1 files changed, 99 insertions, 129 deletions

diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 1346d698463a..7869ad12bc6e 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -149,7 +149,7 @@ static int check_compressed_csum(struct btrfs_inode *inode, struct bio *bio,
 	const u32 csum_size = fs_info->csum_size;
 	const u32 sectorsize = fs_info->sectorsize;
 	struct page *page;
-	unsigned long i;
+	unsigned int i;
 	char *kaddr;
 	u8 csum[BTRFS_CSUM_SIZE];
 	struct compressed_bio *cb = bio->bi_private;
@@ -172,10 +172,9 @@ static int check_compressed_csum(struct btrfs_inode *inode, struct bio *bio,
 		/* Hash through the page sector by sector */
 		for (pg_offset = 0; pg_offset < bytes_left;
 		     pg_offset += sectorsize) {
-			kaddr = kmap_atomic(page);
+			kaddr = page_address(page);
 			crypto_shash_digest(shash, kaddr + pg_offset,
 					    sectorsize, csum);
-			kunmap_atomic(kaddr);
 
 			if (memcmp(&csum, cb_sum, csum_size) != 0) {
 				btrfs_print_data_csum_error(inode, disk_start,
@@ -208,7 +207,7 @@ static void end_compressed_bio_read(struct bio *bio)
 	struct compressed_bio *cb = bio->bi_private;
 	struct inode *inode;
 	struct page *page;
-	unsigned long index;
+	unsigned int index;
 	unsigned int mirror = btrfs_io_bio(bio)->mirror_num;
 	int ret = 0;
 
@@ -334,7 +333,7 @@ static void end_compressed_bio_write(struct bio *bio)
 	struct compressed_bio *cb = bio->bi_private;
 	struct inode *inode;
 	struct page *page;
-	unsigned long index;
+	unsigned int index;
 
 	if (bio->bi_status)
 		cb->errors = 1;
@@ -349,12 +348,10 @@ static void end_compressed_bio_write(struct bio *bio)
 	 * call back into the FS and do all the end_io operations
 	 */
 	inode = cb->inode;
-	cb->compressed_pages[0]->mapping = cb->inode->i_mapping;
 	btrfs_record_physical_zoned(inode, cb->start, bio);
-	btrfs_writepage_endio_finish_ordered(cb->compressed_pages[0],
+	btrfs_writepage_endio_finish_ordered(BTRFS_I(inode), NULL,
 			cb->start, cb->start + cb->len - 1,
-			bio->bi_status == BLK_STS_OK);
-	cb->compressed_pages[0]->mapping = NULL;
+			!cb->errors);
 
 	end_compressed_writeback(inode, cb);
 	/* note, our inode could be gone now */
@@ -387,10 +384,10 @@ out:
  * the end io hooks.
  */
 blk_status_t btrfs_submit_compressed_write(struct btrfs_inode *inode, u64 start,
-				 unsigned long len, u64 disk_start,
-				 unsigned long compressed_len,
+				 unsigned int len, u64 disk_start,
+				 unsigned int compressed_len,
 				 struct page **compressed_pages,
-				 unsigned long nr_pages,
+				 unsigned int nr_pages,
 				 unsigned int write_flags,
 				 struct cgroup_subsys_state *blkcg_css)
 {
@@ -427,24 +424,16 @@ blk_status_t btrfs_submit_compressed_write(struct btrfs_inode *inode, u64 start,
 	bio->bi_end_io = end_compressed_bio_write;
 
 	if (use_append) {
-		struct extent_map *em;
-		struct map_lookup *map;
-		struct block_device *bdev;
+		struct btrfs_device *device;
 
-		em = btrfs_get_chunk_map(fs_info, disk_start, PAGE_SIZE);
-		if (IS_ERR(em)) {
+		device = btrfs_zoned_get_device(fs_info, disk_start, PAGE_SIZE);
+		if (IS_ERR(device)) {
 			kfree(cb);
 			bio_put(bio);
 			return BLK_STS_NOTSUPP;
 		}
 
-		map = em->map_lookup;
-		/* We only support single profile for now */
-		ASSERT(map->num_stripes == 1);
-		bdev = map->stripes[0].dev->bdev;
-
-		bio_set_dev(bio, bdev);
-		free_extent_map(em);
+		bio_set_dev(bio, device->bdev);
 	}
 
 	if (blkcg_css) {
@@ -515,7 +504,7 @@ blk_status_t btrfs_submit_compressed_write(struct btrfs_inode *inode, u64 start,
 		}
 		if (bytes_left < PAGE_SIZE) {
 			btrfs_info(fs_info,
-					"bytes left %lu compress len %lu nr %lu",
+					"bytes left %lu compress len %u nr %u",
 			       bytes_left, cb->compressed_len, cb->nr_pages);
 		}
 		bytes_left -= PAGE_SIZE;
@@ -575,6 +564,16 @@ static noinline int add_ra_bio_pages(struct inode *inode,
 	if (isize == 0)
 		return 0;
 
+	/*
+	 * For current subpage support, we only support 64K page size,
+	 * which means maximum compressed extent size (128K) is just 2x page
+	 * size.
+	 * This makes readahead less effective, so here disable readahead for
+	 * subpage for now, until full compressed write is supported.
+	 */
+	if (btrfs_sb(inode->i_sb)->sectorsize < PAGE_SIZE)
+		return 0;
+
 	end_index = (i_size_read(inode) - 1) >> PAGE_SHIFT;
 
 	while (last_offset < compressed_end) {
@@ -677,12 +676,13 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct extent_map_tree *em_tree;
 	struct compressed_bio *cb;
-	unsigned long compressed_len;
-	unsigned long nr_pages;
-	unsigned long pg_index;
+	unsigned int compressed_len;
+	unsigned int nr_pages;
+	unsigned int pg_index;
 	struct page *page;
 	struct bio *comp_bio;
 	u64 cur_disk_byte = bio->bi_iter.bi_sector << 9;
+	u64 file_offset;
 	u64 em_len;
 	u64 em_start;
 	struct extent_map *em;
@@ -692,15 +692,17 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 
 	em_tree = &BTRFS_I(inode)->extent_tree;
 
+	file_offset = bio_first_bvec_all(bio)->bv_offset +
+		      page_offset(bio_first_page_all(bio));
+
 	/* we need the actual starting offset of this extent in the file */
 	read_lock(&em_tree->lock);
-	em = lookup_extent_mapping(em_tree,
-				   page_offset(bio_first_page_all(bio)),
-				   fs_info->sectorsize);
+	em = lookup_extent_mapping(em_tree, file_offset, fs_info->sectorsize);
 	read_unlock(&em_tree->lock);
 	if (!em)
 		return BLK_STS_IOERR;
 
+	ASSERT(em->compress_type != BTRFS_COMPRESS_NONE);
 	compressed_len = em->block_len;
 	cb = kmalloc(compressed_bio_size(fs_info, compressed_len), GFP_NOFS);
 	if (!cb)
@@ -731,8 +733,7 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 		goto fail1;
 
 	for (pg_index = 0; pg_index < nr_pages; pg_index++) {
-		cb->compressed_pages[pg_index] = alloc_page(GFP_NOFS |
-							      __GFP_HIGHMEM);
+		cb->compressed_pages[pg_index] = alloc_page(GFP_NOFS);
 		if (!cb->compressed_pages[pg_index]) {
 			faili = pg_index - 1;
 			ret = BLK_STS_RESOURCE;
@@ -1202,9 +1203,6 @@ static unsigned int btrfs_compress_set_level(int type, unsigned level)
  *
  * @total_out is an in/out parameter, must be set to the input length and will
  * be also used to return the total number of compressed bytes
- *
- * @max_out tells us the max number of bytes that we're allowed to
- * stuff into pages
  */
 int btrfs_compress_pages(unsigned int type_level, struct address_space *mapping,
 			 u64 start, struct page **pages,
@@ -1225,20 +1223,6 @@ int btrfs_compress_pages(unsigned int type_level, struct address_space *mapping,
 	return ret;
 }
 
-/*
- * pages_in is an array of pages with compressed data.
- *
- * disk_start is the starting logical offset of this array in the file
- *
- * orig_bio contains the pages from the file that we want to decompress into
- *
- * srclen is the number of bytes in pages_in
- *
- * The basic idea is that we have a bio that was created by readpages.
- * The pages in the bio are for the uncompressed data, and they may not
- * be contiguous.  They all correspond to the range of bytes covered by
- * the compressed extent.
- */
 static int btrfs_decompress_bio(struct compressed_bio *cb)
 {
 	struct list_head *workspace;
@@ -1288,96 +1272,82 @@ void __cold btrfs_exit_compress(void)
 }
 
 /*
- * Copy uncompressed data from working buffer to pages.
+ * Copy decompressed data from working buffer to pages.
+ *
+ * @buf:		The decompressed data buffer
+ * @buf_len:		The decompressed data length
+ * @decompressed:	Number of bytes that are already decompressed inside the
+ * 			compressed extent
+ * @cb:			The compressed extent descriptor
+ * @orig_bio:		The original bio that the caller wants to read for
+ *
+ * An easier to understand graph is like below:
  *
- * buf_start is the byte offset we're of the start of our workspace buffer.
+ * 		|<- orig_bio ->|     |<- orig_bio->|
+ * 	|<-------      full decompressed extent      ----->|
+ * 	|<-----------    @cb range   ---->|
+ * 	|			|<-- @buf_len -->|
+ * 	|<--- @decompressed --->|
  *
- * total_out is the last byte of the buffer
+ * Note that, @cb can be a subpage of the full decompressed extent, but
+ * @cb->start always has the same as the orig_file_offset value of the full
+ * decompressed extent.
+ *
+ * When reading compressed extent, we have to read the full compressed extent,
+ * while @orig_bio may only want part of the range.
+ * Thus this function will ensure only data covered by @orig_bio will be copied
+ * to.
+ *
+ * Return 0 if we have copied all needed contents for @orig_bio.
+ * Return >0 if we need continue decompress.
  */
-int btrfs_decompress_buf2page(const char *buf, unsigned long buf_start,
-			      unsigned long total_out, u64 disk_start,
-			      struct bio *bio)
+int btrfs_decompress_buf2page(const char *buf, u32 buf_len,
+			      struct compressed_bio *cb, u32 decompressed)
 {
-	unsigned long buf_offset;
-	unsigned long current_buf_start;
-	unsigned long start_byte;
-	unsigned long prev_start_byte;
-	unsigned long working_bytes = total_out - buf_start;
-	unsigned long bytes;
-	struct bio_vec bvec = bio_iter_iovec(bio, bio->bi_iter);
-
-	/*
-	 * start byte is the first byte of the page we're currently
-	 * copying into relative to the start of the compressed data.
-	 */
-	start_byte = page_offset(bvec.bv_page) - disk_start;
-
-	/* we haven't yet hit data corresponding to this page */
-	if (total_out <= start_byte)
-		return 1;
-
-	/*
-	 * the start of the data we care about is offset into
-	 * the middle of our working buffer
-	 */
-	if (total_out > start_byte && buf_start < start_byte) {
-		buf_offset = start_byte - buf_start;
-		working_bytes -= buf_offset;
-	} else {
-		buf_offset = 0;
-	}
-	current_buf_start = buf_start;
+	struct bio *orig_bio = cb->orig_bio;
+	/* Offset inside the full decompressed extent */
+	u32 cur_offset;
+
+	cur_offset = decompressed;
+	/* The main loop to do the copy */
+	while (cur_offset < decompressed + buf_len) {
+		struct bio_vec bvec;
+		size_t copy_len;
+		u32 copy_start;
+		/* Offset inside the full decompressed extent */
+		u32 bvec_offset;
+
+		bvec = bio_iter_iovec(orig_bio, orig_bio->bi_iter);
+		/*
+		 * cb->start may underflow, but subtracting that value can still
+		 * give us correct offset inside the full decompressed extent.
+		 */
+		bvec_offset = page_offset(bvec.bv_page) + bvec.bv_offset - cb->start;
 
-	/* copy bytes from the working buffer into the pages */
-	while (working_bytes > 0) {
-		bytes = min_t(unsigned long, bvec.bv_len,
-				PAGE_SIZE - (buf_offset % PAGE_SIZE));
-		bytes = min(bytes, working_bytes);
+		/* Haven't reached the bvec range, exit */
+		if (decompressed + buf_len <= bvec_offset)
+			return 1;
 
-		memcpy_to_page(bvec.bv_page, bvec.bv_offset, buf + buf_offset,
-			       bytes);
-		flush_dcache_page(bvec.bv_page);
-
-		buf_offset += bytes;
-		working_bytes -= bytes;
-		current_buf_start += bytes;
-
-		/* check if we need to pick another page */
-		bio_advance(bio, bytes);
-		if (!bio->bi_iter.bi_size)
-			return 0;
-		bvec = bio_iter_iovec(bio, bio->bi_iter);
-		prev_start_byte = start_byte;
-		start_byte = page_offset(bvec.bv_page) - disk_start;
+		copy_start = max(cur_offset, bvec_offset);
+		copy_len = min(bvec_offset + bvec.bv_len,
+			       decompressed + buf_len) - copy_start;
+		ASSERT(copy_len);
 
 		/*
-		 * We need to make sure we're only adjusting
-		 * our offset into compression working buffer when
-		 * we're switching pages.  Otherwise we can incorrectly
-		 * keep copying when we were actually done.
+		 * Extra range check to ensure we didn't go beyond
+		 * @buf + @buf_len.
 		 */
-		if (start_byte != prev_start_byte) {
-			/*
-			 * make sure our new page is covered by this
-			 * working buffer
-			 */
-			if (total_out <= start_byte)
-				return 1;
+		ASSERT(copy_start - decompressed < buf_len);
+		memcpy_to_page(bvec.bv_page, bvec.bv_offset,
+			       buf + copy_start - decompressed, copy_len);
+		flush_dcache_page(bvec.bv_page);
+		cur_offset += copy_len;
 
-			/*
-			 * the next page in the biovec might not be adjacent
-			 * to the last page, but it might still be found
-			 * inside this working buffer. bump our offset pointer
-			 */
-			if (total_out > start_byte &&
-			    current_buf_start < start_byte) {
-				buf_offset = start_byte - buf_start;
-				working_bytes = total_out - start_byte;
-				current_buf_start = buf_start + buf_offset;
-			}
-		}
+		bio_advance(orig_bio, copy_len);
+		/* Finished the bio */
+		if (!orig_bio->bi_iter.bi_size)
+			return 0;
 	}
-
 	return 1;
 }