new file mode 100644
@@ -0,0 +1,80 @@
+From 5782f0d47df99dcfc743aa138361336e9a4ac966 Mon Sep 17 00:00:00 2001
+From: Gao Xiang <hsiangkao@linux.alibaba.com>
+Date: Fri, 2 Jun 2023 13:52:56 +0800
+Subject: [PATCH 1/4] erofs-utils: fsck: block insane long paths when
+ extracting images
+
+Since some crafted EROFS filesystem images could have insane deep
+hierarchy (or may form directory loops) which triggers the
+PATH_MAX-sized path buffer OR stack overflow.
+
+Actually some crafted images cannot be deemed as real corrupted
+images but over-PATH_MAX paths are not something that we'd like to
+support for now.
+
+CVE: CVE-2023-33551
+Closes: https://nvd.nist.gov/vuln/detail/CVE-2023-33551
+Reported-by: Chaoming Yang <lometsj@live.com>
+Fixes: f44043561491 ("erofs-utils: introduce fsck.erofs")
+Fixes: b11f84f593f9 ("erofs-utils: fsck: convert to use erofs_iterate_dir()")
+Fixes: 412c8f908132 ("erofs-utils: fsck: add --extract=X support to extract to path X")
+Signeo-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
+Link: https://lore.kernel.org/r/20230602055256.18061-1-hsiangkao@linux.alibaba.com
+
+Upstream-Status: Backport [https://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs-utils.git/patch/?id=27aeef179bf17d5f1d98f827e93d24839a6d4176]
+Signed-off-by: Changqing Li <changqing.li@windriver.com>
+---
+ fsck/main.c | 23 +++++++++++++++--------
+ 1 file changed, 15 insertions(+), 8 deletions(-)
+
+diff --git a/fsck/main.c b/fsck/main.c
+index 5a2f659..2b6a6dd 100644
+--- a/fsck/main.c
++++ b/fsck/main.c
+@@ -679,28 +679,35 @@ again:
+ static int erofsfsck_dirent_iter(struct erofs_dir_context *ctx)
+ {
+ int ret;
+- size_t prev_pos = fsckcfg.extract_pos;
++ size_t prev_pos, curr_pos;
+
+ if (ctx->dot_dotdot)
+ return 0;
+
+- if (fsckcfg.extract_path) {
+- size_t curr_pos = prev_pos;
++ prev_pos = fsckcfg.extract_pos;
++ curr_pos = prev_pos;
++
++ if (prev_pos + ctx->de_namelen >= PATH_MAX) {
++ erofs_err("unable to fsck since the path is too long (%u)",
++ curr_pos + ctx->de_namelen);
++ return -EOPNOTSUPP;
++ }
+
++ if (fsckcfg.extract_path) {
+ fsckcfg.extract_path[curr_pos++] = '/';
+ strncpy(fsckcfg.extract_path + curr_pos, ctx->dname,
+ ctx->de_namelen);
+ curr_pos += ctx->de_namelen;
+ fsckcfg.extract_path[curr_pos] = '\0';
+- fsckcfg.extract_pos = curr_pos;
++ } else {
++ curr_pos += ctx->de_namelen;
+ }
+-
++ fsckcfg.extract_pos = curr_pos;
+ ret = erofsfsck_check_inode(ctx->dir->nid, ctx->de_nid);
+
+- if (fsckcfg.extract_path) {
++ if (fsckcfg.extract_path)
+ fsckcfg.extract_path[prev_pos] = '\0';
+- fsckcfg.extract_pos = prev_pos;
+- }
++ fsckcfg.extract_pos = prev_pos;
+ return ret;
+ }
+
+--
+2.25.1
+
new file mode 100644
@@ -0,0 +1,221 @@
+From 8aef6015a03242a7d13467d23ad52b5427bf5247 Mon Sep 17 00:00:00 2001
+From: Yue Hu <huyue2@coolpad.com>
+Date: Wed, 11 Jan 2023 09:49:26 +0800
+Subject: [PATCH] erofs-utils: lib: export parts of erofs_pread()
+
+Export parts of erofs_pread() to avoid duplicated code in
+erofs_verify_inode_data(). Let's make two helpers for this.
+
+Signed-off-by: Yue Hu <huyue2@coolpad.com>
+Link: https://lore.kernel.org/r/ff560da9c798b2ca1f1a663a000501486d865487.1673401718.git.huyue2@coolpad.com
+Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
+
+CVE: CVE-2023-33552
+Upstream-Status: Backport [https://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs-utils.git/commit/?id=4c0fb15a5d85378debe9d10d96cd643d167300ca]
+Signed-off-by: Changqing Li <changqing.li@windriver.com>
+---
+ include/erofs/internal.h | 5 ++
+ lib/data.c | 108 ++++++++++++++++++++++++---------------
+ 2 files changed, 71 insertions(+), 42 deletions(-)
+
+diff --git a/include/erofs/internal.h b/include/erofs/internal.h
+index d3b2986..28d0e68 100644
+--- a/include/erofs/internal.h
++++ b/include/erofs/internal.h
+@@ -335,6 +335,11 @@ int erofs_pread(struct erofs_inode *inode, char *buf,
+ int erofs_map_blocks(struct erofs_inode *inode,
+ struct erofs_map_blocks *map, int flags);
+ int erofs_map_dev(struct erofs_sb_info *sbi, struct erofs_map_dev *map);
++int erofs_read_one_data(struct erofs_map_blocks *map, char *buffer, u64 offset,
++ size_t len);
++int z_erofs_read_one_data(struct erofs_inode *inode,
++ struct erofs_map_blocks *map, char *raw, char *buffer,
++ erofs_off_t skip, erofs_off_t length, bool trimmed);
+
+ static inline int erofs_get_occupied_size(const struct erofs_inode *inode,
+ erofs_off_t *size)
+diff --git a/lib/data.c b/lib/data.c
+index 6bc554d..2a7fdd5 100644
+--- a/lib/data.c
++++ b/lib/data.c
+@@ -158,19 +158,38 @@ int erofs_map_dev(struct erofs_sb_info *sbi, struct erofs_map_dev *map)
+ return 0;
+ }
+
++int erofs_read_one_data(struct erofs_map_blocks *map, char *buffer, u64 offset,
++ size_t len)
++{
++ struct erofs_map_dev mdev;
++ int ret;
++
++ mdev = (struct erofs_map_dev) {
++ .m_deviceid = map->m_deviceid,
++ .m_pa = map->m_pa,
++ };
++ ret = erofs_map_dev(&sbi, &mdev);
++ if (ret)
++ return ret;
++
++ ret = dev_read(mdev.m_deviceid, buffer, mdev.m_pa + offset, len);
++ if (ret < 0)
++ return -EIO;
++ return 0;
++}
++
+ static int erofs_read_raw_data(struct erofs_inode *inode, char *buffer,
+ erofs_off_t size, erofs_off_t offset)
+ {
+ struct erofs_map_blocks map = {
+ .index = UINT_MAX,
+ };
+- struct erofs_map_dev mdev;
+ int ret;
+ erofs_off_t ptr = offset;
+
+ while (ptr < offset + size) {
+ char *const estart = buffer + ptr - offset;
+- erofs_off_t eend;
++ erofs_off_t eend, moff = 0;
+
+ map.m_la = ptr;
+ ret = erofs_map_blocks(inode, &map, 0);
+@@ -179,14 +198,6 @@ static int erofs_read_raw_data(struct erofs_inode *inode, char *buffer,
+
+ DBG_BUGON(map.m_plen != map.m_llen);
+
+- mdev = (struct erofs_map_dev) {
+- .m_deviceid = map.m_deviceid,
+- .m_pa = map.m_pa,
+- };
+- ret = erofs_map_dev(&sbi, &mdev);
+- if (ret)
+- return ret;
+-
+ /* trim extent */
+ eend = min(offset + size, map.m_la + map.m_llen);
+ DBG_BUGON(ptr < map.m_la);
+@@ -204,19 +215,54 @@ static int erofs_read_raw_data(struct erofs_inode *inode, char *buffer,
+ }
+
+ if (ptr > map.m_la) {
+- mdev.m_pa += ptr - map.m_la;
++ moff = ptr - map.m_la;
+ map.m_la = ptr;
+ }
+
+- ret = dev_read(mdev.m_deviceid, estart, mdev.m_pa,
+- eend - map.m_la);
+- if (ret < 0)
+- return -EIO;
++ ret = erofs_read_one_data(&map, estart, moff, eend - map.m_la);
++ if (ret)
++ return ret;
+ ptr = eend;
+ }
+ return 0;
+ }
+
++int z_erofs_read_one_data(struct erofs_inode *inode,
++ struct erofs_map_blocks *map, char *raw, char *buffer,
++ erofs_off_t skip, erofs_off_t length, bool trimmed)
++{
++ struct erofs_map_dev mdev;
++ int ret = 0;
++
++ /* no device id here, thus it will always succeed */
++ mdev = (struct erofs_map_dev) {
++ .m_pa = map->m_pa,
++ };
++ ret = erofs_map_dev(&sbi, &mdev);
++ if (ret) {
++ DBG_BUGON(1);
++ return ret;
++ }
++
++ ret = dev_read(mdev.m_deviceid, raw, mdev.m_pa, map->m_plen);
++ if (ret < 0)
++ return ret;
++
++ ret = z_erofs_decompress(&(struct z_erofs_decompress_req) {
++ .in = raw,
++ .out = buffer,
++ .decodedskip = skip,
++ .inputsize = map->m_plen,
++ .decodedlength = length,
++ .alg = map->m_algorithmformat,
++ .partial_decoding = trimmed ? true :
++ !(map->m_flags & EROFS_MAP_FULL_MAPPED)
++ });
++ if (ret < 0)
++ return ret;
++ return 0;
++}
++
+ static int z_erofs_read_data(struct erofs_inode *inode, char *buffer,
+ erofs_off_t size, erofs_off_t offset)
+ {
+@@ -224,8 +270,7 @@ static int z_erofs_read_data(struct erofs_inode *inode, char *buffer,
+ struct erofs_map_blocks map = {
+ .index = UINT_MAX,
+ };
+- struct erofs_map_dev mdev;
+- bool partial;
++ bool trimmed;
+ unsigned int bufsize = 0;
+ char *raw = NULL;
+ int ret = 0;
+@@ -238,27 +283,17 @@ static int z_erofs_read_data(struct erofs_inode *inode, char *buffer,
+ if (ret)
+ break;
+
+- /* no device id here, thus it will always succeed */
+- mdev = (struct erofs_map_dev) {
+- .m_pa = map.m_pa,
+- };
+- ret = erofs_map_dev(&sbi, &mdev);
+- if (ret) {
+- DBG_BUGON(1);
+- break;
+- }
+-
+ /*
+ * 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) {
+ length = end - map.m_la;
+- partial = true;
++ trimmed = true;
+ } else {
+ DBG_BUGON(end != map.m_la + map.m_llen);
+ length = map.m_llen;
+- partial = !(map.m_flags & EROFS_MAP_FULL_MAPPED);
++ trimmed = false;
+ }
+
+ if (map.m_la < offset) {
+@@ -283,19 +318,8 @@ static int z_erofs_read_data(struct erofs_inode *inode, char *buffer,
+ break;
+ }
+ }
+- ret = dev_read(mdev.m_deviceid, raw, mdev.m_pa, map.m_plen);
+- if (ret < 0)
+- break;
+-
+- ret = z_erofs_decompress(&(struct z_erofs_decompress_req) {
+- .in = raw,
+- .out = buffer + end - offset,
+- .decodedskip = skip,
+- .inputsize = map.m_plen,
+- .decodedlength = length,
+- .alg = map.m_algorithmformat,
+- .partial_decoding = partial
+- });
++ ret = z_erofs_read_one_data(inode, &map, raw,
++ buffer + end - offset, skip, length, trimmed);
+ if (ret < 0)
+ break;
+ }
+--
+2.25.1
+
new file mode 100644
@@ -0,0 +1,97 @@
+From 3a360e01058467573bd7239fa430d8dc5fbd60f4 Mon Sep 17 00:00:00 2001
+From: Yue Hu <huyue2@coolpad.com>
+Date: Wed, 11 Jan 2023 09:49:27 +0800
+Subject: [PATCH 3/4] erofs-utils: fsck: cleanup erofs_verify_inode_data()
+
+Diretly call {z_}erofs_read_one_data() to avoid duplicated code.
+Accordingly, fragment and partial-referenced plusters are also supported
+after this change.
+
+Signed-off-by: Yue Hu <huyue2@coolpad.com>
+Link: https://lore.kernel.org/r/115e61fc9c2d34cab6d3dd78383ac57c94a491fc.1673401718.git.huyue2@coolpad.com
+Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
+
+CVE: CVE-2023-33552
+Upstream-Status: Backport [https://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs-utils.git/commit/?id=87430c69e1d542928c4519e8fabfd6348a741999]
+Signed-off-by: Changqing Li <changqing.li@windriver.com>
+---
+ fsck/main.c | 53 ++++++++++-------------------------------------------
+ 1 file changed, 10 insertions(+), 43 deletions(-)
+
+diff --git a/fsck/main.c b/fsck/main.c
+index 2b6a6dd..92ef17a 100644
+--- a/fsck/main.c
++++ b/fsck/main.c
+@@ -366,7 +366,6 @@ static int erofs_verify_inode_data(struct erofs_inode *inode, int outfd)
+ struct erofs_map_blocks map = {
+ .index = UINT_MAX,
+ };
+- struct erofs_map_dev mdev;
+ int ret = 0;
+ bool compressed;
+ erofs_off_t pos = 0;
+@@ -427,51 +426,19 @@ static int erofs_verify_inode_data(struct erofs_inode *inode, int outfd)
+ BUG_ON(!raw);
+ }
+
+- mdev = (struct erofs_map_dev) {
+- .m_deviceid = map.m_deviceid,
+- .m_pa = map.m_pa,
+- };
+- ret = erofs_map_dev(&sbi, &mdev);
+- if (ret) {
+- erofs_err("failed to map device of m_pa %" PRIu64 ", m_deviceid %u @ nid %llu: %d",
+- map.m_pa, map.m_deviceid, inode->nid | 0ULL,
+- ret);
+- goto out;
+- }
+-
+- if (compressed && map.m_llen > buffer_size) {
+- buffer_size = map.m_llen;
+- buffer = realloc(buffer, buffer_size);
+- BUG_ON(!buffer);
+- }
+-
+- ret = dev_read(mdev.m_deviceid, raw, mdev.m_pa, map.m_plen);
+- if (ret < 0) {
+- erofs_err("failed to read data of m_pa %" PRIu64 ", m_plen %" PRIu64 " @ nid %llu: %d",
+- mdev.m_pa, map.m_plen, inode->nid | 0ULL,
+- ret);
+- goto out;
+- }
+-
+ if (compressed) {
+- struct z_erofs_decompress_req rq = {
+- .in = raw,
+- .out = buffer,
+- .decodedskip = 0,
+- .inputsize = map.m_plen,
+- .decodedlength = map.m_llen,
+- .alg = map.m_algorithmformat,
+- .partial_decoding = 0
+- };
+-
+- ret = z_erofs_decompress(&rq);
+- if (ret < 0) {
+- erofs_err("failed to decompress data of m_pa %" PRIu64 ", m_plen %" PRIu64 " @ nid %llu: %s",
+- mdev.m_pa, map.m_plen,
+- inode->nid | 0ULL, strerror(-ret));
+- goto out;
++ if (map.m_llen > buffer_size) {
++ buffer_size = map.m_llen;
++ buffer = realloc(buffer, buffer_size);
++ BUG_ON(!buffer);
+ }
++ ret = z_erofs_read_one_data(inode, &map, raw, buffer,
++ 0, map.m_llen, false);
++ } else {
++ ret = erofs_read_one_data(&map, raw, 0, map.m_plen);
+ }
++ if (ret)
++ goto out;
+
+ if (outfd >= 0 && write(outfd, compressed ? buffer : raw,
+ map.m_llen) < 0) {
+--
+2.25.1
+
new file mode 100644
@@ -0,0 +1,127 @@
+From b4e155ba759ae389c5f71cd13d97eb3bcf2c1adf Mon Sep 17 00:00:00 2001
+From: Gao Xiang <hsiangkao@linux.alibaba.com>
+Date: Fri, 2 Jun 2023 11:05:19 +0800
+Subject: [PATCH] erofs-utils: fsck: don't allocate/read too large extents
+
+Since some crafted EROFS filesystem images could have insane large
+extents, which causes unexpected bahaviors when extracting data.
+
+Fix it by extracting large extents with a buffer of a reasonable
+maximum size limit and reading multiple times instead.
+
+Note that only `--extract` option is impacted.
+
+CVE: CVE-2023-33552
+Closes: https://nvd.nist.gov/vuln/detail/CVE-2023-33552
+Reported-by: Chaoming Yang <lometsj@live.com>
+Fixes: 412c8f908132 ("erofs-utils: fsck: add --extract=X support to extract to path X")
+Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
+Link: https://lore.kernel.org/r/20230602030519.117071-1-hsiangkao@linux.alibaba.com
+
+CVE: CVE-2023-33552
+Upstream-Status: Backport [https://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs-utils.git/patch/?id=2145dff03dd3f3f74bcda3b52160fbad37f7fcfe]
+Signed-off-by: Changqing Li <changqing.li@windriver.com>
+---
+ fsck/main.c | 64 ++++++++++++++++++++++++++++++++++++++++-------------
+ 1 file changed, 49 insertions(+), 15 deletions(-)
+
+diff --git a/fsck/main.c b/fsck/main.c
+index 92ef17a..1bd1117 100644
+--- a/fsck/main.c
++++ b/fsck/main.c
+@@ -392,6 +392,8 @@ static int erofs_verify_inode_data(struct erofs_inode *inode, int outfd)
+ }
+
+ while (pos < inode->i_size) {
++ unsigned int alloc_rawsize;
++
+ map.m_la = pos;
+ if (compressed)
+ ret = z_erofs_map_blocks_iter(inode, &map,
+@@ -420,10 +422,28 @@ static int erofs_verify_inode_data(struct erofs_inode *inode, int outfd)
+ if (!(map.m_flags & EROFS_MAP_MAPPED) || !fsckcfg.check_decomp)
+ continue;
+
+- if (map.m_plen > raw_size) {
+- raw_size = map.m_plen;
+- raw = realloc(raw, raw_size);
+- BUG_ON(!raw);
++ if (map.m_plen > Z_EROFS_PCLUSTER_MAX_SIZE) {
++ if (compressed) {
++ erofs_err("invalid pcluster size %" PRIu64 " @ offset %" PRIu64 " of nid %" PRIu64,
++ map.m_plen, map.m_la,
++ inode->nid | 0ULL);
++ ret = -EFSCORRUPTED;
++ goto out;
++ }
++ alloc_rawsize = Z_EROFS_PCLUSTER_MAX_SIZE;
++ } else {
++ alloc_rawsize = map.m_plen;
++ }
++
++ if (alloc_rawsize > raw_size) {
++ char *newraw = realloc(raw, alloc_rawsize);
++
++ if (!newraw) {
++ ret = -ENOMEM;
++ goto out;
++ }
++ raw = newraw;
++ raw_size = alloc_rawsize;
+ }
+
+ if (compressed) {
+@@ -434,18 +454,26 @@ static int erofs_verify_inode_data(struct erofs_inode *inode, int outfd)
+ }
+ ret = z_erofs_read_one_data(inode, &map, raw, buffer,
+ 0, map.m_llen, false);
+- } else {
+- ret = erofs_read_one_data(&map, raw, 0, map.m_plen);
+- }
+- if (ret)
+- goto out;
++ if (ret)
++ goto out;
+
+- if (outfd >= 0 && write(outfd, compressed ? buffer : raw,
+- map.m_llen) < 0) {
+- erofs_err("I/O error occurred when verifying data chunk @ nid %llu",
+- inode->nid | 0ULL);
+- ret = -EIO;
+- goto out;
++ if (outfd >= 0 && write(outfd, buffer, map.m_llen) < 0)
++ goto fail_eio;
++ } else {
++ u64 p = 0;
++ do {
++ u64 count = min_t(u64, alloc_rawsize,
++ map.m_llen);
++
++ ret = erofs_read_one_data(&map, raw, p, count);
++ if (ret)
++ goto out;
++
++ if (outfd >= 0 && write(outfd, raw, count) < 0)
++ goto fail_eio;
++ map.m_llen -= count;
++ p += count;
++ } while (map.m_llen);
+ }
+ }
+
+@@ -461,6 +489,12 @@ out:
+ if (buffer)
+ free(buffer);
+ return ret < 0 ? ret : 0;
++
++fail_eio:
++ erofs_err("I/O error occurred when verifying data chunk @ nid %llu",
++ inode->nid | 0ULL);
++ ret = -EIO;
++ goto out;
+ }
+
+ static inline int erofs_extract_dir(struct erofs_inode *inode)
+--
+2.25.1
+
@@ -10,6 +10,10 @@ SRC_URI = "git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs-utils.git;b
file://0001-configure-use-AC_SYS_LARGEFILE.patch \
file://0002-erofs-replace-l-stat64-by-equivalent-l-stat.patch \
file://0003-internal.h-Make-LFS-mandatory-for-all-usecases.patch \
+ file://CVE-2023-33551.patch \
+ file://CVE-2023-33552-1.patch \
+ file://CVE-2023-33552-2.patch \
+ file://CVE-2023-33552-3.patch \
"
UPSTREAM_CHECK_GITTAGREGEX = "v(?P<pver>(\d+(\.\d+)+))"