Peter Xu [Thu, 1 Jul 2021 01:49:13 +0000 (18:49 -0700)]
userfaultfd/selftests: add pagemap uffd-wp test
Add one anonymous specific test to start using pagemap. With pagemap
support, we can directly read the uffd-wp bit from pgtable without
triggering any fault, so it's easier to do sanity checks in unit tests.
Meanwhile this test also leverages the newly introduced MADV_PAGEOUT
madvise function to test swap ptes with uffd-wp bit set, and across
fork()s.
Peter Xu [Thu, 1 Jul 2021 01:49:06 +0000 (18:49 -0700)]
mm/userfaultfd: fail uffd-wp registration if not supported
We should fail uffd-wp registration immediately if the arch does not even
have CONFIG_HAVE_ARCH_USERFAULTFD_WP defined. That'll block also relevant
ioctls on e.g. UFFDIO_WRITEPROTECT because that'll check against
VM_UFFD_WP, which can only be applied with a success registration.
Remove the WP feature bit too for those archs when handling UFFDIO_API
ioctl.
Peter Xu [Thu, 1 Jul 2021 01:49:02 +0000 (18:49 -0700)]
mm/userfaultfd: fix uffd-wp special cases for fork()
We tried to do something similar in b569a1760782 ("userfaultfd: wp: drop
_PAGE_UFFD_WP properly when fork") previously, but it's not doing it all
right.. A few fixes around the code path:
1. We were referencing VM_UFFD_WP vm_flags on the _old_ vma rather
than the new vma. That's overlooked in b569a1760782, so it won't work
as expected. Thanks to the recent rework on fork code
(7a4830c380f3a8b3), we can easily get the new vma now, so switch the
checks to that.
2. Dropping the uffd-wp bit in copy_huge_pmd() could be wrong if the
huge pmd is a migration huge pmd. When it happens, instead of using
pmd_uffd_wp(), we should use pmd_swp_uffd_wp(). The fix is simply to
handle them separately.
3. Forget to carry over uffd-wp bit for a write migration huge pmd
entry. This also happens in copy_huge_pmd(), where we converted a
write huge migration entry into a read one.
4. In copy_nonpresent_pte(), drop uffd-wp if necessary for swap ptes.
5. In copy_present_page() when COW is enforced when fork(), we also
need to pass over the uffd-wp bit if VM_UFFD_WP is armed on the new
vma, and when the pte to be copied has uffd-wp bit set.
Remove the comment in copy_present_pte() about this. It won't help a huge
lot to only comment there, but comment everywhere would be an overkill.
Let's assume the commit messages would help.
Peter Xu [Thu, 1 Jul 2021 01:48:59 +0000 (18:48 -0700)]
mm/thp: simplify copying of huge zero page pmd when fork
Patch series "mm/uffd: Misc fix for uffd-wp and one more test".
This series tries to fix some corner case bugs for uffd-wp on either thp
or fork(). Then it introduced a new test with pagemap/pageout.
Patch layout:
Patch 1: cleanup for THP, it'll slightly simplify the follow up patches
Patch 2-4: misc fixes for uffd-wp here and there; please refer to each patch
Patch 5: add pagemap support for uffd-wp
Patch 6: add pagemap/pageout test for uffd-wp
The last test introduced can also verify some of the fixes in previous
patches, as the test will fail without the fixes. However it's not easy
to verify all the changes in patch 2-4, but hopefully they can still be
properly reviewed.
Note that if considering the ongoing uffd-wp shmem & hugetlbfs work, patch
5 will be incomplete as it's missing e.g. hugetlbfs part or the special
swap pte detection. However that's not needed in this series, and since
that series is still during review, this series does not depend on that
one (the last test only runs with anonymous memory, not file-backed). So
this series can be merged even before that series.
This patch (of 6):
Huge zero page is handled in a special path in copy_huge_pmd(), however it
should share most codes with a normal thp page. Trying to share more code
with it by removing the special path. The only leftover so far is the
huge zero page refcounting (mm_get_huge_zero_page()), because that's
separately done with a global counter.
This prepares for a future patch to modify the huge pmd to be installed,
so that we don't need to duplicate it explicitly into huge zero page case
too.
Peter Xu [Thu, 1 Jul 2021 01:48:55 +0000 (18:48 -0700)]
userfaultfd/selftests: unify error handling
Introduce err()/_err() and replace all the different ways to fail the
program, mostly "fprintf" and "perror" with tons of exit() calls. Always
stop the test program at any failure.
Peter Xu [Thu, 1 Jul 2021 01:48:52 +0000 (18:48 -0700)]
userfaultfd/selftests: only dump counts if mode enabled
WP and MINOR modes are conditionally enabled on specific memory types.
This patch avoids dumping tons of zeros for those cases when the modes are
not supported at all.
Peter Xu [Thu, 1 Jul 2021 01:48:48 +0000 (18:48 -0700)]
userfaultfd/selftests: dropping VERIFY check in locking_thread
It tries to check against all zeros and looped for quite a few times.
However after that we'll verify the same page with count_verify, while
count_verify can never be zero. So it means if it's a zero page we'll
detect it anyways with below code.
There's yet another place we conditionally check the fault flag - just do
it unconditionally.
Peter Xu [Thu, 1 Jul 2021 01:48:44 +0000 (18:48 -0700)]
userfaultfd/selftests: remove the time() check on delayed uffd
There seems to have no guarantee that time() will return the same for the
two calls even if there's no delay, e.g. when a fault is accidentally
crossing the changing of a second. Meanwhile, this message is also not
helping that much since delay could happen with a lot of reasons, e.g.,
schedule latency of resolving thread. It may not mean an issue with uffd.
Neither do I saw this error triggered either in the past runs. Even if it
triggers, it'll be drown in all the rest of test logs. Remove it.
Peter Xu [Thu, 1 Jul 2021 01:48:41 +0000 (18:48 -0700)]
userfaultfd/selftests: use user mode only
Patch series "userfaultfd/selftests: A few cleanups", v2.
I wanted to cleanup userfaultfd.c fault handling for a long time. If it's
not cleaned, when the new code grows the file it'll also grow the size
that needs to be cleaned... This is my attempt to cleanup the userfaultfd
selftest on fault handling, to use an err() macro instead of either
fprintf() or perror() then another exit() call.
The huge cleanup is done in the last patch. The first 4 patches are some
other standalone cleanups for the same file, so I put them together.
This patch (of 5):
Userfaultfd selftest does not need to handle kernel initiated fault. Set
user mode so it can be run even if unprivileged_userfaultfd=0 (which is
the default).
Recent changes by patch "mm/page_alloc: allow high-order pages to be
stored on the per-cpu lists" makes kernels determine whether to use pcp by
pcp_allowed_order(), which breaks soft-offline for hugetlb pages.
Soft-offline dissolves a migration source page, then removes it from buddy
free list, so it's assumed that any subpage of the soft-offlined hugepage
are recognized as a buddy page just after returning from
dissolve_free_huge_page(). pcp_allowed_order() returns true for hugetlb,
so this assumption is no longer true.
So disable pcp during dissolve_free_huge_page() and take_page_off_buddy()
to prevent soft-offlined hugepages from linking to pcp lists.
Soft-offline should not be common events so the impact on performance
should be minimal. And I think that the optimization of Mel's patch could
benefit to hugetlb so zone_pcp_disable() is called only in hwpoison
context.
Mike Kravetz [Thu, 1 Jul 2021 01:48:34 +0000 (18:48 -0700)]
hugetlb: address ref count racing in prep_compound_gigantic_page
In [1], Jann Horn points out a possible race between
prep_compound_gigantic_page and __page_cache_add_speculative. The root
cause of the possible race is prep_compound_gigantic_page uncondittionally
setting the ref count of pages to zero. It does this because
prep_compound_gigantic_page is handed a 'group' of pages from an allocator
and needs to convert that group of pages to a compound page. The ref
count of each page in this 'group' is one as set by the allocator.
However, the ref count of compound page tail pages must be zero.
The potential race comes about when ref counted pages are returned from
the allocator. When this happens, other mm code could also take a
reference on the page. __page_cache_add_speculative is one such example.
Therefore, prep_compound_gigantic_page can not just set the ref count of
pages to zero as it does today. Doing so would lose the reference taken
by any other code. This would lead to BUGs in code checking ref counts
and could possibly even lead to memory corruption.
There are two possible ways to address this issue.
1) Make all allocators of gigantic groups of pages be able to return a
properly constructed compound page.
2) Make prep_compound_gigantic_page be more careful when constructing a
compound page.
This patch takes approach 2.
In prep_compound_gigantic_page, use cmpxchg to only set ref count to zero
if it is one. If the cmpxchg fails, call synchronize_rcu() in the hope
that the extra ref count will be driopped during a rcu grace period. This
is not a performance critical code path and the wait should be
accceptable. If the ref count is still inflated after the grace period,
then undo any modifications made and return an error.
Currently prep_compound_gigantic_page is type void and does not return
errors. Modify the two callers to check for and handle error returns. On
error, the caller must free the 'group' of pages as they can not be used
to form a gigantic page. After freeing pages, the runtime caller
(alloc_fresh_huge_page) will retry the allocation once. Boot time
allocations can not be retried.
The routine prep_compound_page also unconditionally sets the ref count of
compound page tail pages to zero. However, in this case the buddy
allocator is constructing a compound page from freshly allocated pages.
The ref count on those freshly allocated pages is already zero, so the
set_page_count(p, 0) is unnecessary and could lead to confusion. Just
remove it.
Mike Kravetz [Thu, 1 Jul 2021 01:48:31 +0000 (18:48 -0700)]
hugetlb: remove prep_compound_huge_page cleanup
Patch series "Fix prep_compound_gigantic_page ref count adjustment".
These patches address the possible race between
prep_compound_gigantic_page and __page_cache_add_speculative as described
by Jann Horn in [1].
The first patch simply removes the unnecessary/obsolete helper routine
prep_compound_huge_page to make the actual fix a little simpler.
The second patch is the actual fix and has a detailed explanation in the
commit message.
This potential issue has existed for almost 10 years and I am unaware of
anyone actually hitting the race. I did not cc stable, but would be happy
to squash the patches and send to stable if anyone thinks that is a good
idea.
I could not think of a reliable way to recreate the issue for testing.
Rather, I 'simulated errors' to exercise all the error paths.
The routine prep_compound_huge_page is a simple wrapper to call either
prep_compound_gigantic_page or prep_compound_page. However, it is only
called from gather_bootmem_prealloc which only processes gigantic pages.
Eliminate the routine and call prep_compound_gigantic_page directly.
When using HUGETLB_PAGE_FREE_VMEMMAP, the freeing unused vmemmap pages
associated with each HugeTLB page is default off. Now the vmemmap is PMD
mapped. So there is no side effect when this feature is enabled with no
HugeTLB pages in the system. Someone may want to enable this feature in
the compiler time instead of using boot command line. So add a config to
make it default on when someone do not want to enable it via command line.
Muchun Song [Thu, 1 Jul 2021 01:48:22 +0000 (18:48 -0700)]
mm: sparsemem: split the huge PMD mapping of vmemmap pages
Patch series "Split huge PMD mapping of vmemmap pages", v4.
In order to reduce the difficulty of code review in series[1]. We disable
huge PMD mapping of vmemmap pages when that feature is enabled. In this
series, we do not disable huge PMD mapping of vmemmap pages anymore. We
will split huge PMD mapping when needed. When HugeTLB pages are freed
from the pool we do not attempt coalasce and move back to a PMD mapping
because it is much more complex.
In [1], PMD mappings of vmemmap pages were disabled if the the feature
hugetlb_free_vmemmap was enabled. This was done to simplify the initial
implementation of vmmemap freeing for hugetlb pages. Now, remove this
simplification by allowing PMD mapping and switching to PTE mappings as
needed for allocated hugetlb pages.
When a hugetlb page is allocated, the vmemmap page tables are walked to
free vmemmap pages. During this walk, split huge PMD mappings to PTE
mappings as required. In the unlikely case PTE pages can not be
allocated, return error(ENOMEM) and do not optimize vmemmap of the hugetlb
page.
When HugeTLB pages are freed from the pool, we do not attempt to
coalesce and move back to a PMD mapping because it is much more complex.
Mina Almasry [Thu, 1 Jul 2021 01:48:19 +0000 (18:48 -0700)]
mm, hugetlb: fix racy resv_huge_pages underflow on UFFDIO_COPY
On UFFDIO_COPY, if we fail to copy the page contents while holding the
hugetlb_fault_mutex, we will drop the mutex and return to the caller after
allocating a page that consumed a reservation. In this case there may be
a fault that double consumes the reservation. To handle this, we free the
allocated page, fix the reservations, and allocate a temporary hugetlb
page and return that to the caller. When the caller does the copy outside
of the lock, we again check the cache, and allocate a page consuming the
reservation, and copy over the contents.
Test:
Hacked the code locally such that resv_huge_pages underflows produce
a warning and the copy_huge_page_from_user() always fails, then:
./tools/testing/selftests/vm/userfaultfd hugetlb_shared 10
2 /tmp/kokonut_test/huge/userfaultfd_test && echo test success
./tools/testing/selftests/vm/userfaultfd hugetlb 10
2 /tmp/kokonut_test/huge/userfaultfd_test && echo test success
Both tests succeed and produce no warnings. After the
test runs number of free/resv hugepages is correct.
Nanyong Sun [Thu, 1 Jul 2021 01:48:16 +0000 (18:48 -0700)]
khugepaged: selftests: remove debug_cow
The debug_cow attribute had been removed since commit 4958e4d86ecb01 ("mm:
thp: remove debug_cow switch"), so remove it in selftest code too,
otherwise the khugepaged test will fail.
At the time being, vmalloc and vmap only support huge pages which are
leaf at PMD level.
Here the PMD level is 4M, it doesn't correspond to any supported
page size.
For now, implement use of 16k and 512k pages which is done
at PTE level.
Support of 8M pages will be implemented later, it requires use of
hugepd tables.
To allow this, the architecture provides two functions:
- arch_vmap_pte_range_map_size() which tells vmap_pte_range() what
page size to use. A stub returning PAGE_SIZE is provided when the
architecture doesn't provide this function.
- arch_vmap_pte_supported_shift() which tells __vmalloc_node_range()
what page shift to use for a given area size. A stub returning
PAGE_SHIFT is provided when the architecture doesn't provide this
function.
This patch (of 5):
At the time being, arch_make_huge_pte() has the following prototype:
Miaohe Lin [Thu, 1 Jul 2021 01:47:57 +0000 (18:47 -0700)]
mm/huge_memory.c: don't discard hugepage if other processes are mapping it
If other processes are mapping any other subpages of the hugepage, i.e.
in pte-mapped thp case, page_mapcount() will return 1 incorrectly. Then
we would discard the page while other processes are still mapping it. Fix
it by using total_mapcount() which can tell whether other processes are
still mapping it.
Miaohe Lin [Thu, 1 Jul 2021 01:47:53 +0000 (18:47 -0700)]
mm/huge_memory.c: remove unnecessary tlb_remove_page_size() for huge zero pmd
Commit aa88b68c3b1d ("thp: keep huge zero page pinned until tlb flush")
introduced tlb_remove_page() for huge zero page to keep it pinned until
flush is complete and prevents the page from being split under us. But
huge zero page is kept pinned until all relevant mm_users reach zero since
the commit 6fcb52a56ff6 ("thp: reduce usage of huge zero page's atomic
counter"). So tlb_remove_page_size() for huge zero pmd is unnecessary
now.
Miaohe Lin [Thu, 1 Jul 2021 01:47:50 +0000 (18:47 -0700)]
mm/huge_memory.c: add missing read-only THP checking in transparent_hugepage_enabled()
Since commit 99cb0dbd47a1 ("mm,thp: add read-only THP support for
(non-shmem) FS"), read-only THP file mapping is supported. But it forgot
to add checking for it in transparent_hugepage_enabled(). To fix it, we
add checking for read-only THP file mapping and also introduce helper
transhuge_vma_enabled() to check whether thp is enabled for specified vma
to reduce duplicated code. We rename transparent_hugepage_enabled to
transparent_hugepage_active to make the code easier to follow as suggested
by David Hildenbrand.
Patch series "Cleanup and fixup for huge_memory:, v3.
This series contains cleanups to remove dedicated macro and remove
unnecessary tlb_remove_page_size() for huge zero pmd. Also this adds
missing read-only THP checking for transparent_hugepage_enabled() and
avoids discarding hugepage if other processes are mapping it. More
details can be found in the respective changelogs.
Thi patch (of 5):
Rewrite the pgoff checking logic to remove macro HPAGE_CACHE_INDEX_MASK
which is only used here to simplify the code.
Shixin Liu [Thu, 1 Jul 2021 01:47:40 +0000 (18:47 -0700)]
mm/debug_vm_pgtable: remove redundant pfn_{pmd/pte}() and fix one comment mistake
Remove redundant pfn_{pmd/pte}() in {pmd/pte}_advanced_tests() and adjust
pfn_pud() in pud_advanced_tests() to make it similar with other two
functions.
In addition, the branch condition should be CONFIG_TRANSPARENT_HUGEPAGE
instead of CONFIG_ARCH_HAS_PTE_DEVMAP.
Shixin Liu [Thu, 1 Jul 2021 01:47:37 +0000 (18:47 -0700)]
mm/debug_vm_pgtable: move {pmd/pud}_huge_tests out of CONFIG_TRANSPARENT_HUGEPAGE
The functions {pmd/pud}_set_huge and {pmd/pud}_clear_huge are not
dependent on THP. Hence move {pmd/pud}_huge_tests out of
CONFIG_TRANSPARENT_HUGEPAGE.
Muchun Song [Thu, 1 Jul 2021 01:47:33 +0000 (18:47 -0700)]
mm: hugetlb: introduce nr_free_vmemmap_pages in the struct hstate
All the infrastructure is ready, so we introduce nr_free_vmemmap_pages
field in the hstate to indicate how many vmemmap pages associated with a
HugeTLB page that can be freed to buddy allocator. And initialize it in
the hugetlb_vmemmap_init(). This patch is actual enablement of the
feature.
There are only (RESERVE_VMEMMAP_SIZE / sizeof(struct page)) struct page
structs that can be used when CONFIG_HUGETLB_PAGE_FREE_VMEMMAP, so add a
BUILD_BUG_ON to catch invalid usage of the tail struct page.
Muchun Song [Thu, 1 Jul 2021 01:47:25 +0000 (18:47 -0700)]
mm: hugetlb: add a kernel parameter hugetlb_free_vmemmap
Add a kernel parameter hugetlb_free_vmemmap to enable the feature of
freeing unused vmemmap pages associated with each hugetlb page on boot.
We disable PMD mapping of vmemmap pages for x86-64 arch when this feature
is enabled. Because vmemmap_remap_free() depends on vmemmap being base
page mapped.
Muchun Song [Thu, 1 Jul 2021 01:47:21 +0000 (18:47 -0700)]
mm: hugetlb: alloc the vmemmap pages associated with each HugeTLB page
When we free a HugeTLB page to the buddy allocator, we need to allocate
the vmemmap pages associated with it. However, we may not be able to
allocate the vmemmap pages when the system is under memory pressure. In
this case, we just refuse to free the HugeTLB page. This changes behavior
in some corner cases as listed below:
1) Failing to free a huge page triggered by the user (decrease nr_pages).
User needs to try again later.
2) Failing to free a surplus huge page when freed by the application.
Try again later when freeing a huge page next time.
3) Failing to dissolve a free huge page on ZONE_MOVABLE via
offline_pages().
This can happen when we have plenty of ZONE_MOVABLE memory, but
not enough kernel memory to allocate vmemmmap pages. We may even
be able to migrate huge page contents, but will not be able to
dissolve the source huge page. This will prevent an offline
operation and is unfortunate as memory offlining is expected to
succeed on movable zones. Users that depend on memory hotplug
to succeed for movable zones should carefully consider whether the
memory savings gained from this feature are worth the risk of
possibly not being able to offline memory in certain situations.
4) Failing to dissolve a huge page on CMA/ZONE_MOVABLE via
alloc_contig_range() - once we have that handling in place. Mainly
affects CMA and virtio-mem.
Similar to 3). virito-mem will handle migration errors gracefully.
CMA might be able to fallback on other free areas within the CMA
region.
Vmemmap pages are allocated from the page freeing context. In order for
those allocations to be not disruptive (e.g. trigger oom killer)
__GFP_NORETRY is used. hugetlb_lock is dropped for the allocation because
a non sleeping allocation would be too fragile and it could fail too
easily under memory pressure. GFP_ATOMIC or other modes to access memory
reserves is not used because we want to prevent consuming reserves under
heavy hugetlb freeing.
Muchun Song [Thu, 1 Jul 2021 01:47:17 +0000 (18:47 -0700)]
mm: hugetlb: defer freeing of HugeTLB pages
In the subsequent patch, we should allocate the vmemmap pages when freeing
a HugeTLB page. But update_and_free_page() can be called under any
context, so we cannot use GFP_KERNEL to allocate vmemmap pages. However,
we can defer the actual freeing in a kworker to prevent from using
GFP_ATOMIC to allocate the vmemmap pages.
The __update_and_free_page() is where the call to allocate vmemmmap pages
will be inserted.
Muchun Song [Thu, 1 Jul 2021 01:47:13 +0000 (18:47 -0700)]
mm: hugetlb: free the vmemmap pages associated with each HugeTLB page
Every HugeTLB has more than one struct page structure. We __know__ that
we only use the first 4 (__NR_USED_SUBPAGE) struct page structures to
store metadata associated with each HugeTLB.
There are a lot of struct page structures associated with each HugeTLB
page. For tail pages, the value of compound_head is the same. So we can
reuse first page of tail page structures. We map the virtual addresses of
the remaining pages of tail page structures to the first tail page struct,
and then free these page frames. Therefore, we need to reserve two pages
as vmemmap areas.
When we allocate a HugeTLB page from the buddy, we can free some vmemmap
pages associated with each HugeTLB page. It is more appropriate to do it
in the prep_new_huge_page().
The free_vmemmap_pages_per_hpage(), which indicates how many vmemmap pages
associated with a HugeTLB page can be freed, returns zero for now, which
means the feature is disabled. We will enable it once all the
infrastructure is there.
Muchun Song [Thu, 1 Jul 2021 01:47:09 +0000 (18:47 -0700)]
mm: hugetlb: gather discrete indexes of tail page
For HugeTLB page, there are more metadata to save in the struct page. But
the head struct page cannot meet our needs, so we have to abuse other tail
struct page to store the metadata. In order to avoid conflicts caused by
subsequent use of more tail struct pages, we can gather these discrete
indexes of tail struct page. In this case, it will be easier to add a new
tail page index later.
Muchun Song [Thu, 1 Jul 2021 01:47:04 +0000 (18:47 -0700)]
mm: hugetlb: introduce a new config HUGETLB_PAGE_FREE_VMEMMAP
The option HUGETLB_PAGE_FREE_VMEMMAP allows for the freeing of some
vmemmap pages associated with pre-allocated HugeTLB pages. For example,
on X86_64 6 vmemmap pages of size 4KB each can be saved for each 2MB
HugeTLB page. 4094 vmemmap pages of size 4KB each can be saved for each
1GB HugeTLB page.
When a HugeTLB page is allocated or freed, the vmemmap array representing
the range associated with the page will need to be remapped. When a page
is allocated, vmemmap pages are freed after remapping. When a page is
freed, previously discarded vmemmap pages must be allocated before
remapping.
The config option is introduced early so that supporting code can be
written to depend on the option. The initial version of the code only
provides support for x86-64.
If config HAVE_BOOTMEM_INFO_NODE is enabled, the freeing vmemmap page code
denpend on it to free vmemmap pages. Otherwise, just use
free_reserved_page() to free vmemmmap pages. The routine
register_page_bootmem_info() is used to register bootmem info. Therefore,
make sure register_page_bootmem_info is enabled if
HUGETLB_PAGE_FREE_VMEMMAP is defined.
Muchun Song [Thu, 1 Jul 2021 01:47:00 +0000 (18:47 -0700)]
mm: memory_hotplug: factor out bootmem core functions to bootmem_info.c
Patch series "Free some vmemmap pages of HugeTLB page", v23.
This patch series will free some vmemmap pages(struct page structures)
associated with each HugeTLB page when preallocated to save memory.
In order to reduce the difficulty of the first version of code review. In
this version, we disable PMD/huge page mapping of vmemmap if this feature
was enabled. This acutely eliminates a bunch of the complex code doing
page table manipulation. When this patch series is solid, we cam add the
code of vmemmap page table manipulation in the future.
The struct page structures (page structs) are used to describe a physical
page frame. By default, there is an one-to-one mapping from a page frame
to it's corresponding page struct.
The HugeTLB pages consist of multiple base page size pages and is
supported by many architectures. See hugetlbpage.rst in the Documentation
directory for more details. On the x86 architecture, HugeTLB pages of
size 2MB and 1GB are currently supported. Since the base page size on x86
is 4KB, a 2MB HugeTLB page consists of 512 base pages and a 1GB HugeTLB
page consists of 4096 base pages. For each base page, there is a
corresponding page struct.
Within the HugeTLB subsystem, only the first 4 page structs are used to
contain unique information about a HugeTLB page. HUGETLB_CGROUP_MIN_ORDER
provides this upper limit. The only 'useful' information in the remaining
page structs is the compound_head field, and this field is the same for
all tail pages.
By removing redundant page structs for HugeTLB pages, memory can returned
to the buddy allocator for other uses.
When the system boot up, every 2M HugeTLB has 512 struct page structs which
size is 8 pages(sizeof(struct page) * 512 / PAGE_SIZE).
The value of page->compound_head is the same for all tail pages. The
first page of page structs (page 0) associated with the HugeTLB page
contains the 4 page structs necessary to describe the HugeTLB. The only
use of the remaining pages of page structs (page 1 to page 7) is to point
to page->compound_head. Therefore, we can remap pages 2 to 7 to page 1.
Only 2 pages of page structs will be used for each HugeTLB page. This
will allow us to free the remaining 6 pages to the buddy allocator.
When a HugeTLB is freed to the buddy system, we should allocate 6 pages
for vmemmap pages and restore the previous mapping relationship.
Apart from 2MB HugeTLB page, we also have 1GB HugeTLB page. It is similar
to the 2MB HugeTLB page. We also can use this approach to free the
vmemmap pages.
In this case, for the 1GB HugeTLB page, we can save 4094 pages. This is a
very substantial gain. On our server, run some SPDK/QEMU applications
which will use 1024GB HugeTLB page. With this feature enabled, we can
save ~16GB (1G hugepage)/~12GB (2MB hugepage) memory.
Because there are vmemmap page tables reconstruction on the
freeing/allocating path, it increases some overhead. Here are some
overhead analysis.
1) Allocating 10240 2MB HugeTLB pages.
a) With this patch series applied:
# time echo 10240 > /proc/sys/vm/nr_hugepages
Summary: The overhead of __free_hugepage is about ~2-3x slower than before.
Although the overhead has increased, the overhead is not significant.
Like Mike said, "However, remember that the majority of use cases create
HugeTLB pages at or shortly after boot time and add them to the pool. So,
additional overhead is at pool creation time. There is no change to
'normal run time' operations of getting a page from or returning a page to
the pool (think page fault/unmap)".
Despite the overhead and in addition to the memory gains from this series.
The following data is obtained by Joao Martins. Very thanks to his
effort.
There's an additional benefit which is page (un)pinners will see an improvement
and Joao presumes because there are fewer memmap pages and thus the tail/head
pages are staying in cache more often.
Out of the box Joao saw (when comparing linux-next against linux-next +
this series) with gup_test and pinning a 16G HugeTLB file (with 1G pages):
Usually any tight loop fetching compound_head(), or reading tail pages
data (e.g. compound_head) benefit a lot. There's some unpinning
inefficiencies Joao was fixing[2], but with that in added it shows even
more:
Move bootmem info registration common API to individual bootmem_info.c.
And we will use {get,put}_page_bootmem() to initialize the page for the
vmemmap pages or free the vmemmap pages to buddy in the later patch. So
move them out of CONFIG_MEMORY_HOTPLUG_SPARSE. This is just code movement
without any functional change.
Merge tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4
Pull ext4 updates from Ted Ts'o:
"In addition to bug fixes and cleanups, there are two new features for
ext4 in 5.14:
- Allow applications to poll on changes to
/sys/fs/ext4/*/errors_count
- Add the ioctl EXT4_IOC_CHECKPOINT which allows the journal to be
checkpointed, truncated and discarded or zero'ed"
* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (32 commits)
jbd2: export jbd2_journal_[un]register_shrinker()
ext4: notify sysfs on errors_count value change
fs: remove bdev_try_to_free_page callback
ext4: remove bdev_try_to_free_page() callback
jbd2: simplify journal_clean_one_cp_list()
jbd2,ext4: add a shrinker to release checkpointed buffers
jbd2: remove redundant buffer io error checks
jbd2: don't abort the journal when freeing buffers
jbd2: ensure abort the journal if detect IO error when writing original buffer back
jbd2: remove the out label in __jbd2_journal_remove_checkpoint()
ext4: no need to verify new add extent block
jbd2: clean up misleading comments for jbd2_fc_release_bufs
ext4: add check to prevent attempting to resize an fs with sparse_super2
ext4: consolidate checks for resize of bigalloc into ext4_resize_begin
ext4: remove duplicate definition of ext4_xattr_ibody_inline_set()
ext4: fsmap: fix the block/inode bitmap comment
ext4: fix comment for s_hash_unsigned
ext4: use local variable ei instead of EXT4_I() macro
ext4: fix avefreec in find_group_orlov
ext4: correct the cache_nr in tracepoint ext4_es_shrink_exit
...
Merge tag 'for-5.14/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm
Pull device mapper updates from Mike Snitzer:
- Various DM persistent-data library improvements and fixes that
benefit both the DM thinp and cache targets.
- A few small DM kcopyd efficiency improvements.
- Significant zoned related block core, DM core and DM zoned target
changes that culminate with adding zoned append emulation (which is
required to properly fix DM crypt's zoned support).
- Various DM writecache target changes that improve efficiency. Adds an
optional "metadata_only" feature that only promotes bios flagged with
REQ_META. But the most significant improvement is writecache's
ability to pause writeback, for a confiurable time, if/when the
working set is larger than the cache (and the cache is full) -- this
ensures performance is no worse than the slower origin device.
* tag 'for-5.14/dm-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: (35 commits)
dm writecache: make writeback pause configurable
dm writecache: pause writeback if cache full and origin being written directly
dm io tracker: factor out IO tracker
dm btree remove: assign new_root only when removal succeeds
dm zone: fix dm_revalidate_zones() memory allocation
dm ps io affinity: remove redundant continue statement
dm writecache: add optional "metadata_only" parameter
dm writecache: add "cleaner" and "max_age" to Documentation
dm writecache: write at least 4k when committing
dm writecache: flush origin device when writing and cache is full
dm writecache: have ssd writeback wait if the kcopyd workqueue is busy
dm writecache: use list_move instead of list_del/list_add in writecache_writeback()
dm writecache: commit just one block, not a full page
dm writecache: remove unused gfp_t argument from wc_add_block()
dm crypt: Fix zoned block device support
dm: introduce zone append emulation
dm: rearrange core declarations for extended use from dm-zone.c
block: introduce BIO_ZONE_WRITE_LOCKED bio flag
block: introduce bio zone helpers
block: improve handling of all zones reset operation
...
Linus Torvalds [Wed, 30 Jun 2021 22:51:09 +0000 (15:51 -0700)]
Merge tag 'net-next-5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next
Pull networking updates from Jakub Kicinski:
"Core:
- BPF:
- add syscall program type and libbpf support for generating
instructions and bindings for in-kernel BPF loaders (BPF loaders
for BPF), this is a stepping stone for signed BPF programs
- infrastructure to migrate TCP child sockets from one listener to
another in the same reuseport group/map to improve flexibility
of service hand-off/restart
- add broadcast support to XDP redirect
- allow bypass of the lockless qdisc to improving performance (for
pktgen: +23% with one thread, +44% with 2 threads)
- add a simpler version of "DO_ONCE()" which does not require jump
labels, intended for slow-path usage
- virtio/vsock: introduce SOCK_SEQPACKET support
- add getsocketopt to retrieve netns cookie
- ip: treat lowest address of a IPv4 subnet as ordinary unicast
address allowing reclaiming of precious IPv4 addresses
- ipv6: use prandom_u32() for ID generation
- ip: add support for more flexible field selection for hashing
across multi-path routes (w/ offload to mlxsw)
- icmp: add support for extended RFC 8335 PROBE (ping)
- seg6: add support for SRv6 End.DT46 behavior
- mptcp:
- DSS checksum support (RFC 8684) to detect middlebox meddling
- support Connection-time 'C' flag
- time stamping support
- sctp: packetization Layer Path MTU Discovery (RFC 8899)
- xfrm: speed up state addition with seq set
- WiFi:
- hidden AP discovery on 6 GHz and other HE 6 GHz improvements
- aggregation handling improvements for some drivers
- minstrel improvements for no-ack frames
- deferred rate control for TXQs to improve reaction times
- switch from round robin to virtual time-based airtime scheduler
- Google vNIC (gve): new ring/descriptor format support
- Qualcomm mobile (rmnet & ipa): inline checksum offload support
- MediaTek WiFi (mt76)
- mt7915 MSI support
- mt7915 Tx status reporting
- mt7915 thermal sensors support
- mt7921 decapsulation offload
- mt7921 enable runtime pm and deep sleep
- Realtek WiFi (rtw88)
- beacon filter support
- Tx antenna path diversity support
- firmware crash information via devcoredump
- Qualcomm WiFi (wcn36xx)
- Wake-on-WLAN support with magic packets and GTK rekeying
- Micrel PHY (ksz886x/ksz8081): add cable test support"
* tag 'net-next-5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (2168 commits)
tcp: change ICSK_CA_PRIV_SIZE definition
tcp_yeah: check struct yeah size at compile time
gve: DQO: Fix off by one in gve_rx_dqo()
stmmac: intel: set PCI_D3hot in suspend
stmmac: intel: Enable PHY WOL option in EHL
net: stmmac: option to enable PHY WOL with PMT enabled
net: say "local" instead of "static" addresses in ndo_dflt_fdb_{add,del}
net: use netdev_info in ndo_dflt_fdb_{add,del}
ptp: Set lookup cookie when creating a PTP PPS source.
net: sock: add trace for socket errors
net: sock: introduce sk_error_report
net: dsa: replay the local bridge FDB entries pointing to the bridge dev too
net: dsa: ensure during dsa_fdb_offload_notify that dev_hold and dev_put are on the same dev
net: dsa: include fdb entries pointing to bridge in the host fdb list
net: dsa: include bridge addresses which are local in the host fdb list
net: dsa: sync static FDB entries on foreign interfaces to hardware
net: dsa: install the host MDB and FDB entries in the master's RX filter
net: dsa: reference count the FDB addresses at the cross-chip notifier level
net: dsa: introduce a separate cross-chip notifier type for host FDBs
net: dsa: reference count the MDB entries at the cross-chip notifier level
...
Linus Torvalds [Wed, 30 Jun 2021 22:37:49 +0000 (15:37 -0700)]
Merge tag 'sched-urgent-2021-06-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Ingo Molnar:
- Fix a small inconsistency (bug) in load tracking, caught by a new
warning that several people reported.
- Flip CONFIG_SCHED_CORE to default-disabled, and update the Kconfig
help text.
* tag 'sched-urgent-2021-06-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/core: Disable CONFIG_SCHED_CORE by default
sched/fair: Ensure _sum and _avg values stay consistent
Linus Torvalds [Wed, 30 Jun 2021 22:22:05 +0000 (15:22 -0700)]
Merge tag 'audit-pr-20210629' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/audit
Pull audit updates from Paul Moore:
"Another merge window, another small audit pull request.
Four patches in total: one is cosmetic, one removes an unnecessary
initialization, one renames some enum values to prevent name
collisions, and one converts list_del()/list_add() to list_move().
None of these are earth shattering and all pass the audit-testsuite
tests while merging cleanly on top of your tree from earlier today"
* tag 'audit-pr-20210629' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/audit:
audit: remove unnecessary 'ret' initialization
audit: remove trailing spaces and tabs
audit: Use list_move instead of list_del/list_add
audit: Rename enum audit_state constants to avoid AUDIT_DISABLED redefinition
audit: add blank line after variable declarations
Linus Torvalds [Wed, 30 Jun 2021 21:55:42 +0000 (14:55 -0700)]
Merge tag 'selinux-pr-20210629' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux
Pull SELinux updates from Paul Moore:
- The slow_avc_audit() function is now non-blocking so we can remove
the AVC_NONBLOCKING tricks; this also includes the 'flags' variant of
avc_has_perm().
- Use kmemdup() instead of kcalloc()+copy when copying parts of the
SELinux policydb.
- The InfiniBand device name is now passed by reference when possible
in the SELinux code, removing a strncpy().
- Minor cleanups including: constification of avtab function args,
removal of useless LSM/XFRM function args, SELinux kdoc fixes, and
removal of redundant assignments.
* tag 'selinux-pr-20210629' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux:
selinux: kill 'flags' argument in avc_has_perm_flags() and avc_audit()
selinux: slow_avc_audit has become non-blocking
selinux: Fix kernel-doc
selinux: use __GFP_NOWARN with GFP_NOWAIT in the AVC
lsm_audit,selinux: pass IB device name by reference
selinux: Remove redundant assignment to rc
selinux: Corrected comment to match kernel-doc comment
selinux: delete selinux_xfrm_policy_lookup() useless argument
selinux: constify some avtab function arguments
selinux: simplify duplicate_policydb_cond_list() by using kmemdup()
Linus Torvalds [Wed, 30 Jun 2021 21:33:25 +0000 (14:33 -0700)]
Merge tag 'clang-features-v5.14-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux
Pull clang feature updates from Kees Cook:
- Add CC_HAS_NO_PROFILE_FN_ATTR in preparation for PGO support in the
face of the noinstr attribute, paving the way for PGO and fixing
GCOV. (Nick Desaulniers)
- x86_64 LTO coverage is expanded to 32-bit x86. (Nathan Chancellor)
- Small fixes to CFI. (Mark Rutland, Nathan Chancellor)
* tag 'clang-features-v5.14-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
qemu_fw_cfg: Make fw_cfg_rev_attr a proper kobj_attribute
Kconfig: Introduce ARCH_WANTS_NO_INSTR and CC_HAS_NO_PROFILE_FN_ATTR
compiler_attributes.h: cleanups for GCC 4.9+
compiler_attributes.h: define __no_profile, add to noinstr
x86, lto: Enable Clang LTO for 32-bit as well
CFI: Move function_nocfi() into compiler.h
MAINTAINERS: Add Clang CFI section
Hao Xu [Sun, 27 Jun 2021 21:37:30 +0000 (05:37 +0800)]
io_uring: spin in iopoll() only when reqs are in a single queue
We currently spin in iopoll() when requests to be iopolled are for
same file(device), while one device may have multiple hardware queues.
given an example:
hw_queue_0 | hw_queue_1
req(30us) req(10us)
If we first spin on iopolling for the hw_queue_0. the avg latency would
be (30us + 30us) / 2 = 30us. While if we do round robin, the avg
latency would be (30us + 10us) / 2 = 20us since we reap the request in
hw_queue_1 in time. So it's better to do spinning only when requests
are in same hardware queue.
Pavel Begunkov [Sat, 26 Jun 2021 20:40:49 +0000 (21:40 +0100)]
io_uring: pre-initialise some of req fields
Most of requests are allocated from an internal cache, so it's waste of
time fully initialising them every time. Instead, let's pre-init some of
the fields we can during initial allocation (e.g. kmalloc(), see
io_alloc_req()) and keep them valid on request recycling. There are four
of them in this patch:
->ctx is always stays the same
->link is NULL on free, it's an invariant
->result is not even needed to init, just a precaution
->async_data we now clean in io_dismantle_req() as it's likely to
never be allocated.
Pavel Begunkov [Sat, 26 Jun 2021 20:40:47 +0000 (21:40 +0100)]
io_uring: optimise hot path restricted checks
Move likely/unlikely from io_check_restriction() to specifically
ctx->restricted check, because doesn't do what it supposed to and make
the common path take an extra jump.
Pavel Begunkov [Sat, 26 Jun 2021 20:40:46 +0000 (21:40 +0100)]
io_uring: remove not needed PF_EXITING check
Since cancellation got moved before exit_signals(), there is no one left
who can call io_run_task_work() with PF_EXIING set, so remove the check.
Note that __io_req_task_submit() still needs a similar check.
Pavel Begunkov [Sat, 26 Jun 2021 20:40:45 +0000 (21:40 +0100)]
io_uring: mainstream sqpoll task_work running
task_works are widely used, so place io_run_task_work() directly into
the main path of io_sq_thread(), and remove it from other places where
it's not needed anymore.
Pavel Begunkov [Sat, 26 Jun 2021 20:40:44 +0000 (21:40 +0100)]
io_uring: refactor io_arm_poll_handler()
gcc 11 goes a weird path and duplicates most of io_arm_poll_handler()
for READ and WRITE cases. Help it and move all pollin vs pollout
specific bits under a single if-else, so there is no temptation for this
kind of unfolding.
before vs after:
text data bss dec hex filename
85362 12650 8 98020 17ee4 ./fs/io_uring.o
85186 12650 8 97844 17e34 ./fs/io_uring.o
Olivier Langlois [Tue, 22 Jun 2021 12:17:39 +0000 (05:17 -0700)]
io_uring: reduce latency by reissueing the operation
It is quite frequent that when an operation fails and returns EAGAIN,
the data becomes available between that failure and the call to
vfs_poll() done by io_arm_poll_handler().
Detecting the situation and reissuing the operation is much faster
than going ahead and push the operation to the io-wq.
Performance improvement testing has been performed with:
Single thread, 1 TCP connection receiving a 5 Mbps stream, no sqpoll.
4 measurements have been taken:
1. The time it takes to process a read request when data is already available
2. The time it takes to process by calling twice io_issue_sqe() after vfs_poll() indicated that data was available
3. The time it takes to execute io_queue_async_work()
4. The time it takes to complete a read request asynchronously
2.25% of all the read operations did use the new path.
async time completion
average 24670.4758861127
min 10758
max 102612
stddev 3483.92416873804
Conclusion:
On average reissuing the sqe with the patch code is 1.1uSec faster and
in the worse case scenario 59uSec faster than placing the request on
io-wq
On average completion time by reissuing the sqe with the patch code is
16.79uSec faster and in the worse case scenario 73.8uSec faster than
async completion.
Pavel Begunkov [Thu, 24 Jun 2021 14:09:55 +0000 (15:09 +0100)]
io_uring: don't change sqpoll creds if not needed
SQPOLL doesn't need to change creds if it's not submitting requests.
Move creds overriding into __io_sq_thread() after checking if there are
SQEs pending.
Linus Torvalds [Wed, 30 Jun 2021 19:21:16 +0000 (12:21 -0700)]
Merge tag 'for-5.14/drivers-2021-06-29' of git://git.kernel.dk/linux-block
Pull block driver updates from Jens Axboe:
"Pretty calm round, mostly just NVMe and a bit of MD:
- NVMe updates (via Christoph)
- improve the APST configuration algorithm (Alexey Bogoslavsky)
- look for StorageD3Enable on companion ACPI device
(Mario Limonciello)
- allow selecting the network interface for TCP connections
(Martin Belanger)
- misc cleanups (Amit Engel, Chaitanya Kulkarni, Colin Ian King,
Christoph)
- move the ACPI StorageD3 code to drivers/acpi/ and add quirks
for certain AMD CPUs (Mario Limonciello)
- zoned device support for nvmet (Chaitanya Kulkarni)
- fix the rules for changing the serial number in nvmet
(Noam Gottlieb)
- various small fixes and cleanups (Dan Carpenter, JK Kim,
Chaitanya Kulkarni, Hannes Reinecke, Wesley Sheng, Geert
Uytterhoeven, Daniel Wagner)
* tag 'for-5.14/drivers-2021-06-29' of git://git.kernel.dk/linux-block: (78 commits)
nvmet: use NVMET_MAX_NAMESPACES to set nn value
loop: Fix missing discard support when using LOOP_CONFIGURE
nvme.h: add missing nvme_lba_range_type endianness annotations
nvme: remove zeroout memset call for struct
nvme-pci: remove zeroout memset call for struct
nvmet: remove zeroout memset call for struct
nvmet: add ZBD over ZNS backend support
nvmet: add Command Set Identifier support
nvmet: add nvmet_req_bio put helper for backends
nvmet: add req cns error complete helper
block: export blk_next_bio()
nvmet: remove local variable
nvmet: use nvme status value directly
nvmet: use u32 type for the local variable nsid
nvmet: use u32 for nvmet_subsys max_nsid
nvmet: use req->cmd directly in file-ns fast path
nvmet: use req->cmd directly in bdev-ns fast path
nvmet: make ver stable once connection established
nvmet: allow mn change if subsys not discovered
nvmet: make sn stable once connection was established
...
Linus Torvalds [Wed, 30 Jun 2021 19:12:56 +0000 (12:12 -0700)]
Merge tag 'for-5.14/block-2021-06-29' of git://git.kernel.dk/linux-block
Pull core block updates from Jens Axboe:
- disk events cleanup (Christoph)
- gendisk and request queue allocation simplifications (Christoph)
- bdev_disk_changed cleanups (Christoph)
- IO priority improvements (Bart)
- Chained bio completion trace fix (Edward)
- blk-wbt fixes (Jan)
- blk-wbt enable/disable fix (Zhang)
- Scheduler dispatch improvements (Jan, Ming)
- Shared tagset scheduler improvements (John)
- BFQ updates (Paolo, Luca, Pietro)
- BFQ lock inversion fix (Jan)
- Documentation improvements (Kir)
- CLONE_IO block cgroup fix (Tejun)
- Remove of ancient and deprecated block dump feature (zhangyi)
- Discard merge fix (Ming)
- Misc fixes or followup fixes (Colin, Damien, Dan, Long, Max, Thomas,
Yang)
* tag 'for-5.14/block-2021-06-29' of git://git.kernel.dk/linux-block: (129 commits)
block: fix discard request merge
block/mq-deadline: Remove a WARN_ON_ONCE() call
blk-mq: update hctx->dispatch_busy in case of real scheduler
blk: Fix lock inversion between ioc lock and bfqd lock
bfq: Remove merged request already in bfq_requests_merged()
block: pass a gendisk to bdev_disk_changed
block: move bdev_disk_changed
block: add the events* attributes to disk_attrs
block: move the disk events code to a separate file
block: fix trace completion for chained bio
block/partitions/msdos: Fix typo inidicator -> indicator
block, bfq: reset waker pointer with shared queues
block, bfq: check waker only for queues with no in-flight I/O
block, bfq: avoid delayed merge of async queues
block, bfq: boost throughput by extending queue-merging times
block, bfq: consider also creation time in delayed stable merge
block, bfq: fix delayed stable merge check
block, bfq: let also stably merged queues enjoy weight raising
blk-wbt: make sure throttle is enabled properly
blk-wbt: introduce a new disable state to prevent false positive by rwb_enabled()
...
Linus Torvalds [Wed, 30 Jun 2021 18:31:32 +0000 (11:31 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid
Pull HID updates from Jiri Kosina:
- patch series that ensures that hid-multitouch driver disables touch
and button-press reporting on hid-mt devices during suspend when the
device is not configured as a wakeup-source, from Hans de Goede
- support for ISH DMA on Intel EHL platform, from Even Xu
- support for Renoir and Cezanne SoCs, Ambient Light Sensor and Human
Presence Detection sensor for amd-sfh driver, from Basavaraj Natikar
- other assorted code cleanups and device-specific fixes/quirks
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid: (45 commits)
HID: thrustmaster: Switch to kmemdup() when allocate change_request
HID: multitouch: Disable event reporting on suspend when the device is not a wakeup-source
HID: logitech-dj: Implement may_wakeup ll-driver callback
HID: usbhid: Implement may_wakeup ll-driver callback
HID: core: Add hid_hw_may_wakeup() function
HID: input: Add support for Programmable Buttons
HID: wacom: Correct base usage for capacitive ExpressKey status bits
HID: amd_sfh: Add initial support for HPD sensor
HID: amd_sfh: Extend ALS support for newer AMD platform
HID: amd_sfh: Extend driver capabilities for multi-generation support
HID: surface-hid: Fix get-report request
HID: sony: fix freeze when inserting ghlive ps3/wii dongles
HID: usbkbd: Avoid GFP_ATOMIC when GFP_KERNEL is possible
HID: amd_sfh: change in maintainer
HID: intel-ish-hid: ipc: Specify that EHL no cache snooping
HID: intel-ish-hid: ishtp: Add dma_no_cache_snooping() callback
HID: intel-ish-hid: Set ISH driver depends on x86
HID: hid-input: add Surface Go battery quirk
HID: intel-ish-hid: Fix minor typos in comments
HID: usbmouse: Avoid GFP_ATOMIC when GFP_KERNEL is possible
...
Alexey Bayduraev [Wed, 30 Jun 2021 15:54:48 +0000 (18:54 +0300)]
tools lib: Adopt bitmap_intersects() operation from the kernel sources
Adopt bitmap_intersects() routine that tests whether bitmaps bitmap1 and
bitmap2 intersects. This routine will be used during thread masks
initialization.
Linus Torvalds [Wed, 30 Jun 2021 18:27:49 +0000 (11:27 -0700)]
Merge tag 'edac_updates_for_v5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras
Pull EDAC updates from Tony Luck:
"Various fixes and support for new CPUs:
- Clean up error messages from thunderx_edac
- Add MODULE_DEVICE_TABLE to ti_edac so it will autoload
- Use %pR to print resources in aspeed_edac
- Add Yazen Ghannam as MAINTAINER for AMD edac drivers
- Fix Ice Lake and Sapphire Rapids drivers to report correct "near"
or "far" device for errors in 2LM configurations
- Add support of on package high bandwidth memory in Sapphire Rapids
- New CPU support for three CPUs supporting in-band ECC (IOT SKUs for
ICL-NNPI, Tiger Lake and Alder Lake)
- Don't even try to load Intel EDAC drivers when running as a guest
- Fix Kconfig dependency on X86_MCE_INTEL for EDAC_IGEN6"
* tag 'edac_updates_for_v5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
EDAC/igen6: fix core dependency
EDAC/Intel: Do not load EDAC driver when running as a guest
EDAC/igen6: Add Intel Alder Lake SoC support
EDAC/igen6: Add Intel Tiger Lake SoC support
EDAC/igen6: Add Intel ICL-NNPI SoC support
EDAC/i10nm: Add support for high bandwidth memory
EDAC/i10nm: Add detection of memory levels for ICX/SPR servers
EDAC/skx_common: Add new ADXL components for 2-level memory
MAINTAINERS: Make Yazen Ghannam maintainer for EDAC-AMD64
EDAC/aspeed: Use proper format string for printing resource
EDAC/ti: Add missing MODULE_DEVICE_TABLE
EDAC/thunderx: Remove irrelevant variable from error messages
Linus Torvalds [Wed, 30 Jun 2021 18:23:33 +0000 (11:23 -0700)]
Merge tag 'tpmdd-next-v5.14-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd
Pull tpm driver updates from Jarkko Sakkinen:
"Bug fixes for TPM"
[ This isn't actually the whole contents of the tag and thus doesn't
contain Jarkko's signature - I dropped the two top commits that added
support for signing modules using elliptic curve keys because there's
a new series for that that fixes a few confising things - Linus ]
* tag 'tpmdd-next-v5.14-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd:
tpm: Replace WARN_ONCE() with dev_err_once() in tpm_tis_status()
tpm_tis: Use DEFINE_RES_MEM() to simplify code
tpm: fix some doc warnings in tpm1-cmd.c
tpm_tis_spi: add missing SPI device ID entries
tpm: add longer timeout for TPM2_CC_VERIFY_SIGNATURE
char: tpm: move to use request_irq by IRQF_NO_AUTOEN flag
tpm_tis_spi: set default probe function if device id not match
tpm_crb: Use IOMEM_ERR_PTR when function returns iomem
Linus Torvalds [Wed, 30 Jun 2021 18:15:39 +0000 (11:15 -0700)]
Merge tag 'platform-drivers-x86-v5.14-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86
Pull x86 platform driver updates from Hans de Goede:
"Highlights:
- New think-lmi driver adding support for changing Lenovo Thinkpad
BIOS settings from within Linux using the standard firmware-
attributes class sysfs API
- MS Surface aggregator-cdev now also supports forwarding events to
user-space (for debugging / new driver development purposes only)
- New intel_skl_int3472 driver this provides the necessary glue to
translate ACPI table information to GPIOs, regulators, etc. for
camera sensors on Intel devices with IPU3 attached MIPI cameras
- A whole bunch of other fixes + device-specific quirk additions
- New devm_work_autocancel() devm-helpers.h function"
* tag 'platform-drivers-x86-v5.14-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86: (83 commits)
platform/x86: dell-wmi-sysman: Change user experience when Admin/System Password is modified
platform/x86: intel_skl_int3472: Uninitialized variable in skl_int3472_handle_gpio_resources()
platform/x86: think-lmi: Move kfree(setting->possible_values) to tlmi_attr_setting_release()
platform/x86: think-lmi: Split current_value to reflect only the value
platform/x86: think-lmi: Fix issues with duplicate attributes
platform/x86: think-lmi: Return EINVAL when kbdlang gets set to a 0 length string
platform/x86: intel_cht_int33fe: Move to its own subfolder
platform/x86: intel_skl_int3472: Move to intel/ subfolder
platform/x86: intel_skl_int3472: Provide skl_int3472_unregister_clock()
platform/x86: intel_skl_int3472: Provide skl_int3472_unregister_regulator()
platform/x86: intel_skl_int3472: Use ACPI GPIO resource directly
platform/x86: intel_skl_int3472: Fix dependencies (drop CLKDEV_LOOKUP)
platform/x86: intel_skl_int3472: Free ACPI device resources after use
platform/x86: Remove "default n" entries
platform/x86: ISST: Use numa node id for cpu pci dev mapping
platform/x86: ISST: Optimize CPU to PCI device mapping
tools/power/x86/intel-speed-select: v1.10 release
tools/power/x86/intel-speed-select: Fix uncore memory frequency display
extcon: extcon-max8997: Simplify driver using devm
extcon: extcon-max8997: Fix IRQ freeing at error path
...
Linus Torvalds [Wed, 30 Jun 2021 18:09:37 +0000 (11:09 -0700)]
Merge tag 'for-linus-5.14-1' of git://github.com/cminyard/linux-ipmi
Pull IPMI driver updates from Corey Minyard:
"Mostly a restructure of the kcs_bmc driver to make it easier to use
with different types of devices, and just to clean things up and
improve things.
Also some bug fixes for the kcs_bmc driver.
One fix to the IPMI watchdog to stop the timer when the action is
none. Not a big deal, but it's the right thing to do"
* tag 'for-linus-5.14-1' of git://github.com/cminyard/linux-ipmi:
ipmi: kcs_bmc_aspeed: Fix less than zero comparison of a unsigned int
ipmi: kcs_bmc_aspeed: Optionally apply status address
ipmi: kcs_bmc_aspeed: Fix IBFIE typo from datasheet
ipmi: kcs_bmc_aspeed: Implement KCS SerIRQ configuration
dt-bindings: ipmi: Add optional SerIRQ property to ASPEED KCS devices
dt-bindings: ipmi: Convert ASPEED KCS binding to schema
ipmi: kcs_bmc: Add serio adaptor
ipmi: kcs_bmc: Enable IBF on open
ipmi: kcs_bmc: Allow clients to control KCS IRQ state
ipmi: kcs_bmc: Decouple the IPMI chardev from the core
ipmi: kcs_bmc: Strip private client data from struct kcs_bmc
ipmi: kcs_bmc: Split headers into device and client
ipmi: kcs_bmc: Turn the driver data-structures inside-out
ipmi: kcs_bmc: Split out kcs_bmc_cdev_ipmi
ipmi: kcs_bmc: Rename {read,write}_{status,data}() functions
ipmi: kcs_bmc: Make status update atomic
ipmi: kcs_bmc_aspeed: Use of match data to extract KCS properties
ipmi/watchdog: Stop watchdog timer when the current action is 'none'
Wei Li [Tue, 29 Jun 2021 14:14:20 +0000 (22:14 +0800)]
MIPS: Fix PKMAP with 32-bit MIPS huge page support
When 32-bit MIPS huge page support is enabled, we halve the number of
pointers a PTE page holds, making its last half go to waste.
Correspondingly, we should halve the number of kmap entries, as we just
initialized only a single pte table for that in pagetable_init().
Fixes: 35476311e529 ("MIPS: Add partial 32-bit huge page support") Signed-off-by: Wei Li <[email protected]> Signed-off-by: Thomas Bogendoerfer <[email protected]>
1.Add a new TCU channel as the percpu timer of core1, this is to
prepare for the subsequent SMP support. The newly added channel
will not adversely affect the current single-core state.
2.Adjust the position of TCU node to make it consistent with the
order in jz4780.dtsi file.
Code which runs with interrupts enabled should be moved above
__end_soft_masked where possible, because maskable interrupts that hit
below that symbol will need to consult the soft mask table, which is an
extra cost.
Nicholas Piggin [Wed, 30 Jun 2021 07:46:20 +0000 (17:46 +1000)]
powerpc/64s/interrupt: clean up interrupt return labels
Normal kernel-interrupt exits can get interrupt_return_srr_user_restart
in their backtrace, which is an unusual and notable function, and it is
part of the user-interrupt exit path, which is doubly confusing.
Add non-local labels for both user and kernel interrupt exit cases to
address this and make the user and kernel cases more symmetric. Also get
rid of an unused label.
Nicholas Piggin [Wed, 30 Jun 2021 07:46:19 +0000 (17:46 +1000)]
powerpc/64/interrupt: add missing kprobe annotations on interrupt exit symbols
If one interrupt exit symbol must not be kprobed, none of them can be,
without more justification for why it's safe. Disallow kprobing on any
of the (non-local) labels in the exit paths.
Nicholas Piggin [Wed, 30 Jun 2021 07:46:18 +0000 (17:46 +1000)]
powerpc/64: enable MSR[EE] in irq replay pt_regs
Similar to commit 2b48e96be2f9f ("powerpc/64: fix irq replay
pt_regs->softe value"), enable MSR_EE in pt_regs->msr. This makes the
regs look more normal. It also allows some extra debug checks to be
added to interrupt handler entry.
Nicholas Piggin [Wed, 30 Jun 2021 07:46:17 +0000 (17:46 +1000)]
powerpc/64s/interrupt: preserve regs->softe for NMI interrupts
If an NMI interrupt hits in an implicit soft-masked region, regs->softe
is modified to reflect that. This may not be necessary for correctness
at the moment, but it is less surprising and it's unhelpful when
debugging or adding checks.
Make sure this is changed back to how it was found before returning.
Nicholas Piggin [Wed, 30 Jun 2021 07:46:16 +0000 (17:46 +1000)]
powerpc/64s: add a table of implicit soft-masked addresses
Commit 9d1988ca87dd ("powerpc/64: treat low kernel text as irqs
soft-masked") ends up catching too much code, including ret_from_fork,
and parts of interrupt and syscall return that do not expect to be
interrupts to be soft-masked. If an interrupt gets marked pending,
and then the code proceeds out of the implicit soft-masked region it
will fail to deal with the pending interrupt.
Fix this by adding a new table of addresses which explicitly marks
the regions of code that are soft masked. This table is only checked
for interrupts that below __end_soft_masked, so most kernel interrupts
will not have the overhead of the table search.
Nicholas Piggin [Wed, 30 Jun 2021 07:46:15 +0000 (17:46 +1000)]
powerpc/64e: remove implicit soft-masking and interrupt exit restart logic
The implicit soft-masking to speed up interrupt return was going to be
used by 64e as well, but it has not been extensively tested on that
platform and is not considered ready. It was intended to be disabled
before merge. Disable it for now.
Most of the restart code is common with 64s, so with more correctness
and performance testing this could be re-enabled again by adding the
extra soft-mask checks to interrupt handlers and flipping
exit_must_hard_disable().
CONFIG_RELOCATABLE=y causes build warnings from unresolved relocations.
Fix these by using TOC addressing for these cases.
Commit 24d33ac5b8ff ("powerpc/64s: Make prom_init require RELOCATABLE")
caused some 64e configs to select RELOCATABLE resulting in these
warnings, but the underlying issue was already there.
The early bad fault or key fault test in do_hash_fault() ends up calling
into ___do_page_fault without having gone through an interrupt handler
wrapper (except the initial _RAW one). This can end up calling local irq
functions while the interrupt has not been reconciled, which will likely
cause crashes and it trips up on a later patch that adds more assertions.
pkey_exec_prot from selftests causes this path to be executed.
There is no real reason to run the in_nmi() test should be performed
before the key fault check. In fact if a perf interrupt in the hash
fault code did a stack walk that was made to take a key fault somehow
then running ___do_page_fault could possibly cause another hash fault
causing problems. Move the in_nmi() test first, and then do everything
else inside the regular interrupt handler function.
Jiri Kosina [Wed, 30 Jun 2021 07:15:15 +0000 (09:15 +0200)]
Merge branch 'for-5.14/multitouch' into for-linus
- patch series that ensures that hid-multitouch driver disables touch and
button-press reporting on hid-mt devices during suspend when the device is
not configured as a wakeup-source, from Hans de Goede
Jiri Kosina [Wed, 30 Jun 2021 07:12:44 +0000 (09:12 +0200)]
Merge branch 'for-5.14/logitech' into for-linus
- support for LCD menu keys + LCD brightness control on the Logitech Z-10
speakers (with LCD) which use the same protocol as the G15 keyboards
from Hans de Goede
Jiri Kosina [Wed, 30 Jun 2021 07:03:51 +0000 (09:03 +0200)]
Merge branch 'for-5.14/core' into for-linus
- device unbinding locking fix from Dmitry Torokhov
- support for programmable buttons (mapping to KEY_MACRO# event codes)
from Thomas Weißschuh
- various other small fixes and code style improvements
projects / linux.git / log