Skip to contentSkip to conversation

mm/migrate_device: Try to handle swapcache pages.

PATCH Superseded [email protected] 2 participants 5 messages May 7, 2023 – May 8, 2023

Subsystems (2)

v1 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 mm/migrate_device: Try to handle swapcache pages.

From: Mika Penttilä <[email protected]>

Migrating file pages and swapcache pages into device memory is not supported. The decision is done based on page_mapping(). For now, swapcache pages are not migrated.

Things can however be improved, for swapcache pages. Try to get rid of the swap cache, and if successful, go ahead as with other anonymous pages.

As a detail, do not mess with shmem pages, as they handle swap internally.

Cc Alistair Popple <[email protected]>
Cc John Hubbard <[email protected]>
Cc Ralph Campbell <[email protected]>
---
 mm/migrate_device.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)
mm/migrate_device.c +10 -2
index d30c9de60b0d..e8169c58c341 100644
@@ -12,6 +12,7 @@
 #include <linux/mmu_notifier.h>
 #include <linux/oom.h>
 #include <linux/pagewalk.h>
+#include <linux/shmem_fs.h>
 #include <linux/rmap.h>
 #include <linux/swapops.h>
 #include <asm/tlbflush.h>
@@ -750,10 +751,17 @@ static void __migrate_device_pages(unsigned long *src_pfns,
 			/*
 			 * For now only support anonymous memory migrating to
 			 * device private or coherent memory.
+			 *
+			 * Try to get rid of swap cache if possible.
+			 * Leave shmem pages alone, they handle swap internally
+			 *
 			 */
 			if (mapping) {
-				src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
-				continue;
+				if (shmem_mapping(mapping) ||
+				    !folio_free_swap(page_folio(page))) {
+					src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
+					continue;
+				}
 			}
 		} else if (is_zone_device_page(newpage)) {
 			/*
Signature · 1 line
2.17.1
#1 lore raw
Quoted 34 lines · [email protected] writes:

From: Mika Penttilä <[email protected]>

Migrating file pages and swapcache pages into device memory is not supported. The decision is done based on page_mapping(). For now, swapcache pages are not migrated.

Things can however be improved, for swapcache pages. Try to get rid of the swap cache, and if successful, go ahead as with other anonymous pages.

As a detail, do not mess with shmem pages, as they handle swap internally.

Cc Alistair Popple <[email protected]>
Cc John Hubbard <[email protected]>
Cc Ralph Campbell <[email protected]>
---
 mm/migrate_device.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)
mm/migrate_device.c +1 -0
index d30c9de60b0d..e8169c58c341 100644
@@ -12,6 +12,7 @@
 #include <linux/mmu_notifier.h>
 #include <linux/oom.h>
 #include <linux/pagewalk.h>
+#include <linux/shmem_fs.h>
 #include <linux/rmap.h>
 #include <linux/swapops.h>
 #include <asm/tlbflush.h>
@@ -750,10 +751,17 @@ static void __migrate_device_pages(unsigned long *src_pfns,
 			/*
 			 * For now only support anonymous memory migrating to
 			 * device private or coherent memory.

mm/migrate_device.c static void __migrate_device_pages(unsigned long *src_pfns, · line 761

Quoted 10 lines
mm/migrate_device.c +6 -2
+			 *
+			 * Try to get rid of swap cache if possible.
+			 * Leave shmem pages alone, they handle swap internally
+			 *
 			 */
 			if (mapping) {
-				src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
-				continue;
+				if (shmem_mapping(mapping) ||
+				    !folio_free_swap(page_folio(page))) {

Should we filter out file pages firstly?

mm/migrate_device.c static void __migrate_device_pages(unsigned long *src_pfns, · line 767

Quoted 6 lines
mm/migrate_device.c +3 -0
+					src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
+					continue;
+				}
 			}
 		} else if (is_zone_device_page(newpage)) {
 			/*

Best Regards,
Huang, Ying

#2 lore raw

H,

Quoted 40 lines · Ying
Quoted 38 lines · [email protected] writes:

From: Mika Penttilä <[email protected]>

Migrating file pages and swapcache pages into device memory is not supported. The decision is done based on page_mapping(). For now, swapcache pages are not migrated.

Things can however be improved, for swapcache pages. Try to get rid of the swap cache, and if successful, go ahead as with other anonymous pages.

As a detail, do not mess with shmem pages, as they handle swap internally.

Cc Alistair Popple <[email protected]>
Cc John Hubbard <[email protected]>
Cc Ralph Campbell <[email protected]>
---
  mm/migrate_device.c | 12 ++++++++++--
  1 file changed, 10 insertions(+), 2 deletions(-)
mm/migrate_device.c +5 -0
index d30c9de60b0d..e8169c58c341 100644
@@ -12,6 +12,7 @@
  #include <linux/mmu_notifier.h>
  #include <linux/oom.h>
  #include <linux/pagewalk.h>
+#include <linux/shmem_fs.h>
  #include <linux/rmap.h>
  #include <linux/swapops.h>
  #include <asm/tlbflush.h>
@@ -750,10 +751,17 @@ static void __migrate_device_pages(unsigned long *src_pfns,
  			/*
  			 * For now only support anonymous memory migrating to
  			 * device private or coherent memory.
+			 *
+			 * Try to get rid of swap cache if possible.
+			 * Leave shmem pages alone, they handle swap internally
+			 *
Quoted 8 lines

mm/migrate_device.c static void __migrate_device_pages(unsigned long *src_pfns, · line 761

Quoted 6 lines
mm/migrate_device.c +2 -2
  			 */
  			if (mapping) {
-				src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
-				continue;
+				if (shmem_mapping(mapping) ||
+				    !folio_free_swap(page_folio(page))) {

Should we filter out file pages firstly?

folio_free_swap() checks first and returns false if not swapcache page (ie including normal file pages) so think that is enough, but maybe for clarity and not depending on it, could be good explicitly check for swapcache before calling folio_free_swap().

Quoted 11 lines

mm/migrate_device.c static void __migrate_device_pages(unsigned long *src_pfns, · line 767

Quoted 7 lines
mm/migrate_device.c +3 -0
+					src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
+					continue;
+				}
  			}
  		} else if (is_zone_device_page(newpage)) {
  			/*

Best Regards,
Huang, Ying

Thanks,
Mika

#3 lore raw
Quoted 52 lines · Mika Penttilä

H,

Quoted 39 lines · Ying
Quoted 37 lines · [email protected] writes:

From: Mika Penttil <[email protected]>

Migrating file pages and swapcache pages into device memory is not supported. The decision is done based on page_mapping(). For now, swapcache pages are not migrated.

Things can however be improved, for swapcache pages. Try to get rid of the swap cache, and if successful, go ahead as with other anonymous pages.

As a detail, do not mess with shmem pages, as they handle swap internally.

Cc Alistair Popple <[email protected]>
Cc John Hubbard <[email protected]>
Cc Ralph Campbell <[email protected]>
---
  mm/migrate_device.c | 12 ++++++++++--
  1 file changed, 10 insertions(+), 2 deletions(-)
mm/migrate_device.c +4 -0
index d30c9de60b0d..e8169c58c341 100644
@@ -12,6 +12,7 @@
  #include <linux/mmu_notifier.h>
  #include <linux/oom.h>
  #include <linux/pagewalk.h>
+#include <linux/shmem_fs.h>
  #include <linux/rmap.h>
  #include <linux/swapops.h>
  #include <asm/tlbflush.h>
@@ -750,10 +751,17 @@ static void __migrate_device_pages(unsigned long *src_pfns,
  			/*
  			 * For now only support anonymous memory migrating to
  			 * device private or coherent memory.
+			 *
+			 * Try to get rid of swap cache if possible.
+			 * Leave shmem pages alone, they handle swap internally
Quoted 8 lines

mm/migrate_device.c static void __migrate_device_pages(unsigned long *src_pfns, · line 761

Quoted 7 lines
mm/migrate_device.c +3 -2
+			 *
  			 */
  			if (mapping) {
-				src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
-				continue;
+				if (shmem_mapping(mapping) ||
+				    !folio_free_swap(page_folio(page))) {

Should we filter out file pages firstly?

Quoted 2 lines

folio_free_swap() checks first and returns false if not swapcache page (ie including normal file pages) so think that is enough,

You are right. folio_free_swap() will call folio_test_swapcache(), and which will check folio_test_swapbacked().

And, IIUC, we don't need to check shmem too. Because folio_free_swap() will not return true for shmem pages.

Quoted 2 lines

but maybe for clarity and not depending on it, could be good explicitly check for swapcache before calling folio_free_swap().

Yes. That will be more clear. I suggest to use folio_test_anon(), that will filter out shmem pages too.

Quoted 7 lines
Quoted 7 lines

mm/migrate_device.c static void __migrate_device_pages(unsigned long *src_pfns, · line 767

Quoted 7 lines
mm/migrate_device.c +3 -0
+					src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
+					continue;
+				}
  			}
  		} else if (is_zone_device_page(newpage)) {
  			/*

Best Regards,
Huang, Ying

#4 lore raw

Hi,

Quoted 54 lines · Ying
Quoted 52 lines · Mika Penttilä

H,

Quoted 47 lines · Ying
Quoted 34 lines · [email protected] writes:

From: Mika Penttil <[email protected]>

Migrating file pages and swapcache pages into device memory is not supported. The decision is done based on page_mapping(). For now, swapcache pages are not migrated.

Things can however be improved, for swapcache pages. Try to get rid of the swap cache, and if successful, go ahead as with other anonymous pages.

As a detail, do not mess with shmem pages, as they handle swap internally.

Cc Alistair Popple <[email protected]>
Cc John Hubbard <[email protected]>
Cc Ralph Campbell <[email protected]>
---
   mm/migrate_device.c | 12 ++++++++++--
   1 file changed, 10 insertions(+), 2 deletions(-)
mm/migrate_device.c +1 -0
index d30c9de60b0d..e8169c58c341 100644
@@ -12,6 +12,7 @@
   #include <linux/mmu_notifier.h>
   #include <linux/oom.h>
   #include <linux/pagewalk.h>
+#include <linux/shmem_fs.h>
   #include <linux/rmap.h>
   #include <linux/swapops.h>
   #include <asm/tlbflush.h>
@@ -750,10 +751,17 @@ static void __migrate_device_pages(unsigned long *src_pfns,
   			/*
   			 * For now only support anonymous memory migrating to
   			 * device private or coherent memory.

mm/migrate_device.c static void __migrate_device_pages(unsigned long *src_pfns, · line 761

Quoted 10 lines
mm/migrate_device.c +6 -2
+			 *
+			 * Try to get rid of swap cache if possible.
+			 * Leave shmem pages alone, they handle swap internally
+			 *
   			 */
   			if (mapping) {
-				src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
-				continue;
+				if (shmem_mapping(mapping) ||
+				    !folio_free_swap(page_folio(page))) {

Should we filter out file pages firstly?

Quoted 8 lines
Quoted 2 lines

folio_free_swap() checks first and returns false if not swapcache page (ie including normal file pages) so think that is enough,

You are right. folio_free_swap() will call folio_test_swapcache(), and which will check folio_test_swapbacked().

And, IIUC, we don't need to check shmem too. Because folio_free_swap() will not return true for shmem pages.

Ah yes true, for not written out shmem page folio_free_swap() returns false (because !swapcache). And written out (but in swap cache) returns false because folio_swapped() == true

Quoted 6 lines
Quoted 3 lines

but maybe for clarity and not depending on it, could be good explicitly check for swapcache before calling folio_free_swap().

Yes. That will be more clear. I suggest to use folio_test_anon(), that will filter out shmem pages too.

Agreed.

Quoted 12 lines
Quoted 8 lines
Quoted 8 lines

mm/migrate_device.c static void __migrate_device_pages(unsigned long *src_pfns, · line 767

Quoted 8 lines
mm/migrate_device.c +3 -0
+					src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
+					continue;
+				}
   			}
   		} else if (is_zone_device_page(newpage)) {
   			/*

Best Regards,
Huang, Ying

Thanks,
Mika