[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-block] [PATCH for-2.6?] qemu-iotests: iotests: fail hard if no
From: |
Sascha Silbe |
Subject: |
Re: [Qemu-block] [PATCH for-2.6?] qemu-iotests: iotests: fail hard if not run via "check" |
Date: |
Tue, 19 Apr 2016 14:22:43 +0200 |
User-agent: |
Notmuch/0.19+1~g6b3e223 (http://notmuchmail.org) Emacs/24.3.1 (x86_64-pc-linux-gnu) |
Dear Max,
Max Reitz <address@hidden> writes:
> On 14.04.2016 13:32, Sascha Silbe wrote:
[tests/qemu-iotests/iotests.py]
[...]
>> def main(supported_fmts=[], supported_oses=['linux']):
>> '''Run tests'''
>>
>> + if test_dir is None or qemu_default_machine is None:
>
> I think checking test_dir would have been sufficient; this makes it look
> like it might want to be a complete list of variables that have to be
> set, but then "cachemode" is missing.
Markus Armbruster basically pointed out the same issue, so how about
this version:
# We are using TEST_DIR and QEMU_DEFAULT_MACHINE as proxies to
# indicate that we're not being run via "check". There may be
# other things set up by "check" that individual test cases rely
# on.
if test_dir is None or qemu_default_machine is None:
sys.stderr.write('Please run this test via the "check" script\n')
sys.exit(os.EX_USAGE)
>> + sys.stderr.write('Please run this test via ./check\n')
>
> Or not ./, for people who have separate build and source trees, you
> don't want to invoke check in the source tree. ;-)
Yeah, was thinking about that, but ultimately considered ./check to be
the best indication of a) "check" being the name of a script and b)
residing in the same directory as the test. But I don't care much about
this, so see the above version.
Sascha
--
Softwareentwicklung Sascha Silbe, Niederhofenstraße 5/1, 71229 Leonberg
https://se-silbe.de/
USt-IdNr. DE281696641
Re: [Qemu-block] [Qemu-devel] [PATCH for-2.6?] qemu-iotests: iotests: fail hard if not run via "check", Markus Armbruster, 2016/04/18