dax: introduce DAX_RECOVERY_WRITE dax access mode

Up till now, dax_direct_access() is used implicitly for normal
access, but for the purpose of recovery write, dax range with
poison is requested. To make the interface clear, introduce
enum dax_access_mode {
DAX_ACCESS,
DAX_RECOVERY_WRITE,
}
where DAX_ACCESS is used for normal dax access, and
DAX_RECOVERY_WRITE is used for dax recovery write.

Suggested-by: Dan Williams <[e-mail]>
Signed-off-by: Jane Chu <[e-mail]>
Reviewed-by: Christoph Hellwig <[e-mail]>
Cc: Mike Snitzer <[e-mail]>
Reviewed-by: Vivek Goyal <[e-mail]>
Link: https://lore.kernel.org/r/[e-mail]
Signed-off-by: Dan Williams <[e-mail]>

drivers/dax/super.c

···
  * @dax_dev: a dax_device instance representing the logical memory range
  * @pgoff: offset in pages from the start of the device to translate
  * @nr_pages: number of consecutive pages caller can handle relative to @pfn
+ * @mode: indicator on normal access or recovery write
  * @kaddr: output parameter that returns a virtual address mapping of pfn
  * @pfn: output parameter that returns an absolute pfn translation of @pgoff
  *
···
  * pages accessible at the device relative @pgoff.
  */
 long dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff, long nr_pages,
-		void **kaddr, pfn_t *pfn)
+		enum dax_access_mode mode, void **kaddr, pfn_t *pfn)
 {
 	long avail;
 
···
 		return -EINVAL;
 
 	avail = dax_dev->ops->direct_access(dax_dev, pgoff, nr_pages,
-			kaddr, pfn);
+			mode, kaddr, pfn);
 	if (!avail)
 		return -ERANGE;
 	return min(avail, nr_pages);

drivers/md/dm-linear.c

···
 }
 
 static long linear_dax_direct_access(struct dm_target *ti, pgoff_t pgoff,
-		long nr_pages, void **kaddr, pfn_t *pfn)
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		pfn_t *pfn)
 {
 	struct dax_device *dax_dev = linear_dax_pgoff(ti, &pgoff);
 
-	return dax_direct_access(dax_dev, pgoff, nr_pages, kaddr, pfn);
+	return dax_direct_access(dax_dev, pgoff, nr_pages, mode, kaddr, pfn);
 }
 
 static int linear_dax_zero_page_range(struct dm_target *ti, pgoff_t pgoff,

drivers/md/dm-log-writes.c

···
 }
 
 static long log_writes_dax_direct_access(struct dm_target *ti, pgoff_t pgoff,
-					 long nr_pages, void **kaddr, pfn_t *pfn)
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		pfn_t *pfn)
 {
 	struct dax_device *dax_dev = log_writes_dax_pgoff(ti, &pgoff);
 
-	return dax_direct_access(dax_dev, pgoff, nr_pages, kaddr, pfn);
+	return dax_direct_access(dax_dev, pgoff, nr_pages, mode, kaddr, pfn);
 }
 
 static int log_writes_dax_zero_page_range(struct dm_target *ti, pgoff_t pgoff,

drivers/md/dm-stripe.c

···
 }
 
 static long stripe_dax_direct_access(struct dm_target *ti, pgoff_t pgoff,
-		long nr_pages, void **kaddr, pfn_t *pfn)
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		pfn_t *pfn)
 {
 	struct dax_device *dax_dev = stripe_dax_pgoff(ti, &pgoff);
 
-	return dax_direct_access(dax_dev, pgoff, nr_pages, kaddr, pfn);
+	return dax_direct_access(dax_dev, pgoff, nr_pages, mode, kaddr, pfn);
 }
 
 static int stripe_dax_zero_page_range(struct dm_target *ti, pgoff_t pgoff,

drivers/md/dm-target.c

···
 #include <linux/init.h>
 #include <linux/kmod.h>
 #include <linux/bio.h>
+#include <linux/dax.h>
 
 #define DM_MSG_PREFIX "target"
 
···
 }
 
 static long io_err_dax_direct_access(struct dm_target *ti, pgoff_t pgoff,
-		long nr_pages, void **kaddr, pfn_t *pfn)
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		pfn_t *pfn)
 {
 	return -EIO;
 }

drivers/md/dm-writecache.c

···
 
 	id = dax_read_lock();
 
-	da = dax_direct_access(wc->ssd_dev->dax_dev, offset, p, &wc->memory_map, &pfn);
+	da = dax_direct_access(wc->ssd_dev->dax_dev, offset, p, DAX_ACCESS,
+			&wc->memory_map, &pfn);
 	if (da < 0) {
 		wc->memory_map = NULL;
 		r = da;
···
 		i = 0;
 		do {
 			long daa;
-			daa = dax_direct_access(wc->ssd_dev->dax_dev, offset + i, p - i,
-						NULL, &pfn);
+			daa = dax_direct_access(wc->ssd_dev->dax_dev, offset + i,
+					p - i, DAX_ACCESS, NULL, &pfn);
 			if (daa <= 0) {
 				r = daa ? daa : -EINVAL;
 				goto err3;

drivers/md/dm.c

···
 }
 
 static long dm_dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff,
-				 long nr_pages, void **kaddr, pfn_t *pfn)
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		pfn_t *pfn)
 {
 	struct mapped_device *md = dax_get_private(dax_dev);
 	sector_t sector = pgoff * PAGE_SECTORS;
···
 	if (len < 1)
 		goto out;
 	nr_pages = min(len, nr_pages);
-	ret = ti->type->direct_access(ti, pgoff, nr_pages, kaddr, pfn);
+	ret = ti->type->direct_access(ti, pgoff, nr_pages, mode, kaddr, pfn);
 
  out:
 	dm_put_live_table(md, srcu_idx);

drivers/nvdimm/pmem.c

···
 
 /* see "strong" declaration in tools/testing/nvdimm/pmem-dax.c */
 __weak long __pmem_direct_access(struct pmem_device *pmem, pgoff_t pgoff,
-		long nr_pages, void **kaddr, pfn_t *pfn)
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		pfn_t *pfn)
 {
 	resource_size_t offset = PFN_PHYS(pgoff) + pmem->data_offset;
 
···
 }
 
 static long pmem_dax_direct_access(struct dax_device *dax_dev,
-		pgoff_t pgoff, long nr_pages, void **kaddr, pfn_t *pfn)
+		pgoff_t pgoff, long nr_pages, enum dax_access_mode mode,
+		void **kaddr, pfn_t *pfn)
 {
 	struct pmem_device *pmem = dax_get_private(dax_dev);
 
-	return __pmem_direct_access(pmem, pgoff, nr_pages, kaddr, pfn);
+	return __pmem_direct_access(pmem, pgoff, nr_pages, mode, kaddr, pfn);
 }
 
 static const struct dax_operations pmem_dax_ops = {

drivers/nvdimm/pmem.h

···
 #include <linux/pfn_t.h>
 #include <linux/fs.h>
 
+enum dax_access_mode;
+
 /* this definition is in it's own header for tools/testing/nvdimm to consume */
 struct pmem_device {
 	/* One contiguous memory region per device */
···
 };
 
 long __pmem_direct_access(struct pmem_device *pmem, pgoff_t pgoff,
-		long nr_pages, void **kaddr, pfn_t *pfn);
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		pfn_t *pfn);
 
 #ifdef CONFIG_MEMORY_FAILURE
 static inline bool test_and_clear_pmem_poison(struct page *page)

drivers/s390/block/dcssblk.c

···
 static void dcssblk_release(struct gendisk *disk, fmode_t mode);
 static void dcssblk_submit_bio(struct bio *bio);
 static long dcssblk_dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff,
-		long nr_pages, void **kaddr, pfn_t *pfn);
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		pfn_t *pfn);
 
 static char dcssblk_segments[DCSSBLK_PARM_LEN] = "\0";
 
···
 	long rc;
 	void *kaddr;
 
-	rc = dax_direct_access(dax_dev, pgoff, nr_pages, &kaddr, NULL);
+	rc = dax_direct_access(dax_dev, pgoff, nr_pages, DAX_ACCESS,
+			&kaddr, NULL);
 	if (rc < 0)
 		return rc;
 	memset(kaddr, 0, nr_pages << PAGE_SHIFT);
···
 
 static long
 dcssblk_dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff,
-		long nr_pages, void **kaddr, pfn_t *pfn)
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		pfn_t *pfn)
 {
 	struct dcssblk_dev_info *dev_info = dax_get_private(dax_dev);
 

fs/dax.c

···
 	int id;
 
 	id = dax_read_lock();
-	rc = dax_direct_access(iter->iomap.dax_dev, pgoff, 1, &kaddr, NULL);
+	rc = dax_direct_access(iter->iomap.dax_dev, pgoff, 1, DAX_ACCESS,
+				&kaddr, NULL);
 	if (rc < 0) {
 		dax_read_unlock(id);
 		return rc;
···
 
 	id = dax_read_lock();
 	length = dax_direct_access(iomap->dax_dev, pgoff, PHYS_PFN(size),
-				   NULL, pfnp);
+				   DAX_ACCESS, NULL, pfnp);
 	if (length < 0) {
 		rc = length;
 		goto out;
···
 	void *kaddr;
 	long ret;
 
-	ret = dax_direct_access(dax_dev, pgoff, 1, &kaddr, NULL);
+	ret = dax_direct_access(dax_dev, pgoff, 1, DAX_ACCESS, &kaddr, NULL);
 	if (ret > 0) {
 		memset(kaddr + offset, 0, size);
 		dax_flush(dax_dev, kaddr + offset, size);
···
 		}
 
 		map_len = dax_direct_access(dax_dev, pgoff, PHYS_PFN(size),
-				&kaddr, NULL);
+				DAX_ACCESS, &kaddr, NULL);
 		if (map_len < 0) {
 			ret = map_len;
 			break;

fs/fuse/dax.c

···
 	INIT_DELAYED_WORK(&fcd->free_work, fuse_dax_free_mem_worker);
 
 	id = dax_read_lock();
-	nr_pages = dax_direct_access(fcd->dev, 0, PHYS_PFN(dax_size), NULL,
-				     NULL);
+	nr_pages = dax_direct_access(fcd->dev, 0, PHYS_PFN(dax_size),
+			DAX_ACCESS, NULL, NULL);
 	dax_read_unlock(id);
 	if (nr_pages < 0) {
 		pr_debug("dax_direct_access() returned %ld\n", nr_pages);

fs/fuse/virtio_fs.c

···
  * offset.
  */
 static long virtio_fs_direct_access(struct dax_device *dax_dev, pgoff_t pgoff,
-				    long nr_pages, void **kaddr, pfn_t *pfn)
+				    long nr_pages, enum dax_access_mode mode,
+				    void **kaddr, pfn_t *pfn)
 {
 	struct virtio_fs *fs = dax_get_private(dax_dev);
 	phys_addr_t offset = PFN_PHYS(pgoff);
···
 	long rc;
 	void *kaddr;
 
-	rc = dax_direct_access(dax_dev, pgoff, nr_pages, &kaddr, NULL);
+	rc = dax_direct_access(dax_dev, pgoff, nr_pages, DAX_ACCESS, &kaddr,
+			       NULL);
 	if (rc < 0)
 		return rc;
 	memset(kaddr, 0, nr_pages << PAGE_SHIFT);

include/linux/dax.h

···
 struct iomap_iter;
 struct iomap;
 
+enum dax_access_mode {
+	DAX_ACCESS,
+	DAX_RECOVERY_WRITE,
+};
+
 struct dax_operations {
 	/*
 	 * direct_access: translate a device-relative
···
 	 * number of pages available for DAX at that pfn.
 	 */
 	long (*direct_access)(struct dax_device *, pgoff_t, long,
-			void **, pfn_t *);
+			enum dax_access_mode, void **, pfn_t *);
 	/*
 	 * Validate whether this device is usable as an fsdax backing
 	 * device.
···
 bool dax_alive(struct dax_device *dax_dev);
 void *dax_get_private(struct dax_device *dax_dev);
 long dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff, long nr_pages,
-		void **kaddr, pfn_t *pfn);
+		enum dax_access_mode mode, void **kaddr, pfn_t *pfn);
 size_t dax_copy_from_iter(struct dax_device *dax_dev, pgoff_t pgoff, void *addr,
 		size_t bytes, struct iov_iter *i);
 size_t dax_copy_to_iter(struct dax_device *dax_dev, pgoff_t pgoff, void *addr,

include/linux/device-mapper.h

···
 struct dm_report_zones_args;
 struct mapped_device;
 struct bio_vec;
+enum dax_access_mode;
 
 /*
  * Type of table, mapped_device's mempool and request_queue
···
  * >= 0 : the number of bytes accessible at the address
  */
 typedef long (*dm_dax_direct_access_fn) (struct dm_target *ti, pgoff_t pgoff,
-		long nr_pages, void **kaddr, pfn_t *pfn);
+		long nr_pages, enum dax_access_mode node, void **kaddr,
+		pfn_t *pfn);
 typedef int (*dm_dax_zero_page_range_fn)(struct dm_target *ti, pgoff_t pgoff,
 		size_t nr_pages);
 

tools/testing/nvdimm/pmem-dax.c

···
  */
 #include "test/nfit_test.h"
 #include <linux/blkdev.h>
+#include <linux/dax.h>
 #include <pmem.h>
 #include <nd.h>
 
 long __pmem_direct_access(struct pmem_device *pmem, pgoff_t pgoff,
-		long nr_pages, void **kaddr, pfn_t *pfn)
+		long nr_pages, enum dax_access_mode mode, void **kaddr,
+		pfn_t *pfn)
 {
 	resource_size_t offset = PFN_PHYS(pgoff) + pmem->data_offset;