Re: [PATCH] Fix resume on x86-32 machines

From: Jarkko Nikula <jarkko.nikula@linux.intel.com>
To: Pavel Machek <pavel@ucw.cz>, Andy Lutomirski <luto@amacapital.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
	Zhang Rui <rui.zhang@intel.com>,
	Andrew Lutomirski <luto@kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	"Rafael J. Wysocki" <rjw@rjwysocki.net>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	the arch/x86 maintainers <x86@kernel.org>
Subject: Re: [PATCH] Fix resume on x86-32 machines
Date: Mon, 11 Dec 2017 11:25:25 +0200	[thread overview]
Message-ID: <1a34404c-9634-34ff-a32b-81970aa79e63@linux.intel.com> (raw)
In-Reply-To: <20171210222049.GA6587@amd>

On 12/11/2017 12:20 AM, Pavel Machek wrote:
> On Sun 2017-12-10 13:58:23, Andy Lutomirski wrote:
>>> On Dec 10, 2017, at 1:38 PM, Pavel Machek <pavel@ucw.cz> wrote:
>>>
>>>
>>> After 4.15-rc2, suspend stopped working on Thinkpad X60. 5b06bbc
>>> (unintentionally?) reordered stuff with respect to
>>> fix_processor_context() on 32-bit and 64-bit. We undo that change on
>>> 32-bit.
>>>
>>
>> Can you explain what was wrong with the reordering?  Your patch certainly *looks* incorrect.
>>
> 
> No, I can't, sorry.
> 
>> I'm guessing that the real issue is that 32-bit needs %fs restored
>>   early for TLS.
> 
> Maybe. I can test patches...
> 
> I don't think it would be good idea to revert
> 5b06bbcfc2c621da3009da8decb7511500c293ed, but since it introduced
> regression in -rc2, I believe we should fix the regression now, and
> then we can try to provide cleaner solution.
> 
I can confirm Pavel's findings. Commit 5b06bbcfc2c6 ("x86/power: Fix 
some ordering bugs in __restore_processor_context()") broke the 
suspend/resume on 32-bit kernel.

v4.15-rc3 works either by reverting the commit or by Pavel's patch. 
Fortunately Pavel's patch still keeps the 64-bit suspend/resume ok.

-- 
Jarkko