Re: [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Vlastimil Babka <vbabka@suse.cz>]

On 5/31/23 14:55, Christoph Hellwig wrote:
> software_resume can be called either from an init call in the boot code,
> or from sysfs once the system has finished booting, and the two
> invocation methods this can't race with each other.
> 
> For the latter case we did just parse the suspend device manually, while
> the former might not have one.  Split software_resume so that the search
> only happens for the boot case, which also means the special lockdep
> nesting annotation can go away as the system transition mutex can be
> taken a little later and doesn't have the sysfs locking nest inside it.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> Acked-by: Rafael J. Wysocki <rafael@kernel.org>

This caused a regression for me in 6.5-rc1+, fix below.

----8<----
From 95a310ae6cfae9b3cab61e54a1bce488c3ab93a1 Mon Sep 17 00:00:00 2001
From: Vlastimil Babka <vbabka@suse.cz>
Date: Wed, 2 Aug 2023 15:46:18 +0200
Subject: [PATCH] PM: hibernate: fix resume_store() return value when
 hibernation not available

On a laptop with hibernation set up but not actively used, and with
secure boot and lockdown enabled kernel, 6.5-rc1 gets stuck on boot with
the following repeated messages:

  A start job is running for Resume from hibernation using device /dev/system/swap (24s / no limit)
  lockdown_is_locked_down: 25311154 callbacks suppressed
  Lockdown: systemd-hiberna: hibernation is restricted; see man kernel_lockdown.7
  ...

Checking the resume code leads to commit cc89c63e2fe3 ("PM: hibernate:
move finding the resume device out of software_resume") which
inadvertently changed the return value from resume_store() to 0 when
!hibernation_available(). This apparently translates to userspace
write() returning 0 as in number of bytes written, and userspace looping
indefinitely in the attempt to write the intended value.

Fix this by returning the full number of bytes that were to be written,
as that's what was done before the commit.

Fixes: cc89c63e2fe3 ("PM: hibernate: move finding the resume device out of software_resume")
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
---
 kernel/power/hibernate.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index e1b4bfa938dd..2b4a946a6ff5 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -1166,7 +1166,7 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	int error;
 
 	if (!hibernation_available())
-		return 0;
+		return n;
 
 	if (len && buf[len-1] == '\n')
 		len--;
-- 
2.41.0