From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1752350AbaIITMZ (ORCPT <rfc822;w@1wt.eu>);
	Tue, 9 Sep 2014 15:12:25 -0400
Received: from mail-pd0-f173.google.com ([209.85.192.173]:33165 "EHLO
	mail-pd0-f173.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751734AbaIITMV convert rfc822-to-8bit (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Tue, 9 Sep 2014 15:12:21 -0400
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: 8BIT
To: Tomeu Vizoso <tomeu.vizoso@collabora.com>
From: Mike Turquette <mturquette@linaro.org>
In-Reply-To: <1410271497-27148-1-git-send-email-tomeu.vizoso@collabora.com>
Cc: linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
        shawn.guo@freescale.com
References: <1410271329-26637-1-git-send-email-tomeu.vizoso@collabora.com>
 <1410271497-27148-1-git-send-email-tomeu.vizoso@collabora.com>
Message-ID: <20140909191205.19023.61366@quantum>
User-Agent: alot/0.3.5
Subject: Re: [PATCH v10 2/9] clk: Move all drivers to use internal API
Date: Tue, 09 Sep 2014 12:12:05 -0700
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

Quoting Tomeu Vizoso (2014-09-09 07:04:57)
> In preparation to change the public API to return a per-user clk structure,
> remove any usage of this public API from the clock implementations.
> 
> The reason for having this in a separate commit from the one that introduces
> the implementation of the new functions is to separate the changes generated
> with Coccinelle from the rest, and keep the patches' size reasonable.
> 
> Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
> Tested-by: Boris Brezillon <boris.brezillon@free-electrons.com>
> Tested-by: Heiko Stuebner <heiko@sntech.de>
> Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com>
> 
> ---
> 
> v10: * Add a few more files to be converted
>      * Re-generate the patch on top of the latest changes

Hi Tomeu,

Generating this on top of linux-next is a no-go. I can't apply it to my
tree. The best thing is to generate it on top of -rc4, and that is what
I will merge.

Running the script against linux-next is still very useful and lets us
patch up the stuff that is not going through the clk tree. E.g. the LPSS
driver is already in mainline, so just running the semantic patch
against -rc4 is sufficient for it. However a patch like Shawn's "ARM:
imx: add an exclusive gate clock type" came in through the i.MX tree and
we'll need to patch it after the fact.

The best way to do that is for me to host a branch with just your
changes in it that everyone can pull in as a dependency with the same
commit ids.

<snip>

> diff --git a/drivers/acpi/acpi_lpss.c b/drivers/acpi/acpi_lpss.c
> index bcbdbd2..f4c6ccf 100644
> --- a/drivers/acpi/acpi_lpss.c
> +++ b/drivers/acpi/acpi_lpss.c
> @@ -11,7 +11,6 @@
>   */
>  
>  #include <linux/acpi.h>
> -#include <linux/clk.h>
>  #include <linux/clkdev.h>
>  #include <linux/clk-provider.h>
>  #include <linux/err.h>
> @@ -78,7 +77,7 @@ struct lpss_private_data {
>         void __iomem *mmio_base;
>         resource_size_t mmio_size;
>         unsigned int fixed_clk_rate;
> -       struct clk *clk;
> +       struct clk_core *clk;
>         const struct lpss_device_desc *dev_desc;
>         u32 prv_reg_ctx[LPSS_PRV_REG_COUNT];
>  };
> @@ -229,7 +228,7 @@ static int register_device_clock(struct acpi_device *adev,
>  {
>         const struct lpss_device_desc *dev_desc = pdata->dev_desc;
>         const char *devname = dev_name(&adev->dev);
> -       struct clk *clk = ERR_PTR(-ENODEV);
> +       struct clk_core *clk = ERR_PTR(-ENODEV);
>         struct lpss_clk_data *clk_data;
>         const char *parent, *clk_name;
>         void __iomem *prv_base;

I think the following hunk is missing from your change:

--- a/drivers/acpi/acpi_lpss.c
+++ b/drivers/acpi/acpi_lpss.c
@@ -57,7 +57,7 @@ ACPI_MODULE_NAME("acpi_lpss");
 struct lpss_shared_clock {
        const char *name;
        unsigned long rate;
-       struct clk *clk;
+       struct clk_core *clk;
 };


Otherwise register_device_clock will blow up because we are assigning a
struct clk * to a struct clk_core *.

Do you mind testing with ARCH=x86_64 and allmodconfig? That will help
catch issues like this.

Regards,
Mike

From mboxrd@z Thu Jan  1 00:00:00 1970
From: mturquette@linaro.org (Mike Turquette)
Date: Tue, 09 Sep 2014 12:12:05 -0700
Subject: [PATCH v10 2/9] clk: Move all drivers to use internal API
In-Reply-To: <1410271497-27148-1-git-send-email-tomeu.vizoso@collabora.com>
References: <1410271329-26637-1-git-send-email-tomeu.vizoso@collabora.com>
 <1410271497-27148-1-git-send-email-tomeu.vizoso@collabora.com>
Message-ID: <20140909191205.19023.61366@quantum>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Quoting Tomeu Vizoso (2014-09-09 07:04:57)
> In preparation to change the public API to return a per-user clk structure,
> remove any usage of this public API from the clock implementations.
> 
> The reason for having this in a separate commit from the one that introduces
> the implementation of the new functions is to separate the changes generated
> with Coccinelle from the rest, and keep the patches' size reasonable.
> 
> Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
> Tested-by: Boris Brezillon <boris.brezillon@free-electrons.com>
> Tested-by: Heiko Stuebner <heiko@sntech.de>
> Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com>
> 
> ---
> 
> v10: * Add a few more files to be converted
>      * Re-generate the patch on top of the latest changes

Hi Tomeu,

Generating this on top of linux-next is a no-go. I can't apply it to my
tree. The best thing is to generate it on top of -rc4, and that is what
I will merge.

Running the script against linux-next is still very useful and lets us
patch up the stuff that is not going through the clk tree. E.g. the LPSS
driver is already in mainline, so just running the semantic patch
against -rc4 is sufficient for it. However a patch like Shawn's "ARM:
imx: add an exclusive gate clock type" came in through the i.MX tree and
we'll need to patch it after the fact.

The best way to do that is for me to host a branch with just your
changes in it that everyone can pull in as a dependency with the same
commit ids.

<snip>

> diff --git a/drivers/acpi/acpi_lpss.c b/drivers/acpi/acpi_lpss.c
> index bcbdbd2..f4c6ccf 100644
> --- a/drivers/acpi/acpi_lpss.c
> +++ b/drivers/acpi/acpi_lpss.c
> @@ -11,7 +11,6 @@
>   */
>  
>  #include <linux/acpi.h>
> -#include <linux/clk.h>
>  #include <linux/clkdev.h>
>  #include <linux/clk-provider.h>
>  #include <linux/err.h>
> @@ -78,7 +77,7 @@ struct lpss_private_data {
>         void __iomem *mmio_base;
>         resource_size_t mmio_size;
>         unsigned int fixed_clk_rate;
> -       struct clk *clk;
> +       struct clk_core *clk;
>         const struct lpss_device_desc *dev_desc;
>         u32 prv_reg_ctx[LPSS_PRV_REG_COUNT];
>  };
> @@ -229,7 +228,7 @@ static int register_device_clock(struct acpi_device *adev,
>  {
>         const struct lpss_device_desc *dev_desc = pdata->dev_desc;
>         const char *devname = dev_name(&adev->dev);
> -       struct clk *clk = ERR_PTR(-ENODEV);
> +       struct clk_core *clk = ERR_PTR(-ENODEV);
>         struct lpss_clk_data *clk_data;
>         const char *parent, *clk_name;
>         void __iomem *prv_base;

I think the following hunk is missing from your change:

--- a/drivers/acpi/acpi_lpss.c
+++ b/drivers/acpi/acpi_lpss.c
@@ -57,7 +57,7 @@ ACPI_MODULE_NAME("acpi_lpss");
 struct lpss_shared_clock {
        const char *name;
        unsigned long rate;
-       struct clk *clk;
+       struct clk_core *clk;
 };


Otherwise register_device_clock will blow up because we are assigning a
struct clk * to a struct clk_core *.

Do you mind testing with ARCH=x86_64 and allmodconfig? That will help
catch issues like this.

Regards,
Mike