pgd_entry() and pud_entry() were removed by commit 0b1fbfe50006c410 ("mm/pagewalk: remove pgd_entry() and pud_entry()") because there were no users. We're about to add users so reintroduce them, along with p4d_entry() as we now have 5 levels of tables. Note that commit a00cc7d9dd93d66a ("mm, x86: add support for PUD-sized transparent hugepages") already re-added pud_entry() but with different semantics to the other callbacks. This commit reverts the semantics back to match the other callbacks. To support hmm.c which now uses the new semantics of pud_entry() a new member ('action') of struct mm_walk is added which allows the callbacks to either descend (ACTION_SUBTREE, the default), skip (ACTION_CONTINUE) or repeat the callback (ACTION_AGAIN). hmm.c is then updated to call pud_trans_huge_lock() itself and make use of the splitting/retry logic of the core code. After this change pud_entry() is called for all entries, not just transparent huge pages. [arnd@arndb.de: fix unused variable warning] Link: http://lkml.kernel.org/r/20200107204607.1533842-1-arnd@arndb.de Link: http://lkml.kernel.org/r/20191218162402.45610-12-steven.price@arm.com Signed-off-by: Steven Price <steven.price@arm.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Albert Ou <aou@eecs.berkeley.edu> Cc: Alexandre Ghiti <alex@ghiti.fr> Cc: Andy Lutomirski <luto@kernel.org> Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: David S. Miller <davem@davemloft.net> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Hogan <jhogan@kernel.org> Cc: James Morse <james.morse@arm.com> Cc: Jerome Glisse <jglisse@redhat.com> Cc: "Liang, Kan" <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Paul Burton <paul.burton@mips.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Paul Walmsley <paul.walmsley@sifive.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vasily Gorbik <gor@linux.ibm.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Will Deacon <will@kernel.org> Cc: Zong Li <zong.li@sifive.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Showing the changes to include/linux (1 of 3 files). Show all
Unified · Split
| @@ -8,15 +8,15 @@ struct mm_walk; · 7 unmodified lines | |||
| 8 | 8 | ||
| 9 | /** | 9 | /** |
| 10 | * mm_walk_ops - callbacks for walk_page_range | 10 | * mm_walk_ops - callbacks for walk_page_range |
| 11 | * @pud_entry: if set, called for each non-empty PUD (2nd-level) entry | 11 | * @pgd_entry: if set, called for each non-empty PGD (top-level) entry |
| 12 | * this handler should only handle pud_trans_huge() puds. | 12 | * @p4d_entry: if set, called for each non-empty P4D entry |
| 13 | * the pmd_entry or pte_entry callbacks will be used for | 13 | * @pud_entry: if set, called for each non-empty PUD entry |
| 14 | * regular PUDs. | 14 | * @pmd_entry: if set, called for each non-empty PMD entry |
| 15 | * @pmd_entry: if set, called for each non-empty PMD (3rd-level) entry | ||
| 16 | * this handler is required to be able to handle | 15 | * this handler is required to be able to handle |
| 17 | * pmd_trans_huge() pmds. They may simply choose to | 16 | * pmd_trans_huge() pmds. They may simply choose to |
| 18 | * split_huge_page() instead of handling it explicitly. | 17 | * split_huge_page() instead of handling it explicitly. |
| 19 | * @pte_entry: if set, called for each non-empty PTE (4th-level) entry | 18 | * @pte_entry: if set, called for each non-empty PTE (lowest-level) |
| 19 | * entry | ||
| 20 | * @pte_hole: if set, called for each hole at all levels | 20 | * @pte_hole: if set, called for each hole at all levels |
| 21 | * @hugetlb_entry: if set, called for each hugetlb entry | 21 | * @hugetlb_entry: if set, called for each hugetlb entry |
| 22 | * @test_walk: caller specific callback function to determine whether | 22 | * @test_walk: caller specific callback function to determine whether |
| @@ -27,8 +27,15 @@ struct mm_walk; · 4 unmodified lines | |||
| 27 | * @pre_vma: if set, called before starting walk on a non-null vma. | 27 | * @pre_vma: if set, called before starting walk on a non-null vma. |
| 28 | * @post_vma: if set, called after a walk on a non-null vma, provided | 28 | * @post_vma: if set, called after a walk on a non-null vma, provided |
| 29 | * that @pre_vma and the vma walk succeeded. | 29 | * that @pre_vma and the vma walk succeeded. |
| 30 | * | ||
| 31 | * p?d_entry callbacks are called even if those levels are folded on a | ||
| 32 | * particular architecture/configuration. | ||
| 30 | */ | 33 | */ |
| 31 | struct mm_walk_ops { | 34 | struct mm_walk_ops { |
| 35 | int (*pgd_entry)(pgd_t *pgd, unsigned long addr, | ||
| 36 | unsigned long next, struct mm_walk *walk); | ||
| 37 | int (*p4d_entry)(p4d_t *p4d, unsigned long addr, | ||
| 38 | unsigned long next, struct mm_walk *walk); | ||
| 32 | int (*pud_entry)(pud_t *pud, unsigned long addr, | 39 | int (*pud_entry)(pud_t *pud, unsigned long addr, |
| 33 | unsigned long next, struct mm_walk *walk); | 40 | unsigned long next, struct mm_walk *walk); |
| 34 | int (*pmd_entry)(pmd_t *pmd, unsigned long addr, | 41 | int (*pmd_entry)(pmd_t *pmd, unsigned long addr, |
| @@ -47,11 +54,25 @@ struct mm_walk_ops { · 12 unmodified lines | |||
| 47 | void (*post_vma)(struct mm_walk *walk); | 54 | void (*post_vma)(struct mm_walk *walk); |
| 48 | }; | 55 | }; |
| 49 | 56 | ||
| 57 | /* | ||
| 58 | * Action for pud_entry / pmd_entry callbacks. | ||
| 59 | * ACTION_SUBTREE is the default | ||
| 60 | */ | ||
| 61 | enum page_walk_action { | ||
| 62 | /* Descend to next level, splitting huge pages if needed and possible */ | ||
| 63 | ACTION_SUBTREE = 0, | ||
| 64 | /* Continue to next entry at this level (ignoring any subtree) */ | ||
| 65 | ACTION_CONTINUE = 1, | ||
| 66 | /* Call again for this entry */ | ||
| 67 | ACTION_AGAIN = 2 | ||
| 68 | }; | ||
| 69 | |||
| 50 | /** | 70 | /** |
| 51 | * mm_walk - walk_page_range data | 71 | * mm_walk - walk_page_range data |
| 52 | * @ops: operation to call during the walk | 72 | * @ops: operation to call during the walk |
| 53 | * @mm: mm_struct representing the target process of page table walk | 73 | * @mm: mm_struct representing the target process of page table walk |
| 54 | * @vma: vma currently walked (NULL if walking outside vmas) | 74 | * @vma: vma currently walked (NULL if walking outside vmas) |
| 75 | * @action: next action to perform (see enum page_walk_action) | ||
| 55 | * @private: private data for callbacks' usage | 76 | * @private: private data for callbacks' usage |
| 56 | * | 77 | * |
| 57 | * (see the comment on walk_page_range() for more details) | 78 | * (see the comment on walk_page_range() for more details) |
| @@ -60,6 +81,7 @@ struct mm_walk { · 2 unmodified lines | |||
| 60 | const struct mm_walk_ops *ops; | 81 | const struct mm_walk_ops *ops; |
| 61 | struct mm_struct *mm; | 82 | struct mm_struct *mm; |
| 62 | struct vm_area_struct *vma; | 83 | struct vm_area_struct *vma; |
| 84 | enum page_walk_action action; | ||
| 63 | void *private; | 85 | void *private; |
| 64 | }; | 86 | }; |
| 65 | 87 | ||