In speculative fault path, while doing page table lookup, offset
is obtained at each level and value at that offset is read and
checks are perfomed on it, later to get next level offset we read
from previous level offset again. A concurrent page table reclaimation
operation could result in change in value at this offset, and we go
ahead and access it, this would result in reading an invalid entry.
Fix this by reading from previous level offset again and comparing
before performing next level access.
Bug: 221005439
Change-Id: I66b3d24ae79c7ee5ccce4ba7a94f028f4cf3fda0
Signed-off-by: Vijayanand Jitta <quic_vjitta@quicinc.com>
Introduce vma_can_speculate(), which allows speculative handling for
VMAs mapping supported file types.
From do_handle_mm_fault(), speculative handling will follow through
__handle_mm_fault(), handle_pte_fault() and do_fault().
At this point, we expect speculative faults to continue through one of:
- do_read_fault(), fully implemented;
- do_cow_fault(), which might abort if missing anon vmas,
- do_shared_fault(), not implemented yet
(would require ->page_mkwrite() changes).
vma_can_speculate() provides an early abort for the do_shared_fault() case,
limiting the time spent on trying that unimplemented case.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20210407014502.24091-31-michel@lespinasse.org/
Conflicts:
include/linux/vm_event_item.h
mm/vmstat.c
1. SPF_ATTEMPT_FILE is taken from https://lore.kernel.org/all/20210407014502.24091-36-michel@lespinasse.org/
since the patch posted upstream at the time had a different structure
with stats for anonymouse and file-backed pagefaults introduced in a
separate patch.
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: I3a28af63b41b649f02f8b73d53f6494ad114ee5a
Call the vm_ops->map_pages method within an rcu read locked section.
In the speculative case, verify the mmap sequence lock at the start of
the section. A match guarantees that the original vma is still valid
at that time, and that the associated vma->vm_file stays valid while
the vm_ops->map_pages() method is running.
Do not test vmf->pmd in the speculative case - we only speculate when
a page table already exists, and and this saves us from having to handle
synchronization around the vmf->pmd read.
Change xfs_filemap_map_pages() account for the fact that it can not
block anymore, as it is now running within an rcu read lock.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20210407014502.24091-28-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Id771c1e6fa9b883595a48d4df63f448a05916eda
In the speculative case, we want to avoid direct pmd checks (which
would require some extra synchronization to be safe), and rely on
pte_map_lock which will both lock the page table and verify that the
pmd has not changed from its initial value.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20210407014502.24091-27-michel@lespinasse.org/
Conflicts:
mm/memory.c
1. Merge conflict due to new vmf->prealloc_pte usage in finish_fault.
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: If6046592083eaf12caf5c51c3fbb287a4dfa1ace
In the speculative case, call the vm_ops->fault() method from within
an rcu read locked section, and verify the mmap sequence lock at the
start of the section. A match guarantees that the original vma is still
valid at that time, and that the associated vma->vm_file stays valid
while the vm_ops->fault() method is running.
Note that this implies that speculative faults can not sleep within
the vm_ops->fault method. We will only attempt to fetch existing pages
from the page cache during speculative faults; any miss (or prefetch)
will be handled by falling back to non-speculative fault handling.
The speculative handling case also does not preallocate page tables,
as it is always called with a pre-existing page table.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20210407014502.24091-25-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: I995ba94d8e96014ef83ac93fe5a4669afcde34b9
In handle_pte_fault(), allow speculative execution to proceed.
Use pte_spinlock() to validate the mmap sequence count when locking
the page table.
If speculative execution proceeds through do_wp_page(), ensure that we
end up in the wp_page_reuse() or wp_page_copy() paths, rather than
wp_pfn_shared() or wp_page_shared() (both unreachable as we only
handle anon vmas so far) or handle_userfault() (needs an explicit
abort to handle non-speculatively).
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20220128131006.67712-28-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Ia45d095ec7b8e23f1c5d68b7a7f572a3f6f6df97
Change wp_page_copy() to handle the speculative case. This involves
aborting speculative faults if they have to allocate an anon_vma,
read-locking the mmu_notifier_lock to avoid races with
mmu_notifier_register(), and using pte_map_lock() instead of
pte_offset_map_lock() to complete the page fault.
Also change call sites to clear vmf->pte after unmapping the page table,
in order to satisfy pte_map_lock()'s preconditions.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20220128131006.67712-27-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Icd2188e9facf5a7fea42000a2808bcda1ad6f0fc
Change handle_pte_fault() to allow speculative fault execution to proceed
through do_numa_page().
do_swap_page() does not implement speculative execution yet, so it
needs to abort with VM_FAULT_RETRY in that case.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20220128131006.67712-22-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: I0390331facc9ecd37534012abdd9f255ab5bbb12
in x86 fault handler, only attempt spf if the vma is anonymous.
In do_handle_mm_fault(), let speculative page faults proceed as long
as they fall into anonymous vmas. This enables the speculative
handling code in __handle_mm_fault() and do_anonymous_page().
In handle_pte_fault(), if vmf->pte is set (the original pte was not
pte_none), catch speculative faults and return VM_FAULT_RETRY as
those cases are not implemented yet. Also assert that do_fault()
is not reached in the speculative case.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20220128131006.67712-20-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: I875106fcfa1084f570c2bf8f24a129bdce55316b
Change do_anonymous_page() to handle the speculative case.
This involves aborting speculative faults if they have to allocate a new
anon_vma, and using pte_map_lock() instead of pte_offset_map_lock()
to complete the page fault.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20220128131006.67712-19-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: I5ad955323faabc142c21f62415db039ac889066a
pte_map_lock() and pte_spinlock() are used by fault handlers to ensure
the pte is mapped and locked before they commit the faulted page to the
mm's address space at the end of the fault.
The functions differ in their preconditions; pte_map_lock() expects
the pte to be unmapped prior to the call, while pte_spinlock() expects
it to be already mapped.
In the speculative fault case, the functions verify, after locking the pte,
that the mmap sequence count has not changed since the start of the fault,
and thus that no mmap lock writers have been running concurrently with
the fault. After that point the page table lock serializes any further
races with concurrent mmap lock writers.
If the mmap sequence count check fails, both functions will return false
with the pte being left unmapped and unlocked.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20220128131006.67712-18-michel@lespinasse.org/
Conflicts:
include/linux/mm.h
1. Fixed pte_map_lock and pte_spinlock macros not to fail when
CONFIG_SPECULATIVE_PAGE_FAULT=n
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Ibd7ccc2ead4fdf29f28c7657b312b2f677ac8836
The speculative path calls speculative_page_walk_begin() before walking
the page table tree to prevent page table reclamation. The logic is
otherwise similar to the non-speculative path, but with additional
restrictions: in the speculative path, we do not handle huge pages or
wiring new pages tables.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20220128131006.67712-17-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: If099534da8b0ac105bbaa5ea4714a6654032592a
Move the code that initializes vmf->pte and vmf->orig_pte from
handle_pte_fault() to its single call site in __handle_mm_fault().
This ensures vmf->pte is now initialized together with the higher levels
of the page table hierarchy. This also prepares for speculative page fault
handling, where the entire page table walk (higher levels down to ptes)
needs special care in the speculative case.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20220128131006.67712-16-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Id550086fe568331aa71c91468f8314faad993b20
Speculative page faults will use these to protect against races with
page table reclamation.
This could always be handled by disabling local IRQs as the fast GUP
code does; however speculative page faults do not need to protect
against races with THP page splitting, so a weaker rcu read lock is
sufficient in the MMU_GATHER_RCU_TABLE_FREE case.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://lore.kernel.org/all/20220128131006.67712-15-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: I3efe5fc6a5a49d537cf33e8093daeea42550077a
Add a new do_handle_mm_fault function, which extends the existing
handle_mm_fault() API by adding an mmap sequence count, to be used
in the FAULT_FLAG_SPECULATIVE case.
In the initial implementation, FAULT_FLAG_SPECULATIVE always fails
(by returning VM_FAULT_RETRY).
The existing handle_mm_fault() API is kept as a wrapper around
do_handle_mm_fault() so that we do not have to immediately update
every handle_mm_fault() call site.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Conflicts:
mm/memory.c
1. Trivial merge conflict due to folios.
Link: https://lore.kernel.org/all/20220128131006.67712-10-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Ic07b6d84af3e5d1fcc856e0968f1a6dd1544fa88
In do_anonymous_page(), we have separate cases for the zero page vs
allocating new anonymous pages. However, once the pte entry has been
computed, the rest of the handling (mapping and locking the page table,
checking that we didn't lose a race with another page fault handler, etc)
is identical between the two cases.
This change reduces the code duplication between the two cases.
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Conflicts:
mm/memory.c
1. Trivial merge conflict caused by folios in mem_cgroup_charge call.
Link: https://lore.kernel.org/all/20220128131006.67712-6-michel@lespinasse.org/
Bug: 161210518
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Ic19579571925878d632e43aa40b9f50cdf473ee6
This reverts commit 77dfeaa02d.
Throttling can now be done using hist triggers and synthetic events
Bug: 145972256
Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
Change-Id: I39c284040e2fdb815cda980f3a40ef188e59287c
When handling shmem page fault the THP with corrupted subpage could be
PMD mapped if certain conditions are satisfied. But kernel is supposed
to send SIGBUS when trying to map hwpoisoned page.
There are two paths which may do PMD map: fault around and regular
fault.
Before commit f9ce0be71d ("mm: Cleanup faultaround and finish_fault()
codepaths") the thing was even worse in fault around path. The THP
could be PMD mapped as long as the VMA fits regardless what subpage is
accessed and corrupted. After this commit as long as head page is not
corrupted the THP could be PMD mapped.
In the regular fault path the THP could be PMD mapped as long as the
corrupted page is not accessed and the VMA fits.
This loophole could be fixed by iterating every subpage to check if any
of them is hwpoisoned or not, but it is somewhat costly in page fault
path.
So introduce a new page flag called HasHWPoisoned on the first tail
page. It indicates the THP has hwpoisoned subpage(s). It is set if any
subpage of THP is found hwpoisoned by memory failure and after the
refcount is bumped successfully, then cleared when the THP is freed or
split.
The soft offline path doesn't need this since soft offline handler just
marks a subpage hwpoisoned when the subpage is migrated successfully.
But shmem THP didn't get split then migrated at all.
Link: https://lkml.kernel.org/r/20211020210755.23964-3-shy828301@gmail.com
Fixes: 800d8c63b2 ("shmem: add huge pages support")
Signed-off-by: Yang Shi <shy828301@gmail.com>
Reviewed-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Suggested-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Peter Xu <peterx@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix the coherency management of mmap'd data such that 3rd-party changes
become visible as soon as possible after the callback notification is
delivered by the fileserver. This is done by the following means:
(1) When we break a callback on a vnode specified by the CB.CallBack call
from the server, we queue a work item (vnode->cb_work) to go and
clobber all the PTEs mapping to that inode.
This causes the CPU to trip through the ->map_pages() and
->page_mkwrite() handlers if userspace attempts to access the page(s)
again.
(Ideally, this would be done in the service handler for CB.CallBack,
but the server is waiting for our reply before considering, and we
have a list of vnodes, all of which need breaking - and the process of
getting the mmap_lock and stripping the PTEs on all CPUs could be
quite slow.)
(2) Call afs_validate() from the ->map_pages() handler to check to see if
the file has changed and to get a new callback promise from the
server.
Also handle the fileserver telling us that it's dropping all callbacks,
possibly after it's been restarted by sending us a CB.InitCallBackState*
call by the following means:
(3) Maintain a per-cell list of afs files that are currently mmap'd
(cell->fs_open_mmaps).
(4) Add a work item to each server that is invoked if there are any open
mmaps when CB.InitCallBackState happens. This work item goes through
the aforementioned list and invokes the vnode->cb_work work item for
each one that is currently using this server.
This causes the PTEs to be cleared, causing ->map_pages() or
->page_mkwrite() to be called again, thereby calling afs_validate()
again.
I've chosen to simply strip the PTEs at the point of notification reception
rather than invalidate all the pages as well because (a) it's faster, (b)
we may get a notification for other reasons than the data being altered (in
which case we don't want to clobber the pagecache) and (c) we need to ask
the server to find out - and I don't want to wait for the reply before
holding up userspace.
This was tested using the attached test program:
#include <stdbool.h>
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <fcntl.h>
#include <sys/mman.h>
int main(int argc, char *argv[])
{
size_t size = getpagesize();
unsigned char *p;
bool mod = (argc == 3);
int fd;
if (argc != 2 && argc != 3) {
fprintf(stderr, "Format: %s <file> [mod]\n", argv[0]);
exit(2);
}
fd = open(argv[1], mod ? O_RDWR : O_RDONLY);
if (fd < 0) {
perror(argv[1]);
exit(1);
}
p = mmap(NULL, size, mod ? PROT_READ|PROT_WRITE : PROT_READ,
MAP_SHARED, fd, 0);
if (p == MAP_FAILED) {
perror("mmap");
exit(1);
}
for (;;) {
if (mod) {
p[0]++;
msync(p, size, MS_ASYNC);
fsync(fd);
}
printf("%02x", p[0]);
fflush(stdout);
sleep(1);
}
}
It runs in two modes: in one mode, it mmaps a file, then sits in a loop
reading the first byte, printing it and sleeping for a second; in the
second mode it mmaps a file, then sits in a loop incrementing the first
byte and flushing, then printing and sleeping.
Two instances of this program can be run on different machines, one doing
the reading and one doing the writing. The reader should see the changes
made by the writer, but without this patch, they aren't because validity
checking is being done lazily - only on entry to the filesystem.
Testing the InitCallBackState change is more complicated. The server has
to be taken offline, the saved callback state file removed and then the
server restarted whilst the reading-mode program continues to run. The
client machine then has to poke the server to trigger the InitCallBackState
call.
Signed-off-by: David Howells <dhowells@redhat.com>
Tested-by: Markus Suvanto <markus.suvanto@gmail.com>
cc: linux-afs@lists.infradead.org
Link: https://lore.kernel.org/r/163111668833.283156.382633263709075739.stgit@warthog.procyon.org.uk/
Some devices require exclusive write access to shared virtual memory (SVM)
ranges to perform atomic operations on that memory. This requires CPU
page tables to be updated to deny access whilst atomic operations are
occurring.
In order to do this introduce a new swap entry type
(SWP_DEVICE_EXCLUSIVE). When a SVM range needs to be marked for exclusive
access by a device all page table mappings for the particular range are
replaced with device exclusive swap entries. This causes any CPU access
to the page to result in a fault.
Faults are resovled by replacing the faulting entry with the original
mapping. This results in MMU notifiers being called which a driver uses
to update access permissions such as revoking atomic access. After
notifiers have been called the device will no longer have exclusive access
to the region.
Walking of the page tables to find the target pages is handled by
get_user_pages() rather than a direct page table walk. A direct page
table walk similar to what migrate_vma_collect()/unmap() does could also
have been utilised. However this resulted in more code similar in
functionality to what get_user_pages() provides as page faulting is
required to make the PTEs present and to break COW.
[dan.carpenter@oracle.com: fix signedness bug in make_device_exclusive_range()]
Link: https://lkml.kernel.org/r/YNIz5NVnZ5GiZ3u1@mwanda
Link: https://lkml.kernel.org/r/20210616105937.23201-8-apopple@nvidia.com
Signed-off-by: Alistair Popple <apopple@nvidia.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Ben Skeggs <bskeggs@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Cc: Peter Xu <peterx@redhat.com>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Shakeel Butt <shakeelb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "Add support for SVM atomics in Nouveau", v11.
Introduction
============
Some devices have features such as atomic PTE bits that can be used to
implement atomic access to system memory. To support atomic operations to
a shared virtual memory page such a device needs access to that page which
is exclusive of the CPU. This series introduces a mechanism to
temporarily unmap pages granting exclusive access to a device.
These changes are required to support OpenCL atomic operations in Nouveau
to shared virtual memory (SVM) regions allocated with the
CL_MEM_SVM_ATOMICS clSVMAlloc flag. A more complete description of the
OpenCL SVM feature is available at
https://www.khronos.org/registry/OpenCL/specs/3.0-unified/html/
OpenCL_API.html#_shared_virtual_memory .
Implementation
==============
Exclusive device access is implemented by adding a new swap entry type
(SWAP_DEVICE_EXCLUSIVE) which is similar to a migration entry. The main
difference is that on fault the original entry is immediately restored by
the fault handler instead of waiting.
Restoring the entry triggers calls to MMU notifers which allows a device
driver to revoke the atomic access permission from the GPU prior to the
CPU finalising the entry.
Patches
=======
Patches 1 & 2 refactor existing migration and device private entry
functions.
Patches 3 & 4 rework try_to_unmap_one() by splitting out unrelated
functionality into separate functions - try_to_migrate_one() and
try_to_munlock_one().
Patch 5 renames some existing code but does not introduce functionality.
Patch 6 is a small clean-up to swap entry handling in copy_pte_range().
Patch 7 contains the bulk of the implementation for device exclusive
memory.
Patch 8 contains some additions to the HMM selftests to ensure everything
works as expected.
Patch 9 is a cleanup for the Nouveau SVM implementation.
Patch 10 contains the implementation of atomic access for the Nouveau
driver.
Testing
=======
This has been tested with upstream Mesa 21.1.0 and a simple OpenCL program
which checks that GPU atomic accesses to system memory are atomic.
Without this series the test fails as there is no way of write-protecting
the page mapping which results in the device clobbering CPU writes. For
reference the test is available at
https://ozlabs.org/~apopple/opencl_svm_atomics/
Further testing has been performed by adding support for testing exclusive
access to the hmm-tests kselftests.
This patch (of 10):
Remove multiple similar inline functions for dealing with different types
of special swap entries.
Both migration and device private swap entries use the swap offset to
store a pfn. Instead of multiple inline functions to obtain a struct page
for each swap entry type use a common function pfn_swap_entry_to_page().
Also open-code the various entry_to_pfn() functions as this results is
shorter code that is easier to understand.
Link: https://lkml.kernel.org/r/20210616105937.23201-1-apopple@nvidia.com
Link: https://lkml.kernel.org/r/20210616105937.23201-2-apopple@nvidia.com
Signed-off-by: Alistair Popple <apopple@nvidia.com>
Reviewed-by: Ralph Campbell <rcampbell@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Ben Skeggs <bskeggs@redhat.com>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pach series "mm: thp: use generic THP migration for NUMA hinting fault", v3.
When the THP NUMA fault support was added THP migration was not supported
yet. So the ad hoc THP migration was implemented in NUMA fault handling.
Since v4.14 THP migration has been supported so it doesn't make too much
sense to still keep another THP migration implementation rather than using
the generic migration code. It is definitely a maintenance burden to keep
two THP migration implementation for different code paths and it is more
error prone. Using the generic THP migration implementation allows us
remove the duplicate code and some hacks needed by the old ad hoc
implementation.
A quick grep shows x86_64, PowerPC (book3s), ARM64 ans S390 support both
THP and NUMA balancing. The most of them support THP migration except for
S390. Zi Yan tried to add THP migration support for S390 before but it
was not accepted due to the design of S390 PMD. For the discussion,
please see: https://lkml.org/lkml/2018/4/27/953.
Per the discussion with Gerald Schaefer in v1 it is acceptible to skip
huge PMD for S390 for now.
I saw there were some hacks about gup from git history, but I didn't
figure out if they have been removed or not since I just found FOLL_NUMA
code in the current gup implementation and they seems useful.
Patch #1 ~ #2 are preparation patches.
Patch #3 is the real meat.
Patch #4 ~ #6 keep consistent counters and behaviors with before.
Patch #7 skips change huge PMD to prot_none if thp migration is not supported.
Test
----
Did some tests to measure the latency of do_huge_pmd_numa_page. The test
VM has 80 vcpus and 64G memory. The test would create 2 processes to
consume 128G memory together which would incur memory pressure to cause
THP splits. And it also creates 80 processes to hog cpu, and the memory
consumer processes are bound to different nodes periodically in order to
increase NUMA faults.
The below test script is used:
echo 3 > /proc/sys/vm/drop_caches
# Run stress-ng for 24 hours
./stress-ng/stress-ng --vm 2 --vm-bytes 64G --timeout 24h &
PID=$!
./stress-ng/stress-ng --cpu $NR_CPUS --timeout 24h &
# Wait for vm stressors forked
sleep 5
PID_1=`pgrep -P $PID | awk 'NR == 1'`
PID_2=`pgrep -P $PID | awk 'NR == 2'`
JOB1=`pgrep -P $PID_1`
JOB2=`pgrep -P $PID_2`
# Bind load jobs to different nodes periodically to force generate
# cross node memory access
while [ -d "/proc/$PID" ]
do
taskset -apc 8 $JOB1
taskset -apc 8 $JOB2
sleep 300
taskset -apc 58 $JOB1
taskset -apc 58 $JOB2
sleep 300
done
With the above test the histogram of latency of do_huge_pmd_numa_page is
as shown below. Since the number of do_huge_pmd_numa_page varies
drastically for each run (should be due to scheduler), so I converted the
raw number to percentage.
patched base
@us[stress-ng]:
[0] 3.57% 0.16%
[1] 55.68% 18.36%
[2, 4) 10.46% 40.44%
[4, 8) 7.26% 17.82%
[8, 16) 21.12% 13.41%
[16, 32) 1.06% 4.27%
[32, 64) 0.56% 4.07%
[64, 128) 0.16% 0.35%
[128, 256) < 0.1% < 0.1%
[256, 512) < 0.1% < 0.1%
[512, 1K) < 0.1% < 0.1%
[1K, 2K) < 0.1% < 0.1%
[2K, 4K) < 0.1% < 0.1%
[4K, 8K) < 0.1% < 0.1%
[8K, 16K) < 0.1% < 0.1%
[16K, 32K) < 0.1% < 0.1%
[32K, 64K) < 0.1% < 0.1%
Per the result, patched kernel is even slightly better than the base
kernel. I think this is because the lock contention against THP split is
less than base kernel due to the refactor.
To exclude the affect from THP split, I also did test w/o memory pressure.
No obvious regression is spotted. The below is the test result *w/o*
memory pressure.
patched base
@us[stress-ng]:
[0] 7.97% 18.4%
[1] 69.63% 58.24%
[2, 4) 4.18% 2.63%
[4, 8) 0.22% 0.17%
[8, 16) 1.03% 0.92%
[16, 32) 0.14% < 0.1%
[32, 64) < 0.1% < 0.1%
[64, 128) < 0.1% < 0.1%
[128, 256) < 0.1% < 0.1%
[256, 512) 0.45% 1.19%
[512, 1K) 15.45% 17.27%
[1K, 2K) < 0.1% < 0.1%
[2K, 4K) < 0.1% < 0.1%
[4K, 8K) < 0.1% < 0.1%
[8K, 16K) 0.86% 0.88%
[16K, 32K) < 0.1% 0.15%
[32K, 64K) < 0.1% < 0.1%
[64K, 128K) < 0.1% < 0.1%
[128K, 256K) < 0.1% < 0.1%
The series also survived a series of tests that exercise NUMA balancing
migrations by Mel.
This patch (of 7):
Add orig_pmd to struct vm_fault so the "orig_pmd" parameter used by huge
page fault could be removed, just like its PTE counterpart does.
Link: https://lkml.kernel.org/r/20210518200801.7413-1-shy828301@gmail.com
Link: https://lkml.kernel.org/r/20210518200801.7413-2-shy828301@gmail.com
Signed-off-by: Yang Shi <shy828301@gmail.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Gerald Schaefer <gerald.schaefer@linux.ibm.com>
Cc: Heiko Carstens <hca@linux.ibm.com>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We tried to do something similar in b569a17607 ("userfaultfd: wp: drop
_PAGE_UFFD_WP properly when fork") previously, but it's not doing it all
right.. A few fixes around the code path:
1. We were referencing VM_UFFD_WP vm_flags on the _old_ vma rather
than the new vma. That's overlooked in b569a17607, so it won't work
as expected. Thanks to the recent rework on fork code
(7a4830c380), we can easily get the new vma now, so switch the
checks to that.
2. Dropping the uffd-wp bit in copy_huge_pmd() could be wrong if the
huge pmd is a migration huge pmd. When it happens, instead of using
pmd_uffd_wp(), we should use pmd_swp_uffd_wp(). The fix is simply to
handle them separately.
3. Forget to carry over uffd-wp bit for a write migration huge pmd
entry. This also happens in copy_huge_pmd(), where we converted a
write huge migration entry into a read one.
4. In copy_nonpresent_pte(), drop uffd-wp if necessary for swap ptes.
5. In copy_present_page() when COW is enforced when fork(), we also
need to pass over the uffd-wp bit if VM_UFFD_WP is armed on the new
vma, and when the pte to be copied has uffd-wp bit set.
Remove the comment in copy_present_pte() about this. It won't help a huge
lot to only comment there, but comment everywhere would be an overkill.
Let's assume the commit messages would help.
[peterx@redhat.com: fix a few thp pmd missing uffd-wp bit]
Link: https://lkml.kernel.org/r/20210428225030.9708-4-peterx@redhat.com
Link: https://lkml.kernel.org/r/20210428225030.9708-3-peterx@redhat.com
Fixes: b569a17607 ("userfaultfd: wp: drop _PAGE_UFFD_WP properly when fork")
Signed-off-by: Peter Xu <peterx@redhat.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: Mike Rapoport <rppt@linux.vnet.ibm.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Axel Rasmussen <axelrasmussen@google.com>
Cc: Brian Geffon <bgeffon@google.com>
Cc: "Dr . David Alan Gilbert" <dgilbert@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Joe Perches <joe@perches.com>
Cc: Kirill A. Shutemov <kirill@shutemov.name>
Cc: Lokesh Gidra <lokeshgidra@google.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Mina Almasry <almasrymina@google.com>
Cc: Oliver Upton <oupton@google.com>
Cc: Shaohua Li <shli@fb.com>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Wang Qing <wangqing@vivo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
With commit 09854ba94c ("mm: do_wp_page() simplification"), after COW,
the idle swap cache page (neither the page nor the corresponding swap
entry is mapped by any process) will be left in the LRU list, even if it's
in the active list or the head of the inactive list. So, the page
reclaimer may take quite some overhead to reclaim these actually unused
pages.
To help the page reclaiming, in this patch, after COW, the idle swap cache
page will be tried to be freed. To avoid to introduce much overhead to
the hot COW code path,
a) there's almost zero overhead for non-swap case via checking
PageSwapCache() firstly.
b) the page lock is acquired via trylock only.
To test the patch, we used pmbench memory accessing benchmark with
working-set larger than available memory on a 2-socket Intel server with a
NVMe SSD as swap device. Test results shows that the pmbench score
increases up to 23.8% with the decreased size of swap cache and swapin
throughput.
Link: https://lkml.kernel.org/r/20210601053143.1380078-1-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Suggested-by: Johannes Weiner <hannes@cmpxchg.org> [use free_swap_cache()]
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Peter Xu <peterx@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@surriel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Tim Chen <tim.c.chen@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There is a race between THP unmapping and truncation, when truncate sees
pmd_none() and skips the entry, after munmap's zap_huge_pmd() cleared
it, but before its page_remove_rmap() gets to decrement
compound_mapcount: generating false "BUG: Bad page cache" reports that
the page is still mapped when deleted. This commit fixes that, but not
in the way I hoped.
The first attempt used try_to_unmap(page, TTU_SYNC|TTU_IGNORE_MLOCK)
instead of unmap_mapping_range() in truncate_cleanup_page(): it has
often been an annoyance that we usually call unmap_mapping_range() with
no pages locked, but there apply it to a single locked page.
try_to_unmap() looks more suitable for a single locked page.
However, try_to_unmap_one() contains a VM_BUG_ON_PAGE(!pvmw.pte,page):
it is used to insert THP migration entries, but not used to unmap THPs.
Copy zap_huge_pmd() and add THP handling now? Perhaps, but their TLB
needs are different, I'm too ignorant of the DAX cases, and couldn't
decide how far to go for anon+swap. Set that aside.
The second attempt took a different tack: make no change in truncate.c,
but modify zap_huge_pmd() to insert an invalidated huge pmd instead of
clearing it initially, then pmd_clear() between page_remove_rmap() and
unlocking at the end. Nice. But powerpc blows that approach out of the
water, with its serialize_against_pte_lookup(), and interesting pgtable
usage. It would need serious help to get working on powerpc (with a
minor optimization issue on s390 too). Set that aside.
Just add an "if (page_mapped(page)) synchronize_rcu();" or other such
delay, after unmapping in truncate_cleanup_page()? Perhaps, but though
that's likely to reduce or eliminate the number of incidents, it would
give less assurance of whether we had identified the problem correctly.
This successful iteration introduces "unmap_mapping_page(page)" instead
of try_to_unmap(), and goes the usual unmap_mapping_range_tree() route,
with an addition to details. Then zap_pmd_range() watches for this
case, and does spin_unlock(pmd_lock) if so - just like
page_vma_mapped_walk() now does in the PVMW_SYNC case. Not pretty, but
safe.
Note that unmap_mapping_page() is doing a VM_BUG_ON(!PageLocked) to
assert its interface; but currently that's only used to make sure that
page->mapping is stable, and zap_pmd_range() doesn't care if the page is
locked or not. Along these lines, in invalidate_inode_pages2_range()
move the initial unmap_mapping_range() out from under page lock, before
then calling unmap_mapping_page() under page lock if still mapped.
Link: https://lkml.kernel.org/r/a2a4a148-cdd8-942c-4ef8-51b77f643dbe@google.com
Fixes: fc127da085 ("truncate: handle file thp")
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Jue Wang <juew@google.com>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Peter Xu <peterx@redhat.com>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Wang Yugui <wangyugui@e16-tech.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
One giant leap, all the way up to 5.13-rc1
Also take the opportunity to re-align (a.k.a. fix a couple of previous
merge conflict fix-up issues) which occurred during this merge-window.
Fixes: 4797acfb9c ("Merge 16b3d0cf5b Merge tag 'sched-core-2021-04-28' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip into android-mainline")
Fixes: 92f282f338 ("Merge 8ca5297e7e Merge tag 'kconfig-v5.13' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild into android-mainline")
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Change-Id: Ie9389f595776e8f66bba6eaf0fa7a3587c6a5749
When I was implementing a latency analyzer tool by using task->delays
and other things, I found an issue in delayacct. The issue is it should
clear the target's flag instead of current's in delayacct_blkio_end().
When I git blame delayacct, I found there're some similar issues we have
fixed in delayacct_blkio_end().
- Commit c96f5471ce ("delayacct: Account blkio completion on the
correct task") fixed the issue that it should account blkio
completion on the target task instead of current.
- Commit b512719f77 ("delayacct: fix crash in delayacct_blkio_end()
after delayacct init failure") fixed the issue that it should check
target task's delays instead of current task'.
It seems that delayacct_blkio_{begin, end} are error prone.
So I introduce a new paratmeter - the target task 'p' - to these
helpers. After that change, the callsite will specifilly set the right
task, which should make it less error prone.
Link: https://lkml.kernel.org/r/20210414083720.24083-1-laoar.shao@gmail.com
Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Josh Snyder <joshs@netflix.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
