October 2025 - Linux-stable-mirror

FAILED: patch "[PATCH] NFSD: Fix last write offset handling in layoutcommit" failed to apply to 6.12-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.12-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(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.12.y git checkout FETCH_HEAD git cherry-pick -x d68886bae76a4b9b3484d23e5b7df086f940fa38 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025101657-preseason-garnish-c1e0@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From d68886bae76a4b9b3484d23e5b7df086f940fa38 Mon Sep 17 00:00:00 2001 From: Sergey Bashirov <sergeybashirov(a)gmail.com> Date: Mon, 21 Jul 2025 21:40:56 +0300 Subject: [PATCH] NFSD: Fix last write offset handling in layoutcommit The data type of loca_last_write_offset is newoffset4 and is switched on a boolean value, no_newoffset, that indicates if a previous write occurred or not. If no_newoffset is FALSE, an offset is not given. This means that client does not try to update the file size. Thus, server should not try to calculate new file size and check if it fits into the segment range. See RFC 8881, section 12.5.4.2. Sometimes the current incorrect logic may cause clients to hang when trying to sync an inode. If layoutcommit fails, the client marks the inode as dirty again. Fixes: 9cf514ccfacb ("nfsd: implement pNFS operations") Cc: stable(a)vger.kernel.org Co-developed-by: Konstantin Evtushenko <koevtushenko(a)yandex.com> Signed-off-by: Konstantin Evtushenko <koevtushenko(a)yandex.com> Signed-off-by: Sergey Bashirov <sergeybashirov(a)gmail.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Reviewed-by: Jeff Layton <jlayton(a)kernel.org> Signed-off-by: Chuck Lever <chuck.lever(a)oracle.com> diff --git a/fs/nfsd/blocklayout.c b/fs/nfsd/blocklayout.c index 4c936132eb44..0822d8a119c6 100644 --- a/fs/nfsd/blocklayout.c +++ b/fs/nfsd/blocklayout.c @@ -118,7 +118,6 @@ nfsd4_block_commit_blocks(struct inode *inode, struct nfsd4_layoutcommit *lcp, struct iomap *iomaps, int nr_iomaps) { struct timespec64 mtime = inode_get_mtime(inode); - loff_t new_size = lcp->lc_last_wr + 1; struct iattr iattr = { .ia_valid = 0 }; int error; @@ -128,9 +127,9 @@ nfsd4_block_commit_blocks(struct inode *inode, struct nfsd4_layoutcommit *lcp, iattr.ia_valid |= ATTR_ATIME | ATTR_CTIME | ATTR_MTIME; iattr.ia_atime = iattr.ia_ctime = iattr.ia_mtime = lcp->lc_mtime; - if (new_size > i_size_read(inode)) { + if (lcp->lc_size_chg) { iattr.ia_valid |= ATTR_SIZE; - iattr.ia_size = new_size; + iattr.ia_size = lcp->lc_newsize; } error = inode->i_sb->s_export_op->commit_blocks(inode, iomaps, diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 656b2e7d8840..7043fc475458 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -2475,7 +2475,6 @@ nfsd4_layoutcommit(struct svc_rqst *rqstp, const struct nfsd4_layout_seg *seg = &lcp->lc_seg; struct svc_fh *current_fh = &cstate->current_fh; const struct nfsd4_layout_ops *ops; - loff_t new_size = lcp->lc_last_wr + 1; struct inode *inode; struct nfs4_layout_stateid *ls; __be32 nfserr; @@ -2491,13 +2490,21 @@ nfsd4_layoutcommit(struct svc_rqst *rqstp, goto out; inode = d_inode(current_fh->fh_dentry); - nfserr = nfserr_inval; - if (new_size <= seg->offset) - goto out; - if (new_size > seg->offset + seg->length) - goto out; - if (!lcp->lc_newoffset && new_size > i_size_read(inode)) - goto out; + lcp->lc_size_chg = false; + if (lcp->lc_newoffset) { + loff_t new_size = lcp->lc_last_wr + 1; + + nfserr = nfserr_inval; + if (new_size <= seg->offset) + goto out; + if (new_size > seg->offset + seg->length) + goto out; + + if (new_size > i_size_read(inode)) { + lcp->lc_size_chg = true; + lcp->lc_newsize = new_size; + } + } nfserr = nfsd4_preprocess_layout_stateid(rqstp, cstate, &lcp->lc_sid, false, lcp->lc_layout_type, @@ -2513,13 +2520,6 @@ nfsd4_layoutcommit(struct svc_rqst *rqstp, /* LAYOUTCOMMIT does not require any serialization */ mutex_unlock(&ls->ls_mutex); - if (new_size > i_size_read(inode)) { - lcp->lc_size_chg = true; - lcp->lc_newsize = new_size; - } else { - lcp->lc_size_chg = false; - } - nfserr = ops->proc_layoutcommit(inode, rqstp, lcp); nfs4_put_stid(&ls->ls_stid); out:

2 weeks, 3 days

2
7
0 0

FAILED: patch "[PATCH] NFSD: Fix last write offset handling in layoutcommit" failed to apply to 6.17-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.17-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(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.17.y git checkout FETCH_HEAD git cherry-pick -x d68886bae76a4b9b3484d23e5b7df086f940fa38 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025101656-unfrosted-rasping-f7dc@gregkh' --subject-prefix 'PATCH 6.17.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From d68886bae76a4b9b3484d23e5b7df086f940fa38 Mon Sep 17 00:00:00 2001 From: Sergey Bashirov <sergeybashirov(a)gmail.com> Date: Mon, 21 Jul 2025 21:40:56 +0300 Subject: [PATCH] NFSD: Fix last write offset handling in layoutcommit The data type of loca_last_write_offset is newoffset4 and is switched on a boolean value, no_newoffset, that indicates if a previous write occurred or not. If no_newoffset is FALSE, an offset is not given. This means that client does not try to update the file size. Thus, server should not try to calculate new file size and check if it fits into the segment range. See RFC 8881, section 12.5.4.2. Sometimes the current incorrect logic may cause clients to hang when trying to sync an inode. If layoutcommit fails, the client marks the inode as dirty again. Fixes: 9cf514ccfacb ("nfsd: implement pNFS operations") Cc: stable(a)vger.kernel.org Co-developed-by: Konstantin Evtushenko <koevtushenko(a)yandex.com> Signed-off-by: Konstantin Evtushenko <koevtushenko(a)yandex.com> Signed-off-by: Sergey Bashirov <sergeybashirov(a)gmail.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Reviewed-by: Jeff Layton <jlayton(a)kernel.org> Signed-off-by: Chuck Lever <chuck.lever(a)oracle.com> diff --git a/fs/nfsd/blocklayout.c b/fs/nfsd/blocklayout.c index 4c936132eb44..0822d8a119c6 100644 --- a/fs/nfsd/blocklayout.c +++ b/fs/nfsd/blocklayout.c @@ -118,7 +118,6 @@ nfsd4_block_commit_blocks(struct inode *inode, struct nfsd4_layoutcommit *lcp, struct iomap *iomaps, int nr_iomaps) { struct timespec64 mtime = inode_get_mtime(inode); - loff_t new_size = lcp->lc_last_wr + 1; struct iattr iattr = { .ia_valid = 0 }; int error; @@ -128,9 +127,9 @@ nfsd4_block_commit_blocks(struct inode *inode, struct nfsd4_layoutcommit *lcp, iattr.ia_valid |= ATTR_ATIME | ATTR_CTIME | ATTR_MTIME; iattr.ia_atime = iattr.ia_ctime = iattr.ia_mtime = lcp->lc_mtime; - if (new_size > i_size_read(inode)) { + if (lcp->lc_size_chg) { iattr.ia_valid |= ATTR_SIZE; - iattr.ia_size = new_size; + iattr.ia_size = lcp->lc_newsize; } error = inode->i_sb->s_export_op->commit_blocks(inode, iomaps, diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 656b2e7d8840..7043fc475458 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -2475,7 +2475,6 @@ nfsd4_layoutcommit(struct svc_rqst *rqstp, const struct nfsd4_layout_seg *seg = &lcp->lc_seg; struct svc_fh *current_fh = &cstate->current_fh; const struct nfsd4_layout_ops *ops; - loff_t new_size = lcp->lc_last_wr + 1; struct inode *inode; struct nfs4_layout_stateid *ls; __be32 nfserr; @@ -2491,13 +2490,21 @@ nfsd4_layoutcommit(struct svc_rqst *rqstp, goto out; inode = d_inode(current_fh->fh_dentry); - nfserr = nfserr_inval; - if (new_size <= seg->offset) - goto out; - if (new_size > seg->offset + seg->length) - goto out; - if (!lcp->lc_newoffset && new_size > i_size_read(inode)) - goto out; + lcp->lc_size_chg = false; + if (lcp->lc_newoffset) { + loff_t new_size = lcp->lc_last_wr + 1; + + nfserr = nfserr_inval; + if (new_size <= seg->offset) + goto out; + if (new_size > seg->offset + seg->length) + goto out; + + if (new_size > i_size_read(inode)) { + lcp->lc_size_chg = true; + lcp->lc_newsize = new_size; + } + } nfserr = nfsd4_preprocess_layout_stateid(rqstp, cstate, &lcp->lc_sid, false, lcp->lc_layout_type, @@ -2513,13 +2520,6 @@ nfsd4_layoutcommit(struct svc_rqst *rqstp, /* LAYOUTCOMMIT does not require any serialization */ mutex_unlock(&ls->ls_mutex); - if (new_size > i_size_read(inode)) { - lcp->lc_size_chg = true; - lcp->lc_newsize = new_size; - } else { - lcp->lc_size_chg = false; - } - nfserr = ops->proc_layoutcommit(inode, rqstp, lcp); nfs4_put_stid(&ls->ls_stid); out:

2 weeks, 3 days

2
4
0 0

FAILED: patch "[PATCH] xfs: fix log CRC mismatches between i386 and other" failed to apply to 5.15-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.15-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(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y git checkout FETCH_HEAD git cherry-pick -x e747883c7d7306acb4d683038d881528fbfbe749 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025101616-negligee-wrongness-0478@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From e747883c7d7306acb4d683038d881528fbfbe749 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig <hch(a)lst.de> Date: Mon, 15 Sep 2025 06:20:30 -0700 Subject: [PATCH] xfs: fix log CRC mismatches between i386 and other architectures When mounting file systems with a log that was dirtied on i386 on other architectures or vice versa, log recovery is unhappy: [ 11.068052] XFS (vdb): Torn write (CRC failure) detected at log block 0x2. Truncating head block from 0xc. This is because the CRCs generated by i386 and other architectures always diff. The reason for that is that sizeof(struct xlog_rec_header) returns different values for i386 vs the rest (324 vs 328), because the struct is not sizeof(uint64_t) aligned, and i386 has odd struct size alignment rules. This issue goes back to commit 13cdc853c519 ("Add log versioning, and new super block field for the log stripe") in the xfs-import tree, which adds log v2 support and the h_size field that causes the unaligned size. At that time it only mattered for the crude debug only log header checksum, but with commit 0e446be44806 ("xfs: add CRC checks to the log") it became a real issue for v5 file system, because now there is a proper CRC, and regular builds actually expect it match. Fix this by allowing checksums with and without the padding. Fixes: 0e446be44806 ("xfs: add CRC checks to the log") Cc: <stable(a)vger.kernel.org> # v3.8 Signed-off-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Carlos Maiolino <cem(a)kernel.org> diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h index a42a83211724..fd00b77af32b 100644 --- a/fs/xfs/libxfs/xfs_log_format.h +++ b/fs/xfs/libxfs/xfs_log_format.h @@ -173,12 +173,40 @@ typedef struct xlog_rec_header { __be32 h_prev_block; /* block number to previous LR : 4 */ __be32 h_num_logops; /* number of log operations in this LR : 4 */ __be32 h_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; - /* new fields */ + + /* fields added by the Linux port: */ __be32 h_fmt; /* format of log record : 4 */ uuid_t h_fs_uuid; /* uuid of FS : 16 */ + + /* fields added for log v2: */ __be32 h_size; /* iclog size : 4 */ + + /* + * When h_size added for log v2 support, it caused structure to have + * a different size on i386 vs all other architectures because the + * sum of the size ofthe member is not aligned by that of the largest + * __be64-sized member, and i386 has really odd struct alignment rules. + * + * Due to the way the log headers are placed out on-disk that alone is + * not a problem becaue the xlog_rec_header always sits alone in a + * BBSIZEs area, and the rest of that area is padded with zeroes. + * But xlog_cksum used to calculate the checksum based on the structure + * size, and thus gives different checksums for i386 vs the rest. + * We now do two checksum validation passes for both sizes to allow + * moving v5 file systems with unclean logs between i386 and other + * (little-endian) architectures. + */ + __u32 h_pad0; } xlog_rec_header_t; +#ifdef __i386__ +#define XLOG_REC_SIZE offsetofend(struct xlog_rec_header, h_size) +#define XLOG_REC_SIZE_OTHER sizeof(struct xlog_rec_header) +#else +#define XLOG_REC_SIZE sizeof(struct xlog_rec_header) +#define XLOG_REC_SIZE_OTHER offsetofend(struct xlog_rec_header, h_size) +#endif /* __i386__ */ + typedef struct xlog_rec_ext_header { __be32 xh_cycle; /* write cycle of log : 4 */ __be32 xh_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; /* : 256 */ diff --git a/fs/xfs/libxfs/xfs_ondisk.h b/fs/xfs/libxfs/xfs_ondisk.h index 5ed44fdf7491..7bfa3242e2c5 100644 --- a/fs/xfs/libxfs/xfs_ondisk.h +++ b/fs/xfs/libxfs/xfs_ondisk.h @@ -174,6 +174,8 @@ xfs_check_ondisk_structs(void) XFS_CHECK_STRUCT_SIZE(struct xfs_rud_log_format, 16); XFS_CHECK_STRUCT_SIZE(struct xfs_map_extent, 32); XFS_CHECK_STRUCT_SIZE(struct xfs_phys_extent, 16); + XFS_CHECK_STRUCT_SIZE(struct xlog_rec_header, 328); + XFS_CHECK_STRUCT_SIZE(struct xlog_rec_ext_header, 260); XFS_CHECK_OFFSET(struct xfs_bui_log_format, bui_extents, 16); XFS_CHECK_OFFSET(struct xfs_cui_log_format, cui_extents, 16); diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 7c590ecdf865..2978de9da38e 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1568,13 +1568,13 @@ xlog_cksum( struct xlog *log, struct xlog_rec_header *rhead, char *dp, - int size) + unsigned int hdrsize, + unsigned int size) { uint32_t crc; /* first generate the crc for the record header ... */ - crc = xfs_start_cksum_update((char *)rhead, - sizeof(struct xlog_rec_header), + crc = xfs_start_cksum_update((char *)rhead, hdrsize, offsetof(struct xlog_rec_header, h_crc)); /* ... then for additional cycle data for v2 logs ... */ @@ -1818,7 +1818,7 @@ xlog_sync( /* calculcate the checksum */ iclog->ic_header.h_crc = xlog_cksum(log, &iclog->ic_header, - iclog->ic_datap, size); + iclog->ic_datap, XLOG_REC_SIZE, size); /* * Intentionally corrupt the log record CRC based on the error injection * frequency, if defined. This facilitates testing log recovery in the diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index a9a7a271c15b..0cfc654d8e87 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -499,8 +499,8 @@ xlog_recover_finish( extern void xlog_recover_cancel(struct xlog *); -extern __le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, - char *dp, int size); +__le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, + char *dp, unsigned int hdrsize, unsigned int size); extern struct kmem_cache *xfs_log_ticket_cache; struct xlog_ticket *xlog_ticket_alloc(struct xlog *log, int unit_bytes, diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 0a4db8efd903..549d60959aee 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -2894,9 +2894,24 @@ xlog_recover_process( int pass, struct list_head *buffer_list) { - __le32 expected_crc = rhead->h_crc, crc; + __le32 expected_crc = rhead->h_crc, crc, other_crc; - crc = xlog_cksum(log, rhead, dp, be32_to_cpu(rhead->h_len)); + crc = xlog_cksum(log, rhead, dp, XLOG_REC_SIZE, + be32_to_cpu(rhead->h_len)); + + /* + * Look at the end of the struct xlog_rec_header definition in + * xfs_log_format.h for the glory details. + */ + if (expected_crc && crc != expected_crc) { + other_crc = xlog_cksum(log, rhead, dp, XLOG_REC_SIZE_OTHER, + be32_to_cpu(rhead->h_len)); + if (other_crc == expected_crc) { + xfs_notice_once(log->l_mp, + "Fixing up incorrect CRC due to padding."); + crc = other_crc; + } + } /* * Nothing else to do if this is a CRC verification pass. Just return

2 weeks, 3 days

2
2
0 0

FAILED: patch "[PATCH] xfs: fix log CRC mismatches between i386 and other" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.1-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(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.1.y git checkout FETCH_HEAD git cherry-pick -x e747883c7d7306acb4d683038d881528fbfbe749 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025101615-delirious-unicycle-3115@gregkh' --subject-prefix 'PATCH 6.1.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From e747883c7d7306acb4d683038d881528fbfbe749 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig <hch(a)lst.de> Date: Mon, 15 Sep 2025 06:20:30 -0700 Subject: [PATCH] xfs: fix log CRC mismatches between i386 and other architectures When mounting file systems with a log that was dirtied on i386 on other architectures or vice versa, log recovery is unhappy: [ 11.068052] XFS (vdb): Torn write (CRC failure) detected at log block 0x2. Truncating head block from 0xc. This is because the CRCs generated by i386 and other architectures always diff. The reason for that is that sizeof(struct xlog_rec_header) returns different values for i386 vs the rest (324 vs 328), because the struct is not sizeof(uint64_t) aligned, and i386 has odd struct size alignment rules. This issue goes back to commit 13cdc853c519 ("Add log versioning, and new super block field for the log stripe") in the xfs-import tree, which adds log v2 support and the h_size field that causes the unaligned size. At that time it only mattered for the crude debug only log header checksum, but with commit 0e446be44806 ("xfs: add CRC checks to the log") it became a real issue for v5 file system, because now there is a proper CRC, and regular builds actually expect it match. Fix this by allowing checksums with and without the padding. Fixes: 0e446be44806 ("xfs: add CRC checks to the log") Cc: <stable(a)vger.kernel.org> # v3.8 Signed-off-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Carlos Maiolino <cem(a)kernel.org> diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h index a42a83211724..fd00b77af32b 100644 --- a/fs/xfs/libxfs/xfs_log_format.h +++ b/fs/xfs/libxfs/xfs_log_format.h @@ -173,12 +173,40 @@ typedef struct xlog_rec_header { __be32 h_prev_block; /* block number to previous LR : 4 */ __be32 h_num_logops; /* number of log operations in this LR : 4 */ __be32 h_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; - /* new fields */ + + /* fields added by the Linux port: */ __be32 h_fmt; /* format of log record : 4 */ uuid_t h_fs_uuid; /* uuid of FS : 16 */ + + /* fields added for log v2: */ __be32 h_size; /* iclog size : 4 */ + + /* + * When h_size added for log v2 support, it caused structure to have + * a different size on i386 vs all other architectures because the + * sum of the size ofthe member is not aligned by that of the largest + * __be64-sized member, and i386 has really odd struct alignment rules. + * + * Due to the way the log headers are placed out on-disk that alone is + * not a problem becaue the xlog_rec_header always sits alone in a + * BBSIZEs area, and the rest of that area is padded with zeroes. + * But xlog_cksum used to calculate the checksum based on the structure + * size, and thus gives different checksums for i386 vs the rest. + * We now do two checksum validation passes for both sizes to allow + * moving v5 file systems with unclean logs between i386 and other + * (little-endian) architectures. + */ + __u32 h_pad0; } xlog_rec_header_t; +#ifdef __i386__ +#define XLOG_REC_SIZE offsetofend(struct xlog_rec_header, h_size) +#define XLOG_REC_SIZE_OTHER sizeof(struct xlog_rec_header) +#else +#define XLOG_REC_SIZE sizeof(struct xlog_rec_header) +#define XLOG_REC_SIZE_OTHER offsetofend(struct xlog_rec_header, h_size) +#endif /* __i386__ */ + typedef struct xlog_rec_ext_header { __be32 xh_cycle; /* write cycle of log : 4 */ __be32 xh_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; /* : 256 */ diff --git a/fs/xfs/libxfs/xfs_ondisk.h b/fs/xfs/libxfs/xfs_ondisk.h index 5ed44fdf7491..7bfa3242e2c5 100644 --- a/fs/xfs/libxfs/xfs_ondisk.h +++ b/fs/xfs/libxfs/xfs_ondisk.h @@ -174,6 +174,8 @@ xfs_check_ondisk_structs(void) XFS_CHECK_STRUCT_SIZE(struct xfs_rud_log_format, 16); XFS_CHECK_STRUCT_SIZE(struct xfs_map_extent, 32); XFS_CHECK_STRUCT_SIZE(struct xfs_phys_extent, 16); + XFS_CHECK_STRUCT_SIZE(struct xlog_rec_header, 328); + XFS_CHECK_STRUCT_SIZE(struct xlog_rec_ext_header, 260); XFS_CHECK_OFFSET(struct xfs_bui_log_format, bui_extents, 16); XFS_CHECK_OFFSET(struct xfs_cui_log_format, cui_extents, 16); diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 7c590ecdf865..2978de9da38e 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1568,13 +1568,13 @@ xlog_cksum( struct xlog *log, struct xlog_rec_header *rhead, char *dp, - int size) + unsigned int hdrsize, + unsigned int size) { uint32_t crc; /* first generate the crc for the record header ... */ - crc = xfs_start_cksum_update((char *)rhead, - sizeof(struct xlog_rec_header), + crc = xfs_start_cksum_update((char *)rhead, hdrsize, offsetof(struct xlog_rec_header, h_crc)); /* ... then for additional cycle data for v2 logs ... */ @@ -1818,7 +1818,7 @@ xlog_sync( /* calculcate the checksum */ iclog->ic_header.h_crc = xlog_cksum(log, &iclog->ic_header, - iclog->ic_datap, size); + iclog->ic_datap, XLOG_REC_SIZE, size); /* * Intentionally corrupt the log record CRC based on the error injection * frequency, if defined. This facilitates testing log recovery in the diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index a9a7a271c15b..0cfc654d8e87 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -499,8 +499,8 @@ xlog_recover_finish( extern void xlog_recover_cancel(struct xlog *); -extern __le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, - char *dp, int size); +__le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, + char *dp, unsigned int hdrsize, unsigned int size); extern struct kmem_cache *xfs_log_ticket_cache; struct xlog_ticket *xlog_ticket_alloc(struct xlog *log, int unit_bytes, diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 0a4db8efd903..549d60959aee 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -2894,9 +2894,24 @@ xlog_recover_process( int pass, struct list_head *buffer_list) { - __le32 expected_crc = rhead->h_crc, crc; + __le32 expected_crc = rhead->h_crc, crc, other_crc; - crc = xlog_cksum(log, rhead, dp, be32_to_cpu(rhead->h_len)); + crc = xlog_cksum(log, rhead, dp, XLOG_REC_SIZE, + be32_to_cpu(rhead->h_len)); + + /* + * Look at the end of the struct xlog_rec_header definition in + * xfs_log_format.h for the glory details. + */ + if (expected_crc && crc != expected_crc) { + other_crc = xlog_cksum(log, rhead, dp, XLOG_REC_SIZE_OTHER, + be32_to_cpu(rhead->h_len)); + if (other_crc == expected_crc) { + xfs_notice_once(log->l_mp, + "Fixing up incorrect CRC due to padding."); + crc = other_crc; + } + } /* * Nothing else to do if this is a CRC verification pass. Just return

2 weeks, 3 days

2
2
0 0

FAILED: patch "[PATCH] xfs: fix log CRC mismatches between i386 and other" failed to apply to 6.6-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.6-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(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.6.y git checkout FETCH_HEAD git cherry-pick -x e747883c7d7306acb4d683038d881528fbfbe749 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025101615-hunter-bonfire-f320@gregkh' --subject-prefix 'PATCH 6.6.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From e747883c7d7306acb4d683038d881528fbfbe749 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig <hch(a)lst.de> Date: Mon, 15 Sep 2025 06:20:30 -0700 Subject: [PATCH] xfs: fix log CRC mismatches between i386 and other architectures When mounting file systems with a log that was dirtied on i386 on other architectures or vice versa, log recovery is unhappy: [ 11.068052] XFS (vdb): Torn write (CRC failure) detected at log block 0x2. Truncating head block from 0xc. This is because the CRCs generated by i386 and other architectures always diff. The reason for that is that sizeof(struct xlog_rec_header) returns different values for i386 vs the rest (324 vs 328), because the struct is not sizeof(uint64_t) aligned, and i386 has odd struct size alignment rules. This issue goes back to commit 13cdc853c519 ("Add log versioning, and new super block field for the log stripe") in the xfs-import tree, which adds log v2 support and the h_size field that causes the unaligned size. At that time it only mattered for the crude debug only log header checksum, but with commit 0e446be44806 ("xfs: add CRC checks to the log") it became a real issue for v5 file system, because now there is a proper CRC, and regular builds actually expect it match. Fix this by allowing checksums with and without the padding. Fixes: 0e446be44806 ("xfs: add CRC checks to the log") Cc: <stable(a)vger.kernel.org> # v3.8 Signed-off-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Carlos Maiolino <cem(a)kernel.org> diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h index a42a83211724..fd00b77af32b 100644 --- a/fs/xfs/libxfs/xfs_log_format.h +++ b/fs/xfs/libxfs/xfs_log_format.h @@ -173,12 +173,40 @@ typedef struct xlog_rec_header { __be32 h_prev_block; /* block number to previous LR : 4 */ __be32 h_num_logops; /* number of log operations in this LR : 4 */ __be32 h_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; - /* new fields */ + + /* fields added by the Linux port: */ __be32 h_fmt; /* format of log record : 4 */ uuid_t h_fs_uuid; /* uuid of FS : 16 */ + + /* fields added for log v2: */ __be32 h_size; /* iclog size : 4 */ + + /* + * When h_size added for log v2 support, it caused structure to have + * a different size on i386 vs all other architectures because the + * sum of the size ofthe member is not aligned by that of the largest + * __be64-sized member, and i386 has really odd struct alignment rules. + * + * Due to the way the log headers are placed out on-disk that alone is + * not a problem becaue the xlog_rec_header always sits alone in a + * BBSIZEs area, and the rest of that area is padded with zeroes. + * But xlog_cksum used to calculate the checksum based on the structure + * size, and thus gives different checksums for i386 vs the rest. + * We now do two checksum validation passes for both sizes to allow + * moving v5 file systems with unclean logs between i386 and other + * (little-endian) architectures. + */ + __u32 h_pad0; } xlog_rec_header_t; +#ifdef __i386__ +#define XLOG_REC_SIZE offsetofend(struct xlog_rec_header, h_size) +#define XLOG_REC_SIZE_OTHER sizeof(struct xlog_rec_header) +#else +#define XLOG_REC_SIZE sizeof(struct xlog_rec_header) +#define XLOG_REC_SIZE_OTHER offsetofend(struct xlog_rec_header, h_size) +#endif /* __i386__ */ + typedef struct xlog_rec_ext_header { __be32 xh_cycle; /* write cycle of log : 4 */ __be32 xh_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; /* : 256 */ diff --git a/fs/xfs/libxfs/xfs_ondisk.h b/fs/xfs/libxfs/xfs_ondisk.h index 5ed44fdf7491..7bfa3242e2c5 100644 --- a/fs/xfs/libxfs/xfs_ondisk.h +++ b/fs/xfs/libxfs/xfs_ondisk.h @@ -174,6 +174,8 @@ xfs_check_ondisk_structs(void) XFS_CHECK_STRUCT_SIZE(struct xfs_rud_log_format, 16); XFS_CHECK_STRUCT_SIZE(struct xfs_map_extent, 32); XFS_CHECK_STRUCT_SIZE(struct xfs_phys_extent, 16); + XFS_CHECK_STRUCT_SIZE(struct xlog_rec_header, 328); + XFS_CHECK_STRUCT_SIZE(struct xlog_rec_ext_header, 260); XFS_CHECK_OFFSET(struct xfs_bui_log_format, bui_extents, 16); XFS_CHECK_OFFSET(struct xfs_cui_log_format, cui_extents, 16); diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 7c590ecdf865..2978de9da38e 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1568,13 +1568,13 @@ xlog_cksum( struct xlog *log, struct xlog_rec_header *rhead, char *dp, - int size) + unsigned int hdrsize, + unsigned int size) { uint32_t crc; /* first generate the crc for the record header ... */ - crc = xfs_start_cksum_update((char *)rhead, - sizeof(struct xlog_rec_header), + crc = xfs_start_cksum_update((char *)rhead, hdrsize, offsetof(struct xlog_rec_header, h_crc)); /* ... then for additional cycle data for v2 logs ... */ @@ -1818,7 +1818,7 @@ xlog_sync( /* calculcate the checksum */ iclog->ic_header.h_crc = xlog_cksum(log, &iclog->ic_header, - iclog->ic_datap, size); + iclog->ic_datap, XLOG_REC_SIZE, size); /* * Intentionally corrupt the log record CRC based on the error injection * frequency, if defined. This facilitates testing log recovery in the diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index a9a7a271c15b..0cfc654d8e87 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -499,8 +499,8 @@ xlog_recover_finish( extern void xlog_recover_cancel(struct xlog *); -extern __le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, - char *dp, int size); +__le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, + char *dp, unsigned int hdrsize, unsigned int size); extern struct kmem_cache *xfs_log_ticket_cache; struct xlog_ticket *xlog_ticket_alloc(struct xlog *log, int unit_bytes, diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 0a4db8efd903..549d60959aee 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -2894,9 +2894,24 @@ xlog_recover_process( int pass, struct list_head *buffer_list) { - __le32 expected_crc = rhead->h_crc, crc; + __le32 expected_crc = rhead->h_crc, crc, other_crc; - crc = xlog_cksum(log, rhead, dp, be32_to_cpu(rhead->h_len)); + crc = xlog_cksum(log, rhead, dp, XLOG_REC_SIZE, + be32_to_cpu(rhead->h_len)); + + /* + * Look at the end of the struct xlog_rec_header definition in + * xfs_log_format.h for the glory details. + */ + if (expected_crc && crc != expected_crc) { + other_crc = xlog_cksum(log, rhead, dp, XLOG_REC_SIZE_OTHER, + be32_to_cpu(rhead->h_len)); + if (other_crc == expected_crc) { + xfs_notice_once(log->l_mp, + "Fixing up incorrect CRC due to padding."); + crc = other_crc; + } + } /* * Nothing else to do if this is a CRC verification pass. Just return

2 weeks, 3 days

2
2
0 0

FAILED: patch "[PATCH] xfs: fix log CRC mismatches between i386 and other" failed to apply to 6.12-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.12-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(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.12.y git checkout FETCH_HEAD git cherry-pick -x e747883c7d7306acb4d683038d881528fbfbe749 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025101614-down-dicing-0c7e@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From e747883c7d7306acb4d683038d881528fbfbe749 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig <hch(a)lst.de> Date: Mon, 15 Sep 2025 06:20:30 -0700 Subject: [PATCH] xfs: fix log CRC mismatches between i386 and other architectures When mounting file systems with a log that was dirtied on i386 on other architectures or vice versa, log recovery is unhappy: [ 11.068052] XFS (vdb): Torn write (CRC failure) detected at log block 0x2. Truncating head block from 0xc. This is because the CRCs generated by i386 and other architectures always diff. The reason for that is that sizeof(struct xlog_rec_header) returns different values for i386 vs the rest (324 vs 328), because the struct is not sizeof(uint64_t) aligned, and i386 has odd struct size alignment rules. This issue goes back to commit 13cdc853c519 ("Add log versioning, and new super block field for the log stripe") in the xfs-import tree, which adds log v2 support and the h_size field that causes the unaligned size. At that time it only mattered for the crude debug only log header checksum, but with commit 0e446be44806 ("xfs: add CRC checks to the log") it became a real issue for v5 file system, because now there is a proper CRC, and regular builds actually expect it match. Fix this by allowing checksums with and without the padding. Fixes: 0e446be44806 ("xfs: add CRC checks to the log") Cc: <stable(a)vger.kernel.org> # v3.8 Signed-off-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Carlos Maiolino <cem(a)kernel.org> diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h index a42a83211724..fd00b77af32b 100644 --- a/fs/xfs/libxfs/xfs_log_format.h +++ b/fs/xfs/libxfs/xfs_log_format.h @@ -173,12 +173,40 @@ typedef struct xlog_rec_header { __be32 h_prev_block; /* block number to previous LR : 4 */ __be32 h_num_logops; /* number of log operations in this LR : 4 */ __be32 h_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; - /* new fields */ + + /* fields added by the Linux port: */ __be32 h_fmt; /* format of log record : 4 */ uuid_t h_fs_uuid; /* uuid of FS : 16 */ + + /* fields added for log v2: */ __be32 h_size; /* iclog size : 4 */ + + /* + * When h_size added for log v2 support, it caused structure to have + * a different size on i386 vs all other architectures because the + * sum of the size ofthe member is not aligned by that of the largest + * __be64-sized member, and i386 has really odd struct alignment rules. + * + * Due to the way the log headers are placed out on-disk that alone is + * not a problem becaue the xlog_rec_header always sits alone in a + * BBSIZEs area, and the rest of that area is padded with zeroes. + * But xlog_cksum used to calculate the checksum based on the structure + * size, and thus gives different checksums for i386 vs the rest. + * We now do two checksum validation passes for both sizes to allow + * moving v5 file systems with unclean logs between i386 and other + * (little-endian) architectures. + */ + __u32 h_pad0; } xlog_rec_header_t; +#ifdef __i386__ +#define XLOG_REC_SIZE offsetofend(struct xlog_rec_header, h_size) +#define XLOG_REC_SIZE_OTHER sizeof(struct xlog_rec_header) +#else +#define XLOG_REC_SIZE sizeof(struct xlog_rec_header) +#define XLOG_REC_SIZE_OTHER offsetofend(struct xlog_rec_header, h_size) +#endif /* __i386__ */ + typedef struct xlog_rec_ext_header { __be32 xh_cycle; /* write cycle of log : 4 */ __be32 xh_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; /* : 256 */ diff --git a/fs/xfs/libxfs/xfs_ondisk.h b/fs/xfs/libxfs/xfs_ondisk.h index 5ed44fdf7491..7bfa3242e2c5 100644 --- a/fs/xfs/libxfs/xfs_ondisk.h +++ b/fs/xfs/libxfs/xfs_ondisk.h @@ -174,6 +174,8 @@ xfs_check_ondisk_structs(void) XFS_CHECK_STRUCT_SIZE(struct xfs_rud_log_format, 16); XFS_CHECK_STRUCT_SIZE(struct xfs_map_extent, 32); XFS_CHECK_STRUCT_SIZE(struct xfs_phys_extent, 16); + XFS_CHECK_STRUCT_SIZE(struct xlog_rec_header, 328); + XFS_CHECK_STRUCT_SIZE(struct xlog_rec_ext_header, 260); XFS_CHECK_OFFSET(struct xfs_bui_log_format, bui_extents, 16); XFS_CHECK_OFFSET(struct xfs_cui_log_format, cui_extents, 16); diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 7c590ecdf865..2978de9da38e 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1568,13 +1568,13 @@ xlog_cksum( struct xlog *log, struct xlog_rec_header *rhead, char *dp, - int size) + unsigned int hdrsize, + unsigned int size) { uint32_t crc; /* first generate the crc for the record header ... */ - crc = xfs_start_cksum_update((char *)rhead, - sizeof(struct xlog_rec_header), + crc = xfs_start_cksum_update((char *)rhead, hdrsize, offsetof(struct xlog_rec_header, h_crc)); /* ... then for additional cycle data for v2 logs ... */ @@ -1818,7 +1818,7 @@ xlog_sync( /* calculcate the checksum */ iclog->ic_header.h_crc = xlog_cksum(log, &iclog->ic_header, - iclog->ic_datap, size); + iclog->ic_datap, XLOG_REC_SIZE, size); /* * Intentionally corrupt the log record CRC based on the error injection * frequency, if defined. This facilitates testing log recovery in the diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index a9a7a271c15b..0cfc654d8e87 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -499,8 +499,8 @@ xlog_recover_finish( extern void xlog_recover_cancel(struct xlog *); -extern __le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, - char *dp, int size); +__le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, + char *dp, unsigned int hdrsize, unsigned int size); extern struct kmem_cache *xfs_log_ticket_cache; struct xlog_ticket *xlog_ticket_alloc(struct xlog *log, int unit_bytes, diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 0a4db8efd903..549d60959aee 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -2894,9 +2894,24 @@ xlog_recover_process( int pass, struct list_head *buffer_list) { - __le32 expected_crc = rhead->h_crc, crc; + __le32 expected_crc = rhead->h_crc, crc, other_crc; - crc = xlog_cksum(log, rhead, dp, be32_to_cpu(rhead->h_len)); + crc = xlog_cksum(log, rhead, dp, XLOG_REC_SIZE, + be32_to_cpu(rhead->h_len)); + + /* + * Look at the end of the struct xlog_rec_header definition in + * xfs_log_format.h for the glory details. + */ + if (expected_crc && crc != expected_crc) { + other_crc = xlog_cksum(log, rhead, dp, XLOG_REC_SIZE_OTHER, + be32_to_cpu(rhead->h_len)); + if (other_crc == expected_crc) { + xfs_notice_once(log->l_mp, + "Fixing up incorrect CRC due to padding."); + crc = other_crc; + } + } /* * Nothing else to do if this is a CRC verification pass. Just return

2 weeks, 3 days

2
2
0 0

FAILED: patch "[PATCH] xfs: fix log CRC mismatches between i386 and other" failed to apply to 6.17-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.17-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(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.17.y git checkout FETCH_HEAD git cherry-pick -x e747883c7d7306acb4d683038d881528fbfbe749 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025101613-unworldly-lumpiness-b4a6@gregkh' --subject-prefix 'PATCH 6.17.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From e747883c7d7306acb4d683038d881528fbfbe749 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig <hch(a)lst.de> Date: Mon, 15 Sep 2025 06:20:30 -0700 Subject: [PATCH] xfs: fix log CRC mismatches between i386 and other architectures When mounting file systems with a log that was dirtied on i386 on other architectures or vice versa, log recovery is unhappy: [ 11.068052] XFS (vdb): Torn write (CRC failure) detected at log block 0x2. Truncating head block from 0xc. This is because the CRCs generated by i386 and other architectures always diff. The reason for that is that sizeof(struct xlog_rec_header) returns different values for i386 vs the rest (324 vs 328), because the struct is not sizeof(uint64_t) aligned, and i386 has odd struct size alignment rules. This issue goes back to commit 13cdc853c519 ("Add log versioning, and new super block field for the log stripe") in the xfs-import tree, which adds log v2 support and the h_size field that causes the unaligned size. At that time it only mattered for the crude debug only log header checksum, but with commit 0e446be44806 ("xfs: add CRC checks to the log") it became a real issue for v5 file system, because now there is a proper CRC, and regular builds actually expect it match. Fix this by allowing checksums with and without the padding. Fixes: 0e446be44806 ("xfs: add CRC checks to the log") Cc: <stable(a)vger.kernel.org> # v3.8 Signed-off-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Carlos Maiolino <cem(a)kernel.org> diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h index a42a83211724..fd00b77af32b 100644 --- a/fs/xfs/libxfs/xfs_log_format.h +++ b/fs/xfs/libxfs/xfs_log_format.h @@ -173,12 +173,40 @@ typedef struct xlog_rec_header { __be32 h_prev_block; /* block number to previous LR : 4 */ __be32 h_num_logops; /* number of log operations in this LR : 4 */ __be32 h_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; - /* new fields */ + + /* fields added by the Linux port: */ __be32 h_fmt; /* format of log record : 4 */ uuid_t h_fs_uuid; /* uuid of FS : 16 */ + + /* fields added for log v2: */ __be32 h_size; /* iclog size : 4 */ + + /* + * When h_size added for log v2 support, it caused structure to have + * a different size on i386 vs all other architectures because the + * sum of the size ofthe member is not aligned by that of the largest + * __be64-sized member, and i386 has really odd struct alignment rules. + * + * Due to the way the log headers are placed out on-disk that alone is + * not a problem becaue the xlog_rec_header always sits alone in a + * BBSIZEs area, and the rest of that area is padded with zeroes. + * But xlog_cksum used to calculate the checksum based on the structure + * size, and thus gives different checksums for i386 vs the rest. + * We now do two checksum validation passes for both sizes to allow + * moving v5 file systems with unclean logs between i386 and other + * (little-endian) architectures. + */ + __u32 h_pad0; } xlog_rec_header_t; +#ifdef __i386__ +#define XLOG_REC_SIZE offsetofend(struct xlog_rec_header, h_size) +#define XLOG_REC_SIZE_OTHER sizeof(struct xlog_rec_header) +#else +#define XLOG_REC_SIZE sizeof(struct xlog_rec_header) +#define XLOG_REC_SIZE_OTHER offsetofend(struct xlog_rec_header, h_size) +#endif /* __i386__ */ + typedef struct xlog_rec_ext_header { __be32 xh_cycle; /* write cycle of log : 4 */ __be32 xh_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE]; /* : 256 */ diff --git a/fs/xfs/libxfs/xfs_ondisk.h b/fs/xfs/libxfs/xfs_ondisk.h index 5ed44fdf7491..7bfa3242e2c5 100644 --- a/fs/xfs/libxfs/xfs_ondisk.h +++ b/fs/xfs/libxfs/xfs_ondisk.h @@ -174,6 +174,8 @@ xfs_check_ondisk_structs(void) XFS_CHECK_STRUCT_SIZE(struct xfs_rud_log_format, 16); XFS_CHECK_STRUCT_SIZE(struct xfs_map_extent, 32); XFS_CHECK_STRUCT_SIZE(struct xfs_phys_extent, 16); + XFS_CHECK_STRUCT_SIZE(struct xlog_rec_header, 328); + XFS_CHECK_STRUCT_SIZE(struct xlog_rec_ext_header, 260); XFS_CHECK_OFFSET(struct xfs_bui_log_format, bui_extents, 16); XFS_CHECK_OFFSET(struct xfs_cui_log_format, cui_extents, 16); diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 7c590ecdf865..2978de9da38e 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1568,13 +1568,13 @@ xlog_cksum( struct xlog *log, struct xlog_rec_header *rhead, char *dp, - int size) + unsigned int hdrsize, + unsigned int size) { uint32_t crc; /* first generate the crc for the record header ... */ - crc = xfs_start_cksum_update((char *)rhead, - sizeof(struct xlog_rec_header), + crc = xfs_start_cksum_update((char *)rhead, hdrsize, offsetof(struct xlog_rec_header, h_crc)); /* ... then for additional cycle data for v2 logs ... */ @@ -1818,7 +1818,7 @@ xlog_sync( /* calculcate the checksum */ iclog->ic_header.h_crc = xlog_cksum(log, &iclog->ic_header, - iclog->ic_datap, size); + iclog->ic_datap, XLOG_REC_SIZE, size); /* * Intentionally corrupt the log record CRC based on the error injection * frequency, if defined. This facilitates testing log recovery in the diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index a9a7a271c15b..0cfc654d8e87 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -499,8 +499,8 @@ xlog_recover_finish( extern void xlog_recover_cancel(struct xlog *); -extern __le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, - char *dp, int size); +__le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, + char *dp, unsigned int hdrsize, unsigned int size); extern struct kmem_cache *xfs_log_ticket_cache; struct xlog_ticket *xlog_ticket_alloc(struct xlog *log, int unit_bytes, diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 0a4db8efd903..549d60959aee 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -2894,9 +2894,24 @@ xlog_recover_process( int pass, struct list_head *buffer_list) { - __le32 expected_crc = rhead->h_crc, crc; + __le32 expected_crc = rhead->h_crc, crc, other_crc; - crc = xlog_cksum(log, rhead, dp, be32_to_cpu(rhead->h_len)); + crc = xlog_cksum(log, rhead, dp, XLOG_REC_SIZE, + be32_to_cpu(rhead->h_len)); + + /* + * Look at the end of the struct xlog_rec_header definition in + * xfs_log_format.h for the glory details. + */ + if (expected_crc && crc != expected_crc) { + other_crc = xlog_cksum(log, rhead, dp, XLOG_REC_SIZE_OTHER, + be32_to_cpu(rhead->h_len)); + if (other_crc == expected_crc) { + xfs_notice_once(log->l_mp, + "Fixing up incorrect CRC due to padding."); + crc = other_crc; + } + } /* * Nothing else to do if this is a CRC verification pass. Just return

2 weeks, 3 days

2
2
0 0

[PATCH 6.6 1/2] block: fix race between set_blocksize and read paths

by Mahmoud Adam

From: "Darrick J. Wong" <djwong(a)kernel.org> commit c0e473a0d226479e8e925d5ba93f751d8df628e9 upstream. With the new large sector size support, it's now the case that set_blocksize can change i_blksize and the folio order in a manner that conflicts with a concurrent reader and causes a kernel crash. Specifically, let's say that udev-worker calls libblkid to detect the labels on a block device. The read call can create an order-0 folio to read the first 4096 bytes from the disk. But then udev is preempted. Next, someone tries to mount an 8k-sectorsize filesystem from the same block device. The filesystem calls set_blksize, which sets i_blksize to 8192 and the minimum folio order to 1. Now udev resumes, still holding the order-0 folio it allocated. It then tries to schedule a read bio and do_mpage_readahead tries to create bufferheads for the folio. Unfortunately, blocks_per_folio == 0 because the page size is 4096 but the blocksize is 8192 so no bufferheads are attached and the bh walk never sets bdev. We then submit the bio with a NULL block device and crash. Therefore, truncate the page cache after flushing but before updating i_blksize. However, that's not enough -- we also need to lock out file IO and page faults during the update. Take both the i_rwsem and the invalidate_lock in exclusive mode for invalidations, and in shared mode for read/write operations. I don't know if this is the correct fix, but xfs/259 found it. Signed-off-by: Darrick J. Wong <djwong(a)kernel.org> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Reviewed-by: Luis Chamberlain <mcgrof(a)kernel.org> Tested-by: Shin'ichiro Kawasaki <shinichiro.kawasaki(a)wdc.com> Link: https://lore.kernel.org/r/174543795699.4139148.2086129139322431423.stgit@fr… Signed-off-by: Jens Axboe <axboe(a)kernel.dk> [use bdev->bd_inode instead] Signed-off-by: Mahmoud Adam <mngyadam(a)amazon.de> --- Fixes CVE-2025-38073. block/bdev.c | 17 +++++++++++++++++ block/blk-zoned.c | 5 ++++- block/fops.c | 16 ++++++++++++++++ block/ioctl.c | 6 ++++++ 4 files changed, 43 insertions(+), 1 deletion(-) diff --git a/block/bdev.c b/block/bdev.c index 5a54977518eeae..a8357b72a27b86 100644 --- a/block/bdev.c +++ b/block/bdev.c @@ -147,9 +147,26 @@ int set_blocksize(struct block_device *bdev, int size) /* Don't change the size if it is same as current */ if (bdev->bd_inode->i_blkbits != blksize_bits(size)) { + /* + * Flush and truncate the pagecache before we reconfigure the + * mapping geometry because folio sizes are variable now. If a + * reader has already allocated a folio whose size is smaller + * than the new min_order but invokes readahead after the new + * min_order becomes visible, readahead will think there are + * "zero" blocks per folio and crash. Take the inode and + * invalidation locks to avoid racing with + * read/write/fallocate. + */ + inode_lock(bdev->bd_inode); + filemap_invalidate_lock(bdev->bd_inode->i_mapping); + sync_blockdev(bdev); + kill_bdev(bdev); + bdev->bd_inode->i_blkbits = blksize_bits(size); kill_bdev(bdev); + filemap_invalidate_unlock(bdev->bd_inode->i_mapping); + inode_unlock(bdev->bd_inode); } return 0; } diff --git a/block/blk-zoned.c b/block/blk-zoned.c index 619ee41a51cc8c..644bfa1f6753ea 100644 --- a/block/blk-zoned.c +++ b/block/blk-zoned.c @@ -401,6 +401,7 @@ int blkdev_zone_mgmt_ioctl(struct block_device *bdev, blk_mode_t mode, op = REQ_OP_ZONE_RESET; /* Invalidate the page cache, including dirty pages. */ + inode_lock(bdev->bd_inode); filemap_invalidate_lock(bdev->bd_inode->i_mapping); ret = blkdev_truncate_zone_range(bdev, mode, &zrange); if (ret) @@ -423,8 +424,10 @@ int blkdev_zone_mgmt_ioctl(struct block_device *bdev, blk_mode_t mode, GFP_KERNEL); fail: - if (cmd == BLKRESETZONE) + if (cmd == BLKRESETZONE) { filemap_invalidate_unlock(bdev->bd_inode->i_mapping); + inode_unlock(bdev->bd_inode); + } return ret; } diff --git a/block/fops.c b/block/fops.c index 7c257eb3564d0c..088143fa9ac9e1 100644 --- a/block/fops.c +++ b/block/fops.c @@ -681,7 +681,14 @@ static ssize_t blkdev_write_iter(struct kiocb *iocb, struct iov_iter *from) ret = direct_write_fallback(iocb, from, ret, blkdev_buffered_write(iocb, from)); } else { + /* + * Take i_rwsem and invalidate_lock to avoid racing with + * set_blocksize changing i_blkbits/folio order and punching + * out the pagecache. + */ + inode_lock_shared(bd_inode); ret = blkdev_buffered_write(iocb, from); + inode_unlock_shared(bd_inode); } if (ret > 0) @@ -693,6 +700,7 @@ static ssize_t blkdev_write_iter(struct kiocb *iocb, struct iov_iter *from) static ssize_t blkdev_read_iter(struct kiocb *iocb, struct iov_iter *to) { struct block_device *bdev = I_BDEV(iocb->ki_filp->f_mapping->host); + struct inode *bd_inode = bdev->bd_inode; loff_t size = bdev_nr_bytes(bdev); loff_t pos = iocb->ki_pos; size_t shorted = 0; @@ -728,7 +736,13 @@ static ssize_t blkdev_read_iter(struct kiocb *iocb, struct iov_iter *to) goto reexpand; } + /* + * Take i_rwsem and invalidate_lock to avoid racing with set_blocksize + * changing i_blkbits/folio order and punching out the pagecache. + */ + inode_lock_shared(bd_inode); ret = filemap_read(iocb, to, ret); + inode_unlock_shared(bd_inode); reexpand: if (unlikely(shorted)) @@ -771,6 +785,7 @@ static long blkdev_fallocate(struct file *file, int mode, loff_t start, if ((start | len) & (bdev_logical_block_size(bdev) - 1)) return -EINVAL; + inode_lock(inode); filemap_invalidate_lock(inode->i_mapping); /* @@ -811,6 +826,7 @@ static long blkdev_fallocate(struct file *file, int mode, loff_t start, fail: filemap_invalidate_unlock(inode->i_mapping); + inode_unlock(inode); return error; } diff --git a/block/ioctl.c b/block/ioctl.c index 231537f79a8cb4..024767fa1e52d5 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -114,6 +114,7 @@ static int blk_ioctl_discard(struct block_device *bdev, blk_mode_t mode, end > bdev_nr_bytes(bdev)) return -EINVAL; + inode_lock(inode); filemap_invalidate_lock(inode->i_mapping); err = truncate_bdev_range(bdev, mode, start, end - 1); if (err) @@ -121,6 +122,7 @@ static int blk_ioctl_discard(struct block_device *bdev, blk_mode_t mode, err = blkdev_issue_discard(bdev, start >> 9, len >> 9, GFP_KERNEL); fail: filemap_invalidate_unlock(inode->i_mapping); + inode_unlock(inode); return err; } @@ -146,12 +148,14 @@ static int blk_ioctl_secure_erase(struct block_device *bdev, blk_mode_t mode, end > bdev_nr_bytes(bdev)) return -EINVAL; + inode_lock(bdev->bd_inode); filemap_invalidate_lock(bdev->bd_inode->i_mapping); err = truncate_bdev_range(bdev, mode, start, end - 1); if (!err) err = blkdev_issue_secure_erase(bdev, start >> 9, len >> 9, GFP_KERNEL); filemap_invalidate_unlock(bdev->bd_inode->i_mapping); + inode_unlock(bdev->bd_inode); return err; } @@ -184,6 +188,7 @@ static int blk_ioctl_zeroout(struct block_device *bdev, blk_mode_t mode, return -EINVAL; /* Invalidate the page cache, including dirty pages */ + inode_lock(inode); filemap_invalidate_lock(inode->i_mapping); err = truncate_bdev_range(bdev, mode, start, end); if (err) @@ -194,6 +199,7 @@ static int blk_ioctl_zeroout(struct block_device *bdev, blk_mode_t mode, fail: filemap_invalidate_unlock(inode->i_mapping); + inode_unlock(inode); return err; } -- 2.47.3 Amazon Web Services Development Center Germany GmbH Tamara-Danz-Str. 13 10243 Berlin Geschaeftsfuehrung: Christian Schlaeger Eingetragen am Amtsgericht Charlottenburg unter HRB 257764 B Sitz: Berlin Ust-ID: DE 365 538 597

2 weeks, 3 days

1
1
0 0

FAILED: patch "[PATCH] Revert "io_uring/rw: drop -EOPNOTSUPP check in" failed to apply to 6.12-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.12-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(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.12.y git checkout FETCH_HEAD git cherry-pick -x 927069c4ac2cd1a37efa468596fb5b8f86db9df0 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025102039-bonelike-vocation-0372@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 927069c4ac2cd1a37efa468596fb5b8f86db9df0 Mon Sep 17 00:00:00 2001 From: Jens Axboe <axboe(a)kernel.dk> Date: Mon, 13 Oct 2025 12:05:31 -0600 Subject: [PATCH] Revert "io_uring/rw: drop -EOPNOTSUPP check in __io_complete_rw_common()" This reverts commit 90bfb28d5fa8127a113a140c9791ea0b40ab156a. Kevin reports that this commit causes an issue for him with LVM snapshots, most likely because of turning off NOWAIT support while a snapshot is being created. This makes -EOPNOTSUPP bubble back through the completion handler, where io_uring read/write handling should just retry it. Reinstate the previous check removed by the referenced commit. Cc: stable(a)vger.kernel.org Fixes: 90bfb28d5fa8 ("io_uring/rw: drop -EOPNOTSUPP check in __io_complete_rw_common()") Reported-by: Salvatore Bonaccorso <carnil(a)debian.org> Reported-by: Kevin Lumik <kevin(a)xf.ee> Link: https://lore.kernel.org/io-uring/cceb723c-051b-4de2-9a4c-4aa82e1619ee@kerne… Signed-off-by: Jens Axboe <axboe(a)kernel.dk> diff --git a/io_uring/rw.c b/io_uring/rw.c index 08882648d569..a0f9d2021e3f 100644 --- a/io_uring/rw.c +++ b/io_uring/rw.c @@ -542,7 +542,7 @@ static void __io_complete_rw_common(struct io_kiocb *req, long res) { if (res == req->cqe.res) return; - if (res == -EAGAIN && io_rw_should_reissue(req)) { + if ((res == -EOPNOTSUPP || res == -EAGAIN) && io_rw_should_reissue(req)) { req->flags |= REQ_F_REISSUE | REQ_F_BL_NO_RECYCLE; } else { req_set_fail(req);

2 weeks, 3 days

3
2
0 0

[PATCH v3 1/5] mm: fix off-by-one error in VMA count limit checks

by Kalesh Singh

The VMA count limit check in do_mmap() and do_brk_flags() uses a strict inequality (>), which allows a process's VMA count to exceed the configured sysctl_max_map_count limit by one. A process with mm->map_count == sysctl_max_map_count will incorrectly pass this check and then exceed the limit upon allocation of a new VMA when its map_count is incremented. Other VMA allocation paths, such as split_vma(), already use the correct, inclusive (>=) comparison. Fix this bug by changing the comparison to be inclusive in do_mmap() and do_brk_flags(), bringing them in line with the correct behavior of other allocation paths. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: <stable(a)vger.kernel.org> Cc: Andrew Morton <akpm(a)linux-foundation.org> Cc: David Hildenbrand <david(a)redhat.com> Cc: "Liam R. Howlett" <Liam.Howlett(a)oracle.com> Cc: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Cc: Mike Rapoport <rppt(a)kernel.org> Cc: Minchan Kim <minchan(a)kernel.org> Cc: Pedro Falcato <pfalcato(a)suse.de> Reviewed-by: David Hildenbrand <david(a)redhat.com> Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Reviewed-by: Pedro Falcato <pfalcato(a)suse.de> Acked-by: SeongJae Park <sj(a)kernel.org> Signed-off-by: Kalesh Singh <kaleshsingh(a)google.com> --- Changes in v3: - Collect Reviewed-by and Acked-by tags. Changes in v2: - Fix mmap check, per Pedro mm/mmap.c | 2 +- mm/vma.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/mmap.c b/mm/mmap.c index 644f02071a41..da2cbdc0f87b 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -374,7 +374,7 @@ unsigned long do_mmap(struct file *file, unsigned long addr, return -EOVERFLOW; /* Too many mappings? */ - if (mm->map_count > sysctl_max_map_count) + if (mm->map_count >= sysctl_max_map_count) return -ENOMEM; /* diff --git a/mm/vma.c b/mm/vma.c index a2e1ae954662..fba68f13e628 100644 --- a/mm/vma.c +++ b/mm/vma.c @@ -2797,7 +2797,7 @@ int do_brk_flags(struct vma_iterator *vmi, struct vm_area_struct *vma, if (!may_expand_vm(mm, vm_flags, len >> PAGE_SHIFT)) return -ENOMEM; - if (mm->map_count > sysctl_max_map_count) + if (mm->map_count >= sysctl_max_map_count) return -ENOMEM; if (security_vm_enough_memory_mm(mm, len >> PAGE_SHIFT)) -- 2.51.0.760.g7b8bcc2412-goog

2 weeks, 3 days

5
11
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror October 2025