1 files changed, 52 insertions, 95 deletions

diff --git a/fs/dax.c b/fs/dax.c
index ea0c35794bf9..289e6254aa30 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -20,12 +20,11 @@
 #include <linux/sched/signal.h>
 #include <linux/uio.h>
 #include <linux/vmstat.h>
-#include <linux/pfn_t.h>
 #include <linux/sizes.h>
 #include <linux/mmu_notifier.h>
 #include <linux/iomap.h>
 #include <linux/rmap.h>
-#include <asm/pgalloc.h>
+#include <linux/pgalloc.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/fs_dax.h>
@@ -76,9 +75,9 @@ static struct folio *dax_to_folio(void *entry)
 	return page_folio(pfn_to_page(dax_to_pfn(entry)));
 }
 
-static void *dax_make_entry(pfn_t pfn, unsigned long flags)
+static void *dax_make_entry(unsigned long pfn, unsigned long flags)
 {
-	return xa_mk_value(flags | (pfn_t_to_pfn(pfn) << DAX_SHIFT));
+	return xa_mk_value(flags | (pfn << DAX_SHIFT));
 }
 
 static bool dax_is_locked(void *entry)
@@ -449,9 +448,6 @@ static void dax_associate_entry(void *entry, struct address_space *mapping,
 	if (dax_is_zero_entry(entry) || dax_is_empty_entry(entry))
 		return;
 
-	if (IS_ENABLED(CONFIG_FS_DAX_LIMITED))
-		return;
-
 	index = linear_page_index(vma, address & ~(size - 1));
 	if (shared && (folio->mapping || dax_folio_is_shared(folio))) {
 		if (folio->mapping)
@@ -474,9 +470,6 @@ static void dax_disassociate_entry(void *entry, struct address_space *mapping,
 {
 	struct folio *folio = dax_to_folio(entry);
 
-	if (IS_ENABLED(CONFIG_FS_DAX_LIMITED))
-		return;
-
 	if (dax_is_zero_entry(entry) || dax_is_empty_entry(entry))
 		return;
 
@@ -719,7 +712,7 @@ retry:
 
 		if (order > 0)
 			flags |= DAX_PMD;
-		entry = dax_make_entry(pfn_to_pfn_t(0), flags);
+		entry = dax_make_entry(0, flags);
 		dax_lock_entry(xas, entry);
 		if (xas_error(xas))
 			goto out_unlock;
@@ -768,12 +761,6 @@ struct page *dax_layout_busy_page_range(struct address_space *mapping,
 	pgoff_t end_idx;
 	XA_STATE(xas, &mapping->i_pages, start_idx);
 
-	/*
-	 * In the 'limited' case get_user_pages() for dax is disabled.
-	 */
-	if (IS_ENABLED(CONFIG_FS_DAX_LIMITED))
-		return NULL;
-
 	if (!dax_mapping(mapping))
 		return NULL;
 
@@ -1053,7 +1040,7 @@ static bool dax_fault_is_synchronous(const struct iomap_iter *iter,
  * appropriate.
  */
 static void *dax_insert_entry(struct xa_state *xas, struct vm_fault *vmf,
-		const struct iomap_iter *iter, void *entry, pfn_t pfn,
+		const struct iomap_iter *iter, void *entry, unsigned long pfn,
 		unsigned long flags)
 {
 	struct address_space *mapping = vmf->vma->vm_file->f_mapping;
@@ -1251,7 +1238,7 @@ int dax_writeback_mapping_range(struct address_space *mapping,
 EXPORT_SYMBOL_GPL(dax_writeback_mapping_range);
 
 static int dax_iomap_direct_access(const struct iomap *iomap, loff_t pos,
-		size_t size, void **kaddr, pfn_t *pfnp)
+		size_t size, void **kaddr, unsigned long *pfnp)
 {
 	pgoff_t pgoff = dax_iomap_pgoff(iomap, pos);
 	int id, rc = 0;
@@ -1269,7 +1256,7 @@ static int dax_iomap_direct_access(const struct iomap *iomap, loff_t pos,
 	rc = -EINVAL;
 	if (PFN_PHYS(length) < size)
 		goto out;
-	if (pfn_t_to_pfn(*pfnp) & (PHYS_PFN(size)-1))
+	if (*pfnp & (PHYS_PFN(size)-1))
 		goto out;
 
 	rc = 0;
@@ -1373,12 +1360,12 @@ static vm_fault_t dax_load_hole(struct xa_state *xas, struct vm_fault *vmf,
 {
 	struct inode *inode = iter->inode;
 	unsigned long vaddr = vmf->address;
-	pfn_t pfn = pfn_to_pfn_t(my_zero_pfn(vaddr));
+	unsigned long pfn = my_zero_pfn(vaddr);
 	vm_fault_t ret;
 
 	*entry = dax_insert_entry(xas, vmf, iter, *entry, pfn, DAX_ZERO_PAGE);
 
-	ret = vmf_insert_page_mkwrite(vmf, pfn_t_to_page(pfn), false);
+	ret = vmf_insert_page_mkwrite(vmf, pfn_to_page(pfn), false);
 	trace_dax_load_hole(inode, vmf, ret);
 	return ret;
 }
@@ -1388,51 +1375,24 @@ static vm_fault_t dax_pmd_load_hole(struct xa_state *xas, struct vm_fault *vmf,
 		const struct iomap_iter *iter, void **entry)
 {
 	struct address_space *mapping = vmf->vma->vm_file->f_mapping;
-	unsigned long pmd_addr = vmf->address & PMD_MASK;
-	struct vm_area_struct *vma = vmf->vma;
 	struct inode *inode = mapping->host;
-	pgtable_t pgtable = NULL;
 	struct folio *zero_folio;
-	spinlock_t *ptl;
-	pmd_t pmd_entry;
-	pfn_t pfn;
+	vm_fault_t ret;
 
 	zero_folio = mm_get_huge_zero_folio(vmf->vma->vm_mm);
 
-	if (unlikely(!zero_folio))
-		goto fallback;
-
-	pfn = page_to_pfn_t(&zero_folio->page);
-	*entry = dax_insert_entry(xas, vmf, iter, *entry, pfn,
-				  DAX_PMD | DAX_ZERO_PAGE);
-
-	if (arch_needs_pgtable_deposit()) {
-		pgtable = pte_alloc_one(vma->vm_mm);
-		if (!pgtable)
-			return VM_FAULT_OOM;
-	}
-
-	ptl = pmd_lock(vmf->vma->vm_mm, vmf->pmd);
-	if (!pmd_none(*(vmf->pmd))) {
-		spin_unlock(ptl);
-		goto fallback;
+	if (unlikely(!zero_folio)) {
+		trace_dax_pmd_load_hole_fallback(inode, vmf, zero_folio, *entry);
+		return VM_FAULT_FALLBACK;
 	}
 
-	if (pgtable) {
-		pgtable_trans_huge_deposit(vma->vm_mm, vmf->pmd, pgtable);
-		mm_inc_nr_ptes(vma->vm_mm);
-	}
-	pmd_entry = folio_mk_pmd(zero_folio, vmf->vma->vm_page_prot);
-	set_pmd_at(vmf->vma->vm_mm, pmd_addr, vmf->pmd, pmd_entry);
-	spin_unlock(ptl);
-	trace_dax_pmd_load_hole(inode, vmf, zero_folio, *entry);
-	return VM_FAULT_NOPAGE;
+	*entry = dax_insert_entry(xas, vmf, iter, *entry, folio_pfn(zero_folio),
+				  DAX_PMD | DAX_ZERO_PAGE);
 
-fallback:
-	if (pgtable)
-		pte_free(vma->vm_mm, pgtable);
-	trace_dax_pmd_load_hole_fallback(inode, vmf, zero_folio, *entry);
-	return VM_FAULT_FALLBACK;
+	ret = vmf_insert_folio_pmd(vmf, zero_folio, false);
+	if (ret == VM_FAULT_NOPAGE)
+		trace_dax_pmd_load_hole(inode, vmf, zero_folio, *entry);
+	return ret;
 }
 #else
 static vm_fault_t dax_pmd_load_hole(struct xa_state *xas, struct vm_fault *vmf,
@@ -1547,7 +1507,7 @@ static int dax_zero_iter(struct iomap_iter *iter, bool *did_zero)
 
 	/* already zeroed?  we're done. */
 	if (srcmap->type == IOMAP_HOLE || srcmap->type == IOMAP_UNWRITTEN)
-		return iomap_iter_advance(iter, &length);
+		return iomap_iter_advance(iter, length);
 
 	/*
 	 * invalidate the pages whose sharing state is to be changed
@@ -1576,10 +1536,10 @@ static int dax_zero_iter(struct iomap_iter *iter, bool *did_zero)
 		if (ret < 0)
 			return ret;
 
-		ret = iomap_iter_advance(iter, &length);
+		ret = iomap_iter_advance(iter, length);
 		if (ret)
 			return ret;
-	} while (length > 0);
+	} while ((length = iomap_length(iter)) > 0);
 
 	if (did_zero)
 		*did_zero = true;
@@ -1637,7 +1597,7 @@ static int dax_iomap_iter(struct iomap_iter *iomi, struct iov_iter *iter)
 
 		if (iomap->type == IOMAP_HOLE || iomap->type == IOMAP_UNWRITTEN) {
 			done = iov_iter_zero(min(length, end - pos), iter);
-			return iomap_iter_advance(iomi, &done);
+			return iomap_iter_advance(iomi, done);
 		}
 	}
 
@@ -1721,12 +1681,12 @@ static int dax_iomap_iter(struct iomap_iter *iomi, struct iov_iter *iter)
 			xfer = dax_copy_to_iter(dax_dev, pgoff, kaddr,
 					map_len, iter);
 
-		length = xfer;
-		ret = iomap_iter_advance(iomi, &length);
+		ret = iomap_iter_advance(iomi, xfer);
 		if (!ret && xfer == 0)
 			ret = -EFAULT;
 		if (xfer < map_len)
 			break;
+		length = iomap_length(iomi);
 	}
 	dax_read_unlock(id);
 
@@ -1756,13 +1716,16 @@ dax_iomap_rw(struct kiocb *iocb, struct iov_iter *iter,
 	loff_t done = 0;
 	int ret;
 
+	if (WARN_ON_ONCE(iocb->ki_flags & IOCB_ATOMIC))
+		return -EIO;
+
 	if (!iomi.len)
 		return 0;
 
 	if (iov_iter_rw(iter) == WRITE) {
 		lockdep_assert_held_write(&iomi.inode->i_rwsem);
 		iomi.flags |= IOMAP_WRITE;
-	} else {
+	} else if (!sb_rdonly(iomi.inode->i_sb)) {
 		lockdep_assert_held(&iomi.inode->i_rwsem);
 	}
 
@@ -1791,7 +1754,8 @@ static vm_fault_t dax_fault_return(int error)
  * insertion for now and return the pfn so that caller can insert it after the
  * fsync is done.
  */
-static vm_fault_t dax_fault_synchronous_pfnp(pfn_t *pfnp, pfn_t pfn)
+static vm_fault_t dax_fault_synchronous_pfnp(unsigned long *pfnp,
+					unsigned long pfn)
 {
 	if (WARN_ON_ONCE(!pfnp))
 		return VM_FAULT_SIGBUS;
@@ -1839,7 +1803,7 @@ static vm_fault_t dax_fault_cow_page(struct vm_fault *vmf,
  * @pmd:	distinguish whether it is a pmd fault
  */
 static vm_fault_t dax_fault_iter(struct vm_fault *vmf,
-		const struct iomap_iter *iter, pfn_t *pfnp,
+		const struct iomap_iter *iter, unsigned long *pfnp,
 		struct xa_state *xas, void **entry, bool pmd)
 {
 	const struct iomap *iomap = &iter->iomap;
@@ -1850,7 +1814,7 @@ static vm_fault_t dax_fault_iter(struct vm_fault *vmf,
 	unsigned long entry_flags = pmd ? DAX_PMD : 0;
 	struct folio *folio;
 	int ret, err = 0;
-	pfn_t pfn;
+	unsigned long pfn;
 	void *kaddr;
 
 	if (!pmd && vmf->cow_page)
@@ -1887,16 +1851,15 @@ static vm_fault_t dax_fault_iter(struct vm_fault *vmf,
 
 	folio_ref_inc(folio);
 	if (pmd)
-		ret = vmf_insert_folio_pmd(vmf, pfn_folio(pfn_t_to_pfn(pfn)),
-					write);
+		ret = vmf_insert_folio_pmd(vmf, pfn_folio(pfn), write);
 	else
-		ret = vmf_insert_page_mkwrite(vmf, pfn_t_to_page(pfn), write);
+		ret = vmf_insert_page_mkwrite(vmf, pfn_to_page(pfn), write);
 	folio_put(folio);
 
 	return ret;
 }
 
-static vm_fault_t dax_iomap_pte_fault(struct vm_fault *vmf, pfn_t *pfnp,
+static vm_fault_t dax_iomap_pte_fault(struct vm_fault *vmf, unsigned long *pfnp,
 			       int *iomap_errp, const struct iomap_ops *ops)
 {
 	struct address_space *mapping = vmf->vma->vm_file->f_mapping;
@@ -1937,7 +1900,7 @@ static vm_fault_t dax_iomap_pte_fault(struct vm_fault *vmf, pfn_t *pfnp,
 	 * the PTE we need to set up.  If so just return and the fault will be
 	 * retried.
 	 */
-	if (pmd_trans_huge(*vmf->pmd) || pmd_devmap(*vmf->pmd)) {
+	if (pmd_trans_huge(*vmf->pmd)) {
 		ret = VM_FAULT_NOPAGE;
 		goto unlock_entry;
 	}
@@ -1956,10 +1919,8 @@ static vm_fault_t dax_iomap_pte_fault(struct vm_fault *vmf, pfn_t *pfnp,
 			ret |= VM_FAULT_MAJOR;
 		}
 
-		if (!(ret & VM_FAULT_ERROR)) {
-			u64 length = PAGE_SIZE;
-			iter.status = iomap_iter_advance(&iter, &length);
-		}
+		if (!(ret & VM_FAULT_ERROR))
+			iter.status = iomap_iter_advance(&iter, PAGE_SIZE);
 	}
 
 	if (iomap_errp)
@@ -2008,7 +1969,7 @@ static bool dax_fault_check_fallback(struct vm_fault *vmf, struct xa_state *xas,
 	return false;
 }
 
-static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp,
+static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, unsigned long *pfnp,
 			       const struct iomap_ops *ops)
 {
 	struct address_space *mapping = vmf->vma->vm_file->f_mapping;
@@ -2060,8 +2021,7 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp,
 	 * the PMD we need to set up.  If so just return and the fault will be
 	 * retried.
 	 */
-	if (!pmd_none(*vmf->pmd) && !pmd_trans_huge(*vmf->pmd) &&
-			!pmd_devmap(*vmf->pmd)) {
+	if (!pmd_none(*vmf->pmd) && !pmd_trans_huge(*vmf->pmd)) {
 		ret = 0;
 		goto unlock_entry;
 	}
@@ -2072,10 +2032,8 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp,
 			continue; /* actually breaks out of the loop */
 
 		ret = dax_fault_iter(vmf, &iter, pfnp, &xas, &entry, true);
-		if (ret != VM_FAULT_FALLBACK) {
-			u64 length = PMD_SIZE;
-			iter.status = iomap_iter_advance(&iter, &length);
-		}
+		if (ret != VM_FAULT_FALLBACK)
+			iter.status = iomap_iter_advance(&iter, PMD_SIZE);
 	}
 
 unlock_entry:
@@ -2090,7 +2048,7 @@ out:
 	return ret;
 }
 #else
-static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp,
+static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, unsigned long *pfnp,
 			       const struct iomap_ops *ops)
 {
 	return VM_FAULT_FALLBACK;
@@ -2111,7 +2069,8 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp,
  * successfully.
  */
 vm_fault_t dax_iomap_fault(struct vm_fault *vmf, unsigned int order,
-		    pfn_t *pfnp, int *iomap_errp, const struct iomap_ops *ops)
+			unsigned long *pfnp, int *iomap_errp,
+			const struct iomap_ops *ops)
 {
 	if (order == 0)
 		return dax_iomap_pte_fault(vmf, pfnp, iomap_errp, ops);
@@ -2131,8 +2090,8 @@ EXPORT_SYMBOL_GPL(dax_iomap_fault);
  * This function inserts a writeable PTE or PMD entry into the page tables
  * for an mmaped DAX file.  It also marks the page cache entry as dirty.
  */
-static vm_fault_t
-dax_insert_pfn_mkwrite(struct vm_fault *vmf, pfn_t pfn, unsigned int order)
+static vm_fault_t dax_insert_pfn_mkwrite(struct vm_fault *vmf,
+					unsigned long pfn, unsigned int order)
 {
 	struct address_space *mapping = vmf->vma->vm_file->f_mapping;
 	XA_STATE_ORDER(xas, &mapping->i_pages, vmf->pgoff, order);
@@ -2154,7 +2113,7 @@ dax_insert_pfn_mkwrite(struct vm_fault *vmf, pfn_t pfn, unsigned int order)
 	xas_set_mark(&xas, PAGECACHE_TAG_DIRTY);
 	dax_lock_entry(&xas, entry);
 	xas_unlock_irq(&xas);
-	folio = pfn_folio(pfn_t_to_pfn(pfn));
+	folio = pfn_folio(pfn);
 	folio_ref_inc(folio);
 	if (order == 0)
 		ret = vmf_insert_page_mkwrite(vmf, &folio->page, true);
@@ -2181,7 +2140,7 @@ dax_insert_pfn_mkwrite(struct vm_fault *vmf, pfn_t pfn, unsigned int order)
  * table entry.
  */
 vm_fault_t dax_finish_sync_fault(struct vm_fault *vmf, unsigned int order,
-		pfn_t pfn)
+		unsigned long pfn)
 {
 	int err;
 	loff_t start = ((loff_t)vmf->pgoff) << PAGE_SHIFT;
@@ -2200,7 +2159,6 @@ static int dax_range_compare_iter(struct iomap_iter *it_src,
 	const struct iomap *smap = &it_src->iomap;
 	const struct iomap *dmap = &it_dest->iomap;
 	loff_t pos1 = it_src->pos, pos2 = it_dest->pos;
-	u64 dest_len;
 	void *saddr, *daddr;
 	int id, ret;
 
@@ -2233,10 +2191,9 @@ static int dax_range_compare_iter(struct iomap_iter *it_src,
 	dax_read_unlock(id);
 
 advance:
-	dest_len = len;
-	ret = iomap_iter_advance(it_src, &len);
+	ret = iomap_iter_advance(it_src, len);
 	if (!ret)
-		ret = iomap_iter_advance(it_dest, &dest_len);
+		ret = iomap_iter_advance(it_dest, len);
 	return ret;
 
 out_unlock: