All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] mm: Disable NUMA_BALANCING_DEFAULT_ENABLED and TRANSPARENT_HUGEPAGE on PREEMPT_RT
@ 2021-10-28 14:33 Sebastian Andrzej Siewior
  2021-10-28 14:37 ` Mel Gorman
  2021-10-28 15:54 ` David Hildenbrand
  0 siblings, 2 replies; 4+ messages in thread
From: Sebastian Andrzej Siewior @ 2021-10-28 14:33 UTC (permalink / raw)
  To: Andrew Morton, Mel Gorman
  Cc: linux-mm, Vlastimil Babka, Peter Zijlstra, Thomas Gleixner

TRANSPARENT_HUGEPAGE:
There are potential non-deterministic delays to an RT thread if a critical
memory region is not THP-aligned and a non-RT buffer is located in the same
hugepage-aligned region. It's also possible for an unrelated thread to migrate
pages belonging to an RT task incurring unexpected page faults due to memory
defragmentation even if khugepaged is disabled.

Regular HUGEPAGEs are not affected by this can be used.

NUMA_BALANCING:
There is a non-deterministic delay to mark PTEs PROT_NONE to gather NUMA fault
samples, increased page faults of regions even if mlocked and non-deterministic
delays when migrating pages.

[Mel Gorman worded 99% of the commit description].

Link: https://lore.kernel.org/all/20200304091159.GN3818@techsingularity.net/
Link: https://lore.kernel.org/all/20211026165100.ahz5bkx44lrrw5pt@linutronix.de/
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 init/Kconfig | 2 +-
 mm/Kconfig   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/init/Kconfig b/init/Kconfig
index edc0a0228f143..f83e7a0478f7c 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -911,7 +911,7 @@ config NUMA_BALANCING
 	bool "Memory placement aware NUMA scheduler"
 	depends on ARCH_SUPPORTS_NUMA_BALANCING
 	depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
-	depends on SMP && NUMA && MIGRATION
+	depends on SMP && NUMA && MIGRATION && !PREEMPT_RT
 	help
 	  This option adds support for automatic NUMA aware memory/task placement.
 	  The mechanism is quite primitive and is based on migrating memory when
diff --git a/mm/Kconfig b/mm/Kconfig
index c150a0c6fce2c..5c5508fafcec5 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -374,7 +374,7 @@ config NOMMU_INITIAL_TRIM_EXCESS
 
 config TRANSPARENT_HUGEPAGE
 	bool "Transparent Hugepage Support"
-	depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE
+	depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE && !PREEMPT_RT
 	select COMPACTION
 	select XARRAY_MULTI
 	help
-- 
2.33.1



^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH] mm: Disable NUMA_BALANCING_DEFAULT_ENABLED and TRANSPARENT_HUGEPAGE on PREEMPT_RT
  2021-10-28 14:33 [PATCH] mm: Disable NUMA_BALANCING_DEFAULT_ENABLED and TRANSPARENT_HUGEPAGE on PREEMPT_RT Sebastian Andrzej Siewior
@ 2021-10-28 14:37 ` Mel Gorman
  2021-10-28 15:54 ` David Hildenbrand
  1 sibling, 0 replies; 4+ messages in thread
From: Mel Gorman @ 2021-10-28 14:37 UTC (permalink / raw)
  To: Sebastian Andrzej Siewior
  Cc: Andrew Morton, linux-mm, Vlastimil Babka, Peter Zijlstra,
	Thomas Gleixner

On Thu, Oct 28, 2021 at 04:33:27PM +0200, Sebastian Andrzej Siewior wrote:
> TRANSPARENT_HUGEPAGE:
> There are potential non-deterministic delays to an RT thread if a critical
> memory region is not THP-aligned and a non-RT buffer is located in the same
> hugepage-aligned region. It's also possible for an unrelated thread to migrate
> pages belonging to an RT task incurring unexpected page faults due to memory
> defragmentation even if khugepaged is disabled.
> 
> Regular HUGEPAGEs are not affected by this can be used.
> 
> NUMA_BALANCING:
> There is a non-deterministic delay to mark PTEs PROT_NONE to gather NUMA fault
> samples, increased page faults of regions even if mlocked and non-deterministic
> delays when migrating pages.
> 
> [Mel Gorman worded 99% of the commit description].
> 
> Link: https://lore.kernel.org/all/20200304091159.GN3818@techsingularity.net/
> Link: https://lore.kernel.org/all/20211026165100.ahz5bkx44lrrw5pt@linutronix.de/
> Cc: Mel Gorman <mgorman@techsingularity.net>
> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>

Acked-by: Mel Gorman <mgorman@techsingularity.net>

-- 
Mel Gorman
SUSE Labs


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] mm: Disable NUMA_BALANCING_DEFAULT_ENABLED and TRANSPARENT_HUGEPAGE on PREEMPT_RT
  2021-10-28 14:33 [PATCH] mm: Disable NUMA_BALANCING_DEFAULT_ENABLED and TRANSPARENT_HUGEPAGE on PREEMPT_RT Sebastian Andrzej Siewior
  2021-10-28 14:37 ` Mel Gorman
@ 2021-10-28 15:54 ` David Hildenbrand
  2021-10-28 16:17   ` Sebastian Andrzej Siewior
  1 sibling, 1 reply; 4+ messages in thread
From: David Hildenbrand @ 2021-10-28 15:54 UTC (permalink / raw)
  To: Sebastian Andrzej Siewior, Andrew Morton, Mel Gorman
  Cc: linux-mm, Vlastimil Babka, Peter Zijlstra, Thomas Gleixner

On 28.10.21 16:33, Sebastian Andrzej Siewior wrote:
> TRANSPARENT_HUGEPAGE:
> There are potential non-deterministic delays to an RT thread if a critical
> memory region is not THP-aligned and a non-RT buffer is located in the same
> hugepage-aligned region. It's also possible for an unrelated thread to migrate
> pages belonging to an RT task incurring unexpected page faults due to memory
> defragmentation even if khugepaged is disabled.
> 
> Regular HUGEPAGEs are not affected by this can be used.
> 
> NUMA_BALANCING:
> There is a non-deterministic delay to mark PTEs PROT_NONE to gather NUMA fault
> samples, increased page faults of regions even if mlocked and non-deterministic
> delays when migrating pages.
> 
> [Mel Gorman worded 99% of the commit description].
> 
> Link: https://lore.kernel.org/all/20200304091159.GN3818@techsingularity.net/
> Link: https://lore.kernel.org/all/20211026165100.ahz5bkx44lrrw5pt@linutronix.de/
> Cc: Mel Gorman <mgorman@techsingularity.net>
> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
> ---
>  init/Kconfig | 2 +-
>  mm/Kconfig   | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/init/Kconfig b/init/Kconfig
> index edc0a0228f143..f83e7a0478f7c 100644
> --- a/init/Kconfig
> +++ b/init/Kconfig
> @@ -911,7 +911,7 @@ config NUMA_BALANCING
>  	bool "Memory placement aware NUMA scheduler"
>  	depends on ARCH_SUPPORTS_NUMA_BALANCING
>  	depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
> -	depends on SMP && NUMA && MIGRATION
> +	depends on SMP && NUMA && MIGRATION && !PREEMPT_RT
>  	help
>  	  This option adds support for automatic NUMA aware memory/task placement.
>  	  The mechanism is quite primitive and is based on migrating memory when
> diff --git a/mm/Kconfig b/mm/Kconfig
> index c150a0c6fce2c..5c5508fafcec5 100644
> --- a/mm/Kconfig
> +++ b/mm/Kconfig
> @@ -374,7 +374,7 @@ config NOMMU_INITIAL_TRIM_EXCESS
>  
>  config TRANSPARENT_HUGEPAGE
>  	bool "Transparent Hugepage Support"
> -	depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE
> +	depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE && !PREEMPT_RT
>  	select COMPACTION
>  	select XARRAY_MULTI
>  	help
> 

Makes perfect sense to me. I was also wondering how "dangerous"
MIGRATION (and everything depending on it, like CMA, compaction,
MEMORY_HOTREMOVE, ...) is for RT recently.

Reviewed-by: David Hildenbrand <david@redhat.com>

-- 
Thanks,

David / dhildenb



^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] mm: Disable NUMA_BALANCING_DEFAULT_ENABLED and TRANSPARENT_HUGEPAGE on PREEMPT_RT
  2021-10-28 15:54 ` David Hildenbrand
@ 2021-10-28 16:17   ` Sebastian Andrzej Siewior
  0 siblings, 0 replies; 4+ messages in thread
From: Sebastian Andrzej Siewior @ 2021-10-28 16:17 UTC (permalink / raw)
  To: David Hildenbrand
  Cc: Andrew Morton, Mel Gorman, linux-mm, Vlastimil Babka,
	Peter Zijlstra, Thomas Gleixner

On 2021-10-28 17:54:01 [+0200], David Hildenbrand wrote:
> Makes perfect sense to me. I was also wondering how "dangerous"
> MIGRATION (and everything depending on it, like CMA, compaction,
> MEMORY_HOTREMOVE, ...) is for RT recently.

for compatcion there is
  6923aa0d8c629 ("mm/compaction: Disable compact_unevictable_allowed on RT")

to honour mlock() otherwise it can end unpleasant. But yeah, any page
migrations outside of the "setup/ init" phase of the RT task are bad.

Sebastian


^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2021-10-28 16:17 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-28 14:33 [PATCH] mm: Disable NUMA_BALANCING_DEFAULT_ENABLED and TRANSPARENT_HUGEPAGE on PREEMPT_RT Sebastian Andrzej Siewior
2021-10-28 14:37 ` Mel Gorman
2021-10-28 15:54 ` David Hildenbrand
2021-10-28 16:17   ` Sebastian Andrzej Siewior

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.