Commit fbf4432f authored by Linus Torvalds's avatar Linus Torvalds

Merge branch 'akpm' (patches from Andrew)

Merge more updates from Andrew Morton:

 - most of the rest of MM

 - a small number of misc things

 - lib/ updates

 - checkpatch

 - autofs updates

 - ipc/ updates

* emailed patches from Andrew Morton <[email protected]>: (126 commits)
  ipc: optimize semget/shmget/msgget for lots of keys
  ipc/sem: play nicer with large nsops allocations
  ipc/sem: drop sem_checkid helper
  ipc: convert kern_ipc_perm.refcount from atomic_t to refcount_t
  ipc: convert sem_undo_list.refcnt from atomic_t to refcount_t
  ipc: convert ipc_namespace.count from atomic_t to refcount_t
  kcov: support compat processes
  sh: defconfig: cleanup from old Kconfig options
  mn10300: defconfig: cleanup from old Kconfig options
  m32r: defconfig: cleanup from old Kconfig options
  drivers/pps: use surrounding "if PPS" to remove numerous dependency checks
  drivers/pps: aesthetic tweaks to PPS-related content
  cpumask: make cpumask_next() out-of-line
  kmod: move #ifdef CONFIG_MODULES wrapper to Makefile
  kmod: split off umh headers into its own file
  MAINTAINERS: clarify kmod is just a kernel module loader
  kmod: split out umh code into its own file
  test_kmod: flip INT checks to be consistent
  test_kmod: remove paranoid UINT_MAX check on uint range processing
  vfat: deduplicate hex2bin()
  ...
parents c054be10 0cfb6aee
......@@ -13,8 +13,12 @@ Optional properties:
Example:
pps {
compatible = "pps-gpio";
gpios = <&gpio2 6 0>;
pinctrl-names = "default";
pinctrl-0 = <&pinctrl_pps>;
gpios = <&gpio1 26 GPIO_ACTIVE_HIGH>;
assert-falling-edge;
compatible = "pps-gpio";
status = "okay";
};
......@@ -48,12 +48,12 @@ problem:
time_pps_create().
This implies that the source has a /dev/... entry. This assumption is
ok for the serial and parallel port, where you can do something
OK for the serial and parallel port, where you can do something
useful besides(!) the gathering of timestamps as it is the central
task for a PPS-API. But this assumption does not work for a single
task for a PPS API. But this assumption does not work for a single
purpose GPIO line. In this case even basic file-related functionality
(like read() and write()) makes no sense at all and should not be a
precondition for the use of a PPS-API.
precondition for the use of a PPS API.
The problem can be simply solved if you consider that a PPS source is
not always connected with a GPS data source.
......@@ -88,13 +88,13 @@ Coding example
--------------
To register a PPS source into the kernel you should define a struct
pps_source_info_s as follows:
pps_source_info as follows:
static struct pps_source_info pps_ktimer_info = {
.name = "ktimer",
.path = "",
.mode = PPS_CAPTUREASSERT | PPS_OFFSETASSERT | \
PPS_ECHOASSERT | \
.mode = PPS_CAPTUREASSERT | PPS_OFFSETASSERT |
PPS_ECHOASSERT |
PPS_CANWAIT | PPS_TSFMT_TSPEC,
.echo = pps_ktimer_echo,
.owner = THIS_MODULE,
......@@ -108,13 +108,13 @@ initialization routine as follows:
The pps_register_source() prototype is:
int pps_register_source(struct pps_source_info_s *info, int default_params)
int pps_register_source(struct pps_source_info *info, int default_params)
where "info" is a pointer to a structure that describes a particular
PPS source, "default_params" tells the system what the initial default
parameters for the device should be (it is obvious that these parameters
must be a subset of ones defined in the struct
pps_source_info_s which describe the capabilities of the driver).
pps_source_info which describe the capabilities of the driver).
Once you have registered a new PPS source into the system you can
signal an assert event (for example in the interrupt handler routine)
......@@ -142,8 +142,10 @@ If the SYSFS filesystem is enabled in the kernel it provides a new class:
Every directory is the ID of a PPS sources defined in the system and
inside you find several files:
$ ls /sys/class/pps/pps0/
assert clear echo mode name path [email protected] uevent
$ ls -F /sys/class/pps/pps0/
assert dev mode path [email protected]
clear echo name power/ uevent
Inside each "assert" and "clear" file you can find the timestamp and a
sequence number:
......@@ -154,32 +156,32 @@ sequence number:
Where before the "#" is the timestamp in seconds; after it is the
sequence number. Other files are:
* echo: reports if the PPS source has an echo function or not;
* echo: reports if the PPS source has an echo function or not;
* mode: reports available PPS functioning modes;
* mode: reports available PPS functioning modes;
* name: reports the PPS source's name;
* name: reports the PPS source's name;
* path: reports the PPS source's device path, that is the device the
PPS source is connected to (if it exists).
* path: reports the PPS source's device path, that is the device the
PPS source is connected to (if it exists).
Testing the PPS support
-----------------------
In order to test the PPS support even without specific hardware you can use
the ktimer driver (see the client subsection in the PPS configuration menu)
the pps-ktimer driver (see the client subsection in the PPS configuration menu)
and the userland tools available in your distribution's pps-tools package,
http://linuxpps.org , or https://github.com/ago/pps-tools .
http://linuxpps.org , or https://github.com/redlab-i/pps-tools.
Once you have enabled the compilation of ktimer just modprobe it (if
Once you have enabled the compilation of pps-ktimer just modprobe it (if
not statically compiled):
# modprobe ktimer
# modprobe pps-ktimer
and the run ppstest as follow:
$ ./ppstest /dev/pps0
$ ./ppstest /dev/pps1
trying PPS source "/dev/pps1"
found PPS source "/dev/pps1"
ok, found 1 source(s), now start fetching data...
......@@ -187,7 +189,7 @@ and the run ppstest as follow:
source 0 - assert 1186592700.388931295, sequence: 365 - clear 0.000000000, sequence: 0
source 0 - assert 1186592701.389032765, sequence: 366 - clear 0.000000000, sequence: 0
Please, note that to compile userland programs you need the file timepps.h .
Please note that to compile userland programs, you need the file timepps.h.
This is available in the pps-tools repository mentioned above.
......
......@@ -193,6 +193,39 @@ Example::
for (node = rb_first(&mytree); node; node = rb_next(node))
printk("key=%s\n", rb_entry(node, struct mytype, node)->keystring);
Cached rbtrees
--------------
Computing the leftmost (smallest) node is quite a common task for binary
search trees, such as for traversals or users relying on a the particular
order for their own logic. To this end, users can use 'struct rb_root_cached'
to optimize O(logN) rb_first() calls to a simple pointer fetch avoiding
potentially expensive tree iterations. This is done at negligible runtime
overhead for maintanence; albeit larger memory footprint.
Similar to the rb_root structure, cached rbtrees are initialized to be
empty via:
struct rb_root_cached mytree = RB_ROOT_CACHED;
Cached rbtree is simply a regular rb_root with an extra pointer to cache the
leftmost node. This allows rb_root_cached to exist wherever rb_root does,
which permits augmented trees to be supported as well as only a few extra
interfaces:
struct rb_node *rb_first_cached(struct rb_root_cached *tree);
void rb_insert_color_cached(struct rb_node *, struct rb_root_cached *, bool);
void rb_erase_cached(struct rb_node *node, struct rb_root_cached *);
Both insert and erase calls have their respective counterpart of augmented
trees:
void rb_insert_augmented_cached(struct rb_node *node, struct rb_root_cached *,
bool, struct rb_augment_callbacks *);
void rb_erase_augmented_cached(struct rb_node *, struct rb_root_cached *,
struct rb_augment_callbacks *);
Support for Augmented rbtrees
-----------------------------
......
Heterogeneous Memory Management (HMM)
Transparently allow any component of a program to use any memory region of said
program with a device without using device specific memory allocator. This is
becoming a requirement to simplify the use of advance heterogeneous computing
where GPU, DSP or FPGA are use to perform various computations.
This document is divided as follow, in the first section i expose the problems
related to the use of a device specific allocator. The second section i expose
the hardware limitations that are inherent to many platforms. The third section
gives an overview of HMM designs. The fourth section explains how CPU page-
table mirroring works and what is HMM purpose in this context. Fifth section
deals with how device memory is represented inside the kernel. Finaly the last
section present the new migration helper that allow to leverage the device DMA
engine.
1) Problems of using device specific memory allocator:
2) System bus, device memory characteristics
3) Share address space and migration
4) Address space mirroring implementation and API
5) Represent and manage device memory from core kernel point of view
6) Migrate to and from device memory
7) Memory cgroup (memcg) and rss accounting
-------------------------------------------------------------------------------
1) Problems of using device specific memory allocator:
Device with large amount of on board memory (several giga bytes) like GPU have
historically manage their memory through dedicated driver specific API. This
creates a disconnect between memory allocated and managed by device driver and
regular application memory (private anonymous, share memory or regular file
back memory). From here on i will refer to this aspect as split address space.
I use share address space to refer to the opposite situation ie one in which
any memory region can be use by device transparently.
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
equal from device point of view which complicate large program that rely on a
wide set of libraries.
Concretly this means that code that wants to leverage device like GPU need to
copy object between genericly allocated memory (malloc, mmap private/share/)
and memory allocated through the device driver API (this still end up with an
mmap but of the device file).
For flat dataset (array, grid, image, ...) this isn't too hard to achieve but
complex data-set (list, tree, ...) are hard to get right. Duplicating a complex
data-set need 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 duplicate
data-set.
Split address space also means that library can not transparently use data they
are getting from core program or other library and thus each library might have
to duplicate its input data-set using specific memory allocator. Large project
suffer from this and waste resources because of the various memory copy.
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
combinatorial explosions in the library entry points.
Finaly with the advance of high level language constructs (in C++ but in other
language too) it is now possible for compiler to leverage GPU or other devices
without even the programmer knowledge. Some of compiler identified patterns are
only do-able with a share address. It is as well more reasonable to use a share
address space for all the other patterns.
-------------------------------------------------------------------------------
2) System bus, device memory characteristics
System bus cripple share address due to few limitations. Most system bus only
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 than not it is not cache coherent.
If we only consider the PCIE bus than device can access main memory (often
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
in the other direction the CPUs can only access a limited range of the device
memory and can not perform atomic operations on it. Thus device memory can not
be consider like regular memory from kernel point of view.
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).
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.
Some platform are developing new system bus or additions/modifications to PCIE
to address some of those limitations (OpenCAPI, CCIX). They mainly allow two
way cache coherency between CPU and device and allow all atomic operations the
architecture supports. Saddly not all platform are following this trends and
some major architecture are left without hardware solutions to those problems.
So for share address space to make sense not only we must allow device 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).
-------------------------------------------------------------------------------
3) Share address space and migration
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
address point to same memory and this for any valid main memory address in
the process address space.
To achieve this, HMM offer a set of helpers to populate the device page table
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
allow a buffer (or use a pool of pre-allocated buffer) and write GPU specifics
commands in it to perform the update (unmap, cache invalidations and flush,
...). This can not be done through common code for all device. Hence why HMM
provides helpers to factor out everything that can be while leaving the gory
details to the device driver.
The second mechanism HMM provide is a new kind of ZONE_DEVICE memory that does
allow to allocate a struct page for each page of the device memory. Those page
are special because the CPU can not map them. They however allow to migrate
main memory to device memory using exhisting migration mechanism and everything
looks like if page was swap out to disk from CPU point of view. Using a struct
page gives the easiest and cleanest integration with existing mm mechanisms.
Again here HMM only provide helpers, first to hotplug new ZONE_DEVICE memory
for the device memory and second to perform migration. Policy decision 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
back to main memory ie when a page backing an given address A is migrated from
a main memory page to a device page then any CPU access to address A trigger a
page fault and initiate a migration back to main memory.
With this two features, HMM not only allow a device to mirror a process address
space and keeps both CPU and device page table synchronize, but also allow to
leverage device memory by migrating part of data-set that is actively use by a
device.
-------------------------------------------------------------------------------
4) Address space mirroring implementation and API
Address space mirroring main objective is to allow to duplicate range of CPU
page table into a device page table and HMM helps keeping both synchronize. A
device driver that want to mirror a process address space must start with the
registration of an hmm_mirror struct:
int hmm_mirror_register(struct hmm_mirror *mirror,
struct mm_struct *mm);
int hmm_mirror_register_locked(struct hmm_mirror *mirror,
struct mm_struct *mm);
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
to propagate CPU page table:
struct hmm_mirror_ops {
/* sync_cpu_device_pagetables() - synchronize page tables
*
* @mirror: pointer to struct hmm_mirror
* @update_type: type of update that occurred to the CPU page table
* @start: virtual start address of the range to update
* @end: virtual end address of the range to update
*
* This callback ultimately originates from mmu_notifiers when the CPU
* page table is updated. The device driver must update its page table
* in response to this callback. The update argument tells what action
* to perform.
*
* The device driver must not return from this callback until the device
* page tables are completely updated (TLBs flushed, etc); this is a
* synchronous call.
*/
void (*update)(struct hmm_mirror *mirror,
enum hmm_update action,
unsigned long start,
unsigned long end);
};
Device driver must perform update to the range following action (turn range
read only, or fully unmap, ...). Once driver callback returns the device must
be done with the update.
When device driver wants to populate a range of virtual address it can use
either:
int hmm_vma_get_pfns(struct vm_area_struct *vma,
struct hmm_range *range,
unsigned long start,
unsigned long end,
hmm_pfn_t *pfns);
int hmm_vma_fault(struct vm_area_struct *vma,
struct hmm_range *range,
unsigned long start,
unsigned long end,
hmm_pfn_t *pfns,
bool write,
bool block);
First one (hmm_vma_get_pfns()) will only fetch present CPU page table entry and
will not trigger a page fault on missing or non present entry. The second one
do trigger page fault on missing or read only entry if write parameter is true.
Page fault 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
that array correspond to an address in the virtual range. HMM provide a set of
flags to help driver identify special CPU page table entries.
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 :
int driver_populate_range(...)
{
struct hmm_range range;
...
again:
ret = hmm_vma_get_pfns(vma, &range, start, end, pfns);
if (ret)
return ret;
take_lock(driver->update);
if (!hmm_vma_range_done(vma, &range)) {
release_lock(driver->update);
goto again;
}
// Use pfns array content to update device page table
release_lock(driver->update);
return 0;
}
The driver->update lock is the same lock that driver takes inside its update()
callback. That lock must be call before hmm_vma_range_done() to avoid 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
simpler API and also to be able to perform optimization latter own like doing
concurrent device update in multi-devices scenario.
HMM also serve as an impedence missmatch between how CPU page table update are
done (by CPU write to the page table and TLB flushes) from how device update
their own page table. Device update is a multi-step process, first appropriate
commands are write to a buffer, then this buffer is schedule for execution on
the device. It is only once the device has executed commands in the buffer that
the update is done. Creating and scheduling update command buffer can happen
concurrently for multiple devices. Waiting for each device to report commands
as executed is serialize (there is no point in doing this concurrently).
-------------------------------------------------------------------------------
5) Represent and manage device memory from core kernel point of view
Several differents design were try to support device memory. First one use
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
that were back by device memory. It turns out that this ended up replicating
most of the fields of struct page and also needed many kernel code path 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
but only care about struct page contents. Because of this HMM switchted to
directly using struct page for device memory which left most kernel code path
un-aware of the difference. We only need to make sure that no one ever try to
map those page from the CPU side.
HMM provide 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:
struct hmm_devmem *hmm_devmem_add(const struct hmm_devmem_ops *ops,
struct device *device,
unsigned long size);
void hmm_devmem_remove(struct hmm_devmem *devmem);
The hmm_devmem_ops is where most of the important things are:
struct hmm_devmem_ops {
void (*free)(struct hmm_devmem *devmem, struct page *page);
int (*fault)(struct hmm_devmem *devmem,
struct vm_area_struct *vma,
unsigned long addr,
struct page *page,
unsigned flags,
pmd_t *pmdp);
};
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
second callback happens whenever CPU try to access a device page which it can
not do. This second callback must trigger a migration back to system memory.
-------------------------------------------------------------------------------
6) Migrate to and from device memory
Because CPU can not access device memory, migration must use device DMA engine
to perform copy from and to device memory. For this we need a new migration
helper:
int migrate_vma(const struct migrate_vma_ops *ops,
struct vm_area_struct *vma,
unsigned long mentries,
unsigned long start,
unsigned long end,
unsigned long *src,
unsigned long *dst,
void *private);
Unlike other migration function it works on a range of virtual address, there
is 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
make the whole excersie pointless. The second reason is because driver trigger
such migration base on range of address the device is actively accessing.
The migrate_vma_ops struct define two callbacks. First one (alloc_and_copy())
control destination memory allocation and copy operation. Second one is there
to allow device driver to perform cleanup operation after migration.
struct migrate_vma_ops {
void (*alloc_and_copy)(struct vm_area_struct *vma,
const unsigned long *src,
unsigned long *dst,
unsigned long start,
unsigned long end,
void *private);
void (*finalize_and_map)(struct vm_area_struct *vma,
const unsigned long *src,
const unsigned long *dst,
unsigned long start,
unsigned long end,
void *private);
};
It is important to stress that this migration helpers allow for hole in the
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
but just skip over those pages.
The alloc_and_copy() might as well decide to not migrate all pages in the
range (for reasons under the callback control). For those the callback just
have to leave the corresponding dst entry empty.
Finaly the migration of the struct page might fails (for file back page) for
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
migrated. Note those page were still copied to new page and thus we wasted
bandwidth but this is considered as a rare event and a price that we are
willing to pay to keep all the code simpler.
-------------------------------------------------------------------------------