From 967fa6b9d1a96b37836e11908b7379867810953b Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 30 Aug 2023 22:20:14 +0900 Subject: [PATCH 01/74] exfat: vfs: get rid of old '->iterate' directory operation All users now just use '->iterate_shared()', which only takes the directory inode lock for reading. Filesystems that never got convered to shared mode now instead use a wrapper that drops the lock, re-takes it in write mode, calls the old function, and then downgrades the lock back to read mode. This way the VFS layer and other callers no longer need to care about filesystems that never got converted to the modern era. The filesystems that use the new wrapper are ceph, coda, exfat, jfs, ntfs, ocfs2, overlayfs, and vboxsf. Honestly, several of them look like they really could just iterate their directories in shared mode and skip the wrapper entirely, but the point of this change is to not change semantics or fix filesystems that haven't been fixed in the last 7+ years, but to finally get rid of the dual iterators. Signed-off-by: Linus Torvalds Signed-off-by: Christian Brauner Signed-off-by: Namjae Jeon --- dir.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/dir.c b/dir.c index bf8eb5d1b1c2..a9dc79e0e6d4 100644 --- a/dir.c +++ b/dir.c @@ -307,10 +307,17 @@ out: return err; } +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 5, 0) +WRAP_DIR_ITER(exfat_iterate) // FIXME! +#endif const struct file_operations exfat_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 5, 0) + .iterate_shared = shared_exfat_iterate, +#else .iterate = exfat_iterate, +#endif .unlocked_ioctl = exfat_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = exfat_compat_ioctl, From cc9879a6ce62c90737c19635bfc74c3c93e2b597 Mon Sep 17 00:00:00 2001 From: Jan Cincera Date: Mon, 4 Sep 2023 11:05:23 +0900 Subject: [PATCH 02/74] exfat: add ioctls for accessing attributes Add GET and SET attributes ioctls to enable attribute modification. We already do this in FAT and a few userspace utils made for it would benefit from this also working on exFAT, namely fatattr. Signed-off-by: Jan Cincera Signed-off-by: Namjae Jeon --- exfat_fs.h | 6 +++ file.c | 132 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 138 insertions(+) diff --git a/exfat_fs.h b/exfat_fs.h index b79f2fe2bea5..97ee6843c17e 100644 --- a/exfat_fs.h +++ b/exfat_fs.h @@ -162,6 +162,12 @@ enum { #define DIR_CACHE_SIZE \ (DIV_ROUND_UP(EXFAT_DEN_TO_B(ES_MAX_ENTRY_NUM), SECTOR_SIZE) + 1) +/* + * attribute ioctls, same as their FAT equivalents. + */ +#define EXFAT_IOCTL_GET_ATTRIBUTES _IOR('r', 0x10, __u32) +#define EXFAT_IOCTL_SET_ATTRIBUTES _IOW('r', 0x11, __u32) + struct exfat_dentry_namebuf { char *lfn; int lfnbuf_len; /* usually MAX_UNINAME_BUF_SIZE */ diff --git a/file.c b/file.c index e35998f5aa8c..216c38b981e1 100644 --- a/file.c +++ b/file.c @@ -9,8 +9,11 @@ #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 11, 0) #include #endif +#include #include #include +#include +#include #include "exfat_raw.h" #include "exfat_fs.h" @@ -390,6 +393,130 @@ out: return error; } +/* + * modified ioctls from fat/file.c by Welmer Almesberger + */ +static int exfat_ioctl_get_attributes(struct inode *inode, u32 __user *user_attr) +{ + u32 attr; + +#if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 7, 0) + inode_lock_shared(inode); +#else + mutex_lock(&inode->i_mutex); +#endif + attr = exfat_make_attr(inode); +#if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 7, 0) + inode_unlock_shared(inode); +#else + mutex_unlock(&inode->i_mutex); +#endif + + return put_user(attr, user_attr); +} + +static int exfat_ioctl_set_attributes(struct file *file, u32 __user *user_attr) +{ + struct inode *inode = file_inode(file); + struct exfat_sb_info *sbi = EXFAT_SB(inode->i_sb); + int is_dir = S_ISDIR(inode->i_mode); + u32 attr, oldattr; + struct iattr ia; + int err; + + err = get_user(attr, user_attr); + if (err) + goto out; + + err = mnt_want_write_file(file); + if (err) + goto out; + +#if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 4, 0) + inode_lock(inode); +#else + mutex_lock(&inode->i_mutex); +#endif + + oldattr = exfat_make_attr(inode); + + /* + * Mask attributes so we don't set reserved fields. + */ + attr &= (ATTR_READONLY | ATTR_HIDDEN | ATTR_SYSTEM | ATTR_ARCHIVE); + attr |= (is_dir ? ATTR_SUBDIR : 0); + + /* Equivalent to a chmod() */ + ia.ia_valid = ATTR_MODE | ATTR_CTIME; +#if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 12, 0) + ia.ia_ctime = current_time(inode); +#else + ia.ia_ctime = current_fs_time(inode->i_sb); +#endif + if (is_dir) + ia.ia_mode = exfat_make_mode(sbi, attr, 0777); + else + ia.ia_mode = exfat_make_mode(sbi, attr, 0666 | (inode->i_mode & 0111)); + + /* The root directory has no attributes */ + if (inode->i_ino == EXFAT_ROOT_INO && attr != ATTR_SUBDIR) { + err = -EINVAL; + goto out_unlock_inode; + } + + if (((attr | oldattr) & ATTR_SYSTEM) && + !capable(CAP_LINUX_IMMUTABLE)) { + err = -EPERM; + goto out_unlock_inode; + } + + /* + * The security check is questionable... We single + * out the RO attribute for checking by the security + * module, just because it maps to a file mode. + */ +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 0, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 3, 0) + err = security_inode_setattr(file_mnt_idmap(file), + file->f_path.dentry, &ia); +#else + err = security_inode_setattr(file_mnt_user_ns(file), + file->f_path.dentry, &ia); +#endif +#else + err = security_inode_setattr(file->f_path.dentry, &ia); +#endif + if (err) + goto out_unlock_inode; + +#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 12, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 3, 0) + /* This MUST be done before doing anything irreversible... */ + err = exfat_setattr(file_mnt_idmap(file), file->f_path.dentry, &ia); +#else + err = exfat_setattr(file_mnt_user_ns(file), file->f_path.dentry, &ia); +#endif +#else + err = exfat_setattr(file->f_path.dentry, &ia); +#endif + if (err) + goto out_unlock_inode; + + fsnotify_change(file->f_path.dentry, ia.ia_valid); + + exfat_save_attr(inode, attr); + mark_inode_dirty(inode); +out_unlock_inode: +#if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 4, 0) + inode_unlock(inode); +#else + mutex_unlock(&inode->i_mutex); +#endif + mnt_drop_write_file(file); +out: + return err; +} + static int exfat_ioctl_fitrim(struct inode *inode, unsigned long arg) { #if LINUX_VERSION_CODE < KERNEL_VERSION(5, 19, 0) @@ -432,8 +559,13 @@ static int exfat_ioctl_fitrim(struct inode *inode, unsigned long arg) long exfat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { struct inode *inode = file_inode(filp); + u32 __user *user_attr = (u32 __user *)arg; switch (cmd) { + case EXFAT_IOCTL_GET_ATTRIBUTES: + return exfat_ioctl_get_attributes(inode, user_attr); + case EXFAT_IOCTL_SET_ATTRIBUTES: + return exfat_ioctl_set_attributes(filp, user_attr); case FITRIM: return exfat_ioctl_fitrim(inode, arg); default: From f6c705d1d91cc6624b70864a32ab55f09bdf5798 Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Mon, 4 Sep 2023 11:08:43 +0900 Subject: [PATCH 03/74] exfat: support handle zero-size directory After repairing a corrupted file system with exfatprogs' fsck.exfat, zero-size directories may result. It is also possible to create zero-size directories in other exFAT implementation, such as Paragon ufsd dirver. As described in the specification, the lower directory size limits is 0 bytes. Without this commit, sub-directories and files cannot be created under a zero-size directory, and it cannot be removed. Signed-off-by: Yuezhang Mo Reviewed-by: Andy Wu Reviewed-by: Aoyama Wataru Signed-off-by: Namjae Jeon --- namei.c | 29 ++++++++++++++++++++++------- 1 file changed, 22 insertions(+), 7 deletions(-) diff --git a/namei.c b/namei.c index e6971c758892..3429a8e47495 100644 --- a/namei.c +++ b/namei.c @@ -378,14 +378,20 @@ static int exfat_find_empty_entry(struct inode *inode, if (exfat_check_max_dentries(inode)) return -ENOSPC; - /* we trust p_dir->size regardless of FAT type */ - if (exfat_find_last_cluster(sb, p_dir, &last_clu)) - return -EIO; - /* * Allocate new cluster to this directory */ - exfat_chain_set(&clu, last_clu + 1, 0, p_dir->flags); + if (ei->start_clu != EXFAT_EOF_CLUSTER) { + /* we trust p_dir->size regardless of FAT type */ + if (exfat_find_last_cluster(sb, p_dir, &last_clu)) + return -EIO; + + exfat_chain_set(&clu, last_clu + 1, 0, p_dir->flags); + } else { + /* This directory is empty */ + exfat_chain_set(&clu, EXFAT_EOF_CLUSTER, 0, + ALLOC_NO_FAT_CHAIN); + } /* allocate a cluster */ ret = exfat_alloc_cluster(inode, 1, &clu, IS_DIRSYNC(inode)); @@ -395,6 +401,11 @@ static int exfat_find_empty_entry(struct inode *inode, if (exfat_zeroed_cluster(inode, clu.dir)) return -EIO; + if (ei->start_clu == EXFAT_EOF_CLUSTER) { + ei->start_clu = clu.dir; + p_dir->dir = clu.dir; + } + /* append to the FAT chain */ if (clu.flags != p_dir->flags) { /* no-fat-chain bit is disabled, @@ -709,7 +720,7 @@ static int exfat_find(struct inode *dir, struct qstr *qname, info->type = exfat_get_entry_type(ep); info->attr = le16_to_cpu(ep->dentry.file.attr); info->size = le64_to_cpu(ep2->dentry.stream.valid_size); - if ((info->type == TYPE_FILE) && (info->size == 0)) { + if (info->size == 0) { info->flags = ALLOC_NO_FAT_CHAIN; info->start_clu = EXFAT_EOF_CLUSTER; } else { @@ -1000,6 +1011,9 @@ static int exfat_check_dir_empty(struct super_block *sb, dentries_per_clu = sbi->dentries_per_clu; + if (p_dir->dir == EXFAT_EOF_CLUSTER) + return 0; + exfat_chain_dup(&clu, p_dir); while (clu.dir != EXFAT_EOF_CLUSTER) { @@ -1383,7 +1397,8 @@ static int __exfat_rename(struct inode *old_parent_inode, } /* Free the clusters if new_inode is a dir(as if exfat_rmdir) */ - if (new_entry_type == TYPE_DIR) { + if (new_entry_type == TYPE_DIR && + new_ei->start_clu != EXFAT_EOF_CLUSTER) { /* new_ei, new_clu_to_free */ struct exfat_chain new_clu_to_free; From 5bb0e9f3ec0983e2dc77e8d4bb8b43b1bcfe280c Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Tue, 5 Sep 2023 11:26:20 +0900 Subject: [PATCH 04/74] exfat: support create zero-size directory This commit adds mount option 'zero_size_dir'. If this option enabled, don't allocate a cluster to directory when creating it, and set the directory size to 0. On Windows, a cluster is allocated for a directory when it is created, so the mount option is disabled by default. Signed-off-by: Yuezhang Mo Reviewed-by: Andy Wu Reviewed-by: Aoyama Wataru Signed-off-by: Namjae Jeon --- dir.c | 12 ++++++------ exfat_fs.h | 2 ++ namei.c | 7 +++++-- super.c | 12 ++++++++++++ 4 files changed, 25 insertions(+), 8 deletions(-) diff --git a/dir.c b/dir.c index a9dc79e0e6d4..f8a04e16dc88 100644 --- a/dir.c +++ b/dir.c @@ -425,11 +425,13 @@ static void exfat_set_entry_type(struct exfat_dentry *ep, unsigned int type) } static void exfat_init_stream_entry(struct exfat_dentry *ep, - unsigned char flags, unsigned int start_clu, - unsigned long long size) + unsigned int start_clu, unsigned long long size) { exfat_set_entry_type(ep, TYPE_STREAM); - ep->dentry.stream.flags = flags; + if (size == 0) + ep->dentry.stream.flags = ALLOC_FAT_CHAIN; + else + ep->dentry.stream.flags = ALLOC_NO_FAT_CHAIN; ep->dentry.stream.start_clu = cpu_to_le32(start_clu); ep->dentry.stream.valid_size = cpu_to_le64(size); ep->dentry.stream.size = cpu_to_le64(size); @@ -505,9 +507,7 @@ int exfat_init_dir_entry(struct inode *inode, struct exfat_chain *p_dir, if (!ep) return -EIO; - exfat_init_stream_entry(ep, - (type == TYPE_FILE) ? ALLOC_FAT_CHAIN : ALLOC_NO_FAT_CHAIN, - start_clu, size); + exfat_init_stream_entry(ep, start_clu, size); exfat_update_bh(bh, IS_DIRSYNC(inode)); brelse(bh); diff --git a/exfat_fs.h b/exfat_fs.h index 97ee6843c17e..311d9f28e25f 100644 --- a/exfat_fs.h +++ b/exfat_fs.h @@ -259,6 +259,8 @@ struct exfat_mount_options { discard:1, /* Issue discard requests on deletions */ keep_last_dots:1; /* Keep trailing periods in paths */ int time_offset; /* Offset of timestamps from UTC (in minutes) */ + /* Support creating zero-size directory, default: false */ + bool zero_size_dir; }; /* diff --git a/namei.c b/namei.c index 3429a8e47495..d43af646e229 100644 --- a/namei.c +++ b/namei.c @@ -545,7 +545,7 @@ static int exfat_add_entry(struct inode *inode, const char *path, goto out; } - if (type == TYPE_DIR) { + if (type == TYPE_DIR && !sbi->options.zero_size_dir) { ret = exfat_alloc_new_dir(inode, &clu); if (ret) goto out; @@ -578,7 +578,10 @@ static int exfat_add_entry(struct inode *inode, const char *path, info->num_subdirs = 0; } else { info->attr = ATTR_SUBDIR; - info->start_clu = start_clu; + if (sbi->options.zero_size_dir) + info->start_clu = EXFAT_EOF_CLUSTER; + else + info->start_clu = start_clu; info->size = clu_size; info->num_subdirs = EXFAT_MIN_SUBDIR; } diff --git a/super.c b/super.c index 665c296e4ad4..89f60d722c99 100644 --- a/super.c +++ b/super.c @@ -195,6 +195,8 @@ static int exfat_show_options(struct seq_file *m, struct dentry *root) seq_puts(m, ",sys_tz"); else if (opts->time_offset) seq_printf(m, ",time_offset=%d", opts->time_offset); + if (opts->zero_size_dir) + seq_puts(m, ",zero_size_dir"); return 0; } @@ -278,6 +280,7 @@ enum { Opt_keep_last_dots, Opt_sys_tz, Opt_time_offset, + Opt_zero_size_dir, /* Deprecated options */ Opt_utf8, @@ -323,6 +326,7 @@ static const struct fs_parameter_spec exfat_param_specs[] = { fsparam_flag("keep_last_dots", Opt_keep_last_dots), fsparam_flag("sys_tz", Opt_sys_tz), fsparam_s32("time_offset", Opt_time_offset), + fsparam_flag("zero_size_dir", Opt_zero_size_dir), __fsparam(NULL, "utf8", Opt_utf8, fs_param_deprecated, NULL), __fsparam(NULL, "debug", Opt_debug, fs_param_deprecated, @@ -403,6 +407,9 @@ static int exfat_parse_param(struct fs_context *fc, struct fs_parameter *param) return -EINVAL; opts->time_offset = result.int_32; break; + case Opt_zero_size_dir: + opts->zero_size_dir = true; + break; case Opt_utf8: case Opt_debug: case Opt_namecase: @@ -435,6 +442,7 @@ enum { Opt_debug, Opt_namecase, Opt_codepage, + Opt_zero_size_dir, Opt_fs, }; @@ -454,6 +462,7 @@ static const match_table_t exfat_tokens = { {Opt_namecase, "namecase=%u"}, {Opt_debug, "debug"}, {Opt_utf8, "utf8"}, + {Opt_zero_size_dir, "zero_size_dir"}, {Opt_err, NULL} }; @@ -545,6 +554,9 @@ static int parse_options(struct super_block *sb, char *options, int silent, case Opt_namecase: case Opt_codepage: break; + case Opt_zero_size_dir: + opts->zero_size_dir = true; + break; default: if (!silent) { exfat_err(sb, From 8279a1a0bc27b03565da6775454e09a520b13a09 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Mon, 11 Sep 2023 16:59:54 +0900 Subject: [PATCH 05/74] exfat: github action: remove liunx-4.1 source to get more disk space Signed-off-by: Namjae Jeon --- .github/workflows/c-cpp.yml | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/.github/workflows/c-cpp.yml b/.github/workflows/c-cpp.yml index 375c800ac23c..6587fd98bd41 100644 --- a/.github/workflows/c-cpp.yml +++ b/.github/workflows/c-cpp.yml @@ -20,6 +20,7 @@ jobs: wget https://mirrors.edge.kernel.org/pub/linux/kernel/v4.x/linux-4.1.36.tar.gz tar xf linux-4.1.36.tar.gz mv linux-4.1.36 linux-stable + rm -rf linux-4.1.36.tar.gz - name: Prerequisite for xfstests testing run: | sudo apt-get install linux-headers-$(uname -r) @@ -47,7 +48,9 @@ jobs: make -j$((`nproc`+1)) fs/exfat/exfat.ko - name: Run xfstests testsuite run: | - cd ../linux-exfat-oot + cd .. + rm -rf linux-stable + cd linux-exfat-oot make > /dev/null sudo make install > /dev/null sudo insmod exfat.ko From 950e271397ef414c402e653e0117e1f884e06fe1 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Mon, 30 Oct 2023 21:08:39 +0900 Subject: [PATCH 06/74] exfat: use fat ioctls definitions from include/uapi/linux/msdos_fs.h Signed-off-by: Namjae Jeon --- dir.c | 8 ++++---- exfat_fs.h | 18 ++++++------------ exfat_raw.h | 17 +++++++++-------- file.c | 16 +++++++++------- inode.c | 6 +++--- namei.c | 16 ++++++++-------- super.c | 4 ++-- 7 files changed, 41 insertions(+), 44 deletions(-) diff --git a/dir.c b/dir.c index f8a04e16dc88..4d5b863b4fde 100644 --- a/dir.c +++ b/dir.c @@ -288,7 +288,7 @@ get_new: mutex_unlock(&EXFAT_SB(sb)->s_lock); if (!dir_emit(ctx, nb->lfn, strlen(nb->lfn), inum, - (de.attr & ATTR_SUBDIR) ? DT_DIR : DT_REG)) + (de.attr & EXFAT_ATTR_SUBDIR) ? DT_DIR : DT_REG)) goto out; ctx->pos = cpos; goto get_new; @@ -366,7 +366,7 @@ unsigned int exfat_get_entry_type(struct exfat_dentry *ep) if (ep->type == EXFAT_VOLUME) return TYPE_VOLUME; if (ep->type == EXFAT_FILE) { - if (le16_to_cpu(ep->dentry.file.attr) & ATTR_SUBDIR) + if (le16_to_cpu(ep->dentry.file.attr) & EXFAT_ATTR_SUBDIR) return TYPE_DIR; return TYPE_FILE; } @@ -417,10 +417,10 @@ static void exfat_set_entry_type(struct exfat_dentry *ep, unsigned int type) ep->type = EXFAT_VOLUME; } else if (type == TYPE_DIR) { ep->type = EXFAT_FILE; - ep->dentry.file.attr = cpu_to_le16(ATTR_SUBDIR); + ep->dentry.file.attr = cpu_to_le16(EXFAT_ATTR_SUBDIR); } else if (type == TYPE_FILE) { ep->type = EXFAT_FILE; - ep->dentry.file.attr = cpu_to_le16(ATTR_ARCHIVE); + ep->dentry.file.attr = cpu_to_le16(EXFAT_ATTR_ARCHIVE); } } diff --git a/exfat_fs.h b/exfat_fs.h index 311d9f28e25f..9c2266afe3f6 100644 --- a/exfat_fs.h +++ b/exfat_fs.h @@ -162,12 +162,6 @@ enum { #define DIR_CACHE_SIZE \ (DIV_ROUND_UP(EXFAT_DEN_TO_B(ES_MAX_ENTRY_NUM), SECTOR_SIZE) + 1) -/* - * attribute ioctls, same as their FAT equivalents. - */ -#define EXFAT_IOCTL_GET_ATTRIBUTES _IOR('r', 0x10, __u32) -#define EXFAT_IOCTL_SET_ATTRIBUTES _IOW('r', 0x11, __u32) - struct exfat_dentry_namebuf { char *lfn; int lfnbuf_len; /* usually MAX_UNINAME_BUF_SIZE */ @@ -390,10 +384,10 @@ static inline int exfat_mode_can_hold_ro(struct inode *inode) static inline mode_t exfat_make_mode(struct exfat_sb_info *sbi, unsigned short attr, mode_t mode) { - if ((attr & ATTR_READONLY) && !(attr & ATTR_SUBDIR)) + if ((attr & EXFAT_ATTR_READONLY) && !(attr & EXFAT_ATTR_SUBDIR)) mode &= ~0222; - if (attr & ATTR_SUBDIR) + if (attr & EXFAT_ATTR_SUBDIR) return (mode & ~sbi->options.fs_dmask) | S_IFDIR; return (mode & ~sbi->options.fs_fmask) | S_IFREG; @@ -405,18 +399,18 @@ static inline unsigned short exfat_make_attr(struct inode *inode) unsigned short attr = EXFAT_I(inode)->attr; if (S_ISDIR(inode->i_mode)) - attr |= ATTR_SUBDIR; + attr |= EXFAT_ATTR_SUBDIR; if (exfat_mode_can_hold_ro(inode) && !(inode->i_mode & 0222)) - attr |= ATTR_READONLY; + attr |= EXFAT_ATTR_READONLY; return attr; } static inline void exfat_save_attr(struct inode *inode, unsigned short attr) { if (exfat_mode_can_hold_ro(inode)) - EXFAT_I(inode)->attr = attr & (ATTR_RWMASK | ATTR_READONLY); + EXFAT_I(inode)->attr = attr & (EXFAT_ATTR_RWMASK | EXFAT_ATTR_READONLY); else - EXFAT_I(inode)->attr = attr & ATTR_RWMASK; + EXFAT_I(inode)->attr = attr & EXFAT_ATTR_RWMASK; } static inline bool exfat_is_last_sector_in_cluster(struct exfat_sb_info *sbi, diff --git a/exfat_raw.h b/exfat_raw.h index 0ece2e43cf49..971a1ccd0e89 100644 --- a/exfat_raw.h +++ b/exfat_raw.h @@ -64,15 +64,16 @@ #define CS_DEFAULT 2 /* file attributes */ -#define ATTR_READONLY 0x0001 -#define ATTR_HIDDEN 0x0002 -#define ATTR_SYSTEM 0x0004 -#define ATTR_VOLUME 0x0008 -#define ATTR_SUBDIR 0x0010 -#define ATTR_ARCHIVE 0x0020 +#define EXFAT_ATTR_READONLY 0x0001 +#define EXFAT_ATTR_HIDDEN 0x0002 +#define EXFAT_ATTR_SYSTEM 0x0004 +#define EXFAT_ATTR_VOLUME 0x0008 +#define EXFAT_ATTR_SUBDIR 0x0010 +#define EXFAT_ATTR_ARCHIVE 0x0020 -#define ATTR_RWMASK (ATTR_HIDDEN | ATTR_SYSTEM | ATTR_VOLUME | \ - ATTR_SUBDIR | ATTR_ARCHIVE) +#define EXFAT_ATTR_RWMASK (EXFAT_ATTR_HIDDEN | EXFAT_ATTR_SYSTEM | \ + EXFAT_ATTR_VOLUME | EXFAT_ATTR_SUBDIR | \ + EXFAT_ATTR_ARCHIVE) #define BOOTSEC_JUMP_BOOT_LEN 3 #define BOOTSEC_FS_NAME_LEN 8 diff --git a/file.c b/file.c index 216c38b981e1..59848fa2793d 100644 --- a/file.c +++ b/file.c @@ -14,6 +14,7 @@ #include #include #include +#include #include "exfat_raw.h" #include "exfat_fs.h" @@ -154,7 +155,7 @@ int __exfat_truncate(struct inode *inode) } if (ei->type == TYPE_FILE) - ei->attr |= ATTR_ARCHIVE; + ei->attr |= EXFAT_ATTR_ARCHIVE; /* * update the directory entry @@ -443,8 +444,9 @@ static int exfat_ioctl_set_attributes(struct file *file, u32 __user *user_attr) /* * Mask attributes so we don't set reserved fields. */ - attr &= (ATTR_READONLY | ATTR_HIDDEN | ATTR_SYSTEM | ATTR_ARCHIVE); - attr |= (is_dir ? ATTR_SUBDIR : 0); + attr &= (EXFAT_ATTR_READONLY | EXFAT_ATTR_HIDDEN | EXFAT_ATTR_SYSTEM | + EXFAT_ATTR_ARCHIVE); + attr |= (is_dir ? EXFAT_ATTR_SUBDIR : 0); /* Equivalent to a chmod() */ ia.ia_valid = ATTR_MODE | ATTR_CTIME; @@ -459,12 +461,12 @@ static int exfat_ioctl_set_attributes(struct file *file, u32 __user *user_attr) ia.ia_mode = exfat_make_mode(sbi, attr, 0666 | (inode->i_mode & 0111)); /* The root directory has no attributes */ - if (inode->i_ino == EXFAT_ROOT_INO && attr != ATTR_SUBDIR) { + if (inode->i_ino == EXFAT_ROOT_INO && attr != EXFAT_ATTR_SUBDIR) { err = -EINVAL; goto out_unlock_inode; } - if (((attr | oldattr) & ATTR_SYSTEM) && + if (((attr | oldattr) & EXFAT_ATTR_SYSTEM) && !capable(CAP_LINUX_IMMUTABLE)) { err = -EPERM; goto out_unlock_inode; @@ -562,9 +564,9 @@ long exfat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) u32 __user *user_attr = (u32 __user *)arg; switch (cmd) { - case EXFAT_IOCTL_GET_ATTRIBUTES: + case FAT_IOCTL_GET_ATTRIBUTES: return exfat_ioctl_get_attributes(inode, user_attr); - case EXFAT_IOCTL_SET_ATTRIBUTES: + case FAT_IOCTL_SET_ATTRIBUTES: return exfat_ioctl_set_attributes(filp, user_attr); case FITRIM: return exfat_ioctl_fitrim(inode, arg); diff --git a/inode.c b/inode.c index 4d4449e1df3d..e4dffd734ba4 100644 --- a/inode.c +++ b/inode.c @@ -436,13 +436,13 @@ static int exfat_write_end(struct file *file, struct address_space *mapping, if (err < len) exfat_write_failed(mapping, pos+len); - if (!(err < 0) && !(ei->attr & ATTR_ARCHIVE)) { + if (!(err < 0) && !(ei->attr & EXFAT_ATTR_ARCHIVE)) { #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) inode->i_mtime = inode->i_ctime = current_time(inode); #else inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; #endif - ei->attr |= ATTR_ARCHIVE; + ei->attr |= EXFAT_ATTR_ARCHIVE; mark_inode_dirty(inode); } @@ -629,7 +629,7 @@ static int exfat_fill_inode(struct inode *inode, struct exfat_dir_entry *info) inode->i_generation = prandom_u32(); #endif - if (info->attr & ATTR_SUBDIR) { /* directory */ + if (info->attr & EXFAT_ATTR_SUBDIR) { /* directory */ inode->i_generation &= ~1; inode->i_mode = exfat_make_mode(sbi, info->attr, 0777); inode->i_op = &exfat_dir_inode_operations; diff --git a/namei.c b/namei.c index d43af646e229..5b2204cb2e4b 100644 --- a/namei.c +++ b/namei.c @@ -572,12 +572,12 @@ static int exfat_add_entry(struct inode *inode, const char *path, info->type = type; if (type == TYPE_FILE) { - info->attr = ATTR_ARCHIVE; + info->attr = EXFAT_ATTR_ARCHIVE; info->start_clu = EXFAT_EOF_CLUSTER; info->size = 0; info->num_subdirs = 0; } else { - info->attr = ATTR_SUBDIR; + info->attr = EXFAT_ATTR_SUBDIR; if (sbi->options.zero_size_dir) info->start_clu = EXFAT_EOF_CLUSTER; else @@ -1177,8 +1177,8 @@ static int exfat_rename_file(struct inode *inode, struct exfat_chain *p_dir, *epnew = *epold; if (exfat_get_entry_type(epnew) == TYPE_FILE) { - epnew->dentry.file.attr |= cpu_to_le16(ATTR_ARCHIVE); - ei->attr |= ATTR_ARCHIVE; + epnew->dentry.file.attr |= cpu_to_le16(EXFAT_ATTR_ARCHIVE); + ei->attr |= EXFAT_ATTR_ARCHIVE; } exfat_update_bh(new_bh, sync); brelse(old_bh); @@ -1209,8 +1209,8 @@ static int exfat_rename_file(struct inode *inode, struct exfat_chain *p_dir, ei->entry = newentry; } else { if (exfat_get_entry_type(epold) == TYPE_FILE) { - epold->dentry.file.attr |= cpu_to_le16(ATTR_ARCHIVE); - ei->attr |= ATTR_ARCHIVE; + epold->dentry.file.attr |= cpu_to_le16(EXFAT_ATTR_ARCHIVE); + ei->attr |= EXFAT_ATTR_ARCHIVE; } exfat_update_bh(old_bh, sync); brelse(old_bh); @@ -1258,8 +1258,8 @@ static int exfat_move_file(struct inode *inode, struct exfat_chain *p_olddir, *epnew = *epmov; if (exfat_get_entry_type(epnew) == TYPE_FILE) { - epnew->dentry.file.attr |= cpu_to_le16(ATTR_ARCHIVE); - ei->attr |= ATTR_ARCHIVE; + epnew->dentry.file.attr |= cpu_to_le16(EXFAT_ATTR_ARCHIVE); + ei->attr |= EXFAT_ATTR_ARCHIVE; } exfat_update_bh(new_bh, IS_DIRSYNC(inode)); brelse(mov_bh); diff --git a/super.c b/super.c index 89f60d722c99..caf8fd4acf33 100644 --- a/super.c +++ b/super.c @@ -640,7 +640,7 @@ static int exfat_read_root(struct inode *inode) inode->i_version++; #endif inode->i_generation = 0; - inode->i_mode = exfat_make_mode(sbi, ATTR_SUBDIR, 0777); + inode->i_mode = exfat_make_mode(sbi, EXFAT_ATTR_SUBDIR, 0777); inode->i_op = &exfat_dir_inode_operations; inode->i_fop = &exfat_dir_operations; @@ -649,7 +649,7 @@ static int exfat_read_root(struct inode *inode) ei->i_size_aligned = i_size_read(inode); ei->i_size_ondisk = i_size_read(inode); - exfat_save_attr(inode, ATTR_SUBDIR); + exfat_save_attr(inode, EXFAT_ATTR_SUBDIR); #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) inode->i_mtime = inode->i_atime = inode->i_ctime = ei->i_crtime = current_time(inode); From 98f36315064b54bf79e5db44839b800a6eccbb98 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 2 Nov 2023 22:52:14 +0900 Subject: [PATCH 07/74] exfat: fs: add CONFIG_BUFFER_HEAD Add a new config option that controls building the buffer_head code, and select it from all file systems and stacking drivers that need it. For the block device nodes and alternative iomap based buffered I/O path is provided when buffer_head support is not enabled, and iomap needs a a small tweak to define the IOMAP_F_BUFFER_HEAD flag to 0 to not call into the buffer_head code when it doesn't exist. Otherwise this is just Kconfig and ifdef changes. Signed-off-by: Christoph Hellwig Reviewed-by: Luis Chamberlain Reviewed-by: Johannes Thumshirn Link: https://lore.kernel.org/r/20230801172201.1923299-7-hch@lst.de Signed-off-by: Jens Axboe Signed-off-by: Namjae Jeon --- Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/Kconfig b/Kconfig index b3e5e9ebabc8..e298907a829a 100644 --- a/Kconfig +++ b/Kconfig @@ -2,6 +2,7 @@ config EXFAT_FS tristate "exFAT filesystem support" + select BUFFER_HEAD select NLS select LEGACY_DIRECT_IO help From 6eb9af0626759d340afd146a7c0941975edd4eea Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Thu, 2 Nov 2023 23:03:27 +0900 Subject: [PATCH 08/74] exfat: ensure that ctime is updated whenever the mtime is When removing entries from a directory, the ctime must also be updated alongside the mtime. Signed-off-by: Jeff Layton Message-Id: <20230705190309.579783-4-jlayton@kernel.org> Signed-off-by: Christian Brauner Signed-off-by: Namjae Jeon --- namei.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/namei.c b/namei.c index 5b2204cb2e4b..01d1aefa1670 100644 --- a/namei.c +++ b/namei.c @@ -903,9 +903,9 @@ static int exfat_unlink(struct inode *dir, struct dentry *dentry) dir->i_version++; #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) - dir->i_mtime = dir->i_atime = current_time(dir); + dir->i_mtime = dir->i_atime = dir->i_ctime = current_time(dir); #else - dir->i_mtime = dir->i_atime = CURRENT_TIME_SEC; + dir->i_mtime = dir->i_atime = dir->i_ctime = CURRENT_TIME_SEC; #endif exfat_truncate_atime(&dir->i_atime); if (IS_DIRSYNC(dir)) @@ -915,9 +915,9 @@ static int exfat_unlink(struct inode *dir, struct dentry *dentry) clear_nlink(inode); #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) - inode->i_mtime = inode->i_atime = current_time(inode); + inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); #else - inode->i_mtime = inode->i_atime = CURRENT_TIME_SEC; + inode->i_mtime = inode->i_atime = dir->i_ctime = CURRENT_TIME_SEC; #endif exfat_truncate_atime(&inode->i_atime); exfat_unhash_inode(inode); @@ -1111,9 +1111,9 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry) dir->i_version++; #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) - dir->i_mtime = dir->i_atime = current_time(dir); + dir->i_mtime = dir->i_atime = dir->i_ctime = current_time(dir); #else - dir->i_mtime = dir->i_atime = CURRENT_TIME_SEC; + dir->i_mtime = dir->i_atime = inode->i_ctime = CURRENT_TIME_SEC; #endif exfat_truncate_atime(&dir->i_atime); if (IS_DIRSYNC(dir)) @@ -1124,9 +1124,9 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry) clear_nlink(inode); #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) - inode->i_mtime = inode->i_atime = current_time(inode); + inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); #else - inode->i_mtime = inode->i_atime = CURRENT_TIME_SEC; + inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; #endif exfat_truncate_atime(&inode->i_atime); exfat_unhash_inode(inode); From 308d7f68ce8b4285ff2eb5795b86f0a7c23678c7 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Fri, 3 Nov 2023 14:15:57 +0900 Subject: [PATCH 09/74] exfat: convert to simple_rename_timestamp A rename potentially involves updating 4 different inode timestamps. Convert to the new simple_rename_timestamp helper function. Signed-off-by: Jeff Layton Message-Id: <20230705190309.579783-10-jlayton@kernel.org> Signed-off-by: Christian Brauner Signed-off-by: Namjae Jeon --- namei.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/namei.c b/namei.c index 01d1aefa1670..ba6c604a48dd 100644 --- a/namei.c +++ b/namei.c @@ -1481,8 +1481,13 @@ static int exfat_rename(struct inode *old_dir, struct dentry *old_dentry, new_dir->i_version++; #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry); + EXFAT_I(new_dir)->i_crtime = current_time(new_dir); +#else new_dir->i_ctime = new_dir->i_mtime = new_dir->i_atime = EXFAT_I(new_dir)->i_crtime = current_time(new_dir); +#endif #else new_dir->i_ctime = new_dir->i_mtime = new_dir->i_atime = EXFAT_I(new_dir)->i_crtime = CURRENT_TIME_SEC; @@ -1514,7 +1519,9 @@ static int exfat_rename(struct inode *old_dir, struct dentry *old_dentry, old_dir->i_version++; #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE <= KERNEL_VERSION(6, 6, 0) old_dir->i_ctime = old_dir->i_mtime = current_time(old_dir); +#endif #else old_dir->i_ctime = old_dir->i_mtime = CURRENT_TIME_SEC; #endif From c49f06ffabbbbc82d6693ae8f43ef2bfb1fa509b Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Thu, 2 Nov 2023 23:15:43 +0900 Subject: [PATCH 10/74] exfat: fs: pass the request_mask to generic_fillattr generic_fillattr just fills in the entire stat struct indiscriminately today, copying data from the inode. There is at least one attribute (STATX_CHANGE_COOKIE) that can have side effects when it is reported, and we're looking at adding more with the addition of multigrain timestamps. Add a request_mask argument to generic_fillattr and have most callers just pass in the value that is passed to getattr. Have other callers (e.g. ksmbd) just pass in STATX_BASIC_STATS. Also move the setting of STATX_CHANGE_COOKIE into generic_fillattr. Acked-by: Joseph Qi Reviewed-by: Xiubo Li Reviewed-by: "Paulo Alcantara (SUSE)" Reviewed-by: Jan Kara Signed-off-by: Jeff Layton Message-Id: <20230807-mgctime-v7-2-d1dec143a704@kernel.org> Signed-off-by: Christian Brauner Signed-off-by: Namjae Jeon --- file.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/file.c b/file.c index 59848fa2793d..c82e5792e504 100644 --- a/file.c +++ b/file.c @@ -269,7 +269,11 @@ int exfat_getattr(struct vfsmount *mnt, struct dentry *dentry, #if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 12, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 3, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + generic_fillattr(&nop_mnt_idmap, request_mask, inode, stat); +#else generic_fillattr(&nop_mnt_idmap, inode, stat); +#endif #else generic_fillattr(&init_user_ns, inode, stat); #endif From 0e040b023e2fb8e39719134533bbf683d262d35c Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Thu, 2 Nov 2023 23:32:26 +0900 Subject: [PATCH 11/74] exfat: convert to ctime accessor functions In later patches, we're going to change how the inode's ctime field is used. Switch to using accessor functions instead of raw accesses of inode->i_ctime. Signed-off-by: Jeff Layton Reviewed-by: Jan Kara Message-Id: <20230705190309.579783-38-jlayton@kernel.org> Signed-off-by: Christian Brauner Signed-off-by: Namjae Jeon --- file.c | 8 ++++++++ inode.c | 12 ++++++++++++ namei.c | 36 ++++++++++++++++++++++++++++++++++++ super.c | 4 ++++ 4 files changed, 60 insertions(+) diff --git a/file.c b/file.c index c82e5792e504..2b74bd03e55e 100644 --- a/file.c +++ b/file.c @@ -30,7 +30,11 @@ static int exfat_cont_expand(struct inode *inode, loff_t size) return err; #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + inode->i_mtime = inode_set_ctime_current(inode); +#else inode->i_ctime = inode->i_mtime = current_time(inode); +#endif #else inode->i_ctime = inode->i_mtime = CURRENT_TIME_SEC; #endif @@ -363,7 +367,11 @@ int exfat_setattr(struct dentry *dentry, struct iattr *attr) if (attr->ia_valid & ATTR_SIZE) #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + inode->i_mtime = inode_set_ctime_current(inode); +#else inode->i_mtime = inode->i_ctime = current_time(inode); +#endif #else inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; #endif diff --git a/inode.c b/inode.c index e4dffd734ba4..89e341e74640 100644 --- a/inode.c +++ b/inode.c @@ -380,7 +380,11 @@ static void exfat_write_failed(struct address_space *mapping, loff_t to) if (to > i_size_read(inode)) { truncate_pagecache(inode, i_size_read(inode)); #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + inode->i_mtime = inode_set_ctime_current(inode); +#else inode->i_mtime = inode->i_ctime = current_time(inode); +#endif #else inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; #endif @@ -438,7 +442,11 @@ static int exfat_write_end(struct file *file, struct address_space *mapping, if (!(err < 0) && !(ei->attr & EXFAT_ATTR_ARCHIVE)) { #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + inode->i_mtime = inode_set_ctime_current(inode); +#else inode->i_mtime = inode->i_ctime = current_time(inode); +#endif #else inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; #endif @@ -659,7 +667,11 @@ static int exfat_fill_inode(struct inode *inode, struct exfat_dir_entry *info) inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> 9; inode->i_mtime = info->mtime; +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + inode_set_ctime_to_ts(inode, info->mtime); +#else inode->i_ctime = info->mtime; +#endif ei->i_crtime = info->crtime; inode->i_atime = info->atime; diff --git a/namei.c b/namei.c index ba6c604a48dd..0dd33a0cfddc 100644 --- a/namei.c +++ b/namei.c @@ -625,7 +625,11 @@ static int exfat_create(struct inode *dir, struct dentry *dentry, umode_t mode, dir->i_version++; #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + dir->i_mtime = inode_set_ctime_current(dir); +#else dir->i_ctime = dir->i_mtime = current_time(dir); +#endif #else dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC; #endif @@ -647,8 +651,12 @@ static int exfat_create(struct inode *dir, struct dentry *dentry, umode_t mode, #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + inode->i_mtime = inode->i_atime = EXFAT_I(inode)->i_crtime = inode_set_ctime_current(inode); +#else inode->i_mtime = inode->i_atime = inode->i_ctime = EXFAT_I(inode)->i_crtime = current_time(inode); +#endif #else inode->i_mtime = inode->i_atime = inode->i_ctime = EXFAT_I(inode)->i_crtime = CURRENT_TIME_SEC; @@ -903,7 +911,11 @@ static int exfat_unlink(struct inode *dir, struct dentry *dentry) dir->i_version++; #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + dir->i_mtime = dir->i_atime = inode_set_ctime_current(dir); +#else dir->i_mtime = dir->i_atime = dir->i_ctime = current_time(dir); +#endif #else dir->i_mtime = dir->i_atime = dir->i_ctime = CURRENT_TIME_SEC; #endif @@ -915,7 +927,11 @@ static int exfat_unlink(struct inode *dir, struct dentry *dentry) clear_nlink(inode); #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + inode->i_mtime = inode->i_atime = inode_set_ctime_current(inode); +#else inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); +#endif #else inode->i_mtime = inode->i_atime = dir->i_ctime = CURRENT_TIME_SEC; #endif @@ -964,7 +980,11 @@ static int exfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + dir->i_mtime = inode_set_ctime_current(dir); +#else dir->i_ctime = dir->i_mtime = current_time(dir); +#endif #else dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC; #endif @@ -986,8 +1006,12 @@ static int exfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) inode->i_version++; #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + inode->i_mtime = inode->i_atime = EXFAT_I(inode)->i_crtime = inode_set_ctime_current(inode); +#else inode->i_mtime = inode->i_atime = inode->i_ctime = EXFAT_I(inode)->i_crtime = current_time(inode); +#endif #else inode->i_mtime = inode->i_atime = inode->i_ctime = EXFAT_I(inode)->i_crtime = CURRENT_TIME_SEC; @@ -1111,7 +1135,11 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry) dir->i_version++; #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + dir->i_mtime = dir->i_atime = inode_set_ctime_current(dir); +#else dir->i_mtime = dir->i_atime = dir->i_ctime = current_time(dir); +#endif #else dir->i_mtime = dir->i_atime = inode->i_ctime = CURRENT_TIME_SEC; #endif @@ -1124,7 +1152,11 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry) clear_nlink(inode); #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + inode->i_mtime = inode->i_atime = inode_set_ctime_current(inode); +#else inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); +#endif #else inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; #endif @@ -1543,8 +1575,12 @@ static int exfat_rename(struct inode *old_dir, struct dentry *old_dentry, WARN_ON(new_inode->i_nlink == 0); } #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + EXFAT_I(new_inode)->i_crtime = current_time(new_inode); +#else new_inode->i_ctime = EXFAT_I(new_inode)->i_crtime = current_time(new_inode); +#endif #else new_inode->i_ctime = EXFAT_I(new_inode)->i_crtime = CURRENT_TIME_SEC; diff --git a/super.c b/super.c index caf8fd4acf33..0b09959c26e6 100644 --- a/super.c +++ b/super.c @@ -651,8 +651,12 @@ static int exfat_read_root(struct inode *inode) exfat_save_attr(inode, EXFAT_ATTR_SUBDIR); #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) + inode->i_mtime = inode->i_atime = ei->i_crtime = inode_set_ctime_current(inode); +#else inode->i_mtime = inode->i_atime = inode->i_ctime = ei->i_crtime = current_time(inode); +#endif #else inode->i_mtime = inode->i_atime = inode->i_ctime = ei->i_crtime = CURRENT_TIME_SEC; From 59b349e724db3d04318735e2484f2f5aab1cdedd Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Thu, 2 Nov 2023 23:54:56 +0900 Subject: [PATCH 12/74] exfat: convert to new timestamp accessors Convert to using the new inode timestamp accessor functions. Signed-off-by: Jeff Layton Link: https://lore.kernel.org/r/20231004185347.80880-31-jlayton@kernel.org Signed-off-by: Christian Brauner Signed-off-by: Namjae Jeon --- exfat_fs.h | 3 +++ file.c | 12 ++++++++++++ inode.c | 34 ++++++++++++++++++++++++++++++++++ misc.c | 10 ++++++++++ namei.c | 43 +++++++++++++++++++++++++++++++++++++++++++ super.c | 5 +++++ 6 files changed, 107 insertions(+) diff --git a/exfat_fs.h b/exfat_fs.h index 9c2266afe3f6..124dfd599c11 100644 --- a/exfat_fs.h +++ b/exfat_fs.h @@ -598,6 +598,9 @@ void __exfat_fs_error(struct super_block *sb, int report, const char *fmt, ...) void exfat_get_entry_time(struct exfat_sb_info *sbi, struct timespec64 *ts, u8 tz, __le16 time, __le16 date, u8 time_cs); void exfat_truncate_atime(struct timespec64 *ts); +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) +void exfat_truncate_inode_atime(struct inode *inode); +#endif void exfat_set_entry_time(struct exfat_sb_info *sbi, struct timespec64 *ts, u8 *tz, __le16 *time, __le16 *date, u8 *time_cs); #else diff --git a/file.c b/file.c index 2b74bd03e55e..72879e405531 100644 --- a/file.c +++ b/file.c @@ -31,7 +31,11 @@ static int exfat_cont_expand(struct inode *inode, loff_t size) #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + inode_set_mtime_to_ts(inode, inode_set_ctime_current(inode)); +#else inode->i_mtime = inode_set_ctime_current(inode); +#endif #else inode->i_ctime = inode->i_mtime = current_time(inode); #endif @@ -368,7 +372,11 @@ int exfat_setattr(struct dentry *dentry, struct iattr *attr) if (attr->ia_valid & ATTR_SIZE) #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + inode_set_mtime_to_ts(inode, inode_set_ctime_current(inode)); +#else inode->i_mtime = inode_set_ctime_current(inode); +#endif #else inode->i_mtime = inode->i_ctime = current_time(inode); #endif @@ -376,6 +384,9 @@ int exfat_setattr(struct dentry *dentry, struct iattr *attr) inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; #endif +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + exfat_truncate_inode_atime(inode); +#else #if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 12, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 3, 0) setattr_copy(&nop_mnt_idmap, inode, attr); @@ -386,6 +397,7 @@ int exfat_setattr(struct dentry *dentry, struct iattr *attr) setattr_copy(inode, attr); #endif exfat_truncate_atime(&inode->i_atime); +#endif if (attr->ia_valid & ATTR_SIZE) { error = exfat_block_truncate_page(inode, attr->ia_size); diff --git a/inode.c b/inode.c index 89e341e74640..7d7f2cff8a05 100644 --- a/inode.c +++ b/inode.c @@ -28,6 +28,9 @@ int __exfat_write_inode(struct inode *inode, int sync) struct exfat_sb_info *sbi = EXFAT_SB(sb); struct exfat_inode_info *ei = EXFAT_I(inode); bool is_dir = (ei->type == TYPE_DIR) ? true : false; +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + struct timespec64 ts; +#endif if (inode->i_ino == EXFAT_ROOT_INO) return 0; @@ -57,6 +60,20 @@ int __exfat_write_inode(struct inode *inode, int sync) &ep->dentry.file.create_time, &ep->dentry.file.create_date, &ep->dentry.file.create_time_cs); +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + exfat_set_entry_time(sbi, &ts, + &ep->dentry.file.modify_tz, + &ep->dentry.file.modify_time, + &ep->dentry.file.modify_date, + &ep->dentry.file.modify_time_cs); + inode_set_mtime_to_ts(inode, ts); + exfat_set_entry_time(sbi, &ts, + &ep->dentry.file.access_tz, + &ep->dentry.file.access_time, + &ep->dentry.file.access_date, + NULL); + inode_set_atime_to_ts(inode, ts); +#else exfat_set_entry_time(sbi, &inode->i_mtime, &ep->dentry.file.modify_tz, &ep->dentry.file.modify_time, @@ -67,6 +84,7 @@ int __exfat_write_inode(struct inode *inode, int sync) &ep->dentry.file.access_time, &ep->dentry.file.access_date, NULL); +#endif /* File size should be zero if there is no cluster allocated */ on_disk_size = i_size_read(inode); @@ -381,7 +399,11 @@ static void exfat_write_failed(struct address_space *mapping, loff_t to) truncate_pagecache(inode, i_size_read(inode)); #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + inode_set_mtime_to_ts(inode, inode_set_ctime_current(inode)); +#else inode->i_mtime = inode_set_ctime_current(inode); +#endif #else inode->i_mtime = inode->i_ctime = current_time(inode); #endif @@ -443,7 +465,11 @@ static int exfat_write_end(struct file *file, struct address_space *mapping, if (!(err < 0) && !(ei->attr & EXFAT_ATTR_ARCHIVE)) { #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + inode_set_mtime_to_ts(inode, inode_set_ctime_current(inode)); +#else inode->i_mtime = inode_set_ctime_current(inode); +#endif #else inode->i_mtime = inode->i_ctime = current_time(inode); #endif @@ -666,14 +692,22 @@ static int exfat_fill_inode(struct inode *inode, struct exfat_dir_entry *info) exfat_save_attr(inode, info->attr); inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> 9; +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + inode_set_mtime_to_ts(inode, info->mtime); +#else inode->i_mtime = info->mtime; +#endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) inode_set_ctime_to_ts(inode, info->mtime); #else inode->i_ctime = info->mtime; #endif ei->i_crtime = info->crtime; +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + inode_set_atime_to_ts(inode, info->atime); +#else inode->i_atime = info->atime; +#endif return 0; } diff --git a/misc.c b/misc.c index e7ae6f629d6e..a744a1fb1e33 100644 --- a/misc.c +++ b/misc.c @@ -163,6 +163,16 @@ void exfat_truncate_atime(struct timespec *ts) ts->tv_nsec = 0; } +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) +void exfat_truncate_inode_atime(struct inode *inode) +{ + struct timespec64 atime = inode_get_atime(inode); + + exfat_truncate_atime(&atime); + inode_set_atime_to_ts(inode, atime); +} +#endif + u16 exfat_calc_chksum16(void *data, int len, u16 chksum, int type) { int i; diff --git a/namei.c b/namei.c index 0dd33a0cfddc..ef9a6df20998 100644 --- a/namei.c +++ b/namei.c @@ -626,7 +626,11 @@ static int exfat_create(struct inode *dir, struct dentry *dentry, umode_t mode, #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + inode_set_mtime_to_ts(dir, inode_set_ctime_current(dir)); +#else dir->i_mtime = inode_set_ctime_current(dir); +#endif #else dir->i_ctime = dir->i_mtime = current_time(dir); #endif @@ -650,6 +654,10 @@ static int exfat_create(struct inode *dir, struct dentry *dentry, umode_t mode, inode->i_version++; #endif +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + EXFAT_I(inode)->i_crtime = simple_inode_init_ts(inode); + exfat_truncate_inode_atime(inode); +#else #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) inode->i_mtime = inode->i_atime = EXFAT_I(inode)->i_crtime = inode_set_ctime_current(inode); @@ -662,6 +670,7 @@ static int exfat_create(struct inode *dir, struct dentry *dentry, umode_t mode, EXFAT_I(inode)->i_crtime = CURRENT_TIME_SEC; #endif exfat_truncate_atime(&inode->i_atime); +#endif /* timestamp is already written, so mark_inode_dirty() is unneeded. */ d_instantiate(dentry, inode); @@ -910,6 +919,11 @@ static int exfat_unlink(struct inode *dir, struct dentry *dentry) #else dir->i_version++; #endif + +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + simple_inode_init_ts(dir); + exfat_truncate_inode_atime(dir); +#else #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) dir->i_mtime = dir->i_atime = inode_set_ctime_current(dir); @@ -920,12 +934,17 @@ static int exfat_unlink(struct inode *dir, struct dentry *dentry) dir->i_mtime = dir->i_atime = dir->i_ctime = CURRENT_TIME_SEC; #endif exfat_truncate_atime(&dir->i_atime); +#endif if (IS_DIRSYNC(dir)) exfat_sync_inode(dir); else mark_inode_dirty(dir); clear_nlink(inode); +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + simple_inode_init_ts(inode); + exfat_truncate_inode_atime(inode); +#else #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) inode->i_mtime = inode->i_atime = inode_set_ctime_current(inode); @@ -936,6 +955,7 @@ static int exfat_unlink(struct inode *dir, struct dentry *dentry) inode->i_mtime = inode->i_atime = dir->i_ctime = CURRENT_TIME_SEC; #endif exfat_truncate_atime(&inode->i_atime); +#endif exfat_unhash_inode(inode); #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 16, 0) exfat_d_version_set(dentry, inode_query_iversion(dir)); @@ -981,7 +1001,11 @@ static int exfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + inode_set_mtime_to_ts(dir, inode_set_ctime_current(dir)); +#else dir->i_mtime = inode_set_ctime_current(dir); +#endif #else dir->i_ctime = dir->i_mtime = current_time(dir); #endif @@ -1005,6 +1029,10 @@ static int exfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) #else inode->i_version++; #endif +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + EXFAT_I(inode)->i_crtime = simple_inode_init_ts(inode); + exfat_truncate_inode_atime(inode); +#else #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) inode->i_mtime = inode->i_atime = EXFAT_I(inode)->i_crtime = inode_set_ctime_current(inode); @@ -1017,6 +1045,7 @@ static int exfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) EXFAT_I(inode)->i_crtime = CURRENT_TIME_SEC; #endif exfat_truncate_atime(&inode->i_atime); +#endif /* timestamp is already written, so mark_inode_dirty() is unneeded. */ d_instantiate(dentry, inode); @@ -1134,6 +1163,10 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry) #else dir->i_version++; #endif +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + simple_inode_init_ts(dir); + exfat_truncate_inode_atime(dir); +#else #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) dir->i_mtime = dir->i_atime = inode_set_ctime_current(dir); @@ -1144,6 +1177,7 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry) dir->i_mtime = dir->i_atime = inode->i_ctime = CURRENT_TIME_SEC; #endif exfat_truncate_atime(&dir->i_atime); +#endif if (IS_DIRSYNC(dir)) exfat_sync_inode(dir); else @@ -1151,6 +1185,10 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry) drop_nlink(dir); clear_nlink(inode); +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + simple_inode_init_ts(inode); + exfat_truncate_inode_atime(inode); +#else #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) inode->i_mtime = inode->i_atime = inode_set_ctime_current(inode); @@ -1161,6 +1199,7 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry) inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; #endif exfat_truncate_atime(&inode->i_atime); +#endif exfat_unhash_inode(inode); #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 16, 0) exfat_d_version_set(dentry, inode_query_iversion(dir)); @@ -1524,7 +1563,11 @@ static int exfat_rename(struct inode *old_dir, struct dentry *old_dentry, new_dir->i_ctime = new_dir->i_mtime = new_dir->i_atime = EXFAT_I(new_dir)->i_crtime = CURRENT_TIME_SEC; #endif +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + exfat_truncate_inode_atime(new_dir); +#else exfat_truncate_atime(&new_dir->i_atime); +#endif if (IS_DIRSYNC(new_dir)) exfat_sync_inode(new_dir); else diff --git a/super.c b/super.c index 0b09959c26e6..decf1c68323b 100644 --- a/super.c +++ b/super.c @@ -650,6 +650,10 @@ static int exfat_read_root(struct inode *inode) ei->i_size_ondisk = i_size_read(inode); exfat_save_attr(inode, EXFAT_ATTR_SUBDIR); +#if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + ei->i_crtime = simple_inode_init_ts(inode); + exfat_truncate_inode_atime(inode); +#else #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0) #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 6, 0) inode->i_mtime = inode->i_atime = ei->i_crtime = inode_set_ctime_current(inode); @@ -662,6 +666,7 @@ static int exfat_read_root(struct inode *inode) CURRENT_TIME_SEC; #endif exfat_truncate_atime(&inode->i_atime); +#endif return 0; } From 327699f41c880e71baf82b87d60bac73f78dda87 Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Fri, 3 Nov 2023 08:51:38 +0900 Subject: [PATCH 13/74] exfat: fix setting uninitialized time to ctime/atime An uninitialized time is set to ctime/atime in __exfat_write_inode(). It causes xfstests generic/003 and generic/192 to fail. And since there will be a time gap between setting ctime/atime to the inode and writing back the inode, so ctime/atime should not be set again when writing back the inode. Fixes: 4c72a36edd54 ("exfat: convert to new timestamp accessors") Signed-off-by: Yuezhang Mo Reviewed-by: Andy Wu Reviewed-by: Aoyama Wataru Reviewed-by: Sungjong Seo Signed-off-by: Namjae Jeon --- inode.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/inode.c b/inode.c index 7d7f2cff8a05..479db4299c83 100644 --- a/inode.c +++ b/inode.c @@ -61,18 +61,18 @@ int __exfat_write_inode(struct inode *inode, int sync) &ep->dentry.file.create_date, &ep->dentry.file.create_time_cs); #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + ts = inode_get_mtime(inode); exfat_set_entry_time(sbi, &ts, &ep->dentry.file.modify_tz, &ep->dentry.file.modify_time, &ep->dentry.file.modify_date, &ep->dentry.file.modify_time_cs); - inode_set_mtime_to_ts(inode, ts); + ts = inode_get_atime(inode); exfat_set_entry_time(sbi, &ts, &ep->dentry.file.access_tz, &ep->dentry.file.access_time, &ep->dentry.file.access_date, NULL); - inode_set_atime_to_ts(inode, ts); #else exfat_set_entry_time(sbi, &inode->i_mtime, &ep->dentry.file.modify_tz, From 3ad149047d1acf158f327b3f0279d3d8b03bc894 Mon Sep 17 00:00:00 2001 From: Yuezhang Mo Date: Fri, 3 Nov 2023 08:52:47 +0900 Subject: [PATCH 14/74] exfat: fix ctime is not updated Commit 4c72a36edd54 removed attr_copy() from exfat_set_attr(). It causes xfstests generic/221 to fail. In xfstests generic/221, it tests ctime should be updated even if futimens() update atime only. But in this case, ctime will not be updated if attr_copy() is removed. attr_copy() may also update other attributes, and removing it may cause other bugs, so this commit restores to call attr_copy() in exfat_set_attr(). Fixes: 4c72a36edd54 ("exfat: convert to new timestamp accessors") Signed-off-by: Yuezhang Mo Reviewed-by: Andy Wu Reviewed-by: Aoyama Wataru Reviewed-by: Sungjong Seo Signed-off-by: Namjae Jeon --- file.c | 1 + 1 file changed, 1 insertion(+) diff --git a/file.c b/file.c index 72879e405531..749c349cbfba 100644 --- a/file.c +++ b/file.c @@ -385,6 +385,7 @@ int exfat_setattr(struct dentry *dentry, struct iattr *attr) #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(6, 7, 0) + setattr_copy(&nop_mnt_idmap, inode, attr); exfat_truncate_inode_atime(inode); #else #if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 12, 0) From 2a0d7a848f60be6a16202ac60328d9ae17757665 Mon Sep 17 00:00:00 2001 From: John Sanpe Date: Fri, 29 Dec 2023 23:04:09 +0900 Subject: [PATCH 15/74] exfat: using hweight instead of internal logic Replace the internal table lookup algorithm with the hweight library, which has instruction set acceleration capabilities. Use it to increase the length of a single calculation of the exfat_find_free_bitmap function to the long type. Signed-off-by: John Sanpe Signed-off-by: Namjae Jeon --- balloc.c | 48 +++++++++++++++++++++--------------------------- 1 file changed, 21 insertions(+), 27 deletions(-) diff --git a/balloc.c b/balloc.c index c226d5d1c36b..e602bea72226 100644 --- a/balloc.c +++ b/balloc.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 11, 0) #include @@ -17,6 +18,16 @@ #include "exfat_raw.h" #include "exfat_fs.h" +#if BITS_PER_LONG == 32 +#define __le_long __le32 +#define lel_to_cpu(A) le32_to_cpu(A) +#elif BITS_PER_LONG == 64 +#define __le_long __le64 +#define lel_to_cpu(A) le64_to_cpu(A) +#else +#error "BITS_PER_LONG not 32 or 64" +#endif + static const unsigned char free_bit[] = { 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 4, 0, 1, 0, 2,/* 0 ~ 19*/ 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 5, 0, 1, 0, 2, 0, 1, 0, 3,/* 20 ~ 39*/ @@ -33,22 +44,6 @@ static const unsigned char free_bit[] = { 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0 /*240 ~ 254*/ }; -static const unsigned char used_bit[] = { - 0, 1, 1, 2, 1, 2, 2, 3, 1, 2, 2, 3, 2, 3, 3, 4, 1, 2, 2, 3,/* 0 ~ 19*/ - 2, 3, 3, 4, 2, 3, 3, 4, 3, 4, 4, 5, 1, 2, 2, 3, 2, 3, 3, 4,/* 20 ~ 39*/ - 2, 3, 3, 4, 3, 4, 4, 5, 2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5,/* 40 ~ 59*/ - 4, 5, 5, 6, 1, 2, 2, 3, 2, 3, 3, 4, 2, 3, 3, 4, 3, 4, 4, 5,/* 60 ~ 79*/ - 2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6, 2, 3, 3, 4,/* 80 ~ 99*/ - 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6, 3, 4, 4, 5, 4, 5, 5, 6,/*100 ~ 119*/ - 4, 5, 5, 6, 5, 6, 6, 7, 1, 2, 2, 3, 2, 3, 3, 4, 2, 3, 3, 4,/*120 ~ 139*/ - 3, 4, 4, 5, 2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6,/*140 ~ 159*/ - 2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6, 3, 4, 4, 5,/*160 ~ 179*/ - 4, 5, 5, 6, 4, 5, 5, 6, 5, 6, 6, 7, 2, 3, 3, 4, 3, 4, 4, 5,/*180 ~ 199*/ - 3, 4, 4, 5, 4, 5, 5, 6, 3, 4, 4, 5, 4, 5, 5, 6, 4, 5, 5, 6,/*200 ~ 219*/ - 5, 6, 6, 7, 3, 4, 4, 5, 4, 5, 5, 6, 4, 5, 5, 6, 5, 6, 6, 7,/*220 ~ 239*/ - 4, 5, 5, 6, 5, 6, 6, 7, 5, 6, 6, 7, 6, 7, 7, 8 /*240 ~ 255*/ -}; - /* * Allocation Bitmap Management Functions */ @@ -255,25 +250,24 @@ int exfat_count_used_clusters(struct super_block *sb, unsigned int *ret_count) unsigned int count = 0; unsigned int i, map_i = 0, map_b = 0; unsigned int total_clus = EXFAT_DATA_CLUSTER_COUNT(sbi); - unsigned int last_mask = total_clus & BITS_PER_BYTE_MASK; - unsigned char clu_bits; - const unsigned char last_bit_mask[] = {0, 0b00000001, 0b00000011, - 0b00000111, 0b00001111, 0b00011111, 0b00111111, 0b01111111}; + unsigned int last_mask = total_clus & (BITS_PER_LONG - 1); + unsigned long *bitmap, clu_bits; total_clus &= ~last_mask; - for (i = 0; i < total_clus; i += BITS_PER_BYTE) { - clu_bits = *(sbi->vol_amap[map_i]->b_data + map_b); - count += used_bit[clu_bits]; - if (++map_b >= (unsigned int)sb->s_blocksize) { + for (i = 0; i < total_clus; i += BITS_PER_LONG) { + bitmap = (void *)(sbi->vol_amap[map_i]->b_data + map_b); + count += hweight_long(*bitmap); + map_b += sizeof(long); + if (map_b >= (unsigned int)sb->s_blocksize) { map_i++; map_b = 0; } } if (last_mask) { - clu_bits = *(sbi->vol_amap[map_i]->b_data + map_b); - clu_bits &= last_bit_mask[last_mask]; - count += used_bit[clu_bits]; + bitmap = (void *)(sbi->vol_amap[map_i]->b_data + map_b); + clu_bits = lel_to_cpu(*(__le_long *)bitmap); + count += hweight_long(clu_bits & BITMAP_LAST_WORD_MASK(last_mask)); } *ret_count = count; From b1cc7d5aa34b0fdd67df01fc1696e0a25ab1d08c Mon Sep 17 00:00:00 2001 From: John Sanpe Date: Mon, 11 Dec 2023 09:26:13 +0900 Subject: [PATCH 16/74] exfat: using ffs instead of internal logic Replaced the internal table lookup algorithm with ffs of the bitops library with better performance. Use it to increase the single processing length of the exfat_find_free_bitmap function, from single-byte search to long type. Signed-off-by: John Sanpe Signed-off-by: Namjae Jeon --- balloc.c | 41 +++++++++++++++-------------------------- exfat_fs.h | 3 +-- 2 files changed, 16 insertions(+), 28 deletions(-) diff --git a/balloc.c b/balloc.c index e602bea72226..63967f60ce84 100644 --- a/balloc.c +++ b/balloc.c @@ -21,29 +21,15 @@ #if BITS_PER_LONG == 32 #define __le_long __le32 #define lel_to_cpu(A) le32_to_cpu(A) +#define cpu_to_lel(A) cpu_to_le32(A) #elif BITS_PER_LONG == 64 #define __le_long __le64 #define lel_to_cpu(A) le64_to_cpu(A) +#define cpu_to_lel(A) cpu_to_le64(A) #else #error "BITS_PER_LONG not 32 or 64" #endif -static const unsigned char free_bit[] = { - 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 4, 0, 1, 0, 2,/* 0 ~ 19*/ - 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 5, 0, 1, 0, 2, 0, 1, 0, 3,/* 20 ~ 39*/ - 0, 1, 0, 2, 0, 1, 0, 4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2,/* 40 ~ 59*/ - 0, 1, 0, 6, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 4,/* 60 ~ 79*/ - 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 5, 0, 1, 0, 2,/* 80 ~ 99*/ - 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 4, 0, 1, 0, 2, 0, 1, 0, 3,/*100 ~ 119*/ - 0, 1, 0, 2, 0, 1, 0, 7, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2,/*120 ~ 139*/ - 0, 1, 0, 4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 5,/*140 ~ 159*/ - 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 4, 0, 1, 0, 2,/*160 ~ 179*/ - 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 6, 0, 1, 0, 2, 0, 1, 0, 3,/*180 ~ 199*/ - 0, 1, 0, 2, 0, 1, 0, 4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2,/*200 ~ 219*/ - 0, 1, 0, 5, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 4,/*220 ~ 239*/ - 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0 /*240 ~ 254*/ -}; - /* * Allocation Bitmap Management Functions */ @@ -206,32 +192,35 @@ unsigned int exfat_find_free_bitmap(struct super_block *sb, unsigned int clu) { unsigned int i, map_i, map_b, ent_idx; unsigned int clu_base, clu_free; - unsigned char k, clu_mask; + unsigned long clu_bits, clu_mask; struct exfat_sb_info *sbi = EXFAT_SB(sb); + __le_long bitval; WARN_ON(clu < EXFAT_FIRST_CLUSTER); - ent_idx = CLUSTER_TO_BITMAP_ENT(clu); - clu_base = BITMAP_ENT_TO_CLUSTER(ent_idx & ~(BITS_PER_BYTE_MASK)); + ent_idx = ALIGN_DOWN(CLUSTER_TO_BITMAP_ENT(clu), BITS_PER_LONG); + clu_base = BITMAP_ENT_TO_CLUSTER(ent_idx); clu_mask = IGNORED_BITS_REMAINED(clu, clu_base); map_i = BITMAP_OFFSET_SECTOR_INDEX(sb, ent_idx); map_b = BITMAP_OFFSET_BYTE_IN_SECTOR(sb, ent_idx); for (i = EXFAT_FIRST_CLUSTER; i < sbi->num_clusters; - i += BITS_PER_BYTE) { - k = *(sbi->vol_amap[map_i]->b_data + map_b); + i += BITS_PER_LONG) { + bitval = *(__le_long *)(sbi->vol_amap[map_i]->b_data + map_b); if (clu_mask > 0) { - k |= clu_mask; + bitval |= cpu_to_lel(clu_mask); clu_mask = 0; } - if (k < 0xFF) { - clu_free = clu_base + free_bit[k]; + if (lel_to_cpu(bitval) != ULONG_MAX) { + clu_bits = lel_to_cpu(bitval); + clu_free = clu_base + ffz(clu_bits); if (clu_free < sbi->num_clusters) return clu_free; } - clu_base += BITS_PER_BYTE; + clu_base += BITS_PER_LONG; + map_b += sizeof(long); - if (++map_b >= sb->s_blocksize || + if (map_b >= sb->s_blocksize || clu_base >= sbi->num_clusters) { if (++map_i >= sbi->map_sectors) { clu_base = EXFAT_FIRST_CLUSTER; diff --git a/exfat_fs.h b/exfat_fs.h index 124dfd599c11..6bf1c67daefb 100644 --- a/exfat_fs.h +++ b/exfat_fs.h @@ -148,8 +148,7 @@ enum { #define BITMAP_OFFSET_BIT_IN_SECTOR(sb, ent) (ent & BITS_PER_SECTOR_MASK(sb)) #define BITMAP_OFFSET_BYTE_IN_SECTOR(sb, ent) \ ((ent / BITS_PER_BYTE) & ((sb)->s_blocksize - 1)) -#define BITS_PER_BYTE_MASK 0x7 -#define IGNORED_BITS_REMAINED(clu, clu_base) ((1 << ((clu) - (clu_base))) - 1) +#define IGNORED_BITS_REMAINED(clu, clu_base) ((1UL << ((clu) - (clu_base))) - 1) #define ES_ENTRY_NUM(name_len) (ES_IDX_LAST_FILENAME(name_len) + 1) /* 19 entries = 1 file entry + 1 stream entry + 17 filename entries */ From e9aa9b7d679302247de92e552dc250268298f18d Mon Sep 17 00:00:00 2001 From: HiGarfield Date: Sat, 30 Dec 2023 01:05:35 +0800 Subject: [PATCH 17/74] exfat: fix ALIGN_DOWN undefined error --- balloc.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/balloc.c b/balloc.c index 63967f60ce84..5aa9b4ab80de 100644 --- a/balloc.c +++ b/balloc.c @@ -30,6 +30,10 @@ #error "BITS_PER_LONG not 32 or 64" #endif +#ifndef ALIGN_DOWN +#define ALIGN_DOWN(x, a) __ALIGN_KERNEL((x) - ((a) - 1), (a)) +#endif + /* * Allocation Bitmap Management Functions */ From ee8b3d9e325878d8163892114593e51256dcdc50 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Sat, 30 Dec 2023 10:13:29 +0900 Subject: [PATCH 18/74] exfat: github action: run buiuld and tests on for-kernel-version-from-4.1.0 branch Signed-off-by: Namjae Jeon --- .github/workflows/c-cpp.yml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/.github/workflows/c-cpp.yml b/.github/workflows/c-cpp.yml index 6587fd98bd41..277727301c8e 100644 --- a/.github/workflows/c-cpp.yml +++ b/.github/workflows/c-cpp.yml @@ -2,9 +2,9 @@ name: linux-exfat-oot CI on: push: - branches: [ "master" ] + branches: [ "for-kernel-version-from-4.1.0" ] pull_request: - branches: [ "master" ] + branches: [ "for-kernel-version-from-4.1.0" ] jobs: build: From ae794c3ccff37f8f81e8cb019bd128a6f9c46632 Mon Sep 17 00:00:00 2001 From: Chetan C R Date: Tue, 6 Feb 2024 10:22:07 +0530 Subject: [PATCH 19/74] dt-bindings: clock: Add support for SDCC BCR id for SDM660 Add clock id for sdcc1 BCR for SDM660. Change-Id: Ifaf0e13a5b0ad821b4be711d7a3c49f838b38c40 Signed-off-by: Chetan C R --- include/dt-bindings/clock/qcom,gcc-sdm660.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/dt-bindings/clock/qcom,gcc-sdm660.h b/include/dt-bindings/clock/qcom,gcc-sdm660.h index 33c615c35ba2..05868793e44a 100644 --- a/include/dt-bindings/clock/qcom,gcc-sdm660.h +++ b/include/dt-bindings/clock/qcom,gcc-sdm660.h @@ -1,6 +1,7 @@ /* SPDX-License-Identifier: GPL-2.0-only */ /* * Copyright (c) 2016-2017, 2019-2020, The Linux Foundation. All rights reserved. + * Copyright (c) 2024, Qualcomm Innovation Center, Inc. All rights reserved. */ #ifndef _DT_BINDINGS_CLK_MSM_GCC_660_H @@ -195,5 +196,6 @@ #define GCC_USB_20_BCR 6 #define GCC_USB_30_BCR 7 #define GCC_USB_PHY_CFG_AHB2PHY_BCR 8 +#define GCC_SDCC1_BCR 9 #endif From 808439035306ca364642ff3017cb22e04c88fb06 Mon Sep 17 00:00:00 2001 From: Chetan C R Date: Tue, 6 Feb 2024 10:35:28 +0530 Subject: [PATCH 20/74] clk: qcom: gcc-sdm660: Add SDCC1 BCR clocks for SDM660 Add SDCC1 BCR clocks support for SDM660. Change-Id: I7a634c0c5f216976784fa6a1e2a571a45bfd9fdc Signed-off-by: Chetan C R --- drivers/clk/qcom/gcc-sdm660.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/clk/qcom/gcc-sdm660.c b/drivers/clk/qcom/gcc-sdm660.c index ea79f1386629..e344dec1b532 100644 --- a/drivers/clk/qcom/gcc-sdm660.c +++ b/drivers/clk/qcom/gcc-sdm660.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0-only /* * Copyright (c) 2016-2017, 2019-2021, The Linux Foundation. All rights reserved. + * Copyright (c) 2024, Qualcomm Innovation Center, Inc. All rights reserved. */ #include @@ -2679,6 +2680,7 @@ static struct clk_regmap *gcc_660_clocks[] = { static const struct qcom_reset_map gcc_660_resets[] = { [GCC_QUSB2PHY_PRIM_BCR] = { 0x12000 }, [GCC_QUSB2PHY_SEC_BCR] = { 0x12004 }, + [GCC_SDCC1_BCR] = { 0x16000 }, [GCC_UFS_BCR] = { 0x75000 }, [GCC_USB3_DP_PHY_BCR] = { 0x50028 }, [GCC_USB3_PHY_BCR] = { 0x50020 }, From e9c3b19682eb270d9fd85a2b3e2d215c74f1463c Mon Sep 17 00:00:00 2001 From: Sarannya S Date: Fri, 22 Sep 2023 13:13:58 +0530 Subject: [PATCH 21/74] soc: qcom: smem: Add boundary checks for partitions Add condition check to make sure that the end address of private entry does not go out of partition. Change-Id: I88b3c69d86d90905b214c13a8c632b134b487a49 Signed-off-by: Sarannya S Signed-off-by: Pranav Mahesh Phansalkar --- drivers/soc/qcom/smem.c | 111 +++++++++++++++++++++++++++------------- 1 file changed, 76 insertions(+), 35 deletions(-) diff --git a/drivers/soc/qcom/smem.c b/drivers/soc/qcom/smem.c index ee005e1a8552..353c1320e3e6 100644 --- a/drivers/soc/qcom/smem.c +++ b/drivers/soc/qcom/smem.c @@ -2,6 +2,7 @@ /* * Copyright (c) 2015, Sony Mobile Communications AB. * Copyright (c) 2012-2013, 2018-2020 The Linux Foundation. All rights reserved. + * Copyright (c) 2023, Qualcomm Innovation Center, Inc. All rights reserved. */ #include @@ -85,6 +86,17 @@ /* Max number of processors/hosts in a system */ #define SMEM_HOST_COUNT 11 +/* Entry range check + * ptr >= start : Checks if ptr is greater than the start of access region + * ptr + size >= ptr: Check for integer overflow (On 32bit system where ptr + * and size are 32bits, ptr + size can wrap around to be a small integer) + * ptr + size <= end: Checks if ptr+size is less than the end of access region + */ +#define IN_PARTITION_RANGE(ptr, size, start, end) \ + (((void *)(ptr) >= (void *)(start)) && \ + (((void *)(ptr) + (size)) >= (void *)(ptr)) && \ + (((void *)(ptr) + (size)) <= (void *)(end))) + /** * struct smem_proc_comm - proc_comm communication struct (legacy) * @command: current command to be executed @@ -359,6 +371,7 @@ static int qcom_smem_alloc_private(struct qcom_smem *smem, size_t size) { struct smem_private_entry *hdr, *end; + struct smem_private_entry *next_hdr; struct smem_partition_header *phdr; size_t alloc_size; void *cached; @@ -371,18 +384,25 @@ static int qcom_smem_alloc_private(struct qcom_smem *smem, end = phdr_to_last_uncached_entry(phdr); cached = phdr_to_last_cached_entry(phdr); - if (WARN_ON((void *)end > p_end || (void *)cached > p_end)) + if (WARN_ON(!IN_PARTITION_RANGE(end, 0, phdr, cached) || + cached > p_end)) return -EINVAL; - while (hdr < end) { + while ((hdr < end) && ((hdr + 1) < end)) { if (hdr->canary != SMEM_PRIVATE_CANARY) goto bad_canary; if (le16_to_cpu(hdr->item) == item) return -EEXIST; - hdr = uncached_entry_next(hdr); + next_hdr = uncached_entry_next(hdr); + + if (WARN_ON(next_hdr <= hdr)) + return -EINVAL; + + hdr = next_hdr; } - if (WARN_ON((void *)hdr > p_end)) + + if (WARN_ON((void *)hdr > (void *)end)) return -EINVAL; /* Check that we don't grow into the cached region */ @@ -540,84 +560,105 @@ static void *qcom_smem_get_private(struct qcom_smem *smem, unsigned item, size_t *size) { - struct smem_private_entry *e, *end; + struct smem_private_entry *e, *uncached_end, *cached_end; + struct smem_private_entry *next_e; struct smem_partition_header *phdr; void *item_ptr, *p_end; + size_t entry_size = 0; + u32 partition_size; size_t cacheline; u32 padding_data; u32 e_size; phdr = p_desc->virt_base; + partition_size = le32_to_cpu(p_desc->size); p_end = (void *)phdr + p_desc->size; cacheline = p_desc->cacheline; e = phdr_to_first_uncached_entry(phdr); - end = phdr_to_last_uncached_entry(phdr); + uncached_end = phdr_to_last_uncached_entry(phdr); + cached_end = phdr_to_last_cached_entry(phdr); - if (WARN_ON((void *)end > p_end)) + if (WARN_ON(!IN_PARTITION_RANGE(uncached_end, 0, phdr, cached_end) + || (void *)cached_end > p_end)) return ERR_PTR(-EINVAL); - while (e < end) { + while ((e < uncached_end) && ((e + 1) < uncached_end)) { if (e->canary != SMEM_PRIVATE_CANARY) goto invalid_canary; if (le16_to_cpu(e->item) == item) { - if (size != NULL) { - e_size = le32_to_cpu(e->size); - padding_data = le16_to_cpu(e->padding_data); + e_size = le32_to_cpu(e->size); + padding_data = le16_to_cpu(e->padding_data); - if (e_size < p_desc->size - && padding_data < e_size) - *size = e_size - padding_data; - else - return ERR_PTR(-EINVAL); - } + if (e_size < partition_size && padding_data < e_size) + entry_size = e_size - padding_data; + else + return ERR_PTR(-EINVAL); item_ptr = uncached_entry_to_item(e); - if (WARN_ON(item_ptr > p_end)) + + if (WARN_ON(!IN_PARTITION_RANGE(item_ptr, entry_size, e, + uncached_end))) return ERR_PTR(-EINVAL); + if (size != NULL) + *size = entry_size; + return item_ptr; } - e = uncached_entry_next(e); + next_e = uncached_entry_next(e); + if (WARN_ON(next_e <= e)) + return ERR_PTR(-EINVAL); + + e = next_e; } - if (WARN_ON((void *)e > p_end)) + if (WARN_ON((void *)e > (void *)uncached_end)) return ERR_PTR(-EINVAL); /* Item was not found in the uncached list, search the cached list */ - e = phdr_to_first_cached_entry(phdr, cacheline); - end = phdr_to_last_cached_entry(phdr); + if (cached_end == p_end) + return ERR_PTR(-ENOENT); - if (WARN_ON((void *)e < (void *)phdr || (void *)end > p_end)) + e = phdr_to_first_cached_entry(phdr, cacheline); + + if (WARN_ON(!IN_PARTITION_RANGE(cached_end, 0, uncached_end, p_end) || + !IN_PARTITION_RANGE(e, sizeof(*e), cached_end, p_end))) return ERR_PTR(-EINVAL); - while (e > end) { + while (e > cached_end) { if (e->canary != SMEM_PRIVATE_CANARY) goto invalid_canary; if (le16_to_cpu(e->item) == item) { - if (size != NULL) { - e_size = le32_to_cpu(e->size); - padding_data = le16_to_cpu(e->padding_data); + e_size = le32_to_cpu(e->size); + padding_data = le16_to_cpu(e->padding_data); - if (e_size < p_desc->size - && padding_data < e_size) - *size = e_size - padding_data; - else - return ERR_PTR(-EINVAL); - } + if (e_size < partition_size && padding_data < e_size) + entry_size = e_size - padding_data; + else + return ERR_PTR(-EINVAL); item_ptr = cached_entry_to_item(e); - if (WARN_ON(item_ptr < (void *)phdr)) + if (WARN_ON(!IN_PARTITION_RANGE(item_ptr, entry_size, + cached_end, e))) return ERR_PTR(-EINVAL); + if (size != NULL) + *size = entry_size; + return item_ptr; } - e = cached_entry_next(e, cacheline); + next_e = cached_entry_next(e, cacheline); + if (WARN_ON(next_e >= e)) + return ERR_PTR(-EINVAL); + + e = next_e; } + if (WARN_ON((void *)e < (void *)phdr)) return ERR_PTR(-EINVAL); From fd110f42e3679af082c427b7a2eeb942d5af470f Mon Sep 17 00:00:00 2001 From: qizhong cheng Date: Mon, 11 Dec 2023 17:49:23 +0800 Subject: [PATCH 22/74] PCI: mediatek: Clear interrupt status before dispatching handler MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ Upstream commit 4e11c29873a8a296a20f99b3e03095e65ebf897d ] We found a failure when using the iperf tool during WiFi performance testing, where some MSIs were received while clearing the interrupt status, and these MSIs cannot be serviced. The interrupt status can be cleared even if the MSI status remains pending. As such, given the edge-triggered interrupt type, its status should be cleared before being dispatched to the handler of the underling device. [kwilczynski: commit log, code comment wording] Link: https://lore.kernel.org/linux-pci/20231211094923.31967-1-jianjun.wang@mediatek.com Fixes: 43e6409db64d ("PCI: mediatek: Add MSI support for MT2712 and MT7622") Signed-off-by: qizhong cheng Signed-off-by: Jianjun Wang Signed-off-by: Krzysztof Wilczyński [bhelgaas: rewrap comment] Signed-off-by: Bjorn Helgaas Reviewed-by: AngeloGioacchino Del Regno Cc: Signed-off-by: Sasha Levin --- drivers/pci/controller/pcie-mediatek.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/pci/controller/pcie-mediatek.c b/drivers/pci/controller/pcie-mediatek.c index 066e9e00de11..0d5be37660aa 100644 --- a/drivers/pci/controller/pcie-mediatek.c +++ b/drivers/pci/controller/pcie-mediatek.c @@ -605,14 +605,20 @@ static void mtk_pcie_intr_handler(struct irq_desc *desc) if (status & MSI_STATUS){ unsigned long imsi_status; + /* + * The interrupt status can be cleared even if the + * MSI status remains pending. As such, given the + * edge-triggered interrupt type, its status should + * be cleared before being dispatched to the + * handler of the underlying device. + */ + writel(MSI_STATUS, port->base + PCIE_INT_STATUS); while ((imsi_status = readl(port->base + PCIE_IMSI_STATUS))) { for_each_set_bit(bit, &imsi_status, MTK_MSI_IRQS_NUM) { virq = irq_find_mapping(port->inner_domain, bit); generic_handle_irq(virq); } } - /* Clear MSI interrupt status */ - writel(MSI_STATUS, port->base + PCIE_INT_STATUS); } } From cde2b87517dcf1e2aa887ad8a327b022e663dcea Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Thu, 30 Jan 2020 22:15:28 -0800 Subject: [PATCH 23/74] include/linux/units.h: add helpers for kelvin to/from Celsius conversion [ Upstream commit 23331e4893614deb555c65cdf115c8a28ed32471 ] Patch series "add header file for kelvin to/from Celsius conversion helpers", v4. There are several helper macros to convert kelvin to/from Celsius in for thermal drivers. These are useful for any other drivers or subsystems, but it's odd to include just for the helpers. This adds a new that provides the equivalent inline functions for any drivers or subsystems, and switches all the users of conversion helpers in to use helpers. This patch (of 12): There are several helper macros to convert kelvin to/from Celsius in for thermal drivers. These are useful for any other drivers or subsystems, but it's odd to include just for the helpers. This adds a new that provides the equivalent inline functions for any drivers or subsystems. It is intended to replace the helpers in . Link: http://lkml.kernel.org/r/1576386975-7941-2-git-send-email-akinobu.mita@gmail.com Signed-off-by: Akinobu Mita Reviewed-by: Andy Shevchenko Cc: Sujith Thomas Cc: Darren Hart Cc: Zhang Rui Cc: Daniel Lezcano Cc: Amit Kucheria Cc: Jean Delvare Cc: Guenter Roeck Cc: Keith Busch Cc: Jens Axboe Cc: Christoph Hellwig Cc: Sagi Grimberg Cc: Kalle Valo Cc: Stanislaw Gruszka Cc: Johannes Berg Cc: Emmanuel Grumbach Cc: Luca Coelho Cc: Jonathan Cameron Cc: Hartmut Knaack Cc: Lars-Peter Clausen Cc: Peter Meerwald-Stadler Cc: Andy Shevchenko Cc: Jonathan Cameron Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Stable-dep-of: 3ef79cd14122 ("serial: sc16is7xx: set safe default SPI clock frequency") Signed-off-by: Sasha Levin --- include/linux/units.h | 84 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 84 insertions(+) create mode 100644 include/linux/units.h diff --git a/include/linux/units.h b/include/linux/units.h new file mode 100644 index 000000000000..aaf716364ec3 --- /dev/null +++ b/include/linux/units.h @@ -0,0 +1,84 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _LINUX_UNITS_H +#define _LINUX_UNITS_H + +#include + +#define ABSOLUTE_ZERO_MILLICELSIUS -273150 + +static inline long milli_kelvin_to_millicelsius(long t) +{ + return t + ABSOLUTE_ZERO_MILLICELSIUS; +} + +static inline long millicelsius_to_milli_kelvin(long t) +{ + return t - ABSOLUTE_ZERO_MILLICELSIUS; +} + +#define MILLIDEGREE_PER_DEGREE 1000 +#define MILLIDEGREE_PER_DECIDEGREE 100 + +static inline long kelvin_to_millicelsius(long t) +{ + return milli_kelvin_to_millicelsius(t * MILLIDEGREE_PER_DEGREE); +} + +static inline long millicelsius_to_kelvin(long t) +{ + t = millicelsius_to_milli_kelvin(t); + + return DIV_ROUND_CLOSEST(t, MILLIDEGREE_PER_DEGREE); +} + +static inline long deci_kelvin_to_celsius(long t) +{ + t = milli_kelvin_to_millicelsius(t * MILLIDEGREE_PER_DECIDEGREE); + + return DIV_ROUND_CLOSEST(t, MILLIDEGREE_PER_DEGREE); +} + +static inline long celsius_to_deci_kelvin(long t) +{ + t = millicelsius_to_milli_kelvin(t * MILLIDEGREE_PER_DEGREE); + + return DIV_ROUND_CLOSEST(t, MILLIDEGREE_PER_DECIDEGREE); +} + +/** + * deci_kelvin_to_millicelsius_with_offset - convert Kelvin to Celsius + * @t: temperature value in decidegrees Kelvin + * @offset: difference between Kelvin and Celsius in millidegrees + * + * Return: temperature value in millidegrees Celsius + */ +static inline long deci_kelvin_to_millicelsius_with_offset(long t, long offset) +{ + return t * MILLIDEGREE_PER_DECIDEGREE - offset; +} + +static inline long deci_kelvin_to_millicelsius(long t) +{ + return milli_kelvin_to_millicelsius(t * MILLIDEGREE_PER_DECIDEGREE); +} + +static inline long millicelsius_to_deci_kelvin(long t) +{ + t = millicelsius_to_milli_kelvin(t); + + return DIV_ROUND_CLOSEST(t, MILLIDEGREE_PER_DECIDEGREE); +} + +static inline long kelvin_to_celsius(long t) +{ + return t + DIV_ROUND_CLOSEST(ABSOLUTE_ZERO_MILLICELSIUS, + MILLIDEGREE_PER_DEGREE); +} + +static inline long celsius_to_kelvin(long t) +{ + return t - DIV_ROUND_CLOSEST(ABSOLUTE_ZERO_MILLICELSIUS, + MILLIDEGREE_PER_DEGREE); +} + +#endif /* _LINUX_UNITS_H */ From 0e7f574162e2f15f4b63f9892906f6a9afe4429f Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Tue, 8 Dec 2020 17:41:42 +0100 Subject: [PATCH 24/74] units: Add Watt units [ Upstream commit 2ee5f8f05949735fa2f4c463a5e13fcb3660c719 ] As there are the temperature units, let's add the Watt macros definition. Signed-off-by: Daniel Lezcano Reviewed-by: Lukasz Luba Signed-off-by: Rafael J. Wysocki Stable-dep-of: 3ef79cd14122 ("serial: sc16is7xx: set safe default SPI clock frequency") Signed-off-by: Sasha Levin --- include/linux/units.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/include/linux/units.h b/include/linux/units.h index aaf716364ec3..92c234e71cab 100644 --- a/include/linux/units.h +++ b/include/linux/units.h @@ -4,6 +4,10 @@ #include +#define MILLIWATT_PER_WATT 1000L +#define MICROWATT_PER_MILLIWATT 1000L +#define MICROWATT_PER_WATT 1000000L + #define ABSOLUTE_ZERO_MILLICELSIUS -273150 static inline long milli_kelvin_to_millicelsius(long t) From 08333e4c4f3ffe6f9d916cc111d9f3429a6aa6c2 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Tue, 7 Sep 2021 19:57:44 -0700 Subject: [PATCH 25/74] units: change from 'L' to 'UL' [ Upstream commit c9221919a2d2df5741ab074dfec5bdfc6f1e043b ] Patch series "Add Hz macros", v3. There are multiple definitions of the HZ_PER_MHZ or HZ_PER_KHZ in the different drivers. Instead of duplicating this definition again and again, add one in the units.h header to be reused in all the place the redefiniton occurs. At the same time, change the type of the Watts, as they can not be negative. This patch (of 10): The users of the macros are safe to be assigned with an unsigned instead of signed as the variables using them are themselves unsigned. Link: https://lkml.kernel.org/r/20210816114732.1834145-1-daniel.lezcano@linaro.org Link: https://lkml.kernel.org/r/20210816114732.1834145-2-daniel.lezcano@linaro.org Signed-off-by: Daniel Lezcano Cc: Andy Shevchenko Cc: Jonathan Cameron Cc: Christian Eggers Cc: Lukasz Luba Cc: MyungJoo Ham Cc: Kyungmin Park Cc: Lars-Peter Clausen Cc: Peter Meerwald Cc: Zhang Rui Cc: Guenter Roeck Cc: Miquel Raynal Cc: Maxime Coquelin Cc: "Rafael J. Wysocki" Cc: Daniel Lezcano Cc: Chanwoo Choi Cc: Jonathan Cameron Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Stable-dep-of: 3ef79cd14122 ("serial: sc16is7xx: set safe default SPI clock frequency") Signed-off-by: Sasha Levin --- include/linux/units.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/include/linux/units.h b/include/linux/units.h index 92c234e71cab..4a23e39acc7b 100644 --- a/include/linux/units.h +++ b/include/linux/units.h @@ -4,9 +4,9 @@ #include -#define MILLIWATT_PER_WATT 1000L -#define MICROWATT_PER_MILLIWATT 1000L -#define MICROWATT_PER_WATT 1000000L +#define MILLIWATT_PER_WATT 1000UL +#define MICROWATT_PER_MILLIWATT 1000UL +#define MICROWATT_PER_WATT 1000000UL #define ABSOLUTE_ZERO_MILLICELSIUS -273150 From fe79b37c6a59c623c9e64e53d5d9a0fffcfe977f Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Tue, 7 Sep 2021 19:57:48 -0700 Subject: [PATCH 26/74] units: add the HZ macros [ Upstream commit e2c77032fcbe515194107994d12cd72ddb77b022 ] The macros for the unit conversion for frequency are duplicated in different places. Provide these macros in the 'units' header, so they can be reused. Link: https://lkml.kernel.org/r/20210816114732.1834145-3-daniel.lezcano@linaro.org Signed-off-by: Daniel Lezcano Reviewed-by: Christian Eggers Reviewed-by: Andy Shevchenko Cc: Chanwoo Choi Cc: Guenter Roeck Cc: Jonathan Cameron Cc: Jonathan Cameron Cc: Kyungmin Park Cc: Lars-Peter Clausen Cc: Lukasz Luba Cc: Maxime Coquelin Cc: Miquel Raynal Cc: MyungJoo Ham Cc: Peter Meerwald Cc: "Rafael J. Wysocki" Cc: Zhang Rui Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Stable-dep-of: 3ef79cd14122 ("serial: sc16is7xx: set safe default SPI clock frequency") Signed-off-by: Sasha Levin --- include/linux/units.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/include/linux/units.h b/include/linux/units.h index 4a23e39acc7b..a0af6d2ef4e5 100644 --- a/include/linux/units.h +++ b/include/linux/units.h @@ -4,6 +4,10 @@ #include +#define HZ_PER_KHZ 1000UL +#define KHZ_PER_MHZ 1000UL +#define HZ_PER_MHZ 1000000UL + #define MILLIWATT_PER_WATT 1000UL #define MICROWATT_PER_MILLIWATT 1000UL #define MICROWATT_PER_WATT 1000000UL From 5733959d6770324020e30dd9313cbeac0aec07ef Mon Sep 17 00:00:00 2001 From: Hugo Villeneuve Date: Thu, 21 Dec 2023 18:18:10 -0500 Subject: [PATCH 27/74] serial: sc16is7xx: set safe default SPI clock frequency [ Upstream commit 3ef79cd1412236d884ab0c46b4d1921380807b48 ] 15 MHz is supported only by 76x variants. If the SPI clock frequency is not specified, use a safe default clock value of 4 MHz that is supported by all variants. Also use HZ_PER_MHZ macro to improve readability. Fixes: 2c837a8a8f9f ("sc16is7xx: spi interface is added") Cc: Signed-off-by: Hugo Villeneuve Link: https://lore.kernel.org/r/20231221231823.2327894-4-hugo@hugovil.com Signed-off-by: Greg Kroah-Hartman Signed-off-by: Sasha Levin --- drivers/tty/serial/sc16is7xx.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/sc16is7xx.c b/drivers/tty/serial/sc16is7xx.c index c65f9194abb0..07898a46fd96 100644 --- a/drivers/tty/serial/sc16is7xx.c +++ b/drivers/tty/serial/sc16is7xx.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #define SC16IS7XX_NAME "sc16is7xx" @@ -1408,7 +1409,7 @@ static int sc16is7xx_spi_probe(struct spi_device *spi) spi->bits_per_word = 8; /* only supports mode 0 on SC16IS762 */ spi->mode = spi->mode ? : SPI_MODE_0; - spi->max_speed_hz = spi->max_speed_hz ? : 15000000; + spi->max_speed_hz = spi->max_speed_hz ? : 4 * HZ_PER_MHZ; ret = spi_setup(spi); if (ret) return ret; From b0465a4897047ece1e4275fefc88a66ee41c5379 Mon Sep 17 00:00:00 2001 From: Andrzej Hajda Date: Mon, 13 Jul 2020 16:43:21 +0200 Subject: [PATCH 28/74] driver core: add device probe log helper [ Upstream commit a787e5400a1ceeb0ef92d71ec43aeb35b1fa1334 ] During probe every time driver gets resource it should usually check for error printk some message if it is not -EPROBE_DEFER and return the error. This pattern is simple but requires adding few lines after any resource acquisition code, as a result it is often omitted or implemented only partially. dev_err_probe helps to replace such code sequences with simple call, so code: if (err != -EPROBE_DEFER) dev_err(dev, ...); return err; becomes: return dev_err_probe(dev, err, ...); Signed-off-by: Andrzej Hajda Reviewed-by: Rafael J. Wysocki Reviewed-by: Mark Brown Reviewed-by: Andy Shevchenko Link: https://lore.kernel.org/r/20200713144324.23654-2-a.hajda@samsung.com Signed-off-by: Greg Kroah-Hartman Stable-dep-of: 6d710b769c1f ("serial: sc16is7xx: add check for unsupported SPI modes during probe") Signed-off-by: Sasha Levin --- drivers/base/core.c | 42 ++++++++++++++++++++++++++++++++++++++++++ include/linux/device.h | 3 +++ 2 files changed, 45 insertions(+) diff --git a/drivers/base/core.c b/drivers/base/core.c index 6e380ad9d08a..b66647277d52 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -3334,6 +3334,48 @@ define_dev_printk_level(_dev_info, KERN_INFO); #endif +/** + * dev_err_probe - probe error check and log helper + * @dev: the pointer to the struct device + * @err: error value to test + * @fmt: printf-style format string + * @...: arguments as specified in the format string + * + * This helper implements common pattern present in probe functions for error + * checking: print debug or error message depending if the error value is + * -EPROBE_DEFER and propagate error upwards. + * It replaces code sequence: + * if (err != -EPROBE_DEFER) + * dev_err(dev, ...); + * else + * dev_dbg(dev, ...); + * return err; + * with + * return dev_err_probe(dev, err, ...); + * + * Returns @err. + * + */ +int dev_err_probe(const struct device *dev, int err, const char *fmt, ...) +{ + struct va_format vaf; + va_list args; + + va_start(args, fmt); + vaf.fmt = fmt; + vaf.va = &args; + + if (err != -EPROBE_DEFER) + dev_err(dev, "error %d: %pV", err, &vaf); + else + dev_dbg(dev, "error %d: %pV", err, &vaf); + + va_end(args); + + return err; +} +EXPORT_SYMBOL_GPL(dev_err_probe); + static inline bool fwnode_is_primary(struct fwnode_handle *fwnode) { return fwnode && !IS_ERR(fwnode->secondary); diff --git a/include/linux/device.h b/include/linux/device.h index bccd367c11de..0714d6e5d500 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -1581,6 +1581,9 @@ do { \ WARN_ONCE(condition, "%s %s: " format, \ dev_driver_string(dev), dev_name(dev), ## arg) +extern __printf(3, 4) +int dev_err_probe(const struct device *dev, int err, const char *fmt, ...); + /* Create alias, so I can be autoloaded. */ #define MODULE_ALIAS_CHARDEV(major,minor) \ MODULE_ALIAS("char-major-" __stringify(major) "-" __stringify(minor)) From 93d63ccd20b55ae704d2c89664375e6f98552ee6 Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Tue, 27 Oct 2020 10:57:23 +0100 Subject: [PATCH 29/74] spi: introduce SPI_MODE_X_MASK macro [ Upstream commit 029b42d8519cef70c4fb5fcaccd08f1053ed2bf0 ] Provide a macro to filter all SPI_MODE_0,1,2,3 mode in one run. The latest SPI framework will parse the devicetree in following call sequence: of_register_spi_device() -> of_spi_parse_dt() So, driver do not need to pars the devicetree and will get prepared flags in the probe. On one hand it is good far most drivers. On other hand some drivers need to filter flags provide by SPI framework and apply know to work flags. This drivers may use SPI_MODE_X_MASK to filter MODE flags and set own, known flags: spi->flags &= ~SPI_MODE_X_MASK; spi->flags |= SPI_MODE_0; Signed-off-by: Oleksij Rempel Link: https://lore.kernel.org/r/20201027095724.18654-2-o.rempel@pengutronix.de Signed-off-by: Mark Brown Stable-dep-of: 6d710b769c1f ("serial: sc16is7xx: add check for unsupported SPI modes during probe") Signed-off-by: Sasha Levin --- include/linux/spi/spi.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index 16158fe097a8..23a232d6db69 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -153,6 +153,7 @@ struct spi_device { #define SPI_MODE_1 (0|SPI_CPHA) #define SPI_MODE_2 (SPI_CPOL|0) #define SPI_MODE_3 (SPI_CPOL|SPI_CPHA) +#define SPI_MODE_X_MASK (SPI_CPOL|SPI_CPHA) #define SPI_CS_HIGH 0x04 /* chipselect active high? */ #define SPI_LSB_FIRST 0x08 /* per-word bits-on-wire */ #define SPI_3WIRE 0x10 /* SI/SO signals shared */ From cccdd04643a7d21d0ca5d8205b15ec0ea1be1c6b Mon Sep 17 00:00:00 2001 From: Hugo Villeneuve Date: Thu, 21 Dec 2023 18:18:09 -0500 Subject: [PATCH 30/74] serial: sc16is7xx: add check for unsupported SPI modes during probe [ Upstream commit 6d710b769c1f5f0d55c9ad9bb49b7dce009ec103 ] The original comment is confusing because it implies that variants other than the SC16IS762 supports other SPI modes beside SPI_MODE_0. Extract from datasheet: The SC16IS762 differs from the SC16IS752 in that it supports SPI clock speeds up to 15 Mbit/s instead of the 4 Mbit/s supported by the SC16IS752... In all other aspects, the SC16IS762 is functionally and electrically the same as the SC16IS752. The same is also true of the SC16IS760 variant versus the SC16IS740 and SC16IS750 variants. For all variants, only SPI mode 0 is supported. Change comment and abort probing if the specified SPI mode is not SPI_MODE_0. Fixes: 2c837a8a8f9f ("sc16is7xx: spi interface is added") Cc: Signed-off-by: Hugo Villeneuve Link: https://lore.kernel.org/r/20231221231823.2327894-3-hugo@hugovil.com Signed-off-by: Greg Kroah-Hartman Signed-off-by: Sasha Levin --- drivers/tty/serial/sc16is7xx.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/sc16is7xx.c b/drivers/tty/serial/sc16is7xx.c index 07898a46fd96..800cb94e4b91 100644 --- a/drivers/tty/serial/sc16is7xx.c +++ b/drivers/tty/serial/sc16is7xx.c @@ -1407,7 +1407,10 @@ static int sc16is7xx_spi_probe(struct spi_device *spi) /* Setup SPI bus */ spi->bits_per_word = 8; - /* only supports mode 0 on SC16IS762 */ + /* For all variants, only mode 0 is supported */ + if ((spi->mode & SPI_MODE_X_MASK) != SPI_MODE_0) + return dev_err_probe(&spi->dev, -EINVAL, "Unsupported SPI mode\n"); + spi->mode = spi->mode ? : SPI_MODE_0; spi->max_speed_hz = spi->max_speed_hz ? : 4 * HZ_PER_MHZ; ret = spi_setup(spi); From 5b6a7f323b533e5ab07e93633ad9644b41b6df42 Mon Sep 17 00:00:00 2001 From: Suraj Jitindar Singh Date: Wed, 13 Dec 2023 16:16:35 +1100 Subject: [PATCH 31/74] ext4: allow for the last group to be marked as trimmed commit 7c784d624819acbeefb0018bac89e632467cca5a upstream. The ext4 filesystem tracks the trim status of blocks at the group level. When an entire group has been trimmed then it is marked as such and subsequent trim invocations with the same minimum trim size will not be attempted on that group unless it is marked as able to be trimmed again such as when a block is freed. Currently the last group can't be marked as trimmed due to incorrect logic in ext4_last_grp_cluster(). ext4_last_grp_cluster() is supposed to return the zero based index of the last cluster in a group. This is then used by ext4_try_to_trim_range() to determine if the trim operation spans the entire group and as such if the trim status of the group should be recorded. ext4_last_grp_cluster() takes a 0 based group index, thus the valid values for grp are 0..(ext4_get_groups_count - 1). Any group index less than (ext4_get_groups_count - 1) is not the last group and must have EXT4_CLUSTERS_PER_GROUP(sb) clusters. For the last group we need to calculate the number of clusters based on the number of blocks in the group. Finally subtract 1 from the number of clusters as zero based indexing is expected. Rearrange the function slightly to make it clear what we are calculating and returning. Reproducer: // Create file system where the last group has fewer blocks than // blocks per group $ mkfs.ext4 -b 4096 -g 8192 /dev/nvme0n1 8191 $ mount /dev/nvme0n1 /mnt Before Patch: $ fstrim -v /mnt /mnt: 25.9 MiB (27156480 bytes) trimmed // Group not marked as trimmed so second invocation still discards blocks $ fstrim -v /mnt /mnt: 25.9 MiB (27156480 bytes) trimmed After Patch: fstrim -v /mnt /mnt: 25.9 MiB (27156480 bytes) trimmed // Group marked as trimmed so second invocation DOESN'T discard any blocks fstrim -v /mnt /mnt: 0 B (0 bytes) trimmed Fixes: 45e4ab320c9b ("ext4: move setting of trimmed bit into ext4_try_to_trim_range()") Cc: # 4.19+ Signed-off-by: Suraj Jitindar Singh Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/20231213051635.37731-1-surajjs@amazon.com Signed-off-by: Theodore Ts'o Signed-off-by: Greg Kroah-Hartman --- fs/ext4/mballoc.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 8875fac9f958..0572ae09c729 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -5192,11 +5192,16 @@ __acquires(bitlock) static ext4_grpblk_t ext4_last_grp_cluster(struct super_block *sb, ext4_group_t grp) { - if (grp < ext4_get_groups_count(sb)) - return EXT4_CLUSTERS_PER_GROUP(sb) - 1; - return (ext4_blocks_count(EXT4_SB(sb)->s_es) - - ext4_group_first_block_no(sb, grp) - 1) >> - EXT4_CLUSTER_BITS(sb); + unsigned long nr_clusters_in_group; + + if (grp < (ext4_get_groups_count(sb) - 1)) + nr_clusters_in_group = EXT4_CLUSTERS_PER_GROUP(sb); + else + nr_clusters_in_group = (ext4_blocks_count(EXT4_SB(sb)->s_es) - + ext4_group_first_block_no(sb, grp)) + >> EXT4_CLUSTER_BITS(sb); + + return nr_clusters_in_group - 1; } static bool ext4_trim_interrupted(void) From 15a67115d487ea5cb8213915a4f75f58adb87cbc Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Thu, 7 Dec 2023 18:36:57 +0800 Subject: [PATCH 32/74] crypto: api - Disallow identical driver names commit 27016f75f5ed47e2d8e0ca75a8ff1f40bc1a5e27 upstream. Disallow registration of two algorithms with identical driver names. Cc: Reported-by: Ovidiu Panait Signed-off-by: Herbert Xu Signed-off-by: Greg Kroah-Hartman --- crypto/algapi.c | 1 + 1 file changed, 1 insertion(+) diff --git a/crypto/algapi.c b/crypto/algapi.c index 346557a3fc0b..a1ceca922841 100644 --- a/crypto/algapi.c +++ b/crypto/algapi.c @@ -231,6 +231,7 @@ static struct crypto_larval *__crypto_register_alg(struct crypto_alg *alg) } if (!strcmp(q->cra_driver_name, alg->cra_name) || + !strcmp(q->cra_driver_name, alg->cra_driver_name) || !strcmp(q->cra_name, alg->cra_driver_name)) goto err; } From 489506a2a0cbbfc7065d4d18ec6bb9baa3818c62 Mon Sep 17 00:00:00 2001 From: Hongchen Zhang Date: Thu, 16 Nov 2023 08:56:09 +0800 Subject: [PATCH 33/74] PM: hibernate: Enforce ordering during image compression/decompression commit 71cd7e80cfde548959952eac7063aeaea1f2e1c6 upstream. An S4 (suspend to disk) test on the LoongArch 3A6000 platform sometimes fails with the following error messaged in the dmesg log: Invalid LZO compressed length That happens because when compressing/decompressing the image, the synchronization between the control thread and the compress/decompress/crc thread is based on a relaxed ordering interface, which is unreliable, and the following situation may occur: CPU 0 CPU 1 save_image_lzo lzo_compress_threadfn atomic_set(&d->stop, 1); atomic_read(&data[thr].stop) data[thr].cmp = data[thr].cmp_len; WRITE data[thr].cmp_len Then CPU0 gets a stale cmp_len and writes it to disk. During resume from S4, wrong cmp_len is loaded. To maintain data consistency between the two threads, use the acquire/release variants of atomic set and read operations. Fixes: 081a9d043c98 ("PM / Hibernate: Improve performance of LZO/plain hibernation, checksum image") Cc: All applicable Signed-off-by: Hongchen Zhang Co-developed-by: Weihao Li Signed-off-by: Weihao Li [ rjw: Subject rewrite and changelog edits ] Signed-off-by: Rafael J. Wysocki Signed-off-by: Greg Kroah-Hartman --- kernel/power/swap.c | 38 +++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-) diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 9db7f2f93fae..a818ad6ade9c 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -596,11 +596,11 @@ static int crc32_threadfn(void *data) unsigned i; while (1) { - wait_event(d->go, atomic_read(&d->ready) || + wait_event(d->go, atomic_read_acquire(&d->ready) || kthread_should_stop()); if (kthread_should_stop()) { d->thr = NULL; - atomic_set(&d->stop, 1); + atomic_set_release(&d->stop, 1); wake_up(&d->done); break; } @@ -609,7 +609,7 @@ static int crc32_threadfn(void *data) for (i = 0; i < d->run_threads; i++) *d->crc32 = crc32_le(*d->crc32, d->unc[i], *d->unc_len[i]); - atomic_set(&d->stop, 1); + atomic_set_release(&d->stop, 1); wake_up(&d->done); } return 0; @@ -639,12 +639,12 @@ static int lzo_compress_threadfn(void *data) struct cmp_data *d = data; while (1) { - wait_event(d->go, atomic_read(&d->ready) || + wait_event(d->go, atomic_read_acquire(&d->ready) || kthread_should_stop()); if (kthread_should_stop()) { d->thr = NULL; d->ret = -1; - atomic_set(&d->stop, 1); + atomic_set_release(&d->stop, 1); wake_up(&d->done); break; } @@ -653,7 +653,7 @@ static int lzo_compress_threadfn(void *data) d->ret = lzo1x_1_compress(d->unc, d->unc_len, d->cmp + LZO_HEADER, &d->cmp_len, d->wrk); - atomic_set(&d->stop, 1); + atomic_set_release(&d->stop, 1); wake_up(&d->done); } return 0; @@ -791,7 +791,7 @@ static int save_image_lzo(struct swap_map_handle *handle, data[thr].unc_len = off; - atomic_set(&data[thr].ready, 1); + atomic_set_release(&data[thr].ready, 1); wake_up(&data[thr].go); } @@ -799,12 +799,12 @@ static int save_image_lzo(struct swap_map_handle *handle, break; crc->run_threads = thr; - atomic_set(&crc->ready, 1); + atomic_set_release(&crc->ready, 1); wake_up(&crc->go); for (run_threads = thr, thr = 0; thr < run_threads; thr++) { wait_event(data[thr].done, - atomic_read(&data[thr].stop)); + atomic_read_acquire(&data[thr].stop)); atomic_set(&data[thr].stop, 0); ret = data[thr].ret; @@ -843,7 +843,7 @@ static int save_image_lzo(struct swap_map_handle *handle, } } - wait_event(crc->done, atomic_read(&crc->stop)); + wait_event(crc->done, atomic_read_acquire(&crc->stop)); atomic_set(&crc->stop, 0); } @@ -1124,12 +1124,12 @@ static int lzo_decompress_threadfn(void *data) struct dec_data *d = data; while (1) { - wait_event(d->go, atomic_read(&d->ready) || + wait_event(d->go, atomic_read_acquire(&d->ready) || kthread_should_stop()); if (kthread_should_stop()) { d->thr = NULL; d->ret = -1; - atomic_set(&d->stop, 1); + atomic_set_release(&d->stop, 1); wake_up(&d->done); break; } @@ -1142,7 +1142,7 @@ static int lzo_decompress_threadfn(void *data) flush_icache_range((unsigned long)d->unc, (unsigned long)d->unc + d->unc_len); - atomic_set(&d->stop, 1); + atomic_set_release(&d->stop, 1); wake_up(&d->done); } return 0; @@ -1330,7 +1330,7 @@ static int load_image_lzo(struct swap_map_handle *handle, } if (crc->run_threads) { - wait_event(crc->done, atomic_read(&crc->stop)); + wait_event(crc->done, atomic_read_acquire(&crc->stop)); atomic_set(&crc->stop, 0); crc->run_threads = 0; } @@ -1366,7 +1366,7 @@ static int load_image_lzo(struct swap_map_handle *handle, pg = 0; } - atomic_set(&data[thr].ready, 1); + atomic_set_release(&data[thr].ready, 1); wake_up(&data[thr].go); } @@ -1385,7 +1385,7 @@ static int load_image_lzo(struct swap_map_handle *handle, for (run_threads = thr, thr = 0; thr < run_threads; thr++) { wait_event(data[thr].done, - atomic_read(&data[thr].stop)); + atomic_read_acquire(&data[thr].stop)); atomic_set(&data[thr].stop, 0); ret = data[thr].ret; @@ -1416,7 +1416,7 @@ static int load_image_lzo(struct swap_map_handle *handle, ret = snapshot_write_next(snapshot); if (ret <= 0) { crc->run_threads = thr + 1; - atomic_set(&crc->ready, 1); + atomic_set_release(&crc->ready, 1); wake_up(&crc->go); goto out_finish; } @@ -1424,13 +1424,13 @@ static int load_image_lzo(struct swap_map_handle *handle, } crc->run_threads = thr; - atomic_set(&crc->ready, 1); + atomic_set_release(&crc->ready, 1); wake_up(&crc->go); } out_finish: if (crc->run_threads) { - wait_event(crc->done, atomic_read(&crc->stop)); + wait_event(crc->done, atomic_read_acquire(&crc->stop)); atomic_set(&crc->stop, 0); } stop = ktime_get(); From eafd83b92f6c044007a3591cbd476bcf90455990 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sat, 2 Dec 2023 09:01:54 +0800 Subject: [PATCH 34/74] hwrng: core - Fix page fault dead lock on mmap-ed hwrng commit 78aafb3884f6bc6636efcc1760c891c8500b9922 upstream. There is a dead-lock in the hwrng device read path. This triggers when the user reads from /dev/hwrng into memory also mmap-ed from /dev/hwrng. The resulting page fault triggers a recursive read which then dead-locks. Fix this by using a stack buffer when calling copy_to_user. Reported-by: Edward Adam Davis Reported-by: syzbot+c52ab18308964d248092@syzkaller.appspotmail.com Fixes: 9996508b3353 ("hwrng: core - Replace u32 in driver API with byte array") Cc: Signed-off-by: Herbert Xu Signed-off-by: Greg Kroah-Hartman --- drivers/char/hw_random/core.c | 36 +++++++++++++++++++++-------------- 1 file changed, 22 insertions(+), 14 deletions(-) diff --git a/drivers/char/hw_random/core.c b/drivers/char/hw_random/core.c index c9757fa2d308..d932dd95b04e 100644 --- a/drivers/char/hw_random/core.c +++ b/drivers/char/hw_random/core.c @@ -24,10 +24,13 @@ #include #include #include +#include #include #define RNG_MODULE_NAME "hw_random" +#define RNG_BUFFER_SIZE (SMP_CACHE_BYTES < 32 ? 32 : SMP_CACHE_BYTES) + static struct hwrng *current_rng; /* the current rng has been explicitly chosen by user via sysfs */ static int cur_rng_set_by_user; @@ -59,7 +62,7 @@ static inline int rng_get_data(struct hwrng *rng, u8 *buffer, size_t size, static size_t rng_buffer_size(void) { - return SMP_CACHE_BYTES < 32 ? 32 : SMP_CACHE_BYTES; + return RNG_BUFFER_SIZE; } static void add_early_randomness(struct hwrng *rng) @@ -202,6 +205,7 @@ static inline int rng_get_data(struct hwrng *rng, u8 *buffer, size_t size, static ssize_t rng_dev_read(struct file *filp, char __user *buf, size_t size, loff_t *offp) { + u8 buffer[RNG_BUFFER_SIZE]; ssize_t ret = 0; int err = 0; int bytes_read, len; @@ -229,34 +233,37 @@ static ssize_t rng_dev_read(struct file *filp, char __user *buf, if (bytes_read < 0) { err = bytes_read; goto out_unlock_reading; - } - data_avail = bytes_read; - } - - if (!data_avail) { - if (filp->f_flags & O_NONBLOCK) { + } else if (bytes_read == 0 && + (filp->f_flags & O_NONBLOCK)) { err = -EAGAIN; goto out_unlock_reading; } - } else { - len = data_avail; + + data_avail = bytes_read; + } + + len = data_avail; + if (len) { if (len > size) len = size; data_avail -= len; - if (copy_to_user(buf + ret, rng_buffer + data_avail, - len)) { + memcpy(buffer, rng_buffer + data_avail, len); + } + mutex_unlock(&reading_mutex); + put_rng(rng); + + if (len) { + if (copy_to_user(buf + ret, buffer, len)) { err = -EFAULT; - goto out_unlock_reading; + goto out; } size -= len; ret += len; } - mutex_unlock(&reading_mutex); - put_rng(rng); if (need_resched()) schedule_timeout_interruptible(1); @@ -267,6 +274,7 @@ static ssize_t rng_dev_read(struct file *filp, char __user *buf, } } out: + memzero_explicit(buffer, sizeof(buffer)); return ret ? : err; out_unlock_reading: From 4ce844d0f50f789cc70de7bd02511df0f40c64b6 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Fri, 26 Jan 2024 10:36:31 -0700 Subject: [PATCH 35/74] powerpc: Use always instead of always-y in for crtsavres.o This commit is for linux-4.19.y only, it has no direct upstream equivalent. Prior to commit 5f2fb52fac15 ("kbuild: rename hostprogs-y/always to hostprogs/always-y"), always-y did not exist, making the backport of mainline commit 1b1e38002648 ("powerpc: add crtsavres.o to always-y instead of extra-y") to linux-4.19.y as commit b7b85ec5ec15 ("powerpc: add crtsavres.o to always-y instead of extra-y") incorrect, breaking the build with linkers that need crtsavres.o: ld.lld: error: cannot open arch/powerpc/lib/crtsavres.o: No such file or directory Backporting the aforementioned kbuild commit is not suitable for stable due to its size and number of conflicts, so transform the always-y usage to an equivalent form using always, which resolves the build issues. Fixes: b7b85ec5ec15 ("powerpc: add crtsavres.o to always-y instead of extra-y") Signed-off-by: Nathan Chancellor Signed-off-by: Greg Kroah-Hartman --- arch/powerpc/lib/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile index 6f1e57182876..f0aa6fc8c6b2 100644 --- a/arch/powerpc/lib/Makefile +++ b/arch/powerpc/lib/Makefile @@ -21,8 +21,8 @@ obj-$(CONFIG_PPC32) += div64.o copy_32.o crtsavres.o strlen_32.o # 64-bit linker creates .sfpr on demand for final link (vmlinux), # so it is only needed for modules, and only for older linkers which # do not support --save-restore-funcs -ifeq ($(call ld-ifversion, -lt, 225000000, y),y) -always-$(CONFIG_PPC64) += crtsavres.o +ifeq ($(call ld-ifversion, -lt, 225000000, y)$(CONFIG_PPC64),yy) +always += crtsavres.o endif obj-$(CONFIG_PPC_BOOK3S_64) += copyuser_power7.o copypage_power7.o \ From 229ce47cbfdc7d3a9415eb676abbfb77d676cb08 Mon Sep 17 00:00:00 2001 From: Xiaolei Wang Date: Fri, 15 Dec 2023 10:00:49 +0800 Subject: [PATCH 36/74] rpmsg: virtio: Free driver_override when rpmsg_remove() commit d5362c37e1f8a40096452fc201c30e705750e687 upstream. Free driver_override when rpmsg_remove(), otherwise the following memory leak will occur: unreferenced object 0xffff0000d55d7080 (size 128): comm "kworker/u8:2", pid 56, jiffies 4294893188 (age 214.272s) hex dump (first 32 bytes): 72 70 6d 73 67 5f 6e 73 00 00 00 00 00 00 00 00 rpmsg_ns........ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [<000000009c94c9c1>] __kmem_cache_alloc_node+0x1f8/0x320 [<000000002300d89b>] __kmalloc_node_track_caller+0x44/0x70 [<00000000228a60c3>] kstrndup+0x4c/0x90 [<0000000077158695>] driver_set_override+0xd0/0x164 [<000000003e9c4ea5>] rpmsg_register_device_override+0x98/0x170 [<000000001c0c89a8>] rpmsg_ns_register_device+0x24/0x30 [<000000008bbf8fa2>] rpmsg_probe+0x2e0/0x3ec [<00000000e65a68df>] virtio_dev_probe+0x1c0/0x280 [<00000000443331cc>] really_probe+0xbc/0x2dc [<00000000391064b1>] __driver_probe_device+0x78/0xe0 [<00000000a41c9a5b>] driver_probe_device+0xd8/0x160 [<000000009c3bd5df>] __device_attach_driver+0xb8/0x140 [<0000000043cd7614>] bus_for_each_drv+0x7c/0xd4 [<000000003b929a36>] __device_attach+0x9c/0x19c [<00000000a94e0ba8>] device_initial_probe+0x14/0x20 [<000000003c999637>] bus_probe_device+0xa0/0xac Signed-off-by: Xiaolei Wang Fixes: b0b03b811963 ("rpmsg: Release rpmsg devices in backends") Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20231215020049.78750-1-xiaolei.wang@windriver.com Signed-off-by: Mathieu Poirier Signed-off-by: Greg Kroah-Hartman --- drivers/rpmsg/virtio_rpmsg_bus.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/rpmsg/virtio_rpmsg_bus.c b/drivers/rpmsg/virtio_rpmsg_bus.c index 664f957012cd..aab990339051 100644 --- a/drivers/rpmsg/virtio_rpmsg_bus.c +++ b/drivers/rpmsg/virtio_rpmsg_bus.c @@ -381,6 +381,7 @@ static void virtio_rpmsg_release_device(struct device *dev) struct rpmsg_device *rpdev = to_rpmsg_device(dev); struct virtio_rpmsg_channel *vch = to_virtio_rpmsg_channel(rpdev); + kfree(rpdev->driver_override); kfree(vch); } From 771df0145297a1a9f1e7f799da43f8b0f8850e7c Mon Sep 17 00:00:00 2001 From: "Maciej S. Szmigiero" Date: Thu, 25 Jan 2024 19:06:54 +0100 Subject: [PATCH 37/74] x86/CPU/AMD: Fix disabling XSAVES on AMD family 0x17 due to erratum The stable kernel version backport of the patch disabling XSAVES on AMD Zen family 0x17 applied this change to the wrong function (init_amd_k6()), one which isn't called for Zen CPUs. Move the erratum to the init_amd_zn() function instead. Add an explicit family 0x17 check to the erratum so nothing will break if someone naively makes this kernel version call init_amd_zn() also for family 0x19 in the future (as the current upstream code does). Fixes: f028a7db9824 ("x86/CPU/AMD: Disable XSAVES on AMD family 0x17") Signed-off-by: Maciej S. Szmigiero Signed-off-by: Greg Kroah-Hartman --- arch/x86/kernel/cpu/amd.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 84667781c41d..5b75a4ff6802 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -271,15 +271,6 @@ static void init_amd_k6(struct cpuinfo_x86 *c) return; } #endif - /* - * Work around Erratum 1386. The XSAVES instruction malfunctions in - * certain circumstances on Zen1/2 uarch, and not all parts have had - * updated microcode at the time of writing (March 2023). - * - * Affected parts all have no supervisor XSAVE states, meaning that - * the XSAVEC instruction (which works fine) is equivalent. - */ - clear_cpu_cap(c, X86_FEATURE_XSAVES); } static void init_amd_k7(struct cpuinfo_x86 *c) @@ -979,6 +970,17 @@ static void init_amd_zn(struct cpuinfo_x86 *c) if (c->x86 == 0x19 && !cpu_has(c, X86_FEATURE_BTC_NO)) set_cpu_cap(c, X86_FEATURE_BTC_NO); } + + /* + * Work around Erratum 1386. The XSAVES instruction malfunctions in + * certain circumstances on Zen1/2 uarch, and not all parts have had + * updated microcode at the time of writing (March 2023). + * + * Affected parts all have no supervisor XSAVE states, meaning that + * the XSAVEC instruction (which works fine) is equivalent. + */ + if (c->x86 == 0x17) + clear_cpu_cap(c, X86_FEATURE_XSAVES); } static bool cpu_has_zenbleed_microcode(void) From 63cc5eb9b4102474423c4d6563fdf2ebbedf437f Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Wed, 3 Jan 2024 21:02:16 +0100 Subject: [PATCH 38/74] parisc/firmware: Fix F-extend for PDC addresses commit 735ae74f73e55c191d48689bd11ff4a06ea0508f upstream. When running with narrow firmware (64-bit kernel using a 32-bit firmware), extend PDC addresses into the 0xfffffff0.00000000 region instead of the 0xf0f0f0f0.00000000 region. This fixes the power button on the C3700 machine in qemu (64-bit CPU with 32-bit firmware), and my assumption is that the previous code was really never used (because most 64-bit machines have a 64-bit firmware), or it just worked on very old machines because they may only decode 40-bit of virtual addresses. Cc: stable@vger.kernel.org Signed-off-by: Helge Deller Signed-off-by: Greg Kroah-Hartman --- arch/parisc/kernel/firmware.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/parisc/kernel/firmware.c b/arch/parisc/kernel/firmware.c index 906b7c882587..82f264621845 100644 --- a/arch/parisc/kernel/firmware.c +++ b/arch/parisc/kernel/firmware.c @@ -127,10 +127,10 @@ static unsigned long f_extend(unsigned long address) #ifdef CONFIG_64BIT if(unlikely(parisc_narrow_firmware)) { if((address & 0xff000000) == 0xf0000000) - return 0xf0f0f0f000000000UL | (u32)address; + return (0xfffffff0UL << 32) | (u32)address; if((address & 0xf0000000) == 0xf0000000) - return 0xffffffff00000000UL | (u32)address; + return (0xffffffffUL << 32) | (u32)address; } #endif return address; From d31978bfec1d251a75d4a038e564ef2ff9d8be40 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Wed, 26 Aug 2020 13:44:59 +0300 Subject: [PATCH 39/74] driver core: Annotate dev_err_probe() with __must_check commit e1f82a0dcf388d98bcc7ad195c03bd812405e6b2 upstream. We have got already new users of this API which interpret it differently and miss the opportunity to optimize their code. In order to avoid similar cases in the future, annotate dev_err_probe() with __must_check. Fixes: a787e5400a1c ("driver core: add device probe log helper") Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20200826104459.81979-1-andriy.shevchenko@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- include/linux/device.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/device.h b/include/linux/device.h index 0714d6e5d500..2e5926792f4e 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -1582,7 +1582,7 @@ do { \ dev_driver_string(dev), dev_name(dev), ## arg) extern __printf(3, 4) -int dev_err_probe(const struct device *dev, int err, const char *fmt, ...); +int __must_check dev_err_probe(const struct device *dev, int err, const char *fmt, ...); /* Create alias, so I can be autoloaded. */ #define MODULE_ALIAS_CHARDEV(major,minor) \ From 98fcd3b663835bc7518d65d053419c640f3515a3 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Thu, 18 Jan 2024 06:19:57 +1000 Subject: [PATCH 40/74] nouveau/vmm: don't set addr on the fail path to avoid warning commit cacea81390fd8c8c85404e5eb2adeb83d87a912e upstream. nvif_vmm_put gets called if addr is set, but if the allocation fails we don't need to call put, otherwise we get a warning like [523232.435671] ------------[ cut here ]------------ [523232.435674] WARNING: CPU: 8 PID: 1505697 at drivers/gpu/drm/nouveau/nvif/vmm.c:68 nvif_vmm_put+0x72/0x80 [nouveau] [523232.435795] Modules linked in: uinput rfcomm snd_seq_dummy snd_hrtimer nf_conntrack_netbios_ns nf_conntrack_broadcast nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ip_set nf_tables nfnetlink qrtr bnep sunrpc binfmt_misc intel_rapl_msr intel_rapl_common intel_uncore_frequency intel_uncore_frequency_common isst_if_common iwlmvm nfit libnvdimm vfat fat x86_pkg_temp_thermal intel_powerclamp mac80211 snd_soc_avs snd_soc_hda_codec coretemp snd_hda_ext_core snd_soc_core snd_hda_codec_realtek kvm_intel snd_hda_codec_hdmi snd_compress snd_hda_codec_generic ac97_bus snd_pcm_dmaengine snd_hda_intel libarc4 snd_intel_dspcfg snd_intel_sdw_acpi snd_hda_codec kvm iwlwifi snd_hda_core btusb snd_hwdep btrtl snd_seq btintel irqbypass btbcm rapl snd_seq_device eeepc_wmi btmtk intel_cstate iTCO_wdt cfg80211 snd_pcm asus_wmi bluetooth intel_pmc_bxt iTCO_vendor_support snd_timer ledtrig_audio pktcdvd snd mei_me [523232.435828] sparse_keymap intel_uncore i2c_i801 platform_profile wmi_bmof mei pcspkr ioatdma soundcore i2c_smbus rfkill idma64 dca joydev acpi_tad loop zram nouveau drm_ttm_helper ttm video drm_exec drm_gpuvm gpu_sched crct10dif_pclmul i2c_algo_bit nvme crc32_pclmul crc32c_intel drm_display_helper polyval_clmulni nvme_core polyval_generic e1000e mxm_wmi cec ghash_clmulni_intel r8169 sha512_ssse3 nvme_common wmi pinctrl_sunrisepoint uas usb_storage ip6_tables ip_tables fuse [523232.435849] CPU: 8 PID: 1505697 Comm: gnome-shell Tainted: G W 6.6.0-rc7-nvk-uapi+ #12 [523232.435851] Hardware name: System manufacturer System Product Name/ROG STRIX X299-E GAMING II, BIOS 1301 09/24/2021 [523232.435852] RIP: 0010:nvif_vmm_put+0x72/0x80 [nouveau] [523232.435934] Code: 00 00 48 89 e2 be 02 00 00 00 48 c7 04 24 00 00 00 00 48 89 44 24 08 e8 fc bf ff ff 85 c0 75 0a 48 c7 43 08 00 00 00 00 eb b3 <0f> 0b eb f2 e8 f5 c9 b2 e6 0f 1f 44 00 00 90 90 90 90 90 90 90 90 [523232.435936] RSP: 0018:ffffc900077ffbd8 EFLAGS: 00010282 [523232.435937] RAX: 00000000fffffffe RBX: ffffc900077ffc00 RCX: 0000000000000010 [523232.435938] RDX: 0000000000000010 RSI: ffffc900077ffb38 RDI: ffffc900077ffbd8 [523232.435940] RBP: ffff888e1c4f2140 R08: 0000000000000000 R09: 0000000000000000 [523232.435940] R10: 0000000000000000 R11: 0000000000000000 R12: ffff888503811800 [523232.435941] R13: ffffc900077ffca0 R14: ffff888e1c4f2140 R15: ffff88810317e1e0 [523232.435942] FS: 00007f933a769640(0000) GS:ffff88905fa00000(0000) knlGS:0000000000000000 [523232.435943] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [523232.435944] CR2: 00007f930bef7000 CR3: 00000005d0322001 CR4: 00000000003706e0 [523232.435945] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [523232.435946] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [523232.435964] Call Trace: [523232.435965] [523232.435966] ? nvif_vmm_put+0x72/0x80 [nouveau] [523232.436051] ? __warn+0x81/0x130 [523232.436055] ? nvif_vmm_put+0x72/0x80 [nouveau] [523232.436138] ? report_bug+0x171/0x1a0 [523232.436142] ? handle_bug+0x3c/0x80 [523232.436144] ? exc_invalid_op+0x17/0x70 [523232.436145] ? asm_exc_invalid_op+0x1a/0x20 [523232.436149] ? nvif_vmm_put+0x72/0x80 [nouveau] [523232.436230] ? nvif_vmm_put+0x64/0x80 [nouveau] [523232.436342] nouveau_vma_del+0x80/0xd0 [nouveau] [523232.436506] nouveau_vma_new+0x1a0/0x210 [nouveau] [523232.436671] nouveau_gem_object_open+0x1d0/0x1f0 [nouveau] [523232.436835] drm_gem_handle_create_tail+0xd1/0x180 [523232.436840] drm_prime_fd_to_handle_ioctl+0x12e/0x200 [523232.436844] ? __pfx_drm_prime_fd_to_handle_ioctl+0x10/0x10 [523232.436847] drm_ioctl_kernel+0xd3/0x180 [523232.436849] drm_ioctl+0x26d/0x4b0 [523232.436851] ? __pfx_drm_prime_fd_to_handle_ioctl+0x10/0x10 [523232.436855] nouveau_drm_ioctl+0x5a/0xb0 [nouveau] [523232.437032] __x64_sys_ioctl+0x94/0xd0 [523232.437036] do_syscall_64+0x5d/0x90 [523232.437040] ? syscall_exit_to_user_mode+0x2b/0x40 [523232.437044] ? do_syscall_64+0x6c/0x90 [523232.437046] entry_SYSCALL_64_after_hwframe+0x6e/0xd8 Reported-by: Faith Ekstrand Cc: stable@vger.kernel.org Signed-off-by: Dave Airlie Link: https://patchwork.freedesktop.org/patch/msgid/20240117213852.295565-1-airlied@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/gpu/drm/nouveau/nouveau_vmm.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/nouveau/nouveau_vmm.c b/drivers/gpu/drm/nouveau/nouveau_vmm.c index 2032c3e4f6e5..fe8b88c6e2f9 100644 --- a/drivers/gpu/drm/nouveau/nouveau_vmm.c +++ b/drivers/gpu/drm/nouveau/nouveau_vmm.c @@ -107,6 +107,9 @@ nouveau_vma_new(struct nouveau_bo *nvbo, struct nouveau_vmm *vmm, } else { ret = nvif_vmm_get(&vmm->vmm, PTES, false, mem->mem.page, 0, mem->mem.size, &tmp); + if (ret) + goto done; + vma->addr = tmp.addr; } From b715d543d42e8e4ddd645193410cd4511fc46d6c Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 9 Sep 2020 09:37:40 +0200 Subject: [PATCH 41/74] Revert "driver core: Annotate dev_err_probe() with __must_check" commit f601e8f37c2c1c52f2923fffc48204a7f7dc023d upstream. This reverts commit e1f82a0dcf388d98bcc7ad195c03bd812405e6b2 as it's already starting to cause build warnings in linux-next for things that are "obviously correct". It's up to driver authors do "do the right thing" here with this function, and if they don't want to call it as the last line of a function, that's up to them, otherwise code that looks like: ret = dev_err_probe(..., ret, ...); does look really "odd". Reported-by: Stephen Rothwell Reported-by: Krzysztof Kozlowski Fixes: e1f82a0dcf38 ("driver core: Annotate dev_err_probe() with __must_check") Cc: Andy Shevchenko Signed-off-by: Greg Kroah-Hartman --- include/linux/device.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/device.h b/include/linux/device.h index 2e5926792f4e..0714d6e5d500 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -1582,7 +1582,7 @@ do { \ dev_driver_string(dev), dev_name(dev), ## arg) extern __printf(3, 4) -int __must_check dev_err_probe(const struct device *dev, int err, const char *fmt, ...); +int dev_err_probe(const struct device *dev, int err, const char *fmt, ...); /* Create alias, so I can be autoloaded. */ #define MODULE_ALIAS_CHARDEV(major,minor) \ From d2d0b95ca1b5fefa3deed444a803c9f809db66cf Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Mon, 14 Aug 2023 15:41:00 +0100 Subject: [PATCH 42/74] block: Remove special-casing of compound pages commit 1b151e2435fc3a9b10c8946c6aebe9f3e1938c55 upstream. The special casing was originally added in pre-git history; reproducing the commit log here: > commit a318a92567d77 > Author: Andrew Morton > Date: Sun Sep 21 01:42:22 2003 -0700 > > [PATCH] Speed up direct-io hugetlbpage handling > > This patch short-circuits all the direct-io page dirtying logic for > higher-order pages. Without this, we pointlessly bounce BIOs up to > keventd all the time. In the last twenty years, compound pages have become used for more than just hugetlb. Rewrite these functions to operate on folios instead of pages and remove the special case for hugetlbfs; I don't think it's needed any more (and if it is, we can put it back in as a call to folio_test_hugetlb()). This was found by inspection; as far as I can tell, this bug can lead to pages used as the destination of a direct I/O read not being marked as dirty. If those pages are then reclaimed by the MM without being dirtied for some other reason, they won't be written out. Then when they're faulted back in, they will not contain the data they should. It'll take a pretty unusual setup to produce this problem with several races all going the wrong way. This problem predates the folio work; it could for example have been triggered by mmaping a THP in tmpfs and using that as the target of an O_DIRECT read. Fixes: 800d8c63b2e98 ("shmem: add huge pages support") Cc: Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Jens Axboe Signed-off-by: Greg Kroah-Hartman --- block/bio.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/block/bio.c b/block/bio.c index 7858b2d23916..476a88e11715 100644 --- a/block/bio.c +++ b/block/bio.c @@ -1592,8 +1592,7 @@ void bio_set_pages_dirty(struct bio *bio) int i; bio_for_each_segment_all(bvec, bio, i) { - if (!PageCompound(bvec->bv_page)) - set_page_dirty_lock(bvec->bv_page); + set_page_dirty_lock(bvec->bv_page); } } EXPORT_SYMBOL_GPL(bio_set_pages_dirty); @@ -1652,7 +1651,7 @@ void bio_check_pages_dirty(struct bio *bio) int i; bio_for_each_segment_all(bvec, bio, i) { - if (!PageDirty(bvec->bv_page) && !PageCompound(bvec->bv_page)) + if (!PageDirty(bvec->bv_page)) goto defer; } From cf07cb793264fd3c459918dda7e839d6a279493c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Micha=C5=82=20Miros=C5=82aw?= Date: Fri, 28 Aug 2020 18:14:35 +0200 Subject: [PATCH 43/74] driver code: print symbolic error code MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit 693a8e936590f93451e6f5a3d748616f5a59c80b upstream. dev_err_probe() prepends the message with an error code. Let's make it more readable by translating the code to a more recognisable symbol. Fixes: a787e5400a1c ("driver core: add device probe log helper") Signed-off-by: Michał Mirosław Link: https://lore.kernel.org/r/ea3f973e4708919573026fdce52c264db147626d.1598630856.git.mirq-linux@rere.qmqm.pl Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/base/core.c b/drivers/base/core.c index b66647277d52..8a53e346e7aa 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -3366,9 +3366,9 @@ int dev_err_probe(const struct device *dev, int err, const char *fmt, ...) vaf.va = &args; if (err != -EPROBE_DEFER) - dev_err(dev, "error %d: %pV", err, &vaf); + dev_err(dev, "error %pe: %pV", ERR_PTR(err), &vaf); else - dev_dbg(dev, "error %d: %pV", err, &vaf); + dev_dbg(dev, "error %pe: %pV", ERR_PTR(err), &vaf); va_end(args); From 4d61ff79b439fe9cd5eaa3363a25853f230e2026 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Wed, 9 Sep 2020 11:53:43 +0200 Subject: [PATCH 44/74] drivers: core: fix kernel-doc markup for dev_err_probe() commit 074b3aad307de6126fbac1fff4996d1034b48fee upstream. There are two literal blocks there. Fix the markups, in order to produce the right html output and solve those warnings: ./drivers/base/core.c:4218: WARNING: Unexpected indentation. ./drivers/base/core.c:4222: WARNING: Definition list ends without a blank line; unexpected unindent. ./drivers/base/core.c:4223: WARNING: Block quote ends without a blank line; unexpected unindent. Fixes: a787e5400a1c ("driver core: add device probe log helper") Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/base/core.c b/drivers/base/core.c index 8a53e346e7aa..5fd2b887cc04 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -3344,13 +3344,15 @@ define_dev_printk_level(_dev_info, KERN_INFO); * This helper implements common pattern present in probe functions for error * checking: print debug or error message depending if the error value is * -EPROBE_DEFER and propagate error upwards. - * It replaces code sequence: + * It replaces code sequence:: * if (err != -EPROBE_DEFER) * dev_err(dev, ...); * else * dev_dbg(dev, ...); * return err; - * with + * + * with:: + * * return dev_err_probe(dev, err, ...); * * Returns @err. From 27aea64838914c6122db5b8bd4bed865c9736f22 Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Thu, 18 Jan 2024 12:32:10 +0800 Subject: [PATCH 45/74] net/smc: fix illegal rmb_desc access in SMC-D connection dump [ Upstream commit dbc153fd3c142909e564bb256da087e13fbf239c ] A crash was found when dumping SMC-D connections. It can be reproduced by following steps: - run nginx/wrk test: smc_run nginx smc_run wrk -t 16 -c 1000 -d -H 'Connection: Close' - continuously dump SMC-D connections in parallel: watch -n 1 'smcss -D' BUG: kernel NULL pointer dereference, address: 0000000000000030 CPU: 2 PID: 7204 Comm: smcss Kdump: loaded Tainted: G E 6.7.0+ #55 RIP: 0010:__smc_diag_dump.constprop.0+0x5e5/0x620 [smc_diag] Call Trace: ? __die+0x24/0x70 ? page_fault_oops+0x66/0x150 ? exc_page_fault+0x69/0x140 ? asm_exc_page_fault+0x26/0x30 ? __smc_diag_dump.constprop.0+0x5e5/0x620 [smc_diag] ? __kmalloc_node_track_caller+0x35d/0x430 ? __alloc_skb+0x77/0x170 smc_diag_dump_proto+0xd0/0xf0 [smc_diag] smc_diag_dump+0x26/0x60 [smc_diag] netlink_dump+0x19f/0x320 __netlink_dump_start+0x1dc/0x300 smc_diag_handler_dump+0x6a/0x80 [smc_diag] ? __pfx_smc_diag_dump+0x10/0x10 [smc_diag] sock_diag_rcv_msg+0x121/0x140 ? __pfx_sock_diag_rcv_msg+0x10/0x10 netlink_rcv_skb+0x5a/0x110 sock_diag_rcv+0x28/0x40 netlink_unicast+0x22a/0x330 netlink_sendmsg+0x1f8/0x420 __sock_sendmsg+0xb0/0xc0 ____sys_sendmsg+0x24e/0x300 ? copy_msghdr_from_user+0x62/0x80 ___sys_sendmsg+0x7c/0xd0 ? __do_fault+0x34/0x160 ? do_read_fault+0x5f/0x100 ? do_fault+0xb0/0x110 ? __handle_mm_fault+0x2b0/0x6c0 __sys_sendmsg+0x4d/0x80 do_syscall_64+0x69/0x180 entry_SYSCALL_64_after_hwframe+0x6e/0x76 It is possible that the connection is in process of being established when we dump it. Assumed that the connection has been registered in a link group by smc_conn_create() but the rmb_desc has not yet been initialized by smc_buf_create(), thus causing the illegal access to conn->rmb_desc. So fix it by checking before dump. Fixes: 4b1b7d3b30a6 ("net/smc: add SMC-D diag support") Signed-off-by: Wen Gu Reviewed-by: Dust Li Reviewed-by: Wenjia Zhang Signed-off-by: David S. Miller Signed-off-by: Sasha Levin --- net/smc/smc_diag.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/smc/smc_diag.c b/net/smc/smc_diag.c index 6c4a7a5938b7..3ab14678afff 100644 --- a/net/smc/smc_diag.c +++ b/net/smc/smc_diag.c @@ -167,7 +167,7 @@ static int __smc_diag_dump(struct sock *sk, struct sk_buff *skb, } if (smc->conn.lgr && smc->conn.lgr->is_smcd && (req->diag_ext & (1 << (SMC_DIAG_DMBINFO - 1))) && - !list_empty(&smc->conn.lgr->list)) { + !list_empty(&smc->conn.lgr->list) && smc->conn.rmb_desc) { struct smc_connection *conn = &smc->conn; struct smcd_diag_dmbinfo dinfo; From 91759822dd336c20f817e6fd59cccee3952599f7 Mon Sep 17 00:00:00 2001 From: Lin Ma Date: Thu, 18 Jan 2024 21:03:06 +0800 Subject: [PATCH 46/74] vlan: skip nested type that is not IFLA_VLAN_QOS_MAPPING [ Upstream commit 6c21660fe221a15c789dee2bc2fd95516bc5aeaf ] In the vlan_changelink function, a loop is used to parse the nested attributes IFLA_VLAN_EGRESS_QOS and IFLA_VLAN_INGRESS_QOS in order to obtain the struct ifla_vlan_qos_mapping. These two nested attributes are checked in the vlan_validate_qos_map function, which calls nla_validate_nested_deprecated with the vlan_map_policy. However, this deprecated validator applies a LIBERAL strictness, allowing the presence of an attribute with the type IFLA_VLAN_QOS_UNSPEC. Consequently, the loop in vlan_changelink may parse an attribute of type IFLA_VLAN_QOS_UNSPEC and believe it carries a payload of struct ifla_vlan_qos_mapping, which is not necessarily true. To address this issue and ensure compatibility, this patch introduces two type checks that skip attributes whose type is not IFLA_VLAN_QOS_MAPPING. Fixes: 07b5b17e157b ("[VLAN]: Use rtnl_link API") Signed-off-by: Lin Ma Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240118130306.1644001-1-linma@zju.edu.cn Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- net/8021q/vlan_netlink.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/8021q/vlan_netlink.c b/net/8021q/vlan_netlink.c index 74042b9d7f73..231eae2f1685 100644 --- a/net/8021q/vlan_netlink.c +++ b/net/8021q/vlan_netlink.c @@ -120,12 +120,16 @@ static int vlan_changelink(struct net_device *dev, struct nlattr *tb[], } if (data[IFLA_VLAN_INGRESS_QOS]) { nla_for_each_nested(attr, data[IFLA_VLAN_INGRESS_QOS], rem) { + if (nla_type(attr) != IFLA_VLAN_QOS_MAPPING) + continue; m = nla_data(attr); vlan_dev_set_ingress_priority(dev, m->to, m->from); } } if (data[IFLA_VLAN_EGRESS_QOS]) { nla_for_each_nested(attr, data[IFLA_VLAN_EGRESS_QOS], rem) { + if (nla_type(attr) != IFLA_VLAN_QOS_MAPPING) + continue; m = nla_data(attr); err = vlan_dev_set_egress_priority(dev, m->from, m->to); if (err) From 84e9d10419f6f4f3f3cd8f9aaf44a48719aa4b1b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Jan 2024 18:36:25 +0000 Subject: [PATCH 47/74] llc: make llc_ui_sendmsg() more robust against bonding changes [ Upstream commit dad555c816a50c6a6a8a86be1f9177673918c647 ] syzbot was able to trick llc_ui_sendmsg(), allocating an skb with no headroom, but subsequently trying to push 14 bytes of Ethernet header [1] Like some others, llc_ui_sendmsg() releases the socket lock before calling sock_alloc_send_skb(). Then it acquires it again, but does not redo all the sanity checks that were performed. This fix: - Uses LL_RESERVED_SPACE() to reserve space. - Check all conditions again after socket lock is held again. - Do not account Ethernet header for mtu limitation. [1] skbuff: skb_under_panic: text:ffff800088baa334 len:1514 put:14 head:ffff0000c9c37000 data:ffff0000c9c36ff2 tail:0x5dc end:0x6c0 dev:bond0 kernel BUG at net/core/skbuff.c:193 ! Internal error: Oops - BUG: 00000000f2000800 [#1] PREEMPT SMP Modules linked in: CPU: 0 PID: 6875 Comm: syz-executor.0 Not tainted 6.7.0-rc8-syzkaller-00101-g0802e17d9aca-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 11/17/2023 pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : skb_panic net/core/skbuff.c:189 [inline] pc : skb_under_panic+0x13c/0x140 net/core/skbuff.c:203 lr : skb_panic net/core/skbuff.c:189 [inline] lr : skb_under_panic+0x13c/0x140 net/core/skbuff.c:203 sp : ffff800096f97000 x29: ffff800096f97010 x28: ffff80008cc8d668 x27: dfff800000000000 x26: ffff0000cb970c90 x25: 00000000000005dc x24: ffff0000c9c36ff2 x23: ffff0000c9c37000 x22: 00000000000005ea x21: 00000000000006c0 x20: 000000000000000e x19: ffff800088baa334 x18: 1fffe000368261ce x17: ffff80008e4ed000 x16: ffff80008a8310f8 x15: 0000000000000001 x14: 1ffff00012df2d58 x13: 0000000000000000 x12: 0000000000000000 x11: 0000000000000001 x10: 0000000000ff0100 x9 : e28a51f1087e8400 x8 : e28a51f1087e8400 x7 : ffff80008028f8d0 x6 : 0000000000000000 x5 : 0000000000000001 x4 : 0000000000000001 x3 : ffff800082b78714 x2 : 0000000000000001 x1 : 0000000100000000 x0 : 0000000000000089 Call trace: skb_panic net/core/skbuff.c:189 [inline] skb_under_panic+0x13c/0x140 net/core/skbuff.c:203 skb_push+0xf0/0x108 net/core/skbuff.c:2451 eth_header+0x44/0x1f8 net/ethernet/eth.c:83 dev_hard_header include/linux/netdevice.h:3188 [inline] llc_mac_hdr_init+0x110/0x17c net/llc/llc_output.c:33 llc_sap_action_send_xid_c+0x170/0x344 net/llc/llc_s_ac.c:85 llc_exec_sap_trans_actions net/llc/llc_sap.c:153 [inline] llc_sap_next_state net/llc/llc_sap.c:182 [inline] llc_sap_state_process+0x1ec/0x774 net/llc/llc_sap.c:209 llc_build_and_send_xid_pkt+0x12c/0x1c0 net/llc/llc_sap.c:270 llc_ui_sendmsg+0x7bc/0xb1c net/llc/af_llc.c:997 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg net/socket.c:745 [inline] sock_sendmsg+0x194/0x274 net/socket.c:767 splice_to_socket+0x7cc/0xd58 fs/splice.c:881 do_splice_from fs/splice.c:933 [inline] direct_splice_actor+0xe4/0x1c0 fs/splice.c:1142 splice_direct_to_actor+0x2a0/0x7e4 fs/splice.c:1088 do_splice_direct+0x20c/0x348 fs/splice.c:1194 do_sendfile+0x4bc/0xc70 fs/read_write.c:1254 __do_sys_sendfile64 fs/read_write.c:1322 [inline] __se_sys_sendfile64 fs/read_write.c:1308 [inline] __arm64_sys_sendfile64+0x160/0x3b4 fs/read_write.c:1308 __invoke_syscall arch/arm64/kernel/syscall.c:37 [inline] invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:51 el0_svc_common+0x130/0x23c arch/arm64/kernel/syscall.c:136 do_el0_svc+0x48/0x58 arch/arm64/kernel/syscall.c:155 el0_svc+0x54/0x158 arch/arm64/kernel/entry-common.c:678 el0t_64_sync_handler+0x84/0xfc arch/arm64/kernel/entry-common.c:696 el0t_64_sync+0x190/0x194 arch/arm64/kernel/entry.S:595 Code: aa1803e6 aa1903e7 a90023f5 94792f6a (d4210000) Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-and-tested-by: syzbot+2a7024e9502df538e8ef@syzkaller.appspotmail.com Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240118183625.4007013-1-edumazet@google.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- net/llc/af_llc.c | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index 5cba9199c3c9..a5c104ea477c 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -926,14 +926,15 @@ copy_uaddr: */ static int llc_ui_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) { + DECLARE_SOCKADDR(struct sockaddr_llc *, addr, msg->msg_name); struct sock *sk = sock->sk; struct llc_sock *llc = llc_sk(sk); - DECLARE_SOCKADDR(struct sockaddr_llc *, addr, msg->msg_name); int flags = msg->msg_flags; int noblock = flags & MSG_DONTWAIT; + int rc = -EINVAL, copied = 0, hdrlen, hh_len; struct sk_buff *skb = NULL; + struct net_device *dev; size_t size = 0; - int rc = -EINVAL, copied = 0, hdrlen; dprintk("%s: sending from %02X to %02X\n", __func__, llc->laddr.lsap, llc->daddr.lsap); @@ -953,22 +954,29 @@ static int llc_ui_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) if (rc) goto out; } - hdrlen = llc->dev->hard_header_len + llc_ui_header_len(sk, addr); + dev = llc->dev; + hh_len = LL_RESERVED_SPACE(dev); + hdrlen = llc_ui_header_len(sk, addr); size = hdrlen + len; - if (size > llc->dev->mtu) - size = llc->dev->mtu; + size = min_t(size_t, size, READ_ONCE(dev->mtu)); copied = size - hdrlen; rc = -EINVAL; if (copied < 0) goto out; release_sock(sk); - skb = sock_alloc_send_skb(sk, size, noblock, &rc); + skb = sock_alloc_send_skb(sk, hh_len + size, noblock, &rc); lock_sock(sk); if (!skb) goto out; - skb->dev = llc->dev; + if (sock_flag(sk, SOCK_ZAPPED) || + llc->dev != dev || + hdrlen != llc_ui_header_len(sk, addr) || + hh_len != LL_RESERVED_SPACE(dev) || + size > READ_ONCE(dev->mtu)) + goto out; + skb->dev = dev; skb->protocol = llc_proto_type(addr->sllc_arphrd); - skb_reserve(skb, hdrlen); + skb_reserve(skb, hh_len + hdrlen); rc = memcpy_from_msg(skb_put(skb, copied), msg, copied); if (rc) goto out; From 165ad1e22779685c3ed3dd349c6c4c632309cc62 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 18 Jan 2024 17:55:15 -0800 Subject: [PATCH 48/74] llc: Drop support for ETH_P_TR_802_2. [ Upstream commit e3f9bed9bee261e3347131764e42aeedf1ffea61 ] syzbot reported an uninit-value bug below. [0] llc supports ETH_P_802_2 (0x0004) and used to support ETH_P_TR_802_2 (0x0011), and syzbot abused the latter to trigger the bug. write$tun(r0, &(0x7f0000000040)={@val={0x0, 0x11}, @val, @mpls={[], @llc={@snap={0xaa, 0x1, ')', "90e5dd"}}}}, 0x16) llc_conn_handler() initialises local variables {saddr,daddr}.mac based on skb in llc_pdu_decode_sa()/llc_pdu_decode_da() and passes them to __llc_lookup(). However, the initialisation is done only when skb->protocol is htons(ETH_P_802_2), otherwise, __llc_lookup_established() and __llc_lookup_listener() will read garbage. The missing initialisation existed prior to commit 211ed865108e ("net: delete all instances of special processing for token ring"). It removed the part to kick out the token ring stuff but forgot to close the door allowing ETH_P_TR_802_2 packets to sneak into llc_rcv(). Let's remove llc_tr_packet_type and complete the deprecation. [0]: BUG: KMSAN: uninit-value in __llc_lookup_established+0xe9d/0xf90 __llc_lookup_established+0xe9d/0xf90 __llc_lookup net/llc/llc_conn.c:611 [inline] llc_conn_handler+0x4bd/0x1360 net/llc/llc_conn.c:791 llc_rcv+0xfbb/0x14a0 net/llc/llc_input.c:206 __netif_receive_skb_one_core net/core/dev.c:5527 [inline] __netif_receive_skb+0x1a6/0x5a0 net/core/dev.c:5641 netif_receive_skb_internal net/core/dev.c:5727 [inline] netif_receive_skb+0x58/0x660 net/core/dev.c:5786 tun_rx_batched+0x3ee/0x980 drivers/net/tun.c:1555 tun_get_user+0x53af/0x66d0 drivers/net/tun.c:2002 tun_chr_write_iter+0x3af/0x5d0 drivers/net/tun.c:2048 call_write_iter include/linux/fs.h:2020 [inline] new_sync_write fs/read_write.c:491 [inline] vfs_write+0x8ef/0x1490 fs/read_write.c:584 ksys_write+0x20f/0x4c0 fs/read_write.c:637 __do_sys_write fs/read_write.c:649 [inline] __se_sys_write fs/read_write.c:646 [inline] __x64_sys_write+0x93/0xd0 fs/read_write.c:646 do_syscall_x64 arch/x86/entry/common.c:51 [inline] do_syscall_64+0x44/0x110 arch/x86/entry/common.c:82 entry_SYSCALL_64_after_hwframe+0x63/0x6b Local variable daddr created at: llc_conn_handler+0x53/0x1360 net/llc/llc_conn.c:783 llc_rcv+0xfbb/0x14a0 net/llc/llc_input.c:206 CPU: 1 PID: 5004 Comm: syz-executor994 Not tainted 6.6.0-syzkaller-14500-g1c41041124bd #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/09/2023 Fixes: 211ed865108e ("net: delete all instances of special processing for token ring") Reported-by: syzbot+b5ad66046b913bc04c6f@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=b5ad66046b913bc04c6f Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20240119015515.61898-1-kuniyu@amazon.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- include/net/llc_pdu.h | 6 ++---- net/llc/llc_core.c | 7 ------- 2 files changed, 2 insertions(+), 11 deletions(-) diff --git a/include/net/llc_pdu.h b/include/net/llc_pdu.h index 49aa79c7b278..581cd37aa98b 100644 --- a/include/net/llc_pdu.h +++ b/include/net/llc_pdu.h @@ -262,8 +262,7 @@ static inline void llc_pdu_header_init(struct sk_buff *skb, u8 type, */ static inline void llc_pdu_decode_sa(struct sk_buff *skb, u8 *sa) { - if (skb->protocol == htons(ETH_P_802_2)) - memcpy(sa, eth_hdr(skb)->h_source, ETH_ALEN); + memcpy(sa, eth_hdr(skb)->h_source, ETH_ALEN); } /** @@ -275,8 +274,7 @@ static inline void llc_pdu_decode_sa(struct sk_buff *skb, u8 *sa) */ static inline void llc_pdu_decode_da(struct sk_buff *skb, u8 *da) { - if (skb->protocol == htons(ETH_P_802_2)) - memcpy(da, eth_hdr(skb)->h_dest, ETH_ALEN); + memcpy(da, eth_hdr(skb)->h_dest, ETH_ALEN); } /** diff --git a/net/llc/llc_core.c b/net/llc/llc_core.c index 64d4bef04e73..4900a27b5176 100644 --- a/net/llc/llc_core.c +++ b/net/llc/llc_core.c @@ -135,22 +135,15 @@ static struct packet_type llc_packet_type __read_mostly = { .func = llc_rcv, }; -static struct packet_type llc_tr_packet_type __read_mostly = { - .type = cpu_to_be16(ETH_P_TR_802_2), - .func = llc_rcv, -}; - static int __init llc_init(void) { dev_add_pack(&llc_packet_type); - dev_add_pack(&llc_tr_packet_type); return 0; } static void __exit llc_exit(void) { dev_remove_pack(&llc_packet_type); - dev_remove_pack(&llc_tr_packet_type); } module_init(llc_init); From 344350bfa3b4b37d7c3d5a00536e6fbf0e953fbf Mon Sep 17 00:00:00 2001 From: Sharath Srinivasan Date: Fri, 19 Jan 2024 17:48:39 -0800 Subject: [PATCH 49/74] net/rds: Fix UBSAN: array-index-out-of-bounds in rds_cmsg_recv [ Upstream commit 13e788deb7348cc88df34bed736c3b3b9927ea52 ] Syzcaller UBSAN crash occurs in rds_cmsg_recv(), which reads inc->i_rx_lat_trace[j + 1] with index 4 (3 + 1), but with array size of 4 (RDS_RX_MAX_TRACES). Here 'j' is assigned from rs->rs_rx_trace[i] and in-turn from trace.rx_trace_pos[i] in rds_recv_track_latency(), with both arrays sized 3 (RDS_MSG_RX_DGRAM_TRACE_MAX). So fix the off-by-one bounds check in rds_recv_track_latency() to prevent a potential crash in rds_cmsg_recv(). Found by syzcaller: ================================================================= UBSAN: array-index-out-of-bounds in net/rds/recv.c:585:39 index 4 is out of range for type 'u64 [4]' CPU: 1 PID: 8058 Comm: syz-executor228 Not tainted 6.6.0-gd2f51b3516da #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x136/0x150 lib/dump_stack.c:106 ubsan_epilogue lib/ubsan.c:217 [inline] __ubsan_handle_out_of_bounds+0xd5/0x130 lib/ubsan.c:348 rds_cmsg_recv+0x60d/0x700 net/rds/recv.c:585 rds_recvmsg+0x3fb/0x1610 net/rds/recv.c:716 sock_recvmsg_nosec net/socket.c:1044 [inline] sock_recvmsg+0xe2/0x160 net/socket.c:1066 __sys_recvfrom+0x1b6/0x2f0 net/socket.c:2246 __do_sys_recvfrom net/socket.c:2264 [inline] __se_sys_recvfrom net/socket.c:2260 [inline] __x64_sys_recvfrom+0xe0/0x1b0 net/socket.c:2260 do_syscall_x64 arch/x86/entry/common.c:51 [inline] do_syscall_64+0x40/0x110 arch/x86/entry/common.c:82 entry_SYSCALL_64_after_hwframe+0x63/0x6b ================================================================== Fixes: 3289025aedc0 ("RDS: add receive message trace used by application") Reported-by: Chenyuan Yang Closes: https://lore.kernel.org/linux-rdma/CALGdzuoVdq-wtQ4Az9iottBqC5cv9ZhcE5q8N7LfYFvkRsOVcw@mail.gmail.com/ Signed-off-by: Sharath Srinivasan Reviewed-by: Simon Horman Signed-off-by: David S. Miller Signed-off-by: Sasha Levin --- net/rds/af_rds.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/rds/af_rds.c b/net/rds/af_rds.c index cd7e01ea8144..5eb71d276706 100644 --- a/net/rds/af_rds.c +++ b/net/rds/af_rds.c @@ -385,7 +385,7 @@ static int rds_recv_track_latency(struct rds_sock *rs, char __user *optval, rs->rs_rx_traces = trace.rx_traces; for (i = 0; i < rs->rs_rx_traces; i++) { - if (trace.rx_trace_pos[i] > RDS_MSG_RX_DGRAM_TRACE_MAX) { + if (trace.rx_trace_pos[i] >= RDS_MSG_RX_DGRAM_TRACE_MAX) { rs->rs_rx_traces = 0; return -EFAULT; } From 5022b331c041e8c54b9a6a3251579bd1e8c0fc0b Mon Sep 17 00:00:00 2001 From: Petr Pavlu Date: Mon, 22 Jan 2024 16:09:28 +0100 Subject: [PATCH 50/74] tracing: Ensure visibility when inserting an element into tracing_map [ Upstream commit 2b44760609e9eaafc9d234a6883d042fc21132a7 ] Running the following two commands in parallel on a multi-processor AArch64 machine can sporadically produce an unexpected warning about duplicate histogram entries: $ while true; do echo hist:key=id.syscall:val=hitcount > \ /sys/kernel/debug/tracing/events/raw_syscalls/sys_enter/trigger cat /sys/kernel/debug/tracing/events/raw_syscalls/sys_enter/hist sleep 0.001 done $ stress-ng --sysbadaddr $(nproc) The warning looks as follows: [ 2911.172474] ------------[ cut here ]------------ [ 2911.173111] Duplicates detected: 1 [ 2911.173574] WARNING: CPU: 2 PID: 12247 at kernel/trace/tracing_map.c:983 tracing_map_sort_entries+0x3e0/0x408 [ 2911.174702] Modules linked in: iscsi_ibft(E) iscsi_boot_sysfs(E) rfkill(E) af_packet(E) nls_iso8859_1(E) nls_cp437(E) vfat(E) fat(E) ena(E) tiny_power_button(E) qemu_fw_cfg(E) button(E) fuse(E) efi_pstore(E) ip_tables(E) x_tables(E) xfs(E) libcrc32c(E) aes_ce_blk(E) aes_ce_cipher(E) crct10dif_ce(E) polyval_ce(E) polyval_generic(E) ghash_ce(E) gf128mul(E) sm4_ce_gcm(E) sm4_ce_ccm(E) sm4_ce(E) sm4_ce_cipher(E) sm4(E) sm3_ce(E) sm3(E) sha3_ce(E) sha512_ce(E) sha512_arm64(E) sha2_ce(E) sha256_arm64(E) nvme(E) sha1_ce(E) nvme_core(E) nvme_auth(E) t10_pi(E) sg(E) scsi_mod(E) scsi_common(E) efivarfs(E) [ 2911.174738] Unloaded tainted modules: cppc_cpufreq(E):1 [ 2911.180985] CPU: 2 PID: 12247 Comm: cat Kdump: loaded Tainted: G E 6.7.0-default #2 1b58bbb22c97e4399dc09f92d309344f69c44a01 [ 2911.182398] Hardware name: Amazon EC2 c7g.8xlarge/, BIOS 1.0 11/1/2018 [ 2911.183208] pstate: 61400005 (nZCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--) [ 2911.184038] pc : tracing_map_sort_entries+0x3e0/0x408 [ 2911.184667] lr : tracing_map_sort_entries+0x3e0/0x408 [ 2911.185310] sp : ffff8000a1513900 [ 2911.185750] x29: ffff8000a1513900 x28: ffff0003f272fe80 x27: 0000000000000001 [ 2911.186600] x26: ffff0003f272fe80 x25: 0000000000000030 x24: 0000000000000008 [ 2911.187458] x23: ffff0003c5788000 x22: ffff0003c16710c8 x21: ffff80008017f180 [ 2911.188310] x20: ffff80008017f000 x19: ffff80008017f180 x18: ffffffffffffffff [ 2911.189160] x17: 0000000000000000 x16: 0000000000000000 x15: ffff8000a15134b8 [ 2911.190015] x14: 0000000000000000 x13: 205d373432323154 x12: 5b5d313131333731 [ 2911.190844] x11: 00000000fffeffff x10: 00000000fffeffff x9 : ffffd1b78274a13c [ 2911.191716] x8 : 000000000017ffe8 x7 : c0000000fffeffff x6 : 000000000057ffa8 [ 2911.192554] x5 : ffff0012f6c24ec0 x4 : 0000000000000000 x3 : ffff2e5b72b5d000 [ 2911.193404] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0003ff254480 [ 2911.194259] Call trace: [ 2911.194626] tracing_map_sort_entries+0x3e0/0x408 [ 2911.195220] hist_show+0x124/0x800 [ 2911.195692] seq_read_iter+0x1d4/0x4e8 [ 2911.196193] seq_read+0xe8/0x138 [ 2911.196638] vfs_read+0xc8/0x300 [ 2911.197078] ksys_read+0x70/0x108 [ 2911.197534] __arm64_sys_read+0x24/0x38 [ 2911.198046] invoke_syscall+0x78/0x108 [ 2911.198553] el0_svc_common.constprop.0+0xd0/0xf8 [ 2911.199157] do_el0_svc+0x28/0x40 [ 2911.199613] el0_svc+0x40/0x178 [ 2911.200048] el0t_64_sync_handler+0x13c/0x158 [ 2911.200621] el0t_64_sync+0x1a8/0x1b0 [ 2911.201115] ---[ end trace 0000000000000000 ]--- The problem appears to be caused by CPU reordering of writes issued from __tracing_map_insert(). The check for the presence of an element with a given key in this function is: val = READ_ONCE(entry->val); if (val && keys_match(key, val->key, map->key_size)) ... The write of a new entry is: elt = get_free_elt(map); memcpy(elt->key, key, map->key_size); entry->val = elt; The "memcpy(elt->key, key, map->key_size);" and "entry->val = elt;" stores may become visible in the reversed order on another CPU. This second CPU might then incorrectly determine that a new key doesn't match an already present val->key and subsequently insert a new element, resulting in a duplicate. Fix the problem by adding a write barrier between "memcpy(elt->key, key, map->key_size);" and "entry->val = elt;", and for good measure, also use WRITE_ONCE(entry->val, elt) for publishing the element. The sequence pairs with the mentioned "READ_ONCE(entry->val);" and the "val->key" check which has an address dependency. The barrier is placed on a path executed when adding an element for a new key. Subsequent updates targeting the same key remain unaffected. From the user's perspective, the issue was introduced by commit c193707dde77 ("tracing: Remove code which merges duplicates"), which followed commit cbf4100efb8f ("tracing: Add support to detect and avoid duplicates"). The previous code operated differently; it inherently expected potential races which result in duplicates but merged them later when they occurred. Link: https://lore.kernel.org/linux-trace-kernel/20240122150928.27725-1-petr.pavlu@suse.com Fixes: c193707dde77 ("tracing: Remove code which merges duplicates") Signed-off-by: Petr Pavlu Acked-by: Tom Zanussi Signed-off-by: Steven Rostedt (Google) Signed-off-by: Sasha Levin --- kernel/trace/tracing_map.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/kernel/trace/tracing_map.c b/kernel/trace/tracing_map.c index 83c2a0598c64..33c463967bb3 100644 --- a/kernel/trace/tracing_map.c +++ b/kernel/trace/tracing_map.c @@ -574,7 +574,12 @@ __tracing_map_insert(struct tracing_map *map, void *key, bool lookup_only) } memcpy(elt->key, key, map->key_size); - entry->val = elt; + /* + * Ensure the initialization is visible and + * publish the elt. + */ + smp_wmb(); + WRITE_ONCE(entry->val, elt); atomic64_inc(&map->hits); return entry->val; From 52c46caf283bfa3016a5e41363df93c02037f788 Mon Sep 17 00:00:00 2001 From: Salvatore Dipietro Date: Fri, 19 Jan 2024 11:01:33 -0800 Subject: [PATCH 51/74] tcp: Add memory barrier to tcp_push() [ Upstream commit 7267e8dcad6b2f9fce05a6a06335d7040acbc2b6 ] On CPUs with weak memory models, reads and updates performed by tcp_push to the sk variables can get reordered leaving the socket throttled when it should not. The tasklet running tcp_wfree() may also not observe the memory updates in time and will skip flushing any packets throttled by tcp_push(), delaying the sending. This can pathologically cause 40ms extra latency due to bad interactions with delayed acks. Adding a memory barrier in tcp_push removes the bug, similarly to the previous commit bf06200e732d ("tcp: tsq: fix nonagle handling"). smp_mb__after_atomic() is used to not incur in unnecessary overhead on x86 since not affected. Patch has been tested using an AWS c7g.2xlarge instance with Ubuntu 22.04 and Apache Tomcat 9.0.83 running the basic servlet below: import java.io.IOException; import java.io.OutputStreamWriter; import java.io.PrintWriter; import javax.servlet.ServletException; import javax.servlet.http.HttpServlet; import javax.servlet.http.HttpServletRequest; import javax.servlet.http.HttpServletResponse; public class HelloWorldServlet extends HttpServlet { @Override protected void doGet(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException { response.setContentType("text/html;charset=utf-8"); OutputStreamWriter osw = new OutputStreamWriter(response.getOutputStream(),"UTF-8"); String s = "a".repeat(3096); osw.write(s,0,s.length()); osw.flush(); } } Load was applied using wrk2 (https://github.com/kinvolk/wrk2) from an AWS c6i.8xlarge instance. Before the patch an additional 40ms latency from P99.99+ values is observed while, with the patch, the extra latency disappears. No patch and tcp_autocorking=1 ./wrk -t32 -c128 -d40s --latency -R10000 http://172.31.60.173:8080/hello/hello ... 50.000% 0.91ms 75.000% 1.13ms 90.000% 1.46ms 99.000% 1.74ms 99.900% 1.89ms 99.990% 41.95ms <<< 40+ ms extra latency 99.999% 48.32ms 100.000% 48.96ms With patch and tcp_autocorking=1 ./wrk -t32 -c128 -d40s --latency -R10000 http://172.31.60.173:8080/hello/hello ... 50.000% 0.90ms 75.000% 1.13ms 90.000% 1.45ms 99.000% 1.72ms 99.900% 1.83ms 99.990% 2.11ms <<< no 40+ ms extra latency 99.999% 2.53ms 100.000% 2.62ms Patch has been also tested on x86 (m7i.2xlarge instance) which it is not affected by this issue and the patch doesn't introduce any additional delay. Fixes: 7aa5470c2c09 ("tcp: tsq: move tsq_flags close to sk_wmem_alloc") Signed-off-by: Salvatore Dipietro Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20240119190133.43698-1-dipiets@amazon.com Signed-off-by: Paolo Abeni Signed-off-by: Sasha Levin --- net/ipv4/tcp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 00648a478c6a..712186336997 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -733,6 +733,7 @@ static void tcp_push(struct sock *sk, int flags, int mss_now, if (!test_bit(TSQ_THROTTLED, &sk->sk_tsq_flags)) { NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAUTOCORKING); set_bit(TSQ_THROTTLED, &sk->sk_tsq_flags); + smp_mb__after_atomic(); } /* It is possible TX completion already happened * before we set TSQ_THROTTLED. From 8ea171e43782a4f4d785ee9a3218860129a121a9 Mon Sep 17 00:00:00 2001 From: Peiyong Wang Date: Tue, 5 Mar 2024 01:46:22 +0000 Subject: [PATCH 52/74] BACKPORT: net: core: enable SO_BINDTODEVICE for non-root users Currently, SO_BINDTODEVICE requires CAP_NET_RAW. This change allows a non-root user to bind a socket to an interface if it is not already bound. This is useful to allow an application to bind itself to a specific VRF for outgoing or incoming connections. Currently, an application wanting to manage connections through several VRF need to be privileged. Previously, IP_UNICAST_IF and IPV6_UNICAST_IF were added for Wine (76e21053b5bf3 and c4062dfc425e9) specifically for use by non-root processes. However, they are restricted to sendmsg() and not usable with TCP. Allowing SO_BINDTODEVICE would allow TCP clients to get the same privilege. As for TCP servers, outside the VRF use case, SO_BINDTODEVICE would only further restrict connections a server could accept. When an application is restricted to a VRF (with `ip vrf exec`), the socket is bound to an interface at creation and therefore, a non-privileged call to SO_BINDTODEVICE to escape the VRF fails. When an application bound a socket to SO_BINDTODEVICE and transmit it to a non-privileged process through a Unix socket, a tentative to change the bound device also fails. Before: >>> import socket >>> s=socket.socket(socket.AF_INET, socket.SOCK_STREAM) >>> s.setsockopt(socket.SOL_SOCKET, socket.SO_BINDTODEVICE, b"dummy0") Traceback (most recent call last): File "", line 1, in PermissionError: [Errno 1] Operation not permitted After: >>> import socket >>> s=socket.socket(socket.AF_INET, socket.SOCK_STREAM) >>> s.setsockopt(socket.SOL_SOCKET, socket.SO_BINDTODEVICE, b"dummy0") >>> s.setsockopt(socket.SOL_SOCKET, socket.SO_BINDTODEVICE, b"dummy0") Traceback (most recent call last): File "", line 1, in PermissionError: [Errno 1] Operation not permitted Bug: 323792489 Signed-off-by: Vincent Bernat Reviewed-by: David Ahern Signed-off-by: David S. Miller (cherry picked from commit c427bfec18f2190b8f4718785ee8ed2db4f84ee6) Change-Id: Ie3f4c536b78da12dd961dc681c6dfb0cdf1a06b7 Signed-off-by: Peiyong Wang --- net/core/sock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/sock.c b/net/core/sock.c index e1379df5d8d0..4b143f3ae1d9 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -531,7 +531,7 @@ static int sock_setbindtodevice(struct sock *sk, char __user *optval, /* Sorry... */ ret = -EPERM; - if (!ns_capable(net->user_ns, CAP_NET_RAW)) + if (sk->sk_bound_dev_if && !ns_capable(net->user_ns, CAP_NET_RAW)) goto out; ret = -EINVAL; From 02965842883d4b8f8ecb5c4a9ac134f32054ed7e Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 4 Jan 2024 12:16:03 +0000 Subject: [PATCH 53/74] Reapply "cred: switch to using atomic_long_t" This reverts commit b2ea47a79d30760d6346e00530c7eef96b25eb6b as it is needed in Android systems. It still breaks the ABI, but that is no longer an issue on this branch. Bug: 317347552 Change-Id: I9c7f8901c5aabf6e0378b5218f587d6c7335522f Signed-off-by: Greg Kroah-Hartman --- include/linux/cred.h | 6 ++-- kernel/cred.c | 66 ++++++++++++++++++++++---------------------- 2 files changed, 36 insertions(+), 36 deletions(-) diff --git a/include/linux/cred.h b/include/linux/cred.h index 4b081e4911c8..62509a19ea08 100644 --- a/include/linux/cred.h +++ b/include/linux/cred.h @@ -114,7 +114,7 @@ static inline int groups_search(const struct group_info *group_info, kgid_t grp) * same context as task->real_cred. */ struct cred { - atomic_t usage; + atomic_long_t usage; #ifdef CONFIG_DEBUG_CREDENTIALS atomic_t subscribers; /* number of processes subscribed */ void *put_addr; @@ -231,7 +231,7 @@ static inline bool cap_ambient_invariant_ok(const struct cred *cred) */ static inline struct cred *get_new_cred(struct cred *cred) { - atomic_inc(&cred->usage); + atomic_long_inc(&cred->usage); return cred; } @@ -275,7 +275,7 @@ static inline void put_cred(const struct cred *_cred) if (cred) { validate_creds(cred); - if (atomic_dec_and_test(&(cred)->usage)) + if (atomic_long_dec_and_test(&(cred)->usage)) __put_cred(cred); } } diff --git a/kernel/cred.c b/kernel/cred.c index a9f0f8b21d8c..8c58f0f63af2 100644 --- a/kernel/cred.c +++ b/kernel/cred.c @@ -101,17 +101,17 @@ static void put_cred_rcu(struct rcu_head *rcu) #ifdef CONFIG_DEBUG_CREDENTIALS if (cred->magic != CRED_MAGIC_DEAD || - atomic_read(&cred->usage) != 0 || + atomic_long_read(&cred->usage) != 0 || read_cred_subscribers(cred) != 0) panic("CRED: put_cred_rcu() sees %p with" - " mag %x, put %p, usage %d, subscr %d\n", + " mag %x, put %p, usage %ld, subscr %d\n", cred, cred->magic, cred->put_addr, - atomic_read(&cred->usage), + atomic_long_read(&cred->usage), read_cred_subscribers(cred)); #else - if (atomic_read(&cred->usage) != 0) - panic("CRED: put_cred_rcu() sees %p with usage %d\n", - cred, atomic_read(&cred->usage)); + if (atomic_long_read(&cred->usage) != 0) + panic("CRED: put_cred_rcu() sees %p with usage %ld\n", + cred, atomic_long_read(&cred->usage)); #endif security_cred_free(cred); @@ -134,11 +134,11 @@ static void put_cred_rcu(struct rcu_head *rcu) */ void __put_cred(struct cred *cred) { - kdebug("__put_cred(%p{%d,%d})", cred, - atomic_read(&cred->usage), + kdebug("__put_cred(%p{%ld,%d})", cred, + atomic_long_read(&cred->usage), read_cred_subscribers(cred)); - BUG_ON(atomic_read(&cred->usage) != 0); + BUG_ON(atomic_long_read(&cred->usage) != 0); #ifdef CONFIG_DEBUG_CREDENTIALS BUG_ON(read_cred_subscribers(cred) != 0); cred->magic = CRED_MAGIC_DEAD; @@ -161,8 +161,8 @@ void exit_creds(struct task_struct *tsk) { struct cred *cred; - kdebug("exit_creds(%u,%p,%p,{%d,%d})", tsk->pid, tsk->real_cred, tsk->cred, - atomic_read(&tsk->cred->usage), + kdebug("exit_creds(%u,%p,%p,{%ld,%d})", tsk->pid, tsk->real_cred, tsk->cred, + atomic_long_read(&tsk->cred->usage), read_cred_subscribers(tsk->cred)); cred = (struct cred *) tsk->real_cred; @@ -197,7 +197,7 @@ const struct cred *get_task_cred(struct task_struct *task) do { cred = __task_cred((task)); BUG_ON(!cred); - } while (!atomic_inc_not_zero(&((struct cred *)cred)->usage)); + } while (!atomic_long_inc_not_zero(&((struct cred *)cred)->usage)); rcu_read_unlock(); return cred; @@ -215,7 +215,7 @@ struct cred *cred_alloc_blank(void) if (!new) return NULL; - atomic_set(&new->usage, 1); + atomic_long_set(&new->usage, 1); #ifdef CONFIG_DEBUG_CREDENTIALS new->magic = CRED_MAGIC; #endif @@ -262,7 +262,7 @@ struct cred *prepare_creds(void) memcpy(new, old, sizeof(struct cred)); new->non_rcu = 0; - atomic_set(&new->usage, 1); + atomic_long_set(&new->usage, 1); set_cred_subscribers(new, 0); get_group_info(new->group_info); get_uid(new->user); @@ -338,8 +338,8 @@ int copy_creds(struct task_struct *p, unsigned long clone_flags) p->real_cred = get_cred(p->cred); get_cred(p->cred); alter_cred_subscribers(p->cred, 2); - kdebug("share_creds(%p{%d,%d})", - p->cred, atomic_read(&p->cred->usage), + kdebug("share_creds(%p{%ld,%d})", + p->cred, atomic_long_read(&p->cred->usage), read_cred_subscribers(p->cred)); atomic_inc(&p->cred->user->processes); return 0; @@ -429,8 +429,8 @@ int commit_creds(struct cred *new) struct task_struct *task = current; const struct cred *old = task->real_cred; - kdebug("commit_creds(%p{%d,%d})", new, - atomic_read(&new->usage), + kdebug("commit_creds(%p{%ld,%d})", new, + atomic_long_read(&new->usage), read_cred_subscribers(new)); BUG_ON(task->cred != old); @@ -439,7 +439,7 @@ int commit_creds(struct cred *new) validate_creds(old); validate_creds(new); #endif - BUG_ON(atomic_read(&new->usage) < 1); + BUG_ON(atomic_long_read(&new->usage) < 1); get_cred(new); /* we will require a ref for the subj creds too */ @@ -512,14 +512,14 @@ EXPORT_SYMBOL(commit_creds); */ void abort_creds(struct cred *new) { - kdebug("abort_creds(%p{%d,%d})", new, - atomic_read(&new->usage), + kdebug("abort_creds(%p{%ld,%d})", new, + atomic_long_read(&new->usage), read_cred_subscribers(new)); #ifdef CONFIG_DEBUG_CREDENTIALS BUG_ON(read_cred_subscribers(new) != 0); #endif - BUG_ON(atomic_read(&new->usage) < 1); + BUG_ON(atomic_long_read(&new->usage) < 1); put_cred(new); } EXPORT_SYMBOL(abort_creds); @@ -535,8 +535,8 @@ const struct cred *override_creds(const struct cred *new) { const struct cred *old = current->cred; - kdebug("override_creds(%p{%d,%d})", new, - atomic_read(&new->usage), + kdebug("override_creds(%p{%ld,%d})", new, + atomic_long_read(&new->usage), read_cred_subscribers(new)); validate_creds(old); @@ -558,8 +558,8 @@ const struct cred *override_creds(const struct cred *new) rcu_assign_pointer(current->cred, new); alter_cred_subscribers(old, -1); - kdebug("override_creds() = %p{%d,%d}", old, - atomic_read(&old->usage), + kdebug("override_creds() = %p{%ld,%d}", old, + atomic_long_read(&old->usage), read_cred_subscribers(old)); return old; } @@ -576,8 +576,8 @@ void revert_creds(const struct cred *old) { const struct cred *override = current->cred; - kdebug("revert_creds(%p{%d,%d})", old, - atomic_read(&old->usage), + kdebug("revert_creds(%p{%ld,%d})", old, + atomic_long_read(&old->usage), read_cred_subscribers(old)); validate_creds(old); @@ -637,7 +637,7 @@ struct cred *prepare_kernel_cred(struct task_struct *daemon) *new = *old; new->non_rcu = 0; - atomic_set(&new->usage, 1); + atomic_long_set(&new->usage, 1); set_cred_subscribers(new, 0); get_uid(new->user); get_user_ns(new->user_ns); @@ -760,8 +760,8 @@ static void dump_invalid_creds(const struct cred *cred, const char *label, cred == tsk->cred ? "[eff]" : ""); printk(KERN_ERR "CRED: ->magic=%x, put_addr=%p\n", cred->magic, cred->put_addr); - printk(KERN_ERR "CRED: ->usage=%d, subscr=%d\n", - atomic_read(&cred->usage), + printk(KERN_ERR "CRED: ->usage=%ld, subscr=%d\n", + atomic_long_read(&cred->usage), read_cred_subscribers(cred)); printk(KERN_ERR "CRED: ->*uid = { %d,%d,%d,%d }\n", from_kuid_munged(&init_user_ns, cred->uid), @@ -833,9 +833,9 @@ EXPORT_SYMBOL(__validate_process_creds); */ void validate_creds_for_do_exit(struct task_struct *tsk) { - kdebug("validate_creds_for_do_exit(%p,%p{%d,%d})", + kdebug("validate_creds_for_do_exit(%p,%p{%ld,%d})", tsk->real_cred, tsk->cred, - atomic_read(&tsk->cred->usage), + atomic_long_read(&tsk->cred->usage), read_cred_subscribers(tsk->cred)); __validate_process_creds(tsk, __FILE__, __LINE__); From 08d0950cd1e0165da7971753e38b0ee575aeb9d4 Mon Sep 17 00:00:00 2001 From: Giuliano Procida Date: Fri, 15 Mar 2024 12:38:56 +0000 Subject: [PATCH 54/74] Revert "Reapply "cred: switch to using atomic_long_t"" This reverts commit 02965842883d4b8f8ecb5c4a9ac134f32054ed7e. Reason for revert: suspect in b/329612465 failures Change-Id: Ie523860ae1c8eb91f19f42ef496a5ce3897b921d --- include/linux/cred.h | 6 ++-- kernel/cred.c | 66 ++++++++++++++++++++++---------------------- 2 files changed, 36 insertions(+), 36 deletions(-) diff --git a/include/linux/cred.h b/include/linux/cred.h index 62509a19ea08..4b081e4911c8 100644 --- a/include/linux/cred.h +++ b/include/linux/cred.h @@ -114,7 +114,7 @@ static inline int groups_search(const struct group_info *group_info, kgid_t grp) * same context as task->real_cred. */ struct cred { - atomic_long_t usage; + atomic_t usage; #ifdef CONFIG_DEBUG_CREDENTIALS atomic_t subscribers; /* number of processes subscribed */ void *put_addr; @@ -231,7 +231,7 @@ static inline bool cap_ambient_invariant_ok(const struct cred *cred) */ static inline struct cred *get_new_cred(struct cred *cred) { - atomic_long_inc(&cred->usage); + atomic_inc(&cred->usage); return cred; } @@ -275,7 +275,7 @@ static inline void put_cred(const struct cred *_cred) if (cred) { validate_creds(cred); - if (atomic_long_dec_and_test(&(cred)->usage)) + if (atomic_dec_and_test(&(cred)->usage)) __put_cred(cred); } } diff --git a/kernel/cred.c b/kernel/cred.c index 8c58f0f63af2..a9f0f8b21d8c 100644 --- a/kernel/cred.c +++ b/kernel/cred.c @@ -101,17 +101,17 @@ static void put_cred_rcu(struct rcu_head *rcu) #ifdef CONFIG_DEBUG_CREDENTIALS if (cred->magic != CRED_MAGIC_DEAD || - atomic_long_read(&cred->usage) != 0 || + atomic_read(&cred->usage) != 0 || read_cred_subscribers(cred) != 0) panic("CRED: put_cred_rcu() sees %p with" - " mag %x, put %p, usage %ld, subscr %d\n", + " mag %x, put %p, usage %d, subscr %d\n", cred, cred->magic, cred->put_addr, - atomic_long_read(&cred->usage), + atomic_read(&cred->usage), read_cred_subscribers(cred)); #else - if (atomic_long_read(&cred->usage) != 0) - panic("CRED: put_cred_rcu() sees %p with usage %ld\n", - cred, atomic_long_read(&cred->usage)); + if (atomic_read(&cred->usage) != 0) + panic("CRED: put_cred_rcu() sees %p with usage %d\n", + cred, atomic_read(&cred->usage)); #endif security_cred_free(cred); @@ -134,11 +134,11 @@ static void put_cred_rcu(struct rcu_head *rcu) */ void __put_cred(struct cred *cred) { - kdebug("__put_cred(%p{%ld,%d})", cred, - atomic_long_read(&cred->usage), + kdebug("__put_cred(%p{%d,%d})", cred, + atomic_read(&cred->usage), read_cred_subscribers(cred)); - BUG_ON(atomic_long_read(&cred->usage) != 0); + BUG_ON(atomic_read(&cred->usage) != 0); #ifdef CONFIG_DEBUG_CREDENTIALS BUG_ON(read_cred_subscribers(cred) != 0); cred->magic = CRED_MAGIC_DEAD; @@ -161,8 +161,8 @@ void exit_creds(struct task_struct *tsk) { struct cred *cred; - kdebug("exit_creds(%u,%p,%p,{%ld,%d})", tsk->pid, tsk->real_cred, tsk->cred, - atomic_long_read(&tsk->cred->usage), + kdebug("exit_creds(%u,%p,%p,{%d,%d})", tsk->pid, tsk->real_cred, tsk->cred, + atomic_read(&tsk->cred->usage), read_cred_subscribers(tsk->cred)); cred = (struct cred *) tsk->real_cred; @@ -197,7 +197,7 @@ const struct cred *get_task_cred(struct task_struct *task) do { cred = __task_cred((task)); BUG_ON(!cred); - } while (!atomic_long_inc_not_zero(&((struct cred *)cred)->usage)); + } while (!atomic_inc_not_zero(&((struct cred *)cred)->usage)); rcu_read_unlock(); return cred; @@ -215,7 +215,7 @@ struct cred *cred_alloc_blank(void) if (!new) return NULL; - atomic_long_set(&new->usage, 1); + atomic_set(&new->usage, 1); #ifdef CONFIG_DEBUG_CREDENTIALS new->magic = CRED_MAGIC; #endif @@ -262,7 +262,7 @@ struct cred *prepare_creds(void) memcpy(new, old, sizeof(struct cred)); new->non_rcu = 0; - atomic_long_set(&new->usage, 1); + atomic_set(&new->usage, 1); set_cred_subscribers(new, 0); get_group_info(new->group_info); get_uid(new->user); @@ -338,8 +338,8 @@ int copy_creds(struct task_struct *p, unsigned long clone_flags) p->real_cred = get_cred(p->cred); get_cred(p->cred); alter_cred_subscribers(p->cred, 2); - kdebug("share_creds(%p{%ld,%d})", - p->cred, atomic_long_read(&p->cred->usage), + kdebug("share_creds(%p{%d,%d})", + p->cred, atomic_read(&p->cred->usage), read_cred_subscribers(p->cred)); atomic_inc(&p->cred->user->processes); return 0; @@ -429,8 +429,8 @@ int commit_creds(struct cred *new) struct task_struct *task = current; const struct cred *old = task->real_cred; - kdebug("commit_creds(%p{%ld,%d})", new, - atomic_long_read(&new->usage), + kdebug("commit_creds(%p{%d,%d})", new, + atomic_read(&new->usage), read_cred_subscribers(new)); BUG_ON(task->cred != old); @@ -439,7 +439,7 @@ int commit_creds(struct cred *new) validate_creds(old); validate_creds(new); #endif - BUG_ON(atomic_long_read(&new->usage) < 1); + BUG_ON(atomic_read(&new->usage) < 1); get_cred(new); /* we will require a ref for the subj creds too */ @@ -512,14 +512,14 @@ EXPORT_SYMBOL(commit_creds); */ void abort_creds(struct cred *new) { - kdebug("abort_creds(%p{%ld,%d})", new, - atomic_long_read(&new->usage), + kdebug("abort_creds(%p{%d,%d})", new, + atomic_read(&new->usage), read_cred_subscribers(new)); #ifdef CONFIG_DEBUG_CREDENTIALS BUG_ON(read_cred_subscribers(new) != 0); #endif - BUG_ON(atomic_long_read(&new->usage) < 1); + BUG_ON(atomic_read(&new->usage) < 1); put_cred(new); } EXPORT_SYMBOL(abort_creds); @@ -535,8 +535,8 @@ const struct cred *override_creds(const struct cred *new) { const struct cred *old = current->cred; - kdebug("override_creds(%p{%ld,%d})", new, - atomic_long_read(&new->usage), + kdebug("override_creds(%p{%d,%d})", new, + atomic_read(&new->usage), read_cred_subscribers(new)); validate_creds(old); @@ -558,8 +558,8 @@ const struct cred *override_creds(const struct cred *new) rcu_assign_pointer(current->cred, new); alter_cred_subscribers(old, -1); - kdebug("override_creds() = %p{%ld,%d}", old, - atomic_long_read(&old->usage), + kdebug("override_creds() = %p{%d,%d}", old, + atomic_read(&old->usage), read_cred_subscribers(old)); return old; } @@ -576,8 +576,8 @@ void revert_creds(const struct cred *old) { const struct cred *override = current->cred; - kdebug("revert_creds(%p{%ld,%d})", old, - atomic_long_read(&old->usage), + kdebug("revert_creds(%p{%d,%d})", old, + atomic_read(&old->usage), read_cred_subscribers(old)); validate_creds(old); @@ -637,7 +637,7 @@ struct cred *prepare_kernel_cred(struct task_struct *daemon) *new = *old; new->non_rcu = 0; - atomic_long_set(&new->usage, 1); + atomic_set(&new->usage, 1); set_cred_subscribers(new, 0); get_uid(new->user); get_user_ns(new->user_ns); @@ -760,8 +760,8 @@ static void dump_invalid_creds(const struct cred *cred, const char *label, cred == tsk->cred ? "[eff]" : ""); printk(KERN_ERR "CRED: ->magic=%x, put_addr=%p\n", cred->magic, cred->put_addr); - printk(KERN_ERR "CRED: ->usage=%ld, subscr=%d\n", - atomic_long_read(&cred->usage), + printk(KERN_ERR "CRED: ->usage=%d, subscr=%d\n", + atomic_read(&cred->usage), read_cred_subscribers(cred)); printk(KERN_ERR "CRED: ->*uid = { %d,%d,%d,%d }\n", from_kuid_munged(&init_user_ns, cred->uid), @@ -833,9 +833,9 @@ EXPORT_SYMBOL(__validate_process_creds); */ void validate_creds_for_do_exit(struct task_struct *tsk) { - kdebug("validate_creds_for_do_exit(%p,%p{%ld,%d})", + kdebug("validate_creds_for_do_exit(%p,%p{%d,%d})", tsk->real_cred, tsk->cred, - atomic_long_read(&tsk->cred->usage), + atomic_read(&tsk->cred->usage), read_cred_subscribers(tsk->cred)); __validate_process_creds(tsk, __FILE__, __LINE__); From c8b16e45250e860cf791911d0ffa96822515970f Mon Sep 17 00:00:00 2001 From: Michael Bestas Date: Mon, 15 Apr 2024 23:19:16 +0300 Subject: [PATCH 55/74] Revert "erofs: fix lz4 inplace decompression" This has been applied on top of a wrong revert. The proper backport will be applied after reintroducing the reverted commit. This reverts commit 1a9efb0ce229854b8c4d4d83dedd80da59e2eee9. Change-Id: I6231a3ce50ae0dec9a5633784877decc1047c327 --- fs/erofs/decompressor.c | 24 ++++++++++-------------- 1 file changed, 10 insertions(+), 14 deletions(-) diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c index 04468c05cba7..5396cf0ca1d4 100644 --- a/fs/erofs/decompressor.c +++ b/fs/erofs/decompressor.c @@ -23,8 +23,7 @@ struct z_erofs_decompressor { */ int (*prepare_destpages)(struct z_erofs_decompress_req *rq, struct list_head *pagepool); - int (*decompress)(struct z_erofs_decompress_req *rq, u8 *out, - u8 *obase); + int (*decompress)(struct z_erofs_decompress_req *rq, u8 *out); char *name; }; @@ -151,13 +150,10 @@ static void *generic_copy_inplace_data(struct z_erofs_decompress_req *rq, return tmp; } -static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out, - u8 *obase) +static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out) { - const uint nrpages_out = PAGE_ALIGN(rq->pageofs_out + - rq->outputsize) >> PAGE_SHIFT; unsigned int inputmargin, inlen; - u8 *src, *src2; + u8 *src; bool copied, support_0padding; int ret; @@ -165,7 +161,6 @@ static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out, return -EOPNOTSUPP; src = kmap_atomic(*rq->in); - src2 = src; inputmargin = 0; support_0padding = false; @@ -188,15 +183,16 @@ static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out, if (rq->inplace_io) { const uint oend = (rq->pageofs_out + rq->outputsize) & ~PAGE_MASK; + const uint nr = PAGE_ALIGN(rq->pageofs_out + + rq->outputsize) >> PAGE_SHIFT; + if (rq->partial_decoding || !support_0padding || - rq->out[nrpages_out - 1] != rq->in[0] || + rq->out[nr - 1] != rq->in[0] || rq->inputsize - oend < LZ4_DECOMPRESS_INPLACE_MARGIN(inlen)) { src = generic_copy_inplace_data(rq, src, inputmargin); inputmargin = 0; copied = true; - } else { - src = obase + ((nrpages_out - 1) << PAGE_SHIFT); } } @@ -226,7 +222,7 @@ static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out, if (copied) erofs_put_pcpubuf(src); else - kunmap_atomic(src2); + kunmap_atomic(src); return ret; } @@ -296,7 +292,7 @@ static int z_erofs_decompress_generic(struct z_erofs_decompress_req *rq, return PTR_ERR(dst); rq->inplace_io = false; - ret = alg->decompress(rq, dst, NULL); + ret = alg->decompress(rq, dst); if (!ret) copy_from_pcpubuf(rq->out, dst, rq->pageofs_out, rq->outputsize); @@ -330,7 +326,7 @@ static int z_erofs_decompress_generic(struct z_erofs_decompress_req *rq, dst_maptype = 2; dstmap_out: - ret = alg->decompress(rq, dst + rq->pageofs_out, dst); + ret = alg->decompress(rq, dst + rq->pageofs_out); if (!dst_maptype) kunmap_atomic(dst); From 22154fcf0b8c36c03da668d8629fbd47f84ec750 Mon Sep 17 00:00:00 2001 From: Michael Bestas Date: Mon, 15 Apr 2024 23:20:28 +0300 Subject: [PATCH 56/74] Reapply "erofs: support decompress big pcluster for lz4 backend" This reverts commit 950e1e4a2d76f45fb4e22d7a38441217c77cda8b. Change-Id: I75bf2ef77e881cb862af3a5598ef5b6feb74511a --- fs/erofs/decompressor.c | 214 +++++++++++++++++++++++----------------- fs/erofs/internal.h | 15 +++ 2 files changed, 136 insertions(+), 93 deletions(-) diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c index 5396cf0ca1d4..8193c14bb111 100644 --- a/fs/erofs/decompressor.c +++ b/fs/erofs/decompressor.c @@ -123,44 +123,85 @@ static int z_erofs_lz4_prepare_destpages(struct z_erofs_decompress_req *rq, return kaddr ? 1 : 0; } -static void *generic_copy_inplace_data(struct z_erofs_decompress_req *rq, - u8 *src, unsigned int pageofs_in) +static void *z_erofs_handle_inplace_io(struct z_erofs_decompress_req *rq, + void *inpage, unsigned int *inputmargin, int *maptype, + bool support_0padding) { - /* - * if in-place decompression is ongoing, those decompressed - * pages should be copied in order to avoid being overlapped. - */ - struct page **in = rq->in; - u8 *const tmp = erofs_get_pcpubuf(1); - u8 *tmpp = tmp; - unsigned int inlen = rq->inputsize - pageofs_in; - unsigned int count = min_t(uint, inlen, PAGE_SIZE - pageofs_in); + unsigned int nrpages_in, nrpages_out; + unsigned int ofull, oend, inputsize, total, i, j; + struct page **in; + void *src, *tmp; - while (tmpp < tmp + inlen) { - if (!src) - src = kmap_atomic(*in); - memcpy(tmpp, src + pageofs_in, count); - kunmap_atomic(src); - src = NULL; - tmpp += count; - pageofs_in = 0; - count = PAGE_SIZE; - ++in; + inputsize = rq->inputsize; + nrpages_in = PAGE_ALIGN(inputsize) >> PAGE_SHIFT; + oend = rq->pageofs_out + rq->outputsize; + ofull = PAGE_ALIGN(oend); + nrpages_out = ofull >> PAGE_SHIFT; + + if (rq->inplace_io) { + if (rq->partial_decoding || !support_0padding || + ofull - oend < LZ4_DECOMPRESS_INPLACE_MARGIN(inputsize)) + goto docopy; + + for (i = 0; i < nrpages_in; ++i) { + DBG_BUGON(rq->in[i] == NULL); + for (j = 0; j < nrpages_out - nrpages_in + i; ++j) + if (rq->out[j] == rq->in[i]) + goto docopy; + } } - return tmp; + + if (nrpages_in <= 1) { + *maptype = 0; + return inpage; + } + kunmap_atomic(inpage); + might_sleep(); + src = erofs_vm_map_ram(rq->in, nrpages_in); + if (!src) + return ERR_PTR(-ENOMEM); + *maptype = 1; + return src; + +docopy: + /* Or copy compressed data which can be overlapped to per-CPU buffer */ + in = rq->in; + src = erofs_get_pcpubuf(nrpages_in); + if (!src) { + DBG_BUGON(1); + kunmap_atomic(inpage); + return ERR_PTR(-EFAULT); + } + + tmp = src; + total = rq->inputsize; + while (total) { + unsigned int page_copycnt = + min_t(unsigned int, total, PAGE_SIZE - *inputmargin); + + if (!inpage) + inpage = kmap_atomic(*in); + memcpy(tmp, inpage + *inputmargin, page_copycnt); + kunmap_atomic(inpage); + inpage = NULL; + tmp += page_copycnt; + total -= page_copycnt; + ++in; + *inputmargin = 0; + } + *maptype = 2; + return src; } static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out) { - unsigned int inputmargin, inlen; - u8 *src; - bool copied, support_0padding; - int ret; + unsigned int inputmargin; + u8 *headpage, *src; + bool support_0padding; + int ret, maptype; - if (rq->inputsize > PAGE_SIZE) - return -EOPNOTSUPP; - - src = kmap_atomic(*rq->in); + DBG_BUGON(*rq->in == NULL); + headpage = kmap_atomic(*rq->in); inputmargin = 0; support_0padding = false; @@ -168,49 +209,36 @@ static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out) if (erofs_sb_has_lz4_0padding(EROFS_SB(rq->sb))) { support_0padding = true; - while (!src[inputmargin & ~PAGE_MASK]) + while (!headpage[inputmargin & ~PAGE_MASK]) if (!(++inputmargin & ~PAGE_MASK)) break; if (inputmargin >= rq->inputsize) { - kunmap_atomic(src); + kunmap_atomic(headpage); return -EIO; } } - copied = false; - inlen = rq->inputsize - inputmargin; - if (rq->inplace_io) { - const uint oend = (rq->pageofs_out + - rq->outputsize) & ~PAGE_MASK; - const uint nr = PAGE_ALIGN(rq->pageofs_out + - rq->outputsize) >> PAGE_SHIFT; - - if (rq->partial_decoding || !support_0padding || - rq->out[nr - 1] != rq->in[0] || - rq->inputsize - oend < - LZ4_DECOMPRESS_INPLACE_MARGIN(inlen)) { - src = generic_copy_inplace_data(rq, src, inputmargin); - inputmargin = 0; - copied = true; - } - } + rq->inputsize -= inputmargin; + src = z_erofs_handle_inplace_io(rq, headpage, &inputmargin, &maptype, + support_0padding); + if (IS_ERR(src)) + return PTR_ERR(src); /* legacy format could compress extra data in a pcluster. */ if (rq->partial_decoding || !support_0padding) ret = LZ4_decompress_safe_partial(src + inputmargin, out, - inlen, rq->outputsize, - rq->outputsize); + rq->inputsize, rq->outputsize, rq->outputsize); else ret = LZ4_decompress_safe(src + inputmargin, out, - inlen, rq->outputsize); + rq->inputsize, rq->outputsize); if (ret != rq->outputsize) { erofs_err(rq->sb, "failed to decompress %d in[%u, %u] out[%u]", - ret, inlen, inputmargin, rq->outputsize); + ret, rq->inputsize, inputmargin, rq->outputsize); print_hex_dump(KERN_DEBUG, "[ in]: ", DUMP_PREFIX_OFFSET, - 16, 1, src + inputmargin, inlen, true); + 16, 1, src + inputmargin, rq->inputsize, true); print_hex_dump(KERN_DEBUG, "[out]: ", DUMP_PREFIX_OFFSET, 16, 1, out, rq->outputsize, true); @@ -219,10 +247,16 @@ static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out) ret = -EIO; } - if (copied) - erofs_put_pcpubuf(src); - else + if (maptype == 0) { kunmap_atomic(src); + } else if (maptype == 1) { + vm_unmap_ram(src, PAGE_ALIGN(rq->inputsize) >> PAGE_SHIFT); + } else if (maptype == 2) { + erofs_put_pcpubuf(src); + } else { + DBG_BUGON(1); + return -EFAULT; + } return ret; } @@ -272,57 +306,51 @@ static int z_erofs_decompress_generic(struct z_erofs_decompress_req *rq, const struct z_erofs_decompressor *alg = decompressors + rq->alg; unsigned int dst_maptype; void *dst; - int ret, i; + int ret; - if (nrpages_out == 1 && !rq->inplace_io) { - DBG_BUGON(!*rq->out); - dst = kmap_atomic(*rq->out); - dst_maptype = 0; - goto dstmap_out; - } + /* two optimized fast paths only for non bigpcluster cases yet */ + if (rq->inputsize <= PAGE_SIZE) { + if (nrpages_out == 1 && !rq->inplace_io) { + DBG_BUGON(!*rq->out); + dst = kmap_atomic(*rq->out); + dst_maptype = 0; + goto dstmap_out; + } - /* - * For the case of small output size (especially much less - * than PAGE_SIZE), memcpy the decompressed data rather than - * compressed data is preferred. - */ - if (rq->outputsize <= PAGE_SIZE * 7 / 8) { - dst = erofs_get_pcpubuf(1); - if (IS_ERR(dst)) - return PTR_ERR(dst); + /* + * For the case of small output size (especially much less + * than PAGE_SIZE), memcpy the decompressed data rather than + * compressed data is preferred. + */ + if (rq->outputsize <= PAGE_SIZE * 7 / 8) { + dst = erofs_get_pcpubuf(1); + if (IS_ERR(dst)) + return PTR_ERR(dst); - rq->inplace_io = false; - ret = alg->decompress(rq, dst); - if (!ret) - copy_from_pcpubuf(rq->out, dst, rq->pageofs_out, - rq->outputsize); + rq->inplace_io = false; + ret = alg->decompress(rq, dst); + if (!ret) + copy_from_pcpubuf(rq->out, dst, rq->pageofs_out, + rq->outputsize); - erofs_put_pcpubuf(dst); - return ret; + erofs_put_pcpubuf(dst); + return ret; + } } + /* general decoding path which can be used for all cases */ ret = alg->prepare_destpages(rq, pagepool); - if (ret < 0) { + if (ret < 0) return ret; - } else if (ret) { + if (ret) { dst = page_address(*rq->out); dst_maptype = 1; goto dstmap_out; } - i = 0; - while (1) { - dst = vm_map_ram(rq->out, nrpages_out, -1, PAGE_KERNEL); - - /* retry two more times (totally 3 times) */ - if (dst || ++i >= 3) - break; - vm_unmap_aliases(); - } - + dst = erofs_vm_map_ram(rq->out, nrpages_out); if (!dst) return -ENOMEM; - dst_maptype = 2; dstmap_out: diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index 814541bf0cce..50d359101642 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -396,6 +396,21 @@ int erofs_namei(struct inode *dir, struct qstr *name, /* dir.c */ extern const struct file_operations erofs_dir_fops; +static inline void *erofs_vm_map_ram(struct page **pages, unsigned int count) +{ + int retried = 0; + + while (1) { + void *p = vm_map_ram(pages, count, -1, PAGE_KERNEL); + + /* retry two more times (totally 3 times) */ + if (p || ++retried >= 3) + return p; + vm_unmap_aliases(); + } + return NULL; +} + /* pcpubuf.c */ void *erofs_get_pcpubuf(unsigned int requiredpages); void erofs_put_pcpubuf(void *ptr); From 054e3f52db818b64f4d98267655b200f00a1fa62 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Tue, 15 Jan 2019 01:39:29 +0800 Subject: [PATCH 57/74] staging: erofs: add document This documents key feature, usage, and on-disk design of erofs. Reviewed-by: Chao Yu Cc: Change-Id: I9b3b4f301e43d33b9c535acf1bafa2e3cf861a76 Signed-off-by: Gao Xiang Signed-off-by: Greg Kroah-Hartman --- .../erofs/Documentation/filesystems/erofs.txt | 206 ++++++++++++++++++ 1 file changed, 206 insertions(+) create mode 100644 drivers/staging/erofs/Documentation/filesystems/erofs.txt diff --git a/drivers/staging/erofs/Documentation/filesystems/erofs.txt b/drivers/staging/erofs/Documentation/filesystems/erofs.txt new file mode 100644 index 000000000000..803988d74c21 --- /dev/null +++ b/drivers/staging/erofs/Documentation/filesystems/erofs.txt @@ -0,0 +1,206 @@ +Overview +======== + +EROFS file-system stands for Enhanced Read-Only File System. Different +from other read-only file systems, it aims to be designed for flexibility, +scalability, but be kept simple and high performance. + +It is designed as a better filesystem solution for the following scenarios: + - read-only storage media or + + - part of a fully trusted read-only solution, which means it needs to be + immutable and bit-for-bit identical to the official golden image for + their releases due to security and other considerations and + + - hope to save some extra storage space with guaranteed end-to-end performance + by using reduced metadata and transparent file compression, especially + for those embedded devices with limited memory (ex, smartphone); + +Here is the main features of EROFS: + - Little endian on-disk design; + + - Currently 4KB block size (nobh) and therefore maximum 16TB address space; + + - Metadata & data could be mixed by design; + + - 2 inode versions for different requirements: + v1 v2 + Inode metadata size: 32 bytes 64 bytes + Max file size: 4 GB 16 EB (also limited by max. vol size) + Max uids/gids: 65536 4294967296 + File creation time: no yes (64 + 32-bit timestamp) + Max hardlinks: 65536 4294967296 + Metadata reserved: 4 bytes 14 bytes + + - Support extended attributes (xattrs) as an option; + + - Support xattr inline and tail-end data inline for all files; + + - Support transparent file compression as an option: + LZ4 algorithm with 4 KB fixed-output compression for high performance; + +The following git tree provides the file system user-space tools under +development (ex, formatting tool mkfs.erofs): +>> git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs-utils.git + +Bugs and patches are welcome, please kindly help us and send to the following +linux-erofs mailing list: +>> linux-erofs mailing list + +Note that EROFS is still working in progress as a Linux staging driver, +Cc the staging mailing list as well is highly recommended: +>> Linux Driver Project Developer List + +Mount options +============= + +fault_injection=%d Enable fault injection in all supported types with + specified injection rate. Supported injection type: + Type_Name Type_Value + FAULT_KMALLOC 0x000000001 +(no)user_xattr Setup Extended User Attributes. Note: xattr is enabled + by default if CONFIG_EROFS_FS_XATTR is selected. +(no)acl Setup POSIX Access Control List. Note: acl is enabled + by default if CONFIG_EROFS_FS_POSIX_ACL is selected. + +On-disk details +=============== + +Summary +------- +Different from other read-only file systems, an EROFS volume is designed +to be as simple as possible: + + |-> aligned with the block size + ____________________________________________________________ + | |SB| | ... | Metadata | ... | Data | Metadata | ... | Data | + |_|__|_|_____|__________|_____|______|__________|_____|______| + 0 +1K + +All data areas should be aligned with the block size, but metadata areas +may not. All metadatas can be now observed in two different spaces (views): + 1. Inode metadata space + Each valid inode should be aligned with an inode slot, which is a fixed + value (32 bytes) and designed to be kept in line with v1 inode size. + + Each inode can be directly found with the following formula: + inode offset = meta_blkaddr * block_size + 32 * nid + + |-> aligned with 8B + |-> followed closely + + meta_blkaddr blocks |-> another slot + _____________________________________________________________________ + | ... | inode | xattrs | extents | data inline | ... | inode ... + |________|_______|(optional)|(optional)|__(optional)_|_____|__________ + |-> aligned with the inode slot size + . . + . . + . . + . . + . . + . . + .____________________________________________________|-> aligned with 4B + | xattr_ibody_header | shared xattrs | inline xattrs | + |____________________|_______________|_______________| + |-> 12 bytes <-|->x * 4 bytes<-| . + . . . + . . . + . . . + ._______________________________.______________________. + | id | id | id | id | ... | id | ent | ... | ent| ... | + |____|____|____|____|______|____|_____|_____|____|_____| + |-> aligned with 4B + |-> aligned with 4B + + Inode could be 32 or 64 bytes, which can be distinguished from a common + field which all inode versions have -- i_advise: + + __________________ __________________ + | i_advise | | i_advise | + |__________________| |__________________| + | ... | | ... | + | | | | + |__________________| 32 bytes | | + | | + |__________________| 64 bytes + + Xattrs, extents, data inline are followed by the corresponding inode with + proper alignes, and they could be optional for different data mappings, + _currently_ there are totally 3 valid data mappings supported: + + 1) flat file data without data inline (no extent); + 2) fixed-output size data compression (must have extents); + 3) flat file data with tail-end data inline (no extent); + + The size of the optional xattrs is indicated by i_xattr_count in inode + header. Large xattrs or xattrs shared by many different files can be + stored in shared xattrs metadata rather than inlined right after inode. + + 2. Shared xattrs metadata space + Shared xattrs space is similar to the above inode space, started with + a specific block indicated by xattr_blkaddr, organized one by one with + proper align. + + Each share xattr can also be directly found by the following formula: + xattr offset = xattr_blkaddr * block_size + 4 * xattr_id + + |-> aligned by 4 bytes + + xattr_blkaddr blocks |-> aligned with 4 bytes + _________________________________________________________________________ + | ... | xattr_entry | xattr data | ... | xattr_entry | xattr data ... + |________|_____________|_____________|_____|______________|_______________ + +Directories +----------- +All directories are now organized in a compact on-disk format. Note that +each directory block is divided into index and name areas in order to support +random file lookup, and all directory entries are _strictly_ recorded in +alphabetical order in order to support improved prefix binary search +algorithm (could refer to the related source code). + + ___________________________ + / | + / ______________|________________ + / / | nameoff1 | nameoffN-1 + ____________.______________._______________v________________v__________ +| dirent | dirent | ... | dirent | filename | filename | ... | filename | +|___.0___|____1___|_____|___N-1__|____0_____|____1_____|_____|___N-1____| + \ ^ + \ | * could have + \ | trailing '\0' + \________________________| nameoff0 + + Directory block + +Note that apart from the offset of the first filename, nameoff0 also indicates +the total number of directory entries in this block since it is no need to +introduce another on-disk field at all. + +Compression +----------- +Currently, EROFS supports 4KB fixed-output clustersize transparent file +compression, as illustrated below: + + |---- Variant-Length Extent ----|-------- VLE --------|----- VLE ----- + clusterofs clusterofs clusterofs + | | | logical data +_________v_______________________________v_____________________v_______________ +... | . | | . | | . | ... +____|____.________|_____________|________.____|_____________|__.__________|____ + |-> cluster <-|-> cluster <-|-> cluster <-|-> cluster <-|-> cluster <-| + size size size size size + . . . . + . . . . + . . . . + _______._____________._____________._____________._____________________ + ... | | | | ... physical data + _______|_____________|_____________|_____________|_____________________ + |-> cluster <-|-> cluster <-|-> cluster <-| + size size size + +Currently each on-disk physical cluster can contain 4KB (un)compressed data +at most. For each logical cluster, there is a corresponding on-disk index to +describe its cluster type, physical cluster address, etc. + +See "struct z_erofs_vle_decompressed_index" in erofs_fs.h for more details. + From e2ff41e85a8b50622a360735129629724b15d4ac Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Tue, 29 Jan 2019 16:35:20 +0800 Subject: [PATCH 58/74] staging: erofs: complete POSIX ACL support Let's add .get_acl() to read the file's acl from its xattrs to make POSIX ACL usable. Here is the on-disk detail, fullname: system.posix_acl_access struct erofs_xattr_entry: .e_name_len = 0 .e_name_index = EROFS_XATTR_INDEX_POSIX_ACL_ACCESS (2) fullname: system.posix_acl_default struct erofs_xattr_entry: .e_name_len = 0 .e_name_index = EROFS_XATTR_INDEX_POSIX_ACL_DEFAULT (3) Reviewed-by: Chao Yu Change-Id: I4ec29f5f22615c441e359249a5ca0917a7e8bb50 Signed-off-by: Gao Xiang Signed-off-by: Greg Kroah-Hartman --- drivers/staging/erofs/Documentation/filesystems/erofs.txt | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/staging/erofs/Documentation/filesystems/erofs.txt b/drivers/staging/erofs/Documentation/filesystems/erofs.txt index 803988d74c21..961ec4da7705 100644 --- a/drivers/staging/erofs/Documentation/filesystems/erofs.txt +++ b/drivers/staging/erofs/Documentation/filesystems/erofs.txt @@ -36,6 +36,8 @@ Here is the main features of EROFS: - Support xattr inline and tail-end data inline for all files; + - Support POSIX.1e ACLs by using xattrs; + - Support transparent file compression as an option: LZ4 algorithm with 4 KB fixed-output compression for high performance; From c657b5f0e2aeff3323e7e8045d2dd5182468af72 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Mon, 25 Mar 2019 11:40:09 +0800 Subject: [PATCH 59/74] staging: erofs: support IO read error injection Used to simulate disk IO read error for testing fatal error tolerance. Here are the details, 1) use bio->bi_private to indicate super_block for non-compressed bios since some (mainly meta) pages can be of the corresponding bdev inode; 2) get super_block dynamically for compressed bios, therefore it could not inject bios full of staging pages, yet it doesn't affect the normal usage. Change-Id: I82830edeac0074abf83c3758e0b692b9bd1a96ae Signed-off-by: Gao Xiang Reviewed-by: Chao Yu Signed-off-by: Greg Kroah-Hartman --- drivers/staging/erofs/Documentation/filesystems/erofs.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/staging/erofs/Documentation/filesystems/erofs.txt b/drivers/staging/erofs/Documentation/filesystems/erofs.txt index 961ec4da7705..74cf84ac48a3 100644 --- a/drivers/staging/erofs/Documentation/filesystems/erofs.txt +++ b/drivers/staging/erofs/Documentation/filesystems/erofs.txt @@ -60,6 +60,7 @@ fault_injection=%d Enable fault injection in all supported types with specified injection rate. Supported injection type: Type_Name Type_Value FAULT_KMALLOC 0x000000001 + FAULT_READ_IO 0x000000002 (no)user_xattr Setup Extended User Attributes. Note: xattr is enabled by default if CONFIG_EROFS_FS_XATTR is selected. (no)acl Setup POSIX Access Control List. Note: acl is enabled From e0450b8428ccee51b1ccc5b9fade4c683d333b85 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Wed, 31 Jul 2019 23:57:44 +0800 Subject: [PATCH 60/74] staging: erofs: kill CONFIG_EROFS_FS_USE_VM_MAP_RAM Turn into a module parameter ("use_vmap") as it can be set at runtime. Suggested-by: David Sterba Reviewed-by: Chao Yu Change-Id: Iff911fc7c0009f78b5be85b8617e262311904eda Signed-off-by: Gao Xiang Link: https://lore.kernel.org/r/20190731155752.210602-15-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman --- drivers/staging/erofs/Documentation/filesystems/erofs.txt | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/staging/erofs/Documentation/filesystems/erofs.txt b/drivers/staging/erofs/Documentation/filesystems/erofs.txt index 74cf84ac48a3..04cf47865c50 100644 --- a/drivers/staging/erofs/Documentation/filesystems/erofs.txt +++ b/drivers/staging/erofs/Documentation/filesystems/erofs.txt @@ -66,6 +66,10 @@ fault_injection=%d Enable fault injection in all supported types with (no)acl Setup POSIX Access Control List. Note: acl is enabled by default if CONFIG_EROFS_FS_POSIX_ACL is selected. +Module parameters +================= +use_vmap=[0|1] Use vmap() instead of vm_map_ram() (default 0). + On-disk details =============== From 549499161a456944d61c3f833996c772138f4d34 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Wed, 31 Jul 2019 23:57:49 +0800 Subject: [PATCH 61/74] staging: erofs: turn cache strategies into mount options Kill all kconfig cache strategies and turn them into mount options "cache_strategy={disable|readahead|readaround}". As the first step, cached pages can still be usable after cache is disabled by remounting, and these pages will be fallen out over time, which can be refined in the later version if some requirement is needed. Update related document as well. Suggested-by: Theodore Ts'o Reviewed-by: Chao Yu Change-Id: I8753a190949a73b2864e5561a7b6d059da4e0684 Signed-off-by: Gao Xiang Link: https://lore.kernel.org/r/20190731155752.210602-20-gaoxiang25@huawei.com Signed-off-by: Greg Kroah-Hartman --- .../staging/erofs/Documentation/filesystems/erofs.txt | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/staging/erofs/Documentation/filesystems/erofs.txt b/drivers/staging/erofs/Documentation/filesystems/erofs.txt index 04cf47865c50..0eab600ca7ca 100644 --- a/drivers/staging/erofs/Documentation/filesystems/erofs.txt +++ b/drivers/staging/erofs/Documentation/filesystems/erofs.txt @@ -65,6 +65,16 @@ fault_injection=%d Enable fault injection in all supported types with by default if CONFIG_EROFS_FS_XATTR is selected. (no)acl Setup POSIX Access Control List. Note: acl is enabled by default if CONFIG_EROFS_FS_POSIX_ACL is selected. +cache_strategy=%s Select a strategy for cached decompression from now on: + disabled: In-place I/O decompression only; + readahead: Cache the last incomplete compressed physical + cluster for further reading. It still does + in-place I/O decompression for the rest + compressed physical clusters; + readaround: Cache the both ends of incomplete compressed + physical clusters for further reading. + It still does in-place I/O decompression + for the rest compressed physical clusters. Module parameters ================= From c6487f63b3fb9e4cf7a0d862c8e2ece072ccee54 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Fri, 23 Aug 2019 05:36:59 +0800 Subject: [PATCH 62/74] erofs: move erofs out of staging EROFS filesystem has been merged into linux-staging for a year. EROFS is designed to be a better solution of saving extra storage space with guaranteed end-to-end performance for read-only files with the help of reduced metadata, fixed-sized output compression and decompression inplace technologies. In the past year, EROFS was greatly improved by many people as a staging driver, self-tested, betaed by a large number of our internal users, successfully applied to almost all in-service HUAWEI smartphones as the part of EMUI 9.1 and proven to be stable enough to be moved out of staging. EROFS is a self-contained filesystem driver. Although there are still some TODOs to be more generic, we have a dedicated team actively keeping on working on EROFS in order to make it better with the evolution of Linux kernel as the other in-kernel filesystems. As Pavel suggested, it's better to do as one commit since git can do moves and all histories will be saved in this way. Let's promote it from staging and enhance it more actively as a "real" part of kernel for more wider scenarios! Cc: Greg Kroah-Hartman Cc: Alexander Viro Cc: Andrew Morton Cc: Stephen Rothwell Cc: Theodore Ts'o Cc: Pavel Machek Cc: David Sterba Cc: Amir Goldstein Cc: Christoph Hellwig Cc: Darrick J . Wong Cc: Dave Chinner Cc: Jaegeuk Kim Cc: Jan Kara Cc: Richard Weinberger Cc: Linus Torvalds Cc: Chao Yu Cc: Miao Xie Cc: Li Guifu Cc: Fang Wei Change-Id: I16111e83bd97a64b13fa67354f2ee437d7b16fc5 Signed-off-by: Gao Xiang Link: https://lore.kernel.org/r/20190822213659.5501-1-hsiangkao@aol.com Signed-off-by: Greg Kroah-Hartman --- .../filesystems/erofs.txt | 4 -- MAINTAINERS | 14 +++--- fs/erofs/TODO | 45 ------------------- fs/erofs/erofs_fs.h | 1 - include/uapi/linux/magic.h | 1 + 5 files changed, 8 insertions(+), 57 deletions(-) rename {drivers/staging/erofs/Documentation => Documentation}/filesystems/erofs.txt (98%) delete mode 100644 fs/erofs/TODO diff --git a/drivers/staging/erofs/Documentation/filesystems/erofs.txt b/Documentation/filesystems/erofs.txt similarity index 98% rename from drivers/staging/erofs/Documentation/filesystems/erofs.txt rename to Documentation/filesystems/erofs.txt index 0eab600ca7ca..38aa9126ec98 100644 --- a/drivers/staging/erofs/Documentation/filesystems/erofs.txt +++ b/Documentation/filesystems/erofs.txt @@ -49,10 +49,6 @@ Bugs and patches are welcome, please kindly help us and send to the following linux-erofs mailing list: >> linux-erofs mailing list -Note that EROFS is still working in progress as a Linux staging driver, -Cc the staging mailing list as well is highly recommended: ->> Linux Driver Project Developer List - Mount options ============= diff --git a/MAINTAINERS b/MAINTAINERS index 75aec8d6f154..365dbf1cfa5e 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5488,6 +5488,13 @@ T: git git://git.kernel.org/pub/scm/linux/kernel/git/kristoffer/linux-hpc.git F: drivers/video/fbdev/s1d13xxxfb.c F: include/video/s1d13xxxfb.h +EROFS FILE SYSTEM +M: Gao Xiang +M: Chao Yu +L: linux-erofs@lists.ozlabs.org +S: Maintained +F: fs/erofs/ + ERRSEQ ERROR TRACKING INFRASTRUCTURE M: Jeff Layton S: Maintained @@ -13829,13 +13836,6 @@ M: H Hartley Sweeten S: Odd Fixes F: drivers/staging/comedi/ -STAGING - EROFS FILE SYSTEM -M: Gao Xiang -M: Chao Yu -L: linux-erofs@lists.ozlabs.org -S: Maintained -F: drivers/staging/erofs/ - STAGING - FLARION FT1000 DRIVERS M: Marek Belisko S: Odd Fixes diff --git a/fs/erofs/TODO b/fs/erofs/TODO deleted file mode 100644 index f99ddb842f99..000000000000 --- a/fs/erofs/TODO +++ /dev/null @@ -1,45 +0,0 @@ - -EROFS is still working in progress, thus it is not suitable -for all productive uses. play at your own risk :) - -TODO List: - - add the missing error handling code - (mainly existed in xattr and decompression submodules); - - - finalize erofs ondisk format design (which means that - minor on-disk revisions could happen later); - - - documentation and detailed technical analysis; - - - general code review and clean up - (including confusing variable names and code snippets); - - - support larger compressed clustersizes for selection - (currently erofs only works as expected with the page-sized - compressed cluster configuration, usually 4KB); - - - support more lossless data compression algorithms - in addition to LZ4 algorithms in VLE approach; - - - data deduplication and other useful features. - -erofs-mkfs (preview version) binaries for i386 / x86_64 are available at: - - https://github.com/hsiangkao/erofs_mkfs_binary - -It is still in progress opening mkfs source code to public, -in any case an open-source mkfs will be released in the near future. - - -Code, suggestions, etc, are welcome. Please feel free to -ask and send patches, - -To: - linux-erofs mailing list - Gao Xiang - Chao Yu - -Cc: (for linux-kernel upstream patches) - Greg Kroah-Hartman - linux-staging mailing list - diff --git a/fs/erofs/erofs_fs.h b/fs/erofs/erofs_fs.h index 32f79b14ef50..799a119af165 100644 --- a/fs/erofs/erofs_fs.h +++ b/fs/erofs/erofs_fs.h @@ -8,7 +8,6 @@ #ifndef __EROFS_FS_H #define __EROFS_FS_H -#define EROFS_SUPER_MAGIC_V1 0xE0F5E1E2 #define EROFS_SUPER_OFFSET 1024 #define EROFS_FEATURE_COMPAT_SB_CHKSUM 0x00000001 diff --git a/include/uapi/linux/magic.h b/include/uapi/linux/magic.h index 291f3f49d8ab..7b5252f1595d 100644 --- a/include/uapi/linux/magic.h +++ b/include/uapi/linux/magic.h @@ -19,6 +19,7 @@ #define SQUASHFS_MAGIC 0x73717368 #define ECRYPTFS_SUPER_MAGIC 0xf15f #define EFS_SUPER_MAGIC 0x414A53 +#define EROFS_SUPER_MAGIC_V1 0xE0F5E1E2 #define EXT2_SUPER_MAGIC 0xEF53 #define EXT3_SUPER_MAGIC 0xEF53 #define XENFS_SUPER_MAGIC 0xabba1974 From 3058b46d31850c2e7e1a135bec57c5e4317bc278 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Thu, 19 Sep 2019 14:28:38 +0800 Subject: [PATCH 63/74] MAINTAINERS: erofs: complete sub-entries for erofs Add a formal git tree and missing files for erofs after moving out of staging for everyone to blame in order for better improvement. Acked-by: Chao Yu Link: https://lore.kernel.org/r/20190919062838.106423-1-gaoxiang25@huawei.com Change-Id: I2ccd90f3ddb80d02b3ac6a3cae1efb28e15d6bf2 Signed-off-by: Gao Xiang --- MAINTAINERS | 3 +++ 1 file changed, 3 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 365dbf1cfa5e..8bcc238c84ad 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5493,7 +5493,10 @@ M: Gao Xiang M: Chao Yu L: linux-erofs@lists.ozlabs.org S: Maintained +T: git git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs.git +F: Documentation/filesystems/erofs.txt F: fs/erofs/ +F: include/trace/events/erofs.h ERRSEQ ERROR TRACKING INFRASTRUCTURE M: Jeff Layton From 9d8a8268ce54d370ef0c186a4cdbfe50cf01cccb Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Sat, 28 Mar 2020 12:00:36 +0800 Subject: [PATCH 64/74] MAINTAINERS: erofs: update my email address This email address will not be available in a few days. Update my own email address to xiang@kernel.org, which should be available all the time. Link: https://lore.kernel.org/r/20200328040036.117974-1-gaoxiang25@huawei.com Acked-by: Chao Yu Change-Id: Iec37d7160e6f7c82f0b0a3703ecd901f0b4e4b0e Signed-off-by: Gao Xiang --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 8bcc238c84ad..65dd8fc47008 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5489,7 +5489,7 @@ F: drivers/video/fbdev/s1d13xxxfb.c F: include/video/s1d13xxxfb.h EROFS FILE SYSTEM -M: Gao Xiang +M: Gao Xiang M: Chao Yu L: linux-erofs@lists.ozlabs.org S: Maintained From 4054f34e317bd4d6cfc01502fc7220644aa99c37 Mon Sep 17 00:00:00 2001 From: Hongyu Jin Date: Fri, 1 Apr 2022 19:55:27 +0800 Subject: [PATCH 65/74] BACKPORT: erofs: fix use-after-free of on-stack io[] The root cause is the race as follows: Thread #1 Thread #2(irq ctx) z_erofs_runqueue() struct z_erofs_decompressqueue io_A[]; submit bio A z_erofs_decompress_kickoff(,,1) z_erofs_decompressqueue_endio(bio A) z_erofs_decompress_kickoff(,,-1) spin_lock_irqsave() atomic_add_return() io_wait_event() -> pending_bios is already 0 [end of function] wake_up_locked(io_A[]) // crash Referenced backtrace in kernel 5.4: [ 10.129422] Unable to handle kernel paging request at virtual address eb0454a4 [ 10.364157] CPU: 0 PID: 709 Comm: getprop Tainted: G WC O 5.4.147-ab09225 #1 [ 11.556325] [] (__wake_up_common) from [] (__wake_up_locked+0x40/0x48) [ 11.565487] [] (__wake_up_locked) from [] (z_erofs_vle_unzip_kickoff+0x6c/0xc0) [ 11.575438] [] (z_erofs_vle_unzip_kickoff) from [] (z_erofs_vle_read_endio+0x16c/0x17c) [ 11.586082] [] (z_erofs_vle_read_endio) from [] (clone_endio+0xb4/0x1d0) [ 11.595428] [] (clone_endio) from [] (blk_update_request+0x150/0x4dc) [ 11.604516] [] (blk_update_request) from [] (mmc_blk_cqe_complete_rq+0x144/0x15c) [ 11.614640] [] (mmc_blk_cqe_complete_rq) from [] (blk_done_softirq+0xb0/0xcc) [ 11.624419] [] (blk_done_softirq) from [] (__do_softirq+0x184/0x56c) [ 11.633419] [] (__do_softirq) from [] (irq_exit+0xd4/0x138) [ 11.641640] [] (irq_exit) from [] (__handle_domain_irq+0x94/0xd0) [ 11.650381] [] (__handle_domain_irq) from [] (gic_handle_irq+0x50/0xd4) [ 11.659641] [] (gic_handle_irq) from [] (__irq_svc+0x70/0xb0) Bug: 246657836 Change-Id: Ieebf1c5abb48723538d05a5e65b5179a382dab3f (cherry picked from commit 60b30050116c0351b90154044345c1b53ae1f323) [Hongyu: Resolved minor conflict in fs/erofs/zdata.c ] Signed-off-by: Hongyu Jin Reviewed-by: Gao Xiang Reviewed-by: Chao Yu Link: https://lore.kernel.org/r/20220401115527.4935-1-hongyu.jin.cn@gmail.com Signed-off-by: Gao Xiang --- fs/erofs/zdata.c | 12 ++++-------- fs/erofs/zdata.h | 2 +- 2 files changed, 5 insertions(+), 9 deletions(-) diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 35a08f3a53d0..e1c0cb1b2477 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -786,12 +786,9 @@ static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io, /* wake up the caller thread for sync decompression */ if (sync) { - unsigned long flags; - - spin_lock_irqsave(&io->u.wait.lock, flags); if (!atomic_add_return(bios, &io->pending_bios)) - wake_up_locked(&io->u.wait); - spin_unlock_irqrestore(&io->u.wait.lock, flags); + complete(&io->u.done); + return; } @@ -1210,7 +1207,7 @@ jobqueue_init(struct super_block *sb, } else { fg_out: q = fgq; - init_waitqueue_head(&fgq->u.wait); + init_completion(&fgq->u.done); atomic_set(&fgq->pending_bios, 0); } q->sb = sb; @@ -1373,8 +1370,7 @@ static void z_erofs_runqueue(struct super_block *sb, return; /* wait until all bios are completed */ - io_wait_event(io[JQ_SUBMIT].u.wait, - !atomic_read(&io[JQ_SUBMIT].pending_bios)); + wait_for_completion_io(&io[JQ_SUBMIT].u.done); /* handle synchronous decompress queue in the caller context */ z_erofs_decompress_queue(&io[JQ_SUBMIT], pagepool); diff --git a/fs/erofs/zdata.h b/fs/erofs/zdata.h index 3a008f1b9f78..1027ffe744a2 100644 --- a/fs/erofs/zdata.h +++ b/fs/erofs/zdata.h @@ -89,7 +89,7 @@ struct z_erofs_decompressqueue { z_erofs_next_pcluster_t head; union { - wait_queue_head_t wait; + struct completion done; struct work_struct work; } u; }; From 1d696af89e906432831554e452cae231d3f18517 Mon Sep 17 00:00:00 2001 From: Guoqing Jiang Date: Mon, 1 Jun 2020 21:47:38 -0700 Subject: [PATCH 66/74] include/linux/pagemap.h: introduce attach/detach_page_private MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Patch series "Introduce attach/detach_page_private to cleanup code". This patch (of 10): The logic in attach_page_buffers and __clear_page_buffers are quite paired, but 1. they are located in different files. 2. attach_page_buffers is implemented in buffer_head.h, so it could be used by other files. But __clear_page_buffers is static function in buffer.c and other potential users can't call the function, md-bitmap even copied the function. So, introduce the new attach/detach_page_private to replace them. With the new pair of function, we will remove the usage of attach_page_buffers and __clear_page_buffers in next patches. Thanks for suggestions about the function name from Alexander Viro, Andreas Grünbacher, Christoph Hellwig and Matthew Wilcox. Suggested-by: Matthew Wilcox Change-Id: I46d97d65243c78cb22ad26da092d9a386ca0e6c0 Signed-off-by: Guoqing Jiang Signed-off-by: Andrew Morton Reviewed-by: Andrew Morton Cc: "Darrick J. Wong" Cc: William Kucharski Cc: "Kirill A. Shutemov" Cc: Andreas Gruenbacher Cc: Yang Shi Cc: Yafang Shao Cc: Song Liu Cc: Chris Mason Cc: Josef Bacik Cc: David Sterba Cc: Alexander Viro Cc: Jaegeuk Kim Cc: Chao Yu Cc: Christoph Hellwig Cc: Anton Altaparmakov Cc: Mike Marshall Cc: Martin Brandenburg Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Roman Gushchin Cc: Andreas Dilger Cc: Chao Yu Cc: Dave Chinner Link: http://lkml.kernel.org/r/20200517214718.468-1-guoqing.jiang@cloud.ionos.com Link: http://lkml.kernel.org/r/20200517214718.468-2-guoqing.jiang@cloud.ionos.com Signed-off-by: Linus Torvalds --- include/linux/pagemap.h | 37 +++++++++++++++++++++++++++++++++++++ 1 file changed, 37 insertions(+) diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index d9217645791a..659a40a9cabd 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -220,6 +220,43 @@ static inline int page_cache_add_speculative(struct page *page, int count) return 1; } +/** + * attach_page_private - Attach private data to a page. + * @page: Page to attach data to. + * @data: Data to attach to page. + * + * Attaching private data to a page increments the page's reference count. + * The data must be detached before the page will be freed. + */ +static inline void attach_page_private(struct page *page, void *data) +{ + get_page(page); + set_page_private(page, (unsigned long)data); + SetPagePrivate(page); +} + +/** + * detach_page_private - Detach private data from a page. + * @page: Page to detach data from. + * + * Removes the data that was previously attached to the page and decrements + * the refcount on the page. + * + * Return: Data that was attached to the page. + */ +static inline void *detach_page_private(struct page *page) +{ + void *data = (void *)page_private(page); + + if (!PagePrivate(page)) + return NULL; + ClearPagePrivate(page); + set_page_private(page, 0); + put_page(page); + + return data; +} + #ifdef CONFIG_NUMA extern struct page *__page_cache_alloc(gfp_t gfp); #else From 959804efaf77b38e520b2d3642e44eb7798e9eaf Mon Sep 17 00:00:00 2001 From: Michael Bestas Date: Tue, 16 Apr 2024 02:45:16 +0300 Subject: [PATCH 67/74] erofs: Switch to attach/detach_page_private Matches upstream behaviour Change-Id: Ia85c5472aef465ec5beffc494bf4a4c6edc0bd16 --- fs/erofs/zdata.c | 17 ++++++----------- 1 file changed, 6 insertions(+), 11 deletions(-) diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index e1c0cb1b2477..5b91602e751d 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -331,11 +331,8 @@ int erofs_try_to_free_all_cached_pages(struct erofs_sb_info *sbi, /* barrier is implied in the following 'unlock_page' */ WRITE_ONCE(pcl->compressed_pages[i], NULL); - - set_page_private(page, 0); - ClearPagePrivate(page); + detach_page_private(page); unlock_page(page); - put_page(page); } return 0; } @@ -358,11 +355,8 @@ int erofs_try_to_free_cached_page(struct address_space *mapping, } erofs_workgroup_unfreeze(&pcl->obj, 1); - if (ret) { - set_page_private(page, 0); - ClearPagePrivate(page); - put_page(page); - } + if (ret) + detach_page_private(page); } return ret; } @@ -1184,8 +1178,9 @@ out_tocache: set_page_private(page, Z_EROFS_SHORTLIVED_PAGE); goto out; } - set_page_private(page, (unsigned long)pcl); - SetPagePrivate(page); + attach_page_private(page, pcl); + /* drop a refcount added by allocpage (then we have 2 refs here) */ + put_page(page); out: /* the only exit (for tracing and debugging) */ return page; From 3aad597d4f55d0170eb3d1ed4de40855f7751561 Mon Sep 17 00:00:00 2001 From: Michael Bestas Date: Tue, 16 Apr 2024 03:25:04 +0300 Subject: [PATCH 68/74] Revert "erofs: add REQ_RAHEAD flag to readahead requests" Reason for revert: Not supported on <=5.4 This reverts commit a6b8e4a11a99cc58edd622016f47a548254da5b2. Change-Id: I5b4b5f00fe750980d8c92c336e333142a72788ce --- fs/erofs/data.c | 2 +- fs/erofs/zdata.c | 16 ++++++---------- 2 files changed, 7 insertions(+), 11 deletions(-) diff --git a/fs/erofs/data.c b/fs/erofs/data.c index dc264a819dc2..06a3d672ab46 100644 --- a/fs/erofs/data.c +++ b/fs/erofs/data.c @@ -208,7 +208,7 @@ submit_bio_retry: bio_set_dev(bio, sb->s_bdev); bio->bi_iter.bi_sector = (sector_t)blknr << LOG_SECTORS_PER_BLOCK; - bio->bi_opf = REQ_OP_READ | (ra ? REQ_RAHEAD : 0); + bio->bi_opf = REQ_OP_READ; } err = bio_add_page(bio, page, PAGE_SIZE, 0); diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 5b91602e751d..c5e28d2ca9db 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -216,7 +216,6 @@ struct z_erofs_decompress_frontend { struct z_erofs_collector clt; struct erofs_map_blocks map; - bool readahead; /* used for applying cache strategy on the fly */ bool backmost; erofs_off_t headoffset; @@ -1251,7 +1250,7 @@ static void move_to_bypass_jobqueue(struct z_erofs_pcluster *pcl, } static void z_erofs_submit_queue(struct super_block *sb, - struct z_erofs_decompress_frontend *f, + z_erofs_next_pcluster_t owned_head, struct list_head *pagepool, struct z_erofs_decompressqueue *fgq, bool *force_fg) @@ -1260,7 +1259,6 @@ static void z_erofs_submit_queue(struct super_block *sb, z_erofs_next_pcluster_t qtail[NR_JOBQUEUES]; struct z_erofs_decompressqueue *q[NR_JOBQUEUES]; void *bi_private; - z_erofs_next_pcluster_t owned_head = f->clt.owned_head; /* since bio will be NULL, no need to initialize last_index */ pgoff_t last_index; unsigned int nr_bios = 0; @@ -1316,8 +1314,6 @@ submit_bio_retry: LOG_SECTORS_PER_BLOCK; bio->bi_private = bi_private; bio->bi_opf = REQ_OP_READ; - if (f->readahead) - bio->bi_opf |= REQ_RAHEAD; ++nr_bios; } @@ -1349,14 +1345,14 @@ submit_bio_retry: } static void z_erofs_runqueue(struct super_block *sb, - struct z_erofs_decompress_frontend *f, + struct z_erofs_collector *clt, struct list_head *pagepool, bool force_fg) { struct z_erofs_decompressqueue io[NR_JOBQUEUES]; - if (f->clt.owned_head == Z_EROFS_PCLUSTER_TAIL) + if (clt->owned_head == Z_EROFS_PCLUSTER_TAIL) return; - z_erofs_submit_queue(sb, f, pagepool, io, &force_fg); + z_erofs_submit_queue(sb, clt->owned_head, pagepool, io, &force_fg); /* handle bypass queue (no i/o pclusters) immediately */ z_erofs_decompress_queue(&io[JQ_BYPASS], pagepool); @@ -1386,7 +1382,7 @@ static int z_erofs_readpage(struct file *file, struct page *page) (void)z_erofs_collector_end(&f.clt); /* if some compressed cluster ready, need submit them anyway */ - z_erofs_runqueue(inode->i_sb, &f, &pagepool, true); + z_erofs_runqueue(inode->i_sb, &f.clt, &pagepool, true); if (err) erofs_err(inode->i_sb, "failed to read, err [%d]", err); @@ -1456,7 +1452,7 @@ static int z_erofs_readpages(struct file *filp, struct address_space *mapping, (void)z_erofs_collector_end(&f.clt); - z_erofs_runqueue(inode->i_sb, &f, &pagepool, sync); + z_erofs_runqueue(inode->i_sb, &f.clt, &pagepool, sync); if (f.map.mpage) put_page(f.map.mpage); From a5af5361108e42bbdda5dc33d95922585f3790de Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 9 Jun 2020 11:05:54 +0200 Subject: [PATCH 69/74] ANDROID: GKI: enable CONFIG_EROFS_FS=y Some vendors are using erofs, so instead of having to support this as a module, build it into the kernel for everyone to use. Bug: 157965270 Signed-off-by: Greg Kroah-Hartman Change-Id: I8f8e953c13991dea50031fb4ff01ee99fbf09b88 --- arch/arm64/configs/gki_defconfig | 1 + arch/x86/configs/gki_defconfig | 1 + 2 files changed, 2 insertions(+) diff --git a/arch/arm64/configs/gki_defconfig b/arch/arm64/configs/gki_defconfig index b094b4f26a2d..8580473accad 100644 --- a/arch/arm64/configs/gki_defconfig +++ b/arch/arm64/configs/gki_defconfig @@ -438,6 +438,7 @@ CONFIG_TMPFS_POSIX_ACL=y CONFIG_PSTORE=y CONFIG_PSTORE_CONSOLE=y CONFIG_PSTORE_RAM=y +CONFIG_EROFS_FS=y CONFIG_NLS_CODEPAGE_437=y CONFIG_NLS_CODEPAGE_737=y CONFIG_NLS_CODEPAGE_775=y diff --git a/arch/x86/configs/gki_defconfig b/arch/x86/configs/gki_defconfig index 95288e35d135..4d6c836cd701 100644 --- a/arch/x86/configs/gki_defconfig +++ b/arch/x86/configs/gki_defconfig @@ -373,6 +373,7 @@ CONFIG_TMPFS_POSIX_ACL=y CONFIG_PSTORE=y CONFIG_PSTORE_CONSOLE=y CONFIG_PSTORE_RAM=y +CONFIG_EROFS_FS=y CONFIG_NLS_CODEPAGE_437=y CONFIG_NLS_CODEPAGE_737=y CONFIG_NLS_CODEPAGE_775=y From 260d03bf36763b7b15df8099534255ab500fb2a8 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 21 Apr 2020 12:09:13 +0200 Subject: [PATCH 70/74] sched: Provide sched_set_fifo() SCHED_FIFO (or any static priority scheduler) is a broken scheduler model; it is fundamentally incapable of resource management, the one thing an OS is actually supposed to do. It is impossible to compose static priority workloads. One cannot take two well designed and functional static priority workloads and mash them together and still expect them to work. Therefore it doesn't make sense to expose the priority field; the kernel is fundamentally incapable of setting a sensible value, it needs systems knowledge that it doesn't have. Take away sched_setschedule() / sched_setattr() from modules and replace them with: - sched_set_fifo(p); create a FIFO task (at prio 50) - sched_set_fifo_low(p); create a task higher than NORMAL, which ends up being a FIFO task at prio 1. - sched_set_normal(p, nice); (re)set the task to normal This stops the proliferation of randomly chosen, and irrelevant, FIFO priorities that dont't really mean anything anyway. The system administrator/integrator, whoever has insight into the actual system design and requirements (userspace) can set-up appropriate priorities if and when needed. Cc: airlied@redhat.com Cc: alexander.deucher@amd.com Cc: awalls@md.metrocast.net Cc: axboe@kernel.dk Cc: broonie@kernel.org Cc: daniel.lezcano@linaro.org Cc: gregkh@linuxfoundation.org Cc: hannes@cmpxchg.org Cc: herbert@gondor.apana.org.au Cc: hverkuil@xs4all.nl Cc: john.stultz@linaro.org Cc: nico@fluxnic.net Cc: paulmck@kernel.org Cc: rafael.j.wysocki@intel.com Cc: rmk+kernel@arm.linux.org.uk Cc: sudeep.holla@arm.com Cc: tglx@linutronix.de Cc: ulf.hansson@linaro.org Cc: wim@linux-watchdog.org Change-Id: I52ed4f1253e82ba3e8f40f3aa1aff62580163f25 Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Ingo Molnar Tested-by: Paul E. McKenney --- include/linux/sched.h | 3 +++ kernel/sched/core.c | 47 +++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 50 insertions(+) diff --git a/include/linux/sched.h b/include/linux/sched.h index 89bd6a449033..590e9d0a57c9 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1859,6 +1859,9 @@ extern int idle_cpu(int cpu); extern int available_idle_cpu(int cpu); extern int sched_setscheduler(struct task_struct *, int, const struct sched_param *); extern int sched_setscheduler_nocheck(struct task_struct *, int, const struct sched_param *); +extern int sched_set_fifo(struct task_struct *p); +extern int sched_set_fifo_low(struct task_struct *p); +extern int sched_set_normal(struct task_struct *p, int nice); extern int sched_setattr(struct task_struct *, const struct sched_attr *); extern int sched_setattr_nocheck(struct task_struct *, const struct sched_attr *); extern struct task_struct *idle_task(int cpu); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 45270c04437f..de5e1adec8fe 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -5310,6 +5310,8 @@ static int _sched_setscheduler(struct task_struct *p, int policy, * @policy: new policy. * @param: structure containing the new RT priority. * + * Use sched_set_fifo(), read its comment. + * * Return: 0 on success. An error code otherwise. * * NOTE that the task may be already dead. @@ -5352,6 +5354,51 @@ int sched_setscheduler_nocheck(struct task_struct *p, int policy, } EXPORT_SYMBOL_GPL(sched_setscheduler_nocheck); +/* + * SCHED_FIFO is a broken scheduler model; that is, it is fundamentally + * incapable of resource management, which is the one thing an OS really should + * be doing. + * + * This is of course the reason it is limited to privileged users only. + * + * Worse still; it is fundamentally impossible to compose static priority + * workloads. You cannot take two correctly working static prio workloads + * and smash them together and still expect them to work. + * + * For this reason 'all' FIFO tasks the kernel creates are basically at: + * + * MAX_RT_PRIO / 2 + * + * The administrator _MUST_ configure the system, the kernel simply doesn't + * know enough information to make a sensible choice. + */ +int sched_set_fifo(struct task_struct *p) +{ + struct sched_param sp = { .sched_priority = MAX_RT_PRIO / 2 }; + return sched_setscheduler_nocheck(p, SCHED_FIFO, &sp); +} +EXPORT_SYMBOL_GPL(sched_set_fifo); + +/* + * For when you don't much care about FIFO, but want to be above SCHED_NORMAL. + */ +int sched_set_fifo_low(struct task_struct *p) +{ + struct sched_param sp = { .sched_priority = 1 }; + return sched_setscheduler_nocheck(p, SCHED_FIFO, &sp); +} +EXPORT_SYMBOL_GPL(sched_set_fifo_low); + +int sched_set_normal(struct task_struct *p, int nice) +{ + struct sched_attr attr = { + .sched_policy = SCHED_NORMAL, + .sched_nice = nice, + }; + return sched_setattr_nocheck(p, &attr); +} +EXPORT_SYMBOL_GPL(sched_set_normal); + static int do_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) { From 5b68ef13a8f73091ce8796957352c0aae8ff07e8 Mon Sep 17 00:00:00 2001 From: Sandeep Dhavale Date: Wed, 8 Feb 2023 17:33:22 +0800 Subject: [PATCH 71/74] BACKPORT: erofs: add per-cpu threads for decompression as an option Using per-cpu thread pool we can reduce the scheduling latency compared to workqueue implementation. With this patch scheduling latency and variation is reduced as per-cpu threads are high priority kthread_workers. The results were evaluated on arm64 Android devices running 5.10 kernel. The table below shows resulting improvements of total scheduling latency for the same app launch benchmark runs with 50 iterations. Scheduling latency is the latency between when the task (workqueue kworker vs kthread_worker) became eligible to run to when it actually started running. +-------------------------+-----------+----------------+---------+ | | workqueue | kthread_worker | diff | +-------------------------+-----------+----------------+---------+ | Average (us) | 15253 | 2914 | -80.89% | | Median (us) | 14001 | 2912 | -79.20% | | Minimum (us) | 3117 | 1027 | -67.05% | | Maximum (us) | 30170 | 3805 | -87.39% | | Standard deviation (us) | 7166 | 359 | | +-------------------------+-----------+----------------+---------+ Background: Boot times and cold app launch benchmarks are very important to the Android ecosystem as they directly translate to responsiveness from user point of view. While EROFS provides a lot of important features like space savings, we saw some performance penalty in cold app launch benchmarks in few scenarios. Analysis showed that the significant variance was coming from the scheduling cost while decompression cost was more or less the same. Having per-cpu thread pool we can see from the above table that this variation is reduced by ~80% on average. This problem was discussed at LPC 2022. Link to LPC 2022 slides and talk at [1] [1] https://lpc.events/event/16/contributions/1338/ [ Gao Xiang: At least, we have to add this until WQ_UNBOUND workqueue issue [2] on many arm64 devices is resolved. ] [2] https://lore.kernel.org/r/CAJkfWY490-m6wNubkxiTPsW59sfsQs37Wey279LmiRxKt7aQYg@mail.gmail.com Bug: 271636421 Bug: 278520205 Test: launch_cvd Change-Id: I9dce2bfd6f40ec6a210161b80cee7c0417b4edb3 Signed-off-by: Sandeep Dhavale Signed-off-by: Gao Xiang Link: https://lore.kernel.org/r/20230208093322.75816-1-hsiangkao@linux.alibaba.com (cherry picked from commit 3fffb589b9a6e331e39cb75373ee7691acd7b109) [dhavale: Fixed minor conflict as upstream now has zdata.h folded in zdata.c] Signed-off-by: Sandeep Dhavale (cherry picked from commit 566a7f6c6b3f5f13b766fe749bbdb45918b029ac) [dhavale: Fixed minor conflicts in Kconfig and zdata.c] (cherry picked from commit 2de95f5d183c2174c9380a902919c8e59e380293) --- fs/erofs/Kconfig | 17 +++++ fs/erofs/zdata.c | 185 ++++++++++++++++++++++++++++++++++++++++++----- fs/erofs/zdata.h | 2 + 3 files changed, 186 insertions(+), 18 deletions(-) diff --git a/fs/erofs/Kconfig b/fs/erofs/Kconfig index 858b3339f381..662660852397 100644 --- a/fs/erofs/Kconfig +++ b/fs/erofs/Kconfig @@ -76,3 +76,20 @@ config EROFS_FS_ZIP If you don't want to enable compression feature, say N. +config EROFS_FS_PCPU_KTHREAD + bool "EROFS per-cpu decompression kthread workers" + depends on EROFS_FS_ZIP + help + Saying Y here enables per-CPU kthread workers pool to carry out + async decompression for low latencies on some architectures. + + If unsure, say N. + +config EROFS_FS_PCPU_KTHREAD_HIPRI + bool "EROFS high priority per-CPU kthread workers" + depends on EROFS_FS_ZIP && EROFS_FS_PCPU_KTHREAD + help + This permits EROFS to configure per-CPU kthread workers to run + at higher priority. + + If unsure, say N. diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index c5e28d2ca9db..19045a574783 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -6,7 +6,7 @@ #include "zdata.h" #include "compress.h" #include - +#include #include /* @@ -124,24 +124,128 @@ typedef tagptr1_t compressed_page_t; static struct workqueue_struct *z_erofs_workqueue __read_mostly; -void z_erofs_exit_zip_subsystem(void) +#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD +static struct kthread_worker __rcu **z_erofs_pcpu_workers; + +static void erofs_destroy_percpu_workers(void) { - destroy_workqueue(z_erofs_workqueue); - z_erofs_destroy_pcluster_pool(); + struct kthread_worker *worker; + unsigned int cpu; + + for_each_possible_cpu(cpu) { + worker = rcu_dereference_protected( + z_erofs_pcpu_workers[cpu], 1); + rcu_assign_pointer(z_erofs_pcpu_workers[cpu], NULL); + if (worker) + kthread_destroy_worker(worker); + } + kfree(z_erofs_pcpu_workers); } -static inline int z_erofs_init_workqueue(void) +static struct kthread_worker *erofs_init_percpu_worker(int cpu) { - const unsigned int onlinecpus = num_possible_cpus(); + struct kthread_worker *worker = + kthread_create_worker_on_cpu(cpu, 0, "erofs_worker/%u", cpu); - /* - * no need to spawn too many threads, limiting threads could minimum - * scheduling overhead, perhaps per-CPU threads should be better? - */ - z_erofs_workqueue = alloc_workqueue("erofs_unzipd", - WQ_UNBOUND | WQ_HIGHPRI, - onlinecpus + onlinecpus / 4); - return z_erofs_workqueue ? 0 : -ENOMEM; + if (IS_ERR(worker)) + return worker; + if (IS_ENABLED(CONFIG_EROFS_FS_PCPU_KTHREAD_HIPRI)) + sched_set_fifo_low(worker->task); + else + sched_set_normal(worker->task, 0); + return worker; +} + +static int erofs_init_percpu_workers(void) +{ + struct kthread_worker *worker; + unsigned int cpu; + + z_erofs_pcpu_workers = kcalloc(num_possible_cpus(), + sizeof(struct kthread_worker *), GFP_ATOMIC); + if (!z_erofs_pcpu_workers) + return -ENOMEM; + + for_each_online_cpu(cpu) { /* could miss cpu{off,on}line? */ + worker = erofs_init_percpu_worker(cpu); + if (!IS_ERR(worker)) + rcu_assign_pointer(z_erofs_pcpu_workers[cpu], worker); + } + return 0; +} +#else +static inline void erofs_destroy_percpu_workers(void) {} +static inline int erofs_init_percpu_workers(void) { return 0; } +#endif + +#if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_EROFS_FS_PCPU_KTHREAD) +static DEFINE_SPINLOCK(z_erofs_pcpu_worker_lock); +static enum cpuhp_state erofs_cpuhp_state; + +static int erofs_cpu_online(unsigned int cpu) +{ + struct kthread_worker *worker, *old; + + worker = erofs_init_percpu_worker(cpu); + if (IS_ERR(worker)) + return PTR_ERR(worker); + + spin_lock(&z_erofs_pcpu_worker_lock); + old = rcu_dereference_protected(z_erofs_pcpu_workers[cpu], + lockdep_is_held(&z_erofs_pcpu_worker_lock)); + if (!old) + rcu_assign_pointer(z_erofs_pcpu_workers[cpu], worker); + spin_unlock(&z_erofs_pcpu_worker_lock); + if (old) + kthread_destroy_worker(worker); + return 0; +} + +static int erofs_cpu_offline(unsigned int cpu) +{ + struct kthread_worker *worker; + + spin_lock(&z_erofs_pcpu_worker_lock); + worker = rcu_dereference_protected(z_erofs_pcpu_workers[cpu], + lockdep_is_held(&z_erofs_pcpu_worker_lock)); + rcu_assign_pointer(z_erofs_pcpu_workers[cpu], NULL); + spin_unlock(&z_erofs_pcpu_worker_lock); + + synchronize_rcu(); + if (worker) + kthread_destroy_worker(worker); + return 0; +} + +static int erofs_cpu_hotplug_init(void) +{ + int state; + + state = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN, + "fs/erofs:online", erofs_cpu_online, erofs_cpu_offline); + if (state < 0) + return state; + + erofs_cpuhp_state = state; + return 0; +} + +static void erofs_cpu_hotplug_destroy(void) +{ + if (erofs_cpuhp_state) + cpuhp_remove_state_nocalls(erofs_cpuhp_state); +} +#else /* !CONFIG_HOTPLUG_CPU || !CONFIG_EROFS_FS_PCPU_KTHREAD */ +static inline int erofs_cpu_hotplug_init(void) { return 0; } +static inline void erofs_cpu_hotplug_destroy(void) {} +#endif + +void z_erofs_exit_zip_subsystem(void) +{ + erofs_cpu_hotplug_destroy(); + erofs_destroy_percpu_workers(); + destroy_workqueue(z_erofs_workqueue); + z_erofs_destroy_pcluster_pool(); } int __init z_erofs_init_zip_subsystem(void) @@ -149,10 +253,29 @@ int __init z_erofs_init_zip_subsystem(void) int err = z_erofs_create_pcluster_pool(); if (err) - return err; - err = z_erofs_init_workqueue(); + goto out_error_pcluster_pool; + + z_erofs_workqueue = alloc_workqueue("erofs_worker", + WQ_UNBOUND | WQ_HIGHPRI, num_possible_cpus()); + if (!z_erofs_workqueue) + goto out_error_workqueue_init; + + err = erofs_init_percpu_workers(); if (err) - z_erofs_destroy_pcluster_pool(); + goto out_error_pcpu_worker; + + err = erofs_cpu_hotplug_init(); + if (err < 0) + goto out_error_cpuhp_init; + return err; + +out_error_cpuhp_init: + erofs_destroy_percpu_workers(); +out_error_pcpu_worker: + destroy_workqueue(z_erofs_workqueue); +out_error_workqueue_init: + z_erofs_destroy_pcluster_pool(); +out_error_pcluster_pool: return err; } @@ -772,6 +895,12 @@ err_out: } static void z_erofs_decompressqueue_work(struct work_struct *work); +#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD +static void z_erofs_decompressqueue_kthread_work(struct kthread_work *work) +{ + z_erofs_decompressqueue_work((struct work_struct *)work); +} +#endif static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io, bool sync, int bios) { @@ -789,7 +918,22 @@ static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io, return; /* Use workqueue and sync decompression for atomic contexts only */ if (in_atomic() || irqs_disabled()) { +#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD + struct kthread_worker *worker; + + rcu_read_lock(); + worker = rcu_dereference( + z_erofs_pcpu_workers[raw_smp_processor_id()]); + if (!worker) { + INIT_WORK(&io->u.work, z_erofs_decompressqueue_work); + queue_work(z_erofs_workqueue, &io->u.work); + } else { + kthread_queue_work(worker, &io->u.kthread_work); + } + rcu_read_unlock(); +#else queue_work(z_erofs_workqueue, &io->u.work); +#endif sbi->readahead_sync_decompress = true; return; } @@ -1197,7 +1341,12 @@ jobqueue_init(struct super_block *sb, *fg = true; goto fg_out; } +#ifdef CONFIG_EROFS_FS_PCPU_KTHREAD + kthread_init_work(&q->u.kthread_work, + z_erofs_decompressqueue_kthread_work); +#else INIT_WORK(&q->u.work, z_erofs_decompressqueue_work); +#endif } else { fg_out: q = fgq; @@ -1335,7 +1484,7 @@ submit_bio_retry: /* * although background is preferred, no one is pending for submission. - * don't issue workqueue for decompression but drop it directly instead. + * don't issue decompression but drop it directly instead. */ if (!*force_fg && !nr_bios) { kvfree(q[JQ_SUBMIT]); diff --git a/fs/erofs/zdata.h b/fs/erofs/zdata.h index 1027ffe744a2..3ada1ae86b38 100644 --- a/fs/erofs/zdata.h +++ b/fs/erofs/zdata.h @@ -6,6 +6,7 @@ #ifndef __EROFS_FS_ZDATA_H #define __EROFS_FS_ZDATA_H +#include #include "internal.h" #include "zpvec.h" @@ -91,6 +92,7 @@ struct z_erofs_decompressqueue { union { struct completion done; struct work_struct work; + struct kthread_work kthread_work; } u; }; From 8fdcc15c1447a8851593c22ebf6c04ab62b3841f Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 16 Feb 2023 15:13:04 +0300 Subject: [PATCH 72/74] UPSTREAM: erofs: fix an error code in z_erofs_init_zip_subsystem() Return -ENOMEM if alloc_workqueue() fails. Don't return success. Fixes: d8a650adf429 ("erofs: add per-cpu threads for decompression as an option") Bug: 271636421 Bug: 278520205 Tests: launch_cvd Change-Id: I49dcf8094655db47ac71ae8967b5402373fc7adc Signed-off-by: Dan Carpenter Reviewed-by: Gao Xiang Link: https://lore.kernel.org/r/Y+4d0FRsUq8jPoOu@kili Signed-off-by: Gao Xiang (cherry picked from commit 8d1b80a79452630f157bf634ae9cfcd9f4eed161) Signed-off-by: Sandeep Dhavale (cherry picked from commit 5004cc655730355d681b333ff5eafd62a4ee02f2) (cherry picked from commit b7f5039f59d8bee886f584c65ca30f14be4b642c) --- fs/erofs/zdata.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 19045a574783..c2f566b06e78 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -257,8 +257,10 @@ int __init z_erofs_init_zip_subsystem(void) z_erofs_workqueue = alloc_workqueue("erofs_worker", WQ_UNBOUND | WQ_HIGHPRI, num_possible_cpus()); - if (!z_erofs_workqueue) + if (!z_erofs_workqueue) { + err = -ENOMEM; goto out_error_workqueue_init; + } err = erofs_init_percpu_workers(); if (err) From d9887724e80d1e60546d7548b81b5962cf298c5f Mon Sep 17 00:00:00 2001 From: Sandeep Dhavale Date: Wed, 1 Mar 2023 19:03:41 +0000 Subject: [PATCH 73/74] ANDROID: Enable percpu high priority kthreads for erofs This change enables 2 configs CONFIG_EROFS_FS_PCPU_KTHREAD CONFIG_EROFS_FS_PCPU_KTHREAD_HIPRI With these changes scheduling latency is reduced by ~80% Bug: 271636421 Bug: 278520205 Tests: launch_cvd Change-Id: I91076edd3c7dcbb3497ca0accf98af28c43f4ffb Signed-off-by: Sandeep Dhavale (cherry picked from commit f8e756a6f7019d6c367844ecc449e052c2173f93) (cherry picked from commit ff1fcec189e004540d2abc41bc20102b126e4d7f) --- arch/arm64/configs/gki_defconfig | 2 ++ arch/x86/configs/gki_defconfig | 2 ++ 2 files changed, 4 insertions(+) diff --git a/arch/arm64/configs/gki_defconfig b/arch/arm64/configs/gki_defconfig index 8580473accad..d43010fc8dea 100644 --- a/arch/arm64/configs/gki_defconfig +++ b/arch/arm64/configs/gki_defconfig @@ -439,6 +439,8 @@ CONFIG_PSTORE=y CONFIG_PSTORE_CONSOLE=y CONFIG_PSTORE_RAM=y CONFIG_EROFS_FS=y +CONFIG_EROFS_FS_PCPU_KTHREAD=y +CONFIG_EROFS_FS_PCPU_KTHREAD_HIPRI=y CONFIG_NLS_CODEPAGE_437=y CONFIG_NLS_CODEPAGE_737=y CONFIG_NLS_CODEPAGE_775=y diff --git a/arch/x86/configs/gki_defconfig b/arch/x86/configs/gki_defconfig index 4d6c836cd701..32390caf4b8e 100644 --- a/arch/x86/configs/gki_defconfig +++ b/arch/x86/configs/gki_defconfig @@ -374,6 +374,8 @@ CONFIG_PSTORE=y CONFIG_PSTORE_CONSOLE=y CONFIG_PSTORE_RAM=y CONFIG_EROFS_FS=y +CONFIG_EROFS_FS_PCPU_KTHREAD=y +CONFIG_EROFS_FS_PCPU_KTHREAD_HIPRI=y CONFIG_NLS_CODEPAGE_437=y CONFIG_NLS_CODEPAGE_737=y CONFIG_NLS_CODEPAGE_775=y From a961e47da245aa82d2a85d8a8afc3f7d8d533630 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Wed, 6 Dec 2023 12:55:34 +0800 Subject: [PATCH 74/74] BACKPORT: erofs: fix lz4 inplace decompression commit 3c12466b6b7bf1e56f9b32c366a3d83d87afb4de upstream. Currently EROFS can map another compressed buffer for inplace decompression, that was used to handle the cases that some pages of compressed data are actually not in-place I/O. However, like most simple LZ77 algorithms, LZ4 expects the compressed data is arranged at the end of the decompressed buffer and it explicitly uses memmove() to handle overlapping: __________________________________________________________ |_ direction of decompression --> ____ |_ compressed data _| Although EROFS arranges compressed data like this, it typically maps two individual virtual buffers so the relative order is uncertain. Previously, it was hardly observed since LZ4 only uses memmove() for short overlapped literals and x86/arm64 memmove implementations seem to completely cover it up and they don't have this issue. Juhyung reported that EROFS data corruption can be found on a new Intel x86 processor. After some analysis, it seems that recent x86 processors with the new FSRM feature expose this issue with "rep movsb". Let's strictly use the decompressed buffer for lz4 inplace decompression for now. Later, as an useful improvement, we could try to tie up these two buffers together in the correct order. Reported-and-tested-by: Juhyung Park Closes: https://lore.kernel.org/r/CAD14+f2AVKf8Fa2OO1aAUdDNTDsVzzR6ctU_oJSmTyd6zSYR2Q@mail.gmail.com Fixes: 0ffd71bcc3a0 ("staging: erofs: introduce LZ4 decompression inplace") Fixes: 598162d05080 ("erofs: support decompress big pcluster for lz4 backend") Cc: stable # 5.4+ Tested-by: Yifan Zhao Change-Id: Ib7a578283e33f0329ae2133223878ddf0738aba4 Signed-off-by: Gao Xiang Link: https://lore.kernel.org/r/20231206045534.3920847-1-hsiangkao@linux.alibaba.com Signed-off-by: Greg Kroah-Hartman [mkbestas: Adapt for android <=5.4 kernel which contains backports that caused various conflicts] Signed-off-by: Michael Bestas --- fs/erofs/decompressor.c | 35 ++++++++++++++++++----------------- 1 file changed, 18 insertions(+), 17 deletions(-) diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c index 8193c14bb111..fbf646e9c70e 100644 --- a/fs/erofs/decompressor.c +++ b/fs/erofs/decompressor.c @@ -23,7 +23,7 @@ struct z_erofs_decompressor { */ int (*prepare_destpages)(struct z_erofs_decompress_req *rq, struct list_head *pagepool); - int (*decompress)(struct z_erofs_decompress_req *rq, u8 *out); + int (*decompress)(struct z_erofs_decompress_req *rq, u8 *dst); char *name; }; @@ -124,11 +124,11 @@ static int z_erofs_lz4_prepare_destpages(struct z_erofs_decompress_req *rq, } static void *z_erofs_handle_inplace_io(struct z_erofs_decompress_req *rq, - void *inpage, unsigned int *inputmargin, int *maptype, - bool support_0padding) + void *inpage, void *out, unsigned int *inputmargin, + int *maptype, bool support_0padding) { unsigned int nrpages_in, nrpages_out; - unsigned int ofull, oend, inputsize, total, i, j; + unsigned int ofull, oend, inputsize, total, i; struct page **in; void *src, *tmp; @@ -143,12 +143,13 @@ static void *z_erofs_handle_inplace_io(struct z_erofs_decompress_req *rq, ofull - oend < LZ4_DECOMPRESS_INPLACE_MARGIN(inputsize)) goto docopy; - for (i = 0; i < nrpages_in; ++i) { - DBG_BUGON(rq->in[i] == NULL); - for (j = 0; j < nrpages_out - nrpages_in + i; ++j) - if (rq->out[j] == rq->in[i]) - goto docopy; - } + for (i = 0; i < nrpages_in; ++i) + if (rq->out[nrpages_out - nrpages_in + i] != + rq->in[i]) + goto docopy; + kunmap_atomic(inpage); + *maptype = 3; + return out + ((nrpages_out - nrpages_in) << PAGE_SHIFT); } if (nrpages_in <= 1) { @@ -156,7 +157,6 @@ static void *z_erofs_handle_inplace_io(struct z_erofs_decompress_req *rq, return inpage; } kunmap_atomic(inpage); - might_sleep(); src = erofs_vm_map_ram(rq->in, nrpages_in); if (!src) return ERR_PTR(-ENOMEM); @@ -193,10 +193,10 @@ docopy: return src; } -static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out) +static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *dst) { unsigned int inputmargin; - u8 *headpage, *src; + u8 *out, *headpage, *src; bool support_0padding; int ret, maptype; @@ -220,11 +220,12 @@ static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out) } rq->inputsize -= inputmargin; - src = z_erofs_handle_inplace_io(rq, headpage, &inputmargin, &maptype, - support_0padding); + src = z_erofs_handle_inplace_io(rq, headpage, dst, &inputmargin, + &maptype, support_0padding); if (IS_ERR(src)) return PTR_ERR(src); + out = dst + rq->pageofs_out; /* legacy format could compress extra data in a pcluster. */ if (rq->partial_decoding || !support_0padding) ret = LZ4_decompress_safe_partial(src + inputmargin, out, @@ -253,7 +254,7 @@ static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out) vm_unmap_ram(src, PAGE_ALIGN(rq->inputsize) >> PAGE_SHIFT); } else if (maptype == 2) { erofs_put_pcpubuf(src); - } else { + } else if (maptype != 3) { DBG_BUGON(1); return -EFAULT; } @@ -354,7 +355,7 @@ static int z_erofs_decompress_generic(struct z_erofs_decompress_req *rq, dst_maptype = 2; dstmap_out: - ret = alg->decompress(rq, dst + rq->pageofs_out); + ret = alg->decompress(rq, dst); if (!dst_maptype) kunmap_atomic(dst);