public inbox for linux-pm@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] PM: hibernate: clean up sync_read handling in snapshot_write_next
@ 2023-09-22 15:53 Brian Geffon
  2023-09-22 16:07 ` Brian Geffon
  0 siblings, 1 reply; 4+ messages in thread
From: Brian Geffon @ 2023-09-22 15:53 UTC (permalink / raw)
  To: Rafael J . Wysocki
  Cc: Pavel Machek, Len Brown, linux-pm, linux-kernel,
	Matthias Kaehlcke, Brian Geffon, stable

In snapshot_write_next sync_read is set and unset in three different
spots unnecessiarly. As a result there is a subtle bug where the first
page after the meta data has been loaded unconditionally sets sync_read
to 0. If this first pfn was actually a highmem page then the returned
buffer will be the global "buffer," and the page needs to be loaded
synchronously.

That is, I'm not sure we can always assume the following to be safe:
		handle->buffer = get_buffer(&orig_bm, &ca);
		handle->sync_read = 0;

Because get_buffer can call get_highmem_page_buffer which can
return 'buffer'

The easiest way to address this is just set sync_read before
snapshot_write_next returns if handle->buffer == buffer.

Signed-off-by: Brian Geffon <bgeffon@google.com>
Cc: stable@vger.kernel.org
---
 kernel/power/snapshot.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 190ed707ddcc..362e6bae5891 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -2780,8 +2780,6 @@ int snapshot_write_next(struct snapshot_handle *handle)
 	if (handle->cur > 1 && handle->cur > nr_meta_pages + nr_copy_pages + nr_zero_pages)
 		return 0;
 
-	handle->sync_read = 1;
-
 	if (!handle->cur) {
 		if (!buffer)
 			/* This makes the buffer be freed by swsusp_free() */
@@ -2824,7 +2822,6 @@ int snapshot_write_next(struct snapshot_handle *handle)
 			memory_bm_position_reset(&zero_bm);
 			restore_pblist = NULL;
 			handle->buffer = get_buffer(&orig_bm, &ca);
-			handle->sync_read = 0;
 			if (IS_ERR(handle->buffer))
 				return PTR_ERR(handle->buffer);
 		}
@@ -2834,9 +2831,8 @@ int snapshot_write_next(struct snapshot_handle *handle)
 		handle->buffer = get_buffer(&orig_bm, &ca);
 		if (IS_ERR(handle->buffer))
 			return PTR_ERR(handle->buffer);
-		if (handle->buffer != buffer)
-			handle->sync_read = 0;
 	}
+	handle->sync_read = (handle->buffer == buffer);
 	handle->cur++;
 
 	/* Zero pages were not included in the image, memset it and move on. */
-- 
2.42.0.515.g380fc7ccd1-goog


^ permalink raw reply related	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2023-09-26 20:03 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-09-22 15:53 [PATCH] PM: hibernate: clean up sync_read handling in snapshot_write_next Brian Geffon
2023-09-22 16:07 ` Brian Geffon
2023-09-26 18:24   ` Rafael J. Wysocki
2023-09-26 20:02     ` Brian Geffon

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox