MIME-Version: 1.0
References: <20211115030155.9395-1-yangcong5@huaqin.corp-partner.google.com>
In-Reply-To: <20211115030155.9395-1-yangcong5@huaqin.corp-partner.google.com>
From:   Doug Anderson <dianders@chromium.org>
Date:   Mon, 15 Nov 2021 07:40:15 -0800
Message-ID: <CAD=FV=XkYKsQUkZxV2hpvTfq2u4n7Cf_Mk3WaSx5COJvsjG-sw@mail.gmail.com>
Subject: Re: [PATCH] arm64: dts: sc7180: Fix ps8640 power sequence for
 Homestar rev4
To:     yangcong <yangcong5@huaqin.corp-partner.google.com>
Cc:     philipchen@chromium.org, agross@kernel.org,
        bjorn.andersson@linaro.org, robh+dt@kernel.org,
        swboyd@chromium.org, devicetree@vger.kernel.org,
        linux-arm-msm@vger.kernel.org, linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Hi,

On Sun, Nov 14, 2021 at 7:02 PM yangcong
<yangcong5@huaqin.corp-partner.google.com> wrote:
>
> When powering up the ps8640, we need to deassert PD right
> after we turn on the vdd33 regulator. Unfortunately, the vdd33
> regulator takes some time (~4ms) to turn on. Add in the delay
> for the vdd33 regulator so that when the driver deasserts PD
> that the regulator has had time to ramp.
>
> Signed-off-by: yangcong <yangcong5@huaqin.corp-partner.google.com>
> ---
>  arch/arm64/boot/dts/qcom/sc7180-trogdor-homestar-r4.dts | 4 ++++
>  1 file changed, 4 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/qcom/sc7180-trogdor-homestar-r4.dts b/arch/arm64/boot/dts/qcom/sc7180-trogdor-homestar-r4.dts
> index fb27106bbb4a..0de0c97f5728 100644
> --- a/arch/arm64/boot/dts/qcom/sc7180-trogdor-homestar-r4.dts
> +++ b/arch/arm64/boot/dts/qcom/sc7180-trogdor-homestar-r4.dts
> @@ -15,3 +15,7 @@ / {
>         model = "Google Homestar (rev4+)";
>         compatible = "google,homestar", "qcom,sc7180";
>  };
> +
> +&pp3300_brij_ps8640 {
> +       regulator-enable-ramp-delay = <4000>;
> +};

It should have been pointed out somewhere in the post, but this patch
depends on Philip's most recent, which is:

https://lore.kernel.org/r/20211029152647.v3.4.If7aaa8e36f1269acae5488035bd62ce543756bf8@changeid/

This seems fine to me. Apparently this 4ms delay isn't needed
lazor-rev9 (at least they were happy with the signals on that board)
which is why this is a homestar change. That can make sense since the
components on the two boards may be different.

In any case:

Reviewed-by: Douglas Anderson <dianders@chromium.org>