Re: [PATCH v2 08/12] soc: mediatek: pm-domains: Add subsystem clocks

From: Matthias Brugger <matthias.bgg@gmail.com>
To: Enric Balletbo i Serra <enric.balletbo@collabora.com>,
	linux-kernel@vger.kernel.org
Cc: drinkcat@chromium.org, weiyi.lu@mediatek.com,
	fparent@baylibre.com, Matthias Brugger <mbrugger@suse.com>,
	linux-mediatek@lists.infradead.org, hsinyi@chromium.org,
	Collabora Kernel ML <kernel@collabora.com>,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v2 08/12] soc: mediatek: pm-domains: Add subsystem clocks
Date: Fri, 2 Oct 2020 11:04:49 +0200	[thread overview]
Message-ID: <d6d132ba-1ce1-f537-57ba-4b6bed784b48@gmail.com> (raw)
In-Reply-To: <20201001160154.3587848-9-enric.balletbo@collabora.com>

On 01/10/2020 18:01, Enric Balletbo i Serra wrote:
> From: Matthias Brugger <mbrugger@suse.com>
> 
> For the bus protection operations, some subsystem clocks need to be enabled
> before releasing the protection. This patch identifies the subsystem clocks
> by it's name.
> 
> Suggested-by: Weiyi Lu <weiyi.lu@mediatek.com>
> [Adapted the patch to the mtk-pm-domains driver]
> Signed-off-by: Matthias Brugger <mbrugger@suse.com>
> Signed-off-by: Enric Balletbo i Serra <enric.balletbo@collabora.com>
> ---
> 
> Changes in v2:
> - Use dev_err_probe if getting clocks fails, so an error is not printed
>    if is deferred.
> 
>   drivers/soc/mediatek/mtk-pm-domains.c | 89 ++++++++++++++++++++++-----
>   1 file changed, 75 insertions(+), 14 deletions(-)
> 
> diff --git a/drivers/soc/mediatek/mtk-pm-domains.c b/drivers/soc/mediatek/mtk-pm-domains.c
> index e0a52d489fea..2075072f16da 100644
> --- a/drivers/soc/mediatek/mtk-pm-domains.c
> +++ b/drivers/soc/mediatek/mtk-pm-domains.c
[...]
>   
> +	for (i = 0; i < pd->num_subsys_clks; i++) {
> +		struct clk *clk = of_clk_get(node, i + clk_ind);
> +
> +		if (IS_ERR(clk)) {
> +			ret = PTR_ERR(clk);
> +			dev_err_probe(scpsys->dev, PTR_ERR(clk),
> +				      "%pOFn: failed to get clk at index %d: %d\n",
> +				      node, i + clk_ind, ret);
> +			goto err_put_subsys_clocks;
> +		}
> +
> +		pd->subsys_clks[i].clk = clk;
> +	}
> +
> +	ret = clk_bulk_prepare(pd->num_subsys_clks, pd->subsys_clks);
> +	if (ret)
> +		goto err_put_subsys_clocks;
> +

Not sure if it really matters, but logically we should prepare the basic clocks 
before we prepare the subsystem clocks. And fix the error handling accordingly.

Regards,
Matthias

>   	ret = clk_bulk_prepare(pd->num_clks, pd->clks);
>   	if (ret)
> -		goto err_put_clocks;
> +		goto err_unprepare_subsys_clocks;
>   
>   	/*
>   	 * Initially turn on all domains to make the domains usable
> @@ -462,6 +513,12 @@ static int scpsys_add_one_domain(struct scpsys *scpsys, struct device_node *node
>   
>   err_unprepare_clocks:
>   	clk_bulk_unprepare(pd->num_clks, pd->clks);
> +err_unprepare_subsys_clocks:
> +	clk_bulk_unprepare(pd->num_subsys_clks, pd->subsys_clks);
> +err_put_subsys_clocks:
> +	clk_bulk_put(pd->num_subsys_clks, pd->subsys_clks);
> +	devm_kfree(scpsys->dev, pd->subsys_clks);
> +	pd->num_subsys_clks = 0;
>   err_put_clocks:
>   	clk_bulk_put(pd->num_clks, pd->clks);
>   	devm_kfree(scpsys->dev, pd->clks);
> @@ -541,6 +598,10 @@ static void scpsys_remove_one_domain(struct scpsys_domain *pd)
>   	clk_bulk_unprepare(pd->num_clks, pd->clks);
>   	clk_bulk_put(pd->num_clks, pd->clks);
>   	pd->num_clks = 0;
> +
> +	clk_bulk_unprepare(pd->num_subsys_clks, pd->subsys_clks);
> +	clk_bulk_put(pd->num_subsys_clks, pd->subsys_clks);
> +	pd->num_subsys_clks = 0;
>   }
>   
>   static void scpsys_domain_cleanup(struct scpsys *scpsys)
> 

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel