Re: [PATCH 1/4] [1/4] mmc: core: expose MMC_CAP2_CQE* to dt

From: Linus Walleij <linus.walleij@linaro.org>
To: Chun-Hung Wu <chun-hung.wu@mediatek.com>
Cc: Chaotian Jing <chaotian.jing@mediatek.com>,
	Ulf Hansson <ulf.hansson@linaro.org>,
	Rob Herring <robh+dt@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Matthias Brugger <matthias.bgg@gmail.com>,
	Pavel Machek <pavel@ucw.cz>,
	Kate Stewart <kstewart@linuxfoundation.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Martin Blumenstingl <martin.blumenstingl@googlemail.com>,
	Pan Bian <bianpan2016@163.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Allison Randal <allison@lohutok.net>,
	Mathieu Malaterre <malat@debian.org>,
	Stanley Chu <stanley.chu@mediatek.com>,
	Kuohong Wang <kuohong.wang@mediatek.com>,
	kernel-team@android.com,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	linux-mmc <linux-mmc@vger.kernel.org>,
	"moderated list:ARM/Mediatek SoC support" 
	<linux-mediatek@lists.infradead.org>,
	"open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS" 
	<devicetree@vger.kernel.org>,
	wsd_upstream@mediatek.com,
	Linux ARM <linux-arm-kernel@lists.infradead.org>
Subject: Re: [PATCH 1/4] [1/4] mmc: core: expose MMC_CAP2_CQE* to dt
Date: Fri, 21 Feb 2020 15:33:58 +0100	[thread overview]
Message-ID: <CACRpkdauCe-zAcsO5q_79+Ux_a2=bktA5UqKjWs4XdBJZ3wWeg@mail.gmail.com> (raw)
In-Reply-To: <1581921111-24172-2-git-send-email-chun-hung.wu@mediatek.com>

On Mon, Feb 17, 2020 at 7:32 AM Chun-Hung Wu <chun-hung.wu@mediatek.com> wrote:

> Expose MMC_CAP2_CQE and MMC_CAP2_CQE_DCMD
> to host->caps2 if
> 1. "supports-cqe" is defined in dt and
> 2. "disable-cqe-dcmd" is not defined in dt.
>
> ---
>  drivers/mmc/core/host.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
>
> diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c
> index 105b7a7..efb0dbe 100644
> --- a/drivers/mmc/core/host.c
> +++ b/drivers/mmc/core/host.c
> @@ -319,6 +319,14 @@ int mmc_of_parse(struct mmc_host *host)
>                 host->caps2 |= MMC_CAP2_NO_SD;
>         if (device_property_read_bool(dev, "no-mmc"))
>                 host->caps2 |= MMC_CAP2_NO_MMC;
> +       if (device_property_read_bool(dev, "supports-cqe"))
> +               host->caps2 |= MMC_CAP2_CQE;

I don't understand why this is even a DT property as it should
be clear from the hosts compatible whether it supports CQE or
not. But it's too late to do anything about that I suppose, and
I just assume there is something I don't understand here.

> +       /* Must be after "supports-cqe" check */
> +       if (!device_property_read_bool(dev, "disable-cqe-dcmd")) {
> +               if (host->caps2 & MMC_CAP2_CQE)
> +                       host->caps2 |= MMC_CAP2_CQE_DCMD;
> +       }

This is the right place to do this I suppose. Disabling CQE
selectively is something you might wanna do for debugging.
Acked-by: Linus Walleij <linus.walleij@linaro.org>

I see that some drivers are already parsing this DT property
on their own, should we follow up with patches so that these:

$ git grep 'supports-cqe'
drivers/mmc/host/sdhci-brcmstb.c:       if
(device_property_read_bool(&pdev->dev, "supports-cqe")) {
drivers/mmc/host/sdhci-msm.c:   if (of_property_read_bool(node, "supports-cqe"))
drivers/mmc/host/sdhci-tegra.c: if
(device_property_read_bool(host->mmc->parent, "supports-cqe"))

Make use of the central parsing instead?

Yours,
Linus Walleij