Re: [PATCH v2 0/2] fsmonitor inline / testing cleanup

From: Taylor Blau <me@ttaylorr.com>
Date: Oct 22, 2020, 18:38 UTC
Message-ID: <20201022183822.GA781760@nand.local>
In-Reply-To: <xmqqblgum7qk.fsf@gitster.c.googlers.com>

On Thu, Oct 22, 2020 at 11:32:51AM -0700, Junio C Hamano wrote:

Show 16 quoted lines

> Taylor Blau <me@ttaylorr.com> writes:
> > I'm still iffy on whether or not this series makes sense to apply
> > without the rest of the code that depends on it, but I'll leave that up
> > to Junio whether he wants to take the series as it is now, or wait for
> > other patches to come in on top.
>
> Sorry but I am not sure what you mean by "the code that depends on
> it".  Are these two functions unused anywhere in the code?  If so,
> the right way to clean them up may not be to turn them from inline
> to a proper definition, but to remove them ;-).
>
> If they have existing callers and it can be demonstrated that their
> callers do not benefit from them being inline, that by itself is a
> worthy clean-up, without adding any more callers, no?
>
> Confused...

Sorry for the confusion. I mean the following:

  - These functions have existing callers that Nipunn claims do not need
    to be explicitly inlined.

  - These functions are being moved to be part of the fsmonitor public
    interface (presumably so that new callers can be added).

...And I was wondering whether you wanted to wait for new callers before applying these to your tree.

Thanks, Taylor