Re: FAILED: patch "[PATCH] dm writecache: fix writing beyond end of underlying device" failed to apply to 4.19-stable tree

[Mikulas Patocka <mpatocka@xxxxxxxxxx>]

On Mon, 1 Mar 2021, gregkh@xxxxxxxxxxxxxxxxxxx wrote:

> 
> The patch below does not apply to the 4.19-stable tree.
> If someone wants it applied there, or to any other stable or longterm
> tree, then please email the backport, including the original git commit
> id to <stable@xxxxxxxxxxxxxxx>.
> 
> thanks,
> 
> greg k-h

Hi

This is backport of the patch 4134455f2aafdfeab50cabb4cccb35e916034b93 for
the stable branch 4.19.

Mikulas



commit 4134455f2aafdfeab50cabb4cccb35e916034b93
Author: Mikulas Patocka <mpatocka@xxxxxxxxxx>
Date:   Tue Feb 9 10:56:20 2021 -0500

    dm writecache: fix writing beyond end of underlying device when shrinking
    
    Do not attempt to write any data beyond the end of the underlying data
    device while shrinking it.
    
    The DM writecache device must be suspended when the underlying data
    device is shrunk.
    
    Signed-off-by: Mikulas Patocka <mpatocka@xxxxxxxxxx>
    Cc: stable@xxxxxxxxxxxxxxx
    Signed-off-by: Mike Snitzer <snitzer@xxxxxxxxxx>

---
 drivers/md/dm-writecache.c |   18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

Index: linux-stable/drivers/md/dm-writecache.c
===================================================================
--- linux-stable.orig/drivers/md/dm-writecache.c	2021-07-18 21:05:45.000000000 +0200
+++ linux-stable/drivers/md/dm-writecache.c	2021-07-18 21:09:52.000000000 +0200
@@ -142,6 +142,7 @@ struct dm_writecache {
 	size_t metadata_sectors;
 	size_t n_blocks;
 	uint64_t seq_count;
+	sector_t data_device_sectors;
 	void *block_start;
 	struct wc_entry *entries;
 	unsigned block_size;
@@ -929,6 +930,8 @@ static void writecache_resume(struct dm_
 
 	wc_lock(wc);
 
+	wc->data_device_sectors = i_size_read(wc->dev->bdev->bd_inode) >> SECTOR_SHIFT;
+
 	if (WC_MODE_PMEM(wc)) {
 		persistent_memory_invalidate_cache(wc->memory_map, wc->memory_map_size);
 	} else {
@@ -1499,6 +1502,10 @@ static bool wc_add_block(struct writebac
 	void *address = memory_data(wc, e);
 
 	persistent_memory_flush_cache(address, block_size);
+
+	if (unlikely(bio_end_sector(&wb->bio) >= wc->data_device_sectors))
+		return true;
+
 	return bio_add_page(&wb->bio, persistent_memory_page(address),
 			    block_size, persistent_memory_page_offset(address)) != 0;
 }
@@ -1571,6 +1578,9 @@ static void __writecache_writeback_pmem(
 		if (writecache_has_error(wc)) {
 			bio->bi_status = BLK_STS_IOERR;
 			bio_endio(&wb->bio);
+		} else if (unlikely(!bio_sectors(&wb->bio))) {
+			bio->bi_status = BLK_STS_OK;
+			bio_endio(&wb->bio);
 		} else {
 			submit_bio(&wb->bio);
 		}
@@ -1614,6 +1624,14 @@ static void __writecache_writeback_ssd(s
 			e = f;
 		}
 
+		if (unlikely(to.sector + to.count > wc->data_device_sectors)) {
+			if (to.sector >= wc->data_device_sectors) {
+				writecache_copy_endio(0, 0, c);
+				continue;
+			}
+			from.count = to.count = wc->data_device_sectors - to.sector;
+		}
+
 		dm_kcopyd_copy(wc->dm_kcopyd, &from, 1, &to, 0, writecache_copy_endio, c);
 
 		__writeback_throttle(wc, wbl);