The quilt patch titled
Subject: mm: compaction: use the proper flag to determine watermarks
has been removed from the -mm tree. Its filename was
mm-compaction-use-the-proper-flag-to-determine-watermarks.patch
This patch was dropped because it was merged into the mm-hotfixes-stable branch
of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
------------------------------------------------------
From: yangge <yangge1116@xxxxxxx>
Subject: mm: compaction: use the proper flag to determine watermarks
Date: Sat, 25 Jan 2025 14:53:57 +0800
There are 4 NUMA nodes on my machine, and each NUMA node has 32GB of
memory. I have configured 16GB of CMA memory on each NUMA node, and
starting a 32GB virtual machine with device passthrough is extremely slow,
taking almost an hour.
Long term GUP cannot allocate memory from CMA area, so a maximum of 16 GB
of no-CMA memory on a NUMA node can be used as virtual machine memory.
There is 16GB of free CMA memory on a NUMA node, which is sufficient to
pass the order-0 watermark check, causing the __compaction_suitable()
function to consistently return true.
For costly allocations, if the __compaction_suitable() function always
returns true, it causes the __alloc_pages_slowpath() function to fail to
exit at the appropriate point. This prevents timely fallback to
allocating memory on other nodes, ultimately resulting in excessively long
virtual machine startup times.
Call trace:
__alloc_pages_slowpath
if (compact_result == COMPACT_SKIPPED ||
compact_result == COMPACT_DEFERRED)
goto nopage; // should exit __alloc_pages_slowpath() from here
We could use the real unmovable allocation context to have
__zone_watermark_unusable_free() subtract CMA pages, and thus we won't
pass the order-0 check anymore once the non-CMA part is exhausted. There
is some risk that in some different scenario the compaction could in fact
migrate pages from the exhausted non-CMA part of the zone to the CMA part
and succeed, and we'll skip it instead. But only __GFP_NORETRY
allocations should be affected in the immediate "goto nopage" when
compaction is skipped, others will attempt with DEF_COMPACT_PRIORITY
anyway and won't fail without trying to compact-migrate the non-CMA
pageblocks into CMA pageblocks first, so it should be fine.
After this fix, it only takes a few tens of seconds to start a 32GB
virtual machine with device passthrough functionality.
Link: https://lore.kernel.org/lkml/1736335854-548-1-git-send-email-yangge1116@xxxxxxx/
Link: https://lkml.kernel.org/r/1737788037-8439-1-git-send-email-yangge1116@xxxxxxx
Signed-off-by: yangge <yangge1116@xxxxxxx>
Acked-by: Vlastimil Babka <vbabka@xxxxxxx>
Reviewed-by: Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx>
Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Barry Song <21cnbao@xxxxxxxxx>
Cc: David Hildenbrand <david@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/compaction.c | 29 +++++++++++++++++++++++++----
1 file changed, 25 insertions(+), 4 deletions(-)
--- a/mm/compaction.c~mm-compaction-use-the-proper-flag-to-determine-watermarks
+++ a/mm/compaction.c
@@ -2491,7 +2491,8 @@ bool compaction_zonelist_suitable(struct
*/
static enum compact_result
compaction_suit_allocation_order(struct zone *zone, unsigned int order,
- int highest_zoneidx, unsigned int alloc_flags)
+ int highest_zoneidx, unsigned int alloc_flags,
+ bool async)
{
unsigned long watermark;
@@ -2500,6 +2501,23 @@ compaction_suit_allocation_order(struct
alloc_flags))
return COMPACT_SUCCESS;
+ /*
+ * For unmovable allocations (without ALLOC_CMA), check if there is enough
+ * free memory in the non-CMA pageblocks. Otherwise compaction could form
+ * the high-order page in CMA pageblocks, which would not help the
+ * allocation to succeed. However, limit the check to costly order async
+ * compaction (such as opportunistic THP attempts) because there is the
+ * possibility that compaction would migrate pages from non-CMA to CMA
+ * pageblock.
+ */
+ if (order > PAGE_ALLOC_COSTLY_ORDER && async &&
+ !(alloc_flags & ALLOC_CMA)) {
+ watermark = low_wmark_pages(zone) + compact_gap(order);
+ if (!__zone_watermark_ok(zone, 0, watermark, highest_zoneidx,
+ 0, zone_page_state(zone, NR_FREE_PAGES)))
+ return COMPACT_SKIPPED;
+ }
+
if (!compaction_suitable(zone, order, highest_zoneidx))
return COMPACT_SKIPPED;
@@ -2535,7 +2553,8 @@ compact_zone(struct compact_control *cc,
if (!is_via_compact_memory(cc->order)) {
ret = compaction_suit_allocation_order(cc->zone, cc->order,
cc->highest_zoneidx,
- cc->alloc_flags);
+ cc->alloc_flags,
+ cc->mode == MIGRATE_ASYNC);
if (ret != COMPACT_CONTINUE)
return ret;
}
@@ -3038,7 +3057,8 @@ static bool kcompactd_node_suitable(pg_d
ret = compaction_suit_allocation_order(zone,
pgdat->kcompactd_max_order,
- highest_zoneidx, ALLOC_WMARK_MIN);
+ highest_zoneidx, ALLOC_WMARK_MIN,
+ false);
if (ret == COMPACT_CONTINUE)
return true;
}
@@ -3079,7 +3099,8 @@ static void kcompactd_do_work(pg_data_t
continue;
ret = compaction_suit_allocation_order(zone,
- cc.order, zoneid, ALLOC_WMARK_MIN);
+ cc.order, zoneid, ALLOC_WMARK_MIN,
+ false);
if (ret != COMPACT_CONTINUE)
continue;
_
Patches currently in -mm which might be from yangge1116@xxxxxxx are
