Recalculate per-cpu page allocator batch and high limits after deferred meminit v2
Subsystems (4)
cover Recalculate per-cpu page allocator batch and high limits after deferred meminit v2
This is an updated series that addresses some review feedback and an LKP warning. I did not preserve Michal Hocko's ack for the fix as it has changed. This series replaces the following patches in mmotm
mm-pcp-share-common-code-between-memory-hotplug-and-percpu-sysctl-handler.patch mm-meminit-recalculate-pcpu-batch-and-high-limits-after-init-completes.patch mm-pcpu-make-zone-pcp-updates-and-reset-internal-to-the-mm.patch
Changelog since V1
o Fix a "might sleep" warning
o Reorder for easier backporting
A private report stated that system CPU usage was excessive on an AMD EPYC 2 machine while building kernels with much longer build times than expected. The issue is partially explained by high zone lock contention due to the per-cpu page allocator batch and high limits being calculated incorrectly. This series addresses a large chunk of the problem. Patch 1 is the real fix and the other two are cosmetic issues noticed while implementing the fix.
include/linux/mm.h | 3 --- mm/internal.h | 3 +++ mm/page_alloc.c | 31 ++++++++++++++++++++----------- 3 files changed, 23 insertions(+), 14 deletions(-)
Signature · 1 line
2.16.4
PATCH 1/3 mm, meminit: Recalculate pcpu batch and high limits after init completes
Deferred memory initialisation updates zone->managed_pages during the initialisation phase but before that finishes, the per-cpu page allocator (pcpu) calculates the number of pages allocated/freed in batches as well as the maximum number of pages allowed on a per-cpu list. As zone->managed_pages is not up to date yet, the pcpu initialisation calculates inappropriately low batch and high values.
This increases zone lock contention quite severely in some cases with the degree of severity depending on how many CPUs share a local zone and the size of the zone. A private report indicated that kernel build times were excessive with extremely high system CPU usage. A perf profile indicated that a large chunk of time was lost on zone->lock contention.
This patch recalculates the pcpu batch and high values after deferred initialisation completes for every populated zone in the system. It was tested on a 2-socket AMD EPYC 2 machine using a kernel compilation workload -- allmodconfig and all available CPUs.
mmtests configuration: config-workload-kernbench-max Configuration was modified to build on a fresh XFS partition.
kernbench
5.4.0-rc3 5.4.0-rc3
vanilla resetpcpu-v2
Amean user-256 13249.50 ( 0.00%) 16401.31 * -23.79%*
Amean syst-256 14760.30 ( 0.00%) 4448.39 * 69.86%*
Amean elsp-256 162.42 ( 0.00%) 119.13 * 26.65%*
Stddev user-256 42.97 ( 0.00%) 19.15 ( 55.43%)
Stddev syst-256 336.87 ( 0.00%) 6.71 ( 98.01%)
Stddev elsp-256 2.46 ( 0.00%) 0.39 ( 84.03%) 5.4.0-rc3 5.4.0-rc3
vanilla resetpcpu-v2
Duration User 39766.24 49221.79
Duration System 44298.10 13361.67Duration Elapsed 519.11 388.87
The patch reduces system CPU usage by 69.86% and total build time by 26.65%. The variance of system CPU usage is also much reduced.
Before, this was the breakdown of batch and high values over all zones was.
256 batch: 1
256 batch: 63
512 batch: 7
256 high: 0
256 high: 378
512 high: 42512 pcpu pagesets had a batch limit of 7 and a high limit of 42. After the patch
256 batch: 1
768 batch: 63
256 high: 0
768 high: 378--- mm/page_alloc.c | 8 ++++++++ 1 file changed, 8 insertions(+)
mm/page_alloc.c +8 -0
@@ -1947,6 +1947,14 @@ void __init page_alloc_init_late(void)/* Block until all are initialised */ wait_for_completion (&pgdat_init_all_done_comp); +/* +* The number of managed pages has changed due to the initialisation +* so the pcpu batch and high limits needs to be updated or the limits +* will be artificially small. +*/ +for_each_populated_zone (zone) +zone_pcp_update (zone); +/* * We initialized the rest of the deferred pages. Permanently disable * on-demand struct page initialization.
Signature · 1 line
2.16.4
PATCH 2/3 mm, pcp: Share common code between memory hotplug and percpu sysctl handler
In reply to Mel GormanBoth the percpu_pagelist_fraction sysctl handler and memory hotplug have a common requirement of updating the pcpu page allocation batch and high values. Split the relevant helper to share common code.
No functional change.
--- mm/page_alloc.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-)
mm/page_alloc.c +12 -11
@@ -7991,6 +7991,15 @@ int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *table, int write,return 0 ; } +static void __zone_pcp_update (struct zone *zone) +{ +unsigned int cpu; + +for_each_possible_cpu (cpu) +pageset_set_high_and_batch (zone, +per_cpu_ptr (zone->pageset, cpu)); +} +/* * percpu_pagelist_fraction - changes the pcp->high for each zone on each * cpu. It is the fraction of total pages in each zone that a hot per cpu
@@ -8022,13 +8031,8 @@ int percpu_pagelist_fraction_sysctl_handler(struct ctl_table *table, int write,if (percpu_pagelist_fraction == old_percpu_pagelist_fraction) goto out; -for_each_populated_zone (zone) { -unsigned int cpu; - -for_each_possible_cpu (cpu) -pageset_set_high_and_batch (zone, -per_cpu_ptr (zone->pageset, cpu)); - } +for_each_populated_zone (zone) +__zone_pcp_update (zone); out:mutex_unlock (&pcp_batch_high_lock);return ret;
@@ -8527,11 +8531,8 @@ void free_contig_range(unsigned long pfn, unsigned int nr_pages)*/ void __meminit zone_pcp_update (struct zone *zone) { -unsigned cpu;mutex_lock (&pcp_batch_high_lock); -for_each_possible_cpu (cpu) -pageset_set_high_and_batch (zone, -per_cpu_ptr (zone->pageset, cpu)); +__zone_pcp_update (zone);mutex_unlock (&pcp_batch_high_lock); }#endif
Signature · 1 line
2.16.4
PATCH 3/3 mm, pcpu: Make zone pcp updates and reset internal to the mm
In reply to Mel GormanMemory hotplug needs to be able to reset and reinit the pcpu allocator batch and high limits but this action is internal to the VM. Move the declaration to internal.h
--- include/linux/mm.h | 3 --- mm/internal.h | 3 +++ 2 files changed, 3 insertions(+), 3 deletions(-)
include/linux/mm.h +0 -3
@@ -2219,9 +2219,6 @@ void warn_alloc(gfp_t gfp_mask, nodemask_t *nodemask, const char *fmt, ...);extern void setup_per_cpu_pageset (void ); -extern void zone_pcp_update (struct zone *zone); -extern void zone_pcp_reset (struct zone *zone); -/* page_alloc.c */ extern int min_free_kbytes;extern int watermark_boost_factor;
mm/internal.h +3 -0
@@ -165,6 +165,9 @@ extern void post_alloc_hook(struct page *page, unsigned int order, gfp_t gfp_flags);extern int user_min_free_kbytes; +extern void zone_pcp_update (struct zone *zone); +extern void zone_pcp_reset (struct zone *zone); +#if definedCONFIG_COMPACTION || definedCONFIG_CMA /*
Signature · 1 line
2.16.4
Re: [PATCH 1/3] mm, meminit: Recalculate pcpu batch and high limits after init completes
In reply to Mel GormanQuoted 51 lines · Mel Gorman
Deferred memory initialisation updates
zone->managed_pagesduring the initialisation phase but before that finishes, the per-cpu page allocator (pcpu) calculates the number of pages allocated/freed in batches as well as the maximum number of pages allowed on a per-cpu list. Aszone->managed_pagesis not up to date yet, the pcpu initialisation calculates inappropriately low batch and high values.This increases zone lock contention quite severely in some cases with the degree of severity depending on how many CPUs share a local zone and the size of the zone. A private report indicated that kernel build times were excessive with extremely high system CPU usage. A perf profile indicated that a large chunk of time was lost on
zone->lockcontention.This patch recalculates the pcpu batch and high values after deferred initialisation completes for every populated zone in the system. It was tested on a 2-socket AMD EPYC 2 machine using a kernel compilation workload -- allmodconfig and all available CPUs.
mmtests configuration: config-workload-kernbench-max Configuration was modified to build on a fresh XFS partition.
kernbench
5.4.0-rc3 5.4.0-rc3 vanilla resetpcpu-v2 Amean user-256 13249.50 ( 0.00%) 16401.31 * -23.79%* Amean syst-256 14760.30 ( 0.00%) 4448.39 * 69.86%* Amean elsp-256 162.42 ( 0.00%) 119.13 * 26.65%* Stddev user-256 42.97 ( 0.00%) 19.15 ( 55.43%) Stddev syst-256 336.87 ( 0.00%) 6.71 ( 98.01%) Stddev elsp-256 2.46 ( 0.00%) 0.39 ( 84.03%)5.4.0-rc3 5.4.0-rc3 vanilla resetpcpu-v2 Duration User 39766.24 49221.79 Duration System 44298.10 13361.67Duration Elapsed 519.11 388.87
The patch reduces system CPU usage by 69.86% and total build time by 26.65%. The variance of system CPU usage is also much reduced.
Before, this was the breakdown of batch and high values over all zones was.
256 batch: 1 256 batch: 63 512 batch: 7 256 high: 0 256 high: 378 512 high: 42512 pcpu pagesets had a batch limit of 7 and a high limit of 42. After the patch
Quoted 7 lines
256 batch: 1 768 batch: 63 256 high: 0 768 high: 378Cc [email protected] # v4.1+Signed-off-by Mel Gorman <[email protected]>
Quoted 25 lines
--- mm/page_alloc.c | 8 ++++++++ 1 file changed, 8 insertions(+)mm/page_alloc.c +8
-0@@ -1947,6 +1947,14 @@ void __init page_alloc_init_late(void)/* Block until all are initialised */ wait_for_completion (&pgdat_init_all_done_comp); +/* +* The number of managed pages has changed due to the initialisation +* so the pcpu batch and high limits needs to be updated or the limits +* will be artificially small. +*/ +for_each_populated_zone (zone) +zone_pcp_update (zone); +/* * We initialized the rest of the deferred pages. Permanently disable * on-demand struct page initialization. Signature · 1 line
2.16.4
Signature · 2 lines
Michal Hocko SUSE Labs
Re: [PATCH 1/3] mm, meminit: Recalculate pcpu batch and high limits after init completes
In reply to Mel GormanQuoted 51 lines · Mel Gorman
Deferred memory initialisation updates
zone->managed_pagesduring the initialisation phase but before that finishes, the per-cpu page allocator (pcpu) calculates the number of pages allocated/freed in batches as well as the maximum number of pages allowed on a per-cpu list. Aszone->managed_pagesis not up to date yet, the pcpu initialisation calculates inappropriately low batch and high values.This increases zone lock contention quite severely in some cases with the degree of severity depending on how many CPUs share a local zone and the size of the zone. A private report indicated that kernel build times were excessive with extremely high system CPU usage. A perf profile indicated that a large chunk of time was lost on
zone->lockcontention.This patch recalculates the pcpu batch and high values after deferred initialisation completes for every populated zone in the system. It was tested on a 2-socket AMD EPYC 2 machine using a kernel compilation workload -- allmodconfig and all available CPUs.
mmtests configuration: config-workload-kernbench-max Configuration was modified to build on a fresh XFS partition.
kernbench
5.4.0-rc3 5.4.0-rc3 vanilla resetpcpu-v2 Amean user-256 13249.50 ( 0.00%) 16401.31 * -23.79%* Amean syst-256 14760.30 ( 0.00%) 4448.39 * 69.86%* Amean elsp-256 162.42 ( 0.00%) 119.13 * 26.65%* Stddev user-256 42.97 ( 0.00%) 19.15 ( 55.43%) Stddev syst-256 336.87 ( 0.00%) 6.71 ( 98.01%) Stddev elsp-256 2.46 ( 0.00%) 0.39 ( 84.03%)5.4.0-rc3 5.4.0-rc3 vanilla resetpcpu-v2 Duration User 39766.24 49221.79 Duration System 44298.10 13361.67Duration Elapsed 519.11 388.87
The patch reduces system CPU usage by 69.86% and total build time by 26.65%. The variance of system CPU usage is also much reduced.
Before, this was the breakdown of batch and high values over all zones was.
256 batch: 1 256 batch: 63 512 batch: 7 256 high: 0 256 high: 378 512 high: 42512 pcpu pagesets had a batch limit of 7 and a high limit of 42. After the patch
Quoted 7 lines
256 batch: 1 768 batch: 63 256 high: 0 768 high: 378Cc [email protected] # v4.1+Signed-off-by Mel Gorman <[email protected]>
Quoted 24 lines
--- mm/page_alloc.c | 8 ++++++++ 1 file changed, 8 insertions(+)mm/page_alloc.c +8
-0@@ -1947,6 +1947,14 @@ void __init page_alloc_init_late(void)/* Block until all are initialised */ wait_for_completion (&pgdat_init_all_done_comp); +/* +* The number of managed pages has changed due to the initialisation +* so the pcpu batch and high limits needs to be updated or the limits +* will be artificially small. +*/ +for_each_populated_zone (zone) +zone_pcp_update (zone); +/* * We initialized the rest of the deferred pages. Permanently disable * on-demand struct page initialization.
Re: [PATCH 3/3] mm, pcpu: Make zone pcp updates and reset internal to the mm
In reply to Mel GormanQuoted 6 lines · Mel Gorman
Memory hotplug needs to be able to reset and reinit the pcpu allocator batch and high limits but this action is internal to the VM. Move the declaration to
internal.hSigned-off-by Mel Gorman <[email protected]>Acked-by Michal Hocko <[email protected]>
Re: [PATCH 2/3] mm, pcp: Share common code between memory hotplug and percpu sysctl handler
In reply to Mel GormanQuoted 8 lines · Mel Gorman
Both the
percpu_pagelist_fractionsysctl handler and memory hotplug have a common requirement of updating the pcpu page allocation batch and high values. Split the relevant helper to share common code.No functional change.
Signed-off-by Mel Gorman <[email protected]>Acked-by Michal Hocko <[email protected]>
Re: [PATCH 1/3] mm, meminit: Recalculate pcpu batch and high limits after init completes
In reply to Mel GormanQuoted 89 lines · Mel Gorman
Deferred memory initialisation updates
zone->managed_pagesduring the initialisation phase but before that finishes, the per-cpu page allocator (pcpu) calculates the number of pages allocated/freed in batches as well as the maximum number of pages allowed on a per-cpu list. Aszone->managed_pagesis not up to date yet, the pcpu initialisation calculates inappropriately low batch and high values.This increases zone lock contention quite severely in some cases with the degree of severity depending on how many CPUs share a local zone and the size of the zone. A private report indicated that kernel build times were excessive with extremely high system CPU usage. A perf profile indicated that a large chunk of time was lost on
zone->lockcontention.This patch recalculates the pcpu batch and high values after deferred initialisation completes for every populated zone in the system. It was tested on a 2-socket AMD EPYC 2 machine using a kernel compilation workload -- allmodconfig and all available CPUs.
mmtests configuration: config-workload-kernbench-max Configuration was modified to build on a fresh XFS partition.
kernbench
5.4.0-rc3 5.4.0-rc3 vanilla resetpcpu-v2 Amean user-256 13249.50 ( 0.00%) 16401.31 * -23.79%* Amean syst-256 14760.30 ( 0.00%) 4448.39 * 69.86%* Amean elsp-256 162.42 ( 0.00%) 119.13 * 26.65%* Stddev user-256 42.97 ( 0.00%) 19.15 ( 55.43%) Stddev syst-256 336.87 ( 0.00%) 6.71 ( 98.01%) Stddev elsp-256 2.46 ( 0.00%) 0.39 ( 84.03%)5.4.0-rc3 5.4.0-rc3 vanilla resetpcpu-v2 Duration User 39766.24 49221.79 Duration System 44298.10 13361.67Duration Elapsed 519.11 388.87
The patch reduces system CPU usage by 69.86% and total build time by 26.65%. The variance of system CPU usage is also much reduced.
Before, this was the breakdown of batch and high values over all zones was.
256 batch: 1 256 batch: 63 512 batch: 7 256 high: 0 256 high: 378 512 high: 42512 pcpu pagesets had a batch limit of 7 and a high limit of 42. After the patch
256 batch: 1 768 batch: 63 256 high: 0 768 high: 378Cc [email protected] # v4.1+Signed-off-by Mel Gorman <[email protected]>---
mm/page_alloc.c| 8 ++++++++1 file changed, 8 insertions(+)mm/page_alloc.c +8
-0@@ -1947,6 +1947,14 @@ void __init page_alloc_init_late(void)/* Block until all are initialised */ wait_for_completion (&pgdat_init_all_done_comp); +/* +* The number of managed pages has changed due to the initialisation +* so the pcpu batch and high limits needs to be updated or the limits +* will be artificially small. +*/ +for_each_populated_zone (zone) +zone_pcp_update (zone); +/* * We initialized the rest of the deferred pages. Permanently disable * on-demand struct page initialization. Signature · 1 line
2.16.4
Warnings from linux-next,
[ 14.265911][ T659] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:935
[ 14.265992][ T659] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 659, name: pgdatinit8
[ 14.266044][ T659] 1 lock held by pgdatinit8/659:
[ 14.266075][ T659] #0: c000201ffca87b40 (&(&pgdat->node_size_lock)->rlock){....}, at: deferred_init_memmap+0xc4/0x26c
[ 14.266160][ T659] irq event stamp: 26
[ 14.266194][ T659] hardirqs last enabled at (25): [<c000000000950584>] _raw_spin_unlock_irq+0x44/0x80
[ 14.266246][ T659] hardirqs last disabled at (26): [<c0000000009502ec>] _raw_spin_lock_irqsave+0x3c/0xa0
[ 14.266299][ T659] softirqs last enabled at (0): [<c0000000000ff8d0>] copy_process+0x720/0x19b0
[ 14.266339][ T659] softirqs last disabled at (0): [<0000000000000000>] 0x0
[ 14.266400][ T659] CPU: 64 PID: 659 Comm: pgdatinit8 Not tainted 5.4.0-rc4-next-20191021 #1
[ 14.266462][ T659] Call Trace:
[ 14.266494][ T659] [c00000003d8efae0] [c000000000921cf4] dump_stack+0xe8/0x164 (unreliable)
[ 14.266538][ T659] [c00000003d8efb30] [c000000000157c54] ___might_sleep+0x334/0x370
[ 14.266577][ T659] [c00000003d8efbb0] [c00000000094a784] __mutex_lock+0x84/0xb20
[ 14.266627][ T659] [c00000003d8efcc0] [c000000000954038] zone_pcp_update+0x34/0x64
[ 14.266677][ T659] [c00000003d8efcf0] [c000000000b9e6bc] deferred_init_memmap+0x1b8/0x26c
[ 14.266740][ T659] [c00000003d8efdb0] [c000000000149528] kthread+0x1a8/0x1b0
[ 14.266792][ T659] [c00000003d8efe20] [c00000000000b748] ret_from_kernel_thread+0x5c/0x74
[ 14.268288][ T659] node 8 initialised, 1879186 pages in 12200ms
[ 14.268527][ T659] pgdatinit8 (659) used greatest stack depth: 27984 bytes left
[ 15.589983][ T658] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:935
[ 15.590041][ T658] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 658, name: pgdatinit0
[ 15.590078][ T658] 1 lock held by pgdatinit0/658:
[ 15.590108][ T658] #0: c000001fff5c7b40 (&(&pgdat->node_size_lock)->rlock){....}, at: deferred_init_memmap+0xc4/0x26c
[ 15.590192][ T658] irq event stamp: 18
[ 15.590224][ T658] hardirqs last enabled at (17): [<c000000000950654>] _raw_spin_unlock_irqrestore+0x94/0xd0
[ 15.590283][ T658] hardirqs last disabled at (18): [<c0000000009502ec>] _raw_spin_lock_irqsave+0x3c/0xa0
[ 15.590332][ T658] softirqs last enabled at (0): [<c0000000000ff8d0>] copy_process+0x720/0x19b0
[ 15.590379][ T658] softirqs last disabled at (0): [<0000000000000000>] 0x0
[ 15.590414][ T658] CPU: 8 PID: 658 Comm: pgdatinit0 Tainted: G W 5.4.0-rc4-next-20191021 #1
[ 15.590460][ T658] Call Trace:
[ 15.590491][ T658] [c00000003d8cfae0] [c000000000921cf4] dump_stack+0xe8/0x164 (unreliable)
[ 15.590541][ T658] [c00000003d8cfb30] [c000000000157c54] ___might_sleep+0x334/0x370
[ 15.590588][ T658] [c00000003d8cfbb0] [c00000000094a784] __mutex_lock+0x84/0xb20
[ 15.590643][ T658] [c00000003d8cfcc0] [c000000000954038] zone_pcp_update+0x34/0x64
[ 15.590689][ T658] [c00000003d8cfcf0] [c000000000b9e6bc] deferred_init_memmap+0x1b8/0x26c
[ 15.590739][ T658] [c00000003d8cfdb0] [c000000000149528] kthread+0x1a8/0x1b0
[ 15.590790][ T658] [c00000003d8cfe20] [c00000000000b748] ret_from_kernel_thread+0x5c/0x74Re: [PATCH 1/3] mm, meminit: Recalculate pcpu batch and high limits after init completes
Quoted 3 lines · Qian Cai
Quoted 3 lines · Mel Gorman
i[...]
Quoted 26 lines
Quoted 23 lines
mm/page_alloc.c +8
-0@@ -1947,6 +1947,14 @@ void __init page_alloc_init_late(void)/* Block until all are initialised */ wait_for_completion (&pgdat_init_all_done_comp); +/* +* The number of managed pages has changed due to the initialisation +* so the pcpu batch and high limits needs to be updated or the limits +* will be artificially small. +*/ +for_each_populated_zone (zone) +zone_pcp_update (zone); +/* * We initialized the rest of the deferred pages. Permanently disable * on-demand struct page initialization. Signature · 1 line
2.16.4Warnings from linux-next,
Quoted 4 lines
[ 14.265911][ T659] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:935
[ 14.265992][ T659] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 659, name: pgdatinit8
[ 14.266044][ T659] 1 lock held by pgdatinit8/659:
[ 14.266075][ T659] #0: c000201ffca87b40 (&(&pgdat->node_size_lock)->rlock){....}, at: deferred_init_memmap+0xc4/0x26cThis is really surprising to say the least. I do not see any spinlock held here. Besides that we do sleep in wait_for_completion already. Is it possible that the patch has been misplaced? zone_pcp_update is called from page_alloc_init_late which is a different context than deferred_init_memmap which runs in a separate kthread.
Quoted 34 lines
[ 14.266160][ T659] irq event stamp: 26
[ 14.266194][ T659] hardirqs last enabled at (25): [<c000000000950584>] _raw_spin_unlock_irq+0x44/0x80
[ 14.266246][ T659] hardirqs last disabled at (26): [<c0000000009502ec>] _raw_spin_lock_irqsave+0x3c/0xa0
[ 14.266299][ T659] softirqs last enabled at (0): [<c0000000000ff8d0>] copy_process+0x720/0x19b0
[ 14.266339][ T659] softirqs last disabled at (0): [<0000000000000000>] 0x0
[ 14.266400][ T659] CPU: 64 PID: 659 Comm: pgdatinit8 Not tainted 5.4.0-rc4-next-20191021 #1
[ 14.266462][ T659] Call Trace:
[ 14.266494][ T659] [c00000003d8efae0] [c000000000921cf4] dump_stack+0xe8/0x164 (unreliable)
[ 14.266538][ T659] [c00000003d8efb30] [c000000000157c54] ___might_sleep+0x334/0x370
[ 14.266577][ T659] [c00000003d8efbb0] [c00000000094a784] __mutex_lock+0x84/0xb20
[ 14.266627][ T659] [c00000003d8efcc0] [c000000000954038] zone_pcp_update+0x34/0x64
[ 14.266677][ T659] [c00000003d8efcf0] [c000000000b9e6bc] deferred_init_memmap+0x1b8/0x26c
[ 14.266740][ T659] [c00000003d8efdb0] [c000000000149528] kthread+0x1a8/0x1b0
[ 14.266792][ T659] [c00000003d8efe20] [c00000000000b748] ret_from_kernel_thread+0x5c/0x74
[ 14.268288][ T659] node 8 initialised, 1879186 pages in 12200ms
[ 14.268527][ T659] pgdatinit8 (659) used greatest stack depth: 27984 bytes left
[ 15.589983][ T658] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:935
[ 15.590041][ T658] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 658, name: pgdatinit0
[ 15.590078][ T658] 1 lock held by pgdatinit0/658:
[ 15.590108][ T658] #0: c000001fff5c7b40 (&(&pgdat->node_size_lock)->rlock){....}, at: deferred_init_memmap+0xc4/0x26c
[ 15.590192][ T658] irq event stamp: 18
[ 15.590224][ T658] hardirqs last enabled at (17): [<c000000000950654>] _raw_spin_unlock_irqrestore+0x94/0xd0
[ 15.590283][ T658] hardirqs last disabled at (18): [<c0000000009502ec>] _raw_spin_lock_irqsave+0x3c/0xa0
[ 15.590332][ T658] softirqs last enabled at (0): [<c0000000000ff8d0>] copy_process+0x720/0x19b0
[ 15.590379][ T658] softirqs last disabled at (0): [<0000000000000000>] 0x0
[ 15.590414][ T658] CPU: 8 PID: 658 Comm: pgdatinit0 Tainted: G W 5.4.0-rc4-next-20191021 #1
[ 15.590460][ T658] Call Trace:
[ 15.590491][ T658] [c00000003d8cfae0] [c000000000921cf4] dump_stack+0xe8/0x164 (unreliable)
[ 15.590541][ T658] [c00000003d8cfb30] [c000000000157c54] ___might_sleep+0x334/0x370
[ 15.590588][ T658] [c00000003d8cfbb0] [c00000000094a784] __mutex_lock+0x84/0xb20
[ 15.590643][ T658] [c00000003d8cfcc0] [c000000000954038] zone_pcp_update+0x34/0x64
[ 15.590689][ T658] [c00000003d8cfcf0] [c000000000b9e6bc] deferred_init_memmap+0x1b8/0x26c
[ 15.590739][ T658] [c00000003d8cfdb0] [c000000000149528] kthread+0x1a8/0x1b0
[ 15.590790][ T658] [c00000003d8cfe20] [c00000000000b748] ret_from_kernel_thread+0x5c/0x74Signature · 2 lines
Michal Hocko SUSE Labs
Re: [PATCH 1/3] mm, meminit: Recalculate pcpu batch and high limits after init completes
In reply to Qian CaiQuoted 5 lines · Qian Cai
Warnings from linux-next,
[ 14.265911][ T659] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:935 [ 14.265992][ T659] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 659, name: pgdatinit8 [ 14.266044][ T659] 1 lock held by pgdatinit8/659:
Fixed in v2 posted this morning. It should hit linux-next eventually.
Signature · 2 lines
Mel Gorman SUSE Labs
PATCH 1/1 mm, meminit: Recalculate pcpu batch and high limits after init completes -fix
In reply to Mel GormanLKP reported the following build problem from two hunks that did not survive the reshuffling of the series reordering.
ld: mm/page_alloc.o: in function `page_alloc_init_late':
mm/page_alloc.c:1956: undefined reference to `zone_pcp_update'
--- mm/page_alloc.c | 2 -- 1 file changed, 2 deletions(-)
mm/page_alloc.c +0 -2
@@ -8524,7 +8524,6 @@ void free_contig_range(unsigned long pfn, unsigned int nr_pages)WARN (count !=0 ,"%d pages are still in use!\n" , count); } -#ifdef CONFIG_MEMORY_HOTPLUG /* * The zone indicated has a new number of managed_pages; batch sizes and percpu * page high values need to be recalulated.
@@ -8535,7 +8534,6 @@ void __meminit zone_pcp_update(struct zone *zone)__zone_pcp_update (zone);mutex_unlock (&pcp_batch_high_lock); } -#endif void zone_pcp_reset (struct zone *zone) {
Re: [PATCH 1/3] mm, meminit: Recalculate pcpu batch and high limits after init completes
In reply to a message not in the archiveQuoted 44 lines · Sasha Levin
Hi,
[This is an automated email]
This commit has been processed because it contains a -stable tag. The stable tag indicates that it's relevant for the following trees: 4.1+
The bot has tested the following trees: v5.3.7, v4.19.80, v4.14.150, v4.9.197, v4.4.197.
v5.3.7: Build OK!
v4.19.80: Build failed! Errors:v4.14.150: Failed to apply! Possible dependencies:
3c2c648842843("mm/page_alloc.c: fix typos in comments")
66e8b438bd5c7("mm/memblock.c: make the index explicit argument offor_each_memblock_type")c9e97a1997fbf("mm: initialize pages on demand during boot")v4.9.197: Failed to apply! Possible dependencies:
3c2c648842843("mm/page_alloc.c: fix typos in comments")
66e8b438bd5c7("mm/memblock.c: make the index explicit argument offor_each_memblock_type")c9e97a1997fbf("mm: initialize pages on demand during boot")v4.4.197: Failed to apply! Possible dependencies:
0a687aace3b8e("mm,oom: do not loop !__GFP_FS allocation if the OOM killer is disabled")0caeef63e6d2f("libnvdimm: Add a poison list and export badblocks")0e749e54244ee("dax: increase granularity ofdax_clear_blocks()operations")34c0fd540e79f("mm, dax, pmem: introducepfn_t")
3c2c648842843("mm/page_alloc.c: fix typos in comments")
3da88fb3bacfa("mm, oom: moveGFP_NOFScheck toout_of_memory")4b94ffdc4163b("x86, mm: introducevmem_altmapto augmentvmemmap_populate()")5020e285856cb("mm, oom: give__GFP_NOFAILallocations access to memory reserves")52db400fcd502("pmem, dax: clean upclear_pmem()")
66e8b438bd5c7("mm/memblock.c: make the index explicit argument offor_each_memblock_type")7cf91a98e607c("mm/compaction: speed uppageblock_pfn_to_page()when zone is contiguous")87ba05dff3510("libnvdimm: don't fail init for full badblocks list")8c9c1701c7c23("mm/memblock: introducefor_each_memblock_type()")9476df7d80dfc("mm: introducefind_dev_pagemap()")
ad9a8bde2cb19("libnvdimm, pmem: move definition ofnvdimm_namespace_add_poisontond.h")b2e0d1625e193("dax: fix lifetime of in-kernel dax mappings withdax_map_atomic()")b95f5f4391fad("libnvdimm: convert to statically allocated badblocks")ba6c19fd113a3("include/linux/memblock.h: Clean up code for several trivial details")c9e97a1997fbf("mm: initialize pages on demand during boot")
Quoted 4 lines
NOTE: The patch will not be queued to stable trees until it is upstream.
How should we proceed with this patch?
What were the 4.19.80 build errors?
For the older kernels, it would have to be confirmed those kernels are definietly affected. The test machines I tried fails to even boot on those kernels so I need to find a NUMA machine that is old enough to boot those kernels and confirmed affected by the bug before determining what the backport needs to look like.
Signature · 2 lines
Mel Gorman SUSE Labs