Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20200625170434.635114-1-masahiroy@kernel.org> <20200625170434.635114-5-masahiroy@kernel.org>
In-Reply-To: <20200625170434.635114-5-masahiroy@kernel.org>
From:   Rob Herring <robh+dt@kernel.org>
Date:   Mon, 29 Jun 2020 14:49:55 -0600
Message-ID: <CAL_JsqL4pTFK_pSNn1cSvjzwdg71dVzM3P06BYYEwDj2t+swPA@mail.gmail.com>
Subject: Re: [PATCH 4/4] dt-bindings: split DT schema check rules
To:     Masahiro Yamada <masahiroy@kernel.org>
Cc:     devicetree@vger.kernel.org, Frank Rowand <frowand.list@gmail.com>,
        Linux Kbuild mailing list <linux-kbuild@vger.kernel.org>,
        Michal Marek <michal.lkml@markovi.net>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Jun 25, 2020 at 11:05 AM Masahiro Yamada <masahiroy@kernel.org> wrote:
>
> When building %.dt.yaml from %.dts, two things happen in a row:
>
>  [1] Run DTC to convert %.dts into %.dt.yaml
>
>  [2] Run dt-validate against %.dt.yaml
>
> Currently, when any .yaml schema file is updated, processed-schema.yaml
> is regenerated, then both [1] and [2] are rerun for all .dts files.
>
> Actually, we do not need to rerun [1] since the original .dts is not
> updated.

I have plans (and an intern working on it) to integrate the schema
checks into dtc. That's going to make turning on the schema checks
just a flag to dtc. I'm not sure if adding the complexity here is
worth it as I'd expect much of this patch to go away again.

Is there any negative impact on the absolute clean build time? I'm
more concerned about that than optimizing rerunning.

Rob