soc: mediatek: SCPSYS: use module_platform_driver
PATCH v2 1/1 soc: mediatek: SCPSYS: use module_platform_driver
With regulator support to the scpsys driver needs to be able to defer probe. Probe deferral is incompatible with module_platform_driver_probe, so use module_platform_driver. module_platform_driver_probe sets the suppress_bind_attrs. To prevent unbinding the device with module_platform_driver we have to set the flag explicitly in the driver now.
---
Changes since v1:
- Add suppress_bind_attrs attribute
drivers/soc/mediatek/mtk-scpsys.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/soc/mediatek/mtk-scpsys.c +3 -1
@@ -542,11 +542,13 @@ static const struct of_device_id of_scpsys_match_tbl[] = { };static struct platform_driver scpsys_drv = { + .probe = scpsys_probe, .driver = { .name ="mtk-scpsys" , + .suppress_bind_attrs = true, .owner =THIS_MODULE , .of_match_table =of_match_ptr (of_scpsys_match_tbl), }, }; -module_platform_driver_probe (scpsys_drv, scpsys_probe); +module_platform_driver (scpsys_drv);
Signature · 1 line
2.6.2
PATCH v2 1/1 soc: mediatek: SCPSYS: use module_platform_driver
Quoted 10 lines · On Wednesday 16 December 2015 12:00:56 Sascha Hauer wrote:
With regulator support to the scpsys driver needs to be able to defer probe. Probe deferral is incompatible with
module_platform_driver_probe, so usemodule_platform_driver.module_platform_driver_probesets thesuppress_bind_attrs. To prevent unbinding the device withmodule_platform_driverwe have to set the flag explicitly in the driver now.Signed-off-by Sascha Hauer <[email protected]>---
Looks good now,
PATCH v2 1/1 soc: mediatek: SCPSYS: use module_platform_driver
Quoted 8 lines · Arnd Bergmann
Quoted 7 lines · On Wednesday 16 December 2015 12:00:56 Sascha Hauer wrote:
With regulator support to the scpsys driver needs to be able to defer probe. Probe deferral is incompatible with
module_platform_driver_probe, so usemodule_platform_driver.module_platform_driver_probesets thesuppress_bind_attrs. To prevent unbinding the device withmodule_platform_driverwe have to set the flag explicitly in the driver now.
Quoted 8 lines
Quoted 3 lines
Signed-off-by Sascha Hauer <[email protected]>---
Looks good now,
Acked-by Arnd Bergmann <[email protected]>
Applied to v4.4-next/soc
Sascha next time please don't forget to send Mediatek related patches to the corresponding mailinglist as well.
Thanks,
Matthias
PATCH v2 1/1 soc: mediatek: SCPSYS: use module_platform_driver
Hi Sasha, Matthias,
Quoted 17 lines · Matthias Brugger
Quoted 14 lines · Arnd Bergmann
Quoted 11 lines · On Wednesday 16 December 2015 12:00:56 Sascha Hauer wrote:
With regulator support to the scpsys driver needs to be able to defer probe. Probe deferral is incompatible with
module_platform_driver_probe, so usemodule_platform_driver.module_platform_driver_probesets thesuppress_bind_attrs. To prevent unbinding the device withmodule_platform_driverwe have to set the flag explicitly in the driver now.Signed-off-by Sascha Hauer <[email protected]>---
Quoted 6 lines
Quoted 4 lines
Looks good now,
Acked-by Arnd Bergmann <[email protected]>Applied to v4.4-next/soc
This patch does not compile on my system.
With "CONFIG_DEBUG_SECTION_MISMATCH=y" I see:
WARNING: vmlinux.o(.data+0x387c8): Section mismatch in reference from the variable scpsys_drv to the function .init.text:scpsys_probe() The variable scpsys_drv references
the function __init scpsys_probe()
If the reference is valid then annotate the
variable with __init* or __refdata (see linux/init.h) or name the variable:*_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console
FATAL: modpost: Section mismatches detected. Set CONFIG_SECTION_MISMATCH_WARN_ONLY=y to allow them.
The problem is that scpsys_probe() is still __init, but it references scp_domain_data[] which is __initconst.
However, scpsys_drv is not __initdata, so it cannot contain a pointer to an __init function.
The only way I know to fix this is by dropping the __init & __initconst. This is rather unfortunate.
Is there another way where we can use -EPROBE_DEFER and still preserve __init / __initconst ?
Also, since scpsys' Kconfig is "bool" not "tristate" it isn't a real module so I think we should use:builtin_platform_driver not module_platform_driver
CC'ing Paul Gortmaker who seems to be on a mass builtin_platform_driver conversion at the moment, see:
https://lkml.org/lkml/2015/12/9/952
-Dan
PATCH v2 1/1 soc: mediatek: SCPSYS: use module_platform_driver
[Re: [PATCH v2] soc: mediatek: SCPSYS: use module_platform_driver] On 17/12/2015 (Thu 10:52) Daniel Kurtz wrote:
[...]
Quoted 9 lines
Also, since scpsys'
Kconfigis "bool" not "tristate" it isn't a real module so I think we should use:builtin_platform_drivernotmodule_platform_driverCC'ing Paul Gortmaker who seems to be on a mass
builtin_platform_driverconversion at the moment, see:
https://lkml.org/lkml/2015/12/9/952
Good guess. In my giant queue I have that change. I just haven't got around to the requisite triple check and multi arch rebuild tests and put out a series for soc yet. Here is the short list for soc:
drivers-soc-make-mediatek-mtk-scpsys.c-explicitly-no.patch drivers-soc-make-qcom-spm.c-explicitly-non-modular.patch drivers-soc-make-samsung-exynos-srom.c.c-explicitly-.patch drivers-soc-make-sunxi_sram.c-explicitly-non-modular.patch drivers-soc-tegra-make-fuse-tegra.c-explicitly-non-m.patch
...and here is the patch you predicted. If someone wants to fold the change into their work then great; I'll be happy to drop mine and be with one less patch to herd along.
Paul.
--
From eb32cc2c1db8192c08dd64af69cb172b3be4c98b Mon Sep 17 00:00:00 2001
From: Paul Gortmaker <[email protected]>
Date: Fri, 4 Sep 2015 19:33:54 -0400
Subject: [PATCH] drivers/soc: make mediatek/mtk-scpsys.c explicitly
non-modular
The Kconfig currently controlling compilation of this code is:
drivers/soc/mediatek/Kconfig:config MTK_SCPSYSdrivers/soc/mediatek/Kconfig: bool "MediaTek SCPSYS Support"
...meaning that it currently is not being built as a module by anyone.
Lets remove the couple traces of modularity so that when reading the driver there is no doubt it is builtin-only.
Since module_platform_driver() uses the same init level priority as builtin_platform_driver() the init ordering remains unchanged with this commit.
drivers/soc/mediatek/mtk-scpsys.c +2 -3
@@ -15,7 +15,7 @@#include <linux/io.h> #include <linux/kernel.h> #include <linux/mfd/syscon.h> -#include <linux/module.h> +#include <linux/init.h> #include <linux/of_device.h> #include <linux/platform_device.h> #include <linux/pm_domain.h>
@@ -484,5 +484,4 @@ static struct platform_driver scpsys_drv = { .of_match_table = of_match_ptr(of_scpsys_match_tbl), }, }; - -module_platform_driver_probe (scpsys_drv, scpsys_probe); +builtin_platform_driver_probe (scpsys_drv, scpsys_probe);
Signature · 1 line
2.5.0
Matthias Brugger
maintainerPATCH v2 1/1 soc: mediatek: SCPSYS: use module_platform_driver
Hi Paul,
Quoted 24 lines · Paul Gortmaker
[Re: [PATCH v2] soc: mediatek: SCPSYS: use
module_platform_driver] On 17/12/2015 (Thu 10:52) Daniel Kurtz wrote:[...]
Quoted 9 lines
Also, since scpsys'
Kconfigis "bool" not "tristate" it isn't a real module so I think we should use:builtin_platform_drivernotmodule_platform_driverCC'ing Paul Gortmaker who seems to be on a mass
builtin_platform_driverconversion at the moment, see:
https://lkml.org/lkml/2015/12/9/952Good guess. In my giant queue I have that change. I just haven't got around to the requisite triple check and multi arch rebuild tests and put out a series for soc yet. Here is the short list for soc:
drivers-soc-make-mediatek-mtk-scpsys.c-explicitly-no.patch drivers-soc-make-qcom-spm.c-explicitly-non-modular.patch drivers-soc-make-samsung-exynos-srom.c.c-explicitly-.patch drivers-soc-make-sunxi_sram.c-explicitly-non-modular.patch drivers-soc-tegra-make-fuse-tegra.c-explicitly-non-m.patch
Quoted 4 lines
...and here is the patch you predicted. If someone wants to fold the change into their work then great; I'll be happy to drop mine and be with one less patch to herd along.
I take this through my tree.
Thanks a lot.
Matthias
Quoted 53 lines
Paul.
--From
eb32cc2c1db8192c08dd64af69cb172b3be4c98bMon Sep 17 00:00:00 2001
From: Paul Gortmaker <[email protected]>
Date: Fri, 4 Sep 2015 19:33:54 -0400
Subject: [PATCH]drivers/soc: makemediatek/mtk-scpsys.cexplicitly
non-modularThe
Kconfigcurrently controlling compilation of this code is:drivers/soc/mediatek/Kconfig:config
MTK_SCPSYSdrivers/soc/mediatek/Kconfig: bool "MediaTek SCPSYS Support"...meaning that it currently is not being built as a module by anyone.
Lets remove the couple traces of modularity so that when reading the driver there is no doubt it is builtin-only.
Since
module_platform_driver()uses the same init level priority asbuiltin_platform_driver()the init ordering remains unchanged with this commit.Cc Matthias Brugger <[email protected]>Cc Sascha Hauer <[email protected]>Cc Daniel Kurtz <[email protected]>Cc Paul Gortmaker <[email protected]>Cc Olof Johansson <[email protected]>Cc linux-arm-kernel at lists.infradead.orgCc linux-mediatek at lists.infradead.orgSigned-off-by Paul Gortmaker <[email protected]>drivers/soc/mediatek/mtk-scpsys.c +2
-3@@ -15,7 +15,7 @@#include <linux/io.h> #include <linux/kernel.h> #include <linux/mfd/syscon.h> -#include <linux/module.h> +#include <linux/init.h> #include <linux/of_device.h> #include <linux/platform_device.h> #include <linux/pm_domain.h> @@ -484,5 +484,4 @@ static struct platform_driver scpsys_drv = { .of_match_table = of_match_ptr(of_scpsys_match_tbl), }, }; - -module_platform_driver_probe (scpsys_drv, scpsys_probe); +builtin_platform_driver_probe (scpsys_drv, scpsys_probe);