From: Kevin Wolf <kwolf@redhat.com>
To: Max Reitz <mreitz@redhat.com>
Cc: qemu-block@nongnu.org, jsnow@redhat.com, eblake@redhat.com,
pkrempa@redhat.com, qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH 08/14] qemu-iotests: Rewrite 206 for blockdev-create job
Date: Tue, 29 May 2018 20:49:04 +0200 [thread overview]
Message-ID: <20180529184904.GJ4756@localhost.localdomain> (raw)
In-Reply-To: <8b64d45f-0e9c-107f-a55a-bb3880c60eb6@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 6965 bytes --]
Am 29.05.2018 um 14:27 hat Max Reitz geschrieben:
> On 2018-05-25 18:33, Kevin Wolf wrote:
> > This rewrites the test case 206 to work with the new x-blockdev-create
> > job rather than the old synchronous version of the command.
> >
> > All of the test cases stay the same as before, but in order to be able
> > to implement proper job handling, the test case is rewritten in Python.
> >
> > Signed-off-by: Kevin Wolf <kwolf@redhat.com>
> > ---
> > tests/qemu-iotests/206 | 705 +++++++++++++++++-------------------------
> > tests/qemu-iotests/206.out | 241 +++++++++------
> > tests/qemu-iotests/group | 2 +-
> > tests/qemu-iotests/iotests.py | 15 +
> > 4 files changed, 448 insertions(+), 515 deletions(-)
> >
> > diff --git a/tests/qemu-iotests/206 b/tests/qemu-iotests/206
> > index 0a18b2b19a..9a305302d1 100755
> > --- a/tests/qemu-iotests/206
> > +++ b/tests/qemu-iotests/206
>
> [...]
>
> > +import iotests
> > +from iotests import imgfmt
> > +
> > +iotests.verify_image_format(supported_fmts=['qcow2'])
> > +
> > +def blockdev_create(vm, options):
> > + result = vm.qmp_log('x-blockdev-create', job_id='job0', options=options)
> > +
> > + if 'return' in result:
> > + assert result['return'] == {}
> > + vm.run_job('job0')
> > + iotests.log("")
> > +
> > +with iotests.FilePath('t.qcow2') as disk_path, \
> > + iotests.FilePath('t.qcow2.base') as backing_path, \
> > + iotests.VM() as vm:
> > +
> > + vm.add_object('secret,id=keysec0,data="foo"')
>
> I don't know how subprocess.Popen() works, but are you sure you aren't
> encrypting with '"foo"' now? (i.e. literally that key, including quotes)
That's correct. Anything wrong with it?
(Okay, you're right. I did fix it in 210, but forgot it here...)
> > +
> > + #
> > + # Successful image creation (defaults)
> > + #
> > + iotests.log("=== Successful image creation (defaults) ===")
>
> OK, the comment makes sense for visual separation. But so would leaving
> three empty lines. *cough*
I tried vertical spacing first, but it didn't work well for me. What
made the difference is the syntax highlighting of comments vs. code.
YMMV.
> > + iotests.log("")
> > +
> > + size = 128 * 1024 * 1024
> > +
> > + vm.launch()
> > + blockdev_create(vm, { 'driver': 'file',
> > + 'filename': disk_path,
> > + 'size': 0 })
> > +
> > + vm.qmp_log('blockdev-add', driver='file', filename=disk_path,
> > + node_name='imgfile')
> > +
> > + blockdev_create(vm, { 'driver': imgfmt,
> > + 'file': 'imgfile',
> > + 'size': size })
> > + vm.shutdown()
> > +
> > + iotests.img_info_log(disk_path)
> > +
>
> [...]
>
> > + #
> > + # Successful image creation (v2 non-default options)
> > + #
> > + iotests.log("=== Successful image creation (v2 non-default options) ===")
> > + iotests.log("")
> > +
> > + vm.launch()
> > + blockdev_create(vm, { 'driver': 'file',
> > + 'filename': disk_path,
> > + 'size': 0 })
> > +
> > + blockdev_create(vm, { 'driver': imgfmt,
> > + 'file': {
> > + 'driver': 'file',
> > + 'filename': disk_path,
> > + },
> > + 'size': size,
> > + 'backing-file': backing_path,
>
> Change from the bash version: backing_path does not exist here. Not
> sure if that was intentional.
No, it was not. It's actually a good test case, though. Creating an
image at backing_path would be easy enough, but maybe we should just
leave it?
> > + 'backing-fmt': 'qcow2',
> > + 'version': 'v2',
> > + 'cluster-size': 512 })
> > + vm.shutdown()
> > +
> > + iotests.img_info_log(disk_path)
>
> [...]
>
> > + #
> > + # Invalid sizes
> > + #
> > + iotests.log("=== Invalid sizes ===")
> > +
> > + # TODO Negative image sizes aren't handled correctly, but this is a problem
> > + # with QAPI's implementation of the 'size' type and affects other commands
> > + # as well. Once this is fixed, we may want to add a test case here.
> > + #
> > + # 1. Misaligned image size
> > + # 2. 2^64 - 512
> > + # 3. 2^63 = 8 EB (qemu-img enforces image sizes less than this)
> > + # 4. 2^63 - 512 (generally valid, but qcow2 can't handle images this size)
> > +
> > + vm.add_blockdev('driver=file,filename=%s,node-name=node0' % (disk_path))
> > +
> > + vm.launch()
> > + blockdev_create(vm, { 'driver': imgfmt,
> > + 'file': 'node0',
> > + 'size': 1234 })
> > + blockdev_create(vm, { 'driver': imgfmt,
> > + 'file': 'node0',
> > + 'size': 18446744073709551104 })
> > + blockdev_create(vm, { 'driver': imgfmt,
> > + 'file': 'node0',
> > + 'size': 9223372036854775808 })
> > + blockdev_create(vm, { 'driver': imgfmt,
> > + 'file': 'node0',
> > + 'size': 9223372036854775296})
>
> Missing space before the closing fancy bracket, critical!
>
> [...]
>
> > diff --git a/tests/qemu-iotests/iotests.py b/tests/qemu-iotests/iotests.py
> > index 20ce5a0cf0..f0f4ef32f0 100644
> > --- a/tests/qemu-iotests/iotests.py
> > +++ b/tests/qemu-iotests/iotests.py
> > @@ -416,6 +416,21 @@ class VM(qtest.QEMUQtestMachine):
> > log(result)
> > return result
> >
> > + def run_job(self, job):
>
> Is there any reason this function did not get its own patch?
>
> Also, this is a function specifically for jobs that need a manual
> dismiss but have auto-finalize. That should be noted somewhere (ideally
> in the function's name, but spontaneously I don't know how).
If you want me to move it into a separate patch, maybe just make it more
reusable with parameters auto_finalize=True, auto_dismiss=false.
> > + while True:
> > + for ev in self.get_qmp_events_filtered(wait=True):
> > + if ev['event'] == 'JOB_STATUS_CHANGE':
> > + if ev['data']['status'] == 'aborting':
> > + result = self.qmp('query-jobs')
> > + for j in result['return']:
> > + log('Job failed: %s' % (j.get('error', None)))
>
> I can understand that you didn't want to use just result['return'][0],
> but if you do iterate, you should probably emit the job ID as well.
If you have multiple jobs running, it doesn't work correctly anyway. :-)
Kevin
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 801 bytes --]
next prev parent reply other threads:[~2018-05-29 18:49 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-05-25 16:33 [Qemu-devel] [PATCH 00/14] block: Make blockdev-create a job and stable API Kevin Wolf
2018-05-25 16:33 ` [Qemu-devel] [PATCH 01/14] vdi: Fix vdi_co_do_create() return value Kevin Wolf
2018-05-29 10:38 ` Max Reitz
2018-05-29 14:45 ` [Qemu-devel] [Qemu-block] " Jeff Cody
2018-05-25 16:33 ` [Qemu-devel] [PATCH 02/14] vhdx: Fix vhdx_co_create() " Kevin Wolf
2018-05-29 10:40 ` Max Reitz
2018-05-29 14:44 ` [Qemu-devel] [Qemu-block] " Jeff Cody
2018-05-25 16:33 ` [Qemu-devel] [PATCH 03/14] job: Add error message for failing jobs Kevin Wolf
2018-05-29 11:01 ` Max Reitz
2018-05-29 14:43 ` [Qemu-devel] [Qemu-block] " Jeff Cody
2018-05-29 19:54 ` Kevin Wolf
2018-05-25 16:33 ` [Qemu-devel] [PATCH 04/14] block/create: Make x-blockdev-create a job Kevin Wolf
2018-05-29 11:38 ` Max Reitz
2018-05-29 15:27 ` [Qemu-devel] [Qemu-block] " Jeff Cody
2018-05-29 15:40 ` Kevin Wolf
2018-05-25 16:33 ` [Qemu-devel] [PATCH 05/14] qemu-iotests: Add VM.get_qmp_events_filtered() Kevin Wolf
2018-05-29 11:41 ` Max Reitz
2018-05-25 16:33 ` [Qemu-devel] [PATCH 06/14] qemu-iotests: Add VM.qmp_log() Kevin Wolf
2018-05-29 11:48 ` Max Reitz
2018-05-29 12:12 ` Kevin Wolf
2018-05-29 12:15 ` Max Reitz
2018-05-29 12:39 ` Kevin Wolf
2018-05-29 12:41 ` Max Reitz
2018-05-29 18:31 ` [Qemu-devel] [Qemu-block] " Jeff Cody
2018-05-25 16:33 ` [Qemu-devel] [PATCH 07/14] qemu-iotests: Add iotests.img_info_log() Kevin Wolf
2018-05-29 11:56 ` Max Reitz
2018-05-25 16:33 ` [Qemu-devel] [PATCH 08/14] qemu-iotests: Rewrite 206 for blockdev-create job Kevin Wolf
2018-05-29 12:27 ` Max Reitz
2018-05-29 18:49 ` Kevin Wolf [this message]
2018-05-25 16:33 ` [Qemu-devel] [PATCH 09/14] qemu-iotests: Rewrite 207 " Kevin Wolf
2018-05-29 12:44 ` Max Reitz
2018-05-29 12:47 ` Max Reitz
2018-05-29 17:52 ` [Qemu-devel] [Qemu-block] " Jeff Cody
2018-05-25 16:33 ` [Qemu-devel] [PATCH 10/14] qemu-iotests: Rewrite 210 " Kevin Wolf
2018-05-29 13:02 ` Max Reitz
2018-05-29 18:23 ` [Qemu-devel] [Qemu-block] " Jeff Cody
2018-05-25 16:33 ` [Qemu-devel] [PATCH 11/14] qemu-iotests: Rewrite 211 " Kevin Wolf
2018-05-29 13:12 ` Max Reitz
2018-05-25 16:33 ` [Qemu-devel] [PATCH 12/14] qemu-iotests: Rewrite 212 " Kevin Wolf
2018-05-29 13:21 ` Max Reitz
2018-05-25 16:33 ` [Qemu-devel] [PATCH 13/14] qemu-iotests: Rewrite 213 " Kevin Wolf
2018-05-29 13:27 ` Max Reitz
2018-05-25 16:33 ` [Qemu-devel] [PATCH 14/14] block/create: Mark blockdev-create stable Kevin Wolf
2018-05-29 13:30 ` Max Reitz
2018-05-29 18:25 ` [Qemu-devel] [Qemu-block] " Jeff Cody
2018-05-25 16:52 ` [Qemu-devel] [PATCH 00/14] block: Make blockdev-create a job and stable API no-reply
2018-05-25 18:13 ` Eric Blake
2018-05-28 8:42 ` Kevin Wolf
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=20180529184904.GJ4756@localhost.localdomain \
--to=kwolf@redhat.com \
--cc=eblake@redhat.com \
--cc=jsnow@redhat.com \
--cc=mreitz@redhat.com \
--cc=pkrempa@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.org \
/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).