out       | 12 ++++++------
>>   tests/qemu-iotests/082.out       | 14 +++++++-------
>>   tests/qemu-iotests/198.out       |  4 ++--
>>   tests/qemu-iotests/287           |  8 ++++----
>>   tests/qemu-iotests/common.filter |  7 +++++++
>>   tests/qemu-iotests/common.rc     | 14 +++++++++++++-
>>   7 files changed, 40 insertions(+), 21 deletions(-)
> 
> [...]
> 
>> diff --git a/tests/qemu-iotests/common.filter b/tests/qemu-iotests/common.filter
>> index 268b749e2f..78efe3e4dd 100644
>> --- a/tests/qemu-iotests/common.filter
>> +++ b/tests/qemu-iotests/common.filter
>> @@ -247,6 +247,7 @@ _filter_img_info()
>>           -e "/block_state_zero: \\(on\\|off\\)/d" \
>>           -e "/log_size: [0-9]\\+/d" \
>>           -e "s/iters: [0-9]\\+/iters: 1024/" \
>> +        -e 's/\(compression type: \)\(zlib\|zstd\)/\1COMPRESSION_TYPE/' \
>>           -e "s/uuid: [-a-f0-9]\\+/uuid: 00000000-0000-0000-0000-000000000000/" | \
>>       while IFS='' read -r line; do
>>           if [[ $format_specific == 1 ]]; then
>> @@ -332,5 +333,11 @@ for fname in fnames:
>>   sys.stdout.write(result)'
>>   }
>> +_filter_qcow2_compression_type_bit()
>> +{
>> +    $SED -e 's/\(incompatible_features\s\+\)\[3\(, \)\?/\1[/' \
>> +         -e 's/\(incompatible_features.*\), 3\]/\1]/'
> 
> What about “incompatble_features   [2, 3, 4]”?

Will add.

> 
> I’d like to propose adding some form of filtering parameter to qcow2.py which allows filtering a specific bit from the qcow2_format.Flags64 representation, but that seems rather difficult, actually...
> 
>> +}
>> +
>>   # make sure this script returns success
>>   true
>> diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
>> index ee4b9d795e..813b51ee03 100644
>> --- a/tests/qemu-iotests/common.rc
>> +++ b/tests/qemu-iotests/common.rc
>> @@ -697,6 +697,7 @@ _img_info()
>>               -e "s#$TEST_DIR#TEST_DIR#g" \
>>               -e "s#$SOCK_DIR/fuse-#TEST_DIR/#g" \
>>               -e "s#$IMGFMT#IMGFMT#g" \
>> +            -e 's/\(compression type: \)\(zlib\|zstd\)/\1COMPRESSION_TYPE/' \
>>               -e "/^disk size:/ D" \
>>               -e "/actual-size/ D" | \
>>           while IFS='' read -r line; do
>> @@ -996,12 +997,23 @@ _require_one_device_of()
>>   _qcow2_dump_header()
>>   {
>> +    if [[ "$1" == "--no-filter-compression" ]]; then
>> +        local filter_compression=0
>> +        shift
>> +    else
>> +        local filter_compression=1
>> +    fi
>> +
>>       img="$1"
>>       if [ -z "$img" ]; then
>>           img="$TEST_IMG"
>>       fi
>> -    $PYTHON qcow2.py "$img" dump-header
>> +    if [[ $filter_compression == 0 ]]; then
>> +        $PYTHON qcow2.py "$img" dump-header
>> +    else
>> +        $PYTHON qcow2.py "$img" dump-header | _filter_qcow2_compression_type_bit
>> +    fi
>>   }
>>   # make sure this script returns success
> 
> Could have been done more extensibly for the future (i.e. a loop over the parameters, and a variable to invoke all applicable filters), but, well.  Not much reason to think about a future that we’re not sure will ever happen.
> 
> Max
> 

-- 
Best regards,
Vladimir