September 2023 - Kernel - mailweb.openeuler.org

[PATCH openEuler-1.0-LTS] crypto: hisilicon/qm - prevent soft lockup in qm_poll_qp()'s loop
by w00416078 19 Sep '23

19 Sep '23

From: Yu'an Wang <wangyuan46(a)huawei.com> driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I82LC6 CVE: NA -------------------------------- The function qm_poll_qp() may take a while due to complex req_cb, so soft lockup may occur in kernel with preemption disabled. Add a cond_resched() to prevent that. Signed-off-by: Yu'an Wang <wangyuan46(a)huawei.com> --- drivers/crypto/hisilicon/qm.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/crypto/hisilicon/qm.c b/drivers/crypto/hisilicon/qm.c index e0fd83465dce..739b1a6565fd 100644 --- a/drivers/crypto/hisilicon/qm.c +++ b/drivers/crypto/hisilicon/qm.c @@ -540,6 +540,8 @@ static void qm_poll_qp(struct hisi_qp *qp, struct hisi_qm *qm) qm_db(qm, qp->qp_id, QM_DOORBELL_CMD_CQ, qp->qp_status.cq_head, 0); atomic_dec(&qp->qp_status.used); + + cond_resched(); } /* set c_flag */ qm_db(qm, qp->qp_id, QM_DOORBELL_CMD_CQ, -- 2.30.0

2 1

[PATCH openEuler-1.0-LTS] media: ttusb-dec: fix memory leak in ttusb_dec_exit_dvb()
by ChenXiaoSong 19 Sep '23

19 Sep '23

From: Hyunwoo Kim <imv4bel(a)gmail.com> stable inclusion from stable-v4.19.285 commit 3e5af0745a4702ab0df2f880bfe0431eb30f9164 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I635HP CVE: CVE-2022-45887 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=… -------------------------------- [ Upstream commit 517a281338322ff8293f988771c98aaa7205e457 ] Since dvb_frontend_detach() is not called in ttusb_dec_exit_dvb(), which is called when the device is disconnected, dvb_frontend_free() is not finally called. This causes a memory leak just by repeatedly plugging and unplugging the device. Fix this issue by adding dvb_frontend_detach() to ttusb_dec_exit_dvb(). Link: https://lore.kernel.org/linux-media/20221117045925.14297-5-imv4bel@gmail.com Signed-off-by: Hyunwoo Kim <imv4bel(a)gmail.com> Signed-off-by: Mauro Carvalho Chehab <mchehab(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: ChenXiaoSong <chenxiaosong2(a)huawei.com> --- drivers/media/usb/ttusb-dec/ttusb_dec.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/media/usb/ttusb-dec/ttusb_dec.c b/drivers/media/usb/ttusb-dec/ttusb_dec.c index f34efa7c61b4..c915e555897b 100644 --- a/drivers/media/usb/ttusb-dec/ttusb_dec.c +++ b/drivers/media/usb/ttusb-dec/ttusb_dec.c @@ -1561,8 +1561,7 @@ static void ttusb_dec_exit_dvb(struct ttusb_dec *dec) dvb_dmx_release(&dec->demux); if (dec->fe) { dvb_unregister_frontend(dec->fe); - if (dec->fe->ops.release) - dec->fe->ops.release(dec->fe); + dvb_frontend_detach(dec->fe); } dvb_unregister_adapter(&dec->adapter); } -- 2.31.1

2 1

[PATCH OLK-5.10] ext4: fix rec_len verify error
by Baokun Li 19 Sep '23

19 Sep '23

From: Shida Zhang <zhangshida(a)kylinos.cn> mainline inclusion from mainline-v6.6-rc2 commit 7fda67e8c3ab6069f75888f67958a6d30454a9f6 category: bugfix bugzilla: 189039, https://gitee.com/openeuler/kernel/issues/I7OXK8 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- With the configuration PAGE_SIZE 64k and filesystem blocksize 64k, a problem occurred when more than 13 million files were directly created under a directory: EXT4-fs error (device xx): ext4_dx_csum_set:492: inode #xxxx: comm xxxxx: dir seems corrupt? Run e2fsck -D. EXT4-fs error (device xx): ext4_dx_csum_verify:463: inode #xxxx: comm xxxxx: dir seems corrupt? Run e2fsck -D. EXT4-fs error (device xx): dx_probe:856: inode #xxxx: block 8188: comm xxxxx: Directory index failed checksum When enough files are created, the fake_dirent->reclen will be 0xffff. it doesn't equal to the blocksize 65536, i.e. 0x10000. But it is not the same condition when blocksize equals to 4k. when enough files are created, the fake_dirent->reclen will be 0x1000. it equals to the blocksize 4k, i.e. 0x1000. The problem seems to be related to the limitation of the 16-bit field when the blocksize is set to 64k. To address this, helpers like ext4_rec_len_{from,to}_disk has already been introduced to complete the conversion between the encoded and the plain form of rec_len. So fix this one by using the helper, and all the other in this file too. Cc: stable(a)kernel.org Fixes: dbe89444042a ("ext4: Calculate and verify checksums for htree nodes") Suggested-by: Andreas Dilger <adilger(a)dilger.ca> Suggested-by: Darrick J. Wong <djwong(a)kernel.org> Signed-off-by: Shida Zhang <zhangshida(a)kylinos.cn> Reviewed-by: Andreas Dilger <adilger(a)dilger.ca> Reviewed-by: Darrick J. Wong <djwong(a)kernel.org> Link: https://lore.kernel.org/r/20230803060938.1929759-1-zhangshida@kylinos.cn Signed-off-by: Theodore Ts'o <tytso(a)mit.edu> Signed-off-by: Baokun Li <libaokun1(a)huawei.com> --- fs/ext4/namei.c | 26 +++++++++++++++----------- 1 file changed, 15 insertions(+), 11 deletions(-) diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 746aed40a3c8..a01aba88c39d 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -344,17 +344,17 @@ static struct ext4_dir_entry_tail *get_dirent_tail(struct inode *inode, struct buffer_head *bh) { struct ext4_dir_entry_tail *t; + int blocksize = EXT4_BLOCK_SIZE(inode->i_sb); #ifdef PARANOID struct ext4_dir_entry *d, *top; d = (struct ext4_dir_entry *)bh->b_data; top = (struct ext4_dir_entry *)(bh->b_data + - (EXT4_BLOCK_SIZE(inode->i_sb) - - sizeof(struct ext4_dir_entry_tail))); - while (d < top && d->rec_len) + (blocksize - sizeof(struct ext4_dir_entry_tail))); + while (d < top && ext4_rec_len_from_disk(d->rec_len, blocksize)) d = (struct ext4_dir_entry *)(((void *)d) + - le16_to_cpu(d->rec_len)); + ext4_rec_len_from_disk(d->rec_len, blocksize)); if (d != top) return NULL; @@ -365,7 +365,8 @@ static struct ext4_dir_entry_tail *get_dirent_tail(struct inode *inode, #endif if (t->det_reserved_zero1 || - le16_to_cpu(t->det_rec_len) != sizeof(struct ext4_dir_entry_tail) || + (ext4_rec_len_from_disk(t->det_rec_len, blocksize) != + sizeof(struct ext4_dir_entry_tail)) || t->det_reserved_zero2 || t->det_reserved_ft != EXT4_FT_DIR_CSUM) return NULL; @@ -446,13 +447,14 @@ static struct dx_countlimit *get_dx_countlimit(struct inode *inode, struct ext4_dir_entry *dp; struct dx_root_info *root; int count_offset; + int blocksize = EXT4_BLOCK_SIZE(inode->i_sb); + unsigned int rlen = ext4_rec_len_from_disk(dirent->rec_len, blocksize); - if (le16_to_cpu(dirent->rec_len) == EXT4_BLOCK_SIZE(inode->i_sb)) + if (rlen == blocksize) count_offset = 8; - else if (le16_to_cpu(dirent->rec_len) == 12) { + else if (rlen == 12) { dp = (struct ext4_dir_entry *)(((void *)dirent) + 12); - if (le16_to_cpu(dp->rec_len) != - EXT4_BLOCK_SIZE(inode->i_sb) - 12) + if (ext4_rec_len_from_disk(dp->rec_len, blocksize) != blocksize - 12) return NULL; root = (struct dx_root_info *)(((void *)dp + 12)); if (root->reserved_zero || @@ -1261,6 +1263,7 @@ static int dx_make_map(struct inode *dir, struct buffer_head *bh, unsigned int buflen = bh->b_size; char *base = bh->b_data; struct dx_hash_info h = *hinfo; + int blocksize = EXT4_BLOCK_SIZE(dir->i_sb); if (ext4_has_metadata_csum(dir->i_sb)) buflen -= sizeof(struct ext4_dir_entry_tail); @@ -1274,11 +1277,12 @@ static int dx_make_map(struct inode *dir, struct buffer_head *bh, map_tail--; map_tail->hash = h.hash; map_tail->offs = ((char *) de - base)>>2; - map_tail->size = le16_to_cpu(de->rec_len); + map_tail->size = ext4_rec_len_from_disk(de->rec_len, + blocksize); count++; cond_resched(); } - de = ext4_next_entry(de, dir->i_sb->s_blocksize); + de = ext4_next_entry(de, blocksize); } return count; } -- 2.31.1

2 1

[PATCH openEuler-23.09 0/2] remote_pager: fix msg_handler_peer.c build failed
by Wupeng Ma 19 Sep '23

19 Sep '23

From: Ma Wupeng <mawupeng1(a)huawei.com> remote_pager: fix msg_handler_peer.c build failed. Chunsheng Luo (2): mmap: export __do_mmap_mm symbol remote_pager: fix msg_handler_peer.c build failed drivers/remote_pager/Kconfig | 9 ++ drivers/remote_pager/Makefile | 1 + drivers/remote_pager/main.c | 7 - drivers/remote_pager/msg_handler_peer.c | 197 +++++++++++++----------- mm/mmap.c | 1 + 5 files changed, 117 insertions(+), 98 deletions(-) -- 2.25.1

2 3

[PATCH OLK-5.10] Add new config 'CONFIG_EXT4_ERROR_REPORT' to control ext3/4 error reporting
by Baokun Li 19 Sep '23

19 Sep '23

From: Zhihao Cheng <chengzhihao1(a)huawei.com> hulk inclusion category: bugfix bugzilla: 187975, https://gitee.com/openeuler/kernel/issues/I7T77P -------------------------------- Add new config 'CONFIG_EXT4_ERROR_REPORT' to control ext3/4 error reporting. Signed-off-by: Zhihao Cheng <chengzhihao1(a)huawei.com> Signed-off-by: Baokun Li <libaokun1(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + fs/ext4/Kconfig | 8 ++++++++ fs/ext4/ext4.h | 2 ++ fs/ext4/super.c | 19 +++++++++++++++++++ 5 files changed, 31 insertions(+) diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index f055d8e93bc4..514d35e099f7 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -6211,6 +6211,7 @@ CONFIG_EXT4_FS=m CONFIG_EXT4_USE_FOR_EXT2=y CONFIG_EXT4_FS_POSIX_ACL=y CONFIG_EXT4_FS_SECURITY=y +CONFIG_EXT4_ERROR_REPORT=y # CONFIG_EXT4_DEBUG is not set CONFIG_JBD2=m # CONFIG_JBD2_DEBUG is not set diff --git a/arch/x86/configs/openeuler_defconfig b/arch/x86/configs/openeuler_defconfig index 9adedd9d615a..203c5e353d94 100644 --- a/arch/x86/configs/openeuler_defconfig +++ b/arch/x86/configs/openeuler_defconfig @@ -7308,6 +7308,7 @@ CONFIG_EXT4_FS=m CONFIG_EXT4_USE_FOR_EXT2=y CONFIG_EXT4_FS_POSIX_ACL=y CONFIG_EXT4_FS_SECURITY=y +CONFIG_EXT4_ERROR_REPORT=y # CONFIG_EXT4_DEBUG is not set CONFIG_JBD2=m # CONFIG_JBD2_DEBUG is not set diff --git a/fs/ext4/Kconfig b/fs/ext4/Kconfig index 86699c8cab28..ae108d47ff00 100644 --- a/fs/ext4/Kconfig +++ b/fs/ext4/Kconfig @@ -117,3 +117,11 @@ config EXT4_KUNIT_TESTS to the KUnit documentation in Documentation/dev-tools/kunit/. If unsure, say N. + +config EXT4_ERROR_REPORT + bool "Ext4 error reporting by netlink" + depends on EXT4_FS && NET + default n + help + Implement the ext3/ext4 file system error report. Report error to + userspace by netlink diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 56776dea9cd1..74f72a22f5ec 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -46,6 +46,7 @@ #include <linux/compiler.h> +#ifdef CONFIG_EXT4_ERROR_REPORT #define NL_EXT4_ERROR_GROUP 1 #define EXT4_ERROR_MAGIC 0xAE32014U struct ext4_err_msg { @@ -54,6 +55,7 @@ struct ext4_err_msg { unsigned long s_flags; int ext4_errno; }; +#endif /* * The fourth extended filesystem constants/structures diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 4afc9dab14cf..d24816506339 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -55,9 +55,11 @@ #include "mballoc.h" #include "fsmap.h" +#ifdef CONFIG_EXT4_ERROR_REPORT #include <uapi/linux/netlink.h> #include <net/sock.h> #include <net/net_namespace.h> +#endif #define CREATE_TRACE_POINTS #include <trace/events/ext4.h> @@ -90,8 +92,10 @@ static void ext4_unregister_li_request(struct super_block *sb); static void ext4_clear_request_list(void); static struct inode *ext4_get_journal_inode(struct super_block *sb, unsigned int journal_inum); +#ifdef CONFIG_EXT4_ERROR_REPORT static void ext4_netlink_send_info(struct super_block *sb, int ext4_errno); static struct sock *ext4nl; +#endif /* * Lock ordering @@ -616,6 +620,7 @@ static void save_error_info(struct super_block *sb, int error, spin_unlock(&sbi->s_error_lock); } +#ifdef CONFIG_EXT4_ERROR_REPORT static void ext4_netlink_send_info(struct super_block *sb, int ext4_errno) { int size; @@ -651,6 +656,7 @@ static void ext4_netlink_send_info(struct super_block *sb, int ext4_errno) kfree_skb(skb); } } +#endif /* Deal with the reporting of failure conditions on a filesystem such as * inconsistencies detected or read IO failures. @@ -713,11 +719,16 @@ static void ext4_handle_error(struct super_block *sb, bool force_ro, int error, sb->s_id); } +#ifdef CONFIG_EXT4_ERROR_REPORT if (sb_rdonly(sb)) return; if (continue_fs) goto out; +#else + if (sb_rdonly(sb) || continue_fs) + return; +#endif ext4_msg(sb, KERN_CRIT, "Remounting filesystem read-only"); @@ -727,8 +738,10 @@ static void ext4_handle_error(struct super_block *sb, bool force_ro, int error, */ smp_wmb(); sb->s_flags |= SB_RDONLY; +#ifdef CONFIG_EXT4_ERROR_REPORT out: ext4_netlink_send_info(sb, force_ro ? 2 : 1); +#endif } static void flush_stashed_error_work(struct work_struct *work) @@ -6855,7 +6868,9 @@ wait_queue_head_t ext4__ioend_wq[EXT4_WQ_HASH_SZ]; static int __init ext4_init_fs(void) { int i, err; +#ifdef CONFIG_EXT4_ERROR_REPORT struct netlink_kernel_cfg cfg = {.groups = NL_EXT4_ERROR_GROUP,}; +#endif ratelimit_state_init(&ext4_mount_msg_ratelimit, 30 * HZ, 64); ext4_li_info = NULL; @@ -6908,9 +6923,11 @@ static int __init ext4_init_fs(void) if (err) goto out; +#ifdef CONFIG_EXT4_ERROR_REPORT ext4nl = netlink_kernel_create(&init_net, NETLINK_FILESYSTEM, &cfg); if (!ext4nl) printk(KERN_ERR "EXT4-fs: Cannot create netlink socket.\n"); +#endif return 0; out: unregister_as_ext2(); @@ -6951,7 +6968,9 @@ static void __exit ext4_exit_fs(void) ext4_exit_post_read_processing(); ext4_exit_es(); ext4_exit_pending(); +#ifdef CONFIG_EXT4_ERROR_REPORT netlink_kernel_release(ext4nl); +#endif } MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others"); -- 2.31.1

2 1

[PATCH 1/2] mmap: export __do_mmap_mm symbol
by Wupeng Ma 19 Sep '23

19 Sep '23

From: Chunsheng Luo <luochunsheng(a)huawei.com> euleros inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I7WLVX --------------------------------------------- export __do_mmap_mm symbol by referring to the earlier version Signed-off-by: Chunsheng Luo <luochunsheng(a)huawei.com> --- mm/mmap.c | 1 + 1 file changed, 1 insertion(+) diff --git a/mm/mmap.c b/mm/mmap.c index 2aef07b8a85e..819c37e7eb1f 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1469,6 +1469,7 @@ unsigned long __do_mmap_mm(struct mm_struct *mm, struct file *file, unsigned lon *populate = len; return addr; } +EXPORT_SYMBOL(__do_mmap_mm); unsigned long do_mmap(struct file *file, unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, -- 2.33.0

1 1

[PATCH OLK-5.10] sched/qos: Fix warning in CPU hotplug scenarios
by Xia Fukun 19 Sep '23

19 Sep '23

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7ZMCB CVE: NA -------------------------------- CPU hotplug callbacks race against distribute_cfs_runtime(), when the QOS_SCHED feature is enabled, there may be situations where the cfs_rq-> runtime_remaining == 1 and cfs_rq is QOS_THROTTLED. Turn off the Qos_throttle when the CPU is offline. No longer allocate time to cfs_rq in this scenario to fix the warning. Fixes: 4eb6eb7941dc ("sched/qos: Don't unthrottle cfs_rq when cfs_rq is throttled by qos") Signed-off-by: Xia Fukun <xiafukun(a)huawei.com> --- kernel/sched/fair.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index daa853b19853..b8bf7acb9f9a 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5267,6 +5267,19 @@ static void distribute_cfs_runtime(struct cfs_bandwidth *cfs_b) if (!cfs_rq_throttled(cfs_rq)) goto next; + /* + * CPU hotplug callbacks race against distribute_cfs_runtime() + * when the QOS_SCHED feature is enabled, there may be + * situations where the runtime_remaining > 0. + * Qos_sched does not care whether the cfs_rq has time left, + * so no longer allocate time to cfs_rq in this scenario. + */ +#ifdef CONFIG_QOS_SCHED + if (cfs_rq->throttled == QOS_THROTTLED && + cfs_rq->runtime_remaining > 0) + goto next; +#endif + /* By the above check, this should never be true */ SCHED_WARN_ON(cfs_rq->runtime_remaining > 0); @@ -7923,6 +7936,10 @@ static __always_inline bool check_qos_cfs_rq(struct cfs_rq *cfs_rq) if (unlikely(cfs_rq && is_offline_level(cfs_rq->tg->qos_level) && !sched_idle_cpu(smp_processor_id()) && cfs_rq->h_nr_running == cfs_rq->idle_h_nr_running)) { + + if (!rq_of(cfs_rq)->online) + return false; + throttle_qos_cfs_rq(cfs_rq); return true; } -- 2.34.1

2 1

[PATCH openEuler-23.09] mm/mlock: return EINVAL for illegal user memory range in mlock
by Wupeng Ma 19 Sep '23

19 Sep '23

From: Ma Wupeng <mawupeng1(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7TW89 -------------------------------- While testing mlock, we have a problem if the len of mlock is ULONG_MAX. The return value of mlock is zero. But nothing will be locked since the len in do_mlock overflows to zero due to the following code in mlock: len = PAGE_ALIGN(len + (offset_in_page(start))); The same problem happens in munlock. Since TASK_SIZE is the maximum user space address. The start or len of mlock shouldn't be bigger than this. Function access_ok can be used to check this issue, so return -EINVAL if bigger. Signed-off-by: Ma Wupeng <mawupeng1(a)huawei.com> --- mm/mlock.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/mm/mlock.c b/mm/mlock.c index 40b43f8740df..e90139d42f88 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -479,8 +479,6 @@ static int apply_vma_lock_flags(unsigned long start, size_t len, end = start + len; if (end < start) return -EINVAL; - if (end == start) - return 0; vma = vma_iter_load(&vmi); if (!vma) return -ENOMEM; @@ -574,9 +572,15 @@ static __must_check int do_mlock(unsigned long start, size_t len, vm_flags_t fla if (!can_do_mlock()) return -EPERM; + if (!len) + return 0; + len = PAGE_ALIGN(len + (offset_in_page(start))); start &= PAGE_MASK; + if (!len) + return -EINVAL; + lock_limit = rlimit(RLIMIT_MEMLOCK); lock_limit >>= PAGE_SHIFT; locked = len >> PAGE_SHIFT; @@ -634,8 +638,13 @@ SYSCALL_DEFINE2(munlock, unsigned long, start, size_t, len) start = untagged_addr(start); + if (!len) + return 0; + len = PAGE_ALIGN(len + (offset_in_page(start))); start &= PAGE_MASK; + if (!len) + return -EINVAL; if (mmap_write_lock_killable(current->mm)) return -EINTR; -- 2.25.1

2 1

[PATCH OLK-5.10] uacce: modify the configuration mode of device isolation stragety
by Wenkai Lin 19 Sep '23

19 Sep '23

From: Qi Tao <taoqi10(a)huawei.com> driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I82ARL CVE: NA ------------------------------------------------------------------ cancel the reference counting of the accelerator device so that the device (with hardware errors) isolation stragety can also be modified while the accelerator is running a task. Signed-off-by: Qi Tao <taoqi10(a)huawei.com> Signed-off-by: JiangShui Yang <yangjiangshui(a)h-partners.com> --- drivers/misc/uacce/uacce.c | 5 ----- include/linux/uacce.h | 2 -- 2 files changed, 7 deletions(-) diff --git a/drivers/misc/uacce/uacce.c b/drivers/misc/uacce/uacce.c index 21a4f5892aff..7a64eed01df7 100644 --- a/drivers/misc/uacce/uacce.c +++ b/drivers/misc/uacce/uacce.c @@ -88,7 +88,6 @@ static int uacce_put_queue(struct uacce_queue *q) uacce->ops->put_queue(q); q->state = UACCE_Q_ZOMBIE; - atomic_dec(&uacce->ref); return 0; } @@ -348,7 +347,6 @@ static int uacce_fops_open(struct inode *inode, struct file *filep) goto out_with_bond; } - atomic_inc(&uacce->ref); init_waitqueue_head(&q->wait); filep->private_data = q; q->state = UACCE_Q_INIT; @@ -821,9 +819,6 @@ static ssize_t isolate_strategy_store(struct device *dev, struct device_attribut if (val > UACCE_MAX_ERR_THRESHOLD) return -EINVAL; - if (atomic_read(&uacce->ref)) - return -EBUSY; - ret = uacce->ops->isolate_err_threshold_write(uacce, val); if (ret) return ret; diff --git a/include/linux/uacce.h b/include/linux/uacce.h index 8187c1bda236..9b0c04a9cff7 100644 --- a/include/linux/uacce.h +++ b/include/linux/uacce.h @@ -148,7 +148,6 @@ struct uacce_queue { * @mutex: protects uacce operation * @priv: private pointer of the uacce * @queues: list of queues - * @ref: reference of the uacce */ struct uacce_device { const char *algs; @@ -164,7 +163,6 @@ struct uacce_device { struct device dev; struct mutex mutex; void *priv; - atomic_t ref; struct uacce_err_isolate *isolate; struct list_head queues; }; -- 2.30.0

2 1

[PATCH OLK-5.10 v2] jbd2: Fix potential data lost in recovering journal raced with synchronizing fs bdev
by Zhihao Cheng 19 Sep '23

19 Sep '23

maillist inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I828EV CVE: NA Reference: https://lore.kernel.org/linux-ext4/20230911155822.kbg2xayc7ys24kay@quack3/T… -------------------------------- JBD2 makes sure journal data is fallen on fs device by sync_blockdev(), however, other process could intercept the EIO information from bdev's mapping, which leads journal recovering successful even EIO occurs during data written back to fs device. We found this problem in our product, iscsi + multipath is chosen for block device of ext4. Unstable network may trigger kpartx to rescan partitions in device mapper layer. Detailed process is shown as following: mount kpartx irq jbd2_journal_recover do_one_pass memcpy(nbh->b_data, obh->b_data) // copy data to fs dev from journal mark_buffer_dirty // mark bh dirty vfs_read generic_file_read_iter // dio filemap_write_and_wait_range __filemap_fdatawrite_range do_writepages block_write_full_folio submit_bh_wbc >> EIO occurs in disk << end_buffer_async_write mark_buffer_write_io_error mapping_set_error set_bit(AS_EIO, &mapping->flags) // set! filemap_check_errors test_and_clear_bit(AS_EIO, &mapping->flags) // clear! err2 = sync_blockdev filemap_write_and_wait filemap_check_errors test_and_clear_bit(AS_EIO, &mapping->flags) // false err2 = 0 Filesystem is mounted successfully even data from journal is failed written into disk, and ext4/ocfs2 could become corrupted. Fix it by comparing the wb_err state in fs block device before recovering and after recovering. Fetch a reproducer in [Link]. Link: https://bugzilla.kernel.org/show_bug.cgi?id=217888 Cc: stable(a)vger.kernel.org Signed-off-by: Zhihao Cheng <chengzhihao1(a)huawei.com> Signed-off-by: Zhang Yi <yi.zhang(a)huawei.com> --- v1->v3: Initialize wb_err. Untialized wb_err could be same with mapping->wb_err(eg. EIO without ERRSEQ_SEEN). When EIO occurs again, mapping->wb_err won't be changed, and wb_err is still same with mapping->wb_err. fs/jbd2/recovery.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index 9e4537349380..cdf4c553f058 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -292,6 +292,8 @@ int jbd2_journal_recover(journal_t *journal) journal_superblock_t * sb; struct recovery_info info; + errseq_t wb_err; + struct address_space *mapping; memset(&info, 0, sizeof(info)); sb = journal->j_superblock; @@ -309,6 +311,9 @@ int jbd2_journal_recover(journal_t *journal) return 0; } + wb_err = 0; + mapping = journal->j_fs_dev->bd_inode->i_mapping; + errseq_check_and_advance(&mapping->wb_err, &wb_err); err = do_one_pass(journal, &info, PASS_SCAN); if (!err) err = do_one_pass(journal, &info, PASS_REVOKE); @@ -327,6 +332,9 @@ int jbd2_journal_recover(journal_t *journal) jbd2_journal_clear_revoke(journal); err2 = sync_blockdev(journal->j_fs_dev); + if (!err) + err = err2; + err2 = errseq_check_and_advance(&mapping->wb_err, &wb_err); if (!err) err = err2; /* Make sure all replayed data is on permanent storage */ -- 2.31.1

2 1

2024

2023

2022

2021

2020

2019

Kernel September 2023