Commit e8eddfd2 authored by Jérôme Glisse's avatar Jérôme Glisse Committed by Linus Torvalds

Documentation/vm/hmm.txt: typos and syntaxes fixes

This fix typos and syntaxes, thanks to Randy Dunlap for pointing them out
(they were all my faults).

Link: http://lkml.kernel.org/r/20180409151859.4713-1-jglisse@redhat.comSigned-off-by: default avatarJérôme Glisse <jglisse@redhat.com>
Reviewed-by: default avatarRalph Campbell <rcampbell@nvidia.com>
Reviewed-by: default avatarRandy Dunlap <rdunlap@infradead.org>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent 9d8a463a
Heterogeneous Memory Management (HMM) Heterogeneous Memory Management (HMM)
Provide infrastructure and helpers to integrate non conventional memory (device Provide infrastructure and helpers to integrate non-conventional memory (device
memory like GPU on board memory) into regular kernel code path. Corner stone of memory like GPU on board memory) into regular kernel path, with the cornerstone
this being specialize struct page for such memory (see sections 5 to 7 of this of this being specialized struct page for such memory (see sections 5 to 7 of
document). this document).
HMM also provide optional helpers for SVM (Share Virtual Memory) ie allowing a HMM also provides optional helpers for SVM (Share Virtual Memory), i.e.,
device to transparently access program address coherently with the CPU meaning allowing a device to transparently access program address coherently with the
that any valid pointer on the CPU is also a valid pointer for the device. This CPU meaning that any valid pointer on the CPU is also a valid pointer for the
is becoming a mandatory to simplify the use of advance heterogeneous computing device. This is becoming mandatory to simplify the use of advanced hetero-
where GPU, DSP, or FPGA are used to perform various computations on behalf of geneous computing where GPU, DSP, or FPGA are used to perform various
a process. computations on behalf of a process.
This document is divided as follows: in the first section I expose the problems 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 related to using device specific memory allocators. In the second section, I
expose the hardware limitations that are inherent to many platforms. The third expose the hardware limitations that are inherent to many platforms. The third
section gives an overview of the HMM design. The fourth section explains how section gives an overview of the HMM design. The fourth section explains how
CPU page-table mirroring works and what is HMM's purpose in this context. The CPU page-table mirroring works and the purpose of HMM in this context. The
fifth section deals with how device memory is represented inside the kernel. fifth section deals with how device memory is represented inside the kernel.
Finally, the last section presents a new migration helper that allows lever- Finally, the last section presents a new migration helper that allows lever-
aging the device DMA engine. aging the device DMA engine.
...@@ -35,7 +35,7 @@ aging the device DMA engine. ...@@ -35,7 +35,7 @@ aging the device DMA engine.
1) Problems of using a device specific memory allocator: 1) Problems of using a device specific memory allocator:
Devices with a large amount of on board memory (several giga bytes) like GPUs Devices with a large amount of on board memory (several gigabytes) like GPUs
have historically managed their memory through dedicated driver specific APIs. have historically managed their memory through dedicated driver specific APIs.
This creates a disconnect between memory allocated and managed by a device This creates a disconnect between memory allocated and managed by a device
driver and regular application memory (private anonymous, shared memory, or driver and regular application memory (private anonymous, shared memory, or
...@@ -44,29 +44,29 @@ address space. I use shared address space to refer to the opposite situation: ...@@ -44,29 +44,29 @@ address space. I use shared address space to refer to the opposite situation:
i.e., one in which any application memory region can be used by a device i.e., one in which any application memory region can be used by a device
transparently. transparently.
Split address space because device can only access memory allocated through the Split address space happens because device can only access memory allocated
device specific API. This implies that all memory objects in a program are not through device specific API. This implies that all memory objects in a program
equal from the device point of view which complicates large programs that rely are not equal from the device point of view which complicates large programs
on a wide set of libraries. that rely on a wide set of libraries.
Concretly this means that code that wants to leverage devices like GPUs need to Concretely this means that code that wants to leverage devices like GPUs needs
copy object between genericly allocated memory (malloc, mmap private/share/) to copy object between generically allocated memory (malloc, mmap private, mmap
and memory allocated through the device driver API (this still end up with an share) and memory allocated through the device driver API (this still ends up
mmap but of the device file). with an mmap but of the device file).
For flat data-sets (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-sets (list, tree, ...) are hard to get right. Duplicating a complex data sets (list, tree, ...) are hard to get right. Duplicating a
complex data-set needs to re-map all the pointer relations between each of its complex data set needs to re-map all the pointer relations between each of its
elements. This is error prone and program gets harder to debug because of the elements. This is error prone and program gets harder to debug because of the
duplicate data-set and addresses. duplicate data set and addresses.
Split address space also means that libraries can not transparently use data Split address space also means that libraries cannot transparently use data
they are getting from the core program or another library and thus each library they are getting from the core program or another library and thus each library
might have to duplicate its input data-set using the device specific memory might have to duplicate its input data set using the device specific memory
allocator. Large projects suffer from this and waste resources because of the allocator. Large projects suffer from this and waste resources because of the
various memory copies. 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 allocated 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 explosion in the library entry points. combinatorial explosion in the library entry points.
...@@ -81,16 +81,16 @@ a shared address space for all other patterns. ...@@ -81,16 +81,16 @@ a shared address space for all other patterns.
2) I/O bus, device memory characteristics 2) I/O bus, device memory characteristics
I/O buses cripple shared address due to few limitations. Most I/O buses only I/O buses cripple shared address spaces due to a few limitations. Most I/O
allow basic memory access from device to main memory, even cache coherency is buses only allow basic memory access from device to main memory; even cache
often optional. Access to device memory from CPU is even more limited. More coherency is often optional. Access to device memory from CPU is even more
often than not, it is not cache coherent. limited. More often than not, it is not cache coherent.
If we only consider the PCIE bus, then a 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 operations 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 CPU 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 cannot perform atomic operations on it. Thus device memory cannot
be considered the same as regular memory from the 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
...@@ -99,14 +99,14 @@ The final limitation is latency. Access to main memory from the device has an ...@@ -99,14 +99,14 @@ The final limitation is latency. Access to main memory from the device has an
order of magnitude higher latency than when the device accesses its own memory. order of magnitude higher latency than when the device accesses its own memory.
Some platforms are developing new I/O buses or additions/modifications to PCIE Some platforms are developing new I/O buses or additions/modifications to PCIE
to address some of these 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 platforms are following this trend and architecture supports. Sadly, not all platforms are following this trend and
some major architectures are left without hardware solutions to these problems. some major architectures are left without hardware solutions to these problems.
So for shared address space to make sense, not only must we allow device to So for shared address space to make sense, not only must we allow devices to
access any memory memory but we must also permit any memory to be migrated to access any memory but we must also permit any memory to be migrated to device
device memory while device is using it (blocking CPU access while it happens). memory while device is using it (blocking CPU access while it happens).
------------------------------------------------------------------------------- -------------------------------------------------------------------------------
...@@ -123,13 +123,13 @@ while keeping track of CPU page table updates. Device page table updates are ...@@ -123,13 +123,13 @@ 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
allocate a buffer (or use a pool of pre-allocated buffers) and write GPU allocate a buffer (or use a pool of pre-allocated buffers) and write GPU
specific commands in it to perform the update (unmap, cache invalidations, and specific commands in it to perform the update (unmap, cache invalidations, and
flush, ...). This can not be done through common code for all devices. Hence flush, ...). This cannot be done through common code for all devices. Hence
why HMM provides helpers to factor out everything that can be while leaving the why HMM provides helpers to factor out everything that can be while leaving the
hardware specific details to the device driver. hardware specific details to the device driver.
The second mechanism HMM provides, is a new kind of ZONE_DEVICE memory that The second mechanism HMM provides is a new kind of ZONE_DEVICE memory that
allows allocating a struct page for each page of the device memory. Those pages allows allocating a struct page for each page of the device memory. Those pages
are special because the CPU can not map them. However, they allow migrating are special because the CPU cannot map them. However, they allow migrating
main memory to device memory using existing migration mechanisms and everything main memory to device memory using existing migration mechanisms and everything
looks like a page is swapped out to disk from the CPU point of view. Using a looks like a page is swapped out to disk from the CPU point of view. Using a
struct page gives the easiest and cleanest integration with existing mm mech- struct page gives the easiest and cleanest integration with existing mm mech-
...@@ -144,7 +144,7 @@ address A triggers a page fault and initiates a migration back to main memory. ...@@ -144,7 +144,7 @@ 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 these two features, HMM not only allows a device to mirror process address
space and keeping both CPU and device page table synchronized, but also lever- space and keeping both CPU and device page table synchronized, but also lever-
ages device memory by migrating the part of the data-set that is actively being ages device memory by migrating the part of the data set that is actively being
used by the device. used by the device.
...@@ -154,7 +154,7 @@ used by the device. ...@@ -154,7 +154,7 @@ used by the device.
Address space mirroring's main objective is to allow duplication of a range of Address space mirroring's main objective is to allow duplication of a range of
CPU page table into a device page table; HMM helps keep both synchronized. 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 wants to mirror a process address space must start with the
registration of an hmm_mirror struct: registration of an hmm_mirror struct:
int hmm_mirror_register(struct hmm_mirror *mirror, int hmm_mirror_register(struct hmm_mirror *mirror,
...@@ -162,7 +162,7 @@ registration of an hmm_mirror struct: ...@@ -162,7 +162,7 @@ registration of an hmm_mirror struct:
int hmm_mirror_register_locked(struct hmm_mirror *mirror, int hmm_mirror_register_locked(struct hmm_mirror *mirror,
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 used when the driver is already holding mmap_sem
of the mm in write mode. The mirror struct has a set of callbacks that are used of the mm in write mode. The mirror struct has a set of callbacks that are used
to propagate CPU page tables: to propagate CPU page tables:
...@@ -210,8 +210,8 @@ use either: ...@@ -210,8 +210,8 @@ use either:
bool block); bool block);
The first one (hmm_vma_get_pfns()) will only fetch present CPU page table The first one (hmm_vma_get_pfns()) will only fetch present CPU page table
entries and will not trigger a page fault on missing or non present entries. entries and will not trigger a page fault on missing or non-present entries.
The second one does trigger a page fault on missing or read only entry if the The second one does trigger a page fault on missing or read-only entry if the
write parameter is true. Page faults use the generic mm page fault code path write parameter is true. Page faults use the generic mm page fault code path
just like a CPU page fault. just like a CPU page fault.
...@@ -251,10 +251,10 @@ HMM implements all this on top of the mmu_notifier API because we wanted a ...@@ -251,10 +251,10 @@ HMM implements all this on top of the mmu_notifier API because we wanted a
simpler API and also to be able to perform optimizations latter on like doing simpler API and also to be able to perform optimizations latter on like doing
concurrent device updates in multi-devices scenario. concurrent device updates in multi-devices scenario.
HMM also serves as an impedence mismatch between how CPU page table updates HMM also serves as an impedance mismatch between how CPU page table updates
are done (by CPU write to the page table and TLB flushes) and how devices are done (by CPU write to the page table and TLB flushes) and how devices
update their own page table. Device updates are a multi-step process. First, update their own page table. Device updates are a multi-step process. First,
appropriate commands are writen to a buffer, then this buffer is scheduled for appropriate commands are written to a buffer, then this buffer is scheduled for
execution on the device. It is only once the device has executed commands in execution on the device. It is only once the device has executed commands in
the buffer that the update is done. Creating and scheduling the update command the buffer that the update is done. Creating and scheduling the update command
buffer can happen concurrently for multiple devices. Waiting for each device to buffer can happen concurrently for multiple devices. Waiting for each device to
...@@ -302,7 +302,7 @@ The hmm_devmem_ops is where most of the important things are: ...@@ -302,7 +302,7 @@ 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
dropped. This means the device page is now free and no longer used by anyone. dropped. This means the device page is now free and no longer used by anyone.
The second callback happens whenever the CPU tries to access a device page The second callback happens whenever the CPU tries to access a device page
which it can not do. This second callback must trigger a migration back to which it cannot do. This second callback must trigger a migration back to
system memory. system memory.
...@@ -310,7 +310,7 @@ system memory. ...@@ -310,7 +310,7 @@ system memory.
6) Migration to and from device memory 6) Migration to and from device memory
Because the CPU can not access device memory, migration must use the device DMA Because the CPU cannot access device memory, migration must use the device DMA
engine to perform copy from and to device memory. For this we need a new engine to perform copy from and to device memory. For this we need a new
migration helper: migration helper:
...@@ -326,7 +326,7 @@ migration helper: ...@@ -326,7 +326,7 @@ migration helper:
Unlike other migration functions it works on a range of virtual address, there Unlike other migration functions it works on a range of virtual address, there
are 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
makes the whole exersize pointless. The second reason is because the makes the whole exercise pointless. The second reason is because the
migration might be for a range of addresses the device is actively accessing. migration might be for a range of addresses the device is actively accessing.
The migrate_vma_ops struct defines two callbacks. First one (alloc_and_copy()) The migrate_vma_ops struct defines two callbacks. First one (alloc_and_copy())
...@@ -375,7 +375,7 @@ file backed page or shmem if device page is used for shared memory). This is a ...@@ -375,7 +375,7 @@ file backed page or shmem if device page is used for shared memory). This is a
deliberate choice to keep existing applications, that might start using device deliberate choice to keep existing applications, that might start using device
memory without knowing about it, running unimpacted. memory without knowing about it, running unimpacted.
A Drawback is that the OOM killer might kill an application using a lot of A drawback is that the OOM killer might kill an application using a lot of
device memory and not a lot of regular system memory and thus not freeing much device memory and not a lot of regular system memory and thus not freeing much
system memory. We want to gather more real world experience on how applications system memory. We want to gather more real world experience on how applications
and 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
...@@ -385,7 +385,7 @@ deciding to account device memory differently. ...@@ -385,7 +385,7 @@ deciding to account device memory differently.
Same decision was made for memory cgroup. Device memory pages 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 cannot 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 used and its impact on memory get more experience in how device memory is used and its impact on memory
resource control. resource control.
......
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