Hi Gaoxiang,

Thanks for your review and suggestions. It's great to communicate with outstanding community developers. I have learned a lot and look forward to contribute more patches in the future.

Regards,
Jianan


 
From: Gao Xiang
Date: 2020-10-17 00:17
To: Huang Jianan
CC: linux-erofs; guoweichao; zhangshiming; huangjianan
Subject: Re: [PATCH 5/5] erofs-utils: support read compressed file
Hi Jianan,
 
On Thu, Oct 15, 2020 at 09:39:59PM +0800, Huang Jianan wrote:
> Signed-off-by: Huang Jianan <huangjianan@oppo.com>
> Signed-off-by: Guo Weichao <guoweichao@oppo.com>
 
I read the logic below, and generally it looks good! <thumb>
 
I'm now handling some minor thing about this and I will
update as the following patch to speed up the development.
 
> ---
 
...
 
> diff --git a/fuse/decompress.c b/fuse/decompress.c
> new file mode 100644
> index 0000000..e2df3ce
> --- /dev/null
> +++ b/fuse/decompress.c
 
we might need to move this file to lib/decompress.c
 
> @@ -0,0 +1,86 @@
> +/* SPDX-License-Identifier: GPL-2.0+ */
> +/*
> + * Copyright (C), 2008-2020, OPPO Mobile Comm Corp., Ltd.
> + * Created by Huang Jianan <huangjianan@oppo.com>
> + */
> +
> +#include <stdlib.h>
> +#include <lz4.h>
> +
> +#include "erofs/internal.h"
> +#include "erofs/err.h"
> +#include "decompress.h"
> +#include "logging.h"
> +#include "init.h"
> +
> +static int z_erofs_shifted_transform(struct z_erofs_decompress_req *rq)
> +{
> + char *dest = rq->out + rq->ofs_out;
 
I think ofs_out can be omited, since we only care about the dest buffer here.
so (new) rq->out == (old)rq->out + rq->ofs_out.
 
> + char *src = rq->in + rq->ofs_head;
 
need to judge if the outputsize is larger than EROFS_BLKSIZ, since it's
undefined on-disk behavior, and can be implemented in the future
on-disk update.
 
> +
> + memcpy(dest, src, rq->outputsize - rq->ofs_head);
> +
> + return 0;
> +}
 
And this function can be folded into z_erofs_decompress(), because the
implementation here is much simple enough.
 
> +
> +static int z_erofs_decompress_generic(struct z_erofs_decompress_req *rq)
 
it might be renamed into z_erofs_decompress_lz4.
 
> +{
> + int ret = 0;
> + char *dest = rq->out + rq->ofs_out;
> + char *src = rq->in;
> + char *buff = NULL;
> + bool support_0padding = false;
> + unsigned int inputmargin = 0;
> +
> + if (sbk->feature_incompat & EROFS_FEATURE_INCOMPAT_LZ4_0PADDING) {
> + support_0padding = true;
> +
> + while (!src[inputmargin & ~PAGE_MASK])
> + if (!(++inputmargin & ~PAGE_MASK))
> + break;
> +
> + if (inputmargin >= rq->inputsize)
> + return -EIO;
> + }
> +
> + if (rq->ofs_head) {
 
we might need to rethink about the name of ofs_head.
 
> + buff = malloc(rq->outputsize);
> + if (!buff)
> + return -ENOMEM;
> + dest = buff;
 
need some cleanup, and dest variable is assigned for several times,
but without making the logic simpler.
 
> + }
> +
> + if (rq->partial_decoding || !support_0padding)
> + ret = LZ4_decompress_safe_partial(src + inputmargin, dest,
> +   rq->inputsize - inputmargin,
> +   rq->outputsize, rq->outputsize);
> + else
> + ret = LZ4_decompress_safe(src + inputmargin, dest,
> +   rq->inputsize - inputmargin,
> +   rq->outputsize);
> +
> + if (ret != (int)rq->outputsize) {
> + ret = -EIO;
> + goto out;
> + }
> +
> + if (rq->ofs_head) {
> + src = dest + rq->ofs_head;
> + dest = rq->out + rq->ofs_out;
> + memcpy(dest, src, rq->outputsize - rq->ofs_head);
> + }
> +
> +out:
> + if (buff)
> + free(buff);
> +
> + return ret;
> +}
> +
> +int z_erofs_decompress(struct z_erofs_decompress_req *rq)
> +{
> + if (rq->alg == Z_EROFS_COMPRESSION_SHIFTED)
> + return z_erofs_shifted_transform(rq);
> +
> + return z_erofs_decompress_generic(rq);
> +}
> diff --git a/fuse/decompress.h b/fuse/decompress.h
> new file mode 100644
> index 0000000..cd395c3
> --- /dev/null
> +++ b/fuse/decompress.h
 
move this file to include/erofs/decompress.h
 
> @@ -0,0 +1,37 @@
> +/* SPDX-License-Identifier: GPL-2.0+ */
> +/*
> + * Copyright (C), 2008-2020, OPPO Mobile Comm Corp., Ltd.
> + * Created by Huang Jianan <huangjianan@oppo.com>
> + */
> +
> +#ifndef __EROFS_DECOMPRESS_H
> +#define __EROFS_DECOMPRESS_H
> +
> +#include "erofs/internal.h"
> +
> +enum {
> + Z_EROFS_COMPRESSION_SHIFTED = Z_EROFS_COMPRESSION_MAX,
> + Z_EROFS_COMPRESSION_RUNTIME_MAX
> +};
> +
> +struct z_erofs_decompress_req {
> + char *in, *out;
> +
> + size_t ofs_out, ofs_head;
> + unsigned int inputsize, outputsize;
> +
> + /* indicate the algorithm will be used for decompression */
> + unsigned int alg;
> + bool partial_decoding;
> +};
> +
> +#ifdef LZ4_ENABLED
> +int z_erofs_decompress(struct z_erofs_decompress_req *rq);
> +#else
> +int z_erofs_decompress(struct z_erofs_decompress_req *rq)
> +{
> + return -EOPNOTSUPP;
> +}
> +#endif
> +
> +#endif
> diff --git a/fuse/dentry.h b/fuse/dentry.h
> index ee2144d..f89c506 100644
> --- a/fuse/dentry.h
> +++ b/fuse/dentry.h
> @@ -10,10 +10,11 @@
>  #include <stdint.h>
>  #include "erofs/internal.h"
>  
> +/* fixme: Deal with names that exceed the allocated size */
>  #ifdef __64BITS
> -#define DCACHE_ENTRY_NAME_LEN       40
> +#define DCACHE_ENTRY_NAME_LEN       EROFS_NAME_LEN
>  #else
> -#define DCACHE_ENTRY_NAME_LEN       48
> +#define DCACHE_ENTRY_NAME_LEN       EROFS_NAME_LEN
>  #endif
 
not related to this patch, does it relate to another preexist
bug about erofsfuse codebase?
 
>  
>  /* This struct declares a node of a k-tree.  Every node has a pointer to one of
> diff --git a/fuse/init.c b/fuse/init.c
> index 8198fa7..e9cc9f8 100644
> --- a/fuse/init.c
> +++ b/fuse/init.c
> @@ -17,7 +17,23 @@
>  
>  
>  struct erofs_super_block super;
> -static struct erofs_super_block *sbk = &super;
> +struct erofs_super_block *sbk = &super;
> +
> +static bool check_layout_compatibility(struct erofs_super_block *sb,
> +        struct erofs_super_block *dsb)
> +{
> + const unsigned int feature = le32_to_cpu(dsb->feature_incompat);
> +
> + sb->feature_incompat = feature;
> +
> + /* check if current kernel meets all mandatory requirements */
> + if (feature & (~EROFS_ALL_FEATURE_INCOMPAT)) {
> + loge("unidentified incompatible feature %x, please upgrade kernel version",
> +      feature & ~EROFS_ALL_FEATURE_INCOMPAT);
> + return false;
> + }
> + return true;
> +}
>  
>  int erofs_init_super(void)
>  {
> @@ -40,6 +56,9 @@ int erofs_init_super(void)
>  return -EINVAL;
>  }
>  
> + if (!check_layout_compatibility(sbk, sb))
> + return -EINVAL;
> +
>  sbk->checksum = le32_to_cpu(sb->checksum);
>  sbk->feature_compat = le32_to_cpu(sb->feature_compat);
>  sbk->blkszbits = sb->blkszbits;
> @@ -56,6 +75,7 @@ int erofs_init_super(void)
>  sbk->root_nid = le16_to_cpu(sb->root_nid);
>  
>  logp("%-15s:0x%X", STR(magic), SUPER_MEM(magic));
> + logp("%-15s:0x%X", STR(feature_incompat), SUPER_MEM(feature_incompat));
>  logp("%-15s:0x%X", STR(feature_compat), SUPER_MEM(feature_compat));
>  logp("%-15s:%u",   STR(blkszbits), SUPER_MEM(blkszbits));
>  logp("%-15s:%u",   STR(root_nid), SUPER_MEM(root_nid));
> diff --git a/fuse/init.h b/fuse/init.h
> index d7a97b5..3fc4eb5 100644
> --- a/fuse/init.h
> +++ b/fuse/init.h
> @@ -13,6 +13,8 @@
>  
>  #define BOOT_SECTOR_SIZE 0x400
>  
> +extern struct erofs_super_block *sbk;
> +
>  int erofs_init_super(void);
>  erofs_nid_t erofs_get_root_nid(void);
>  erofs_off_t nid2addr(erofs_nid_t nid);
> diff --git a/fuse/namei.c b/fuse/namei.c
> index 7ed1168..510fcfd 100644
> --- a/fuse/namei.c
> +++ b/fuse/namei.c
> @@ -49,7 +49,7 @@ static inline dev_t new_decode_dev(u32 dev)
>  
>  int erofs_iget_by_nid(erofs_nid_t nid, struct erofs_vnode *vi)
>  {
> - int ret;
> + int ret, ifmt;
>  char buf[EROFS_BLKSIZ];
>  struct erofs_inode_compact *v1;
>  const erofs_off_t addr = nid2addr(nid);
> @@ -60,6 +60,11 @@ int erofs_iget_by_nid(erofs_nid_t nid, struct erofs_vnode *vi)
>  return -EIO;
>  
>  v1 = (struct erofs_inode_compact *)buf;
> + /* fixme: support extended inode */
> + ifmt = le16_to_cpu(v1->i_format);
> + if (__inode_version(ifmt) != EROFS_INODE_LAYOUT_COMPACT)
> + return -EOPNOTSUPP;
> +
>  vi->datalayout = __inode_data_mapping(le16_to_cpu(v1->i_format));
>  vi->inode_isize = sizeof(struct erofs_inode_compact);
>  vi->xattr_isize = erofs_xattr_ibody_size(v1->i_xattr_icount);
> @@ -88,6 +93,10 @@ int erofs_iget_by_nid(erofs_nid_t nid, struct erofs_vnode *vi)
>  return -EIO;
>  }
>  
> + vi->z_inited = false;
> + if (erofs_inode_is_data_compressed(vi->datalayout))
> + z_erofs_fill_inode(vi);
> +
>  return 0;
>  }
>  
> diff --git a/fuse/read.c b/fuse/read.c
> index 3ce5c4f..cc0781f 100644
> --- a/fuse/read.c
> +++ b/fuse/read.c
> @@ -16,6 +16,7 @@
>  #include "namei.h"
>  #include "disk_io.h"
>  #include "init.h"
> +#include "decompress.h"
>  
>  size_t erofs_read_data(struct erofs_vnode *vnode, char *buffer,
>         size_t size, off_t offset)
> @@ -78,6 +79,73 @@ finished:
>  
>  }
>  
> +size_t erofs_read_data_compression(struct erofs_vnode *vnode, char *buffer,
> +        size_t size, off_t offset)
> +{
> + int ret;
> + size_t end, count, ofs, sum = size;
> + struct erofs_map_blocks map = {
> + .index = UINT_MAX,
> + };
> + bool partial;
> + unsigned int algorithmformat;
> + char raw[EROFS_BLKSIZ];
> +
> + while (sum) {
> + end = offset + sum;
> + map.m_la = end - 1;
> +
> + ret = z_erofs_map_blocks_iter(vnode, &map);
> + if (ret)
> + return ret;
> +
> + if (!(map.m_flags & EROFS_MAP_MAPPED)) {
> + sum -= map.m_llen;
> + continue;
> + }
> +
> + ret = dev_read(raw, EROFS_BLKSIZ, map.m_pa);
> + if (ret < 0 || (size_t)ret != EROFS_BLKSIZ)
> + return -EIO;
> +
> + algorithmformat = map.m_flags & EROFS_MAP_ZIPPED ?
> + Z_EROFS_COMPRESSION_LZ4 :
> + Z_EROFS_COMPRESSION_SHIFTED;
> +
> + if (end >= map.m_la + map.m_llen) {
> + count = map.m_llen;
> + partial = !(map.m_flags & EROFS_MAP_FULL_MAPPED);
> + } else {
> + count = end - map.m_la;
> + partial = true;
> + }
 
it would be better to make the logic more explicitly, like:
 
-               if (end >= map.m_la + map.m_llen) {
-                       count = map.m_llen;
-                       partial = !(map.m_flags & EROFS_MAP_FULL_MAPPED);
-               } else {
+               /*
+                * trim to the needed size if the returned extent is quite
+                * larger than requested, and set up partial flag as well.
+                */
+               if (end < map.m_la + map.m_llen) {
                        count = end - map.m_la;
                        partial = true;
+               } else {
+                       ASSERT(end == map.m_la + map_m_llen);
+                       count = map.m_llen;
+                       partial = !(map.m_flags & EROFS_MAP_FULL_MAPPED);
                }
 
> +
> + if ((off_t)map.m_la < offset) {
> + ofs = offset - map.m_la;
> + sum = 0;
> + } else {
> + ofs = 0;
> + sum -= count;
> + }
 
a bit weird here, anyway, let me clean up here as well.
 
> +
> + ret = z_erofs_decompress(&(struct z_erofs_decompress_req) {
> + .in = raw,
> + .out = buffer,
> + .ofs_out = sum,
> + .ofs_head = ofs,
> + .inputsize = EROFS_BLKSIZ,
> + .outputsize = count,
> + .alg = algorithmformat,
> + .partial_decoding = partial
> + });
> + if (ret < 0)
> + return ret;
> + }
> +
> + logi("nid:%u size=%zd offset=%llu done",
> +      vnode->nid, size, (long long)offset);
> + return size;
> +}
>  
>  int erofs_read(const char *path, char *buffer, size_t size, off_t offset,
>         struct fuse_file_info *fi)
> @@ -107,7 +175,8 @@ int erofs_read(const char *path, char *buffer, size_t size, off_t offset,
>  
>  case EROFS_INODE_FLAT_COMPRESSION_LEGACY:
>  case EROFS_INODE_FLAT_COMPRESSION:
> - /* Fixme: */
> + return erofs_read_data_compression(&v, buffer, size, offset);
> +
>  default:
>  return -EINVAL;
>  }
> diff --git a/fuse/zmap.c b/fuse/zmap.c
> new file mode 100644
> index 0000000..022ca1b
> --- /dev/null
> +++ b/fuse/zmap.c
 
let's move this file to lib/ as well.
 
> @@ -0,0 +1,416 @@
> +// SPDX-License-Identifier: GPL-2.0-only
 
Let's relicense this file to GPL-2.0+ for erofs-utils
(since I'm the original author, and assume you agree on this as well..)
 
> +/*
> + * Many parts of codes are copied from Linux kernel/fs/erofs.
> + *
> + * Copyright (C) 2018-2019 HUAWEI, Inc.
> + *             https://www.huawei.com/
> + * Created by Gao Xiang <gaoxiang25@huawei.com>
> + * Modified by Huang Jianan <huangjianan@oppo.com>
> + */
> +
> +#include "init.h"
> +#include "disk_io.h"
> +#include "logging.h"
> +
> +int z_erofs_fill_inode(struct erofs_vnode *vi)
> +{
> + if (vi->datalayout == EROFS_INODE_FLAT_COMPRESSION_LEGACY) {
> + vi->z_advise = 0;
> + vi->z_algorithmtype[0] = 0;
> + vi->z_algorithmtype[1] = 0;
> + vi->z_logical_clusterbits = LOG_BLOCK_SIZE;
> + vi->z_physical_clusterbits[0] = vi->z_logical_clusterbits;
> + vi->z_physical_clusterbits[1] = vi->z_logical_clusterbits;
> + vi->z_inited = true;
> + }
> +
> + return 0;
> +}
> +
> +static int z_erofs_fill_inode_lazy(struct erofs_vnode *vi)
> +{
> + int ret;
> + erofs_off_t pos;
> + struct z_erofs_map_header *h;
> + char buf[8];
> +
> + if (vi->z_inited)
> + return 0;
> +
> + DBG_BUGON(vi->datalayout == EROFS_INODE_FLAT_COMPRESSION_LEGACY);
> +
> + pos = round_up(nid2addr(vi->nid) + vi->inode_isize + vi->xattr_isize, 8);
> +
> + ret = dev_read(buf, 8, pos);
> + if (ret < 0 && (uint32_t)ret != 8)
> + return -EIO;
> +
> + h = (struct z_erofs_map_header *)buf;
> + vi->z_advise = le16_to_cpu(h->h_advise);
> + vi->z_algorithmtype[0] = h->h_algorithmtype & 15;
> + vi->z_algorithmtype[1] = h->h_algorithmtype >> 4;
> +
> + if (vi->z_algorithmtype[0] >= Z_EROFS_COMPRESSION_MAX) {
> + loge("unknown compression format %u for nid %llu",
> +      vi->z_algorithmtype[0], vi->nid);
> + return -EOPNOTSUPP;
> + }
> +
> + vi->z_logical_clusterbits = LOG_BLOCK_SIZE + (h->h_clusterbits & 7);
> + vi->z_physical_clusterbits[0] = vi->z_logical_clusterbits +
> + ((h->h_clusterbits >> 3) & 3);
> +
> + if (vi->z_physical_clusterbits[0] != LOG_BLOCK_SIZE) {
> + loge("unsupported physical clusterbits %u for nid %llu",
> +      vi->z_physical_clusterbits[0], vi->nid);
> + return -EOPNOTSUPP;
> + }
> +
> + vi->z_physical_clusterbits[1] = vi->z_logical_clusterbits +
> + ((h->h_clusterbits >> 5) & 7);
> + vi->z_inited = true;
> +
> + return 0;
> +}
> +
> +struct z_erofs_maprecorder {
> + struct erofs_vnode *vnode;
> + struct erofs_map_blocks *map;
> + void *kaddr;
> +
> + unsigned long lcn;
> + /* compression extent information gathered */
> + u8  type;
> + u16 clusterofs;
> + u16 delta[2];
> + erofs_blk_t pblk;
> +};
> +
> +static int z_erofs_reload_indexes(struct z_erofs_maprecorder *m,
> +   erofs_blk_t eblk)
> +{
> + int ret;
> + struct erofs_map_blocks *const map = m->map;
> + char *mpage = map->mpage;
> +
> + if (map->index == eblk)
> + return 0;
> +
> + ret = dev_read(mpage, EROFS_BLKSIZ, blknr_to_addr(eblk));
> + if (ret < 0 && (uint32_t)ret != EROFS_BLKSIZ)
> + return -EIO;
> +
> + map->index = eblk;
> +
> + return 0;
> +}
> +
> +static int legacy_load_cluster_from_disk(struct z_erofs_maprecorder *m,
> + unsigned long lcn)
> +{
> + struct erofs_vnode *const vi = m->vnode;
> + const erofs_off_t ibase = nid2addr(vi->nid);
> + const erofs_off_t pos =
> + Z_EROFS_VLE_LEGACY_INDEX_ALIGN(ibase + vi->inode_isize +
> +        vi->xattr_isize) +
> + lcn * sizeof(struct z_erofs_vle_decompressed_index);
> + struct z_erofs_vle_decompressed_index *di;
> + unsigned int advise, type;
> + int err;
> +
> + err = z_erofs_reload_indexes(m, erofs_blknr(pos));
> + if (err)
> + return err;
> +
> + m->lcn = lcn;
> + di = m->kaddr + erofs_blkoff(pos);
> +
> + advise = le16_to_cpu(di->di_advise);
> + type = (advise >> Z_EROFS_VLE_DI_CLUSTER_TYPE_BIT) &
> + ((1 << Z_EROFS_VLE_DI_CLUSTER_TYPE_BITS) - 1);
> + switch (type) {
> + case Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD:
> + m->clusterofs = 1 << vi->z_logical_clusterbits;
> + m->delta[0] = le16_to_cpu(di->di_u.delta[0]);
> + m->delta[1] = le16_to_cpu(di->di_u.delta[1]);
> + break;
> + case Z_EROFS_VLE_CLUSTER_TYPE_PLAIN:
> + case Z_EROFS_VLE_CLUSTER_TYPE_HEAD:
> + m->clusterofs = le16_to_cpu(di->di_clusterofs);
> + m->pblk = le32_to_cpu(di->di_u.blkaddr);
> + break;
> + default:
> + DBG_BUGON(1);
> + return -EOPNOTSUPP;
> + }
> + m->type = type;
> + return 0;
> +}
> +
> +static unsigned int decode_compactedbits(unsigned int lobits,
> + unsigned int lomask,
> + u8 *in, unsigned int pos, u8 *type)
> +{
> + const unsigned int v = get_unaligned_le32(in + pos / 8) >> (pos & 7);
> + const unsigned int lo = v & lomask;
> +
> + *type = (v >> lobits) & 3;
> + return lo;
> +}
> +
> +static int unpack_compacted_index(struct z_erofs_maprecorder *m,
> +   unsigned int amortizedshift,
> +   unsigned int eofs)
> +{
> + struct erofs_vnode *const vi = m->vnode;
> + const unsigned int lclusterbits = vi->z_logical_clusterbits;
> + const unsigned int lomask = (1 << lclusterbits) - 1;
> + unsigned int vcnt, base, lo, encodebits, nblk;
> + int i;
> + u8 *in, type;
> +
> + if (1 << amortizedshift == 4)
> + vcnt = 2;
> + else if (1 << amortizedshift == 2 && lclusterbits == 12)
> + vcnt = 16;
> + else
> + return -EOPNOTSUPP;
> +
> + encodebits = ((vcnt << amortizedshift) - sizeof(__le32)) * 8 / vcnt;
> + base = round_down(eofs, vcnt << amortizedshift);
> + in = m->kaddr + base;
> +
> + i = (eofs - base) >> amortizedshift;
> +
> + lo = decode_compactedbits(lclusterbits, lomask,
> +   in, encodebits * i, &type);
> + m->type = type;
> + if (type == Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD) {
> + m->clusterofs = 1 << lclusterbits;
> + if (i + 1 != (int)vcnt) {
> + m->delta[0] = lo;
> + return 0;
> + }
> + /*
> + * since the last lcluster in the pack is special,
> + * of which lo saves delta[1] rather than delta[0].
> + * Hence, get delta[0] by the previous lcluster indirectly.
> + */
> + lo = decode_compactedbits(lclusterbits, lomask,
> +   in, encodebits * (i - 1), &type);
> + if (type != Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD)
> + lo = 0;
> + m->delta[0] = lo + 1;
> + return 0;
> + }
> + m->clusterofs = lo;
> + m->delta[0] = 0;
> + /* figout out blkaddr (pblk) for HEAD lclusters */
> + nblk = 1;
> + while (i > 0) {
> + --i;
> + lo = decode_compactedbits(lclusterbits, lomask,
> +   in, encodebits * i, &type);
> + if (type == Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD)
> + i -= lo;
> +
> + if (i >= 0)
> + ++nblk;
> + }
> + in += (vcnt << amortizedshift) - sizeof(__le32);
> + m->pblk = le32_to_cpu(*(__le32 *)in) + nblk;
> + return 0;
> +}
> +
> +static int compacted_load_cluster_from_disk(struct z_erofs_maprecorder *m,
> +     unsigned long lcn)
> +{
> + struct erofs_vnode *const vi = m->vnode;
> + const unsigned int lclusterbits = vi->z_logical_clusterbits;
> + const erofs_off_t ebase = round_up(nid2addr(vi->nid) + vi->inode_isize +
> +    vi->xattr_isize, 8) +
> + sizeof(struct z_erofs_map_header);
> + const unsigned int totalidx = DIV_ROUND_UP(vi->i_size, EROFS_BLKSIZ);
> + unsigned int compacted_4b_initial, compacted_2b;
> + unsigned int amortizedshift;
> + erofs_off_t pos;
> + int err;
> +
> + if (lclusterbits != 12)
> + return -EOPNOTSUPP;
> +
> + if (lcn >= totalidx)
> + return -EINVAL;
> +
> + m->lcn = lcn;
> + /* used to align to 32-byte (compacted_2b) alignment */
> + compacted_4b_initial = (32 - ebase % 32) / 4;
> + if (compacted_4b_initial == 32 / 4)
> + compacted_4b_initial = 0;
> +
> + if (vi->z_advise & Z_EROFS_ADVISE_COMPACTED_2B)
> + compacted_2b = rounddown(totalidx - compacted_4b_initial, 16);
> + else
> + compacted_2b = 0;
> +
> + pos = ebase;
> + if (lcn < compacted_4b_initial) {
> + amortizedshift = 2;
> + goto out;
> + }
> + pos += compacted_4b_initial * 4;
> + lcn -= compacted_4b_initial;
> +
> + if (lcn < compacted_2b) {
> + amortizedshift = 1;
> + goto out;
> + }
> + pos += compacted_2b * 2;
> + lcn -= compacted_2b;
> + amortizedshift = 2;
> +out:
> + pos += lcn * (1 << amortizedshift);
> + err = z_erofs_reload_indexes(m, erofs_blknr(pos));
> + if (err)
> + return err;
> + return unpack_compacted_index(m, amortizedshift, erofs_blkoff(pos));
> +}
> +
> +static int z_erofs_load_cluster_from_disk(struct z_erofs_maprecorder *m,
> +   unsigned int lcn)
> +{
> + const unsigned int datamode = m->vnode->datalayout;
> +
> + if (datamode == EROFS_INODE_FLAT_COMPRESSION_LEGACY)
> + return legacy_load_cluster_from_disk(m, lcn);
> +
> + if (datamode == EROFS_INODE_FLAT_COMPRESSION)
> + return compacted_load_cluster_from_disk(m, lcn);
> +
> + return -EINVAL;
> +}
> +
> +static int z_erofs_extent_lookback(struct z_erofs_maprecorder *m,
> +    unsigned int lookback_distance)
> +{
> + struct erofs_vnode *const vi = m->vnode;
> + struct erofs_map_blocks *const map = m->map;
> + const unsigned int lclusterbits = vi->z_logical_clusterbits;
> + unsigned long lcn = m->lcn;
> + int err;
> +
> + if (lcn < lookback_distance) {
> + loge("bogus lookback distance @ nid %llu", vi->nid);
> + DBG_BUGON(1);
> + return -EFSCORRUPTED;
> + }
> +
> + /* load extent head logical cluster if needed */
> + lcn -= lookback_distance;
> + err = z_erofs_load_cluster_from_disk(m, lcn);
> + if (err)
> + return err;
> +
> + switch (m->type) {
> + case Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD:
> + if (!m->delta[0]) {
> + loge("invalid lookback distance 0 @ nid %llu",
> +   vi->nid);
> + DBG_BUGON(1);
> + return -EFSCORRUPTED;
> + }
> + return z_erofs_extent_lookback(m, m->delta[0]);
> + case Z_EROFS_VLE_CLUSTER_TYPE_PLAIN:
> + map->m_flags &= ~EROFS_MAP_ZIPPED;
> + case Z_EROFS_VLE_CLUSTER_TYPE_HEAD:
> + map->m_la = (lcn << lclusterbits) | m->clusterofs;
> + break;
> + default:
> + loge("unknown type %u @ lcn %lu of nid %llu",
> +      m->type, lcn, vi->nid);
> + DBG_BUGON(1);
> + return -EOPNOTSUPP;
> + }
> + return 0;
> +}
> +
> +int z_erofs_map_blocks_iter(struct erofs_vnode *vi,
> +     struct erofs_map_blocks *map)
> +{
> + struct z_erofs_maprecorder m = {
> + .vnode = vi,
> + .map = map,
> + .kaddr = map->mpage,
> + };
> + int err = 0;
> + unsigned int lclusterbits, endoff;
> + unsigned long long ofs, end;
> +
> + /* when trying to read beyond EOF, leave it unmapped */
> + if (map->m_la >= vi->i_size) {
> + map->m_llen = map->m_la + 1 - vi->i_size;
> + map->m_la = vi->i_size;
> + map->m_flags = 0;
> + goto out;
> + }
> +
> + err = z_erofs_fill_inode_lazy(vi);
> + if (err)
> + goto out;
> +
> + lclusterbits = vi->z_logical_clusterbits;
> + ofs = map->m_la;
> + m.lcn = ofs >> lclusterbits;
> + endoff = ofs & ((1 << lclusterbits) - 1);
> +
> + err = z_erofs_load_cluster_from_disk(&m, m.lcn);
> + if (err)
> + goto out;
> +
> + map->m_flags = EROFS_MAP_ZIPPED; /* by default, compressed */
> + end = (m.lcn + 1ULL) << lclusterbits;
> + switch (m.type) {
> + case Z_EROFS_VLE_CLUSTER_TYPE_PLAIN:
> + if (endoff >= m.clusterofs)
> + map->m_flags &= ~EROFS_MAP_ZIPPED;
> + case Z_EROFS_VLE_CLUSTER_TYPE_HEAD:
> + if (endoff >= m.clusterofs) {
> + map->m_la = (m.lcn << lclusterbits) | m.clusterofs;
> + break;
> + }
> + /* m.lcn should be >= 1 if endoff < m.clusterofs */
> + if (!m.lcn) {
> + loge("invalid logical cluster 0 at nid %llu",
> +      vi->nid);
> + err = -EFSCORRUPTED;
> + goto out;
> + }
> + end = (m.lcn << lclusterbits) | m.clusterofs;
> + map->m_flags |= EROFS_MAP_FULL_MAPPED;
> + m.delta[0] = 1;
> + case Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD:
> + /* get the correspoinding first chunk */
> + err = z_erofs_extent_lookback(&m, m.delta[0]);
> + if (err)
> + goto out;
> + break;
> + default:
> + loge("unknown type %u @ offset %llu of nid %llu",
> +      m.type, ofs, vi->nid);
> + err = -EOPNOTSUPP;
> + goto out;
> + }
> +
> + map->m_llen = end - map->m_la;
> + map->m_plen = 1 << lclusterbits;
> + map->m_pa = blknr_to_addr(m.pblk);
> + map->m_flags |= EROFS_MAP_MAPPED;
> +
> +out:
> + logd("m_la %llu m_pa %llu m_llen %llu m_plen %llu m_flags 0%o",
> +      map->m_la, map->m_pa,
> +      map->m_llen, map->m_plen, map->m_flags);
> +
> + DBG_BUGON(err < 0 && err != -ENOMEM);
> + return err;
> +}
> diff --git a/include/erofs/defs.h b/include/erofs/defs.h
> index a9c769e..06d29a9 100644
> --- a/include/erofs/defs.h
> +++ b/include/erofs/defs.h
> @@ -173,5 +173,18 @@ typedef int64_t         s64;
>  #define __maybe_unused      __attribute__((__unused__))
>  #endif
>  
> +struct __una_u32 { u32 x; } __packed;
> +
> +static inline u32 __get_unaligned_cpu32(const void *p)
> +{
> + const struct __una_u32 *ptr = (const struct __una_u32 *)p;
> + return ptr->x;
> +}
> +
> +static inline u32 get_unaligned_le32(const void *p)
> +{
> + return __get_unaligned_cpu32((const u8 *)p);
> +}
 
need to handle big-endian, anyway, minor for now.
 
Thanks,
Gao Xiang