The main goal of this series is to get the I/O tests passing
100% with LUKS when run with './check -luks'. It also adds a
few more combinations to the LUKS/dmcrypt interoperability
test.

To make LUKS testing not quite as slow, we drop the PBKDF
iteration count down to a very small value. This doesn't
remove all overhead, as formatting the volume will always
measure PBKDF timing over a 1 second interval.

Changed in v6:

 - Fix quoting of $TEST_IMG (Eric/Max)
 - Remove accidental use of tabs (Max)
 - Drop unused SYSEMU_EXTRA_ARGS (Max)
 - Remove accidental import of 'time' module (Max)

Changed in v5:

 - Rebase to kevin/block  git tree instead of master
 - Fix LUKS compat in new test 181

Changed in v4:

 - Fix misc mistakes in syntax conversion (Max)
 - Drop changes to 120 & mark it unsupported (Max)
 - Use _unsupported_fmt (Max)

Changed in v3:

 - Fix some typos in commit message(s) (Eric)

Changed in v2:

 - Split off patch that change check.time recording since
   it was not a direct dependancy
 - Skip new 159 & 170 tests which don't work due to qemu-img
   dd limitations

Daniel P. Berrange (5):
  iotests: skip 159 & 170 with luks format
  iotests: fix remainining tests to work with LUKS
  iotests: reduce PBKDF iterations when testing LUKS
  iotests: add more LUKS hash combination tests
  iotests: chown LUKS device before qemu-io launches

 tests/qemu-iotests/033           |   12 +-
 tests/qemu-iotests/120           |    1 +
 tests/qemu-iotests/140           |    9 +-
 tests/qemu-iotests/145           |   19 +-
 tests/qemu-iotests/149           |   25 +-
 tests/qemu-iotests/149.out       | 1002 ++++++++++++++++++++++++++++----------
 tests/qemu-iotests/157           |   17 +-
 tests/qemu-iotests/157.out       |   16 +-
 tests/qemu-iotests/159           |    1 +
 tests/qemu-iotests/170           |    1 +
 tests/qemu-iotests/174           |    2 +-
 tests/qemu-iotests/181           |   21 +-
 tests/qemu-iotests/common.filter |    3 +-
 tests/qemu-iotests/common.qemu   |    9 +-
 tests/qemu-iotests/common.rc     |    3 +
 15 files changed, 841 insertions(+), 300 deletions(-)

-- 
2.9.3

Hi,

This series seems to have some coding style problems. See output below for
more information:

Message-id: 20170626123510.20134-1-berrange@redhat.com
Type: series
Subject: [Qemu-devel] [PATCH v6 0/5] Improve I/O tests coverage of LUKS driver

=== TEST SCRIPT BEGIN ===
#!/bin/bash

BASE=base
n=1
total=$(git log --oneline $BASE.. | wc -l)
failed=0

git config --local diff.renamelimit 0
git config --local diff.renames True

commits="$(git log --format=%H --reverse $BASE..)"
for c in $commits; do
    echo "Checking PATCH $n/$total: $(git log -n 1 --format=%s $c)..."
    if ! git show $c --format=email | ./scripts/checkpatch.pl --mailback -; then
        failed=1
        echo
    fi
    n=$((n+1))
done

exit $failed
=== TEST SCRIPT END ===

Updating 3c8cf5a9c21ff8782164d1def7f44bd888713384
Traceback (most recent call last):
  File "./patchew-cli", line 430, in test_one
    git_clone_repo(clone, r["repo"], r["head"], logf)
  File "./patchew-cli", line 48, in git_clone_repo
    stdout=logf, stderr=logf)
  File "/usr/lib64/python3.5/subprocess.py", line 266, in check_call
    retcode = call(*popenargs, **kwargs)
  File "/usr/lib64/python3.5/subprocess.py", line 249, in call
    return p.wait(timeout=timeout)
  File "/usr/lib64/python3.5/subprocess.py", line 1389, in wait
    (pid, sts) = self._try_wait(0)
  File "/usr/lib64/python3.5/subprocess.py", line 1339, in _try_wait
    (pid, sts) = os.waitpid(self.pid, wait_flags)
KeyboardInterrupt



---
Email generated automatically by Patchew [http://patchew.org/].
Please send your feedback to patchew-devel@freelists.org

On Tue, 06/27 07:43, no-reply@patchew.org wrote:
> Updating 3c8cf5a9c21ff8782164d1def7f44bd888713384
> Traceback (most recent call last):
>   File "./patchew-cli", line 430, in test_one
>     git_clone_repo(clone, r["repo"], r["head"], logf)
>   File "./patchew-cli", line 48, in git_clone_repo
>     stdout=logf, stderr=logf)
>   File "/usr/lib64/python3.5/subprocess.py", line 266, in check_call
>     retcode = call(*popenargs, **kwargs)
>   File "/usr/lib64/python3.5/subprocess.py", line 249, in call
>     return p.wait(timeout=timeout)
>   File "/usr/lib64/python3.5/subprocess.py", line 1389, in wait
>     (pid, sts) = self._try_wait(0)
>   File "/usr/lib64/python3.5/subprocess.py", line 1339, in _try_wait
>     (pid, sts) = os.waitpid(self.pid, wait_flags)
> KeyboardInterrupt

My mistake, sorry for the noise.

Fam

On 2017-06-26 14:35, Daniel P. Berrange wrote:
> The main goal of this series is to get the I/O tests passing
> 100% with LUKS when run with './check -luks'. It also adds a
> few more combinations to the LUKS/dmcrypt interoperability
> test.
> 
> To make LUKS testing not quite as slow, we drop the PBKDF
> iteration count down to a very small value. This doesn't
> remove all overhead, as formatting the volume will always
> measure PBKDF timing over a 1 second interval.

Thanks, applied to my block branch:

https://github.com/XanClic/qemu/commits/block

Max