| Age | Commit message (Collapse) | Author |
|---|
|
Drop the wholly unnecessary set_vma_sealed() helper(), which is used only
once, and place VMA_ITERATOR() declarations in the correct place.
Retain vma_is_sealed(), and use it instead of the confusingly named
can_modify_vma(), so it's abundantly clear what's being tested, rather
then a nebulous sense of 'can the VMA be modified'.
No functional change intended.
Link: https://lkml.kernel.org/r/98cf28d04583d632a6eb698e9ad23733bb6af26b.1753431105.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Reviewed-by: Pedro Falcato <pfalcato@suse.de>
Acked-by: David Hildenbrand <david@redhat.com>
Acked-by: Jeff Xu <jeffxu@chromium.org>
Cc: Jann Horn <jannh@google.com>
Cc: Kees Cook <kees@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The madvise() logic is inexplicably performed in mm/mseal.c - this ought
to be located in mm/madvise.c.
Additionally can_modify_vma_madv() is inconsistently named and, in
combination with is_ro_anon(), is very confusing logic.
Put a static function in mm/madvise.c instead - can_madvise_modify() -
that spells out exactly what's happening. Also explicitly check for an
anon VMA.
Also add commentary to explain what's going on.
Essentially - we disallow discarding of data in mseal()'d mappings in
instances where the user couldn't otherwise write to that data.
We retain the existing behaviour here regarding MAP_PRIVATE mappings of
file-backed mappings, which entails some complexity - while this, strictly
speaking - appears to violate mseal() semantics, it may interact badly
with users which expect to be able to madvise(MADV_DONTNEED) .text
mappings for instance.
We may revisit this at a later date.
No functional change intended.
Link: https://lkml.kernel.org/r/492a98d9189646e92c8f23f4cce41ed323fe01df.1753431105.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Reviewed-by: Pedro Falcato <pfalcato@suse.de>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Jann Horn <jannh@google.com>
Cc: Jeff Xu <jeffxu@chromium.org>
Cc: Kees Cook <kees@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
With maple_tree supporting vma tree traversal under RCU and per-vma locks,
/proc/pid/maps can be read while holding individual vma locks instead of
locking the entire address space.
A completely lockless approach (walking vma tree under RCU) would be quite
complex with the main issue being get_vma_name() using callbacks which
might not work correctly with a stable vma copy, requiring original
(unstable) vma - see special_mapping_name() for example.
When per-vma lock acquisition fails, we take the mmap_lock for reading,
lock the vma, release the mmap_lock and continue. This fallback to mmap
read lock guarantees the reader to make forward progress even during lock
contention. This will interfere with the writer but for a very short time
while we are acquiring the per-vma lock and only when there was contention
on the vma reader is interested in.
We shouldn't see a repeated fallback to mmap read locks in practice, as
this require a very unlikely series of lock contentions (for instance due
to repeated vma split operations). However even if this did somehow
happen, we would still progress.
One case requiring special handling is when a vma changes between the time
it was found and the time it got locked. A problematic case would be if a
vma got shrunk so that its vm_start moved higher in the address space and
a new vma was installed at the beginning:
reader found: |--------VMA A--------|
VMA is modified: |-VMA B-|----VMA A----|
reader locks modified VMA A
reader reports VMA A: | gap |----VMA A----|
This would result in reporting a gap in the address space that does not
exist. To prevent this we retry the lookup after locking the vma, however
we do that only when we identify a gap and detect that the address space
was changed after we found the vma.
This change is designed to reduce mmap_lock contention and prevent a
process reading /proc/pid/maps files (often a low priority task, such as
monitoring/data collection services) from blocking address space updates.
Note that this change has a userspace visible disadvantage: it allows for
sub-page data tearing as opposed to the previous mechanism where data
tearing could happen only between pages of generated output data. Since
current userspace considers data tearing between pages to be acceptable,
we assume is will be able to handle sub-page data tearing as well.
Link: https://lkml.kernel.org/r/20250719182854.3166724-7-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Hildenbrand <david@redhat.com>
Cc: Jann Horn <jannh@google.com>
Cc: Jeongjun Park <aha310510@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Kalesh Singh <kaleshsingh@google.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: "Paul E . McKenney" <paulmck@kernel.org>
Cc: Peter Xu <peterx@redhat.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Thomas Weißschuh <linux@weissschuh.net>
Cc: T.J. Mercier <tjmercier@google.com>
Cc: Ye Bin <yebin10@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The single instance in which we use this function doesn't actually need to
change VMA flags, so remove this parameter and update the caller
accordingly.
[lorenzo.stoakes@oracle.com: correct comment]
Link: https://lkml.kernel.org/r/77f45b2e-a748-4635-9381-a5051091087f@lucifer.local
Link: https://lkml.kernel.org/r/20250714135839.178032-1-lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Jann Horn <jannh@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Instead, let's just allow for specifying through flags whether we want to
have bits merged into the original PTE.
For the madvise() case, simplify by having only a single parameter for
merging young+dirty. For madvise_cold_or_pageout_pte_range() merging the
dirty bit is not required, but also not harmful. This code is not that
performance critical after all to really force all micro-optimizations.
As we now have two pte_t * parameters, use PageTable() to make sure we are
actually given a pointer at a copy of the PTE, not a pointer into an
actual page table.
Link: https://lkml.kernel.org/r/20250702104926.212243-5-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Dev Jain <dev.jain@arm.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Byungchul Park <byungchul@sk.com>
Cc: Gregory Price <gourry@gourry.net>
Cc: "Huang, Ying" <ying.huang@linux.alibaba.com>
Cc: Jann Horn <jannh@google.com>
Cc: Joshua Hahn <joshua.hahnjy@gmail.com>
Cc: Lance Yang <lance.yang@linux.dev>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Mathew Brost <matthew.brost@intel.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Rakie Kim <rakie.kim@sk.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Many users (including upcoming ones) don't really need the flags etc, and
can live with the possible overhead of a function call.
So let's provide a basic, non-inlined folio_pte_batch(), to avoid code
bloat while still providing a variant that optimizes out all flag checks
at runtime. folio_pte_batch_flags() will get inlined into
folio_pte_batch(), optimizing out any conditionals that depend on input
flags.
folio_pte_batch() will behave like folio_pte_batch_flags() when no flags
are specified. It's okay to add new users of folio_pte_batch_flags(), but
using folio_pte_batch() if applicable is preferred.
So, before this change, folio_pte_batch() was inlined into the C file
optimized by propagating constants within the resulting object file.
With this change, we now also have a folio_pte_batch() that is optimized
by propagating all constants. But instead of having one instance per
object file, we have a single shared one.
In zap_present_ptes(), where we care about performance, the compiler
already seem to generate a call to a common inlined folio_pte_batch()
variant, shared with fork() code. So calling the new non-inlined variant
should not make a difference.
While at it, drop the "addr" parameter that is unused.
Link: https://lkml.kernel.org/r/20250702104926.212243-4-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Suggested-by: Andrew Morton <akpm@linux-foundation.org>
Link: https://lore.kernel.org/linux-mm/20250503182858.5a02729fcffd6d4723afcfc2@linux-foundation.org/
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: Dev Jain <dev.jain@arm.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Byungchul Park <byungchul@sk.com>
Cc: Gregory Price <gourry@gourry.net>
Cc: "Huang, Ying" <ying.huang@linux.alibaba.com>
Cc: Jann Horn <jannh@google.com>
Cc: Joshua Hahn <joshua.hahnjy@gmail.com>
Cc: Lance Yang <lance.yang@linux.dev>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Mathew Brost <matthew.brost@intel.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Rakie Kim <rakie.kim@sk.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "mm: folio_pte_batch() improvements", v2.
Ever since we added folio_pte_batch() for fork() + munmap() purposes, a
lot more users appeared (and more are being proposed), and more
functionality was added.
Most of the users only need basic functionality, and could benefit from a
non-inlined version.
So let's clean up folio_pte_batch() and split it into a basic
folio_pte_batch() (no flags) and a more advanced folio_pte_batch_ext().
Using either variant will now look much cleaner.
This series will likely conflict with some changes in some (old+new)
folio_pte_batch() users, but conflicts should be trivial to resolve.
This patch (of 4):
Respecting these PTE bits is the exception, so let's invert the meaning.
With this change, most callers don't have to pass any flags. This is a
preparation for splitting folio_pte_batch() into a non-inlined variant
that doesn't consume any flags.
Long-term, we want folio_pte_batch() to probably ignore most common PTE
bits (e.g., write/dirty/young/soft-dirty) that are not relevant for most
page table walkers: uffd-wp and protnone might be bits to consider in the
future. Only walkers that care about them can opt-in to respect them.
No functional change intended.
Link: https://lkml.kernel.org/r/20250702104926.212243-2-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Lance Yang <lance.yang@linux.dev>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: Dev Jain <dev.jain@arm.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Byungchul Park <byungchul@sk.com>
Cc: Gregory Price <gourry@gourry.net>
Cc: "Huang, Ying" <ying.huang@linux.alibaba.com>
Cc: Jann Horn <jannh@google.com>
Cc: Joshua Hahn <joshua.hahnjy@gmail.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Mathew Brost <matthew.brost@intel.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Rakie Kim <rakie.kim@sk.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Use madvise_lock()/madvise_unlock() in madvise_set_anon_name() in the same
way as in do_madvise(). This narrows the lock scope a bit and reuses
existing functionality. get_lock_mode() already picks the correct
MADVISE_MMAP_WRITE_LOCK mode for __MADV_SET_ANON_VMA_NAME so we can just
remove the explicit assignment.
There is a user visible change in that the prctl(PR_SET_VMA,
PR_SET_VMA_ANON_NAME...) might now return -EINTR.
Link: https://lkml.kernel.org/r/20250624-anon_name_cleanup-v2-4-600075462a11@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Suren Baghdasaryan <surenb@google.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Colin Cross <ccross@google.com>
Cc: Jann Horn <jannh@google.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: SeongJae Park <sj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Preparatory change so that we can use madvise_lock()/unlock() in the
function without forward declarations or more thorough shuffling.
No functional change. Move as a separate commit helps git heuristics to
detect it properly.
Link: https://lkml.kernel.org/r/20250624-anon_name_cleanup-v2-3-600075462a11@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Suren Baghdasaryan <surenb@google.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Colin Cross <ccross@google.com>
Cc: Jann Horn <jannh@google.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: SeongJae Park <sj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Setting anon_name is done via madvise_set_anon_name() and behaves a lot of
like other madvise operations. However, apparently because madvise() has
lacked the 4th argument and prctl() not, the userspace entry point has
been implemented via prctl(PR_SET_VMA, ...) and handled first by
prctl_set_vma().
Currently prctl_set_vma() lives in kernel/sys.c but setting the
vma->anon_name is mm-specific code so extract it to a new
set_anon_vma_name() function under mm. mm/madvise.c seems to be the most
straightforward place as that's where madvise_set_anon_name() lives. Stop
declaring the latter in mm.h and instead declare set_anon_vma_name().
Link: https://lkml.kernel.org/r/20250624-anon_name_cleanup-v2-2-600075462a11@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: David Hildenbrand <david@redhat.com>
Tested-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Suren Baghdasaryan <surenb@google.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Colin Cross <ccross@google.com>
Cc: Jann Horn <jannh@google.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: SeongJae Park <sj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "madvise anon_name cleanups", v2.
While reviewing Lorenzo's madvise cleanups I've noticed that we can handle
anon_name in madvise code much better, so sending that as patch 1.
Initially I wanted to do first move the existing logic from
madvise_vma_behavior() to madvise_update_vma() as a separate patch before
the actual simplification but that would require adding
anon_vma_name_put() in error handling paths only to be removed again, so
it's a single patch to avoid churn.
It's also an opportunity to move some mm code from prctl under mm, hence
patch 2. After code moving preparation in patch 3, also unify madvise
lock handling for madvise_set_anon_name() in patch 4.
This patch (of 4):
Since the introduction in 9a10064f5625 ("mm: add a field to store names
for private anonymous memory") the code to set anon_name on a vma has been
using madvise_update_vma() to call replace_anon_vma_name(). Since the
former is called also by a number of other madvise behaviours that do not
set a new anon_name, they have been passing the existing anon_name of the
vma to make replace_anon_vma_name() a no-op.
This is rather wasteful as it needs anon_vma_name_eq() to determine the
no-op situations, and checks for when replace_anon_vma_name() is allowed
(the vma is anon/shmem) duplicate the checks already done earlier in
madvise_vma_behavior(). It has also lead to commit 942341dcc574 ("mm: fix
use-after-free when anon vma name is used after vma is freed") adding
anon_name refcount get/put operations exactly to the cases that actually
do not change anon_name - just so the replace_anon_vma_name() can keep
safely determining it has nothing to do.
The recent madvise cleanups made this suboptimal handling very obvious,
but happily also allow for an easy fix. madvise_update_vma() now has the
complete information whether it's been called to set a new anon_name, so
stop passing it the existing vma's name and doing the refcount get/put in
its only caller madvise_vma_behavior().
In madvise_update_vma() itself, limit calling of replace_anon_vma_name()
only to cases where we are setting a new name, otherwise we know it's a
no-op. We can rely solely on the __MADV_SET_ANON_VMA_NAME behaviour and
can remove the duplicate checks for vma being anon/shmem that were done
already in madvise_vma_behavior().
Additionally, by using vma_modify_flags() when not modifying the
anon_name, avoid explicitly passing the existing vma's anon_name and
storing a pointer to it in struct madv_behavior or a local variable. This
prevents the danger of accessing a freed anon_name after vma merging,
previously fixed by commit 942341dcc574.
Link: https://lkml.kernel.org/r/20250624-anon_name_cleanup-v2-0-600075462a11@suse.cz
Link: https://lkml.kernel.org/r/20250624-anon_name_cleanup-v2-1-600075462a11@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Suren Baghdasaryan <surenb@google.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Tested-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Colin Cross <ccross@google.com>
Cc: Jann Horn <jannh@google.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: SeongJae Park <sj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The madvise code has for the longest time had very confusing code around
the 'prev' VMA pointer passed around various functions which, in all cases
except madvise_update_vma(), is unused and instead simply updated as soon
as the function is invoked.
To compound the confusion, the prev pointer is also used to indicate to
the caller that the mmap lock has been dropped and that we can therefore
not safely access the end of the current VMA (which might have been
updated by madvise_update_vma()).
Clear up this confusion by not setting prev = vma anywhere except in
madvise_walk_vmas(), update all references to prev which will always be
equal to vma after madvise_vma_behavior() is invoked, and adding a flag to
indicate that the lock has been dropped to make this explicit.
Additionally, drop a redundant BUG_ON() from madvise_collapse(), which is
simply reiterating the BUG_ON(mmap_locked) above it (note that BUG_ON() is
not appropriate here, but we leave existing code as-is).
We finally adjust the madvise_walk_vmas() logic to be a little clearer -
delaying the assignment of the end of the range to the start of the new
range until the last moment and handling the lock being dropped scenario
immediately.
Additionally add some explanatory comments.
[lorenzo.stoakes@oracle.com: fix very subtle bug]
Link: https://lkml.kernel.org/r/dca94cde-8afb-4eab-8e57-3f508624d670@lucifer.local
Link: https://lkml.kernel.org/r/63d281c5df2e64225ab5b4bda398b45e22818701.1750433500.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: SeongJae Park <sj@kernel.org>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Barry Song <baohua@kernel.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: Dev Jain <dev.jain@arm.com>
Cc: Jann Horn <jannh@google.com>
Cc: Lance Yang <ioworker0@gmail.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Mariano Pache <npache@redhat.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Doing so means we can get rid of all the weird struct vm_area_struct
**prev stuff, everything becomes consistent and in future if we want to
make change to behaviour there's a single place where all relevant state
is stored.
This also allows us to update try_vma_read_lock() to be a little more
succinct and set up state for us, as well as cleaning up
madvise_update_vma().
We also update the debug assertion prior to madvise_update_vma() to assert
that this is a write operation as correctly pointed out by Barry in the
relevant thread.
We can't reasonably update the madvise functions that live outside of
mm/madvise.c so we leave those as-is.
Link: https://lkml.kernel.org/r/7b345ab82ef51e551f8bc0c4f7be25712871629d.1750433500.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Acked-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: SeongJae Park <sj@kernel.org>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Barry Song <baohua@kernel.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: Dev Jain <dev.jain@arm.com>
Cc: Jann Horn <jannh@google.com>
Cc: Lance Yang <ioworker0@gmail.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Mariano Pache <npache@redhat.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Rather than updating start and a confusing local parameter 'tmp' in
madvise_walk_vmas(), instead store the current range being operated upon
in the struct madvise_behavior helper object in a range pair and use this
consistently in all operations.
This makes it clearer what is going on and opens the door to further
cleanup now we store state regarding what is currently being operated upon
here.
Link: https://lkml.kernel.org/r/518480ceb48553d3c280bc2b0b5e77bbad840147.1750433500.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Barry Song <baohua@kernel.org>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Dev Jain <dev.jain@arm.com>
Cc: Jann Horn <jannh@google.com>
Cc: Lance Yang <ioworker0@gmail.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Mariano Pache <npache@redhat.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
There's no need to thread a pointer to the mm_struct nor have different
functions signatures for each behaviour, instead store state in the struct
madvise_behavior object consistently and use it for all madvise() actions.
Link: https://lkml.kernel.org/r/a47d850b0111735e026d438c3300c0e3b7f439f4.1750433500.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Barry Song <baohua@kernel.org>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Dev Jain <dev.jain@arm.com>
Cc: Jann Horn <jannh@google.com>
Cc: Lance Yang <ioworker0@gmail.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Mariano Pache <npache@redhat.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "madvise cleanup", v2.
This is a series of patches that helps address a number of historic
problems in the madvise() implementation:
* Eliminate the visitor pattern and having the code which is implemented
for both the anon_vma_name implementation and ordinary madvise()
operations use the same madvise_vma_behavior() implementation.
* Thread state through the madvise_behavior state object - this object,
very usefully introduced by SJ, is already used to transmit state
through operations. This series extends this by having all madvise()
operations use this, including anon_vma_name.
* Thread range, VMA state through madvise_behavior - This helps avoid a
lot of the confusing code around range and VMA state and again keeps
things consistent and with a single 'source of truth'.
* Addressing the very strange behaviour around the passed around struct
vm_area_struct **prev pointer - all read-only users do absolutely
nothing with the prev pointer. The only function that uses it is
madvise_update_vma(), and in all cases prev is always reset to VMA.
Fix this by no longer having aything but madvise_update_vma()
reference prev, and having madvise_walk_vmas() update prev in each
instance. Additionally make it clear that the meaningful change in vma
state is when madvise_update_vma() potentially merges a VMA, so
explicitly retrieve the VMA in this case.
* Update and clarify the madvise_walk_vmas() function - this is a source
of a great deal of confusion, so simplify, stop using prev = NULL to
signify that the mmap lock has been dropped (!) and make that explicit,
and add some comments to explain what's going on.
This patch (of 5):
Now we have the madvise_behavior helper struct we no longer need to mess
around with void* pointers in order to propagate anon_vma_name, and this
means we can get rid of the confusing and inconsistent visitor pattern
implementation in madvise_vma_anon_name().
This means we now have a single state object that threads through most of
madvise()'s logic and a single code path which executes the majority of
madvise() behaviour (we maintain separate logic for failure injection and
memory population for the time being).
We are able to remove the visitor pattern by handling the anon_vma_name
setting logic via an internal madvise flag - __MADV_SET_ANON_VMA_NAME.
This uses a negative value so it isn't reasonable that we will ever add
this as a UAPI flag.
Additionally, the madvise_behavior_valid() check ensures that
user-specified behaviours are strictly only those we permit which, of
course, this flag will be excluded from.
We are able to propagate the anon_vma_name object through use of the
madvise_behavior helper struct.
Doing this results in a can_modify_vma_madv() check for anonymous VMA name
changes, however this will cause no issues as this operation is not
prohibited.
We can also then reuse more code and drop the redundant
madvise_vma_anon_name() function altogether.
Additionally separate out behaviours that update VMAs from those that do
not.
Link: https://lkml.kernel.org/r/cover.1750433500.git.lorenzo.stoakes@oracle.com
Link: https://lkml.kernel.org/r/c5094bfccb41ecd19d4e9bcaa1c4a11e00158bba.1750433500.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Barry Song <baohua@kernel.org>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Dev Jain <dev.jain@arm.com>
Cc: Jann Horn <jannh@google.com>
Cc: Lance Yang <ioworker0@gmail.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Mariano Pache <npache@redhat.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Now that DAX and all other reference counts to ZONE_DEVICE pages are
managed normally there is no need for the special devmap PTE/PMD/PUD page
table bits. So drop all references to these, freeing up a software
defined page table bit on architectures supporting it.
Link: https://lkml.kernel.org/r/6389398c32cc9daa3dfcaa9f79c7972525d310ce.1750323463.git-series.apopple@nvidia.com
Signed-off-by: Alistair Popple <apopple@nvidia.com>
Acked-by: Will Deacon <will@kernel.org> # arm64
Acked-by: David Hildenbrand <david@redhat.com>
Suggested-by: Chunyan Zhang <zhang.lyra@gmail.com>
Reviewed-by: Björn Töpel <bjorn@rivosinc.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Cc: Balbir Singh <balbirs@nvidia.com>
Cc: Björn Töpel <bjorn@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Deepak Gupta <debug@rivosinc.com>
Cc: Gerald Schaefer <gerald.schaefer@linux.ibm.com>
Cc: Inki Dae <m.szyprowski@samsung.com>
Cc: John Groves <john@groves.net>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The core kernel code is currently very inconsistent in its use of
vm_flags_t vs. unsigned long. This prevents us from changing the type of
vm_flags_t in the future and is simply not correct, so correct this.
While this results in rather a lot of churn, it is a critical
pre-requisite for a future planned change to VMA flag type.
Additionally, update VMA userland tests to account for the changes.
To make review easier and to break things into smaller parts, driver and
architecture-specific changes is left for a subsequent commit.
The code has been adjusted to cascade the changes across all calling code
as far as is needed.
We will adjust architecture-specific and driver code in a subsequent patch.
Overall, this patch does not introduce any functional change.
Link: https://lkml.kernel.org/r/d1588e7bb96d1ea3fe7b9df2c699d5b4592d901d.1750274467.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Acked-by: Kees Cook <kees@kernel.org>
Acked-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Acked-by: Jan Kara <jack@suse.cz>
Acked-by: Christian Brauner <brauner@kernel.org>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: Pedro Falcato <pfalcato@suse.de>
Acked-by: Zi Yan <ziy@nvidia.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com>
Cc: Jann Horn <jannh@google.com>
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
MADV_FREE is another option, besides MADV_DONTNEED, for dynamic memory
freeing in user-space native or Java heap memory management. For example,
jemalloc can be configured to use MADV_FREE, and recent versions of the
Android Java heap have also increasingly adopted MADV_FREE. Supporting
per-VMA locking for MADV_FREE thus appears increasingly necessary.
We have replaced walk_page_range() with walk_page_range_vma(). Along with
the proposed madvise_lock_mode by Lorenzo, the necessary infrastructure is
now in place to begin exploring per-VMA locking support for MADV_FREE and
potentially other madvise using walk_page_range_vma().
This patch adds support for the PGWALK_VMA_RDLOCK walk_lock mode in
walk_page_range_vma(), and leverages madvise_lock_mode from madv_behavior
to select the appropriate walk_lock—either mmap_lock or per-VMA
lock—based on the context.
Because we now dynamically update the walk_ops->walk_lock field, we must
ensure this is thread-safe. The madvise_free_walk_ops is now defined as a
stack variable instead of a global constant.
Link: https://lkml.kernel.org/r/20250611104745.57405-1-21cnbao@gmail.com
Signed-off-by: Barry Song <v-songbaohua@oppo.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Acked-by: SeongJae Park <sj@kernel.org>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Jann Horn <jannh@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Lokesh Gidra <lokeshgidra@google.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Tangquan Zheng <zhengtangquan@oppo.com>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Certain madvise operations, especially MADV_DONTNEED, occur far more
frequently than other madvise options, particularly in native and Java
heaps for dynamic memory management.
Currently, the mmap_lock is always held during these operations, even when
unnecessary. This causes lock contention and can lead to severe priority
inversion, where low-priority threads—such as Android's
HeapTaskDaemon— hold the lock and block higher-priority threads.
This patch enables the use of per-VMA locks when the advised range lies
entirely within a single VMA, avoiding the need for full VMA traversal.
In practice, userspace heaps rarely issue MADV_DONTNEED across multiple
VMAs.
Tangquan's testing shows that over 99.5% of memory reclaimed by Android
benefits from this per-VMA lock optimization. After extended runtime,
217,735 madvise calls from HeapTaskDaemon used the per-VMA path, while
only 1,231 fell back to mmap_lock.
To simplify handling, the implementation falls back to the standard
mmap_lock if userfaultfd is enabled on the VMA, avoiding the complexity of
userfaultfd_remove().
Many thanks to Lorenzo's work[1] on "mm/madvise: support VMA read locks
for MADV_DONTNEED[_LOCKED]"
Then use this mechanism to permit VMA locking to be done later in the
madvise() logic and also to allow altering of the locking mode to permit
falling back to an mmap read lock if required."
One important point, as pointed out by Jann[2], is that
untagged_addr_remote() requires holding mmap_lock. This is because
address tagging on x86 and RISC-V is quite complex.
Until untagged_addr_remote() becomes atomic—which seems unlikely in the
near future—we cannot support per-VMA locks for remote processes. So
for now, only local processes are supported.
Lance said:
: Just to put some numbers on it, I ran a micro-benchmark with 100
: parallel threads, where each thread calls madvise() on its own 1GiB
: chunk of 64KiB mTHP-backed memory. The performance gain is huge:
:
: 1) MADV_DONTNEED saw its average time drop from 0.0508s to 0.0270s
: (~47% faster)
:
: 2) MADV_FREE saw its average time drop from 0.3078s to 0.1095s (~64%
: faster)
[lorenzo.stoakes@oracle.com: avoid any chance of uninitialised pointer deref]
Link: https://lkml.kernel.org/r/309d22ca-6cd9-4601-8402-d441a07d9443@lucifer.local
Link: https://lore.kernel.org/all/0b96ce61-a52c-4036-b5b6-5c50783db51f@lucifer.local/ [1]
Link: https://lore.kernel.org/all/CAG48ez11zi-1jicHUZtLhyoNPGGVB+ROeAJCUw48bsjk4bbEkA@mail.gmail.com/ [2]
Link: https://lkml.kernel.org/r/20250607220150.2980-1-21cnbao@gmail.com
Signed-off-by: Barry Song <v-songbaohua@oppo.com>
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Acked-by: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Jann Horn <jannh@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Lokesh Gidra <lokeshgidra@google.com>
Cc: Tangquan Zheng <zhengtangquan@oppo.com>
Cc: Lance Yang <ioworker0@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
We've already found the VMA within madvise_walk_vmas() before calling
specific madvise behavior functions like madvise_free_single_vma(). So
calling walk_page_range() and doing find_vma() again seems unnecessary.
It also prevents potential optimizations in those madvise callbacks,
particularly the use of dedicated per-VMA locking.
[v-songbaohua@oppo.com: revert the walk_page_range_vma change for MADV_GUARD_INSTALL]
Link: https://lkml.kernel.org/r/20250609105513.10901-1-21cnbao@gmail.com
Link: https://lkml.kernel.org/r/20250605083144.43046-1-21cnbao@gmail.com
Signed-off-by: Barry Song <v-songbaohua@oppo.com>
Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: Harry Yoo <harry.yoo@oracle.com>
Reviewed-by: Dev Jain <dev.jain@arm.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Ryan Roberts <ryan.roberts@arm.com>
Tested-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Jann Horn <jannh@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Lokesh Gidra <lokeshgidra@google.com>
Cc: Tangquan Zheng <zhengtangquan@oppo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Commit 3ea277194daa ("mm, mprotect: flush TLB if potentially racing with a
parallel reclaim leaving stale TLB entries") described a theoretical race
as such:
"""
Nadav Amit identified a theoretical race between page reclaim and mprotect
due to TLB flushes being batched outside of the PTL being held.
He described the race as follows:
CPU0 CPU1
---- ----
user accesses memory using RW PTE
[PTE now cached in TLB]
try_to_unmap_one()
==> ptep_get_and_clear()
==> set_tlb_ubc_flush_pending()
mprotect(addr, PROT_READ)
==> change_pte_range()
==> [ PTE non-present - no flush ]
user writes using cached RW PTE
...
try_to_unmap_flush()
The same type of race exists for reads when protecting for PROT_NONE and
also exists for operations that can leave an old TLB entry behind such as
munmap, mremap and madvise.
"""
The solution was to introduce flush_tlb_batched_pending() and call it
under the PTL from mprotect/madvise/munmap/mremap to complete any pending
tlb flushes.
However, while madvise_free_pte_range() and
madvise_cold_or_pageout_pte_range() were both retro-fitted to call
flush_tlb_batched_pending() immediately after initially acquiring the PTL,
they both temporarily release the PTL to split a large folio if they
stumble upon one. In this case, where re-acquiring the PTL
flush_tlb_batched_pending() must be called again, but it previously was
not. Let's fix that.
There are 2 Fixes: tags here: the first is the commit that fixed
madvise_free_pte_range(). The second is the commit that added
madvise_cold_or_pageout_pte_range(), which looks like it copy/pasted the
faulty pattern from madvise_free_pte_range().
This is a theoretical bug discovered during code review.
Link: https://lkml.kernel.org/r/20250606092809.4194056-1-ryan.roberts@arm.com
Fixes: 3ea277194daa ("mm, mprotect: flush TLB if potentially racing with a parallel reclaim leaving stale TLB entries")
Fixes: 9c276cc65a58 ("mm: introduce MADV_COLD")
Signed-off-by: Ryan Roberts <ryan.roberts@arm.com>
Reviewed-by: Jann Horn <jannh@google.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Mel Gorman <mgorman <mgorman@suse.de>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
When unwinding race on -ERESTARTNOINTR handling of process_madvise(),
madvise_lock() failure is ignored. Check the failure and abort remaining
works in the case.
Link: https://lkml.kernel.org/r/20250602174926.1074-1-sj@kernel.org
Fixes: 4000e3d0a367 ("mm/madvise: remove redundant mmap_lock operations from process_madvise()")
Signed-off-by: SeongJae Park <sj@kernel.org>
Reported-by: Barry Song <21cnbao@gmail.com>
Closes: https://lore.kernel.org/CAGsJ_4xJXXO0G+4BizhohSZ4yDteziPw43_uF8nPXPWxUVChzw@mail.gmail.com
Reviewed-by: Jann Horn <jannh@google.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Reviewed-by: Barry Song <baohua@kernel.org>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
MADV_DONTNEED[_LOCKED] handling for [process_]madvise() flushes tlb for
each vma of each address range. Update the logic to do tlb flushes in a
batched way. Initialize an mmu_gather object from do_madvise() and
vector_madvise(), which are the entry level functions for
[process_]madvise(), respectively. And pass those objects to the function
for per-vma work, via madvise_behavior struct. Make the per-vma logic not
flushes tlb on their own but just saves the tlb entries to the received
mmu_gather object. For this internal logic change, make
zap_page_range_single_batched() non-static and use it directly from
madvise_dontneed_single_vma(). Finally, the entry level functions flush
the tlb entries that gathered for the entire user request, at once.
Link: https://lkml.kernel.org/r/20250410000022.1901-5-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
MADV_FREE handling for [process_]madvise() flushes tlb for each vma of
each address range. Update the logic to do tlb flushes in a batched way.
Initialize an mmu_gather object from do_madvise() and vector_madvise(),
which are the entry level functions for [process_]madvise(), respectively.
And pass those objects to the function for per-vma work, via
madvise_behavior struct. Make the per-vma logic not flushes tlb on their
own but just saves the tlb entries to the received mmu_gather object.
Finally, the entry level functions flush the tlb entries that gathered for
the entire user request, at once.
Link: https://lkml.kernel.org/r/20250410000022.1901-3-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "mm/madvise: batch tlb flushes for MADV_DONTNEED and
MADV_FREE", v3.
When process_madvise() is called to do MADV_DONTNEED[_LOCKED] or MADV_FREE
with multiple address ranges, tlb flushes happen for each of the given
address ranges. Because such tlb flushes are for the same process, doing
those in a batch is more efficient while still being safe. Modify
process_madvise() entry level code path to do such batched tlb flushes,
while the internal unmap logic do only gathering of the tlb entries to
flush.
In more detail, modify the entry functions to initialize an mmu_gather
object and pass it to the internal logic. And make the internal logic do
only gathering of the tlb entries to flush into the received mmu_gather
object. After all internal function calls are done, the entry functions
flush the gathered tlb entries at once.
Because process_madvise() and madvise() share the internal unmap logic,
make same change to madvise() entry code together, to make code consistent
and cleaner. It is only for keeping the code clean, and shouldn't degrade
madvise(). It could rather provide a potential tlb flushes reduction
benefit for a case that there are multiple vmas for the given address
range. It is only a side effect from an effort to keep code clean, so we
don't measure it separately.
Similar optimizations might be applicable to other madvise behavior such
as MADV_COLD and MADV_PAGEOUT. Those are simply out of the scope of this
patch series, though.
Patches Sequence
================
The first patch defines a new data structure for managing information that
is required for batched tlb flushes (mmu_gather and behavior), and update
code paths for MADV_DONTNEED[_LOCKED] and MADV_FREE handling internal
logic to receive it.
The second patch batches tlb flushes for MADV_FREE handling for both
madvise() and process_madvise().
Remaining two patches are for MADV_DONTNEED[_LOCKED] tlb flushes batching.
The third patch splits zap_page_range_single() for batching of
MADV_DONTNEED[_LOCKED] handling. The fourth patch batches tlb flushes for
the hint using the sub-logic that the third patch split out, and the
helpers for batched tlb flushes that introduced for the MADV_FREE case, by
the second patch.
Test Results
============
I measured the latency to apply MADV_DONTNEED advice to 256 MiB memory
using multiple process_madvise() calls. I apply the advice in 4 KiB sized
regions granularity, but with varying batch size per process_madvise()
call (vlen) from 1 to 1024. The source code for the measurement is
available at GitHub[1]. To reduce measurement errors, I did the
measurement five times.
The measurement results are as below. 'sz_batch' column shows the batch
size of process_madvise() calls. 'Before' and 'After' columns show the
average of latencies in nanoseconds that measured five times on kernels
that built without and with the tlb flushes batching of this series
(patches 3 and 4), respectively. For the baseline, mm-new tree of
2025-04-09[2] has been used, after reverting the second version of this
patch series and adding a temporal fix for !CONFIG_DEBUG_VM build
failure[3]. 'B-stdev' and 'A-stdev' columns show ratios of latency
measurements standard deviation to average in percent for 'Before' and
'After', respectively. 'Latency_reduction' shows the reduction of the
latency that the 'After' has achieved compared to 'Before', in percent.
Higher 'Latency_reduction' values mean more efficiency improvements.
sz_batch Before B-stdev After A-stdev Latency_reduction
1 146386348 2.78 111327360.6 3.13 23.95
2 108222130 1.54 72131173.6 2.39 33.35
4 93617846.8 2.76 51859294.4 2.50 44.61
8 80555150.4 2.38 44328790 1.58 44.97
16 77272777 1.62 37489433.2 1.16 51.48
32 76478465.2 2.75 33570506 3.48 56.10
64 75810266.6 1.15 27037652.6 1.61 64.34
128 73222748 3.86 25517629.4 3.30 65.15
256 72534970.8 2.31 25002180.4 0.94 65.53
512 71809392 5.12 24152285.4 2.41 66.37
1024 73281170.2 4.53 24183615 2.09 67.00
Unexpectedly the latency has reduced (improved) even with batch size one.
I think some of compiler optimizations have affected that, like also
observed with the first version of this patch series.
So, please focus on the proportion between the improvement and the batch
size. As expected, tlb flushes batching provides latency reduction that
proportional to the batch size. The efficiency gain ranges from about 33
percent with batch size 2, and up to 67 percent with batch size 1,024.
Please note that this is a very simple microbenchmark, so real efficiency
gain on real workload could be very different.
This patch (of 4):
To implement batched tlb flushes for MADV_DONTNEED[_LOCKED] and MADV_FREE,
an mmu_gather object in addition to the behavior integer need to be passed
to the internal logics. Using a struct can make it easy without
increasing the number of parameters of all code paths towards the internal
logic. Define a struct for the purpose and use it on the code path that
starts from madvise_do_behavior() and ends on madvise_dontneed_free().
Note that this changes madvise_walk_vmas() visitor type signature, too.
Specifically, it changes its 'arg' type from 'unsigned long' to the new
struct pointer.
Link: https://lkml.kernel.org/r/20250410000022.1901-1-sj@kernel.org
Link: https://lkml.kernel.org/r/20250410000022.1901-2-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Liam R. Howlett <howlett@gmail.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: SeongJae Park <sj@kernel.org>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Because madise_should_skip() logic is factored out, making
madvise_do_behavior() calculates 'len' on its own rather then receiving it
as a parameter makes code simpler. Remove the parameter.
Link: https://lkml.kernel.org/r/20250312164750.59215-5-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Cc: David Hildenbrand <david@redhat.com>
Cc: Liam R. Howlett <howlett@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The logic for checking if a given madvise() request for a single memory
range can skip real work, namely madvise_do_behavior(), is duplicated in
do_madvise() and vector_madvise(). Split out the logic to a function and
reuse it.
Link: https://lkml.kernel.org/r/20250312164750.59215-4-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Cc: David Hildenbrand <david@redhat.com>
Cc: Liam R. Howlett <howlett@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
madvise_do_behavior() has a long open-coded 'behavior' check for
MADV_POPULATE_{READ,WRITE}. It adds multiple layers[1] and make the code
arguably take longer time to read. Like is_memory_failure(), split out
the check to a separate function. This is not technically removing the
additional layer but discourage further extending the switch-case. Also
it makes madvise_do_behavior() code shorter and therefore easier to read.
[1] https://lore.kernel.org/bd6d0bf1-c79e-46bd-a810-9791efb9ad73@lucifer.local
Link: https://lkml.kernel.org/r/20250312164750.59215-3-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Cc: David Hildenbrand <david@redhat.com>
Cc: Liam R. Howlett <howlett@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "mm/madvise: cleanup requests validations and classifications".
Cleanup madvise entry level code for cleaner request validations and
classifications.
This patch (of 4):
To reduce redundant open-coded checks of CONFIG_MEMORY_FAILURE and
MADV_{HWPOISON,SOFT_OFFLINE} in madvise_[un]lock(), is_memory_failure() is
introduced. madvise_do_behavior() is still doing the same open-coded
check, though. Use is_memory_failure() instead.
To avoid build failure on !CONFIG_MEMORY_FAILURE case, implement an empty
madvise_inject_error() under the config. Also move the definition of
is_memory_failure() inside #ifdef CONFIG_MEMORY_FAILURE clause for
madvise_inject_error() definition, to reduce duplicated ifdef clauses.
Link: https://lkml.kernel.org/r/20250312164750.59215-1-sj@kernel.org
Link: https://lkml.kernel.org/r/20250312164750.59215-2-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Cc: David Hildenbrand <david@redhat.com>
Cc: Liam R. Howlett <howlett@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Let's reuse our new MM ownership tracking infrastructure for large folios
to make folio_likely_mapped_shared() never return false negatives -- never
indicating "not mapped shared" although the folio *is* mapped shared.
With that, we can rename it to folio_maybe_mapped_shared() and get rid of
the dependency on the mapcount of the first folio page.
The semantics are now arguably clearer: no mixture of "false negatives"
and "false positives", only the remaining possibility for "false
positives".
Thoroughly document the new semantics. We might now detect that a large
folio is "maybe mapped shared" although it *no longer* is -- but once was.
Now, if more than two MMs mapped a folio at the same time, and the MM
mapping the folio exclusively at the end is not one tracked in the two
folio MM slots, we will detect the folio as "maybe mapped shared".
For anonymous folios, usually (except weird corner cases) all PTEs that
target a "maybe mapped shared" folio are R/O. As soon as a child process
would write to them (iow, actively use them), we would CoW and effectively
replace these PTEs. Most cases (below) are not expected to really matter
with large anonymous folios for this reason.
Most importantly, there will be no change at all for:
* small folios
* hugetlb folios
* PMD-mapped PMD-sized THPs (single mapping)
This change has the potential to affect existing callers of
folio_likely_mapped_shared() -> folio_maybe_mapped_shared():
(1) fs/proc/task_mmu.c: no change (hugetlb)
(2) khugepaged counts PTEs that target shared folios towards
max_ptes_shared (default: HPAGE_PMD_NR / 2), meaning we could skip a
collapse where we would have previously collapsed. This only applies
to anonymous folios and is not expected to matter in practice.
Worth noting that this change sorts out case (A) documented in
commit 1bafe96e89f0 ("mm/khugepaged: replace page_mapcount() check by
folio_likely_mapped_shared()") by removing the possibility for "false
negatives".
(3) MADV_COLD / MADV_PAGEOUT / MADV_FREE will not try splitting
PTE-mapped THPs that are considered shared but not fully covered by
the requested range, consequently not processing them.
PMD-mapped PMD-sized THP are not affected, or when all PTEs are
covered. These functions are usually only called on anon/file folios
that are exclusively mapped most of the time (no other file mappings
or no fork()), so the "false negatives" are not expected to matter in
practice.
(4) mbind() / migrate_pages() / move_pages() will refuse to migrate
shared folios unless MPOL_MF_MOVE_ALL is effective (requires
CAP_SYS_NICE). We will now reject some folios that could be migrated.
Similar to (3), especially with MPOL_MF_MOVE_ALL, so this is not
expected to matter in practice.
Note that cpuset_migrate_mm_workfn() calls do_migrate_pages() with
MPOL_MF_MOVE_ALL.
(5) NUMA hinting
mm/migrate.c:migrate_misplaced_folio_prepare() will skip file
folios that are probably shared libraries (-> "mapped shared" and
executable). This check would have detected it as a shared library at
some point (at least 3 MMs mapping it), so detecting it afterwards
does not sound wrong (still a shared library). Not expected to
matter.
mm/memory.c:numa_migrate_check() will indicate TNF_SHARED in
MAP_SHARED file mappings when encountering a shared folio. Similar
reasoning, not expected to matter.
mm/mprotect.c:change_pte_range() will skip folios detected as
shared in CoW mappings. Similarly, this is not expected to matter in
practice, but if it would ever be a problem we could relax that check
a bit (e.g., basing it on the average page-mapcount in a folio),
because it was only an optimization when many (e.g., 288) processes
were mapping the same folios -- see commit 859d4adc3415 ("mm: numa: do
not trap faults on shared data section pages.")
(6) mm/rmap.c:folio_referenced_one() will skip exclusive swapbacked
folios in dying processes. Applies to anonymous folios only. Without
"false negatives", we'll now skip all actually shared ones. Skipping
ones that are actually exclusive won't really matter, it's a pure
optimization, and is not expected to matter in practice.
In theory, one can detect the problematic scenario: folio_mapcount() > 0
and no folio MM slot is occupied ("state unknown"). One could reset the
MM slots while doing an rmap walk, which migration / folio split already
do when setting everything up. Further, when batching PTEs we might
naturally learn about a owner (e.g., folio_mapcount() == nr_ptes) and
could update the owner. However, we'll defer that until the scenarios
where it would really matter are clear.
Link: https://lkml.kernel.org/r/20250303163014.1128035-15-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Cc: Andy Lutomirks^H^Hski <luto@kernel.org>
Cc: Borislav Betkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jann Horn <jannh@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Lance Yang <ioworker0@gmail.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Matthew Wilcow (Oracle) <willy@infradead.org>
Cc: Michal Koutn <mkoutny@suse.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: tejun heo <tj@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Zefan Li <lizefan.x@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "mm: permit guard regions for file-backed/shmem mappings".
The guard regions feature was initially implemented to support anonymous
mappings only, excluding shmem.
This was done so as to introduce the feature carefully and incrementally
and to be conservative when considering the various caveats and corner
cases that are applicable to file-backed mappings but not to anonymous
ones.
Now this feature has landed in 6.13, it is time to revisit this and to
extend this functionality to file-backed and shmem mappings.
In order to make this maximally useful, and since one may map file-backed
mappings read-only (for instance ELF images), we also remove the
restriction on read-only mappings and permit the establishment of guard
regions in any non-hugetlb, non-mlock()'d mapping.
It is permissible to permit the establishment of guard regions in
read-only mappings because the guard regions only reduce access to the
mapping, and when removed simply reinstate the existing attributes of the
underlying VMA, meaning no access violations can occur.
While the change in kernel code introduced in this series is small, the
majority of the effort here is spent in extending the testing to assert
that the feature works correctly across numerous file-backed mapping
scenarios.
Every single guard region self-test performed against anonymous memory
(which is relevant and not anon-only) has now been updated to also be
performed against shmem and a mapping of a file in the working directory.
This confirms that all cases also function correctly for file-backed guard
regions.
In addition a number of other tests are added for specific file-backed
mapping scenarios.
There are a number of other concerns that one might have with regard to
guard regions, addressed below:
Readahead
~~~~~~~~~
Readahead is a process through which the page cache is populated on the
assumption that sequential reads will occur, thus amortising I/O and,
through a clever use of the PG_readahead folio flag establishing during
major fault and checked upon minor fault, provides for asynchronous I/O to
occur as dat is processed, reducing I/O stalls as data is faulted in.
Guard regions do not alter this mechanism which operates at the folio and
fault level, but does of course prevent the faulting of folios that would
otherwise be mapped.
In the instance of a major fault prior to a guard region, synchronous
readahead will occur including populating folios in the page cache which
the guard regions will, in the case of the mapping in question, prevent
access to.
In addition, if PG_readahead is placed in a folio that is now
inaccessible, this will prevent asynchronous readahead from occurring as
it would otherwise do.
However, there are mechanisms for heuristically resetting this within
readahead regardless, which will 'recover' correct readahead behaviour.
Readahead presumes sequential data access, the presence of a guard region
clearly indicates that, at least in the guard region, no such sequential
access will occur, as it cannot occur there.
So this should have very little impact on any real workload. The far more
important point is as to whether readahead causes incorrect or
inappropriate mapping of ranges disallowed by the presence of guard
regions - this is not the case, as readahead does not 'pre-fault' memory
in this fashion.
At any rate, any mechanism which would attempt to do so would hit the
usual page fault paths, which correctly handle PTE markers as with
anonymous mappings.
Fault-Around
~~~~~~~~~~~~
The fault-around logic, in a similar vein to readahead, attempts to
improve efficiency with regard to file-backed memory mappings, however it
differs in that it does not try to fetch folios into the page cache that
are about to be accessed, but rather pre-maps a range of folios around the
faulting address.
Guard regions making use of PTE markers makes this relatively trivial, as
this case is already handled - see filemap_map_folio_range() and
filemap_map_order0_folio() - in both instances, the solution is to simply
keep the established page table mappings and let the fault handler take
care of PTE markers, as per the comment:
/*
* NOTE: If there're PTE markers, we'll leave them to be
* handled in the specific fault path, and it'll prohibit
* the fault-around logic.
*/
This works, as establishing guard regions results in page table mappings
with PTE markers, and clearing them removes them.
Truncation
~~~~~~~~~~
File truncation will not eliminate existing guard regions, as the
truncation operation will ultimately zap the range via
unmap_mapping_range(), which specifically excludes PTE markers.
Zapping
~~~~~~~
Zapping is, as with anonymous mappings, handled by zap_nonpresent_ptes(),
which specifically deals with guard entries, leaving them intact except in
instances such as process teardown or munmap() where they need to be
removed.
Reclaim
~~~~~~~
When reclaim is performed on file-backed folios, it ultimately invokes
try_to_unmap_one() via the rmap. If the folio is non-large, then
map_pte() will ultimately abort the operation for the guard region
mapping. If large, then check_pte() will determine that this is a
non-device private entry/device-exclusive entry 'swap' PTE and thus abort
the operation in that instance.
Therefore, no odd things happen in the instance of reclaim being attempted
upon a file-backed guard region.
Hole Punching
~~~~~~~~~~~~~
This updates the page cache and ultimately invokes unmap_mapping_range(),
which explicitly leaves PTE markers in place.
Because the establishment of guard regions zapped any existing mappings to
file-backed folios, once the guard regions are removed then the
hole-punched region will be faulted in as usual and everything will behave
as expected.
One thing to note with this series is that it now implies file-backed
VMAs which install guard regions will now have an anon_vma installed if
not already present (i.e. if not post-CoW MAP_PRIVATE).
I have audited kernel source for instances of vma->anon_vma checks and
found nowhere where this would be problematic for pure file-backed
mappings.
I also discussed (off-list) with Matthew who confirmed he can't see any
issue with this.
In effect, we treat these VMAs as if they are MAP_PRIVATE, only with 0
CoW'd pages. As a result, the rmap never has a reason to reference the
anon_vma from folios at any point and thus no unexpected or weird
behaviour results.
The anon_vma logic tries to avoid unnecessary anon_vma propagation on
fork so we ought to at least minimise overhead.
However, this is still overhead, and unwelcome overhead.
The whole reason we do this (in madvise_guard_install()) is to ensure
that fork _copies page tables_. Otherwise, in vma_needs_copy(),
nothing will indicate that we should do so.
This was already an unpleasant thing to have to do, but without a new
VMA flag we really have no reasonable means of ensuring this happens.
Going forward, I intend to add a new VMA flag, VM_MAYBE_GUARDED or
something like this.
This would have specific behaviour - for the purposes of merging, it
would be ignored. However on both split and merge, it will be
propagated. It is therefore 'sticky'.
This is to avoid having to traverse page tables to determine which
parts of a VMA contain guard regions and of course to maintain the
desirable qualities of guard regions - the lack of VMA propagation (+
thus slab allocations of VMAs).
Adding this flag and adjusting vma_needs_copy() to reference it would
resolve the issue.
However :) we have a VMA flag space issue, so it'd render this a 64-bit
feature only.
Having discussed with Matthew a plan by which to perhaps extend
available flags for 32-bit we may going forward be able to avoid this.
But this may be a longer term project.
In the meantime, we'd have to resort to the anon_vma hack for 32-bit,
using the flag for 64-bit. The issue with this however is if we do
then intend to allow the flag to enable /proc/$pid/maps visibility
(something this could allow), it would also end up being 64-bit only
which would be a pity.
Regardless - I wanted to highlight this behaviour as it is perhaps
somewhat surprising.
This patch (of 4):
There is no reason to disallow guard regions in file-backed mappings -
readahead and fault-around both function correctly in the presence of PTE
markers, equally other operations relating to memory-mapped files function
correctly.
Additionally, read-only mappings if introducing guard-regions, only
restrict the mapping further, which means there is no violation of any
access rights by permitting this to be so.
Removing this restriction allows for read-only mapped files (such as
executable files) correctly which would otherwise not be permitted.
Link: https://lkml.kernel.org/r/cover.1739469950.git.lorenzo.stoakes@oracle.com
Link: https://lkml.kernel.org/r/d885cb259174736c2830a5dfe07f81b214ef3faa.1739469950.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Jann Horn <jannh@google.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Kalesh Singh <kaleshsingh@google.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: "Paul E . McKenney" <paulmck@kernel.org>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Optimize redundant mmap lock operations from process_madvise() by directly
doing the mmap locking first, and then the remaining works for all ranges
in the loop.
[akpm@linux-foundation.org: update comment, per Lorenzo]
Link: https://lkml.kernel.org/r/20250206061517.2958-5-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Reviewed-by: Liam R. Howlett <howlett@gmail.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Split out the madvise behavior applying logic from do_madvise() to make
it easier to reuse from the following change.
Link: https://lkml.kernel.org/r/20250206061517.2958-4-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Liam R. Howlett <howlett@gmail.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Split out the madvise parameters validation logic from do_madvise(), for
easy reuse of the logic from a future change.
Link: https://lkml.kernel.org/r/20250206061517.2958-3-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Davidlohr Bueso <dave@stgolabs.net>
Reviewed-by: Liam R. Howlett <howlett@gmail.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "mm/madvise: remove redundant mmap_lock operations from
process_madvise()".
process_madvise() calls do_madvise() for each address range. Then, each
do_madvise() invocation holds and releases same mmap_lock. Optimize the
redundant lock operations by splitting do_madvise() internal logic
including the mmap_lock operations, and calling the small logic directly
from process_madvise() in a sequence that removes the redundant locking.
As a result of this change, process_madvise() becomes more efficient and
less racy in terms of its results and latency.
Note that the potential downside of this series is that other mmap_lock
holders may take more time due to the increased length of mmap_lock
critical section for process_madvise() calls. But there is maximum limit
in the kernel space (IOV_MAX), and userspace can control the critical
section length by setting the request size. Hence, the downside would be
limited and controllable.
Evaluation
==========
I measured the time to apply MADV_DONTNEED advice to 256 MiB memory using
multiple madvise() calls, 4 KiB per each call. I also do the same with
process_madvise(), but with varying batch size (vlen) from 1 to 1024. The
source code for the measurement is available at GitHub[1]. Because the
microbenchmark result is not that stable, I ran each configuration five
times and use the average.
The measurement results are as below. 'sz_batches' column shows the batch
size of process_madvise() calls. '0' batch size is for madvise() calls
case. 'before' and 'after' columns are the measured time to apply
MADV_DONTNEED to the 256 MiB memory buffer in nanoseconds, on kernels that
built without and with the last patch of this series, respectively. So
lower value means better efficiency. 'after/before' column is the ratio
of 'after' to 'before'.
sz_batches before after after/before
0 146294215.2 121280536.2 0.829017989769427
1 165851018.8 136305598.2 0.821855658085351
2 129469321.2 103740383.6 0.801273866569094
4 110369232.4 87835896.2 0.795836795182785
8 102906232.4 77420920.2 0.752344327397609
16 97551017.4 74959714.4 0.768415506038587
32 94809848.2 71200848.4 0.750985786305689
64 96087575.6 72593180 0.755489765942227
128 96154163.8 68517055.4 0.712575022154163
256 92901257.6 69054216.6 0.743307662177439
512 93646170.8 67053296.2 0.716028168874151
1024 92663219.2 70168196.8 0.75723892830177
Despite the unstable nature of the test program, the trend is as we
expect. The measurement shows this patchset reduces the process_madvise()
latency, proportional to the batching size. The latency gain was about
20% with the batch size 2, and it has increased to about 28% with the
batch size 512, since more number of mmap locking is reduced with larger
batch size.
Note that the standard devitation of the measurements for each sz_batches
configuration ranged from 1.9% to 7.2%. That is, this result is not very
stable. The average of the standard deviations for different batch sizes
were 4.62% and 4.70% for the 'before' and 'after' kernel measurements.
Also note that this patch has somehow decreased latencies of madvise() and
single batch size process_madvise(). Seems this code path is small enough
to significantly be affected by compiler optimizations including inlining
of split-out functions. Please focus on only the improvement amount that
changed by the batch size.
[1] https://github.com/sjp38/eval_proc_madvise
This patch (of 4):
Split out the madvise behavior-dependent mmap_lock operations from
do_madvise(), for easier reuse of the logic in an upcoming change.
[lorenzo.stoakes@oracle.com: fix madvise_[un]lock() issue]
Link: https://lkml.kernel.org/r/2f448f7b-1da7-4099-aa9e-0179d47fde40@lucifer.local
[akpm@linux-foundation.org: coding-style cleanups]
Link: https://lkml.kernel.org/r/20250206061517.2958-1-sj@kernel.org
Link: https://lkml.kernel.org/r/20250206061517.2958-2-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Davidlohr Bueso <dave@stgolabs.net>
Reviewed-by: Liam R. Howlett <howlett@gmail.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: SeongJae Park <sj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Add a sanity check to madvise_dontneed_free() to address a corner case in
madvise where a race condition causes the current vma being processed to
be backed by a different page size.
During a madvise(MADV_DONTNEED) call on a memory region registered with a
userfaultfd, there's a period of time where the process mm lock is
temporarily released in order to send a UFFD_EVENT_REMOVE and let
userspace handle the event. During this time, the vma covering the
current address range may change due to an explicit mmap done concurrently
by another thread.
If, after that change, the memory region, which was originally backed by
4KB pages, is now backed by hugepages, the end address is rounded down to
a hugepage boundary to avoid data loss (see "Fixes" below). This rounding
may cause the end address to be truncated to the same address as the
start.
Make this corner case follow the same semantics as in other similar cases
where the requested region has zero length (ie. return 0).
This will make madvise_walk_vmas() continue to the next vma in the range
(this time holding the process mm lock) which, due to the prev pointer
becoming stale because of the vma change, will be the same hugepage-backed
vma that was just checked before. The next time madvise_dontneed_free()
runs for this vma, if the start address isn't aligned to a hugepage
boundary, it'll return -EINVAL, which is also in line with the madvise
api.
From userspace perspective, madvise() will return EINVAL because the start
address isn't aligned according to the new vma alignment requirements
(hugepage), even though it was correctly page-aligned when the call was
issued.
Link: https://lkml.kernel.org/r/20250203075206.1452208-1-rcn@igalia.com
Fixes: 8ebe0a5eaaeb ("mm,madvise,hugetlb: fix unexpected data loss with MADV_DONTNEED on hugetlbfs")
Signed-off-by: Ricardo Cañuelo Navarro <rcn@igalia.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Florent Revest <revest@google.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Now in order to pursue high performance, applications mostly use some
high-performance user-mode memory allocators, such as jemalloc or
tcmalloc. These memory allocators use madvise(MADV_DONTNEED or MADV_FREE)
to release physical memory, but neither MADV_DONTNEED nor MADV_FREE will
release page table memory, which may cause huge page table memory usage.
The following are a memory usage snapshot of one process which actually
happened on our server:
VIRT: 55t
RES: 590g
VmPTE: 110g
In this case, most of the page table entries are empty. For such a PTE
page where all entries are empty, we can actually free it back to the
system for others to use.
As a first step, this commit aims to synchronously free the empty PTE
pages in madvise(MADV_DONTNEED) case. We will detect and free empty PTE
pages in zap_pte_range(), and will add zap_details.reclaim_pt to exclude
cases other than madvise(MADV_DONTNEED).
Once an empty PTE is detected, we first try to hold the pmd lock within
the pte lock. If successful, we clear the pmd entry directly (fast path).
Otherwise, we wait until the pte lock is released, then re-hold the pmd
and pte locks and loop PTRS_PER_PTE times to check pte_none() to re-detect
whether the PTE page is empty and free it (slow path).
For other cases such as madvise(MADV_FREE), consider scanning and freeing
empty PTE pages asynchronously in the future.
The following code snippet can show the effect of optimization:
mmap 50G
while (1) {
for (; i < 1024 * 25; i++) {
touch 2M memory
madvise MADV_DONTNEED 2M
}
}
As we can see, the memory usage of VmPTE is reduced:
before after
VIRT 50.0 GB 50.0 GB
RES 3.1 MB 3.1 MB
VmPTE 102640 KB 240 KB
[zhengqi.arch@bytedance.com: fix uninitialized symbol 'ptl']
Link: https://lkml.kernel.org/r/20241206112348.51570-1-zhengqi.arch@bytedance.com
Link: https://lore.kernel.org/linux-mm/224e6a4e-43b5-4080-bdd8-b0a6fb2f0853@stanley.mountain/
Link: https://lkml.kernel.org/r/92aba2b319a734913f18ba41e7d86a265f0b84e2.1733305182.git.zhengqi.arch@bytedance.com
Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jann Horn <jannh@google.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Peter Xu <peterx@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Cc: Zach O'Keefe <zokeefe@google.com>
Cc: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Implement a new lightweight guard page feature, that is regions of
userland virtual memory that, when accessed, cause a fatal signal to
arise.
Currently users must establish PROT_NONE ranges to achieve this.
However this is very costly memory-wise - we need a VMA for each and every
one of these regions AND they become unmergeable with surrounding VMAs.
In addition repeated mmap() calls require repeated kernel context switches
and contention of the mmap lock to install these ranges, potentially also
having to unmap memory if installed over existing ranges.
The lightweight guard approach eliminates the VMA cost altogether - rather
than establishing a PROT_NONE VMA, it operates at the level of page table
entries - establishing PTE markers such that accesses to them cause a
fault followed by a SIGSGEV signal being raised.
This is achieved through the PTE marker mechanism, which we have already
extended to provide PTE_MARKER_GUARD, which we installed via the generic
page walking logic which we have extended for this purpose.
These guard ranges are established with MADV_GUARD_INSTALL. If the range
in which they are installed contain any existing mappings, they will be
zapped, i.e. free the range and unmap memory (thus mimicking the
behaviour of MADV_DONTNEED in this respect).
Any existing guard entries will be left untouched. There is therefore no
nesting of guarded pages.
Guarded ranges are NOT cleared by MADV_DONTNEED nor MADV_FREE (in both
instances the memory range may be reused at which point a user would
expect guards to still be in place), but they are cleared via
MADV_GUARD_REMOVE, process teardown or unmapping of memory ranges.
The guard property can be removed from ranges via MADV_GUARD_REMOVE. The
ranges over which this is applied, should they contain non-guard entries,
will be untouched, with only guard entries being cleared.
We permit this operation on anonymous memory only, and only VMAs which are
non-special, non-huge and not mlock()'d (if we permitted this we'd have to
drop locked pages which would be rather counterintuitive).
Racing page faults can cause repeated attempts to install guard pages that
are interrupted, result in a zap, and this process can end up being
repeated. If this happens more than would be expected in normal
operation, we rescind locks and retry the whole thing, which avoids lock
contention in this scenario.
Link: https://lkml.kernel.org/r/6aafb5821bf209f277dfae0787abb2ef87a37542.1730123433.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Suggested-by: Vlastimil Babka <vbabka@suse.cz>
Suggested-by: Jann Horn <jannh@google.com>
Suggested-by: David Hildenbrand <david@redhat.com>
Suggested-by: Vlastimil Babka <vbabka@suse.cz>
Suggested-by: Jann Horn <jannh@google.com>
Suggested-by: David Hildenbrand <david@redhat.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Arnd Bergmann <arnd@kernel.org>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Chris Zankel <chris@zankel.net>
Cc: Helge Deller <deller@gmx.de>
Cc: James E.J. Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Jeff Xu <jeffxu@chromium.org>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Liam R. Howlett <Liam.Howlett@Oracle.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Paul E. McKenney <paulmck@kernel.org>
Cc: Richard Henderson <richard.henderson@linaro.org>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Shuah Khan <skhan@linuxfoundation.org>
Cc: Sidhartha Kumar <sidhartha.kumar@oracle.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The process_madvise() call was introduced in commit ecb8ac8b1f14
("mm/madvise: introduce process_madvise() syscall: an external memory
hinting API") as a means of performing madvise() operations on another
process.
However, as it provides the means by which to perform multiple madvise()
operations in a batch via an iovec, it is useful to utilise the same
interface for performing operations on the current process rather than a
remote one.
Commit 22af8caff7d1 ("mm/madvise: process_madvise() drop capability check
if same mm") removed the need for a caller invoking process_madvise() on
its own pidfd to possess the CAP_SYS_NICE capability, however this leaves
the restrictions on operation in place.
Resolve this by only applying the restriction on operations when accessing
a remote process.
Moving forward we plan to implement a simpler means of specifying this
condition other than needing to establish a self pidfd, perhaps in the
form of a sentinel pidfd.
Also take the opportunity to refactor the system call implementation
abstracting the vectorised operation.
Link: https://lkml.kernel.org/r/20240926151019.82902-1-lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Acked-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Christian Brauner <brauner@kernel.org>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Pedro Falcato <pedro.falcato@gmail.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
mm_access() can return NULL if the mm is not found, but this is handled
the same as an error in all callers, with some translating this into an
-ESRCH error.
Only proc_mem_open() returns NULL if no mm is found, however in this case
it is clearer and makes more sense to explicitly handle the error.
Additionally we take the opportunity to refactor the function to eliminate
unnecessary nesting.
Simplify things by simply returning -ESRCH if no mm is found - this both
eliminates confusing use of the IS_ERR_OR_NULL() macro, and simplifies
callers which would return -ESRCH by returning this error directly.
[lorenzo.stoakes@oracle.com: prefer neater pointer error comparison]
Link: https://lkml.kernel.org/r/2fae1834-749a-45e1-8594-5e5979cf7103@lucifer.local
Link: https://lkml.kernel.org/r/20240924201023.193135-1-lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Suggested-by: Arnd Bergmann <arnd@arndb.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull MM updates from Andrew Morton:
"Along with the usual shower of singleton patches, notable patch series
in this pull request are:
- "Align kvrealloc() with krealloc()" from Danilo Krummrich. Adds
consistency to the APIs and behaviour of these two core allocation
functions. This also simplifies/enables Rustification.
- "Some cleanups for shmem" from Baolin Wang. No functional changes -
mode code reuse, better function naming, logic simplifications.
- "mm: some small page fault cleanups" from Josef Bacik. No
functional changes - code cleanups only.
- "Various memory tiering fixes" from Zi Yan. A small fix and a
little cleanup.
- "mm/swap: remove boilerplate" from Yu Zhao. Code cleanups and
simplifications and .text shrinkage.
- "Kernel stack usage histogram" from Pasha Tatashin and Shakeel
Butt. This is a feature, it adds new feilds to /proc/vmstat such as
$ grep kstack /proc/vmstat
kstack_1k 3
kstack_2k 188
kstack_4k 11391
kstack_8k 243
kstack_16k 0
which tells us that 11391 processes used 4k of stack while none at
all used 16k. Useful for some system tuning things, but
partivularly useful for "the dynamic kernel stack project".
- "kmemleak: support for percpu memory leak detect" from Pavel
Tikhomirov. Teaches kmemleak to detect leaksage of percpu memory.
- "mm: memcg: page counters optimizations" from Roman Gushchin. "3
independent small optimizations of page counters".
- "mm: split PTE/PMD PT table Kconfig cleanups+clarifications" from
David Hildenbrand. Improves PTE/PMD splitlock detection, makes
powerpc/8xx work correctly by design rather than by accident.
- "mm: remove arch_make_page_accessible()" from David Hildenbrand.
Some folio conversions which make arch_make_page_accessible()
unneeded.
- "mm, memcg: cg2 memory{.swap,}.peak write handlers" fro David
Finkel. Cleans up and fixes our handling of the resetting of the
cgroup/process peak-memory-use detector.
- "Make core VMA operations internal and testable" from Lorenzo
Stoakes. Rationalizaion and encapsulation of the VMA manipulation
APIs. With a view to better enable testing of the VMA functions,
even from a userspace-only harness.
- "mm: zswap: fixes for global shrinker" from Takero Funaki. Fix
issues in the zswap global shrinker, resulting in improved
performance.
- "mm: print the promo watermark in zoneinfo" from Kaiyang Zhao. Fill
in some missing info in /proc/zoneinfo.
- "mm: replace follow_page() by folio_walk" from David Hildenbrand.
Code cleanups and rationalizations (conversion to folio_walk())
resulting in the removal of follow_page().
- "improving dynamic zswap shrinker protection scheme" from Nhat
Pham. Some tuning to improve zswap's dynamic shrinker. Significant
reductions in swapin and improvements in performance are shown.
- "mm: Fix several issues with unaccepted memory" from Kirill
Shutemov. Improvements to the new unaccepted memory feature,
- "mm/mprotect: Fix dax puds" from Peter Xu. Implements mprotect on
DAX PUDs. This was missing, although nobody seems to have notied
yet.
- "Introduce a store type enum for the Maple tree" from Sidhartha
Kumar. Cleanups and modest performance improvements for the maple
tree library code.
- "memcg: further decouple v1 code from v2" from Shakeel Butt. Move
more cgroup v1 remnants away from the v2 memcg code.
- "memcg: initiate deprecation of v1 features" from Shakeel Butt.
Adds various warnings telling users that memcg v1 features are
deprecated.
- "mm: swap: mTHP swap allocator base on swap cluster order" from
Chris Li. Greatly improves the success rate of the mTHP swap
allocation.
- "mm: introduce numa_memblks" from Mike Rapoport. Moves various
disparate per-arch implementations of numa_memblk code into generic
code.
- "mm: batch free swaps for zap_pte_range()" from Barry Song. Greatly
improves the performance of munmap() of swap-filled ptes.
- "support large folio swap-out and swap-in for shmem" from Baolin
Wang. With this series we no longer split shmem large folios into
simgle-page folios when swapping out shmem.
- "mm/hugetlb: alloc/free gigantic folios" from Yu Zhao. Nice
performance improvements and code reductions for gigantic folios.
- "support shmem mTHP collapse" from Baolin Wang. Adds support for
khugepaged's collapsing of shmem mTHP folios.
- "mm: Optimize mseal checks" from Pedro Falcato. Fixes an mprotect()
performance regression due to the addition of mseal().
- "Increase the number of bits available in page_type" from Matthew
Wilcox. Increases the number of bits available in page_type!
- "Simplify the page flags a little" from Matthew Wilcox. Many legacy
page flags are now folio flags, so the page-based flags and their
accessors/mutators can be removed.
- "mm: store zero pages to be swapped out in a bitmap" from Usama
Arif. An optimization which permits us to avoid writing/reading
zero-filled zswap pages to backing store.
- "Avoid MAP_FIXED gap exposure" from Liam Howlett. Fixes a race
window which occurs when a MAP_FIXED operqtion is occurring during
an unrelated vma tree walk.
- "mm: remove vma_merge()" from Lorenzo Stoakes. Major rotorooting of
the vma_merge() functionality, making ot cleaner, more testable and
better tested.
- "misc fixups for DAMON {self,kunit} tests" from SeongJae Park.
Minor fixups of DAMON selftests and kunit tests.
- "mm: memory_hotplug: improve do_migrate_range()" from Kefeng Wang.
Code cleanups and folio conversions.
- "Shmem mTHP controls and stats improvements" from Ryan Roberts.
Cleanups for shmem controls and stats.
- "mm: count the number of anonymous THPs per size" from Barry Song.
Expose additional anon THP stats to userspace for improved tuning.
- "mm: finish isolate/putback_lru_page()" from Kefeng Wang: more
folio conversions and removal of now-unused page-based APIs.
- "replace per-quota region priorities histogram buffer with
per-context one" from SeongJae Park. DAMON histogram
rationalization.
- "Docs/damon: update GitHub repo URLs and maintainer-profile" from
SeongJae Park. DAMON documentation updates.
- "mm/vdpa: correct misuse of non-direct-reclaim __GFP_NOFAIL and
improve related doc and warn" from Jason Wang: fixes usage of page
allocator __GFP_NOFAIL and GFP_ATOMIC flags.
- "mm: split underused THPs" from Yu Zhao. Improve THP=always policy.
This was overprovisioning THPs in sparsely accessed memory areas.
- "zram: introduce custom comp backends API" frm Sergey Senozhatsky.
Add support for zram run-time compression algorithm tuning.
- "mm: Care about shadow stack guard gap when getting an unmapped
area" from Mark Brown. Fix up the various arch_get_unmapped_area()
implementations to better respect guard areas.
- "Improve mem_cgroup_iter()" from Kinsey Ho. Improve the reliability
of mem_cgroup_iter() and various code cleanups.
- "mm: Support huge pfnmaps" from Peter Xu. Extends the usage of huge
pfnmap support.
- "resource: Fix region_intersects() vs add_memory_driver_managed()"
from Huang Ying. Fix a bug in region_intersects() for systems with
CXL memory.
- "mm: hwpoison: two more poison recovery" from Kefeng Wang. Teaches
a couple more code paths to correctly recover from the encountering
of poisoned memry.
- "mm: enable large folios swap-in support" from Barry Song. Support
the swapin of mTHP memory into appropriately-sized folios, rather
than into single-page folios"
* tag 'mm-stable-2024-09-20-02-31' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (416 commits)
zram: free secondary algorithms names
uprobes: turn xol_area->pages[2] into xol_area->page
uprobes: introduce the global struct vm_special_mapping xol_mapping
Revert "uprobes: use vm_special_mapping close() functionality"
mm: support large folios swap-in for sync io devices
mm: add nr argument in mem_cgroup_swapin_uncharge_swap() helper to support large folios
mm: fix swap_read_folio_zeromap() for large folios with partial zeromap
mm/debug_vm_pgtable: Use pxdp_get() for accessing page table entries
set_memory: add __must_check to generic stubs
mm/vma: return the exact errno in vms_gather_munmap_vmas()
memcg: cleanup with !CONFIG_MEMCG_V1
mm/show_mem.c: report alloc tags in human readable units
mm: support poison recovery from copy_present_page()
mm: support poison recovery from do_cow_fault()
resource, kunit: add test case for region_intersects()
resource: make alloc_free_mem_region() works for iomem_resource
mm: z3fold: deprecate CONFIG_Z3FOLD
vfio/pci: implement huge_fault support
mm/arm64: support large pfn mappings
mm/x86: support large pfn mappings
...
|
|
In commit 96cfe2c0fd23 ("mm/madvise: replace ptrace attach requirement for
process_madvise") process_madvise() was updated to require the caller to
possess the CAP_SYS_NICE capability to perform the operation, in addition
to a check against PTRACE_MODE_READ performed by mm_access().
The mm_access() function explicitly checks to see if the address space of
the process being referenced is the current one, in which case no check is
performed.
We, however, do not do this when checking the CAP_SYS_NICE capability. This
means that we insist on the caller possessing this capability in order to
perform madvise() operations on its own address space, which seems
nonsensical.
Simply add a check to allow for an invocation of this function with pidfd
set to the current process without elevation.
Link: https://lkml.kernel.org/r/20240913140628.77047-1-lorenzo.stoakes@oracle.com
Fixes: 96cfe2c0fd23 ("mm/madvise: replace ptrace attach requirement for process_madvise")
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@Oracle.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Replace can_modify_mm_madv() with a single vma variant, and associated
checks in madvise.
While we're at it, also invert the order of checks in:
if (unlikely(is_ro_anon(vma) && !can_modify_vma(vma))
Checking if we can modify the vma itself (through vm_flags) is certainly
cheaper than is_ro_anon() due to arch_vma_access_permitted() looking at
e.g pkeys registers (with extra branches) in some architectures.
This patch allows for partial madvise success when finding a sealed VMA,
which historically has been allowed in Linux.
Link: https://lkml.kernel.org/r/20240817-mseal-depessimize-v3-5-d8d2e037df30@gmail.com
Signed-off-by: Pedro Falcato <pedro.falcato@gmail.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@Oracle.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Jeff Xu <jeffxu@chromium.org>
Cc: Kees Cook <kees@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/crng/random
Pull random number generator updates from Jason Donenfeld:
"This adds getrandom() support to the vDSO.
First, it adds a new kind of mapping to mmap(2), MAP_DROPPABLE, which
lets the kernel zero out pages anytime under memory pressure, which
enables allocating memory that never gets swapped to disk but also
doesn't count as being mlocked.
Then, the vDSO implementation of getrandom() is introduced in a
generic manner and hooked into random.c.
Next, this is implemented on x86. (Also, though it's not ready for
this pull, somebody has begun an arm64 implementation already)
Finally, two vDSO selftests are added.
There are also two housekeeping cleanup commits"
* tag 'random-6.11-rc1-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/crng/random:
MAINTAINERS: add random.h headers to RNG subsection
random: note that RNDGETPOOL was removed in 2.6.9-rc2
selftests/vDSO: add tests for vgetrandom
x86: vdso: Wire up getrandom() vDSO implementation
random: introduce generic vDSO getrandom() implementation
mm: add MAP_DROPPABLE for designating always lazily freeable mappings
|
|
The vDSO getrandom() implementation works with a buffer allocated with a
new system call that has certain requirements:
- It shouldn't be written to core dumps.
* Easy: VM_DONTDUMP.
- It should be zeroed on fork.
* Easy: VM_WIPEONFORK.
- It shouldn't be written to swap.
* Uh-oh: mlock is rlimited.
* Uh-oh: mlock isn't inherited by forks.
- It shouldn't reserve actual memory, but it also shouldn't crash when
page faulting in memory if none is available
* Uh-oh: VM_NORESERVE means segfaults.
It turns out that the vDSO getrandom() function has three really nice
characteristics that we can exploit to solve this problem:
1) Due to being wiped during fork(), the vDSO code is already robust to
having the contents of the pages it reads zeroed out midway through
the function's execution.
2) In the absolute worst case of whatever contingency we're coding for,
we have the option to fallback to the getrandom() syscall, and
everything is fine.
3) The buffers the function uses are only ever useful for a maximum of
60 seconds -- a sort of cache, rather than a long term allocation.
These characteristics mean that we can introduce VM_DROPPABLE, which
has the following semantics:
a) It never is written out to swap.
b) Under memory pressure, mm can just drop the pages (so that they're
zero when read back again).
c) It is inherited by fork.
d) It doesn't count against the mlock budget, since nothing is locked.
e) If there's not enough memory to service a page fault, it's not fatal,
and no signal is sent.
This way, allocations used by vDSO getrandom() can use:
VM_DROPPABLE | VM_DONTDUMP | VM_WIPEONFORK | VM_NORESERVE
And there will be no problem with OOMing, crashing on overcommitment,
using memory when not in use, not wiping on fork(), coredumps, or
writing out to swap.
In order to let vDSO getrandom() use this, expose these via mmap(2) as
MAP_DROPPABLE.
Note that this involves removing the MADV_FREE special case from
sort_folio(), which according to Yu Zhao is unnecessary and will simply
result in an extra call to shrink_folio_list() in the worst case. The
chunk removed reenables the swapbacked flag, which we don't want for
VM_DROPPABLE, and we can't conditionalize it here because there isn't a
vma reference available.
Finally, the provided self test ensures that this is working as desired.
Cc: linux-mm@kvack.org
Acked-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
The soft hwpoison injector via madvise(MADV_HWPOISON) operates in a
synchrous way in a sense, the injector is also a process under test, and
should it have the poisoned page mapped in its address space, it should
get killed as much as in a real UE situation. Doing so align with what
the madvise(2) man page says: " "This operation may result in the calling
process receiving a SIGBUS and the page being unmapped."
Link: https://lkml.kernel.org/r/20240524215306.2705454-3-jane.chu@oracle.com
Signed-off-by: Jane Chu <jane.chu@oracle.com>
Reviewed-by: Oscar Salvador <oalvador@suse.de>
Acked-by: Miaohe Lin <linmiaohe@huawei.com>
Cc: Naoya Horiguchi <nao.horiguchi@gmail.com>
Cc: Oscar Salvador <osalvador@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The new mseal() is an syscall on 64 bit CPU, and with following signature:
int mseal(void addr, size_t len, unsigned long flags)
addr/len: memory range.
flags: reserved.
mseal() blocks following operations for the given memory range.
1> Unmapping, moving to another location, and shrinking the size,
via munmap() and mremap(), can leave an empty space, therefore can
be replaced with a VMA with a new set of attributes.
2> Moving or expanding a different VMA into the current location,
via mremap().
3> Modifying a VMA via mmap(MAP_FIXED).
4> Size expansion, via mremap(), does not appear to pose any specific
risks to sealed VMAs. It is included anyway because the use case is
unclear. In any case, users can rely on merging to expand a sealed VMA.
5> mprotect() and pkey_mprotect().
6> Some destructive madvice() behaviors (e.g. MADV_DONTNEED) for anonymous
memory, when users don't have write permission to the memory. Those
behaviors can alter region contents by discarding pages, effectively a
memset(0) for anonymous memory.
Following input during RFC are incooperated into this patch:
Jann Horn: raising awareness and providing valuable insights on the
destructive madvise operations.
Linus Torvalds: assisting in defining system call signature and scope.
Liam R. Howlett: perf optimization.
Theo de Raadt: sharing the experiences and insight gained from
implementing mimmutable() in OpenBSD.
Finally, the idea that inspired this patch comes from Stephen Röttger's
work in Chrome V8 CFI.
[jeffxu@chromium.org: add branch prediction hint, per Pedro]
Link: https://lkml.kernel.org/r/20240423192825.1273679-2-jeffxu@chromium.org
Link: https://lkml.kernel.org/r/20240415163527.626541-3-jeffxu@chromium.org
Signed-off-by: Jeff Xu <jeffxu@chromium.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Pedro Falcato <pedro.falcato@gmail.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Guenter Roeck <groeck@chromium.org>
Cc: Jann Horn <jannh@google.com>
Cc: Jeff Xu <jeffxu@google.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Jorge Lucangeli Obes <jorgelo@chromium.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Muhammad Usama Anjum <usama.anjum@collabora.com>
Cc: Pedro Falcato <pedro.falcato@gmail.com>
Cc: Stephen Röttger <sroettger@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Amer Al Shanawany <amer.shanawany@gmail.com>
Cc: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
All reclaim_pages() callers are setting 'ignore_references' parameter
'true'. In other words, the parameter is not really being used. Remove
the argument to make it simple.
Link: https://lkml.kernel.org/r/20240429224451.67081-4-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
This patch optimizes lazyfreeing with PTE-mapped mTHP[1] (Inspired by
David Hildenbrand[2]). We aim to avoid unnecessary folio splitting if the
large folio is fully mapped within the target range.
If a large folio is locked or shared, or if we fail to split it, we just
leave it in place and advance to the next PTE in the range. But note that
the behavior is changed; previously, any failure of this sort would cause
the entire operation to give up. As large folios become more common,
sticking to the old way could result in wasted opportunities.
On an Intel I5 CPU, lazyfreeing a 1GiB VMA backed by PTE-mapped folios of
the same size results in the following runtimes for madvise(MADV_FREE) in
seconds (shorter is better):
Folio Size | Old | New | Change
------------------------------------------
4KiB | 0.590251 | 0.590259 | 0%
16KiB | 2.990447 | 0.185655 | -94%
32KiB | 2.547831 | 0.104870 | -95%
64KiB | 2.457796 | 0.052812 | -97%
128KiB | 2.281034 | 0.032777 | -99%
256KiB | 2.230387 | 0.017496 | -99%
512KiB | 2.189106 | 0.010781 | -99%
1024KiB | 2.183949 | 0.007753 | -99%
2048KiB | 0.002799 | 0.002804 | 0%
[1] https://lkml.kernel.org/r/20231207161211.2374093-5-ryan.roberts@arm.com
[2] https://lore.kernel.org/linux-mm/20240214204435.167852-1-david@redhat.com
Link: https://lkml.kernel.org/r/20240418134435.6092-5-ioworker0@gmail.com
Signed-off-by: Lance Yang <ioworker0@gmail.com>
Reviewed-by: Ryan Roberts <ryan.roberts@arm.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Barry Song <21cnbao@gmail.com>
Cc: Jeff Xie <xiehuan09@gmail.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: Yin Fengwei <fengwei.yin@intel.com>
Cc: Zach O'Keefe <zokeefe@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
