This is a note to let you know that I've just added the patch titled
memblock: align freed memory map on pageblock boundaries with SPARSEMEM
to the 5.4-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
memblock-align-freed-memory-map-on-pageblock-boundaries-with-sparsemem.patch
and it can be found in the queue-5.4 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
>From foo@baz Wed Dec 15 03:02:39 PM CET 2021
From: Mark-PK Tsai <mark-pk.tsai@xxxxxxxxxxxx>
Date: Mon, 13 Dec 2021 16:57:07 +0800
Subject: memblock: align freed memory map on pageblock boundaries with SPARSEMEM
To: <stable@xxxxxxxxxxxxxxx>
Cc: <rppt@xxxxxxxxxx>, <akpm@xxxxxxxxxxxxxxxxxxxx>, <linux-arm-kernel@xxxxxxxxxxxxxxxxxxx>, <linux-kernel@xxxxxxxxxxxxxxx>, <linux-mm@xxxxxxxxx>, <linux@xxxxxxxxxxxxxxx>, <rppt@xxxxxxxxxxxxx>, <tony@xxxxxxxxxxx>, <wangkefeng.wang@xxxxxxxxxx>, <mark-pk.tsai@xxxxxxxxxxxx>, <yj.chiang@xxxxxxxxxxxx>
Message-ID: <20211213085710.28962-3-mark-pk.tsai@xxxxxxxxxxxx>
From: Mike Rapoport <rppt@xxxxxxxxxxxxx>
commit f921f53e089a12a192808ac4319f28727b35dc0f upstream.
When CONFIG_SPARSEMEM=y the ranges of the memory map that are freed are not
aligned to the pageblock boundaries which breaks assumptions about
homogeneity of the memory map throughout core mm code.
Make sure that the freed memory map is always aligned on pageblock
boundaries regardless of the memory model selection.
Signed-off-by: Mike Rapoport <rppt@xxxxxxxxxxxxx>
Tested-by: Tony Lindgren <tony@xxxxxxxxxxx>
Link: https://lore.kernel.org/lkml/20210630071211.21011-1-rppt@xxxxxxxxxx/
[backport upstream modification in mm/memblock.c to arch/arm/mm/init.c]
Signed-off-by: Mark-PK Tsai <mark-pk.tsai@xxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
arch/arm/mm/init.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -371,14 +371,14 @@ static void __init free_unused_memmap(vo
*/
start = min(start,
ALIGN(prev_end, PAGES_PER_SECTION));
-#else
+#endif
/*
* Align down here since many operations in VM subsystem
* presume that there are no holes in the memory map inside
* a pageblock
*/
start = round_down(start, pageblock_nr_pages);
-#endif
+
/*
* If we had a previous bank, and there is a space
* between the current bank and the previous, free it.
@@ -396,9 +396,11 @@ static void __init free_unused_memmap(vo
}
#ifdef CONFIG_SPARSEMEM
- if (!IS_ALIGNED(prev_end, PAGES_PER_SECTION))
+ if (!IS_ALIGNED(prev_end, PAGES_PER_SECTION)) {
+ prev_end = ALIGN(prev_end, pageblock_nr_pages);
free_memmap(prev_end,
ALIGN(prev_end, PAGES_PER_SECTION));
+ }
#endif
}
Patches currently in stable-queue which might be from mark-pk.tsai@xxxxxxxxxxxx are
queue-5.4/arm-extend-pfn_valid-to-take-into-account-freed-memory-map-alignment.patch
queue-5.4/arm-ioremap-don-t-abuse-pfn_valid-to-check-if-pfn-is-in-ram.patch
queue-5.4/memblock-free_unused_memmap-use-pageblock-units-instead-of-max_order.patch
queue-5.4/memblock-align-freed-memory-map-on-pageblock-boundaries-with-sparsemem.patch
queue-5.4/memblock-ensure-there-is-no-overflow-in-memblock_overlaps_region.patch
