public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: David Brownell <david-b@pacbell.net>
To: Pavel Machek <pavel@ucw.cz>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Linux Kernel list <linux-kernel@vger.kernel.org>,
	Patrick Mochel <mochel@digitalimplant.org>
Subject: Re: Solving suspend-level confusion
Date: Sat, 7 Aug 2004 16:23:16 -0700	[thread overview]
Message-ID: <200408071623.16829.david-b@pacbell.net> (raw)
In-Reply-To: <20040806211033.GE30518@elf.ucw.cz>

[-- Attachment #1: Type: text/plain, Size: 792 bytes --]


> I believe right solution is to pass system state into suspend(), and
> provide system_to_pci_state(), so that drivers that only care about
> PCI state can use.

So long as tools (gcc by preference, or sparse) warn about API changes,
I can be happy.  See the attached "pmcore-0807.patch", a subset of
yours, against a recent rc3.

I'd rather address the PCI issue with Ben's renumbering:

enum system_state {
        PM_SUSPEND_ON = 0,
        PM_SUSPEND_STANDBY = 1,
        PM_SUSPEND_MEM = 3,
        PM_SUSPEND_DISK = 4,
        PM_SUSPEND_MAX,
};

That way sysfs won't get all wierd with PCI, and by the way
no drivers should need to change ... :)


> static void ide_device_shutdown(struct device *dev)

Ah, that needs attention too.  Should USB do anything special
there?

- Dave




[-- Attachment #2: pmcore-0807.patch --]
[-- Type: text/x-diff, Size: 4172 bytes --]

Cleanup of some of the "new 2.6" suspend call path, using
"enum system_state" rather than "u32".  This mostly just
documents existing practice for non-PCI drivers.


--- 1.16/include/linux/pm.h	Thu Jul  1 22:23:53 2004
+++ edited/include/linux/pm.h	Sat Aug  7 14:32:44 2004
@@ -193,11 +193,11 @@
 extern void (*pm_idle)(void);
 extern void (*pm_power_off)(void);
 
-enum {
-	PM_SUSPEND_ON,
-	PM_SUSPEND_STANDBY,
-	PM_SUSPEND_MEM,
-	PM_SUSPEND_DISK,
+enum system_state {
+	PM_SUSPEND_ON = 0,
+	PM_SUSPEND_STANDBY = 1,
+	PM_SUSPEND_MEM = 2,
+	PM_SUSPEND_DISK = 3,
 	PM_SUSPEND_MAX,
 };
 
@@ -241,11 +241,13 @@
 
 extern void device_pm_set_parent(struct device * dev, struct device * parent);
 
-extern int device_suspend(u32 state);
-extern int device_power_down(u32 state);
+/*
+ * apply system suspend policy to all devices
+ */
+extern int device_suspend(enum system_state reason);
+extern int device_power_down(enum system_state reason);
 extern void device_power_up(void);
 extern void device_resume(void);
-
 
 #endif /* __KERNEL__ */
 
--- 1.86/kernel/power/swsusp.c	Thu Jul  1 22:23:48 2004
+++ edited/kernel/power/swsusp.c	Sat Aug  7 14:32:44 2004
@@ -699,7 +699,7 @@
 	else
 #endif
 	{
-		device_suspend(3);
+		device_suspend(PM_SUSPEND_DISK);
 		device_shutdown();
 		machine_power_off();
 	}
@@ -720,7 +720,7 @@
 	mb();
 	spin_lock_irq(&suspend_pagedir_lock);	/* Done to disable interrupts */ 
 
-	device_power_down(3);
+	device_power_down(PM_SUSPEND_DISK);
 	PRINTK( "Waiting for DMAs to settle down...\n");
 	mdelay(1000);	/* We do not want some readahead with DMA to corrupt our memory, right?
 			   Do it with disabled interrupts for best effect. That way, if some
@@ -789,7 +789,7 @@
 {
 	int is_problem;
 	read_swapfiles();
-	device_power_down(3);
+	device_power_down(PM_SUSPEND_DISK);
 	is_problem = suspend_prepare_image();
 	device_power_up();
 	spin_unlock_irq(&suspend_pagedir_lock);
@@ -845,7 +845,7 @@
 		disable_nonboot_cpus();
 		/* Save state of all device drivers, and stop them. */
 		printk("Suspending devices... ");
-		if ((res = device_suspend(3))==0) {
+		if ((res = device_suspend(PM_SUSPEND_DISK))==0) {
 			/* If stopping device drivers worked, we proceed basically into
 			 * suspend_save_image.
 			 *
@@ -1200,7 +1200,7 @@
 		goto read_failure;
 	/* FIXME: Should we stop processes here, just to be safer? */
 	disable_nonboot_cpus();
-	device_suspend(3);
+	device_suspend(PM_SUSPEND_DISK);
 	do_magic(1);
 	panic("This never returns");
 
--- 1.7/drivers/base/power/power.h	Wed Jun  9 23:34:24 2004
+++ edited/drivers/base/power/power.h	Sat Aug  7 14:32:44 2004
@@ -66,14 +66,14 @@
 /*
  * suspend.c
  */
-extern int suspend_device(struct device *, u32);
+extern int suspend_device(struct device *, enum system_state);
 
 
 /*
  * runtime.c
  */
 
-extern int dpm_runtime_suspend(struct device *, u32);
+extern int dpm_runtime_suspend(struct device *, enum system_state);
 extern void dpm_runtime_resume(struct device *);
 
 #else /* CONFIG_PM */
@@ -88,7 +88,7 @@
 
 }
 
-static inline int dpm_runtime_suspend(struct device * dev, u32 state)
+static inline int dpm_runtime_suspend(struct device * dev, enum system_state state)
 {
 	return 0;
 }
--- 1.32/drivers/base/power/shutdown.c	Wed Jun  9 23:34:24 2004
+++ edited/drivers/base/power/shutdown.c	Sat Aug  7 14:32:44 2004
@@ -29,7 +29,7 @@
 			dev->driver->shutdown(dev);
 		return 0;
 	}
-	return dpm_runtime_suspend(dev, dev->detach_state);
+	return dpm_runtime_suspend(dev, (enum system_state) dev->detach_state);
 }
 
 
--- 1.16/drivers/base/power/suspend.c	Wed Jun  9 23:34:24 2004
+++ edited/drivers/base/power/suspend.c	Sat Aug  7 14:32:44 2004
@@ -35,7 +35,7 @@
  *	@state:	Power state device is entering.
  */
 
-int suspend_device(struct device * dev, u32 state)
+int suspend_device(struct device * dev, enum system_state state)
 {
 	int error = 0;
 
@@ -70,7 +70,7 @@
  *
  */
 
-int device_suspend(u32 state)
+int device_suspend(enum system_state state)
 {
 	int error = 0;
 
@@ -112,7 +112,7 @@
  *	done, power down system devices.
  */
 
-int device_power_down(u32 state)
+int device_power_down(enum system_state state)
 {
 	int error = 0;
 	struct device * dev;

  reply	other threads:[~2004-08-07 23:51 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-07-30 16:44 Solving suspend-level confusion Pavel Machek
2004-07-30 22:39 ` Benjamin Herrenschmidt
2004-07-30 23:06   ` Pavel Machek
2004-07-31  4:02 ` David Brownell
2004-07-31  4:36   ` Pavel Machek
2004-07-31  5:49   ` Benjamin Herrenschmidt
2004-07-31 14:23     ` David Brownell
2004-07-31 17:01       ` Oliver Neukum
2004-07-31 17:51         ` David Brownell
2004-08-01  0:41         ` David Brownell
2004-08-01  1:34           ` Benjamin Herrenschmidt
2004-08-02 16:38             ` David Brownell
2004-08-03  0:38               ` Benjamin Herrenschmidt
2004-08-04  2:28                 ` David Brownell
2004-08-04  2:26                   ` Nigel Cunningham
2004-08-04  2:53                     ` Benjamin Herrenschmidt
2004-08-04  2:52                       ` Nigel Cunningham
2004-08-04  4:14                         ` Benjamin Herrenschmidt
2004-08-04  4:25                           ` Nigel Cunningham
2004-08-04  4:52                             ` Benjamin Herrenschmidt
2004-08-04  4:54                               ` Nigel Cunningham
2004-08-04  5:03                                 ` Benjamin Herrenschmidt
2004-08-04  5:05                                   ` Nigel Cunningham
2004-08-05 10:05                                   ` Nigel Cunningham
2004-08-05 22:31                                   ` Nigel Cunningham
2004-08-06  0:39                                     ` Benjamin Herrenschmidt
2004-08-06 21:30                                     ` Pavel Machek
2004-08-06 21:29                           ` Pavel Machek
2004-08-06 22:27                             ` Benjamin Herrenschmidt
2004-08-06 22:37                               ` Pavel Machek
2004-08-06 21:26                         ` Pavel Machek
2004-08-05  1:29                     ` David Brownell
2004-08-05 10:19                       ` Nigel Cunningham
2004-08-06  0:32                         ` David Brownell
     [not found]                           ` <1091772799.2532.50.camel@laptop.cunninghams>
2004-08-07 22:24                             ` David Brownell
2004-08-04  2:56                   ` Benjamin Herrenschmidt
2004-08-04  3:30                     ` David Brownell
2004-08-04  4:19                       ` Benjamin Herrenschmidt
2004-08-04  4:47                       ` What PM should be and do (Was Re: Solving suspend-level confusion) Nigel Cunningham
2004-08-04  4:53                         ` Benjamin Herrenschmidt
2004-08-04  4:59                           ` Nigel Cunningham
2004-08-08 16:54                             ` Pavel Machek
2004-08-08 21:55                               ` Nigel Cunningham
2004-08-09  8:42                                 ` Pavel Machek
2004-08-05 18:19                         ` Greg KH
2004-08-05 22:14                           ` Nigel Cunningham
2004-08-07  0:08                             `  Éric Brunet
2004-08-08 19:48                               ` Pavel Machek
2004-08-11 21:23                             ` Greg KH
2004-08-08  0:54                         ` David Brownell
2004-08-06 21:21               ` Solving suspend-level confusion Pavel Machek
2004-07-31 21:09       ` Benjamin Herrenschmidt
2004-08-02 16:40         ` David Brownell
2004-08-03  0:50           ` Benjamin Herrenschmidt
2004-08-07 23:30             ` David Brownell
2004-08-06 21:10           ` Pavel Machek
2004-08-07 23:23             ` David Brownell [this message]
2004-08-08 17:16               ` Pavel Machek
2004-08-06 20:04       ` Pavel Machek
2004-08-07 22:14         ` David Brownell
2004-08-07 23:53           ` Benjamin Herrenschmidt

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=200408071623.16829.david-b@pacbell.net \
    --to=david-b@pacbell.net \
    --cc=benh@kernel.crashing.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mochel@digitalimplant.org \
    --cc=pavel@ucw.cz \
    /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