convert HS-MMC driver to hwmod + runtime PM
Subsystems (6)
cover convert HS-MMC driver to hwmod + runtime PM
This series converts the OMAP HS-MMC driver to use omap_hwmod + runtime PM API.
Depends on MMC hwmods available in 'pm-wip/hwmods' branch of my git tree[1] as well as previously posted runtime PM series:
[PATCH/RFC 0/2] initial runtime PM layer for OMAP
The easies way to experiment/test is to use my 'pm-wip/mmc' branch which has all the dependencies, and is based on omap/for-next'. It has been tested by merging with current PM branch.
[1] http://git.kernel.org/?p=linux/kernel/git/khilman/linux-omap-pm.git
Kevin Hilman (3):
MMC: OMAP HS-MMC: convert to dev_pm_ops
OMAP3EVM: MMC: enable power-saving mode
OMAP2/3 MMC: initial conversion to runtime PM
Paul Walmsley (1):
OMAP: MMC (core): split device registration by OMAP
arch/arm/mach-omap1/devices.c | 45 ++++++- arch/arm/mach-omap2/board-omap3evm.c | 1 + arch/arm/mach-omap2/devices.c | 111 ++++++++------- arch/arm/mach-omap2/hsmmc.c | 12 +- arch/arm/plat-omap/devices.c | 50 ------- arch/arm/plat-omap/include/plat/mmc.h | 5 + drivers/mmc/host/omap_hsmmc.c | 241 ++++++++++++++++----------------- 7 files changed, 233 insertions(+), 232 deletions(-)
PATCH 1/4 MMC: OMAP HS-MMC: convert to dev_pm_ops
Convert PM operations to use dev_pm_ops. This will facilitate the runtime PM coversion which will add to dev_pm_ops hooks.
Note that dev_pm_ops version of the suspend hook no longer takes a 'state' argument. However, the MMC core function mmc_suspend_host() still takes a 'state' argument, but it is unused, so a dummy state variable was created to pass to the MMC core.
In the future, the MMC core should be converted to drop this state argument and the rest of the MMC drivers could be easily converted to dev_pm_ops as well.
--- drivers/mmc/host/omap_hsmmc.c | 15 +++++++++++---- 1 files changed, 11 insertions(+), 4 deletions(-)
drivers/mmc/host/omap_hsmmc.c +11 -4
@@ -2238,10 +2238,12 @@ static int omap_hsmmc_remove(struct platform_device *pdev) }#ifdef CONFIG_PM -static int omap_hsmmc_suspend (struct platform_device *pdev,pm_message_t state) +static int omap_hsmmc_suspend (struct device *dev) {int ret =0 ; +struct platform_device *pdev =to_platform_device (dev);struct omap_hsmmc_host *host =platform_get_drvdata (pdev); +pm_message_t state =PMSG_SUSPEND ;/* unused by MMC core */ if (host && host->suspended)return 0 ;
@@ -2290,9 +2292,10 @@ static int omap_hsmmc_suspend(struct platform_device *pdev, pm_message_t state) }/* Routine to resume the MMC device */ -static int omap_hsmmc_resume (struct platform_device *pdev) +static int omap_hsmmc_resume (struct device *dev) {int ret =0 ; +struct platform_device *pdev =to_platform_device (dev);struct omap_hsmmc_host *host =platform_get_drvdata (pdev);if (host && !host->suspended)
@@ -2343,13 +2346,17 @@ clk_en_err:#define omap_hsmmc_resume NULL#endif -static struct platform_driver omap_hsmmc_driver = { - .remove = omap_hsmmc_remove, +static struct dev_pm_ops omap_hsmmc_dev_pm_ops = { .suspend = omap_hsmmc_suspend, .resume = omap_hsmmc_resume, +}; + +static struct platform_driver omap_hsmmc_driver = { + .remove = omap_hsmmc_remove, .driver = { .name =DRIVER_NAME , .owner =THIS_MODULE , + .pm = &omap_hsmmc_dev_pm_ops, }, };
Signature · 1 line
1.6.6
PATCH 2/4 OMAP3EVM: MMC: enable power-saving mode
In reply to Kevin Hilman--- arch/arm/mach-omap2/board-omap3evm.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-)
arch/arm/mach-omap2/board-omap3evm.c +1 -0
@@ -191,6 +191,7 @@ static struct omap2_hsmmc_info mmc[] = { .wires = 4, .gpio_cd = -EINVAL, .gpio_wp = 63, + .power_saving = true, }, {}/* Terminator */ };
Signature · 1 line
1.6.6
PATCH 3/4 OMAP: MMC (core): split device registration by OMAP
In reply to Kevin HilmanFrom: Paul Walmsley <[email protected]>
Split OMAP-common MMC device registration into OMAP1 and OMAP2 components. This is in preparation for the omap_device conversion of the HSMMC portions.
Original version by Paul: http://marc.info/?l=linux-omap&m=124419789024565&w=2
Updated for current linux-omap by Kevin.
--- arch/arm/mach-omap1/devices.c | 45 ++++++++++++++++++++++++++++++++++++- arch/arm/mach-omap2/devices.c | 45 ++++++++++++++++++++++++++++++++++++- arch/arm/plat-omap/devices.c | 50 ----------------------------------------- 3 files changed, 88 insertions(+), 52 deletions(-)
arch/arm/mach-omap1/devices.c +44 -1
@@ -160,6 +160,49 @@ static inline void omap1_mmc_mux(struct omap_mmc_platform_data *mmc_controller, } } +#define OMAP_MMC_NR_RES 2 + +/* +* Register MMC devices. Called from mach-omap1 and mach-omap2 device init. +*/ +int __init omap1_mmc_add (const char *name,int id,unsigned long base, +unsigned long size,unsigned int irq, +struct omap_mmc_platform_data *data) +{ +struct platform_device *pdev; +struct resource res[OMAP_MMC_NR_RES ]; +int ret; + + pdev =platform_device_alloc (name, id); +if (!pdev) +return -ENOMEM ; + +memset (res,0 ,OMAP_MMC_NR_RES *sizeof (struct resource )); + res[0 ].start = base; + res[0 ].end = base + size -1 ; + res[0 ].flags =IORESOURCE_MEM ; + res[1 ].start = res[1 ].end = irq; + res[1 ].flags =IORESOURCE_IRQ ; + + ret =platform_device_add_resources (pdev, res,ARRAY_SIZE (res)); +if (ret ==0 ) + ret =platform_device_add_data (pdev, data,sizeof (*data)); +if (ret) + goto fail; + + ret =platform_device_add (pdev); +if (ret) + goto fail; + +/* return device handle to board setup code */ + data->dev = &pdev->dev; +return 0 ; + +fail: +platform_device_put (pdev); +return ret; +} +void __init omap1_init_mmc (struct omap_mmc_platform_data **mmc_data,int nr_controllers) {
@@ -190,7 +233,7 @@ void __init omap1_init_mmc(struct omap_mmc_platform_data **mmc_data, } size =OMAP1_MMC_SIZE ; -omap_mmc_add ("mmci-omap" , i, base, size, irq, mmc_data[i]); +omap1_mmc_add ("mmci-omap" , i, base, size, irq, mmc_data[i]); }; }
arch/arm/mach-omap2/devices.c +44 -1
57 lines not on this page · raw message
arch/arm/plat-omap/devices.c +0 -50
56 lines not on this page · raw message
Signature · 1 line
1.6.6
PATCH 4/4 OMAP2/3 MMC: initial conversion to runtime PM
In reply to Kevin HilmanConvert the HSMMC driver to use the runtime PM layer. A notable aspect of this is that the use of the dev_attr data from the omap_hwmod allows the redaction of all of the integration-specific hacks inside this driver. Regulator control has not yet been converted; the driver still uses the platform_data set_power hook.
In converting to runtime PM layer, the clock frameworks is no longer used for fclk and iclk. Instead, the runtime PM get and put functions are used. These result in calls into the OMAP runtime PM core which internally uses the omap_device API to disable/re-enable the device. (Note that the 2430 debounce clock is not currently handled here, only the fclk and iclk are managed.)
In addition, the context_loss tracking has been removed from the driver and is now handled by omap_device + runtime PM core. The driver's ->runtime_suspend() hook will be called before device is disabled, and the driver's ->runtime_resume() hook will be called if context has been lost.
Based on an initial conversion of this driver to use omap_device by by Paul Walmsely: http://marc.info/?l=linux-omap&m=124419789124570&w=2
and then converted to use runtime PM API instead of omap_device API.
The omap_hsmmc driver has some hacks in this version to compensate for the fact that the main runtime PM core prevents runtime transitions during suspend. This prevens us from using common hooks for runtime PM and static PM. Current workaround is to hack in some extra put/get calls in the suspend/resume path while this issue is being discussed on linux-pm.
--- arch/arm/mach-omap2/devices.c | 152 +++++++++-------------- arch/arm/mach-omap2/hsmmc.c | 12 +- arch/arm/plat-omap/include/plat/mmc.h | 5 + drivers/mmc/host/omap_hsmmc.c | 226 +++++++++++++++------------------ 4 files changed, 176 insertions(+), 219 deletions(-)
arch/arm/mach-omap2/devices.c +58 -94
@@ -15,6 +15,7 @@#include <linux/platform_device.h> #include <linux/io.h> #include <linux/clk.h> +#include <linux/err.h> #include <mach/hardware.h> #include <asm/mach-types.h>
@@ -29,6 +30,9 @@#include "mux.h" +#include <plat/omap_device.h> +#include <plat/omap_hwmod.h> +#if defined(CONFIG_VIDEO_OMAP2 ) || defined(CONFIG_VIDEO_OMAP2_MODULE ) static struct resource cam_resources[] = {
@@ -588,6 +592,18 @@ static inline void omap_hsmmc_reset(void) {}#if defined(CONFIG_MMC_OMAP ) || defined(CONFIG_MMC_OMAP_MODULE ) || \ defined(CONFIG_MMC_OMAP_HS ) || defined(CONFIG_MMC_OMAP_HS_MODULE ) +static struct omap_device_pm_latency omap_hsmmc_latency[] = { + [0 ] = { + .deactivate_func = omap_device_idle_hwmods, + .activate_func = omap_device_enable_hwmods, + .flags =OMAP_DEVICE_LATENCY_AUTO_ADJUST , + }, +/* +* XXX There should also be an entry here to power off/on the +* MMC regulators/PBIAS cells, etc. +*/ +}; +static inline void omap2_mmc_mux (struct omap_mmc_platform_data *mmc_controller,int controller_nr) {
@@ -686,110 +702,58 @@ static inline void omap2_mmc_mux(struct omap_mmc_platform_data *mmc_controller, } } -#define OMAP_MMC_NR_RES 2 - -/* -* Register MMC devices. Called from mach-omap1 and mach-omap2 device init. -*/ -int __init omap2_mmc_add (const char *name,int id,unsigned long base, -unsigned long size,unsigned int irq, -struct omap_mmc_platform_data *data) -{ -struct platform_device *pdev; -struct resource res[OMAP_MMC_NR_RES ]; -int ret; - - pdev =platform_device_alloc (name, id); -if (!pdev) -return -ENOMEM ; - -memset (res,0 ,OMAP_MMC_NR_RES *sizeof (struct resource )); - res[0 ].start = base; - res[0 ].end = base + size -1 ; - res[0 ].flags =IORESOURCE_MEM ; - res[1 ].start = res[1 ].end = irq; - res[1 ].flags =IORESOURCE_IRQ ; - - ret =platform_device_add_resources (pdev, res,ARRAY_SIZE (res)); -if (ret ==0 ) - ret =platform_device_add_data (pdev, data,sizeof (*data)); -if (ret) - goto fail; - - ret =platform_device_add (pdev); -if (ret) - goto fail; - -/* return device handle to board setup code */ - data->dev = &pdev->dev; -return 0 ; - -fail: -platform_device_put (pdev); -return ret; -} +#define MAX_OMAP_MMC_HWMOD_NAME_LEN 16void __init omap2_init_mmc (struct omap_mmc_platform_data **mmc_data, -int nr_controllers) +int nr_controllers) { -int i; +struct omap_hwmod *oh; +struct omap_device *od; +struct omap_device_pm_latency *ohl; +char oh_name[MAX_OMAP_MMC_HWMOD_NAME_LEN ];char *name; +int i, l; +int ohl_cnt =0 ; -for (i =0 ; i < nr_controllers; i++) { -unsigned long base, size; -unsigned int irq =0 ; +if (cpu_is_omap2420 ()) { + name ="mmci-omap" ; + }else { + name ="mmci-omap-hs" ; + ohl = omap_hsmmc_latency; + ohl_cnt =ARRAY_SIZE (omap_hsmmc_latency); + } -if (!mmc_data[i]) +/* +* XXX This isn't a good way to set these up. What if a board +* uses MMC2 but not MMC1? +*/ +for (i =1 ; i <= nr_controllers; i++) { +int idx = i -1 ; + + l =snprintf (oh_name,MAX_OMAP_MMC_HWMOD_NAME_LEN , +"mmc%d_hwmod" , i); +WARN (l >=MAX_OMAP_MMC_HWMOD_NAME_LEN , +"String buffer overflow in MMC%d device setup\n" , i); + oh =omap_hwmod_lookup (oh_name); +if (!oh) { +pr_err ("Could not look up %s\n" , oh_name);continue ; + } -omap2_mmc_mux (mmc_data[i], i); + mmc_data[idx]->dev_attr = oh->dev_attr; -switch (i) { -case 0 : - base =OMAP2_MMC1_BASE ; - irq =INT_24XX_MMC_IRQ ; -break ; -case 1 : - base =OMAP2_MMC2_BASE ; - irq =INT_24XX_MMC2_IRQ ; -break ; -case 2 : -if (!cpu_is_omap44xx () && !cpu_is_omap34xx ()) -return ; - base =OMAP3_MMC3_BASE ; - irq =INT_34XX_MMC3_IRQ ; -break ; -case 3 : -if (!cpu_is_omap44xx ()) -return ; - base =OMAP4_MMC4_BASE +OMAP4_MMC_REG_OFFSET ; - irq =INT_44XX_MMC4_IRQ ; -break ; -case 4 : -if (!cpu_is_omap44xx ()) -return ; - base =OMAP4_MMC5_BASE +OMAP4_MMC_REG_OFFSET ; - irq =INT_44XX_MMC5_IRQ ; -break ; -default : -continue ; - } +omap2_mmc_mux (mmc_data[idx], idx); + od =omap_device_build (name, idx, oh, mmc_data[idx], +sizeof (struct omap_mmc_platform_data ), + ohl, ohl_cnt); +WARN (IS_ERR (od),"Could not build omap_device for %s %s\n" , + name, oh_name); -if (cpu_is_omap2420 ()) { - size =OMAP2420_MMC_SIZE ; - name ="mmci-omap" ; - }else if (cpu_is_omap44xx ()) { -if (i <3 ) { - base +=OMAP4_MMC_REG_OFFSET ; - irq +=IRQ_GIC_START ; - } - size =OMAP4_HSMMC_SIZE ; - name ="mmci-omap-hs" ; - }else { - size =OMAP3_HSMMC_SIZE ; - name ="mmci-omap-hs" ; - } -omap2_mmc_add (name, i, base, size, irq, mmc_data[i]); +/* +* return device handle to board setup code +* XXX Can this be removed? +*/ + mmc_data[idx]->dev = &od->pdev.dev; }; }
arch/arm/mach-omap2/hsmmc.c +8 -4
@@ -1,8 +1,8 @@/* -* linux/arch/arm/mach-omap2/hsmmc.c +* OMAP HS-MMC platform init * -* Copyright (C) 2007-2008 Texas Instruments -* Copyright (C) 2008 Nokia Corporation +* Copyright (C) 2007-2008,2010 Texas Instruments +* Copyright (C) 2008-2009 Nokia Corporation * Author: Texas Instruments * * This program is free software; you can redistribute it and/or modify
@@ -17,6 +17,7 @@#include <plat/control.h> #include <plat/mmc.h> #include <plat/omap-pm.h> +#include <plat/omap_device.h> #include "hsmmc.h"
@@ -145,6 +146,7 @@ void __init omap2_hsmmc_init(struct omap2_hsmmc_info *controllers) {struct omap2_hsmmc_info *c;int nr_hsmmc =ARRAY_SIZE (hsmmc_data); +int controller_cnt =0 ;if (cpu_is_omap2430 ()) { control_pbias_offset =OMAP243X_CONTROL_PBIAS_LITE ;
@@ -167,6 +169,8 @@ void __init omap2_hsmmc_init(struct omap2_hsmmc_info *controllers)continue ; } + controller_cnt++; + mmc =kzalloc (sizeof (struct omap_mmc_platform_data ),GFP_KERNEL );if (!mmc) {
@@ -246,7 +250,7 @@ void __init omap2_hsmmc_init(struct omap2_hsmmc_info *controllers) hsmmc_data[c->mmc -1 ] = mmc; } -omap2_init_mmc (hsmmc_data,OMAP34XX_NR_MMC ); +omap2_init_mmc (hsmmc_data, controller_cnt);/* pass the device nodes back to board setup code */ for (c = controllers; c->mmc; c++) {
arch/arm/plat-omap/include/plat/mmc.h +5 -0
@@ -14,8 +14,10 @@#include <linux/types.h> #include <linux/device.h> #include <linux/mmc/host.h> +#include <linux/platform_device.h> #include <plat/board.h> +#include <plat/omap_hwmod.h> #define OMAP15XX_NR_MMC 1#define OMAP16XX_NR_MMC 2
@@ -79,6 +81,9 @@ struct omap_mmc_platform_data {u64 dma_mask; +/* integration attributes from the omap_hwmod layer */ +struct mmc_dev_attr *dev_attr; +struct omap_mmc_slot_data {/* 4 wire signaling is optional, and is used for SD/SDIO/HSMMC;
drivers/mmc/host/omap_hsmmc.c +105 -121
444 lines not on this page · raw message
Signature · 1 line
1.6.6
Re: [PATCH/RFC 4/4] OMAP2/3 MMC: initial conversion to runtime PM
Quoted 30 lines · Kevin Hilman
Convert the HSMMC driver to use the runtime PM layer. A notable aspect of this is that the use of the
dev_attrdata from theomap_hwmodallows the redaction of all of the integration-specific hacks inside this driver. Regulator control has not yet been converted; the driver still uses theplatform_dataset_powerhook.In converting to runtime PM layer, the clock frameworks is no longer used for fclk and iclk. Instead, the runtime PM get and put functions are used. These result in calls into the OMAP runtime PM core which internally uses the
omap_deviceAPI to disable/re-enable the device. (Note that the 2430 debounce clock is not currently handled here, only the fclk and iclk are managed.)In addition, the
context_losstracking has been removed from the driver and is now handled byomap_device+ runtime PM core. The driver's ->runtime_suspend() hook will be called before device is disabled, and the driver's ->runtime_resume() hook will be called if context has been lost.Based on an initial conversion of this driver to use
omap_deviceby by Paul Walmsely: http://marc.info/?l=linux-omap&m=124419789124570&w=2
and then converted to use runtime PM API instead ofomap_deviceAPI.The
omap_hsmmcdriver has some hacks in this version to compensate for the fact that the main runtime PM core prevents runtime transitions during suspend. This prevens us from using common hooks for runtime PM and static PM. Current workaround is to hack in some extra put/get calls in the suspend/resume path while this issue is being discussed on linux-pm.
Quoted 8 lines
Cc Paul Walmsley <[email protected]>Signed-off-by Kevin Hilman <[email protected]>--- arch/arm/mach-omap2/devices.c | 152 +++++++++-------------- arch/arm/mach-omap2/hsmmc.c | 12 +- arch/arm/plat-omap/include/plat/mmc.h | 5 + drivers/mmc/host/omap_hsmmc.c | 226 +++++++++++++++------------------ 4 files changed, 176 insertions(+), 219 deletions(-)
Why is PM runtime not simply PM?
Can pm_runtime_put_sync() and pm_runtime_get_sync() have less obscure names?
<snip>
Re: [PATCH/RFC 4/4] OMAP2/3 MMC: initial conversion to runtime PM
Quoted 40 lines · Adrian Hunter
Quoted 30 lines · Kevin Hilman
Convert the HSMMC driver to use the runtime PM layer. A notable aspect of this is that the use of the
dev_attrdata from theomap_hwmodallows the redaction of all of the integration-specific hacks inside this driver. Regulator control has not yet been converted; the driver still uses theplatform_dataset_powerhook.In converting to runtime PM layer, the clock frameworks is no longer used for fclk and iclk. Instead, the runtime PM get and put functions are used. These result in calls into the OMAP runtime PM core which internally uses the
omap_deviceAPI to disable/re-enable the device. (Note that the 2430 debounce clock is not currently handled here, only the fclk and iclk are managed.)In addition, the
context_losstracking has been removed from the driver and is now handled byomap_device+ runtime PM core. The driver's ->runtime_suspend() hook will be called before device is disabled, and the driver's ->runtime_resume() hook will be called if context has been lost.Based on an initial conversion of this driver to use
omap_deviceby by Paul Walmsely: http://marc.info/?l=linux-omap&m=124419789124570&w=2
and then converted to use runtime PM API instead ofomap_deviceAPI.The
omap_hsmmcdriver has some hacks in this version to compensate for the fact that the main runtime PM core prevents runtime transitions during suspend. This prevens us from using common hooks for runtime PM and static PM. Current workaround is to hack in some extra put/get calls in the suspend/resume path while this issue is being discussed on linux-pm.Quoted 8 lines
Cc Paul Walmsley <[email protected]>Signed-off-by Kevin Hilman <[email protected]>--- arch/arm/mach-omap2/devices.c | 152 +++++++++-------------- arch/arm/mach-omap2/hsmmc.c | 12 +- arch/arm/plat-omap/include/plat/mmc.h | 5 + drivers/mmc/host/omap_hsmmc.c | 226 +++++++++++++++------------------ 4 files changed, 176 insertions(+), 219 deletions(-)
Quoted 3 lines
Why is PM runtime not simply PM?
Can
pm_runtime_put_sync()andpm_runtime_get_sync()have less obscure names?
The runtime PM names are not my creation. This is a new PM framework in mainline that I am taking advantage of.
See:
http://elinux.org/OMAP_Power_Management#Run-time_PM
Kevin
RE: [PATCH/RFC 4/4] OMAP2/3 MMC: initial conversion to runtime PM
In reply to Kevin Hilman<snip>
drivers/mmc/host/omap_hsmmc.c err_irq_cd_init: · line 2130
Quoted 4 lines
err_irq:drivers/mmc/host/omap_hsmmc.c +1
-1mmc_host_disable (host->mmc); -clk_disable (host->iclk); +pm_runtime_suspend (host->dev);
Why not pm_runtime_put_sync() here? It can replace the calls:mmc_host_disable(host->mmc); clk_disable(host->iclk);
Quoted 7 lines
drivers/mmc/host/omap_hsmmc.c +1
-0clk_put (host->fclk);clk_put (host->iclk); +if (host->got_dbclk) {clk_disable (host->dbclk);clk_put (host->dbclk);
drivers/mmc/host/omap_hsmmc.c static int omap_hsmmc_remove(struct platform_device · line 2169
Quoted 7 lines
drivers/mmc/host/omap_hsmmc.c +2
-1@@ -2216,7 +2164,8 @@ static int omap_hsmmc_remove(struct platform_device *pdev)flush_scheduled_work ();mmc_host_disable (host->mmc); -clk_disable (host->iclk); +pm_runtime_suspend (host->dev); +
Ditto
Quoted 46 lines
(quoted diff) +0
-0clk_put(host->fclk); clk_put(host->iclk);drivers/mmc/host/omap_hsmmc.c +12
-5if (host->got_dbclk) {@@ -2272,7 +2221,7 @@ static int omap_hsmmc_suspend(struct device *dev)OMAP_HSMMC_WRITE (host->base,HCTL ,OMAP_HSMMC_READ (host->base,HCTL ) & ~SDBP );mmc_host_disable (host->mmc); -clk_disable (host->iclk); +if (host->got_dbclk)clk_disable (host->dbclk); }else {@@ -2287,6 +2236,11 @@ static int omap_hsmmc_suspend(struct device *dev)mmc_host_disable (host->mmc); } +/* +* HACK: "extra" put to compensate for DPM core keeping +* runtime PM disabled. -- khilman +*/ +pm_runtime_put_sync (host->dev); }return ret; }@@ -2302,12 +2256,14 @@ static int omap_hsmmc_resume(struct device *dev)return 0 ;if (host) { - ret =clk_enable (host->iclk); -if (ret) - goto clk_en_err; +/* +* HACK: "extra" get to compensate for DPM core keeping +* runtime PM disabled. -- khilman +*/ +pm_runtime_get_sync (host->dev);if (mmc_host_enable (host->mmc) !=0 ) { -clk_disable (host->iclk); +pm_runtime_suspend (host->dev); goto clk_en_err; }@@ -2346,9 +2302,37 @@ clk_en_err:#define omap_hsmmc_resume NULL#endif
drivers/mmc/host/omap_hsmmc.c clk_en_err: · line 2313
Quoted 10 lines
drivers/mmc/host/omap_hsmmc.c +9
-0+/* called just before device is disabled */ +static int omap_hsmmc_runtime_suspend (struct device *dev) +{ +struct omap_hsmmc_host *host; + +dev_dbg (dev,"%s\n" , __func__); + + host =platform_get_drvdata (to_platform_device (dev)); +omap_hsmmc_context_save (host);
The context_save fn is now empty. How does it help here?
Quoted 32 lines
drivers/mmc/host/omap_hsmmc.c +19
-1+ +return 0 ; +} + +/* called after device is (re)enabled, ONLY if context was lost */ +static int omap_hsmmc_runtime_resume (struct device *dev) +{ +struct omap_hsmmc_host *host; + +dev_dbg (dev,"%s\n" , __func__); + + host =platform_get_drvdata (to_platform_device (dev)); +omap_hsmmc_context_restore (host); + +return 0 ; +} +static struct dev_pm_ops omap_hsmmc_dev_pm_ops = { .suspend = omap_hsmmc_suspend, .resume = omap_hsmmc_resume, + .runtime_suspend = omap_hsmmc_runtime_suspend, + .runtime_resume = omap_hsmmc_runtime_resume, };static struct platform_driver omap_hsmmc_driver = { --1.6.6
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH/RFC 4/4] OMAP2/3 MMC: initial conversion to runtime PM
Quoted 10 lines · Madhusudhan
<snip>
drivers/mmc/host/omap_hsmmc.c err_irq_cd_init: · line 2130
Quoted 4 lines
err_irq:drivers/mmc/host/omap_hsmmc.c +1
-1mmc_host_disable (host->mmc); -clk_disable (host->iclk); +pm_runtime_suspend (host->dev);Why not
pm_runtime_put_sync()here? It can replace the calls:mmc_host_disable(host->mmc);clk_disable(host->iclk);
Not knowing a ton about this driver, I'd rather keep the mmc_host_disable() since it matches the mmc_host_enable() earlier.
Quoted 65 lines
Quoted 7 lines
drivers/mmc/host/omap_hsmmc.c +1
-0clk_put (host->fclk);clk_put (host->iclk); +if (host->got_dbclk) {clk_disable (host->dbclk);clk_put (host->dbclk);drivers/mmc/host/omap_hsmmc.c static int omap_hsmmc_remove(struct platform_device · line 2169
Quoted 7 lines
drivers/mmc/host/omap_hsmmc.c +2
-1@@ -2216,7 +2164,8 @@ static int omap_hsmmc_remove(struct platform_device *pdev)flush_scheduled_work ();mmc_host_disable (host->mmc); -clk_disable (host->iclk); +pm_runtime_suspend (host->dev); +Ditto
Quoted 47 lines
(quoted diff) +0
-0clk_put(host->fclk); clk_put(host->iclk);drivers/mmc/host/omap_hsmmc.c +12
-5if (host->got_dbclk) {@@ -2272,7 +2221,7 @@ static int omap_hsmmc_suspend(struct device *dev)OMAP_HSMMC_WRITE (host->base,HCTL ,OMAP_HSMMC_READ (host->base,HCTL ) & ~SDBP );mmc_host_disable (host->mmc); -clk_disable (host->iclk); +if (host->got_dbclk)clk_disable (host->dbclk); }else {@@ -2287,6 +2236,11 @@ static int omap_hsmmc_suspend(struct device *dev)mmc_host_disable (host->mmc); } +/* +* HACK: "extra" put to compensate for DPM core keeping +* runtime PM disabled. -- khilman +*/ +pm_runtime_put_sync (host->dev); }return ret; }@@ -2302,12 +2256,14 @@ static int omap_hsmmc_resume(struct device *dev)return 0 ;if (host) { - ret =clk_enable (host->iclk); -if (ret) - goto clk_en_err; +/* +* HACK: "extra" get to compensate for DPM core keeping +* runtime PM disabled. -- khilman +*/ +pm_runtime_get_sync (host->dev);if (mmc_host_enable (host->mmc) !=0 ) { -clk_disable (host->iclk); +pm_runtime_suspend (host->dev); goto clk_en_err; }@@ -2346,9 +2302,37 @@ clk_en_err:#define omap_hsmmc_resume NULL#endif
Quoted 11 lines
drivers/mmc/host/omap_hsmmc.c clk_en_err: · line 2313
Quoted 9 lines
drivers/mmc/host/omap_hsmmc.c +9
-0+/* called just before device is disabled */ +static int omap_hsmmc_runtime_suspend (struct device *dev) +{ +struct omap_hsmmc_host *host; + +dev_dbg (dev,"%s\n" , __func__); + + host =platform_get_drvdata (to_platform_device (dev)); +omap_hsmmc_context_save (host);The
context_savefn is now empty. How does it help here?
It doesn't add anything, but I put it there to show that this is the only place that context save would be needed, and as an example to other drivers.
Kevin
Quoted 32 lines
Quoted 32 lines
drivers/mmc/host/omap_hsmmc.c +19
-1+ +return 0 ; +} + +/* called after device is (re)enabled, ONLY if context was lost */ +static int omap_hsmmc_runtime_resume (struct device *dev) +{ +struct omap_hsmmc_host *host; + +dev_dbg (dev,"%s\n" , __func__); + + host =platform_get_drvdata (to_platform_device (dev)); +omap_hsmmc_context_restore (host); + +return 0 ; +} +static struct dev_pm_ops omap_hsmmc_dev_pm_ops = { .suspend = omap_hsmmc_suspend, .resume = omap_hsmmc_resume, + .runtime_suspend = omap_hsmmc_runtime_suspend, + .runtime_resume = omap_hsmmc_runtime_resume, };static struct platform_driver omap_hsmmc_driver = { --1.6.6
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 11 lines · Kevin Hilman
This series converts the OMAP HS-MMC driver to use
omap_hwmod+ runtime PM API.Depends on MMC hwmods available in 'pm-wip/hwmods' branch of my git tree[1] as well as previously posted runtime PM series:
[PATCH/RFC 0/2] initial runtime PM layer for OMAP
The easies way to experiment/test is to use my 'pm-wip/mmc' branch which has all the dependencies, and is based on omap/for-next'. It has been tested by merging with current PM branch.
One more thing for those wanting to test this:
You can start from omap3_pm_defconfig and then CONFIG_PM_RUNTIME needs to be enabled.
Kevin
Quoted 18 lines
[1] http://git.kernel.org/?p=linux/kernel/git/khilman/linux-omap-pm.git
Kevin Hilman (3):
MMC: OMAP HS-MMC: convert todev_pm_ops
OMAP3EVM: MMC: enable power-saving mode
OMAP2/3 MMC: initial conversion to runtime PMPaul Walmsley (1):
OMAP: MMC (core): split device registration by OMAParch/arm/mach-omap1/devices.c | 45 ++++++- arch/arm/mach-omap2/board-omap3evm.c | 1 + arch/arm/mach-omap2/devices.c | 111 ++++++++------- arch/arm/mach-omap2/hsmmc.c | 12 +- arch/arm/plat-omap/devices.c | 50 ------- arch/arm/plat-omap/include/plat/mmc.h | 5 + drivers/mmc/host/omap_hsmmc.c | 241 ++++++++++++++++----------------- 7 files changed, 233 insertions(+), 232 deletions(-)
Quoted 6 lines · Kevin Hilman
This series converts the OMAP HS-MMC driver to use
omap_hwmod+ runtime PM API.Depends on MMC hwmods available in 'pm-wip/hwmods' branch of my git tree[1] as well as previously posted runtime PM series:
Quoted 8 lines
[PATCH/RFC 0/2] initial runtime PM layer for OMAP
The easies way to experiment/test is to use my 'pm-wip/mmc' branch which has all the dependencies, and is based on omap/for-next'. It has been tested by merging with current PM branch.
[1] http://git.kernel.org/?p=linux/kernel/git/khilman/linux-omap-pm.git
A question for those of you who actually understand the MMC driver... I'm having problems getting my head around the current PM stuff in the MMC driver. My primary question is:
Why does the suspend hook need to re-enable the device before suspending?
When using runtime PM, the MMC device is disabled including clocks off and regulator off (if power_savings == true) when there is no activity.
Then, in the static suspend hook, it's re-enabled (including taking it out of off, re-enabling regulators etc) only to be quickly disabled again. This seems horribly inefficient.
I admit to not understanding the MMC layer terribly well, so can someone enlighten me as to what is going on here?
What I am testing here is a patch on top of this series (below) that adds a check to the static suspend hook. If the device is already runtime suspended, then the suspend and resume hooks should be noop.
This appears to work just fine while testing on omap3evm just doing simple read/write tests before an after suspend resume.
Note that if you want to test this patch, it also depends on this patch to runtime PM from the linux-pm list:
https://lists.linux-foundation.org/pipermail/linux-pm/2010-February/024275.html
These are all included in an updated version of my pm-wip/mmc branch for ease of testing. Merge it with the current PM branch, enable CONFIG_PM_RUNTIME and test away.
Kevin
commit 166cba7679fa267ee6e6eb39fd1e871ede5ded16
Author: Kevin Hilman <[email protected]> Date: Tue Feb 23 16:21:56 2010 -0800
MMC: omap_hsmmc: check for runtime-suspend in static suspend
drivers/mmc/host/omap_hsmmc.c +6 -12
@@ -2222,6 +2222,9 @@ static int omap_hsmmc_suspend(struct device *dev)if (host && host->suspended)return 0 ; +if (pm_is_runtime_suspended (host->dev)) +return 0 ; +if (host) { host->suspended =1 ;if (host->pdata->suspend) {
@@ -2260,12 +2263,6 @@ static int omap_hsmmc_suspend(struct device *dev) }mmc_host_disable (host->mmc); } - -/* -* HACK: "extra" put to compensate for DPM core keeping -* runtime PM disabled. -- khilman -*/ -pm_runtime_put_sync (host->dev); }return ret; }
@@ -2280,13 +2277,10 @@ static int omap_hsmmc_resume(struct device *dev)if (host && !host->suspended)return 0 ; -if (host) { -/* -* HACK: "extra" get to compensate for DPM core keeping -* runtime PM disabled. -- khilman -*/ -pm_runtime_get_sync (host->dev); +if (pm_is_runtime_suspended (host->dev)) +return 0 ; +if (host) {if (mmc_host_enable (host->mmc) !=0 ) {pm_runtime_suspend (host->dev); goto clk_en_err;
Quoted 26 lines
-----Original Message-----
From: [email protected] [mailto:linux-omap- [email protected]] On Behalf Of Kevin Hilman
Sent: Tuesday, February 23, 2010 6:51 PM
To: [email protected]
Subject: Re: [PATCH/RFC 0/4] convert HS-MMC driver to hwmod + runtime PMQuoted 6 lines · Kevin Hilman
This series converts the OMAP HS-MMC driver to use
omap_hwmod+ runtime PM API.Depends on MMC hwmods available in 'pm-wip/hwmods' branch of my git tree[1] as well as previously posted runtime PM series:
Quoted 8 lines
[PATCH/RFC 0/2] initial runtime PM layer for OMAP
The easies way to experiment/test is to use my 'pm-wip/mmc' branch which has all the dependencies, and is based on omap/for-next'. It has been tested by merging with current PM branch.
[1] http://git.kernel.org/?p=linux/kernel/git/khilman/linux-omap-pm.git
Quoted 7 lines
A question for those of you who actually understand the MMC driver... I'm having problems getting my head around the current PM stuff in the MMC driver. My primary question is:
Why does the suspend hook need to re-enable the device before suspending?
In the scenario where there is no activity on the bus the MMC clocks are kept disabled. Now in the suspend path the MMC core will issue certain CMDs like CMD7(to end the suspend path) to deselect the card(more of a protocol stuff). Hence the host need to be in enabled state before letting the core know that there is a suspend request.
Quoted 9 lines
When using runtime PM, the MMC device is disabled including clocks off and regulator off (if
power_savings== true) when there is no activity.Then, in the static suspend hook, it's re-enabled (including taking it out of
off, re-enabling regulators etc) only to be quickly disabled again. This seems horribly inefficient.
This is exactly for the reason I mentioned above.
Quoted 10 lines
I admit to not understanding the MMC layer terribly well, so can someone enlighten me as to what is going on here?
What I am testing here is a patch on top of this series (below) that adds a check to the static suspend hook. If the device is already runtime suspended, then the suspend and resume hooks should be noop.
This appears to work just fine while testing on omap3evm just doing simple read/write tests before an after suspend resume.
I did some basic testing with your previously posted patches. But my testing was incomplete because on Zoom2 because for some reason the OFF mode was not working even without your patches.
My concern was more with respect to OFF mode in idle path since your patches removed context restore calls if I recall correctly. Are you able to hit CORE OFF and then come back and do the read/write transfers in idle as well as suspend/resume path?
Regards,
Madhu
Quoted 68 lines
Note that if you want to test this patch, it also depends on this patch to runtime PM from the linux-pm list:
https://lists.linux-foundation.org/pipermail/linux-pm/2010-
February/024275.htmlThese are all included in an updated version of my pm-wip/mmc branch for ease of testing. Merge it with the current PM branch, enable
CONFIG_PM_RUNTIMEand test away.Kevin
commit
166cba7679fa267ee6e6eb39fd1e871ede5ded16Author: Kevin Hilman <[email protected]> Date: Tue Feb 23 16:21:56 2010 -0800MMC:
omap_hsmmc: check for runtime-suspend in static suspenddrivers/mmc/host/omap_hsmmc.c +6
-13@@ -2222,6 +2222,9 @@ static int omap_hsmmc_suspend(struct device *dev)if (host && host->suspended)return 0 ; +if (pm_is_runtime_suspended (host->dev)) +return 0 ; +if (host) { host->suspended =1 ;if (host->pdata->suspend) {@@ -2260,12 +2263,6 @@ static int omap_hsmmc_suspend(struct device *dev) }mmc_host_disable (host->mmc); } - -/* -* HACK: "extra" put to compensate for DPM core keeping -* runtime PM disabled. -- khilman -*/ -pm_runtime_put_sync (host->dev); }return ret; }@@ -2280,13 +2277,10 @@ static int omap_hsmmc_resume(struct device *dev)if (host && !host->suspended)return 0 ; -if (host) { -/* -* HACK: "extra" get to compensate for DPM core keeping -* runtime PM disabled. -- khilman -*/ -pm_runtime_get_sync (host->dev); +if (pm_is_runtime_suspended (host->dev)) +return 0 ; +if (host) {if (mmc_host_enable (host->mmc) !=0 ) {pm_runtime_suspend (host->dev); goto clk_en_err; --To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 32 lines · Madhusudhan
Quoted 24 lines
-----Original Message-----
From: [email protected] [mailto:linux-omap- [email protected]] On Behalf Of Kevin Hilman
Sent: Tuesday, February 23, 2010 6:51 PM
To: [email protected]
Subject: Re: [PATCH/RFC 0/4] convert HS-MMC driver to hwmod + runtime PMQuoted 6 lines · Kevin Hilman
This series converts the OMAP HS-MMC driver to use
omap_hwmod+ runtime PM API.Depends on MMC hwmods available in 'pm-wip/hwmods' branch of my git tree[1] as well as previously posted runtime PM series:
Quoted 8 lines
[PATCH/RFC 0/2] initial runtime PM layer for OMAP
The easies way to experiment/test is to use my 'pm-wip/mmc' branch which has all the dependencies, and is based on omap/for-next'. It has been tested by merging with current PM branch.
[1] http://git.kernel.org/?p=linux/kernel/git/khilman/linux-omap-pm.git
Quoted 7 lines
A question for those of you who actually understand the MMC driver... I'm having problems getting my head around the current PM stuff in the MMC driver. My primary question is:
Why does the suspend hook need to re-enable the device before suspending?
Quoted 5 lines
In the scenario where there is no activity on the bus the MMC clocks are kept disabled. Now in the suspend path the MMC core will issue certain CMDs like CMD7(to end the suspend path) to deselect the card(more of a protocol stuff). Hence the host need to be in enabled state before letting the core know that there is a suspend request.
I guess what I'm wondering is whether we can (should) have the exact same path for runtime PM (idle) and static PM (suspend/resume.)
My current approach is that the suspend is a NOP if the device is already runtime suspended.
Quoted 19 lines
Quoted 9 lines
When using runtime PM, the MMC device is disabled including clocks off and regulator off (if
power_savings== true) when there is no activity.Then, in the static suspend hook, it's re-enabled (including taking it out of
off, re-enabling regulators etc) only to be quickly disabled again. This seems horribly inefficient.This is exactly for the reason I mentioned above.
Quoted 7 lines
I admit to not understanding the MMC layer terribly well, so can someone enlighten me as to what is going on here?
What I am testing here is a patch on top of this series (below) that adds a check to the static suspend hook. If the device is already runtime suspended, then the suspend and resume hooks should be noop.
Quoted 7 lines
Quoted 3 lines
This appears to work just fine while testing on omap3evm just doing simple read/write tests before an after suspend resume.
I did some basic testing with your previously posted patches. But my testing was incomplete because on Zoom2 because for some reason the OFF mode was not working even without your patches.
OK, I will try on Zoom2. I've only tested this so far on OMAP3EVM.
Quoted 2 lines
My concern was more with respect to OFF mode in idle path since your patches removed context restore calls if I recall correctly.
No, I didn't remove restore. I just moved it into the runtime PM resume callback.
Quoted 2 lines
Are you able to hit CORE OFF and then come back and do the read/write transfers in idle as well as suspend/resume path?
Yes.
Kevin
Quoted 69 lines
Quoted 69 lines
Note that if you want to test this patch, it also depends on this patch to runtime PM from the linux-pm list:
https://lists.linux-foundation.org/pipermail/linux-pm/2010-
February/024275.htmlThese are all included in an updated version of my pm-wip/mmc branch for ease of testing. Merge it with the current PM branch, enable
CONFIG_PM_RUNTIMEand test away.Kevin
commit
166cba7679fa267ee6e6eb39fd1e871ede5ded16Author: Kevin Hilman <[email protected]> Date: Tue Feb 23 16:21:56 2010 -0800MMC:
omap_hsmmc: check for runtime-suspend in static suspenddrivers/mmc/host/omap_hsmmc.c +6
-13@@ -2222,6 +2222,9 @@ static int omap_hsmmc_suspend(struct device *dev)if (host && host->suspended)return 0 ; +if (pm_is_runtime_suspended (host->dev)) +return 0 ; +if (host) { host->suspended =1 ;if (host->pdata->suspend) {@@ -2260,12 +2263,6 @@ static int omap_hsmmc_suspend(struct device *dev) }mmc_host_disable (host->mmc); } - -/* -* HACK: "extra" put to compensate for DPM core keeping -* runtime PM disabled. -- khilman -*/ -pm_runtime_put_sync (host->dev); }return ret; }@@ -2280,13 +2277,10 @@ static int omap_hsmmc_resume(struct device *dev)if (host && !host->suspended)return 0 ; -if (host) { -/* -* HACK: "extra" get to compensate for DPM core keeping -* runtime PM disabled. -- khilman -*/ -pm_runtime_get_sync (host->dev); +if (pm_is_runtime_suspended (host->dev)) +return 0 ; +if (host) {if (mmc_host_enable (host->mmc) !=0 ) {pm_runtime_suspend (host->dev); goto clk_en_err; --To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 38 lines · Kevin Hilman
Quoted 27 lines · Madhusudhan
Quoted 27 lines
-----Original Message-----
From: [email protected] [mailto:linux-omap- [email protected]] On Behalf Of Kevin Hilman
Sent: Tuesday, February 23, 2010 6:51 PM
To: [email protected]
Subject: Re: [PATCH/RFC 0/4] convert HS-MMC driver to hwmod + runtime PMQuoted 6 lines · Kevin Hilman
This series converts the OMAP HS-MMC driver to use
omap_hwmod+ runtime PM API.Depends on MMC hwmods available in 'pm-wip/hwmods' branch of my git tree[1] as well as previously posted runtime PM series:
Quoted 8 lines
[PATCH/RFC 0/2] initial runtime PM layer for OMAP
The easies way to experiment/test is to use my 'pm-wip/mmc' branch which has all the dependencies, and is based on omap/for-next'. It has been tested by merging with current PM branch.
[1] http://git.kernel.org/?p=linux/kernel/git/khilman/linux-omap-pm.git
A question for those of you who actually understand the MMC driver... I'm having problems getting my head around the current PM stuff in the MMC driver. My primary question is:
Quoted 8 lines
Quoted 3 lines
Why does the suspend hook need to re-enable the device before suspending?
In the scenario where there is no activity on the bus the MMC clocks are kept disabled. Now in the suspend path the MMC core will issue certain CMDs like CMD7(to end the suspend path) to deselect the card(more of a protocol stuff). Hence the host need to be in enabled state before letting the core know that there is a suspend request.
Quoted 5 lines
I guess what I'm wondering is whether we can (should) have the exact same path for runtime PM (idle) and static PM (suspend/resume.)
My current approach is that the suspend is a NOP if the device is already runtime suspended.
One difference is that the card-detect irq is disabled during suspend. That means that removeable cards must have their block devices removed too (or not if "unsafe-resume" is configured). The advantage of OFF-mode is that it will wake-up if someone opens the cover, whereas suspend will not.
A similar issue may develop with SDIO which may want to wake-up the device from OFF-mode (e.g. wireless card connected via SDIO)
So is the difference that OFF-mode has wakeups but suspend does not?
Quoted 115 lines
Quoted 19 lines
Quoted 10 lines
When using runtime PM, the MMC device is disabled including clocks off and regulator off (if
power_savings== true) when there is no activity.Then, in the static suspend hook, it's re-enabled (including taking it out of
off, re-enabling regulators etc) only to be quickly disabled again. This seems horribly inefficient.This is exactly for the reason I mentioned above.
Quoted 7 lines
I admit to not understanding the MMC layer terribly well, so can someone enlighten me as to what is going on here?
What I am testing here is a patch on top of this series (below) that adds a check to the static suspend hook. If the device is already runtime suspended, then the suspend and resume hooks should be noop.
Quoted 6 lines
Quoted 3 lines
This appears to work just fine while testing on omap3evm just doing simple read/write tests before an after suspend resume.
I did some basic testing with your previously posted patches. But my testing was incomplete because on Zoom2 because for some reason the OFF mode was not working even without your patches.
OK, I will try on Zoom2. I've only tested this so far on OMAP3EVM.
Quoted 2 lines
My concern was more with respect to OFF mode in idle path since your patches removed context restore calls if I recall correctly.
No, I didn't remove restore. I just moved it into the runtime PM resume callback.
Quoted 2 lines
Are you able to hit CORE OFF and then come back and do the read/write transfers in idle as well as suspend/resume path?
Yes.
Kevin
Quoted 65 lines
Quoted 65 lines
Note that if you want to test this patch, it also depends on this patch to runtime PM from the linux-pm list:
https://lists.linux-foundation.org/pipermail/linux-pm/2010-
February/024275.htmlThese are all included in an updated version of my pm-wip/mmc branch for ease of testing. Merge it with the current PM branch, enable
CONFIG_PM_RUNTIMEand test away.Kevin
commit
166cba7679fa267ee6e6eb39fd1e871ede5ded16Author: Kevin Hilman <[email protected]> Date: Tue Feb 23 16:21:56 2010 -0800MMC:
omap_hsmmc: check for runtime-suspend in static suspenddrivers/mmc/host/omap_hsmmc.c +6
-12@@ -2222,6 +2222,9 @@ static int omap_hsmmc_suspend(struct device *dev)if (host && host->suspended)return 0 ; +if (pm_is_runtime_suspended (host->dev)) +return 0 ; +if (host) { host->suspended =1 ;if (host->pdata->suspend) {@@ -2260,12 +2263,6 @@ static int omap_hsmmc_suspend(struct device *dev) }mmc_host_disable (host->mmc); } - -/* -* HACK: "extra" put to compensate for DPM core keeping -* runtime PM disabled. -- khilman -*/ -pm_runtime_put_sync (host->dev); }return ret; }@@ -2280,13 +2277,10 @@ static int omap_hsmmc_resume(struct device *dev)if (host && !host->suspended)return 0 ; -if (host) { -/* -* HACK: "extra" get to compensate for DPM core keeping -* runtime PM disabled. -- khilman -*/ -pm_runtime_get_sync (host->dev); +if (pm_is_runtime_suspended (host->dev)) +return 0 ; +if (host) {if (mmc_host_enable (host->mmc) !=0 ) {pm_runtime_suspend (host->dev); goto clk_en_err;Quoted 4 lines
Quoted 4 lines
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html