qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: John Snow <jsnow@redhat.com>
To: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>,
	Eric Blake <eblake@redhat.com>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
	"qemu-block@nongnu.org" <qemu-block@nongnu.org>
Cc: "kwolf@redhat.com" <kwolf@redhat.com>,
	Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>,
	Denis Lunev <den@virtuozzo.com>,
	"mreitz@redhat.com" <mreitz@redhat.com>
Subject: Re: [Qemu-devel] [PATCH v6 1/6] iotests: allow Valgrind checking all QEMU processes
Date: Thu, 29 Aug 2019 13:33:00 -0400	[thread overview]
Message-ID: <115c74bc-5e73-b825-ac2f-7b8dff06d064@redhat.com> (raw)
In-Reply-To: <be87cbd6-4b12-0222-3825-90ef8538d7ac@virtuozzo.com>



On 8/29/19 6:50 AM, Andrey Shinkevich wrote:
> 
> 
> On 29/08/2019 03:30, Eric Blake wrote:
>> On 8/28/19 5:58 PM, John Snow wrote:
>>
>>>> +++ b/tests/qemu-iotests/common.rc
>>>> @@ -60,61 +60,132 @@ if ! . ./common.config
>>>>       exit 1
>>>>   fi
>>>>   
>>>> +# Unset the variables to turn Valgrind off for specific processes, e.g.
>>
>> That's not unsetting, that's setting to the empty string.
>>
> 
> Thanks Eric, I will make the correction of the comment. Any string other 
> than "y", including the empty one, fits.
> 
>>>> +# $ VALGRIND_QEMU_IO= ./check -qcow2 -valgrind 015
>>>> +
>>>> +: ${VALGRIND_QEMU_VM='y'}
>>>> +: ${VALGRIND_QEMU_IMG='y'}
>>>> +: ${VALGRIND_QEMU_IO='y'}
>>>> +: ${VALGRIND_QEMU_NBD='y'}
>>>> +: ${VALGRIND_QEMU_VXHS='y'}
>>>> +
>>>
> 
> I am going to make the change:
> 
> : ${VALGRIND_QEMU_VM=$VALGRIND_QEMU}
> : ${VALGRIND_QEMU_IMG=$VALGRIND_QEMU}
> : ${VALGRIND_QEMU_IO=$VALGRIND_QEMU}
> : ${VALGRIND_QEMU_NBD=$VALGRIND_QEMU}
> : ${VALGRIND_QEMU_VXHS=$VALGRIND_QEMU}
> 
> and get rid of the local VALGRIND_ON="${VALGRIND_QEMU}"
> 
> so that the code will be optimized.
> 

Seems good!

>>> I have to admit to you that I'm not familiar with this trick. I'm
>>> looking it up and I see := documented, but not = alone.
>>
>> It's been a repeated complaint to the bash developer that the manual is
>> doing a disservice to its users by not documenting ${var=val} in an
>> easily searchable form.  It IS documented, but only by virtue of
>> ${var:=val} occurring under a section header that states:
>>
>>         When not performing substring expansion,  using  the  forms
>> documented
>>         below  (e.g.,  :-),  bash  tests for a parameter that is unset or
>> null.
>>         Omitting the colon results in a test  only  for  a  parameter
>> that  is
>>         unset.
>>
>> So the choice is whether you want to special case a variable set to an
>> empty string the same as an unset variable, or the same as a variable
>> with a non-empty value.
>>
> 
> Thank you all for your reviews and comments. The purpose why I omitted 
> the colon is to allow a user writing the shorter command syntax like
> $ VALGRIND_QEMU_IO= ./check -valgrind <test#>
> rather than
> $ VALGRIND_QEMU_IO=" 'no' or 'off' or else anything other than 'y' " 
> ./check -valgrind <test#>
> so, no need to strike the Shift key twice and guess at what else is 
> acceptable to type )))
> 
> The variable default value 'y' looks good to me to implement the new 
> functionality that is compatible with the existing one when we just set 
> the '-valgrind' switch. The general idea behind using the Valgrind is to 
> make a careful search for memory issues. Once found, a user can tune the 
> particular test with extra variables to save their development/testing 
> time as John suggested. Also, no need to specify all the five long name 
> variables each time a user writes the command if default values aren't set.
> 
> I am flexible to make a change that is good for all. So, what solution 
> will we come to?
> 

I don't actually really have a preference here; it's development and
testing infrastructure. As long as it is POSIX portable, I'm happy. If
we goof it up, we'll find out eventually. If we don't, well. Just more
evidence we need more non-Linux contributors.

--js


  reply	other threads:[~2019-08-29 17:33 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-26 15:50 [Qemu-devel] [PATCH v6 0/6] Allow Valgrind checking all QEMU processes Andrey Shinkevich
2019-08-26 15:50 ` [Qemu-devel] [PATCH v6 1/6] iotests: allow " Andrey Shinkevich
2019-08-28 22:58   ` John Snow
2019-08-29  0:30     ` Eric Blake
2019-08-29 10:50       ` Andrey Shinkevich
2019-08-29 17:33         ` John Snow [this message]
2019-08-26 15:50 ` [Qemu-devel] [PATCH v6 2/6] iotests: exclude killed processes from running under Valgrind Andrey Shinkevich
2019-08-28 23:00   ` John Snow
2019-08-26 15:50 ` [Qemu-devel] [PATCH v6 3/6] iotests: Add casenotrun report to bash tests Andrey Shinkevich
2019-08-26 15:50 ` [Qemu-devel] [PATCH v6 4/6] iotests: Valgrind fails with nonexistent directory Andrey Shinkevich
2019-08-26 15:50 ` [Qemu-devel] [PATCH v6 5/6] iotests: extended timeout under Valgrind Andrey Shinkevich
2019-08-26 15:50 ` [Qemu-devel] [PATCH v6 6/6] iotests: extend sleeping time " Andrey Shinkevich
2019-08-28 23:02   ` John Snow

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=115c74bc-5e73-b825-ac2f-7b8dff06d064@redhat.com \
    --to=jsnow@redhat.com \
    --cc=andrey.shinkevich@virtuozzo.com \
    --cc=den@virtuozzo.com \
    --cc=eblake@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=mreitz@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).