The patch titled
Subject: mm: numa: slow PTE scan rate if migration failures occur
has been added to the -mm tree. Its filename is
mm-numa-slow-pte-scan-rate-if-migration-failures-occur.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mm-numa-slow-pte-scan-rate-if-migration-failures-occur.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mm-numa-slow-pte-scan-rate-if-migration-failures-occur.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Mel Gorman <mgorman@xxxxxxx>
Subject: mm: numa: slow PTE scan rate if migration failures occur
Dave Chinner reported the following on https://lkml.org/lkml/2015/3/1/226
Across the board the 4.0-rc1 numbers are much slower, and the degradation
is far worse when using the large memory footprint configs. Perf points
straight at the cause - this is from 4.0-rc1 on the "-o bhash=101073" config:
- 56.07% 56.07% [kernel] [k] default_send_IPI_mask_sequence_phys
- default_send_IPI_mask_sequence_phys
- 99.99% physflat_send_IPI_mask
- 99.37% native_send_call_func_ipi
smp_call_function_many
- native_flush_tlb_others
- 99.85% flush_tlb_page
ptep_clear_flush
try_to_unmap_one
rmap_walk
try_to_unmap
migrate_pages
migrate_misplaced_page
- handle_mm_fault
- 99.73% __do_page_fault
trace_do_page_fault
do_async_page_fault
+ async_page_fault
0.63% native_send_call_func_single_ipi
generic_exec_single
smp_call_function_single
This is showing excessive migration activity even though excessive
migrations are meant to get throttled. Normally, the scan rate is tuned
on a per-task basis depending on the locality of faults. However, if
migrations fail for any reason then the PTE scanner may scan faster if the
faults continue to be remote. This means there is higher system CPU
overhead and fault trapping at exactly the time we know that migrations
cannot happen. This patch tracks when migration failures occur and slows
the PTE scanner.
Signed-off-by: Mel Gorman <mgorman@xxxxxxx>
Reported-by: Dave Chinner <david@xxxxxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Aneesh Kumar <aneesh.kumar@xxxxxxxxxxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/sched.h | 9 +++++----
kernel/sched/fair.c | 8 ++++++--
mm/huge_memory.c | 3 ++-
mm/memory.c | 3 ++-
4 files changed, 15 insertions(+), 8 deletions(-)
diff -puN include/linux/sched.h~mm-numa-slow-pte-scan-rate-if-migration-failures-occur include/linux/sched.h
--- a/include/linux/sched.h~mm-numa-slow-pte-scan-rate-if-migration-failures-occur
+++ a/include/linux/sched.h
@@ -1625,11 +1625,11 @@ struct task_struct {
/*
* numa_faults_locality tracks if faults recorded during the last
- * scan window were remote/local. The task scan period is adapted
- * based on the locality of the faults with different weights
- * depending on whether they were shared or private faults
+ * scan window were remote/local or failed to migrate. The task scan
+ * period is adapted based on the locality of the faults with different
+ * weights depending on whether they were shared or private faults
*/
- unsigned long numa_faults_locality[2];
+ unsigned long numa_faults_locality[3];
unsigned long numa_pages_migrated;
#endif /* CONFIG_NUMA_BALANCING */
@@ -1719,6 +1719,7 @@ struct task_struct {
#define TNF_NO_GROUP 0x02
#define TNF_SHARED 0x04
#define TNF_FAULT_LOCAL 0x08
+#define TNF_MIGRATE_FAIL 0x10
#ifdef CONFIG_NUMA_BALANCING
extern void task_numa_fault(int last_node, int node, int pages, int flags);
diff -puN kernel/sched/fair.c~mm-numa-slow-pte-scan-rate-if-migration-failures-occur kernel/sched/fair.c
--- a/kernel/sched/fair.c~mm-numa-slow-pte-scan-rate-if-migration-failures-occur
+++ a/kernel/sched/fair.c
@@ -1609,9 +1609,11 @@ static void update_task_scan_period(stru
/*
* If there were no record hinting faults then either the task is
* completely idle or all activity is areas that are not of interest
- * to automatic numa balancing. Scan slower
+ * to automatic numa balancing. Related to that, if there were failed
+ * migration then it implies we are migrating too quickly or the local
+ * node is overloaded. In either case, scan slower
*/
- if (local + shared == 0) {
+ if (local + shared == 0 || p->numa_faults_locality[2]) {
p->numa_scan_period = min(p->numa_scan_period_max,
p->numa_scan_period << 1);
@@ -2080,6 +2082,8 @@ void task_numa_fault(int last_cpupid, in
if (migrated)
p->numa_pages_migrated += pages;
+ if (flags & TNF_MIGRATE_FAIL)
+ p->numa_faults_locality[2] += pages;
p->numa_faults[task_faults_idx(NUMA_MEMBUF, mem_node, priv)] += pages;
p->numa_faults[task_faults_idx(NUMA_CPUBUF, cpu_node, priv)] += pages;
diff -puN mm/huge_memory.c~mm-numa-slow-pte-scan-rate-if-migration-failures-occur mm/huge_memory.c
--- a/mm/huge_memory.c~mm-numa-slow-pte-scan-rate-if-migration-failures-occur
+++ a/mm/huge_memory.c
@@ -1350,7 +1350,8 @@ int do_huge_pmd_numa_page(struct mm_stru
if (migrated) {
flags |= TNF_MIGRATED;
page_nid = target_nid;
- }
+ } else
+ flags |= TNF_MIGRATE_FAIL;
goto out;
clear_pmdnuma:
diff -puN mm/memory.c~mm-numa-slow-pte-scan-rate-if-migration-failures-occur mm/memory.c
--- a/mm/memory.c~mm-numa-slow-pte-scan-rate-if-migration-failures-occur
+++ a/mm/memory.c
@@ -3103,7 +3103,8 @@ static int do_numa_page(struct mm_struct
if (migrated) {
page_nid = target_nid;
flags |= TNF_MIGRATED;
- }
+ } else
+ flags |= TNF_MIGRATE_FAIL;
out:
if (page_nid != -1)
_
Patches currently in -mm which might be from mgorman@xxxxxxx are
mm-page_alloc-call-kernel_map_pages-in-unset_migrateype_isolate.patch
mm-numa-group-related-processes-based-on-vma-flags-instead-of-page-table-flags.patch
mm-numa-preserve-pte-write-permissions-across-a-numa-hinting-fault.patch
mm-numa-slow-pte-scan-rate-if-migration-failures-occur.patch
mm-numa-mark-huge-ptes-young-when-clearing-numa-hinting-faults.patch
cxgb4-drop-__gfp_nofail-allocation.patch
jbd2-revert-must-not-fail-allocation-loops-back-to-gfp_nofail.patch
mm-cma-change-fallback-behaviour-for-cma-freepage.patch
mm-page_alloc-factor-out-fallback-freepage-checking.patch
mm-compaction-enhance-compaction-finish-condition.patch
mm-compaction-enhance-compaction-finish-condition-fix.patch
mm-refactor-do_wp_page-extract-the-reuse-case.patch
mm-refactor-do_wp_page-rewrite-the-unlock-flow.patch
mm-refactor-do_wp_page-extract-the-page-copy-flow.patch
mm-refactor-do_wp_page-handling-of-shared-vma-into-a-function.patch
mm-remove-gfp_thisnode.patch
mm-thp-really-limit-transparent-hugepage-allocation-to-local-node.patch
kernel-cpuset-remove-exception-for-__gfp_thisnode.patch
mm-clarify-__gfp_nofail-deprecation-status.patch
sparc-clarify-__gfp_nofail-allocation.patch
mm-numa-remove-migrate_ratelimited.patch
mm-consolidate-all-page-flags-helpers-in-linux-page-flagsh.patch
page-flags-trivial-cleanup-for-pagetrans-helpers.patch
page-flags-introduce-page-flags-policies-wrt-compound-pages.patch
page-flags-define-pg_locked-behavior-on-compound-pages.patch
page-flags-define-behavior-of-fs-io-related-flags-on-compound-pages.patch
page-flags-define-behavior-of-lru-related-flags-on-compound-pages.patch
page-flags-define-behavior-slb-related-flags-on-compound-pages.patch
page-flags-define-behavior-of-xen-related-flags-on-compound-pages.patch
page-flags-define-pg_reserved-behavior-on-compound-pages.patch
page-flags-define-pg_swapbacked-behavior-on-compound-pages.patch
page-flags-define-pg_swapcache-behavior-on-compound-pages.patch
page-flags-define-pg_mlocked-behavior-on-compound-pages.patch
page-flags-define-pg_uncached-behavior-on-compound-pages.patch
page-flags-define-pg_uptodate-behavior-on-compound-pages.patch
page-flags-look-on-head-page-if-the-flag-is-encoded-in-page-mapping.patch
mm-sanitize-page-mapping-for-tail-pages.patch
allow-compaction-of-unevictable-pages.patch
mm-change-deactivate_page-with-deactivate_file_page.patch
mm-vmscan-fix-the-page-state-calculation-in-too_many_isolated.patch
mm-move-lazy-free-pages-to-inactive-list.patch
linux-next.patch
do_shared_fault-check-that-mmap_sem-is-held.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
