qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Max Reitz <mreitz@redhat.com>
To: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>,
	qemu-block@nongnu.org
Cc: kwolf@redhat.com, den@openvz.org, jsnow@redhat.com,
	qemu-devel@nongnu.org
Subject: Re: [PATCH 02/14] iotests.py: qemu_img*("create"): support IMGOPTS='compression_type=zstd'
Date: Fri, 16 Jul 2021 12:07:15 +0200	[thread overview]
Message-ID: <5e3574ea-2d7a-d159-3882-3e627841f368@redhat.com> (raw)
In-Reply-To: <20210705091549.178335-3-vsementsov@virtuozzo.com>

On 05.07.21 11:15, Vladimir Sementsov-Ogievskiy wrote:
> Adding support of IMGOPTS (like in bash tests) allows user to pass a
> lot of different options. Still, some may require additional logic.
>
> Now we want compression_type option, so add some smart logic around it:
> ignore compression_type=zstd in IMGOPTS, if test want qcow2 in
> compatibility mode. As well, ignore compression_type for non-qcow2
> formats.
>
> Note that we may instead add support only to qemu_img_create(), but
> that works bad:
>
> 1. We'll have to update a lot of tests to use qemu_img_create instead
>     of qemu_img('create'). (still, we may want do it anyway, but no
>     reason to create a dependancy between task of supporting IMGOPTS and
>     updating a lot of tests)
>
> 2. Some tests use qemu_img_pipe('create', ..) - even more work on
>     updating

I feel compelled to again say that we had a series that did exactly 
that.  But of course, now that so much time has passed, overhauling it 
would require quite a bit of work.

> 3. Even if we update all tests to go through qemu_img_create, we'll
>     need a way to avoid creating new tests using qemu_img*('create') -
>     add assertions.. That doesn't seem good.

That almost sounds like you remember my series, because:

https://lists.nongnu.org/archive/html/qemu-block/2019-10/msg00135.html

;)

> So, let's add support of IMGOPTS to most generic
> qemu_img_pipe_and_status().
>
> Signed-off-by: Vladimir Sementsov-Ogievskiy<vsementsov@virtuozzo.com>
> ---
>   tests/qemu-iotests/iotests.py | 48 ++++++++++++++++++++++++++++++++++-
>   1 file changed, 47 insertions(+), 1 deletion(-)
>
> diff --git a/tests/qemu-iotests/iotests.py b/tests/qemu-iotests/iotests.py
> index 0d99dd841f..80f0cb4f42 100644
> --- a/tests/qemu-iotests/iotests.py
> +++ b/tests/qemu-iotests/iotests.py
> @@ -16,6 +16,7 @@
>   # along with this program.  If not, see<http://www.gnu.org/licenses/>.
>   #
>   
> +import argparse
>   import atexit
>   import bz2
>   from collections import OrderedDict
> @@ -41,6 +42,19 @@
>   from qemu.machine import qtest
>   from qemu.qmp import QMPMessage
>   
> +
> +def optstr2dict(opts: str) -> Dict[str, str]:
> +    if not opts:
> +        return {}
> +
> +    return {arr[0]: arr[1] for arr in
> +            (opt.split('=', 1) for opt in opts.strip().split(','))}
> +
> +
> +def dict2optstr(opts: Dict[str, str]) -> str:
> +    return ','.join(f'{k}={v}' for k, v in opts.items())
> +
> +
>   # Use this logger for logging messages directly from the iotests module
>   logger = logging.getLogger('qemu.iotests')
>   logger.addHandler(logging.NullHandler())
> @@ -57,6 +71,8 @@
>   if os.environ.get('QEMU_IMG_OPTIONS'):
>       qemu_img_args += os.environ['QEMU_IMG_OPTIONS'].strip().split(' ')
>   
> +imgopts = optstr2dict(os.environ.get('IMGOPTS', ''))
> +
>   qemu_io_args = [os.environ.get('QEMU_IO_PROG', 'qemu-io')]
>   if os.environ.get('QEMU_IO_OPTIONS'):
>       qemu_io_args += os.environ['QEMU_IO_OPTIONS'].strip().split(' ')
> @@ -121,11 +137,41 @@ def qemu_tool_pipe_and_status(tool: str, args: Sequence[str],
>                                  {-subp.returncode}: {cmd}\n')
>           return (output, subp.returncode)
>   
> +def qemu_img_create_prepare_args(args: List[str]) -> List[str]:
> +    if not args or args[0] != 'create':
> +        return list(args)
> +    args = args[1:]
> +
> +    p = argparse.ArgumentParser(allow_abbrev=False)
> +    # -o option may be specified several times
> +    p.add_argument('-o', action='append', default=[])
> +    p.add_argument('-f')
> +    parsed, remaining = p.parse_known_args(args)
> +
> +    opts = optstr2dict(','.join(parsed.o))
> +
> +    compat = 'compat' in opts and opts['compat'][0] == '0'

I suppose `opts['compat'][0] == '0'` is supposed to check for compat=0.10?

If so, then why not just check `opts['compat'] == '0.10'` to be 
clearer?  I don’t think we allow any other compat=0* values, and I have 
no reason to believe we ever will.

Also, I think compat=v2 is valid, too.  (And I think calling this 
variable “v2” would also make more sense than “compat”.)

> +    for k, v in imgopts.items():
> +        if k in opts:
> +            continue
> +        if k == 'compression_type' and (compat or parsed.f != 'qcow2'):
> +            continue
> +        opts[k] = v

Could also be done with something like

imgopts = os.environ.get('IMGOPTS')
opts = optstr2dict(','.join(([imgopts] if imgopts else []) + parsed.o))

if parsed.f != 'qcow2' or (opts.get('compat') in ['v2', '0.10']):
     opts.pop('compression_type', None)

(Never tested, of course)

Because optstr2dict() prioritizes later options over earlier ones. 
(Which is good, because that’s also qemu-img’s behavior.)

*shrug*

> +
> +    result = ['create']
> +    if parsed.f is not None:
> +        result += ['-f', parsed.f]

Can this even be None?  I hope none of our tests do this.

> +    if opts:
> +        result += ['-o', dict2optstr(opts)]
> +    result += remaining
> +
> +    return result
> +
>   def qemu_img_pipe_and_status(*args: str) -> Tuple[str, int]:
>       """
>       Run qemu-img and return both its output and its exit code
>       """
> -    full_args = qemu_img_args + list(args)
> +    full_args = qemu_img_args + qemu_img_create_prepare_args(list(args))
>       return qemu_tool_pipe_and_status('qemu-img', full_args)
>   
>   def qemu_img(*args: str) -> int:

There’s also qemu_img_verbose() which I think should be amended the same 
way (even if it’s currently never used for 'create').

Max



  reply	other threads:[~2021-07-16 10:08 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-05  9:15 [PATCH 00/14] iotests: support zstd Vladimir Sementsov-Ogievskiy
2021-07-05  9:15 ` [PATCH 01/14] iotests.py: img_info_log(): rename imgopts argument Vladimir Sementsov-Ogievskiy
2021-07-16 10:07   ` Max Reitz
2021-07-05  9:15 ` [PATCH 02/14] iotests.py: qemu_img*("create"): support IMGOPTS='compression_type=zstd' Vladimir Sementsov-Ogievskiy
2021-07-16 10:07   ` Max Reitz [this message]
2021-07-16 10:34     ` Vladimir Sementsov-Ogievskiy
2021-07-19 12:58       ` Vladimir Sementsov-Ogievskiy
2021-07-19 13:51         ` Vladimir Sementsov-Ogievskiy
2021-07-05  9:15 ` [PATCH 03/14] iotest 303: explicit compression type Vladimir Sementsov-Ogievskiy
2021-07-16 10:38   ` Max Reitz
2021-07-05  9:15 ` [PATCH 04/14] iotest 065: " Vladimir Sementsov-Ogievskiy
2021-07-16 11:01   ` Max Reitz
2021-07-05  9:15 ` [PATCH 05/14] iotests.py: filter compression type out Vladimir Sementsov-Ogievskiy
2021-07-16 11:15   ` Max Reitz
2021-07-16 11:32     ` Vladimir Sementsov-Ogievskiy
2021-07-05  9:15 ` [PATCH 06/14] iotest 302: use img_info_log() helper Vladimir Sementsov-Ogievskiy
2021-07-05 11:02   ` Vladimir Sementsov-Ogievskiy
2021-07-16 11:39   ` Max Reitz
2021-07-16 12:32     ` Vladimir Sementsov-Ogievskiy
2021-07-05  9:15 ` [PATCH 07/14] qcow2: simple case support for downgrading of qcow2 images with zstd Vladimir Sementsov-Ogievskiy
2021-07-16 12:08   ` Max Reitz
2021-07-05  9:15 ` [PATCH 08/14] iotests/common.rc: _make_test_img(): smarter compressiont_type handling Vladimir Sementsov-Ogievskiy
2021-07-16 12:38   ` Max Reitz
2021-07-16 14:24     ` Vladimir Sementsov-Ogievskiy
2021-07-16 14:46       ` Max Reitz
2021-07-05  9:15 ` [PATCH 09/14] iotests/common.rc: introduce _qcow2_dump_header helper Vladimir Sementsov-Ogievskiy
2021-07-16 12:56   ` Max Reitz
2021-07-05  9:15 ` [PATCH 10/14] iotests: massive use _qcow2_dump_header Vladimir Sementsov-Ogievskiy
2021-07-16 13:04   ` Max Reitz
2021-07-05  9:15 ` [PATCH 11/14] iotests: bash tests: filter compression type Vladimir Sementsov-Ogievskiy
2021-07-16 13:17   ` Max Reitz
2021-07-16 14:35     ` Vladimir Sementsov-Ogievskiy
2021-07-05  9:15 ` [PATCH 12/14] iotests 60: more accurate set dirty bit in qcow2 header Vladimir Sementsov-Ogievskiy
2021-07-16 13:20   ` Max Reitz
2021-07-05  9:15 ` [PATCH 13/14] iotest 39: use _qcow2_dump_header Vladimir Sementsov-Ogievskiy
2021-07-16 13:31   ` Max Reitz
2021-07-05  9:15 ` [PATCH 14/14] iotest 214: explicit compression type Vladimir Sementsov-Ogievskiy
2021-07-16 13:35   ` Max Reitz

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5e3574ea-2d7a-d159-3882-3e627841f368@redhat.com \
    --to=mreitz@redhat.com \
    --cc=den@openvz.org \
    --cc=jsnow@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=vsementsov@virtuozzo.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).