kernel.git/mm/shmem.c, branch linux-5.5.y

mm/shmem.c: thp, shmem: fix conflict of above-47bit hint address and PMD alignment

2020-01-14T02:19:01Z

Shmem/tmpfs tries to provide THP-friendly mappings if huge pages are enabled. But it doesn't work well with above-47bit hint address. Normally, the kernel doesn't create userspace mappings above 47-bit, even if the machine allows this (such as with 5-level paging on x86-64). Not all user space is ready to handle wide addresses. It's known that at least some JIT compilers use higher bits in pointers to encode their information. Userspace can ask for allocation from full address space by specifying hint address (with or without MAP_FIXED) above 47-bits. If the application doesn't need a particular address, but wants to allocate from whole address space it can specify -1 as a hint address. Unfortunately, this trick breaks THP alignment in shmem/tmp: shmem_get_unmapped_area() would not try to allocate PMD-aligned area if *any* hint address specified. This can be fixed by requesting the aligned area if the we failed to allocated at user-specified hint address. The request with inflated length will also take the user-specified hint address. This way we will not lose an allocation request from the full address space. [kirill@shutemov.name: fold in a fixup] Link: http://lkml.kernel.org/r/20191223231309.t6bh5hkbmokihpfu@box Link: http://lkml.kernel.org/r/20191220142548.7118-3-kirill.shutemov@linux.intel.com Fixes: b569bab78d8d ("x86/mm: Prepare to expose larger address space to userspace") Signed-off-by: Kirill A. Shutemov Cc: "Willhalm, Thomas" Cc: Dan Williams Cc: "Bruggeman, Otto G" Cc: "Aneesh Kumar K . V" Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

mm/shmem.c: cast the type of unmap_start to u64

2019-12-01T20:59:10Z

In 64bit system. sb->s_maxbytes of shmem filesystem is MAX_LFS_FILESIZE, which equal LLONG_MAX. If offset > LLONG_MAX - PAGE_SIZE, offset + len < LLONG_MAX in shmem_fallocate, which will pass the checking in vfs_fallocate. /* Check for wrap through zero too */ if (((offset + len) > inode->i_sb->s_maxbytes) || ((offset + len) < 0)) return -EFBIG; loff_t unmap_start = round_up(offset, PAGE_SIZE) in shmem_fallocate causes a overflow. Syzkaller reports a overflow problem in mm/shmem: UBSAN: Undefined behaviour in mm/shmem.c:2014:10 signed integer overflow: '9223372036854775807 + 1' cannot be represented in type 'long long int' CPU: 0 PID:17076 Comm: syz-executor0 Not tainted 4.1.46+ #1 Hardware name: linux, dummy-virt (DT) Call trace: dump_backtrace+0x0/0x2c8 arch/arm64/kernel/traps.c:100 show_stack+0x20/0x30 arch/arm64/kernel/traps.c:238 __dump_stack lib/dump_stack.c:15 [inline] ubsan_epilogue+0x18/0x70 lib/ubsan.c:164 handle_overflow+0x158/0x1b0 lib/ubsan.c:195 shmem_fallocate+0x6d0/0x820 mm/shmem.c:2104 vfs_fallocate+0x238/0x428 fs/open.c:312 SYSC_fallocate fs/open.c:335 [inline] SyS_fallocate+0x54/0xc8 fs/open.c:239 The highest bit of unmap_start will be appended with sign bit 1 (overflow) when calculate shmem_falloc.start: shmem_falloc.start = unmap_start >> PAGE_SHIFT. Fix it by casting the type of unmap_start to u64, when right shifted. This bug is found in LTS Linux 4.1. It also seems to exist in mainline. Link: http://lkml.kernel.org/r/1573867464-5107-1-git-send-email-chenjun102@huawei.com Signed-off-by: Chen Jun Reviewed-by: Andrew Morton Cc: Hugh Dickins Cc: Qian Cai Cc: Kefeng Wang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

mm: shmem: use proper gfp flags for shmem_writepage()

2019-12-01T20:59:10Z

The shmem_writepage() uses GFP_ATOMIC to allocate swap cache. GFP_ATOMIC used to mean __GFP_HIGH, but now it means __GFP_HIGH | __GFP_ATOMIC | __GFP_KSWAPD_RECLAIM. However, shmem_writepage() should write out to swap only in response to memory pressure, so __GFP_KSWAPD_RECLAIM looks useless since the caller may be kswapd itself or in direct reclaim already. In addition, XArray node allocations from PF_MEMALLOC contexts could completely exhaust the page allocator, __GFP_NOMEMALLOC stops emergency reserves from being allocated. Here just copy the gfp flags used by add_to_swap(). Hugh: "a cleanup to make the two calls look the same when they don't need to be different (whereas the call from __read_swap_cache_async() rightly uses a lower priority gfp)". Link: http://lkml.kernel.org/r/1572991351-86061-1-git-send-email-yang.shi@linux.alibaba.com Signed-off-by: Yang Shi Acked-by: Hugh Dickins Cc: Michal Hocko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

mm/shmem.c: make array 'values' static const, makes object smaller

2019-12-01T20:59:10Z

Don't populate the array 'values' on the stack but instead make it static const. Makes the object code smaller by 111 bytes. Before: text data bss dec hex filename 108612 11169 512 120293 1d5e5 mm/shmem.o After: text data bss dec hex filename 108437 11233 512 120182 1d576 mm/shmem.o (gcc version 9.2.1, amd64) Link: http://lkml.kernel.org/r/20190906143012.28698-1-colin.king@canonical.com Signed-off-by: Colin Ian King Reviewed-by: Andrew Morton Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

mm, memfd: fix COW issue on MAP_PRIVATE and F_SEAL_FUTURE_WRITE mappings

2019-12-01T20:59:03Z

F_SEAL_FUTURE_WRITE has unexpected behavior when used with MAP_PRIVATE: A private mapping created after the memfd file that gets sealed with F_SEAL_FUTURE_WRITE loses the copy-on-write at fork behavior, meaning children and parent share the same memory, even though the mapping is private. The reason for this is due to the code below: static int shmem_mmap(struct file *file, struct vm_area_struct *vma) { struct shmem_inode_info *info = SHMEM_I(file_inode(file)); if (info->seals & F_SEAL_FUTURE_WRITE) { /* * New PROT_WRITE and MAP_SHARED mmaps are not allowed when * "future write" seal active. */ if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_WRITE)) return -EPERM; /* * Since the F_SEAL_FUTURE_WRITE seals allow for a MAP_SHARED * read-only mapping, take care to not allow mprotect to revert * protections. */ vma->vm_flags &= ~(VM_MAYWRITE); } ... } And for the mm to know if a mapping is copy-on-write: static inline bool is_cow_mapping(vm_flags_t flags) { return (flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE; } The patch fixes the issue by making the mprotect revert protection happen only for shared mappings. For private mappings, using mprotect will have no effect on the seal behavior. The F_SEAL_FUTURE_WRITE feature was introduced in v5.1 so v5.3.x stable kernels would need a backport. [akpm@linux-foundation.org: reflow comment, per Christoph] Link: http://lkml.kernel.org/r/20191107195355.80608-1-joel@joelfernandes.org Fixes: ab3948f58ff84 ("mm/memfd: add an F_SEAL_FUTURE_WRITE seal to memfd") Signed-off-by: Nicolas Geoffray Signed-off-by: Joel Fernandes (Google) Cc: Hugh Dickins Cc: Shuah Khan Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

shmem: pin the file in shmem_fault() if mmap_sem is dropped

2019-12-01T14:29:18Z

syzbot found the following crash: BUG: KASAN: use-after-free in perf_trace_lock_acquire+0x401/0x530 include/trace/events/lock.h:13 Read of size 8 at addr ffff8880a5cf2c50 by task syz-executor.0/26173 CPU: 0 PID: 26173 Comm: syz-executor.0 Not tainted 5.3.0-rc6 #146 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: perf_trace_lock_acquire+0x401/0x530 include/trace/events/lock.h:13 trace_lock_acquire include/trace/events/lock.h:13 [inline] lock_acquire+0x2de/0x410 kernel/locking/lockdep.c:4411 __raw_spin_lock include/linux/spinlock_api_smp.h:142 [inline] _raw_spin_lock+0x2f/0x40 kernel/locking/spinlock.c:151 spin_lock include/linux/spinlock.h:338 [inline] shmem_fault+0x5ec/0x7b0 mm/shmem.c:2034 __do_fault+0x111/0x540 mm/memory.c:3083 do_shared_fault mm/memory.c:3535 [inline] do_fault mm/memory.c:3613 [inline] handle_pte_fault mm/memory.c:3840 [inline] __handle_mm_fault+0x2adf/0x3f20 mm/memory.c:3964 handle_mm_fault+0x1b5/0x6b0 mm/memory.c:4001 do_user_addr_fault arch/x86/mm/fault.c:1441 [inline] __do_page_fault+0x536/0xdd0 arch/x86/mm/fault.c:1506 do_page_fault+0x38/0x590 arch/x86/mm/fault.c:1530 page_fault+0x39/0x40 arch/x86/entry/entry_64.S:1202 It happens if the VMA got unmapped under us while we dropped mmap_sem and inode got freed. Pinning the file if we drop mmap_sem fixes the issue. Link: http://lkml.kernel.org/r/20190927083908.rhifa4mmaxefc24r@box Signed-off-by: Kirill A. Shutemov Reported-by: syzbot+03ee87124ee05af991bd@syzkaller.appspotmail.com Acked-by: Johannes Weiner Reviewed-by: Matthew Wilcox (Oracle) Cc: Hillf Danton Cc: Hugh Dickins Cc: Josef Bacik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

Merge branch 'work.mount3' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

2019-10-10T15:16:44Z

Pull mount fixes from Al Viro: "A couple of regressions from the mount series" * 'work.mount3' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: vfs: add missing blkdev_put() in get_tree_bdev() shmem: fix LSM options parsing

shmem: fix LSM options parsing

2019-10-10T02:48:01Z

->parse_monolithic() there forgets to call security_sb_eat_lsm_opts() Signed-off-by: Al Viro

Merge branch 'hugepage-fallbacks' (hugepatch patches from David Rientjes)

2019-09-28T21:26:47Z

Merge hugepage allocation updates from David Rientjes: "We (mostly Linus, Andrea, and myself) have been discussing offlist how to implement a sane default allocation strategy for hugepages on NUMA platforms. With these reverts in place, the page allocator will happily allocate a remote hugepage immediately rather than try to make a local hugepage available. This incurs a substantial performance degradation when memory compaction would have otherwise made a local hugepage available. This series reverts those reverts and attempts to propose a more sane default allocation strategy specifically for hugepages. Andrea acknowledges this is likely to fix the swap storms that he originally reported that resulted in the patches that removed __GFP_THISNODE from hugepage allocations. The immediate goal is to return 5.3 to the behavior the kernel has implemented over the past several years so that remote hugepages are not immediately allocated when local hugepages could have been made available because the increased access latency is untenable. The next goal is to introduce a sane default allocation strategy for hugepages allocations in general regardless of the configuration of the system so that we prevent thrashing of local memory when compaction is unlikely to succeed and can prefer remote hugepages over remote native pages when the local node is low on memory." Note on timing: this reverts the hugepage VM behavior changes that got introduced fairly late in the 5.3 cycle, and that fixed a huge performance regression for certain loads that had been around since 4.18. Andrea had this note: "The regression of 4.18 was that it was taking hours to start a VM where 3.10 was only taking a few seconds, I reported all the details on lkml when it was finally tracked down in August 2018. https://lore.kernel.org/linux-mm/20180820032640.9896-2-aarcange@redhat.com/ __GFP_THISNODE in MADV_HUGEPAGE made the above enterprise vfio workload degrade like in the "current upstream" above. And it still would have been that bad as above until 5.3-rc5" where the bad behavior ends up happening as you fill up a local node, and without that change, you'd get into the nasty swap storm behavior due to compaction working overtime to make room for more memory on the nodes. As a result 5.3 got the two performance fix reverts in rc5. However, David Rientjes then noted that those performance fixes in turn regressed performance for other loads - although not quite to the same degree. He suggested reverting the reverts and instead replacing them with two small changes to how hugepage allocations are done (patch descriptions rephrased by me): - "avoid expensive reclaim when compaction may not succeed": just admit that the allocation failed when you're trying to allocate a huge-page and compaction wasn't successful. - "allow hugepage fallback to remote nodes when madvised": when that node-local huge-page allocation failed, retry without forcing the local node. but by then I judged it too late to replace the fixes for a 5.3 release. So 5.3 was released with behavior that harked back to the pre-4.18 logic. But now we're in the merge window for 5.4, and we can see if this alternate model fixes not just the horrendous swap storm behavior, but also restores the performance regression that the late reverts caused. Fingers crossed. * emailed patches from David Rientjes : mm, page_alloc: allow hugepage fallback to remote nodes when madvised mm, page_alloc: avoid expensive reclaim when compaction may not succeed Revert "Revert "Revert "mm, thp: consolidate THP gfp handling into alloc_hugepage_direct_gfpmask"" Revert "Revert "mm, thp: restore node-local hugepage allocations""

Revert "Revert "Revert "mm, thp: consolidate THP gfp handling into alloc_hugepage_direct_gfpmask""

2019-09-28T21:05:38Z

This reverts commit 92717d429b38e4f9f934eed7e605cc42858f1839. Since commit a8282608c88e ("Revert "mm, thp: restore node-local hugepage allocations"") is reverted in this series, it is better to restore the previous 5.2 behavior between the thp allocation and the page allocator rather than to attempt any consolidation or cleanup for a policy that is now reverted. It's less risky during an rc cycle and subsequent patches in this series further modify the same policy that the pre-5.3 behavior implements. Consolidation and cleanup can be done subsequent to a sane default page allocation strategy, so this patch reverts a cleanup done on a strategy that is now reverted and thus is the least risky option. Signed-off-by: David Rientjes Cc: Andrea Arcangeli Cc: Michal Hocko Cc: Mel Gorman Cc: Vlastimil Babka Cc: Stefan Priebe - Profihost AG Cc: "Kirill A. Shutemov" Cc: Andrew Morton Signed-off-by: Linus Torvalds