This moves the actually exported functions towards the end of the file,
and reorders some functions to be in more logical blocks as a preparation
for moving various stubs inline into the main functionality using
IS_ENABLED().
mm: rename CONFIG_HAVE_GENERIC_GUP to CONFIG_HAVE_FAST_GUP
We only support the generic GUP now, so rename the config option to
be more clear, and always use the mm/Kconfig definition of the
symbol and select it from the arch Kconfigs.
The mips code is mostly equivalent to the generic one, minus various
bugfixes and an arch override for gup_fast_permitted.
Note that this defines ARCH_HAS_PTE_SPECIAL for mips as mips has
pte_special and pte_mkspecial implemented and used in the existing gup
code. They are no-op stubs, though which makes me a little unsure if this
is really right thing to do.
Note that this also adds back a missing cpu_has_dc_aliases check for
__get_user_pages_fast, which the old code was only doing for
get_user_pages_fast. This clearly looks like an oversight, as any
condition that makes get_user_pages_fast unsafe also applies to
__get_user_pages_fast.
mm: lift the x86_32 PAE version of gup_get_pte to common code
The split low/high access is the only non-READ_ONCE version of gup_get_pte
that did show up in the various arch implemenations. Lift it to common
code and drop the ifdef based arch override.
Pass in the already calculated end value instead of recomputing it, and
leave the end > start check in the callers instead of duplicating them in
the arch code.
Waiman Long [Fri, 12 Jul 2019 03:56:38 +0000 (20:56 -0700)]
mm, memcg: add a memcg_slabinfo debugfs file
There are concerns about memory leaks from extensive use of memory cgroups
as each memory cgroup creates its own set of kmem caches. There is a
possiblity that the memcg kmem caches may remain even after the memory
cgroups have been offlined. Therefore, it will be useful to show the
status of each of memcg kmem caches.
This patch introduces a new <debugfs>/memcg_slabinfo file which is
somewhat similar to /proc/slabinfo in format, but lists only information
about kmem caches that have child memcg kmem caches. Information
available in /proc/slabinfo are not repeated in memcg_slabinfo.
Roman Gushchin [Fri, 12 Jul 2019 03:56:34 +0000 (20:56 -0700)]
mm: memcg/slab: reparent memcg kmem_caches on cgroup removal
Let's reparent non-root kmem_caches on memcg offlining. This allows us to
release the memory cgroup without waiting for the last outstanding kernel
object (e.g. dentry used by another application).
Since the parent cgroup is already charged, everything we need to do is to
splice the list of kmem_caches to the parent's kmem_caches list, swap the
memcg pointer, drop the css refcounter for each kmem_cache and adjust the
parent's css refcounter.
Please, note that kmem_cache->memcg_params.memcg isn't a stable pointer
anymore. It's safe to read it under rcu_read_lock(), cgroup_mutex held,
or any other way that protects the memory cgroup from being released.
We can race with the slab allocation and deallocation paths. It's not a
big problem: parent's charge and slab global stats are always correct, and
we don't care anymore about the child usage and global stats. The child
cgroup is already offline, so we don't use or show it anywhere.
Local slab stats (NR_SLAB_RECLAIMABLE and NR_SLAB_UNRECLAIMABLE) aren't
used anywhere except count_shadow_nodes(). But even there it won't break
anything: after reparenting "nodes" will be 0 on child level (because
we're already reparenting shrinker lists), and on parent level page stats
always were 0, and this patch won't change anything.
Roman Gushchin [Fri, 12 Jul 2019 03:56:31 +0000 (20:56 -0700)]
mm: memcg/slab: stop setting page->mem_cgroup pointer for slab pages
Every slab page charged to a non-root memory cgroup has a pointer to the
memory cgroup and holds a reference to it, which protects a non-empty
memory cgroup from being released. At the same time the page has a
pointer to the corresponding kmem_cache, and also hold a reference to the
kmem_cache. And kmem_cache by itself holds a reference to the cgroup.
So there is clearly some redundancy, which allows to stop setting the
page->mem_cgroup pointer and rely on getting memcg pointer indirectly via
kmem_cache. Further it will allow to change this pointer easier, without
a need to go over all charged pages.
So let's stop setting page->mem_cgroup pointer for slab pages, and stop
using the css refcounter directly for protecting the memory cgroup from
going away. Instead rely on kmem_cache as an intermediate object.
Make sure that vmstats and shrinker lists are working as previously, as
well as /proc/kpagecgroup interface.
Currently each charged slab page holds a reference to the cgroup to which
it's charged. Kmem_caches are held by the memcg and are released all
together with the memory cgroup. It means that none of kmem_caches are
released unless at least one reference to the memcg exists, which is very
far from optimal.
Let's rework it in a way that allows releasing individual kmem_caches as
soon as the cgroup is offline, the kmem_cache is empty and there are no
pending allocations.
To make it possible, let's introduce a new percpu refcounter for non-root
kmem caches. The counter is initialized to the percpu mode, and is
switched to the atomic mode during kmem_cache deactivation. The counter
is bumped for every charged page and also for every running allocation.
So the kmem_cache can't be released unless all allocations complete.
To shutdown non-active empty kmem_caches, let's reuse the work queue,
previously used for the kmem_cache deactivation. Once the reference
counter reaches 0, let's schedule an asynchronous kmem_cache release.
* I used the following simple approach to test the performance
(stolen from another patchset by T. Harding):
time find / -name fname-no-exist
echo 2 > /proc/sys/vm/drop_caches
repeat 10 times
Results:
orig patched
real 0m1.455s real 0m1.355s
user 0m0.206s user 0m0.219s
sys 0m0.855s sys 0m0.807s
real 0m1.487s real 0m1.699s
user 0m0.221s user 0m0.256s
sys 0m0.806s sys 0m0.948s
real 0m1.515s real 0m1.505s
user 0m0.183s user 0m0.215s
sys 0m0.876s sys 0m0.858s
real 0m1.291s real 0m1.380s
user 0m0.193s user 0m0.198s
sys 0m0.843s sys 0m0.786s
real 0m1.364s real 0m1.374s
user 0m0.180s user 0m0.182s
sys 0m0.868s sys 0m0.806s
real 0m1.352s real 0m1.312s
user 0m0.201s user 0m0.212s
sys 0m0.820s sys 0m0.761s
real 0m1.302s real 0m1.349s
user 0m0.205s user 0m0.203s
sys 0m0.803s sys 0m0.792s
real 0m1.334s real 0m1.301s
user 0m0.194s user 0m0.201s
sys 0m0.806s sys 0m0.779s
real 0m1.426s real 0m1.434s
user 0m0.216s user 0m0.181s
sys 0m0.824s sys 0m0.864s
real 0m1.350s real 0m1.295s
user 0m0.200s user 0m0.190s
sys 0m0.842s sys 0m0.811s
So it looks like the difference is not noticeable in this test.
Roman Gushchin [Fri, 12 Jul 2019 03:56:24 +0000 (20:56 -0700)]
mm: memcg/slab: synchronize access to kmem_cache dying flag using a spinlock
Currently the memcg_params.dying flag and the corresponding workqueue used
for the asynchronous deactivation of kmem_caches is synchronized using the
slab_mutex.
It makes impossible to check this flag from the irq context, which will be
required in order to implement asynchronous release of kmem_caches.
So let's switch over to the irq-save flavor of the spinlock-based
synchronization.
Roman Gushchin [Fri, 12 Jul 2019 03:56:20 +0000 (20:56 -0700)]
mm: memcg/slab: don't check the dying flag on kmem_cache creation
There is no point in checking the root_cache->memcg_params.dying flag on
kmem_cache creation path. New allocations shouldn't be performed using a
dead root kmem_cache, so no new memcg kmem_cache creation can be scheduled
after the flag is set. And if it was scheduled before,
flush_memcg_workqueue() will wait for it anyway.
Roman Gushchin [Fri, 12 Jul 2019 03:56:16 +0000 (20:56 -0700)]
mm: memcg/slab: unify SLAB and SLUB page accounting
Currently the page accounting code is duplicated in SLAB and SLUB
internals. Let's move it into new (un)charge_slab_page helpers in the
slab_common.c file. These helpers will be responsible for statistics
(global and memcg-aware) and memcg charging. So they are replacing direct
memcg_(un)charge_slab() calls.
Let's separate the page counter modification code out of
__memcg_kmem_uncharge() in a way similar to what
__memcg_kmem_charge() and __memcg_kmem_charge_memcg() work.
This will allow to reuse this code later using a new
memcg_kmem_uncharge_memcg() wrapper, which calls
__memcg_kmem_uncharge_memcg() if memcg_kmem_enabled()
check is passed.
Currently SLUB uses a work scheduled after an RCU grace period to
deactivate a non-root kmem_cache. This mechanism can be reused for
kmem_caches release, but requires generalization for SLAB case.
Introduce kmemcg_cache_deactivate() function, which calls
allocator-specific __kmem_cache_deactivate() and schedules execution of
__kmem_cache_deactivate_after_rcu() with all necessary locks in a worker
context after an rcu grace period.
Here is the new calling scheme:
kmemcg_cache_deactivate()
__kmemcg_cache_deactivate() SLAB/SLUB-specific
kmemcg_rcufn() rcu
kmemcg_workfn() work
__kmemcg_cache_deactivate_after_rcu() SLAB/SLUB-specific
Roman Gushchin [Fri, 12 Jul 2019 03:56:06 +0000 (20:56 -0700)]
mm: memcg/slab: rename slab delayed deactivation functions and fields
The delayed work/rcu deactivation infrastructure of non-root kmem_caches
can be also used for asynchronous release of these objects. Let's get rid
of the word "deactivation" in corresponding names to make the code look
better after generalization.
It's easier to make the renaming first, so that the generalized code will
look consistent from scratch.
Roman Gushchin [Fri, 12 Jul 2019 03:56:02 +0000 (20:56 -0700)]
mm: memcg/slab: postpone kmem_cache memcg pointer initialization to memcg_link_cache()
Patch series "mm: reparent slab memory on cgroup removal", v7.
# Why do we need this?
We've noticed that the number of dying cgroups is steadily growing on most
of our hosts in production. The following investigation revealed an issue
in the userspace memory reclaim code [1], accounting of kernel stacks [2],
and also the main reason: slab objects.
The underlying problem is quite simple: any page charged to a cgroup holds
a reference to it, so the cgroup can't be reclaimed unless all charged
pages are gone. If a slab object is actively used by other cgroups, it
won't be reclaimed, and will prevent the origin cgroup from being
reclaimed.
Slab objects, and first of all vfs cache, is shared between cgroups, which
are using the same underlying fs, and what's even more important, it's
shared between multiple generations of the same workload. So if something
is running periodically every time in a new cgroup (like how systemd
works), we do accumulate multiple dying cgroups.
Strictly speaking pagecache isn't different here, but there is a key
difference: we disable protection and apply some extra pressure on LRUs of
dying cgroups, and these LRUs contain all charged pages. My experiments
show that with the disabled kernel memory accounting the number of dying
cgroups stabilizes at a relatively small number (~100, depends on memory
pressure and cgroup creation rate), and with kernel memory accounting it
grows pretty steadily up to several thousands.
Memory cgroups are quite complex and big objects (mostly due to percpu
stats), so it leads to noticeable memory losses. Memory occupied by dying
cgroups is measured in hundreds of megabytes. I've even seen a host with
more than 100Gb of memory wasted for dying cgroups. It leads to a
degradation of performance with the uptime, and generally limits the usage
of cgroups.
My previous attempt [3] to fix the problem by applying extra pressure on
slab shrinker lists caused a regressions with xfs and ext4, and has been
reverted [4]. The following attempts to find the right balance [5, 6]
were not successful.
So instead of trying to find a maybe non-existing balance, let's do
reparent accounted slab caches to the parent cgroup on cgroup removal.
# Implementation approach
There is however a significant problem with reparenting of slab memory:
there is no list of charged pages. Some of them are in shrinker lists,
but not all. Introducing of a new list is really not an option.
But fortunately there is a way forward: every slab page has a stable
pointer to the corresponding kmem_cache. So the idea is to reparent
kmem_caches instead of slab pages.
It's actually simpler and cheaper, but requires some underlying changes:
1) Make kmem_caches to hold a single reference to the memory cgroup,
instead of a separate reference per every slab page.
2) Stop setting page->mem_cgroup pointer for memcg slab pages and use
page->kmem_cache->memcg indirection instead. It's used only on
slab page release, so performance overhead shouldn't be a big issue.
3) Introduce a refcounter for non-root slab caches. It's required to
be able to destroy kmem_caches when they become empty and release
the associated memory cgroup.
There is a bonus: currently we release all memcg kmem_caches all together
with the memory cgroup itself. This patchset allows individual
kmem_caches to be released as soon as they become inactive and free.
Some additional implementation details are provided in corresponding
commit messages.
# Results
Below is the average number of dying cgroups on two groups of our
production hosts. They do run some sort of web frontend workload, the
memory pressure is moderate. As we can see, with the kernel memory
reparenting the number stabilizes in 60s range; however with the original
version it grows almost linearly and doesn't show any signs of plateauing.
The difference in slab and percpu usage between patched and unpatched
versions also grows linearly. In 7 days it exceeded 200Mb.
[1]: commit 68600f623d69 ("mm: don't miss the last page because of round-off error")
[2]: commit 9b6f7e163cd0 ("mm: rework memcg kernel stack accounting")
[3]: commit 172b06c32b94 ("mm: slowly shrink slabs with a relatively small number of objects")
[4]: commit a9a238e83fbb ("Revert "mm: slowly shrink slabs with a relatively small number of objects")
[5]: https://lkml.org/lkml/2019/1/28/1865
[6]: https://marc.info/?l=linux-mm&m=155064763626437&w=2
This patch (of 10):
Initialize kmem_cache->memcg_params.memcg pointer in memcg_link_cache()
rather than in init_memcg_params().
Once kmem_cache will hold a reference to the memory cgroup, it will
simplify the refcounting.
For non-root kmem_caches memcg_link_cache() is always called before the
kmem_cache becomes visible to a user, so it's safe.
Johannes Weiner [Fri, 12 Jul 2019 03:55:59 +0000 (20:55 -0700)]
mm: memcontrol: dump memory.stat during cgroup OOM
The current cgroup OOM memory info dump doesn't include all the memory
we are tracking, nor does it give insight into what the VM tried to do
leading up to the OOM. All that useful info is in memory.stat.
Furthermore, the recursive printing for every child cgroup can
generate absurd amounts of data on the console for larger cgroup
trees, and it's not like we provide a per-cgroup breakdown during
global OOM kills.
When an OOM kill is triggered, print one set of recursive memory.stat
items at the level whose limit triggered the OOM condition.
The memory controller in cgroup v2 exposes memory.events file for each
memcg which shows the number of times events like low, high, max, oom
and oom_kill have happened for the whole tree rooted at that memcg.
Users can also poll or register notification to monitor the changes in
that file. Any event at any level of the tree rooted at memcg will
notify all the listeners along the path till root_mem_cgroup. There are
existing users which depend on this behavior.
However there are users which are only interested in the events
happening at a specific level of the memcg tree and not in the events in
the underlying tree rooted at that memcg. One such use-case is a
centralized resource monitor which can dynamically adjust the limits of
the jobs running on a system. The jobs can create their sub-hierarchy
for their own sub-tasks. The centralized monitor is only interested in
the events at the top level memcgs of the jobs as it can then act and
adjust the limits of the jobs. Using the current memory.events for such
centralized monitor is very inconvenient. The monitor will keep
receiving events which it is not interested and to find if the received
event is interesting, it has to read memory.event files of the next
level and compare it with the top level one. So, let's introduce
memory.events.local to the memcg which shows and notify for the events
at the memcg level.
Now, does memory.stat and memory.pressure need their local versions. IMHO
no due to the no internal process contraint of the cgroup v2. The
memory.stat file of the top level memcg of a job shows the stats and
vmevents of the whole tree. The local stats or vmevents of the top level
memcg will only change if there is a process running in that memcg but v2
does not allow that. Similarly for memory.pressure there will not be any
process in the internal nodes and thus no chance of local pressure.
memcg, fsnotify: no oom-kill for remote memcg charging
Commit d46eb14b735b ("fs: fsnotify: account fsnotify metadata to
kmemcg") added remote memcg charging for fanotify and inotify event
objects. The aim was to charge the memory to the listener who is
interested in the events but without triggering the OOM killer.
Otherwise there would be security concerns for the listener.
At the time, oom-kill trigger was not in the charging path. A parallel
work added the oom-kill back to charging path i.e. commit 29ef680ae7c2
("memcg, oom: move out_of_memory back to the charge path"). So to not
trigger oom-killer in the remote memcg, explicitly add
__GFP_RETRY_MAYFAIL to the fanotigy and inotify event allocations.
The documentation of __GFP_RETRY_MAYFAIL clearly mentioned that the OOM
killer will not be triggered and indeed the page alloc does not invoke OOM
killer for such allocations. However we do trigger memcg OOM killer for
__GFP_RETRY_MAYFAIL. Fix that. This flag will used later to not trigger
oom-killer in the charging path for fanotify and inotify event
allocations.
mm/mincore.c: fix race between swapoff and mincore
Via commit 4b3ef9daa4fc ("mm/swap: split swap cache into 64MB trunks"),
after swapoff, the address_space associated with the swap device will be
freed. So swap_address_space() users which touch the address_space need
some kind of mechanism to prevent the address_space from being freed
during accessing.
When mincore processes an unmapped range for swapped shmem pages, it
doesn't hold the lock to prevent swap device from being swapped off. So
the following race is possible:
The address space may be accessed after being freed.
To fix the race, get_swap_device()/put_swap_device() is used to enclose
find_get_page() to check whether the swap entry is valid and prevent the
swap device from being swapoff during accessing.
Aaron Lu [Fri, 12 Jul 2019 03:55:41 +0000 (20:55 -0700)]
mm, swap: use rbtree for swap_extent
swap_extent is used to map swap page offset to backing device's block
offset. For a continuous block range, one swap_extent is used and all
these swap_extents are managed in a linked list.
These swap_extents are used by map_swap_entry() during swap's read and
write path. To find out the backing device's block offset for a page
offset, the swap_extent list will be traversed linearly, with
curr_swap_extent being used as a cache to speed up the search.
This works well as long as swap_extents are not huge or when the number
of processes that access swap device are few, but when the swap device
has many extents and there are a number of processes accessing the swap
device concurrently, it can be a problem. On one of our servers, the
disk's remaining size is tight:
$df -h
Filesystem Size Used Avail Use% Mounted on
... ...
/dev/nvme0n1p1 1.8T 1.3T 504G 72% /home/t4
When creating a 80G swapfile there, there are as many as 84656 swap
extents. The end result is, kernel spends abou 30% time in
map_swap_entry() and swap throughput is only 70MB/s.
As a comparison, when I used smaller sized swapfile, like 4G whose
swap_extent dropped to 2000, swap throughput is back to 400-500MB/s and
map_swap_entry() is about 3%.
One downside of using rbtree for swap_extent is, 'struct rbtree' takes
24 bytes while 'struct list_head' takes 16 bytes, that's 8 bytes more
for each swap_extent. For a swapfile that has 80k swap_extents, that
means 625KiB more memory consumed.
Test:
Since it's not possible to reboot that server, I can not test this patch
diretly there. Instead, I tested it on another server with NVMe disk.
I created a 20G swapfile on an NVMe backed XFS fs. By default, the
filesystem is quite clean and the created swapfile has only 2 extents.
Testing vanilla and this patch shows no obvious performance difference
when swapfile is not fragmented.
To see the patch's effects, I used some tweaks to manually fragment the
swapfile by breaking the extent at 1M boundary. This made the swapfile
have 20K extents.
swapout: the throughput of swap out, in KB/s, higher is better 1st
map_swap_entry: cpu cycles percent sampled by perf swapin: the
throughput of swap in, in KB/s, higher is better. 2nd map_swap_entry:
cpu cycles percent sampled by perf
nr_task=1 doesn't show any difference, this is due to the curr_swap_extent
can be effectively used to cache the correct swap extent for single task
workload.
mm/swap_state.c: simplify total_swapcache_pages() with get_swap_device()
total_swapcache_pages() may race with swapper_spaces[] allocation and
freeing. Previously, this is protected with a swapper_spaces[] specific
RCU mechanism. To simplify the logic/code complexity, it is replaced with
get/put_swap_device(). The code line number is reduced too. Although not
so important, the swapoff() performance improves too because one
synchronize_rcu() call during swapoff() is deleted.
mm, swap: fix race between swapoff and some swap operations
When swapin is performed, after getting the swap entry information from
the page table, system will swap in the swap entry, without any lock held
to prevent the swap device from being swapoff. This may cause the race
like below,
Because swapoff is usually done when system shutdown only, the race may
not hit many people in practice. But it is still a race need to be fixed.
To fix the race, get_swap_device() is added to check whether the specified
swap entry is valid in its swap device. If so, it will keep the swap
entry valid via preventing the swap device from being swapoff, until
put_swap_device() is called.
Because swapoff() is very rare code path, to make the normal path runs as
fast as possible, rcu_read_lock/unlock() and synchronize_rcu() instead of
reference count is used to implement get/put_swap_device(). >From
get_swap_device() to put_swap_device(), RCU reader side is locked, so
synchronize_rcu() in swapoff() will wait until put_swap_device() is
called.
In addition to swap_map, cluster_info, etc. data structure in the struct
swap_info_struct, the swap cache radix tree will be freed after swapoff,
so this patch fixes the race between swap cache looking up and swapoff
too.
Races between some other swap cache usages and swapoff are fixed too via
calling synchronize_rcu() between clearing PageSwapCache() and freeing
swap cache data structure.
Another possible method to fix this is to use preempt_off() +
stop_machine() to prevent the swap device from being swapoff when its data
structure is being accessed. The overhead in hot-path of both methods is
similar. The advantages of RCU based method are,
1. stop_machine() may disturb the normal execution code path on other
CPUs.
2. File cache uses RCU to protect its radix tree. If the similar
mechanism is used for swap cache too, it is easier to share code
between them.
3. RCU is used to protect swap cache in total_swapcache_pages() and
exit_swap_address_space() already. The two mechanisms can be
merged to simplify the logic.
Yang Shi [Fri, 12 Jul 2019 03:55:29 +0000 (20:55 -0700)]
mm/filemap.c: correct the comment about VM_FAULT_RETRY
Commit 6b4c9f446981 ("filemap: drop the mmap_sem for all blocking
operations") changed when mmap_sem is dropped during filemap page fault
and when returning VM_FAULT_RETRY.
Fix the callback 9p passes to read_cache_page to actually have the
proper type expected. Casting around function pointers can easily
hide typing bugs, and defeats control flow protection.
jffs2: pass the correct prototype to read_cache_page
Fix the callback jffs2 passes to read_cache_page to actually have the
proper type expected. Casting around function pointers can easily hide
typing bugs, and defeats control flow protection.
mm/filemap.c: fix an overly long line in read_cache_page
Patch series "fix filler_t callback type mismatches", v2.
Casting mapping->a_ops->readpage to filler_t causes an indirect call
type mismatch with Control-Flow Integrity checking. This change fixes
the mismatch in read_cache_page_gfp and read_mapping_page by adding
using a NULL filler argument as an indication to call ->readpage
directly, and by passing the right parameter callbacks in nfs and jffs2.
mm, debug_pagealloc: use a page type instead of page_ext flag
When debug_pagealloc is enabled, we currently allocate the page_ext
array to mark guard pages with the PAGE_EXT_DEBUG_GUARD flag. Now that
we have the page_type field in struct page, we can use that instead, as
guard pages are neither PageSlab nor mapped to userspace. This reduces
memory overhead when debug_pagealloc is enabled and there are no other
features requiring the page_ext array.
mm, page_alloc: more extensive free page checking with debug_pagealloc
The page allocator checks struct pages for expected state (mapcount,
flags etc) as pages are being allocated (check_new_page()) and freed
(free_pages_check()) to provide some defense against errors in page
allocator users.
Prior commits 479f854a207c ("mm, page_alloc: defer debugging checks of
pages allocated from the PCP") and 4db7548ccbd9 ("mm, page_alloc: defer
debugging checks of freed pages until a PCP drain") this has happened
for order-0 pages as they were allocated from or freed to the per-cpu
caches (pcplists). Since those are fast paths, the checks are now
performed only when pages are moved between pcplists and global free
lists. This however lowers the chances of catching errors soon enough.
In order to increase the chances of the checks to catch errors, the
kernel has to be rebuilt with CONFIG_DEBUG_VM, which also enables
multiple other internal debug checks (VM_BUG_ON() etc), which is
suboptimal when the goal is to catch errors in mm users, not in mm code
itself.
To catch some wrong users of the page allocator we have
CONFIG_DEBUG_PAGEALLOC, which is designed to have virtually no overhead
unless enabled at boot time. Memory corruptions when writing to freed
pages have often the same underlying errors (use-after-free, double free)
as corrupting the corresponding struct pages, so this existing debugging
functionality is a good fit to extend by also perform struct page checks
at least as often as if CONFIG_DEBUG_VM was enabled.
Specifically, after this patch, when debug_pagealloc is enabled on boot,
and CONFIG_DEBUG_VM disabled, pages are checked when allocated from or
freed to the pcplists *in addition* to being moved between pcplists and
free lists. When both debug_pagealloc and CONFIG_DEBUG_VM are enabled,
pages are checked when being moved between pcplists and free lists *in
addition* to when allocated from or freed to the pcplists.
When debug_pagealloc is not enabled on boot, the overhead in fast paths
should be virtually none thanks to the use of static key.
mm, debug_pagelloc: use static keys to enable debugging
Patch series "debug_pagealloc improvements".
I have been recently debugging some pcplist corruptions, where it would be
useful to perform struct page checks immediately as pages are allocated
from and freed to pcplists, which is now only possible by rebuilding the
kernel with CONFIG_DEBUG_VM (details in Patch 2 changelog).
To make this kind of debugging simpler in future on a distro kernel, I
have improved CONFIG_DEBUG_PAGEALLOC so that it has even smaller overhead
when not enabled at boot time (Patch 1) and also when enabled (Patch 3),
and extended it to perform the struct page checks more often when enabled
(Patch 2). Now it can be configured in when building a distro kernel
without extra overhead, and debugging page use after free or double free
can be enabled simply by rebooting with debug_pagealloc=on.
This patch (of 3):
CONFIG_DEBUG_PAGEALLOC has been redesigned by 031bc5743f15
("mm/debug-pagealloc: make debug-pagealloc boottime configurable") to
allow being always enabled in a distro kernel, but only perform its
expensive functionality when booted with debug_pagelloc=on. We can
further reduce the overhead when not boot-enabled (including page
allocator fast paths) using static keys. This patch introduces one for
debug_pagealloc core functionality, and another for the optional guard
page functionality (enabled by booting with debug_guardpage_minorder=X).
Nicolas Boichat [Fri, 12 Jul 2019 03:55:03 +0000 (20:55 -0700)]
mm/failslab.c: by default, do not fail allocations with direct reclaim only
When failslab was originally written, the intention of the
"ignore-gfp-wait" flag default value ("N") was to fail GFP_ATOMIC
allocations. Those were defined as (__GFP_HIGH), and the code would test
for __GFP_WAIT (0x10u).
However, since then, __GFP_WAIT was replaced by __GFP_RECLAIM
(___GFP_DIRECT_RECLAIM|___GFP_KSWAPD_RECLAIM), and GFP_ATOMIC is now
defined as (__GFP_HIGH|__GFP_ATOMIC|__GFP_KSWAPD_RECLAIM).
This means that when the flag is false, almost no allocation ever fails
(as even GFP_ATOMIC allocations contain ___GFP_KSWAPD_RECLAIM).
Restore the original intent of the code, by ignoring calls that directly
reclaim only (__GFP_DIRECT_RECLAIM), and thus, failing GFP_ATOMIC calls
again by default.
Previously totalram_pages was the global variable. Currently,
totalram_pages is the static inline function from the include/linux/mm.h
However, the function is also marked as EXPORT_SYMBOL, which is at best an
odd combination. Because there is no point for the static inline function
from a public header to be exported, this commit removes the
EXPORT_SYMBOL() marking. It will be still possible to use the function in
modules because all the symbols it depends on are exported.
Mike Rapoport [Fri, 12 Jul 2019 03:54:31 +0000 (20:54 -0700)]
sparc: remove ARCH_SELECT_MEMORY_MODEL
The ARCH_SELECT_MEMORY_MODEL option is enabled only for 64-bit. However,
64-bit configuration also enables ARCH_SPARSEMEM_DEFAULT and there is no
ARCH_FLATMEM_ENABLE in arch/sparc/Kconfig.
With such settings, the dependencies in mm/Kconfig are always evaluated to
SPARSEMEM=y for 64-bit and to FLATMEM=y for 32-bit.
The ARCH_SELECT_MEMORY_MODEL option in arch/sparc/Kconfig does not affect
anything and can be removed.
Mike Rapoport [Fri, 12 Jul 2019 03:54:27 +0000 (20:54 -0700)]
s390: remove ARCH_SELECT_MEMORY_MODEL
The only reason s390 has ARCH_SELECT_MEMORY_MODEL option in
arch/s390/Kconfig is an ancient compile error with allnoconfig which was
fixed by commit 97195d6b411f ("[S390] fix sparsemem related compile error
with allnoconfig on s390") by adding the ARCH_SELECT_MEMORY_MODEL option.
Since then a lot have changed and now allnoconfig builds just fine without
ARCH_SELECT_MEMORY_MODEL, so it can be removed.
Mike Rapoport [Fri, 12 Jul 2019 03:54:24 +0000 (20:54 -0700)]
arm: remove ARCH_SELECT_MEMORY_MODEL
Patch series "remove ARCH_SELECT_MEMORY_MODEL where it has no effect".
For several architectures the ARCH_SELECT_MEMORY_MODEL has no real effect
because the dependencies for the memory model are always evaluated to a
single value.
Remove the ARCH_SELECT_MEMORY_MODEL from the Kconfigs for these
architectures.
This patch (of 3):
The ARCH_SELECT_MEMORY_MODEL in arch/arm/Kconfig is enabled only when
ARCH_SPARSEMEM_ENABLE=y. But in this case, ARCH_SPARSEMEM_DEFAULT is also
enabled and this in turn enables SPARSEMEM_MANUAL.
Since there is no definition of ARCH_FLATMEM_ENABLE in arch/arm/Kconfig,
SPARSEMEM_MANUAL is the only enabled memory model, hence the final
selection will evaluate to SPARSEMEM=y.
Since ARCH_SPARSEMEM_ENABLE is set to 'y' only by several sub-arch
configurations, the default for must sub-arches would be the falback to
FLATMEM regardless of ARCH_SELECT_MEMORY_MODEL.
Marco Elver [Fri, 12 Jul 2019 03:54:18 +0000 (20:54 -0700)]
mm/kasan: add object validation in ksize()
ksize() has been unconditionally unpoisoning the whole shadow memory
region associated with an allocation. This can lead to various undetected
bugs, for example, double-kzfree().
Specifically, kzfree() uses ksize() to determine the actual allocation
size, and subsequently zeroes the memory. Since ksize() used to just
unpoison the whole shadow memory region, no invalid free was detected.
This patch addresses this as follows:
1. Add a check in ksize(), and only then unpoison the memory region.
2. Preserve kasan_unpoison_slab() semantics by explicitly unpoisoning
the shadow memory region using the size obtained from __ksize().
Tested:
1. With SLAB allocator: a) normal boot without warnings; b) verified the
added double-kzfree() is detected.
2. With SLUB allocator: a) normal boot without warnings; b) verified the
added double-kzfree() is detected.
Marco Elver [Fri, 12 Jul 2019 03:54:14 +0000 (20:54 -0700)]
mm/slab: refactor common ksize KASAN logic into slab_common.c
This refactors common code of ksize() between the various allocators into
slab_common.c: __ksize() is the allocator-specific implementation without
instrumentation, whereas ksize() includes the required KASAN logic.
Marco Elver [Fri, 12 Jul 2019 03:54:03 +0000 (20:54 -0700)]
mm/kasan: introduce __kasan_check_{read,write}
Patch series "mm/kasan: Add object validation in ksize()", v3.
This patch (of 5):
This introduces __kasan_check_{read,write}. __kasan_check functions may
be used from anywhere, even compilation units that disable instrumentation
selectively.
This change eliminates the need for the __KASAN_INTERNAL definition.
Marco Elver [Fri, 12 Jul 2019 03:54:00 +0000 (20:54 -0700)]
asm-generic, x86: add bitops instrumentation for KASAN
This adds a new header to asm-generic to allow optionally instrumenting
architecture-specific asm implementations of bitops.
This change includes the required change for x86 as reference and
changes the kernel API doc to point to bitops-instrumented.h instead.
Rationale: the functions in x86's bitops.h are no longer the kernel API
functions, but instead the arch_ prefixed functions, which are then
instrumented via bitops-instrumented.h.
Other architectures can similarly add support for asm implementations of
bitops.
The documentation text was derived from x86 and existing bitops
asm-generic versions: 1) references to x86 have been removed; 2) as a
result, some of the text had to be reworded for clarity and consistency.
Tested using lib/test_kasan with bitops tests (pre-requisite patch).
Bugzilla ref: https://bugzilla.kernel.org/show_bug.cgi?id=198439
Marco Elver [Fri, 12 Jul 2019 03:53:56 +0000 (20:53 -0700)]
x86: use static_cpu_has in uaccess region to avoid instrumentation
This patch is a pre-requisite for enabling KASAN bitops instrumentation;
using static_cpu_has instead of boot_cpu_has avoids instrumentation of
test_bit inside the uaccess region. With instrumentation, the KASAN
check would otherwise be flagged by objtool.
For consistency, kernel/signal.c was changed to mirror this change,
however, is never instrumented with KASAN (currently unsupported under
x86 32bit).
Marco Elver [Fri, 12 Jul 2019 03:53:49 +0000 (20:53 -0700)]
mm/kasan: print frame description for stack bugs
This adds support for printing stack frame description on invalid stack
accesses. The frame description is embedded by the compiler, which is
parsed and then pretty-printed.
Currently, we can only print the stack frame info for accesses to the
task's own stack, but not accesses to other tasks' stacks.
Example of what it looks like:
page dumped because: kasan: bad access detected
addr ffff8880673ef98a is located in stack of task insmod/2008 at offset 106 in frame:
kasan_stack_oob+0x0/0xf5 [test_kasan]
this frame has 2 objects:
[32, 36) 'i'
[96, 106) 'stack_array'
Wikipedia now has a main article to "tracing garbage collector" topic.
Change the URL and use the reStructuredText syntax for hyperlinks and add
more details about the use of the tool. Add a section about how to use
the kmemleak-test module to test the memory leak scanning.
mm/kmemleak.c: change error at _write when kmemleak is disabled
According to POSIX, EBUSY means that the "device or resource is busy", and
this can lead to people thinking that the file
`/sys/kernel/debug/kmemleak/` is somehow locked or being used by other
process. Change this error code to a more appropriate one.
in_softirq() is a wrong predicate to check if we are in a softirq
context. It also returns true if we have BH disabled, so objects are
falsely stamped with "softirq" comm. The correct predicate is
in_serving_softirq().
If user does cat from /sys/kernel/debug/kmemleak previously they would
see this, which is clearly wrong, this is system call context (see the
comm):
slub: don't panic for memcg kmem cache creation failure
Currently for CONFIG_SLUB, if a memcg kmem cache creation is failed and
the corresponding root kmem cache has SLAB_PANIC flag, the kernel will
be crashed. This is unnecessary as the kernel can handle the creation
failures of memcg kmem caches. Additionally CONFIG_SLAB does not
implement this behavior. So, to keep the behavior consistent between
SLAB and SLUB, removing the panic for memcg kmem cache creation
failures. The root kmem cache creation failure for SLAB_PANIC correctly
panics for both SLAB and SLUB.
mm/slab: sanity-check page type when looking up cache
This avoids any possible type confusion when looking up an object. For
example, if a non-slab were to be passed to kfree(), the invalid
slab_cache pointer (i.e. overlapped with some other value from the
struct page union) would be used for subsequent slab manipulations that
could lead to further memory corruption.
Since the page is already in cache, adding the PageSlab() check will
have nearly zero cost, so add a check and WARN() to virt_to_cache().
Additionally replaces an open-coded virt_to_cache(). To support the
failure mode this also updates all callers of virt_to_cache() and
cache_from_obj() to handle a NULL cache pointer return value (though
note that several already handle this case gracefully).
mm/slab: validate cache membership under freelist hardening
Patch series "mm/slab: Improved sanity checking".
This adds defenses against slab cache confusion (as seen in real-world
exploits[1]) and gracefully handles type confusions when trying to look
up slab caches from an arbitrary page. (Also is patch 3: new LKDTM
tests for these defenses as well as for the existing double-free
detection.
This patch (of 3):
When building under CONFIG_SLAB_FREELIST_HARDENING, it makes sense to
perform sanity-checking on the assumed slab cache during
kmem_cache_free() to make sure the kernel doesn't mix freelists across
slab caches and corrupt memory (as seen in the exploitation of flaws
like CVE-2018-9568[1]). Note that the prior code might WARN() but still
corrupt memory (i.e. return the assumed cache instead of the owned
cache).
There is no noticeable performance impact (changes are within noise).
Measuring parallel kernel builds, I saw the following with
CONFIG_SLAB_FREELIST_HARDENED, before and after this patch:
ocfs2: use kmemdup rather than duplicating its implementation
kmemdup is introduced to duplicate a region of memory in a neat way.
Rather than kmalloc/kzalloc + memcpy, which the programmer needs to
write the size twice (sometimes lead to mistakes), kmemdup improves
readability, leads to smaller code and also reduce the chances of
mistakes.
Suggestion to use kmemdup rather than using kmalloc/kzalloc + memcpy.
ocfs: no need to check return value of debugfs_create functions
When calling debugfs functions, there is no need to ever check the
return value. The function can work or not, but the code logic should
never do something different based on this.
Also, because there is no need to save the file dentry, remove all of
the variables that were being saved, and just recursively delete the
whole directory when shutting down, saving a lot of logic and local
variables.
Gang He [Fri, 12 Jul 2019 03:53:09 +0000 (20:53 -0700)]
ocfs2: add first lock wait time in locking_state
ocfs2 file system uses locking_state file under debugfs to dump each
ocfs2 file system's dlm lock resources, but the users ever encountered
some hang(deadlock) problems in ocfs2 file system. I'd like to add
first lock wait time in locking_state file, which can help the upper
scripts detect these deadlock problems via comparing the first lock wait
time with the current time.
Gang He [Fri, 12 Jul 2019 03:53:05 +0000 (20:53 -0700)]
ocfs2: add locking filter debugfs file
Add locking filter debugfs file, which is used to filter lock resources
dump from locking_state debugfs file. We use d_filter_secs field to
filter lock resources dump, the default d_filter_secs(0) value filters
nothing, otherwise, only dump the last N seconds active lock resources.
This enhancement can avoid dumping lots of old records. The
d_filter_secs value can be changed via locking_filter file.
Gang He [Fri, 12 Jul 2019 03:53:02 +0000 (20:53 -0700)]
ocfs2: add last unlock times in locking_state
ocfs2 file system uses locking_state file under debugfs to dump each
ocfs2 file system's dlm lock resources, but the dlm lock resources in
memory are becoming more and more after the files were touched by the
user. it will become a bit difficult to analyze these dlm lock resource
records in locking_state file by the upper scripts, though some files
are not active for now, which were accessed long time ago.
Then, I'd like to add last pr/ex unlock times in locking_state file for
each dlm lock resource record, the the upper scripts can use last unlock
time to filter inactive dlm lock resource record.
One of the more common cases of allocation size calculations is finding
the size of a structure that has a zero-sized array at the end, along
with memory for some number of elements for that array. For example:
CONFIG_BACKLIGHT_LCD_SUPPORT was removed in 8c5dc8d9f19c ("video:
backlight: Remove useless BACKLIGHT_LCD_SUPPORT kernel symbol").
Options protected by CONFIG_BACKLIGHT_LCD_SUPPORT are now available
directly.
Colin Ian King [Fri, 12 Jul 2019 03:52:42 +0000 (20:52 -0700)]
scripts/spelling.txt: add more spellings to spelling.txt
Here are some of the more common spelling mistakes and typos that I've
found while fixing up spelling mistakes in the kernel over the past few
months. Developers keep on coming up with more inventive ways to spell
words.
Evan Green [Fri, 12 Jul 2019 03:52:39 +0000 (20:52 -0700)]
scripts/decode_stacktrace: Accept dash/underscore in modules
The manpage for modprobe mentions that dashes and underscores are
treated interchangeably in module names. The stack trace dumps seem to
print module names with underscores. Use bash to replace _ with the
pattern [-_] so that file names with dashes or underscores can be found.
For example, this line:
[ 27.919759] hda_widget_sysfs_init+0x2b8/0x3a5 [snd_hda_core]
Chris Paterson [Fri, 12 Jul 2019 03:52:36 +0000 (20:52 -0700)]
scripts/spelling.txt: add spelling fix for prohibited
Misspelling 'prohibited' is quite common in the real world, although
surprisingly not so much in the Linux Kernel. In addition to fixing the
typo we may as well add it to the spelling checker.
Paul Walmsley [Fri, 12 Jul 2019 03:52:33 +0000 (20:52 -0700)]
scripts/spelling.txt: drop "sepc" from the misspelling list
The RISC-V architecture has a register named the "Supervisor Exception
Program Counter", or "sepc". This abbreviation triggers checkpatch.pl's
misspelling detector, resulting in noise in the checkpatch output. The
risk that this noise could cause more useful warnings to be missed seems
to outweigh the harm of an occasional misspelling of "spec". Thus drop
the "sepc" entry from the misspelling list.
nilfs2: do not use unexported cpu_to_le32()/le32_to_cpu() in uapi header
cpu_to_le32/le32_to_cpu is defined in include/linux/byteorder/generic.h,
which is not exported to user-space.
UAPI headers must use the ones prefixed with double-underscore.
Detected by compile-testing exported headers:
include/linux/nilfs2_ondisk.h: In function `nilfs_checkpoint_set_snapshot':
include/linux/nilfs2_ondisk.h:536:17: error: implicit declaration of function `cpu_to_le32' [-Werror=implicit-function-declaration]
cp->cp_flags = cpu_to_le32(le32_to_cpu(cp->cp_flags) | \
^
include/linux/nilfs2_ondisk.h:552:1: note: in expansion of macro `NILFS_CHECKPOINT_FNS'
NILFS_CHECKPOINT_FNS(SNAPSHOT, snapshot)
^~~~~~~~~~~~~~~~~~~~
include/linux/nilfs2_ondisk.h:536:29: error: implicit declaration of function `le32_to_cpu' [-Werror=implicit-function-declaration]
cp->cp_flags = cpu_to_le32(le32_to_cpu(cp->cp_flags) | \
^
include/linux/nilfs2_ondisk.h:552:1: note: in expansion of macro `NILFS_CHECKPOINT_FNS'
NILFS_CHECKPOINT_FNS(SNAPSHOT, snapshot)
^~~~~~~~~~~~~~~~~~~~
include/linux/nilfs2_ondisk.h: In function `nilfs_segment_usage_set_clean':
include/linux/nilfs2_ondisk.h:622:19: error: implicit declaration of function `cpu_to_le64' [-Werror=implicit-function-declaration]
su->su_lastmod = cpu_to_le64(0);
^~~~~~~~~~~
Henry Burns [Fri, 12 Jul 2019 03:52:14 +0000 (20:52 -0700)]
mm/z3fold.c: lock z3fold page before __SetPageMovable()
Following zsmalloc.c's example we call trylock_page() and unlock_page().
Also make z3fold_page_migrate() assert that newpage is passed in locked,
as per the documentation.
mm/memcontrol: fix wrong statistics in memory.stat
When we calculate total statistics for memcg1_stats and memcg1_events,
we use the the index 'i' in the for loop as the events index. Actually
we should use memcg1_stats[i] and memcg1_events[i] as the events index.
mm/nvdimm: add is_ioremap_addr and use that to check ioremap address
Architectures like powerpc use different address range to map ioremap
and vmalloc range. The memunmap() check used by the nvdimm layer was
wrongly using is_vmalloc_addr() to check for ioremap range which fails
for ppc64. This result in ppc64 not freeing the ioremap mapping. The
side effect of this is an unbind failure during module unload with
papr_scm nvdimm driver
Kuo-Hsin Yang [Fri, 12 Jul 2019 03:52:04 +0000 (20:52 -0700)]
mm: vmscan: scan anonymous pages on file refaults
When file refaults are detected and there are many inactive file pages,
the system never reclaim anonymous pages, the file pages are dropped
aggressively when there are still a lot of cold anonymous pages and
system thrashes. This issue impacts the performance of applications
with large executable, e.g. chrome.
With this patch, when file refault is detected, inactive_list_is_low()
always returns true for file pages in get_scan_count() to enable
scanning anonymous pages.
The problem can be reproduced by the following test program.
---8<---
void fallocate_file(const char *filename, off_t size)
{
struct stat st;
int fd;
if (!stat(filename, &st) && st.st_size >= size)
return;
long *alloc_anon(long size)
{
long *start = malloc(size);
memset(start, 1, size);
return start;
}
long access_file(const char *filename, long size, long rounds)
{
int fd, i;
volatile char *start1, *end1, *start2;
const int page_size = getpagesize();
long sum = 0;
/*
* Some applications, e.g. chrome, use a lot of executable file
* pages, map some of the pages with PROT_EXEC flag to simulate
* the behavior.
*/
start1 = mmap(NULL, size / 2, PROT_READ | PROT_EXEC, MAP_SHARED,
fd, 0);
if (start1 == MAP_FAILED) {
perror("mmap");
exit(1);
}
end1 = start1 + size / 2;
Running the test program on 2GB RAM VM with kernel 5.2.0-rc5, the program
fills ram with 2048 MB memory, access a 200 MB file for 10 times. Without
this patch, the file cache is dropped aggresively and every access to the
file is from disk.
$ ./thrash 2048 200 10
Allocate 2048 MB anonymous pages
Access 200 MB file pages
File access time, round 0: 2.489316 (sec)
File access time, round 1: 2.581277 (sec)
File access time, round 2: 2.487624 (sec)
File access time, round 3: 2.449100 (sec)
File access time, round 4: 2.420423 (sec)
File access time, round 5: 2.343411 (sec)
File access time, round 6: 2.454833 (sec)
File access time, round 7: 2.483398 (sec)
File access time, round 8: 2.572701 (sec)
File access time, round 9: 2.493014 (sec)
With this patch, these file pages can be cached.
$ ./thrash 2048 200 10
Allocate 2048 MB anonymous pages
Access 200 MB file pages
File access time, round 0: 2.475189 (sec)
File access time, round 1: 2.440777 (sec)
File access time, round 2: 2.411671 (sec)
File access time, round 3: 1.955267 (sec)
File access time, round 4: 0.029924 (sec)
File access time, round 5: 0.000808 (sec)
File access time, round 6: 0.000771 (sec)
File access time, round 7: 0.000746 (sec)
File access time, round 8: 0.000738 (sec)
File access time, round 9: 0.000747 (sec)
Checked the swap out stats during the test [1], 19006 pages swapped out
with this patch, 3418 pages swapped out without this patch. There are
more swap out, but I think it's within reasonable range when file backed
data set doesn't fit into the memory.
With 4 processes accessing non-overlapping parts of a large file, 30316
pages swapped out with this patch, 5152 pages swapped out without this
patch. The swapout number is small comparing to pgpgin.
Merge tag 'tag-chrome-platform-for-v5.3' of git://git.kernel.org/pub/scm/linux/kernel/git/chrome-platform/linux
Pull chrome platform updates from Benson Leung
"CrOS EC:
- Add new CrOS ISHTP transport protocol
- Add proper documentation for debugfs entries and expose resume and
uptime files
- Select LPC transport protocol variant at runtime.
- Add lid angle sensor driver
- Fix oops on suspend/resume for lightbar driver
- Set CrOS SPI transport protol in realtime
Wilco EC:
- Add telemetry char device interface
- Add support for event handling
- Add new sysfs attributes
Misc:
- Contains ib-mfd-cros-v5.3 immutable branch from mfd, with
cros_ec_commands.h header freshly synced with Chrome OS's EC
project"
* tag 'tag-chrome-platform-for-v5.3' of git://git.kernel.org/pub/scm/linux/kernel/git/chrome-platform/linux: (54 commits)
mfd / platform: cros_ec_debugfs: Expose resume result via debugfs
platform/chrome: lightbar: Get drvdata from parent in suspend/resume
iio: cros_ec: Add lid angle driver
platform/chrome: wilco_ec: Add circular buffer as event queue
platform/chrome: cros_ec_lpc_mec: Fix kernel-doc comment first line
platform/chrome: cros_ec_lpc: Choose Microchip EC at runtime
platform/chrome: cros_ec_lpc: Merge cros_ec_lpc and cros_ec_lpc_reg
Input: cros_ec_keyb: mask out extra flags in event_type
platform/chrome: wilco_ec: Fix unreleased lock in event_read()
platform/chrome: cros_ec_debugfs: cros_ec_uptime_fops can be static
platform/chrome: cros_ec_debugfs: Add debugfs ABI documentation
platform/chrome: cros_ec_debugfs: Fix kernel-doc comment first line
platform/chrome: cros_ec_debugfs: Add debugfs entry to retrieve EC uptime
mfd: cros_ec: Update I2S API
mfd: cros_ec: Add Management API entry points
mfd: cros_ec: Add SKU ID and Secure storage API
mfd: cros_ec: Add API for rwsig
mfd: cros_ec: Add API for Fingerprint support
mfd: cros_ec: Add API for Touchpad support
mfd: cros_ec: Add API for EC-EC communication
...
- Initial schemas for networking bindings. This includes ethernet, phy
and mdio common bindings with several Allwinner and stmmac converted
to the schema.
- Conversion of more Arm top-level SoC/board bindings to DT schema
- Conversion of PSCI binding to DT schema
- Rework Arm CPU schema to coexist with other CPU schemas
- Add a bunch of missing vendor prefixes and new ones for SoChip,
Sipeed, Kontron, B&R Industrial Automation GmbH, and Espressif
- Add Mediatek UART RX wakeup support to binding
- Add reset to ST UART binding
- Remove some Linuxisms from the endianness common-properties.txt
binding
- Make the flattened DT read-only after init
- Ignore disabled reserved memory nodes
- Clean-up some dead code in FDT functions
* tag 'devicetree-for-5.3' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux: (56 commits)
dt-bindings: vendor-prefixes: add Sipeed
dt-bindings: vendor-prefixes: add SoChip
dt-bindings: 83xx-512x-pci: Drop cell-index property
dt-bindings: serial: add documentation for Rx in-band wakeup support
dt-bindings: arm: Convert RDA Micro board/soc bindings to json-schema
of: unittest: simplify getting the adapter of a client
of/fdt: pass early_init_dt_reserve_memory_arch() with bool type nomap
of/platform: Drop superfluous cast in of_device_make_bus_id()
dt-bindings: usb: ehci: Fix example warnings
dt-bindings: net: Use phy-mode instead of phy-connection-type
dt-bindings: simple-framebuffer: Add requirement for pipelines
dt-bindings: display: Fix simple-framebuffer example
dt-bindings: net: mdio: Add child nodes
dt-bindings: net: mdio: Add address and size cells
dt-bindings: net: mdio: Add a nodename pattern
dt-bindings: mtd: sunxi-nand: Drop 'maxItems' from child 'reg' property
dt-bindings: arm: Limit cpus schema to only check Arm 'cpu' nodes
dt-bindings: backlight: lm3630a: correct schema validation
dt-bindings: net: dwmac: Deprecate the PHY reset properties
dt-bindings: net: sun8i-emac: Convert the binding to a schemas
...
Merge tag 'mmc-v5.3' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc
Pull MMC updates from Ulf Hansson:
"MMC core:
- Let the dma map ops deal with bouncing and drop dma_max_pfn() from
the dma-mapping interface for ARM
- Convert the generic MMC DT doc to YAML schemas
- Drop questionable support for powered-on re-init of SDIO cards at
runtime resume and for SDIO HW reset
- Prevent questionable re-init of powered-on removable SDIO cards at
system resume
- Cleanup and clarify some SDIO core code
MMC host:
- tmio: Make runtime PM enablement more flexible for variants
- tmio/renesas_sdhi: Rename DT doc tmio_mmc.txt to renesas,sdhi.txt
to clarify
- sdhci-pci: Add support for Intel EHL
- sdhci-pci-o2micro: Enable support for 8-bit bus
- sdhci-msm: Prevent acquiring a mutex while holding a spin_lock
- sdhci-of-esdhc: Improve clock management and tuning
- sdhci_am654: Enable support for 4 and 8-bit bus on J721E
- sdhci-sprd: Use pinctrl for a proper signal voltage switch
- sdhci-sprd: Add support for HS400 enhanced strobe mode
- sdhci-sprd: Enable PHY DLL and allow delay config to stabilize the
clock
- sdhci-sprd: Add support for optional gate clock
- sunxi-mmc: Convert DT doc to YAML schemas
- meson-gx: Add support for broken DRAM access for DMA
MEMSTICK core:
- Fixup error path of memstick_init()"
* tag 'mmc-v5.3' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc: (52 commits)
mmc: sdhci_am654: Add dependency on MMC_SDHCI_AM654
mmc: alcor: remove a redundant greater or equal to zero comparison
mmc: sdhci-msm: fix mutex while in spinlock
mmc: sdhci_am654: Make some symbols static
dma-mapping: remove dma_max_pfn
mmc: core: let the dma map ops handle bouncing
dt-binding: mmc: rename tmio_mmc.txt to renesas,sdhi.txt
mmc: sdhci-sprd: Add pin control support for voltage switch
dt-bindings: mmc: sprd: Add pinctrl support
mmc: sdhci-sprd: Add start_signal_voltage_switch ops
mmc: sdhci-pci: Add support for Intel EHL
mmc: tmio: Use dma_max_mapping_size() instead of a workaround
mmc: sdio: Drop unused in-parameter from mmc_sdio_init_card()
mmc: sdio: Drop unused in-parameter to mmc_sdio_reinit_card()
mmc: sdio: Don't re-initialize powered-on removable SDIO cards at resume
mmc: sdio: Drop powered-on re-init at runtime resume and HW reset
mmc: sdio: Move comment about re-initialization to mmc_sdio_reinit_card()
mmc: sdio: Drop mmc_claim|release_host() in mmc_sdio_power_restore()
mmc: sdio: Turn sdio_run_irqs() into static
mmc: sdhci: Fix indenting on SDHCI_CTRL_8BITBUS
...
Merge tag 'usb-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Pull USB / PHY updates from Greg KH:
"Here is the big USB and PHY driver pull request for 5.3-rc1.
Lots of stuff here, all of which has been in linux-next for a while
with no reported issues. Nothing is earth-shattering, just constant
forward progress for more devices supported and cleanups and small
fixes:
- USB gadget driver updates and fixes
- new USB gadget driver for some hardware, followed by a quick revert
of those patches as they were not ready to be merged...
- PHY driver updates
- Lots of new driver additions and cleanups with a few fixes mixed
in"
* tag 'usb-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (145 commits)
Revert "usb: gadget: storage: Remove warning message"
Revert "dt-bindings: add binding for USBSS-DRD controller."
Revert "usb:gadget Separated decoding functions from dwc3 driver."
Revert "usb:gadget Patch simplify usb_decode_set_clear_feature function."
Revert "usb:gadget Simplify usb_decode_get_set_descriptor function."
Revert "usb:cdns3 Add Cadence USB3 DRD Driver"
Revert "usb:cdns3 Fix for stuck packets in on-chip OUT buffer."
usb :fsl: Change string format for errata property
usb: host: Stops USB controller init if PLL fails to lock
usb: linux/fsl_device: Add platform member has_fsl_erratum_a006918
usb: phy: Workaround for USB erratum-A005728
usb: fsl: Set USB_EN bit to select ULPI phy
usb: Handle USB3 remote wakeup for LPM enabled devices correctly
drivers/usb/typec/tps6598x.c: fix 4CC cmd write
drivers/usb/typec/tps6598x.c: fix portinfo width
usb: storage: scsiglue: Do not skip VPD if try_vpd_pages is set
usb: renesas_usbhs: add a workaround for a race condition of workqueue
usb: gadget: udc: renesas_usb3: remove redundant assignment to ret
usb: dwc2: use a longer AHB idle timeout in dwc2_core_reset()
USB: gadget: function: fix issue Unneeded variable: "value"
...
Merge tag 'tty-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Pull tty / serial driver updates from Greg KH:
"Here is the "large" TTY and Serial driver update for 5.3-rc1.
It's in the negative number of lines overall as we removed an obsolete
serial driver that was causing problems for some people who were
trying to clean up some apis (the mpsc.c driver, which only worked for
some pre-production hardware that no one has anymore.)
Other than that, lots of tiny changes, cleaning up small things along
with some platform-specific serial driver updates.
All of these have been in linux-next for a while now with no reported
issues"
* tag 'tty-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty: (68 commits)
tty: serial: fsl_lpuart: add imx8qxp support
serial: imx: set_termios(): preserve RTS state
serial: imx: set_termios(): clarify RTS/CTS bits calculation
serial: imx: set_termios(): factor-out 'ucr2' initial value
serial: sh-sci: Terminate TX DMA during buffer flushing
serial: sh-sci: Fix TX DMA buffer flushing and workqueue races
serial: mpsc: Remove obsolete MPSC driver
serial: 8250: 8250_core: Fix missing unlock on error in serial8250_register_8250_port()
serial: stm32: add RX and TX FIFO flush
serial: stm32: add support of RX FIFO threshold
serial: stm32: add support of TX FIFO threshold
serial: stm32: update PIO transmission
serial: stm32: add support of timeout interrupt for RX
Revert "serial: 8250: Don't service RX FIFO if interrupts are disabled"
tty/serial/8250: use mctrl_gpio helpers
serial: mctrl_gpio: Check if GPIO property exisits before requesting it
serial: 8250: pericom_do_set_divisor can be static
tty: serial_core: Set port active bit in uart_port_activate
serial: 8250: Add MSR/MCR TIOCM conversion wrapper functions
serial: 8250: factor out serial8250_{set,clear}_THRI() helpers
...
Merge tag 'staging-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
Pull staging and IIO driver updates from Greg KH:
"Here is the big Staging and IIO driver update for 5.3-rc1.
Lots of new IIO drivers are in here, along with loads of tiny staging
driver cleanups and fixes. Overall we almost break even with the same
lines added as removed.
Full details are in the shortlog, they are too large to list here.
All of these changes have been in linux-next for a while with no
reported issues"
* tag 'staging-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging: (608 commits)
staging: kpc2000: simplify comparison to NULL in fileops.c
staging: kpc2000: simplify comparison to NULL in dma.c
staging: kpc2000: simplify comparison to NULL in kpc2000_spi.c
staging: rtl8723bs: hal: remove redundant assignment to packetType
staging: rtl8723bs: Change return type of hal_btcoex_IsBtDisabled()
staging: rtl8723bs: Remove rtw_btcoex_DisplayBtCoexInfo()
staging: rtl8723bs: Remove function rtw_btcoex_GetDBG()
staging: rtl8723bs: Remove function rtw_btcoex_SetDBG()
staging: rtl8723bs: Remove rtw_btcoex_IsBTCoexCtrlAMPDUSize()
staging: rtl8723bs: Remove rtw_btcoex_BtInfoNotify()
staging: rtl8723bs: Remove rtw_btcoex_ScanNotify()
staging: rtl8723bs: Remove rtw_btcoex_SetSingleAntPath()
staging: rtl8723bs: Remove rtw_btcoex_SetPGAntNum()
staging: rtl8192e: remove redundant initialization of rtstatus
staging: rtl8723bs: Remove rtw_btcoex_GetRaMask()
staging: rtl8723bs: Remove rtw_btcoex_SetChipType()
staging: rtl8723bs: Remove rtw_btcoex_ConnectNotify()
staging: rtl8723bs: Remove rtw_btcoex_SetBTCoexist()
staging: rtl8723bs: Remove rtw_btcoex_IsBtDisabled()
staging: rtl8723bs: Remove rtw_btcoex_IsBtControlLps()
...
Merge tag 'char-misc-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
Pull char / misc driver updates from Greg KH:
"Here is the "large" pull request for char and misc and other assorted
smaller driver subsystems for 5.3-rc1.
It seems that this tree is becoming the funnel point of lots of
smaller driver subsystems, which is fine for me, but that's why it is
getting larger over time and does not just contain stuff under
drivers/char/ and drivers/misc.
Lots of small updates all over the place here from different driver
subsystems:
- habana driver updates
- coresight driver updates
- documentation file movements and updates
- Android binder fixes and updates
- extcon driver updates
- google firmware driver updates
- fsi driver updates
- smaller misc and char driver updates
- soundwire driver updates
- nvmem driver updates
- w1 driver fixes
All of these have been in linux-next for a while with no reported
issues"
* tag 'char-misc-5.3-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (188 commits)
coresight: Do not default to CPU0 for missing CPU phandle
dt-bindings: coresight: Change CPU phandle to required property
ocxl: Allow contexts to be attached with a NULL mm
fsi: sbefifo: Don't fail operations when in SBE IPL state
coresight: tmc: Smatch: Fix potential NULL pointer dereference
coresight: etm3x: Smatch: Fix potential NULL pointer dereference
coresight: Potential uninitialized variable in probe()
coresight: etb10: Do not call smp_processor_id from preemptible
coresight: tmc-etf: Do not call smp_processor_id from preemptible
coresight: tmc-etr: alloc_perf_buf: Do not call smp_processor_id from preemptible
coresight: tmc-etr: Do not call smp_processor_id() from preemptible
docs: misc-devices: convert files without extension to ReST
fpga: dfl: fme: align PR buffer size per PR datawidth
fpga: dfl: fme: remove copy_to_user() in ioctl for PR
fpga: dfl-fme-mgr: fix FME_PR_INTFC_ID register address.
intel_th: msu: Start read iterator from a non-empty window
intel_th: msu: Split sgt array and pointer in multiwindow mode
intel_th: msu: Support multipage blocks
intel_th: pci: Add Ice Lake NNPI support
intel_th: msu: Fix single mode with disabled IOMMU
...
projects / linux.git / log