arm64: reduce section size for sparsemem

PATCH In review Sudarshan Rajagopalan 3 participants 6 messages Jan 8, 2021 – Jan 21, 2021 arm64

Share & tools

Sudarshan Rajagopalan

author

2021-01-08 23:17 UTC #0 lore raw

cover arm64: reduce section size for sparsemem

This patch is the follow-up from the discussions in the thread [1]. Reducing the section size has the merit of reducing wastage of reserved memory for huge memory holes in sparsemem model. Also with smaller section size gives more grunularity and agility for memory hot(un)plugging.

This patch tends to use the suggestion from David Hildenbrand in thread [1] to set the least possible SECTION_SIZE_BITS for 4K, 16K and 64K page granule. That is 27 (128MB) for 4K/16K and 29 (512MB) for 64K page granule.

[1] https://lore.kernel.org/lkml/[email protected]/T/#m8ee60ae69db5e9eb06ca7999c43828d49ccb9626

Sudarshan Rajagopalan (1):
arm64: reduce section size for sparsemem

 arch/arm64/include/asm/sparsemem.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

Signature · 2 lines

Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

Sudarshan Rajagopalan

author

2021-01-08 23:17 UTC #1 lore raw

PATCH 1/1 arm64: reduce section size for sparsemem

Reducing the section size helps reduce wastage of reserved memory for huge memory holes in sparsemem model. But having a much smaller section size bits could break PMD mappings for vmemmap and wouldn't accomodate the highest order page for certain page size granule configs. It is determined that SECTION_SIZE_BITS of 27 (128MB) could be ideal default value for 4K_PAGES that gives least section size without breaking PMD based vmemmap mappings. For simplicity, 16K_PAGES could follow the same as 4K_PAGES. And the least SECTION_SIZE_BITS for 64K_PAGES is 29 that could accomodate MAX_ORDER.

Signed-off-by Sudarshan Rajagopalan <[email protected]>

Cc Will Deacon <[email protected]>

Cc Anshuman Khandual <[email protected]>

Cc Mike Rapoport <[email protected]>

Cc Mark Rutland <[email protected]>

Cc Suren Baghdasaryan <[email protected]>

---
 arch/arm64/include/asm/sparsemem.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

arch/arm64/include/asm/sparsemem.h +8 -2

index 1f43fcc79738..ff08ff6b677c 100644
@@ -7,7 +7,13 @@
 
 #ifdef CONFIG_SPARSEMEM
 #define MAX_PHYSMEM_BITS	CONFIG_ARM64_PA_BITS
-#define SECTION_SIZE_BITS	30
-#endif
+
+#if defined(CONFIG_ARM64_4K_PAGES) || defined(CONFIG_ARM64_16K_PAGES)
+#define SECTION_SIZE_BITS 27
+#else
+#define SECTION_SIZE_BITS 29
+#endif /* CONFIG_ARM64_4K_PAGES || CONFIG_ARM64_16K_PAGES */
+
+#endif /* CONFIG_SPARSEMEM*/
 
 #endif

Signature · 2 lines

Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

Anshuman Khandual

2021-01-11 11:09 UTC #2 lore raw

+ Catalin

Hello Sudershan,

Could you please change the subject line above as follows for better classifications and clarity.

arm64/sparsemem: Reduce SECTION_SIZE_BITS

Quoted 2 lines · Sudarshan Rajagopalan

Reducing the section size helps reduce wastage of reserved memory for huge memory holes in sparsemem model. But having a much smaller

There are two distinct benefits of reducing SECTION_SIZE_BITS.

- Improve memory hotplug granularity
- Reduce reserved memory wastage for vmmemmap mappings for sections with large memory holes

Quoted 2 lines

section size bits could break PMD mappings for vmemmap and wouldn't accomodate the highest order page for certain page size granule configs.

There are constrains in reducing SECTION_SIZE_BIT like

- Should accommodate highest order page for a given config
- Should not break PMD mapping in vmemmap for 4K pages
- Should not consume too many page->flags bits reducing space for other info

Both benefits and constraints should be described in the commit message for folks to understand the rationale clearly at a later point in time.

Quoted 1 line

It is determined that SECTION_SIZE_BITS of 27 (128MB) could be ideal

Probably needs some description how we arrived here.

Quoted 4 lines

default value for 4K_PAGES that gives least section size without breaking PMD based vmemmap mappings. For simplicity, 16K_PAGES could follow the same as 4K_PAGES. And the least SECTION_SIZE_BITS for 64K_PAGES is 29 that could accomodate MAX_ORDER.

Did not see this patch earlier and hence ended up writing yet another one. Here is the draft commit message from that patch, please feel free to use in part or full. But please do include the benefits, the constraints and the rationale for arriving at these figures.

---------
memory_block_size_bytes() determines the memory hotplug granularity i.e the amount of memory which can be hot added or hot removed from the kernel. The generic value here being MIN_MEMORY_BLOCK_SIZE (1UL << SECTION_SIZE_BITS) for memory_block_size_bytes() on platforms like arm64 that does not override.

Current SECTION_SIZE_BITS is 30 i.e 1GB which is large and a reduction here increases memory hotplug granularity, thus improving its agility. A reduced section size also reduces memory wastage in vmemmmap mapping for sections with large memory holes. A section size bits selection must follow.

(MAX_ORDER - 1 + PAGE_SHIFT) <= SECTION_SIZE_BITS

CONFIG_FORCE_MAX_ZONEORDER is always defined on arm64 and just following it would help achieve the smallest section size.

SECTION_SIZE_BITS = (CONFIG_FORCE_MAX_ZONEORDER - 1 + PAGE_SHIFT)

SECTION_SIZE_BITS = 22 (11 - 1 + 12) i.e 4MB for 4K pages
SECTION_SIZE_BITS = 24 (11 - 1 + 14) i.e 16MB for 16K pages without THP SECTION_SIZE_BITS = 25 (12 - 1 + 14) i.e 32MB for 16K pages with THP SECTION_SIZE_BITS = 26 (11 - 1 + 16) i.e 64MB for 64K pages without THP SECTION_SIZE_BITS = 29 (14 - 1 + 16) i.e 512MB for 64K pages with THP

But there are other problems. Reducing the section size too much would over populate /sys/devices/system/memory/ and also consume too many page->flags bits in the !vmemmap case. Also section size needs to be multiple of 128MB to have PMD based vmemmap mapping with CONFIG_ARM64_4K_PAGES.

Given these constraints, lets just reduce the section size to 128MB for 4K and 16K base page size configs and to 512MB for 64K base page size config.
-------------

Quoted 8 lines

Signed-off-by Sudarshan Rajagopalan <[email protected]>
Suggested-by David Hildenbrand <[email protected]>
Cc Will Deacon <[email protected]>
Cc Anshuman Khandual <[email protected]>
Cc Mike Rapoport <[email protected]>
Cc Mark Rutland <[email protected]>
Cc Suren Baghdasaryan <[email protected]>

A nit. Please add all relevant mailing lists like LAKML, MM along with other developers here in the CC list, so that it would never be missed.

Quoted 9 lines

---
 arch/arm64/include/asm/sparsemem.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

arch/arm64/include/asm/sparsemem.h +0 -0

index 1f43fcc79738..ff08ff6b677c 100644

arch/arm64/include/asm/sparsemem.h line 11

Quoted 7 lines

arch/arm64/include/asm/sparsemem.h +2 -2

@@ -7,7 +7,13 @@
 
 #ifdef CONFIG_SPARSEMEM
 #define MAX_PHYSMEM_BITS	CONFIG_ARM64_PA_BITS
-#define SECTION_SIZE_BITS	30
-#endif
+
+#if defined(CONFIG_ARM64_4K_PAGES) || defined(CONFIG_ARM64_16K_PAGES)

Please add a comment, something like

/*
* Section size must be at least 128MB for 4K base
* page size config. Otherwise PMD based huge page
* entries could not be created for vmemmap mappings.
* 16K follows 4K for simplicity.
*/

arch/arm64/include/asm/sparsemem.h line 13

Quoted 2 lines

arch/arm64/include/asm/sparsemem.h +2 -0
+#define SECTION_SIZE_BITS 27
+#else

Please add a comment, something like

/*
* Section size must be at least 512MB for 64K base
* page size config. Otherwise it will be less than
* (MAX_ORDER - 1) and the build process will fail. */

arch/arm64/include/asm/sparsemem.h line 15

Quoted 2 lines

arch/arm64/include/asm/sparsemem.h +2 -0

+#define SECTION_SIZE_BITS 29
+#endif /* CONFIG_ARM64_4K_PAGES || CONFIG_ARM64_16K_PAGES */

This condition here can be flipped (#ifdef CONFIG_ARM64_64K_PAGES) to be bit cleaner and possibly to reduce the code.

arch/arm64/include/asm/sparsemem.h line 19

Quoted 5 lines

arch/arm64/include/asm/sparsemem.h +2 -0
+
+#endif /* CONFIG_SPARSEMEM*/
 
 #endif

- Anshuman

Sudarshan Rajagopalan

author

2021-01-21 05:39 UTC #3 lore raw

Quoted 151 lines · Anshuman Khandual

+ Catalin
Hello Sudershan,
Could you please change the subject line above as follows for better classifications and clarity.
arm64/sparsemem: Reduce SECTION_SIZE_BITS
Quoted 2 lines · Sudarshan Rajagopalan
Reducing the section size helps reduce wastage of reserved memory for huge memory holes in sparsemem model. But having a much smaller
There are two distinct benefits of reducing SECTION_SIZE_BITS.
- Improve memory hotplug granularity
- Reduce reserved memory wastage for vmmemmap mappings for sections with large memory holes
Quoted 2 lines
section size bits could break PMD mappings for vmemmap and wouldn't accomodate the highest order page for certain page size granule configs.
There are constrains in reducing SECTION_SIZE_BIT like
- Should accommodate highest order page for a given config
- Should not break PMD mapping in vmemmap for 4K pages
- Should not consume too many page->flags bits reducing space for other info
Both benefits and constraints should be described in the commit message for folks to understand the rationale clearly at a later point in time.
Quoted 1 line
It is determined that SECTION_SIZE_BITS of 27 (128MB) could be ideal
Probably needs some description how we arrived here.
Quoted 4 lines
default value for 4K_PAGES that gives least section size without breaking PMD based vmemmap mappings. For simplicity, 16K_PAGES could follow the same as 4K_PAGES. And the least SECTION_SIZE_BITS for 64K_PAGES is 29 that could accomodate MAX_ORDER.
Did not see this patch earlier and hence ended up writing yet another one. Here is the draft commit message from that patch, please feel free to use in part or full. But please do include the benefits, the constraints and the rationale for arriving at these figures.
---------
memory_block_size_bytes() determines the memory hotplug granularity
i.e the
amount of memory which can be hot added or hot removed from the
kernel. The
generic value here being MIN_MEMORY_BLOCK_SIZE (1UL <<
SECTION_SIZE_BITS)
for memory_block_size_bytes() on platforms like arm64 that does
not override.
Current SECTION_SIZE_BITS is 30 i.e 1GB which is large and a
reduction here
increases memory hotplug granularity, thus improving its agility. A
reduced
section size also reduces memory wastage in vmemmmap mapping for
sections
with large memory holes. A section size bits selection must follow.
(MAX_ORDER - 1 + PAGE_SHIFT) <= SECTION_SIZE_BITS
CONFIG_FORCE_MAX_ZONEORDER is always defined on arm64 and just
following it
would help achieve the smallest section size.
SECTION_SIZE_BITS = (CONFIG_FORCE_MAX_ZONEORDER - 1 + PAGE_SHIFT)
SECTION_SIZE_BITS = 22 (11 - 1 + 12) i.e 4MB for 4K pages
SECTION_SIZE_BITS = 24 (11 - 1 + 14) i.e 16MB for 16K pages
without THP
    SECTION_SIZE_BITS = 25 (12 - 1 + 14) i.e 32MB  for 16K pages with 
THP
    SECTION_SIZE_BITS = 26 (11 - 1 + 16) i.e 64MB  for 64K pages 
without THP
    SECTION_SIZE_BITS = 29 (14 - 1 + 16) i.e 512MB for 64K pages with 
THP
But there are other problems. Reducing the section size too much
would over
populate /sys/devices/system/memory/ and also consume too many
page->flags
bits in the !vmemmap case. Also section size needs to be multiple
of 128MB
to have PMD based vmemmap mapping with CONFIG_ARM64_4K_PAGES.
Given these constraints, lets just reduce the section size to 128MB
for 4K
and 16K base page size configs and to 512MB for 64K base page size
config.
    -------------
Quoted 8 lines
Signed-off-by Sudarshan Rajagopalan <[email protected]>
Suggested-by David Hildenbrand <[email protected]>
Cc Will Deacon <[email protected]>
Cc Anshuman Khandual <[email protected]>
Cc Mike Rapoport <[email protected]>
Cc Mark Rutland <[email protected]>
Cc Suren Baghdasaryan <[email protected]>
A nit. Please add all relevant mailing lists like LAKML, MM along with other developers here in the CC list, so that it would never be missed.
Quoted 9 lines
---
 arch/arm64/include/asm/sparsemem.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
arch/arm64/include/asm/sparsemem.h +0 -0
index 1f43fcc79738..ff08ff6b677c 100644
arch/arm64/include/asm/sparsemem.h line 11
Quoted 7 lines
arch/arm64/include/asm/sparsemem.h +2 -2
@@ -7,7 +7,13 @@
 
 #ifdef CONFIG_SPARSEMEM
 #define MAX_PHYSMEM_BITS	CONFIG_ARM64_PA_BITS
-#define SECTION_SIZE_BITS	30
-#endif
+
+#if defined(CONFIG_ARM64_4K_PAGES) || defined(CONFIG_ARM64_16K_PAGES)
Please add a comment, something like
/*
* Section size must be at least 128MB for 4K base
* page size config. Otherwise PMD based huge page
* entries could not be created for vmemmap mappings.
* 16K follows 4K for simplicity.
*/
arch/arm64/include/asm/sparsemem.h line 13
Quoted 2 lines
arch/arm64/include/asm/sparsemem.h +2 -0
+#define SECTION_SIZE_BITS 27
+#else
Please add a comment, something like
/*
* Section size must be at least 512MB for 64K base
* page size config. Otherwise it will be less than
* (MAX_ORDER - 1) and the build process will fail. */
arch/arm64/include/asm/sparsemem.h line 15
Quoted 2 lines
arch/arm64/include/asm/sparsemem.h +2 -0
+#define SECTION_SIZE_BITS 29
+#endif /* CONFIG_ARM64_4K_PAGES || CONFIG_ARM64_16K_PAGES */
This condition here can be flipped (#ifdef CONFIG_ARM64_64K_PAGES) to be bit cleaner and possibly to reduce the code.

Quoted 7 lines

arch/arm64/include/asm/sparsemem.h line 19
Quoted 5 lines
arch/arm64/include/asm/sparsemem.h +2 -0
+
+#endif /* CONFIG_SPARSEMEM*/
 
 #endif
- Anshuman

Thanks Anshuman for the comments and suggestions. I have used the excerpt you had provided before for the commit message with slight modifications. I have pushed the new patch with your comments addressed -
https://lore.kernel.org/lkml/43843c5e092bfe3ec4c41e3c8c78a7ee35b69bb0.1611206601.git.sudaraja@codeaurora.org/

Sudarshan

--
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, a Linux Foundation Collaborative Project

Will Deacon

maintainer

2021-01-20 17:52 UTC #4 lore raw

In reply to Sudarshan Rajagopalan

Quoted 29 lines · Sudarshan Rajagopalan

Reducing the section size helps reduce wastage of reserved memory for huge memory holes in sparsemem model. But having a much smaller section size bits could break PMD mappings for vmemmap and wouldn't accomodate the highest order page for certain page size granule configs. It is determined that SECTION_SIZE_BITS of 27 (128MB) could be ideal default value for 4K_PAGES that gives least section size without breaking PMD based vmemmap mappings. For simplicity, 16K_PAGES could follow the same as 4K_PAGES. And the least SECTION_SIZE_BITS for 64K_PAGES is 29 that could accomodate MAX_ORDER.
Signed-off-by Sudarshan Rajagopalan <[email protected]>
Suggested-by David Hildenbrand <[email protected]>
Cc Will Deacon <[email protected]>
Cc Anshuman Khandual <[email protected]>
Cc Mike Rapoport <[email protected]>
Cc Mark Rutland <[email protected]>
Cc Suren Baghdasaryan <[email protected]>
---
 arch/arm64/include/asm/sparsemem.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
arch/arm64/include/asm/sparsemem.h +0 -0
index 1f43fcc79738..ff08ff6b677c 100644
@@ -7,7 +7,13 @@
 
 #ifdef CONFIG_SPARSEMEM
 #define MAX_PHYSMEM_BITS	CONFIG_ARM64_PA_BITS

arch/arm64/include/asm/sparsemem.h line 17

Quoted 10 lines

arch/arm64/include/asm/sparsemem.h +8 -2

-#define SECTION_SIZE_BITS	30
-#endif
+
+#if defined(CONFIG_ARM64_4K_PAGES) || defined(CONFIG_ARM64_16K_PAGES)
+#define SECTION_SIZE_BITS 27
+#else
+#define SECTION_SIZE_BITS 29
+#endif /* CONFIG_ARM64_4K_PAGES || CONFIG_ARM64_16K_PAGES */
+
+#endif /* CONFIG_SPARSEMEM*/

Please can you repost this in light of the comments from Anshuman?

Thanks,

Will

_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Sudarshan Rajagopalan

author

2021-01-20 19:23 UTC #5 lore raw

Quoted 43 lines · Will Deacon

Quoted 29 lines · Sudarshan Rajagopalan
Reducing the section size helps reduce wastage of reserved memory for huge memory holes in sparsemem model. But having a much smaller section size bits could break PMD mappings for vmemmap and wouldn't accomodate the highest order page for certain page size granule configs. It is determined that SECTION_SIZE_BITS of 27 (128MB) could be ideal default value for 4K_PAGES that gives least section size without breaking PMD based vmemmap mappings. For simplicity, 16K_PAGES could follow the same as 4K_PAGES. And the least SECTION_SIZE_BITS for 64K_PAGES is 29 that could accomodate MAX_ORDER.
Signed-off-by Sudarshan Rajagopalan <[email protected]>
Suggested-by David Hildenbrand <[email protected]>
Cc Will Deacon <[email protected]>
Cc Anshuman Khandual <[email protected]>
Cc Mike Rapoport <[email protected]>
Cc Mark Rutland <[email protected]>
Cc Suren Baghdasaryan <[email protected]>
---
 arch/arm64/include/asm/sparsemem.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
arch/arm64/include/asm/sparsemem.h +0 -0
index 1f43fcc79738..ff08ff6b677c 100644
@@ -7,7 +7,13 @@
 
 #ifdef CONFIG_SPARSEMEM
 #define MAX_PHYSMEM_BITS	CONFIG_ARM64_PA_BITS
arch/arm64/include/asm/sparsemem.h line 17
Quoted 10 lines
arch/arm64/include/asm/sparsemem.h +8 -2
-#define SECTION_SIZE_BITS	30
-#endif
+
+#if defined(CONFIG_ARM64_4K_PAGES) || defined(CONFIG_ARM64_16K_PAGES)
+#define SECTION_SIZE_BITS 27
+#else
+#define SECTION_SIZE_BITS 29
+#endif /* CONFIG_ARM64_4K_PAGES || CONFIG_ARM64_16K_PAGES */
+
+#endif /* CONFIG_SPARSEMEM*/
Please can you repost this in light of the comments from Anshuman?

Quoted 3 lines

Thanks,
Will

Sure Will. We were held up with some other critical tasks.. will repost the patch by EOD after addressing Anshuman's comments.

Signature · 4 lines

Sudarshan

--
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, a Linux Foundation Collaborative Project