The patch titled
Subject: dax: stop using VM_MIXEDMAP for dax
has been added to the -mm tree. Its filename is
dax-stop-using-vm_mixedmap-for-dax.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/dax-stop-using-vm_mixedmap-for-dax.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/dax-stop-using-vm_mixedmap-for-dax.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Dan Williams <dan.j.williams@xxxxxxxxx>
Subject: dax: stop using VM_MIXEDMAP for dax
VM_MIXEDMAP is used by dax to direct mm paths like vm_normal_page() that
the memory page it is dealing with is not typical memory from the linear
map. The get_user_pages_fast() path, since it does not resolve the vma,
is already using {pte,pmd}_devmap() as a stand-in for VM_MIXEDMAP, so we
use that as a VM_MIXEDMAP replacement in some locations. In the cases
where there is no pte to consult we fallback to using vma_is_dax() to
detect the VM_MIXEDMAP special case.
Now that we always have pages for DAX we can stop setting VM_MIXEDMAP.
This also means we no longer need to worry about safely manipulating
vm_flags in a future where we support dynamically changing the dax mode of
a file.
Link: http://lkml.kernel.org/r/150664807800.36094.3685385297224300424.stgit@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Signed-off-by: Dan Williams <dan.j.williams@xxxxxxxxx>
Cc: Jan Kara <jack@xxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxx>
Cc: Jeff Moyer <jmoyer@xxxxxxxxxx>
Cc: Christoph Hellwig <hch@xxxxxx>
Cc: Ross Zwisler <ross.zwisler@xxxxxxxxxxxxxxx>
Cc: "Kirill A. Shutemov" <kirill.shutemov@xxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
drivers/dax/device.c | 2 +-
fs/ext2/file.c | 1 -
fs/ext4/file.c | 2 +-
fs/xfs/xfs_file.c | 2 +-
mm/huge_memory.c | 8 ++++----
mm/ksm.c | 3 +++
mm/madvise.c | 2 +-
mm/memory.c | 20 ++++++++++++++++++--
mm/migrate.c | 3 ++-
mm/mlock.c | 3 ++-
mm/mmap.c | 3 ++-
11 files changed, 35 insertions(+), 14 deletions(-)
diff -puN drivers/dax/device.c~dax-stop-using-vm_mixedmap-for-dax drivers/dax/device.c
--- a/drivers/dax/device.c~dax-stop-using-vm_mixedmap-for-dax
+++ a/drivers/dax/device.c
@@ -450,7 +450,7 @@ static int dax_mmap(struct file *filp, s
return rc;
vma->vm_ops = &dax_vm_ops;
- vma->vm_flags |= VM_MIXEDMAP | VM_HUGEPAGE;
+ vma->vm_flags |= VM_HUGEPAGE;
return 0;
}
diff -puN fs/ext2/file.c~dax-stop-using-vm_mixedmap-for-dax fs/ext2/file.c
--- a/fs/ext2/file.c~dax-stop-using-vm_mixedmap-for-dax
+++ a/fs/ext2/file.c
@@ -125,7 +125,6 @@ static int ext2_file_mmap(struct file *f
file_accessed(file);
vma->vm_ops = &ext2_dax_vm_ops;
- vma->vm_flags |= VM_MIXEDMAP;
return 0;
}
#else
diff -puN fs/ext4/file.c~dax-stop-using-vm_mixedmap-for-dax fs/ext4/file.c
--- a/fs/ext4/file.c~dax-stop-using-vm_mixedmap-for-dax
+++ a/fs/ext4/file.c
@@ -352,7 +352,7 @@ static int ext4_file_mmap(struct file *f
file_accessed(file);
if (IS_DAX(file_inode(file))) {
vma->vm_ops = &ext4_dax_vm_ops;
- vma->vm_flags |= VM_MIXEDMAP | VM_HUGEPAGE;
+ vma->vm_flags |= VM_HUGEPAGE;
} else {
vma->vm_ops = &ext4_file_vm_ops;
}
diff -puN fs/xfs/xfs_file.c~dax-stop-using-vm_mixedmap-for-dax fs/xfs/xfs_file.c
--- a/fs/xfs/xfs_file.c~dax-stop-using-vm_mixedmap-for-dax
+++ a/fs/xfs/xfs_file.c
@@ -1134,7 +1134,7 @@ xfs_file_mmap(
file_accessed(filp);
vma->vm_ops = &xfs_file_vm_ops;
if (IS_DAX(file_inode(filp)))
- vma->vm_flags |= VM_MIXEDMAP | VM_HUGEPAGE;
+ vma->vm_flags |= VM_HUGEPAGE;
return 0;
}
diff -puN mm/huge_memory.c~dax-stop-using-vm_mixedmap-for-dax mm/huge_memory.c
--- a/mm/huge_memory.c~dax-stop-using-vm_mixedmap-for-dax
+++ a/mm/huge_memory.c
@@ -765,11 +765,11 @@ int vmf_insert_pfn_pmd(struct vm_area_st
* but we need to be consistent with PTEs and architectures that
* can't support a 'special' bit.
*/
- BUG_ON(!(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)));
+ BUG_ON(!((vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))
+ || pfn_t_devmap(pfn)));
BUG_ON((vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)) ==
(VM_PFNMAP|VM_MIXEDMAP));
BUG_ON((vma->vm_flags & VM_PFNMAP) && is_cow_mapping(vma->vm_flags));
- BUG_ON(!pfn_t_devmap(pfn));
if (addr < vma->vm_start || addr >= vma->vm_end)
return VM_FAULT_SIGBUS;
@@ -824,11 +824,11 @@ int vmf_insert_pfn_pud(struct vm_area_st
* but we need to be consistent with PTEs and architectures that
* can't support a 'special' bit.
*/
- BUG_ON(!(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)));
+ BUG_ON(!((vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))
+ || pfn_t_devmap(pfn)));
BUG_ON((vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)) ==
(VM_PFNMAP|VM_MIXEDMAP));
BUG_ON((vma->vm_flags & VM_PFNMAP) && is_cow_mapping(vma->vm_flags));
- BUG_ON(!pfn_t_devmap(pfn));
if (addr < vma->vm_start || addr >= vma->vm_end)
return VM_FAULT_SIGBUS;
diff -puN mm/ksm.c~dax-stop-using-vm_mixedmap-for-dax mm/ksm.c
--- a/mm/ksm.c~dax-stop-using-vm_mixedmap-for-dax
+++ a/mm/ksm.c
@@ -2361,6 +2361,9 @@ int ksm_madvise(struct vm_area_struct *v
VM_HUGETLB | VM_MIXEDMAP))
return 0; /* just ignore the advice */
+ if (vma_is_dax(vma))
+ return 0;
+
#ifdef VM_SAO
if (*vm_flags & VM_SAO)
return 0;
diff -puN mm/madvise.c~dax-stop-using-vm_mixedmap-for-dax mm/madvise.c
--- a/mm/madvise.c~dax-stop-using-vm_mixedmap-for-dax
+++ a/mm/madvise.c
@@ -95,7 +95,7 @@ static long madvise_behavior(struct vm_a
new_flags |= VM_DONTDUMP;
break;
case MADV_DODUMP:
- if (new_flags & VM_SPECIAL) {
+ if (vma_is_dax(vma) || (new_flags & VM_SPECIAL)) {
error = -EINVAL;
goto out;
}
diff -puN mm/memory.c~dax-stop-using-vm_mixedmap-for-dax mm/memory.c
--- a/mm/memory.c~dax-stop-using-vm_mixedmap-for-dax
+++ a/mm/memory.c
@@ -830,6 +830,8 @@ struct page *_vm_normal_page(struct vm_a
return vma->vm_ops->find_special_page(vma, addr);
if (vma->vm_flags & (VM_PFNMAP | VM_MIXEDMAP))
return NULL;
+ if (pte_devmap(pte))
+ return NULL;
if (is_zero_pfn(pfn))
return NULL;
@@ -917,6 +919,8 @@ struct page *vm_normal_page_pmd(struct v
}
}
+ if (pmd_devmap(pmd))
+ return NULL;
if (is_zero_pfn(pfn))
return NULL;
if (unlikely(pfn > highest_memmap_pfn))
@@ -1227,7 +1231,7 @@ int copy_page_range(struct mm_struct *ds
* efficient than faulting.
*/
if (!(vma->vm_flags & (VM_HUGETLB | VM_PFNMAP | VM_MIXEDMAP)) &&
- !vma->anon_vma)
+ !vma->anon_vma && !vma_is_dax(vma))
return 0;
if (is_vm_hugetlb_page(vma))
@@ -1896,12 +1900,24 @@ int vm_insert_pfn_prot(struct vm_area_st
}
EXPORT_SYMBOL(vm_insert_pfn_prot);
+static bool vm_mixed_ok(struct vm_area_struct *vma, pfn_t pfn)
+{
+ /* these checks mirror the abort conditions in vm_normal_page */
+ if (vma->vm_flags & VM_MIXEDMAP)
+ return true;
+ if (pfn_t_devmap(pfn))
+ return true;
+ if (is_zero_pfn(pfn_t_to_pfn(pfn)))
+ return true;
+ return false;
+}
+
static int __vm_insert_mixed(struct vm_area_struct *vma, unsigned long addr,
pfn_t pfn, bool mkwrite)
{
pgprot_t pgprot = vma->vm_page_prot;
- BUG_ON(!(vma->vm_flags & VM_MIXEDMAP));
+ BUG_ON(!vm_mixed_ok(vma, pfn));
if (addr < vma->vm_start || addr >= vma->vm_end)
return -EFAULT;
diff -puN mm/migrate.c~dax-stop-using-vm_mixedmap-for-dax mm/migrate.c
--- a/mm/migrate.c~dax-stop-using-vm_mixedmap-for-dax
+++ a/mm/migrate.c
@@ -2928,7 +2928,8 @@ int migrate_vma(const struct migrate_vma
/* Sanity check the arguments */
start &= PAGE_MASK;
end &= PAGE_MASK;
- if (!vma || is_vm_hugetlb_page(vma) || (vma->vm_flags & VM_SPECIAL))
+ if (!vma || is_vm_hugetlb_page(vma) || (vma->vm_flags & VM_SPECIAL)
+ || vma_is_dax(dma))
return -EINVAL;
if (start < vma->vm_start || start >= vma->vm_end)
return -EINVAL;
diff -puN mm/mlock.c~dax-stop-using-vm_mixedmap-for-dax mm/mlock.c
--- a/mm/mlock.c~dax-stop-using-vm_mixedmap-for-dax
+++ a/mm/mlock.c
@@ -520,7 +520,8 @@ static int mlock_fixup(struct vm_area_st
vm_flags_t old_flags = vma->vm_flags;
if (newflags == vma->vm_flags || (vma->vm_flags & VM_SPECIAL) ||
- is_vm_hugetlb_page(vma) || vma == get_gate_vma(current->mm))
+ is_vm_hugetlb_page(vma) || vma == get_gate_vma(current->mm) ||
+ vma_is_dax(vma))
/* don't set VM_LOCKED or VM_LOCKONFAULT and don't count */
goto out;
diff -puN mm/mmap.c~dax-stop-using-vm_mixedmap-for-dax mm/mmap.c
--- a/mm/mmap.c~dax-stop-using-vm_mixedmap-for-dax
+++ a/mm/mmap.c
@@ -1723,7 +1723,8 @@ out:
vm_stat_account(mm, vm_flags, len >> PAGE_SHIFT);
if (vm_flags & VM_LOCKED) {
if (!((vm_flags & VM_SPECIAL) || is_vm_hugetlb_page(vma) ||
- vma == get_gate_vma(current->mm)))
+ vma == get_gate_vma(current->mm) ||
+ vma_is_dax(vma)))
mm->locked_vm += (len >> PAGE_SHIFT);
else
vma->vm_flags &= VM_LOCKED_CLEAR_MASK;
_
Patches currently in -mm which might be from dan.j.williams@xxxxxxxxx are
dax-quiet-bdev_dax_supported.patch
dax-disable-filesystem-dax-on-devices-that-do-not-map-pages.patch
dax-stop-using-vm_mixedmap-for-dax.patch
dax-stop-using-vm_hugepage-for-dax.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
