linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: james.morse@arm.com (James Morse)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v4 1/3] PM / Hibernate: Allow arch code to influence CPUs disabled during hibernate
Date: Thu, 07 Jul 2016 09:29:36 +0100	[thread overview]
Message-ID: <577E12F0.9040403@arm.com> (raw)
In-Reply-To: <54369969.i5TPzUzNII@vostro.rjw.lan>

On 06/07/16 01:38, Rafael J. Wysocki wrote:
> On Monday, July 04, 2016 03:52:28 PM James Morse wrote:
>> Architecture code may need to do extra work when secondary CPUs are
>> disabled during hibernate and resume. This may include pushing sleeping
>> CPUs into a deeper power-saving state, or influencing which CPU resume
>> occurs on.
>>
>> Define a macro arch_hibernation_disable_cpus(), which defaults to
>> calling disable_nonboot_cpus() if undefined. Architectures that
>> need to do extra work around these calls can use this to influence
>> the CPU down calls.
>> The macros should be defined in asm/suspend.h, and
>> ARCH_HIBERNATION_CPU_HOOKS should be added to Kconfig.

>> diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
>> index fca9254280ee..855a3a2374c8 100644
>> --- a/kernel/power/hibernate.c
>> +++ b/kernel/power/hibernate.c
>> @@ -31,8 +31,16 @@
>>  #include <linux/ktime.h>
>>  #include <trace/events/power.h>
>>  
>> +#ifdef CONFIG_ARCH_HIBERNATION_CPU_HOOKS
>> +/* Arch definition of the arch_hibernation_disable_cpus() macros? */
>> +#include <asm/suspend.h>
>> +#endif
>> +
>>  #include "power.h"
>>  
>> +#ifndef arch_hibernation_disable_cpus
>> +#define arch_hibernation_disable_cpus(x) disable_nonboot_cpus()
>> +#endif
>>  
>>  static int nocompress;
>>  static int noresume;
>> @@ -279,7 +287,7 @@ static int create_image(int platform_mode)
>>  	if (error || hibernation_test(TEST_PLATFORM))
>>  		goto Platform_finish;
>>  
>> -	error = disable_nonboot_cpus();
>> +	error = arch_hibernation_disable_cpus(true);
>>  	if (error || hibernation_test(TEST_CPUS))
>>  		goto Enable_cpus;
>>  
>> @@ -433,7 +441,7 @@ static int resume_target_kernel(bool platform_mode)
>>  	if (error)
>>  		goto Cleanup;
>>  
>> -	error = disable_nonboot_cpus();
>> +	error = arch_hibernation_disable_cpus(false);
> 
> Why "false"?

To indicate whether this is suspend or resume. On suspend we just call
disable_nonboot_cpus(), this ensures frozen_cpus and the potential races with
userspace are covered properly. At this point we don't care which CPU it picks.

On resume we know which CPU we want, so cpu_down() all the others. I thought the
frozen_cpus and user-space race wouldn't be a problem here, but Lorenzo
suggested it may confuse some device drivers to receive a CPU_DOWN_PREPARE etc
followed by CPU_UP_PREPARE_FROZEN etc.

I haven't found any drivers in the tree where this would be a problem (~95% of
notifiers either mask out the frozen bits, or fall-through in those cases). But
I'm still going through the list...


> 
>>  	if (error)
>>  		goto Enable_cpus;
>>  
>> @@ -551,7 +559,7 @@ int hibernation_platform_enter(void)
>>  	if (error)
>>  		goto Platform_finish;
>>  
>> -	error = disable_nonboot_cpus();
>> +	error = arch_hibernation_disable_cpus(true);
> 
> I have the same question about this hunk I had before.
> 
> Is it really necessary to do the arch thing here?

Ah, sorry I didn't understand what this did before. This is used when ACPI
drives hibernate/resume instead of swsusp_arch_suspend().

No, its not needed.


> It shouldn't really matter AFAICS.
> 
>>  	if (error)
>>  		goto Enable_cpus;

Thanks,

James

  reply	other threads:[~2016-07-07  8:29 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-07-04 14:52 [PATCH v4 0/3] PM / Hibernate: Allow arch code to influence CPUs disabled during hibernate James Morse
2016-07-04 14:52 ` [PATCH v4 1/3] " James Morse
2016-07-05 12:28   ` Rafael J. Wysocki
2016-07-06  9:16     ` James Morse
2016-07-06 21:11       ` Rafael J. Wysocki
2016-07-06  0:38   ` Rafael J. Wysocki
2016-07-07  8:29     ` James Morse [this message]
2016-07-04 14:52 ` [PATCH v4 2/3] arm64: hibernate: Resume when hibernate image created on non-boot CPU James Morse
2016-07-05 17:49   ` Catalin Marinas
2016-08-17 10:03     ` James Morse
2016-07-04 14:52 ` [PATCH v4 3/3] Revert "arm64: hibernate: Refuse to hibernate if the boot cpu is offline" James Morse
2016-07-05 17:49   ` Catalin Marinas

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=577E12F0.9040403@arm.com \
    --to=james.morse@arm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    /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;
as well as URLs for NNTP newsgroup(s).