linux-stable.git/include/linux/mm_types.h, branch v5.15.208 Linux kernel stable tree mm/hugetlb: fix excessive IPI broadcasts when unsharing PMD tables using mmu_gather 2026-04-18T08:33:18+00:00 David Hildenbrand (Red Hat) david@kernel.org 2025-12-23T21:40:37+00:00 ce4cf24761467756462c0e1dbe4a5e9c9eb520fc commit 8ce720d5bd91e9dc16db3604aa4b1bf76770a9a1 upstream. As reported, ever since commit 1013af4f585f ("mm/hugetlb: fix huge_pmd_unshare() vs GUP-fast race") we can end up in some situations where we perform so many IPI broadcasts when unsharing hugetlb PMD page tables that it severely regresses some workloads. In particular, when we fork()+exit(), or when we munmap() a large area backed by many shared PMD tables, we perform one IPI broadcast per unshared PMD table. There are two optimizations to be had: (1) When we process (unshare) multiple such PMD tables, such as during exit(), it is sufficient to send a single IPI broadcast (as long as we respect locking rules) instead of one per PMD table. Locking prevents that any of these PMD tables could get reused before we drop the lock. (2) When we are not the last sharer (> 2 users including us), there is no need to send the IPI broadcast. The shared PMD tables cannot become exclusive (fully unshared) before an IPI will be broadcasted by the last sharer. Concurrent GUP-fast could walk into a PMD table just before we unshared it. It could then succeed in grabbing a page from the shared page table even after munmap() etc succeeded (and supressed an IPI). But there is not difference compared to GUP-fast just sleeping for a while after grabbing the page and re-enabling IRQs. Most importantly, GUP-fast will never walk into page tables that are no-longer shared, because the last sharer will issue an IPI broadcast. (if ever required, checking whether the PUD changed in GUP-fast after grabbing the page like we do in the PTE case could handle this) So let's rework PMD sharing TLB flushing + IPI sync to use the mmu_gather infrastructure so we can implement these optimizations and demystify the code at least a bit. Extend the mmu_gather infrastructure to be able to deal with our special hugetlb PMD table sharing implementation. To make initialization of the mmu_gather easier when working on a single VMA (in particular, when dealing with hugetlb), provide tlb_gather_mmu_vma(). We'll consolidate the handling for (full) unsharing of PMD tables in tlb_unshare_pmd_ptdesc() and tlb_flush_unshared_tables(), and track in "struct mmu_gather" whether we had (full) unsharing of PMD tables. Because locking is very special (concurrent unsharing+reuse must be prevented), we disallow deferring flushing to tlb_finish_mmu() and instead require an explicit earlier call to tlb_flush_unshared_tables(). From hugetlb code, we call huge_pmd_unshare_flush() where we make sure that the expected lock protecting us from concurrent unsharing+reuse is still held. Check with a VM_WARN_ON_ONCE() in tlb_finish_mmu() that tlb_flush_unshared_tables() was properly called earlier. Document it all properly. Notes about tlb_remove_table_sync_one() interaction with unsharing: There are two fairly tricky things: (1) tlb_remove_table_sync_one() is a NOP on architectures without CONFIG_MMU_GATHER_RCU_TABLE_FREE. Here, the assumption is that the previous TLB flush would send an IPI to all relevant CPUs. Careful: some architectures like x86 only send IPIs to all relevant CPUs when tlb->freed_tables is set. The relevant architectures should be selecting MMU_GATHER_RCU_TABLE_FREE, but x86 might not do that in stable kernels and it might have been problematic before this patch. Also, the arch flushing behavior (independent of IPIs) is different when tlb->freed_tables is set. Do we have to enlighten them to also take care of tlb->unshared_tables? So far we didn't care, so hopefully we are fine. Of course, we could be setting tlb->freed_tables as well, but that might then unnecessarily flush too much, because the semantics of tlb->freed_tables are a bit fuzzy. This patch changes nothing in this regard. (2) tlb_remove_table_sync_one() is not a NOP on architectures with CONFIG_MMU_GATHER_RCU_TABLE_FREE that actually don't need a sync. Take x86 as an example: in the common case (!pv, !X86_FEATURE_INVLPGB) we still issue IPIs during TLB flushes and don't actually need the second tlb_remove_table_sync_one(). This optimized can be implemented on top of this, by checking e.g., in tlb_remove_table_sync_one() whether we really need IPIs. But as described in (1), it really must honor tlb->freed_tables then to send IPIs to all relevant CPUs. Notes on TLB flushing changes: (1) Flushing for non-shared PMD tables We're converting from flush_hugetlb_tlb_range() to tlb_remove_huge_tlb_entry(). Given that we properly initialize the MMU gather in tlb_gather_mmu_vma() to be hugetlb aware, similar to __unmap_hugepage_range(), that should be fine. (2) Flushing for shared PMD tables We're converting from various things (flush_hugetlb_tlb_range(), tlb_flush_pmd_range(), flush_tlb_range()) to tlb_flush_pmd_range(). tlb_flush_pmd_range() achieves the same that tlb_remove_huge_tlb_entry() would achieve in these scenarios. Note that tlb_remove_huge_tlb_entry() also calls __tlb_remove_tlb_entry(), however that is only implemented on powerpc, which does not support PMD table sharing. Similar to (1), tlb_gather_mmu_vma() should make sure that TLB flushing keeps on working as expected. Further, note that the ptdesc_pmd_pts_dec() in huge_pmd_share() is not a concern, as we are holding the i_mmap_lock the whole time, preventing concurrent unsharing. That ptdesc_pmd_pts_dec() usage will be removed separately as a cleanup later. There are plenty more cleanups to be had, but they have to wait until this is fixed. [david@kernel.org: fix kerneldoc] Link: https://lkml.kernel.org/r/f223dd74-331c-412d-93fc-69e360a5006c@kernel.org Link: https://lkml.kernel.org/r/20251223214037.580860-5-david@kernel.org Fixes: 1013af4f585f ("mm/hugetlb: fix huge_pmd_unshare() vs GUP-fast race") Signed-off-by: David Hildenbrand (Red Hat) <david@kernel.org> Reported-by: "Uschakow, Stanislav" <suschako@amazon.de> Closes: https://lore.kernel.org/all/4d3878531c76479d9f8ca9789dc6485d@amazon.de/ Tested-by: Laurence Oberman <loberman@redhat.com> Acked-by: Harry Yoo <harry.yoo@oracle.com> Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com> Cc: Lance Yang <lance.yang@linux.dev> Cc: Liu Shixin <liushixin2@huawei.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Rik van Riel <riel@surriel.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> [ David: We don't have ptdesc and the wrappers, so work directly on page->pt_share_count and pass "struct page" instead of "struct ptdesc". CONFIG_HUGETLB_PMD_PAGE_TABLE_SHARING is still called CONFIG_ARCH_WANT_HUGE_PMD_SHARE and is set even without CONFIG_HUGETLB_PAGE. We don't have 550a7d60bd5e ("mm, hugepages: add mremap() support for hugepage backed vma"), so move_hugetlb_page_tables() does not exist. We don't have 40549ba8f8e0 ("hugetlb: use new vma_lock for pmd sharing synchronization") so changes in mm/rmap.c looks quite different. We don't have 4ddb4d91b82f ("hugetlb: do not update address in huge_pmd_unshare"), so huge_pmd_unshare() still gets a pointer to an address. Some smaller contextual stuff. ] Signed-off-by: David Hildenbrand (Arm) <david@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 8ce720d5bd91e9dc16db3604aa4b1bf76770a9a1 upstream.

As reported, ever since commit 1013af4f585f ("mm/hugetlb: fix
huge_pmd_unshare() vs GUP-fast race") we can end up in some situations
where we perform so many IPI broadcasts when unsharing hugetlb PMD page
tables that it severely regresses some workloads.

In particular, when we fork()+exit(), or when we munmap() a large
area backed by many shared PMD tables, we perform one IPI broadcast per
unshared PMD table.

There are two optimizations to be had:

(1) When we process (unshare) multiple such PMD tables, such as during
    exit(), it is sufficient to send a single IPI broadcast (as long as
    we respect locking rules) instead of one per PMD table.

    Locking prevents that any of these PMD tables could get reused before
    we drop the lock.

(2) When we are not the last sharer (> 2 users including us), there is
    no need to send the IPI broadcast. The shared PMD tables cannot
    become exclusive (fully unshared) before an IPI will be broadcasted
    by the last sharer.

    Concurrent GUP-fast could walk into a PMD table just before we
    unshared it. It could then succeed in grabbing a page from the
    shared page table even after munmap() etc succeeded (and supressed
    an IPI). But there is not difference compared to GUP-fast just
    sleeping for a while after grabbing the page and re-enabling IRQs.

    Most importantly, GUP-fast will never walk into page tables that are
    no-longer shared, because the last sharer will issue an IPI
    broadcast.

    (if ever required, checking whether the PUD changed in GUP-fast
     after grabbing the page like we do in the PTE case could handle
     this)

So let's rework PMD sharing TLB flushing + IPI sync to use the mmu_gather
infrastructure so we can implement these optimizations and demystify the
code at least a bit. Extend the mmu_gather infrastructure to be able to
deal with our special hugetlb PMD table sharing implementation.

To make initialization of the mmu_gather easier when working on a single
VMA (in particular, when dealing with hugetlb), provide
tlb_gather_mmu_vma().

We'll consolidate the handling for (full) unsharing of PMD tables in
tlb_unshare_pmd_ptdesc() and tlb_flush_unshared_tables(), and track
in "struct mmu_gather" whether we had (full) unsharing of PMD tables.

Because locking is very special (concurrent unsharing+reuse must be
prevented), we disallow deferring flushing to tlb_finish_mmu() and instead
require an explicit earlier call to tlb_flush_unshared_tables().

From hugetlb code, we call huge_pmd_unshare_flush() where we make sure
that the expected lock protecting us from concurrent unsharing+reuse is
still held.

Check with a VM_WARN_ON_ONCE() in tlb_finish_mmu() that
tlb_flush_unshared_tables() was properly called earlier.

Document it all properly.

Notes about tlb_remove_table_sync_one() interaction with unsharing:

There are two fairly tricky things:

(1) tlb_remove_table_sync_one() is a NOP on architectures without
    CONFIG_MMU_GATHER_RCU_TABLE_FREE.

    Here, the assumption is that the previous TLB flush would send an
    IPI to all relevant CPUs. Careful: some architectures like x86 only
    send IPIs to all relevant CPUs when tlb->freed_tables is set.

    The relevant architectures should be selecting
    MMU_GATHER_RCU_TABLE_FREE, but x86 might not do that in stable
    kernels and it might have been problematic before this patch.

    Also, the arch flushing behavior (independent of IPIs) is different
    when tlb->freed_tables is set. Do we have to enlighten them to also
    take care of tlb->unshared_tables? So far we didn't care, so
    hopefully we are fine. Of course, we could be setting
    tlb->freed_tables as well, but that might then unnecessarily flush
    too much, because the semantics of tlb->freed_tables are a bit
    fuzzy.

    This patch changes nothing in this regard.

(2) tlb_remove_table_sync_one() is not a NOP on architectures with
    CONFIG_MMU_GATHER_RCU_TABLE_FREE that actually don't need a sync.

    Take x86 as an example: in the common case (!pv, !X86_FEATURE_INVLPGB)
    we still issue IPIs during TLB flushes and don't actually need the
    second tlb_remove_table_sync_one().

    This optimized can be implemented on top of this, by checking e.g., in
    tlb_remove_table_sync_one() whether we really need IPIs. But as
    described in (1), it really must honor tlb->freed_tables then to
    send IPIs to all relevant CPUs.

Notes on TLB flushing changes:

(1) Flushing for non-shared PMD tables

    We're converting from flush_hugetlb_tlb_range() to
    tlb_remove_huge_tlb_entry(). Given that we properly initialize the
    MMU gather in tlb_gather_mmu_vma() to be hugetlb aware, similar to
    __unmap_hugepage_range(), that should be fine.

(2) Flushing for shared PMD tables

    We're converting from various things (flush_hugetlb_tlb_range(),
    tlb_flush_pmd_range(), flush_tlb_range()) to tlb_flush_pmd_range().

    tlb_flush_pmd_range() achieves the same that
    tlb_remove_huge_tlb_entry() would achieve in these scenarios.
    Note that tlb_remove_huge_tlb_entry() also calls
    __tlb_remove_tlb_entry(), however that is only implemented on
    powerpc, which does not support PMD table sharing.

    Similar to (1), tlb_gather_mmu_vma() should make sure that TLB
    flushing keeps on working as expected.

Further, note that the ptdesc_pmd_pts_dec() in huge_pmd_share() is not a
concern, as we are holding the i_mmap_lock the whole time, preventing
concurrent unsharing. That ptdesc_pmd_pts_dec() usage will be removed
separately as a cleanup later.

There are plenty more cleanups to be had, but they have to wait until
this is fixed.

[david@kernel.org: fix kerneldoc]
  Link: https://lkml.kernel.org/r/f223dd74-331c-412d-93fc-69e360a5006c@kernel.org
Link: https://lkml.kernel.org/r/20251223214037.580860-5-david@kernel.org
Fixes: 1013af4f585f ("mm/hugetlb: fix huge_pmd_unshare() vs GUP-fast race")
Signed-off-by: David Hildenbrand (Red Hat) <david@kernel.org>
Reported-by: "Uschakow, Stanislav" <suschako@amazon.de>
Closes: https://lore.kernel.org/all/4d3878531c76479d9f8ca9789dc6485d@amazon.de/
Tested-by: Laurence Oberman <loberman@redhat.com>
Acked-by: Harry Yoo <harry.yoo@oracle.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Lance Yang <lance.yang@linux.dev>
Cc: Liu Shixin <liushixin2@huawei.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Rik van Riel <riel@surriel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
[ David: We don't have ptdesc and the wrappers, so work directly on
  page->pt_share_count and pass "struct page" instead of "struct ptdesc".
  CONFIG_HUGETLB_PMD_PAGE_TABLE_SHARING is still called
  CONFIG_ARCH_WANT_HUGE_PMD_SHARE and is set even without
  CONFIG_HUGETLB_PAGE. We don't have 550a7d60bd5e ("mm, hugepages: add
  mremap() support for hugepage backed vma"), so move_hugetlb_page_tables()
  does not exist. We don't have 40549ba8f8e0 ("hugetlb: use new vma_lock
  for pmd sharing synchronization") so changes in mm/rmap.c looks quite
  different. We don't have 4ddb4d91b82f ("hugetlb: do not update address
  in huge_pmd_unshare"), so huge_pmd_unshare() still gets a pointer to
  an address. Some smaller contextual stuff. ]
Signed-off-by: David Hildenbrand (Arm) <david@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
mm: hugetlb: independent PMD page table shared count 2025-06-27T10:05:35+00:00 Liu Shixin liushixin2@huawei.com 2024-12-16T07:11:47+00:00 8410996eb6fea116fe1483ed977aacf580eee7b4 commit 59d9094df3d79443937add8700b2ef1a866b1081 upstream. The folio refcount may be increased unexpectly through try_get_folio() by caller such as split_huge_pages. In huge_pmd_unshare(), we use refcount to check whether a pmd page table is shared. The check is incorrect if the refcount is increased by the above caller, and this can cause the page table leaked: BUG: Bad page state in process sh pfn:109324 page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x66 pfn:0x109324 flags: 0x17ffff800000000(node=0|zone=2|lastcpupid=0xfffff) page_type: f2(table) raw: 017ffff800000000 0000000000000000 0000000000000000 0000000000000000 raw: 0000000000000066 0000000000000000 00000000f2000000 0000000000000000 page dumped because: nonzero mapcount ... CPU: 31 UID: 0 PID: 7515 Comm: sh Kdump: loaded Tainted: G B 6.13.0-rc2master+ #7 Tainted: [B]=BAD_PAGE Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015 Call trace: show_stack+0x20/0x38 (C) dump_stack_lvl+0x80/0xf8 dump_stack+0x18/0x28 bad_page+0x8c/0x130 free_page_is_bad_report+0xa4/0xb0 free_unref_page+0x3cc/0x620 __folio_put+0xf4/0x158 split_huge_pages_all+0x1e0/0x3e8 split_huge_pages_write+0x25c/0x2d8 full_proxy_write+0x64/0xd8 vfs_write+0xcc/0x280 ksys_write+0x70/0x110 __arm64_sys_write+0x24/0x38 invoke_syscall+0x50/0x120 el0_svc_common.constprop.0+0xc8/0xf0 do_el0_svc+0x24/0x38 el0_svc+0x34/0x128 el0t_64_sync_handler+0xc8/0xd0 el0t_64_sync+0x190/0x198 The issue may be triggered by damon, offline_page, page_idle, etc, which will increase the refcount of page table. 1. The page table itself will be discarded after reporting the "nonzero mapcount". 2. The HugeTLB page mapped by the page table miss freeing since we treat the page table as shared and a shared page table will not be unmapped. Fix it by introducing independent PMD page table shared count. As described by comment, pt_index/pt_mm/pt_frag_refcount are used for s390 gmap, x86 pgds and powerpc, pt_share_count is used for x86/arm64/riscv pmds, so we can reuse the field as pt_share_count. Link: https://lkml.kernel.org/r/20241216071147.3984217-1-liushixin2@huawei.com Fixes: 39dde65c9940 ("[PATCH] shared page table for hugetlb page") Signed-off-by: Liu Shixin <liushixin2@huawei.com> Cc: Kefeng Wang <wangkefeng.wang@huawei.com> Cc: Ken Chen <kenneth.w.chen@intel.com> Cc: Muchun Song <muchun.song@linux.dev> Cc: Nanyong Sun <sunnanyong@huawei.com> Cc: Jane Chu <jane.chu@oracle.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> [backport note: struct ptdesc did not exist yet, stuff it equivalently into struct page instead] Signed-off-by: Jann Horn <jannh@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 59d9094df3d79443937add8700b2ef1a866b1081 upstream.

The folio refcount may be increased unexpectly through try_get_folio() by
caller such as split_huge_pages.  In huge_pmd_unshare(), we use refcount
to check whether a pmd page table is shared.  The check is incorrect if
the refcount is increased by the above caller, and this can cause the page
table leaked:

 BUG: Bad page state in process sh  pfn:109324
 page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x66 pfn:0x109324
 flags: 0x17ffff800000000(node=0|zone=2|lastcpupid=0xfffff)
 page_type: f2(table)
 raw: 017ffff800000000 0000000000000000 0000000000000000 0000000000000000
 raw: 0000000000000066 0000000000000000 00000000f2000000 0000000000000000
 page dumped because: nonzero mapcount
 ...
 CPU: 31 UID: 0 PID: 7515 Comm: sh Kdump: loaded Tainted: G    B              6.13.0-rc2master+ #7
 Tainted: [B]=BAD_PAGE
 Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
 Call trace:
  show_stack+0x20/0x38 (C)
  dump_stack_lvl+0x80/0xf8
  dump_stack+0x18/0x28
  bad_page+0x8c/0x130
  free_page_is_bad_report+0xa4/0xb0
  free_unref_page+0x3cc/0x620
  __folio_put+0xf4/0x158
  split_huge_pages_all+0x1e0/0x3e8
  split_huge_pages_write+0x25c/0x2d8
  full_proxy_write+0x64/0xd8
  vfs_write+0xcc/0x280
  ksys_write+0x70/0x110
  __arm64_sys_write+0x24/0x38
  invoke_syscall+0x50/0x120
  el0_svc_common.constprop.0+0xc8/0xf0
  do_el0_svc+0x24/0x38
  el0_svc+0x34/0x128
  el0t_64_sync_handler+0xc8/0xd0
  el0t_64_sync+0x190/0x198

The issue may be triggered by damon, offline_page, page_idle, etc, which
will increase the refcount of page table.

1. The page table itself will be discarded after reporting the
   "nonzero mapcount".

2. The HugeTLB page mapped by the page table miss freeing since we
   treat the page table as shared and a shared page table will not be
   unmapped.

Fix it by introducing independent PMD page table shared count.  As
described by comment, pt_index/pt_mm/pt_frag_refcount are used for s390
gmap, x86 pgds and powerpc, pt_share_count is used for x86/arm64/riscv
pmds, so we can reuse the field as pt_share_count.

Link: https://lkml.kernel.org/r/20241216071147.3984217-1-liushixin2@huawei.com
Fixes: 39dde65c9940 ("[PATCH] shared page table for hugetlb page")
Signed-off-by: Liu Shixin <liushixin2@huawei.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Ken Chen <kenneth.w.chen@intel.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Nanyong Sun <sunnanyong@huawei.com>
Cc: Jane Chu <jane.chu@oracle.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
[backport note: struct ptdesc did not exist yet, stuff it equivalently
into struct page instead]
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
page_pool: add interface to manipulate frag count in page pool 2021-08-09T22:49:00+00:00 Yunsheng Lin linyunsheng@huawei.com 2021-08-06T02:46:20+00:00 0e9d2a0a3a836c37528899010e73b5be8111753e For 32 bit systems with 64 bit dma, dma_addr[1] is used to store the upper 32 bit dma addr, those system should be rare those days. For normal system, the dma_addr[1] in 'struct page' is not used, so we can reuse dma_addr[1] for storing frag count, which means how many frags this page might be splited to. In order to simplify the page frag support in the page pool, the PAGE_POOL_DMA_USE_PP_FRAG_COUNT macro is added to indicate the 32 bit systems with 64 bit dma, and the page frag support in page pool is disabled for such system. The newly added page_pool_set_frag_count() is called to reserve the maximum frag count before any page frag is passed to the user. The page_pool_atomic_sub_frag_count_return() is called when user is done with the page frag. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> 
For 32 bit systems with 64 bit dma, dma_addr[1] is used to
store the upper 32 bit dma addr, those system should be rare
those days.

For normal system, the dma_addr[1] in 'struct page' is not
used, so we can reuse dma_addr[1] for storing frag count,
which means how many frags this page might be splited to.

In order to simplify the page frag support in the page pool,
the PAGE_POOL_DMA_USE_PP_FRAG_COUNT macro is added to indicate
the 32 bit systems with 64 bit dma, and the page frag support
in page pool is disabled for such system.

The newly added page_pool_set_frag_count() is called to reserve
the maximum frag count before any page frag is passed to the
user. The page_pool_atomic_sub_frag_count_return() is called
when user is done with the page frag.

Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Merge branch 'akpm' (patches from Andrew) 2021-07-02T19:08:10+00:00 Linus Torvalds torvalds@linux-foundation.org 2021-07-02T19:08:10+00:00 71bd9341011f626d692aabe024f099820f02c497 Merge more updates from Andrew Morton: "190 patches. Subsystems affected by this patch series: mm (hugetlb, userfaultfd, vmscan, kconfig, proc, z3fold, zbud, ras, mempolicy, memblock, migration, thp, nommu, kconfig, madvise, memory-hotplug, zswap, zsmalloc, zram, cleanups, kfence, and hmm), procfs, sysctl, misc, core-kernel, lib, lz4, checkpatch, init, kprobes, nilfs2, hfs, signals, exec, kcov, selftests, compress/decompress, and ipc" * emailed patches from Andrew Morton <akpm@linux-foundation.org>: (190 commits) ipc/util.c: use binary search for max_idx ipc/sem.c: use READ_ONCE()/WRITE_ONCE() for use_global_lock ipc: use kmalloc for msg_queue and shmid_kernel ipc sem: use kvmalloc for sem_undo allocation lib/decompressors: remove set but not used variabled 'level' selftests/vm/pkeys: exercise x86 XSAVE init state selftests/vm/pkeys: refill shadow register after implicit kernel write selftests/vm/pkeys: handle negative sys_pkey_alloc() return code selftests/vm/pkeys: fix alloc_random_pkey() to make it really, really random kcov: add __no_sanitize_coverage to fix noinstr for all architectures exec: remove checks in __register_bimfmt() x86: signal: don't do sas_ss_reset() until we are certain that sigframe won't be abandoned hfsplus: report create_date to kstat.btime hfsplus: remove unnecessary oom message nilfs2: remove redundant continue statement in a while-loop kprobes: remove duplicated strong free_insn_page in x86 and s390 init: print out unknown kernel parameters checkpatch: do not complain about positive return values starting with EPOLL checkpatch: improve the indented label test checkpatch: scripts/spdxcheck.py now requires python3 ... 
Merge more updates from Andrew Morton:
 "190 patches.

  Subsystems affected by this patch series: mm (hugetlb, userfaultfd,
  vmscan, kconfig, proc, z3fold, zbud, ras, mempolicy, memblock,
  migration, thp, nommu, kconfig, madvise, memory-hotplug, zswap,
  zsmalloc, zram, cleanups, kfence, and hmm), procfs, sysctl, misc,
  core-kernel, lib, lz4, checkpatch, init, kprobes, nilfs2, hfs,
  signals, exec, kcov, selftests, compress/decompress, and ipc"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (190 commits)
  ipc/util.c: use binary search for max_idx
  ipc/sem.c: use READ_ONCE()/WRITE_ONCE() for use_global_lock
  ipc: use kmalloc for msg_queue and shmid_kernel
  ipc sem: use kvmalloc for sem_undo allocation
  lib/decompressors: remove set but not used variabled 'level'
  selftests/vm/pkeys: exercise x86 XSAVE init state
  selftests/vm/pkeys: refill shadow register after implicit kernel write
  selftests/vm/pkeys: handle negative sys_pkey_alloc() return code
  selftests/vm/pkeys: fix alloc_random_pkey() to make it really, really random
  kcov: add __no_sanitize_coverage to fix noinstr for all architectures
  exec: remove checks in __register_bimfmt()
  x86: signal: don't do sas_ss_reset() until we are certain that sigframe won't be abandoned
  hfsplus: report create_date to kstat.btime
  hfsplus: remove unnecessary oom message
  nilfs2: remove redundant continue statement in a while-loop
  kprobes: remove duplicated strong free_insn_page in x86 and s390
  init: print out unknown kernel parameters
  checkpatch: do not complain about positive return values starting with EPOLL
  checkpatch: improve the indented label test
  checkpatch: scripts/spdxcheck.py now requires python3
  ...
mm: fix spelling mistakes 2021-07-01T18:06:02+00:00 Zhen Lei thunder.leizhen@huawei.com 2021-07-01T01:53:17+00:00 041711ce7cdf023f53d76f64d82b75210248e18d Fix some spelling mistakes in comments: each having differents usage ==> each has a different usage statments ==> statements adresses ==> addresses aggresive ==> aggressive datas ==> data posion ==> poison higer ==> higher precisly ==> precisely wont ==> won't We moves tha ==> We move the endianess ==> endianness Link: https://lkml.kernel.org/r/20210519065853.7723-2-thunder.leizhen@huawei.com Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com> Reviewed-by: Souptick Joarder <jrdr.linux@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> 
Fix some spelling mistakes in comments:
each having differents usage ==> each has a different usage
statments ==> statements
adresses ==> addresses
aggresive ==> aggressive
datas ==> data
posion ==> poison
higer ==> higher
precisly ==> precisely
wont ==> won't
We moves tha ==> We move the
endianess ==> endianness

Link: https://lkml.kernel.org/r/20210519065853.7723-2-thunder.leizhen@huawei.com
Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Reviewed-by: Souptick Joarder <jrdr.linux@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge tag 'net-next-5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next 2021-06-30T22:51:09+00:00 Linus Torvalds torvalds@linux-foundation.org 2021-06-30T22:51:09+00:00 dbe69e43372212527abf48609aba7fc39a6daa27 Pull networking updates from Jakub Kicinski: "Core: - BPF: - add syscall program type and libbpf support for generating instructions and bindings for in-kernel BPF loaders (BPF loaders for BPF), this is a stepping stone for signed BPF programs - infrastructure to migrate TCP child sockets from one listener to another in the same reuseport group/map to improve flexibility of service hand-off/restart - add broadcast support to XDP redirect - allow bypass of the lockless qdisc to improving performance (for pktgen: +23% with one thread, +44% with 2 threads) - add a simpler version of "DO_ONCE()" which does not require jump labels, intended for slow-path usage - virtio/vsock: introduce SOCK_SEQPACKET support - add getsocketopt to retrieve netns cookie - ip: treat lowest address of a IPv4 subnet as ordinary unicast address allowing reclaiming of precious IPv4 addresses - ipv6: use prandom_u32() for ID generation - ip: add support for more flexible field selection for hashing across multi-path routes (w/ offload to mlxsw) - icmp: add support for extended RFC 8335 PROBE (ping) - seg6: add support for SRv6 End.DT46 behavior - mptcp: - DSS checksum support (RFC 8684) to detect middlebox meddling - support Connection-time 'C' flag - time stamping support - sctp: packetization Layer Path MTU Discovery (RFC 8899) - xfrm: speed up state addition with seq set - WiFi: - hidden AP discovery on 6 GHz and other HE 6 GHz improvements - aggregation handling improvements for some drivers - minstrel improvements for no-ack frames - deferred rate control for TXQs to improve reaction times - switch from round robin to virtual time-based airtime scheduler - add trace points: - tcp checksum errors - openvswitch - action execution, upcalls - socket errors via sk_error_report Device APIs: - devlink: add rate API for hierarchical control of max egress rate of virtual devices (VFs, SFs etc.) - don't require RCU read lock to be held around BPF hooks in NAPI context - page_pool: generic buffer recycling New hardware/drivers: - mobile: - iosm: PCIe Driver for Intel M.2 Modem - support for Qualcomm MSM8998 (ipa) - WiFi: Qualcomm QCN9074 and WCN6855 PCI devices - sparx5: Microchip SparX-5 family of Enterprise Ethernet switches - Mellanox BlueField Gigabit Ethernet (control NIC of the DPU) - NXP SJA1110 Automotive Ethernet 10-port switch - Qualcomm QCA8327 switch support (qca8k) - Mikrotik 10/25G NIC (atl1c) Driver changes: - ACPI support for some MDIO, MAC and PHY devices from Marvell and NXP (our first foray into MAC/PHY description via ACPI) - HW timestamping (PTP) support: bnxt_en, ice, sja1105, hns3, tja11xx - Mellanox/Nvidia NIC (mlx5) - NIC VF offload of L2 bridging - support IRQ distribution to Sub-functions - Marvell (prestera): - add flower and match all - devlink trap - link aggregation - Netronome (nfp): connection tracking offload - Intel 1GE (igc): add AF_XDP support - Marvell DPU (octeontx2): ingress ratelimit offload - Google vNIC (gve): new ring/descriptor format support - Qualcomm mobile (rmnet & ipa): inline checksum offload support - MediaTek WiFi (mt76) - mt7915 MSI support - mt7915 Tx status reporting - mt7915 thermal sensors support - mt7921 decapsulation offload - mt7921 enable runtime pm and deep sleep - Realtek WiFi (rtw88) - beacon filter support - Tx antenna path diversity support - firmware crash information via devcoredump - Qualcomm WiFi (wcn36xx) - Wake-on-WLAN support with magic packets and GTK rekeying - Micrel PHY (ksz886x/ksz8081): add cable test support" * tag 'net-next-5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (2168 commits) tcp: change ICSK_CA_PRIV_SIZE definition tcp_yeah: check struct yeah size at compile time gve: DQO: Fix off by one in gve_rx_dqo() stmmac: intel: set PCI_D3hot in suspend stmmac: intel: Enable PHY WOL option in EHL net: stmmac: option to enable PHY WOL with PMT enabled net: say "local" instead of "static" addresses in ndo_dflt_fdb_{add,del} net: use netdev_info in ndo_dflt_fdb_{add,del} ptp: Set lookup cookie when creating a PTP PPS source. net: sock: add trace for socket errors net: sock: introduce sk_error_report net: dsa: replay the local bridge FDB entries pointing to the bridge dev too net: dsa: ensure during dsa_fdb_offload_notify that dev_hold and dev_put are on the same dev net: dsa: include fdb entries pointing to bridge in the host fdb list net: dsa: include bridge addresses which are local in the host fdb list net: dsa: sync static FDB entries on foreign interfaces to hardware net: dsa: install the host MDB and FDB entries in the master's RX filter net: dsa: reference count the FDB addresses at the cross-chip notifier level net: dsa: introduce a separate cross-chip notifier type for host FDBs net: dsa: reference count the MDB entries at the cross-chip notifier level ... 
Pull networking updates from Jakub Kicinski:
 "Core:

   - BPF:
      - add syscall program type and libbpf support for generating
        instructions and bindings for in-kernel BPF loaders (BPF loaders
        for BPF), this is a stepping stone for signed BPF programs
      - infrastructure to migrate TCP child sockets from one listener to
        another in the same reuseport group/map to improve flexibility
        of service hand-off/restart
      - add broadcast support to XDP redirect

   - allow bypass of the lockless qdisc to improving performance (for
     pktgen: +23% with one thread, +44% with 2 threads)

   - add a simpler version of "DO_ONCE()" which does not require jump
     labels, intended for slow-path usage

   - virtio/vsock: introduce SOCK_SEQPACKET support

   - add getsocketopt to retrieve netns cookie

   - ip: treat lowest address of a IPv4 subnet as ordinary unicast
     address allowing reclaiming of precious IPv4 addresses

   - ipv6: use prandom_u32() for ID generation

   - ip: add support for more flexible field selection for hashing
     across multi-path routes (w/ offload to mlxsw)

   - icmp: add support for extended RFC 8335 PROBE (ping)

   - seg6: add support for SRv6 End.DT46 behavior

   - mptcp:
      - DSS checksum support (RFC 8684) to detect middlebox meddling
      - support Connection-time 'C' flag
      - time stamping support

   - sctp: packetization Layer Path MTU Discovery (RFC 8899)

   - xfrm: speed up state addition with seq set

   - WiFi:
      - hidden AP discovery on 6 GHz and other HE 6 GHz improvements
      - aggregation handling improvements for some drivers
      - minstrel improvements for no-ack frames
      - deferred rate control for TXQs to improve reaction times
      - switch from round robin to virtual time-based airtime scheduler

   - add trace points:
      - tcp checksum errors
      - openvswitch - action execution, upcalls
      - socket errors via sk_error_report

  Device APIs:

   - devlink: add rate API for hierarchical control of max egress rate
     of virtual devices (VFs, SFs etc.)

   - don't require RCU read lock to be held around BPF hooks in NAPI
     context

   - page_pool: generic buffer recycling

  New hardware/drivers:

   - mobile:
      - iosm: PCIe Driver for Intel M.2 Modem
      - support for Qualcomm MSM8998 (ipa)

   - WiFi: Qualcomm QCN9074 and WCN6855 PCI devices

   - sparx5: Microchip SparX-5 family of Enterprise Ethernet switches

   - Mellanox BlueField Gigabit Ethernet (control NIC of the DPU)

   - NXP SJA1110 Automotive Ethernet 10-port switch

   - Qualcomm QCA8327 switch support (qca8k)

   - Mikrotik 10/25G NIC (atl1c)

  Driver changes:

   - ACPI support for some MDIO, MAC and PHY devices from Marvell and
     NXP (our first foray into MAC/PHY description via ACPI)

   - HW timestamping (PTP) support: bnxt_en, ice, sja1105, hns3, tja11xx

   - Mellanox/Nvidia NIC (mlx5)
      - NIC VF offload of L2 bridging
      - support IRQ distribution to Sub-functions

   - Marvell (prestera):
      - add flower and match all
      - devlink trap
      - link aggregation

   - Netronome (nfp): connection tracking offload

   - Intel 1GE (igc): add AF_XDP support

   - Marvell DPU (octeontx2): ingress ratelimit offload

   - Google vNIC (gve): new ring/descriptor format support

   - Qualcomm mobile (rmnet & ipa): inline checksum offload support

   - MediaTek WiFi (mt76)
      - mt7915 MSI support
      - mt7915 Tx status reporting
      - mt7915 thermal sensors support
      - mt7921 decapsulation offload
      - mt7921 enable runtime pm and deep sleep

   - Realtek WiFi (rtw88)
      - beacon filter support
      - Tx antenna path diversity support
      - firmware crash information via devcoredump

   - Qualcomm WiFi (wcn36xx)
      - Wake-on-WLAN support with magic packets and GTK rekeying

   - Micrel PHY (ksz886x/ksz8081): add cable test support"

* tag 'net-next-5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (2168 commits)
  tcp: change ICSK_CA_PRIV_SIZE definition
  tcp_yeah: check struct yeah size at compile time
  gve: DQO: Fix off by one in gve_rx_dqo()
  stmmac: intel: set PCI_D3hot in suspend
  stmmac: intel: Enable PHY WOL option in EHL
  net: stmmac: option to enable PHY WOL with PMT enabled
  net: say "local" instead of "static" addresses in ndo_dflt_fdb_{add,del}
  net: use netdev_info in ndo_dflt_fdb_{add,del}
  ptp: Set lookup cookie when creating a PTP PPS source.
  net: sock: add trace for socket errors
  net: sock: introduce sk_error_report
  net: dsa: replay the local bridge FDB entries pointing to the bridge dev too
  net: dsa: ensure during dsa_fdb_offload_notify that dev_hold and dev_put are on the same dev
  net: dsa: include fdb entries pointing to bridge in the host fdb list
  net: dsa: include bridge addresses which are local in the host fdb list
  net: dsa: sync static FDB entries on foreign interfaces to hardware
  net: dsa: install the host MDB and FDB entries in the master's RX filter
  net: dsa: reference count the FDB addresses at the cross-chip notifier level
  net: dsa: introduce a separate cross-chip notifier type for host FDBs
  net: dsa: reference count the MDB entries at the cross-chip notifier level
  ...
mm: gup: pack has_pinned in MMF_HAS_PINNED 2021-06-29T17:53:48+00:00 Andrea Arcangeli aarcange@redhat.com 2021-06-29T02:36:40+00:00 a458b76a4171f893efa7657dc079924580a8746a has_pinned 32bit can be packed in the MMF_HAS_PINNED bit as a noop cleanup. Any atomic_inc/dec to the mm cacheline shared by all threads in pin-fast would reintroduce a loss of SMP scalability to pin-fast, so there's no future potential usefulness to keep an atomic in the mm for this. set_bit(MMF_HAS_PINNED) will be theoretically a bit slower than WRITE_ONCE (atomic_set is equivalent to WRITE_ONCE), but the set_bit (just like atomic_set after this commit) has to be still issued only once per "mm", so the difference between the two will be lost in the noise. will-it-scale "mmap2" shows no change in performance with enterprise config as expected. will-it-scale "pin_fast" retains the > 4000% SMP scalability performance improvement against upstream as expected. This is a noop as far as overall performance and SMP scalability are concerned. [peterx@redhat.com: pack has_pinned in MMF_HAS_PINNED] Link: https://lkml.kernel.org/r/YJqWESqyxa8OZA+2@t490s [akpm@linux-foundation.org: coding style fixes] [peterx@redhat.com: fix build for task_mmu.c, introduce mm_set_has_pinned_flag, fix comments] Link: https://lkml.kernel.org/r/20210507150553.208763-4-peterx@redhat.com Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com> Reviewed-by: John Hubbard <jhubbard@nvidia.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: Jann Horn <jannh@google.com> Cc: Jason Gunthorpe <jgg@nvidia.com> Cc: Kirill Shutemov <kirill@shutemov.name> Cc: Kirill Tkhai <ktkhai@virtuozzo.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> 
has_pinned 32bit can be packed in the MMF_HAS_PINNED bit as a noop
cleanup.

Any atomic_inc/dec to the mm cacheline shared by all threads in pin-fast
would reintroduce a loss of SMP scalability to pin-fast, so there's no
future potential usefulness to keep an atomic in the mm for this.

set_bit(MMF_HAS_PINNED) will be theoretically a bit slower than WRITE_ONCE
(atomic_set is equivalent to WRITE_ONCE), but the set_bit (just like
atomic_set after this commit) has to be still issued only once per "mm",
so the difference between the two will be lost in the noise.

will-it-scale "mmap2" shows no change in performance with enterprise
config as expected.

will-it-scale "pin_fast" retains the > 4000% SMP scalability performance
improvement against upstream as expected.

This is a noop as far as overall performance and SMP scalability are
concerned.

[peterx@redhat.com: pack has_pinned in MMF_HAS_PINNED]
  Link: https://lkml.kernel.org/r/YJqWESqyxa8OZA+2@t490s
[akpm@linux-foundation.org: coding style fixes]
[peterx@redhat.com: fix build for task_mmu.c, introduce mm_set_has_pinned_flag, fix comments]

Link: https://lkml.kernel.org/r/20210507150553.208763-4-peterx@redhat.com
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: John Hubbard <jhubbard@nvidia.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Jann Horn <jannh@google.com>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: Kirill Shutemov <kirill@shutemov.name>
Cc: Kirill Tkhai <ktkhai@virtuozzo.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2021-06-19T02:47:02+00:00 Jakub Kicinski kuba@kernel.org 2021-06-19T02:47:02+00:00 adc2e56ebe6377f5c032d96aee0feac30a640453 Trivial conflicts in net/can/isotp.c and tools/testing/selftests/net/mptcp/mptcp_connect.sh scaled_ppm_to_ppb() was moved from drivers/ptp/ptp_clock.c to include/linux/ptp_clock_kernel.h in -next so re-apply the fix there. Signed-off-by: Jakub Kicinski <kuba@kernel.org> 
Trivial conflicts in net/can/isotp.c and
tools/testing/selftests/net/mptcp/mptcp_connect.sh

scaled_ppm_to_ppb() was moved from drivers/ptp/ptp_clock.c
to include/linux/ptp_clock_kernel.h in -next so re-apply
the fix there.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
mm: relocate 'write_protect_seq' in struct mm_struct 2021-06-12T20:28:50+00:00 Feng Tang feng.tang@intel.com 2021-06-11T01:54:42+00:00 2e3025434a6ba090c85871a1d4080ff784109e1f 0day robot reported a 9.2% regression for will-it-scale mmap1 test case[1], caused by commit 57efa1fe5957 ("mm/gup: prevent gup_fast from racing with COW during fork"). Further debug shows the regression is due to that commit changes the offset of hot fields 'mmap_lock' inside structure 'mm_struct', thus some cache alignment changes. From the perf data, the contention for 'mmap_lock' is very severe and takes around 95% cpu cycles, and it is a rw_semaphore struct rw_semaphore { atomic_long_t count; /* 8 bytes */ atomic_long_t owner; /* 8 bytes */ struct optimistic_spin_queue osq; /* spinner MCS lock */ ... Before commit 57efa1fe5957 adds the 'write_protect_seq', it happens to have a very optimal cache alignment layout, as Linus explained: "and before the addition of the 'write_protect_seq' field, the mmap_sem was at offset 120 in 'struct mm_struct'. Which meant that count and owner were in two different cachelines, and then when you have contention and spend time in rwsem_down_write_slowpath(), this is probably *exactly* the kind of layout you want. Because first the rwsem_write_trylock() will do a cmpxchg on the first cacheline (for the optimistic fast-path), and then in the case of contention, rwsem_down_write_slowpath() will just access the second cacheline. Which is probably just optimal for a load that spends a lot of time contended - new waiters touch that first cacheline, and then they queue themselves up on the second cacheline." After the commit, the rw_semaphore is at offset 128, which means the 'count' and 'owner' fields are now in the same cacheline, and causes more cache bouncing. Currently there are 3 "#ifdef CONFIG_XXX" before 'mmap_lock' which will affect its offset: CONFIG_MMU CONFIG_MEMBARRIER CONFIG_HAVE_ARCH_COMPAT_MMAP_BASES The layout above is on 64 bits system with 0day's default kernel config (similar to RHEL-8.3's config), in which all these 3 options are 'y'. And the layout can vary with different kernel configs. Relayouting a structure is usually a double-edged sword, as sometimes it can helps one case, but hurt other cases. For this case, one solution is, as the newly added 'write_protect_seq' is a 4 bytes long seqcount_t (when CONFIG_DEBUG_LOCK_ALLOC=n), placing it into an existing 4 bytes hole in 'mm_struct' will not change other fields' alignment, while restoring the regression. Link: https://lore.kernel.org/lkml/20210525031636.GB7744@xsang-OptiPlex-9020/ [1] Reported-by: kernel test robot <oliver.sang@intel.com> Signed-off-by: Feng Tang <feng.tang@intel.com> Reviewed-by: John Hubbard <jhubbard@nvidia.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Cc: Peter Xu <peterx@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> 
0day robot reported a 9.2% regression for will-it-scale mmap1 test
case[1], caused by commit 57efa1fe5957 ("mm/gup: prevent gup_fast from
racing with COW during fork").

Further debug shows the regression is due to that commit changes the
offset of hot fields 'mmap_lock' inside structure 'mm_struct', thus some
cache alignment changes.

From the perf data, the contention for 'mmap_lock' is very severe and
takes around 95% cpu cycles, and it is a rw_semaphore

        struct rw_semaphore {
                atomic_long_t count;	/* 8 bytes */
                atomic_long_t owner;	/* 8 bytes */
                struct optimistic_spin_queue osq; /* spinner MCS lock */
                ...

Before commit 57efa1fe5957 adds the 'write_protect_seq', it happens to
have a very optimal cache alignment layout, as Linus explained:

 "and before the addition of the 'write_protect_seq' field, the
  mmap_sem was at offset 120 in 'struct mm_struct'.

  Which meant that count and owner were in two different cachelines,
  and then when you have contention and spend time in
  rwsem_down_write_slowpath(), this is probably *exactly* the kind
  of layout you want.

  Because first the rwsem_write_trylock() will do a cmpxchg on the
  first cacheline (for the optimistic fast-path), and then in the
  case of contention, rwsem_down_write_slowpath() will just access
  the second cacheline.

  Which is probably just optimal for a load that spends a lot of
  time contended - new waiters touch that first cacheline, and then
  they queue themselves up on the second cacheline."

After the commit, the rw_semaphore is at offset 128, which means the
'count' and 'owner' fields are now in the same cacheline, and causes
more cache bouncing.

Currently there are 3 "#ifdef CONFIG_XXX" before 'mmap_lock' which will
affect its offset:

  CONFIG_MMU
  CONFIG_MEMBARRIER
  CONFIG_HAVE_ARCH_COMPAT_MMAP_BASES

The layout above is on 64 bits system with 0day's default kernel config
(similar to RHEL-8.3's config), in which all these 3 options are 'y'.
And the layout can vary with different kernel configs.

Relayouting a structure is usually a double-edged sword, as sometimes it
can helps one case, but hurt other cases.  For this case, one solution
is, as the newly added 'write_protect_seq' is a 4 bytes long seqcount_t
(when CONFIG_DEBUG_LOCK_ALLOC=n), placing it into an existing 4 bytes
hole in 'mm_struct' will not change other fields' alignment, while
restoring the regression.

Link: https://lore.kernel.org/lkml/20210525031636.GB7744@xsang-OptiPlex-9020/ [1]
Reported-by: kernel test robot <oliver.sang@intel.com>
Signed-off-by: Feng Tang <feng.tang@intel.com>
Reviewed-by: John Hubbard <jhubbard@nvidia.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Cc: Peter Xu <peterx@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm: add a signature in struct page 2021-06-07T21:11:47+00:00 Matteo Croce mcroce@microsoft.com 2021-06-07T19:02:36+00:00 c07aea3ef4d4076f18f567b98ed01e082e02ed51 This is needed by the page_pool to avoid recycling a page not allocated via page_pool. The page->signature field is aliased to page->lru.next and page->compound_head, but it can't be set by mistake because the signature value is a bad pointer, and can't trigger a false positive in PageTail() because the last bit is 0. Co-developed-by: Matthew Wilcox (Oracle) <willy@infradead.org> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Signed-off-by: Matteo Croce <mcroce@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net> 
This is needed by the page_pool to avoid recycling a page not allocated
via page_pool.

The page->signature field is aliased to page->lru.next and
page->compound_head, but it can't be set by mistake because the
signature value is a bad pointer, and can't trigger a false positive
in PageTail() because the last bit is 0.

Co-developed-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Matteo Croce <mcroce@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>