mm: convert totalram_pages, totalhigh_pages and managed_pages to atomic.

Remove managed_page_count_lock spinlock and instead use atomic variables.

Remove managed_page_count_lock spinlock and instead use atomic variables.

Remove managed_page_count_lock spinlock and instead use atomic variables.

Remove managed_page_count_lock spinlock and instead use atomic variables.

Remove managed_page_count_lock spinlock and instead use atomic variables.

Remove managed_page_count_lock spinlock and instead use atomic variables.

Remove managed_page_count_lock spinlock and instead use atomic variables.

Quoted 2 lines · Arun KS

Remove managed_page_count_lock spinlock and instead use atomic variables.

I assume this has been auto-generated. If yes, it would be better to mention the script so that people can review it and regenerate for comparision. Such a large change is hard to review manually.

--
Michal Hocko
SUSE Labs

Quoted 2 lines · Arun KS

Remove managed_page_count_lock spinlock and instead use atomic variables.

I assume this has been auto-generated. If yes, it would be better to mention the script so that people can review it and regenerate for comparision. Such a large change is hard to review manually.

--
Michal Hocko
SUSE Labs

Quoted 2 lines · Arun KS

Remove managed_page_count_lock spinlock and instead use atomic variables.

Perhaps better to define and use macros for the accesses instead of specific uses of atomic_long_<inc/dec/read>

Something like:

#define totalram_pages() (unsigned long)atomic_long_read(&_totalram_pages)
#define totalram_pages_inc() (unsigned long)atomic_long_inc(&_totalram_pages)
#define totalram_pages_dec() (unsigned long)atomic_long_dec(&_totalram_pages)

Quoted 2 lines · Arun KS

Remove managed_page_count_lock spinlock and instead use atomic variables.

Perhaps better to define and use macros for the accesses instead of specific uses of atomic_long_<inc/dec/read>

Something like:

#define totalram_pages() (unsigned long)atomic_long_read(&_totalram_pages)
#define totalram_pages_inc() (unsigned long)atomic_long_inc(&_totalram_pages)
#define totalram_pages_dec() (unsigned long)atomic_long_dec(&_totalram_pages)

Quoted 2 lines

Remove managed_page_count_lock spinlock and instead use atomic variables.

Perhaps better to define and use macros for the accesses instead of specific uses of atomic_long_<inc/dec/read>

Something like:

#define totalram_pages() (unsigned long)atomic_long_read(&_totalram_pages)

#define totalram_pages_inc() (unsigned long)atomic_long_inc(&_totalram_pages)
#define totalram_pages_dec() (unsigned long)atomic_long_dec(&_totalram_pages)

Quoted 5 lines · Michal Hocko

Quoted 2 lines · Arun KS

Remove managed_page_count_lock spinlock and instead use atomic variables.

Hello Michal,

Quoted 3 lines

I assume this has been auto-generated. If yes, it would be better to mention the script so that people can review it and regenerate for comparision. Such a large change is hard to review manually.

Changes were made partially with script. For totalram_pages and totalhigh_pages,

find dir -type f -exec sed -i 's/totalram_pages/atomic_long_read(\&totalram_pages)/g' {} \; find dir -type f -exec sed -i 's/totalhigh_pages/atomic_long_read(\&totalhigh_pages)/g' {} \;

For managed_pages it was mostly manual edits after using, find mm/ -type f -exec sed -i 's/zone->managed_pages/atomic_long_read(\&zone->managed_pages)/g' {} \;

Remove managed_page_count_lock spinlock and instead use atomic variables.

Suggested-by Michal Hocko <mhocko-IBi9RG/[email protected]>

Suggested-by Vlastimil Babka <[email protected]>

Signed-off-by Arun KS <[email protected]>

---
As discussed here,
https://patchwork.kernel.org/patch/10627521/#22261253

Remove managed_page_count_lock spinlock and instead use atomic variables.

Suggested-by Michal Hocko <[email protected]>

Suggested-by Vlastimil Babka <[email protected]>

Signed-off-by Arun KS <[email protected]>

---
As discussed here,
https://patchwork.kernel.org/patch/10627521/#22261253

[Trimmed CC list + Julia - there is indeed no need to CC everybody maintain a file you are updating for the change like this]

Quoted 20 lines · Arun Sudhilal

Quoted 5 lines · Michal Hocko

Quoted 2 lines · Arun KS

Remove managed_page_count_lock spinlock and instead use atomic variables.

Hello Michal,

Quoted 3 lines

I assume this has been auto-generated. If yes, it would be better to mention the script so that people can review it and regenerate for comparision. Such a large change is hard to review manually.

Changes were made partially with script. For totalram_pages and totalhigh_pages,

find dir -type f -exec sed -i 's/totalram_pages/atomic_long_read(\&totalram_pages)/g' {} \; find dir -type f -exec sed -i 's/totalhigh_pages/atomic_long_read(\&totalhigh_pages)/g' {} \;

Quoted 4 lines

For managed_pages it was mostly manual edits after using, find mm/ -type f -exec sed -i 's/zone->managed_pages/atomic_long_read(\&zone->managed_pages)/g' {} \;

I guess we should be able to use coccinelle for this kind of change and reduce the amount of manual intervention to absolute minimum.

Quoted 29 lines · Michal Hocko

[Trimmed CC list + Julia - there is indeed no need to CC everybody maintain a file you are updating for the change like this]

Quoted 24 lines · Arun Sudhilal

Quoted 5 lines · Michal Hocko

Quoted 2 lines · Arun KS

Remove managed_page_count_lock spinlock and instead use atomic variables.

Hello Michal,

Quoted 3 lines

I assume this has been auto-generated. If yes, it would be better to mention the script so that people can review it and regenerate for comparision. Such a large change is hard to review manually.

Changes were made partially with script. For totalram_pages and totalhigh_pages,

find dir -type f -exec sed -i 's/totalram_pages/atomic_long_read(\&totalram_pages)/g' {} \; find dir -type f -exec sed -i 's/totalhigh_pages/atomic_long_read(\&totalhigh_pages)/g' {} \;

For managed_pages it was mostly manual edits after using, find mm/ -type f -exec sed -i 's/zone->managed_pages/atomic_long_read(\&zone->managed_pages)/g' {} \;

Quoted 2 lines

I guess we should be able to use coccinelle for this kind of change and reduce the amount of manual intervention to absolute minimum.

Coccinelle looks like it would be desirable, especially in case the word zone is not always used.

Arun, please feel free to contact me if you want to try it and need help.

julia

Quoted 6 lines · Arun KS

Quoted 2 lines

Remove managed_page_count_lock spinlock and instead use atomic variables.

Perhaps better to define and use macros for the accesses instead of specific uses of atomic_long_<inc/dec/read>

Quoted 4 lines

Something like:

#define totalram_pages() (unsigned
long)atomic_long_read(&_totalram_pages)

or proper static inline
this code isn't so low level for breaking include dependencies with macro

[dropped bloated cc - my server rejects this mess]

Quoted 9 lines · Konstantin Khlebnikov

Quoted 6 lines · Arun KS

Quoted 2 lines

Remove managed_page_count_lock spinlock and instead use atomic variables.

Perhaps better to define and use macros for the accesses instead of specific uses of atomic_long_<inc/dec/read>

Quoted 7 lines

Quoted 4 lines

Something like:

#define totalram_pages() (unsigned
long)atomic_long_read(&_totalram_pages)

or proper static inline
this code isn't so low level for breaking include dependencies with macro

BTW, I noticed a few places in the patch that did multiple evaluations of totalram_pages. It might be worth fixing those prior to doing the conversion, too. e.g.:

if (totalram_pages > something)
foobar(totalram_pages); <- value may have changed here

should, instead, be:

var = totalram_pages; <- get stable view of the value if (var > something)

    foobar(var);

-Kees

Quoted 1 line

[dropped bloated cc - my server rejects this mess]

Thank you -- I was struggling to figure out the best way to reply to this. :)

-Kees

On 2018-10-24 01:34, Kees Cook wrote:

Quoted 2 lines

Thank you -- I was struggling to figure out the best way to reply to this. :)

I'm sorry for the trouble caused. Sent the email using, git send-email --to-cmd="scripts/get_maintainer.pl -i" 0001-convert-totalram_pages-totalhigh_pages-and-managed_p.patch

Is this not a recommended approach?

Quoted 1 line · Arun KS

On 2018-10-24 01:34, Kees Cook wrote:

[...]

Quoted 6 lines

Quoted 2 lines

Thank you -- I was struggling to figure out the best way to reply to this. :)

I'm sorry for the trouble caused. Sent the email using, git send-email --to-cmd="scripts/get_maintainer.pl -i" 0001-convert-totalram_pages-totalhigh_pages-and-managed_p.patch

Quoted 1 line

Is this not a recommended approach?

Not really for tree wide mechanical changes. It is much more preferrable IMHO to only CC people who should review the intention of the change rather than each and every maintainer whose code is going to be changed. This is a case by case thing of course but as soon as you see a giant CC list from get_maintainer.pl then you should try to think twice to use it. If not sure, just ask on the mailing list.

Quoted 9 lines · Michal Hocko

Quoted 1 line · Arun KS

On 2018-10-24 01:34, Kees Cook wrote:

[...]

Quoted 6 lines

Quoted 2 lines

Thank you -- I was struggling to figure out the best way to reply to this. :)

I'm sorry for the trouble caused. Sent the email using, git send-email --to-cmd="scripts/get_maintainer.pl -i" 0001-convert-totalram_pages-totalhigh_pages-and-managed_p.patch

Quoted 8 lines

Quoted 1 line

Is this not a recommended approach?

Not really for tree wide mechanical changes. It is much more preferrable IMHO to only CC people who should review the intention of the change rather than each and every maintainer whose code is going to be changed. This is a case by case thing of course but as soon as you see a giant CC list from get_maintainer.pl then you should try to think twice to use it. If not sure, just ask on the mailing list.

Generally, it's better to use scripts to control
the --to-cmd and --cc-cmd options.

Quoted 19 lines · Joe Perches

Quoted 17 lines · Michal Hocko

Quoted 1 line · Arun KS

On 2018-10-24 01:34, Kees Cook wrote:

[...]

Quoted 7 lines

Quoted 2 lines

Thank you -- I was struggling to figure out the best way to reply to this. :)

I'm sorry for the trouble caused. Sent the email using, git send-email --to-cmd="scripts/get_maintainer.pl -i" 0001-convert-totalram_pages-totalhigh_pages-and-managed_p.patch

Is this not a recommended approach?

Not really for tree wide mechanical changes. It is much more preferrable IMHO to only CC people who should review the intention of the change rather than each and every maintainer whose code is going to be changed. This is a case by case thing of course but as soon as you see a giant CC list from get_maintainer.pl then you should try to think twice to use it. If not sure, just ask on the mailing list.

Quoted 2 lines

Generally, it's better to use scripts to control
the --to-cmd and --cc-cmd options.

I would argue that it is better to use a common sense much more than scripts.

Quoted 29 lines · Michal Hocko

[Trimmed CC list + Julia - there is indeed no need to CC everybody maintain a file you are updating for the change like this]

Quoted 24 lines · Arun Sudhilal

Quoted 5 lines · Michal Hocko

Quoted 2 lines · Arun KS

Remove managed_page_count_lock spinlock and instead use atomic variables.

Hello Michal,

Quoted 3 lines

I assume this has been auto-generated. If yes, it would be better to mention the script so that people can review it and regenerate for comparision. Such a large change is hard to review manually.

Changes were made partially with script. For totalram_pages and totalhigh_pages,

find dir -type f -exec sed -i 's/totalram_pages/atomic_long_read(\&totalram_pages)/g' {} \; find dir -type f -exec sed -i 's/totalhigh_pages/atomic_long_read(\&totalhigh_pages)/g' {} \;

For managed_pages it was mostly manual edits after using, find mm/ -type f -exec sed -i 's/zone->managed_pages/atomic_long_read(\&zone->managed_pages)/g' {} \;

Quoted 2 lines

I guess we should be able to use coccinelle for this kind of change and reduce the amount of manual intervention to absolute minimum.

Coccinelle looks like it would be desirable, especially in case the word zone is not always used.

Arun, please feel free to contact me if you want to try it and need help.

Quoted 1 line

Quoted 1 line

atomic_long_read(&rule starting on line 1:totalram_pages)

Quoted 1 line

Quoted 1 line

atomic_long_read(&rule starting on line 1:totalram_pages)

julia

Quoted 20 lines · Michal Hocko

Quoted 19 lines · Joe Perches

Quoted 17 lines · Michal Hocko

Quoted 1 line · Arun KS

On 2018-10-24 01:34, Kees Cook wrote:

[...]

Quoted 7 lines

Quoted 2 lines

Thank you -- I was struggling to figure out the best way to reply to this. :)

I'm sorry for the trouble caused. Sent the email using, git send-email --to-cmd="scripts/get_maintainer.pl -i" 0001-convert-totalram_pages-totalhigh_pages-and-managed_p.patch

Is this not a recommended approach?

Not really for tree wide mechanical changes. It is much more preferrable IMHO to only CC people who should review the intention of the change rather than each and every maintainer whose code is going to be changed. This is a case by case thing of course but as soon as you see a giant CC list from get_maintainer.pl then you should try to think twice to use it. If not sure, just ask on the mailing list.

Quoted 5 lines

Quoted 2 lines

Generally, it's better to use scripts to control
the --to-cmd and --cc-cmd options.

I would argue that it is better to use a common sense much more than scripts.

Common sense isn't common.

Remove managed_page_count_lock spinlock and instead use atomic variables.

Suggested-by Michal Hocko <mhocko-IBi9RG/[email protected]>

Suggested-by Vlastimil Babka <[email protected]>

Signed-off-by Arun KS <[email protected]>

---
As discussed here,
https://patchwork.kernel.org/patch/10627521/#22261253
---

---
 arch/csky/mm/init.c                           |  4 +-
 arch/powerpc/platforms/pseries/cmm.c          | 11 ++--
 arch/s390/mm/init.c                           |  2 +-
 arch/um/kernel/mem.c                          |  4 +-
 arch/x86/kernel/cpu/microcode/core.c          |  5 +-
 drivers/char/agp/backend.c                    |  4 +-
 drivers/gpu/drm/amd/amdkfd/kfd_crat.c         |  2 +-
 drivers/gpu/drm/i915/i915_gem.c               |  2 +-
 drivers/gpu/drm/i915/selftests/i915_gem_gtt.c |  4 +-
 drivers/hv/hv_balloon.c                       | 19 +++----
 drivers/md/dm-bufio.c                         |  5 +-
 drivers/md/dm-crypt.c                         |  4 +-
 drivers/md/dm-integrity.c                     |  4 +-
 drivers/md/dm-stats.c                         |  3 +-
 drivers/media/platform/mtk-vpu/mtk_vpu.c      |  3 +-
 drivers/misc/vmw_balloon.c                    |  2 +-
 drivers/parisc/ccio-dma.c                     |  5 +-
 drivers/parisc/sba_iommu.c                    |  5 +-
 drivers/staging/android/ion/ion_system_heap.c |  2 +-
 drivers/xen/xen-selfballoon.c                 |  7 +--
 fs/ceph/super.h                               |  3 +-
 fs/file_table.c                               |  9 ++--
 fs/fuse/inode.c                               |  4 +-
 fs/nfs/write.c                                |  3 +-
 fs/nfsd/nfscache.c                            |  3 +-
 fs/ntfs/malloc.h                              |  2 +-
 fs/proc/base.c                                |  3 +-
 include/linux/highmem.h                       |  2 +-
 include/linux/mm.h                            |  2 +-
 include/linux/mmzone.h                        | 10 +---
 include/linux/swap.h                          |  2 +-
 kernel/fork.c                                 |  6 +--
 kernel/kexec_core.c                           |  5 +-
 kernel/power/snapshot.c                       |  2 +-
 lib/show_mem.c                                |  3 +-
 mm/highmem.c                                  |  2 +-
 mm/huge_memory.c                              |  2 +-
 mm/kasan/quarantine.c                         |  4 +-
 mm/memblock.c                                 |  6 +--
 mm/memory_hotplug.c                           |  4 +-
 mm/mm_init.c                                  |  3 +-
 mm/oom_kill.c                                 |  2 +-
 mm/page_alloc.c                               | 75 ++++++++++++++-------------
 mm/shmem.c                                    | 12 +++--
 mm/slab.c                                     |  3 +-
 mm/swap.c                                     |  3 +-
 mm/util.c                                     |  2 +-
 mm/vmalloc.c                                  |  4 +-
 mm/vmstat.c                                   |  4 +-
 mm/workingset.c                               |  2 +-
 mm/zswap.c                                    |  2 +-
 net/dccp/proto.c                              |  6 +--
 net/decnet/dn_route.c                         |  2 +-
 net/ipv4/tcp_metrics.c                        |  2 +-
 net/netfilter/nf_conntrack_core.c             |  6 +--
 net/netfilter/xt_hashlimit.c                  |  4 +-
 net/sctp/protocol.c                           |  6 +--
 security/integrity/ima/ima_kexec.c            |  2 +-
 58 files changed, 171 insertions(+), 143 deletions(-)

arch/csky/mm/init.c +1 -1

index dc07c07..3f4d35e 100644
@@ -71,7 +71,7 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 		ClearPageReserved(virt_to_page(start));
 		init_page_count(virt_to_page(start));
 		free_page(start);
-		totalram_pages++;
+		atomic_long_inc(&totalram_pages);
 	}
 }
 #endif

Remove managed_page_count_lock spinlock and instead use atomic variables.

Suggested-by Michal Hocko <[email protected]>

Suggested-by Vlastimil Babka <[email protected]>

Signed-off-by Arun KS <[email protected]>

---
As discussed here,
https://patchwork.kernel.org/patch/10627521/#22261253
---

---
 arch/csky/mm/init.c                           |  4 +-
 arch/powerpc/platforms/pseries/cmm.c          | 11 ++--
 arch/s390/mm/init.c                           |  2 +-
 arch/um/kernel/mem.c                          |  4 +-
 arch/x86/kernel/cpu/microcode/core.c          |  5 +-
 drivers/char/agp/backend.c                    |  4 +-
 drivers/gpu/drm/amd/amdkfd/kfd_crat.c         |  2 +-
 drivers/gpu/drm/i915/i915_gem.c               |  2 +-
 drivers/gpu/drm/i915/selftests/i915_gem_gtt.c |  4 +-
 drivers/hv/hv_balloon.c                       | 19 +++----
 drivers/md/dm-bufio.c                         |  5 +-
 drivers/md/dm-crypt.c                         |  4 +-
 drivers/md/dm-integrity.c                     |  4 +-
 drivers/md/dm-stats.c                         |  3 +-
 drivers/media/platform/mtk-vpu/mtk_vpu.c      |  3 +-
 drivers/misc/vmw_balloon.c                    |  2 +-
 drivers/parisc/ccio-dma.c                     |  5 +-
 drivers/parisc/sba_iommu.c                    |  5 +-
 drivers/staging/android/ion/ion_system_heap.c |  2 +-
 drivers/xen/xen-selfballoon.c                 |  7 +--
 fs/ceph/super.h                               |  3 +-
 fs/file_table.c                               |  9 ++--
 fs/fuse/inode.c                               |  4 +-
 fs/nfs/write.c                                |  3 +-
 fs/nfsd/nfscache.c                            |  3 +-
 fs/ntfs/malloc.h                              |  2 +-
 fs/proc/base.c                                |  3 +-
 include/linux/highmem.h                       |  2 +-
 include/linux/mm.h                            |  2 +-
 include/linux/mmzone.h                        | 10 +---
 include/linux/swap.h                          |  2 +-
 kernel/fork.c                                 |  6 +--
 kernel/kexec_core.c                           |  5 +-
 kernel/power/snapshot.c                       |  2 +-
 lib/show_mem.c                                |  3 +-
 mm/highmem.c                                  |  2 +-
 mm/huge_memory.c                              |  2 +-
 mm/kasan/quarantine.c                         |  4 +-
 mm/memblock.c                                 |  6 +--
 mm/memory_hotplug.c                           |  4 +-
 mm/mm_init.c                                  |  3 +-
 mm/oom_kill.c                                 |  2 +-
 mm/page_alloc.c                               | 75 ++++++++++++++-------------
 mm/shmem.c                                    | 12 +++--
 mm/slab.c                                     |  3 +-
 mm/swap.c                                     |  3 +-
 mm/util.c                                     |  2 +-
 mm/vmalloc.c                                  |  4 +-
 mm/vmstat.c                                   |  4 +-
 mm/workingset.c                               |  2 +-
 mm/zswap.c                                    |  2 +-
 net/dccp/proto.c                              |  6 +--
 net/decnet/dn_route.c                         |  2 +-
 net/ipv4/tcp_metrics.c                        |  2 +-
 net/netfilter/nf_conntrack_core.c             |  6 +--
 net/netfilter/xt_hashlimit.c                  |  4 +-
 net/sctp/protocol.c                           |  6 +--
 security/integrity/ima/ima_kexec.c            |  2 +-
 58 files changed, 171 insertions(+), 143 deletions(-)

arch/csky/mm/init.c +1 -1

index dc07c07..3f4d35e 100644
@@ -71,7 +71,7 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 		ClearPageReserved(virt_to_page(start));
 		init_page_count(virt_to_page(start));
 		free_page(start);
-		totalram_pages++;
+		atomic_long_inc(&totalram_pages);
 	}
 }
 #endif

Remove managed_page_count_lock spinlock and instead use atomic variables.

Suggested-by Michal Hocko <[email protected]>

Suggested-by Vlastimil Babka <[email protected]>

Signed-off-by Arun KS <[email protected]>

---
As discussed here,
https://patchwork.kernel.org/patch/10627521/#22261253
---

---
 arch/csky/mm/init.c                           |  4 +-
 arch/powerpc/platforms/pseries/cmm.c          | 11 ++--
 arch/s390/mm/init.c                           |  2 +-
 arch/um/kernel/mem.c                          |  4 +-
 arch/x86/kernel/cpu/microcode/core.c          |  5 +-
 drivers/char/agp/backend.c                    |  4 +-
 drivers/gpu/drm/amd/amdkfd/kfd_crat.c         |  2 +-
 drivers/gpu/drm/i915/i915_gem.c               |  2 +-
 drivers/gpu/drm/i915/selftests/i915_gem_gtt.c |  4 +-
 drivers/hv/hv_balloon.c                       | 19 +++----
 drivers/md/dm-bufio.c                         |  5 +-
 drivers/md/dm-crypt.c                         |  4 +-
 drivers/md/dm-integrity.c                     |  4 +-
 drivers/md/dm-stats.c                         |  3 +-
 drivers/media/platform/mtk-vpu/mtk_vpu.c      |  3 +-
 drivers/misc/vmw_balloon.c                    |  2 +-
 drivers/parisc/ccio-dma.c                     |  5 +-
 drivers/parisc/sba_iommu.c                    |  5 +-
 drivers/staging/android/ion/ion_system_heap.c |  2 +-
 drivers/xen/xen-selfballoon.c                 |  7 +--
 fs/ceph/super.h                               |  3 +-
 fs/file_table.c                               |  9 ++--
 fs/fuse/inode.c                               |  4 +-
 fs/nfs/write.c                                |  3 +-
 fs/nfsd/nfscache.c                            |  3 +-
 fs/ntfs/malloc.h                              |  2 +-
 fs/proc/base.c                                |  3 +-
 include/linux/highmem.h                       |  2 +-
 include/linux/mm.h                            |  2 +-
 include/linux/mmzone.h                        | 10 +---
 include/linux/swap.h                          |  2 +-
 kernel/fork.c                                 |  6 +--
 kernel/kexec_core.c                           |  5 +-
 kernel/power/snapshot.c                       |  2 +-
 lib/show_mem.c                                |  3 +-
 mm/highmem.c                                  |  2 +-
 mm/huge_memory.c                              |  2 +-
 mm/kasan/quarantine.c                         |  4 +-
 mm/memblock.c                                 |  6 +--
 mm/memory_hotplug.c                           |  4 +-
 mm/mm_init.c                                  |  3 +-
 mm/oom_kill.c                                 |  2 +-
 mm/page_alloc.c                               | 75 ++++++++++++++-------------
 mm/shmem.c                                    | 12 +++--
 mm/slab.c                                     |  3 +-
 mm/swap.c                                     |  3 +-
 mm/util.c                                     |  2 +-
 mm/vmalloc.c                                  |  4 +-
 mm/vmstat.c                                   |  4 +-
 mm/workingset.c                               |  2 +-
 mm/zswap.c                                    |  2 +-
 net/dccp/proto.c                              |  6 +--
 net/decnet/dn_route.c                         |  2 +-
 net/ipv4/tcp_metrics.c                        |  2 +-
 net/netfilter/nf_conntrack_core.c             |  6 +--
 net/netfilter/xt_hashlimit.c                  |  4 +-
 net/sctp/protocol.c                           |  6 +--
 security/integrity/ima/ima_kexec.c            |  2 +-
 58 files changed, 171 insertions(+), 143 deletions(-)

arch/csky/mm/init.c +2 -2

index dc07c07..3f4d35e 100644
@@ -71,7 +71,7 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 		ClearPageReserved(virt_to_page(start));
 		init_page_count(virt_to_page(start));
 		free_page(start);
-		totalram_pages++;
+		atomic_long_inc(&totalram_pages);
 	}
 }
 #endif

@@ -88,7 +88,7 @@ void free_initmem(void)
 		ClearPageReserved(virt_to_page(addr));
 		init_page_count(virt_to_page(addr));
 		free_page(addr);
-		totalram_pages++;
+		atomic_long_inc(&totalram_pages);
 		addr += PAGE_SIZE;
 	}
 

arch/powerpc/platforms/pseries/cmm.c +6 -5

index 25427a4..85fe503 100644
@@ -208,7 +208,7 @@ static long cmm_alloc_pages(long nr)
 
 		pa->page[pa->index++] = addr;
 		loaned_pages++;
-		totalram_pages--;
+		atomic_long_dec(&totalram_pages);
 		spin_unlock(&cmm_lock);
 		nr--;
 	}

@@ -247,7 +247,7 @@ static long cmm_free_pages(long nr)
 		free_page(addr);
 		loaned_pages--;
 		nr--;
-		totalram_pages++;
+		atomic_long_inc(&totalram_pages);
 	}
 	spin_unlock(&cmm_lock);
 	cmm_dbg("End request with %ld pages unfulfilled\n", nr);

@@ -291,7 +291,8 @@ static void cmm_get_mpp(void)
 	int rc;
 	struct hvcall_mpp_data mpp_data;
 	signed long active_pages_target, page_loan_request, target;
-	signed long total_pages = totalram_pages + loaned_pages;
+	signed long total_pages = atomic_long_read(&totalram_pages) +
+				  loaned_pages;
 	signed long min_mem_pages = (min_mem_mb * 1024 * 1024) / PAGE_SIZE;
 
 	rc = h_get_mpp(&mpp_data);

@@ -322,7 +323,7 @@ static void cmm_get_mpp(void)
 
 	cmm_dbg("delta = %ld, loaned = %lu, target = %lu, oom = %lu, totalram = %lu\n",
 		page_loan_request, loaned_pages, loaned_pages_target,
-		oom_freed_pages, totalram_pages);
+		oom_freed_pages, atomic_long_read(&totalram_pages));
 }
 
 static struct notifier_block cmm_oom_nb = {

@@ -581,7 +582,7 @@ static int cmm_mem_going_offline(void *arg)
 			free_page(pa_curr->page[idx]);
 			freed++;
 			loaned_pages--;
-			totalram_pages++;
+			atomic_long_inc(&totalram_pages);
 			pa_curr->page[idx] = pa_last->page[--pa_last->index];
 			if (pa_last->index == 0) {
 				if (pa_curr == pa_last)

arch/s390/mm/init.c +1 -1

index 76d0708..d6529e8 100644
@@ -59,7 +59,7 @@ static void __init setup_zero_pages(void)
 	order = 7;
 
 	/* Limit number of empty zero pages for small memory sizes */
-	while (order > 2 && (totalram_pages >> 10) < (1UL << order))
+	while (order > 2 && (atomic_long_read(&totalram_pages) >> 10) < (1UL << order))
 		order--;
 
 	empty_zero_page = __get_free_pages(GFP_KERNEL | __GFP_ZERO, order);

arch/um/kernel/mem.c +2 -2

index 1067469..da78a06 100644
@@ -51,8 +51,8 @@ void __init mem_init(void)
 
 	/* this will put all low memory onto the freelists */
 	memblock_free_all();
-	max_low_pfn = totalram_pages;
-	max_pfn = totalram_pages;
+	max_low_pfn = atomic_long_read(&totalram_pages);
+	max_pfn = atomic_long_read(&totalram_pages);
 	mem_init_print_info(NULL);
 	kmalloc_ok = 1;
 }

arch/x86/kernel/cpu/microcode/core.c +3 -2

index 2637ff0..4ccc8dd 100644
@@ -435,8 +435,9 @@ static ssize_t microcode_write(struct file *file, const char __user *buf,
 {
 	ssize_t ret = -EINVAL;
 
-	if ((len >> PAGE_SHIFT) > totalram_pages) {
-		pr_err("too much data (max %ld pages)\n", totalram_pages);
+	if ((len >> PAGE_SHIFT) > atomic_long_read(&totalram_pages)) {
+		pr_err("too much data (max %ld pages)\n",
+				atomic_long_read(&totalram_pages));
 		return ret;
 	}
 

drivers/char/agp/backend.c +2 -2

index 38ffb28..2753e1d 100644
@@ -115,9 +115,9 @@ static int agp_find_max(void)
 	long memory, index, result;
 
 #if PAGE_SHIFT < 20
-	memory = totalram_pages >> (20 - PAGE_SHIFT);
+	memory = atomic_long_read(&totalram_pages) >> (20 - PAGE_SHIFT);
 #else
-	memory = totalram_pages << (PAGE_SHIFT - 20);
+	memory = atomic_long_read(&totalram_pages) << (PAGE_SHIFT - 20);
 #endif
 	index = 1;
 

drivers/gpu/drm/amd/amdkfd/kfd_crat.c +1 -1

index 56412b0..ca18502 100644
@@ -848,7 +848,7 @@ static int kfd_fill_mem_info_for_cpu(int numa_node_id, int *avail_size,
 	 */
 	pgdat = NODE_DATA(numa_node_id);
 	for (zone_type = 0; zone_type < MAX_NR_ZONES; zone_type++)
-		mem_in_bytes += pgdat->node_zones[zone_type].managed_pages;
+		mem_in_bytes += atomic_long_read(&pgdat->node_zones[zone_type].managed_pages);
 	mem_in_bytes <<= PAGE_SHIFT;
 
 	sub_type_hdr->length_low = lower_32_bits(mem_in_bytes);

drivers/gpu/drm/i915/i915_gem.c +1 -1

index 0c8aa57..b4c245b 100644
@@ -2539,7 +2539,7 @@ static int i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 	 * If there's no chance of allocating enough pages for the whole
 	 * object, bail early.
 	 */
-	if (page_count > totalram_pages)
+	if (page_count > atomic_long_read(&totalram_pages))
 		return -ENOMEM;
 
 	st = kmalloc(sizeof(*st), GFP_KERNEL);

drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +2 -2

index 8e2e269..9ea10eb 100644
@@ -170,7 +170,7 @@ static int igt_ppgtt_alloc(void *arg)
 	 * This should ensure that we do not run into the oomkiller during
 	 * the test and take down the machine wilfully.
 	 */
-	limit = totalram_pages << PAGE_SHIFT;
+	limit = atomic_long_read(&totalram_pages) << PAGE_SHIFT;
 	limit = min(ppgtt->vm.total, limit);
 
 	/* Check we can allocate the entire range */

@@ -1244,7 +1244,7 @@ static int exercise_mock(struct drm_i915_private *i915,
 				     u64 hole_start, u64 hole_end,
 				     unsigned long end_time))
 {
-	const u64 limit = totalram_pages << PAGE_SHIFT;
+	const u64 limit = atomic_long_read(&totalram_pages) << PAGE_SHIFT;
 	struct i915_gem_context *ctx;
 	struct i915_hw_ppgtt *ppgtt;
 	IGT_TIMEOUT(end_time);

drivers/hv/hv_balloon.c +10 -9

index c5bc0b5..4498c94 100644
@@ -1092,6 +1092,7 @@ static void process_info(struct hv_dynmem_device *dm, struct dm_info_msg *msg)
 static unsigned long compute_balloon_floor(void)
 {
 	unsigned long min_pages;
+	unsigned long totalram = (unsigned long)atomic_long_read(&totalram_pages);
 #define MB2PAGES(mb) ((mb) << (20 - PAGE_SHIFT))
 	/* Simple continuous piecewiese linear function:
 	 *  max MiB -> min MiB  gradient

@@ -1104,16 +1105,16 @@ static unsigned long compute_balloon_floor(void)
 	 *    8192       744    (1/16)
 	 *   32768      1512	(1/32)
 	 */
-	if (totalram_pages < MB2PAGES(128))
-		min_pages = MB2PAGES(8) + (totalram_pages >> 1);
-	else if (totalram_pages < MB2PAGES(512))
-		min_pages = MB2PAGES(40) + (totalram_pages >> 2);
-	else if (totalram_pages < MB2PAGES(2048))
-		min_pages = MB2PAGES(104) + (totalram_pages >> 3);
-	else if (totalram_pages < MB2PAGES(8192))
-		min_pages = MB2PAGES(232) + (totalram_pages >> 4);
+	if (totalram < MB2PAGES(128))
+		min_pages = MB2PAGES(8) + (totalram >> 1);
+	else if (totalram < MB2PAGES(512))
+		min_pages = MB2PAGES(40) + (totalram >> 2);
+	else if (totalram < MB2PAGES(2048))
+		min_pages = MB2PAGES(104) + (totalram >> 3);
+	else if (totalram < MB2PAGES(8192))
+		min_pages = MB2PAGES(232) + (totalram >> 4);
 	else
-		min_pages = MB2PAGES(488) + (totalram_pages >> 5);
+		min_pages = MB2PAGES(488) + (totalram >> 5);
 #undef MB2PAGES
 	return min_pages;
 }

drivers/md/dm-bufio.c +3 -2

index dc385b7..6d61259 100644
@@ -1887,8 +1887,9 @@ static int __init dm_bufio_init(void)
 	dm_bufio_allocated_vmalloc = 0;
 	dm_bufio_current_allocated = 0;
 
-	mem = (__u64)mult_frac(totalram_pages - totalhigh_pages,
-			       DM_BUFIO_MEMORY_PERCENT, 100) << PAGE_SHIFT;
+	mem = (__u64)mult_frac(atomic_long_read(&totalram_pages) -
+				atomic_long_read(&totalhigh_pages),
+				DM_BUFIO_MEMORY_PERCENT, 100) << PAGE_SHIFT;
 
 	if (mem > ULONG_MAX)
 		mem = ULONG_MAX;

drivers/md/dm-crypt.c +3 -1

index 0481223..1c58f4c 100644
@@ -2158,7 +2158,9 @@ static int crypt_wipe_key(struct crypt_config *cc)
 
 static void crypt_calculate_pages_per_client(void)
 {
-	unsigned long pages = (totalram_pages - totalhigh_pages) * DM_CRYPT_MEMORY_PERCENT / 100;
+	unsigned long pages = (atomic_long_read(&totalram_pages) -
+				atomic_long_read(&totalhigh_pages)) *
+				DM_CRYPT_MEMORY_PERCENT / 100;
 
 	if (!dm_crypt_clients_n)
 		return;

drivers/md/dm-integrity.c +3 -1

index bb3096b..d91c931 100644
@@ -2843,7 +2843,9 @@ static int create_journal(struct dm_integrity_c *ic, char **error)
 	journal_pages = roundup((__u64)ic->journal_sections * ic->journal_section_sectors,
 				PAGE_SIZE >> SECTOR_SHIFT) >> (PAGE_SHIFT - SECTOR_SHIFT);
 	journal_desc_size = journal_pages * sizeof(struct page_list);
-	if (journal_pages >= totalram_pages - totalhigh_pages || journal_desc_size > ULONG_MAX) {
+	if (journal_pages >= atomic_long_read(&totalram_pages) -
+			atomic_long_read(&totalhigh_pages) ||
+			journal_desc_size > ULONG_MAX) {
 		*error = "Journal doesn't fit into memory";
 		r = -ENOMEM;
 		goto bad;

drivers/md/dm-stats.c +2 -1

index 21de30b..f154a07 100644
@@ -85,7 +85,8 @@ static bool __check_shared_memory(size_t alloc_size)
 	a = shared_memory_amount + alloc_size;
 	if (a < shared_memory_amount)
 		return false;
-	if (a >> PAGE_SHIFT > totalram_pages / DM_STATS_MEMORY_FACTOR)
+	if (a >> PAGE_SHIFT > atomic_long_read(&totalram_pages) /
+					DM_STATS_MEMORY_FACTOR)
 		return false;
 #ifdef CONFIG_MMU
 	if (a > (VMALLOC_END - VMALLOC_START) / DM_STATS_VMALLOC_FACTOR)

drivers/media/platform/mtk-vpu/mtk_vpu.c +2 -1

index 616f78b..ee3654a 100644
@@ -855,7 +855,8 @@ static int mtk_vpu_probe(struct platform_device *pdev)
 	/* Set PTCM to 96K and DTCM to 32K */
 	vpu_cfg_writel(vpu, 0x2, VPU_TCM_CFG);
 
-	vpu->enable_4GB = !!(totalram_pages > (SZ_2G >> PAGE_SHIFT));
+	vpu->enable_4GB = !!(atomic_long_read(&totalram_pages) >
+					(SZ_2G >> PAGE_SHIFT));
 	dev_info(dev, "4GB mode %u\n", vpu->enable_4GB);
 
 	if (vpu->enable_4GB) {

drivers/misc/vmw_balloon.c +1 -1

index 9b0b3fa..0ac0fee 100644
@@ -570,7 +570,7 @@ static int vmballoon_send_get_target(struct vmballoon *b)
 	unsigned long status;
 	unsigned long limit;
 
-	limit = totalram_pages;
+	limit = atomic_long_read(&totalram_pages);
 
 	/* Ensure limit fits in 32-bits */
 	if (limit != (u32)limit)

drivers/parisc/ccio-dma.c +3 -2

index 6148236..705df1a 100644
@@ -1255,7 +1255,8 @@ void __init ccio_cujo20_fixup(struct parisc_device *cujo, u32 iovp)
 	** Hot-Plug/Removal of PCI cards. (aka PCI OLARD).
 	*/
 
-	iova_space_size = (u32) (totalram_pages / count_parisc_driver(&ccio_driver));
+	iova_space_size = (u32) (atomic_long_read(&totalram_pages) /
+				count_parisc_driver(&ccio_driver));
 
 	/* limit IOVA space size to 1MB-1GB */
 

@@ -1294,7 +1295,7 @@ void __init ccio_cujo20_fixup(struct parisc_device *cujo, u32 iovp)
 
 	DBG_INIT("%s() hpa 0x%p mem %luMB IOV %dMB (%d bits)\n",
 			__func__, ioc->ioc_regs,
-			(unsigned long) totalram_pages >> (20 - PAGE_SHIFT),
+			(unsigned long) atomic_long_read(&totalram_pages) >> (20 - PAGE_SHIFT),
 			iova_space_size>>20,
 			iov_order + PAGE_SHIFT);
 

drivers/parisc/sba_iommu.c +3 -2

index 11de0ec..02f4ce9 100644
@@ -1419,7 +1419,8 @@ static int setup_ibase_imask_callback(struct device *dev, void *data)
 	** for DMA hints - ergo only 30 bits max.
 	*/
 
-	iova_space_size = (u32) (totalram_pages/global_ioc_cnt);
+	iova_space_size = (u32) (atomic_long_read(&totalram_pages)/
+						global_ioc_cnt);
 
 	/* limit IOVA space size to 1MB-1GB */
 	if (iova_space_size < (1 << (20 - PAGE_SHIFT))) {

@@ -1444,7 +1445,7 @@ static int setup_ibase_imask_callback(struct device *dev, void *data)
 	DBG_INIT("%s() hpa 0x%lx mem %ldMB IOV %dMB (%d bits)\n",
 			__func__,
 			ioc->ioc_hpa,
-			(unsigned long) totalram_pages >> (20 - PAGE_SHIFT),
+			(unsigned long) atomic_long_read(&totalram_pages) >> (20 - PAGE_SHIFT),
 			iova_space_size>>20,
 			iov_order + PAGE_SHIFT);
 

drivers/staging/android/ion/ion_system_heap.c +1 -1

index 548bb02..64bd925 100644
@@ -110,7 +110,7 @@ static int ion_system_heap_allocate(struct ion_heap *heap,
 	unsigned long size_remaining = PAGE_ALIGN(size);
 	unsigned int max_order = orders[0];
 
-	if (size / PAGE_SIZE > totalram_pages / 2)
+	if (size / PAGE_SIZE > atomic_long_read(&totalram_pages) / 2)
 		return -ENOMEM;
 
 	INIT_LIST_HEAD(&pages);

drivers/xen/xen-selfballoon.c +4 -3

index 5165aa8..0b925fd 100644
@@ -189,7 +189,7 @@ static void selfballoon_process(struct work_struct *work)
 	bool reset_timer = false;
 
 	if (xen_selfballooning_enabled) {
-		cur_pages = totalram_pages;
+		cur_pages = atomic_long_read(&totalram_pages);
 		tgt_pages = cur_pages; /* default is no change */
 		goal_pages = vm_memory_committed() +
 				totalreserve_pages +

@@ -227,7 +227,8 @@ static void selfballoon_process(struct work_struct *work)
 		if (tgt_pages < floor_pages)
 			tgt_pages = floor_pages;
 		balloon_set_new_target(tgt_pages +
-			balloon_stats.current_pages - totalram_pages);
+			balloon_stats.current_pages -
+			atomic_long_read(&totalram_pages));
 		reset_timer = true;
 	}
 #ifdef CONFIG_FRONTSWAP

@@ -569,7 +570,7 @@ int xen_selfballoon_init(bool use_selfballooning, bool use_frontswap_selfshrink)
 	 * much more reliably and response faster in some cases.
 	 */
 	if (!selfballoon_reserved_mb) {
-		reserve_pages = totalram_pages / 10;
+		reserve_pages = atomic_long_read(&totalram_pages) / 10;
 		selfballoon_reserved_mb = PAGES2MB(reserve_pages);
 	}
 	schedule_delayed_work(&selfballoon_worker, selfballoon_interval * HZ);

fs/ceph/super.h +2 -1

index 582e28f..92f56d3 100644
@@ -807,7 +807,8 @@ static inline int default_congestion_kb(void)
 	 * This allows larger machines to have larger/more transfers.
 	 * Limit the default to 256M
 	 */
-	congestion_kb = (16*int_sqrt(totalram_pages)) << (PAGE_SHIFT-10);
+	congestion_kb = (16*int_sqrt(atomic_long_read(&totalram_pages))) <<
+								(PAGE_SHIFT-10);
 	if (congestion_kb > 256*1024)
 		congestion_kb = 256*1024;
 

fs/file_table.c +6 -3

index e03c8d1..5dde5c3 100644
@@ -383,10 +383,13 @@ void __init files_init(void)
 void __init files_maxfiles_init(void)
 {
 	unsigned long n;
-	unsigned long memreserve = (totalram_pages - nr_free_pages()) * 3/2;
+	unsigned long memreserve = (atomic_long_read(&totalram_pages) -
+						nr_free_pages()) * 3/2;
 
-	memreserve = min(memreserve, totalram_pages - 1);
-	n = ((totalram_pages - memreserve) * (PAGE_SIZE / 1024)) / 10;
+	memreserve = min(memreserve,
+			(unsigned long)atomic_long_read(&totalram_pages) - 1);
+	n = ((atomic_long_read(&totalram_pages) - memreserve) *
+					(PAGE_SIZE / 1024)) / 10;
 
 	files_stat.max_files = max_t(unsigned long, n, NR_FILE);
 }

fs/fuse/inode.c +2 -2

index 4727ef6..acdbaf7 100644
@@ -825,8 +825,8 @@ static struct dentry *fuse_get_parent(struct dentry *child)
 static void sanitize_global_limit(unsigned *limit)
 {
 	if (*limit == 0)
-		*limit = ((totalram_pages << PAGE_SHIFT) >> 13) /
-			 sizeof(struct fuse_req);
+		*limit = ((atomic_long_read(&totalram_pages)
+			 << PAGE_SHIFT) >> 13) / sizeof(struct fuse_req);
 
 	if (*limit >= 1 << 16)
 		*limit = (1 << 16) - 1;

fs/nfs/write.c +2 -1

index 586726a..e3663b7 100644
@@ -2121,7 +2121,8 @@ int __init nfs_init_writepagecache(void)
 	 * This allows larger machines to have larger/more transfers.
 	 * Limit the default to 256M
 	 */
-	nfs_congestion_kb = (16*int_sqrt(totalram_pages)) << (PAGE_SHIFT-10);
+	nfs_congestion_kb = (16*int_sqrt(atomic_long_read(&totalram_pages))) <<
+								(PAGE_SHIFT-10);
 	if (nfs_congestion_kb > 256*1024)
 		nfs_congestion_kb = 256*1024;
 

fs/nfsd/nfscache.c +2 -1

index e2fe0e9..e877558 100644
@@ -99,7 +99,8 @@ static unsigned long nfsd_reply_cache_scan(struct shrinker *shrink,
 nfsd_cache_size_limit(void)
 {
 	unsigned int limit;
-	unsigned long low_pages = totalram_pages - totalhigh_pages;
+	unsigned long low_pages = atomic_long_read(&totalram_pages) -
+					atomic_long_read(&totalhigh_pages);
 
 	limit = (16 * int_sqrt(low_pages)) << (PAGE_SHIFT-10);
 	return min_t(unsigned int, limit, 256*1024);

fs/ntfs/malloc.h +1 -1

index ab172e5..4ae6bbe 100644
@@ -47,7 +47,7 @@ static inline void *__ntfs_malloc(unsigned long size, gfp_t gfp_mask)
 		return kmalloc(PAGE_SIZE, gfp_mask & ~__GFP_HIGHMEM);
 		/* return (void *)__get_free_page(gfp_mask); */
 	}
-	if (likely((size >> PAGE_SHIFT) < totalram_pages))
+	if (likely((size >> PAGE_SHIFT) < atomic_long_read(&totalram_pages)))
 		return __vmalloc(size, gfp_mask, PAGE_KERNEL);
 	return NULL;
 }

fs/proc/base.c +2 -1

index ce34654..9ef26dc 100644
@@ -530,7 +530,8 @@ static ssize_t lstats_write(struct file *file, const char __user *buf,
 static int proc_oom_score(struct seq_file *m, struct pid_namespace *ns,
 			  struct pid *pid, struct task_struct *task)
 {
-	unsigned long totalpages = totalram_pages + total_swap_pages;
+	unsigned long totalpages = atomic_long_read(&totalram_pages) +
+							total_swap_pages;
 	unsigned long points = 0;
 
 	points = oom_badness(task, NULL, NULL, totalpages) *

include/linux/highmem.h +1 -1

index 0690679..84edaa2 100644
@@ -36,7 +36,7 @@ static inline void invalidate_kernel_vmap_range(void *vaddr, int size)
 
 /* declarations for linux/mm/highmem.c */
 unsigned int nr_free_highpages(void);
-extern unsigned long totalhigh_pages;
+extern atomic_long_t totalhigh_pages;
 
 void kmap_flush_unused(void);
 

include/linux/mm.h +1 -1

index fcf9cc9..af952fc 100644
@@ -48,7 +48,7 @@ static inline void set_max_mapnr(unsigned long limit)
 static inline void set_max_mapnr(unsigned long limit) { }
 #endif
 
-extern unsigned long totalram_pages;
+extern atomic_long_t totalram_pages;
 extern void * high_memory;
 extern int page_cluster;
 

include/linux/mmzone.h +2 -8

index 8555509..2639b05 100644
@@ -428,14 +428,8 @@ struct zone {
 	 * Write access to present_pages at runtime should be protected by
 	 * mem_hotplug_begin/end(). Any reader who can't tolerant drift of
 	 * present_pages should get_online_mems() to get a stable value.
-	 *
-	 * Read access to managed_pages should be safe because it's unsigned
-	 * long. Write access to zone->managed_pages and totalram_pages are
-	 * protected by managed_page_count_lock at runtime. Idealy only
-	 * adjust_managed_page_count() should be used instead of directly
-	 * touching zone->managed_pages and totalram_pages.
 	 */
-	unsigned long		managed_pages;
+	atomic_long_t		managed_pages;
 	unsigned long		spanned_pages;
 	unsigned long		present_pages;
 

@@ -814,7 +808,7 @@ static inline bool is_dev_zone(const struct zone *zone)
  */
 static inline bool managed_zone(struct zone *zone)
 {
-	return zone->managed_pages;
+	return atomic_long_read(&zone->managed_pages);
 }
 
 /* Returns true if a zone has memory */

include/linux/swap.h +1 -1

index d098743..b34c6e7 100644
@@ -309,7 +309,7 @@ struct vma_swap_readahead {
 } while (0)
 
 /* linux/mm/page_alloc.c */
-extern unsigned long totalram_pages;
+extern atomic_long_t totalram_pages;
 extern unsigned long totalreserve_pages;
 extern unsigned long nr_free_buffer_pages(void);
 extern unsigned long nr_free_pagecache_pages(void);

kernel/fork.c +3 -3

index 2f78d32..b6068c5 100644
@@ -744,11 +744,11 @@ static void set_max_threads(unsigned int max_threads_suggested)
 	 * The number of threads shall be limited such that the thread
 	 * structures may only consume a small part of the available memory.
 	 */
-	if (fls64(totalram_pages) + fls64(PAGE_SIZE) > 64)
+	if (fls64(atomic_long_read(&totalram_pages)) + fls64(PAGE_SIZE) > 64)
 		threads = MAX_THREADS;
 	else
-		threads = div64_u64((u64) totalram_pages * (u64) PAGE_SIZE,
-				    (u64) THREAD_SIZE * 8UL);
+		threads = div64_u64((u64) atomic_long_read(&totalram_pages) *
+				(u64) PAGE_SIZE, (u64) THREAD_SIZE * 8UL);
 
 	if (threads > max_threads_suggested)
 		threads = max_threads_suggested;

kernel/kexec_core.c +3 -2

index 86ef06d..ed85ddd 100644
@@ -217,13 +217,14 @@ int sanity_check_segment_list(struct kimage *image)
 	 * wasted allocating pages, which can cause a soft lockup.
 	 */
 	for (i = 0; i < nr_segments; i++) {
-		if (PAGE_COUNT(image->segment[i].memsz) > totalram_pages / 2)
+		if (PAGE_COUNT(image->segment[i].memsz) >
+				atomic_long_read(&totalram_pages) / 2)
 			return -EINVAL;
 
 		total_pages += PAGE_COUNT(image->segment[i].memsz);
 	}
 
-	if (total_pages > totalram_pages / 2)
+	if (total_pages > atomic_long_read(&totalram_pages) / 2)
 		return -EINVAL;
 
 	/*

kernel/power/snapshot.c +1 -1

index b0308a2..142a3c76 100644
@@ -105,7 +105,7 @@ void __init hibernate_reserved_size_init(void)
 
 void __init hibernate_image_size_init(void)
 {
-	image_size = ((totalram_pages * 2) / 5) * PAGE_SIZE;
+	image_size = ((atomic_long_read(&totalram_pages) * 2) / 5) * PAGE_SIZE;
 }
 
 /*

lib/show_mem.c +2 -1

index 0beaa1d..0701f63 100644
@@ -28,7 +28,8 @@ void show_mem(unsigned int filter, nodemask_t *nodemask)
 				continue;
 
 			total += zone->present_pages;
-			reserved += zone->present_pages - zone->managed_pages;
+			reserved += zone->present_pages -
+				atomic_long_read(&zone->managed_pages);
 
 			if (is_highmem_idx(zoneid))
 				highmem += zone->present_pages;

mm/highmem.c +1 -1

index 59db322..93a45c0 100644
@@ -105,7 +105,7 @@ static inline wait_queue_head_t *get_pkmap_wait_queue_head(unsigned int color)
 }
 #endif
 
-unsigned long totalhigh_pages __read_mostly;
+atomic_long_t totalhigh_pages __read_mostly;
 EXPORT_SYMBOL(totalhigh_pages);
 
 

mm/huge_memory.c +1 -1

index d394d18..f2f18b5 100644
@@ -420,7 +420,7 @@ static int __init hugepage_init(void)
 	 * where the extra memory used could hurt more than TLB overhead
 	 * is likely to save.  The admin can still enable it through /sys.
 	 */
-	if (totalram_pages < (512 << (20 - PAGE_SHIFT))) {
+	if (atomic_long_read(&totalram_pages) < (512 << (20 - PAGE_SHIFT))) {
 		transparent_hugepage_flags = 0;
 		return 0;
 	}

mm/kasan/quarantine.c +2 -2

index b209dba..4d36aed 100644
@@ -236,8 +236,8 @@ void quarantine_reduce(void)
 	 * Update quarantine size in case of hotplug. Allocate a fraction of
 	 * the installed memory to quarantine minus per-cpu queue limits.
 	 */
-	total_size = (READ_ONCE(totalram_pages) << PAGE_SHIFT) /
-		QUARANTINE_FRACTION;
+	total_size = (READ_ONCE(atomic_long_read(&totalram_pages)) <<
+			PAGE_SHIFT) / QUARANTINE_FRACTION;
 	percpu_quarantines = QUARANTINE_PERCPU_SIZE * num_online_cpus();
 	new_quarantine_size = (total_size < percpu_quarantines) ?
 		0 : total_size - percpu_quarantines;

mm/memblock.c +3 -3

index eddcac2..43f53e9 100644
@@ -1627,7 +1627,7 @@ void __init __memblock_free_late(phys_addr_t base, phys_addr_t size)
 
 	for (; cursor < end; cursor++) {
 		memblock_free_pages(pfn_to_page(cursor), cursor, 0);
-		totalram_pages++;
+		atomic_long_inc(&totalram_pages);
 	}
 }
 

@@ -2001,7 +2001,7 @@ void reset_node_managed_pages(pg_data_t *pgdat)
 	struct zone *z;
 
 	for (z = pgdat->node_zones; z < pgdat->node_zones + MAX_NR_ZONES; z++)
-		z->managed_pages = 0;
+		atomic_long_set(&z->managed_pages, 0);
 }
 
 void __init reset_all_zones_managed_pages(void)

@@ -2029,7 +2029,7 @@ unsigned long __init memblock_free_all(void)
 	reset_all_zones_managed_pages();
 
 	pages = free_low_memory_core_early();
-	totalram_pages += pages;
+	atomic_long_add(pages, &totalram_pages);
 
 	return pages;
 }

mm/memory_hotplug.c +2 -2

index dbbb945..0725984 100644
@@ -657,10 +657,10 @@ void __online_page_free(struct page *page)
 static int generic_online_page(struct page *page, unsigned int order)
 {
 	__free_pages_core(page, order);
-	totalram_pages += (1UL << order);
+	atomic_long_add((1UL << order), &totalram_pages);
 #ifdef CONFIG_HIGHMEM
 	if (PageHighMem(page))
-		totalhigh_pages += (1UL << order);
+		atomic_long_add((1UL << order), &totalhigh_pages);
 #endif
 	return 0;
 }

mm/mm_init.c +2 -1

index 6838a53..93a6611 100644
@@ -146,7 +146,8 @@ static void __meminit mm_compute_batch(void)
 	s32 batch = max_t(s32, nr*2, 32);
 
 	/* batch size set to 0.4% of (total memory/#cpus), or max int32 */
-	memsized_batch = min_t(u64, (totalram_pages/nr)/256, 0x7fffffff);
+	memsized_batch = min_t(u64, (atomic_long_read(&totalram_pages)/nr)/256,
+								0x7fffffff);
 
 	vm_committed_as_batch = max_t(s32, memsized_batch, batch);
 }

mm/oom_kill.c +1 -1

index 6589f60..1a37d68 100644
@@ -269,7 +269,7 @@ static enum oom_constraint constrained_alloc(struct oom_control *oc)
 	}
 
 	/* Default to all available memory */
-	oc->totalpages = totalram_pages + total_swap_pages;
+	oc->totalpages = atomic_long_read(&totalram_pages) + total_swap_pages;
 
 	if (!IS_ENABLED(CONFIG_NUMA))
 		return CONSTRAINT_NONE;

mm/page_alloc.c +39 -36

index 4bd858d..c7b26e3 100644
@@ -121,10 +121,7 @@
 };
 EXPORT_SYMBOL(node_states);
 
-/* Protect totalram_pages and zone->managed_pages */
-static DEFINE_SPINLOCK(managed_page_count_lock);
-
-unsigned long totalram_pages __read_mostly;
+atomic_long_t totalram_pages __read_mostly;
 unsigned long totalreserve_pages __read_mostly;
 unsigned long totalcma_pages __read_mostly;
 

@@ -1275,7 +1272,7 @@ void __free_pages_core(struct page *page, unsigned int order)
 		set_page_count(p, 0);
 	}
 
-	page_zone(page)->managed_pages += nr_pages;
+	atomic_long_add(nr_pages, &page_zone(page)->managed_pages);
 	set_page_refcounted(page);
 	__free_pages(page, order);
 }

@@ -2254,7 +2251,8 @@ static void reserve_highatomic_pageblock(struct page *page, struct zone *zone,
 	 * Limit the number reserved to 1 pageblock or roughly 1% of a zone.
 	 * Check is race-prone but harmless.
 	 */
-	max_managed = (zone->managed_pages / 100) + pageblock_nr_pages;
+	max_managed = (atomic_long_read(&zone->managed_pages) / 100) +
+						pageblock_nr_pages;
 	if (zone->nr_reserved_highatomic >= max_managed)
 		return;
 

@@ -4658,7 +4656,7 @@ static unsigned long nr_free_zone_pages(int offset)
 	struct zonelist *zonelist = node_zonelist(numa_node_id(), GFP_KERNEL);
 
 	for_each_zone_zonelist(zone, z, zonelist, offset) {
-		unsigned long size = zone->managed_pages;
+		unsigned long size = atomic_long_read(&zone->managed_pages);
 		unsigned long high = high_wmark_pages(zone);
 		if (size > high)
 			sum += size - high;

@@ -4744,11 +4742,15 @@ long si_mem_available(void)
 
 void si_meminfo(struct sysinfo *val)
 {
-	val->totalram = totalram_pages;
+	val->totalram = atomic_long_read(&totalram_pages);
 	val->sharedram = global_node_page_state(NR_SHMEM);
 	val->freeram = global_zone_page_state(NR_FREE_PAGES);
 	val->bufferram = nr_blockdev_pages();
-	val->totalhigh = totalhigh_pages;
+#ifdef CONFIG_HIGHMEM
+	val->totalhigh = atomic_long_read(&totalhigh_pages);
+#else
+	val->totalhigh = 0;
+#endif
 	val->freehigh = nr_free_highpages();
 	val->mem_unit = PAGE_SIZE;
 }

@@ -4765,7 +4767,7 @@ void si_meminfo_node(struct sysinfo *val, int nid)
 	pg_data_t *pgdat = NODE_DATA(nid);
 
 	for (zone_type = 0; zone_type < MAX_NR_ZONES; zone_type++)
-		managed_pages += pgdat->node_zones[zone_type].managed_pages;
+		managed_pages += atomic_long_read(&pgdat->node_zones[zone_type].managed_pages);
 	val->totalram = managed_pages;
 	val->sharedram = node_page_state(pgdat, NR_SHMEM);
 	val->freeram = sum_zone_node_page_state(nid, NR_FREE_PAGES);

@@ -4774,7 +4776,7 @@ void si_meminfo_node(struct sysinfo *val, int nid)
 		struct zone *zone = &pgdat->node_zones[zone_type];
 
 		if (is_highmem(zone)) {
-			managed_highpages += zone->managed_pages;
+			managed_highpages += atomic_long_read(&zone->managed_pages);
 			free_highpages += zone_page_state(zone, NR_FREE_PAGES);
 		}
 	}

@@ -4981,7 +4983,7 @@ void show_free_areas(unsigned int filter, nodemask_t *nodemask)
 			K(zone_page_state(zone, NR_ZONE_UNEVICTABLE)),
 			K(zone_page_state(zone, NR_ZONE_WRITE_PENDING)),
 			K(zone->present_pages),
-			K(zone->managed_pages),
+			K(atomic_long_read(&zone->managed_pages)),
 			K(zone_page_state(zone, NR_MLOCK)),
 			zone_page_state(zone, NR_KERNEL_STACK_KB),
 			K(zone_page_state(zone, NR_PAGETABLE)),

@@ -5643,7 +5645,7 @@ static int zone_batchsize(struct zone *zone)
 	 * The per-cpu-pages pools are set to around 1000th of the
 	 * size of the zone.
 	 */
-	batch = zone->managed_pages / 1024;
+	batch = atomic_long_read(&zone->managed_pages) / 1024;
 	/* But no more than a meg. */
 	if (batch * PAGE_SIZE > 1024 * 1024)
 		batch = (1024 * 1024) / PAGE_SIZE;

@@ -5754,7 +5756,7 @@ static void pageset_set_high_and_batch(struct zone *zone,
 {
 	if (percpu_pagelist_fraction)
 		pageset_set_high(pcp,
-			(zone->managed_pages /
+			(atomic_long_read(&zone->managed_pages) /
 				percpu_pagelist_fraction));
 	else
 		pageset_set_batch(pcp, zone_batchsize(zone));

@@ -6309,7 +6311,7 @@ static void __meminit pgdat_init_internals(struct pglist_data *pgdat)
 static void __meminit zone_init_internals(struct zone *zone, enum zone_type idx, int nid,
 							unsigned long remaining_pages)
 {
-	zone->managed_pages = remaining_pages;
+	atomic_long_set(&zone->managed_pages, remaining_pages);
 	zone_set_nid(zone, nid);
 	zone->name = zone_names[idx];
 	zone->zone_pgdat = NODE_DATA(nid);

@@ -7061,14 +7063,12 @@ static int __init cmdline_parse_movablecore(char *p)
 
 void adjust_managed_page_count(struct page *page, long count)
 {
-	spin_lock(&managed_page_count_lock);
-	page_zone(page)->managed_pages += count;
-	totalram_pages += count;
+	atomic_long_add(count, &page_zone(page)->managed_pages);
+	atomic_long_add(count, &totalram_pages);
 #ifdef CONFIG_HIGHMEM
 	if (PageHighMem(page))
-		totalhigh_pages += count;
+		atomic_long_add(count, &totalhigh_pages);
 #endif
-	spin_unlock(&managed_page_count_lock);
 }
 EXPORT_SYMBOL(adjust_managed_page_count);
 

@@ -7109,9 +7109,9 @@ unsigned long free_reserved_area(void *start, void *end, int poison, char *s)
 void free_highmem_page(struct page *page)
 {
 	__free_reserved_page(page);
-	totalram_pages++;
-	page_zone(page)->managed_pages++;
-	totalhigh_pages++;
+	atomic_long_inc(&totalram_pages);
+	atomic_long_inc(&page_zone(page)->managed_pages);
+	atomic_long_inc(&totalhigh_pages);
 }
 #endif
 

@@ -7160,10 +7160,10 @@ void __init mem_init_print_info(const char *str)
 		physpages << (PAGE_SHIFT - 10),
 		codesize >> 10, datasize >> 10, rosize >> 10,
 		(init_data_size + init_code_size) >> 10, bss_size >> 10,
-		(physpages - totalram_pages - totalcma_pages) << (PAGE_SHIFT - 10),
+		(physpages - atomic_long_read(&totalram_pages) - totalcma_pages) << (PAGE_SHIFT - 10),
 		totalcma_pages << (PAGE_SHIFT - 10),
 #ifdef	CONFIG_HIGHMEM
-		totalhigh_pages << (PAGE_SHIFT - 10),
+		atomic_long_read(&totalhigh_pages) << (PAGE_SHIFT - 10),
 #endif
 		str ? ", " : "", str ? str : "");
 }

@@ -7253,8 +7253,8 @@ static void calculate_totalreserve_pages(void)
 			/* we treat the high watermark as reserved pages. */
 			max += high_wmark_pages(zone);
 
-			if (max > zone->managed_pages)
-				max = zone->managed_pages;
+			if (max > atomic_long_read(&zone->managed_pages))
+				max = atomic_long_read(&zone->managed_pages);
 
 			pgdat->totalreserve_pages += max;
 

@@ -7278,7 +7278,7 @@ static void setup_per_zone_lowmem_reserve(void)
 	for_each_online_pgdat(pgdat) {
 		for (j = 0; j < MAX_NR_ZONES; j++) {
 			struct zone *zone = pgdat->node_zones + j;
-			unsigned long managed_pages = zone->managed_pages;
+			unsigned long managed_pages = atomic_long_read(&zone->managed_pages);
 
 			zone->lowmem_reserve[j] = 0;
 

@@ -7296,7 +7296,7 @@ static void setup_per_zone_lowmem_reserve(void)
 					lower_zone->lowmem_reserve[j] =
 						managed_pages / sysctl_lowmem_reserve_ratio[idx];
 				}
-				managed_pages += lower_zone->managed_pages;
+				managed_pages += atomic_long_read(&lower_zone->managed_pages);
 			}
 		}
 	}

@@ -7315,14 +7315,14 @@ static void __setup_per_zone_wmarks(void)
 	/* Calculate total number of !ZONE_HIGHMEM pages */
 	for_each_zone(zone) {
 		if (!is_highmem(zone))
-			lowmem_pages += zone->managed_pages;
+			lowmem_pages += atomic_long_read(&zone->managed_pages);
 	}
 
 	for_each_zone(zone) {
 		u64 tmp;
 
 		spin_lock_irqsave(&zone->lock, flags);
-		tmp = (u64)pages_min * zone->managed_pages;
+		tmp = (u64)pages_min * atomic_long_read(&zone->managed_pages);
 		do_div(tmp, lowmem_pages);
 		if (is_highmem(zone)) {
 			/*

@@ -7336,7 +7336,8 @@ static void __setup_per_zone_wmarks(void)
 			 */
 			unsigned long min_pages;
 
-			min_pages = zone->managed_pages / 1024;
+			min_pages = atomic_long_read(&zone->managed_pages) /
+									1024;
 			min_pages = clamp(min_pages, SWAP_CLUSTER_MAX, 128UL);
 			zone->watermark[WMARK_MIN] = min_pages;
 		} else {

@@ -7353,7 +7354,7 @@ static void __setup_per_zone_wmarks(void)
 		 * ensure a minimum size on small systems.
 		 */
 		tmp = max_t(u64, tmp >> 2,
-			    mult_frac(zone->managed_pages,
+			    mult_frac(atomic_long_read(&zone->managed_pages),
 				      watermark_scale_factor, 10000));
 
 		zone->watermark[WMARK_LOW]  = min_wmark_pages(zone) + tmp;

@@ -7483,7 +7484,8 @@ static void setup_min_unmapped_ratio(void)
 		pgdat->min_unmapped_pages = 0;
 
 	for_each_zone(zone)
-		zone->zone_pgdat->min_unmapped_pages += (zone->managed_pages *
+		zone->zone_pgdat->min_unmapped_pages +=
+				(atomic_long_read(&zone->managed_pages) *
 				sysctl_min_unmapped_ratio) / 100;
 }
 

@@ -7511,8 +7513,9 @@ static void setup_min_slab_ratio(void)
 		pgdat->min_slab_pages = 0;
 
 	for_each_zone(zone)
-		zone->zone_pgdat->min_slab_pages += (zone->managed_pages *
-				sysctl_min_slab_ratio) / 100;
+		zone->zone_pgdat->min_slab_pages +=
+			(atomic_long_read(&zone->managed_pages) *
+			sysctl_min_slab_ratio) / 100;
 }
 
 int sysctl_min_slab_ratio_sysctl_handler(struct ctl_table *table, int write,

mm/shmem.c +9 -3

index a6964ba..edd55db 100644
@@ -109,12 +109,18 @@ struct shmem_falloc {
 #ifdef CONFIG_TMPFS
 static unsigned long shmem_default_max_blocks(void)
 {
-	return totalram_pages / 2;
+	return atomic_long_read(&totalram_pages) / 2;
 }
 
 static unsigned long shmem_default_max_inodes(void)
 {
-	return min(totalram_pages - totalhigh_pages, totalram_pages / 2);
+	return min((unsigned long)atomic_long_read(&totalram_pages) -
+#ifdef CONFIG_HIGHMEM
+		(unsigned long) atomic_long_read(&totalhigh_pages),
+#else
+		0,
+#endif
+		(unsigned long)atomic_long_read(&totalram_pages) / 2);
 }
 #endif
 

@@ -3274,7 +3280,7 @@ static int shmem_parse_options(char *options, struct shmem_sb_info *sbinfo,
 			size = memparse(value,&rest);
 			if (*rest == '%') {
 				size <<= PAGE_SHIFT;
-				size *= totalram_pages;
+				size *= atomic_long_read(&totalram_pages);
 				do_div(size, 100);
 				rest++;
 			}

mm/slab.c +2 -1

index 2a5654b..70252b0 100644
@@ -1248,7 +1248,8 @@ void __init kmem_cache_init(void)
 	 * page orders on machines with more than 32MB of memory if
 	 * not overridden on the command line.
 	 */
-	if (!slab_max_order_set && totalram_pages > (32 << 20) >> PAGE_SHIFT)
+	if (!slab_max_order_set && atomic_long_read(&totalram_pages) >
+						(32 << 20) >> PAGE_SHIFT)
 		slab_max_order = SLAB_MAX_ORDER_HI;
 
 	/* Bootstrap is tricky, because several objects are allocated

mm/swap.c +2 -1

index aa48371..e85bc4a 100644
@@ -1023,7 +1023,8 @@ unsigned pagevec_lookup_range_nr_tag(struct pagevec *pvec,
  */
 void __init swap_setup(void)
 {
-	unsigned long megs = totalram_pages >> (20 - PAGE_SHIFT);
+	unsigned long megs = atomic_long_read(&totalram_pages) >>
+						(20 - PAGE_SHIFT);
 
 	/* Use a smaller cluster for small-memory machines */
 	if (megs < 16)

mm/util.c +1 -1

index 7f1f165..a3ae8ee 100644
@@ -600,7 +600,7 @@ unsigned long vm_commit_limit(void)
 	if (sysctl_overcommit_kbytes)
 		allowed = sysctl_overcommit_kbytes >> (PAGE_SHIFT - 10);
 	else
-		allowed = ((totalram_pages - hugetlb_total_pages())
+		allowed = ((atomic_long_read(&totalram_pages) - hugetlb_total_pages())
 			   * sysctl_overcommit_ratio / 100);
 	allowed += total_swap_pages;
 

mm/vmalloc.c +2 -2

index 97d4b25..f177af8 100644
@@ -1634,7 +1634,7 @@ void *vmap(struct page **pages, unsigned int count,
 
 	might_sleep();
 
-	if (count > totalram_pages)
+	if (count > atomic_long_read(&totalram_pages))
 		return NULL;
 
 	size = (unsigned long)count << PAGE_SHIFT;

@@ -1739,7 +1739,7 @@ void *__vmalloc_node_range(unsigned long size, unsigned long align,
 	unsigned long real_size = size;
 
 	size = PAGE_ALIGN(size);
-	if (!size || (size >> PAGE_SHIFT) > totalram_pages)
+	if (!size || (size >> PAGE_SHIFT) > atomic_long_read(&totalram_pages))
 		goto fail;
 
 	area = __get_vm_area_node(size, align, VM_ALLOC | VM_UNINITIALIZED |

mm/vmstat.c +2 -2

index 6038ce5..20551e8 100644
@@ -227,7 +227,7 @@ int calculate_normal_threshold(struct zone *zone)
 	 * 125		1024		10	16-32 GB	9
 	 */
 
-	mem = zone->managed_pages >> (27 - PAGE_SHIFT);
+	mem = atomic_long_read(&zone->managed_pages) >> (27 - PAGE_SHIFT);
 
 	threshold = 2 * fls(num_online_cpus()) * (1 + fls(mem));
 

@@ -1569,7 +1569,7 @@ static void zoneinfo_show_print(struct seq_file *m, pg_data_t *pgdat,
 		   high_wmark_pages(zone),
 		   zone->spanned_pages,
 		   zone->present_pages,
-		   zone->managed_pages);
+		   atomic_long_read(&zone->managed_pages));
 
 	seq_printf(m,
 		   "\n        protection: (%ld",

mm/workingset.c +1 -1

index b15799d..dcd4e16 100644
@@ -550,7 +550,7 @@ static int __init workingset_init(void)
 	 * double the initial memory by using totalram_pages as-is.
 	 */
 	timestamp_bits = BITS_PER_LONG - EVICTION_SHIFT;
-	max_order = fls_long(totalram_pages - 1);
+	max_order = fls_long(atomic_long_read(&totalram_pages) - 1);
 	if (max_order > timestamp_bits)
 		bucket_order = max_order - timestamp_bits;
 	pr_info("workingset: timestamp_bits=%d max_order=%d bucket_order=%u\n",

mm/zswap.c +1 -1

index cd91fd9..5d2d7b9 100644
@@ -219,7 +219,7 @@ struct zswap_tree {
 
 static bool zswap_is_full(void)
 {
-	return totalram_pages * zswap_max_pool_percent / 100 <
+	return atomic_long_read(&totalram_pages) * zswap_max_pool_percent / 100 <
 		DIV_ROUND_UP(zswap_pool_total_size, PAGE_SIZE);
 }
 

net/dccp/proto.c +3 -3

index 875858c..4a92d11 100644
@@ -1154,10 +1154,10 @@ static int __init dccp_init(void)
 	 *
 	 * The methodology is similar to that of the buffer cache.
 	 */
-	if (totalram_pages >= (128 * 1024))
-		goal = totalram_pages >> (21 - PAGE_SHIFT);
+	if (atomic_long_read(&totalram_pages) >= (128 * 1024))
+		goal = atomic_long_read(&totalram_pages) >> (21 - PAGE_SHIFT);
 	else
-		goal = totalram_pages >> (23 - PAGE_SHIFT);
+		goal = atomic_long_read(&totalram_pages) >> (23 - PAGE_SHIFT);
 
 	if (thash_entries)
 		goal = (thash_entries *

net/decnet/dn_route.c +1 -1

index 1c002c0..bb49b0f 100644
@@ -1866,7 +1866,7 @@ void __init dn_route_init(void)
 	dn_route_timer.expires = jiffies + decnet_dst_gc_interval * HZ;
 	add_timer(&dn_route_timer);
 
-	goal = totalram_pages >> (26 - PAGE_SHIFT);
+	goal = atomic_long_read(&totalram_pages) >> (26 - PAGE_SHIFT);
 
 	for(order = 0; (1UL << order) < goal; order++)
 		/* NOTHING */;

net/ipv4/tcp_metrics.c +1 -1

index 03b51cd..d91bdab 100644
@@ -1000,7 +1000,7 @@ static int __net_init tcp_net_metrics_init(struct net *net)
 
 	slots = tcpmhash_entries;
 	if (!slots) {
-		if (totalram_pages >= 128 * 1024)
+		if (atomic_long_read(&totalram_pages) >= 128 * 1024)
 			slots = 16 * 1024;
 		else
 			slots = 8 * 1024;

net/netfilter/nf_conntrack_core.c +3 -3

index ca1168d..3285df9 100644
@@ -2267,11 +2267,11 @@ int nf_conntrack_init_start(void)
 		 * >= 4GB machines have 65536 buckets.
 		 */
 		nf_conntrack_htable_size
-			= (((totalram_pages << PAGE_SHIFT) / 16384)
+			= (((atomic_long_read(&totalram_pages) << PAGE_SHIFT) / 16384)
 			   / sizeof(struct hlist_head));
-		if (totalram_pages > (4 * (1024 * 1024 * 1024 / PAGE_SIZE)))
+		if (atomic_long_read(&totalram_pages) > (4 * (1024 * 1024 * 1024 / PAGE_SIZE)))
 			nf_conntrack_htable_size = 65536;
-		else if (totalram_pages > (1024 * 1024 * 1024 / PAGE_SIZE))
+		else if (atomic_long_read(&totalram_pages) > (1024 * 1024 * 1024 / PAGE_SIZE))
 			nf_conntrack_htable_size = 16384;
 		if (nf_conntrack_htable_size < 32)
 			nf_conntrack_htable_size = 32;

net/netfilter/xt_hashlimit.c +2 -2

index 3e7d259..3c79a0f 100644
@@ -279,9 +279,9 @@ static int htable_create(struct net *net, struct hashlimit_cfg3 *cfg,
 	if (cfg->size) {
 		size = cfg->size;
 	} else {
-		size = (totalram_pages << PAGE_SHIFT) / 16384 /
+		size = (atomic_long_read(&totalram_pages) << PAGE_SHIFT) / 16384 /
 		       sizeof(struct hlist_head);
-		if (totalram_pages > 1024 * 1024 * 1024 / PAGE_SIZE)
+		if (atomic_long_read(&totalram_pages) > 1024 * 1024 * 1024 / PAGE_SIZE)
 			size = 8192;
 		if (size < 16)
 			size = 16;

net/sctp/protocol.c +3 -3

index 9b277bd..4ca4def 100644
@@ -1426,10 +1426,10 @@ static __init int sctp_init(void)
 	 * The methodology is similar to that of the tcp hash tables.
 	 * Though not identical.  Start by getting a goal size
 	 */
-	if (totalram_pages >= (128 * 1024))
-		goal = totalram_pages >> (22 - PAGE_SHIFT);
+	if (atomic_long_read(&totalram_pages) >= (128 * 1024))
+		goal = atomic_long_read(&totalram_pages) >> (22 - PAGE_SHIFT);
 	else
-		goal = totalram_pages >> (24 - PAGE_SHIFT);
+		goal = atomic_long_read(&totalram_pages) >> (24 - PAGE_SHIFT);
 
 	/* Then compute the page order for said goal */
 	order = get_order(goal);

security/integrity/ima/ima_kexec.c +0 -0

index 16bd187..8bb32ad 100644

Remove managed_page_count_lock spinlock and instead use atomic variables.

Suggested-by Michal Hocko <[email protected]>

Suggested-by Vlastimil Babka <[email protected]>

Signed-off-by Arun KS <[email protected]>

---
As discussed here,
https://patchwork.kernel.org/patch/10627521/#22261253
---

---
 arch/csky/mm/init.c                           |  4 +-
 arch/powerpc/platforms/pseries/cmm.c          | 11 ++--
 arch/s390/mm/init.c                           |  2 +-
 arch/um/kernel/mem.c                          |  4 +-
 arch/x86/kernel/cpu/microcode/core.c          |  5 +-
 drivers/char/agp/backend.c                    |  4 +-
 drivers/gpu/drm/amd/amdkfd/kfd_crat.c         |  2 +-
 drivers/gpu/drm/i915/i915_gem.c               |  2 +-
 drivers/gpu/drm/i915/selftests/i915_gem_gtt.c |  4 +-
 drivers/hv/hv_balloon.c                       | 19 +++----
 drivers/md/dm-bufio.c                         |  5 +-
 drivers/md/dm-crypt.c                         |  4 +-
 drivers/md/dm-integrity.c                     |  4 +-
 drivers/md/dm-stats.c                         |  3 +-
 drivers/media/platform/mtk-vpu/mtk_vpu.c      |  3 +-
 drivers/misc/vmw_balloon.c                    |  2 +-
 drivers/parisc/ccio-dma.c                     |  5 +-
 drivers/parisc/sba_iommu.c                    |  5 +-
 drivers/staging/android/ion/ion_system_heap.c |  2 +-
 drivers/xen/xen-selfballoon.c                 |  7 +--
 fs/ceph/super.h                               |  3 +-
 fs/file_table.c                               |  9 ++--
 fs/fuse/inode.c                               |  4 +-
 fs/nfs/write.c                                |  3 +-
 fs/nfsd/nfscache.c                            |  3 +-
 fs/ntfs/malloc.h                              |  2 +-
 fs/proc/base.c                                |  3 +-
 include/linux/highmem.h                       |  2 +-
 include/linux/mm.h                            |  2 +-
 include/linux/mmzone.h                        | 10 +---
 include/linux/swap.h                          |  2 +-
 kernel/fork.c                                 |  6 +--
 kernel/kexec_core.c                           |  5 +-
 kernel/power/snapshot.c                       |  2 +-
 lib/show_mem.c                                |  3 +-
 mm/highmem.c                                  |  2 +-
 mm/huge_memory.c                              |  2 +-
 mm/kasan/quarantine.c                         |  4 +-
 mm/memblock.c                                 |  6 +--
 mm/memory_hotplug.c                           |  4 +-
 mm/mm_init.c                                  |  3 +-
 mm/oom_kill.c                                 |  2 +-
 mm/page_alloc.c                               | 75 ++++++++++++++-------------
 mm/shmem.c                                    | 12 +++--
 mm/slab.c                                     |  3 +-
 mm/swap.c                                     |  3 +-
 mm/util.c                                     |  2 +-
 mm/vmalloc.c                                  |  4 +-
 mm/vmstat.c                                   |  4 +-
 mm/workingset.c                               |  2 +-
 mm/zswap.c                                    |  2 +-
 net/dccp/proto.c                              |  6 +--
 net/decnet/dn_route.c                         |  2 +-
 net/ipv4/tcp_metrics.c                        |  2 +-
 net/netfilter/nf_conntrack_core.c             |  6 +--
 net/netfilter/xt_hashlimit.c                  |  4 +-
 net/sctp/protocol.c                           |  6 +--
 security/integrity/ima/ima_kexec.c            |  2 +-
 58 files changed, 171 insertions(+), 143 deletions(-)

arch/csky/mm/init.c +2 -2

index dc07c07..3f4d35e 100644
@@ -71,7 +71,7 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 		ClearPageReserved(virt_to_page(start));
 		init_page_count(virt_to_page(start));
 		free_page(start);
-		totalram_pages++;
+		atomic_long_inc(&totalram_pages);
 	}
 }
 #endif

@@ -88,7 +88,7 @@ void free_initmem(void)
 		ClearPageReserved(virt_to_page(addr));
 		init_page_count(virt_to_page(addr));
 		free_page(addr);
-		totalram_pages++;
+		atomic_long_inc(&totalram_pages);
 		addr += PAGE_SIZE;
 	}
 

arch/powerpc/platforms/pseries/cmm.c +6 -5

index 25427a4..85fe503 100644
@@ -208,7 +208,7 @@ static long cmm_alloc_pages(long nr)
 
 		pa->page[pa->index++] = addr;
 		loaned_pages++;
-		totalram_pages--;
+		atomic_long_dec(&totalram_pages);
 		spin_unlock(&cmm_lock);
 		nr--;
 	}

@@ -247,7 +247,7 @@ static long cmm_free_pages(long nr)
 		free_page(addr);
 		loaned_pages--;
 		nr--;
-		totalram_pages++;
+		atomic_long_inc(&totalram_pages);
 	}
 	spin_unlock(&cmm_lock);
 	cmm_dbg("End request with %ld pages unfulfilled\n", nr);

@@ -291,7 +291,8 @@ static void cmm_get_mpp(void)
 	int rc;
 	struct hvcall_mpp_data mpp_data;
 	signed long active_pages_target, page_loan_request, target;
-	signed long total_pages = totalram_pages + loaned_pages;
+	signed long total_pages = atomic_long_read(&totalram_pages) +
+				  loaned_pages;
 	signed long min_mem_pages = (min_mem_mb * 1024 * 1024) / PAGE_SIZE;
 
 	rc = h_get_mpp(&mpp_data);

@@ -322,7 +323,7 @@ static void cmm_get_mpp(void)
 
 	cmm_dbg("delta = %ld, loaned = %lu, target = %lu, oom = %lu, totalram = %lu\n",
 		page_loan_request, loaned_pages, loaned_pages_target,
-		oom_freed_pages, totalram_pages);
+		oom_freed_pages, atomic_long_read(&totalram_pages));
 }
 
 static struct notifier_block cmm_oom_nb = {

@@ -581,7 +582,7 @@ static int cmm_mem_going_offline(void *arg)
 			free_page(pa_curr->page[idx]);
 			freed++;
 			loaned_pages--;
-			totalram_pages++;
+			atomic_long_inc(&totalram_pages);
 			pa_curr->page[idx] = pa_last->page[--pa_last->index];
 			if (pa_last->index == 0) {
 				if (pa_curr == pa_last)

arch/s390/mm/init.c +1 -1

index 76d0708..d6529e8 100644
@@ -59,7 +59,7 @@ static void __init setup_zero_pages(void)
 	order = 7;
 
 	/* Limit number of empty zero pages for small memory sizes */
-	while (order > 2 && (totalram_pages >> 10) < (1UL << order))
+	while (order > 2 && (atomic_long_read(&totalram_pages) >> 10) < (1UL << order))
 		order--;
 
 	empty_zero_page = __get_free_pages(GFP_KERNEL | __GFP_ZERO, order);

arch/um/kernel/mem.c +2 -2

index 1067469..da78a06 100644
@@ -51,8 +51,8 @@ void __init mem_init(void)
 
 	/* this will put all low memory onto the freelists */
 	memblock_free_all();
-	max_low_pfn = totalram_pages;
-	max_pfn = totalram_pages;
+	max_low_pfn = atomic_long_read(&totalram_pages);
+	max_pfn = atomic_long_read(&totalram_pages);
 	mem_init_print_info(NULL);
 	kmalloc_ok = 1;
 }

arch/x86/kernel/cpu/microcode/core.c +3 -2

index 2637ff0..4ccc8dd 100644
@@ -435,8 +435,9 @@ static ssize_t microcode_write(struct file *file, const char __user *buf,
 {
 	ssize_t ret = -EINVAL;
 
-	if ((len >> PAGE_SHIFT) > totalram_pages) {
-		pr_err("too much data (max %ld pages)\n", totalram_pages);
+	if ((len >> PAGE_SHIFT) > atomic_long_read(&totalram_pages)) {
+		pr_err("too much data (max %ld pages)\n",
+				atomic_long_read(&totalram_pages));
 		return ret;
 	}
 

drivers/char/agp/backend.c +2 -2

index 38ffb28..2753e1d 100644
@@ -115,9 +115,9 @@ static int agp_find_max(void)
 	long memory, index, result;
 
 #if PAGE_SHIFT < 20
-	memory = totalram_pages >> (20 - PAGE_SHIFT);
+	memory = atomic_long_read(&totalram_pages) >> (20 - PAGE_SHIFT);
 #else
-	memory = totalram_pages << (PAGE_SHIFT - 20);
+	memory = atomic_long_read(&totalram_pages) << (PAGE_SHIFT - 20);
 #endif
 	index = 1;
 

drivers/gpu/drm/amd/amdkfd/kfd_crat.c +1 -1

index 56412b0..ca18502 100644
@@ -848,7 +848,7 @@ static int kfd_fill_mem_info_for_cpu(int numa_node_id, int *avail_size,
 	 */
 	pgdat = NODE_DATA(numa_node_id);
 	for (zone_type = 0; zone_type < MAX_NR_ZONES; zone_type++)
-		mem_in_bytes += pgdat->node_zones[zone_type].managed_pages;
+		mem_in_bytes += atomic_long_read(&pgdat->node_zones[zone_type].managed_pages);
 	mem_in_bytes <<= PAGE_SHIFT;
 
 	sub_type_hdr->length_low = lower_32_bits(mem_in_bytes);

drivers/gpu/drm/i915/i915_gem.c +1 -1

index 0c8aa57..b4c245b 100644
@@ -2539,7 +2539,7 @@ static int i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 	 * If there's no chance of allocating enough pages for the whole
 	 * object, bail early.
 	 */
-	if (page_count > totalram_pages)
+	if (page_count > atomic_long_read(&totalram_pages))
 		return -ENOMEM;
 
 	st = kmalloc(sizeof(*st), GFP_KERNEL);

drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +2 -2

index 8e2e269..9ea10eb 100644
@@ -170,7 +170,7 @@ static int igt_ppgtt_alloc(void *arg)
 	 * This should ensure that we do not run into the oomkiller during
 	 * the test and take down the machine wilfully.
 	 */
-	limit = totalram_pages << PAGE_SHIFT;
+	limit = atomic_long_read(&totalram_pages) << PAGE_SHIFT;
 	limit = min(ppgtt->vm.total, limit);
 
 	/* Check we can allocate the entire range */

@@ -1244,7 +1244,7 @@ static int exercise_mock(struct drm_i915_private *i915,
 				     u64 hole_start, u64 hole_end,
 				     unsigned long end_time))
 {
-	const u64 limit = totalram_pages << PAGE_SHIFT;
+	const u64 limit = atomic_long_read(&totalram_pages) << PAGE_SHIFT;
 	struct i915_gem_context *ctx;
 	struct i915_hw_ppgtt *ppgtt;
 	IGT_TIMEOUT(end_time);

drivers/hv/hv_balloon.c +10 -9

index c5bc0b5..4498c94 100644
@@ -1092,6 +1092,7 @@ static void process_info(struct hv_dynmem_device *dm, struct dm_info_msg *msg)
 static unsigned long compute_balloon_floor(void)
 {
 	unsigned long min_pages;
+	unsigned long totalram = (unsigned long)atomic_long_read(&totalram_pages);
 #define MB2PAGES(mb) ((mb) << (20 - PAGE_SHIFT))
 	/* Simple continuous piecewiese linear function:
 	 *  max MiB -> min MiB  gradient

@@ -1104,16 +1105,16 @@ static unsigned long compute_balloon_floor(void)
 	 *    8192       744    (1/16)
 	 *   32768      1512	(1/32)
 	 */
-	if (totalram_pages < MB2PAGES(128))
-		min_pages = MB2PAGES(8) + (totalram_pages >> 1);
-	else if (totalram_pages < MB2PAGES(512))
-		min_pages = MB2PAGES(40) + (totalram_pages >> 2);
-	else if (totalram_pages < MB2PAGES(2048))
-		min_pages = MB2PAGES(104) + (totalram_pages >> 3);
-	else if (totalram_pages < MB2PAGES(8192))
-		min_pages = MB2PAGES(232) + (totalram_pages >> 4);
+	if (totalram < MB2PAGES(128))
+		min_pages = MB2PAGES(8) + (totalram >> 1);
+	else if (totalram < MB2PAGES(512))
+		min_pages = MB2PAGES(40) + (totalram >> 2);
+	else if (totalram < MB2PAGES(2048))
+		min_pages = MB2PAGES(104) + (totalram >> 3);
+	else if (totalram < MB2PAGES(8192))
+		min_pages = MB2PAGES(232) + (totalram >> 4);
 	else
-		min_pages = MB2PAGES(488) + (totalram_pages >> 5);
+		min_pages = MB2PAGES(488) + (totalram >> 5);
 #undef MB2PAGES
 	return min_pages;
 }

drivers/md/dm-bufio.c +3 -2

index dc385b7..6d61259 100644
@@ -1887,8 +1887,9 @@ static int __init dm_bufio_init(void)
 	dm_bufio_allocated_vmalloc = 0;
 	dm_bufio_current_allocated = 0;
 
-	mem = (__u64)mult_frac(totalram_pages - totalhigh_pages,
-			       DM_BUFIO_MEMORY_PERCENT, 100) << PAGE_SHIFT;
+	mem = (__u64)mult_frac(atomic_long_read(&totalram_pages) -
+				atomic_long_read(&totalhigh_pages),
+				DM_BUFIO_MEMORY_PERCENT, 100) << PAGE_SHIFT;
 
 	if (mem > ULONG_MAX)
 		mem = ULONG_MAX;

drivers/md/dm-crypt.c +3 -1

index 0481223..1c58f4c 100644
@@ -2158,7 +2158,9 @@ static int crypt_wipe_key(struct crypt_config *cc)
 
 static void crypt_calculate_pages_per_client(void)
 {
-	unsigned long pages = (totalram_pages - totalhigh_pages) * DM_CRYPT_MEMORY_PERCENT / 100;
+	unsigned long pages = (atomic_long_read(&totalram_pages) -
+				atomic_long_read(&totalhigh_pages)) *
+				DM_CRYPT_MEMORY_PERCENT / 100;
 
 	if (!dm_crypt_clients_n)
 		return;

drivers/md/dm-integrity.c +3 -1

index bb3096b..d91c931 100644
@@ -2843,7 +2843,9 @@ static int create_journal(struct dm_integrity_c *ic, char **error)
 	journal_pages = roundup((__u64)ic->journal_sections * ic->journal_section_sectors,
 				PAGE_SIZE >> SECTOR_SHIFT) >> (PAGE_SHIFT - SECTOR_SHIFT);
 	journal_desc_size = journal_pages * sizeof(struct page_list);
-	if (journal_pages >= totalram_pages - totalhigh_pages || journal_desc_size > ULONG_MAX) {
+	if (journal_pages >= atomic_long_read(&totalram_pages) -
+			atomic_long_read(&totalhigh_pages) ||
+			journal_desc_size > ULONG_MAX) {
 		*error = "Journal doesn't fit into memory";
 		r = -ENOMEM;
 		goto bad;

drivers/md/dm-stats.c +2 -1

index 21de30b..f154a07 100644
@@ -85,7 +85,8 @@ static bool __check_shared_memory(size_t alloc_size)
 	a = shared_memory_amount + alloc_size;
 	if (a < shared_memory_amount)
 		return false;
-	if (a >> PAGE_SHIFT > totalram_pages / DM_STATS_MEMORY_FACTOR)
+	if (a >> PAGE_SHIFT > atomic_long_read(&totalram_pages) /
+					DM_STATS_MEMORY_FACTOR)
 		return false;
 #ifdef CONFIG_MMU
 	if (a > (VMALLOC_END - VMALLOC_START) / DM_STATS_VMALLOC_FACTOR)

drivers/media/platform/mtk-vpu/mtk_vpu.c +2 -1

index 616f78b..ee3654a 100644
@@ -855,7 +855,8 @@ static int mtk_vpu_probe(struct platform_device *pdev)
 	/* Set PTCM to 96K and DTCM to 32K */
 	vpu_cfg_writel(vpu, 0x2, VPU_TCM_CFG);
 
-	vpu->enable_4GB = !!(totalram_pages > (SZ_2G >> PAGE_SHIFT));
+	vpu->enable_4GB = !!(atomic_long_read(&totalram_pages) >
+					(SZ_2G >> PAGE_SHIFT));
 	dev_info(dev, "4GB mode %u\n", vpu->enable_4GB);
 
 	if (vpu->enable_4GB) {

drivers/misc/vmw_balloon.c +1 -1

index 9b0b3fa..0ac0fee 100644
@@ -570,7 +570,7 @@ static int vmballoon_send_get_target(struct vmballoon *b)
 	unsigned long status;
 	unsigned long limit;
 
-	limit = totalram_pages;
+	limit = atomic_long_read(&totalram_pages);
 
 	/* Ensure limit fits in 32-bits */
 	if (limit != (u32)limit)

drivers/parisc/ccio-dma.c +3 -2

index 6148236..705df1a 100644
@@ -1255,7 +1255,8 @@ void __init ccio_cujo20_fixup(struct parisc_device *cujo, u32 iovp)
 	** Hot-Plug/Removal of PCI cards. (aka PCI OLARD).
 	*/
 
-	iova_space_size = (u32) (totalram_pages / count_parisc_driver(&ccio_driver));
+	iova_space_size = (u32) (atomic_long_read(&totalram_pages) /
+				count_parisc_driver(&ccio_driver));
 
 	/* limit IOVA space size to 1MB-1GB */
 

@@ -1294,7 +1295,7 @@ void __init ccio_cujo20_fixup(struct parisc_device *cujo, u32 iovp)
 
 	DBG_INIT("%s() hpa 0x%p mem %luMB IOV %dMB (%d bits)\n",
 			__func__, ioc->ioc_regs,
-			(unsigned long) totalram_pages >> (20 - PAGE_SHIFT),
+			(unsigned long) atomic_long_read(&totalram_pages) >> (20 - PAGE_SHIFT),
 			iova_space_size>>20,
 			iov_order + PAGE_SHIFT);
 

drivers/parisc/sba_iommu.c +3 -2

index 11de0ec..02f4ce9 100644
@@ -1419,7 +1419,8 @@ static int setup_ibase_imask_callback(struct device *dev, void *data)
 	** for DMA hints - ergo only 30 bits max.
 	*/
 
-	iova_space_size = (u32) (totalram_pages/global_ioc_cnt);
+	iova_space_size = (u32) (atomic_long_read(&totalram_pages)/
+						global_ioc_cnt);
 
 	/* limit IOVA space size to 1MB-1GB */
 	if (iova_space_size < (1 << (20 - PAGE_SHIFT))) {

@@ -1444,7 +1445,7 @@ static int setup_ibase_imask_callback(struct device *dev, void *data)
 	DBG_INIT("%s() hpa 0x%lx mem %ldMB IOV %dMB (%d bits)\n",
 			__func__,
 			ioc->ioc_hpa,
-			(unsigned long) totalram_pages >> (20 - PAGE_SHIFT),
+			(unsigned long) atomic_long_read(&totalram_pages) >> (20 - PAGE_SHIFT),
 			iova_space_size>>20,
 			iov_order + PAGE_SHIFT);
 

drivers/staging/android/ion/ion_system_heap.c +1 -1

index 548bb02..64bd925 100644
@@ -110,7 +110,7 @@ static int ion_system_heap_allocate(struct ion_heap *heap,
 	unsigned long size_remaining = PAGE_ALIGN(size);
 	unsigned int max_order = orders[0];
 
-	if (size / PAGE_SIZE > totalram_pages / 2)
+	if (size / PAGE_SIZE > atomic_long_read(&totalram_pages) / 2)
 		return -ENOMEM;
 
 	INIT_LIST_HEAD(&pages);

drivers/xen/xen-selfballoon.c +4 -3

index 5165aa8..0b925fd 100644
@@ -189,7 +189,7 @@ static void selfballoon_process(struct work_struct *work)
 	bool reset_timer = false;
 
 	if (xen_selfballooning_enabled) {
-		cur_pages = totalram_pages;
+		cur_pages = atomic_long_read(&totalram_pages);
 		tgt_pages = cur_pages; /* default is no change */
 		goal_pages = vm_memory_committed() +
 				totalreserve_pages +

@@ -227,7 +227,8 @@ static void selfballoon_process(struct work_struct *work)
 		if (tgt_pages < floor_pages)
 			tgt_pages = floor_pages;
 		balloon_set_new_target(tgt_pages +
-			balloon_stats.current_pages - totalram_pages);
+			balloon_stats.current_pages -
+			atomic_long_read(&totalram_pages));
 		reset_timer = true;
 	}
 #ifdef CONFIG_FRONTSWAP

@@ -569,7 +570,7 @@ int xen_selfballoon_init(bool use_selfballooning, bool use_frontswap_selfshrink)
 	 * much more reliably and response faster in some cases.
 	 */
 	if (!selfballoon_reserved_mb) {
-		reserve_pages = totalram_pages / 10;
+		reserve_pages = atomic_long_read(&totalram_pages) / 10;
 		selfballoon_reserved_mb = PAGES2MB(reserve_pages);
 	}
 	schedule_delayed_work(&selfballoon_worker, selfballoon_interval * HZ);

fs/ceph/super.h +2 -1

index 582e28f..92f56d3 100644
@@ -807,7 +807,8 @@ static inline int default_congestion_kb(void)
 	 * This allows larger machines to have larger/more transfers.
 	 * Limit the default to 256M
 	 */
-	congestion_kb = (16*int_sqrt(totalram_pages)) << (PAGE_SHIFT-10);
+	congestion_kb = (16*int_sqrt(atomic_long_read(&totalram_pages))) <<
+								(PAGE_SHIFT-10);
 	if (congestion_kb > 256*1024)
 		congestion_kb = 256*1024;
 

fs/file_table.c +6 -3

index e03c8d1..5dde5c3 100644
@@ -383,10 +383,13 @@ void __init files_init(void)
 void __init files_maxfiles_init(void)
 {
 	unsigned long n;
-	unsigned long memreserve = (totalram_pages - nr_free_pages()) * 3/2;
+	unsigned long memreserve = (atomic_long_read(&totalram_pages) -
+						nr_free_pages()) * 3/2;
 
-	memreserve = min(memreserve, totalram_pages - 1);
-	n = ((totalram_pages - memreserve) * (PAGE_SIZE / 1024)) / 10;
+	memreserve = min(memreserve,
+			(unsigned long)atomic_long_read(&totalram_pages) - 1);
+	n = ((atomic_long_read(&totalram_pages) - memreserve) *
+					(PAGE_SIZE / 1024)) / 10;
 
 	files_stat.max_files = max_t(unsigned long, n, NR_FILE);
 }

fs/fuse/inode.c +2 -2

index 4727ef6..acdbaf7 100644
@@ -825,8 +825,8 @@ static struct dentry *fuse_get_parent(struct dentry *child)
 static void sanitize_global_limit(unsigned *limit)
 {
 	if (*limit == 0)
-		*limit = ((totalram_pages << PAGE_SHIFT) >> 13) /
-			 sizeof(struct fuse_req);
+		*limit = ((atomic_long_read(&totalram_pages)
+			 << PAGE_SHIFT) >> 13) / sizeof(struct fuse_req);
 
 	if (*limit >= 1 << 16)
 		*limit = (1 << 16) - 1;

fs/nfs/write.c +2 -1

index 586726a..e3663b7 100644
@@ -2121,7 +2121,8 @@ int __init nfs_init_writepagecache(void)
 	 * This allows larger machines to have larger/more transfers.
 	 * Limit the default to 256M
 	 */
-	nfs_congestion_kb = (16*int_sqrt(totalram_pages)) << (PAGE_SHIFT-10);
+	nfs_congestion_kb = (16*int_sqrt(atomic_long_read(&totalram_pages))) <<
+								(PAGE_SHIFT-10);
 	if (nfs_congestion_kb > 256*1024)
 		nfs_congestion_kb = 256*1024;
 

fs/nfsd/nfscache.c +2 -1

index e2fe0e9..e877558 100644
@@ -99,7 +99,8 @@ static unsigned long nfsd_reply_cache_scan(struct shrinker *shrink,
 nfsd_cache_size_limit(void)
 {
 	unsigned int limit;
-	unsigned long low_pages = totalram_pages - totalhigh_pages;
+	unsigned long low_pages = atomic_long_read(&totalram_pages) -
+					atomic_long_read(&totalhigh_pages);
 
 	limit = (16 * int_sqrt(low_pages)) << (PAGE_SHIFT-10);
 	return min_t(unsigned int, limit, 256*1024);

fs/ntfs/malloc.h +1 -1

index ab172e5..4ae6bbe 100644
@@ -47,7 +47,7 @@ static inline void *__ntfs_malloc(unsigned long size, gfp_t gfp_mask)
 		return kmalloc(PAGE_SIZE, gfp_mask & ~__GFP_HIGHMEM);
 		/* return (void *)__get_free_page(gfp_mask); */
 	}
-	if (likely((size >> PAGE_SHIFT) < totalram_pages))
+	if (likely((size >> PAGE_SHIFT) < atomic_long_read(&totalram_pages)))
 		return __vmalloc(size, gfp_mask, PAGE_KERNEL);
 	return NULL;
 }

fs/proc/base.c +2 -1

index ce34654..9ef26dc 100644
@@ -530,7 +530,8 @@ static ssize_t lstats_write(struct file *file, const char __user *buf,
 static int proc_oom_score(struct seq_file *m, struct pid_namespace *ns,
 			  struct pid *pid, struct task_struct *task)
 {
-	unsigned long totalpages = totalram_pages + total_swap_pages;
+	unsigned long totalpages = atomic_long_read(&totalram_pages) +
+							total_swap_pages;
 	unsigned long points = 0;
 
 	points = oom_badness(task, NULL, NULL, totalpages) *

include/linux/highmem.h +1 -1

index 0690679..84edaa2 100644
@@ -36,7 +36,7 @@ static inline void invalidate_kernel_vmap_range(void *vaddr, int size)
 
 /* declarations for linux/mm/highmem.c */
 unsigned int nr_free_highpages(void);
-extern unsigned long totalhigh_pages;
+extern atomic_long_t totalhigh_pages;
 
 void kmap_flush_unused(void);
 

include/linux/mm.h +1 -1

index fcf9cc9..af952fc 100644
@@ -48,7 +48,7 @@ static inline void set_max_mapnr(unsigned long limit)
 static inline void set_max_mapnr(unsigned long limit) { }
 #endif
 
-extern unsigned long totalram_pages;
+extern atomic_long_t totalram_pages;
 extern void * high_memory;
 extern int page_cluster;
 

include/linux/mmzone.h +2 -8

index 8555509..2639b05 100644
@@ -428,14 +428,8 @@ struct zone {
 	 * Write access to present_pages at runtime should be protected by
 	 * mem_hotplug_begin/end(). Any reader who can't tolerant drift of
 	 * present_pages should get_online_mems() to get a stable value.
-	 *
-	 * Read access to managed_pages should be safe because it's unsigned
-	 * long. Write access to zone->managed_pages and totalram_pages are
-	 * protected by managed_page_count_lock at runtime. Idealy only
-	 * adjust_managed_page_count() should be used instead of directly
-	 * touching zone->managed_pages and totalram_pages.
 	 */
-	unsigned long		managed_pages;
+	atomic_long_t		managed_pages;
 	unsigned long		spanned_pages;
 	unsigned long		present_pages;
 

@@ -814,7 +808,7 @@ static inline bool is_dev_zone(const struct zone *zone)
  */
 static inline bool managed_zone(struct zone *zone)
 {
-	return zone->managed_pages;
+	return atomic_long_read(&zone->managed_pages);
 }
 
 /* Returns true if a zone has memory */

include/linux/swap.h +1 -1

index d098743..b34c6e7 100644
@@ -309,7 +309,7 @@ struct vma_swap_readahead {
 } while (0)
 
 /* linux/mm/page_alloc.c */
-extern unsigned long totalram_pages;
+extern atomic_long_t totalram_pages;
 extern unsigned long totalreserve_pages;
 extern unsigned long nr_free_buffer_pages(void);
 extern unsigned long nr_free_pagecache_pages(void);

kernel/fork.c +3 -3

index 2f78d32..b6068c5 100644
@@ -744,11 +744,11 @@ static void set_max_threads(unsigned int max_threads_suggested)
 	 * The number of threads shall be limited such that the thread
 	 * structures may only consume a small part of the available memory.
 	 */
-	if (fls64(totalram_pages) + fls64(PAGE_SIZE) > 64)
+	if (fls64(atomic_long_read(&totalram_pages)) + fls64(PAGE_SIZE) > 64)
 		threads = MAX_THREADS;
 	else
-		threads = div64_u64((u64) totalram_pages * (u64) PAGE_SIZE,
-				    (u64) THREAD_SIZE * 8UL);
+		threads = div64_u64((u64) atomic_long_read(&totalram_pages) *
+				(u64) PAGE_SIZE, (u64) THREAD_SIZE * 8UL);
 
 	if (threads > max_threads_suggested)
 		threads = max_threads_suggested;

kernel/kexec_core.c +3 -2

index 86ef06d..ed85ddd 100644
@@ -217,13 +217,14 @@ int sanity_check_segment_list(struct kimage *image)
 	 * wasted allocating pages, which can cause a soft lockup.
 	 */
 	for (i = 0; i < nr_segments; i++) {
-		if (PAGE_COUNT(image->segment[i].memsz) > totalram_pages / 2)
+		if (PAGE_COUNT(image->segment[i].memsz) >
+				atomic_long_read(&totalram_pages) / 2)
 			return -EINVAL;
 
 		total_pages += PAGE_COUNT(image->segment[i].memsz);
 	}
 
-	if (total_pages > totalram_pages / 2)
+	if (total_pages > atomic_long_read(&totalram_pages) / 2)
 		return -EINVAL;
 
 	/*

kernel/power/snapshot.c +1 -1

index b0308a2..142a3c76 100644
@@ -105,7 +105,7 @@ void __init hibernate_reserved_size_init(void)
 
 void __init hibernate_image_size_init(void)
 {
-	image_size = ((totalram_pages * 2) / 5) * PAGE_SIZE;
+	image_size = ((atomic_long_read(&totalram_pages) * 2) / 5) * PAGE_SIZE;
 }
 
 /*

lib/show_mem.c +2 -1

index 0beaa1d..0701f63 100644
@@ -28,7 +28,8 @@ void show_mem(unsigned int filter, nodemask_t *nodemask)
 				continue;
 
 			total += zone->present_pages;
-			reserved += zone->present_pages - zone->managed_pages;
+			reserved += zone->present_pages -
+				atomic_long_read(&zone->managed_pages);
 
 			if (is_highmem_idx(zoneid))
 				highmem += zone->present_pages;

mm/highmem.c +1 -1

index 59db322..93a45c0 100644
@@ -105,7 +105,7 @@ static inline wait_queue_head_t *get_pkmap_wait_queue_head(unsigned int color)
 }
 #endif
 
-unsigned long totalhigh_pages __read_mostly;
+atomic_long_t totalhigh_pages __read_mostly;
 EXPORT_SYMBOL(totalhigh_pages);
 
 

mm/huge_memory.c +1 -1

index d394d18..f2f18b5 100644
@@ -420,7 +420,7 @@ static int __init hugepage_init(void)
 	 * where the extra memory used could hurt more than TLB overhead
 	 * is likely to save.  The admin can still enable it through /sys.
 	 */
-	if (totalram_pages < (512 << (20 - PAGE_SHIFT))) {
+	if (atomic_long_read(&totalram_pages) < (512 << (20 - PAGE_SHIFT))) {
 		transparent_hugepage_flags = 0;
 		return 0;
 	}

mm/kasan/quarantine.c +2 -2

index b209dba..4d36aed 100644
@@ -236,8 +236,8 @@ void quarantine_reduce(void)
 	 * Update quarantine size in case of hotplug. Allocate a fraction of
 	 * the installed memory to quarantine minus per-cpu queue limits.
 	 */
-	total_size = (READ_ONCE(totalram_pages) << PAGE_SHIFT) /
-		QUARANTINE_FRACTION;
+	total_size = (READ_ONCE(atomic_long_read(&totalram_pages)) <<
+			PAGE_SHIFT) / QUARANTINE_FRACTION;
 	percpu_quarantines = QUARANTINE_PERCPU_SIZE * num_online_cpus();
 	new_quarantine_size = (total_size < percpu_quarantines) ?
 		0 : total_size - percpu_quarantines;

mm/memblock.c +3 -3

index eddcac2..43f53e9 100644
@@ -1627,7 +1627,7 @@ void __init __memblock_free_late(phys_addr_t base, phys_addr_t size)
 
 	for (; cursor < end; cursor++) {
 		memblock_free_pages(pfn_to_page(cursor), cursor, 0);
-		totalram_pages++;
+		atomic_long_inc(&totalram_pages);
 	}
 }
 

@@ -2001,7 +2001,7 @@ void reset_node_managed_pages(pg_data_t *pgdat)
 	struct zone *z;
 
 	for (z = pgdat->node_zones; z < pgdat->node_zones + MAX_NR_ZONES; z++)
-		z->managed_pages = 0;
+		atomic_long_set(&z->managed_pages, 0);
 }
 
 void __init reset_all_zones_managed_pages(void)

@@ -2029,7 +2029,7 @@ unsigned long __init memblock_free_all(void)
 	reset_all_zones_managed_pages();
 
 	pages = free_low_memory_core_early();
-	totalram_pages += pages;
+	atomic_long_add(pages, &totalram_pages);
 
 	return pages;
 }

mm/memory_hotplug.c +2 -2

index dbbb945..0725984 100644
@@ -657,10 +657,10 @@ void __online_page_free(struct page *page)
 static int generic_online_page(struct page *page, unsigned int order)
 {
 	__free_pages_core(page, order);
-	totalram_pages += (1UL << order);
+	atomic_long_add((1UL << order), &totalram_pages);
 #ifdef CONFIG_HIGHMEM
 	if (PageHighMem(page))
-		totalhigh_pages += (1UL << order);
+		atomic_long_add((1UL << order), &totalhigh_pages);
 #endif
 	return 0;
 }

mm/mm_init.c +2 -1

index 6838a53..93a6611 100644
@@ -146,7 +146,8 @@ static void __meminit mm_compute_batch(void)
 	s32 batch = max_t(s32, nr*2, 32);
 
 	/* batch size set to 0.4% of (total memory/#cpus), or max int32 */
-	memsized_batch = min_t(u64, (totalram_pages/nr)/256, 0x7fffffff);
+	memsized_batch = min_t(u64, (atomic_long_read(&totalram_pages)/nr)/256,
+								0x7fffffff);
 
 	vm_committed_as_batch = max_t(s32, memsized_batch, batch);
 }

mm/oom_kill.c +1 -1

index 6589f60..1a37d68 100644
@@ -269,7 +269,7 @@ static enum oom_constraint constrained_alloc(struct oom_control *oc)
 	}
 
 	/* Default to all available memory */
-	oc->totalpages = totalram_pages + total_swap_pages;
+	oc->totalpages = atomic_long_read(&totalram_pages) + total_swap_pages;
 
 	if (!IS_ENABLED(CONFIG_NUMA))
 		return CONSTRAINT_NONE;

mm/page_alloc.c +39 -36

index 4bd858d..c7b26e3 100644
@@ -121,10 +121,7 @@
 };
 EXPORT_SYMBOL(node_states);
 
-/* Protect totalram_pages and zone->managed_pages */
-static DEFINE_SPINLOCK(managed_page_count_lock);
-
-unsigned long totalram_pages __read_mostly;
+atomic_long_t totalram_pages __read_mostly;
 unsigned long totalreserve_pages __read_mostly;
 unsigned long totalcma_pages __read_mostly;
 

@@ -1275,7 +1272,7 @@ void __free_pages_core(struct page *page, unsigned int order)
 		set_page_count(p, 0);
 	}
 
-	page_zone(page)->managed_pages += nr_pages;
+	atomic_long_add(nr_pages, &page_zone(page)->managed_pages);
 	set_page_refcounted(page);
 	__free_pages(page, order);
 }

@@ -2254,7 +2251,8 @@ static void reserve_highatomic_pageblock(struct page *page, struct zone *zone,
 	 * Limit the number reserved to 1 pageblock or roughly 1% of a zone.
 	 * Check is race-prone but harmless.
 	 */
-	max_managed = (zone->managed_pages / 100) + pageblock_nr_pages;
+	max_managed = (atomic_long_read(&zone->managed_pages) / 100) +
+						pageblock_nr_pages;
 	if (zone->nr_reserved_highatomic >= max_managed)
 		return;
 

@@ -4658,7 +4656,7 @@ static unsigned long nr_free_zone_pages(int offset)
 	struct zonelist *zonelist = node_zonelist(numa_node_id(), GFP_KERNEL);
 
 	for_each_zone_zonelist(zone, z, zonelist, offset) {
-		unsigned long size = zone->managed_pages;
+		unsigned long size = atomic_long_read(&zone->managed_pages);
 		unsigned long high = high_wmark_pages(zone);
 		if (size > high)
 			sum += size - high;

@@ -4744,11 +4742,15 @@ long si_mem_available(void)
 
 void si_meminfo(struct sysinfo *val)
 {
-	val->totalram = totalram_pages;
+	val->totalram = atomic_long_read(&totalram_pages);
 	val->sharedram = global_node_page_state(NR_SHMEM);
 	val->freeram = global_zone_page_state(NR_FREE_PAGES);
 	val->bufferram = nr_blockdev_pages();
-	val->totalhigh = totalhigh_pages;
+#ifdef CONFIG_HIGHMEM
+	val->totalhigh = atomic_long_read(&totalhigh_pages);
+#else
+	val->totalhigh = 0;
+#endif
 	val->freehigh = nr_free_highpages();
 	val->mem_unit = PAGE_SIZE;
 }

@@ -4765,7 +4767,7 @@ void si_meminfo_node(struct sysinfo *val, int nid)
 	pg_data_t *pgdat = NODE_DATA(nid);
 
 	for (zone_type = 0; zone_type < MAX_NR_ZONES; zone_type++)
-		managed_pages += pgdat->node_zones[zone_type].managed_pages;
+		managed_pages += atomic_long_read(&pgdat->node_zones[zone_type].managed_pages);
 	val->totalram = managed_pages;
 	val->sharedram = node_page_state(pgdat, NR_SHMEM);
 	val->freeram = sum_zone_node_page_state(nid, NR_FREE_PAGES);

@@ -4774,7 +4776,7 @@ void si_meminfo_node(struct sysinfo *val, int nid)
 		struct zone *zone = &pgdat->node_zones[zone_type];
 
 		if (is_highmem(zone)) {
-			managed_highpages += zone->managed_pages;
+			managed_highpages += atomic_long_read(&zone->managed_pages);
 			free_highpages += zone_page_state(zone, NR_FREE_PAGES);
 		}
 	}

@@ -4981,7 +4983,7 @@ void show_free_areas(unsigned int filter, nodemask_t *nodemask)
 			K(zone_page_state(zone, NR_ZONE_UNEVICTABLE)),
 			K(zone_page_state(zone, NR_ZONE_WRITE_PENDING)),
 			K(zone->present_pages),
-			K(zone->managed_pages),
+			K(atomic_long_read(&zone->managed_pages)),
 			K(zone_page_state(zone, NR_MLOCK)),
 			zone_page_state(zone, NR_KERNEL_STACK_KB),
 			K(zone_page_state(zone, NR_PAGETABLE)),

@@ -5643,7 +5645,7 @@ static int zone_batchsize(struct zone *zone)
 	 * The per-cpu-pages pools are set to around 1000th of the
 	 * size of the zone.
 	 */
-	batch = zone->managed_pages / 1024;
+	batch = atomic_long_read(&zone->managed_pages) / 1024;
 	/* But no more than a meg. */
 	if (batch * PAGE_SIZE > 1024 * 1024)
 		batch = (1024 * 1024) / PAGE_SIZE;

@@ -5754,7 +5756,7 @@ static void pageset_set_high_and_batch(struct zone *zone,
 {
 	if (percpu_pagelist_fraction)
 		pageset_set_high(pcp,
-			(zone->managed_pages /
+			(atomic_long_read(&zone->managed_pages) /
 				percpu_pagelist_fraction));
 	else
 		pageset_set_batch(pcp, zone_batchsize(zone));

@@ -6309,7 +6311,7 @@ static void __meminit pgdat_init_internals(struct pglist_data *pgdat)
 static void __meminit zone_init_internals(struct zone *zone, enum zone_type idx, int nid,
 							unsigned long remaining_pages)
 {
-	zone->managed_pages = remaining_pages;
+	atomic_long_set(&zone->managed_pages, remaining_pages);
 	zone_set_nid(zone, nid);
 	zone->name = zone_names[idx];
 	zone->zone_pgdat = NODE_DATA(nid);

@@ -7061,14 +7063,12 @@ static int __init cmdline_parse_movablecore(char *p)
 
 void adjust_managed_page_count(struct page *page, long count)
 {
-	spin_lock(&managed_page_count_lock);
-	page_zone(page)->managed_pages += count;
-	totalram_pages += count;
+	atomic_long_add(count, &page_zone(page)->managed_pages);
+	atomic_long_add(count, &totalram_pages);
 #ifdef CONFIG_HIGHMEM
 	if (PageHighMem(page))
-		totalhigh_pages += count;
+		atomic_long_add(count, &totalhigh_pages);
 #endif
-	spin_unlock(&managed_page_count_lock);
 }
 EXPORT_SYMBOL(adjust_managed_page_count);
 

@@ -7109,9 +7109,9 @@ unsigned long free_reserved_area(void *start, void *end, int poison, char *s)
 void free_highmem_page(struct page *page)
 {
 	__free_reserved_page(page);
-	totalram_pages++;
-	page_zone(page)->managed_pages++;
-	totalhigh_pages++;
+	atomic_long_inc(&totalram_pages);
+	atomic_long_inc(&page_zone(page)->managed_pages);
+	atomic_long_inc(&totalhigh_pages);
 }
 #endif
 

@@ -7160,10 +7160,10 @@ void __init mem_init_print_info(const char *str)
 		physpages << (PAGE_SHIFT - 10),
 		codesize >> 10, datasize >> 10, rosize >> 10,
 		(init_data_size + init_code_size) >> 10, bss_size >> 10,
-		(physpages - totalram_pages - totalcma_pages) << (PAGE_SHIFT - 10),
+		(physpages - atomic_long_read(&totalram_pages) - totalcma_pages) << (PAGE_SHIFT - 10),
 		totalcma_pages << (PAGE_SHIFT - 10),
 #ifdef	CONFIG_HIGHMEM
-		totalhigh_pages << (PAGE_SHIFT - 10),
+		atomic_long_read(&totalhigh_pages) << (PAGE_SHIFT - 10),
 #endif
 		str ? ", " : "", str ? str : "");
 }

@@ -7253,8 +7253,8 @@ static void calculate_totalreserve_pages(void)
 			/* we treat the high watermark as reserved pages. */
 			max += high_wmark_pages(zone);
 
-			if (max > zone->managed_pages)
-				max = zone->managed_pages;
+			if (max > atomic_long_read(&zone->managed_pages))
+				max = atomic_long_read(&zone->managed_pages);
 
 			pgdat->totalreserve_pages += max;
 

@@ -7278,7 +7278,7 @@ static void setup_per_zone_lowmem_reserve(void)
 	for_each_online_pgdat(pgdat) {
 		for (j = 0; j < MAX_NR_ZONES; j++) {
 			struct zone *zone = pgdat->node_zones + j;
-			unsigned long managed_pages = zone->managed_pages;
+			unsigned long managed_pages = atomic_long_read(&zone->managed_pages);
 
 			zone->lowmem_reserve[j] = 0;
 

@@ -7296,7 +7296,7 @@ static void setup_per_zone_lowmem_reserve(void)
 					lower_zone->lowmem_reserve[j] =
 						managed_pages / sysctl_lowmem_reserve_ratio[idx];
 				}
-				managed_pages += lower_zone->managed_pages;
+				managed_pages += atomic_long_read(&lower_zone->managed_pages);
 			}
 		}
 	}

@@ -7315,14 +7315,14 @@ static void __setup_per_zone_wmarks(void)
 	/* Calculate total number of !ZONE_HIGHMEM pages */
 	for_each_zone(zone) {
 		if (!is_highmem(zone))
-			lowmem_pages += zone->managed_pages;
+			lowmem_pages += atomic_long_read(&zone->managed_pages);
 	}
 
 	for_each_zone(zone) {
 		u64 tmp;
 
 		spin_lock_irqsave(&zone->lock, flags);
-		tmp = (u64)pages_min * zone->managed_pages;
+		tmp = (u64)pages_min * atomic_long_read(&zone->managed_pages);
 		do_div(tmp, lowmem_pages);
 		if (is_highmem(zone)) {
 			/*

@@ -7336,7 +7336,8 @@ static void __setup_per_zone_wmarks(void)
 			 */
 			unsigned long min_pages;
 
-			min_pages = zone->managed_pages / 1024;
+			min_pages = atomic_long_read(&zone->managed_pages) /
+									1024;
 			min_pages = clamp(min_pages, SWAP_CLUSTER_MAX, 128UL);
 			zone->watermark[WMARK_MIN] = min_pages;
 		} else {

@@ -7353,7 +7354,7 @@ static void __setup_per_zone_wmarks(void)
 		 * ensure a minimum size on small systems.
 		 */
 		tmp = max_t(u64, tmp >> 2,
-			    mult_frac(zone->managed_pages,
+			    mult_frac(atomic_long_read(&zone->managed_pages),
 				      watermark_scale_factor, 10000));
 
 		zone->watermark[WMARK_LOW]  = min_wmark_pages(zone) + tmp;

@@ -7483,7 +7484,8 @@ static void setup_min_unmapped_ratio(void)
 		pgdat->min_unmapped_pages = 0;
 
 	for_each_zone(zone)
-		zone->zone_pgdat->min_unmapped_pages += (zone->managed_pages *
+		zone->zone_pgdat->min_unmapped_pages +=
+				(atomic_long_read(&zone->managed_pages) *
 				sysctl_min_unmapped_ratio) / 100;
 }
 

@@ -7511,8 +7513,9 @@ static void setup_min_slab_ratio(void)
 		pgdat->min_slab_pages = 0;
 
 	for_each_zone(zone)
-		zone->zone_pgdat->min_slab_pages += (zone->managed_pages *
-				sysctl_min_slab_ratio) / 100;
+		zone->zone_pgdat->min_slab_pages +=
+			(atomic_long_read(&zone->managed_pages) *
+			sysctl_min_slab_ratio) / 100;
 }
 
 int sysctl_min_slab_ratio_sysctl_handler(struct ctl_table *table, int write,

mm/shmem.c +9 -3

index a6964ba..edd55db 100644
@@ -109,12 +109,18 @@ struct shmem_falloc {
 #ifdef CONFIG_TMPFS
 static unsigned long shmem_default_max_blocks(void)
 {
-	return totalram_pages / 2;
+	return atomic_long_read(&totalram_pages) / 2;
 }
 
 static unsigned long shmem_default_max_inodes(void)
 {
-	return min(totalram_pages - totalhigh_pages, totalram_pages / 2);
+	return min((unsigned long)atomic_long_read(&totalram_pages) -
+#ifdef CONFIG_HIGHMEM
+		(unsigned long) atomic_long_read(&totalhigh_pages),
+#else
+		0,
+#endif
+		(unsigned long)atomic_long_read(&totalram_pages) / 2);
 }
 #endif
 

@@ -3274,7 +3280,7 @@ static int shmem_parse_options(char *options, struct shmem_sb_info *sbinfo,
 			size = memparse(value,&rest);
 			if (*rest == '%') {
 				size <<= PAGE_SHIFT;
-				size *= totalram_pages;
+				size *= atomic_long_read(&totalram_pages);
 				do_div(size, 100);
 				rest++;
 			}

mm/slab.c +2 -1

index 2a5654b..70252b0 100644
@@ -1248,7 +1248,8 @@ void __init kmem_cache_init(void)
 	 * page orders on machines with more than 32MB of memory if
 	 * not overridden on the command line.
 	 */
-	if (!slab_max_order_set && totalram_pages > (32 << 20) >> PAGE_SHIFT)
+	if (!slab_max_order_set && atomic_long_read(&totalram_pages) >
+						(32 << 20) >> PAGE_SHIFT)
 		slab_max_order = SLAB_MAX_ORDER_HI;
 
 	/* Bootstrap is tricky, because several objects are allocated

mm/swap.c +2 -1

index aa48371..e85bc4a 100644
@@ -1023,7 +1023,8 @@ unsigned pagevec_lookup_range_nr_tag(struct pagevec *pvec,
  */
 void __init swap_setup(void)
 {
-	unsigned long megs = totalram_pages >> (20 - PAGE_SHIFT);
+	unsigned long megs = atomic_long_read(&totalram_pages) >>
+						(20 - PAGE_SHIFT);
 
 	/* Use a smaller cluster for small-memory machines */
 	if (megs < 16)

mm/util.c +1 -1

index 7f1f165..a3ae8ee 100644
@@ -600,7 +600,7 @@ unsigned long vm_commit_limit(void)
 	if (sysctl_overcommit_kbytes)
 		allowed = sysctl_overcommit_kbytes >> (PAGE_SHIFT - 10);
 	else
-		allowed = ((totalram_pages - hugetlb_total_pages())
+		allowed = ((atomic_long_read(&totalram_pages) - hugetlb_total_pages())
 			   * sysctl_overcommit_ratio / 100);
 	allowed += total_swap_pages;
 

mm/vmalloc.c +2 -2

index 97d4b25..f177af8 100644
@@ -1634,7 +1634,7 @@ void *vmap(struct page **pages, unsigned int count,
 
 	might_sleep();
 
-	if (count > totalram_pages)
+	if (count > atomic_long_read(&totalram_pages))
 		return NULL;
 
 	size = (unsigned long)count << PAGE_SHIFT;

@@ -1739,7 +1739,7 @@ void *__vmalloc_node_range(unsigned long size, unsigned long align,
 	unsigned long real_size = size;
 
 	size = PAGE_ALIGN(size);
-	if (!size || (size >> PAGE_SHIFT) > totalram_pages)
+	if (!size || (size >> PAGE_SHIFT) > atomic_long_read(&totalram_pages))
 		goto fail;
 
 	area = __get_vm_area_node(size, align, VM_ALLOC | VM_UNINITIALIZED |

mm/vmstat.c +2 -2

index 6038ce5..20551e8 100644
@@ -227,7 +227,7 @@ int calculate_normal_threshold(struct zone *zone)
 	 * 125		1024		10	16-32 GB	9
 	 */
 
-	mem = zone->managed_pages >> (27 - PAGE_SHIFT);
+	mem = atomic_long_read(&zone->managed_pages) >> (27 - PAGE_SHIFT);
 
 	threshold = 2 * fls(num_online_cpus()) * (1 + fls(mem));
 

@@ -1569,7 +1569,7 @@ static void zoneinfo_show_print(struct seq_file *m, pg_data_t *pgdat,
 		   high_wmark_pages(zone),
 		   zone->spanned_pages,
 		   zone->present_pages,
-		   zone->managed_pages);
+		   atomic_long_read(&zone->managed_pages));
 
 	seq_printf(m,
 		   "\n        protection: (%ld",

mm/workingset.c +1 -1

index b15799d..dcd4e16 100644
@@ -550,7 +550,7 @@ static int __init workingset_init(void)
 	 * double the initial memory by using totalram_pages as-is.
 	 */
 	timestamp_bits = BITS_PER_LONG - EVICTION_SHIFT;
-	max_order = fls_long(totalram_pages - 1);
+	max_order = fls_long(atomic_long_read(&totalram_pages) - 1);
 	if (max_order > timestamp_bits)
 		bucket_order = max_order - timestamp_bits;
 	pr_info("workingset: timestamp_bits=%d max_order=%d bucket_order=%u\n",

mm/zswap.c +1 -1

index cd91fd9..5d2d7b9 100644
@@ -219,7 +219,7 @@ struct zswap_tree {
 
 static bool zswap_is_full(void)
 {
-	return totalram_pages * zswap_max_pool_percent / 100 <
+	return atomic_long_read(&totalram_pages) * zswap_max_pool_percent / 100 <
 		DIV_ROUND_UP(zswap_pool_total_size, PAGE_SIZE);
 }
 

net/dccp/proto.c +3 -3

index 875858c..4a92d11 100644
@@ -1154,10 +1154,10 @@ static int __init dccp_init(void)
 	 *
 	 * The methodology is similar to that of the buffer cache.
 	 */
-	if (totalram_pages >= (128 * 1024))
-		goal = totalram_pages >> (21 - PAGE_SHIFT);
+	if (atomic_long_read(&totalram_pages) >= (128 * 1024))
+		goal = atomic_long_read(&totalram_pages) >> (21 - PAGE_SHIFT);
 	else
-		goal = totalram_pages >> (23 - PAGE_SHIFT);
+		goal = atomic_long_read(&totalram_pages) >> (23 - PAGE_SHIFT);
 
 	if (thash_entries)
 		goal = (thash_entries *

net/decnet/dn_route.c +1 -1

index 1c002c0..bb49b0f 100644
@@ -1866,7 +1866,7 @@ void __init dn_route_init(void)
 	dn_route_timer.expires = jiffies + decnet_dst_gc_interval * HZ;
 	add_timer(&dn_route_timer);
 
-	goal = totalram_pages >> (26 - PAGE_SHIFT);
+	goal = atomic_long_read(&totalram_pages) >> (26 - PAGE_SHIFT);
 
 	for(order = 0; (1UL << order) < goal; order++)
 		/* NOTHING */;

net/ipv4/tcp_metrics.c +1 -1

index 03b51cd..d91bdab 100644
@@ -1000,7 +1000,7 @@ static int __net_init tcp_net_metrics_init(struct net *net)
 
 	slots = tcpmhash_entries;
 	if (!slots) {
-		if (totalram_pages >= 128 * 1024)
+		if (atomic_long_read(&totalram_pages) >= 128 * 1024)
 			slots = 16 * 1024;
 		else
 			slots = 8 * 1024;

net/netfilter/nf_conntrack_core.c +3 -3

index ca1168d..3285df9 100644
@@ -2267,11 +2267,11 @@ int nf_conntrack_init_start(void)
 		 * >= 4GB machines have 65536 buckets.
 		 */
 		nf_conntrack_htable_size
-			= (((totalram_pages << PAGE_SHIFT) / 16384)
+			= (((atomic_long_read(&totalram_pages) << PAGE_SHIFT) / 16384)
 			   / sizeof(struct hlist_head));
-		if (totalram_pages > (4 * (1024 * 1024 * 1024 / PAGE_SIZE)))
+		if (atomic_long_read(&totalram_pages) > (4 * (1024 * 1024 * 1024 / PAGE_SIZE)))
 			nf_conntrack_htable_size = 65536;
-		else if (totalram_pages > (1024 * 1024 * 1024 / PAGE_SIZE))
+		else if (atomic_long_read(&totalram_pages) > (1024 * 1024 * 1024 / PAGE_SIZE))
 			nf_conntrack_htable_size = 16384;
 		if (nf_conntrack_htable_size < 32)
 			nf_conntrack_htable_size = 32;

net/netfilter/xt_hashlimit.c +2 -2

index 3e7d259..3c79a0f 100644
@@ -279,9 +279,9 @@ static int htable_create(struct net *net, struct hashlimit_cfg3 *cfg,
 	if (cfg->size) {
 		size = cfg->size;
 	} else {
-		size = (totalram_pages << PAGE_SHIFT) / 16384 /
+		size = (atomic_long_read(&totalram_pages) << PAGE_SHIFT) / 16384 /
 		       sizeof(struct hlist_head);
-		if (totalram_pages > 1024 * 1024 * 1024 / PAGE_SIZE)
+		if (atomic_long_read(&totalram_pages) > 1024 * 1024 * 1024 / PAGE_SIZE)
 			size = 8192;
 		if (size < 16)
 			size = 16;

net/sctp/protocol.c +3 -3

index 9b277bd..4ca4def 100644
@@ -1426,10 +1426,10 @@ static __init int sctp_init(void)
 	 * The methodology is similar to that of the tcp hash tables.
 	 * Though not identical.  Start by getting a goal size
 	 */
-	if (totalram_pages >= (128 * 1024))
-		goal = totalram_pages >> (22 - PAGE_SHIFT);
+	if (atomic_long_read(&totalram_pages) >= (128 * 1024))
+		goal = atomic_long_read(&totalram_pages) >> (22 - PAGE_SHIFT);
 	else
-		goal = totalram_pages >> (24 - PAGE_SHIFT);
+		goal = atomic_long_read(&totalram_pages) >> (24 - PAGE_SHIFT);
 
 	/* Then compute the page order for said goal */
 	order = get_order(goal);

security/integrity/ima/ima_kexec.c +0 -0

index 16bd187..8bb32ad 100644