Re: [PATCH] docs: Fix virtiofsd.1 location

[Philippe Mathieu-Daudé]

On 2/12/20 5:02 PM, Peter Maydell wrote:
> On Wed, 12 Feb 2020 at 14:28, Daniel P. Berrangé <address@hidden> wrote:
> > On Wed, Feb 12, 2020 at 02:18:19PM +0000, Peter Maydell wrote:
> > > On Wed, 12 Feb 2020 at 13:16, Miroslav Rezanina <address@hidden> wrote:
> > > > Patch 6a7e2bbee5 docs: add virtiofsd(1) man page introduced new man
> > > > page virtiofsd.1. Unfortunately, wrong file location is used as
> > > > source for install command. This cause installation of docs fail.
> > > >
> > > > Fixing wrong location so installation is successful.
> > > >
> > > > Signed-off-by: Miroslav Rezanina <address@hidden>
> > >
> > > Reviewed-by: Peter Maydell <address@hidden>
> > >
> > > I noticed this in review of v1 of the patch
> > > https://patchew.org/QEMU/address@hidden/
> > > but missed that it hadn't been fixed in v2/v3.
> >
> > How come the mistake didn't break any of our CI ?  Looks like it
> > should have failed due to non-existing source file

We stopped testing in-tree builds 2 months ago:

commit bc4486fb233573e77b6e9ad6d6379afb5e37ad8c
Author: Paolo Bonzini <address@hidden>
Date:   Wed Dec 11 15:33:49 2019 +0100

    ci: build out-of-tree

    Most developers are using out-of-tree builds and it was discussed 
in the past
    to only allow those.  To prepare for the transition, use 
out-of-tree builds
    in all continuous integration jobs.

> I imagine all our CI setups use separate-build-dir and don't
> test 'make install'. This only shows up if you do build-in-source-tree
> and then 'make install'. (In a separate-build-dir setup, the
> MANUAL_BUILDDIR is the same as the actual build dir,
> which is the same as make's current working directory, so
> forgetting it in the install rune doesn't matter. In a build from
> the source tree, Sphinx requires the output to be to a
> different directory than the source, so we have to create
> a subdirectory to be the MANUAL_BUILDDIR.)
>
> thanks
> -- PMM