Merge tag 'erofs-for-5.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs

Pull erofs updates from Gao Xiang:
 "There are some new features available for this cycle. Firstly, EROFS
  LZMA algorithm support, specifically called MicroLZMA, is available as
  an option for embedded devices, LiveCDs and/or as the secondary
  auxiliary compression algorithm besides the primary algorithm in one
  file.

  In order to better support the LZMA fixed-sized output compression,
  especially for 4KiB pcluster size (which has lowest memory pressure
  thus useful for memory-sensitive scenarios), Lasse introduced a new
  LZMA header/container format called MicroLZMA to minimize the original
  LZMA1 header (for example, we don't need to waste 4-byte dictionary
  size and another 8-byte uncompressed size, which can be calculated by
  fs directly, for each pcluster) and enable EROFS fixed-sized output
  compression.

  Note that MicroLZMA can also be later used by other things in addition
  to EROFS too where wasting minimal amount of space for headers is
  important and it can be only compiled by enabling XZ_DEC_MICROLZMA.
  MicroLZMA has been supported by the latest upstream XZ embedded [1] &
  XZ utils [2], apply the latest related XZ embedded upstream patches by
  the XZ author Lasse here.

  Secondly, multiple device is also supported in this cycle, which is
  designed for multi-layer container images. By working together with
  inter-layer data deduplication and compression, we can achieve the
  next high-performance container image solution. Our team will announce
  the new Nydus container image service [3] implementation with new RAFS
  v6 (EROFS-compatible) format in Open Source Summit 2021 China [4]
  soon.

  Besides, the secondary compression head support and readmore
  decompression strategy are also included in this cycle. There are also
  some minor bugfixes and cleanups, as always.

  Summary:

   - support multiple devices for multi-layer container images;

   - support the secondary compression head;

   - support readmore decompression strategy;

   - support new LZMA algorithm (specifically called MicroLZMA);

   - some bugfixes & cleanups"

* tag 'erofs-for-5.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs:
  erofs: don't trigger WARN() when decompression fails
  erofs: get rid of ->lru usage
  erofs: lzma compression support
  erofs: rename some generic methods in decompressor
  lib/xz, lib/decompress_unxz.c: Fix spelling in comments
  lib/xz: Add MicroLZMA decoder
  lib/xz: Move s->lzma.len = 0 initialization to lzma_reset()
  lib/xz: Validate the value before assigning it to an enum variable
  lib/xz: Avoid overlapping memcpy() with invalid input with in-place decompression
  erofs: introduce readmore decompression strategy
  erofs: introduce the secondary compression head
  erofs: get compression algorithms directly on mapping
  erofs: add multiple device support
  erofs: decouple basic mount options from fs_context
  erofs: remove the fast path of per-CPU buffer decompression

1 files changed, 94 insertions, 11 deletions

diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h
index 9524e155b38f..3265688af7f9 100644
--- a/fs/erofs/internal.h
+++ b/fs/erofs/internal.h
@@ -47,7 +47,16 @@ typedef u64 erofs_off_t;
 /* data type for filesystem-wide blocks number */
 typedef u32 erofs_blk_t;
 
-struct erofs_fs_context {
+struct erofs_device_info {
+	char *path;
+	struct block_device *bdev;
+	struct dax_device *dax_dev;
+
+	u32 blocks;
+	u32 mapped_blkaddr;
+};
+
+struct erofs_mount_opts {
 #ifdef CONFIG_EROFS_FS_ZIP
 	/* current strategy of how to use managed cache */
 	unsigned char cache_strategy;
@@ -60,6 +69,18 @@ struct erofs_fs_context {
 	unsigned int mount_opt;
 };
 
+struct erofs_dev_context {
+	struct idr tree;
+	struct rw_semaphore rwsem;
+
+	unsigned int extra_devices;
+};
+
+struct erofs_fs_context {
+	struct erofs_mount_opts opt;
+	struct erofs_dev_context *devs;
+};
+
 /* all filesystem-wide lz4 configurations */
 struct erofs_sb_lz4_info {
 	/* # of pages needed for EROFS lz4 rolling decompression */
@@ -69,6 +90,7 @@ struct erofs_sb_lz4_info {
 };
 
 struct erofs_sb_info {
+	struct erofs_mount_opts opt;	/* options */
 #ifdef CONFIG_EROFS_FS_ZIP
 	/* list for all registered superblocks, mainly for shrinker */
 	struct list_head list;
@@ -85,12 +107,16 @@ struct erofs_sb_info {
 
 	struct erofs_sb_lz4_info lz4;
 #endif	/* CONFIG_EROFS_FS_ZIP */
+	struct erofs_dev_context *devs;
 	struct dax_device *dax_dev;
-	u32 blocks;
+	u64 total_blocks;
+	u32 primarydevice_blocks;
+
 	u32 meta_blkaddr;
 #ifdef CONFIG_EROFS_FS_XATTR
 	u32 xattr_blkaddr;
 #endif
+	u16 device_id_mask;	/* valid bits of device id to be used */
 
 	/* inode slot unit size in bit shift */
 	unsigned char islotbits;
@@ -108,8 +134,6 @@ struct erofs_sb_info {
 	u8 volume_name[16];             /* volume name */
 	u32 feature_compat;
 	u32 feature_incompat;
-
-	struct erofs_fs_context ctx;	/* options */
 };
 
 #define EROFS_SB(sb) ((struct erofs_sb_info *)(sb)->s_fs_info)
@@ -121,9 +145,9 @@ struct erofs_sb_info {
 #define EROFS_MOUNT_DAX_ALWAYS		0x00000040
 #define EROFS_MOUNT_DAX_NEVER		0x00000080
 
-#define clear_opt(ctx, option)	((ctx)->mount_opt &= ~EROFS_MOUNT_##option)
-#define set_opt(ctx, option)	((ctx)->mount_opt |= EROFS_MOUNT_##option)
-#define test_opt(ctx, option)	((ctx)->mount_opt & EROFS_MOUNT_##option)
+#define clear_opt(opt, option)	((opt)->mount_opt &= ~EROFS_MOUNT_##option)
+#define set_opt(opt, option)	((opt)->mount_opt |= EROFS_MOUNT_##option)
+#define test_opt(opt, option)	((opt)->mount_opt & EROFS_MOUNT_##option)
 
 enum {
 	EROFS_ZIP_CACHE_DISABLED,
@@ -237,6 +261,7 @@ static inline bool erofs_sb_has_##name(struct erofs_sb_info *sbi) \
 EROFS_FEATURE_FUNCS(lz4_0padding, incompat, INCOMPAT_LZ4_0PADDING)
 EROFS_FEATURE_FUNCS(compr_cfgs, incompat, INCOMPAT_COMPR_CFGS)
 EROFS_FEATURE_FUNCS(big_pcluster, incompat, INCOMPAT_BIG_PCLUSTER)
+EROFS_FEATURE_FUNCS(device_table, incompat, INCOMPAT_DEVICE_TABLE)
 EROFS_FEATURE_FUNCS(sb_chksum, compat, COMPAT_SB_CHKSUM)
 
 /* atomic flag definitions */
@@ -307,6 +332,19 @@ static inline unsigned int erofs_inode_datalayout(unsigned int value)
 			      EROFS_I_DATALAYOUT_BITS);
 }
 
+/*
+ * Different from grab_cache_page_nowait(), reclaiming is never triggered
+ * when allocating new pages.
+ */
+static inline
+struct page *erofs_grab_cache_page_nowait(struct address_space *mapping,
+					  pgoff_t index)
+{
+	return pagecache_get_page(mapping, index,
+			FGP_LOCK|FGP_CREAT|FGP_NOFS|FGP_NOWAIT,
+			readahead_gfp_mask(mapping) & ~__GFP_RECLAIM);
+}
+
 extern const struct super_operations erofs_sops;
 
 extern const struct address_space_operations erofs_raw_access_aops;
@@ -338,7 +376,7 @@ extern const struct address_space_operations z_erofs_aops;
  * of the corresponding uncompressed data in the file.
  */
 enum {
-	BH_Zipped = BH_PrivateStart,
+	BH_Encoded = BH_PrivateStart,
 	BH_FullMapped,
 };
 
@@ -346,8 +384,8 @@ enum {
 #define EROFS_MAP_MAPPED	(1 << BH_Mapped)
 /* Located in metadata (could be copied from bd_inode) */
 #define EROFS_MAP_META		(1 << BH_Meta)
-/* The extent has been compressed */
-#define EROFS_MAP_ZIPPED	(1 << BH_Zipped)
+/* The extent is encoded */
+#define EROFS_MAP_ENCODED	(1 << BH_Encoded)
 /* The length of extent is full */
 #define EROFS_MAP_FULL_MAPPED	(1 << BH_FullMapped)
 
@@ -355,6 +393,8 @@ struct erofs_map_blocks {
 	erofs_off_t m_pa, m_la;
 	u64 m_plen, m_llen;
 
+	unsigned short m_deviceid;
+	char m_algorithmformat;
 	unsigned int m_flags;
 
 	struct page *mpage;
@@ -367,6 +407,13 @@ struct erofs_map_blocks {
  * approach instead if possible since it's more metadata lightweight.)
  */
 #define EROFS_GET_BLOCKS_FIEMAP	0x0002
+/* Used to map the whole extent if non-negligible data is requested for LZMA */
+#define EROFS_GET_BLOCKS_READMORE	0x0004
+
+enum {
+	Z_EROFS_COMPRESSION_SHIFTED = Z_EROFS_COMPRESSION_MAX,
+	Z_EROFS_COMPRESSION_RUNTIME_MAX
+};
 
 /* zmap.c */
 extern const struct iomap_ops z_erofs_iomap_report_ops;
@@ -386,9 +433,18 @@ static inline int z_erofs_map_blocks_iter(struct inode *inode,
 }
 #endif	/* !CONFIG_EROFS_FS_ZIP */
 
+struct erofs_map_dev {
+	struct block_device *m_bdev;
+	struct dax_device *m_daxdev;
+
+	erofs_off_t m_pa;
+	unsigned int m_deviceid;
+};
+
 /* data.c */
 extern const struct file_operations erofs_file_fops;
 struct page *erofs_get_meta_page(struct super_block *sb, erofs_blk_t blkaddr);
+int erofs_map_dev(struct super_block *sb, struct erofs_map_dev *dev);
 int erofs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		 u64 start, u64 len);
 
@@ -443,7 +499,14 @@ void erofs_pcpubuf_init(void);
 void erofs_pcpubuf_exit(void);
 
 /* utils.c / zdata.c */
-struct page *erofs_allocpage(struct list_head *pool, gfp_t gfp);
+struct page *erofs_allocpage(struct page **pagepool, gfp_t gfp);
+static inline void erofs_pagepool_add(struct page **pagepool,
+		struct page *page)
+{
+	set_page_private(page, (unsigned long)*pagepool);
+	*pagepool = page;
+}
+void erofs_release_pages(struct page **pagepool);
 
 #ifdef CONFIG_EROFS_FS_ZIP
 int erofs_workgroup_put(struct erofs_workgroup *grp);
@@ -483,6 +546,26 @@ static inline int z_erofs_load_lz4_config(struct super_block *sb,
 }
 #endif	/* !CONFIG_EROFS_FS_ZIP */
 
+#ifdef CONFIG_EROFS_FS_ZIP_LZMA
+int z_erofs_lzma_init(void);
+void z_erofs_lzma_exit(void);
+int z_erofs_load_lzma_config(struct super_block *sb,
+			     struct erofs_super_block *dsb,
+			     struct z_erofs_lzma_cfgs *lzma, int size);
+#else
+static inline int z_erofs_lzma_init(void) { return 0; }
+static inline int z_erofs_lzma_exit(void) { return 0; }
+static inline int z_erofs_load_lzma_config(struct super_block *sb,
+			     struct erofs_super_block *dsb,
+			     struct z_erofs_lzma_cfgs *lzma, int size) {
+	if (lzma) {
+		erofs_err(sb, "lzma algorithm isn't enabled");
+		return -EINVAL;
+	}
+	return 0;
+}
+#endif	/* !CONFIG_EROFS_FS_ZIP */
+
 #define EFSCORRUPTED    EUCLEAN         /* Filesystem is corrupted */
 
 #endif	/* __EROFS_INTERNAL_H */