The quilt patch titled
Subject: fixup! mm/pagewalk: check pfnmap for folio_walk_start()
has been removed from the -mm tree. Its filename was
mm-pagewalk-check-pfnmap-for-folio_walk_start-fix.patch
This patch was dropped because it was folded into mm-pagewalk-check-pfnmap-for-folio_walk_start.patch
------------------------------------------------------
From: Peter Xu <peterx@xxxxxxxxxx>
Subject: fixup! mm/pagewalk: check pfnmap for folio_walk_start()
Date: Thu, 29 Aug 2024 16:22:37 -0400
Per David's suggestion, remove some stale comment in vm_normal_page_pmd()
as we start to have special bit in pmd too. Meanwhile move the
pmd_special() check here from folio_walk_start().
Link: https://lkml.kernel.org/r/20240829202237.2640288-1-peterx@xxxxxxxxxx
Signed-off-by: Peter Xu <peterx@xxxxxxxxxx>
Cc: David Hildenbrand <david@xxxxxxxxxx>
Cc: Jason Gunthorpe <jgg@xxxxxxxxxx>
Cc: Alexander Gordeev <agordeev@xxxxxxxxxxxxx>
Cc: Alex Williamson <alex.williamson@xxxxxxxxxx>
Cc: Aneesh Kumar K.V <aneesh.kumar@xxxxxxxxxxxxx>
Cc: Borislav Petkov <bp@xxxxxxxxx>
Cc: Catalin Marinas <catalin.marinas@xxxxxxx>
Cc: Christian Borntraeger <borntraeger@xxxxxxxxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
Cc: Gavin Shan <gshan@xxxxxxxxxx>
Cc: Gerald Schaefer <gerald.schaefer@xxxxxxxxxxxxx>
Cc: Heiko Carstens <hca@xxxxxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
Cc: Niklas Schnelle <schnelle@xxxxxxxxxxxxx>
Cc: Paolo Bonzini <pbonzini@xxxxxxxxxx>
Cc: Ryan Roberts <ryan.roberts@xxxxxxx>
Cc: Sean Christopherson <seanjc@xxxxxxxxxx>
Cc: Sven Schnelle <svens@xxxxxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Vasily Gorbik <gor@xxxxxxxxxxxxx>
Cc: Will Deacon <will@xxxxxxxxxx>
Cc: Zi Yan <ziy@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/memory.c | 9 ++++-----
mm/pagewalk.c | 2 +-
2 files changed, 5 insertions(+), 6 deletions(-)
--- a/mm/memory.c~mm-pagewalk-check-pfnmap-for-folio_walk_start-fix
+++ a/mm/memory.c
@@ -672,11 +672,10 @@ struct page *vm_normal_page_pmd(struct v
{
unsigned long pfn = pmd_pfn(pmd);
- /*
- * There is no pmd_special() but there may be special pmds, e.g.
- * in a direct-access (dax) mapping, so let's just replicate the
- * !CONFIG_ARCH_HAS_PTE_SPECIAL case from vm_normal_page() here.
- */
+ /* Currently it's only used for huge pfnmaps */
+ if (unlikely(pmd_special(pmd)))
+ return NULL;
+
if (unlikely(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))) {
if (vma->vm_flags & VM_MIXEDMAP) {
if (!pfn_valid(pfn))
--- a/mm/pagewalk.c~mm-pagewalk-check-pfnmap-for-folio_walk_start-fix
+++ a/mm/pagewalk.c
@@ -783,7 +783,7 @@ pmd_table:
fw->pmdp = pmdp;
fw->pmd = pmd;
- if (pmd_none(pmd) || pmd_special(pmd)) {
+ if (pmd_none(pmd)) {
spin_unlock(ptl);
goto not_found;
} else if (!pmd_leaf(pmd)) {
_
Patches currently in -mm which might be from peterx@xxxxxxxxxx are
mm-introduce-arch_supports_huge_pfnmap-and-special-bits-to-pmd-pud.patch
mm-drop-is_huge_zero_pud.patch
mm-mark-special-bits-for-huge-pfn-mappings-when-inject.patch
mm-allow-thp-orders-for-pfnmaps.patch
mm-gup-detect-huge-pfnmap-entries-in-gup-fast.patch
mm-pagewalk-check-pfnmap-for-folio_walk_start.patch
mm-fork-accept-huge-pfnmap-entries.patch
mm-always-define-pxx_pgprot.patch
mm-new-follow_pfnmap-api.patch
kvm-use-follow_pfnmap-api.patch
s390-pci_mmio-use-follow_pfnmap-api.patch
mm-x86-pat-use-the-new-follow_pfnmap-api.patch
vfio-use-the-new-follow_pfnmap-api.patch
acrn-use-the-new-follow_pfnmap-api.patch
mm-access_process_vm-use-the-new-follow_pfnmap-api.patch
mm-remove-follow_pte.patch
mm-x86-support-large-pfn-mappings.patch
mm-arm64-support-large-pfn-mappings.patch
