linux-next: manual merge of the random tree with the mm-stable tree

linux-next: manual merge of the random tree with the mm-stable tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 3069 bytes --]

Hi all,

Today's linux-next merge of the random tree got a conflict in:

  mm/rmap.c

between commits:

  a2a1b9f670f3 ("mm/rmap: remove duplicated exit code in pagewalk loop")
  92493c3b92b1 ("mm: extend rmap flags arguments for folio_add_new_anon_rmap")
  db4fb1039a19 ("mm: use folio_add_new_anon_rmap() if folio_test_anon(folio)==false")

from the mm-stable tree and commit:

  e788371a45ac ("mm: add VM_DROPPABLE for designating always lazily freeable mappings")

from the random tree.

I fixed it up (I think - see below) and can carry the fix as
necessary. This is now fixed as far as linux-next is concerned, but any
non trivial conflicts should be mentioned to your upstream maintainer
when your tree is submitted for merging.  You may also want to consider
cooperating with the maintainer of the conflicting tree to minimise any
particularly complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc mm/rmap.c
index 8616308610b9,d873a3f06506..000000000000
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@@ -1394,27 -1384,22 +1394,27 @@@ void folio_add_anon_rmap_pmd(struct fol
   *
   * Like folio_add_anon_rmap_*() but must only be called on *new* folios.
   * This means the inc-and-test can be bypassed.
 - * The folio does not have to be locked.
 + * The folio doesn't necessarily need to be locked while it's exclusive
 + * unless two threads map it concurrently. However, the folio must be
 + * locked if it's shared.
   *
 - * If the folio is pmd-mappable, it is accounted as a THP.  As the folio
 - * is new, it's assumed to be mapped exclusively by a single process.
 + * If the folio is pmd-mappable, it is accounted as a THP.
   */
  void folio_add_new_anon_rmap(struct folio *folio, struct vm_area_struct *vma,
 -		unsigned long address)
 +		unsigned long address, rmap_t flags)
  {
 -	int nr = folio_nr_pages(folio);
 +	const int nr = folio_nr_pages(folio);
 +	const bool exclusive = flags & RMAP_EXCLUSIVE;
 +	int nr_pmdmapped = 0;
  
  	VM_WARN_ON_FOLIO(folio_test_hugetlb(folio), folio);
 +	VM_WARN_ON_FOLIO(!exclusive && !folio_test_locked(folio), folio);
  	VM_BUG_ON_VMA(address < vma->vm_start ||
  			address + (nr << PAGE_SHIFT) > vma->vm_end, vma);
 -	if (!(vma->vm_flags & VM_DROPPABLE))
 +
- 	if (!folio_test_swapbacked(folio))
++	if (!folio_test_swapbacked(folio) && !(vma->vm_flags & VM_DROPPABLE))
  		__folio_set_swapbacked(folio);
 -	__folio_set_anon(folio, vma, address, true);
 +	__folio_set_anon(folio, vma, address, exclusive);
  
  	if (likely(!folio_test_large(folio))) {
  		/* increment count (starts at -1) */
@@@ -1858,8 -1852,11 +1858,9 @@@ static bool try_to_unmap_one(struct fol
  				 * discarded. Remap the page to page table.
  				 */
  				set_pte_at(mm, address, pvmw.pte, pteval);
- 				folio_set_swapbacked(folio);
+ 				if (!(vma->vm_flags & VM_DROPPABLE))
+ 					folio_set_swapbacked(folio);
 -				ret = false;
 -				page_vma_mapped_walk_done(&pvmw);
 -				break;
 +				goto walk_abort;
  			}
  
  			if (swap_duplicate(entry) < 0) {

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

linux-next: manual merge of the random tree with the mm-stable tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 3222 bytes --]

Hi all,

Today's linux-next merge of the random tree got a conflict in:

  mm/rmap.c

between commits:

  26d21b18d971 ("mm/rmap: remove duplicated exit code in pagewalk loop")
  15bde4abab73 ("mm: extend rmap flags arguments for folio_add_new_anon_rmap")

from the mm-stable tree and commit:

  94beef29e110 ("mm: add MAP_DROPPABLE for designating always lazily freeable mappings")

from the random tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc mm/rmap.c
index 8616308610b9,1f9b5a9cb121..000000000000
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@@ -1394,27 -1384,26 +1394,30 @@@ void folio_add_anon_rmap_pmd(struct fol
   *
   * Like folio_add_anon_rmap_*() but must only be called on *new* folios.
   * This means the inc-and-test can be bypassed.
 - * The folio does not have to be locked.
 + * The folio doesn't necessarily need to be locked while it's exclusive
 + * unless two threads map it concurrently. However, the folio must be
 + * locked if it's shared.
   *
 - * If the folio is pmd-mappable, it is accounted as a THP.  As the folio
 - * is new, it's assumed to be mapped exclusively by a single process.
 + * If the folio is pmd-mappable, it is accounted as a THP.
   */
  void folio_add_new_anon_rmap(struct folio *folio, struct vm_area_struct *vma,
 -		unsigned long address)
 +		unsigned long address, rmap_t flags)
  {
 -	int nr = folio_nr_pages(folio);
 +	const int nr = folio_nr_pages(folio);
 +	const bool exclusive = flags & RMAP_EXCLUSIVE;
 +	int nr_pmdmapped = 0;
  
  	VM_WARN_ON_FOLIO(folio_test_hugetlb(folio), folio);
 +	VM_WARN_ON_FOLIO(!exclusive && !folio_test_locked(folio), folio);
  	VM_BUG_ON_VMA(address < vma->vm_start ||
  			address + (nr << PAGE_SHIFT) > vma->vm_end, vma);
- 
- 	if (!folio_test_swapbacked(folio))
+ 	/*
+ 	 * VM_DROPPABLE mappings don't swap; instead they're just dropped when
+ 	 * under memory pressure.
+ 	 */
 -	if (!(vma->vm_flags & VM_DROPPABLE))
++	if (!folio_test_swapbacked(folio) && !(vma->vm_flags & VM_DROPPABLE))
  		__folio_set_swapbacked(folio);
 -	__folio_set_anon(folio, vma, address, true);
 +	__folio_set_anon(folio, vma, address, exclusive);
  
  	if (likely(!folio_test_large(folio))) {
  		/* increment count (starts at -1) */
@@@ -1858,8 -1862,15 +1867,13 @@@ static bool try_to_unmap_one(struct fol
  				 * discarded. Remap the page to page table.
  				 */
  				set_pte_at(mm, address, pvmw.pte, pteval);
- 				folio_set_swapbacked(folio);
+ 				/*
+ 				 * Unlike MADV_FREE mappings, VM_DROPPABLE ones
+ 				 * never get swap backed on failure to drop.
+ 				 */
+ 				if (!(vma->vm_flags & VM_DROPPABLE))
+ 					folio_set_swapbacked(folio);
 -				ret = false;
 -				page_vma_mapped_walk_done(&pvmw);
 -				break;
 +				goto walk_abort;
  			}
  
  			if (swap_duplicate(entry) < 0) {

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the random tree with the mm-stable tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 3543 bytes --]

Hi all,

On Tue, 9 Jul 2024 17:40:53 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> 
> Today's linux-next merge of the random tree got a conflict in:
> 
>   mm/rmap.c
> 
> between commits:
> 
>   26d21b18d971 ("mm/rmap: remove duplicated exit code in pagewalk loop")
>   15bde4abab73 ("mm: extend rmap flags arguments for folio_add_new_anon_rmap")
> 
> from the mm-stable tree and commit:
> 
>   94beef29e110 ("mm: add MAP_DROPPABLE for designating always lazily freeable mappings")
> 
> from the random tree.
> 
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging.  You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.
> 
> diff --cc mm/rmap.c
> index 8616308610b9,1f9b5a9cb121..000000000000
> --- a/mm/rmap.c
> +++ b/mm/rmap.c
> @@@ -1394,27 -1384,26 +1394,30 @@@ void folio_add_anon_rmap_pmd(struct fol
>    *
>    * Like folio_add_anon_rmap_*() but must only be called on *new* folios.
>    * This means the inc-and-test can be bypassed.
>  - * The folio does not have to be locked.
>  + * The folio doesn't necessarily need to be locked while it's exclusive
>  + * unless two threads map it concurrently. However, the folio must be
>  + * locked if it's shared.
>    *
>  - * If the folio is pmd-mappable, it is accounted as a THP.  As the folio
>  - * is new, it's assumed to be mapped exclusively by a single process.
>  + * If the folio is pmd-mappable, it is accounted as a THP.
>    */
>   void folio_add_new_anon_rmap(struct folio *folio, struct vm_area_struct *vma,
>  -		unsigned long address)
>  +		unsigned long address, rmap_t flags)
>   {
>  -	int nr = folio_nr_pages(folio);
>  +	const int nr = folio_nr_pages(folio);
>  +	const bool exclusive = flags & RMAP_EXCLUSIVE;
>  +	int nr_pmdmapped = 0;
>   
>   	VM_WARN_ON_FOLIO(folio_test_hugetlb(folio), folio);
>  +	VM_WARN_ON_FOLIO(!exclusive && !folio_test_locked(folio), folio);
>   	VM_BUG_ON_VMA(address < vma->vm_start ||
>   			address + (nr << PAGE_SHIFT) > vma->vm_end, vma);
> - 
> - 	if (!folio_test_swapbacked(folio))
> + 	/*
> + 	 * VM_DROPPABLE mappings don't swap; instead they're just dropped when
> + 	 * under memory pressure.
> + 	 */
>  -	if (!(vma->vm_flags & VM_DROPPABLE))
> ++	if (!folio_test_swapbacked(folio) && !(vma->vm_flags & VM_DROPPABLE))
>   		__folio_set_swapbacked(folio);
>  -	__folio_set_anon(folio, vma, address, true);
>  +	__folio_set_anon(folio, vma, address, exclusive);
>   
>   	if (likely(!folio_test_large(folio))) {
>   		/* increment count (starts at -1) */
> @@@ -1858,8 -1862,15 +1867,13 @@@ static bool try_to_unmap_one(struct fol
>   				 * discarded. Remap the page to page table.
>   				 */
>   				set_pte_at(mm, address, pvmw.pte, pteval);
> - 				folio_set_swapbacked(folio);
> + 				/*
> + 				 * Unlike MADV_FREE mappings, VM_DROPPABLE ones
> + 				 * never get swap backed on failure to drop.
> + 				 */
> + 				if (!(vma->vm_flags & VM_DROPPABLE))
> + 					folio_set_swapbacked(folio);
>  -				ret = false;
>  -				page_vma_mapped_walk_done(&pvmw);
>  -				break;
>  +				goto walk_abort;
>   			}
>   
>   			if (swap_duplicate(entry) < 0) {

This is now a conflict between the random tree and Linus' tree.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the random tree with the mm-stable tree

[Jason A. Donenfeld]

Hi Stephen,

On Wed, Jul 24, 2024 at 09:15:54AM +1000, Stephen Rothwell wrote:
> This is now a conflict between the random tree and Linus' tree.

Not too big of a deal, I don't think; I'm just waiting on Linus to pull
the tag I sent last week. Andrew only sent mm this weekend, so I assume
pulling mine is slated for sometime soon after, now that Linus has
pulled the mm patches. So these conflicts should disappear shortly.

Regards,
Jason

linux-next: manual merge of the random tree with the mm-stable tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1294 bytes --]

Hi all,

Today's linux-next merge of the random tree got a conflict in:

  tools/testing/selftests/mm/Makefile

between commit:

  3a103b5315b7 ("selftest: mm: Test if hugepage does not get leaked during __bio_release_pages()")

from the mm-stable tree and commit:

  94beef29e110 ("mm: add MAP_DROPPABLE for designating always lazily freeable mappings")

from the random tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc tools/testing/selftests/mm/Makefile
index e1aa09ddaa3d,e3e5740e13e1..000000000000
--- a/tools/testing/selftests/mm/Makefile
+++ b/tools/testing/selftests/mm/Makefile
@@@ -75,7 -73,7 +75,8 @@@ TEST_GEN_FILES += ksm_functional_test
  TEST_GEN_FILES += mdwe_test
  TEST_GEN_FILES += hugetlb_fault_after_madv
  TEST_GEN_FILES += hugetlb_madv_vs_map
 +TEST_GEN_FILES += hugetlb_dio
+ TEST_GEN_FILES += droppable
  
  ifneq ($(ARCH),arm64)
  TEST_GEN_FILES += soft-dirty

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the random tree with the mm-stable tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1515 bytes --]

Hi all,

On Tue, 9 Jul 2024 17:43:40 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> 
> Today's linux-next merge of the random tree got a conflict in:
> 
>   tools/testing/selftests/mm/Makefile
> 
> between commit:
> 
>   3a103b5315b7 ("selftest: mm: Test if hugepage does not get leaked during __bio_release_pages()")
> 
> from the mm-stable tree and commit:
> 
>   94beef29e110 ("mm: add MAP_DROPPABLE for designating always lazily freeable mappings")
> 
> from the random tree.
> 
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging.  You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.
> 
> diff --cc tools/testing/selftests/mm/Makefile
> index e1aa09ddaa3d,e3e5740e13e1..000000000000
> --- a/tools/testing/selftests/mm/Makefile
> +++ b/tools/testing/selftests/mm/Makefile
> @@@ -75,7 -73,7 +75,8 @@@ TEST_GEN_FILES += ksm_functional_test
>   TEST_GEN_FILES += mdwe_test
>   TEST_GEN_FILES += hugetlb_fault_after_madv
>   TEST_GEN_FILES += hugetlb_madv_vs_map
>  +TEST_GEN_FILES += hugetlb_dio
> + TEST_GEN_FILES += droppable
>   
>   ifneq ($(ARCH),arm64)
>   TEST_GEN_FILES += soft-dirty

This is now a conflict between the random tree and Linus' tree.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]