Re: [PATCH 1/6] scripts/coccinelle: add error-use-after-free.cocci

[Markus Armbruster <armbru@redhat.com>]

Peter Maydell <peter.maydell@linaro.org> writes:

> On Thu, 2 Apr 2020 at 07:55, Markus Armbruster <armbru@redhat.com> wrote:
>> Peter Maydell <peter.maydell@linaro.org> writes:
>> > I use this thing maybe once a month at most, more likely once
>> > every three months, and the documentation is notoriously
>> > impenetrable. I really really don't want to have to start looking in it
>> > and guessing about how the original author ran the script, when
>> > they could have just told me.
>>
>> I'm afraid we're talking part each other.
>
> Perhaps; but I think we're also genuinely disagreeing.
>
>> >>           $  spatch --sp-file scripts/coccinelle/qobject.cocci \
>> >>                     --macro-file scripts/cocci-macro-file.h \
>> >>                     --dir block --in-place
>> >
>> > Yep, that command line would be great to see in the script file.
>>
>> Except for the --dir block part, which is even worse than noise: it
>> suggests this is just for block/, which is wrong.
>
> It tells the reader that the original author only tested
> the script to work inside block/, which is useful information.
> (This is why scripts/coccinelle/cpu-reset.cocci specifies
> --dir target in its command.)
>
>> > The least-effort way for the author of the script to do that is to
>> > simply give the command line they used to run it.
>>
>> If you're capable of writing a Coccinelle script that actually does what
>> you want, you're certainly capable of doing better than blindly paste
>> from your shell history.  Kindly drop the options that are specific to
>> just this particular use of the script.  Keep the ones that future users
>> will actually need.
>
> I'm a future user; in fact I'm the future user whose needs I have
> the best information on. I want to see the whole command, please.

In that case, nothing seems to be left than agree to disagree.

[...]