Bad use of highmem with buffer_migrate_page?

PATCH Laura Abbott 5 participants 8 messages May 10, 2012 – Jul 5, 2012

Subsystems (3)

Memory management Arm Qualcomm mailing list Page allocator

lore

Share & tools

lore

Laura Abbott

author

2012-05-10 20:07 UTC #0 lore raw

Hi,

I did a backport of the Contiguous Memory Allocator to a 3.0.8 tree. I wrote fairly simple test case that, in 1MB chunks, allocs up to 40MB from a reserved area, maps, writes, unmaps and then frees in an infinite loop. When running this with another program in parallel to put some stress on the filesystem, I hit data aborts in the filesystem/journal layer, although not always the same backtrace. As an example:

[<c02907a4>] (__ext4_check_dir_entry+0x20/0x184) from [<c029e1a8>] (add_dirent_to_buf+0x70/0x2ac)
[<c029e1a8>] (add_dirent_to_buf+0x70/0x2ac) from [<c029f3f0>] (ext4_add_entry+0xd8/0x4bc)
[<c029f3f0>] (ext4_add_entry+0xd8/0x4bc) from [<c029fe90>] (ext4_add_nondir+0x14/0x64)
[<c029fe90>] (ext4_add_nondir+0x14/0x64) from [<c02a04c4>] (ext4_create+0xd8/0x120)
[<c02a04c4>] (ext4_create+0xd8/0x120) from [<c022e134>] (vfs_create+0x74/0xa4)
[<c022e134>] (vfs_create+0x74/0xa4) from [<c022ed3c>] (do_last+0x588/0x8d4) [<c022ed3c>] (do_last+0x588/0x8d4) from [<c022fe64>] (path_openat+0xc4/0x394)
[<c022fe64>] (path_openat+0xc4/0x394) from [<c0230214>] (do_filp_open+0x30/0x7c)
[<c0230214>] (do_filp_open+0x30/0x7c) from [<c0220cb4>] (do_sys_open+0xd8/0x174)
[<c0220cb4>] (do_sys_open+0xd8/0x174) from [<c0105ea0>] (ret_fast_syscall+0x0/0x30)

Every panic had the same issue where a struct buffer_head [1] had a b_data that was unexpectedly NULL.

During the course of CMA, buffer_migrate_page could be called to migrate from a CMA page to a new page. buffer_migrate_page calls set_bh_page[2] to set the new page for the buffer_head. If the new page is a highmem page though, the bh->b_data ends up as NULL, which could produce the panics seen above.

This seems to indicate that highmem pages are not not appropriate for use as pages to migrate to. The following made the problem go away for me:

mm/page_alloc.c +1 -1

@@ -5753,7 +5753,7 @@ static struct page *
  __alloc_contig_migrate_alloc(struct page *page, unsigned long private,
                              int **resultp)
  {
-       return alloc_page(GFP_HIGHUSER_MOVABLE);
+       return alloc_page(GFP_USER | __GFP_MOVABLE);
  }

Does this seem like an actual issue or is this an artifact of my backport to 3.0? I'm not familiar enough with the filesystem layer to be able to tell where highmem can actually be used.

Thanks,
Laura

[1] http://lxr.free-electrons.com/source/include/linux/buffer_head.h#L59
[2] http://lxr.free-electrons.com/source/fs/buffer.c?v=3.0#L1441

Signature · 2 lines

Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.

Marek Szyprowski

2012-05-11 08:30 UTC #1 lore raw

Hello,

Quoted 50 lines · Laura Abbott

I did a backport of the Contiguous Memory Allocator to a 3.0.8 tree. I wrote fairly simple test case that, in 1MB chunks, allocs up to 40MB from a reserved area, maps, writes, unmaps and then frees in an infinite loop. When running this with another program in parallel to put some stress on the filesystem, I hit data aborts in the filesystem/journal layer, although not always the same backtrace. As an example:
[<c02907a4>] (__ext4_check_dir_entry+0x20/0x184) from [<c029e1a8>] (add_dirent_to_buf+0x70/0x2ac)
[<c029e1a8>] (add_dirent_to_buf+0x70/0x2ac) from [<c029f3f0>] (ext4_add_entry+0xd8/0x4bc)
[<c029f3f0>] (ext4_add_entry+0xd8/0x4bc) from [<c029fe90>] (ext4_add_nondir+0x14/0x64)
[<c029fe90>] (ext4_add_nondir+0x14/0x64) from [<c02a04c4>] (ext4_create+0xd8/0x120)
[<c02a04c4>] (ext4_create+0xd8/0x120) from [<c022e134>] (vfs_create+0x74/0xa4)
[<c022e134>] (vfs_create+0x74/0xa4) from [<c022ed3c>] (do_last+0x588/0x8d4) [<c022ed3c>] (do_last+0x588/0x8d4) from [<c022fe64>] (path_openat+0xc4/0x394)
[<c022fe64>] (path_openat+0xc4/0x394) from [<c0230214>] (do_filp_open+0x30/0x7c)
[<c0230214>] (do_filp_open+0x30/0x7c) from [<c0220cb4>] (do_sys_open+0xd8/0x174)
[<c0220cb4>] (do_sys_open+0xd8/0x174) from [<c0105ea0>] (ret_fast_syscall+0x0/0x30)
Every panic had the same issue where a struct buffer_head [1] had a b_data that was unexpectedly NULL.
During the course of CMA, buffer_migrate_page could be called to migrate from a CMA page to a new page. buffer_migrate_page calls set_bh_page[2] to set the new page for the buffer_head. If the new page is a highmem page though, the bh->b_data ends up as NULL, which could produce the panics seen above.
This seems to indicate that highmem pages are not not appropriate for use as pages to migrate to. The following made the problem go away for me:
mm/page_alloc.c +1 -1
@@ -5753,7 +5753,7 @@ static struct page *
  __alloc_contig_migrate_alloc(struct page *page, unsigned long private,
                              int **resultp)
  {
-       return alloc_page(GFP_HIGHUSER_MOVABLE);
+       return alloc_page(GFP_USER | __GFP_MOVABLE);
  }
 
 

Quoted 3 lines

Does this seem like an actual issue or is this an artifact of my backport to 3.0? I'm not familiar enough with the filesystem layer to be able to tell where highmem can actually be used.

I will need to investigate this further as this issue doesn't appear on v3.3+ kernels, but I remember I saw something similar when I tried CMA backported to v3.0.

You have pointed to an important issue which we need to solve somehow. CMA wasn't fully tested with high mem and it looks that there are some issues here and there. Your patch will prevent using highmem for any migration triggered by CMA. IMHO this is a bit limited, but right now I have no better idea. For a quick backport it should be ok.

Best regards

Signature · 2 lines

Marek Szyprowski
Samsung Poland R&D Center

Laura Abbott

author

2012-05-11 21:51 UTC #2 lore raw

Quoted 54 lines · Marek Szyprowski

Hello,
Quoted 50 lines · Laura Abbott
I did a backport of the Contiguous Memory Allocator to a 3.0.8 tree. I wrote fairly simple test case that, in 1MB chunks, allocs up to 40MB from a reserved area, maps, writes, unmaps and then frees in an infinite loop. When running this with another program in parallel to put some stress on the filesystem, I hit data aborts in the filesystem/journal layer, although not always the same backtrace. As an example:
[<c02907a4>] (__ext4_check_dir_entry+0x20/0x184) from [<c029e1a8>] (add_dirent_to_buf+0x70/0x2ac)
[<c029e1a8>] (add_dirent_to_buf+0x70/0x2ac) from [<c029f3f0>] (ext4_add_entry+0xd8/0x4bc)
[<c029f3f0>] (ext4_add_entry+0xd8/0x4bc) from [<c029fe90>] (ext4_add_nondir+0x14/0x64)
[<c029fe90>] (ext4_add_nondir+0x14/0x64) from [<c02a04c4>] (ext4_create+0xd8/0x120)
[<c02a04c4>] (ext4_create+0xd8/0x120) from [<c022e134>] (vfs_create+0x74/0xa4)
[<c022e134>] (vfs_create+0x74/0xa4) from [<c022ed3c>] (do_last+0x588/0x8d4) [<c022ed3c>] (do_last+0x588/0x8d4) from [<c022fe64>] (path_openat+0xc4/0x394)
[<c022fe64>] (path_openat+0xc4/0x394) from [<c0230214>] (do_filp_open+0x30/0x7c)
[<c0230214>] (do_filp_open+0x30/0x7c) from [<c0220cb4>] (do_sys_open+0xd8/0x174)
[<c0220cb4>] (do_sys_open+0xd8/0x174) from [<c0105ea0>] (ret_fast_syscall+0x0/0x30)
Every panic had the same issue where a struct buffer_head [1] had a b_data that was unexpectedly NULL.
During the course of CMA, buffer_migrate_page could be called to migrate from a CMA page to a new page. buffer_migrate_page calls set_bh_page[2] to set the new page for the buffer_head. If the new page is a highmem page though, the bh->b_data ends up as NULL, which could produce the panics seen above.
This seems to indicate that highmem pages are not not appropriate for use as pages to migrate to. The following made the problem go away for me:
mm/page_alloc.c +1 -1
@@ -5753,7 +5753,7 @@ static struct page *
   __alloc_contig_migrate_alloc(struct page *page, unsigned long private,
                               int **resultp)
   {
-       return alloc_page(GFP_HIGHUSER_MOVABLE);
+       return alloc_page(GFP_USER | __GFP_MOVABLE);
   }
 
 

Quoted 8 lines

Quoted 3 lines
Does this seem like an actual issue or is this an artifact of my backport to 3.0? I'm not familiar enough with the filesystem layer to be able to tell where highmem can actually be used.
I will need to investigate this further as this issue doesn't appear on v3.3+ kernels, but I remember I saw something similar when I tried CMA backported to v3.0.

The 3.0 kernel was the most stable I had around and easiest to work with. I'll be trying 3.4 sometime in the near future.

Quoted 6 lines

You have pointed to an important issue which we need to solve somehow. CMA wasn't fully tested with high mem and it looks that there are some issues here and there. Your patch will prevent using highmem for any migration triggered by CMA. IMHO this is a bit limited, but right now I have no better idea. For a quick backport it should be ok.

All the systems I'll be testing with will have highmem so I'll continue to report any problems I find. I'll be curious if this will show up again when I try 3.4. If it does, I'll see if I can get evidence of why this is the proper fix or a more nuanced patch.

Quoted 1 line

Best regards

Thanks,
Laura

Signature · 2 lines

Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.

Rabin Vincent

2012-07-05 09:28 UTC #3 lore raw

Quoted 54 lines · Laura Abbott

Quoted 53 lines · Marek Szyprowski
Quoted 52 lines · Laura Abbott
I did a backport of the Contiguous Memory Allocator to a 3.0.8 tree. I wrote fairly simple test case that, in 1MB chunks, allocs up to 40MB from a reserved area, maps, writes, unmaps and then frees in an infinite loop. When running this with another program in parallel to put some stress on the filesystem, I hit data aborts in the filesystem/journal layer, although not always the same backtrace. As an example:
[<c02907a4>] (__ext4_check_dir_entry+0x20/0x184) from [<c029e1a8>] (add_dirent_to_buf+0x70/0x2ac)
[<c029e1a8>] (add_dirent_to_buf+0x70/0x2ac) from [<c029f3f0>] (ext4_add_entry+0xd8/0x4bc)
[<c029f3f0>] (ext4_add_entry+0xd8/0x4bc) from [<c029fe90>] (ext4_add_nondir+0x14/0x64)
[<c029fe90>] (ext4_add_nondir+0x14/0x64) from [<c02a04c4>] (ext4_create+0xd8/0x120)
[<c02a04c4>] (ext4_create+0xd8/0x120) from [<c022e134>] (vfs_create+0x74/0xa4)
[<c022e134>] (vfs_create+0x74/0xa4) from [<c022ed3c>] (do_last+0x588/0x8d4)
[<c022ed3c>] (do_last+0x588/0x8d4) from [<c022fe64>] (path_openat+0xc4/0x394)
[<c022fe64>] (path_openat+0xc4/0x394) from [<c0230214>] (do_filp_open+0x30/0x7c)
[<c0230214>] (do_filp_open+0x30/0x7c) from [<c0220cb4>] (do_sys_open+0xd8/0x174)
[<c0220cb4>] (do_sys_open+0xd8/0x174) from [<c0105ea0>] (ret_fast_syscall+0x0/0x30)
Every panic had the same issue where a struct buffer_head [1] had a b_data that was unexpectedly NULL.
During the course of CMA, buffer_migrate_page could be called to migrate from a CMA page to a new page. buffer_migrate_page calls set_bh_page[2] to set the new page for the buffer_head. If the new page is a highmem page though, the bh->b_data ends up as NULL, which could produce the panics seen above.
This seems to indicate that highmem pages are not not appropriate for use as pages to migrate to. The following made the problem go away for me:
mm/page_alloc.c +1 -1
@@ -5753,7 +5753,7 @@ static struct page *
   __alloc_contig_migrate_alloc(struct page *page, unsigned long private,
                               int **resultp)
   {
-       return alloc_page(GFP_HIGHUSER_MOVABLE);
+       return alloc_page(GFP_USER | __GFP_MOVABLE);
   }
 
 

Quoted 8 lines

Quoted 8 lines
Quoted 3 lines
Does this seem like an actual issue or is this an artifact of my backport to 3.0? I'm not familiar enough with the filesystem layer to be able to tell where highmem can actually be used.
I will need to investigate this further as this issue doesn't appear on v3.3+ kernels, but I remember I saw something similar when I tried CMA backported to v3.0.

The problem is still present on latest mainline. The filesystem layer expects that the pages in the block device's mapping are not in highmem (the mapping's gfp mask is set in bdget()), but CMA replaces lowmem pages with highmem pages leading to the crashes.

The above fix should work, but perhaps the following is preferable since it should allow moving highmem pages to other highmem pages?

mm/page_alloc.c +6 -1

index 4403009..4a4f921 100644
@@ -5635,7 +5635,12 @@ static struct page *
 __alloc_contig_migrate_alloc(struct page *page, unsigned long private,
 			     int **resultp)
 {
-	return alloc_page(GFP_HIGHUSER_MOVABLE);
+	gfp_t gfp_mask = GFP_USER | __GFP_MOVABLE;
+
+	if (PageHighMem(page))
+		gfp_mask |= __GFP_HIGHMEM;
+
+	return alloc_page(gfp_mask);
 }
 
 /* [start, end) must belong to a single zone. */

Marek Szyprowski

2012-07-05 10:05 UTC #4 lore raw

Hello,

Quoted 80 lines · Rabin Vincent

Quoted 54 lines · Laura Abbott
Quoted 53 lines · Marek Szyprowski
Quoted 52 lines · Laura Abbott
I did a backport of the Contiguous Memory Allocator to a 3.0.8 tree. I wrote fairly simple test case that, in 1MB chunks, allocs up to 40MB from a reserved area, maps, writes, unmaps and then frees in an infinite loop. When running this with another program in parallel to put some stress on the filesystem, I hit data aborts in the filesystem/journal layer, although not always the same backtrace. As an example:
[<c02907a4>] (__ext4_check_dir_entry+0x20/0x184) from [<c029e1a8>] (add_dirent_to_buf+0x70/0x2ac)
[<c029e1a8>] (add_dirent_to_buf+0x70/0x2ac) from [<c029f3f0>] (ext4_add_entry+0xd8/0x4bc)
[<c029f3f0>] (ext4_add_entry+0xd8/0x4bc) from [<c029fe90>] (ext4_add_nondir+0x14/0x64)
[<c029fe90>] (ext4_add_nondir+0x14/0x64) from [<c02a04c4>] (ext4_create+0xd8/0x120)
[<c02a04c4>] (ext4_create+0xd8/0x120) from [<c022e134>] (vfs_create+0x74/0xa4)
[<c022e134>] (vfs_create+0x74/0xa4) from [<c022ed3c>] (do_last+0x588/0x8d4)
[<c022ed3c>] (do_last+0x588/0x8d4) from [<c022fe64>] (path_openat+0xc4/0x394)
[<c022fe64>] (path_openat+0xc4/0x394) from [<c0230214>] (do_filp_open+0x30/0x7c)
[<c0230214>] (do_filp_open+0x30/0x7c) from [<c0220cb4>] (do_sys_open+0xd8/0x174)
[<c0220cb4>] (do_sys_open+0xd8/0x174) from [<c0105ea0>] (ret_fast_syscall+0x0/0x30)
Every panic had the same issue where a struct buffer_head [1] had a b_data that was unexpectedly NULL.
During the course of CMA, buffer_migrate_page could be called to migrate from a CMA page to a new page. buffer_migrate_page calls set_bh_page[2] to set the new page for the buffer_head. If the new page is a highmem page though, the bh->b_data ends up as NULL, which could produce the panics seen above.
This seems to indicate that highmem pages are not not appropriate for use as pages to migrate to. The following made the problem go away for me:
mm/page_alloc.c +1 -1
@@ -5753,7 +5753,7 @@ static struct page *
   __alloc_contig_migrate_alloc(struct page *page, unsigned long private,
                               int **resultp)
   {
-       return alloc_page(GFP_HIGHUSER_MOVABLE);
+       return alloc_page(GFP_USER | __GFP_MOVABLE);
   }
 
 
Quoted 8 lines
Quoted 8 lines
Quoted 3 lines
Does this seem like an actual issue or is this an artifact of my backport to 3.0? I'm not familiar enough with the filesystem layer to be able to tell where highmem can actually be used.
I will need to investigate this further as this issue doesn't appear on v3.3+ kernels, but I remember I saw something similar when I tried CMA backported to v3.0.
The problem is still present on latest mainline. The filesystem layer expects that the pages in the block device's mapping are not in highmem (the mapping's gfp mask is set in bdget()), but CMA replaces lowmem pages with highmem pages leading to the crashes.
The above fix should work, but perhaps the following is preferable since it should allow moving highmem pages to other highmem pages?
mm/page_alloc.c +0 -0
index 4403009..4a4f921 100644
@@ -5635,7 +5635,12 @@ static struct page *
 __alloc_contig_migrate_alloc(struct page *page, unsigned long private,
 			     int **resultp)
 {

mm/page_alloc.c static struct page * · line 5646

Quoted 10 lines

mm/page_alloc.c +6 -1

-	return alloc_page(GFP_HIGHUSER_MOVABLE);
+	gfp_t gfp_mask = GFP_USER | __GFP_MOVABLE;
+
+	if (PageHighMem(page))
+		gfp_mask |= __GFP_HIGHMEM;
+
+	return alloc_page(gfp_mask);
 }
 
 /* [start, end) must belong to a single zone. */

The patch looks fine and does it job well. Could you resend it as a complete patch with commit message and signed-off-by/reported-by lines? I will handle merging it to mainline then.

Best regards

Signature · 2 lines

Marek Szyprowski
Samsung Poland R&D Center

Rabin Vincent

2012-07-05 10:45 UTC #5 lore raw

Quoted 26 lines · Marek Szyprowski

Quoted 25 lines · Rabin Vincent
The problem is still present on latest mainline. The filesystem layer expects that the pages in the block device's mapping are not in highmem (the mapping's gfp mask is set in bdget()), but CMA replaces lowmem pages with highmem pages leading to the crashes.
The above fix should work, but perhaps the following is preferable since it should allow moving highmem pages to other highmem pages?
mm/page_alloc.c +6 -1
index 4403009..4a4f921 100644
@@ -5635,7 +5635,12 @@ static struct page *
 __alloc_contig_migrate_alloc(struct page *page, unsigned long private,
 			     int **resultp)
 {
-	return alloc_page(GFP_HIGHUSER_MOVABLE);
+	gfp_t gfp_mask = GFP_USER | __GFP_MOVABLE;
+
+	if (PageHighMem(page))
+		gfp_mask |= __GFP_HIGHMEM;
+
+	return alloc_page(gfp_mask);
 }
 

Quoted 6 lines

mm/page_alloc.c static struct page * · line 5646
Quoted 1 line
mm/page_alloc.c +0 -0
 /* [start, end) must belong to a single zone. */
The patch looks fine and does it job well. Could you resend it as a complete patch with commit message and signed-off-by/reported-by lines? I will handle merging it to mainline then.

Thanks, here it is:

8<----

Quoted 1 line

From 8a94126eb3aa2824866405fb78bb0b8316f8fd00 Mon Sep 17 00:00:00 2001

From: Rabin Vincent <[email protected]>
Date: Thu, 5 Jul 2012 15:52:23 +0530
Subject: [PATCH] mm: cma: don't replace lowmem pages with highmem

The filesystem layer expects pages in the block device's mapping to not be in highmem (the mapping's gfp mask is set in bdget()), but CMA can currently replace lowmem pages with highmem pages, leading to crashes in filesystem code such as the one below:

  Unable to handle kernel NULL pointer dereference at virtual address 00000400
  pgd = c0c98000
  [00000400] *pgd=00c91831, *pte=00000000, *ppte=00000000
  Internal error: Oops: 817 [#1] PREEMPT SMP ARM
  CPU: 0    Not tainted  (3.5.0-rc5+ #80)
  PC is at __memzero+0x24/0x80
  ...
  Process fsstress (pid: 323, stack limit = 0xc0cbc2f0)
  Backtrace:
  [<c010e3f0>] (ext4_getblk+0x0/0x180) from [<c010e58c>] (ext4_bread+0x1c/0x98)
  [<c010e570>] (ext4_bread+0x0/0x98) from [<c0117944>] (ext4_mkdir+0x160/0x3bc)
   r4:c15337f0
  [<c01177e4>] (ext4_mkdir+0x0/0x3bc) from [<c00c29e0>] (vfs_mkdir+0x8c/0x98)
  [<c00c2954>] (vfs_mkdir+0x0/0x98) from [<c00c2a60>] (sys_mkdirat+0x74/0xac)
   r6:00000000 r5:c152eb40 r4:000001ff r3:c14b43f0
  [<c00c29ec>] (sys_mkdirat+0x0/0xac) from [<c00c2ab8>] (sys_mkdir+0x20/0x24)
   r6:beccdcf0 r5:00074000 r4:beccdbbc
  [<c00c2a98>] (sys_mkdir+0x0/0x24) from [<c000e3c0>] (ret_fast_syscall+0x0/0x30)

Fix this by replacing only highmem pages with highmem.

Reported-by Laura Abbott <[email protected]>

Signed-off-by Rabin Vincent <[email protected]>

---
 mm/page_alloc.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

mm/page_alloc.c +6 -1

index 4403009..4a4f921 100644
@@ -5635,7 +5635,12 @@ static struct page *
 __alloc_contig_migrate_alloc(struct page *page, unsigned long private,
 			     int **resultp)
 {
-	return alloc_page(GFP_HIGHUSER_MOVABLE);
+	gfp_t gfp_mask = GFP_USER | __GFP_MOVABLE;
+
+	if (PageHighMem(page))
+		gfp_mask |= __GFP_HIGHMEM;
+
+	return alloc_page(gfp_mask);
 }
 
 /* [start, end) must belong to a single zone. */

Signature · 1 line

1.7.9.5

Cong Wang

Reviewed-by

2012-07-05 13:56 UTC #6 lore raw

Quoted 11 lines · Rabin Vincent

8<----
From 8a94126eb3aa2824866405fb78bb0b8316f8fd00 Mon Sep 17 00:00:00 2001
From: Rabin Vincent <[email protected]>
Date: Thu, 5 Jul 2012 15:52:23 +0530
Subject: [PATCH] mm: cma: don't replace lowmem pages with highmem
The filesystem layer expects pages in the block device's mapping to not be in highmem (the mapping's gfp mask is set in bdget()), but CMA can currently replace lowmem pages with highmem pages, leading to crashes in filesystem code such as the one below:

...

Quoted 2 lines

Fix this by replacing only highmem pages with highmem.

Looks good to me too,

Reviewed-by WANG Cong <[email protected]>

Michał Nazarewicz

Acked-by

2012-07-05 19:21 UTC #7 lore raw

In reply to Rabin Vincent

2012/7/5 Rabin Vincent <[email protected]>:

Quoted 29 lines

From 8a94126eb3aa2824866405fb78bb0b8316f8fd00 Mon Sep 17 00:00:00 2001
From: Rabin Vincent <[email protected]>
Date: Thu, 5 Jul 2012 15:52:23 +0530
Subject: [PATCH] mm: cma: don't replace lowmem pages with highmem
The filesystem layer expects pages in the block device's mapping to not be in highmem (the mapping's gfp mask is set in bdget()), but CMA can currently replace lowmem pages with highmem pages, leading to crashes in filesystem code such as the one below:
  Unable to handle kernel NULL pointer dereference at virtual address 00000400
  pgd = c0c98000
  [00000400] *pgd=00c91831, *pte=00000000, *ppte=00000000
  Internal error: Oops: 817 [#1] PREEMPT SMP ARM
  CPU: 0    Not tainted  (3.5.0-rc5+ #80)
  PC is at __memzero+0x24/0x80
  ...
  Process fsstress (pid: 323, stack limit = 0xc0cbc2f0)
  Backtrace:
  [<c010e3f0>] (ext4_getblk+0x0/0x180) from [<c010e58c>] (ext4_bread+0x1c/0x98)
  [<c010e570>] (ext4_bread+0x0/0x98) from [<c0117944>] (ext4_mkdir+0x160/0x3bc)
   r4:c15337f0
  [<c01177e4>] (ext4_mkdir+0x0/0x3bc) from [<c00c29e0>] (vfs_mkdir+0x8c/0x98)
  [<c00c2954>] (vfs_mkdir+0x0/0x98) from [<c00c2a60>] (sys_mkdirat+0x74/0xac)
   r6:00000000 r5:c152eb40 r4:000001ff r3:c14b43f0
  [<c00c29ec>] (sys_mkdirat+0x0/0xac) from [<c00c2ab8>] (sys_mkdir+0x20/0x24)
   r6:beccdcf0 r5:00074000 r4:beccdbbc
  [<c00c2a98>] (sys_mkdir+0x0/0x24) from [<c000e3c0>] (ret_fast_syscall+0x0/0x30)

Quoted 4 lines

Fix this by replacing only highmem pages with highmem.
Reported-by Laura Abbott <[email protected]>
Signed-off-by Rabin Vincent <[email protected]>

Acked-by Michal Nazarewicz <[email protected]>

Quoted 22 lines

---
 mm/page_alloc.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

mm/page_alloc.c +6 -1

index 4403009..4a4f921 100644
@@ -5635,7 +5635,12 @@ static struct page *
 __alloc_contig_migrate_alloc(struct page *page, unsigned long private,
                             int **resultp)
 {
-       return alloc_page(GFP_HIGHUSER_MOVABLE);
+       gfp_t gfp_mask = GFP_USER | __GFP_MOVABLE;
+
+       if (PageHighMem(page))
+               gfp_mask |= __GFP_HIGHMEM;
+
+       return alloc_page(gfp_mask);
 }
 
 /* [start, end) must belong to a single zone. */