| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <20250504085923.1895402-7-john.g.garry@oracle.com>
Date: Sun, 4 May 2025 08:59:13 +0000
From: John Garry <john.g.garry@...cle.com>
To: brauner@...nel.org, djwong@...nel.org, hch@....de, viro@...iv.linux.org.uk,
jack@...e.cz, cem@...nel.org
Cc: linux-fsdevel@...r.kernel.org, dchinner@...hat.com,
linux-xfs@...r.kernel.org, linux-kernel@...r.kernel.org,
ojaswin@...ux.ibm.com, ritesh.list@...il.com,
martin.petersen@...cle.com, linux-ext4@...r.kernel.org,
linux-block@...r.kernel.org, catherine.hoang@...cle.com,
linux-api@...r.kernel.org, John Garry <john.g.garry@...cle.com>
Subject: [PATCH v11 06/16] xfs: ignore HW which cannot atomic write a single block
From: "Darrick J. Wong" <djwong@...nel.org>
Currently only HW which can write at least 1x block is supported.
For supporting atomic writes > 1x block, a CoW-based method will also be
used and this will not be resticted to using HW which can write >= 1x
block.
However for deciding if HW-based atomic writes can be used, we need to
start adding checks for write length < HW min, which complicates the
code. Indeed, a statx field similar to unit_max_opt should also be
added for this minimum, which is undesirable.
HW which can only write > 1x blocks would be uncommon and quite weird,
so let's just not support it.
Signed-off-by: "Darrick J. Wong" <djwong@...nel.org>
Signed-off-by: John Garry <john.g.garry@...cle.com>
---
fs/xfs/xfs_buf.c | 44 ++++++++++++++++++++++++++++++++++++--------
fs/xfs/xfs_buf.h | 4 ++--
fs/xfs/xfs_inode.h | 14 ++------------
fs/xfs/xfs_super.c | 6 +++---
4 files changed, 43 insertions(+), 25 deletions(-)
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index 292891d6ff69..770dc4ca79e4 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -1714,13 +1714,45 @@ xfs_free_buftarg(
kfree(btp);
}
+/*
+ * Configure this buffer target for hardware-assisted atomic writes if the
+ * underlying block device supports is congruent with the filesystem geometry.
+ */
+static inline void
+xfs_configure_buftarg_atomic_writes(
+ struct xfs_buftarg *btp)
+{
+ struct xfs_mount *mp = btp->bt_mount;
+ unsigned int min_bytes, max_bytes;
+
+ min_bytes = bdev_atomic_write_unit_min_bytes(btp->bt_bdev);
+ max_bytes = bdev_atomic_write_unit_max_bytes(btp->bt_bdev);
+
+ /*
+ * Ignore atomic write geometry that is nonsense or doesn't even cover
+ * a single fsblock.
+ */
+ if (min_bytes > max_bytes ||
+ min_bytes > mp->m_sb.sb_blocksize ||
+ max_bytes < mp->m_sb.sb_blocksize) {
+ min_bytes = 0;
+ max_bytes = 0;
+ }
+
+ btp->bt_bdev_awu_min = min_bytes;
+ btp->bt_bdev_awu_max = max_bytes;
+}
+
+/* Configure a buffer target that abstracts a block device. */
int
-xfs_setsize_buftarg(
+xfs_configure_buftarg(
struct xfs_buftarg *btp,
unsigned int sectorsize)
{
int error;
+ ASSERT(btp->bt_bdev != NULL);
+
/* Set up metadata sector size info */
btp->bt_meta_sectorsize = sectorsize;
btp->bt_meta_sectormask = sectorsize - 1;
@@ -1733,6 +1765,9 @@ xfs_setsize_buftarg(
return -EINVAL;
}
+ if (bdev_can_atomic_write(btp->bt_bdev))
+ xfs_configure_buftarg_atomic_writes(btp);
+
return 0;
}
@@ -1795,13 +1830,6 @@ xfs_alloc_buftarg(
btp->bt_daxdev = fs_dax_get_by_bdev(btp->bt_bdev, &btp->bt_dax_part_off,
mp, ops);
- if (bdev_can_atomic_write(btp->bt_bdev)) {
- btp->bt_bdev_awu_min = bdev_atomic_write_unit_min_bytes(
- btp->bt_bdev);
- btp->bt_bdev_awu_max = bdev_atomic_write_unit_max_bytes(
- btp->bt_bdev);
- }
-
/*
* When allocating the buftargs we have not yet read the super block and
* thus don't know the file system sector size yet.
diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h
index 132210705602..7759fe35d93e 100644
--- a/fs/xfs/xfs_buf.h
+++ b/fs/xfs/xfs_buf.h
@@ -112,7 +112,7 @@ struct xfs_buftarg {
struct percpu_counter bt_readahead_count;
struct ratelimit_state bt_ioerror_rl;
- /* Atomic write unit values */
+ /* Atomic write unit values, bytes */
unsigned int bt_bdev_awu_min;
unsigned int bt_bdev_awu_max;
@@ -374,7 +374,7 @@ struct xfs_buftarg *xfs_alloc_buftarg(struct xfs_mount *mp,
extern void xfs_free_buftarg(struct xfs_buftarg *);
extern void xfs_buftarg_wait(struct xfs_buftarg *);
extern void xfs_buftarg_drain(struct xfs_buftarg *);
-extern int xfs_setsize_buftarg(struct xfs_buftarg *, unsigned int);
+extern int xfs_configure_buftarg(struct xfs_buftarg *, unsigned int);
#define xfs_getsize_buftarg(buftarg) block_size((buftarg)->bt_bdev)
#define xfs_readonly_buftarg(buftarg) bdev_read_only((buftarg)->bt_bdev)
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index bdbbff0d8d99..d7e2b902ef5c 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -356,19 +356,9 @@ static inline bool xfs_inode_has_bigrtalloc(const struct xfs_inode *ip)
(XFS_IS_REALTIME_INODE(ip) ? \
(ip)->i_mount->m_rtdev_targp : (ip)->i_mount->m_ddev_targp)
-static inline bool
-xfs_inode_can_hw_atomic_write(
- struct xfs_inode *ip)
+static inline bool xfs_inode_can_hw_atomic_write(const struct xfs_inode *ip)
{
- struct xfs_mount *mp = ip->i_mount;
- struct xfs_buftarg *target = xfs_inode_buftarg(ip);
-
- if (mp->m_sb.sb_blocksize < target->bt_bdev_awu_min)
- return false;
- if (mp->m_sb.sb_blocksize > target->bt_bdev_awu_max)
- return false;
-
- return true;
+ return xfs_inode_buftarg(ip)->bt_bdev_awu_max > 0;
}
/*
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index 45e188466e51..04e361664710 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -558,7 +558,7 @@ xfs_setup_devices(
{
int error;
- error = xfs_setsize_buftarg(mp->m_ddev_targp, mp->m_sb.sb_sectsize);
+ error = xfs_configure_buftarg(mp->m_ddev_targp, mp->m_sb.sb_sectsize);
if (error)
return error;
@@ -567,7 +567,7 @@ xfs_setup_devices(
if (xfs_has_sector(mp))
log_sector_size = mp->m_sb.sb_logsectsize;
- error = xfs_setsize_buftarg(mp->m_logdev_targp,
+ error = xfs_configure_buftarg(mp->m_logdev_targp,
log_sector_size);
if (error)
return error;
@@ -581,7 +581,7 @@ xfs_setup_devices(
}
mp->m_rtdev_targp = mp->m_ddev_targp;
} else if (mp->m_rtname) {
- error = xfs_setsize_buftarg(mp->m_rtdev_targp,
+ error = xfs_configure_buftarg(mp->m_rtdev_targp,
mp->m_sb.sb_sectsize);
if (error)
return error;
--
2.31.1
Powered by blists - more mailing lists