Jann Horn demonstrated how userfaultfd ioctl UFFDIO_COPY into a private
shmem mapping can add valid PTEs to page table collapse_pte_mapped_thp()
thought it had emptied: page lock on the huge page is enough to protect
against WP faults (which find the PTE has been cleared), but not enough to
protect against userfaultfd. "BUG: Bad rss-counter state" followed.
retract_page_tables() protects against this by checking !vma->anon_vma;
but we know that MADV_COLLAPSE needs to be able to work on private shmem
mappings, even those with an anon_vma prepared for another part of the
mapping; and we know that MADV_COLLAPSE needs to work on shared shmem
mappings which are userfaultfd_armed(). Whether it needs to work on
private shmem mappings which are userfaultfd_armed(), I'm not so sure: but
assume that it does.
Just for this case, take the pmd_lock() two steps earlier: not because it
gives any protection against this case itself, but because ptlock nests
inside it, and it's the dropping of ptlock which let the bug in. In other
cases, continue to minimize the pmd_lock() hold time.
Link: https://lkml.kernel.org/r/4d31abf5-56c0-9f3d-d12f-c9317936691@google.com
Fixes: 1043173eb5 ("mm/khugepaged: collapse_pte_mapped_thp() with mmap_read_lock()")
Signed-off-by: Hugh Dickins <hughd@google.com>
Reported-by: Jann Horn <jannh@google.com>
Closes: https://lore.kernel.org/linux-mm/CAG48ez0FxiRC4d3VTu_a9h=rg5FW-kYD5Rg5xo_RDBM0LTTqZQ@mail.gmail.com/
Acked-by: Peter Xu <peterx@redhat.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
task_work's are typically queued up from IRQ/softirq potentially by a
random CPU like in case of networking. Batch ctx fields bouncing as this
into a separate cache line.
We also move ->cq_timeouts there because waiters have to read and check
it. We can also conditionally hide ->cq_timeouts in the future from the
CQ wait path as a not really useful rudiment.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/b7f3fcb5b6b9cca0238778262c1fdb7ada6286b7.1692916914.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Not many aware, but io_uring submission queue has two levels. The first
level usually appears as sq_array and stores indexes into the actual SQ.
To my knowledge, no one has ever seriously used it, nor liburing exposes
it to users. Add IORING_SETUP_NO_SQARRAY, when set we don't bother
creating and using the sq_array and SQ heads/tails will be pointing
directly into the SQ. Improves memory footprint, in term of both
allocations as well as cache usage, and also should make io_get_sqe()
less branchy in the end.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/0ffa3268a5ef61d326201ff43a233315c96312e0.1692916914.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Queues heads and tails cache line aligned. That makes sq, cq taking 4
lines or 5 lines if we include the rest of struct io_rings (e.g.
sq_flags is frequently accessed).
Since modern io_uring is mostly single threaded, it doesn't make much
send to spread them as such, it wastes space and puts additional pressure
on caches. Put them all into a single line.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/9c8deddf9a7ed32069235a530d1e117fb460bc4c.1692916914.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_do_iopoll() and io_submit_flush_completions() are pretty similar,
both filling CQEs and then free a list of requests. Don't duplicate it
and make iopoll use __io_submit_flush_completions(), which also helps
with inlining and other optimisations.
For that, we need to first find all completed iopoll requests and splice
them from the iopoll list and then pass it down. This adds one extra
list traversal, which should be fine as requests will stay hot in cache.
CQ locking is already conditional, introduce ->lockless_cq and skip
locking for IOPOLL as it's protected by ->uring_lock.
We also add a wakeup optimisation for IOPOLL to __io_cq_unlock_post(),
so it works just like io_cqring_ev_posted_iopoll().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/3840473f5e8a960de35b77292026691880f6bdbc.1692916914.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If the cached cqe check passes in io_get_cqe*() it already means that
the cqe we return is valid and non-zero, however the compiler is unable
to optimise null checks like in io_fill_cqe_req().
Do a bit of trickery, return success/fail boolean from io_get_cqe*()
and store cqe in the cqe parameter. That makes it do the right thing,
erasing the check together with the introduced indirection.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/322ea4d3377d3d4efd8ae90ab8ed28a99f518210.1692916914.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
While looking at io_fill_cqe_req()'s asm I stumbled on our trace points
turning into the chunk below:
trace_io_uring_complete(req->ctx, req, req->cqe.user_data,
req->cqe.res, req->cqe.flags,
req->extra1, req->extra2);
io_uring/io_uring.c:898: trace_io_uring_complete(req->ctx, req, req->cqe.user_data,
movq 232(%rbx), %rdi # req_44(D)->big_cqe.extra2, _5
movq 224(%rbx), %rdx # req_44(D)->big_cqe.extra1, _6
movl 84(%rbx), %r9d # req_44(D)->cqe.D.81184.flags, _7
movl 80(%rbx), %r8d # req_44(D)->cqe.res, _8
movq 72(%rbx), %rcx # req_44(D)->cqe.user_data, _9
movq 88(%rbx), %rsi # req_44(D)->ctx, _10
./arch/x86/include/asm/jump_label.h:27: asm_volatile_goto("1:"
1:jmp .L1772 # objtool NOPs this #
...
It does a jump_label for actual tracing, but those 6 moves will stay
there in the hottest io_uring path. As an optimisation, add a
trace_io_uring_complete_enabled() check, which is also uses jump_labels,
it tricks the compiler into behaving. It removes the junk without
changing anything else int the hot path.
Note: apparently, it's not only me noticing it, and people are also
working it around. We should remove the check when it's solved
generically or rework tracing.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/555d8312644b3776f4be7e23f9b92943875c4bc7.1692916914.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In mpc5xxx_fwnode_get_bus_frequency(), we should add
fwnode_handle_put() when break out of the iteration
fwnode_for_each_parent_node() as it will automatically
increase and decrease the refcounter.
Fixes: de06fba62a ("powerpc/mpc5xxx: Switch mpc5xxx_get_bus_frequency() to use fwnode")
Signed-off-by: Liang He <windhl@126.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20230322030423.1855440-1-windhl@126.com
In 5.10 commit 5e84dd547b ("powerpc/configs/skiroot: Enable some more
hardening options") set SLUB_DEBUG_ON.
When 5.14 came around, commit 792702911f ("slub: force on
no_hash_pointers when slub_debug is enabled") print all the
pointers when SLUB_DEBUG_ON is set. This was fine, but in 5.12 commit
5ead723a20 ("lib/vsprintf: no_hash_pointers prints all addresses as
unhashed") added the warning at boot.
Disable SLAB_DEBUG_ON as we don't want the nasty warning. We have
CONFIG_EXPERT so SLAB_DEBUG is enabled. We do lose the settings in
DEBUG_DEFAULT_FLAGS, but it's not clear that these should have been
always-on anyway.
Signed-off-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20230705023056.16273-1-joel@jms.id.au
When JUMP_LABEL=n, the tracepoint refcount test in the pre-call stores
the refcount value to the stack, so the same value can be used for the
post-call (presumably to avoid racing with the value concurrently
changing).
On little-endian (ELFv2) that might have just worked by luck, because
32(r1) is STK_PARAM(R3) there and so the value save gets clobbered by
the tracing code when it's non-zero, but fortunately r3 is the hcall
number and 0 is an invalid hcall number so it should get clobbered by
another non-zero value. In any case, commit cc1adb5f32
("powerpc/pseries: Use jump labels for hcall tracepoints") removed the
code that actually used the value stored, so now it's just dead code.
It's fragile to be storing to the stack like this, and confusing. Better
remove it.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20230509091600.70994-2-npiggin@gmail.com
Currently the -mtune options are set in the Makefile, depending on what
the compiler supports.
One downside of doing it that way is that the chosen -mtune option is
not recorded in the .config.
Another downside is that if there's ever a need to do more complicated
logic to calculate the correct option, that gets messy in the Makefile.
So move the determination of which -mtune option to use into Kconfig
logic.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20230329234308.2215833-1-mpe@ellerman.id.au
smaps_pte_hole_lookup() is calling shmem_partial_swap_usage() with page
table lock held: but shmem_partial_swap_usage() does cond_resched_rcu() if
need_resched(): "BUG: sleeping function called from invalid context".
Since shmem_partial_swap_usage() is designed to count across a range, but
smaps_pte_hole_lookup() only calls it for a single page slot, just break
out of the loop on the last or only page, before checking need_resched().
Link: https://lkml.kernel.org/r/6fe3b3ec-abdf-332f-5c23-6a3b3a3b11a9@google.com
Fixes: 2301003215 ("mm/smaps: simplify shmem handling of pte holes")
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Peter Xu <peterx@redhat.com>
Cc: <stable@vger.kernel.org> [5.16+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
The cachestat kselftest runs a test on a normal file, which is created
temporarily in the current directory. Among the tests it runs there is a
call to fsync(), which is expected to clean all dirty pages used by the
file.
However the tmpfs filesystem implements fsync() as noop_fsync(), so the
call will not even attempt to clean anything when this test file happens
to live on a tmpfs instance. This happens in an initramfs, or when the
current directory is in /dev/shm or sometimes /tmp.
To avoid this test failing wrongly, use statfs() to check which filesystem
the test file lives on. If that is "tmpfs", we skip the fsync() test.
Since the fsync test is only one part of the "normal file" test, we now
execute this twice, skipping the fsync part on the first call. This way
only the second test, including the fsync part, would be skipped.
Link: https://lkml.kernel.org/r/20230821160534.3414911-3-andre.przywara@arm.com
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Nhat Pham <nphamcs@gmail.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Patch series "selftests: cachestat: fix run on older kernels", v2.
I ran all kernel selftests on some test machine, and stumbled upon
cachestat failing (among others). These patches fix the run on older
kernels and when the current directory is on a tmpfs instance.
This patch (of 2):
As cachestat is a new syscall, it won't be available on older kernels, for
instance those running on a development machine. At the moment the test
reports all tests as "not ok" in this case.
Test for the cachestat syscall availability first, before doing further
tests, and bail out early with a TAP SKIP comment.
This also uses the opportunity to add the proper TAP headers, and add one
check for proper error handling (illegal file descriptor).
Link: https://lkml.kernel.org/r/20230821160534.3414911-1-andre.przywara@arm.com
Link: https://lkml.kernel.org/r/20230821160534.3414911-2-andre.przywara@arm.com
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Nhat Pham <nphamcs@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
The current implementation of append may cause duplicate data and/or
incorrect ranges to be returned to a reader during an update. Although
this has not been reported or seen, disable the append write operation
while the tree is in rcu mode out of an abundance of caution.
During the analysis of the mas_next_slot() the following was
artificially created by separating the writer and reader code:
Writer: reader:
mas_wr_append
set end pivot
updates end metata
Detects write to last slot
last slot write is to start of slot
store current contents in slot
overwrite old end pivot
mas_next_slot():
read end metadata
read old end pivot
return with incorrect range
store new value
Alternatively:
Writer: reader:
mas_wr_append
set end pivot
updates end metata
Detects write to last slot
last lost write to end of slot
store value
mas_next_slot():
read end metadata
read old end pivot
read new end pivot
return with incorrect range
set old end pivot
There may be other accesses that are not safe since we are now updating
both metadata and pointers, so disabling append if there could be rcu
readers is the safest action.
Link: https://lkml.kernel.org/r/20230819004356.1454718-2-Liam.Howlett@oracle.com
Fixes: 54a611b605 ("Maple Tree: add new data structure")
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Commit 98b211d641 ("madvise: convert madvise_free_pte_range() to use a
folio") replaced the page_mapcount() with folio_mapcount() to check
whether the folio is shared by other mapping.
It's not correct for large folios. folio_mapcount() returns the total
mapcount of large folio which is not suitable to detect whether the folio
is shared.
Use folio_estimated_sharers() which returns a estimated number of shares.
That means it's not 100% correct. It should be OK for madvise case here.
User-visible effects is that the THP is skipped when user call madvise.
But the correct behavior is THP should be split and processed then.
NOTE: this change is a temporary fix to reduce the user-visible effects
before the long term fix from David is ready.
Link: https://lkml.kernel.org/r/20230808020917.2230692-4-fengwei.yin@intel.com
Fixes: 98b211d641 ("madvise: convert madvise_free_pte_range() to use a folio")
Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
Reviewed-by: Yu Zhao <yuzhao@google.com>
Reviewed-by: Ryan Roberts <ryan.roberts@arm.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Vishal Moola (Oracle) <vishal.moola@gmail.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Commit fc986a38b6 ("mm: huge_memory: convert madvise_free_huge_pmd to
use a folio") replaced the page_mapcount() with folio_mapcount() to check
whether the folio is shared by other mapping.
It's not correct for large folios. folio_mapcount() returns the total
mapcount of large folio which is not suitable to detect whether the folio
is shared.
Use folio_estimated_sharers() which returns a estimated number of shares.
That means it's not 100% correct. It should be OK for madvise case here.
User-visible effects is that the THP is skipped when user call madvise.
But the correct behavior is THP should be split and processed then.
NOTE: this change is a temporary fix to reduce the user-visible effects
before the long term fix from David is ready.
Link: https://lkml.kernel.org/r/20230808020917.2230692-3-fengwei.yin@intel.com
Fixes: fc986a38b6 ("mm: huge_memory: convert madvise_free_huge_pmd to use a folio")
Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
Reviewed-by: Yu Zhao <yuzhao@google.com>
Reviewed-by: Ryan Roberts <ryan.roberts@arm.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Vishal Moola (Oracle) <vishal.moola@gmail.com>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Patch series "don't use mapcount() to check large folio sharing", v2.
In madvise_cold_or_pageout_pte_range() and madvise_free_pte_range(),
folio_mapcount() is used to check whether the folio is shared. But it's
not correct as folio_mapcount() returns total mapcount of large folio.
Use folio_estimated_sharers() here as the estimated number is enough.
This patchset will fix the cases:
User space application call madvise() with MADV_FREE, MADV_COLD and
MADV_PAGEOUT for specific address range. There are THP mapped to the
range. Without the patchset, the THP is skipped. With the patch, the
THP will be split and handled accordingly.
David reported the cow self test skip some cases because of MADV_PAGEOUT
skip THP:
https://lore.kernel.org/linux-mm/9e92e42d-488f-47db-ac9d-75b24cd0d037@intel.com/T/#mbf0f2ec7fbe45da47526de1d7036183981691e81
and I confirmed this patchset make it work again.
This patch (of 3):
Commit 07e8c82b5e ("madvise: convert madvise_cold_or_pageout_pte_range()
to use folios") replaced the page_mapcount() with folio_mapcount() to
check whether the folio is shared by other mapping.
It's not correct for large folio. folio_mapcount() returns the total
mapcount of large folio which is not suitable to detect whether the folio
is shared.
Use folio_estimated_sharers() which returns a estimated number of shares.
That means it's not 100% correct. It should be OK for madvise case here.
User-visible effects is that the THP is skipped when user call madvise.
But the correct behavior is THP should be split and processed then.
NOTE: this change is a temporary fix to reduce the user-visible effects
before the long term fix from David is ready.
Link: https://lkml.kernel.org/r/20230808020917.2230692-1-fengwei.yin@intel.com
Link: https://lkml.kernel.org/r/20230808020917.2230692-2-fengwei.yin@intel.com
Fixes: 07e8c82b5e ("madvise: convert madvise_cold_or_pageout_pte_range() to use folios")
Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
Reviewed-by: Yu Zhao <yuzhao@google.com>
Reviewed-by: Ryan Roberts <ryan.roberts@arm.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Vishal Moola (Oracle) <vishal.moola@gmail.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Pull nfsd fixes from Chuck Lever:
"Two last-minute one-liners for v6.5-rc. One got lost in the shuffle,
and the other was reported just this morning"
- Close race window when handling FREE_STATEID operations
- Fix regression in /proc/fs/nfsd/v4_end_grace introduced in v6.5-rc"
* tag 'nfsd-6.5-5' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux:
NFSD: Fix a thinko introduced by recent trace point changes
nfsd: Fix race to FREE_STATEID and cl_revoked
Pull spi fixes from Mark Brown:
"A couple more small driver specific fixes for v6.5.
The device mode for Cadence had been broken by some recent updates
done for host mode and large transfers for multi-byte words on stm32
had been broken by an API update in what I think was a rebasing
incident"
* tag 'spi-fix-v6.5-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: spi-cadence: Fix data corruption issues in slave mode
spi: stm32: fix accidential revert to byte-sized transfer splitting
The following declarations have never been implemented since the beginning
of git history, so remove them:
u8 acpiphp_get_attention_status(struct acpiphp_slot *slot);
u8 cpci_get_latch_status(struct slot *slot);
u8 cpci_get_adapter_status(struct slot *slot);
int ibmphp_get_total_hp_slots(void);
void ibmphp_free_ibm_slot(struct slot *);
void pdev_enable_device(struct pci_dev *);
Link: https://lore.kernel.org/r/20230811095933.28652-1-yuehaibing@huawei.com
Signed-off-by: Yue Haibing <yuehaibing@huawei.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
A recent cleanup of soc_dai_link_sanity_check() is responsible for
generating lots of confusing errors before returning -EPROBE_DEFER:
acp5x_mach acp5x_mach.0: ASoC: Component acp5x_i2s_dma.0 not found for link acp5x-8821-play
[...]
acp5x_mach acp5x_mach.0: ASoC: Component spi-VLV1776:00 not found for link acp5x-CS35L41-Stereo
[...]
acp5x_mach acp5x_mach.0: ASoC: Component spi-VLV1776:01 not found for link acp5x-CS35L41-Stereo
Switch back to the initial behaviour of logging those messages on
KERN_DEBUG level instead of KERN_ERR.
While at it, use the correct form of the verb in 'component_not_find'
label.
Fixes: 0e66a2c694 ("ASoC: soc-core.c: cleanup soc_dai_link_sanity_check()")
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com
Link: https://lore.kernel.org/r/20230824193837.369761-1-cristian.ciocaltea@collabora.com
Signed-off-by: Mark Brown <broonie@kernel.org
Add devices IDs for the next LOM generations that will be available on the
next Intel Client platforms. This patch provides the initial support for
these devices.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
