Index: linux-2.4.29/Documentation/Configure.help =================================================================== --- linux-2.4.29.orig/Documentation/Configure.help +++ linux-2.4.29/Documentation/Configure.help @@ -16679,6 +16679,30 @@ CONFIG_EXT2_FS be compiled as a module, and so this could be dangerous. Most everyone wants to say Y here. +Ext2 extended attributes +CONFIG_EXT2_FS_XATTR + Extended attributes are name:value pairs associated with inodes by + the kernel or by users (see the attr(5) manual page, or visit + for details). + + If unsure, say N. + +Ext2 extended attribute block sharing +CONFIG_EXT2_FS_XATTR_SHARING + This options enables code for sharing identical extended attribute + blocks among multiple inodes. + + Usually, say Y. + +Ext2 extended user attributes +CONFIG_EXT2_FS_XATTR_USER + This option enables extended user attributes on ext2. Processes can + associate extended user attributes with inodes to store additional + information such as the character encoding of files, etc. (see the + attr(5) manual page, or visit for details). + + If unsure, say N. + Ext3 journalling file system support (EXPERIMENTAL) CONFIG_EXT3_FS This is the journalling version of the Second extended file system @@ -16711,6 +16735,30 @@ CONFIG_EXT3_FS of your root partition (the one containing the directory /) cannot be compiled as a module, and so this may be dangerous. +Ext3 extended attributes +CONFIG_EXT3_FS_XATTR + Extended attributes are name:value pairs associated with inodes by + the kernel or by users (see the attr(5) manual page, or visit + for details). + + If unsure, say N. + +Ext3 extended attribute block sharing +CONFIG_EXT3_FS_XATTR_SHARING + This options enables code for sharing identical extended attribute + blocks among multiple inodes. + + Usually, say Y. + +Ext3 extended user attributes +CONFIG_EXT3_FS_XATTR_USER + This option enables extended user attributes on ext3. Processes can + associate extended user attributes with inodes to store additional + information such as the character encoding of files, etc. (see the + attr(5) manual page, or visit for details). + + If unsure, say N. + Journal Block Device support (JBD for ext3) (EXPERIMENTAL) CONFIG_JBD This is a generic journalling layer for block devices. It is Index: linux-2.4.29/Documentation/filesystems/Locking =================================================================== --- linux-2.4.29.orig/Documentation/filesystems/Locking +++ linux-2.4.29/Documentation/filesystems/Locking @@ -69,8 +69,8 @@ permission: yes no no getattr: (see below) revalidate: no (see below) setxattr: yes yes no -getxattr: yes yes no -listxattr: yes yes no +getxattr: yes no no +listxattr: yes no no removexattr: yes yes no Additionally, ->rmdir() has i_zombie on victim and so does ->rename() in case when target exists and is a directory. Index: linux-2.4.29/arch/alpha/kernel/entry.S =================================================================== --- linux-2.4.29.orig/arch/alpha/kernel/entry.S +++ linux-2.4.29/arch/alpha/kernel/entry.S @@ -10,7 +10,7 @@ #define SIGCHLD 20 -#define NR_SYSCALLS 382 +#define NR_SYSCALLS 394 /* * These offsets must match with alpha_mv in . @@ -1154,6 +1154,18 @@ sys_call_table: .quad sys_readahead .quad sys_ni_syscall /* 380, sys_security */ .quad sys_tkill + .quad sys_setxattr + .quad sys_lsetxattr + .quad sys_fsetxattr + .quad sys_getxattr /* 385 */ + .quad sys_lgetxattr + .quad sys_fgetxattr + .quad sys_listxattr + .quad sys_llistxattr + .quad sys_flistxattr /* 390 */ + .quad sys_removexattr + .quad sys_lremovexattr + .quad sys_fremovexattr /* Remember to update everything, kids. */ .ifne (. - sys_call_table) - (NR_SYSCALLS * 8) Index: linux-2.4.29/arch/arm/kernel/calls.S =================================================================== --- linux-2.4.29.orig/arch/arm/kernel/calls.S +++ linux-2.4.29/arch/arm/kernel/calls.S @@ -240,18 +240,18 @@ __syscall_start: .long SYMBOL_NAME(sys_ni_syscall) /* Security */ .long SYMBOL_NAME(sys_gettid) /* 225 */ .long SYMBOL_NAME(sys_readahead) - .long SYMBOL_NAME(sys_ni_syscall) /* setxattr */ - .long SYMBOL_NAME(sys_ni_syscall) /* lsetxattr */ - .long SYMBOL_NAME(sys_ni_syscall) /* fsetxattr */ - .long SYMBOL_NAME(sys_ni_syscall) /* getxattr */ -/* 230 */ .long SYMBOL_NAME(sys_ni_syscall) /* lgetxattr */ - .long SYMBOL_NAME(sys_ni_syscall) /* fgetxattr */ - .long SYMBOL_NAME(sys_ni_syscall) /* listxattr */ - .long SYMBOL_NAME(sys_ni_syscall) /* llistxattr */ - .long SYMBOL_NAME(sys_ni_syscall) /* flistxattr */ -/* 235 */ .long SYMBOL_NAME(sys_ni_syscall) /* removexattr */ - .long SYMBOL_NAME(sys_ni_syscall) /* lremovexattr */ - .long SYMBOL_NAME(sys_ni_syscall) /* fremovexattr */ + .long SYMBOL_NAME(sys_setxattr) + .long SYMBOL_NAME(sys_lsetxattr) + .long SYMBOL_NAME(sys_fsetxattr) + .long SYMBOL_NAME(sys_getxattr) +/* 230 */ .long SYMBOL_NAME(sys_lgetxattr) + .long SYMBOL_NAME(sys_fgetxattr) + .long SYMBOL_NAME(sys_listxattr) + .long SYMBOL_NAME(sys_llistxattr) + .long SYMBOL_NAME(sys_flistxattr) +/* 235 */ .long SYMBOL_NAME(sys_removexattr) + .long SYMBOL_NAME(sys_lremovexattr) + .long SYMBOL_NAME(sys_fremovexattr) .long SYMBOL_NAME(sys_tkill) .long SYMBOL_NAME(sys_ni_syscall) /* sendfile64 */ /* 240 */ .long SYMBOL_NAME(sys_ni_syscall) /* futex */ Index: linux-2.4.29/arch/ppc64/kernel/misc.S =================================================================== --- linux-2.4.29.orig/arch/ppc64/kernel/misc.S +++ linux-2.4.29/arch/ppc64/kernel/misc.S @@ -807,7 +807,6 @@ _GLOBAL(sys_call_table32) .llong .sys_madvise /* 205 */ .llong .sys_mincore /* 206 */ .llong .sys_gettid /* 207 */ -#if 0 /* Reserved syscalls */ .llong .sys_tkill /* 208 */ .llong .sys_setxattr .llong .sys_lsetxattr /* 210 */ @@ -821,10 +820,7 @@ _GLOBAL(sys_call_table32) .llong .sys_removexattr .llong .sys_lremovexattr .llong .sys_fremovexattr /* 220 */ - .llong .sys_futex -#endif - .llong .sys_perfmonctl /* Put this here for now ... */ - .rept NR_syscalls-222 + .rept NR_syscalls-221 .llong .sys_ni_syscall .endr #endif @@ -1038,7 +1034,6 @@ _GLOBAL(sys_call_table) .llong .sys_madvise /* 205 */ .llong .sys_mincore /* 206 */ .llong .sys_gettid /* 207 */ -#if 0 /* Reserved syscalls */ .llong .sys_tkill /* 208 */ .llong .sys_setxattr .llong .sys_lsetxattr /* 210 */ @@ -1052,9 +1047,6 @@ _GLOBAL(sys_call_table) .llong .sys_removexattr .llong .sys_lremovexattr .llong .sys_fremovexattr /* 220 */ - .llong .sys_futex -#endif - .llong .sys_perfmonctl /* Put this here for now ... */ - .rept NR_syscalls-222 + .rept NR_syscalls-221 .llong .sys_ni_syscall .endr Index: linux-2.4.29/arch/s390/kernel/entry.S =================================================================== --- linux-2.4.29.orig/arch/s390/kernel/entry.S +++ linux-2.4.29/arch/s390/kernel/entry.S @@ -558,18 +558,18 @@ sys_call_table: .long sys_fcntl64 .long sys_readahead .long sys_ni_syscall - .long sys_ni_syscall /* 224 - reserved for setxattr */ - .long sys_ni_syscall /* 225 - reserved for lsetxattr */ - .long sys_ni_syscall /* 226 - reserved for fsetxattr */ - .long sys_ni_syscall /* 227 - reserved for getxattr */ - .long sys_ni_syscall /* 228 - reserved for lgetxattr */ - .long sys_ni_syscall /* 229 - reserved for fgetxattr */ - .long sys_ni_syscall /* 230 - reserved for listxattr */ - .long sys_ni_syscall /* 231 - reserved for llistxattr */ - .long sys_ni_syscall /* 232 - reserved for flistxattr */ - .long sys_ni_syscall /* 233 - reserved for removexattr */ - .long sys_ni_syscall /* 234 - reserved for lremovexattr */ - .long sys_ni_syscall /* 235 - reserved for fremovexattr */ + .long sys_setxattr + .long sys_lsetxattr /* 225 */ + .long sys_fsetxattr + .long sys_getxattr + .long sys_lgetxattr + .long sys_fgetxattr + .long sys_listxattr /* 230 */ + .long sys_llistxattr + .long sys_flistxattr + .long sys_removexattr + .long sys_lremovexattr + .long sys_fremovexattr /* 235 */ .long sys_gettid .long sys_tkill .rept 255-237 Index: linux-2.4.29/arch/s390x/kernel/entry.S =================================================================== --- linux-2.4.29.orig/arch/s390x/kernel/entry.S +++ linux-2.4.29/arch/s390x/kernel/entry.S @@ -591,18 +591,18 @@ sys_call_table: .long SYSCALL(sys_ni_syscall,sys32_fcntl64_wrapper) .long SYSCALL(sys_readahead,sys32_readahead) .long SYSCALL(sys_ni_syscall,sys_ni_syscall) - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 224 - reserved for setxattr */ - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 225 - reserved for lsetxattr */ - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 226 - reserved for fsetxattr */ - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 227 - reserved for getxattr */ - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 228 - reserved for lgetxattr */ - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 229 - reserved for fgetxattr */ - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 230 - reserved for listxattr */ - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 231 - reserved for llistxattr */ - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 232 - reserved for flistxattr */ - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 233 - reserved for removexattr */ - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 234 - reserved for lremovexattr */ - .long SYSCALL(sys_ni_syscall,sys_ni_syscall) /* 235 - reserved for fremovexattr */ + .long SYSCALL(sys_setxattr,sys32_setxattr_wrapper) + .long SYSCALL(sys_lsetxattr,sys32_lsetxattr_wrapper) /* 225 */ + .long SYSCALL(sys_fsetxattr,sys32_fsetxattr_wrapper) + .long SYSCALL(sys_getxattr,sys32_getxattr_wrapper) + .long SYSCALL(sys_lgetxattr,sys32_lgetxattr_wrapper) + .long SYSCALL(sys_fgetxattr,sys32_fgetxattr_wrapper) + .long SYSCALL(sys_listxattr,sys32_listxattr_wrapper) /* 230 */ + .long SYSCALL(sys_llistxattr,sys32_llistxattr_wrapper) + .long SYSCALL(sys_flistxattr,sys32_flistxattr_wrapper) + .long SYSCALL(sys_removexattr,sys32_removexattr_wrapper) + .long SYSCALL(sys_lremovexattr,sys32_lremovexattr_wrapper) + .long SYSCALL(sys_fremovexattr,sys32_fremovexattr_wrapper)/* 235 */ .long SYSCALL(sys_gettid,sys_gettid) .long SYSCALL(sys_tkill,sys_tkill) .rept 255-237 Index: linux-2.4.29/arch/s390x/kernel/wrapper32.S =================================================================== --- linux-2.4.29.orig/arch/s390x/kernel/wrapper32.S +++ linux-2.4.29/arch/s390x/kernel/wrapper32.S @@ -1107,3 +1107,93 @@ sys32_stime_wrapper: sys32_sysctl_wrapper: llgtr %r2,%r2 # struct __sysctl_args32 * jg sys32_sysctl + + .globl sys32_setxattr_wrapper +sys32_setxattr_wrapper: + llgtr %r2,%r2 # char * + llgtr %r3,%r3 # char * + llgtr %r4,%r4 # void * + llgfr %r5,%r5 # size_t + lgfr %r6,%r6 # int + jg sys_setxattr + + .globl sys32_lsetxattr_wrapper +sys32_lsetxattr_wrapper: + llgtr %r2,%r2 # char * + llgtr %r3,%r3 # char * + llgtr %r4,%r4 # void * + llgfr %r5,%r5 # size_t + lgfr %r6,%r6 # int + jg sys_lsetxattr + + .globl sys32_fsetxattr_wrapper +sys32_fsetxattr_wrapper: + lgfr %r2,%r2 # int + llgtr %r3,%r3 # char * + llgtr %r4,%r4 # void * + llgfr %r5,%r5 # size_t + lgfr %r6,%r6 # int + jg sys_fsetxattr + + .globl sys32_getxattr_wrapper +sys32_getxattr_wrapper: + llgtr %r2,%r2 # char * + llgtr %r3,%r3 # char * + llgtr %r4,%r4 # void * + llgfr %r5,%r5 # size_t + jg sys_getxattr + + .globl sys32_lgetxattr_wrapper +sys32_lgetxattr_wrapper: + llgtr %r2,%r2 # char * + llgtr %r3,%r3 # char * + llgtr %r4,%r4 # void * + llgfr %r5,%r5 # size_t + jg sys_lgetxattr + + .globl sys32_fgetxattr_wrapper +sys32_fgetxattr_wrapper: + lgfr %r2,%r2 # int + llgtr %r3,%r3 # char * + llgtr %r4,%r4 # void * + llgfr %r5,%r5 # size_t + jg sys_fgetxattr + + .globl sys32_listxattr_wrapper +sys32_listxattr_wrapper: + llgtr %r2,%r2 # char * + llgtr %r3,%r3 # char * + llgfr %r4,%r4 # size_t + jg sys_listxattr + + .globl sys32_llistxattr_wrapper +sys32_llistxattr_wrapper: + llgtr %r2,%r2 # char * + llgtr %r3,%r3 # char * + llgfr %r4,%r4 # size_t + jg sys_llistxattr + + .globl sys32_flistxattr_wrapper +sys32_flistxattr_wrapper: + lgfr %r2,%r2 # int + llgtr %r3,%r3 # char * + llgfr %r4,%r4 # size_t + jg sys_flistxattr + + .globl sys32_removexattr_wrapper +sys32_removexattr_wrapper: + llgtr %r2,%r2 # char * + llgtr %r3,%r3 # char * + jg sys_removexattr + + .globl sys32_lremovexattr_wrapper +sys32_lremovexattr_wrapper: + llgtr %r2,%r2 # char * + llgtr %r3,%r3 # char * + jg sys_lremovexattr + + .globl sys32_fremovexattr_wrapper +sys32_fremovexattr_wrapper: + lgfr %r2,%r2 # int + llgtr %r3,%r3 # char * + jg sys_fremovexattr Index: linux-2.4.29/fs/Config.in =================================================================== --- linux-2.4.29.orig/fs/Config.in +++ linux-2.4.29/fs/Config.in @@ -29,6 +29,13 @@ dep_mbool ' Debug Befs' CONFIG_BEFS_DEB dep_tristate 'BFS file system support (EXPERIMENTAL)' CONFIG_BFS_FS $CONFIG_EXPERIMENTAL tristate 'Ext3 journalling file system support' CONFIG_EXT3_FS +dep_mbool ' Ext3 extended attributes' CONFIG_EXT3_FS_XATTR $CONFIG_EXT3_FS +dep_bool ' Ext3 extended attribute block sharing' \ + CONFIG_EXT3_FS_XATTR_SHARING $CONFIG_EXT3_FS_XATTR +dep_bool ' Ext3 extended user attributes' \ + CONFIG_EXT3_FS_XATTR_USER $CONFIG_EXT3_FS_XATTR +dep_bool ' Ext3 trusted extended attributes' \ + CONFIG_EXT3_FS_XATTR_TRUSTED $CONFIG_EXT3_FS_XATTR # CONFIG_JBD could be its own option (even modular), but until there are # other users than ext3, we will simply make it be the same as CONFIG_EXT3_FS # dep_tristate ' Journal Block Device support (JBD for ext3)' CONFIG_JBD $CONFIG_EXT3_FS @@ -92,6 +99,13 @@ dep_mbool ' QNX4FS write support (DANGE tristate 'ROM file system support' CONFIG_ROMFS_FS tristate 'Second extended fs support' CONFIG_EXT2_FS +dep_mbool ' Ext2 extended attributes' CONFIG_EXT2_FS_XATTR $CONFIG_EXT2_FS +dep_bool ' Ext2 extended attribute block sharing' \ + CONFIG_EXT2_FS_XATTR_SHARING $CONFIG_EXT2_FS_XATTR +dep_bool ' Ext2 extended user attributes' \ + CONFIG_EXT2_FS_XATTR_USER $CONFIG_EXT2_FS_XATTR +dep_bool ' Ext2 trusted extended attributes' \ + CONFIG_EXT2_FS_XATTR_TRUSTED $CONFIG_EXT2_FS_XATTR tristate 'System V/Xenix/V7/Coherent file system support' CONFIG_SYSV_FS @@ -171,6 +185,10 @@ else define_tristate CONFIG_ZISOFS_FS n fi +# Meta block cache for Extended Attributes (ext2/ext3) +#tristate 'Meta block cache' CONFIG_FS_MBCACHE +define_tristate CONFIG_FS_MBCACHE y + mainmenu_option next_comment comment 'Partition Types' source fs/partitions/Config.in Index: linux-2.4.29/fs/Makefile =================================================================== --- linux-2.4.29.orig/fs/Makefile +++ linux-2.4.29/fs/Makefile @@ -77,6 +77,9 @@ obj-y += binfmt_script.o obj-$(CONFIG_BINFMT_ELF) += binfmt_elf.o +export-objs += mbcache.o +obj-$(CONFIG_FS_MBCACHE) += mbcache.o + # persistent filesystems obj-y += $(join $(subdir-y),$(subdir-y:%=/%.o)) Index: linux-2.4.29/fs/ext2/Makefile =================================================================== --- linux-2.4.29.orig/fs/ext2/Makefile +++ linux-2.4.29/fs/ext2/Makefile @@ -13,4 +13,9 @@ obj-y := balloc.o bitmap.o dir.o file ioctl.o namei.o super.o symlink.o obj-m := $(O_TARGET) +export-objs += xattr.o +obj-$(CONFIG_EXT2_FS_XATTR) += xattr.o +obj-$(CONFIG_EXT2_FS_XATTR_USER) += xattr_user.o +obj-$(CONFIG_EXT2_FS_XATTR_TRUSTED) += xattr_trusted.o + include $(TOPDIR)/Rules.make Index: linux-2.4.29/fs/ext2/file.c =================================================================== --- linux-2.4.29.orig/fs/ext2/file.c +++ linux-2.4.29/fs/ext2/file.c @@ -20,6 +20,7 @@ #include #include +#include #include /* @@ -51,4 +52,8 @@ struct file_operations ext2_file_operati struct inode_operations ext2_file_inode_operations = { truncate: ext2_truncate, + setxattr: ext2_setxattr, + getxattr: ext2_getxattr, + listxattr: ext2_listxattr, + removexattr: ext2_removexattr, }; Index: linux-2.4.29/fs/ext2/ialloc.c =================================================================== --- linux-2.4.29.orig/fs/ext2/ialloc.c +++ linux-2.4.29/fs/ext2/ialloc.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -167,6 +168,7 @@ void ext2_free_inode (struct inode * ino */ if (!is_bad_inode(inode)) { /* Quota is already initialized in iput() */ + ext2_xattr_delete_inode(inode); DQUOT_FREE_INODE(inode); DQUOT_DROP(inode); } @@ -395,6 +397,10 @@ repeat: inode->i_generation = event++; mark_inode_dirty(inode); +#ifdef CONFIG_EXT2_FS_XATTR + init_rwsem(&inode->u.ext2_i.xattr_sem); +#endif + unlock_super (sb); if(DQUOT_ALLOC_INODE(inode)) { DQUOT_DROP(inode); Index: linux-2.4.29/fs/ext2/inode.c =================================================================== --- linux-2.4.29.orig/fs/ext2/inode.c +++ linux-2.4.29/fs/ext2/inode.c @@ -64,9 +64,7 @@ void ext2_delete_inode (struct inode * i { lock_kernel(); - if (is_bad_inode(inode) || - inode->i_ino == EXT2_ACL_IDX_INO || - inode->i_ino == EXT2_ACL_DATA_INO) + if (is_bad_inode(inode)) goto no_delete; inode->u.ext2_i.i_dtime = CURRENT_TIME; mark_inode_dirty(inode); @@ -917,8 +915,7 @@ void ext2_read_inode (struct inode * ino unsigned long offset; struct ext2_group_desc * gdp; - if ((inode->i_ino != EXT2_ROOT_INO && inode->i_ino != EXT2_ACL_IDX_INO && - inode->i_ino != EXT2_ACL_DATA_INO && + if ((inode->i_ino != EXT2_ROOT_INO && inode->i_ino < EXT2_FIRST_INO(inode->i_sb)) || inode->i_ino > le32_to_cpu(inode->i_sb->u.ext2_sb.s_es->s_inodes_count)) { ext2_error (inode->i_sb, "ext2_read_inode", @@ -1004,10 +1001,7 @@ void ext2_read_inode (struct inode * ino for (block = 0; block < EXT2_N_BLOCKS; block++) inode->u.ext2_i.i_data[block] = raw_inode->i_block[block]; - if (inode->i_ino == EXT2_ACL_IDX_INO || - inode->i_ino == EXT2_ACL_DATA_INO) - /* Nothing to do */ ; - else if (S_ISREG(inode->i_mode)) { + if (S_ISREG(inode->i_mode)) { inode->i_op = &ext2_file_inode_operations; inode->i_fop = &ext2_file_operations; inode->i_mapping->a_ops = &ext2_aops; @@ -1019,15 +1013,20 @@ void ext2_read_inode (struct inode * ino if (ext2_inode_is_fast_symlink(inode)) inode->i_op = &ext2_fast_symlink_inode_operations; else { - inode->i_op = &page_symlink_inode_operations; + inode->i_op = &ext2_symlink_inode_operations; inode->i_mapping->a_ops = &ext2_aops; } - } else + } else { + inode->i_op = &ext2_special_inode_operations; init_special_inode(inode, inode->i_mode, le32_to_cpu(raw_inode->i_block[0])); + } brelse (bh); inode->i_attr_flags = 0; ext2_set_inode_flags(inode); +#ifdef CONFIG_EXT2_FS_XATTR + init_rwsem(&inode->u.ext2_i.xattr_sem); +#endif return; bad_inode: Index: linux-2.4.29/fs/ext2/namei.c =================================================================== --- linux-2.4.29.orig/fs/ext2/namei.c +++ linux-2.4.29/fs/ext2/namei.c @@ -31,6 +31,7 @@ #include #include +#include #include /* @@ -136,7 +137,7 @@ static int ext2_symlink (struct inode * if (l > sizeof (inode->u.ext2_i.i_data)) { /* slow symlink */ - inode->i_op = &page_symlink_inode_operations; + inode->i_op = &ext2_symlink_inode_operations; inode->i_mapping->a_ops = &ext2_aops; err = block_symlink(inode, symname, l); if (err) @@ -345,4 +346,15 @@ struct inode_operations ext2_dir_inode_o rmdir: ext2_rmdir, mknod: ext2_mknod, rename: ext2_rename, + setxattr: ext2_setxattr, + getxattr: ext2_getxattr, + listxattr: ext2_listxattr, + removexattr: ext2_removexattr, +}; + +struct inode_operations ext2_special_inode_operations = { + setxattr: ext2_setxattr, + getxattr: ext2_getxattr, + listxattr: ext2_listxattr, + removexattr: ext2_removexattr, }; Index: linux-2.4.29/fs/ext2/super.c =================================================================== --- linux-2.4.29.orig/fs/ext2/super.c +++ linux-2.4.29/fs/ext2/super.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include #include @@ -125,6 +126,7 @@ void ext2_put_super (struct super_block int db_count; int i; + ext2_xattr_put_super(sb); if (!(sb->s_flags & MS_RDONLY)) { struct ext2_super_block *es = EXT2_SB(sb)->s_es; @@ -175,6 +177,13 @@ static int parse_options (char * options this_char = strtok (NULL, ",")) { if ((value = strchr (this_char, '=')) != NULL) *value++ = 0; +#ifdef CONFIG_EXT2_FS_XATTR_USER + if (!strcmp (this_char, "user_xattr")) + set_opt (*mount_options, XATTR_USER); + else if (!strcmp (this_char, "nouser_xattr")) + clear_opt (*mount_options, XATTR_USER); + else +#endif if (!strcmp (this_char, "bsddf")) clear_opt (*mount_options, MINIX_DF); else if (!strcmp (this_char, "nouid32")) { @@ -446,6 +455,9 @@ struct super_block * ext2_read_super (st blocksize = BLOCK_SIZE; sb->u.ext2_sb.s_mount_opt = 0; +#ifdef CONFIG_EXT2_FS_XATTR_USER + /* set_opt (sb->u.ext2_sb.s_mount_opt, XATTR_USER); */ +#endif if (!parse_options ((char *) data, &sb_block, &resuid, &resgid, &sb->u.ext2_sb.s_mount_opt)) { return NULL; @@ -840,11 +852,21 @@ static DECLARE_FSTYPE_DEV(ext2_fs_type, static int __init init_ext2_fs(void) { - return register_filesystem(&ext2_fs_type); + int error = init_ext2_xattr(); + if (error) + return error; + error = init_ext2_xattr_user(); + if (error) + return error; + error = init_ext2_xattr_trusted(); + if (error) + return error; + return register_filesystem(&ext2_fs_type); } static void __exit exit_ext2_fs(void) { + exit_ext2_xattr(); unregister_filesystem(&ext2_fs_type); } Index: linux-2.4.29/fs/ext2/symlink.c =================================================================== --- linux-2.4.29.orig/fs/ext2/symlink.c +++ linux-2.4.29/fs/ext2/symlink.c @@ -19,6 +19,7 @@ #include #include +#include static int ext2_readlink(struct dentry *dentry, char *buffer, int buflen) { @@ -32,7 +33,20 @@ static int ext2_follow_link(struct dentr return vfs_follow_link(nd, s); } +struct inode_operations ext2_symlink_inode_operations = { + readlink: page_readlink, + follow_link: page_follow_link, + setxattr: ext2_setxattr, + getxattr: ext2_getxattr, + listxattr: ext2_listxattr, + removexattr: ext2_removexattr, +}; + struct inode_operations ext2_fast_symlink_inode_operations = { readlink: ext2_readlink, follow_link: ext2_follow_link, + setxattr: ext2_setxattr, + getxattr: ext2_getxattr, + listxattr: ext2_listxattr, + removexattr: ext2_removexattr, }; Index: linux-2.4.29/fs/ext2/xattr.c =================================================================== --- /dev/null +++ linux-2.4.29/fs/ext2/xattr.c @@ -0,0 +1,1145 @@ +/* + * linux/fs/ext2/xattr.c + * + * Copyright (C) 2001-2003 by Andreas Gruenbacher, + * + * Fix by Harrison Xing . + * Extended attributes for symlinks and special files added per + * suggestion of Luka Renko . + */ + +/* + * Extended attributes are stored on disk blocks allocated outside of + * any inode. The i_file_acl field is then made to point to this allocated + * block. If all extended attributes of an inode are identical, these + * inodes may share the same extended attribute block. Such situations + * are automatically detected by keeping a cache of recent attribute block + * numbers and hashes over the block's contents in memory. + * + * + * Extended attribute block layout: + * + * +------------------+ + * | header | + * | entry 1 | | + * | entry 2 | | growing downwards + * | entry 3 | v + * | four null bytes | + * | . . . | + * | value 1 | ^ + * | value 3 | | growing upwards + * | value 2 | | + * +------------------+ + * + * The block header is followed by multiple entry descriptors. These entry + * descriptors are variable in size, and alligned to EXT2_XATTR_PAD + * byte boundaries. The entry descriptors are sorted by attribute name, + * so that two extended attribute blocks can be compared efficiently. + * + * Attribute values are aligned to the end of the block, stored in + * no specific order. They are also padded to EXT2_XATTR_PAD byte + * boundaries. No additional gaps are left between them. + * + * Locking strategy + * ---------------- + * EXT2_I(inode)->i_file_acl is protected by EXT2_I(inode)->xattr_sem. + * EA blocks are only changed if they are exclusive to an inode, so + * holding xattr_sem also means that nothing but the EA block's reference + * count will change. Multiple writers to an EA block are synchronized + * by the bh lock. No more than a single bh lock is held at any time, + * which avoids deadlocks. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* These symbols may be needed by a module. */ +EXPORT_SYMBOL(ext2_xattr_register); +EXPORT_SYMBOL(ext2_xattr_unregister); +EXPORT_SYMBOL(ext2_xattr_get); +EXPORT_SYMBOL(ext2_xattr_list); +EXPORT_SYMBOL(ext2_xattr_set); + +#define HDR(bh) ((struct ext2_xattr_header *)((bh)->b_data)) +#define ENTRY(ptr) ((struct ext2_xattr_entry *)(ptr)) +#define FIRST_ENTRY(bh) ENTRY(HDR(bh)+1) +#define IS_LAST_ENTRY(entry) (*(__u32 *)(entry) == 0) + +#ifdef EXT2_XATTR_DEBUG +# define ea_idebug(inode, f...) do { \ + printk(KERN_DEBUG "inode %s:%ld: ", \ + kdevname(inode->i_dev), inode->i_ino); \ + printk(f); \ + printk("\n"); \ + } while (0) +# define ea_bdebug(bh, f...) do { \ + printk(KERN_DEBUG "block %s:%ld: ", \ + kdevname(bh->b_dev), bh->b_blocknr); \ + printk(f); \ + printk("\n"); \ + } while (0) +#else +# define ea_idebug(f...) +# define ea_bdebug(f...) +#endif + +static int ext2_xattr_set2(struct inode *, struct buffer_head *, + struct ext2_xattr_header *); + +#ifdef CONFIG_EXT2_FS_XATTR_SHARING + +static int ext2_xattr_cache_insert(struct buffer_head *); +static struct buffer_head *ext2_xattr_cache_find(struct inode *, + struct ext2_xattr_header *, + struct mb_cache_entry **); +static void ext2_xattr_rehash(struct ext2_xattr_header *, + struct ext2_xattr_entry *); + +static struct mb_cache *ext2_xattr_cache; + +#else +# define ext2_xattr_cache_insert(bh) 0 +# define ext2_xattr_cache_find(inode, header, pce) NULL +# define ext2_xattr_rehash(header, entry) while(0) {} +#endif + +struct ext2_xattr_handler *ext2_xattr_handlers[EXT2_XATTR_INDEX_MAX]; +rwlock_t ext2_handler_lock = RW_LOCK_UNLOCKED; + +int +ext2_xattr_register(int name_index, struct ext2_xattr_handler *handler) +{ + int error = -EINVAL; + + if (name_index > 0 && name_index <= EXT2_XATTR_INDEX_MAX) { + write_lock(&ext2_handler_lock); + if (!ext2_xattr_handlers[name_index-1]) { + ext2_xattr_handlers[name_index-1] = handler; + error = 0; + } + write_unlock(&ext2_handler_lock); + } + return error; +} + +void +ext2_xattr_unregister(int name_index, struct ext2_xattr_handler *handler) +{ + if (name_index > 0 || name_index <= EXT2_XATTR_INDEX_MAX) { + write_lock(&ext2_handler_lock); + ext2_xattr_handlers[name_index-1] = NULL; + write_unlock(&ext2_handler_lock); + } +} + +static inline const char * +strcmp_prefix(const char *a, const char *a_prefix) +{ + while (*a_prefix && *a == *a_prefix) { + a++; + a_prefix++; + } + return *a_prefix ? NULL : a; +} + +/* + * Decode the extended attribute name, and translate it into + * the name_index and name suffix. + */ +static struct ext2_xattr_handler * +ext2_xattr_resolve_name(const char **name) +{ + struct ext2_xattr_handler *handler = NULL; + int i; + + if (!*name) + return NULL; + read_lock(&ext2_handler_lock); + for (i=0; iprefix); + if (n) { + handler = ext2_xattr_handlers[i]; + *name = n; + break; + } + } + } + read_unlock(&ext2_handler_lock); + return handler; +} + +static inline struct ext2_xattr_handler * +ext2_xattr_handler(int name_index) +{ + struct ext2_xattr_handler *handler = NULL; + if (name_index > 0 && name_index <= EXT2_XATTR_INDEX_MAX) { + read_lock(&ext2_handler_lock); + handler = ext2_xattr_handlers[name_index-1]; + read_unlock(&ext2_handler_lock); + } + return handler; +} + +/* + * Inode operation getxattr() + * + * dentry->d_inode->i_sem: don't care + * BKL: held + */ +ssize_t +ext2_getxattr(struct dentry *dentry, const char *name, + void *buffer, size_t size) +{ + struct ext2_xattr_handler *handler; + struct inode *inode = dentry->d_inode; + + handler = ext2_xattr_resolve_name(&name); + if (!handler) + return -EOPNOTSUPP; + return handler->get(inode, name, buffer, size); +} + +/* + * Inode operation listxattr() + * + * dentry->d_inode->i_sem: don't care + * BKL: held + */ +ssize_t +ext2_listxattr(struct dentry *dentry, char *buffer, size_t size) +{ + return ext2_xattr_list(dentry->d_inode, buffer, size); +} + +/* + * Inode operation setxattr() + * + * dentry->d_inode->i_sem: down + * BKL: held + */ +int +ext2_setxattr(struct dentry *dentry, const char *name, void *value, + size_t size, int flags) +{ + struct ext2_xattr_handler *handler; + struct inode *inode = dentry->d_inode; + + if (size == 0) + value = ""; /* empty EA, do not remove */ + handler = ext2_xattr_resolve_name(&name); + if (!handler) + return -EOPNOTSUPP; + return handler->set(inode, name, value, size, flags); +} + +/* + * Inode operation removexattr() + * + * dentry->d_inode->i_sem: down + * BKL: held + */ +int +ext2_removexattr(struct dentry *dentry, const char *name) +{ + struct ext2_xattr_handler *handler; + struct inode *inode = dentry->d_inode; + + handler = ext2_xattr_resolve_name(&name); + if (!handler) + return -EOPNOTSUPP; + return handler->set(inode, name, NULL, 0, XATTR_REPLACE); +} + +/* + * ext2_xattr_get() + * + * Copy an extended attribute into the buffer + * provided, or compute the buffer size required. + * Buffer is NULL to compute the size of the buffer required. + * + * Returns a negative error number on failure, or the number of bytes + * used / required on success. + */ +int +ext2_xattr_get(struct inode *inode, int name_index, const char *name, + void *buffer, size_t buffer_size) +{ + struct buffer_head *bh = NULL; + struct ext2_xattr_entry *entry; + size_t size, name_len; + char *end; + int error; + + ea_idebug(inode, "name=%d.%s, buffer=%p, buffer_size=%ld", + name_index, name, buffer, (long)buffer_size); + + if (name == NULL) + return -EINVAL; + down_read(&EXT2_I(inode)->xattr_sem); + error = -ENODATA; + if (!EXT2_I(inode)->i_file_acl) + goto cleanup; + ea_idebug(inode, "reading block %d", EXT2_I(inode)->i_file_acl); + bh = sb_bread(inode->i_sb, EXT2_I(inode)->i_file_acl); + error = -EIO; + if (!bh) + goto cleanup; + ea_bdebug(bh, "b_count=%d, refcount=%d", + atomic_read(&(bh->b_count)), le32_to_cpu(HDR(bh)->h_refcount)); + end = bh->b_data + bh->b_size; + if (HDR(bh)->h_magic != cpu_to_le32(EXT2_XATTR_MAGIC) || + HDR(bh)->h_blocks != cpu_to_le32(1)) { +bad_block: ext2_error(inode->i_sb, "ext2_xattr_get", + "inode %ld: bad block %d", inode->i_ino, + EXT2_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + /* find named attribute */ + name_len = strlen(name); + + error = -ERANGE; + if (name_len > 255) + goto cleanup; + entry = FIRST_ENTRY(bh); + while (!IS_LAST_ENTRY(entry)) { + struct ext2_xattr_entry *next = + EXT2_XATTR_NEXT(entry); + if ((char *)next >= end) + goto bad_block; + if (name_index == entry->e_name_index && + name_len == entry->e_name_len && + memcmp(name, entry->e_name, name_len) == 0) + goto found; + entry = next; + } + /* Check the remaining name entries */ + while (!IS_LAST_ENTRY(entry)) { + struct ext2_xattr_entry *next = + EXT2_XATTR_NEXT(entry); + if ((char *)next >= end) + goto bad_block; + entry = next; + } + if (ext2_xattr_cache_insert(bh)) + ea_idebug(inode, "cache insert failed"); + error = -ENODATA; + goto cleanup; +found: + /* check the buffer size */ + if (entry->e_value_block != 0) + goto bad_block; + size = le32_to_cpu(entry->e_value_size); + if (size > inode->i_sb->s_blocksize || + le16_to_cpu(entry->e_value_offs) + size > inode->i_sb->s_blocksize) + goto bad_block; + + if (ext2_xattr_cache_insert(bh)) + ea_idebug(inode, "cache insert failed"); + if (buffer) { + error = -ERANGE; + if (size > buffer_size) + goto cleanup; + /* return value of attribute */ + memcpy(buffer, bh->b_data + le16_to_cpu(entry->e_value_offs), + size); + } + error = size; + +cleanup: + brelse(bh); + up_read(&EXT2_I(inode)->xattr_sem); + + return error; +} + +/* + * ext2_xattr_list() + * + * Copy a list of attribute names into the buffer + * provided, or compute the buffer size required. + * Buffer is NULL to compute the size of the buffer required. + * + * Returns a negative error number on failure, or the number of bytes + * used / required on success. + */ +int +ext2_xattr_list(struct inode *inode, char *buffer, size_t buffer_size) +{ + struct buffer_head *bh = NULL; + struct ext2_xattr_entry *entry; + size_t size = 0; + char *buf, *end; + int error; + + ea_idebug(inode, "buffer=%p, buffer_size=%ld", + buffer, (long)buffer_size); + + down_read(&EXT2_I(inode)->xattr_sem); + error = 0; + if (!EXT2_I(inode)->i_file_acl) + goto cleanup; + ea_idebug(inode, "reading block %d", EXT2_I(inode)->i_file_acl); + bh = sb_bread(inode->i_sb, EXT2_I(inode)->i_file_acl); + error = -EIO; + if (!bh) + goto cleanup; + ea_bdebug(bh, "b_count=%d, refcount=%d", + atomic_read(&(bh->b_count)), le32_to_cpu(HDR(bh)->h_refcount)); + end = bh->b_data + bh->b_size; + if (HDR(bh)->h_magic != cpu_to_le32(EXT2_XATTR_MAGIC) || + HDR(bh)->h_blocks != cpu_to_le32(1)) { +bad_block: ext2_error(inode->i_sb, "ext2_xattr_list", + "inode %ld: bad block %d", inode->i_ino, + EXT2_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + /* compute the size required for the list of attribute names */ + for (entry = FIRST_ENTRY(bh); !IS_LAST_ENTRY(entry); + entry = EXT2_XATTR_NEXT(entry)) { + struct ext2_xattr_handler *handler; + struct ext2_xattr_entry *next = + EXT2_XATTR_NEXT(entry); + if ((char *)next >= end) + goto bad_block; + + handler = ext2_xattr_handler(entry->e_name_index); + if (handler) + size += handler->list(NULL, inode, entry->e_name, + entry->e_name_len); + } + + if (ext2_xattr_cache_insert(bh)) + ea_idebug(inode, "cache insert failed"); + if (!buffer) { + error = size; + goto cleanup; + } else { + error = -ERANGE; + if (size > buffer_size) + goto cleanup; + } + + /* list the attribute names */ + buf = buffer; + for (entry = FIRST_ENTRY(bh); !IS_LAST_ENTRY(entry); + entry = EXT2_XATTR_NEXT(entry)) { + struct ext2_xattr_handler *handler; + + handler = ext2_xattr_handler(entry->e_name_index); + if (handler) + buf += handler->list(buf, inode, entry->e_name, + entry->e_name_len); + } + error = size; + +cleanup: + brelse(bh); + up_read(&EXT2_I(inode)->xattr_sem); + + return error; +} + +/* + * If the EXT2_FEATURE_COMPAT_EXT_ATTR feature of this file system is + * not set, set it. + */ +static void ext2_xattr_update_super_block(struct super_block *sb) +{ + if (EXT2_HAS_COMPAT_FEATURE(sb, EXT2_FEATURE_COMPAT_EXT_ATTR)) + return; + + lock_super(sb); + EXT2_SB(sb)->s_es->s_feature_compat |= + cpu_to_le32(EXT2_FEATURE_COMPAT_EXT_ATTR); + sb->s_dirt = 1; + mark_buffer_dirty(EXT2_SB(sb)->s_sbh); + unlock_super(sb); +} + +/* + * Release the xattr block BH: If the reference count is > 1, decrement + * it; otherwise free the block. + */ +static void +ext2_xattr_release_block(struct inode *inode, struct buffer_head *bh) +{ + struct mb_cache_entry *ce; + + ce = mb_cache_entry_get(ext2_xattr_cache, bh->b_dev, bh->b_blocknr); + if (HDR(bh)->h_refcount == cpu_to_le32(1)) { + ea_bdebug(bh, "freeing"); + if (ce) + mb_cache_entry_free(ce); + ext2_free_blocks(inode, bh->b_blocknr, 1); + } else { + lock_buffer(bh); + HDR(bh)->h_refcount = cpu_to_le32( + le32_to_cpu(HDR(bh)->h_refcount) - 1); + mark_buffer_dirty(bh); + ea_bdebug(bh, "releasing; refcount now=%d", + le32_to_cpu(HDR(bh)->h_refcount)); + unlock_buffer(bh); + if (ce) + mb_cache_entry_release(ce); + DQUOT_FREE_BLOCK(inode, 1); + } +} + +/* + * ext2_xattr_set() + * + * Create, replace or remove an extended attribute for this inode. Buffer + * is NULL to remove an existing extended attribute, and non-NULL to + * either replace an existing extended attribute, or create a new extended + * attribute. The flags XATTR_REPLACE and XATTR_CREATE + * specify that an extended attribute must exist and must not exist + * previous to the call, respectively. + * + * Returns 0, or a negative error number on failure. + */ +int +ext2_xattr_set(struct inode *inode, int name_index, const char *name, + const void *value, size_t value_len, int flags) +{ + struct super_block *sb = inode->i_sb; + struct buffer_head *bh = NULL; + struct ext2_xattr_header *header = NULL; + struct ext2_xattr_entry *here, *last; + size_t name_len, free, min_offs = sb->s_blocksize; + int not_found = 1, error; + char *end; + + /* + * header -- Points either into bh, or to a temporarily + * allocated buffer. + * here -- The named entry found, or the place for inserting, within + * the block pointed to by header. + * last -- Points right after the last named entry within the block + * pointed to by header. + * min_offs -- The offset of the first value (values are aligned + * towards the end of the block). + * end -- Points right after the block pointed to by header. + */ + + ea_idebug(inode, "name=%d.%s, value=%p, value_len=%ld", + name_index, name, value, (long)value_len); + + if (IS_RDONLY(inode)) + return -EROFS; + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + return -EPERM; + if (value == NULL) + value_len = 0; + if (name == NULL) + return -EINVAL; + name_len = strlen(name); + if (name_len > 255 || value_len > sb->s_blocksize) + return -ERANGE; + down_write(&EXT2_I(inode)->xattr_sem); + if (EXT2_I(inode)->i_file_acl) { + /* The inode already has an extended attribute block. */ + + bh = sb_bread(sb, EXT2_I(inode)->i_file_acl); + error = -EIO; + if (!bh) + goto cleanup; + ea_bdebug(bh, "b_count=%d, refcount=%d", + atomic_read(&(bh->b_count)), + le32_to_cpu(HDR(bh)->h_refcount)); + header = HDR(bh); + end = bh->b_data + bh->b_size; + if (header->h_magic != cpu_to_le32(EXT2_XATTR_MAGIC) || + header->h_blocks != cpu_to_le32(1)) { +bad_block: ext2_error(sb, "ext2_xattr_set", + "inode %ld: bad block %d", inode->i_ino, + EXT2_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + /* Find the named attribute. */ + here = FIRST_ENTRY(bh); + while (!IS_LAST_ENTRY(here)) { + struct ext2_xattr_entry *next = EXT2_XATTR_NEXT(here); + if ((char *)next >= end) + goto bad_block; + if (!here->e_value_block && here->e_value_size) { + size_t offs = le16_to_cpu(here->e_value_offs); + if (offs < min_offs) + min_offs = offs; + } + not_found = name_index - here->e_name_index; + if (!not_found) + not_found = name_len - here->e_name_len; + if (!not_found) + not_found = memcmp(name, here->e_name,name_len); + if (not_found <= 0) + break; + here = next; + } + last = here; + /* We still need to compute min_offs and last. */ + while (!IS_LAST_ENTRY(last)) { + struct ext2_xattr_entry *next = EXT2_XATTR_NEXT(last); + if ((char *)next >= end) + goto bad_block; + if (!last->e_value_block && last->e_value_size) { + size_t offs = le16_to_cpu(last->e_value_offs); + if (offs < min_offs) + min_offs = offs; + } + last = next; + } + + /* Check whether we have enough space left. */ + free = min_offs - ((char*)last - (char*)header) - sizeof(__u32); + } else { + /* We will use a new extended attribute block. */ + free = sb->s_blocksize - + sizeof(struct ext2_xattr_header) - sizeof(__u32); + here = last = NULL; /* avoid gcc uninitialized warning. */ + } + + if (not_found) { + /* Request to remove a nonexistent attribute? */ + error = -ENODATA; + if (flags & XATTR_REPLACE) + goto cleanup; + error = 0; + if (value == NULL) + goto cleanup; + } else { + /* Request to create an existing attribute? */ + error = -EEXIST; + if (flags & XATTR_CREATE) + goto cleanup; + if (!here->e_value_block && here->e_value_size) { + size_t size = le32_to_cpu(here->e_value_size); + + if (le16_to_cpu(here->e_value_offs) + size > + sb->s_blocksize || size > sb->s_blocksize) + goto bad_block; + free += EXT2_XATTR_SIZE(size); + } + free += EXT2_XATTR_LEN(name_len); + } + error = -ENOSPC; + if (free < EXT2_XATTR_LEN(name_len) + EXT2_XATTR_SIZE(value_len)) + goto cleanup; + + /* Here we know that we can set the new attribute. */ + + if (header) { + struct mb_cache_entry *ce; + + ce = mb_cache_entry_get(ext2_xattr_cache, bh->b_dev, + bh->b_blocknr); + if (header->h_refcount == cpu_to_le32(1)) { + if (ce) + mb_cache_entry_free(ce); + ea_bdebug(bh, "modifying in-place"); + lock_buffer(bh); + /* keep the buffer locked while modifying it. */ + } else { + int offset; + + if (ce) + mb_cache_entry_release(ce); + ea_bdebug(bh, "cloning"); + header = kmalloc(bh->b_size, GFP_KERNEL); + error = -ENOMEM; + if (header == NULL) + goto cleanup; + memcpy(header, HDR(bh), bh->b_size); + header->h_refcount = cpu_to_le32(1); + offset = (char *)here - bh->b_data; + here = ENTRY((char *)header + offset); + offset = (char *)last - bh->b_data; + last = ENTRY((char *)header + offset); + } + } else { + /* Allocate a buffer where we construct the new block. */ + header = kmalloc(sb->s_blocksize, GFP_KERNEL); + error = -ENOMEM; + if (header == NULL) + goto cleanup; + memset(header, 0, sb->s_blocksize); + end = (char *)header + sb->s_blocksize; + header->h_magic = cpu_to_le32(EXT2_XATTR_MAGIC); + header->h_blocks = header->h_refcount = cpu_to_le32(1); + last = here = ENTRY(header+1); + } + + /* Iff we are modifying the block in-place, bh is locked here. */ + + if (not_found) { + /* Insert the new name. */ + int size = EXT2_XATTR_LEN(name_len); + int rest = (char *)last - (char *)here; + memmove((char *)here + size, here, rest); + memset(here, 0, size); + here->e_name_index = name_index; + here->e_name_len = name_len; + memcpy(here->e_name, name, name_len); + } else { + if (!here->e_value_block && here->e_value_size) { + char *first_val = (char *)header + min_offs; + int offs = le16_to_cpu(here->e_value_offs); + char *val = (char *)header + offs; + size_t size = EXT2_XATTR_SIZE( + le32_to_cpu(here->e_value_size)); + + if (size == EXT2_XATTR_SIZE(value_len)) { + /* The old and the new value have the same + size. Just replace. */ + here->e_value_size = cpu_to_le32(value_len); + memset(val + size - EXT2_XATTR_PAD, 0, + EXT2_XATTR_PAD); /* Clear pad bytes. */ + memcpy(val, value, value_len); + goto skip_replace; + } + + /* Remove the old value. */ + memmove(first_val + size, first_val, val - first_val); + memset(first_val, 0, size); + here->e_value_offs = 0; + min_offs += size; + + /* Adjust all value offsets. */ + last = ENTRY(header+1); + while (!IS_LAST_ENTRY(last)) { + int o = le16_to_cpu(last->e_value_offs); + if (!last->e_value_block && o < offs) + last->e_value_offs = + cpu_to_le16(o + size); + last = EXT2_XATTR_NEXT(last); + } + } + if (value == NULL) { + /* Remove the old name. */ + int size = EXT2_XATTR_LEN(name_len); + last = ENTRY((char *)last - size); + memmove(here, (char*)here + size, + (char*)last - (char*)here); + memset(last, 0, size); + } + } + + if (value != NULL) { + /* Insert the new value. */ + here->e_value_size = cpu_to_le32(value_len); + if (value_len) { + size_t size = EXT2_XATTR_SIZE(value_len); + char *val = (char *)header + min_offs - size; + here->e_value_offs = + cpu_to_le16((char *)val - (char *)header); + memset(val + size - EXT2_XATTR_PAD, 0, + EXT2_XATTR_PAD); /* Clear the pad bytes. */ + memcpy(val, value, value_len); + } + } + +skip_replace: + if (!IS_LAST_ENTRY(ENTRY(header+1))) + ext2_xattr_rehash(header, here); + if (bh && header == HDR(bh)) { + /* we were modifying in-place. */ + unlock_buffer(bh); + mark_buffer_dirty(bh); + } + error = ext2_xattr_set2(inode, bh, IS_LAST_ENTRY(ENTRY(header+1)) ? + NULL : header); + +cleanup: + brelse(bh); + if (!(bh && header == HDR(bh))) + kfree(header); + up_write(&EXT2_I(inode)->xattr_sem); + + return error; +} + +/* + * Second half of ext2_xattr_set(): Update the file system. + */ +static int +ext2_xattr_set2(struct inode *inode, struct buffer_head *old_bh, + struct ext2_xattr_header *header) +{ + struct super_block *sb = inode->i_sb; + struct buffer_head *new_bh = NULL; + struct mb_cache_entry *ce = NULL; + int error; + + if (header) { + new_bh = ext2_xattr_cache_find(inode, header, &ce); + if (new_bh) { + /* We found an identical block in the cache. */ + if (new_bh == old_bh) + ea_bdebug(new_bh, "keeping this block"); + else { + error = -EDQUOT; + /* How can we enforce the allocation? */ + if (DQUOT_ALLOC_BLOCK(inode, 1)) + goto cleanup; + /* The old block is released after updating + the inode. */ + lock_buffer(new_bh); + HDR(new_bh)->h_refcount = cpu_to_le32(1 + + le32_to_cpu(HDR(new_bh)->h_refcount)); + ea_bdebug(new_bh, "reusing; refcount now=%d", + le32_to_cpu(HDR(new_bh)->h_refcount)); + unlock_buffer(new_bh); + mark_buffer_dirty(new_bh); + } + mb_cache_entry_release(ce); + ce = NULL; + } else if (old_bh && header == HDR(old_bh)) { + /* We were modifying this block in-place. */ + ea_bdebug(bs->bh, "keeping this block"); + new_bh = old_bh; + get_bh(new_bh); + ext2_xattr_cache_insert(new_bh); + } else { + /* We need to allocate a new block */ + int goal = le32_to_cpu(EXT2_SB(inode->i_sb)->s_es-> + s_first_data_block) + + EXT2_I(inode)->i_block_group * + EXT2_BLOCKS_PER_GROUP(inode->i_sb); + /* How can we enforce the allocation? */ + int block = ext2_new_block(inode, goal, 0, 0, &error); + if (error) + goto cleanup; + ea_idebug(inode, "creating block %d", block); + + new_bh = sb_getblk(sb, block); + if (!new_bh) { + ext2_free_blocks(inode, block, 1); + error = -EIO; + goto cleanup; + } + lock_buffer(new_bh); + memcpy(new_bh->b_data, header, new_bh->b_size); + mark_buffer_uptodate(new_bh, 1); + unlock_buffer(new_bh); + ext2_xattr_cache_insert(new_bh); + + ext2_xattr_update_super_block(sb); + mark_buffer_dirty(new_bh); + } + if (IS_SYNC(inode)) { + ll_rw_block(WRITE, 1, &new_bh); + wait_on_buffer(new_bh); + error = -EIO; + if (buffer_req(new_bh) && !buffer_uptodate(new_bh)) + goto cleanup; + } + } + + /* Update the inode. */ + EXT2_I(inode)->i_file_acl = new_bh ? new_bh->b_blocknr : 0; + inode->i_ctime = CURRENT_TIME; + if (IS_SYNC(inode)) { + error = ext2_sync_inode (inode); + if (error) + goto cleanup; + } else + mark_inode_dirty(inode); + + error = 0; + if (old_bh && old_bh != new_bh) { + /* + * If there was an old block and we are no longer using it, + * release the old block. + */ + ext2_xattr_release_block(inode, old_bh); + } + +cleanup: + if (ce) + mb_cache_entry_release(ce); + brelse(new_bh); + + return error; +} + +/* + * ext2_xattr_delete_inode() + * + * Free extended attribute resources associated with this inode. This + * is called immediately before an inode is freed. + */ +void +ext2_xattr_delete_inode(struct inode *inode) +{ + struct buffer_head *bh = NULL; + + if (!EXT2_I(inode)->i_file_acl) + goto cleanup; + bh = sb_bread(inode->i_sb, EXT2_I(inode)->i_file_acl); + if (!bh) { + ext2_error(inode->i_sb, "ext2_xattr_delete_inode", + "inode %ld: block %d read error", inode->i_ino, + EXT2_I(inode)->i_file_acl); + goto cleanup; + } + ea_bdebug(bh, "b_count=%d", atomic_read(&(bh->b_count))); + if (HDR(bh)->h_magic != cpu_to_le32(EXT2_XATTR_MAGIC) || + HDR(bh)->h_blocks != cpu_to_le32(1)) { + ext2_error(inode->i_sb, "ext2_xattr_delete_inode", + "inode %ld: bad block %d", inode->i_ino, + EXT2_I(inode)->i_file_acl); + goto cleanup; + } + ext2_xattr_release_block(inode, bh); + EXT2_I(inode)->i_file_acl = 0; + +cleanup: + brelse(bh); +} + +/* + * ext2_xattr_put_super() + * + * This is called when a file system is unmounted. + */ +void +ext2_xattr_put_super(struct super_block *sb) +{ +#ifdef CONFIG_EXT2_FS_XATTR_SHARING + mb_cache_shrink(ext2_xattr_cache, sb->s_dev); +#endif +} + +#ifdef CONFIG_EXT2_FS_XATTR_SHARING + +/* + * ext2_xattr_cache_insert() + * + * Create a new entry in the extended attribute cache, and insert + * it unless such an entry is already in the cache. + * + * Returns 0, or a negative error number on failure. + */ +static int +ext2_xattr_cache_insert(struct buffer_head *bh) +{ + __u32 hash = le32_to_cpu(HDR(bh)->h_hash); + struct mb_cache_entry *ce; + int error; + + ce = mb_cache_entry_alloc(ext2_xattr_cache); + if (!ce) + return -ENOMEM; + error = mb_cache_entry_insert(ce, bh->b_dev, bh->b_blocknr, &hash); + if (error) { + mb_cache_entry_free(ce); + if (error == -EBUSY) { + ea_bdebug(bh, "already in cache (%d cache entries)", + atomic_read(&ext2_xattr_cache->c_entry_count)); + error = 0; + } + } else { + ea_bdebug(bh, "inserting [%x] (%d cache entries)", (int)hash, + atomic_read(&ext2_xattr_cache->c_entry_count)); + mb_cache_entry_release(ce); + } + return error; +} + +/* + * ext2_xattr_cmp() + * + * Compare two extended attribute blocks for equality. + * + * Returns 0 if the blocks are equal, 1 if they differ, and + * a negative error number on errors. + */ +static int +ext2_xattr_cmp(struct ext2_xattr_header *header1, + struct ext2_xattr_header *header2) +{ + struct ext2_xattr_entry *entry1, *entry2; + + entry1 = ENTRY(header1+1); + entry2 = ENTRY(header2+1); + while (!IS_LAST_ENTRY(entry1)) { + if (IS_LAST_ENTRY(entry2)) + return 1; + if (entry1->e_hash != entry2->e_hash || + entry1->e_name_index != entry2->e_name_index || + entry1->e_name_len != entry2->e_name_len || + entry1->e_value_size != entry2->e_value_size || + memcmp(entry1->e_name, entry2->e_name, entry1->e_name_len)) + return 1; + if (entry1->e_value_block != 0 || entry2->e_value_block != 0) + return -EIO; + if (memcmp((char *)header1 + le16_to_cpu(entry1->e_value_offs), + (char *)header2 + le16_to_cpu(entry2->e_value_offs), + le32_to_cpu(entry1->e_value_size))) + return 1; + + entry1 = EXT2_XATTR_NEXT(entry1); + entry2 = EXT2_XATTR_NEXT(entry2); + } + if (!IS_LAST_ENTRY(entry2)) + return 1; + return 0; +} + +/* + * ext2_xattr_cache_find() + * + * Find an identical extended attribute block. + * + * Returns a locked buffer head to the block found, or NULL if such + * a block was not found or an error occurred. + */ +static struct buffer_head * +ext2_xattr_cache_find(struct inode *inode, struct ext2_xattr_header *header, + struct mb_cache_entry **pce) +{ + __u32 hash = le32_to_cpu(header->h_hash); + struct mb_cache_entry *ce; + + if (!header->h_hash) + return NULL; /* never share */ + ea_idebug(inode, "looking for cached blocks [%x]", (int)hash); +again: + ce = mb_cache_entry_find_first(ext2_xattr_cache, 0, inode->i_dev, hash); + while (ce) { + struct buffer_head *bh; + + if (IS_ERR(ce)) { + if (PTR_ERR(ce) == -EAGAIN) + goto again; + break; + } + + bh = sb_bread(inode->i_sb, ce->e_block); + if (!bh) { + ext2_error(inode->i_sb, "ext2_xattr_cache_find", + "inode %ld: block %ld read error", + inode->i_ino, ce->e_block); + } else if (le32_to_cpu(HDR(bh)->h_refcount) >= + EXT2_XATTR_REFCOUNT_MAX) { + ea_idebug(inode, "block %ld refcount %d>=%d", + (unsigned long) ce->e_block, + le32_to_cpu(HDR(bh)->h_refcount), + EXT2_XATTR_REFCOUNT_MAX); + } else if (ext2_xattr_cmp(header, HDR(bh)) == 0) { + *pce = ce; + return bh; + } + brelse(bh); + + ce = mb_cache_entry_find_next(ce, 0, inode->i_dev, hash); + } + return NULL; +} + +#define NAME_HASH_SHIFT 5 +#define VALUE_HASH_SHIFT 16 + +/* + * ext2_xattr_hash_entry() + * + * Compute the hash of an extended attribute. + */ +static inline void ext2_xattr_hash_entry(struct ext2_xattr_header *header, + struct ext2_xattr_entry *entry) +{ + __u32 hash = 0; + char *name = entry->e_name; + int n; + + for (n=0; n < entry->e_name_len; n++) { + hash = (hash << NAME_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - NAME_HASH_SHIFT)) ^ + *name++; + } + + if (entry->e_value_block == 0 && entry->e_value_size != 0) { + __u32 *value = (__u32 *)((char *)header + + le16_to_cpu(entry->e_value_offs)); + for (n = (le32_to_cpu(entry->e_value_size) + + EXT2_XATTR_ROUND) >> EXT2_XATTR_PAD_BITS; n; n--) { + hash = (hash << VALUE_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - VALUE_HASH_SHIFT)) ^ + le32_to_cpu(*value++); + } + } + entry->e_hash = cpu_to_le32(hash); +} + +#undef NAME_HASH_SHIFT +#undef VALUE_HASH_SHIFT + +#define BLOCK_HASH_SHIFT 16 + +/* + * ext2_xattr_rehash() + * + * Re-compute the extended attribute hash value after an entry has changed. + */ +static void ext2_xattr_rehash(struct ext2_xattr_header *header, + struct ext2_xattr_entry *entry) +{ + struct ext2_xattr_entry *here; + __u32 hash = 0; + + ext2_xattr_hash_entry(header, entry); + here = ENTRY(header+1); + while (!IS_LAST_ENTRY(here)) { + if (!here->e_hash) { + /* Block is not shared if an entry's hash value == 0 */ + hash = 0; + break; + } + hash = (hash << BLOCK_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - BLOCK_HASH_SHIFT)) ^ + le32_to_cpu(here->e_hash); + here = EXT2_XATTR_NEXT(here); + } + header->h_hash = cpu_to_le32(hash); +} + +#undef BLOCK_HASH_SHIFT + +int __init +init_ext2_xattr(void) +{ + ext2_xattr_cache = mb_cache_create("ext2_xattr", NULL, + sizeof(struct mb_cache_entry) + + sizeof(((struct mb_cache_entry *) 0)->e_indexes[0]), 1, 61); + if (!ext2_xattr_cache) + return -ENOMEM; + + return 0; +} + +void +exit_ext2_xattr(void) +{ + mb_cache_destroy(ext2_xattr_cache); +} + +#else /* CONFIG_EXT2_FS_XATTR_SHARING */ + +int __init +init_ext2_xattr(void) +{ + return 0; +} + +#endif /* CONFIG_EXT2_FS_XATTR_SHARING */ Index: linux-2.4.29/fs/ext2/xattr_trusted.c =================================================================== --- /dev/null +++ linux-2.4.29/fs/ext2/xattr_trusted.c @@ -0,0 +1,69 @@ +/* + * linux/fs/ext2/xattr_trusted.c + * Handler for trusted extended attributes. + * + * Copyright (C) 2001 by Andreas Gruenbacher, + */ + +#include +#include +#include +#include + +#define XATTR_TRUSTED_PREFIX "trusted." + +static size_t +ext2_xattr_trusted_list(char *list, struct inode *inode, + const char *name, int name_len) +{ + const int prefix_len = sizeof(XATTR_TRUSTED_PREFIX)-1; + + if (!capable(CAP_SYS_ADMIN)) + return 0; + + if (list) { + memcpy(list, XATTR_TRUSTED_PREFIX, prefix_len); + memcpy(list+prefix_len, name, name_len); + list[prefix_len + name_len] = '\0'; + } + return prefix_len + name_len + 1; +} + +static int +ext2_xattr_trusted_get(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + return ext2_xattr_get(inode, EXT2_XATTR_INDEX_TRUSTED, name, + buffer, size); +} + +static int +ext2_xattr_trusted_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + + if (strcmp(name, "") == 0) + return -EINVAL; + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + return ext2_xattr_set(inode, EXT2_XATTR_INDEX_TRUSTED, name, + value, size, flags); +} + +struct ext2_xattr_handler ext2_xattr_trusted_handler = { + prefix: XATTR_TRUSTED_PREFIX, + list: ext2_xattr_trusted_list, + get: ext2_xattr_trusted_get, + set: ext2_xattr_trusted_set, +}; + +int __init +init_ext2_xattr_trusted(void) +{ + return ext2_xattr_register(EXT2_XATTR_INDEX_TRUSTED, + &ext2_xattr_trusted_handler); +} Index: linux-2.4.29/fs/ext2/xattr_user.c =================================================================== --- /dev/null +++ linux-2.4.29/fs/ext2/xattr_user.c @@ -0,0 +1,88 @@ +/* + * linux/fs/ext2/xattr_user.c + * Handler for extended user attributes. + * + * Copyright (C) 2001 by Andreas Gruenbacher, + */ + +#include +#include +#include +#include +#include + +#ifdef CONFIG_EXT2_FS_POSIX_ACL +# include +#endif + +#define XATTR_USER_PREFIX "user." + +static size_t +ext2_xattr_user_list(char *list, struct inode *inode, + const char *name, int name_len) +{ + const int prefix_len = sizeof(XATTR_USER_PREFIX)-1; + + if (!test_opt(inode->i_sb, XATTR_USER)) + return 0; + + if (list) { + memcpy(list, XATTR_USER_PREFIX, prefix_len); + memcpy(list+prefix_len, name, name_len); + list[prefix_len + name_len] = '\0'; + } + return prefix_len + name_len + 1; +} + +static int +ext2_xattr_user_get(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + int error; + + if (strcmp(name, "") == 0) + return -EINVAL; + if (!test_opt(inode->i_sb, XATTR_USER)) + return -EOPNOTSUPP; + error = permission(inode, MAY_READ); + if (error) + return error; + + return ext2_xattr_get(inode, EXT2_XATTR_INDEX_USER, name, + buffer, size); +} + +static int +ext2_xattr_user_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + int error; + + if (strcmp(name, "") == 0) + return -EINVAL; + if (!test_opt(inode->i_sb, XATTR_USER)) + return -EOPNOTSUPP; + if (!S_ISREG(inode->i_mode) && + (!S_ISDIR(inode->i_mode) || inode->i_mode & S_ISVTX)) + return -EPERM; + error = permission(inode, MAY_WRITE); + if (error) + return error; + + return ext2_xattr_set(inode, EXT2_XATTR_INDEX_USER, name, + value, size, flags); +} + +struct ext2_xattr_handler ext2_xattr_user_handler = { + prefix: XATTR_USER_PREFIX, + list: ext2_xattr_user_list, + get: ext2_xattr_user_get, + set: ext2_xattr_user_set, +}; + +int __init +init_ext2_xattr_user(void) +{ + return ext2_xattr_register(EXT2_XATTR_INDEX_USER, + &ext2_xattr_user_handler); +} Index: linux-2.4.29/fs/ext3/Makefile =================================================================== --- linux-2.4.29.orig/fs/ext3/Makefile +++ linux-2.4.29/fs/ext3/Makefile @@ -1,5 +1,5 @@ # -# Makefile for the linux ext2-filesystem routines. +# Makefile for the linux ext3-filesystem routines. # # Note! Dependencies are done automagically by 'make dep', which also # removes any old dependencies. DON'T put your own dependencies here @@ -13,4 +13,9 @@ obj-y := balloc.o bitmap.o dir.o file ioctl.o namei.o super.o symlink.o obj-m := $(O_TARGET) +export-objs += xattr.o +obj-$(CONFIG_EXT3_FS_XATTR) += xattr.o +obj-$(CONFIG_EXT3_FS_XATTR_USER) += xattr_user.o +obj-$(CONFIG_EXT3_FS_XATTR_TRUSTED) += xattr_trusted.o + include $(TOPDIR)/Rules.make Index: linux-2.4.29/fs/ext3/file.c =================================================================== --- linux-2.4.29.orig/fs/ext3/file.c +++ linux-2.4.29/fs/ext3/file.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include @@ -124,5 +125,9 @@ struct file_operations ext3_file_operati struct inode_operations ext3_file_inode_operations = { truncate: ext3_truncate, /* BKL held */ setattr: ext3_setattr, /* BKL held */ + setxattr: ext3_setxattr, /* BKL held */ + getxattr: ext3_getxattr, /* BKL held */ + listxattr: ext3_listxattr, /* BKL held */ + removexattr: ext3_removexattr, /* BKL held */ }; Index: linux-2.4.29/fs/ext3/ialloc.c =================================================================== --- linux-2.4.29.orig/fs/ext3/ialloc.c +++ linux-2.4.29/fs/ext3/ialloc.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -216,6 +217,7 @@ void ext3_free_inode (handle_t *handle, * as writing the quota to disk may need the lock as well. */ DQUOT_INIT(inode); + ext3_xattr_delete_inode(handle, inode); DQUOT_FREE_INODE(inode); DQUOT_DROP(inode); @@ -509,6 +511,10 @@ repeat: inode->u.ext3_i.i_state = EXT3_STATE_NEW; err = ext3_mark_inode_dirty(handle, inode); if (err) goto fail; + +#ifdef CONFIG_EXT3_FS_XATTR + init_rwsem(&inode->u.ext3_i.xattr_sem); +#endif unlock_super (sb); if(DQUOT_ALLOC_INODE(inode)) { Index: linux-2.4.29/fs/ext3/inode.c =================================================================== --- linux-2.4.29.orig/fs/ext3/inode.c +++ linux-2.4.29/fs/ext3/inode.c @@ -60,7 +60,7 @@ static inline int ext3_inode_is_fast_sym * still needs to be revoked. */ -static int ext3_forget(handle_t *handle, int is_metadata, +int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode, struct buffer_head *bh, int blocknr) { @@ -191,9 +191,7 @@ void ext3_delete_inode (struct inode * i { handle_t *handle; - if (is_bad_inode(inode) || - inode->i_ino == EXT3_ACL_IDX_INO || - inode->i_ino == EXT3_ACL_DATA_INO) + if (is_bad_inode(inode)) goto no_delete; lock_kernel(); @@ -2032,8 +2030,6 @@ int ext3_get_inode_loc (struct inode *in struct ext3_group_desc * gdp; if ((inode->i_ino != EXT3_ROOT_INO && - inode->i_ino != EXT3_ACL_IDX_INO && - inode->i_ino != EXT3_ACL_DATA_INO && inode->i_ino != EXT3_JOURNAL_INO && inode->i_ino < EXT3_FIRST_INO(inode->i_sb)) || inode->i_ino > le32_to_cpu( @@ -2174,10 +2170,7 @@ void ext3_read_inode(struct inode * inod inode->u.ext3_i.i_data[block] = iloc.raw_inode->i_block[block]; INIT_LIST_HEAD(&inode->u.ext3_i.i_orphan); - if (inode->i_ino == EXT3_ACL_IDX_INO || - inode->i_ino == EXT3_ACL_DATA_INO) - /* Nothing to do */ ; - else if (S_ISREG(inode->i_mode)) { + if (S_ISREG(inode->i_mode)) { inode->i_op = &ext3_file_inode_operations; inode->i_fop = &ext3_file_operations; inode->i_mapping->a_ops = &ext3_aops; @@ -2188,14 +2181,19 @@ void ext3_read_inode(struct inode * inod if (ext3_inode_is_fast_symlink(inode)) inode->i_op = &ext3_fast_symlink_inode_operations; else { - inode->i_op = &page_symlink_inode_operations; + inode->i_op = &ext3_symlink_inode_operations; inode->i_mapping->a_ops = &ext3_aops; } - } else + } else { + inode->i_op = &ext3_special_inode_operations; init_special_inode(inode, inode->i_mode, le32_to_cpu(iloc.raw_inode->i_block[0])); + } brelse(iloc.bh); ext3_set_inode_flags(inode); +#ifdef CONFIG_EXT3_FS_XATTR + init_rwsem(&inode->u.ext3_i.xattr_sem); +#endif return; bad_inode: Index: linux-2.4.29/fs/ext3/namei.c =================================================================== --- linux-2.4.29.orig/fs/ext3/namei.c +++ linux-2.4.29/fs/ext3/namei.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include @@ -515,7 +516,7 @@ static int ext3_mkdir(struct inode * dir if (IS_SYNC(dir)) handle->h_sync = 1; - inode = ext3_new_inode (handle, dir, S_IFDIR); + inode = ext3_new_inode (handle, dir, S_IFDIR | mode); err = PTR_ERR(inode); if (IS_ERR(inode)) goto out_stop; @@ -523,7 +524,6 @@ static int ext3_mkdir(struct inode * dir inode->i_op = &ext3_dir_inode_operations; inode->i_fop = &ext3_dir_operations; inode->i_size = inode->u.ext3_i.i_disksize = inode->i_sb->s_blocksize; - inode->i_blocks = 0; dir_block = ext3_bread (handle, inode, 0, 1, &err); if (!dir_block) { inode->i_nlink--; /* is this nlink == 0? */ @@ -550,9 +550,6 @@ static int ext3_mkdir(struct inode * dir BUFFER_TRACE(dir_block, "call ext3_journal_dirty_metadata"); ext3_journal_dirty_metadata(handle, dir_block); brelse (dir_block); - inode->i_mode = S_IFDIR | mode; - if (dir->i_mode & S_ISGID) - inode->i_mode |= S_ISGID; ext3_mark_inode_dirty(handle, inode); err = ext3_add_entry (handle, dentry, inode); if (err) @@ -918,7 +915,7 @@ static int ext3_symlink (struct inode * goto out_stop; if (l > sizeof (inode->u.ext3_i.i_data)) { - inode->i_op = &page_symlink_inode_operations; + inode->i_op = &ext3_symlink_inode_operations; inode->i_mapping->a_ops = &ext3_aops; /* * block_symlink() calls back into ext3_prepare/commit_write. @@ -1121,4 +1118,16 @@ struct inode_operations ext3_dir_inode_o rmdir: ext3_rmdir, /* BKL held */ mknod: ext3_mknod, /* BKL held */ rename: ext3_rename, /* BKL held */ + setxattr: ext3_setxattr, /* BKL held */ + getxattr: ext3_getxattr, /* BKL held */ + listxattr: ext3_listxattr, /* BKL held */ + removexattr: ext3_removexattr, /* BKL held */ }; + +struct inode_operations ext3_special_inode_operations = { + setxattr: ext3_setxattr, /* BKL held */ + getxattr: ext3_getxattr, /* BKL held */ + listxattr: ext3_listxattr, /* BKL held */ + removexattr: ext3_removexattr, /* BKL held */ +}; + Index: linux-2.4.29/fs/ext3/super.c =================================================================== --- linux-2.4.29.orig/fs/ext3/super.c +++ linux-2.4.29/fs/ext3/super.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -406,6 +407,7 @@ void ext3_put_super (struct super_block kdev_t j_dev = sbi->s_journal->j_dev; int i; + ext3_xattr_put_super(sb); journal_destroy(sbi->s_journal); if (!(sb->s_flags & MS_RDONLY)) { EXT3_CLEAR_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); @@ -504,6 +506,7 @@ static int parse_options (char * options int is_remount) { unsigned long *mount_options = &sbi->s_mount_opt; + uid_t *resuid = &sbi->s_resuid; gid_t *resgid = &sbi->s_resgid; char * this_char; @@ -516,6 +519,13 @@ static int parse_options (char * options this_char = strtok (NULL, ",")) { if ((value = strchr (this_char, '=')) != NULL) *value++ = 0; +#ifdef CONFIG_EXT3_FS_XATTR_USER + if (!strcmp (this_char, "user_xattr")) + set_opt (*mount_options, XATTR_USER); + else if (!strcmp (this_char, "nouser_xattr")) + clear_opt (*mount_options, XATTR_USER); + else +#endif if (!strcmp (this_char, "bsddf")) clear_opt (*mount_options, MINIX_DF); else if (!strcmp (this_char, "nouid32")) { @@ -951,6 +961,12 @@ struct super_block * ext3_read_super (st sbi->s_mount_opt = 0; sbi->s_resuid = EXT3_DEF_RESUID; sbi->s_resgid = EXT3_DEF_RESGID; + + /* Default extended attribute flags */ +#ifdef CONFIG_EXT3_FS_XATTR_USER + /* set_opt(sbi->s_mount_opt, XATTR_USER); */ +#endif + if (!parse_options ((char *) data, &sb_block, sbi, &journal_inum, 0)) { sb->s_dev = 0; goto out_fail; @@ -1832,16 +1848,28 @@ static DECLARE_FSTYPE_DEV(ext3_fs_type, static int __init init_ext3_fs(void) { + int error; + #ifdef CONFIG_QUOTA init_dquot_operations(&ext3_qops); old_write_dquot = ext3_qops.write_dquot; ext3_qops.write_dquot = ext3_write_dquot; #endif - return register_filesystem(&ext3_fs_type); + error = init_ext3_xattr(); + if (error) + return error; + error = init_ext3_xattr_user(); + if (error) + return error; + error = init_ext3_xattr_trusted(); + if (error) + return error; + return register_filesystem(&ext3_fs_type); } static void __exit exit_ext3_fs(void) { + exit_ext3_xattr(); unregister_filesystem(&ext3_fs_type); } Index: linux-2.4.29/fs/ext3/symlink.c =================================================================== --- linux-2.4.29.orig/fs/ext3/symlink.c +++ linux-2.4.29/fs/ext3/symlink.c @@ -20,6 +20,7 @@ #include #include #include +#include static int ext3_readlink(struct dentry *dentry, char *buffer, int buflen) { @@ -33,7 +34,20 @@ static int ext3_follow_link(struct dentr return vfs_follow_link(nd, s); } +struct inode_operations ext3_symlink_inode_operations = { + readlink: page_readlink, /* BKL not held. Don't need */ + follow_link: page_follow_link, /* BKL not held. Don't need */ + setxattr: ext3_setxattr, /* BKL held */ + getxattr: ext3_getxattr, /* BKL held */ + listxattr: ext3_listxattr, /* BKL held */ + removexattr: ext3_removexattr, /* BKL held */ +}; + struct inode_operations ext3_fast_symlink_inode_operations = { readlink: ext3_readlink, /* BKL not held. Don't need */ follow_link: ext3_follow_link, /* BKL not held. Don't need */ + setxattr: ext3_setxattr, /* BKL held */ + getxattr: ext3_getxattr, /* BKL held */ + listxattr: ext3_listxattr, /* BKL held */ + removexattr: ext3_removexattr, /* BKL held */ }; Index: linux-2.4.29/fs/ext3/xattr.c =================================================================== --- /dev/null +++ linux-2.4.29/fs/ext3/xattr.c @@ -0,0 +1,1188 @@ +/* + * linux/fs/ext3/xattr.c + * + * Copyright (C) 2001-2003 by Andreas Gruenbacher, + * + * Fix by Harrison Xing . + * Ext3 code with a lot of help from Eric Jarman . + * Extended attributes for symlinks and special files added per + * suggestion of Luka Renko . + */ + +/* + * Extended attributes are stored on disk blocks allocated outside of + * any inode. The i_file_acl field is then made to point to this allocated + * block. If all extended attributes of an inode are identical, these + * inodes may share the same extended attribute block. Such situations + * are automatically detected by keeping a cache of recent attribute block + * numbers and hashes over the block's contents in memory. + * + * + * Extended attribute block layout: + * + * +------------------+ + * | header | + * | entry 1 | | + * | entry 2 | | growing downwards + * | entry 3 | v + * | four null bytes | + * | . . . | + * | value 1 | ^ + * | value 3 | | growing upwards + * | value 2 | | + * +------------------+ + * + * The block header is followed by multiple entry descriptors. These entry + * descriptors are variable in size, and alligned to EXT3_XATTR_PAD + * byte boundaries. The entry descriptors are sorted by attribute name, + * so that two extended attribute blocks can be compared efficiently. + * + * Attribute values are aligned to the end of the block, stored in + * no specific order. They are also padded to EXT3_XATTR_PAD byte + * boundaries. No additional gaps are left between them. + * + * Locking strategy + * ---------------- + * EXT3_I(inode)->i_file_acl is protected by EXT3_I(inode)->xattr_sem. + * EA blocks are only changed if they are exclusive to an inode, so + * holding xattr_sem also means that nothing but the EA block's reference + * count will change. Multiple writers to an EA block are synchronized + * by the bh lock. No more than a single bh lock is held at any time, + * which avoids deadlocks. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define HDR(bh) ((struct ext3_xattr_header *)((bh)->b_data)) +#define ENTRY(ptr) ((struct ext3_xattr_entry *)(ptr)) +#define FIRST_ENTRY(bh) ENTRY(HDR(bh)+1) +#define IS_LAST_ENTRY(entry) (*(__u32 *)(entry) == 0) + +#ifdef EXT3_XATTR_DEBUG +# define ea_idebug(inode, f...) do { \ + printk(KERN_DEBUG "inode %s:%ld: ", \ + kdevname(inode->i_dev), inode->i_ino); \ + printk(f); \ + printk("\n"); \ + } while (0) +# define ea_bdebug(bh, f...) do { \ + printk(KERN_DEBUG "block %s:%ld: ", \ + kdevname(bh->b_dev), bh->b_blocknr); \ + printk(f); \ + printk("\n"); \ + } while (0) +#else +# define ea_idebug(f...) +# define ea_bdebug(f...) +#endif + +static int ext3_xattr_set_handle2(handle_t *, struct inode *, + struct buffer_head *, + struct ext3_xattr_header *); + +#ifdef CONFIG_EXT3_FS_XATTR_SHARING + +static int ext3_xattr_cache_insert(struct buffer_head *); +static struct buffer_head *ext3_xattr_cache_find(struct inode *, + struct ext3_xattr_header *, + struct mb_cache_entry **); +static void ext3_xattr_rehash(struct ext3_xattr_header *, + struct ext3_xattr_entry *); + +static struct mb_cache *ext3_xattr_cache; + +#else +# define ext3_xattr_cache_insert(bh) 0 +# define ext3_xattr_cache_find(inode, header, pce) NULL +# define ext3_xattr_rehash(header, entry) while(0) {} +#endif + +struct ext3_xattr_handler *ext3_xattr_handlers[EXT3_XATTR_INDEX_MAX]; +rwlock_t ext3_handler_lock = RW_LOCK_UNLOCKED; + +int +ext3_xattr_register(int name_index, struct ext3_xattr_handler *handler) +{ + int error = -EINVAL; + + if (name_index > 0 && name_index <= EXT3_XATTR_INDEX_MAX) { + write_lock(&ext3_handler_lock); + if (!ext3_xattr_handlers[name_index-1]) { + ext3_xattr_handlers[name_index-1] = handler; + error = 0; + } + write_unlock(&ext3_handler_lock); + } + return error; +} + +void +ext3_xattr_unregister(int name_index, struct ext3_xattr_handler *handler) +{ + if (name_index > 0 || name_index <= EXT3_XATTR_INDEX_MAX) { + write_lock(&ext3_handler_lock); + ext3_xattr_handlers[name_index-1] = NULL; + write_unlock(&ext3_handler_lock); + } +} + +static inline const char * +strcmp_prefix(const char *a, const char *a_prefix) +{ + while (*a_prefix && *a == *a_prefix) { + a++; + a_prefix++; + } + return *a_prefix ? NULL : a; +} + +/* + * Decode the extended attribute name, and translate it into + * the name_index and name suffix. + */ +static inline struct ext3_xattr_handler * +ext3_xattr_resolve_name(const char **name) +{ + struct ext3_xattr_handler *handler = NULL; + int i; + + if (!*name) + return NULL; + read_lock(&ext3_handler_lock); + for (i=0; iprefix); + if (n) { + handler = ext3_xattr_handlers[i]; + *name = n; + break; + } + } + } + read_unlock(&ext3_handler_lock); + return handler; +} + +static inline struct ext3_xattr_handler * +ext3_xattr_handler(int name_index) +{ + struct ext3_xattr_handler *handler = NULL; + if (name_index > 0 && name_index <= EXT3_XATTR_INDEX_MAX) { + read_lock(&ext3_handler_lock); + handler = ext3_xattr_handlers[name_index-1]; + read_unlock(&ext3_handler_lock); + } + return handler; +} + +/* + * Inode operation getxattr() + * + * dentry->d_inode->i_sem: don't care + * BKL: held + */ +ssize_t +ext3_getxattr(struct dentry *dentry, const char *name, + void *buffer, size_t size) +{ + struct ext3_xattr_handler *handler; + struct inode *inode = dentry->d_inode; + + handler = ext3_xattr_resolve_name(&name); + if (!handler) + return -EOPNOTSUPP; + return handler->get(inode, name, buffer, size); +} + +/* + * Inode operation listxattr() + * + * dentry->d_inode->i_sem: don't care + * BKL: held + */ +ssize_t +ext3_listxattr(struct dentry *dentry, char *buffer, size_t size) +{ + return ext3_xattr_list(dentry->d_inode, buffer, size); +} + +/* + * Inode operation setxattr() + * + * dentry->d_inode->i_sem: down + * BKL: held + */ +int +ext3_setxattr(struct dentry *dentry, const char *name, void *value, + size_t size, int flags) +{ + struct ext3_xattr_handler *handler; + struct inode *inode = dentry->d_inode; + + if (size == 0) + value = ""; /* empty EA, do not remove */ + handler = ext3_xattr_resolve_name(&name); + if (!handler) + return -EOPNOTSUPP; + return handler->set(inode, name, value, size, flags); +} + +/* + * Inode operation removexattr() + * + * dentry->d_inode->i_sem: down + * BKL: held + */ +int +ext3_removexattr(struct dentry *dentry, const char *name) +{ + struct ext3_xattr_handler *handler; + struct inode *inode = dentry->d_inode; + + handler = ext3_xattr_resolve_name(&name); + if (!handler) + return -EOPNOTSUPP; + return handler->set(inode, name, NULL, 0, XATTR_REPLACE); +} + +/* + * ext3_xattr_get() + * + * Copy an extended attribute into the buffer + * provided, or compute the buffer size required. + * Buffer is NULL to compute the size of the buffer required. + * + * Returns a negative error number on failure, or the number of bytes + * used / required on success. + */ +int +ext3_xattr_get(struct inode *inode, int name_index, const char *name, + void *buffer, size_t buffer_size) +{ + struct buffer_head *bh = NULL; + struct ext3_xattr_entry *entry; + size_t size, name_len; + char *end; + int error; + + ea_idebug(inode, "name=%d.%s, buffer=%p, buffer_size=%ld", + name_index, name, buffer, (long)buffer_size); + + if (name == NULL) + return -EINVAL; + down_read(&EXT3_I(inode)->xattr_sem); + error = -ENODATA; + if (!EXT3_I(inode)->i_file_acl) + goto cleanup; + ea_idebug(inode, "reading block %d", EXT3_I(inode)->i_file_acl); + bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl); + error = -EIO; + if (!bh) + goto cleanup; + ea_bdebug(bh, "b_count=%d, refcount=%d", + atomic_read(&(bh->b_count)), le32_to_cpu(HDR(bh)->h_refcount)); + end = bh->b_data + bh->b_size; + if (HDR(bh)->h_magic != cpu_to_le32(EXT3_XATTR_MAGIC) || + HDR(bh)->h_blocks != cpu_to_le32(1)) { +bad_block: ext3_error(inode->i_sb, "ext3_xattr_get", + "inode %ld: bad block %d", inode->i_ino, + EXT3_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + /* find named attribute */ + name_len = strlen(name); + + error = -ERANGE; + if (name_len > 255) + goto cleanup; + entry = FIRST_ENTRY(bh); + while (!IS_LAST_ENTRY(entry)) { + struct ext3_xattr_entry *next = + EXT3_XATTR_NEXT(entry); + if ((char *)next >= end) + goto bad_block; + if (name_index == entry->e_name_index && + name_len == entry->e_name_len && + memcmp(name, entry->e_name, name_len) == 0) + goto found; + entry = next; + } + /* Check the remaining name entries */ + while (!IS_LAST_ENTRY(entry)) { + struct ext3_xattr_entry *next = + EXT3_XATTR_NEXT(entry); + if ((char *)next >= end) + goto bad_block; + entry = next; + } + if (ext3_xattr_cache_insert(bh)) + ea_idebug(inode, "cache insert failed"); + error = -ENODATA; + goto cleanup; +found: + /* check the buffer size */ + if (entry->e_value_block != 0) + goto bad_block; + size = le32_to_cpu(entry->e_value_size); + if (size > inode->i_sb->s_blocksize || + le16_to_cpu(entry->e_value_offs) + size > inode->i_sb->s_blocksize) + goto bad_block; + + if (ext3_xattr_cache_insert(bh)) + ea_idebug(inode, "cache insert failed"); + if (buffer) { + error = -ERANGE; + if (size > buffer_size) + goto cleanup; + /* return value of attribute */ + memcpy(buffer, bh->b_data + le16_to_cpu(entry->e_value_offs), + size); + } + error = size; + +cleanup: + brelse(bh); + up_read(&EXT3_I(inode)->xattr_sem); + + return error; +} + +/* + * ext3_xattr_list() + * + * Copy a list of attribute names into the buffer + * provided, or compute the buffer size required. + * Buffer is NULL to compute the size of the buffer required. + * + * Returns a negative error number on failure, or the number of bytes + * used / required on success. + */ +int +ext3_xattr_list(struct inode *inode, char *buffer, size_t buffer_size) +{ + struct buffer_head *bh = NULL; + struct ext3_xattr_entry *entry; + size_t size = 0; + char *buf, *end; + int error; + + ea_idebug(inode, "buffer=%p, buffer_size=%ld", + buffer, (long)buffer_size); + + down_read(&EXT3_I(inode)->xattr_sem); + error = 0; + if (!EXT3_I(inode)->i_file_acl) + goto cleanup; + ea_idebug(inode, "reading block %d", EXT3_I(inode)->i_file_acl); + bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl); + error = -EIO; + if (!bh) + goto cleanup; + ea_bdebug(bh, "b_count=%d, refcount=%d", + atomic_read(&(bh->b_count)), le32_to_cpu(HDR(bh)->h_refcount)); + end = bh->b_data + bh->b_size; + if (HDR(bh)->h_magic != cpu_to_le32(EXT3_XATTR_MAGIC) || + HDR(bh)->h_blocks != cpu_to_le32(1)) { +bad_block: ext3_error(inode->i_sb, "ext3_xattr_list", + "inode %ld: bad block %d", inode->i_ino, + EXT3_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + /* compute the size required for the list of attribute names */ + for (entry = FIRST_ENTRY(bh); !IS_LAST_ENTRY(entry); + entry = EXT3_XATTR_NEXT(entry)) { + struct ext3_xattr_handler *handler; + struct ext3_xattr_entry *next = + EXT3_XATTR_NEXT(entry); + if ((char *)next >= end) + goto bad_block; + + handler = ext3_xattr_handler(entry->e_name_index); + if (handler) + size += handler->list(NULL, inode, entry->e_name, + entry->e_name_len); + } + + if (ext3_xattr_cache_insert(bh)) + ea_idebug(inode, "cache insert failed"); + if (!buffer) { + error = size; + goto cleanup; + } else { + error = -ERANGE; + if (size > buffer_size) + goto cleanup; + } + + /* list the attribute names */ + buf = buffer; + for (entry = FIRST_ENTRY(bh); !IS_LAST_ENTRY(entry); + entry = EXT3_XATTR_NEXT(entry)) { + struct ext3_xattr_handler *handler; + + handler = ext3_xattr_handler(entry->e_name_index); + if (handler) + buf += handler->list(buf, inode, entry->e_name, + entry->e_name_len); + } + error = size; + +cleanup: + brelse(bh); + up_read(&EXT3_I(inode)->xattr_sem); + + return error; +} + +/* + * If the EXT3_FEATURE_COMPAT_EXT_ATTR feature of this file system is + * not set, set it. + */ +static void ext3_xattr_update_super_block(handle_t *handle, + struct super_block *sb) +{ + if (EXT3_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_EXT_ATTR)) + return; + + lock_super(sb); + if (ext3_journal_get_write_access(handle, EXT3_SB(sb)->s_sbh) == 0) { + EXT3_SB(sb)->s_es->s_feature_compat |= + cpu_to_le32(EXT3_FEATURE_COMPAT_EXT_ATTR); + sb->s_dirt = 1; + ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); + } + unlock_super(sb); +} + +/* + * Release the xattr block BH: If the reference count is > 1, decrement + * it; otherwise free the block. + */ +static void +ext3_xattr_release_block(handle_t *handle, struct inode *inode, + struct buffer_head *bh) +{ + struct mb_cache_entry *ce; + + ce = mb_cache_entry_get(ext3_xattr_cache, bh->b_dev, bh->b_blocknr); + if (HDR(bh)->h_refcount == cpu_to_le32(1)) { + ea_bdebug(bh, "freeing"); + if (ce) + mb_cache_entry_free(ce); + ext3_free_blocks(handle, inode, bh->b_blocknr, 1); + get_bh(bh); + ext3_forget(handle, 1, inode, bh, bh->b_blocknr); + } else { + if (ext3_journal_get_write_access(handle, bh) == 0) { + lock_buffer(bh); + HDR(bh)->h_refcount = cpu_to_le32( + le32_to_cpu(HDR(bh)->h_refcount) - 1); + ext3_journal_dirty_metadata(handle, bh); + if (IS_SYNC(inode)) + handle->h_sync = 1; + DQUOT_FREE_BLOCK(inode, 1); + unlock_buffer(bh); + ea_bdebug(bh, "releasing; refcount now=%d", + le32_to_cpu(HDR(bh)->h_refcount)); + } + if (ce) + mb_cache_entry_release(ce); + } +} + +/* + * ext3_xattr_set_handle() + * + * Create, replace or remove an extended attribute for this inode. Buffer + * is NULL to remove an existing extended attribute, and non-NULL to + * either replace an existing extended attribute, or create a new extended + * attribute. The flags XATTR_REPLACE and XATTR_CREATE + * specify that an extended attribute must exist and must not exist + * previous to the call, respectively. + * + * Returns 0, or a negative error number on failure. + */ +int +ext3_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, + const char *name, const void *value, size_t value_len, + int flags) +{ + struct super_block *sb = inode->i_sb; + struct buffer_head *bh = NULL; + struct ext3_xattr_header *header = NULL; + struct ext3_xattr_entry *here, *last; + size_t name_len, free, min_offs = sb->s_blocksize; + int not_found = 1, error; + char *end; + + /* + * header -- Points either into bh, or to a temporarily + * allocated buffer. + * here -- The named entry found, or the place for inserting, within + * the block pointed to by header. + * last -- Points right after the last named entry within the block + * pointed to by header. + * min_offs -- The offset of the first value (values are aligned + * towards the end of the block). + * end -- Points right after the block pointed to by header. + */ + + ea_idebug(inode, "name=%d.%s, value=%p, value_len=%ld", + name_index, name, value, (long)value_len); + + if (IS_RDONLY(inode)) + return -EROFS; + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + return -EPERM; + if (value == NULL) + value_len = 0; + if (name == NULL) + return -EINVAL; + name_len = strlen(name); + if (name_len > 255 || value_len > sb->s_blocksize) + return -ERANGE; + down_write(&EXT3_I(inode)->xattr_sem); + if (EXT3_I(inode)->i_file_acl) { + /* The inode already has an extended attribute block. */ + bh = sb_bread(sb, EXT3_I(inode)->i_file_acl); + error = -EIO; + if (!bh) + goto cleanup; + ea_bdebug(bh, "b_count=%d, refcount=%d", + atomic_read(&(bh->b_count)), + le32_to_cpu(HDR(bh)->h_refcount)); + header = HDR(bh); + end = bh->b_data + bh->b_size; + if (header->h_magic != cpu_to_le32(EXT3_XATTR_MAGIC) || + header->h_blocks != cpu_to_le32(1)) { +bad_block: ext3_error(sb, "ext3_xattr_set", + "inode %ld: bad block %d", inode->i_ino, + EXT3_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + /* Find the named attribute. */ + here = FIRST_ENTRY(bh); + while (!IS_LAST_ENTRY(here)) { + struct ext3_xattr_entry *next = EXT3_XATTR_NEXT(here); + if ((char *)next >= end) + goto bad_block; + if (!here->e_value_block && here->e_value_size) { + size_t offs = le16_to_cpu(here->e_value_offs); + if (offs < min_offs) + min_offs = offs; + } + not_found = name_index - here->e_name_index; + if (!not_found) + not_found = name_len - here->e_name_len; + if (!not_found) + not_found = memcmp(name, here->e_name,name_len); + if (not_found <= 0) + break; + here = next; + } + last = here; + /* We still need to compute min_offs and last. */ + while (!IS_LAST_ENTRY(last)) { + struct ext3_xattr_entry *next = EXT3_XATTR_NEXT(last); + if ((char *)next >= end) + goto bad_block; + if (!last->e_value_block && last->e_value_size) { + size_t offs = le16_to_cpu(last->e_value_offs); + if (offs < min_offs) + min_offs = offs; + } + last = next; + } + + /* Check whether we have enough space left. */ + free = min_offs - ((char*)last - (char*)header) - sizeof(__u32); + } else { + /* We will use a new extended attribute block. */ + free = sb->s_blocksize - + sizeof(struct ext3_xattr_header) - sizeof(__u32); + here = last = NULL; /* avoid gcc uninitialized warning. */ + } + + if (not_found) { + /* Request to remove a nonexistent attribute? */ + error = -ENODATA; + if (flags & XATTR_REPLACE) + goto cleanup; + error = 0; + if (value == NULL) + goto cleanup; + } else { + /* Request to create an existing attribute? */ + error = -EEXIST; + if (flags & XATTR_CREATE) + goto cleanup; + if (!here->e_value_block && here->e_value_size) { + size_t size = le32_to_cpu(here->e_value_size); + + if (le16_to_cpu(here->e_value_offs) + size > + sb->s_blocksize || size > sb->s_blocksize) + goto bad_block; + free += EXT3_XATTR_SIZE(size); + } + free += EXT3_XATTR_LEN(name_len); + } + error = -ENOSPC; + if (free < EXT3_XATTR_LEN(name_len) + EXT3_XATTR_SIZE(value_len)) + goto cleanup; + + /* Here we know that we can set the new attribute. */ + + if (header) { + struct mb_cache_entry *ce; + + ce = mb_cache_entry_get(ext3_xattr_cache, bh->b_dev, + bh->b_blocknr); + if (header->h_refcount == cpu_to_le32(1)) { + if (ce) + mb_cache_entry_free(ce); + ea_bdebug(bh, "modifying in-place"); + error = ext3_journal_get_write_access(handle, bh); + if (error) + goto cleanup; + lock_buffer(bh); + /* keep the buffer locked while modifying it. */ + } else { + int offset; + + if (ce) + mb_cache_entry_release(ce); + ea_bdebug(bh, "cloning"); + header = kmalloc(bh->b_size, GFP_KERNEL); + error = -ENOMEM; + if (header == NULL) + goto cleanup; + memcpy(header, HDR(bh), bh->b_size); + header->h_refcount = cpu_to_le32(1); + offset = (char *)here - bh->b_data; + here = ENTRY((char *)header + offset); + offset = (char *)last - bh->b_data; + last = ENTRY((char *)header + offset); + } + } else { + /* Allocate a buffer where we construct the new block. */ + header = kmalloc(sb->s_blocksize, GFP_KERNEL); + error = -ENOMEM; + if (header == NULL) + goto cleanup; + memset(header, 0, sb->s_blocksize); + end = (char *)header + sb->s_blocksize; + header->h_magic = cpu_to_le32(EXT3_XATTR_MAGIC); + header->h_blocks = header->h_refcount = cpu_to_le32(1); + last = here = ENTRY(header+1); + } + + /* Iff we are modifying the block in-place, bh is locked here. */ + + if (not_found) { + /* Insert the new name. */ + int size = EXT3_XATTR_LEN(name_len); + int rest = (char *)last - (char *)here; + memmove((char *)here + size, here, rest); + memset(here, 0, size); + here->e_name_index = name_index; + here->e_name_len = name_len; + memcpy(here->e_name, name, name_len); + } else { + if (!here->e_value_block && here->e_value_size) { + char *first_val = (char *)header + min_offs; + int offs = le16_to_cpu(here->e_value_offs); + char *val = (char *)header + offs; + size_t size = EXT3_XATTR_SIZE( + le32_to_cpu(here->e_value_size)); + + if (size == EXT3_XATTR_SIZE(value_len)) { + /* The old and the new value have the same + size. Just replace. */ + here->e_value_size = cpu_to_le32(value_len); + memset(val + size - EXT3_XATTR_PAD, 0, + EXT3_XATTR_PAD); /* Clear pad bytes. */ + memcpy(val, value, value_len); + goto skip_replace; + } + + /* Remove the old value. */ + memmove(first_val + size, first_val, val - first_val); + memset(first_val, 0, size); + here->e_value_offs = 0; + min_offs += size; + + /* Adjust all value offsets. */ + last = ENTRY(header+1); + while (!IS_LAST_ENTRY(last)) { + int o = le16_to_cpu(last->e_value_offs); + if (!last->e_value_block && o < offs) + last->e_value_offs = + cpu_to_le16(o + size); + last = EXT3_XATTR_NEXT(last); + } + } + if (value == NULL) { + /* Remove the old name. */ + int size = EXT3_XATTR_LEN(name_len); + last = ENTRY((char *)last - size); + memmove(here, (char*)here + size, + (char*)last - (char*)here); + memset(last, 0, size); + } + } + + if (value != NULL) { + /* Insert the new value. */ + here->e_value_size = cpu_to_le32(value_len); + if (value_len) { + size_t size = EXT3_XATTR_SIZE(value_len); + char *val = (char *)header + min_offs - size; + here->e_value_offs = + cpu_to_le16((char *)val - (char *)header); + memset(val + size - EXT3_XATTR_PAD, 0, + EXT3_XATTR_PAD); /* Clear the pad bytes. */ + memcpy(val, value, value_len); + } + } + +skip_replace: + if (!IS_LAST_ENTRY(ENTRY(header+1))) + ext3_xattr_rehash(header, here); + if (bh && header == HDR(bh)) { + /* we were modifying in-place. */ + unlock_buffer(bh); + error = ext3_journal_dirty_metadata(handle, bh); + if (error) + goto cleanup; + } + error = ext3_xattr_set_handle2(handle, inode, bh, + IS_LAST_ENTRY(ENTRY(header+1)) ? + NULL : header); + +cleanup: + brelse(bh); + if (!(bh && header == HDR(bh))) + kfree(header); + up_write(&EXT3_I(inode)->xattr_sem); + + return error; +} + +/* + * Second half of ext3_xattr_set_handle(): Update the file system. + */ +static int +ext3_xattr_set_handle2(handle_t *handle, struct inode *inode, + struct buffer_head *old_bh, struct ext3_xattr_header *header) +{ + struct super_block *sb = inode->i_sb; + struct buffer_head *new_bh = NULL; + struct mb_cache_entry *ce = NULL; + int error; + + if (header) { + new_bh = ext3_xattr_cache_find(inode, header, &ce); + if (new_bh) { + /* We found an identical block in the cache. */ + if (new_bh == old_bh) + ea_bdebug(new_bh, "keeping"); + else { + /* The old block is released after updating + the inode. */ + error = -EDQUOT; + if (DQUOT_ALLOC_BLOCK(inode, 1)) + goto cleanup; + error = ext3_journal_get_write_access(handle, + new_bh); + if (error) + goto cleanup; + lock_buffer(new_bh); + HDR(new_bh)->h_refcount = cpu_to_le32(1 + + le32_to_cpu(HDR(new_bh)->h_refcount)); + ea_bdebug(new_bh, "reusing; refcount now=%d", + le32_to_cpu(HDR(new_bh)->h_refcount)); + unlock_buffer(new_bh); + error = ext3_journal_dirty_metadata(handle, + new_bh); + if (error) + goto cleanup; + } + mb_cache_entry_release(ce); + ce = NULL; + } else if (old_bh && header == HDR(old_bh)) { + /* We were modifying this block in-place. */ + ea_bdebug(bs->bh, "keeping this block"); + new_bh = old_bh; + get_bh(new_bh); + ext3_xattr_cache_insert(new_bh); + } else { + /* We need to allocate a new block */ + int goal = le32_to_cpu(EXT3_SB(inode->i_sb)->s_es-> + s_first_data_block) + + EXT3_I(inode)->i_block_group * + EXT3_BLOCKS_PER_GROUP(inode->i_sb); + /* How can we enforce the allocation? */ + int block = ext3_new_block(handle, inode, goal, 0, 0, + &error); + if (error) + goto cleanup; + ea_idebug(inode, "creating block %d", block); + + new_bh = sb_getblk(sb, block); + if (!new_bh) { +getblk_failed: ext3_free_blocks(handle, inode, block, 1); + error = -EIO; + goto cleanup; + } + lock_buffer(new_bh); + error = ext3_journal_get_create_access(handle, new_bh); + if (error) { + unlock_buffer(new_bh); + goto getblk_failed; + } + memcpy(new_bh->b_data, header, new_bh->b_size); + mark_buffer_uptodate(new_bh, 1); + unlock_buffer(new_bh); + ext3_xattr_cache_insert(new_bh); + + ext3_xattr_update_super_block(handle, sb); + error = ext3_journal_dirty_metadata(handle, new_bh); + if (error) + goto cleanup; + } + } + + /* Update the inode. */ + EXT3_I(inode)->i_file_acl = new_bh ? new_bh->b_blocknr : 0; + inode->i_ctime = CURRENT_TIME; + ext3_mark_inode_dirty(handle, inode); + if (IS_SYNC(inode)) + handle->h_sync = 1; + + error = 0; + if (old_bh && old_bh != new_bh) { + /* + * If there was an old block and we are no longer using it, + * release the old block. + */ + ext3_xattr_release_block(handle, inode, old_bh); + } + +cleanup: + if (ce) + mb_cache_entry_release(ce); + brelse(new_bh); + + return error; +} + +/* + * ext3_xattr_set() + * + * Like ext3_xattr_set_handle, but start from an inode. This extended + * attribute modification is a filesystem transaction by itself. + * + * Returns 0, or a negative error number on failure. + */ +int +ext3_xattr_set(struct inode *inode, int name_index, const char *name, + const void *value, size_t value_len, int flags) +{ + handle_t *handle; + int error, error2; + + lock_kernel(); + handle = ext3_journal_start(inode, EXT3_XATTR_TRANS_BLOCKS); + error = PTR_ERR(handle); + if (IS_ERR(handle)) + goto cleanup; + error = ext3_xattr_set_handle(handle, inode, name_index, name, + value, value_len, flags); + error2 = ext3_journal_stop(handle, inode); + if (!error) + error = error2; + +cleanup: + unlock_kernel(); + return error; +} + +/* + * ext3_xattr_delete_inode() + * + * Free extended attribute resources associated with this inode. This + * is called immediately before an inode is freed. + */ +void +ext3_xattr_delete_inode(handle_t *handle, struct inode *inode) +{ + struct buffer_head *bh = NULL; + + if (!EXT3_I(inode)->i_file_acl) + goto cleanup; + bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl); + if (!bh) { + ext3_error(inode->i_sb, "ext3_xattr_delete_inode", + "inode %ld: block %d read error", inode->i_ino, + EXT3_I(inode)->i_file_acl); + goto cleanup; + } + ext3_xattr_release_block(handle, inode, bh); + EXT3_I(inode)->i_file_acl = 0; + +cleanup: + brelse(bh); +} + +/* + * ext3_xattr_put_super() + * + * This is called when a file system is unmounted. + */ +void +ext3_xattr_put_super(struct super_block *sb) +{ +#ifdef CONFIG_EXT3_FS_XATTR_SHARING + mb_cache_shrink(ext3_xattr_cache, sb->s_dev); +#endif +} + +#ifdef CONFIG_EXT3_FS_XATTR_SHARING + +/* + * ext3_xattr_cache_insert() + * + * Create a new entry in the extended attribute cache, and insert + * it unless such an entry is already in the cache. + * + * Returns 0, or a negative error number on failure. + */ +static int +ext3_xattr_cache_insert(struct buffer_head *bh) +{ + __u32 hash = le32_to_cpu(HDR(bh)->h_hash); + struct mb_cache_entry *ce; + int error; + + ce = mb_cache_entry_alloc(ext3_xattr_cache); + if (!ce) + return -ENOMEM; + error = mb_cache_entry_insert(ce, bh->b_dev, bh->b_blocknr, &hash); + if (error) { + mb_cache_entry_free(ce); + if (error == -EBUSY) { + ea_bdebug(bh, "already in cache (%d cache entries)", + atomic_read(&ext3_xattr_cache->c_entry_count)); + error = 0; + } + } else { + ea_bdebug(bh, "inserting [%x] (%d cache entries)", (int)hash, + atomic_read(&ext3_xattr_cache->c_entry_count)); + mb_cache_entry_release(ce); + } + return error; +} + +/* + * ext3_xattr_cmp() + * + * Compare two extended attribute blocks for equality. + * + * Returns 0 if the blocks are equal, 1 if they differ, and + * a negative error number on errors. + */ +static int +ext3_xattr_cmp(struct ext3_xattr_header *header1, + struct ext3_xattr_header *header2) +{ + struct ext3_xattr_entry *entry1, *entry2; + + entry1 = ENTRY(header1+1); + entry2 = ENTRY(header2+1); + while (!IS_LAST_ENTRY(entry1)) { + if (IS_LAST_ENTRY(entry2)) + return 1; + if (entry1->e_hash != entry2->e_hash || + entry1->e_name_index != entry2->e_name_index || + entry1->e_name_len != entry2->e_name_len || + entry1->e_value_size != entry2->e_value_size || + memcmp(entry1->e_name, entry2->e_name, entry1->e_name_len)) + return 1; + if (entry1->e_value_block != 0 || entry2->e_value_block != 0) + return -EIO; + if (memcmp((char *)header1 + le16_to_cpu(entry1->e_value_offs), + (char *)header2 + le16_to_cpu(entry2->e_value_offs), + le32_to_cpu(entry1->e_value_size))) + return 1; + + entry1 = EXT3_XATTR_NEXT(entry1); + entry2 = EXT3_XATTR_NEXT(entry2); + } + if (!IS_LAST_ENTRY(entry2)) + return 1; + return 0; +} + +/* + * ext3_xattr_cache_find() + * + * Find an identical extended attribute block. + * + * Returns a pointer to the block found, or NULL if such a block was + * not found or an error occurred. + */ +static struct buffer_head * +ext3_xattr_cache_find(struct inode *inode, struct ext3_xattr_header *header, + struct mb_cache_entry **pce) +{ + __u32 hash = le32_to_cpu(header->h_hash); + struct mb_cache_entry *ce; + + if (!header->h_hash) + return NULL; /* never share */ + ea_idebug(inode, "looking for cached blocks [%x]", (int)hash); +again: + ce = mb_cache_entry_find_first(ext3_xattr_cache, 0, inode->i_dev, hash); + while (ce) { + struct buffer_head *bh; + + if (IS_ERR(ce)) { + if (PTR_ERR(ce) == -EAGAIN) + goto again; + break; + } + bh = sb_bread(inode->i_sb, ce->e_block); + if (!bh) { + ext3_error(inode->i_sb, "ext3_xattr_cache_find", + "inode %ld: block %ld read error", + inode->i_ino, ce->e_block); + } else if (le32_to_cpu(HDR(bh)->h_refcount) >= + EXT3_XATTR_REFCOUNT_MAX) { + ea_idebug(inode, "block %ld refcount %d>=%d", + (unsigned long) ce->e_block, + le32_to_cpu(HDR(bh)->h_refcount), + EXT3_XATTR_REFCOUNT_MAX); + } else if (ext3_xattr_cmp(header, HDR(bh)) == 0) { + *pce = ce; + return bh; + } + brelse(bh); + ce = mb_cache_entry_find_next(ce, 0, inode->i_dev, hash); + } + return NULL; +} + +#define NAME_HASH_SHIFT 5 +#define VALUE_HASH_SHIFT 16 + +/* + * ext3_xattr_hash_entry() + * + * Compute the hash of an extended attribute. + */ +static inline void ext3_xattr_hash_entry(struct ext3_xattr_header *header, + struct ext3_xattr_entry *entry) +{ + __u32 hash = 0; + char *name = entry->e_name; + int n; + + for (n=0; n < entry->e_name_len; n++) { + hash = (hash << NAME_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - NAME_HASH_SHIFT)) ^ + *name++; + } + + if (entry->e_value_block == 0 && entry->e_value_size != 0) { + __u32 *value = (__u32 *)((char *)header + + le16_to_cpu(entry->e_value_offs)); + for (n = (le32_to_cpu(entry->e_value_size) + + EXT3_XATTR_ROUND) >> EXT3_XATTR_PAD_BITS; n; n--) { + hash = (hash << VALUE_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - VALUE_HASH_SHIFT)) ^ + le32_to_cpu(*value++); + } + } + entry->e_hash = cpu_to_le32(hash); +} + +#undef NAME_HASH_SHIFT +#undef VALUE_HASH_SHIFT + +#define BLOCK_HASH_SHIFT 16 + +/* + * ext3_xattr_rehash() + * + * Re-compute the extended attribute hash value after an entry has changed. + */ +static void ext3_xattr_rehash(struct ext3_xattr_header *header, + struct ext3_xattr_entry *entry) +{ + struct ext3_xattr_entry *here; + __u32 hash = 0; + + ext3_xattr_hash_entry(header, entry); + here = ENTRY(header+1); + while (!IS_LAST_ENTRY(here)) { + if (!here->e_hash) { + /* Block is not shared if an entry's hash value == 0 */ + hash = 0; + break; + } + hash = (hash << BLOCK_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - BLOCK_HASH_SHIFT)) ^ + le32_to_cpu(here->e_hash); + here = EXT3_XATTR_NEXT(here); + } + header->h_hash = cpu_to_le32(hash); +} + +#undef BLOCK_HASH_SHIFT + +int __init +init_ext3_xattr(void) +{ + ext3_xattr_cache = mb_cache_create("ext3_xattr", NULL, + sizeof(struct mb_cache_entry) + + sizeof(((struct mb_cache_entry *) 0)->e_indexes[0]), 1, 61); + if (!ext3_xattr_cache) + return -ENOMEM; + + return 0; +} + +void +exit_ext3_xattr(void) +{ + if (ext3_xattr_cache) + mb_cache_destroy(ext3_xattr_cache); + ext3_xattr_cache = NULL; +} + +#else /* CONFIG_EXT3_FS_XATTR_SHARING */ + +int __init +init_ext3_xattr(void) +{ + return 0; +} + +void +exit_ext3_xattr(void) +{ +} + +#endif /* CONFIG_EXT3_FS_XATTR_SHARING */ Index: linux-2.4.29/fs/ext3/xattr_trusted.c =================================================================== --- /dev/null +++ linux-2.4.29/fs/ext3/xattr_trusted.c @@ -0,0 +1,69 @@ +/* + * linux/fs/ext3/xattr_trusted.c + * Handler for trusted extended attributes. + * + * Copyright (C) 2001 by Andreas Gruenbacher, + */ + +#include +#include +#include +#include +#include + +#define XATTR_TRUSTED_PREFIX "trusted." + +static size_t +ext3_xattr_trusted_list(char *list, struct inode *inode, + const char *name, int name_len) +{ + const int prefix_len = sizeof(XATTR_TRUSTED_PREFIX)-1; + + if (!capable(CAP_SYS_ADMIN)) + return 0; + + if (list) { + memcpy(list, XATTR_TRUSTED_PREFIX, prefix_len); + memcpy(list+prefix_len, name, name_len); + list[prefix_len + name_len] = '\0'; + } + return prefix_len + name_len + 1; +} + +static int +ext3_xattr_trusted_get(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + return ext3_xattr_get(inode, EXT3_XATTR_INDEX_TRUSTED, name, + buffer, size); +} + +static int +ext3_xattr_trusted_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + return ext3_xattr_set(inode, EXT3_XATTR_INDEX_TRUSTED, name, + value, size, flags); +} + +struct ext3_xattr_handler ext3_xattr_trusted_handler = { + prefix: XATTR_TRUSTED_PREFIX, + list: ext3_xattr_trusted_list, + get: ext3_xattr_trusted_get, + set: ext3_xattr_trusted_set, +}; + +int __init +init_ext3_xattr_trusted(void) +{ + return ext3_xattr_register(EXT3_XATTR_INDEX_TRUSTED, + &ext3_xattr_trusted_handler); +} Index: linux-2.4.29/fs/ext3/xattr_user.c =================================================================== --- /dev/null +++ linux-2.4.29/fs/ext3/xattr_user.c @@ -0,0 +1,89 @@ +/* + * linux/fs/ext3/xattr_user.c + * Handler for extended user attributes. + * + * Copyright (C) 2001 by Andreas Gruenbacher, + */ + +#include +#include +#include +#include +#include +#include + +#ifdef CONFIG_EXT3_FS_POSIX_ACL +# include +#endif + +#define XATTR_USER_PREFIX "user." + +static size_t +ext3_xattr_user_list(char *list, struct inode *inode, + const char *name, int name_len) +{ + const int prefix_len = sizeof(XATTR_USER_PREFIX)-1; + + if (!test_opt(inode->i_sb, XATTR_USER)) + return 0; + + if (list) { + memcpy(list, XATTR_USER_PREFIX, prefix_len); + memcpy(list+prefix_len, name, name_len); + list[prefix_len + name_len] = '\0'; + } + return prefix_len + name_len + 1; +} + +static int +ext3_xattr_user_get(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + int error; + + if (strcmp(name, "") == 0) + return -EINVAL; + if (!test_opt(inode->i_sb, XATTR_USER)) + return -EOPNOTSUPP; + error = permission(inode, MAY_READ); + if (error) + return error; + + return ext3_xattr_get(inode, EXT3_XATTR_INDEX_USER, name, + buffer, size); +} + +static int +ext3_xattr_user_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + int error; + + if (strcmp(name, "") == 0) + return -EINVAL; + if ( !S_ISREG(inode->i_mode) && + (!S_ISDIR(inode->i_mode) || inode->i_mode & S_ISVTX)) + return -EPERM; + if (!test_opt(inode->i_sb, XATTR_USER)) + return -EOPNOTSUPP; + error = permission(inode, MAY_WRITE); + if (error) + return error; + + return ext3_xattr_set(inode, EXT3_XATTR_INDEX_USER, name, + value, size, flags); +} + +struct ext3_xattr_handler ext3_xattr_user_handler = { + prefix: XATTR_USER_PREFIX, + list: ext3_xattr_user_list, + get: ext3_xattr_user_get, + set: ext3_xattr_user_set, +}; + +int __init +init_ext3_xattr_user(void) +{ + return ext3_xattr_register(EXT3_XATTR_INDEX_USER, + &ext3_xattr_user_handler); +} Index: linux-2.4.29/fs/jfs/xattr.c =================================================================== --- linux-2.4.29.orig/fs/jfs/xattr.c +++ linux-2.4.29/fs/jfs/xattr.c @@ -892,12 +892,18 @@ ssize_t __jfs_getxattr(struct inode *ino ssize_t jfs_getxattr(struct dentry *dentry, const char *name, void *data, size_t buf_size) { - return __jfs_getxattr(dentry->d_inode, name, data, buf_size); + int err; + + down(&dentry->d_inode->i_sem); + err = __jfs_getxattr(dentry->d_inode, name, data, buf_size); + up(&dentry->d_inode->i_sem); + + return err; } -ssize_t jfs_listxattr(struct dentry * dentry, char *data, size_t buf_size) +static ssize_t __jfs_listxattr(struct inode *inode, char *data, + size_t buf_size) { - struct inode *inode = dentry->d_inode; char *buffer; ssize_t size = 0; int xattr_size; @@ -941,6 +947,17 @@ ssize_t jfs_listxattr(struct dentry * de return size; } +ssize_t jfs_listxattr(struct dentry * dentry, char *data, size_t buf_size) +{ + int err; + + down(&dentry->d_inode->i_sem); + err = __jfs_listxattr(dentry->d_inode, data, buf_size); + up(&dentry->d_inode->i_sem); + + return err; +} + int jfs_removexattr(struct dentry *dentry, const char *name) { return __jfs_setxattr(dentry->d_inode, name, 0, 0, XATTR_REPLACE); Index: linux-2.4.29/fs/mbcache.c =================================================================== --- /dev/null +++ linux-2.4.29/fs/mbcache.c @@ -0,0 +1,677 @@ +/* + * linux/fs/mbcache.c + * (C) 2001-2002 Andreas Gruenbacher, + */ + +/* + * Filesystem Meta Information Block Cache (mbcache) + * + * The mbcache caches blocks of block devices that need to be located + * by their device/block number, as well as by other criteria (such + * as the block's contents). + * + * There can only be one cache entry in a cache per device and block number. + * Additional indexes need not be unique in this sense. The number of + * additional indexes (=other criteria) can be hardwired at compile time + * or specified at cache create time. + * + * Each cache entry is of fixed size. An entry may be `valid' or `invalid' + * in the cache. A valid entry is in the main hash tables of the cache, + * and may also be in the lru list. An invalid entry is not in any hashes + * or lists. + * + * A valid cache entry is only in the lru list if no handles refer to it. + * Invalid cache entries will be freed when the last handle to the cache + * entry is released. Entries that cannot be freed immediately are put + * back on the lru list. + */ + +#include +#include + +#include +#include +#include +#include +#include +#include +#include + + +#ifdef MB_CACHE_DEBUG +# define mb_debug(f...) do { \ + printk(KERN_DEBUG f); \ + printk("\n"); \ + } while (0) +#define mb_assert(c) do { if (!(c)) \ + printk(KERN_ERR "assertion " #c " failed\n"); \ + } while(0) +#else +# define mb_debug(f...) do { } while(0) +# define mb_assert(c) do { } while(0) +#endif +#define mb_error(f...) do { \ + printk(KERN_ERR f); \ + printk("\n"); \ + } while(0) + +#define MB_CACHE_WRITER ((unsigned short)~0U >> 1) + +DECLARE_WAIT_QUEUE_HEAD(mb_cache_queue); + +MODULE_AUTHOR("Andreas Gruenbacher "); +MODULE_DESCRIPTION("Meta block cache (for extended attributes)"); +#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,0) +MODULE_LICENSE("GPL"); +#endif + +EXPORT_SYMBOL(mb_cache_create); +EXPORT_SYMBOL(mb_cache_shrink); +EXPORT_SYMBOL(mb_cache_destroy); +EXPORT_SYMBOL(mb_cache_entry_alloc); +EXPORT_SYMBOL(mb_cache_entry_insert); +EXPORT_SYMBOL(mb_cache_entry_release); +EXPORT_SYMBOL(mb_cache_entry_free); +EXPORT_SYMBOL(mb_cache_entry_get); +#if !defined(MB_CACHE_INDEXES_COUNT) || (MB_CACHE_INDEXES_COUNT > 0) +EXPORT_SYMBOL(mb_cache_entry_find_first); +EXPORT_SYMBOL(mb_cache_entry_find_next); +#endif + +struct mb_cache { + struct list_head c_cache_list; + const char *c_name; + struct mb_cache_op c_op; + atomic_t c_entry_count; + int c_bucket_count; +#ifndef MB_CACHE_INDEXES_COUNT + int c_indexes_count; +#endif + kmem_cache_t *c_entry_cache; + struct list_head *c_block_hash; + struct list_head *c_indexes_hash[0]; +}; + + +/* + * Global data: list of all mbcache's, lru list, and a spinlock for + * accessing cache data structures on SMP machines. The lru list is + * global across all mbcaches. + */ + +static LIST_HEAD(mb_cache_list); +static LIST_HEAD(mb_cache_lru_list); +static spinlock_t mb_cache_spinlock = SPIN_LOCK_UNLOCKED; + +static inline int +mb_cache_indexes(struct mb_cache *cache) +{ +#ifdef MB_CACHE_INDEXES_COUNT + return MB_CACHE_INDEXES_COUNT; +#else + return cache->c_indexes_count; +#endif +} + +/* + * What the mbcache registers as to get shrunk dynamically. + */ + +static void +mb_cache_memory_pressure(int priority, unsigned int gfp_mask); + +static struct cache_definition mb_cache_definition = { + "mb_cache", + mb_cache_memory_pressure +}; + + +static inline int +__mb_cache_entry_is_hashed(struct mb_cache_entry *ce) +{ + return !list_empty(&ce->e_block_list); +} + + +static inline void +__mb_cache_entry_unhash(struct mb_cache_entry *ce) +{ + int n; + + if (__mb_cache_entry_is_hashed(ce)) { + list_del_init(&ce->e_block_list); + for (n=0; ne_cache); n++) + list_del(&ce->e_indexes[n].o_list); + } +} + + +static inline void +__mb_cache_entry_forget(struct mb_cache_entry *ce, int gfp_mask) +{ + struct mb_cache *cache = ce->e_cache; + + mb_assert(!(ce->e_used || ce->e_queued)); + if (cache->c_op.free && cache->c_op.free(ce, gfp_mask)) { + /* free failed -- put back on the lru list + for freeing later. */ + spin_lock(&mb_cache_spinlock); + list_add(&ce->e_lru_list, &mb_cache_lru_list); + spin_unlock(&mb_cache_spinlock); + } else { + kmem_cache_free(cache->c_entry_cache, ce); + atomic_dec(&cache->c_entry_count); + } +} + + +static inline void +__mb_cache_entry_release_unlock(struct mb_cache_entry *ce) +{ + /* Wake up all processes queuing for this cache entry. */ + if (ce->e_queued) + wake_up_all(&mb_cache_queue); + if (ce->e_used >= MB_CACHE_WRITER) + ce->e_used -= MB_CACHE_WRITER; + ce->e_used--; + if (!(ce->e_used || ce->e_queued)) { + if (!__mb_cache_entry_is_hashed(ce)) { + spin_unlock(&mb_cache_spinlock); + __mb_cache_entry_forget(ce, GFP_KERNEL); + return; + } + mb_assert(list_empty(&ce->e_lru_list)); + list_add_tail(&ce->e_lru_list, &mb_cache_lru_list); + } + spin_unlock(&mb_cache_spinlock); +} + + +/* + * mb_cache_memory_pressure() memory pressure callback + * + * This function is called by the kernel memory management when memory + * gets low. + * + * @priority: Amount by which to shrink the cache (0 = highes priority) + * @gfp_mask: (ignored) + */ +static void +mb_cache_memory_pressure(int priority, unsigned int gfp_mask) +{ + LIST_HEAD(free_list); + struct list_head *l, *ltmp; + int count = 0; + + spin_lock(&mb_cache_spinlock); + list_for_each(l, &mb_cache_list) { + struct mb_cache *cache = + list_entry(l, struct mb_cache, c_cache_list); + mb_debug("cache %s (%d)", cache->c_name, + atomic_read(&cache->c_entry_count)); + count += atomic_read(&cache->c_entry_count); + } + mb_debug("trying to free %d of %d entries", + count / (priority ? priority : 1), count); + if (priority) + count /= priority; + while (count-- && !list_empty(&mb_cache_lru_list)) { + struct mb_cache_entry *ce = + list_entry(mb_cache_lru_list.next, + struct mb_cache_entry, e_lru_list); + list_move_tail(&ce->e_lru_list, &free_list); + __mb_cache_entry_unhash(ce); + } + spin_unlock(&mb_cache_spinlock); + list_for_each_safe(l, ltmp, &free_list) { + __mb_cache_entry_forget(list_entry(l, struct mb_cache_entry, + e_lru_list), gfp_mask); + } +} + + +/* + * mb_cache_create() create a new cache + * + * All entries in one cache are equal size. Cache entries may be from + * multiple devices. If this is the first mbcache created, registers + * the cache with kernel memory management. Returns NULL if no more + * memory was available. + * + * @name: name of the cache (informal) + * @cache_op: contains the callback called when freeing a cache entry + * @entry_size: The size of a cache entry, including + * struct mb_cache_entry + * @indexes_count: number of additional indexes in the cache. Must equal + * MB_CACHE_INDEXES_COUNT if the number of indexes is + * hardwired. + * @bucket_count: number of hash buckets + */ +struct mb_cache * +mb_cache_create(const char *name, struct mb_cache_op *cache_op, + size_t entry_size, int indexes_count, int bucket_count) +{ + int m=0, n; + struct mb_cache *cache = NULL; + + if(entry_size < sizeof(struct mb_cache_entry) + + indexes_count * sizeof(((struct mb_cache_entry *) 0)->e_indexes[0])) + return NULL; + + MOD_INC_USE_COUNT; + cache = kmalloc(sizeof(struct mb_cache) + + indexes_count * sizeof(struct list_head), GFP_KERNEL); + if (!cache) + goto fail; + cache->c_name = name; + cache->c_op.free = NULL; + if (cache_op) + cache->c_op.free = cache_op->free; + atomic_set(&cache->c_entry_count, 0); + cache->c_bucket_count = bucket_count; +#ifdef MB_CACHE_INDEXES_COUNT + mb_assert(indexes_count == MB_CACHE_INDEXES_COUNT); +#else + cache->c_indexes_count = indexes_count; +#endif + cache->c_block_hash = kmalloc(bucket_count * sizeof(struct list_head), + GFP_KERNEL); + if (!cache->c_block_hash) + goto fail; + for (n=0; nc_block_hash[n]); + for (m=0; mc_indexes_hash[m] = kmalloc(bucket_count * + sizeof(struct list_head), + GFP_KERNEL); + if (!cache->c_indexes_hash[m]) + goto fail; + for (n=0; nc_indexes_hash[m][n]); + } + cache->c_entry_cache = kmem_cache_create(name, entry_size, 0, + 0 /*SLAB_POISON | SLAB_RED_ZONE*/, NULL, NULL); + if (!cache->c_entry_cache) + goto fail; + + spin_lock(&mb_cache_spinlock); + list_add(&cache->c_cache_list, &mb_cache_list); + spin_unlock(&mb_cache_spinlock); + return cache; + +fail: + if (cache) { + while (--m >= 0) + kfree(cache->c_indexes_hash[m]); + if (cache->c_block_hash) + kfree(cache->c_block_hash); + kfree(cache); + } + MOD_DEC_USE_COUNT; + return NULL; +} + + +/* + * mb_cache_shrink() + * + * Removes all cache entires of a device from the cache. All cache entries + * currently in use cannot be freed, and thus remain in the cache. + * + * @cache: which cache to shrink + * @dev: which device's cache entries to shrink + */ +void +mb_cache_shrink(struct mb_cache *cache, kdev_t dev) +{ + LIST_HEAD(free_list); + struct list_head *l, *ltmp; + + spin_lock(&mb_cache_spinlock); + list_for_each_safe(l, ltmp, &mb_cache_lru_list) { + struct mb_cache_entry *ce = + list_entry(l, struct mb_cache_entry, e_lru_list); + if (ce->e_dev == dev) { + list_move_tail(&ce->e_lru_list, &free_list); + __mb_cache_entry_unhash(ce); + } + } + spin_unlock(&mb_cache_spinlock); + list_for_each_safe(l, ltmp, &free_list) { + __mb_cache_entry_forget(list_entry(l, struct mb_cache_entry, + e_lru_list), GFP_KERNEL); + } +} + + +/* + * mb_cache_destroy() + * + * Shrinks the cache to its minimum possible size (hopefully 0 entries), + * and then destroys it. If this was the last mbcache, un-registers the + * mbcache from kernel memory management. + */ +void +mb_cache_destroy(struct mb_cache *cache) +{ + LIST_HEAD(free_list); + struct list_head *l, *ltmp; + int n; + + spin_lock(&mb_cache_spinlock); + list_for_each_safe(l, ltmp, &mb_cache_lru_list) { + struct mb_cache_entry *ce = + list_entry(l, struct mb_cache_entry, e_lru_list); + if (ce->e_cache == cache) { + list_move_tail(&ce->e_lru_list, &free_list); + __mb_cache_entry_unhash(ce); + } + } + list_del(&cache->c_cache_list); + spin_unlock(&mb_cache_spinlock); + list_for_each_safe(l, ltmp, &free_list) { + __mb_cache_entry_forget(list_entry(l, struct mb_cache_entry, + e_lru_list), GFP_KERNEL); + } + + if (atomic_read(&cache->c_entry_count) > 0) { + mb_error("cache %s: %d orphaned entries", + cache->c_name, + atomic_read(&cache->c_entry_count)); + } + +#if (LINUX_VERSION_CODE < KERNEL_VERSION(2,3,0)) + /* We don't have kmem_cache_destroy() in 2.2.x */ + kmem_cache_shrink(cache->c_entry_cache); +#else + kmem_cache_destroy(cache->c_entry_cache); +#endif + for (n=0; n < mb_cache_indexes(cache); n++) + kfree(cache->c_indexes_hash[n]); + kfree(cache->c_block_hash); + kfree(cache); + + MOD_DEC_USE_COUNT; +} + + +/* + * mb_cache_entry_alloc() + * + * Allocates a new cache entry. The new entry will not be valid initially, + * and thus cannot be looked up yet. It should be filled with data, and + * then inserted into the cache using mb_cache_entry_insert(). Returns NULL + * if no more memory was available. + */ +struct mb_cache_entry * +mb_cache_entry_alloc(struct mb_cache *cache) +{ + struct mb_cache_entry *ce; + + atomic_inc(&cache->c_entry_count); + ce = kmem_cache_alloc(cache->c_entry_cache, GFP_KERNEL); + if (ce) { + INIT_LIST_HEAD(&ce->e_lru_list); + INIT_LIST_HEAD(&ce->e_block_list); + ce->e_cache = cache; + ce->e_used = 1 + MB_CACHE_WRITER; + ce->e_queued = 0; + } + return ce; +} + + +/* + * mb_cache_entry_insert() + * + * Inserts an entry that was allocated using mb_cache_entry_alloc() into + * the cache. After this, the cache entry can be looked up, but is not yet + * in the lru list as the caller still holds a handle to it. Returns 0 on + * success, or -EBUSY if a cache entry for that device + inode exists + * already (this may happen after a failed lookup, if another process has + * inserted the same cache entry in the meantime). + * + * @dev: device the cache entry belongs to + * @block: block number + * @keys: array of additional keys. There must be indexes_count entries + * in the array (as specified when creating the cache). + */ +int +mb_cache_entry_insert(struct mb_cache_entry *ce, kdev_t dev, + unsigned long block, unsigned int keys[]) +{ + struct mb_cache *cache = ce->e_cache; + unsigned int bucket = (HASHDEV(dev) + block) % cache->c_bucket_count; + struct list_head *l; + int error = -EBUSY, n; + + spin_lock(&mb_cache_spinlock); + list_for_each_prev(l, &cache->c_block_hash[bucket]) { + struct mb_cache_entry *ce = + list_entry(l, struct mb_cache_entry, e_block_list); + if (ce->e_dev == dev && ce->e_block == block) + goto out; + } + __mb_cache_entry_unhash(ce); + ce->e_dev = dev; + ce->e_block = block; + list_add(&ce->e_block_list, &cache->c_block_hash[bucket]); + for (n=0; ne_indexes[n].o_key = keys[n]; + bucket = keys[n] % cache->c_bucket_count; + list_add(&ce->e_indexes[n].o_list, + &cache->c_indexes_hash[n][bucket]); + } + error = 0; +out: + spin_unlock(&mb_cache_spinlock); + return error; +} + + +/* + * mb_cache_entry_release() + * + * Release a handle to a cache entry. When the last handle to a cache entry + * is released it is either freed (if it is invalid) or otherwise inserted + * in to the lru list. + */ +void +mb_cache_entry_release(struct mb_cache_entry *ce) +{ + spin_lock(&mb_cache_spinlock); + __mb_cache_entry_release_unlock(ce); +} + + +/* + * mb_cache_entry_free() + */ +void +mb_cache_entry_free(struct mb_cache_entry *ce) +{ + spin_lock(&mb_cache_spinlock); + mb_assert(list_empty(&ce->e_lru_list)); + __mb_cache_entry_unhash(ce); + __mb_cache_entry_release_unlock(ce); +} + + +/* + * mb_cache_entry_get() + * + * Get a cache entry by device / block number. (There can only be one entry + * in the cache per device and block.) Returns NULL if no such cache entry + * exists. The returned cache entry is locked for exclusive access ("single + * writer"). + */ +struct mb_cache_entry * +mb_cache_entry_get(struct mb_cache *cache, kdev_t dev, unsigned long block) +{ + unsigned int bucket = (HASHDEV(dev) + block) % cache->c_bucket_count; + struct list_head *l; + struct mb_cache_entry *ce; + + spin_lock(&mb_cache_spinlock); + list_for_each(l, &cache->c_block_hash[bucket]) { + ce = list_entry(l, struct mb_cache_entry, e_block_list); + if (ce->e_dev == dev && ce->e_block == block) { + DECLARE_WAITQUEUE(wait, current); + + if (!list_empty(&ce->e_lru_list)) + list_del_init(&ce->e_lru_list); + + while (ce->e_used > 0) { + ce->e_queued++; + add_wait_queue(&mb_cache_queue, &wait); + spin_unlock(&mb_cache_spinlock); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule(); + set_current_state(TASK_RUNNING); + remove_wait_queue(&mb_cache_queue, &wait); + spin_lock(&mb_cache_spinlock); + ce->e_queued--; + } + ce->e_used += 1 + MB_CACHE_WRITER; + + if (!__mb_cache_entry_is_hashed(ce)) { + __mb_cache_entry_release_unlock(ce); + return NULL; + } + goto cleanup; + } + } + ce = NULL; + +cleanup: + spin_unlock(&mb_cache_spinlock); + return ce; +} + +#if !defined(MB_CACHE_INDEXES_COUNT) || (MB_CACHE_INDEXES_COUNT > 0) + +static struct mb_cache_entry * +__mb_cache_entry_find(struct list_head *l, struct list_head *head, + int index, kdev_t dev, unsigned int key) +{ + while (l != head) { + struct mb_cache_entry *ce = + list_entry(l, struct mb_cache_entry, + e_indexes[index].o_list); + if (ce->e_dev == dev && ce->e_indexes[index].o_key == key) { + DECLARE_WAITQUEUE(wait, current); + + if (!list_empty(&ce->e_lru_list)) + list_del_init(&ce->e_lru_list); + + /* Incrementing before holding the lock gives readers + priority over writers. */ + ce->e_used++; + while (ce->e_used >= MB_CACHE_WRITER) { + ce->e_queued++; + add_wait_queue(&mb_cache_queue, &wait); + spin_unlock(&mb_cache_spinlock); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule(); + set_current_state(TASK_RUNNING); + remove_wait_queue(&mb_cache_queue, &wait); + spin_lock(&mb_cache_spinlock); + ce->e_queued--; + } + + if (!__mb_cache_entry_is_hashed(ce)) { + __mb_cache_entry_release_unlock(ce); + spin_lock(&mb_cache_spinlock); + return ERR_PTR(-EAGAIN); + } + return ce; + } + l = l->next; + } + return NULL; +} + + +/* + * mb_cache_entry_find_first() + * + * Find the first cache entry on a given device with a certain key in + * an additional index. Additonal matches can be found with + * mb_cache_entry_find_next(). Returns NULL if no match was found. + * + * @cache: the cache to search + * @index: the number of the additonal index to search (0<=indexc_bucket_count; + struct list_head *l; + struct mb_cache_entry *ce; + + mb_assert(index < mb_cache_indexes(cache)); + spin_lock(&mb_cache_spinlock); + l = cache->c_indexes_hash[index][bucket].next; + ce = __mb_cache_entry_find(l, &cache->c_indexes_hash[index][bucket], + index, dev, key); + spin_unlock(&mb_cache_spinlock); + return ce; +} + + +/* + * mb_cache_entry_find_next() + * + * Find the next cache entry on a given device with a certain key in an + * additional index. Returns NULL if no match could be found. The previous + * entry is atomatically released, so that mb_cache_entry_find_next() can + * be called like this: + * + * entry = mb_cache_entry_find_first(); + * while (entry) { + * ... + * entry = mb_cache_entry_find_next(entry, ...); + * } + * + * @prev: The previous match + * @index: the number of the additonal index to search (0<=indexe_cache; + unsigned int bucket = key % cache->c_bucket_count; + struct list_head *l; + struct mb_cache_entry *ce; + + mb_assert(index < mb_cache_indexes(cache)); + spin_lock(&mb_cache_spinlock); + l = prev->e_indexes[index].o_list.next; + ce = __mb_cache_entry_find(l, &cache->c_indexes_hash[index][bucket], + index, dev, key); + __mb_cache_entry_release_unlock(prev); + return ce; +} + +#endif /* !defined(MB_CACHE_INDEXES_COUNT) || (MB_CACHE_INDEXES_COUNT > 0) */ + +static int __init init_mbcache(void) +{ + register_cache(&mb_cache_definition); + return 0; +} + +static void __exit exit_mbcache(void) +{ + unregister_cache(&mb_cache_definition); +} + +module_init(init_mbcache) +module_exit(exit_mbcache) + Index: linux-2.4.29/fs/xattr.c =================================================================== --- linux-2.4.29.orig/fs/xattr.c +++ linux-2.4.29/fs/xattr.c @@ -159,11 +159,9 @@ getxattr(struct dentry *d, char *name, v error = -EOPNOTSUPP; if (d->d_inode->i_op && d->d_inode->i_op->getxattr) { - down(&d->d_inode->i_sem); lock_kernel(); error = d->d_inode->i_op->getxattr(d, kname, kvalue, size); unlock_kernel(); - up(&d->d_inode->i_sem); } if (kvalue && error > 0) @@ -230,11 +228,9 @@ listxattr(struct dentry *d, char *list, error = -EOPNOTSUPP; if (d->d_inode->i_op && d->d_inode->i_op->listxattr) { - down(&d->d_inode->i_sem); lock_kernel(); error = d->d_inode->i_op->listxattr(d, klist, size); unlock_kernel(); - up(&d->d_inode->i_sem); } if (klist && error > 0) Index: linux-2.4.29/include/asm-arm/unistd.h =================================================================== --- linux-2.4.29.orig/include/asm-arm/unistd.h +++ linux-2.4.29/include/asm-arm/unistd.h @@ -250,7 +250,6 @@ #define __NR_security (__NR_SYSCALL_BASE+223) #define __NR_gettid (__NR_SYSCALL_BASE+224) #define __NR_readahead (__NR_SYSCALL_BASE+225) -#if 0 /* allocated in 2.5 */ #define __NR_setxattr (__NR_SYSCALL_BASE+226) #define __NR_lsetxattr (__NR_SYSCALL_BASE+227) #define __NR_fsetxattr (__NR_SYSCALL_BASE+228) @@ -263,7 +262,6 @@ #define __NR_removexattr (__NR_SYSCALL_BASE+235) #define __NR_lremovexattr (__NR_SYSCALL_BASE+236) #define __NR_fremovexattr (__NR_SYSCALL_BASE+237) -#endif #define __NR_tkill (__NR_SYSCALL_BASE+238) #if 0 /* allocated in 2.5 */ #define __NR_sendfile64 (__NR_SYSCALL_BASE+239) Index: linux-2.4.29/include/asm-s390/unistd.h =================================================================== --- linux-2.4.29.orig/include/asm-s390/unistd.h +++ linux-2.4.29/include/asm-s390/unistd.h @@ -213,9 +213,19 @@ #define __NR_getdents64 220 #define __NR_fcntl64 221 #define __NR_readahead 222 -/* - * Numbers 224-235 are reserved for posix acl - */ + +#define __NR_setxattr 224 +#define __NR_lsetxattr 225 +#define __NR_fsetxattr 226 +#define __NR_getxattr 227 +#define __NR_lgetxattr 228 +#define __NR_fgetxattr 229 +#define __NR_listxattr 230 +#define __NR_llistxattr 231 +#define __NR_flistxattr 232 +#define __NR_removexattr 233 +#define __NR_lremovexattr 234 +#define __NR_fremovexattr 235 #define __NR_gettid 236 #define __NR_tkill 237 Index: linux-2.4.29/include/asm-s390x/unistd.h =================================================================== --- linux-2.4.29.orig/include/asm-s390x/unistd.h +++ linux-2.4.29/include/asm-s390x/unistd.h @@ -181,9 +181,19 @@ #define __NR_mincore 218 #define __NR_madvise 219 #define __NR_readahead 222 -/* - * Numbers 224-235 are reserved for posix acl - */ + +#define __NR_setxattr 224 +#define __NR_lsetxattr 225 +#define __NR_fsetxattr 226 +#define __NR_getxattr 227 +#define __NR_lgetxattr 228 +#define __NR_fgetxattr 229 +#define __NR_listxattr 230 +#define __NR_llistxattr 231 +#define __NR_flistxattr 232 +#define __NR_removexattr 233 +#define __NR_lremovexattr 234 +#define __NR_fremovexattr 235 #define __NR_gettid 236 #define __NR_tkill 237 Index: linux-2.4.29/include/linux/cache_def.h =================================================================== --- /dev/null +++ linux-2.4.29/include/linux/cache_def.h @@ -0,0 +1,15 @@ +/* + * linux/cache_def.h + * Handling of caches defined in drivers, filesystems, ... + * + * Copyright (C) 2002 by Andreas Gruenbacher, + */ + +struct cache_definition { + const char *name; + void (*shrink)(int, unsigned int); + struct list_head link; +}; + +extern void register_cache(struct cache_definition *); +extern void unregister_cache(struct cache_definition *); Index: linux-2.4.29/include/linux/ext2_fs.h =================================================================== --- linux-2.4.29.orig/include/linux/ext2_fs.h +++ linux-2.4.29/include/linux/ext2_fs.h @@ -57,8 +57,6 @@ */ #define EXT2_BAD_INO 1 /* Bad blocks inode */ #define EXT2_ROOT_INO 2 /* Root inode */ -#define EXT2_ACL_IDX_INO 3 /* ACL inode */ -#define EXT2_ACL_DATA_INO 4 /* ACL inode */ #define EXT2_BOOT_LOADER_INO 5 /* Boot loader inode */ #define EXT2_UNDEL_DIR_INO 6 /* Undelete directory inode */ @@ -86,7 +84,6 @@ #else # define EXT2_BLOCK_SIZE(s) (EXT2_MIN_BLOCK_SIZE << (s)->s_log_block_size) #endif -#define EXT2_ACLE_PER_BLOCK(s) (EXT2_BLOCK_SIZE(s) / sizeof (struct ext2_acl_entry)) #define EXT2_ADDR_PER_BLOCK(s) (EXT2_BLOCK_SIZE(s) / sizeof (__u32)) #ifdef __KERNEL__ # define EXT2_BLOCK_SIZE_BITS(s) ((s)->s_blocksize_bits) @@ -121,28 +118,6 @@ #endif /* - * ACL structures - */ -struct ext2_acl_header /* Header of Access Control Lists */ -{ - __u32 aclh_size; - __u32 aclh_file_count; - __u32 aclh_acle_count; - __u32 aclh_first_acle; -}; - -struct ext2_acl_entry /* Access Control List Entry */ -{ - __u32 acle_size; - __u16 acle_perms; /* Access permissions */ - __u16 acle_type; /* Type of entry */ - __u16 acle_tag; /* User or group identity */ - __u16 acle_pad1; - __u32 acle_next; /* Pointer on next entry for the */ - /* same inode or on next free entry */ -}; - -/* * Structure of a blocks group descriptor */ struct ext2_group_desc @@ -314,6 +289,7 @@ struct ext2_inode { #define EXT2_MOUNT_ERRORS_PANIC 0x0040 /* Panic on errors */ #define EXT2_MOUNT_MINIX_DF 0x0080 /* Mimics the Minix statfs */ #define EXT2_MOUNT_NO_UID32 0x0200 /* Disable 32-bit UIDs */ +#define EXT2_MOUNT_XATTR_USER 0x4000 /* Extended user attributes */ #define clear_opt(o, opt) o &= ~EXT2_MOUNT_##opt #define set_opt(o, opt) o |= EXT2_MOUNT_##opt @@ -410,6 +386,7 @@ struct ext2_super_block { #ifdef __KERNEL__ #define EXT2_SB(sb) (&((sb)->u.ext2_sb)) +#define EXT2_I(inode) (&((inode)->u.ext2_i)) #else /* Assume that user mode programs are passing in an ext2fs superblock, not * a kernel struct super_block. This will allow us to call the feature-test @@ -480,7 +457,7 @@ struct ext2_super_block { #define EXT2_FEATURE_INCOMPAT_META_BG 0x0010 #define EXT2_FEATURE_INCOMPAT_ANY 0xffffffff -#define EXT2_FEATURE_COMPAT_SUPP 0 +#define EXT2_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR #define EXT2_FEATURE_INCOMPAT_SUPP (EXT2_FEATURE_INCOMPAT_FILETYPE| \ EXT2_FEATURE_INCOMPAT_META_BG) #define EXT2_FEATURE_RO_COMPAT_SUPP (EXT2_FEATURE_RO_COMPAT_SPARSE_SUPER| \ @@ -650,8 +627,10 @@ extern struct address_space_operations e /* namei.c */ extern struct inode_operations ext2_dir_inode_operations; +extern struct inode_operations ext2_special_inode_operations; /* symlink.c */ +extern struct inode_operations ext2_symlink_inode_operations; extern struct inode_operations ext2_fast_symlink_inode_operations; #endif /* __KERNEL__ */ Index: linux-2.4.29/include/linux/ext2_fs_i.h =================================================================== --- linux-2.4.29.orig/include/linux/ext2_fs_i.h +++ linux-2.4.29/include/linux/ext2_fs_i.h @@ -35,6 +35,16 @@ struct ext2_inode_info { __u32 i_prealloc_block; __u32 i_prealloc_count; __u32 i_dir_start_lookup; +#ifdef CONFIG_EXT2_FS_XATTR + /* + * Extended attributes can be read independently of the main file + * data. Taking i_sem even when reading would cause contention + * between readers of EAs and writers of regular file data, so + * instead we synchronize on xattr_sem when reading or changing + * EAs. + */ + struct rw_semaphore xattr_sem; +#endif }; /* Index: linux-2.4.29/include/linux/ext2_xattr.h =================================================================== --- /dev/null +++ linux-2.4.29/include/linux/ext2_xattr.h @@ -0,0 +1,167 @@ +/* + File: linux/ext2_xattr.h + + On-disk format of extended attributes for the ext2 filesystem. + + (C) 2001 Andreas Gruenbacher, +*/ + +#include +#include +#include + +/* Magic value in attribute blocks */ +#define EXT2_XATTR_MAGIC 0xEA020000 + +/* Maximum number of references to one attribute block */ +#define EXT2_XATTR_REFCOUNT_MAX 1024 + +/* Name indexes */ +#define EXT2_XATTR_INDEX_MAX 10 +#define EXT2_XATTR_INDEX_USER 1 +#define EXT2_XATTR_INDEX_POSIX_ACL_ACCESS 2 +#define EXT2_XATTR_INDEX_POSIX_ACL_DEFAULT 3 +#define EXT2_XATTR_INDEX_TRUSTED 4 + +struct ext2_xattr_header { + __u32 h_magic; /* magic number for identification */ + __u32 h_refcount; /* reference count */ + __u32 h_blocks; /* number of disk blocks used */ + __u32 h_hash; /* hash value of all attributes */ + __u32 h_reserved[4]; /* zero right now */ +}; + +struct ext2_xattr_entry { + __u8 e_name_len; /* length of name */ + __u8 e_name_index; /* attribute name index */ + __u16 e_value_offs; /* offset in disk block of value */ + __u32 e_value_block; /* disk block attribute is stored on (n/i) */ + __u32 e_value_size; /* size of attribute value */ + __u32 e_hash; /* hash value of name and value */ + char e_name[0]; /* attribute name */ +}; + +#define EXT2_XATTR_PAD_BITS 2 +#define EXT2_XATTR_PAD (1<e_name_len)) ) +#define EXT2_XATTR_SIZE(size) \ + (((size) + EXT2_XATTR_ROUND) & ~EXT2_XATTR_ROUND) + +#ifdef __KERNEL__ + +# ifdef CONFIG_EXT2_FS_XATTR + +struct ext2_xattr_handler { + char *prefix; + size_t (*list)(char *list, struct inode *inode, const char *name, + int name_len); + int (*get)(struct inode *inode, const char *name, void *buffer, + size_t size); + int (*set)(struct inode *inode, const char *name, const void *buffer, + size_t size, int flags); +}; + +extern int ext2_xattr_register(int, struct ext2_xattr_handler *); +extern void ext2_xattr_unregister(int, struct ext2_xattr_handler *); + +extern int ext2_setxattr(struct dentry *, const char *, void *, size_t, int); +extern ssize_t ext2_getxattr(struct dentry *, const char *, void *, size_t); +extern ssize_t ext2_listxattr(struct dentry *, char *, size_t); +extern int ext2_removexattr(struct dentry *, const char *); + +extern int ext2_xattr_get(struct inode *, int, const char *, void *, size_t); +extern int ext2_xattr_list(struct inode *, char *, size_t); +extern int ext2_xattr_set(struct inode *, int, const char *, const void *, + size_t, int); + +extern void ext2_xattr_delete_inode(struct inode *); +extern void ext2_xattr_put_super(struct super_block *); + +extern int init_ext2_xattr(void) __init; +extern void exit_ext2_xattr(void); + +# else /* CONFIG_EXT2_FS_XATTR */ +# define ext2_setxattr NULL +# define ext2_getxattr NULL +# define ext2_listxattr NULL +# define ext2_removexattr NULL + +static inline int +ext2_xattr_get(struct inode *inode, int name_index, + const char *name, void *buffer, size_t size) +{ + return -EOPNOTSUPP; +} + +static inline int +ext2_xattr_list(struct inode *inode, char *buffer, size_t size) +{ + return -EOPNOTSUPP; +} + +static inline int +ext2_xattr_set(struct inode *inode, int name_index, const char *name, + const void *value, size_t size, int flags) +{ + return -EOPNOTSUPP; +} + +static inline void +ext2_xattr_delete_inode(struct inode *inode) +{ +} + +static inline void +ext2_xattr_put_super(struct super_block *sb) +{ +} + +static inline int +init_ext2_xattr(void) +{ + return 0; +} + +static inline void +exit_ext2_xattr(void) +{ +} + +# endif /* CONFIG_EXT2_FS_XATTR */ + +# ifdef CONFIG_EXT2_FS_XATTR_USER + +extern int init_ext2_xattr_user(void) __init; + +# else /* CONFIG_EXT2_FS_XATTR_USER */ + +static inline int +init_ext2_xattr_user(void) +{ + return 0; +} + +# endif /* CONFIG_EXT2_FS_XATTR_USER */ + +# ifdef CONFIG_EXT2_FS_XATTR_TRUSTED + +extern int init_ext2_xattr_trusted(void) __init; + +# else /* CONFIG_EXT2_FS_XATTR_TRUSTED */ + +static inline int +init_ext2_xattr_trusted(void) +{ + return 0; +} + +# endif /* CONFIG_EXT2_FS_XATTR_TRUSTED */ + +#endif /* __KERNEL__ */ + Index: linux-2.4.29/include/linux/ext3_fs.h =================================================================== --- linux-2.4.29.orig/include/linux/ext3_fs.h +++ linux-2.4.29/include/linux/ext3_fs.h @@ -58,8 +58,6 @@ */ #define EXT3_BAD_INO 1 /* Bad blocks inode */ #define EXT3_ROOT_INO 2 /* Root inode */ -#define EXT3_ACL_IDX_INO 3 /* ACL inode */ -#define EXT3_ACL_DATA_INO 4 /* ACL inode */ #define EXT3_BOOT_LOADER_INO 5 /* Boot loader inode */ #define EXT3_UNDEL_DIR_INO 6 /* Undelete directory inode */ #define EXT3_RESIZE_INO 7 /* Reserved group descriptors inode */ @@ -89,7 +87,6 @@ #else # define EXT3_BLOCK_SIZE(s) (EXT3_MIN_BLOCK_SIZE << (s)->s_log_block_size) #endif -#define EXT3_ACLE_PER_BLOCK(s) (EXT3_BLOCK_SIZE(s) / sizeof (struct ext3_acl_entry)) #define EXT3_ADDR_PER_BLOCK(s) (EXT3_BLOCK_SIZE(s) / sizeof (__u32)) #ifdef __KERNEL__ # define EXT3_BLOCK_SIZE_BITS(s) ((s)->s_blocksize_bits) @@ -124,28 +121,6 @@ #endif /* - * ACL structures - */ -struct ext3_acl_header /* Header of Access Control Lists */ -{ - __u32 aclh_size; - __u32 aclh_file_count; - __u32 aclh_acle_count; - __u32 aclh_first_acle; -}; - -struct ext3_acl_entry /* Access Control List Entry */ -{ - __u32 acle_size; - __u16 acle_perms; /* Access permissions */ - __u16 acle_type; /* Type of entry */ - __u16 acle_tag; /* User or group identity */ - __u16 acle_pad1; - __u32 acle_next; /* Pointer on next entry for the */ - /* same inode or on next free entry */ -}; - -/* * Structure of a blocks group descriptor */ struct ext3_group_desc @@ -339,6 +314,7 @@ struct ext3_inode { #define EXT3_MOUNT_WRITEBACK_DATA 0x0C00 /* No data ordering */ #define EXT3_MOUNT_UPDATE_JOURNAL 0x1000 /* Update the journal format */ #define EXT3_MOUNT_NO_UID32 0x2000 /* Disable 32-bit UIDs */ +#define EXT3_MOUNT_XATTR_USER 0x4000 /* Extended user attributes */ /* Compatibility, for having both ext2_fs.h and ext3_fs.h included at once */ #ifndef _LINUX_EXT2_FS_H @@ -519,7 +495,7 @@ struct ext3_super_block { #define EXT3_FEATURE_INCOMPAT_JOURNAL_DEV 0x0008 /* Journal device */ #define EXT3_FEATURE_INCOMPAT_META_BG 0x0010 -#define EXT3_FEATURE_COMPAT_SUPP 0 +#define EXT3_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR #define EXT3_FEATURE_INCOMPAT_SUPP (EXT3_FEATURE_INCOMPAT_FILETYPE| \ EXT3_FEATURE_INCOMPAT_RECOVER| \ EXT3_FEATURE_INCOMPAT_META_BG) @@ -646,6 +622,7 @@ extern void ext3_check_inodes_bitmap (st extern unsigned long ext3_count_free (struct buffer_head *, unsigned); /* inode.c */ +extern int ext3_forget(handle_t *, int, struct inode *, struct buffer_head *, int); extern struct buffer_head * ext3_getblk (handle_t *, struct inode *, long, int, int *); extern struct buffer_head * ext3_bread (handle_t *, struct inode *, int, int, int *); @@ -713,8 +690,10 @@ extern struct address_space_operations e /* namei.c */ extern struct inode_operations ext3_dir_inode_operations; +extern struct inode_operations ext3_special_inode_operations; /* symlink.c */ +extern struct inode_operations ext3_symlink_inode_operations; extern struct inode_operations ext3_fast_symlink_inode_operations; Index: linux-2.4.29/include/linux/ext3_fs_i.h =================================================================== --- linux-2.4.29.orig/include/linux/ext3_fs_i.h +++ linux-2.4.29/include/linux/ext3_fs_i.h @@ -42,6 +42,16 @@ struct ext3_inode_info { __u32 i_prealloc_count; #endif __u32 i_dir_start_lookup; +#ifdef CONFIG_EXT3_FS_XATTR + /* + * Extended attributes can be read independently of the main file + * data. Taking i_sem even when reading would cause contention + * between readers of EAs and writers of regular file data, so + * instead we synchronize on xattr_sem when reading or changing + * EAs. + */ + struct rw_semaphore xattr_sem; +#endif struct list_head i_orphan; /* unlinked but open inodes */ Index: linux-2.4.29/include/linux/ext3_jbd.h =================================================================== --- linux-2.4.29.orig/include/linux/ext3_jbd.h +++ linux-2.4.29/include/linux/ext3_jbd.h @@ -30,13 +30,19 @@ #define EXT3_SINGLEDATA_TRANS_BLOCKS 8U +/* Extended attributes may touch two data buffers, two bitmap buffers, + * and two group and summaries. */ + +#define EXT3_XATTR_TRANS_BLOCKS 8 + /* Define the minimum size for a transaction which modifies data. This * needs to take into account the fact that we may end up modifying two * quota files too (one for the group, one for the user quota). The * superblock only gets updated once, of course, so don't bother * counting that again for the quota updates. */ -#define EXT3_DATA_TRANS_BLOCKS (3 * EXT3_SINGLEDATA_TRANS_BLOCKS - 2) +#define EXT3_DATA_TRANS_BLOCKS (3 * EXT3_SINGLEDATA_TRANS_BLOCKS + \ + EXT3_XATTR_TRANS_BLOCKS - 2) extern int ext3_writepage_trans_blocks(struct inode *inode); Index: linux-2.4.29/include/linux/ext3_xattr.h =================================================================== --- /dev/null +++ linux-2.4.29/include/linux/ext3_xattr.h @@ -0,0 +1,169 @@ +/* + File: linux/ext3_xattr.h + + On-disk format of extended attributes for the ext3 filesystem. + + (C) 2001 Andreas Gruenbacher, +*/ + +#include +#include +#include + +/* Magic value in attribute blocks */ +#define EXT3_XATTR_MAGIC 0xEA020000 + +/* Maximum number of references to one attribute block */ +#define EXT3_XATTR_REFCOUNT_MAX 1024 + +/* Name indexes */ +#define EXT3_XATTR_INDEX_MAX 10 +#define EXT3_XATTR_INDEX_USER 1 +#define EXT3_XATTR_INDEX_POSIX_ACL_ACCESS 2 +#define EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT 3 +#define EXT3_XATTR_INDEX_TRUSTED 4 + +struct ext3_xattr_header { + __u32 h_magic; /* magic number for identification */ + __u32 h_refcount; /* reference count */ + __u32 h_blocks; /* number of disk blocks used */ + __u32 h_hash; /* hash value of all attributes */ + __u32 h_reserved[4]; /* zero right now */ +}; + +struct ext3_xattr_entry { + __u8 e_name_len; /* length of name */ + __u8 e_name_index; /* attribute name index */ + __u16 e_value_offs; /* offset in disk block of value */ + __u32 e_value_block; /* disk block attribute is stored on (n/i) */ + __u32 e_value_size; /* size of attribute value */ + __u32 e_hash; /* hash value of name and value */ + char e_name[0]; /* attribute name */ +}; + +#define EXT3_XATTR_PAD_BITS 2 +#define EXT3_XATTR_PAD (1<e_name_len)) ) +#define EXT3_XATTR_SIZE(size) \ + (((size) + EXT3_XATTR_ROUND) & ~EXT3_XATTR_ROUND) + +#ifdef __KERNEL__ + +# ifdef CONFIG_EXT3_FS_XATTR + +struct ext3_xattr_handler { + char *prefix; + size_t (*list)(char *list, struct inode *inode, const char *name, + int name_len); + int (*get)(struct inode *inode, const char *name, void *buffer, + size_t size); + int (*set)(struct inode *inode, const char *name, const void *buffer, + size_t size, int flags); +}; + +extern int ext3_xattr_register(int, struct ext3_xattr_handler *); +extern void ext3_xattr_unregister(int, struct ext3_xattr_handler *); + +extern int ext3_setxattr(struct dentry *, const char *, void *, size_t, int); +extern ssize_t ext3_getxattr(struct dentry *, const char *, void *, size_t); +extern ssize_t ext3_listxattr(struct dentry *, char *, size_t); +extern int ext3_removexattr(struct dentry *, const char *); + +extern int ext3_xattr_get(struct inode *, int, const char *, void *, size_t); +extern int ext3_xattr_list(struct inode *, char *, size_t); +extern int ext3_xattr_set_handle(handle_t *handle, struct inode *, int, + const char *, const void *, size_t, int); +extern int ext3_xattr_set(struct inode *, int, const char *, const void *, + size_t, int); + +extern void ext3_xattr_delete_inode(handle_t *, struct inode *); +extern void ext3_xattr_put_super(struct super_block *); + +extern int init_ext3_xattr(void) __init; +extern void exit_ext3_xattr(void); + +# else /* CONFIG_EXT3_FS_XATTR */ +# define ext3_setxattr NULL +# define ext3_getxattr NULL +# define ext3_listxattr NULL +# define ext3_removexattr NULL + +static inline int +ext3_xattr_get(struct inode *inode, int name_index, const char *name, + void *buffer, size_t size) +{ + return -EOPNOTSUPP; +} + +static inline int +ext3_xattr_list(struct inode *inode, void *buffer, size_t size) +{ + return -EOPNOTSUPP; +} + +static inline int +ext3_xattr_set(handle_t *handle, struct inode *inode, int name_index, + const char *name, const void *value, size_t size) +{ + return -EOPNOTSUPP; +} + +static inline void +ext3_xattr_delete_inode(handle_t *handle, struct inode *inode) +{ +} + +static inline void +ext3_xattr_put_super(struct super_block *sb) +{ +} + +static inline int +init_ext3_xattr(void) +{ + return 0; +} + +static inline void +exit_ext3_xattr(void) +{ +} + +# endif /* CONFIG_EXT3_FS_XATTR */ + +# ifdef CONFIG_EXT3_FS_XATTR_USER + +extern int init_ext3_xattr_user(void) __init; + +# else /* CONFIG_EXT3_FS_XATTR_USER */ + +static inline int +init_ext3_xattr_user(void) +{ + return 0; +} + +#endif /* CONFIG_EXT3_FS_XATTR_USER */ + +# ifdef CONFIG_EXT3_FS_XATTR_TRUSTED + +extern int init_ext3_xattr_trusted(void) __init; + +# else /* CONFIG_EXT3_FS_XATTR_TRUSTED */ + +static inline int +init_ext3_xattr_trusted(void) +{ + return 0; +} + +#endif /* CONFIG_EXT3_FS_XATTR_TRUSTED */ + +#endif /* __KERNEL__ */ + Index: linux-2.4.29/include/linux/mbcache.h =================================================================== --- /dev/null +++ linux-2.4.29/include/linux/mbcache.h @@ -0,0 +1,51 @@ +/* + File: linux/mbcache.h + + (C) 2001 by Andreas Gruenbacher, +*/ + +/* Hardwire the number of additional indexes */ +#define MB_CACHE_INDEXES_COUNT 1 + +struct mb_cache_entry; + +struct mb_cache_entry { + struct list_head e_lru_list; + struct mb_cache *e_cache; + unsigned short e_used; + unsigned short e_queued; + kdev_t e_dev; + unsigned long e_block; + struct list_head e_block_list; + struct { + struct list_head o_list; + unsigned int o_key; + } e_indexes[0]; +}; + +struct mb_cache_op { + int (*free)(struct mb_cache_entry *, int); +}; + +/* Functions on caches */ + +struct mb_cache * mb_cache_create(const char *, struct mb_cache_op *, size_t, + int, int); +void mb_cache_shrink(struct mb_cache *, kdev_t); +void mb_cache_destroy(struct mb_cache *); + +/* Functions on cache entries */ + +struct mb_cache_entry *mb_cache_entry_alloc(struct mb_cache *); +int mb_cache_entry_insert(struct mb_cache_entry *, kdev_t, unsigned long, + unsigned int[]); +void mb_cache_entry_release(struct mb_cache_entry *); +void mb_cache_entry_free(struct mb_cache_entry *); +struct mb_cache_entry *mb_cache_entry_get(struct mb_cache *, kdev_t, + unsigned long); +#if !defined(MB_CACHE_INDEXES_COUNT) || (MB_CACHE_INDEXES_COUNT > 0) +struct mb_cache_entry *mb_cache_entry_find_first(struct mb_cache *cache, int, + kdev_t, unsigned int); +struct mb_cache_entry *mb_cache_entry_find_next(struct mb_cache_entry *, int, + kdev_t, unsigned int); +#endif Index: linux-2.4.29/kernel/ksyms.c =================================================================== --- linux-2.4.29.orig/kernel/ksyms.c +++ linux-2.4.29/kernel/ksyms.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include #include @@ -109,6 +110,8 @@ EXPORT_SYMBOL(kmem_cache_shrink); EXPORT_SYMBOL(kmem_cache_alloc); EXPORT_SYMBOL(kmem_cache_free); EXPORT_SYMBOL(kmem_cache_size); +EXPORT_SYMBOL(register_cache); +EXPORT_SYMBOL(unregister_cache); EXPORT_SYMBOL(kmalloc); EXPORT_SYMBOL(kfree); EXPORT_SYMBOL(vfree); Index: linux-2.4.29/mm/vmscan.c =================================================================== --- linux-2.4.29.orig/mm/vmscan.c +++ linux-2.4.29/mm/vmscan.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -86,6 +87,49 @@ int vm_vfs_scan_ratio = 6; int vm_anon_lru = 0; /* + * Handling of caches defined in drivers, filesystems, ... + * + * The cache definition contains a callback for shrinking + * the cache. + * + * The [un]register_cache() functions may only be called when + * the kernel lock is held. The shrink() functions are also + * called with the kernel lock held. + */ +static DECLARE_MUTEX(other_caches_sem); +static LIST_HEAD(other_caches); + +void register_cache(struct cache_definition *cache) +{ + down(&other_caches_sem); + list_add(&cache->link, &other_caches); + up(&other_caches_sem); +} + +void unregister_cache(struct cache_definition *cache) +{ + down(&other_caches_sem); + list_del(&cache->link); + up(&other_caches_sem); +} + +static void shrink_other_caches(unsigned int priority, int gfp_mask) +{ + struct list_head *p; + + if (down_trylock(&other_caches_sem)) + return; + + list_for_each_prev(p, &other_caches) { + struct cache_definition *cache = + list_entry(p, struct cache_definition, link); + + cache->shrink(priority, gfp_mask); + } + up(&other_caches_sem); +} + +/* * The swap-out function returns 1 if it successfully * scanned all the pages it was asked to (`count'). * It returns zero if it couldn't do anything, @@ -544,6 +588,7 @@ page_mapped: #ifdef CONFIG_QUOTA shrink_dqcache_memory(vm_vfs_scan_ratio, gfp_mask); #endif + shrink_other_caches(vm_vfs_scan_ratio, gfp_mask); if (!*failed_swapout) *failed_swapout = !swap_out(classzone);