| Message ID | 20230711202047.3818697-4-willy@infradead.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Avoid the mmap lock for fault-around | expand |
On Tue, Jul 11, 2023 at 1:20 PM Matthew Wilcox (Oracle) <willy@infradead.org> wrote: > > Handle a little more of the page fault path outside the mmap sem. > The hugetlb path doesn't need to check whether the VMA is anonymous; > the VM_HUGETLB flag is only set on hugetlbfs VMAs. There should be no > performance change from the previous commit; this is simply a step to > ease bisection of any problems. > > Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Reviewed-by: Suren Baghdasaryan <surenb@google.com> > --- > mm/hugetlb.c | 6 ++++++ > mm/memory.c | 18 +++++++++--------- > 2 files changed, 15 insertions(+), 9 deletions(-) > > diff --git a/mm/hugetlb.c b/mm/hugetlb.c > index e4a28ce0667f..109e1ff92bc8 100644 > --- a/mm/hugetlb.c > +++ b/mm/hugetlb.c > @@ -6063,6 +6063,12 @@ vm_fault_t hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, > int need_wait_lock = 0; > unsigned long haddr = address & huge_page_mask(h); > > + /* TODO: Handle faults under the VMA lock */ > + if (flags & FAULT_FLAG_VMA_LOCK) { > + vma_end_read(vma); > + return VM_FAULT_RETRY; > + } > + > /* > * Serialize hugepage allocation and instantiation, so that we don't > * get spurious allocation failures if two CPUs race to instantiate > diff --git a/mm/memory.c b/mm/memory.c > index f2dcc695f54e..6eda5c5f2069 100644 > --- a/mm/memory.c > +++ b/mm/memory.c > @@ -4998,10 +4998,10 @@ static vm_fault_t handle_pte_fault(struct vm_fault *vmf) > } > > /* > - * By the time we get here, we already hold the mm semaphore > - * > - * The mmap_lock may have been released depending on flags and our > - * return value. See filemap_fault() and __folio_lock_or_retry(). > + * On entry, we hold either the VMA lock or the mmap_lock > + * (FAULT_FLAG_VMA_LOCK tells you which). If VM_FAULT_RETRY is set in > + * the result, the mmap_lock is not held on exit. See filemap_fault() > + * and __folio_lock_or_retry(). > */ > static vm_fault_t __handle_mm_fault(struct vm_area_struct *vma, > unsigned long address, unsigned int flags) > @@ -5020,6 +5020,11 @@ static vm_fault_t __handle_mm_fault(struct vm_area_struct *vma, > p4d_t *p4d; > vm_fault_t ret; > > + if ((flags & FAULT_FLAG_VMA_LOCK) && !vma_is_anonymous(vma)) { > + vma_end_read(vma); > + return VM_FAULT_RETRY; > + } > + > pgd = pgd_offset(mm, address); > p4d = p4d_alloc(mm, pgd, address); > if (!p4d) > @@ -5247,11 +5252,6 @@ vm_fault_t handle_mm_fault(struct vm_area_struct *vma, unsigned long address, > goto out; > } > > - if ((flags & FAULT_FLAG_VMA_LOCK) && !vma_is_anonymous(vma)) { > - vma_end_read(vma); > - return VM_FAULT_RETRY; > - } > - > /* > * Enable the memcg OOM handling for faults triggered in user > * space. Kernel faults are handled more gracefully. > -- > 2.39.2 >
diff --git a/mm/hugetlb.c b/mm/hugetlb.c index e4a28ce0667f..109e1ff92bc8 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -6063,6 +6063,12 @@ vm_fault_t hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, int need_wait_lock = 0; unsigned long haddr = address & huge_page_mask(h); + /* TODO: Handle faults under the VMA lock */ + if (flags & FAULT_FLAG_VMA_LOCK) { + vma_end_read(vma); + return VM_FAULT_RETRY; + } + /* * Serialize hugepage allocation and instantiation, so that we don't * get spurious allocation failures if two CPUs race to instantiate diff --git a/mm/memory.c b/mm/memory.c index f2dcc695f54e..6eda5c5f2069 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -4998,10 +4998,10 @@ static vm_fault_t handle_pte_fault(struct vm_fault *vmf) } /* - * By the time we get here, we already hold the mm semaphore - * - * The mmap_lock may have been released depending on flags and our - * return value. See filemap_fault() and __folio_lock_or_retry(). + * On entry, we hold either the VMA lock or the mmap_lock + * (FAULT_FLAG_VMA_LOCK tells you which). If VM_FAULT_RETRY is set in + * the result, the mmap_lock is not held on exit. See filemap_fault() + * and __folio_lock_or_retry(). */ static vm_fault_t __handle_mm_fault(struct vm_area_struct *vma, unsigned long address, unsigned int flags) @@ -5020,6 +5020,11 @@ static vm_fault_t __handle_mm_fault(struct vm_area_struct *vma, p4d_t *p4d; vm_fault_t ret; + if ((flags & FAULT_FLAG_VMA_LOCK) && !vma_is_anonymous(vma)) { + vma_end_read(vma); + return VM_FAULT_RETRY; + } + pgd = pgd_offset(mm, address); p4d = p4d_alloc(mm, pgd, address); if (!p4d) @@ -5247,11 +5252,6 @@ vm_fault_t handle_mm_fault(struct vm_area_struct *vma, unsigned long address, goto out; } - if ((flags & FAULT_FLAG_VMA_LOCK) && !vma_is_anonymous(vma)) { - vma_end_read(vma); - return VM_FAULT_RETRY; - } - /* * Enable the memcg OOM handling for faults triggered in user * space. Kernel faults are handled more gracefully.
Handle a little more of the page fault path outside the mmap sem. The hugetlb path doesn't need to check whether the VMA is anonymous; the VM_HUGETLB flag is only set on hugetlbfs VMAs. There should be no performance change from the previous commit; this is simply a step to ease bisection of any problems. Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> --- mm/hugetlb.c | 6 ++++++ mm/memory.c | 18 +++++++++--------- 2 files changed, 15 insertions(+), 9 deletions(-)