Commit 70b085b0 authored by Dave Jiang's avatar Dave Jiang Committed by Dan Williams

device-dax: fix pud fault fallback handling

Jeff Moyer reports:

    With a device dax alignment of 4KB or 2MB, I get sigbus when running
    the attached fio job file for the current kernel (4.11.0-rc1+).  If
    I specify an alignment of 1GB, it works.

    I turned on debug output, and saw that it was failing in the huge
    fault code.

     dax dax1.0: dax_open
     dax dax1.0: dax_mmap
     dax dax1.0: dax_dev_huge_fault: fio: write (0x7f08f0a00000 -
     dax dax1.0: __dax_dev_pud_fault: phys_to_pgoff(0xffffffffcf60)
     dax dax1.0: dax_release

    fio config for reproduce:
    [global]
    ioengine=dev-dax
    direct=0
    filename=/dev/dax0.0
    bs=2m

    [write]
    rw=write

    [read]
    stonewall
    rw=read

The driver fails to fallback when taking a fault that is larger than
the device alignment, or handling a larger fault when a smaller
mapping is already established. While we could support larger
mappings for a device with a smaller alignment, that change is
too large for the immediate fix. The simplest change is to force
fallback until the fault size matches the alignment.
Reported-by: default avatarJeff Moyer <jmoyer@redhat.com>
Signed-off-by: default avatarDave Jiang <dave.jiang@intel.com>
Signed-off-by: default avatarDan Williams <dan.j.williams@intel.com>
parent 0134ed4f
......@@ -518,6 +518,8 @@ static int __dax_dev_pud_fault(struct dax_dev *dax_dev, struct vm_fault *vmf)
phys_addr_t phys;
pgoff_t pgoff;
pfn_t pfn;
unsigned int fault_size = PUD_SIZE;
if (check_vma(dax_dev, vmf->vma, __func__))
return VM_FAULT_SIGBUS;
......@@ -534,6 +536,16 @@ static int __dax_dev_pud_fault(struct dax_dev *dax_dev, struct vm_fault *vmf)
return VM_FAULT_SIGBUS;
}
if (fault_size < dax_region->align)
return VM_FAULT_SIGBUS;
else if (fault_size > dax_region->align)
return VM_FAULT_FALLBACK;
/* if we are outside of the VMA */
if (pud_addr < vmf->vma->vm_start ||
(pud_addr + PUD_SIZE) > vmf->vma->vm_end)
return VM_FAULT_SIGBUS;
pgoff = linear_page_index(vmf->vma, pud_addr);
phys = pgoff_to_phys(dax_dev, pgoff, PUD_SIZE);
if (phys == -1) {
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment