mm/mm_init: Don't cond_resched() in deferred_init_memmap_chunk() if called from deferred_grow_zone()
Subsystems (3)
PATCH v3 1/1 mm/mm_init: Don't cond_resched() in deferred_init_memmap_chunk() if called from deferred_grow_zone()
Commit 3acb913c9d5b ("mm/mm_init: use deferred_init_memmap_chunk() in deferred_grow_zone()") made deferred_grow_zone() call deferred_init_memmap_chunk() within a pgdat_resize_lock() critical section with irqs disabled. It did check for irqs_disabled() in deferred_init_memmap_chunk() to avoid calling cond_resched(). For a PREEMPT_RT kernel build, however, spin_lock_irqsave() does not disable interrupt but rcu_read_lock() is called. This leads to the following bug report.
BUG: sleeping function called from invalid context at mm/mm_init.c:2091
in_atomic(): 0, irqs_disabled(): 0, non_block: 0, pid: 1, name: swapper/0
preempt_count: 0, expected: 0
RCU nest depth: 1, expected: 0
3 locks held by swapper/0/1:
#0: ffff80008471b7a0 (sched_domains_mutex){+.+.}-{4:4}, at: sched_domains_mutex_lock+0x28/0x40
#1: ffff003bdfffef48 (&pgdat->node_size_lock){+.+.}-{3:3}, at: deferred_grow_zone+0x140/0x278
#2: ffff800084acf600 (rcu_read_lock){....}-{1:3}, at: rt_spin_lock+0x1b4/0x408
CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Tainted: G W 6.19.0-rc6-test #1 PREEMPT_{RT,(full)
}
Tainted: [W]=WARN
Call trace:
show_stack+0x20/0x38 (C)
dump_stack_lvl+0xdc/0xf8
dump_stack+0x1c/0x28
__might_resched+0x384/0x530
deferred_init_memmap_chunk+0x560/0x688
deferred_grow_zone+0x190/0x278
_deferred_grow_zone+0x18/0x30
get_page_from_freelist+0x780/0xf78
__alloc_frozen_pages_noprof+0x1dc/0x348
alloc_slab_page+0x30/0x110
allocate_slab+0x98/0x2a0
new_slab+0x4c/0x80
___slab_alloc+0x5a4/0x770
__slab_alloc.constprop.0+0x88/0x1e0
__kmalloc_node_noprof+0x2c0/0x598
__sdt_alloc+0x3b8/0x728
build_sched_domains+0xe0/0x1260
sched_init_domains+0x14c/0x1c8
sched_init_smp+0x9c/0x1d0
kernel_init_freeable+0x218/0x358
kernel_init+0x28/0x208
ret_from_fork+0x10/0x20Fix it adding a new argument to deferred_init_memmap_chunk() to explicitly tell it if cond_resched() is allowed or not instead of relying on some current state information which may vary depending on the exact kernel configuration options that are enabled.
---
[v3]: Add a new can_resched argument to deferred_init_memmap_chunk() as
suggested by Sebastian.
mm/mm_init.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
mm/mm_init.c +6 -6
@@ -2059,7 +2059,7 @@ static unsigned long __init deferred_init_pages(struct zone *zone,*/ static unsigned long __init deferred_init_memmap_chunk (unsigned long start_pfn,unsigned long end_pfn, -struct zone *zone) +struct zone *zone,bool can_resched) {int nid =zone_to_nid (zone);unsigned long nr_pages =0 ;
@@ -2085,10 +2085,10 @@ deferred_init_memmap_chunk(unsigned long start_pfn, unsigned long end_pfn, spfn = chunk_end; -if (irqs_disabled ()) -touch_nmi_watchdog (); -else +if (can_resched)cond_resched (); +else +touch_nmi_watchdog (); } }
@@ -2101,7 +2101,7 @@ deferred_init_memmap_job(unsigned long start_pfn, unsigned long end_pfn, {struct zone *zone = arg; -deferred_init_memmap_chunk (start_pfn, end_pfn, zone); +deferred_init_memmap_chunk (start_pfn, end_pfn, zone, true); } static unsigned int __init
@@ -2216,7 +2216,7 @@ bool __init deferred_grow_zone(struct zone *zone, unsigned int order)for (spfn = first_deferred_pfn, epfn =SECTION_ALIGN_UP (spfn +1 ); nr_pages < nr_pages_needed && spfn <zone_end_pfn (zone); spfn = epfn, epfn +=PAGES_PER_SECTION ) { - nr_pages +=deferred_init_memmap_chunk (spfn, epfn, zone); + nr_pages +=deferred_init_memmap_chunk (spfn, epfn, zone, false); }/*
Signature · 1 line
2.52.0
Andrew Morton
maintainerQuoted 18 lines · On Thu, 22 Jan 2026 13:43:43 -0500 Waiman Long <[email protected]> wrote:
Commit
3acb913c9d5b("mm/mm_init: usedeferred_init_memmap_chunk()indeferred_grow_zone()") madedeferred_grow_zone()calldeferred_init_memmap_chunk()within apgdat_resize_lock()critical section with irqs disabled.It did check for
irqs_disabled()indeferred_init_memmap_chunk()to avoid callingcond_resched(). For aPREEMPT_RTkernel build, however,spin_lock_irqsave()does not disable interrupt butrcu_read_lock()is called. This leads to the following bug report.BUG: sleeping function called from invalid context at mm/mm_init.c:2091 in_atomic(): 0, irqs_disabled(): 0, non_block: 0, pid: 1, name: swapper/0 preempt_count: 0, expected: 0mm/mm_init.c +0
-0@@ -2085,10 +2085,10 @@ deferred_init_memmap_chunk(unsigned long start_pfn, unsigned long end_pfn, spfn = chunk_end;
mm/mm_init.c deferred_init_memmap_chunk(unsigned long start_pfn, unsigned long end_pfn, · line 2093
Quoted 10 lines
mm/mm_init.c +3
-3-if (irqs_disabled ()) -touch_nmi_watchdog (); -else +if (can_resched)cond_resched (); +else +touch_nmi_watchdog (); } }
Disables the cond_resched() in some situations. Can this reintroduce the watchdog warnings which that cond_resched() was intended to prevent?
The cond_resched() was added by <dig, dig> da97f2d56bbd ("mm: call cond_resched() from deferred_init_memmap()").
Pasha's 2020 patch replaced touch_nmi_watchdog() with cond_resched() to prevent RCU stall warnings. So I think the answer to my question is yes, going back to touch_nmi_watchdog() could reintroduce those RCU warnings.
Quoted 21 lines · Andrew Morton
Quoted 19 lines · On Thu, 22 Jan 2026 13:43:43 -0500 Waiman Long <[email protected]> wrote:
Commit
3acb913c9d5b("mm/mm_init: usedeferred_init_memmap_chunk()indeferred_grow_zone()") madedeferred_grow_zone()calldeferred_init_memmap_chunk()within apgdat_resize_lock()critical section with irqs disabled.It did check for
irqs_disabled()indeferred_init_memmap_chunk()to avoid callingcond_resched(). For aPREEMPT_RTkernel build, however,spin_lock_irqsave()does not disable interrupt butrcu_read_lock()is called. This leads to the following bug report.BUG: sleeping function called from invalid context at mm/mm_init.c:2091 in_atomic(): 0, irqs_disabled(): 0, non_block: 0, pid: 1, name: swapper/0 preempt_count: 0, expected: 0mm/mm_init.c +0
-0@@ -2085,10 +2085,10 @@ deferred_init_memmap_chunk(unsigned long start_pfn, unsigned long end_pfn, spfn = chunk_end;
Quoted 12 lines
mm/mm_init.c deferred_init_memmap_chunk(unsigned long start_pfn, unsigned long end_pfn, · line 2093
Quoted 9 lines
mm/mm_init.c +3
-3-if (irqs_disabled ()) -touch_nmi_watchdog (); -else +if (can_resched)cond_resched (); +else +touch_nmi_watchdog (); } }Disables the
cond_resched()in some situations. Can this reintroduce the watchdog warnings which thatcond_resched()was intended to prevent?
cond_resched() is disabled only when it is called from deferred_grow_zone() where a spinlock was acquired with irqs disabled in the case of non-RT kernel and in a rcu_read_lock() acquired with RT kernel. In either case, scheduling out should not be allowed or something bad may happen. I suppose that iterating of pfn's in deferred_grow_zone() requires pgdat_resize_lock() protection.
Quoted 8 lines
The
cond_resched()was added by <dig, dig>da97f2d56bbd("mm: callcond_resched()fromdeferred_init_memmap()").Pasha's 2020 patch replaced
touch_nmi_watchdog()withcond_resched()to prevent RCU stall warnings. So I think the answer to my question is yes, going back totouch_nmi_watchdog()could reintroduce those RCU warnings.
deferred_init_memmap() will still have cond_resched() called in the iteration loop. It had RCU stall problem before without cond_resched() because it needs to iterate all the available memory which can takes a long time if we are talking about TBs of memory.
For deferred_grow_zone(), as long as the number of pfn's that are iterated are not huge, RCU stall warning shouldn't happen.
Cheers,
Longman
Mike Rapoport
maintainerQuoted 35 lines · Andrew Morton
Quoted 18 lines · On Thu, 22 Jan 2026 13:43:43 -0500 Waiman Long <[email protected]> wrote:
Commit
3acb913c9d5b("mm/mm_init: usedeferred_init_memmap_chunk()indeferred_grow_zone()") madedeferred_grow_zone()calldeferred_init_memmap_chunk()within apgdat_resize_lock()critical section with irqs disabled.It did check for
irqs_disabled()indeferred_init_memmap_chunk()to avoid callingcond_resched(). For aPREEMPT_RTkernel build, however,spin_lock_irqsave()does not disable interrupt butrcu_read_lock()is called. This leads to the following bug report.BUG: sleeping function called from invalid context at mm/mm_init.c:2091 in_atomic(): 0, irqs_disabled(): 0, non_block: 0, pid: 1, name: swapper/0 preempt_count: 0, expected: 0mm/mm_init.c +0
-0@@ -2085,10 +2085,10 @@ deferred_init_memmap_chunk(unsigned long start_pfn, unsigned long end_pfn, spfn = chunk_end;mm/mm_init.c deferred_init_memmap_chunk(unsigned long start_pfn, unsigned long end_pfn, · line 2093
Quoted 10 lines
mm/mm_init.c +3
-3-if (irqs_disabled ()) -touch_nmi_watchdog (); -else +if (can_resched)cond_resched (); +else +touch_nmi_watchdog (); } }Disables the
cond_resched()in some situations. Can this reintroduce the watchdog warnings which thatcond_resched()was intended to prevent?
Quoted 7 lines
The
cond_resched()was added by <dig, dig>da97f2d56bbd("mm: callcond_resched()fromdeferred_init_memmap()").Pasha's 2020 patch replaced
touch_nmi_watchdog()withcond_resched()to prevent RCU stall warnings. So I think the answer to my question is yes, going back totouch_nmi_watchdog()could reintroduce those RCU warnings.
Before 3acb913c9d5b ("mm/mm_init: use deferred_init_memmap_chunk() we had touch_nmi_watchdog() in deferred_grow_zone() and cond_resched() in the deferred_init_memmap()->deferred_init_memmap_chunk() that ran in a thread context.
I thought irqs_disabled() would be enough to differentiate these cases because deferred_grow_zone() takes a spinlock, but I missed that with RT spinlock also sleeps.
Using a boolean essentially restores the behaviour we had before the refactoring.
Signature · 2 lines
Sincerely yours, Mike.
On 2026-01-22 13:43:43 [-0500], Waiman Long wrote:
…
Quoted 28 lines
bug report.
BUG: sleeping function called from invalid context at mm/mm_init.c:2091 in_atomic(): 0, irqs_disabled(): 0, non_block: 0, pid: 1, name: swapper/0 preempt_count: 0, expected: 0 RCU nest depth: 1, expected: 0 3 locks held by swapper/0/1: #0: ffff80008471b7a0 (sched_domains_mutex){+.+.}-{4:4}, at: sched_domains_mutex_lock+0x28/0x40 #1: ffff003bdfffef48 (&pgdat->node_size_lock){+.+.}-{3:3}, at: deferred_grow_zone+0x140/0x278 #2: ffff800084acf600 (rcu_read_lock){....}-{1:3}, at: rt_spin_lock+0x1b4/0x408 CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Tainted: G W 6.19.0-rc6-test #1 PREEMPT_{RT,(full) } Tainted: [W]=WARN Call trace: show_stack+0x20/0x38 (C) dump_stack_lvl+0xdc/0xf8 dump_stack+0x1c/0x28 __might_resched+0x384/0x530 deferred_init_memmap_chunk+0x560/0x688 deferred_grow_zone+0x190/0x278 _deferred_grow_zone+0x18/0x30 get_page_from_freelist+0x780/0xf78 __alloc_frozen_pages_noprof+0x1dc/0x348 alloc_slab_page+0x30/0x110 allocate_slab+0x98/0x2a0 new_slab+0x4c/0x80 ___slab_alloc+0x5a4/0x770 __slab_alloc.constprop.0+0x88/0x1e0
Quoted 8 lines
__kmalloc_node_noprof+0x2c0/0x598 __sdt_alloc+0x3b8/0x728 build_sched_domains+0xe0/0x1260 sched_init_domains+0x14c/0x1c8 sched_init_smp+0x9c/0x1d0 kernel_init_freeable+0x218/0x358 kernel_init+0x28/0x208 ret_from_fork+0x10/0x20
I would strip this report because the call chain is simple and there just one so it is not one of many and hard to find.
…
Quoted 4 lines
Fixes 3acb913c9d5b ("mm/mm_init: use deferred_init_memmap_chunk() in deferred_grow_zone()")Suggested-by Sebastian Andrzej Siewior <[email protected]>Signed-off-by Waiman Long <[email protected]>
Sebastian
Quoted 44 lines · Waiman Long
Commit
3acb913c9d5b("mm/mm_init: usedeferred_init_memmap_chunk()indeferred_grow_zone()") madedeferred_grow_zone()calldeferred_init_memmap_chunk()within apgdat_resize_lock()critical section with irqs disabled. It did check forirqs_disabled()indeferred_init_memmap_chunk()to avoid callingcond_resched(). For aPREEMPT_RTkernel build, however,spin_lock_irqsave()does not disable interrupt butrcu_read_lock()is called. This leads to the following bug report.BUG: sleeping function called from invalid context at mm/mm_init.c:2091 in_atomic(): 0, irqs_disabled(): 0, non_block: 0, pid: 1, name: swapper/0 preempt_count: 0, expected: 0 RCU nest depth: 1, expected: 0 3 locks held by swapper/0/1: #0: ffff80008471b7a0 (sched_domains_mutex){+.+.}-{4:4}, at: sched_domains_mutex_lock+0x28/0x40 #1: ffff003bdfffef48 (&pgdat->node_size_lock){+.+.}-{3:3}, at: deferred_grow_zone+0x140/0x278 #2: ffff800084acf600 (rcu_read_lock){....}-{1:3}, at: rt_spin_lock+0x1b4/0x408 CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Tainted: G W 6.19.0-rc6-test #1 PREEMPT_{RT,(full) } Tainted: [W]=WARN Call trace: show_stack+0x20/0x38 (C) dump_stack_lvl+0xdc/0xf8 dump_stack+0x1c/0x28 __might_resched+0x384/0x530 deferred_init_memmap_chunk+0x560/0x688 deferred_grow_zone+0x190/0x278 _deferred_grow_zone+0x18/0x30 get_page_from_freelist+0x780/0xf78 __alloc_frozen_pages_noprof+0x1dc/0x348 alloc_slab_page+0x30/0x110 allocate_slab+0x98/0x2a0 new_slab+0x4c/0x80 ___slab_alloc+0x5a4/0x770 __slab_alloc.constprop.0+0x88/0x1e0 __kmalloc_node_noprof+0x2c0/0x598 __sdt_alloc+0x3b8/0x728 build_sched_domains+0xe0/0x1260 sched_init_domains+0x14c/0x1c8 sched_init_smp+0x9c/0x1d0 kernel_init_freeable+0x218/0x358 kernel_init+0x28/0x208 ret_from_fork+0x10/0x20
Quoted 8 lines
Fix it adding a new argument to
deferred_init_memmap_chunk()to explicitly tell it ifcond_resched()is allowed or not instead of relying on some current state information which may vary depending on the exact kernel configuration options that are enabled.Fixes 3acb913c9d5b ("mm/mm_init: use deferred_init_memmap_chunk() in deferred_grow_zone()")Suggested-by Sebastian Andrzej Siewior <[email protected]>Signed-off-by Waiman Long <[email protected]>
Signature · 2 lines
Sincerely yours, Mike.