Skip to content

L
linux
Commit 44f380fe authored by Mike Rapoport's avatar Mike Rapoport Committed by Jonathan Corbet
Browse files
Options

docs/vm: transhuge.txt: convert to ReST format 

Signed-off-by: default avatarMike Rapoport <rppt@linux.vnet.ibm.com>
Signed-off-by: default avatarJonathan Corbet <corbet@lwn.net>
parent 54319982
Hide whitespace changes
Inline Side-by-side
Showing with 166 additions and 120 deletions
Documentation/vm/transhuge.txt
View file @ 44f380fe
= Transparent Hugepage Support = .. _transhuge:
== Objective == ============================
Transparent Hugepage Support
============================
Objective
=========
Performance critical computing applications dealing with large memory Performance critical computing applications dealing with large memory
working sets are already running on top of libhugetlbfs and in turn working sets are already running on top of libhugetlbfs and in turn
...@@ -33,7 +38,8 @@ are using hugepages but a significant speedup already happens if only ...@@ -33,7 +38,8 @@ are using hugepages but a significant speedup already happens if only
one of the two is using hugepages just because of the fact the TLB one of the two is using hugepages just because of the fact the TLB
miss is going to run faster. miss is going to run faster.
== Design == Design
======
- "graceful fallback": mm components which don't have transparent hugepage - "graceful fallback": mm components which don't have transparent hugepage
knowledge fall back to breaking huge pmd mapping into table of ptes and, knowledge fall back to breaking huge pmd mapping into table of ptes and,
...@@ -88,16 +94,17 @@ Applications that gets a lot of benefit from hugepages and that don't ...@@ -88,16 +94,17 @@ Applications that gets a lot of benefit from hugepages and that don't
risk to lose memory by using hugepages, should use risk to lose memory by using hugepages, should use
madvise(MADV_HUGEPAGE) on their critical mmapped regions. madvise(MADV_HUGEPAGE) on their critical mmapped regions.
== sysfs == sysfs
=====
Transparent Hugepage Support for anonymous memory can be entirely disabled Transparent Hugepage Support for anonymous memory can be entirely disabled
(mostly for debugging purposes) or only enabled inside MADV_HUGEPAGE (mostly for debugging purposes) or only enabled inside MADV_HUGEPAGE
regions (to avoid the risk of consuming more memory resources) or enabled regions (to avoid the risk of consuming more memory resources) or enabled
system wide. This can be achieved with one of: system wide. This can be achieved with one of::
echo always >/sys/kernel/mm/transparent_hugepage/enabled echo always >/sys/kernel/mm/transparent_hugepage/enabled
echo madvise >/sys/kernel/mm/transparent_hugepage/enabled echo madvise >/sys/kernel/mm/transparent_hugepage/enabled
echo never >/sys/kernel/mm/transparent_hugepage/enabled echo never >/sys/kernel/mm/transparent_hugepage/enabled
It's also possible to limit defrag efforts in the VM to generate It's also possible to limit defrag efforts in the VM to generate
anonymous hugepages in case they're not immediately free to madvise anonymous hugepages in case they're not immediately free to madvise
...@@ -108,44 +115,53 @@ use hugepages later instead of regular pages. This isn't always ...@@ -108,44 +115,53 @@ use hugepages later instead of regular pages. This isn't always
guaranteed, but it may be more likely in case the allocation is for a guaranteed, but it may be more likely in case the allocation is for a
MADV_HUGEPAGE region. MADV_HUGEPAGE region.
echo always >/sys/kernel/mm/transparent_hugepage/defrag ::
echo defer >/sys/kernel/mm/transparent_hugepage/defrag
echo defer+madvise >/sys/kernel/mm/transparent_hugepage/defrag echo always >/sys/kernel/mm/transparent_hugepage/defrag
echo madvise >/sys/kernel/mm/transparent_hugepage/defrag echo defer >/sys/kernel/mm/transparent_hugepage/defrag
echo never >/sys/kernel/mm/transparent_hugepage/defrag echo defer+madvise >/sys/kernel/mm/transparent_hugepage/defrag
echo madvise >/sys/kernel/mm/transparent_hugepage/defrag
"always" means that an application requesting THP will stall on allocation echo never >/sys/kernel/mm/transparent_hugepage/defrag
failure and directly reclaim pages and compact memory in an effort to
allocate a THP immediately. This may be desirable for virtual machines always
that benefit heavily from THP use and are willing to delay the VM start means that an application requesting THP will stall on
to utilise them. allocation failure and directly reclaim pages and compact
memory in an effort to allocate a THP immediately. This may be
"defer" means that an application will wake kswapd in the background desirable for virtual machines that benefit heavily from THP
to reclaim pages and wake kcompactd to compact memory so that THP is use and are willing to delay the VM start to utilise them.
available in the near future. It's the responsibility of khugepaged
to then install the THP pages later. defer
means that an application will wake kswapd in the background
"defer+madvise" will enter direct reclaim and compaction like "always", but to reclaim pages and wake kcompactd to compact memory so that
only for regions that have used madvise(MADV_HUGEPAGE); all other regions THP is available in the near future. It's the responsibility
will wake kswapd in the background to reclaim pages and wake kcompactd to of khugepaged to then install the THP pages later.
compact memory so that THP is available in the near future.
defer+madvise
"madvise" will enter direct reclaim like "always" but only for regions will enter direct reclaim and compaction like ``always``, but
that are have used madvise(MADV_HUGEPAGE). This is the default behaviour. only for regions that have used madvise(MADV_HUGEPAGE); all
other regions will wake kswapd in the background to reclaim
"never" should be self-explanatory. pages and wake kcompactd to compact memory so that THP is
available in the near future.
madvise
will enter direct reclaim like ``always`` but only for regions
that are have used madvise(MADV_HUGEPAGE). This is the default
behaviour.
never
should be self-explanatory.
By default kernel tries to use huge zero page on read page fault to By default kernel tries to use huge zero page on read page fault to
anonymous mapping. It's possible to disable huge zero page by writing 0 anonymous mapping. It's possible to disable huge zero page by writing 0
or enable it back by writing 1: or enable it back by writing 1::
echo 0 >/sys/kernel/mm/transparent_hugepage/use_zero_page echo 0 >/sys/kernel/mm/transparent_hugepage/use_zero_page
echo 1 >/sys/kernel/mm/transparent_hugepage/use_zero_page echo 1 >/sys/kernel/mm/transparent_hugepage/use_zero_page
Some userspace (such as a test program, or an optimized memory allocation Some userspace (such as a test program, or an optimized memory allocation
library) may want to know the size (in bytes) of a transparent hugepage: library) may want to know the size (in bytes) of a transparent hugepage::
cat /sys/kernel/mm/transparent_hugepage/hpage_pmd_size cat /sys/kernel/mm/transparent_hugepage/hpage_pmd_size
khugepaged will be automatically started when khugepaged will be automatically started when
transparent_hugepage/enabled is set to "always" or "madvise, and it'll transparent_hugepage/enabled is set to "always" or "madvise, and it'll
...@@ -155,84 +171,86 @@ khugepaged runs usually at low frequency so while one may not want to ...@@ -155,84 +171,86 @@ khugepaged runs usually at low frequency so while one may not want to
invoke defrag algorithms synchronously during the page faults, it invoke defrag algorithms synchronously during the page faults, it
should be worth invoking defrag at least in khugepaged. However it's should be worth invoking defrag at least in khugepaged. However it's
also possible to disable defrag in khugepaged by writing 0 or enable also possible to disable defrag in khugepaged by writing 0 or enable
defrag in khugepaged by writing 1: defrag in khugepaged by writing 1::
echo 0 >/sys/kernel/mm/transparent_hugepage/khugepaged/defrag echo 0 >/sys/kernel/mm/transparent_hugepage/khugepaged/defrag
echo 1 >/sys/kernel/mm/transparent_hugepage/khugepaged/defrag echo 1 >/sys/kernel/mm/transparent_hugepage/khugepaged/defrag
You can also control how many pages khugepaged should scan at each You can also control how many pages khugepaged should scan at each
pass: pass::
/sys/kernel/mm/transparent_hugepage/khugepaged/pages_to_scan /sys/kernel/mm/transparent_hugepage/khugepaged/pages_to_scan
and how many milliseconds to wait in khugepaged between each pass (you and how many milliseconds to wait in khugepaged between each pass (you
can set this to 0 to run khugepaged at 100% utilization of one core): can set this to 0 to run khugepaged at 100% utilization of one core)::
/sys/kernel/mm/transparent_hugepage/khugepaged/scan_sleep_millisecs /sys/kernel/mm/transparent_hugepage/khugepaged/scan_sleep_millisecs
and how many milliseconds to wait in khugepaged if there's an hugepage and how many milliseconds to wait in khugepaged if there's an hugepage
allocation failure to throttle the next allocation attempt. allocation failure to throttle the next allocation attempt::
/sys/kernel/mm/transparent_hugepage/khugepaged/alloc_sleep_millisecs /sys/kernel/mm/transparent_hugepage/khugepaged/alloc_sleep_millisecs
The khugepaged progress can be seen in the number of pages collapsed: The khugepaged progress can be seen in the number of pages collapsed::
/sys/kernel/mm/transparent_hugepage/khugepaged/pages_collapsed /sys/kernel/mm/transparent_hugepage/khugepaged/pages_collapsed
for each pass: for each pass::
/sys/kernel/mm/transparent_hugepage/khugepaged/full_scans /sys/kernel/mm/transparent_hugepage/khugepaged/full_scans
max_ptes_none specifies how many extra small pages (that are ``max_ptes_none`` specifies how many extra small pages (that are
not already mapped) can be allocated when collapsing a group not already mapped) can be allocated when collapsing a group
of small pages into one large page. of small pages into one large page::
/sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_none /sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_none
A higher value leads to use additional memory for programs. A higher value leads to use additional memory for programs.
A lower value leads to gain less thp performance. Value of A lower value leads to gain less thp performance. Value of
max_ptes_none can waste cpu time very little, you can max_ptes_none can waste cpu time very little, you can
ignore it. ignore it.
max_ptes_swap specifies how many pages can be brought in from ``max_ptes_swap`` specifies how many pages can be brought in from
swap when collapsing a group of pages into a transparent huge page. swap when collapsing a group of pages into a transparent huge page::
/sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_swap /sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_swap
A higher value can cause excessive swap IO and waste A higher value can cause excessive swap IO and waste
memory. A lower value can prevent THPs from being memory. A lower value can prevent THPs from being
collapsed, resulting fewer pages being collapsed into collapsed, resulting fewer pages being collapsed into
THPs, and lower memory access performance. THPs, and lower memory access performance.
== Boot parameter == Boot parameter
==============
You can change the sysfs boot time defaults of Transparent Hugepage You can change the sysfs boot time defaults of Transparent Hugepage
Support by passing the parameter "transparent_hugepage=always" or Support by passing the parameter ``transparent_hugepage=always`` or
"transparent_hugepage=madvise" or "transparent_hugepage=never" ``transparent_hugepage=madvise`` or ``transparent_hugepage=never``
(without "") to the kernel command line. to the kernel command line.
== Hugepages in tmpfs/shmem == Hugepages in tmpfs/shmem
========================
You can control hugepage allocation policy in tmpfs with mount option You can control hugepage allocation policy in tmpfs with mount option
"huge=". It can have following values: ``huge=``. It can have following values:
- "always": always
Attempt to allocate huge pages every time we need a new page; Attempt to allocate huge pages every time we need a new page;
- "never": never
Do not allocate huge pages; Do not allocate huge pages;
- "within_size": within_size
Only allocate huge page if it will be fully within i_size. Only allocate huge page if it will be fully within i_size.
Also respect fadvise()/madvise() hints; Also respect fadvise()/madvise() hints;
- "advise: advise
Only allocate huge pages if requested with fadvise()/madvise(); Only allocate huge pages if requested with fadvise()/madvise();
The default policy is "never". The default policy is ``never``.
"mount -o remount,huge= /mountpoint" works fine after mount: remounting ``mount -o remount,huge= /mountpoint`` works fine after mount: remounting
huge=never will not attempt to break up huge pages at all, just stop more ``huge=never`` will not attempt to break up huge pages at all, just stop more
from being allocated. from being allocated.
There's also sysfs knob to control hugepage allocation policy for internal There's also sysfs knob to control hugepage allocation policy for internal
...@@ -243,110 +261,130 @@ MAP_ANONYMOUS), GPU drivers' DRM objects, Ashmem. ...@@ -243,110 +261,130 @@ MAP_ANONYMOUS), GPU drivers' DRM objects, Ashmem.
In addition to policies listed above, shmem_enabled allows two further In addition to policies listed above, shmem_enabled allows two further
values: values:
- "deny": deny
For use in emergencies, to force the huge option off from For use in emergencies, to force the huge option off from
all mounts; all mounts;
- "force": force
Force the huge option on for all - very useful for testing; Force the huge option on for all - very useful for testing;
== Need of application restart == Need of application restart
===========================
The transparent_hugepage/enabled values and tmpfs mount option only affect The transparent_hugepage/enabled values and tmpfs mount option only affect
future behavior. So to make them effective you need to restart any future behavior. So to make them effective you need to restart any
application that could have been using hugepages. This also applies to the application that could have been using hugepages. This also applies to the
regions registered in khugepaged. regions registered in khugepaged.
== Monitoring usage == Monitoring usage
================
The number of anonymous transparent huge pages currently used by the The number of anonymous transparent huge pages currently used by the
system is available by reading the AnonHugePages field in /proc/meminfo. system is available by reading the AnonHugePages field in ``/proc/meminfo``.
To identify what applications are using anonymous transparent huge pages, To identify what applications are using anonymous transparent huge pages,
it is necessary to read /proc/PID/smaps and count the AnonHugePages fields it is necessary to read ``/proc/PID/smaps`` and count the AnonHugePages fields
for each mapping. for each mapping.
The number of file transparent huge pages mapped to userspace is available The number of file transparent huge pages mapped to userspace is available
by reading ShmemPmdMapped and ShmemHugePages fields in /proc/meminfo. by reading ShmemPmdMapped and ShmemHugePages fields in ``/proc/meminfo``.
To identify what applications are mapping file transparent huge pages, it To identify what applications are mapping file transparent huge pages, it
is necessary to read /proc/PID/smaps and count the FileHugeMapped fields is necessary to read ``/proc/PID/smaps`` and count the FileHugeMapped fields
for each mapping. for each mapping.
Note that reading the smaps file is expensive and reading it Note that reading the smaps file is expensive and reading it
frequently will incur overhead. frequently will incur overhead.
There are a number of counters in /proc/vmstat that may be used to There are a number of counters in ``/proc/vmstat`` that may be used to
monitor how successfully the system is providing huge pages for use. monitor how successfully the system is providing huge pages for use.
thp_fault_alloc is incremented every time a huge page is successfully thp_fault_alloc
is incremented every time a huge page is successfully
allocated to handle a page fault. This applies to both the allocated to handle a page fault. This applies to both the
first time a page is faulted and for COW faults. first time a page is faulted and for COW faults.
thp_collapse_alloc is incremented by khugepaged when it has found thp_collapse_alloc
is incremented by khugepaged when it has found
a range of pages to collapse into one huge page and has a range of pages to collapse into one huge page and has
successfully allocated a new huge page to store the data. successfully allocated a new huge page to store the data.
thp_fault_fallback is incremented if a page fault fails to allocate thp_fault_fallback
is incremented if a page fault fails to allocate
a huge page and instead falls back to using small pages. a huge page and instead falls back to using small pages.
thp_collapse_alloc_failed is incremented if khugepaged found a range thp_collapse_alloc_failed
is incremented if khugepaged found a range
of pages that should be collapsed into one huge page but failed of pages that should be collapsed into one huge page but failed
the allocation. the allocation.
thp_file_alloc is incremented every time a file huge page is successfully thp_file_alloc
is incremented every time a file huge page is successfully
allocated. allocated.
thp_file_mapped is incremented every time a file huge page is mapped into thp_file_mapped
is incremented every time a file huge page is mapped into
user address space. user address space.
thp_split_page is incremented every time a huge page is split into base thp_split_page
is incremented every time a huge page is split into base
pages. This can happen for a variety of reasons but a common pages. This can happen for a variety of reasons but a common
reason is that a huge page is old and is being reclaimed. reason is that a huge page is old and is being reclaimed.
This action implies splitting all PMD the page mapped with. This action implies splitting all PMD the page mapped with.
thp_split_page_failed is incremented if kernel fails to split huge thp_split_page_failed
is incremented if kernel fails to split huge
page. This can happen if the page was pinned by somebody. page. This can happen if the page was pinned by somebody.
thp_deferred_split_page is incremented when a huge page is put onto split thp_deferred_split_page
is incremented when a huge page is put onto split
queue. This happens when a huge page is partially unmapped and queue. This happens when a huge page is partially unmapped and
splitting it would free up some memory. Pages on split queue are splitting it would free up some memory. Pages on split queue are
going to be split under memory pressure. going to be split under memory pressure.
thp_split_pmd is incremented every time a PMD split into table of PTEs. thp_split_pmd
is incremented every time a PMD split into table of PTEs.
This can happen, for instance, when application calls mprotect() or This can happen, for instance, when application calls mprotect() or
munmap() on part of huge page. It doesn't split huge page, only munmap() on part of huge page. It doesn't split huge page, only
page table entry. page table entry.
thp_zero_page_alloc is incremented every time a huge zero page is thp_zero_page_alloc
is incremented every time a huge zero page is
successfully allocated. It includes allocations which where successfully allocated. It includes allocations which where
dropped due race with other allocation. Note, it doesn't count dropped due race with other allocation. Note, it doesn't count
every map of the huge zero page, only its allocation. every map of the huge zero page, only its allocation.
thp_zero_page_alloc_failed is incremented if kernel fails to allocate thp_zero_page_alloc_failed
is incremented if kernel fails to allocate
huge zero page and falls back to using small pages. huge zero page and falls back to using small pages.
As the system ages, allocating huge pages may be expensive as the As the system ages, allocating huge pages may be expensive as the
system uses memory compaction to copy data around memory to free a system uses memory compaction to copy data around memory to free a
huge page for use. There are some counters in /proc/vmstat to help huge page for use. There are some counters in ``/proc/vmstat`` to help
monitor this overhead. monitor this overhead.
compact_stall is incremented every time a process stalls to run compact_stall
is incremented every time a process stalls to run
memory compaction so that a huge page is free for use. memory compaction so that a huge page is free for use.
compact_success is incremented if the system compacted memory and compact_success
is incremented if the system compacted memory and
freed a huge page for use. freed a huge page for use.
compact_fail is incremented if the system tries to compact memory compact_fail
is incremented if the system tries to compact memory
but failed. but failed.
compact_pages_moved is incremented each time a page is moved. If compact_pages_moved
is incremented each time a page is moved. If
this value is increasing rapidly, it implies that the system this value is increasing rapidly, it implies that the system
is copying a lot of data to satisfy the huge page allocation. is copying a lot of data to satisfy the huge page allocation.
It is possible that the cost of copying exceeds any savings It is possible that the cost of copying exceeds any savings
from reduced TLB misses. from reduced TLB misses.
compact_pagemigrate_failed is incremented when the underlying mechanism compact_pagemigrate_failed
is incremented when the underlying mechanism
for moving a page failed. for moving a page failed.
compact_blocks_moved is incremented each time memory compaction examines compact_blocks_moved
is incremented each time memory compaction examines
a huge page aligned range of pages. a huge page aligned range of pages.
It is possible to establish how long the stalls were using the function It is possible to establish how long the stalls were using the function
...@@ -354,7 +392,8 @@ tracer to record how long was spent in __alloc_pages_nodemask and ...@@ -354,7 +392,8 @@ tracer to record how long was spent in __alloc_pages_nodemask and
using the mm_page_alloc tracepoint to identify which allocations were using the mm_page_alloc tracepoint to identify which allocations were
for huge pages. for huge pages.
== get_user_pages and follow_page == get_user_pages and follow_page
==============================
get_user_pages and follow_page if run on a hugepage, will return the get_user_pages and follow_page if run on a hugepage, will return the
head or tail pages as usual (exactly as they would do on head or tail pages as usual (exactly as they would do on
...@@ -367,10 +406,11 @@ for the head page and not the tail page), it should be updated to jump ...@@ -367,10 +406,11 @@ for the head page and not the tail page), it should be updated to jump
to check head page instead. Taking reference on any head/tail page would to check head page instead. Taking reference on any head/tail page would
prevent page from being split by anyone. prevent page from being split by anyone.
NOTE: these aren't new constraints to the GUP API, and they match the .. note::
same constrains that applies to hugetlbfs too, so any driver capable these aren't new constraints to the GUP API, and they match the
of handling GUP on hugetlbfs will also work fine on transparent same constrains that applies to hugetlbfs too, so any driver capable
hugepage backed mappings. of handling GUP on hugetlbfs will also work fine on transparent
hugepage backed mappings.
In case you can't handle compound pages if they're returned by In case you can't handle compound pages if they're returned by
follow_page, the FOLL_SPLIT bit can be specified as parameter to follow_page, the FOLL_SPLIT bit can be specified as parameter to
...@@ -383,13 +423,15 @@ hugepages being returned (as it's not only checking the pfn of the ...@@ -383,13 +423,15 @@ hugepages being returned (as it's not only checking the pfn of the
page and pinning it during the copy but it pretends to migrate the page and pinning it during the copy but it pretends to migrate the
memory in regular page sizes and with regular pte/pmd mappings). memory in regular page sizes and with regular pte/pmd mappings).
== Optimizing the applications == Optimizing the applications
===========================
To be guaranteed that the kernel will map a 2M page immediately in any To be guaranteed that the kernel will map a 2M page immediately in any
memory region, the mmap region has to be hugepage naturally memory region, the mmap region has to be hugepage naturally
aligned. posix_memalign() can provide that guarantee. aligned. posix_memalign() can provide that guarantee.
== Hugetlbfs == Hugetlbfs
=========
You can use hugetlbfs on a kernel that has transparent hugepage You can use hugetlbfs on a kernel that has transparent hugepage
support enabled just fine as always. No difference can be noted in support enabled just fine as always. No difference can be noted in
...@@ -397,7 +439,8 @@ hugetlbfs other than there will be less overall fragmentation. All ...@@ -397,7 +439,8 @@ hugetlbfs other than there will be less overall fragmentation. All
usual features belonging to hugetlbfs are preserved and usual features belonging to hugetlbfs are preserved and
unaffected. libhugetlbfs will also work fine as usual. unaffected. libhugetlbfs will also work fine as usual.
== Graceful fallback == Graceful fallback
=================
Code walking pagetables but unaware about huge pmds can simply call Code walking pagetables but unaware about huge pmds can simply call
split_huge_pmd(vma, pmd, addr) where the pmd is the one returned by split_huge_pmd(vma, pmd, addr) where the pmd is the one returned by
...@@ -415,20 +458,21 @@ it tries to swapout the hugepage for example. split_huge_page() can fail ...@@ -415,20 +458,21 @@ it tries to swapout the hugepage for example. split_huge_page() can fail
if the page is pinned and you must handle this correctly. if the page is pinned and you must handle this correctly.
Example to make mremap.c transparent hugepage aware with a one liner Example to make mremap.c transparent hugepage aware with a one liner
change: change::
diff --git a/mm/mremap.c b/mm/mremap.c diff --git a/mm/mremap.c b/mm/mremap.c
--- a/mm/mremap.c --- a/mm/mremap.c
+++ b/mm/mremap.c +++ b/mm/mremap.c
@@ -41,6 +41,7 @@ static pmd_t *get_old_pmd(struct mm_stru @@ -41,6 +41,7 @@ static pmd_t *get_old_pmd(struct mm_stru
return NULL; return NULL;
pmd = pmd_offset(pud, addr); pmd = pmd_offset(pud, addr);
+ split_huge_pmd(vma, pmd, addr); + split_huge_pmd(vma, pmd, addr);
if (pmd_none_or_clear_bad(pmd)) if (pmd_none_or_clear_bad(pmd))
return NULL; return NULL;
== Locking in hugepage aware code == Locking in hugepage aware code
==============================
We want as much code as possible hugepage aware, as calling We want as much code as possible hugepage aware, as calling
split_huge_page() or split_huge_pmd() has a cost. split_huge_page() or split_huge_pmd() has a cost.
...@@ -448,7 +492,8 @@ should just drop the page table lock and fallback to the old code as ...@@ -448,7 +492,8 @@ should just drop the page table lock and fallback to the old code as
before. Otherwise you can proceed to process the huge pmd and the before. Otherwise you can proceed to process the huge pmd and the
hugepage natively. Once finished you can drop the page table lock. hugepage natively. Once finished you can drop the page table lock.
== Refcounts and transparent huge pages == Refcounts and transparent huge pages
====================================
Refcounting on THP is mostly consistent with refcounting on other compound Refcounting on THP is mostly consistent with refcounting on other compound
pages: pages:
...@@ -510,7 +555,8 @@ clear where reference should go after split: it will stay on head page. ...@@ -510,7 +555,8 @@ clear where reference should go after split: it will stay on head page.
Note that split_huge_pmd() doesn't have any limitation on refcounting: Note that split_huge_pmd() doesn't have any limitation on refcounting:
pmd can be split at any point and never fails. pmd can be split at any point and never fails.
== Partial unmap and deferred_split_huge_page() == Partial unmap and deferred_split_huge_page()
============================================
Unmapping part of THP (with munmap() or other way) is not going to free Unmapping part of THP (with munmap() or other way) is not going to free
memory immediately. Instead, we detect that a subpage of THP is not in use memory immediately. Instead, we detect that a subpage of THP is not in use
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7