From: "Rafael J. Wysocki" <rjw@sisk.pl>
To: Greg KH <gregkh@suse.de>, akpm@linux-foundation.org
Cc: david-b@pacbell.net, pavel@ucw.cz, stern@rowland.harvard.edu,
rmk@arm.linux.org.uk, LKML <linux-kernel@vger.kernel.org>
Subject: Re: [RFC][PATCH -mm] PM: Do not use saved_state from struct dev_pm_info on ARM (was: Re: patch pm-remove-saved_state-from-struct-dev_pm_info.patch added to gregkh-2.6 tree)
Date: Thu, 28 Jun 2007 17:48:40 +0200 [thread overview]
Message-ID: <200706281748.41473.rjw@sisk.pl> (raw)
In-Reply-To: <200706272311.04256.rjw@sisk.pl>
On Wednesday, 27 June 2007 23:11, Rafael J. Wysocki wrote:
> On Tuesday, 26 June 2007 23:03, Greg KH wrote:
> > On Tue, Jun 26, 2007 at 11:00:06PM +0200, Rafael J. Wysocki wrote:
> > > On Friday, 15 June 2007 00:57, gregkh@suse.de wrote:
> > > >
> > > > This is a note to let you know that I've just added the patch titled
> > > >
> > > > Subject: PM: Remove saved_state from struct dev_pm_info
> > > >
> > > > to my gregkh-2.6 tree. Its filename is
> > > >
> > > > pm-remove-saved_state-from-struct-dev_pm_info.patch
> > > >
> > > > This tree can be found at
> > > > http://www.kernel.org/pub/linux/kernel/people/gregkh/gregkh-2.6/patches/
> > > >
> > > >
> > > > From rjw@sisk.pl Thu Jun 14 15:47:15 2007
> > > > From: "Rafael J. Wysocki" <rjw@sisk.pl>
> > > > To: Andrew Morton <akpm@linux-foundation.org>
> > > > Subject: PM: Remove saved_state from struct dev_pm_info
> > > > Date: Wed, 13 Jun 2007 15:55:34 +0200
> > > > Cc: Alan Stern <stern@rowland.harvard.edu>, David Brownell <david-b@pacbell.net>, Greg KH <greg@kroah.com>, Pavel Machek <pavel@ucw.cz>
> > > > Content-Disposition: inline
> > > > Message-Id: <200706131555.37277.rjw@sisk.pl>
> > > >
> > > > From: Rafael J. Wysocki <rjw@sisk.pl>
> > > >
> > > > The saved_state member of struct dev_pm_info, defined in include/linux/pm.h, is
> > > > not used anywhere, so it can be removed.
> > >
> > > Unfortunately this is wrong, as I've just found three drivers in the ARM tree
> > > that use it. I don't know why I didn't see them when I was creating the patch,
> > > though. :-(
> >
> > Why not just fix up the ARM drivers instead?
>
> OK, let's try. :-)
>
> The following patch has only been compilation tested.
>
> ---
> From: Rafael J. Wysocki <rjw@sisk.pl>
>
> The saved_state member of 'struct dev_pm_info' that's going to be removed is
> used in arch/arm/common/locomo.c, arch/arm/common/sa1111.c and
> arch/arm/mach-sa1100/neponset.c. Change the code in there to use local
> variables for saving the state of devices during suspend.
>
> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Russell says that the affected drivers were coded to support multiple instances
of the same device which the $subject patch doesn't take into account (still,
as far as the neponset is concerned, I think that there can be only one
instance of NCR_0 in the system).
Updated (compilation tested) patch follows.
Greetings,
Rafael
---
From: Rafael J. Wysocki <rjw@sisk.pl>
The saved_state member of 'struct dev_pm_info' that's going to be removed is
used in arch/arm/common/locomo.c, arch/arm/common/sa1111.c and
arch/arm/mach-sa1100/neponset.c. Change the code in there to use the drivers'
own per-device data structures or static variables for saving the state of
devices during suspend.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
arch/arm/common/locomo.c | 9 +++++++--
arch/arm/common/sa1111.c | 13 ++++++++-----
arch/arm/mach-sa1100/neponset.c | 15 ++++-----------
3 files changed, 19 insertions(+), 18 deletions(-)
Index: linux-2.6.22-rc6/arch/arm/common/locomo.c
===================================================================
--- linux-2.6.22-rc6.orig/arch/arm/common/locomo.c
+++ linux-2.6.22-rc6/arch/arm/common/locomo.c
@@ -60,6 +60,9 @@ struct locomo {
unsigned int irq;
spinlock_t lock;
void __iomem *base;
+#ifdef CONFIG_PM
+ void *saved_state;
+#endif
};
struct locomo_dev_info {
@@ -565,7 +568,7 @@ static int locomo_suspend(struct platfor
if (!save)
return -ENOMEM;
- dev->dev.power.saved_state = (void *) save;
+ lchip->saved_state = save;
spin_lock_irqsave(&lchip->lock, flags);
@@ -606,7 +609,7 @@ static int locomo_resume(struct platform
unsigned long r;
unsigned long flags;
- save = (struct locomo_save_data *) dev->dev.power.saved_state;
+ save = lchip->saved_state;
if (!save)
return 0;
@@ -628,6 +631,8 @@ static int locomo_resume(struct platform
locomo_writel(0x1, lchip->base + LOCOMO_KEYBOARD + LOCOMO_KCMD);
spin_unlock_irqrestore(&lchip->lock, flags);
+
+ lchip->saved_state = NULL;
kfree(save);
return 0;
Index: linux-2.6.22-rc6/arch/arm/common/sa1111.c
===================================================================
--- linux-2.6.22-rc6.orig/arch/arm/common/sa1111.c
+++ linux-2.6.22-rc6/arch/arm/common/sa1111.c
@@ -51,6 +51,9 @@ struct sa1111 {
int irq;
spinlock_t lock;
void __iomem *base;
+#ifdef CONFIG_PM
+ void *saved_state;
+#endif
};
/*
@@ -822,7 +825,7 @@ static int sa1111_suspend(struct platfor
save = kmalloc(sizeof(struct sa1111_save_data), GFP_KERNEL);
if (!save)
return -ENOMEM;
- dev->dev.power.saved_state = save;
+ sachip->saved_state = save;
spin_lock_irqsave(&sachip->lock, flags);
@@ -878,7 +881,7 @@ static int sa1111_resume(struct platform
unsigned long flags, id;
void __iomem *base;
- save = (struct sa1111_save_data *)dev->dev.power.saved_state;
+ save = sachip->saved_state;
if (!save)
return 0;
@@ -923,7 +926,7 @@ static int sa1111_resume(struct platform
spin_unlock_irqrestore(&sachip->lock, flags);
- dev->dev.power.saved_state = NULL;
+ sachip->saved_state = NULL;
kfree(save);
return 0;
@@ -958,8 +961,8 @@ static int sa1111_remove(struct platform
platform_set_drvdata(pdev, NULL);
#ifdef CONFIG_PM
- kfree(pdev->dev.power.saved_state);
- pdev->dev.power.saved_state = NULL;
+ kfree(sachip->saved_state);
+ sachip->saved_state = NULL;
#endif
}
Index: linux-2.6.22-rc6/arch/arm/mach-sa1100/neponset.c
===================================================================
--- linux-2.6.22-rc6.orig/arch/arm/mach-sa1100/neponset.c
+++ linux-2.6.22-rc6/arch/arm/mach-sa1100/neponset.c
@@ -185,28 +185,21 @@ static int __devinit neponset_probe(stru
/*
* LDM power management.
*/
+static unsigned int neponset_saved_state;
+
static int neponset_suspend(struct platform_device *dev, pm_message_t state)
{
/*
* Save state.
*/
- if (!dev->dev.power.saved_state)
- dev->dev.power.saved_state = kmalloc(sizeof(unsigned int), GFP_KERNEL);
- if (!dev->dev.power.saved_state)
- return -ENOMEM;
-
- *(unsigned int *)dev->dev.power.saved_state = NCR_0;
+ neponset_saved_state = NCR_0;
return 0;
}
static int neponset_resume(struct platform_device *dev)
{
- if (dev->dev.power.saved_state) {
- NCR_0 = *(unsigned int *)dev->dev.power.saved_state;
- kfree(dev->dev.power.saved_state);
- dev->dev.power.saved_state = NULL;
- }
+ NCR_0 = neponset_saved_state;
return 0;
}
prev parent reply other threads:[~2007-06-28 15:41 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1181861863@kroah.com>
[not found] ` <200706262300.08543.rjw@sisk.pl>
[not found] ` <20070626210311.GC7814@suse.de>
2007-06-27 21:11 ` Rafael J. Wysocki
2007-06-28 15:48 ` Rafael J. Wysocki [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=200706281748.41473.rjw@sisk.pl \
--to=rjw@sisk.pl \
--cc=akpm@linux-foundation.org \
--cc=david-b@pacbell.net \
--cc=gregkh@suse.de \
--cc=linux-kernel@vger.kernel.org \
--cc=pavel@ucw.cz \
--cc=rmk@arm.linux.org.uk \
--cc=stern@rowland.harvard.edu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®