mbox series

[v7,00/15] Free some vmemmap pages of hugetlb page

Message ID 20201130151838.11208-1-songmuchun@bytedance.com (mailing list archive)
Headers show
Series Free some vmemmap pages of hugetlb page | expand

Message

Muchun Song Nov. 30, 2020, 3:18 p.m. UTC 
Hi all,

This patch series will free some vmemmap pages(struct page structures)
associated with each hugetlbpage when preallocated to save memory.

In order to reduce the difficulty of the first version of code review.
From this version, we disable PMD/huge page mapping of vmemmap if this
feature was enabled. This accutualy eliminate a bunch of the complex code
doing page table manipulation. When this patch series is solid, we cam add
the code of vmemmap page table manipulation in the future.

The struct page structures (page structs) are used to describe a physical
page frame. By default, there is a one-to-one mapping from a page frame to
it's corresponding page struct.

The HugeTLB pages consist of multiple base page size pages and is supported
by many architectures. See hugetlbpage.rst in the Documentation directory
for more details. On the x86 architecture, HugeTLB pages of size 2MB and 1GB
are currently supported. Since the base page size on x86 is 4KB, a 2MB
HugeTLB page consists of 512 base pages and a 1GB HugeTLB page consists of
4096 base pages. For each base page, there is a corresponding page struct.

Within the HugeTLB subsystem, only the first 4 page structs are used to
contain unique information about a HugeTLB page. HUGETLB_CGROUP_MIN_ORDER
provides this upper limit. The only 'useful' information in the remaining
page structs is the compound_head field, and this field is the same for all
tail pages.

By removing redundant page structs for HugeTLB pages, memory can returned to
the buddy allocator for other uses.

When the system boot up, every 2M HugeTLB has 512 struct page structs which
size is 8 pages(sizeof(struct page) * 512 / PAGE_SIZE).

    HugeTLB                  struct pages(8 pages)         page frame(8 pages)
 +-----------+ ---virt_to_page---> +-----------+   mapping to   +-----------+
 |           |                     |     0     | -------------> |     0     |
 |           |                     +-----------+                +-----------+
 |           |                     |     1     | -------------> |     1     |
 |           |                     +-----------+                +-----------+
 |           |                     |     2     | -------------> |     2     |
 |           |                     +-----------+                +-----------+
 |           |                     |     3     | -------------> |     3     |
 |           |                     +-----------+                +-----------+
 |           |                     |     4     | -------------> |     4     |
 |    2MB    |                     +-----------+                +-----------+
 |           |                     |     5     | -------------> |     5     |
 |           |                     +-----------+                +-----------+
 |           |                     |     6     | -------------> |     6     |
 |           |                     +-----------+                +-----------+
 |           |                     |     7     | -------------> |     7     |
 |           |                     +-----------+                +-----------+
 |           |
 |           |
 |           |
 +-----------+

The value of page->compound_head is the same for all tail pages. The first
page of page structs (page 0) associated with the HugeTLB page contains the 4
page structs necessary to describe the HugeTLB. The only use of the remaining
pages of page structs (page 1 to page 7) is to point to page->compound_head.
Therefore, we can remap pages 2 to 7 to page 1. Only 2 pages of page structs
will be used for each HugeTLB page. This will allow us to free the remaining
6 pages to the buddy allocator.

Here is how things look after remapping.

    HugeTLB                  struct pages(8 pages)         page frame(8 pages)
 +-----------+ ---virt_to_page---> +-----------+   mapping to   +-----------+
 |           |                     |     0     | -------------> |     0     |
 |           |                     +-----------+                +-----------+
 |           |                     |     1     | -------------> |     1     |
 |           |                     +-----------+                +-----------+
 |           |                     |     2     | ----------------^ ^ ^ ^ ^ ^
 |           |                     +-----------+                   | | | | |
 |           |                     |     3     | ------------------+ | | | |
 |           |                     +-----------+                     | | | |
 |           |                     |     4     | --------------------+ | | |
 |    2MB    |                     +-----------+                       | | |
 |           |                     |     5     | ----------------------+ | |
 |           |                     +-----------+                         | |
 |           |                     |     6     | ------------------------+ |
 |           |                     +-----------+                           |
 |           |                     |     7     | --------------------------+
 |           |                     +-----------+
 |           |
 |           |
 |           |
 +-----------+

When a HugeTLB is freed to the buddy system, we should allocate 6 pages for
vmemmap pages and restore the previous mapping relationship.

Apart from 2MB HugeTLB page, we also have 1GB HugeTLB page. It is similar
to the 2MB HugeTLB page. We also can use this approach to free the vmemmap
pages.

In this case, for the 1GB HugeTLB page, we can save 4088 pages(There are
4096 pages for struct page structs, we reserve 2 pages for vmemmap and 8
pages for page tables. So we can save 4088 pages). This is a very substantial
gain. On our server, run some SPDK/QEMU applications which will use 1024GB
hugetlbpage. With this feature enabled, we can save ~16GB(1G hugepage)/~11GB
(2MB hugepage, the worst case is 10GB while the best is 12GB) memory.

Because there are vmemmap page tables reconstruction on the freeing/allocating
path, it increases some overhead. Here are some overhead analysis.

1) Allocating 10240 2MB hugetlb pages.

   a) With this patch series applied:
   # time echo 10240 > /proc/sys/vm/nr_hugepages

   real     0m0.166s
   user     0m0.000s
   sys      0m0.166s

   # bpftrace -e 'kprobe:alloc_fresh_huge_page { @start[tid] = nsecs; } kretprobe:alloc_fresh_huge_page /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
   Attaching 2 probes...

   @latency:
   [8K, 16K)           8360 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
   [16K, 32K)          1868 |@@@@@@@@@@@                                         |
   [32K, 64K)            10 |                                                    |
   [64K, 128K)            2 |                                                    |

   b) Without this patch series:
   # time echo 10240 > /proc/sys/vm/nr_hugepages

   real     0m0.066s
   user     0m0.000s
   sys      0m0.066s

   # bpftrace -e 'kprobe:alloc_fresh_huge_page { @start[tid] = nsecs; } kretprobe:alloc_fresh_huge_page /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
   Attaching 2 probes...

   @latency:
   [4K, 8K)           10176 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
   [8K, 16K)             62 |                                                    |
   [16K, 32K)             2 |                                                    |

   Summarize: this feature is about ~2x slower than before.

2) Freeing 10240 2MB hugetlb pages.

   a) With this patch series applied:
   # time echo 0 > /proc/sys/vm/nr_hugepages

   real     0m0.004s
   user     0m0.000s
   sys      0m0.002s

   # bpftrace -e 'kprobe:__free_hugepage { @start[tid] = nsecs; } kretprobe:__free_hugepage /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
   Attaching 2 probes...

   @latency:
   [16K, 32K)         10240 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|

   b) Without this patch series:
   # time echo 0 > /proc/sys/vm/nr_hugepages

   real     0m0.077s
   user     0m0.001s
   sys      0m0.075s

   # bpftrace -e 'kprobe:__free_hugepage { @start[tid] = nsecs; } kretprobe:__free_hugepage /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
   Attaching 2 probes...

   @latency:
   [4K, 8K)            9950 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
   [8K, 16K)            287 |@                                                   |
   [16K, 32K)             3 |                                                    |

   Summarize: The overhead of __free_hugepage is about ~2-4x slower than before.
              But according to the allocation test above, I think that here is
	      also ~2x slower than before.

              But why the 'real' time of patched is smaller than before? Because
	      In this patch series, the freeing hugetlb is asynchronous(through
	      kwoker).

Although the overhead has increased, the overhead is not significant. Like Mike
said, "However, remember that the majority of use cases create hugetlb pages at
or shortly after boot time and add them to the pool. So, additional overhead is
at pool creation time. There is no change to 'normal run time' operations of
getting a page from or returning a page to the pool (think page fault/unmap)".

Todo:
  1. Free all of the tail vmemmap pages
     Now for the 2MB HugrTLB page, we only free 6 vmemmap pages. we really can
     free 7 vmemmap pages. In this case, we can see 8 of the 512 struct page
     structures has beed set PG_head flag. If we can adjust compound_head()
     slightly and make compound_head() return the real head struct page when
     the parameter is the tail struct page but with PG_head flag set.

     In order to make the code evolution route clearer. This feature can can be
     a separate patch after this patchset is solid.
  2. Support for other architectures (e.g. aarch64).
  3. Enable PMD/huge page mapping of vmemmap even if this feature was enabled.

Changelog in v7:
  1. Rebase to linux-next 20201130
  2. Do not use basepage mapping for vmemmap when this feature is disabled.
     Thanks to Oscar and Barry.
  3. Rework some patchs.
     [PATCH v6 08/16] mm/hugetlb: Free the vmemmap pages associated with each hugetlb page
     [PATCH v6 10/16] mm/hugetlb: Allocate the vmemmap pages associated with each hugetlb page

Changelog in v6:
  1. Disable PMD/huge page mapping of vmemmap if this feature was enabled.
  2. Simplify the first version code.

Changelog in v5:
  1. Rework somme comments and code in the [PATCH v4 04/21] and [PATCH v4 05/21].
     Thanks to Mike and Oscar's suggestions.

Changelog in v4:
  1. Move all the vmemmap functions to hugetlb_vmemmap.c.
  2. Make the CONFIG_HUGETLB_PAGE_FREE_VMEMMAP default to y, if we want to
     disable this feature, we should disable it by a boot/kernel command line.
  3. Remove vmemmap_pgtable_{init, deposit, withdraw}() helper functions.
  4. Initialize page table lock for vmemmap through core_initcall mechanism.

  Thanks for Mike and Oscar's suggestions.

Changelog in v3:
  1. Rename some helps function name. Thanks Mike.
  2. Rework some code. Thanks Mike and Oscar.
  3. Remap the tail vmemmap page with PAGE_KERNEL_RO instead of
     PAGE_KERNEL. Thanks Matthew.
  4. Add some overhead analysis in the cover letter.
  5. Use vmemap pmd table lock instead of a hugetlb specific global lock.

Changelog in v2:
  1. Fix do not call dissolve_compound_page in alloc_huge_page_vmemmap().
  2. Fix some typo and code style problems.
  3. Remove unused handle_vmemmap_fault().
  4. Merge some commits to one commit suggested by Mike.

Muchun Song (15):
  mm/memory_hotplug: Move bootmem info registration API to
    bootmem_info.c
  mm/memory_hotplug: Move {get,put}_page_bootmem() to bootmem_info.c
  mm/hugetlb: Introduce a new config HUGETLB_PAGE_FREE_VMEMMAP
  mm/hugetlb: Introduce nr_free_vmemmap_pages in the struct hstate
  mm/bootmem_info: Introduce {free,prepare}_vmemmap_page()
  mm/hugetlb: Disable freeing vmemmap if struct page size is not power
    of two
  x86/mm/64: Disable PMD page mapping of vmemmap
  mm/hugetlb: Free the vmemmap pages associated with each hugetlb page
  mm/hugetlb: Defer freeing of HugeTLB pages
  mm/hugetlb: Allocate the vmemmap pages associated with each hugetlb
    page
  mm/hugetlb: Set the PageHWPoison to the raw error page
  mm/hugetlb: Flush work when dissolving hugetlb page
  mm/hugetlb: Add a kernel parameter hugetlb_free_vmemmap
  mm/hugetlb: Gather discrete indexes of tail page
  mm/hugetlb: Add BUILD_BUG_ON to catch invalid usage of tail struct
    page

 Documentation/admin-guide/kernel-parameters.txt |   9 +
 Documentation/admin-guide/mm/hugetlbpage.rst    |   3 +
 arch/x86/mm/init_64.c                           |  13 +-
 fs/Kconfig                                      |  14 +
 include/linux/bootmem_info.h                    |  64 +++++
 include/linux/hugetlb.h                         |  35 +++
 include/linux/hugetlb_cgroup.h                  |  15 +-
 include/linux/memory_hotplug.h                  |  27 --
 mm/Makefile                                     |   2 +
 mm/bootmem_info.c                               | 124 ++++++++
 mm/hugetlb.c                                    | 144 ++++++++--
 mm/hugetlb_vmemmap.c                            | 367 ++++++++++++++++++++++++
 mm/hugetlb_vmemmap.h                            |  79 +++++
 mm/memory_hotplug.c                             | 116 --------
 mm/sparse.c                                     |   1 +
 15 files changed, 834 insertions(+), 179 deletions(-)
 create mode 100644 include/linux/bootmem_info.h
 create mode 100644 mm/bootmem_info.c
 create mode 100644 mm/hugetlb_vmemmap.c
 create mode 100644 mm/hugetlb_vmemmap.h

Comments

Muchun Song Dec. 3, 2020, 8:35 a.m. UTC | #1 
On Mon, Nov 30, 2020 at 11:19 PM Muchun Song <songmuchun@bytedance.com> wrote:
>
> Hi all,
>
> This patch series will free some vmemmap pages(struct page structures)
> associated with each hugetlbpage when preallocated to save memory.

Hi Mike,

What's your opinion on this version?  Any comments or suggestions?
And hoping you or more people review the series. Thank you very
much.

>
> In order to reduce the difficulty of the first version of code review.
> From this version, we disable PMD/huge page mapping of vmemmap if this
> feature was enabled. This accutualy eliminate a bunch of the complex code
> doing page table manipulation. When this patch series is solid, we cam add
> the code of vmemmap page table manipulation in the future.
>
> The struct page structures (page structs) are used to describe a physical
> page frame. By default, there is a one-to-one mapping from a page frame to
> it's corresponding page struct.
>
> The HugeTLB pages consist of multiple base page size pages and is supported
> by many architectures. See hugetlbpage.rst in the Documentation directory
> for more details. On the x86 architecture, HugeTLB pages of size 2MB and 1GB
> are currently supported. Since the base page size on x86 is 4KB, a 2MB
> HugeTLB page consists of 512 base pages and a 1GB HugeTLB page consists of
> 4096 base pages. For each base page, there is a corresponding page struct.
>
> Within the HugeTLB subsystem, only the first 4 page structs are used to
> contain unique information about a HugeTLB page. HUGETLB_CGROUP_MIN_ORDER
> provides this upper limit. The only 'useful' information in the remaining
> page structs is the compound_head field, and this field is the same for all
> tail pages.
>
> By removing redundant page structs for HugeTLB pages, memory can returned to
> the buddy allocator for other uses.
>
> When the system boot up, every 2M HugeTLB has 512 struct page structs which
> size is 8 pages(sizeof(struct page) * 512 / PAGE_SIZE).
>
>     HugeTLB                  struct pages(8 pages)         page frame(8 pages)
>  +-----------+ ---virt_to_page---> +-----------+   mapping to   +-----------+
>  |           |                     |     0     | -------------> |     0     |
>  |           |                     +-----------+                +-----------+
>  |           |                     |     1     | -------------> |     1     |
>  |           |                     +-----------+                +-----------+
>  |           |                     |     2     | -------------> |     2     |
>  |           |                     +-----------+                +-----------+
>  |           |                     |     3     | -------------> |     3     |
>  |           |                     +-----------+                +-----------+
>  |           |                     |     4     | -------------> |     4     |
>  |    2MB    |                     +-----------+                +-----------+
>  |           |                     |     5     | -------------> |     5     |
>  |           |                     +-----------+                +-----------+
>  |           |                     |     6     | -------------> |     6     |
>  |           |                     +-----------+                +-----------+
>  |           |                     |     7     | -------------> |     7     |
>  |           |                     +-----------+                +-----------+
>  |           |
>  |           |
>  |           |
>  +-----------+
>
> The value of page->compound_head is the same for all tail pages. The first
> page of page structs (page 0) associated with the HugeTLB page contains the 4
> page structs necessary to describe the HugeTLB. The only use of the remaining
> pages of page structs (page 1 to page 7) is to point to page->compound_head.
> Therefore, we can remap pages 2 to 7 to page 1. Only 2 pages of page structs
> will be used for each HugeTLB page. This will allow us to free the remaining
> 6 pages to the buddy allocator.
>
> Here is how things look after remapping.
>
>     HugeTLB                  struct pages(8 pages)         page frame(8 pages)
>  +-----------+ ---virt_to_page---> +-----------+   mapping to   +-----------+
>  |           |                     |     0     | -------------> |     0     |
>  |           |                     +-----------+                +-----------+
>  |           |                     |     1     | -------------> |     1     |
>  |           |                     +-----------+                +-----------+
>  |           |                     |     2     | ----------------^ ^ ^ ^ ^ ^
>  |           |                     +-----------+                   | | | | |
>  |           |                     |     3     | ------------------+ | | | |
>  |           |                     +-----------+                     | | | |
>  |           |                     |     4     | --------------------+ | | |
>  |    2MB    |                     +-----------+                       | | |
>  |           |                     |     5     | ----------------------+ | |
>  |           |                     +-----------+                         | |
>  |           |                     |     6     | ------------------------+ |
>  |           |                     +-----------+                           |
>  |           |                     |     7     | --------------------------+
>  |           |                     +-----------+
>  |           |
>  |           |
>  |           |
>  +-----------+
>
> When a HugeTLB is freed to the buddy system, we should allocate 6 pages for
> vmemmap pages and restore the previous mapping relationship.
>
> Apart from 2MB HugeTLB page, we also have 1GB HugeTLB page. It is similar
> to the 2MB HugeTLB page. We also can use this approach to free the vmemmap
> pages.
>
> In this case, for the 1GB HugeTLB page, we can save 4088 pages(There are
> 4096 pages for struct page structs, we reserve 2 pages for vmemmap and 8
> pages for page tables. So we can save 4088 pages). This is a very substantial
> gain. On our server, run some SPDK/QEMU applications which will use 1024GB
> hugetlbpage. With this feature enabled, we can save ~16GB(1G hugepage)/~11GB
> (2MB hugepage, the worst case is 10GB while the best is 12GB) memory.
>
> Because there are vmemmap page tables reconstruction on the freeing/allocating
> path, it increases some overhead. Here are some overhead analysis.
>
> 1) Allocating 10240 2MB hugetlb pages.
>
>    a) With this patch series applied:
>    # time echo 10240 > /proc/sys/vm/nr_hugepages
>
>    real     0m0.166s
>    user     0m0.000s
>    sys      0m0.166s
>
>    # bpftrace -e 'kprobe:alloc_fresh_huge_page { @start[tid] = nsecs; } kretprobe:alloc_fresh_huge_page /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
>    Attaching 2 probes...
>
>    @latency:
>    [8K, 16K)           8360 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
>    [16K, 32K)          1868 |@@@@@@@@@@@                                         |
>    [32K, 64K)            10 |                                                    |
>    [64K, 128K)            2 |                                                    |
>
>    b) Without this patch series:
>    # time echo 10240 > /proc/sys/vm/nr_hugepages
>
>    real     0m0.066s
>    user     0m0.000s
>    sys      0m0.066s
>
>    # bpftrace -e 'kprobe:alloc_fresh_huge_page { @start[tid] = nsecs; } kretprobe:alloc_fresh_huge_page /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
>    Attaching 2 probes...
>
>    @latency:
>    [4K, 8K)           10176 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
>    [8K, 16K)             62 |                                                    |
>    [16K, 32K)             2 |                                                    |
>
>    Summarize: this feature is about ~2x slower than before.
>
> 2) Freeing 10240 2MB hugetlb pages.
>
>    a) With this patch series applied:
>    # time echo 0 > /proc/sys/vm/nr_hugepages
>
>    real     0m0.004s
>    user     0m0.000s
>    sys      0m0.002s
>
>    # bpftrace -e 'kprobe:__free_hugepage { @start[tid] = nsecs; } kretprobe:__free_hugepage /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
>    Attaching 2 probes...
>
>    @latency:
>    [16K, 32K)         10240 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
>
>    b) Without this patch series:
>    # time echo 0 > /proc/sys/vm/nr_hugepages
>
>    real     0m0.077s
>    user     0m0.001s
>    sys      0m0.075s
>
>    # bpftrace -e 'kprobe:__free_hugepage { @start[tid] = nsecs; } kretprobe:__free_hugepage /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
>    Attaching 2 probes...
>
>    @latency:
>    [4K, 8K)            9950 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
>    [8K, 16K)            287 |@                                                   |
>    [16K, 32K)             3 |                                                    |
>
>    Summarize: The overhead of __free_hugepage is about ~2-4x slower than before.
>               But according to the allocation test above, I think that here is
>               also ~2x slower than before.
>
>               But why the 'real' time of patched is smaller than before? Because
>               In this patch series, the freeing hugetlb is asynchronous(through
>               kwoker).
>
> Although the overhead has increased, the overhead is not significant. Like Mike
> said, "However, remember that the majority of use cases create hugetlb pages at
> or shortly after boot time and add them to the pool. So, additional overhead is
> at pool creation time. There is no change to 'normal run time' operations of
> getting a page from or returning a page to the pool (think page fault/unmap)".
>
> Todo:
>   1. Free all of the tail vmemmap pages
>      Now for the 2MB HugrTLB page, we only free 6 vmemmap pages. we really can
>      free 7 vmemmap pages. In this case, we can see 8 of the 512 struct page
>      structures has beed set PG_head flag. If we can adjust compound_head()
>      slightly and make compound_head() return the real head struct page when
>      the parameter is the tail struct page but with PG_head flag set.
>
>      In order to make the code evolution route clearer. This feature can can be
>      a separate patch after this patchset is solid.
>   2. Support for other architectures (e.g. aarch64).
>   3. Enable PMD/huge page mapping of vmemmap even if this feature was enabled.
>
> Changelog in v7:
>   1. Rebase to linux-next 20201130
>   2. Do not use basepage mapping for vmemmap when this feature is disabled.
>      Thanks to Oscar and Barry.
>   3. Rework some patchs.
>      [PATCH v6 08/16] mm/hugetlb: Free the vmemmap pages associated with each hugetlb page
>      [PATCH v6 10/16] mm/hugetlb: Allocate the vmemmap pages associated with each hugetlb page
>
> Changelog in v6:
>   1. Disable PMD/huge page mapping of vmemmap if this feature was enabled.
>   2. Simplify the first version code.
>
> Changelog in v5:
>   1. Rework somme comments and code in the [PATCH v4 04/21] and [PATCH v4 05/21].
>      Thanks to Mike and Oscar's suggestions.
>
> Changelog in v4:
>   1. Move all the vmemmap functions to hugetlb_vmemmap.c.
>   2. Make the CONFIG_HUGETLB_PAGE_FREE_VMEMMAP default to y, if we want to
>      disable this feature, we should disable it by a boot/kernel command line.
>   3. Remove vmemmap_pgtable_{init, deposit, withdraw}() helper functions.
>   4. Initialize page table lock for vmemmap through core_initcall mechanism.
>
>   Thanks for Mike and Oscar's suggestions.
>
> Changelog in v3:
>   1. Rename some helps function name. Thanks Mike.
>   2. Rework some code. Thanks Mike and Oscar.
>   3. Remap the tail vmemmap page with PAGE_KERNEL_RO instead of
>      PAGE_KERNEL. Thanks Matthew.
>   4. Add some overhead analysis in the cover letter.
>   5. Use vmemap pmd table lock instead of a hugetlb specific global lock.
>
> Changelog in v2:
>   1. Fix do not call dissolve_compound_page in alloc_huge_page_vmemmap().
>   2. Fix some typo and code style problems.
>   3. Remove unused handle_vmemmap_fault().
>   4. Merge some commits to one commit suggested by Mike.
>
> Muchun Song (15):
>   mm/memory_hotplug: Move bootmem info registration API to
>     bootmem_info.c
>   mm/memory_hotplug: Move {get,put}_page_bootmem() to bootmem_info.c
>   mm/hugetlb: Introduce a new config HUGETLB_PAGE_FREE_VMEMMAP
>   mm/hugetlb: Introduce nr_free_vmemmap_pages in the struct hstate
>   mm/bootmem_info: Introduce {free,prepare}_vmemmap_page()
>   mm/hugetlb: Disable freeing vmemmap if struct page size is not power
>     of two
>   x86/mm/64: Disable PMD page mapping of vmemmap
>   mm/hugetlb: Free the vmemmap pages associated with each hugetlb page
>   mm/hugetlb: Defer freeing of HugeTLB pages
>   mm/hugetlb: Allocate the vmemmap pages associated with each hugetlb
>     page
>   mm/hugetlb: Set the PageHWPoison to the raw error page
>   mm/hugetlb: Flush work when dissolving hugetlb page
>   mm/hugetlb: Add a kernel parameter hugetlb_free_vmemmap
>   mm/hugetlb: Gather discrete indexes of tail page
>   mm/hugetlb: Add BUILD_BUG_ON to catch invalid usage of tail struct
>     page
>
>  Documentation/admin-guide/kernel-parameters.txt |   9 +
>  Documentation/admin-guide/mm/hugetlbpage.rst    |   3 +
>  arch/x86/mm/init_64.c                           |  13 +-
>  fs/Kconfig                                      |  14 +
>  include/linux/bootmem_info.h                    |  64 +++++
>  include/linux/hugetlb.h                         |  35 +++
>  include/linux/hugetlb_cgroup.h                  |  15 +-
>  include/linux/memory_hotplug.h                  |  27 --
>  mm/Makefile                                     |   2 +
>  mm/bootmem_info.c                               | 124 ++++++++
>  mm/hugetlb.c                                    | 144 ++++++++--
>  mm/hugetlb_vmemmap.c                            | 367 ++++++++++++++++++++++++
>  mm/hugetlb_vmemmap.h                            |  79 +++++
>  mm/memory_hotplug.c                             | 116 --------
>  mm/sparse.c                                     |   1 +
>  15 files changed, 834 insertions(+), 179 deletions(-)
>  create mode 100644 include/linux/bootmem_info.h
>  create mode 100644 mm/bootmem_info.c
>  create mode 100644 mm/hugetlb_vmemmap.c
>  create mode 100644 mm/hugetlb_vmemmap.h
>
> --
> 2.11.0
>


--
Yours,
Muchun
Mike Kravetz Dec. 3, 2020, 11:48 p.m. UTC | #2 
On 12/3/20 12:35 AM, Muchun Song wrote:
> On Mon, Nov 30, 2020 at 11:19 PM Muchun Song <songmuchun@bytedance.com> wrote:
>>
>> Hi all,
>>
>> This patch series will free some vmemmap pages(struct page structures)
>> associated with each hugetlbpage when preallocated to save memory.
> 
> Hi Mike,
> 
> What's your opinion on this version?  Any comments or suggestions?
> And hoping you or more people review the series. Thank you very
> much.

Sorry Muchun, I have been busy with other things and have not looked at
this new version.  Should have some time soon.

As previously mentioned, I feel qualified to review the hugetlb changes
and some other closely related changes.  However, this patch set is
touching quite a few areas and I do not feel qualified to make authoritative
statements about them all.  I too hope others will take a look.
Muchun Song Dec. 4, 2020, 3:39 a.m. UTC | #3 
On Fri, Dec 4, 2020 at 7:49 AM Mike Kravetz <mike.kravetz@oracle.com> wrote:
>
> On 12/3/20 12:35 AM, Muchun Song wrote:
> > On Mon, Nov 30, 2020 at 11:19 PM Muchun Song <songmuchun@bytedance.com> wrote:
> >>
> >> Hi all,
> >>
> >> This patch series will free some vmemmap pages(struct page structures)
> >> associated with each hugetlbpage when preallocated to save memory.
> >
> > Hi Mike,
> >
> > What's your opinion on this version?  Any comments or suggestions?
> > And hoping you or more people review the series. Thank you very
> > much.
>
> Sorry Muchun, I have been busy with other things and have not looked at
> this new version.  Should have some time soon.

Thanks very much.

>
> As previously mentioned, I feel qualified to review the hugetlb changes
> and some other closely related changes.  However, this patch set is
> touching quite a few areas and I do not feel qualified to make authoritative
> statements about them all.  I too hope others will take a look.

Agree. I also hope others can take a look at other modules(e.g.
sparse-vmemmap, memory-hotplug). Thanks for everyone's efforts
on this.

> --
> Mike Kravetz
Oscar Salvador Dec. 7, 2020, 6:38 p.m. UTC | #4 
On Fri, Dec 04, 2020 at 11:39:31AM +0800, Muchun Song wrote:
> On Fri, Dec 4, 2020 at 7:49 AM Mike Kravetz <mike.kravetz@oracle.com> wrote:
> > As previously mentioned, I feel qualified to review the hugetlb changes
> > and some other closely related changes.  However, this patch set is
> > touching quite a few areas and I do not feel qualified to make authoritative
> > statements about them all.  I too hope others will take a look.
> 
> Agree. I also hope others can take a look at other modules(e.g.
> sparse-vmemmap, memory-hotplug). Thanks for everyone's efforts
> on this.

I got sidetracked by some other stuff but I plan to continue reviewing
this series.

One thing that came to my mind is that if we do as David suggested in
patch#4, and we move it towards the end to actually __enable__ this
once all the infrastructure is there (unless hstate->nr_vmemmap_pages
differs from 0 we should not be doing any work AFAIK), we could also
move patch#6 to the end (right before the enablement), kill patch#7
and only leave patch#13.

The reason for that (killing patch#7 and leaving patch#13 only)
is that it does not make much sense to me to disable PMD-mapped vmemmap
depending on the CONFIG_HUGETLB_xxxxx as that is enabled by default
to replace that later by the boot kernel parameter.
It looks more natural to me to disable it when we introduce the kernel
boot parameter, before the actual enablement of the feature.

As I said, I plan to start the review again, but the order above would
make more sense to me.

thanks
Muchun Song Dec. 8, 2020, 2:26 a.m. UTC | #5 
On Tue, Dec 8, 2020 at 2:38 AM Oscar Salvador <osalvador@suse.de> wrote:
>
> On Fri, Dec 04, 2020 at 11:39:31AM +0800, Muchun Song wrote:
> > On Fri, Dec 4, 2020 at 7:49 AM Mike Kravetz <mike.kravetz@oracle.com> wrote:
> > > As previously mentioned, I feel qualified to review the hugetlb changes
> > > and some other closely related changes.  However, this patch set is
> > > touching quite a few areas and I do not feel qualified to make authoritative
> > > statements about them all.  I too hope others will take a look.
> >
> > Agree. I also hope others can take a look at other modules(e.g.
> > sparse-vmemmap, memory-hotplug). Thanks for everyone's efforts
> > on this.
>
> I got sidetracked by some other stuff but I plan to continue reviewing
> this series.

Many thanks, Oscar.

>
> One thing that came to my mind is that if we do as David suggested in
> patch#4, and we move it towards the end to actually __enable__ this
> once all the infrastructure is there (unless hstate->nr_vmemmap_pages
> differs from 0 we should not be doing any work AFAIK), we could also
> move patch#6 to the end (right before the enablement), kill patch#7
> and only leave patch#13.
>
> The reason for that (killing patch#7 and leaving patch#13 only)
> is that it does not make much sense to me to disable PMD-mapped vmemmap
> depending on the CONFIG_HUGETLB_xxxxx as that is enabled by default
> to replace that later by the boot kernel parameter.
> It looks more natural to me to disable it when we introduce the kernel
> boot parameter, before the actual enablement of the feature.

Thanks for your suggestions. I agree with you. :)

>
> As I said, I plan to start the review again, but the order above would
> make more sense to me.
>
> thanks
>
> --
> Oscar Salvador
> SUSE L3