Skip to content

L
linux
Commit 76ea470c authored by Ralph Campbell's avatar Ralph Campbell Committed by Linus Torvalds
Browse files
Options

mm/hmm: documentation editorial update to HMM documentation 

Update the documentation for HMM to fix minor typos and phrasing to be a
bit more readable.

Link: http://lkml.kernel.org/r/20180323005527.758-2-jglisse@redhat.comSigned-off-by: default avatarRalph Campbell <rcampbell@nvidia.com>
Signed-off-by: default avatarJérôme Glisse <jglisse@redhat.com>
Cc: Stephen  Bates <sbates@raithlin.com>
Cc: Jason Gunthorpe <jgg@mellanox.com>
Cc: Logan Gunthorpe <logang@deltatee.com>
Cc: Evgeny Baskakov <ebaskakov@nvidia.com>
Cc: Mark Hairgrove <mhairgrove@nvidia.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent d51d1e64
Hide whitespace changes
Inline Side-by-side
Showing with 187 additions and 174 deletions
Documentation/vm/hmm.txt
View file @ 76ea470c
Heterogeneous Memory Management (HMM) Heterogeneous Memory Management (HMM)
Transparently allow any component of a program to use any memory region of said Provide infrastructure and helpers to integrate non conventional memory (device
program with a device without using device specific memory allocator. This is memory like GPU on board memory) into regular kernel code path. Corner stone of
becoming a requirement to simplify the use of advance heterogeneous computing this being specialize struct page for such memory (see sections 5 to 7 of this
where GPU, DSP or FPGA are use to perform various computations. document).
This document is divided as follow, in the first section i expose the problems HMM also provide optional helpers for SVM (Share Virtual Memory) ie allowing a
related to the use of a device specific allocator. The second section i expose device to transparently access program address coherently with the CPU meaning
the hardware limitations that are inherent to many platforms. The third section that any valid pointer on the CPU is also a valid pointer for the device. This
gives an overview of HMM designs. The fourth section explains how CPU page- is becoming a mandatory to simplify the use of advance heterogeneous computing
table mirroring works and what is HMM purpose in this context. Fifth section where GPU, DSP, or FPGA are used to perform various computations on behalf of
deals with how device memory is represented inside the kernel. Finaly the last a process.
section present the new migration helper that allow to leverage the device DMA
engine. This document is divided as follows: in the first section I expose the problems
related to using device specific memory allocators. In the second section, I
expose the hardware limitations that are inherent to many platforms. The third
1) Problems of using device specific memory allocator: section gives an overview of the HMM design. The fourth section explains how
2) System bus, device memory characteristics CPU page-table mirroring works and what is HMM's purpose in this context. The
3) Share address space and migration fifth section deals with how device memory is represented inside the kernel.
Finally, the last section presents a new migration helper that allows lever-
aging the device DMA engine.
1) Problems of using a device specific memory allocator:
2) I/O bus, device memory characteristics
3) Shared address space and migration
4) Address space mirroring implementation and API 4) Address space mirroring implementation and API
5) Represent and manage device memory from core kernel point of view 5) Represent and manage device memory from core kernel point of view
6) Migrate to and from device memory 6) Migration to and from device memory
7) Memory cgroup (memcg) and rss accounting 7) Memory cgroup (memcg) and rss accounting
------------------------------------------------------------------------------- -------------------------------------------------------------------------------
1) Problems of using device specific memory allocator: 1) Problems of using a device specific memory allocator:
Device with large amount of on board memory (several giga bytes) like GPU have Devices with a large amount of on board memory (several giga bytes) like GPUs
historically manage their memory through dedicated driver specific API. This have historically managed their memory through dedicated driver specific APIs.
creates a disconnect between memory allocated and managed by device driver and This creates a disconnect between memory allocated and managed by a device
regular application memory (private anonymous, share memory or regular file driver and regular application memory (private anonymous, shared memory, or
back memory). From here on i will refer to this aspect as split address space. regular file backed memory). From here on I will refer to this aspect as split
I use share address space to refer to the opposite situation ie one in which address space. I use shared address space to refer to the opposite situation:
any memory region can be use by device transparently. i.e., one in which any application memory region can be used by a device
transparently.
Split address space because device can only access memory allocated through the Split address space because device can only access memory allocated through the
device specific API. This imply that all memory object in a program are not device specific API. This implies that all memory objects in a program are not
equal from device point of view which complicate large program that rely on a equal from the device point of view which complicates large programs that rely
wide set of libraries. on a wide set of libraries.
Concretly this means that code that wants to leverage device like GPU need to Concretly this means that code that wants to leverage devices like GPUs need to
copy object between genericly allocated memory (malloc, mmap private/share/) copy object between genericly allocated memory (malloc, mmap private/share/)
and memory allocated through the device driver API (this still end up with an and memory allocated through the device driver API (this still end up with an
mmap but of the device file). mmap but of the device file).
For flat dataset (array, grid, image, ...) this isn't too hard to achieve but For flat data-sets (array, grid, image, ...) this isn't too hard to achieve but
complex data-set (list, tree, ...) are hard to get right. Duplicating a complex complex data-sets (list, tree, ...) are hard to get right. Duplicating a
data-set need to re-map all the pointer relations between each of its elements. complex data-set needs to re-map all the pointer relations between each of its
This is error prone and program gets harder to debug because of the duplicate elements. This is error prone and program gets harder to debug because of the
data-set. duplicate data-set and addresses.
Split address space also means that library can not transparently use data they Split address space also means that libraries can not transparently use data
are getting from core program or other library and thus each library might have they are getting from the core program or another library and thus each library
to duplicate its input data-set using specific memory allocator. Large project might have to duplicate its input data-set using the device specific memory
suffer from this and waste resources because of the various memory copy. allocator. Large projects suffer from this and waste resources because of the
various memory copies.
Duplicating each library API to accept as input or output memory allocted by Duplicating each library API to accept as input or output memory allocted by
each device specific allocator is not a viable option. It would lead to a each device specific allocator is not a viable option. It would lead to a
combinatorial explosions in the library entry points. combinatorial explosion in the library entry points.
Finaly with the advance of high level language constructs (in C++ but in other Finally, with the advance of high level language constructs (in C++ but in
language too) it is now possible for compiler to leverage GPU or other devices other languages too) it is now possible for the compiler to leverage GPUs and
without even the programmer knowledge. Some of compiler identified patterns are other devices without programmer knowledge. Some compiler identified patterns
only do-able with a share address. It is as well more reasonable to use a share are only do-able with a shared address space. It is also more reasonable to use
address space for all the other patterns. a shared address space for all other patterns.
------------------------------------------------------------------------------- -------------------------------------------------------------------------------
2) System bus, device memory characteristics 2) I/O bus, device memory characteristics
System bus cripple share address due to few limitations. Most system bus only I/O buses cripple shared address due to few limitations. Most I/O buses only
allow basic memory access from device to main memory, even cache coherency is allow basic memory access from device to main memory, even cache coherency is
often optional. Access to device memory from CPU is even more limited, most often optional. Access to device memory from CPU is even more limited. More
often than not it is not cache coherent. often than not, it is not cache coherent.
If we only consider the PCIE bus than device can access main memory (often If we only consider the PCIE bus, then a device can access main memory (often
through an IOMMU) and be cache coherent with the CPUs. However it only allows through an IOMMU) and be cache coherent with the CPUs. However, it only allows
a limited set of atomic operation from device on main memory. This is worse a limited set of atomic operations from device on main memory. This is worse
in the other direction the CPUs can only access a limited range of the device in the other direction, the CPU can only access a limited range of the device
memory and can not perform atomic operations on it. Thus device memory can not memory and can not perform atomic operations on it. Thus device memory can not
be consider like regular memory from kernel point of view. be considered the same as regular memory from the kernel point of view.
Another crippling factor is the limited bandwidth (~32GBytes/s with PCIE 4.0 Another crippling factor is the limited bandwidth (~32GBytes/s with PCIE 4.0
and 16 lanes). This is 33 times less that fastest GPU memory (1 TBytes/s). and 16 lanes). This is 33 times less than the fastest GPU memory (1 TBytes/s).
The final limitation is latency, access to main memory from the device has an The final limitation is latency. Access to main memory from the device has an
order of magnitude higher latency than when the device access its own memory. order of magnitude higher latency than when the device accesses its own memory.
Some platform are developing new system bus or additions/modifications to PCIE Some platforms are developing new I/O buses or additions/modifications to PCIE
to address some of those limitations (OpenCAPI, CCIX). They mainly allow two to address some of these limitations (OpenCAPI, CCIX). They mainly allow two
way cache coherency between CPU and device and allow all atomic operations the way cache coherency between CPU and device and allow all atomic operations the
architecture supports. Saddly not all platform are following this trends and architecture supports. Saddly, not all platforms are following this trend and
some major architecture are left without hardware solutions to those problems. some major architectures are left without hardware solutions to these problems.
So for share address space to make sense not only we must allow device to So for shared address space to make sense, not only must we allow device to
access any memory memory but we must also permit any memory to be migrated to access any memory memory but we must also permit any memory to be migrated to
device memory while device is using it (blocking CPU access while it happens). device memory while device is using it (blocking CPU access while it happens).
------------------------------------------------------------------------------- -------------------------------------------------------------------------------
3) Share address space and migration 3) Shared address space and migration
HMM intends to provide two main features. First one is to share the address HMM intends to provide two main features. First one is to share the address
space by duplication the CPU page table into the device page table so same space by duplicating the CPU page table in the device page table so the same
address point to same memory and this for any valid main memory address in address points to the same physical memory for any valid main memory address in
the process address space. the process address space.
To achieve this, HMM offer a set of helpers to populate the device page table To achieve this, HMM offers a set of helpers to populate the device page table
while keeping track of CPU page table updates. Device page table updates are while keeping track of CPU page table updates. Device page table updates are
not as easy as CPU page table updates. To update the device page table you must not as easy as CPU page table updates. To update the device page table, you must
allow a buffer (or use a pool of pre-allocated buffer) and write GPU specifics allocate a buffer (or use a pool of pre-allocated buffers) and write GPU
commands in it to perform the update (unmap, cache invalidations and flush, specific commands in it to perform the update (unmap, cache invalidations, and
...). This can not be done through common code for all device. Hence why HMM flush, ...). This can not be done through common code for all devices. Hence
provides helpers to factor out everything that can be while leaving the gory why HMM provides helpers to factor out everything that can be while leaving the
details to the device driver. hardware specific details to the device driver.
The second mechanism HMM provide is a new kind of ZONE_DEVICE memory that does The second mechanism HMM provides, is a new kind of ZONE_DEVICE memory that
allow to allocate a struct page for each page of the device memory. Those page allows allocating a struct page for each page of the device memory. Those pages
are special because the CPU can not map them. They however allow to migrate are special because the CPU can not map them. However, they allow migrating
main memory to device memory using exhisting migration mechanism and everything main memory to device memory using existing migration mechanisms and everything
looks like if page was swap out to disk from CPU point of view. Using a struct looks like a page is swapped out to disk from the CPU point of view. Using a
page gives the easiest and cleanest integration with existing mm mechanisms. struct page gives the easiest and cleanest integration with existing mm mech-
Again here HMM only provide helpers, first to hotplug new ZONE_DEVICE memory anisms. Here again, HMM only provides helpers, first to hotplug new ZONE_DEVICE
for the device memory and second to perform migration. Policy decision of what memory for the device memory and second to perform migration. Policy decisions
and when to migrate things is left to the device driver. of what and when to migrate things is left to the device driver.
Note that any CPU access to a device page trigger a page fault and a migration Note that any CPU access to a device page triggers a page fault and a migration
back to main memory ie when a page backing an given address A is migrated from back to main memory. For example, when a page backing a given CPU address A is
a main memory page to a device page then any CPU access to address A trigger a migrated from a main memory page to a device page, then any CPU access to
page fault and initiate a migration back to main memory. address A triggers a page fault and initiates a migration back to main memory.
With these two features, HMM not only allows a device to mirror process address
With this two features, HMM not only allow a device to mirror a process address space and keeping both CPU and device page table synchronized, but also lever-
space and keeps both CPU and device page table synchronize, but also allow to ages device memory by migrating the part of the data-set that is actively being
leverage device memory by migrating part of data-set that is actively use by a used by the device.
device.
------------------------------------------------------------------------------- -------------------------------------------------------------------------------
4) Address space mirroring implementation and API 4) Address space mirroring implementation and API
Address space mirroring main objective is to allow to duplicate range of CPU Address space mirroring's main objective is to allow duplication of a range of
page table into a device page table and HMM helps keeping both synchronize. A CPU page table into a device page table; HMM helps keep both synchronized. A
device driver that want to mirror a process address space must start with the device driver that want to mirror a process address space must start with the
registration of an hmm_mirror struct: registration of an hmm_mirror struct:
...@@ -155,8 +163,8 @@ registration of an hmm_mirror struct: ...@@ -155,8 +163,8 @@ registration of an hmm_mirror struct:
struct mm_struct *mm); struct mm_struct *mm);
The locked variant is to be use when the driver is already holding the mmap_sem The locked variant is to be use when the driver is already holding the mmap_sem
of the mm in write mode. The mirror struct has a set of callback that are use of the mm in write mode. The mirror struct has a set of callbacks that are used
to propagate CPU page table: to propagate CPU page tables:
struct hmm_mirror_ops { struct hmm_mirror_ops {
/* sync_cpu_device_pagetables() - synchronize page tables /* sync_cpu_device_pagetables() - synchronize page tables
...@@ -181,13 +189,13 @@ to propagate CPU page table: ...@@ -181,13 +189,13 @@ to propagate CPU page table:
unsigned long end); unsigned long end);
}; };
Device driver must perform update to the range following action (turn range The device driver must perform the update action to the range (mark range
read only, or fully unmap, ...). Once driver callback returns the device must read only, or fully unmap, ...). The device must be done with the update before
be done with the update. the driver callback returns.
When device driver wants to populate a range of virtual address it can use When the device driver wants to populate a range of virtual addresses, it can
either: use either:
int hmm_vma_get_pfns(struct vm_area_struct *vma, int hmm_vma_get_pfns(struct vm_area_struct *vma,
struct hmm_range *range, struct hmm_range *range,
unsigned long start, unsigned long start,
...@@ -201,17 +209,19 @@ either: ...@@ -201,17 +209,19 @@ either:
bool write, bool write,
bool block); bool block);
First one (hmm_vma_get_pfns()) will only fetch present CPU page table entry and The first one (hmm_vma_get_pfns()) will only fetch present CPU page table
will not trigger a page fault on missing or non present entry. The second one entries and will not trigger a page fault on missing or non present entries.
do trigger page fault on missing or read only entry if write parameter is true. The second one does trigger a page fault on missing or read only entry if the
Page fault use the generic mm page fault code path just like a CPU page fault. write parameter is true. Page faults use the generic mm page fault code path
just like a CPU page fault.
Both function copy CPU page table into their pfns array argument. Each entry in Both functions copy CPU page table entries into their pfns array argument. Each
that array correspond to an address in the virtual range. HMM provide a set of entry in that array corresponds to an address in the virtual range. HMM
flags to help driver identify special CPU page table entries. provides a set of flags to help the driver identify special CPU page table
entries.
Locking with the update() callback is the most important aspect the driver must Locking with the update() callback is the most important aspect the driver must
respect in order to keep things properly synchronize. The usage pattern is : respect in order to keep things properly synchronized. The usage pattern is:
int driver_populate_range(...) int driver_populate_range(...)
{ {
...@@ -233,43 +243,44 @@ respect in order to keep things properly synchronize. The usage pattern is : ...@@ -233,43 +243,44 @@ respect in order to keep things properly synchronize. The usage pattern is :
return 0; return 0;
} }
The driver->update lock is the same lock that driver takes inside its update() The driver->update lock is the same lock that the driver takes inside its
callback. That lock must be call before hmm_vma_range_done() to avoid any race update() callback. That lock must be held before hmm_vma_range_done() to avoid
with a concurrent CPU page table update. any race with a concurrent CPU page table update.
HMM implements all this on top of the mmu_notifier API because we wanted to a HMM implements all this on top of the mmu_notifier API because we wanted a
simpler API and also to be able to perform optimization latter own like doing simpler API and also to be able to perform optimizations latter on like doing
concurrent device update in multi-devices scenario. concurrent device updates in multi-devices scenario.
HMM also serve as an impedence missmatch between how CPU page table update are HMM also serves as an impedence mismatch between how CPU page table updates
done (by CPU write to the page table and TLB flushes) from how device update are done (by CPU write to the page table and TLB flushes) and how devices
their own page table. Device update is a multi-step process, first appropriate update their own page table. Device updates are a multi-step process. First,
commands are write to a buffer, then this buffer is schedule for execution on appropriate commands are writen to a buffer, then this buffer is scheduled for
the device. It is only once the device has executed commands in the buffer that execution on the device. It is only once the device has executed commands in
the update is done. Creating and scheduling update command buffer can happen the buffer that the update is done. Creating and scheduling the update command
concurrently for multiple devices. Waiting for each device to report commands buffer can happen concurrently for multiple devices. Waiting for each device to
as executed is serialize (there is no point in doing this concurrently). report commands as executed is serialized (there is no point in doing this
concurrently).
------------------------------------------------------------------------------- -------------------------------------------------------------------------------
5) Represent and manage device memory from core kernel point of view 5) Represent and manage device memory from core kernel point of view
Several differents design were try to support device memory. First one use Several different designs were tried to support device memory. First one used
device specific data structure to keep information about migrated memory and a device specific data structure to keep information about migrated memory and
HMM hooked itself in various place of mm code to handle any access to address HMM hooked itself in various places of mm code to handle any access to
that were back by device memory. It turns out that this ended up replicating addresses that were backed by device memory. It turns out that this ended up
most of the fields of struct page and also needed many kernel code path to be replicating most of the fields of struct page and also needed many kernel code
updated to understand this new kind of memory. paths to be updated to understand this new kind of memory.
Thing is most kernel code path never try to access the memory behind a page Most kernel code paths never try to access the memory behind a page
but only care about struct page contents. Because of this HMM switchted to but only care about struct page contents. Because of this, HMM switched to
directly using struct page for device memory which left most kernel code path directly using struct page for device memory which left most kernel code paths
un-aware of the difference. We only need to make sure that no one ever try to unaware of the difference. We only need to make sure that no one ever tries to
map those page from the CPU side. map those pages from the CPU side.
HMM provide a set of helpers to register and hotplug device memory as a new HMM provides a set of helpers to register and hotplug device memory as a new
region needing struct page. This is offer through a very simple API: region needing a struct page. This is offered through a very simple API:
struct hmm_devmem *hmm_devmem_add(const struct hmm_devmem_ops *ops, struct hmm_devmem *hmm_devmem_add(const struct hmm_devmem_ops *ops,
struct device *device, struct device *device,
...@@ -289,18 +300,19 @@ The hmm_devmem_ops is where most of the important things are: ...@@ -289,18 +300,19 @@ The hmm_devmem_ops is where most of the important things are:
}; };
The first callback (free()) happens when the last reference on a device page is The first callback (free()) happens when the last reference on a device page is
drop. This means the device page is now free and no longer use by anyone. The dropped. This means the device page is now free and no longer used by anyone.
second callback happens whenever CPU try to access a device page which it can The second callback happens whenever the CPU tries to access a device page
not do. This second callback must trigger a migration back to system memory. which it can not do. This second callback must trigger a migration back to
system memory.
------------------------------------------------------------------------------- -------------------------------------------------------------------------------
6) Migrate to and from device memory 6) Migration to and from device memory
Because CPU can not access device memory, migration must use device DMA engine Because the CPU can not access device memory, migration must use the device DMA
to perform copy from and to device memory. For this we need a new migration engine to perform copy from and to device memory. For this we need a new
helper: migration helper:
int migrate_vma(const struct migrate_vma_ops *ops, int migrate_vma(const struct migrate_vma_ops *ops,
struct vm_area_struct *vma, struct vm_area_struct *vma,
...@@ -311,15 +323,15 @@ helper: ...@@ -311,15 +323,15 @@ helper:
unsigned long *dst, unsigned long *dst,
void *private); void *private);
Unlike other migration function it works on a range of virtual address, there Unlike other migration functions it works on a range of virtual address, there
is two reasons for that. First device DMA copy has a high setup overhead cost are two reasons for that. First, device DMA copy has a high setup overhead cost
and thus batching multiple pages is needed as otherwise the migration overhead and thus batching multiple pages is needed as otherwise the migration overhead
make the whole excersie pointless. The second reason is because driver trigger makes the whole exersize pointless. The second reason is because the
such migration base on range of address the device is actively accessing. migration might be for a range of addresses the device is actively accessing.
The migrate_vma_ops struct define two callbacks. First one (alloc_and_copy()) The migrate_vma_ops struct defines two callbacks. First one (alloc_and_copy())
control destination memory allocation and copy operation. Second one is there controls destination memory allocation and copy operation. Second one is there
to allow device driver to perform cleanup operation after migration. to allow the device driver to perform cleanup operations after migration.
struct migrate_vma_ops { struct migrate_vma_ops {
void (*alloc_and_copy)(struct vm_area_struct *vma, void (*alloc_and_copy)(struct vm_area_struct *vma,
...@@ -336,19 +348,19 @@ to allow device driver to perform cleanup operation after migration. ...@@ -336,19 +348,19 @@ to allow device driver to perform cleanup operation after migration.
void *private); void *private);
}; };
It is important to stress that this migration helpers allow for hole in the It is important to stress that these migration helpers allow for holes in the
virtual address range. Some pages in the range might not be migrated for all virtual address range. Some pages in the range might not be migrated for all
the usual reasons (page is pin, page is lock, ...). This helper does not fail the usual reasons (page is pinned, page is locked, ...). This helper does not
but just skip over those pages. fail but just skips over those pages.
The alloc_and_copy() might as well decide to not migrate all pages in the The alloc_and_copy() might decide to not migrate all pages in the
range (for reasons under the callback control). For those the callback just range (for reasons under the callback control). For those, the callback just
have to leave the corresponding dst entry empty. has to leave the corresponding dst entry empty.
Finaly the migration of the struct page might fails (for file back page) for Finally, the migration of the struct page might fail (for file backed page) for
various reasons (failure to freeze reference, or update page cache, ...). If various reasons (failure to freeze reference, or update page cache, ...). If
that happens then the finalize_and_map() can catch any pages that was not that happens, then the finalize_and_map() can catch any pages that were not
migrated. Note those page were still copied to new page and thus we wasted migrated. Note those pages were still copied to a new page and thus we wasted
bandwidth but this is considered as a rare event and a price that we are bandwidth but this is considered as a rare event and a price that we are
willing to pay to keep all the code simpler. willing to pay to keep all the code simpler.
...@@ -358,27 +370,27 @@ willing to pay to keep all the code simpler. ...@@ -358,27 +370,27 @@ willing to pay to keep all the code simpler.
7) Memory cgroup (memcg) and rss accounting 7) Memory cgroup (memcg) and rss accounting
For now device memory is accounted as any regular page in rss counters (either For now device memory is accounted as any regular page in rss counters (either
anonymous if device page is use for anonymous, file if device page is use for anonymous if device page is used for anonymous, file if device page is used for
file back page or shmem if device page is use for share memory). This is a file backed page or shmem if device page is used for shared memory). This is a
deliberate choice to keep existing application that might start using device deliberate choice to keep existing applications, that might start using device
memory without knowing about it to keep runing unimpacted. memory without knowing about it, running unimpacted.
Drawbacks is that OOM killer might kill an application using a lot of device A Drawback is that the OOM killer might kill an application using a lot of
memory and not a lot of regular system memory and thus not freeing much system device memory and not a lot of regular system memory and thus not freeing much
memory. We want to gather more real world experience on how application and system memory. We want to gather more real world experience on how applications
system react under memory pressure in the presence of device memory before and system react under memory pressure in the presence of device memory before
deciding to account device memory differently. deciding to account device memory differently.
Same decision was made for memory cgroup. Device memory page are accounted Same decision was made for memory cgroup. Device memory pages are accounted
against same memory cgroup a regular page would be accounted to. This does against same memory cgroup a regular page would be accounted to. This does
simplify migration to and from device memory. This also means that migration simplify migration to and from device memory. This also means that migration
back from device memory to regular memory can not fail because it would back from device memory to regular memory can not fail because it would
go above memory cgroup limit. We might revisit this choice latter on once we go above memory cgroup limit. We might revisit this choice latter on once we
get more experience in how device memory is use and its impact on memory get more experience in how device memory is used and its impact on memory
resource control. resource control.
Note that device memory can never be pin nor by device driver nor through GUP Note that device memory can never be pinned by device driver nor through GUP
and thus such memory is always free upon process exit. Or when last reference and thus such memory is always free upon process exit. Or when last reference
is drop in case of share memory or file back memory. is dropped in case of shared memory or file backed memory.
MAINTAINERS
View file @ 76ea470c
...@@ -6410,6 +6410,7 @@ L: linux-mm@kvack.org ...@@ -6410,6 +6410,7 @@ L: linux-mm@kvack.org
S: Maintained S: Maintained
F: mm/hmm* F: mm/hmm*
F: include/linux/hmm* F: include/linux/hmm*
F: Documentation/vm/hmm.txt
HOST AP DRIVER HOST AP DRIVER
M: Jouni Malinen <j@w1.fi> M: Jouni Malinen <j@w1.fi>
......
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