diff options
Diffstat (limited to 'fs')
135 files changed, 3413 insertions, 1449 deletions
diff --git a/fs/9p/conv.c b/fs/9p/conv.c index 1e898144eb7..56d88c1a09c 100644 --- a/fs/9p/conv.c +++ b/fs/9p/conv.c @@ -673,8 +673,10 @@ struct v9fs_fcall *v9fs_create_tcreate(u32 fid, char *name, u32 perm, u8 mode, struct cbuf *bufp = &buffer; size = 4 + 2 + strlen(name) + 4 + 1; /* fid[4] name[s] perm[4] mode[1] */ - if (extended && extension!=NULL) - size += 2 + strlen(extension); /* extension[s] */ + if (extended) { + size += 2 + /* extension[s] */ + (extension == NULL ? 0 : strlen(extension)); + } fc = v9fs_create_common(bufp, size, TCREATE); if (IS_ERR(fc)) diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 2f580a197b8..eae50c9d6dc 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -434,11 +434,11 @@ static int v9fs_remove(struct inode *dir, struct dentry *file, int rmdir) result = v9fs_t_remove(v9ses, fid, &fcall); if (result < 0) { PRINT_FCALL_ERROR("remove fails", fcall); - } else { - v9fs_put_idpool(fid, &v9ses->fidpool); - v9fs_fid_destroy(v9fid); } + v9fs_put_idpool(fid, &v9ses->fidpool); + v9fs_fid_destroy(v9fid); + kfree(fcall); return result; } diff --git a/fs/Kconfig b/fs/Kconfig index 53f5c6d6112..3f00a9faabc 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1801,6 +1801,7 @@ config CIFS_POSIX config CIFS_DEBUG2 bool "Enable additional CIFS debugging routines" + depends on CIFS help Enabling this option adds a few more debugging routines to the cifs code which slightly increases the size of diff --git a/fs/adfs/super.c b/fs/adfs/super.c index ba1c88af49f..82011019494 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -308,7 +308,7 @@ static struct adfs_discmap *adfs_read_map(struct super_block *sb, struct adfs_di if (adfs_checkmap(sb, dm)) return dm; - adfs_error(sb, NULL, "map corrupted"); + adfs_error(sb, "map corrupted"); error_free: while (--zone >= 0) diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index fcaeead9696..50cfca5c7ef 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -512,7 +512,11 @@ befs_utf2nls(struct super_block *sb, const char *in, wchar_t uni; int unilen, utflen; char *result; - int maxlen = in_len; /* The utf8->nls conversion can't make more chars */ + /* The utf8->nls conversion won't make the final nls string bigger + * than the utf one, but if the string is pure ascii they'll have the + * same width and an extra char is needed to save the additional \0 + */ + int maxlen = in_len + 1; befs_debug(sb, "---> utf2nls()"); @@ -588,7 +592,10 @@ befs_nls2utf(struct super_block *sb, const char *in, wchar_t uni; int unilen, utflen; char *result; - int maxlen = 3 * in_len; + /* There're nls characters that will translate to 3-chars-wide UTF-8 + * characters, a additional byte is needed to save the final \0 + * in special cases */ + int maxlen = (3 * in_len) + 1; befs_debug(sb, "---> nls2utf()\n"); diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index d0434406eae..672a3b90bc5 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -84,7 +84,7 @@ static struct linux_binfmt elf_format = { .min_coredump = ELF_EXEC_PAGESIZE }; -#define BAD_ADDR(x) ((unsigned long)(x) > TASK_SIZE) +#define BAD_ADDR(x) ((unsigned long)(x) >= TASK_SIZE) static int set_brk(unsigned long start, unsigned long end) { @@ -394,7 +394,7 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex, * <= p_memsize so it's only necessary to check p_memsz. */ k = load_addr + eppnt->p_vaddr; - if (k > TASK_SIZE || + if (BAD_ADDR(k) || eppnt->p_filesz > eppnt->p_memsz || eppnt->p_memsz > TASK_SIZE || TASK_SIZE - eppnt->p_memsz < k) { @@ -887,7 +887,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) * allowed task size. Note that p_filesz must always be * <= p_memsz so it is only necessary to check p_memsz. */ - if (k > TASK_SIZE || elf_ppnt->p_filesz > elf_ppnt->p_memsz || + if (BAD_ADDR(k) || elf_ppnt->p_filesz > elf_ppnt->p_memsz || elf_ppnt->p_memsz > TASK_SIZE || TASK_SIZE - elf_ppnt->p_memsz < k) { /* set_brk can never work. Avoid overflows. */ @@ -941,10 +941,9 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) interpreter, &interp_load_addr); if (BAD_ADDR(elf_entry)) { - printk(KERN_ERR "Unable to load interpreter %.128s\n", - elf_interpreter); force_sig(SIGSEGV, current); - retval = -ENOEXEC; /* Nobody gets to see this, but.. */ + retval = IS_ERR((void *)elf_entry) ? + (int)elf_entry : -EINVAL; goto out_free_dentry; } reloc_func_desc = interp_load_addr; @@ -955,8 +954,8 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) } else { elf_entry = loc->elf_ex.e_entry; if (BAD_ADDR(elf_entry)) { - send_sig(SIGSEGV, current, 0); - retval = -ENOEXEC; /* Nobody gets to see this, but.. */ + force_sig(SIGSEGV, current); + retval = -EINVAL; goto out_free_dentry; } } @@ -1186,8 +1185,6 @@ static int maydump(struct vm_area_struct *vma) return 1; } -#define roundup(x, y) ((((x) + ((y) - 1)) / (y)) * (y)) - /* An ELF note in memory */ struct memelfnote { diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index eba4e23b9ca..2f336582922 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -1,6 +1,6 @@ /* binfmt_elf_fdpic.c: FDPIC ELF binary format * - * Copyright (C) 2003, 2004 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2003, 2004, 2006 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * Derived from binfmt_elf.c * @@ -24,7 +24,9 @@ #include <linux/file.h> #include <linux/fcntl.h> #include <linux/slab.h> +#include <linux/pagemap.h> #include <linux/highmem.h> +#include <linux/highuid.h> #include <linux/personality.h> #include <linux/ptrace.h> #include <linux/init.h> @@ -48,45 +50,59 @@ typedef char *elf_caddr_t; #define kdebug(fmt, ...) do {} while(0) #endif +#if 0 +#define kdcore(fmt, ...) printk("FDPIC "fmt"\n" ,##__VA_ARGS__ ) +#else +#define kdcore(fmt, ...) do {} while(0) +#endif + MODULE_LICENSE("GPL"); -static int load_elf_fdpic_binary(struct linux_binprm *bprm, struct pt_regs *regs); -//static int load_elf_fdpic_library(struct file *); -static int elf_fdpic_fetch_phdrs(struct elf_fdpic_params *params, struct file *file); -static int elf_fdpic_map_file(struct elf_fdpic_params *params, - struct file *file, - struct mm_struct *mm, - const char *what); +static int load_elf_fdpic_binary(struct linux_binprm *, struct pt_regs *); +static int elf_fdpic_fetch_phdrs(struct elf_fdpic_params *, struct file *); +static int elf_fdpic_map_file(struct elf_fdpic_params *, struct file *, + struct mm_struct *, const char *); -static int create_elf_fdpic_tables(struct linux_binprm *bprm, - struct mm_struct *mm, - struct elf_fdpic_params *exec_params, - struct elf_fdpic_params *interp_params); +static int create_elf_fdpic_tables(struct linux_binprm *, struct mm_struct *, + struct elf_fdpic_params *, + struct elf_fdpic_params *); #ifndef CONFIG_MMU -static int elf_fdpic_transfer_args_to_stack(struct linux_binprm *bprm, unsigned long *_sp); -static int elf_fdpic_map_file_constdisp_on_uclinux(struct elf_fdpic_params *params, - struct file *file, - struct mm_struct *mm); +static int elf_fdpic_transfer_args_to_stack(struct linux_binprm *, + unsigned long *); +static int elf_fdpic_map_file_constdisp_on_uclinux(struct elf_fdpic_params *, + struct file *, + struct mm_struct *); #endif -static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params, - struct file *file, - struct mm_struct *mm); +static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *, + struct file *, struct mm_struct *); + +#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE) +static int elf_fdpic_core_dump(long, struct pt_regs *, struct file *); +#endif static struct linux_binfmt elf_fdpic_format = { .module = THIS_MODULE, .load_binary = load_elf_fdpic_binary, -// .load_shlib = load_elf_fdpic_library, -// .core_dump = elf_fdpic_core_dump, +#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE) + .core_dump = elf_fdpic_core_dump, +#endif .min_coredump = ELF_EXEC_PAGESIZE, }; -static int __init init_elf_fdpic_binfmt(void) { return register_binfmt(&elf_fdpic_format); } -static void __exit exit_elf_fdpic_binfmt(void) { unregister_binfmt(&elf_fdpic_format); } +static int __init init_elf_fdpic_binfmt(void) +{ + return register_binfmt(&elf_fdpic_format); +} + +static void __exit exit_elf_fdpic_binfmt(void) +{ + unregister_binfmt(&elf_fdpic_format); +} -module_init(init_elf_fdpic_binfmt) -module_exit(exit_elf_fdpic_binfmt) +core_initcall(init_elf_fdpic_binfmt); +module_exit(exit_elf_fdpic_binfmt); static int is_elf_fdpic(struct elfhdr *hdr, struct file *file) { @@ -105,7 +121,8 @@ static int is_elf_fdpic(struct elfhdr *hdr, struct file *file) /* * read the program headers table into memory */ -static int elf_fdpic_fetch_phdrs(struct elf_fdpic_params *params, struct file *file) +static int elf_fdpic_fetch_phdrs(struct elf_fdpic_params *params, + struct file *file) { struct elf32_phdr *phdr; unsigned long size; @@ -121,7 +138,8 @@ static int elf_fdpic_fetch_phdrs(struct elf_fdpic_params *params, struct file *f if (!params->phdrs) return -ENOMEM; - retval = kernel_read(file, params->hdr.e_phoff, (char *) params->phdrs, size); + retval = kernel_read(file, params->hdr.e_phoff, + (char *) params->phdrs, size); if (retval < 0) return retval; @@ -141,17 +159,24 @@ static int elf_fdpic_fetch_phdrs(struct elf_fdpic_params *params, struct file *f } return 0; -} /* end elf_fdpic_fetch_phdrs() */ +} /*****************************************************************************/ /* * load an fdpic binary into various bits of memory */ -static int load_elf_fdpic_binary(struct linux_binprm *bprm, struct pt_regs *regs) +static int load_elf_fdpic_binary(struct linux_binprm *bprm, + struct pt_regs *regs) { struct elf_fdpic_params exec_params, interp_params; struct elf_phdr *phdr; - unsigned long stack_size; + unsigned long stack_size, entryaddr; +#ifndef CONFIG_MMU + unsigned long fullsize; +#endif +#ifdef ELF_FDPIC_PLAT_INIT + unsigned long dynaddr; +#endif struct file *interpreter = NULL; /* to shut gcc up */ char *interpreter_name = NULL; int executable_stack; @@ -212,7 +237,8 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, struct pt_regs *regs goto error; } - retval = kernel_read(interpreter, 0, bprm->buf, BINPRM_BUF_SIZE); + retval = kernel_read(interpreter, 0, bprm->buf, + BINPRM_BUF_SIZE); if (retval < 0) goto error; @@ -295,7 +321,8 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, struct pt_regs *regs ¤t->mm->start_stack, ¤t->mm->start_brk); - retval = setup_arg_pages(bprm, current->mm->start_stack, executable_stack); + retval = setup_arg_pages(bprm, current->mm->start_stack, + executable_stack); if (retval < 0) { send_sig(SIGKILL, current, 0); goto error_kill; @@ -303,7 +330,8 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, struct pt_regs *regs #endif /* load the executable and interpreter into memory */ - retval = elf_fdpic_map_file(&exec_params, bprm->file, current->mm, "executable"); + retval = elf_fdpic_map_file(&exec_params, bprm->file, current->mm, + "executable"); if (retval < 0) goto error_kill; @@ -324,7 +352,8 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, struct pt_regs *regs if (!current->mm->start_brk) current->mm->start_brk = current->mm->end_data; - current->mm->brk = current->mm->start_brk = PAGE_ALIGN(current->mm->start_brk); + current->mm->brk = current->mm->start_brk = + PAGE_ALIGN(current->mm->start_brk); #else /* create a stack and brk area big enough for everyone @@ -336,47 +365,45 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, struct pt_regs *regs stack_size = PAGE_SIZE * 2; down_write(¤t->mm->mmap_sem); - current->mm->start_brk = do_mmap(NULL, - 0, - stack_size, + current->mm->start_brk = do_mmap(NULL, 0, stack_size, PROT_READ | PROT_WRITE | PROT_EXEC, MAP_PRIVATE | MAP_ANON | MAP_GROWSDOWN, 0); - if (IS_ERR((void *) current->mm->start_brk)) { + if (IS_ERR_VALUE(current->mm->start_brk)) { up_write(¤t->mm->mmap_sem); retval = current->mm->start_brk; current->mm->start_brk = 0; goto error_kill; } - if (do_mremap(current->mm->start_brk, - stack_size, - ksize((char *) current->mm->start_brk), - 0, 0 - ) == current->mm->start_brk - ) - stack_size = ksize((char *) current->mm->start_brk); + /* expand the stack mapping to use up the entire allocation granule */ + fullsize = ksize((char *) current->mm->start_brk); + if (!IS_ERR_VALUE(do_mremap(current->mm->start_brk, stack_size, + fullsize, 0, 0))) + stack_size = fullsize; up_write(¤t->mm->mmap_sem); current->mm->brk = current->mm->start_brk; current->mm->context.end_brk = current->mm->start_brk; - current->mm->context.end_brk += (stack_size > PAGE_SIZE) ? (stack_size - PAGE_SIZE) : 0; + current->mm->context.end_brk += + (stack_size > PAGE_SIZE) ? (stack_size - PAGE_SIZE) : 0; current->mm->start_stack = current->mm->start_brk + stack_size; #endif compute_creds(bprm); current->flags &= ~PF_FORKNOEXEC; - if (create_elf_fdpic_tables(bprm, current->mm, &exec_params, &interp_params) < 0) + if (create_elf_fdpic_tables(bprm, current->mm, + &exec_params, &interp_params) < 0) goto error_kill; - kdebug("- start_code %lx", (long) current->mm->start_code); - kdebug("- end_code %lx", (long) current->mm->end_code); - kdebug("- start_data %lx", (long) current->mm->start_data); - kdebug("- end_data %lx", (long) current->mm->end_data); - kdebug("- start_brk %lx", (long) current->mm->start_brk); - kdebug("- brk %lx", (long) current->mm->brk); - kdebug("- start_stack %lx", (long) current->mm->start_stack); + kdebug("- start_code %lx", current->mm->start_code); + kdebug("- end_code %lx", current->mm->end_code); + kdebug("- start_data %lx", current->mm->start_data); + kdebug("- end_data %lx", current->mm->end_data); + kdebug("- start_brk %lx", current->mm->start_brk); + kdebug("- brk %lx", current->mm->brk); + kdebug("- start_stack %lx", current->mm->start_stack); #ifdef ELF_FDPIC_PLAT_INIT /* @@ -385,21 +412,18 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, struct pt_regs *regs * example. This macro performs whatever initialization to * the regs structure is required. */ - ELF_FDPIC_PLAT_INIT(regs, - exec_params.map_addr, - interp_params.map_addr, - interp_params.dynamic_addr ?: exec_params.dynamic_addr - ); + dynaddr = interp_params.dynamic_addr ?: exec_params.dynamic_addr; + ELF_FDPIC_PLAT_INIT(regs, exec_params.map_addr, interp_params.map_addr, + dynaddr); #endif /* everything is now ready... get the userspace context ready to roll */ - start_thread(regs, - interp_params.entry_addr ?: exec_params.entry_addr, - current->mm->start_stack); + entryaddr = interp_params.entry_addr ?: exec_params.entry_addr; + start_thread(regs, entryaddr, current->mm->start_stack); if (unlikely(current->ptrace & PT_PTRACED)) { if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify ((PTRACE_EVENT_EXEC << 8) | SIGTRAP); + ptrace_notify((PTRACE_EVENT_EXEC << 8) | SIGTRAP); else send_sig(SIGTRAP, current, 0); } @@ -419,11 +443,11 @@ error: return retval; /* unrecoverable error - kill the process */ - error_kill: +error_kill: send_sig(SIGSEGV, current, 0); goto error; -} /* end load_elf_fdpic_binary() */ +} /*****************************************************************************/ /* @@ -459,6 +483,7 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, */ hwcap = ELF_HWCAP; k_platform = ELF_PLATFORM; + u_platform = NULL; if (k_platform) { platform_len = strlen(k_platform) + 1; @@ -470,11 +495,11 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, #if defined(__i386__) && defined(CONFIG_SMP) /* in some cases (e.g. Hyper-Threading), we want to avoid L1 evictions - * by the processes running on the same package. One thing we can do - * is to shuffle the initial stack for them. + * by the processes running on the same package. One thing we can do is + * to shuffle the initial stack for them. * - * the conditionals here are unneeded, but kept in to make the - * code behaviour the same as pre change unless we have hyperthreaded + * the conditionals here are unneeded, but kept in to make the code + * behaviour the same as pre change unless we have hyperthreaded * processors. This keeps Mr Marcelo Person happier but should be * removed for 2.5 */ @@ -497,11 +522,13 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, if (interp_params->loadmap) { len = sizeof(struct elf32_fdpic_loadmap); - len += sizeof(struct elf32_fdpic_loadseg) * interp_params->loadmap->nsegs; + len += sizeof(struct elf32_fdpic_loadseg) * + interp_params->loadmap->nsegs; sp = (sp - len) & ~7UL; interp_params->map_addr = sp; - if (copy_to_user((void __user *) sp, interp_params->loadmap, len) != 0) + if (copy_to_user((void __user *) sp, interp_params->loadmap, + len) != 0) return -EFAULT; current->mm->context.interp_fdpic_loadmap = (unsigned long) sp; @@ -525,34 +552,37 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, sp -= sp & 15UL; /* put the ELF interpreter info on the stack */ -#define NEW_AUX_ENT(nr, id, val) \ - do { \ - struct { unsigned long _id, _val; } __user *ent = (void __user *) csp; \ - __put_user((id), &ent[nr]._id); \ - __put_user((val), &ent[nr]._val); \ +#define NEW_AUX_ENT(nr, id, val) \ + do { \ + struct { unsigned long _id, _val; } __user *ent; \ + \ + ent = (void __user *) csp; \ + __put_user((id), &ent[nr]._id); \ + __put_user((val), &ent[nr]._val); \ } while (0) csp -= 2 * sizeof(unsigned long); NEW_AUX_ENT(0, AT_NULL, 0); if (k_platform) { csp -= 2 * sizeof(unsigned long); - NEW_AUX_ENT(0, AT_PLATFORM, (elf_addr_t)(unsigned long) u_platform); + NEW_AUX_ENT(0, AT_PLATFORM, + (elf_addr_t) (unsigned long) u_platform); } csp -= DLINFO_ITEMS * 2 * sizeof(unsigned long); - NEW_AUX_ENT( 0, AT_HWCAP, hwcap); - NEW_AUX_ENT( 1, AT_PAGESZ, PAGE_SIZE); - NEW_AUX_ENT( 2, AT_CLKTCK, CLOCKS_PER_SEC); - NEW_AUX_ENT( 3, AT_PHDR, exec_params->ph_addr); - NEW_AUX_ENT( 4, AT_PHENT, sizeof(struct elf_phdr)); - NEW_AUX_ENT( 5, AT_PHNUM, exec_params->hdr.e_phnum); - NEW_AUX_ENT( 6, AT_BASE, interp_params->elfhdr_addr); - NEW_AUX_ENT( 7, AT_FLAGS, 0); - NEW_AUX_ENT( 8, AT_ENTRY, exec_params->entry_addr); - NEW_AUX_ENT( 9, AT_UID, (elf_addr_t) current->uid); - NEW_AUX_ENT(10, AT_EUID, (elf_addr_t) current->euid); - NEW_AUX_ENT(11, AT_GID, (elf_addr_t) current->gid); - NEW_AUX_ENT(12, AT_EGID, (elf_addr_t) current->egid); + NEW_AUX_ENT( 0, AT_HWCAP, hwcap); + NEW_AUX_ENT( 1, AT_PAGESZ, PAGE_SIZE); + NEW_AUX_ENT( 2, AT_CLKTCK, CLOCKS_PER_SEC); + NEW_AUX_ENT( 3, AT_PHDR, exec_params->ph_addr); + NEW_AUX_ENT( 4, AT_PHENT, sizeof(struct elf_phdr)); + NEW_AUX_ENT( 5, AT_PHNUM, exec_params->hdr.e_phnum); + NEW_AUX_ENT( 6, AT_BASE, interp_params->elfhdr_addr); + NEW_AUX_ENT( 7, AT_FLAGS, 0); + NEW_AUX_ENT( 8, AT_ENTRY, exec_params->entry_addr); + NEW_AUX_ENT( 9, AT_UID, (elf_addr_t) current->uid); + NEW_AUX_ENT(10, AT_EUID, (elf_addr_t) current->euid); + NEW_AUX_ENT(11, AT_GID, (elf_addr_t) current->gid); + NEW_AUX_ENT(12, AT_EGID, (elf_addr_t) current->egid); #ifdef ARCH_DLINFO /* ARCH_DLINFO must come last so platform specific code can enforce @@ -578,7 +608,8 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, #ifdef CONFIG_MMU current->mm->arg_start = bprm->p; #else - current->mm->arg_start = current->mm->start_stack - (MAX_ARG_PAGES * PAGE_SIZE - bprm->p); + current->mm->arg_start = current->mm->start_stack - + (MAX_ARG_PAGES * PAGE_SIZE - bprm->p); #endif p = (char __user *) current->mm->arg_start; @@ -606,7 +637,7 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, mm->start_stack = (unsigned long) sp; return 0; -} /* end create_elf_fdpic_tables() */ +} /*****************************************************************************/ /* @@ -614,7 +645,8 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, * the stack */ #ifndef CONFIG_MMU -static int elf_fdpic_transfer_args_to_stack(struct linux_binprm *bprm, unsigned long *_sp) +static int elf_fdpic_transfer_args_to_stack(struct linux_binprm *bprm, + unsigned long *_sp) { unsigned long index, stop, sp; char *src; @@ -635,9 +667,9 @@ static int elf_fdpic_transfer_args_to_stack(struct linux_binprm *bprm, unsigned *_sp = (*_sp - (MAX_ARG_PAGES * PAGE_SIZE - bprm->p)) & ~15; - out: +out: return ret; -} /* end elf_fdpic_transfer_args_to_stack() */ +} #endif /*****************************************************************************/ @@ -712,17 +744,18 @@ static int elf_fdpic_map_file(struct elf_fdpic_params *params, seg = loadmap->segs; for (loop = loadmap->nsegs; loop > 0; loop--, seg++) { if (params->hdr.e_entry >= seg->p_vaddr && - params->hdr.e_entry < seg->p_vaddr + seg->p_memsz - ) { + params->hdr.e_entry < seg->p_vaddr + seg->p_memsz) { params->entry_addr = - (params->hdr.e_entry - seg->p_vaddr) + seg->addr; + (params->hdr.e_entry - seg->p_vaddr) + + seg->addr; break; } } } /* determine where the program header table has wound up if mapped */ - stop = params->hdr.e_phoff + params->hdr.e_phnum * sizeof (struct elf_phdr); + stop = params->hdr.e_phoff; + stop += params->hdr.e_phnum * sizeof (struct elf_phdr); phdr = params->phdrs; for (loop = 0; loop < params->hdr.e_phnum; loop++, phdr++) { @@ -736,9 +769,11 @@ static int elf_fdpic_map_file(struct elf_fdpic_params *params, seg = loadmap->segs; for (loop = loadmap->nsegs; loop > 0; loop--, seg++) { if (phdr->p_vaddr >= seg->p_vaddr && - phdr->p_vaddr + phdr->p_filesz <= seg->p_vaddr + seg->p_memsz - ) { - params->ph_addr = (phdr->p_vaddr - seg->p_vaddr) + seg->addr + + phdr->p_vaddr + phdr->p_filesz <= + seg->p_vaddr + seg->p_memsz) { + params->ph_addr = + (phdr->p_vaddr - seg->p_vaddr) + + seg->addr + params->hdr.e_phoff - phdr->p_offset; break; } @@ -755,18 +790,22 @@ static int elf_fdpic_map_file(struct elf_fdpic_params *params, seg = loadmap->segs; for (loop = loadmap->nsegs; loop > 0; loop--, seg++) { if (phdr->p_vaddr >= seg->p_vaddr && - phdr->p_vaddr + phdr->p_memsz <= seg->p_vaddr + seg->p_memsz - ) { - params->dynamic_addr = (phdr->p_vaddr - seg->p_vaddr) + seg->addr; - - /* check the dynamic section contains at least one item, and that - * the last item is a NULL entry */ + phdr->p_vaddr + phdr->p_memsz <= + seg->p_vaddr + seg->p_memsz) { + params->dynamic_addr = + (phdr->p_vaddr - seg->p_vaddr) + + seg->addr; + + /* check the dynamic section contains at least + * one item, and that the last item is a NULL + * entry */ if (phdr->p_memsz == 0 || phdr->p_memsz % sizeof(Elf32_Dyn) != 0) goto dynamic_error; tmp = phdr->p_memsz / sizeof(Elf32_Dyn); - if (((Elf32_Dyn *) params->dynamic_addr)[tmp - 1].d_tag != 0) + if (((Elf32_Dyn *) + params->dynamic_addr)[tmp - 1].d_tag != 0) goto dynamic_error; break; } @@ -775,8 +814,8 @@ static int elf_fdpic_map_file(struct elf_fdpic_params *params, } /* now elide adjacent segments in the load map on MMU linux - * - on uClinux the holes between may actually be filled with system stuff or stuff from - * other processes + * - on uClinux the holes between may actually be filled with system + * stuff or stuff from other processes */ #ifdef CONFIG_MMU nloads = loadmap->nsegs; @@ -787,7 +826,9 @@ static int elf_fdpic_map_file(struct elf_fdpic_params *params, if (seg->p_vaddr - mseg->p_vaddr == seg->addr - mseg->addr) { load_addr = PAGE_ALIGN(mseg->addr + mseg->p_memsz); if (load_addr == (seg->addr & PAGE_MASK)) { - mseg->p_memsz += load_addr - (mseg->addr + mseg->p_memsz); + mseg->p_memsz += + load_addr - + (mseg->addr + mseg->p_memsz); mseg->p_memsz += seg->addr & ~PAGE_MASK; mseg->p_memsz += seg->p_memsz; loadmap->nsegs--; @@ -815,20 +856,21 @@ static int elf_fdpic_map_file(struct elf_fdpic_params *params, return 0; - dynamic_error: +dynamic_error: printk("ELF FDPIC %s with invalid DYNAMIC section (inode=%lu)\n", what, file->f_dentry->d_inode->i_ino); return -ELIBBAD; -} /* end elf_fdpic_map_file() */ +} /*****************************************************************************/ /* * map a file with constant displacement under uClinux */ #ifndef CONFIG_MMU -static int elf_fdpic_map_file_constdisp_on_uclinux(struct elf_fdpic_params *params, - struct file *file, - struct mm_struct *mm) +static int elf_fdpic_map_file_constdisp_on_uclinux( + struct elf_fdpic_params *params, + struct file *file, + struct mm_struct *mm) { struct elf32_fdpic_loadseg *seg; struct elf32_phdr *phdr; @@ -839,7 +881,8 @@ static int elf_fdpic_map_file_constdisp_on_uclinux(struct elf_fdpic_params *para load_addr = params->load_addr; seg = params->loadmap->segs; - /* determine the bounds of the contiguous overall allocation we must make */ + /* determine the bounds of the contiguous overall allocation we must + * make */ phdr = params->phdrs; for (loop = 0; loop < params->hdr.e_phnum; loop++, phdr++) { if (params->phdrs[loop].p_type != PT_LOAD) @@ -860,7 +903,7 @@ static int elf_fdpic_map_file_constdisp_on_uclinux(struct elf_fdpic_params *para maddr = do_mmap(NULL, load_addr, top - base, PROT_READ | PROT_WRITE | PROT_EXEC, mflags, 0); up_write(&mm->mmap_sem); - if (IS_ERR((void *) maddr)) + if (IS_ERR_VALUE(maddr)) return (int) maddr; if (load_addr != 0) @@ -878,7 +921,8 @@ static int elf_fdpic_map_file_constdisp_on_uclinux(struct elf_fdpic_params *para seg->p_vaddr = phdr->p_vaddr; seg->p_memsz = phdr->p_memsz; - ret = file->f_op->read(file, (void *) seg->addr, phdr->p_filesz, &fpos); + ret = file->f_op->read(file, (void *) seg->addr, + phdr->p_filesz, &fpos); if (ret < 0) return ret; @@ -895,8 +939,7 @@ static int elf_fdpic_map_file_constdisp_on_uclinux(struct elf_fdpic_params *para if (phdr->p_flags & PF_X) { mm->start_code = seg->addr; mm->end_code = seg->addr + phdr->p_memsz; - } - else if (!mm->start_data) { + } else if (!mm->start_data) { mm->start_data = seg->addr; #ifndef CONFIG_MMU mm->end_data = seg->addr + phdr->p_memsz; @@ -913,7 +956,7 @@ static int elf_fdpic_map_file_constdisp_on_uclinux(struct elf_fdpic_params *para } return 0; -} /* end elf_fdpic_map_file_constdisp_on_uclinux() */ +} #endif /*****************************************************************************/ @@ -974,14 +1017,14 @@ static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params, case ELF_FDPIC_FLAG_CONSTDISP: /* constant displacement - * - can be mapped anywhere, but must be mapped as a unit + * - can be mapped anywhere, but must be mapped as a + * unit */ if (!dvset) { maddr = load_addr; delta_vaddr = phdr->p_vaddr; dvset = 1; - } - else { + } else { maddr = load_addr + phdr->p_vaddr - delta_vaddr; flags |= MAP_FIXED; } @@ -1005,13 +1048,14 @@ static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params, up_write(&mm->mmap_sem); kdebug("mmap[%d] <file> sz=%lx pr=%x fl=%x of=%lx --> %08lx", - loop, phdr->p_memsz + disp, prot, flags, phdr->p_offset - disp, - maddr); + loop, phdr->p_memsz + disp, prot, flags, + phdr->p_offset - disp, maddr); - if (IS_ERR((void *) maddr)) + if (IS_ERR_VALUE(maddr)) return (int) maddr; - if ((params->flags & ELF_FDPIC_FLAG_ARRANGEMENT) == ELF_FDPIC_FLAG_CONTIGUOUS) + if ((params->flags & ELF_FDPIC_FLAG_ARRANGEMENT) == + ELF_FDPIC_FLAG_CONTIGUOUS) load_addr += PAGE_ALIGN(phdr->p_memsz + disp); seg->addr = maddr + disp; @@ -1022,7 +1066,8 @@ static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params, if (phdr->p_offset == 0) params->elfhdr_addr = seg->addr; - /* clear the bit between beginning of mapping and beginning of PT_LOAD */ + /* clear the bit between beginning of mapping and beginning of + * PT_LOAD */ if (prot & PROT_WRITE && disp > 0) { kdebug("clear[%d] ad=%lx sz=%lx", loop, maddr, disp); clear_user((void __user *) maddr, disp); @@ -1038,19 +1083,20 @@ static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params, excess1 = PAGE_SIZE - ((maddr + phdr->p_filesz) & ~PAGE_MASK); #ifdef CONFIG_MMU - if (excess > excess1) { unsigned long xaddr = maddr + phdr->p_filesz + excess1; unsigned long xmaddr; flags |= MAP_FIXED | MAP_ANONYMOUS; down_write(&mm->mmap_sem); - xmaddr = do_mmap(NULL, xaddr, excess - excess1, prot, flags, 0); + xmaddr = do_mmap(NULL, xaddr, excess - excess1, + prot, flags, 0); up_write(&mm->mmap_sem); kdebug("mmap[%d] <anon>" " ad=%lx sz=%lx pr=%x fl=%x of=0 --> %08lx", - loop, xaddr, excess - excess1, prot, flags, xmaddr); + loop, xaddr, excess - excess1, prot, flags, + xmaddr); if (xmaddr != xaddr) return -ENOMEM; @@ -1059,7 +1105,8 @@ static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params, if (prot & PROT_WRITE && excess1 > 0) { kdebug("clear[%d] ad=%lx sz=%lx", loop, maddr + phdr->p_filesz, excess1); - clear_user((void __user *) maddr + phdr->p_filesz, excess1); + clear_user((void __user *) maddr + phdr->p_filesz, + excess1); } #else @@ -1074,8 +1121,7 @@ static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params, if (phdr->p_flags & PF_X) { mm->start_code = maddr; mm->end_code = maddr + phdr->p_memsz; - } - else if (!mm->start_data) { + } else if (!mm->start_data) { mm->start_data = maddr; mm->end_data = maddr + phdr->p_memsz; } @@ -1085,4 +1131,662 @@ static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params, } return 0; -} /* end elf_fdpic_map_file_by_direct_mmap() */ +} + +/*****************************************************************************/ +/* + * ELF-FDPIC core dumper + * + * Modelled on fs/exec.c:aout_core_dump() + * Jeremy Fitzhardinge <jeremy@sw.oz.au> + * + * Modelled on fs/binfmt_elf.c core dumper + */ +#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE) + +/* + * These are the only things you should do on a core-file: use only these + * functions to write out all the necessary info. + */ +static int dump_write(struct file *file, const void *addr, int nr) +{ + return file->f_op->write(file, addr, nr, &file->f_pos) == nr; +} + +static int dump_seek(struct file *file, loff_t off) +{ + if (file->f_op->llseek) { + if (file->f_op->llseek(file, off, SEEK_SET) != off) + return 0; + } else { + file->f_pos = off; + } + return 1; +} + +/* + * Decide whether a segment is worth dumping; default is yes to be + * sure (missing info is worse than too much; etc). + * Personally I'd include everything, and use the coredump limit... + * + * I think we should skip something. But I am not sure how. H.J. + */ +static int maydump(struct vm_area_struct *vma) +{ + /* Do not dump I/O mapped devices or special mappings */ + if (vma->vm_flags & (VM_IO | VM_RESERVED)) { + kdcore("%08lx: %08lx: no (IO)", vma->vm_start, vma->vm_flags); + return 0; + } + + /* If we may not read the contents, don't allow us to dump + * them either. "dump_write()" can't handle it anyway. + */ + if (!(vma->vm_flags & VM_READ)) { + kdcore("%08lx: %08lx: no (!read)", vma->vm_start, vma->vm_flags); + return 0; + } + + /* Dump shared memory only if mapped from an anonymous file. */ + if (vma->vm_flags & VM_SHARED) { + if (vma->vm_file->f_dentry->d_inode->i_nlink == 0) { + kdcore("%08lx: %08lx: no (share)", vma->vm_start, vma->vm_flags); + return 1; + } + + kdcore("%08lx: %08lx: no (share)", vma->vm_start, vma->vm_flags); + return 0; + } + +#ifdef CONFIG_MMU + /* If it hasn't been written to, don't write it out */ + if (!vma->anon_vma) { + kdcore("%08lx: %08lx: no (!anon)", vma->vm_start, vma->vm_flags); + return 0; + } +#endif + + kdcore("%08lx: %08lx: yes", vma->vm_start, vma->vm_flags); + return 1; +} + +/* An ELF note in memory */ +struct memelfnote +{ + const char *name; + int type; + unsigned int datasz; + void *data; +}; + +static int notesize(struct memelfnote *en) +{ + int sz; + + sz = sizeof(struct elf_note); + sz += roundup(strlen(en->name) + 1, 4); + sz += roundup(en->datasz, 4); + + return sz; +} + +/* #define DEBUG */ + +#define DUMP_WRITE(addr, nr) \ + do { if (!dump_write(file, (addr), (nr))) return 0; } while(0) +#define DUMP_SEEK(off) \ + do { if (!dump_seek(file, (off))) return 0; } while(0) + +static int writenote(struct memelfnote *men, struct file *file) +{ + struct elf_note en; + + en.n_namesz = strlen(men->name) + 1; + en.n_descsz = men->datasz; + en.n_type = men->type; + + DUMP_WRITE(&en, sizeof(en)); + DUMP_WRITE(men->name, en.n_namesz); + /* XXX - cast from long long to long to avoid need for libgcc.a */ + DUMP_SEEK(roundup((unsigned long)file->f_pos, 4)); /* XXX */ + DUMP_WRITE(men->data, men->datasz); + DUMP_SEEK(roundup((unsigned long)file->f_pos, 4)); /* XXX */ + + return 1; +} +#undef DUMP_WRITE +#undef DUMP_SEEK + +#define DUMP_WRITE(addr, nr) \ + if ((size += (nr)) > limit || !dump_write(file, (addr), (nr))) \ + goto end_coredump; +#define DUMP_SEEK(off) \ + if (!dump_seek(file, (off))) \ + goto end_coredump; + +static inline void fill_elf_fdpic_header(struct elfhdr *elf, int segs) +{ + memcpy(elf->e_ident, ELFMAG, SELFMAG); + elf->e_ident[EI_CLASS] = ELF_CLASS; + elf->e_ident[EI_DATA] = ELF_DATA; + elf->e_ident[EI_VERSION] = EV_CURRENT; + elf->e_ident[EI_OSABI] = ELF_OSABI; + memset(elf->e_ident+EI_PAD, 0, EI_NIDENT-EI_PAD); + + elf->e_type = ET_CORE; + elf->e_machine = ELF_ARCH; + elf->e_version = EV_CURRENT; + elf->e_entry = 0; + elf->e_phoff = sizeof(struct elfhdr); + elf->e_shoff = 0; + elf->e_flags = ELF_FDPIC_CORE_EFLAGS; + elf->e_ehsize = sizeof(struct elfhdr); + elf->e_phentsize = sizeof(struct elf_phdr); + elf->e_phnum = segs; + elf->e_shentsize = 0; + elf->e_shnum = 0; + elf->e_shstrndx = 0; + return; +} + +static inline void fill_elf_note_phdr(struct elf_phdr *phdr, int sz, loff_t offset) +{ + phdr->p_type = PT_NOTE; + phdr->p_offset = offset; + phdr->p_vaddr = 0; + phdr->p_paddr = 0; + phdr->p_filesz = sz; + phdr->p_memsz = 0; + phdr->p_flags = 0; + phdr->p_align = 0; + return; +} + +static inline void fill_note(struct memelfnote *note, const char *name, int type, + unsigned int sz, void *data) +{ + note->name = name; + note->type = type; + note->datasz = sz; + note->data = data; + return; +} + +/* + * fill up all the fields in prstatus from the given task struct, except + * registers which need to be filled up seperately. + */ +static void fill_prstatus(struct elf_prstatus *prstatus, + struct task_struct *p, long signr) +{ + prstatus->pr_info.si_signo = prstatus->pr_cursig = signr; + prstatus->pr_sigpend = p->pending.signal.sig[0]; + prstatus->pr_sighold = p->blocked.sig[0]; + prstatus->pr_pid = p->pid; + prstatus->pr_ppid = p->parent->pid; + prstatus->pr_pgrp = process_group(p); + prstatus->pr_sid = p->signal->session; + if (thread_group_leader(p)) { + /* + * This is the record for the group leader. Add in the + * cumulative times of previous dead threads. This total + * won't include the time of each live thread whose state + * is included in the core dump. The final total reported + * to our parent process when it calls wait4 will include + * those sums as well as the little bit more time it takes + * this and each other thread to finish dying after the + * core dump synchronization phase. + */ + cputime_to_timeval(cputime_add(p->utime, p->signal->utime), + &prstatus->pr_utime); + cputime_to_timeval(cputime_add(p->stime, p->signal->stime), + &prstatus->pr_stime); + } else { + cputime_to_timeval(p->utime, &prstatus->pr_utime); + cputime_to_timeval(p->stime, &prstatus->pr_stime); + } + cputime_to_timeval(p->signal->cutime, &prstatus->pr_cutime); + cputime_to_timeval(p->signal->cstime, &prstatus->pr_cstime); + + prstatus->pr_exec_fdpic_loadmap = p->mm->context.exec_fdpic_loadmap; + prstatus->pr_interp_fdpic_loadmap = p->mm->context.interp_fdpic_loadmap; +} + +static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p, + struct mm_struct *mm) +{ + unsigned int i, len; + + /* first copy the parameters from user space */ + memset(psinfo, 0, sizeof(struct elf_prpsinfo)); + + len = mm->arg_end - mm->arg_start; + if (len >= ELF_PRARGSZ) + len = ELF_PRARGSZ - 1; + if (copy_from_user(&psinfo->pr_psargs, + (const char __user *) mm->arg_start, len)) + return -EFAULT; + for (i = 0; i < len; i++) + if (psinfo->pr_psargs[i] == 0) + psinfo->pr_psargs[i] = ' '; + psinfo->pr_psargs[len] = 0; + + psinfo->pr_pid = p->pid; + psinfo->pr_ppid = p->parent->pid; + psinfo->pr_pgrp = process_group(p); + psinfo->pr_sid = p->signal->session; + + i = p->state ? ffz(~p->state) + 1 : 0; + psinfo->pr_state = i; + psinfo->pr_sname = (i > 5) ? '.' : "RSDTZW"[i]; + psinfo->pr_zomb = psinfo->pr_sname == 'Z'; + psinfo->pr_nice = task_nice(p); + psinfo->pr_flag = p->flags; + SET_UID(psinfo->pr_uid, p->uid); + SET_GID(psinfo->pr_gid, p->gid); + strncpy(psinfo->pr_fname, p->comm, sizeof(psinfo->pr_fname)); + + return 0; +} + +/* Here is the structure in which status of each thread is captured. */ +struct elf_thread_status +{ + struct list_head list; + struct elf_prstatus prstatus; /* NT_PRSTATUS */ + elf_fpregset_t fpu; /* NT_PRFPREG */ + struct task_struct *thread; +#ifdef ELF_CORE_COPY_XFPREGS + elf_fpxregset_t xfpu; /* NT_PRXFPREG */ +#endif + struct memelfnote notes[3]; + int num_notes; +}; + +/* + * In order to add the specific thread information for the elf file format, + * we need to keep a linked list of every thread's pr_status and then create + * a single section for them in the final core file. + */ +static int elf_dump_thread_status(long signr, struct elf_thread_status *t) +{ + struct task_struct *p = t->thread; + int sz = 0; + + t->num_notes = 0; + + fill_prstatus(&t->prstatus, p, signr); + elf_core_copy_task_regs(p, &t->prstatus.pr_reg); + + fill_note(&t->notes[0], "CORE", NT_PRSTATUS, sizeof(t->prstatus), + &t->prstatus); + t->num_notes++; + sz += notesize(&t->notes[0]); + + t->prstatus.pr_fpvalid = elf_core_copy_task_fpregs(p, NULL, &t->fpu); + if (t->prstatus.pr_fpvalid) { + fill_note(&t->notes[1], "CORE", NT_PRFPREG, sizeof(t->fpu), + &t->fpu); + t->num_notes++; + sz += notesize(&t->notes[1]); + } + +#ifdef ELF_CORE_COPY_XFPREGS + if (elf_core_copy_task_xfpregs(p, &t->xfpu)) { + fill_note(&t->notes[2], "LINUX", NT_PRXFPREG, sizeof(t->xfpu), + &t->xfpu); + t->num_notes++; + sz += notesize(&t->notes[2]); + } +#endif + return sz; +} + +/* + * dump the segments for an MMU process + */ +#ifdef CONFIG_MMU +static int elf_fdpic_dump_segments(struct file *file, struct mm_struct *mm, + size_t *size, unsigned long *limit) +{ + struct vm_area_struct *vma; + + for (vma = current->mm->mmap; vma; vma = vma->vm_next) { + unsigned long addr; + + if (!maydump(vma)) + continue; + + for (addr = vma->vm_start; + addr < vma->vm_end; + addr += PAGE_SIZE + ) { + struct vm_area_struct *vma; + struct page *page; + + if (get_user_pages(current, current->mm, addr, 1, 0, 1, + &page, &vma) <= 0) { + DUMP_SEEK(file->f_pos + PAGE_SIZE); + } + else if (page == ZERO_PAGE(addr)) { + DUMP_SEEK(file->f_pos + PAGE_SIZE); + page_cache_release(page); + } + else { + void *kaddr; + + flush_cache_page(vma, addr, page_to_pfn(page)); + kaddr = kmap(page); + if ((*size += PAGE_SIZE) > *limit || + !dump_write(file, kaddr, PAGE_SIZE) + ) { + kunmap(page); + page_cache_release(page); + return -EIO; + } + kunmap(page); + page_cache_release(page); + } + } + } + + return 0; + +end_coredump: + return -EFBIG; +} +#endif + +/* + * dump the segments for a NOMMU process + */ +#ifndef CONFIG_MMU +static int elf_fdpic_dump_segments(struct file *file, struct mm_struct *mm, + size_t *size, unsigned long *limit) +{ + struct vm_list_struct *vml; + + for (vml = current->mm->context.vmlist; vml; vml = vml->next) { + struct vm_area_struct *vma = vml->vma; + + if (!maydump(vma)) + continue; + + if ((*size += PAGE_SIZE) > *limit) + return -EFBIG; + + if (!dump_write(file, (void *) vma->vm_start, + vma->vm_end - vma->vm_start)) + return -EIO; + } + + return 0; +} +#endif + +/* + * Actual dumper + * + * This is a two-pass process; first we find the offsets of the bits, + * and then they are actually written out. If we run out of core limit + * we just truncate. + */ +static int elf_fdpic_core_dump(long signr, struct pt_regs *regs, + struct file *file) +{ +#define NUM_NOTES 6 + int has_dumped = 0; + mm_segment_t fs; + int segs; + size_t size = 0; + int i; + struct vm_area_struct *vma; + struct elfhdr *elf = NULL; + loff_t offset = 0, dataoff; + unsigned long limit = current->signal->rlim[RLIMIT_CORE].rlim_cur; + int numnote; + struct memelfnote *notes = NULL; + struct elf_prstatus *prstatus = NULL; /* NT_PRSTATUS */ + struct elf_prpsinfo *psinfo = NULL; /* NT_PRPSINFO */ + struct task_struct *g, *p; + LIST_HEAD(thread_list); + struct list_head *t; + elf_fpregset_t *fpu = NULL; +#ifdef ELF_CORE_COPY_XFPREGS + elf_fpxregset_t *xfpu = NULL; +#endif + int thread_status_size = 0; +#ifndef CONFIG_MMU + struct vm_list_struct *vml; +#endif + elf_addr_t *auxv; + + /* + * We no longer stop all VM operations. + * + * This is because those proceses that could possibly change map_count + * or the mmap / vma pages are now blocked in do_exit on current + * finishing this core dump. + * + * Only ptrace can touch these memory addresses, but it doesn't change + * the map_count or the pages allocated. So no possibility of crashing + * exists while dumping the mm->vm_next areas to the core file. + */ + + /* alloc memory for large data structures: too large to be on stack */ + elf = kmalloc(sizeof(*elf), GFP_KERNEL); + if (!elf) + goto cleanup; + prstatus = kzalloc(sizeof(*prstatus), GFP_KERNEL); + if (!prstatus) + goto cleanup; + psinfo = kmalloc(sizeof(*psinfo), GFP_KERNEL); + if (!psinfo) + goto cleanup; + notes = kmalloc(NUM_NOTES * sizeof(struct memelfnote), GFP_KERNEL); + if (!notes) + goto cleanup; + fpu = kmalloc(sizeof(*fpu), GFP_KERNEL); + if (!fpu) + goto cleanup; +#ifdef ELF_CORE_COPY_XFPREGS + xfpu = kmalloc(sizeof(*xfpu), GFP_KERNEL); + if (!xfpu) + goto cleanup; +#endif + + if (signr) { + struct elf_thread_status *tmp; + read_lock(&tasklist_lock); + do_each_thread(g,p) + if (current->mm == p->mm && current != p) { + tmp = kzalloc(sizeof(*tmp), GFP_ATOMIC); + if (!tmp) { + read_unlock(&tasklist_lock); + goto cleanup; + } + INIT_LIST_HEAD(&tmp->list); + tmp->thread = p; + list_add(&tmp->list, &thread_list); + } + while_each_thread(g,p); + read_unlock(&tasklist_lock); + list_for_each(t, &thread_list) { + struct elf_thread_status *tmp; + int sz; + + tmp = list_entry(t, struct elf_thread_status, list); + sz = elf_dump_thread_status(signr, tmp); + thread_status_size += sz; + } + } + + /* now collect the dump for the current */ + fill_prstatus(prstatus, current, signr); + elf_core_copy_regs(&prstatus->pr_reg, regs); + +#ifdef CONFIG_MMU + segs = current->mm->map_count; +#else + segs = 0; + for (vml = current->mm->context.vmlist; vml; vml = vml->next) + segs++; +#endif +#ifdef ELF_CORE_EXTRA_PHDRS + segs += ELF_CORE_EXTRA_PHDRS; +#endif + + /* Set up header */ + fill_elf_fdpic_header(elf, segs + 1); /* including notes section */ + + has_dumped = 1; + current->flags |= PF_DUMPCORE; + + /* + * Set up the notes in similar form to SVR4 core dumps made + * with info from their /proc. + */ + + fill_note(notes + 0, "CORE", NT_PRSTATUS, sizeof(*prstatus), prstatus); + fill_psinfo(psinfo, current->group_leader, current->mm); + fill_note(notes + 1, "CORE", NT_PRPSINFO, sizeof(*psinfo), psinfo); + + numnote = 2; + + auxv = (elf_addr_t *) current->mm->saved_auxv; + + i = 0; + do + i += 2; + while (auxv[i - 2] != AT_NULL); + fill_note(¬es[numnote++], "CORE", NT_AUXV, + i * sizeof(elf_addr_t), auxv); + + /* Try to dump the FPU. */ + if ((prstatus->pr_fpvalid = + elf_core_copy_task_fpregs(current, regs, fpu))) + fill_note(notes + numnote++, + "CORE", NT_PRFPREG, sizeof(*fpu), fpu); +#ifdef ELF_CORE_COPY_XFPREGS + if (elf_core_copy_task_xfpregs(current, xfpu)) + fill_note(notes + numnote++, + "LINUX", NT_PRXFPREG, sizeof(*xfpu), xfpu); +#endif + + fs = get_fs(); + set_fs(KERNEL_DS); + + DUMP_WRITE(elf, sizeof(*elf)); + offset += sizeof(*elf); /* Elf header */ + offset += (segs+1) * sizeof(struct elf_phdr); /* Program headers */ + + /* Write notes phdr entry */ + { + struct elf_phdr phdr; + int sz = 0; + + for (i = 0; i < numnote; i++) + sz += notesize(notes + i); + + sz += thread_status_size; + + fill_elf_note_phdr(&phdr, sz, offset); + offset += sz; + DUMP_WRITE(&phdr, sizeof(phdr)); + } + + /* Page-align dumped data */ + dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE); + + /* write program headers for segments dump */ + for ( +#ifdef CONFIG_MMU + vma = current->mm->mmap; vma; vma = vma->vm_next +#else + vml = current->mm->context.vmlist; vml; vml = vml->next +#endif + ) { + struct elf_phdr phdr; + size_t sz; + +#ifndef CONFIG_MMU + vma = vml->vma; +#endif + + sz = vma->vm_end - vma->vm_start; + + phdr.p_type = PT_LOAD; + phdr.p_offset = offset; + phdr.p_vaddr = vma->vm_start; + phdr.p_paddr = 0; + phdr.p_filesz = maydump(vma) ? sz : 0; + phdr.p_memsz = sz; + offset += phdr.p_filesz; + phdr.p_flags = vma->vm_flags & VM_READ ? PF_R : 0; + if (vma->vm_flags & VM_WRITE) + phdr.p_flags |= PF_W; + if (vma->vm_flags & VM_EXEC) + phdr.p_flags |= PF_X; + phdr.p_align = ELF_EXEC_PAGESIZE; + + DUMP_WRITE(&phdr, sizeof(phdr)); + } + +#ifdef ELF_CORE_WRITE_EXTRA_PHDRS + ELF_CORE_WRITE_EXTRA_PHDRS; +#endif + + /* write out the notes section */ + for (i = 0; i < numnote; i++) + if (!writenote(notes + i, file)) + goto end_coredump; + + /* write out the thread status notes section */ + list_for_each(t, &thread_list) { + struct elf_thread_status *tmp = + list_entry(t, struct elf_thread_status, list); + + for (i = 0; i < tmp->num_notes; i++) + if (!writenote(&tmp->notes[i], file)) + goto end_coredump; + } + + DUMP_SEEK(dataoff); + + if (elf_fdpic_dump_segments(file, current->mm, &size, &limit) < 0) + goto end_coredump; + +#ifdef ELF_CORE_WRITE_EXTRA_DATA + ELF_CORE_WRITE_EXTRA_DATA; +#endif + + if (file->f_pos != offset) { + /* Sanity check */ + printk(KERN_WARNING + "elf_core_dump: file->f_pos (%lld) != offset (%lld)\n", + file->f_pos, offset); + } + +end_coredump: + set_fs(fs); + +cleanup: + while (!list_empty(&thread_list)) { + struct list_head *tmp = thread_list.next; + list_del(tmp); + kfree(list_entry(tmp, struct elf_thread_status, list)); + } + + kfree(elf); + kfree(prstatus); + kfree(psinfo); + kfree(notes); + kfree(fpu); +#ifdef ELF_CORE_COPY_XFPREGS + kfree(xfpu); +#endif + return has_dumped; +#undef NUM_NOTES +} + +#endif /* USE_ELF_CORE_DUMP */ diff --git a/fs/block_dev.c b/fs/block_dev.c index 9633a490dab..045f98854f1 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -739,7 +739,7 @@ static int bd_claim_by_kobject(struct block_device *bdev, void *holder, if (!bo) return -ENOMEM; - mutex_lock(&bdev->bd_mutex); + mutex_lock_nested(&bdev->bd_mutex, BD_MUTEX_PARTITION); res = bd_claim(bdev, holder); if (res || !add_bd_holder(bdev, bo)) free_bd_holder(bo); @@ -764,7 +764,7 @@ static void bd_release_from_kobject(struct block_device *bdev, if (!kobj) return; - mutex_lock(&bdev->bd_mutex); + mutex_lock_nested(&bdev->bd_mutex, BD_MUTEX_PARTITION); bd_release(bdev); if ((bo = del_bd_holder(bdev, kobj))) free_bd_holder(bo); @@ -822,6 +822,22 @@ struct block_device *open_by_devnum(dev_t dev, unsigned mode) EXPORT_SYMBOL(open_by_devnum); +static int +blkdev_get_partition(struct block_device *bdev, mode_t mode, unsigned flags); + +struct block_device *open_partition_by_devnum(dev_t dev, unsigned mode) +{ + struct block_device *bdev = bdget(dev); + int err = -ENOMEM; + int flags = mode & FMODE_WRITE ? O_RDWR : O_RDONLY; + if (bdev) + err = blkdev_get_partition(bdev, mode, flags); + return err ? ERR_PTR(err) : bdev; +} + +EXPORT_SYMBOL(open_partition_by_devnum); + + /* * This routine checks whether a removable media has been changed, * and invalidates all buffer-cache-entries in that case. This @@ -868,7 +884,66 @@ void bd_set_size(struct block_device *bdev, loff_t size) } EXPORT_SYMBOL(bd_set_size); -static int do_open(struct block_device *bdev, struct file *file) +static int __blkdev_put(struct block_device *bdev, unsigned int subclass) +{ + int ret = 0; + struct inode *bd_inode = bdev->bd_inode; + struct gendisk *disk = bdev->bd_disk; + + mutex_lock_nested(&bdev->bd_mutex, subclass); + lock_kernel(); + if (!--bdev->bd_openers) { + sync_blockdev(bdev); + kill_bdev(bdev); + } + if (bdev->bd_contains == bdev) { + if (disk->fops->release) + ret = disk->fops->release(bd_inode, NULL); + } else { + mutex_lock_nested(&bdev->bd_contains->bd_mutex, + subclass + 1); + bdev->bd_contains->bd_part_count--; + mutex_unlock(&bdev->bd_contains->bd_mutex); + } + if (!bdev->bd_openers) { + struct module *owner = disk->fops->owner; + + put_disk(disk); + module_put(owner); + + if (bdev->bd_contains != bdev) { + kobject_put(&bdev->bd_part->kobj); + bdev->bd_part = NULL; + } + bdev->bd_disk = NULL; + bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info; + if (bdev != bdev->bd_contains) + __blkdev_put(bdev->bd_contains, subclass + 1); + bdev->bd_contains = NULL; + } + unlock_kernel(); + mutex_unlock(&bdev->bd_mutex); + bdput(bdev); + return ret; +} + +int blkdev_put(struct block_device *bdev) +{ + return __blkdev_put(bdev, BD_MUTEX_NORMAL); +} +EXPORT_SYMBOL(blkdev_put); + +int blkdev_put_partition(struct block_device *bdev) +{ + return __blkdev_put(bdev, BD_MUTEX_PARTITION); +} +EXPORT_SYMBOL(blkdev_put_partition); + +static int +blkdev_get_whole(struct block_device *bdev, mode_t mode, unsigned flags); + +static int +do_open(struct block_device *bdev, struct file *file, unsigned int subclass) { struct module *owner = NULL; struct gendisk *disk; @@ -885,7 +960,8 @@ static int do_open(struct block_device *bdev, struct file *file) } owner = disk->fops->owner; - mutex_lock(&bdev->bd_mutex); + mutex_lock_nested(&bdev->bd_mutex, subclass); + if (!bdev->bd_openers) { bdev->bd_disk = disk; bdev->bd_contains = bdev; @@ -912,11 +988,11 @@ static int do_open(struct block_device *bdev, struct file *file) ret = -ENOMEM; if (!whole) goto out_first; - ret = blkdev_get(whole, file->f_mode, file->f_flags); + ret = blkdev_get_whole(whole, file->f_mode, file->f_flags); if (ret) goto out_first; bdev->bd_contains = whole; - mutex_lock(&whole->bd_mutex); + mutex_lock_nested(&whole->bd_mutex, BD_MUTEX_WHOLE); whole->bd_part_count++; p = disk->part[part - 1]; bdev->bd_inode->i_data.backing_dev_info = @@ -944,7 +1020,8 @@ static int do_open(struct block_device *bdev, struct file *file) if (bdev->bd_invalidated) rescan_partitions(bdev->bd_disk, bdev); } else { - mutex_lock(&bdev->bd_contains->bd_mutex); + mutex_lock_nested(&bdev->bd_contains->bd_mutex, + BD_MUTEX_PARTITION); bdev->bd_contains->bd_part_count++; mutex_unlock(&bdev->bd_contains->bd_mutex); } @@ -958,7 +1035,7 @@ out_first: bdev->bd_disk = NULL; bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info; if (bdev != bdev->bd_contains) - blkdev_put(bdev->bd_contains); + __blkdev_put(bdev->bd_contains, BD_MUTEX_WHOLE); bdev->bd_contains = NULL; put_disk(disk); module_put(owner); @@ -985,11 +1062,49 @@ int blkdev_get(struct block_device *bdev, mode_t mode, unsigned flags) fake_file.f_dentry = &fake_dentry; fake_dentry.d_inode = bdev->bd_inode; - return do_open(bdev, &fake_file); + return do_open(bdev, &fake_file, BD_MUTEX_NORMAL); } EXPORT_SYMBOL(blkdev_get); +static int +blkdev_get_whole(struct block_device *bdev, mode_t mode, unsigned flags) +{ + /* + * This crockload is due to bad choice of ->open() type. + * It will go away. + * For now, block device ->open() routine must _not_ + * examine anything in 'inode' argument except ->i_rdev. + */ + struct file fake_file = {}; + struct dentry fake_dentry = {}; + fake_file.f_mode = mode; + fake_file.f_flags = flags; + fake_file.f_dentry = &fake_dentry; + fake_dentry.d_inode = bdev->bd_inode; + + return do_open(bdev, &fake_file, BD_MUTEX_WHOLE); +} + +static int +blkdev_get_partition(struct block_device *bdev, mode_t mode, unsigned flags) +{ + /* + * This crockload is due to bad choice of ->open() type. + * It will go away. + * For now, block device ->open() routine must _not_ + * examine anything in 'inode' argument except ->i_rdev. + */ + struct file fake_file = {}; + struct dentry fake_dentry = {}; + fake_file.f_mode = mode; + fake_file.f_flags = flags; + fake_file.f_dentry = &fake_dentry; + fake_dentry.d_inode = bdev->bd_inode; + + return do_open(bdev, &fake_file, BD_MUTEX_PARTITION); +} + static int blkdev_open(struct inode * inode, struct file * filp) { struct block_device *bdev; @@ -1005,7 +1120,7 @@ static int blkdev_open(struct inode * inode, struct file * filp) bdev = bd_acquire(inode); - res = do_open(bdev, filp); + res = do_open(bdev, filp, BD_MUTEX_NORMAL); if (res) return res; @@ -1019,51 +1134,6 @@ static int blkdev_open(struct inode * inode, struct file * filp) return res; } -int blkdev_put(struct block_device *bdev) -{ - int ret = 0; - struct inode *bd_inode = bdev->bd_inode; - struct gendisk *disk = bdev->bd_disk; - - mutex_lock(&bdev->bd_mutex); - lock_kernel(); - if (!--bdev->bd_openers) { - sync_blockdev(bdev); - kill_bdev(bdev); - } - if (bdev->bd_contains == bdev) { - if (disk->fops->release) - ret = disk->fops->release(bd_inode, NULL); - } else { - mutex_lock(&bdev->bd_contains->bd_mutex); - bdev->bd_contains->bd_part_count--; - mutex_unlock(&bdev->bd_contains->bd_mutex); - } - if (!bdev->bd_openers) { - struct module *owner = disk->fops->owner; - - put_disk(disk); - module_put(owner); - - if (bdev->bd_contains != bdev) { - kobject_put(&bdev->bd_part->kobj); - bdev->bd_part = NULL; - } - bdev->bd_disk = NULL; - bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info; - if (bdev != bdev->bd_contains) { - blkdev_put(bdev->bd_contains); - } - bdev->bd_contains = NULL; - } - unlock_kernel(); - mutex_unlock(&bdev->bd_mutex); - bdput(bdev); - return ret; -} - -EXPORT_SYMBOL(blkdev_put); - static int blkdev_close(struct inode * inode, struct file * filp) { struct block_device *bdev = I_BDEV(filp->f_mapping->host); diff --git a/fs/buffer.c b/fs/buffer.c index 3660dcb9759..71649ef9b65 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -470,13 +470,18 @@ out: pass does the actual I/O. */ void invalidate_bdev(struct block_device *bdev, int destroy_dirty_buffers) { + struct address_space *mapping = bdev->bd_inode->i_mapping; + + if (mapping->nrpages == 0) + return; + invalidate_bh_lrus(); /* * FIXME: what about destroy_dirty_buffers? * We really want to use invalidate_inode_pages2() for * that, but not until that's cleaned up. */ - invalidate_inode_pages(bdev->bd_inode->i_mapping); + invalidate_inode_pages(mapping); } /* diff --git a/fs/char_dev.c b/fs/char_dev.c index a4cbc6706ef..3483d3cf808 100644 --- a/fs/char_dev.c +++ b/fs/char_dev.c @@ -182,6 +182,28 @@ int alloc_chrdev_region(dev_t *dev, unsigned baseminor, unsigned count, return 0; } +/** + * register_chrdev() - Register a major number for character devices. + * @major: major device number or 0 for dynamic allocation + * @name: name of this range of devices + * @fops: file operations associated with this devices + * + * If @major == 0 this functions will dynamically allocate a major and return + * its number. + * + * If @major > 0 this function will attempt to reserve a device with the given + * major number and will return zero on success. + * + * Returns a -ve errno on failure. + * + * The name of this device has nothing to do with the name of the device in + * /dev. It only helps to keep track of the different owners of devices. If + * your module name has only one type of devices it's ok to use e.g. the name + * of the module here. + * + * This function registers a range of 256 minor numbers. The first minor number + * is 0. + */ int register_chrdev(unsigned int major, const char *name, const struct file_operations *fops) { diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index a61d17ed182..0feb3bd49cb 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,3 +1,13 @@ +Version 1.45 +------------ +Do not time out lockw calls when using posix extensions. Do not +time out requests if server still responding reasonably fast +on requests on other threads. Improve POSIX locking emulation, +(lock cancel now works, and unlock of merged range works even +to Windows servers now). Fix oops on mount to lanman servers +(win9x, os/2 etc.) when null password. Do not send listxattr +(SMB to query all EAs) if nouser_xattr specified. + Version 1.44 ------------ Rewritten sessionsetup support, including support for legacy SMB diff --git a/fs/cifs/README b/fs/cifs/README index 7986d0d97ac..5f0e1bd64fe 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -408,7 +408,7 @@ A partial list of the supported mount options follows: user_xattr Allow getting and setting user xattrs as OS/2 EAs (extended attributes) to the server (default) e.g. via setfattr and getfattr utilities. - nouser_xattr Do not allow getfattr/setfattr to get/set xattrs + nouser_xattr Do not allow getfattr/setfattr to get/set/list xattrs mapchars Translate six of the seven reserved characters (not backslash) *?<>|: to the remap range (above 0xF000), which also diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index a89efaf78a2..4bc250b2d9f 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -277,7 +277,8 @@ void calc_lanman_hash(struct cifsSesInfo * ses, char * lnm_session_key) return; memset(password_with_pad, 0, CIFS_ENCPWD_SIZE); - strncpy(password_with_pad, ses->password, CIFS_ENCPWD_SIZE); + if(ses->password) + strncpy(password_with_pad, ses->password, CIFS_ENCPWD_SIZE); if((ses->server->secMode & SECMODE_PW_ENCRYPT) == 0) if(extended_security & CIFSSEC_MAY_PLNTXT) { diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index c28ede59994..3cd750029be 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -402,7 +402,6 @@ static struct quotactl_ops cifs_quotactl_ops = { }; #endif -#ifdef CONFIG_CIFS_EXPERIMENTAL static void cifs_umount_begin(struct vfsmount * vfsmnt, int flags) { struct cifs_sb_info *cifs_sb; @@ -422,7 +421,7 @@ static void cifs_umount_begin(struct vfsmount * vfsmnt, int flags) tcon->tidStatus = CifsExiting; up(&tcon->tconSem); - /* cancel_brl_requests(tcon); */ + /* cancel_brl_requests(tcon); */ /* BB mark all brl mids as exiting */ /* cancel_notify_requests(tcon); */ if(tcon->ses && tcon->ses->server) { @@ -438,7 +437,6 @@ static void cifs_umount_begin(struct vfsmount * vfsmnt, int flags) return; } -#endif static int cifs_remount(struct super_block *sb, int *flags, char *data) { @@ -457,9 +455,7 @@ struct super_operations cifs_super_ops = { unless later we add lazy close of inodes or unless the kernel forgets to call us with the same number of releases (closes) as opens */ .show_options = cifs_show_options, -#ifdef CONFIG_CIFS_EXPERIMENTAL .umount_begin = cifs_umount_begin, -#endif .remount_fs = cifs_remount, }; diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 8f75c6f2470..39ee8ef3bde 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -100,5 +100,5 @@ extern ssize_t cifs_getxattr(struct dentry *, const char *, void *, size_t); extern ssize_t cifs_listxattr(struct dentry *, char *, size_t); extern int cifs_ioctl (struct inode * inode, struct file * filep, unsigned int command, unsigned long arg); -#define CIFS_VERSION "1.44" +#define CIFS_VERSION "1.45" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 6d7cf5f3bc0..b24006c47df 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -3,6 +3,7 @@ * * Copyright (C) International Business Machines Corp., 2002,2006 * Author(s): Steve French (sfrench@us.ibm.com) + * Jeremy Allison (jra@samba.org) * * This library is free software; you can redistribute it and/or modify * it under the terms of the GNU Lesser General Public License as published @@ -158,7 +159,8 @@ struct TCP_Server_Info { /* 16th byte of RFC1001 workstation name is always null */ char workstation_RFC1001_name[SERVER_NAME_LEN_WITH_NULL]; __u32 sequence_number; /* needed for CIFS PDU signature */ - char mac_signing_key[CIFS_SESS_KEY_SIZE + 16]; + char mac_signing_key[CIFS_SESS_KEY_SIZE + 16]; + unsigned long lstrp; /* when we got last response from this server */ }; /* @@ -266,14 +268,14 @@ struct cifsTconInfo { }; /* - * This info hangs off the cifsFileInfo structure. This is used to track - * byte stream locks on the file + * This info hangs off the cifsFileInfo structure, pointed to by llist. + * This is used to track byte stream locks on the file */ struct cifsLockInfo { - struct cifsLockInfo *next; - int start; - int length; - int type; + struct list_head llist; /* pointer to next cifsLockInfo */ + __u64 offset; + __u64 length; + __u8 type; }; /* @@ -304,6 +306,8 @@ struct cifsFileInfo { /* lock scope id (0 if none) */ struct file * pfile; /* needed for writepage */ struct inode * pInode; /* needed for oplock break */ + struct semaphore lock_sem; + struct list_head llist; /* list of byte range locks we have. */ unsigned closePend:1; /* file is marked to close */ unsigned invalidHandle:1; /* file closed via session abend */ atomic_t wrtPending; /* handle in use - defer close */ diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index a5ddc62d6fe..b35c55c3c8b 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -50,6 +50,10 @@ extern int SendReceive(const unsigned int /* xid */ , struct cifsSesInfo *, extern int SendReceive2(const unsigned int /* xid */ , struct cifsSesInfo *, struct kvec *, int /* nvec to send */, int * /* type of buf returned */ , const int long_op); +extern int SendReceiveBlockingLock(const unsigned int /* xid */ , struct cifsTconInfo *, + struct smb_hdr * /* input */ , + struct smb_hdr * /* out */ , + int * /* bytes returned */); extern int checkSMBhdr(struct smb_hdr *smb, __u16 mid); extern int checkSMB(struct smb_hdr *smb, __u16 mid, int length); extern int is_valid_oplock_break(struct smb_hdr *smb, struct TCP_Server_Info *); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 19678c575df..075d8fb3d37 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -477,7 +477,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) /* BB get server time for time conversions and add code to use it and timezone since this is not UTC */ - if (rsp->EncryptionKeyLength == CIFS_CRYPTO_KEY_SIZE) { + if (rsp->EncryptionKeyLength == cpu_to_le16(CIFS_CRYPTO_KEY_SIZE)) { memcpy(server->cryptKey, rsp->EncryptionKey, CIFS_CRYPTO_KEY_SIZE); } else if (server->secMode & SECMODE_PW_ENCRYPT) { @@ -1460,8 +1460,13 @@ CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, pSMB->hdr.smb_buf_length += count; pSMB->ByteCount = cpu_to_le16(count); - rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, + if (waitFlag) { + rc = SendReceiveBlockingLock(xid, tcon, (struct smb_hdr *) pSMB, + (struct smb_hdr *) pSMBr, &bytes_returned); + } else { + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, timeout); + } cifs_stats_inc(&tcon->num_locks); if (rc) { cFYI(1, ("Send error in Lock = %d", rc)); @@ -1484,6 +1489,7 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, char *data_offset; struct cifs_posix_lock *parm_data; int rc = 0; + int timeout = 0; int bytes_returned = 0; __u16 params, param_offset, offset, byte_count, count; @@ -1503,7 +1509,6 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, pSMB->MaxSetupCount = 0; pSMB->Reserved = 0; pSMB->Flags = 0; - pSMB->Timeout = 0; pSMB->Reserved2 = 0; param_offset = offsetof(struct smb_com_transaction2_sfi_req, Fid) - 4; offset = param_offset + params; @@ -1529,8 +1534,13 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, (((char *) &pSMB->hdr.Protocol) + offset); parm_data->lock_type = cpu_to_le16(lock_type); - if(waitFlag) + if(waitFlag) { + timeout = 3; /* blocking operation, no timeout */ parm_data->lock_flags = cpu_to_le16(1); + pSMB->Timeout = cpu_to_le32(-1); + } else + pSMB->Timeout = 0; + parm_data->pid = cpu_to_le32(current->tgid); parm_data->start = cpu_to_le64(pLockData->fl_start); parm_data->length = cpu_to_le64(len); /* normalize negative numbers */ @@ -1541,8 +1551,14 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, pSMB->Reserved4 = 0; pSMB->hdr.smb_buf_length += byte_count; pSMB->ByteCount = cpu_to_le16(byte_count); - rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, - (struct smb_hdr *) pSMBr, &bytes_returned, 0); + if (waitFlag) { + rc = SendReceiveBlockingLock(xid, tcon, (struct smb_hdr *) pSMB, + (struct smb_hdr *) pSMBr, &bytes_returned); + } else { + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, + (struct smb_hdr *) pSMBr, &bytes_returned, timeout); + } + if (rc) { cFYI(1, ("Send error in Posix Lock = %d", rc)); } else if (get_flag) { diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 876eb9ef85f..5d394c72686 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -182,6 +182,7 @@ cifs_reconnect(struct TCP_Server_Info *server) while ((server->tcpStatus != CifsExiting) && (server->tcpStatus != CifsGood)) { + try_to_freeze(); if(server->protocolType == IPV6) { rc = ipv6_connect(&server->addr.sockAddr6,&server->ssocket); } else { @@ -612,6 +613,10 @@ multi_t2_fnd: #ifdef CONFIG_CIFS_STATS2 mid_entry->when_received = jiffies; #endif + /* so we do not time out requests to server + which is still responding (since server could + be busy but not dead) */ + server->lstrp = jiffies; break; } } @@ -1266,33 +1271,35 @@ find_unc(__be32 new_target_ip_addr, char *uncName, char *userName) read_lock(&GlobalSMBSeslock); list_for_each(tmp, &GlobalTreeConnectionList) { - cFYI(1, ("Next tcon - ")); + cFYI(1, ("Next tcon")); tcon = list_entry(tmp, struct cifsTconInfo, cifsConnectionList); if (tcon->ses) { if (tcon->ses->server) { cFYI(1, - (" old ip addr: %x == new ip %x ?", + ("old ip addr: %x == new ip %x ?", tcon->ses->server->addr.sockAddr.sin_addr. s_addr, new_target_ip_addr)); if (tcon->ses->server->addr.sockAddr.sin_addr. s_addr == new_target_ip_addr) { - /* BB lock tcon and server and tcp session and increment use count here? */ + /* BB lock tcon, server and tcp session and increment use count here? */ /* found a match on the TCP session */ /* BB check if reconnection needed */ - cFYI(1,("Matched ip, old UNC: %s == new: %s ?", + cFYI(1,("IP match, old UNC: %s new: %s", tcon->treeName, uncName)); if (strncmp (tcon->treeName, uncName, MAX_TREE_SIZE) == 0) { cFYI(1, - ("Matched UNC, old user: %s == new: %s ?", + ("and old usr: %s new: %s", tcon->treeName, uncName)); if (strncmp (tcon->ses->userName, userName, MAX_USERNAME_SIZE) == 0) { read_unlock(&GlobalSMBSeslock); - return tcon;/* also matched user (smb session)*/ + /* matched smb session + (user name */ + return tcon; } } } @@ -1969,7 +1976,18 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, } cFYI(1,("Negotiate caps 0x%x",(int)cap)); - +#ifdef CONFIG_CIFS_DEBUG2 + if(cap & CIFS_UNIX_FCNTL_CAP) + cFYI(1,("FCNTL cap")); + if(cap & CIFS_UNIX_EXTATTR_CAP) + cFYI(1,("EXTATTR cap")); + if(cap & CIFS_UNIX_POSIX_PATHNAMES_CAP) + cFYI(1,("POSIX path cap")); + if(cap & CIFS_UNIX_XATTR_CAP) + cFYI(1,("XATTR cap")); + if(cap & CIFS_UNIX_POSIX_ACL_CAP) + cFYI(1,("POSIX ACL cap")); +#endif /* CIFS_DEBUG2 */ if (CIFSSMBSetFSUnixInfo(xid, tcon, cap)) { cFYI(1,("setting capabilities failed")); } diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index ba4cbe9b068..914239d5363 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -267,6 +267,10 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, pCifsFile->invalidHandle = FALSE; pCifsFile->closePend = FALSE; init_MUTEX(&pCifsFile->fh_sem); + init_MUTEX(&pCifsFile->lock_sem); + INIT_LIST_HEAD(&pCifsFile->llist); + atomic_set(&pCifsFile->wrtPending,0); + /* set the following in open now pCifsFile->pfile = file; */ write_lock(&GlobalSMBSeslock); diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 944d2b9e092..e9c5ba9084f 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -5,6 +5,7 @@ * * Copyright (C) International Business Machines Corp., 2002,2003 * Author(s): Steve French (sfrench@us.ibm.com) + * Jeremy Allison (jra@samba.org) * * This library is free software; you can redistribute it and/or modify * it under the terms of the GNU Lesser General Public License as published @@ -47,6 +48,8 @@ static inline struct cifsFileInfo *cifs_init_private( private_data->netfid = netfid; private_data->pid = current->tgid; init_MUTEX(&private_data->fh_sem); + init_MUTEX(&private_data->lock_sem); + INIT_LIST_HEAD(&private_data->llist); private_data->pfile = file; /* needed for writepage */ private_data->pInode = inode; private_data->invalidHandle = FALSE; @@ -473,6 +476,8 @@ int cifs_close(struct inode *inode, struct file *file) cifs_sb = CIFS_SB(inode->i_sb); pTcon = cifs_sb->tcon; if (pSMBFile) { + struct cifsLockInfo *li, *tmp; + pSMBFile->closePend = TRUE; if (pTcon) { /* no sense reconnecting to close a file that is @@ -496,6 +501,16 @@ int cifs_close(struct inode *inode, struct file *file) pSMBFile->netfid); } } + + /* Delete any outstanding lock records. + We'll lose them when the file is closed anyway. */ + down(&pSMBFile->lock_sem); + list_for_each_entry_safe(li, tmp, &pSMBFile->llist, llist) { + list_del(&li->llist); + kfree(li); + } + up(&pSMBFile->lock_sem); + write_lock(&GlobalSMBSeslock); list_del(&pSMBFile->flist); list_del(&pSMBFile->tlist); @@ -570,6 +585,21 @@ int cifs_closedir(struct inode *inode, struct file *file) return rc; } +static int store_file_lock(struct cifsFileInfo *fid, __u64 len, + __u64 offset, __u8 lockType) +{ + struct cifsLockInfo *li = kmalloc(sizeof(struct cifsLockInfo), GFP_KERNEL); + if (li == NULL) + return -ENOMEM; + li->offset = offset; + li->length = len; + li->type = lockType; + down(&fid->lock_sem); + list_add(&li->llist, &fid->llist); + up(&fid->lock_sem); + return 0; +} + int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) { int rc, xid; @@ -581,6 +611,7 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) struct cifsTconInfo *pTcon; __u16 netfid; __u8 lockType = LOCKING_ANDX_LARGE_FILES; + int posix_locking; length = 1 + pfLock->fl_end - pfLock->fl_start; rc = -EACCES; @@ -639,15 +670,14 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) } netfid = ((struct cifsFileInfo *)file->private_data)->netfid; + posix_locking = (cifs_sb->tcon->ses->capabilities & CAP_UNIX) && + (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(cifs_sb->tcon->fsUnixInfo.Capability)); /* BB add code here to normalize offset and length to account for negative length which we can not accept over the wire */ if (IS_GETLK(cmd)) { - if(experimEnabled && - (cifs_sb->tcon->ses->capabilities & CAP_UNIX) && - (CIFS_UNIX_FCNTL_CAP & - le64_to_cpu(cifs_sb->tcon->fsUnixInfo.Capability))) { + if(posix_locking) { int posix_lock_type; if(lockType & LOCKING_ANDX_SHARED_LOCK) posix_lock_type = CIFS_RDLCK; @@ -683,10 +713,15 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) FreeXid(xid); return rc; } - if (experimEnabled && - (cifs_sb->tcon->ses->capabilities & CAP_UNIX) && - (CIFS_UNIX_FCNTL_CAP & - le64_to_cpu(cifs_sb->tcon->fsUnixInfo.Capability))) { + + if (!numLock && !numUnlock) { + /* if no lock or unlock then nothing + to do since we do not know what it is */ + FreeXid(xid); + return -EOPNOTSUPP; + } + + if (posix_locking) { int posix_lock_type; if(lockType & LOCKING_ANDX_SHARED_LOCK) posix_lock_type = CIFS_RDLCK; @@ -695,18 +730,46 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) if(numUnlock == 1) posix_lock_type = CIFS_UNLCK; - else if(numLock == 0) { - /* if no lock or unlock then nothing - to do since we do not know what it is */ - FreeXid(xid); - return -EOPNOTSUPP; - } + rc = CIFSSMBPosixLock(xid, pTcon, netfid, 0 /* set */, length, pfLock, posix_lock_type, wait_flag); - } else - rc = CIFSSMBLock(xid, pTcon, netfid, length, pfLock->fl_start, - numUnlock, numLock, lockType, wait_flag); + } else { + struct cifsFileInfo *fid = (struct cifsFileInfo *)file->private_data; + + if (numLock) { + rc = CIFSSMBLock(xid, pTcon, netfid, length, pfLock->fl_start, + 0, numLock, lockType, wait_flag); + + if (rc == 0) { + /* For Windows locks we must store them. */ + rc = store_file_lock(fid, length, + pfLock->fl_start, lockType); + } + } else if (numUnlock) { + /* For each stored lock that this unlock overlaps + completely, unlock it. */ + int stored_rc = 0; + struct cifsLockInfo *li, *tmp; + + down(&fid->lock_sem); + list_for_each_entry_safe(li, tmp, &fid->llist, llist) { + if (pfLock->fl_start <= li->offset && + length >= li->length) { + stored_rc = CIFSSMBLock(xid, pTcon, netfid, + li->length, li->offset, + 1, 0, li->type, FALSE); + if (stored_rc) + rc = stored_rc; + + list_del(&li->llist); + kfree(li); + } + } + up(&fid->lock_sem); + } + } + if (pfLock->fl_flags & FL_POSIX) posix_lock_file_wait(file, pfLock); FreeXid(xid); diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index b66eff5dc62..ce87550e918 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -72,6 +72,7 @@ static const struct smb_to_posix_error mapping_table_ERRDOS[] = { {ERRinvlevel,-EOPNOTSUPP}, {ERRdirnotempty, -ENOTEMPTY}, {ERRnotlocked, -ENOLCK}, + {ERRcancelviolation, -ENOLCK}, {ERRalreadyexists, -EEXIST}, {ERRmoredata, -EOVERFLOW}, {ERReasnotsupported,-EOPNOTSUPP}, diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 03bbcb37791..9aeb58a7d36 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -82,7 +82,6 @@ static int construct_dentry(struct qstr *qstring, struct file *file, if(*ptmp_inode == NULL) return rc; rc = 1; - d_instantiate(tmp_dentry, *ptmp_inode); } } else { tmp_dentry = d_alloc(file->f_dentry, qstring); @@ -99,9 +98,7 @@ static int construct_dentry(struct qstr *qstring, struct file *file, tmp_dentry->d_op = &cifs_dentry_ops; if(*ptmp_inode == NULL) return rc; - rc = 1; - d_instantiate(tmp_dentry, *ptmp_inode); - d_rehash(tmp_dentry); + rc = 2; } tmp_dentry->d_time = jiffies; @@ -556,7 +553,7 @@ static int cifs_entry_is_dot(char *current_entry, struct cifsFileInfo *cfile) FIND_FILE_STANDARD_INFO * pFindData = (FIND_FILE_STANDARD_INFO *)current_entry; filename = &pFindData->FileName[0]; - len = le32_to_cpu(pFindData->FileNameLength); + len = pFindData->FileNameLength; } else { cFYI(1,("Unknown findfirst level %d",cfile->srch_inf.info_level)); } @@ -870,6 +867,12 @@ static int cifs_filldir(char *pfindEntry, struct file *file, pfindEntry, &obj_type, rc); else fill_in_inode(tmp_inode, 1 /* NT */, pfindEntry, &obj_type, rc); + + if(rc) /* new inode - needs to be tied to dentry */ { + d_instantiate(tmp_dentry, tmp_inode); + if(rc == 2) + d_rehash(tmp_dentry); + } rc = filldir(direntry,qstring.name,qstring.len,file->f_pos, diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 7202d534ef0..d1705ab8136 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -372,7 +372,7 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, /* no capabilities flags in old lanman negotiation */ - pSMB->old_req.PasswordLength = CIFS_SESS_KEY_SIZE; + pSMB->old_req.PasswordLength = cpu_to_le16(CIFS_SESS_KEY_SIZE); /* BB calculate hash with password */ /* and copy into bcc */ diff --git a/fs/cifs/smberr.h b/fs/cifs/smberr.h index cd41c67ff8d..212c3c29640 100644 --- a/fs/cifs/smberr.h +++ b/fs/cifs/smberr.h @@ -95,6 +95,7 @@ #define ERRinvlevel 124 #define ERRdirnotempty 145 #define ERRnotlocked 158 +#define ERRcancelviolation 173 #define ERRalreadyexists 183 #define ERRbadpipe 230 #define ERRpipebusy 231 diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 17ba329e2b3..48d47b46b1f 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -3,7 +3,8 @@ * * Copyright (C) International Business Machines Corp., 2002,2005 * Author(s): Steve French (sfrench@us.ibm.com) - * + * Jeremy Allison (jra@samba.org) 2006. + * * This library is free software; you can redistribute it and/or modify * it under the terms of the GNU Lesser General Public License as published * by the Free Software Foundation; either version 2.1 of the License, or @@ -36,7 +37,7 @@ extern mempool_t *cifs_mid_poolp; extern kmem_cache_t *cifs_oplock_cachep; static struct mid_q_entry * -AllocMidQEntry(struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) +AllocMidQEntry(const struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) { struct mid_q_entry *temp; @@ -203,6 +204,10 @@ smb_send(struct socket *ssocket, struct smb_hdr *smb_buffer, rc = 0; } + /* Don't want to modify the buffer as a + side effect of this call. */ + smb_buffer->smb_buf_length = smb_buf_length; + return rc; } @@ -217,6 +222,7 @@ smb_send2(struct socket *ssocket, struct kvec *iov, int n_vec, unsigned int len = iov[0].iov_len; unsigned int total_len; int first_vec = 0; + unsigned int smb_buf_length = smb_buffer->smb_buf_length; if(ssocket == NULL) return -ENOTSOCK; /* BB eventually add reconnect code here */ @@ -293,36 +299,15 @@ smb_send2(struct socket *ssocket, struct kvec *iov, int n_vec, } else rc = 0; + /* Don't want to modify the buffer as a + side effect of this call. */ + smb_buffer->smb_buf_length = smb_buf_length; + return rc; } -int -SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, - struct kvec *iov, int n_vec, int * pRespBufType /* ret */, - const int long_op) +static int wait_for_free_request(struct cifsSesInfo *ses, const int long_op) { - int rc = 0; - unsigned int receive_len; - unsigned long timeout; - struct mid_q_entry *midQ; - struct smb_hdr *in_buf = iov[0].iov_base; - - *pRespBufType = CIFS_NO_BUFFER; /* no response buf yet */ - - if ((ses == NULL) || (ses->server == NULL)) { - cifs_small_buf_release(in_buf); - cERROR(1,("Null session")); - return -EIO; - } - - if(ses->server->tcpStatus == CifsExiting) { - cifs_small_buf_release(in_buf); - return -ENOENT; - } - - /* Ensure that we do not send more than 50 overlapping requests - to the same server. We may make this configurable later or - use ses->maxReq */ if(long_op == -1) { /* oplock breaks must not be held up */ atomic_inc(&ses->server->inFlight); @@ -345,53 +330,140 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, } else { if(ses->server->tcpStatus == CifsExiting) { spin_unlock(&GlobalMid_Lock); - cifs_small_buf_release(in_buf); return -ENOENT; } - /* can not count locking commands against total since - they are allowed to block on server */ + /* can not count locking commands against total since + they are allowed to block on server */ - if(long_op < 3) { /* update # of requests on the wire to server */ + if (long_op < 3) atomic_inc(&ses->server->inFlight); - } spin_unlock(&GlobalMid_Lock); break; } } } - /* make sure that we sign in the same order that we send on this socket - and avoid races inside tcp sendmsg code that could cause corruption - of smb data */ - - down(&ses->server->tcpSem); + return 0; +} +static int allocate_mid(struct cifsSesInfo *ses, struct smb_hdr *in_buf, + struct mid_q_entry **ppmidQ) +{ if (ses->server->tcpStatus == CifsExiting) { - rc = -ENOENT; - goto out_unlock2; + return -ENOENT; } else if (ses->server->tcpStatus == CifsNeedReconnect) { cFYI(1,("tcp session dead - return to caller to retry")); - rc = -EAGAIN; - goto out_unlock2; + return -EAGAIN; } else if (ses->status != CifsGood) { /* check if SMB session is bad because we are setting it up */ if((in_buf->Command != SMB_COM_SESSION_SETUP_ANDX) && (in_buf->Command != SMB_COM_NEGOTIATE)) { - rc = -EAGAIN; - goto out_unlock2; + return -EAGAIN; } /* else ok - we are setting up session */ } - midQ = AllocMidQEntry(in_buf, ses); - if (midQ == NULL) { + *ppmidQ = AllocMidQEntry(in_buf, ses); + if (*ppmidQ == NULL) { + return -ENOMEM; + } + return 0; +} + +static int wait_for_response(struct cifsSesInfo *ses, + struct mid_q_entry *midQ, + unsigned long timeout, + unsigned long time_to_wait) +{ + unsigned long curr_timeout; + + for (;;) { + curr_timeout = timeout + jiffies; + wait_event(ses->server->response_q, + (!(midQ->midState == MID_REQUEST_SUBMITTED)) || + time_after(jiffies, curr_timeout) || + ((ses->server->tcpStatus != CifsGood) && + (ses->server->tcpStatus != CifsNew))); + + if (time_after(jiffies, curr_timeout) && + (midQ->midState == MID_REQUEST_SUBMITTED) && + ((ses->server->tcpStatus == CifsGood) || + (ses->server->tcpStatus == CifsNew))) { + + unsigned long lrt; + + /* We timed out. Is the server still + sending replies ? */ + spin_lock(&GlobalMid_Lock); + lrt = ses->server->lstrp; + spin_unlock(&GlobalMid_Lock); + + /* Calculate time_to_wait past last receive time. + Although we prefer not to time out if the + server is still responding - we will time + out if the server takes more than 15 (or 45 + or 180) seconds to respond to this request + and has not responded to any request from + other threads on the client within 10 seconds */ + lrt += time_to_wait; + if (time_after(jiffies, lrt)) { + /* No replies for time_to_wait. */ + cERROR(1,("server not responding")); + return -1; + } + } else { + return 0; + } + } +} + +int +SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, + struct kvec *iov, int n_vec, int * pRespBufType /* ret */, + const int long_op) +{ + int rc = 0; + unsigned int receive_len; + unsigned long timeout; + struct mid_q_entry *midQ; + struct smb_hdr *in_buf = iov[0].iov_base; + + *pRespBufType = CIFS_NO_BUFFER; /* no response buf yet */ + + if ((ses == NULL) || (ses->server == NULL)) { + cifs_small_buf_release(in_buf); + cERROR(1,("Null session")); + return -EIO; + } + + if(ses->server->tcpStatus == CifsExiting) { + cifs_small_buf_release(in_buf); + return -ENOENT; + } + + /* Ensure that we do not send more than 50 overlapping requests + to the same server. We may make this configurable later or + use ses->maxReq */ + + rc = wait_for_free_request(ses, long_op); + if (rc) { + cifs_small_buf_release(in_buf); + return rc; + } + + /* make sure that we sign in the same order that we send on this socket + and avoid races inside tcp sendmsg code that could cause corruption + of smb data */ + + down(&ses->server->tcpSem); + + rc = allocate_mid(ses, in_buf, &midQ); + if (rc) { up(&ses->server->tcpSem); cifs_small_buf_release(in_buf); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } - return -ENOMEM; + /* Update # of requests on wire to server */ + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); + return rc; } rc = cifs_sign_smb2(iov, n_vec, ses->server, &midQ->sequence_number); @@ -406,32 +478,23 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, atomic_dec(&ses->server->inSend); midQ->when_sent = jiffies; #endif - if(rc < 0) { - DeleteMidQEntry(midQ); - up(&ses->server->tcpSem); - cifs_small_buf_release(in_buf); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } - return rc; - } else { - up(&ses->server->tcpSem); - cifs_small_buf_release(in_buf); - } + + up(&ses->server->tcpSem); + cifs_small_buf_release(in_buf); + + if(rc < 0) + goto out; if (long_op == -1) - goto cifs_no_response_exit2; + goto out; else if (long_op == 2) /* writes past end of file can take loong time */ timeout = 180 * HZ; else if (long_op == 1) timeout = 45 * HZ; /* should be greater than servers oplock break timeout (about 43 seconds) */ - else if (long_op > 2) { - timeout = MAX_SCHEDULE_TIMEOUT; - } else + else timeout = 15 * HZ; + /* wait for 15 seconds or until woken up due to response arriving or due to last connection to this server being unmounted */ if (signal_pending(current)) { @@ -441,19 +504,7 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, } /* No user interrupts in wait - wreaks havoc with performance */ - if(timeout != MAX_SCHEDULE_TIMEOUT) { - timeout += jiffies; - wait_event(ses->server->response_q, - (!(midQ->midState & MID_REQUEST_SUBMITTED)) || - time_after(jiffies, timeout) || - ((ses->server->tcpStatus != CifsGood) && - (ses->server->tcpStatus != CifsNew))); - } else { - wait_event(ses->server->response_q, - (!(midQ->midState & MID_REQUEST_SUBMITTED)) || - ((ses->server->tcpStatus != CifsGood) && - (ses->server->tcpStatus != CifsNew))); - } + wait_for_response(ses, midQ, timeout, 10 * HZ); spin_lock(&GlobalMid_Lock); if (midQ->resp_buf) { @@ -481,11 +532,9 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, } spin_unlock(&GlobalMid_Lock); DeleteMidQEntry(midQ); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } + /* Update # of requests on wire to server */ + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); return rc; } @@ -536,24 +585,12 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, cFYI(1,("Bad MID state?")); } } -cifs_no_response_exit2: - DeleteMidQEntry(midQ); - - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } - return rc; +out: -out_unlock2: - up(&ses->server->tcpSem); - cifs_small_buf_release(in_buf); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } + DeleteMidQEntry(midQ); + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); return rc; } @@ -583,85 +620,34 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, /* Ensure that we do not send more than 50 overlapping requests to the same server. We may make this configurable later or use ses->maxReq */ - if(long_op == -1) { - /* oplock breaks must not be held up */ - atomic_inc(&ses->server->inFlight); - } else { - spin_lock(&GlobalMid_Lock); - while(1) { - if(atomic_read(&ses->server->inFlight) >= - cifs_max_pending){ - spin_unlock(&GlobalMid_Lock); -#ifdef CONFIG_CIFS_STATS2 - atomic_inc(&ses->server->num_waiters); -#endif - wait_event(ses->server->request_q, - atomic_read(&ses->server->inFlight) - < cifs_max_pending); -#ifdef CONFIG_CIFS_STATS2 - atomic_dec(&ses->server->num_waiters); -#endif - spin_lock(&GlobalMid_Lock); - } else { - if(ses->server->tcpStatus == CifsExiting) { - spin_unlock(&GlobalMid_Lock); - return -ENOENT; - } - /* can not count locking commands against total since - they are allowed to block on server */ - - if(long_op < 3) { - /* update # of requests on the wire to server */ - atomic_inc(&ses->server->inFlight); - } - spin_unlock(&GlobalMid_Lock); - break; - } - } - } + rc = wait_for_free_request(ses, long_op); + if (rc) + return rc; + /* make sure that we sign in the same order that we send on this socket and avoid races inside tcp sendmsg code that could cause corruption of smb data */ down(&ses->server->tcpSem); - if (ses->server->tcpStatus == CifsExiting) { - rc = -ENOENT; - goto out_unlock; - } else if (ses->server->tcpStatus == CifsNeedReconnect) { - cFYI(1,("tcp session dead - return to caller to retry")); - rc = -EAGAIN; - goto out_unlock; - } else if (ses->status != CifsGood) { - /* check if SMB session is bad because we are setting it up */ - if((in_buf->Command != SMB_COM_SESSION_SETUP_ANDX) && - (in_buf->Command != SMB_COM_NEGOTIATE)) { - rc = -EAGAIN; - goto out_unlock; - } /* else ok - we are setting up session */ - } - midQ = AllocMidQEntry(in_buf, ses); - if (midQ == NULL) { + rc = allocate_mid(ses, in_buf, &midQ); + if (rc) { up(&ses->server->tcpSem); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } - return -ENOMEM; + /* Update # of requests on wire to server */ + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); + return rc; } if (in_buf->smb_buf_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) { - up(&ses->server->tcpSem); cERROR(1, ("Illegal length, greater than maximum frame, %d", in_buf->smb_buf_length)); DeleteMidQEntry(midQ); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } + up(&ses->server->tcpSem); + /* Update # of requests on wire to server */ + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); return -EIO; } @@ -677,27 +663,19 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, atomic_dec(&ses->server->inSend); midQ->when_sent = jiffies; #endif - if(rc < 0) { - DeleteMidQEntry(midQ); - up(&ses->server->tcpSem); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } - return rc; - } else - up(&ses->server->tcpSem); + up(&ses->server->tcpSem); + + if(rc < 0) + goto out; + if (long_op == -1) - goto cifs_no_response_exit; + goto out; else if (long_op == 2) /* writes past end of file can take loong time */ timeout = 180 * HZ; else if (long_op == 1) timeout = 45 * HZ; /* should be greater than servers oplock break timeout (about 43 seconds) */ - else if (long_op > 2) { - timeout = MAX_SCHEDULE_TIMEOUT; - } else + else timeout = 15 * HZ; /* wait for 15 seconds or until woken up due to response arriving or due to last connection to this server being unmounted */ @@ -708,19 +686,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, } /* No user interrupts in wait - wreaks havoc with performance */ - if(timeout != MAX_SCHEDULE_TIMEOUT) { - timeout += jiffies; - wait_event(ses->server->response_q, - (!(midQ->midState & MID_REQUEST_SUBMITTED)) || - time_after(jiffies, timeout) || - ((ses->server->tcpStatus != CifsGood) && - (ses->server->tcpStatus != CifsNew))); - } else { - wait_event(ses->server->response_q, - (!(midQ->midState & MID_REQUEST_SUBMITTED)) || - ((ses->server->tcpStatus != CifsGood) && - (ses->server->tcpStatus != CifsNew))); - } + wait_for_response(ses, midQ, timeout, 10 * HZ); spin_lock(&GlobalMid_Lock); if (midQ->resp_buf) { @@ -748,11 +714,9 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, } spin_unlock(&GlobalMid_Lock); DeleteMidQEntry(midQ); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } + /* Update # of requests on wire to server */ + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); return rc; } @@ -799,23 +763,253 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, cERROR(1,("Bad MID state?")); } } -cifs_no_response_exit: + +out: + DeleteMidQEntry(midQ); + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } + return rc; +} + +/* Send an NT_CANCEL SMB to cause the POSIX blocking lock to return. */ + +static int +send_nt_cancel(struct cifsTconInfo *tcon, struct smb_hdr *in_buf, + struct mid_q_entry *midQ) +{ + int rc = 0; + struct cifsSesInfo *ses = tcon->ses; + __u16 mid = in_buf->Mid; + header_assemble(in_buf, SMB_COM_NT_CANCEL, tcon, 0); + in_buf->Mid = mid; + down(&ses->server->tcpSem); + rc = cifs_sign_smb(in_buf, ses->server, &midQ->sequence_number); + if (rc) { + up(&ses->server->tcpSem); + return rc; + } + rc = smb_send(ses->server->ssocket, in_buf, in_buf->smb_buf_length, + (struct sockaddr *) &(ses->server->addr.sockAddr)); + up(&ses->server->tcpSem); return rc; +} + +/* We send a LOCKINGX_CANCEL_LOCK to cause the Windows + blocking lock to return. */ + +static int +send_lock_cancel(const unsigned int xid, struct cifsTconInfo *tcon, + struct smb_hdr *in_buf, + struct smb_hdr *out_buf) +{ + int bytes_returned; + struct cifsSesInfo *ses = tcon->ses; + LOCK_REQ *pSMB = (LOCK_REQ *)in_buf; + + /* We just modify the current in_buf to change + the type of lock from LOCKING_ANDX_SHARED_LOCK + or LOCKING_ANDX_EXCLUSIVE_LOCK to + LOCKING_ANDX_CANCEL_LOCK. */ + + pSMB->LockType = LOCKING_ANDX_CANCEL_LOCK|LOCKING_ANDX_LARGE_FILES; + pSMB->Timeout = 0; + pSMB->hdr.Mid = GetNextMid(ses->server); + + return SendReceive(xid, ses, in_buf, out_buf, + &bytes_returned, 0); +} -out_unlock: +int +SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon, + struct smb_hdr *in_buf, struct smb_hdr *out_buf, + int *pbytes_returned) +{ + int rc = 0; + int rstart = 0; + unsigned int receive_len; + struct mid_q_entry *midQ; + struct cifsSesInfo *ses; + + if (tcon == NULL || tcon->ses == NULL) { + cERROR(1,("Null smb session")); + return -EIO; + } + ses = tcon->ses; + + if(ses->server == NULL) { + cERROR(1,("Null tcp session")); + return -EIO; + } + + if(ses->server->tcpStatus == CifsExiting) + return -ENOENT; + + /* Ensure that we do not send more than 50 overlapping requests + to the same server. We may make this configurable later or + use ses->maxReq */ + + rc = wait_for_free_request(ses, 3); + if (rc) + return rc; + + /* make sure that we sign in the same order that we send on this socket + and avoid races inside tcp sendmsg code that could cause corruption + of smb data */ + + down(&ses->server->tcpSem); + + rc = allocate_mid(ses, in_buf, &midQ); + if (rc) { + up(&ses->server->tcpSem); + return rc; + } + + if (in_buf->smb_buf_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) { + up(&ses->server->tcpSem); + cERROR(1, ("Illegal length, greater than maximum frame, %d", + in_buf->smb_buf_length)); + DeleteMidQEntry(midQ); + return -EIO; + } + + rc = cifs_sign_smb(in_buf, ses->server, &midQ->sequence_number); + + midQ->midState = MID_REQUEST_SUBMITTED; +#ifdef CONFIG_CIFS_STATS2 + atomic_inc(&ses->server->inSend); +#endif + rc = smb_send(ses->server->ssocket, in_buf, in_buf->smb_buf_length, + (struct sockaddr *) &(ses->server->addr.sockAddr)); +#ifdef CONFIG_CIFS_STATS2 + atomic_dec(&ses->server->inSend); + midQ->when_sent = jiffies; +#endif up(&ses->server->tcpSem); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); + + if(rc < 0) { + DeleteMidQEntry(midQ); + return rc; + } + + /* Wait for a reply - allow signals to interrupt. */ + rc = wait_event_interruptible(ses->server->response_q, + (!(midQ->midState == MID_REQUEST_SUBMITTED)) || + ((ses->server->tcpStatus != CifsGood) && + (ses->server->tcpStatus != CifsNew))); + + /* Were we interrupted by a signal ? */ + if ((rc == -ERESTARTSYS) && + (midQ->midState == MID_REQUEST_SUBMITTED) && + ((ses->server->tcpStatus == CifsGood) || + (ses->server->tcpStatus == CifsNew))) { + + if (in_buf->Command == SMB_COM_TRANSACTION2) { + /* POSIX lock. We send a NT_CANCEL SMB to cause the + blocking lock to return. */ + + rc = send_nt_cancel(tcon, in_buf, midQ); + if (rc) { + DeleteMidQEntry(midQ); + return rc; + } + } else { + /* Windows lock. We send a LOCKINGX_CANCEL_LOCK + to cause the blocking lock to return. */ + + rc = send_lock_cancel(xid, tcon, in_buf, out_buf); + + /* If we get -ENOLCK back the lock may have + already been removed. Don't exit in this case. */ + if (rc && rc != -ENOLCK) { + DeleteMidQEntry(midQ); + return rc; + } + } + + /* Wait 5 seconds for the response. */ + if (wait_for_response(ses, midQ, 5 * HZ, 5 * HZ)==0) { + /* We got the response - restart system call. */ + rstart = 1; + } + } + + spin_lock(&GlobalMid_Lock); + if (midQ->resp_buf) { + spin_unlock(&GlobalMid_Lock); + receive_len = midQ->resp_buf->smb_buf_length; + } else { + cERROR(1,("No response for cmd %d mid %d", + midQ->command, midQ->mid)); + if(midQ->midState == MID_REQUEST_SUBMITTED) { + if(ses->server->tcpStatus == CifsExiting) + rc = -EHOSTDOWN; + else { + ses->server->tcpStatus = CifsNeedReconnect; + midQ->midState = MID_RETRY_NEEDED; + } + } + + if (rc != -EHOSTDOWN) { + if(midQ->midState == MID_RETRY_NEEDED) { + rc = -EAGAIN; + cFYI(1,("marking request for retry")); + } else { + rc = -EIO; + } + } + spin_unlock(&GlobalMid_Lock); + DeleteMidQEntry(midQ); + return rc; } + + if (receive_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) { + cERROR(1, ("Frame too large received. Length: %d Xid: %d", + receive_len, xid)); + rc = -EIO; + } else { /* rcvd frame is ok */ + + if (midQ->resp_buf && out_buf + && (midQ->midState == MID_RESPONSE_RECEIVED)) { + out_buf->smb_buf_length = receive_len; + memcpy((char *)out_buf + 4, + (char *)midQ->resp_buf + 4, + receive_len); + + dump_smb(out_buf, 92); + /* convert the length into a more usable form */ + if((receive_len > 24) && + (ses->server->secMode & (SECMODE_SIGN_REQUIRED | + SECMODE_SIGN_ENABLED))) { + rc = cifs_verify_signature(out_buf, + ses->server->mac_signing_key, + midQ->sequence_number+1); + if(rc) { + cERROR(1,("Unexpected SMB signature")); + /* BB FIXME add code to kill session */ + } + } + + *pbytes_returned = out_buf->smb_buf_length; + + /* BB special case reconnect tid and uid here? */ + rc = map_smb_to_linux_error(out_buf); + /* convert ByteCount if necessary */ + if (receive_len >= + sizeof (struct smb_hdr) - + 4 /* do not count RFC1001 header */ + + (2 * out_buf->WordCount) + 2 /* bcc */ ) + BCC(out_buf) = le16_to_cpu(BCC_LE(out_buf)); + } else { + rc = -EIO; + cERROR(1,("Bad MID state?")); + } + } + DeleteMidQEntry(midQ); + if (rstart && rc == -EACCES) + return -ERESTARTSYS; return rc; } diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index 7754d641775..067648b7179 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -330,11 +330,15 @@ ssize_t cifs_listxattr(struct dentry * direntry, char * data, size_t buf_size) sb = direntry->d_inode->i_sb; if(sb == NULL) return -EIO; - xid = GetXid(); cifs_sb = CIFS_SB(sb); pTcon = cifs_sb->tcon; + if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR) + return -EOPNOTSUPP; + + xid = GetXid(); + full_path = build_path_from_dentry(direntry); if(full_path == NULL) { FreeXid(xid); diff --git a/fs/coda/file.c b/fs/coda/file.c index cc66c681bd1..dbfbcfa5b3c 100644 --- a/fs/coda/file.c +++ b/fs/coda/file.c @@ -136,10 +136,8 @@ int coda_open(struct inode *coda_inode, struct file *coda_file) coda_vfs_stat.open++; cfi = kmalloc(sizeof(struct coda_file_info), GFP_KERNEL); - if (!cfi) { - unlock_kernel(); + if (!cfi) return -ENOMEM; - } lock_kernel(); diff --git a/fs/dcache.c b/fs/dcache.c index c6e3535be19..1b4a3a34ec5 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -38,7 +38,7 @@ int sysctl_vfs_cache_pressure __read_mostly = 100; EXPORT_SYMBOL_GPL(sysctl_vfs_cache_pressure); __cacheline_aligned_in_smp DEFINE_SPINLOCK(dcache_lock); -static seqlock_t rename_lock __cacheline_aligned_in_smp = SEQLOCK_UNLOCKED; +static __cacheline_aligned_in_smp DEFINE_SEQLOCK(rename_lock); EXPORT_SYMBOL(dcache_lock); @@ -1339,10 +1339,10 @@ void d_move(struct dentry * dentry, struct dentry * target) */ if (target < dentry) { spin_lock(&target->d_lock); - spin_lock(&dentry->d_lock); + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); } else { spin_lock(&dentry->d_lock); - spin_lock(&target->d_lock); + spin_lock_nested(&target->d_lock, DENTRY_D_LOCK_NESTED); } /* Move the dentry to the target hash queue, if on different bucket */ diff --git a/fs/direct-io.c b/fs/direct-io.c index 538fb0418fb..5981e17f46f 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -220,7 +220,8 @@ static void dio_complete(struct dio *dio, loff_t offset, ssize_t bytes) if (dio->end_io && dio->result) dio->end_io(dio->iocb, offset, bytes, dio->map_bh.b_private); if (dio->lock_type == DIO_LOCKING) - up_read(&dio->inode->i_alloc_sem); + /* lockdep: non-owner release */ + up_read_non_owner(&dio->inode->i_alloc_sem); } /* @@ -1261,7 +1262,8 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, } if (dio_lock_type == DIO_LOCKING) - down_read(&inode->i_alloc_sem); + /* lockdep: not the owner will release it */ + down_read_non_owner(&inode->i_alloc_sem); } /* diff --git a/fs/efs/symlink.c b/fs/efs/symlink.c index e249cf733a6..1d30d2ff440 100644 --- a/fs/efs/symlink.c +++ b/fs/efs/symlink.c @@ -22,7 +22,7 @@ static int efs_symlink_readpage(struct file *file, struct page *page) err = -ENAMETOOLONG; if (size > 2 * EFS_BLOCKSIZE) - goto fail; + goto fail_notlocked; lock_kernel(); /* read first 512 bytes of link target */ @@ -47,6 +47,7 @@ static int efs_symlink_readpage(struct file *file, struct page *page) return 0; fail: unlock_kernel(); +fail_notlocked: SetPageError(page); kunmap(page); unlock_page(page); diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 9c677bbd0b0..3a3567433b9 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -120,7 +120,7 @@ struct epoll_filefd { */ struct wake_task_node { struct list_head llink; - task_t *task; + struct task_struct *task; wait_queue_head_t *wq; }; @@ -413,7 +413,7 @@ static void ep_poll_safewake(struct poll_safewake *psw, wait_queue_head_t *wq) { int wake_nests = 0; unsigned long flags; - task_t *this_task = current; + struct task_struct *this_task = current; struct list_head *lsthead = &psw->wake_task_list, *lnk; struct wake_task_node *tncur; struct wake_task_node tnode; @@ -1168,7 +1168,7 @@ static int ep_unlink(struct eventpoll *ep, struct epitem *epi) eexit_1: DNPRINTK(3, (KERN_INFO "[%p] eventpoll: ep_unlink(%p, %p) = %d\n", - current, ep, epi->file, error)); + current, ep, epi->ffd.file, error)); return error; } @@ -1236,7 +1236,7 @@ static int ep_poll_callback(wait_queue_t *wait, unsigned mode, int sync, void *k struct eventpoll *ep = epi->ep; DNPRINTK(3, (KERN_INFO "[%p] eventpoll: poll_callback(%p) epi=%p ep=%p\n", - current, epi->file, epi, ep)); + current, epi->ffd.file, epi, ep)); write_lock_irqsave(&ep->lock, flags); diff --git a/fs/exec.c b/fs/exec.c index 8344ba73a2a..54135df2a96 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -486,8 +486,6 @@ struct file *open_exec(const char *name) if (!(nd.mnt->mnt_flags & MNT_NOEXEC) && S_ISREG(inode->i_mode)) { int err = vfs_permission(&nd, MAY_EXEC); - if (!err && !(inode->i_mode & 0111)) - err = -EACCES; file = ERR_PTR(err); if (!err) { file = nameidata_to_filp(&nd, O_RDONLY); @@ -753,7 +751,7 @@ no_thread_group: write_lock_irq(&tasklist_lock); spin_lock(&oldsighand->siglock); - spin_lock(&newsighand->siglock); + spin_lock_nested(&newsighand->siglock, SINGLE_DEPTH_NESTING); rcu_assign_pointer(current->sighand, newsighand); recalc_sigpending(); @@ -922,12 +920,6 @@ int prepare_binprm(struct linux_binprm *bprm) int retval; mode = inode->i_mode; - /* - * Check execute perms again - if the caller has CAP_DAC_OVERRIDE, - * generic_permission lets a non-executable through - */ - if (!(mode & 0111)) /* with at least _one_ execute bit set */ - return -EACCES; if (bprm->file->f_op == NULL) return -EACCES; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 9f43879d6d6..ca5bfb6914d 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -251,6 +251,44 @@ static struct super_operations ext2_sops = { #endif }; +static struct dentry *ext2_get_dentry(struct super_block *sb, void *vobjp) +{ + __u32 *objp = vobjp; + unsigned long ino = objp[0]; + __u32 generation = objp[1]; + struct inode *inode; + struct dentry *result; + + if (ino < EXT2_FIRST_INO(sb) && ino != EXT2_ROOT_INO) + return ERR_PTR(-ESTALE); + if (ino > le32_to_cpu(EXT2_SB(sb)->s_es->s_inodes_count)) + return ERR_PTR(-ESTALE); + + /* iget isn't really right if the inode is currently unallocated!! + * ext2_read_inode currently does appropriate checks, but + * it might be "neater" to call ext2_get_inode first and check + * if the inode is valid..... + */ + inode = iget(sb, ino); + if (inode == NULL) + return ERR_PTR(-ENOMEM); + if (is_bad_inode(inode) || + (generation && inode->i_generation != generation)) { + /* we didn't find the right inode.. */ + iput(inode); + return ERR_PTR(-ESTALE); + } + /* now to find a dentry. + * If possible, get a well-connected one + */ + result = d_alloc_anon(inode); + if (!result) { + iput(inode); + return ERR_PTR(-ENOMEM); + } + return result; +} + /* Yes, most of these are left as NULL!! * A NULL value implies the default, which works with ext2-like file * systems, but can be improved upon. @@ -258,6 +296,7 @@ static struct super_operations ext2_sops = { */ static struct export_operations ext2_export_ops = { .get_parent = ext2_get_parent, + .get_dentry = ext2_get_dentry, }; static unsigned long get_sb_block(void **data) @@ -775,7 +814,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) if (EXT2_INODE_SIZE(sb) == 0) goto cantfind_ext2; sbi->s_inodes_per_block = sb->s_blocksize / EXT2_INODE_SIZE(sb); - if (sbi->s_inodes_per_block == 0) + if (sbi->s_inodes_per_block == 0 || sbi->s_inodes_per_group == 0) goto cantfind_ext2; sbi->s_itb_per_group = sbi->s_inodes_per_group / sbi->s_inodes_per_block; @@ -1157,7 +1196,7 @@ static ssize_t ext2_quota_write(struct super_block *sb, int type, struct buffer_head tmp_bh; struct buffer_head *bh; - mutex_lock(&inode->i_mutex); + mutex_lock_nested(&inode->i_mutex, I_MUTEX_QUOTA); while (towrite > 0) { tocopy = sb->s_blocksize - offset < towrite ? sb->s_blocksize - offset : towrite; diff --git a/fs/ext3/acl.h b/fs/ext3/acl.h index 92d50b53a93..0d1e6279cbf 100644 --- a/fs/ext3/acl.h +++ b/fs/ext3/acl.h @@ -62,9 +62,6 @@ extern int ext3_permission (struct inode *, int, struct nameidata *); extern int ext3_acl_chmod (struct inode *); extern int ext3_init_acl (handle_t *, struct inode *, struct inode *); -extern int init_ext3_acl(void); -extern void exit_ext3_acl(void); - #else /* CONFIG_EXT3_FS_POSIX_ACL */ #include <linux/sched.h> #define ext3_permission NULL diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c index a504a40d6d2..063d994bda0 100644 --- a/fs/ext3/balloc.c +++ b/fs/ext3/balloc.c @@ -1269,12 +1269,12 @@ ext3_fsblk_t ext3_new_blocks(handle_t *handle, struct inode *inode, goal = le32_to_cpu(es->s_first_data_block); group_no = (goal - le32_to_cpu(es->s_first_data_block)) / EXT3_BLOCKS_PER_GROUP(sb); + goal_group = group_no; +retry_alloc: gdp = ext3_get_group_desc(sb, group_no, &gdp_bh); if (!gdp) goto io_error; - goal_group = group_no; -retry: free_blocks = le16_to_cpu(gdp->bg_free_blocks_count); /* * if there is not enough free blocks to make a new resevation @@ -1349,7 +1349,7 @@ retry: if (my_rsv) { my_rsv = NULL; group_no = goal_group; - goto retry; + goto retry_alloc; } /* No space left on the device */ *errp = -ENOSPC; diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index f804d5e9d60..84be02e9365 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -925,7 +925,7 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, set_buffer_new(bh_result); got_it: map_bh(bh_result, inode->i_sb, le32_to_cpu(chain[depth-1].key)); - if (blocks_to_boundary == 0) + if (count > blocks_to_boundary) set_buffer_boundary(bh_result); err = count; /* Clean up and exit */ @@ -1009,11 +1009,14 @@ struct buffer_head *ext3_getblk(handle_t *handle, struct inode *inode, buffer_trace_init(&dummy.b_history); err = ext3_get_blocks_handle(handle, inode, block, 1, &dummy, create, 1); - if (err == 1) { + /* + * ext3_get_blocks_handle() returns number of blocks + * mapped. 0 in case of a HOLE. + */ + if (err > 0) { + if (err > 1) + WARN_ON(1); err = 0; - } else if (err >= 0) { - WARN_ON(1); - err = -EIO; } *errp = err; if (!err && buffer_mapped(&dummy)) { @@ -1158,7 +1161,7 @@ retry: ret = PTR_ERR(handle); goto out; } - if (test_opt(inode->i_sb, NOBH)) + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) ret = nobh_prepare_write(page, from, to, ext3_get_block); else ret = block_prepare_write(page, from, to, ext3_get_block); @@ -1244,7 +1247,7 @@ static int ext3_writeback_commit_write(struct file *file, struct page *page, if (new_i_size > EXT3_I(inode)->i_disksize) EXT3_I(inode)->i_disksize = new_i_size; - if (test_opt(inode->i_sb, NOBH)) + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) ret = nobh_commit_write(file, page, from, to); else ret = generic_commit_write(file, page, from, to); @@ -1494,7 +1497,7 @@ static int ext3_writeback_writepage(struct page *page, goto out_fail; } - if (test_opt(inode->i_sb, NOBH)) + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) ret = nobh_writepage(page, ext3_get_block, wbc); else ret = block_write_full_page(page, ext3_get_block, wbc); @@ -2402,14 +2405,15 @@ static ext3_fsblk_t ext3_get_inode_block(struct super_block *sb, struct buffer_head *bh; struct ext3_group_desc * gdp; - - if ((ino != EXT3_ROOT_INO && ino != EXT3_JOURNAL_INO && - ino != EXT3_RESIZE_INO && ino < EXT3_FIRST_INO(sb)) || - ino > le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)) { - ext3_error(sb, "ext3_get_inode_block", - "bad inode number: %lu", ino); + if (!ext3_valid_inum(sb, ino)) { + /* + * This error is already checked for in namei.c unless we are + * looking at an NFS filehandle, in which case no error + * report is needed + */ return 0; } + block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); if (block_group >= EXT3_SB(sb)->s_groups_count) { ext3_error(sb,"ext3_get_inode_block","group >= groups count"); diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index d9176dba369..2aa7101b27c 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -1000,7 +1000,12 @@ static struct dentry *ext3_lookup(struct inode * dir, struct dentry *dentry, str if (bh) { unsigned long ino = le32_to_cpu(de->inode); brelse (bh); - inode = iget(dir->i_sb, ino); + if (!ext3_valid_inum(dir->i_sb, ino)) { + ext3_error(dir->i_sb, "ext3_lookup", + "bad inode number: %lu", ino); + inode = NULL; + } else + inode = iget(dir->i_sb, ino); if (!inode) return ERR_PTR(-EACCES); @@ -1028,7 +1033,13 @@ struct dentry *ext3_get_parent(struct dentry *child) return ERR_PTR(-ENOENT); ino = le32_to_cpu(de->inode); brelse(bh); - inode = iget(child->d_inode->i_sb, ino); + + if (!ext3_valid_inum(child->d_inode->i_sb, ino)) { + ext3_error(child->d_inode->i_sb, "ext3_get_parent", + "bad inode number: %lu", ino); + inode = NULL; + } else + inode = iget(child->d_inode->i_sb, ino); if (!inode) return ERR_PTR(-EACCES); diff --git a/fs/ext3/super.c b/fs/ext3/super.c index f2dd7133661..3559086eee5 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -554,6 +554,47 @@ static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs) return 0; } + +static struct dentry *ext3_get_dentry(struct super_block *sb, void *vobjp) +{ + __u32 *objp = vobjp; + unsigned long ino = objp[0]; + __u32 generation = objp[1]; + struct inode *inode; + struct dentry *result; + + if (ino < EXT3_FIRST_INO(sb) && ino != EXT3_ROOT_INO) + return ERR_PTR(-ESTALE); + if (ino > le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)) + return ERR_PTR(-ESTALE); + + /* iget isn't really right if the inode is currently unallocated!! + * + * ext3_read_inode will return a bad_inode if the inode had been + * deleted, so we should be safe. + * + * Currently we don't know the generation for parent directory, so + * a generation of 0 means "accept any" + */ + inode = iget(sb, ino); + if (inode == NULL) + return ERR_PTR(-ENOMEM); + if (is_bad_inode(inode) || + (generation && inode->i_generation != generation)) { + iput(inode); + return ERR_PTR(-ESTALE); + } + /* now to find a dentry. + * If possible, get a well-connected one + */ + result = d_alloc_anon(inode); + if (!result) { + iput(inode); + return ERR_PTR(-ENOMEM); + } + return result; +} + #ifdef CONFIG_QUOTA #define QTYPE2NAME(t) ((t)==USRQUOTA?"user":"group") #define QTYPE2MOPT(on, t) ((t)==USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA)) @@ -622,6 +663,7 @@ static struct super_operations ext3_sops = { static struct export_operations ext3_export_ops = { .get_parent = ext3_get_parent, + .get_dentry = ext3_get_dentry, }; enum { @@ -2614,7 +2656,7 @@ static ssize_t ext3_quota_write(struct super_block *sb, int type, struct buffer_head *bh; handle_t *handle = journal_current_handle(); - mutex_lock(&inode->i_mutex); + mutex_lock_nested(&inode->i_mutex, I_MUTEX_QUOTA); while (towrite > 0) { tocopy = sb->s_blocksize - offset < towrite ? sb->s_blocksize - offset : towrite; diff --git a/fs/file.c b/fs/file.c index 55f4e702256..b3c6b82e6a9 100644 --- a/fs/file.c +++ b/fs/file.c @@ -240,13 +240,9 @@ static struct fdtable *alloc_fdtable(int nr) if (!fdt) goto out; - nfds = 8 * L1_CACHE_BYTES; - /* Expand to the max in easy steps */ - while (nfds <= nr) { - nfds = nfds * 2; - if (nfds > NR_OPEN) - nfds = NR_OPEN; - } + nfds = max_t(int, 8 * L1_CACHE_BYTES, roundup_pow_of_two(nr + 1)); + if (nfds > NR_OPEN) + nfds = NR_OPEN; new_openset = alloc_fdset(nfds); new_execset = alloc_fdset(nfds); @@ -277,11 +273,13 @@ static struct fdtable *alloc_fdtable(int nr) } while (nfds <= nr); new_fds = alloc_fd_array(nfds); if (!new_fds) - goto out; + goto out2; fdt->fd = new_fds; fdt->max_fds = nfds; fdt->free_files = NULL; return fdt; +out2: + nfds = fdt->max_fdset; out: if (new_openset) free_fdset(new_openset, nfds); diff --git a/fs/freevxfs/vxfs_lookup.c b/fs/freevxfs/vxfs_lookup.c index 29cce456c7c..43886fa00a2 100644 --- a/fs/freevxfs/vxfs_lookup.c +++ b/fs/freevxfs/vxfs_lookup.c @@ -246,6 +246,8 @@ vxfs_readdir(struct file *fp, void *retp, filldir_t filler) u_long page, npages, block, pblocks, nblocks, offset; loff_t pos; + lock_kernel(); + switch ((long)fp->f_pos) { case 0: if (filler(retp, ".", 1, fp->f_pos, ip->i_ino, DT_DIR) < 0) diff --git a/fs/fuse/control.c b/fs/fuse/control.c index a3bce3a7725..46fe60b2da2 100644 --- a/fs/fuse/control.c +++ b/fs/fuse/control.c @@ -105,7 +105,7 @@ static struct dentry *fuse_ctl_add_dentry(struct dentry *parent, /* * Add a connection to the control filesystem (if it exists). Caller - * must host fuse_mutex + * must hold fuse_mutex */ int fuse_ctl_add_conn(struct fuse_conn *fc) { @@ -139,7 +139,7 @@ int fuse_ctl_add_conn(struct fuse_conn *fc) /* * Remove a connection from the control filesystem (if it exists). - * Caller must host fuse_mutex + * Caller must hold fuse_mutex */ void fuse_ctl_remove_conn(struct fuse_conn *fc) { diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 72a74cde6de..409ce6a7cca 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -14,6 +14,33 @@ #include <linux/sched.h> #include <linux/namei.h> +#if BITS_PER_LONG >= 64 +static inline void fuse_dentry_settime(struct dentry *entry, u64 time) +{ + entry->d_time = time; +} + +static inline u64 fuse_dentry_time(struct dentry *entry) +{ + return entry->d_time; +} +#else +/* + * On 32 bit archs store the high 32 bits of time in d_fsdata + */ +static void fuse_dentry_settime(struct dentry *entry, u64 time) +{ + entry->d_time = time; + entry->d_fsdata = (void *) (unsigned long) (time >> 32); +} + +static u64 fuse_dentry_time(struct dentry *entry) +{ + return (u64) entry->d_time + + ((u64) (unsigned long) entry->d_fsdata << 32); +} +#endif + /* * FUSE caches dentries and attributes with separate timeout. The * time in jiffies until the dentry/attributes are valid is stored in @@ -23,10 +50,13 @@ /* * Calculate the time in jiffies until a dentry/attributes are valid */ -static unsigned long time_to_jiffies(unsigned long sec, unsigned long nsec) +static u64 time_to_jiffies(unsigned long sec, unsigned long nsec) { - struct timespec ts = {sec, nsec}; - return jiffies + timespec_to_jiffies(&ts); + if (sec || nsec) { + struct timespec ts = {sec, nsec}; + return get_jiffies_64() + timespec_to_jiffies(&ts); + } else + return 0; } /* @@ -35,7 +65,8 @@ static unsigned long time_to_jiffies(unsigned long sec, unsigned long nsec) */ static void fuse_change_timeout(struct dentry *entry, struct fuse_entry_out *o) { - entry->d_time = time_to_jiffies(o->entry_valid, o->entry_valid_nsec); + fuse_dentry_settime(entry, + time_to_jiffies(o->entry_valid, o->entry_valid_nsec)); if (entry->d_inode) get_fuse_inode(entry->d_inode)->i_time = time_to_jiffies(o->attr_valid, o->attr_valid_nsec); @@ -47,7 +78,7 @@ static void fuse_change_timeout(struct dentry *entry, struct fuse_entry_out *o) */ void fuse_invalidate_attr(struct inode *inode) { - get_fuse_inode(inode)->i_time = jiffies - 1; + get_fuse_inode(inode)->i_time = 0; } /* @@ -60,7 +91,7 @@ void fuse_invalidate_attr(struct inode *inode) */ static void fuse_invalidate_entry_cache(struct dentry *entry) { - entry->d_time = jiffies - 1; + fuse_dentry_settime(entry, 0); } /* @@ -102,7 +133,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) if (inode && is_bad_inode(inode)) return 0; - else if (time_after(jiffies, entry->d_time)) { + else if (fuse_dentry_time(entry) < get_jiffies_64()) { int err; struct fuse_entry_out outarg; struct fuse_conn *fc; @@ -666,7 +697,7 @@ static int fuse_revalidate(struct dentry *entry) if (!fuse_allow_task(fc, current)) return -EACCES; if (get_node_id(inode) != FUSE_ROOT_ID && - time_before_eq(jiffies, fi->i_time)) + fi->i_time >= get_jiffies_64()) return 0; return fuse_do_getattr(inode); diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 63614ed1633..5c4fcd1dbf5 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -395,14 +395,16 @@ static int fuse_readpages(struct file *file, struct address_space *mapping, struct fuse_readpages_data data; int err; + err = -EIO; if (is_bad_inode(inode)) - return -EIO; + goto clean_pages_up; data.file = file; data.inode = inode; data.req = fuse_get_req(fc); + err = PTR_ERR(data.req); if (IS_ERR(data.req)) - return PTR_ERR(data.req); + goto clean_pages_up; err = read_cache_pages(mapping, pages, fuse_readpages_fill, &data); if (!err) { @@ -412,6 +414,10 @@ static int fuse_readpages(struct file *file, struct address_space *mapping, fuse_put_request(fc, data.req); } return err; + +clean_pages_up: + put_pages_list(pages); + return err; } static size_t fuse_send_write(struct fuse_req *req, struct file *file, diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 0dbf9662184..69c7750d55b 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -59,7 +59,7 @@ struct fuse_inode { struct fuse_req *forget_req; /** Time in jiffies until the file attributes are valid */ - unsigned long i_time; + u64 i_time; }; /** FUSE specific file data */ diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index dcaaabd3b9c..7d25092262a 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -51,7 +51,7 @@ static struct inode *fuse_alloc_inode(struct super_block *sb) return NULL; fi = get_fuse_inode(inode); - fi->i_time = jiffies - 1; + fi->i_time = 0; fi->nodeid = 0; fi->nlookup = 0; fi->forget_req = fuse_request_alloc(); diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 6449cb69796..c3920c96dad 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -83,8 +83,6 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) ret = -ENOMEM; len = vma_len + ((loff_t)vma->vm_pgoff << PAGE_SHIFT); - if (!(vma->vm_flags & VM_WRITE) && len > inode->i_size) - goto out; if (vma->vm_flags & VM_MAYSHARE && hugetlb_reserve_pages(inode, vma->vm_pgoff >> (HPAGE_SHIFT-PAGE_SHIFT), @@ -93,7 +91,7 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) ret = 0; hugetlb_prefault_arch_hook(vma->vm_mm); - if (inode->i_size < len) + if (vma->vm_flags & VM_WRITE && inode->i_size < len) inode->i_size = len; out: mutex_unlock(&inode->i_mutex); diff --git a/fs/inotify_user.c b/fs/inotify_user.c index f2386442ade..017cb0f134d 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -187,7 +187,7 @@ static struct inotify_kernel_event * kernel_event(s32 wd, u32 mask, u32 cookie, { struct inotify_kernel_event *kevent; - kevent = kmem_cache_alloc(event_cachep, GFP_KERNEL); + kevent = kmem_cache_alloc(event_cachep, GFP_NOFS); if (unlikely(!kevent)) return NULL; diff --git a/fs/ioprio.c b/fs/ioprio.c index 93aa5715f22..78b1deae3fa 100644 --- a/fs/ioprio.c +++ b/fs/ioprio.c @@ -44,6 +44,9 @@ static int set_task_ioprio(struct task_struct *task, int ioprio) task->ioprio = ioprio; ioc = task->io_context; + /* see wmb() in current_io_context() */ + smp_read_barrier_depends(); + if (ioc && ioc->set_ioprio) ioc->set_ioprio(ioc, ioprio); @@ -111,9 +114,9 @@ asmlinkage long sys_ioprio_set(int which, int who, int ioprio) continue; ret = set_task_ioprio(p, ioprio); if (ret) - break; + goto free_uid; } while_each_thread(g, p); - +free_uid: if (who) free_uid(user); break; @@ -137,6 +140,29 @@ out: return ret; } +int ioprio_best(unsigned short aprio, unsigned short bprio) +{ + unsigned short aclass = IOPRIO_PRIO_CLASS(aprio); + unsigned short bclass = IOPRIO_PRIO_CLASS(bprio); + + if (!ioprio_valid(aprio)) + return bprio; + if (!ioprio_valid(bprio)) + return aprio; + + if (aclass == IOPRIO_CLASS_NONE) + aclass = IOPRIO_CLASS_BE; + if (bclass == IOPRIO_CLASS_NONE) + bclass = IOPRIO_CLASS_BE; + + if (aclass == bclass) + return min(aprio, bprio); + if (aclass > bclass) + return bprio; + else + return aprio; +} + asmlinkage long sys_ioprio_get(int which, int who) { struct task_struct *g, *p; diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index 0971814c38b..42da6078431 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -261,7 +261,7 @@ void journal_commit_transaction(journal_t *journal) struct buffer_head *bh = jh2bh(jh); jbd_lock_bh_state(bh); - kfree(jh->b_committed_data); + jbd_slab_free(jh->b_committed_data, bh->b_size); jh->b_committed_data = NULL; jbd_unlock_bh_state(bh); } @@ -745,14 +745,14 @@ restart_loop: * Otherwise, we can just throw away the frozen data now. */ if (jh->b_committed_data) { - kfree(jh->b_committed_data); + jbd_slab_free(jh->b_committed_data, bh->b_size); jh->b_committed_data = NULL; if (jh->b_frozen_data) { jh->b_committed_data = jh->b_frozen_data; jh->b_frozen_data = NULL; } } else if (jh->b_frozen_data) { - kfree(jh->b_frozen_data); + jbd_slab_free(jh->b_frozen_data, bh->b_size); jh->b_frozen_data = NULL; } diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index 8c9b28dff11..f66724ce443 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -84,6 +84,7 @@ EXPORT_SYMBOL(journal_force_commit); static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *); static void __journal_abort_soft (journal_t *journal, int errno); +static int journal_create_jbd_slab(size_t slab_size); /* * Helper function used to manage commit timeouts @@ -328,10 +329,10 @@ repeat: char *tmp; jbd_unlock_bh_state(bh_in); - tmp = jbd_rep_kmalloc(bh_in->b_size, GFP_NOFS); + tmp = jbd_slab_alloc(bh_in->b_size, GFP_NOFS); jbd_lock_bh_state(bh_in); if (jh_in->b_frozen_data) { - kfree(tmp); + jbd_slab_free(tmp, bh_in->b_size); goto repeat; } @@ -1069,17 +1070,17 @@ static int load_superblock(journal_t *journal) int journal_load(journal_t *journal) { int err; + journal_superblock_t *sb; err = load_superblock(journal); if (err) return err; + sb = journal->j_superblock; /* If this is a V2 superblock, then we have to check the * features flags on it. */ if (journal->j_format_version >= 2) { - journal_superblock_t *sb = journal->j_superblock; - if ((sb->s_feature_ro_compat & ~cpu_to_be32(JFS_KNOWN_ROCOMPAT_FEATURES)) || (sb->s_feature_incompat & @@ -1090,6 +1091,13 @@ int journal_load(journal_t *journal) } } + /* + * Create a slab for this blocksize + */ + err = journal_create_jbd_slab(cpu_to_be32(sb->s_blocksize)); + if (err) + return err; + /* Let the recovery code check whether it needs to recover any * data from the journal. */ if (journal_recover(journal)) @@ -1612,6 +1620,77 @@ void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry) } /* + * jbd slab management: create 1k, 2k, 4k, 8k slabs as needed + * and allocate frozen and commit buffers from these slabs. + * + * Reason for doing this is to avoid, SLAB_DEBUG - since it could + * cause bh to cross page boundary. + */ + +#define JBD_MAX_SLABS 5 +#define JBD_SLAB_INDEX(size) (size >> 11) + +static kmem_cache_t *jbd_slab[JBD_MAX_SLABS]; +static const char *jbd_slab_names[JBD_MAX_SLABS] = { + "jbd_1k", "jbd_2k", "jbd_4k", NULL, "jbd_8k" +}; + +static void journal_destroy_jbd_slabs(void) +{ + int i; + + for (i = 0; i < JBD_MAX_SLABS; i++) { + if (jbd_slab[i]) + kmem_cache_destroy(jbd_slab[i]); + jbd_slab[i] = NULL; + } +} + +static int journal_create_jbd_slab(size_t slab_size) +{ + int i = JBD_SLAB_INDEX(slab_size); + + BUG_ON(i >= JBD_MAX_SLABS); + + /* + * Check if we already have a slab created for this size + */ + if (jbd_slab[i]) + return 0; + + /* + * Create a slab and force alignment to be same as slabsize - + * this will make sure that allocations won't cross the page + * boundary. + */ + jbd_slab[i] = kmem_cache_create(jbd_slab_names[i], + slab_size, slab_size, 0, NULL, NULL); + if (!jbd_slab[i]) { + printk(KERN_EMERG "JBD: no memory for jbd_slab cache\n"); + return -ENOMEM; + } + return 0; +} + +void * jbd_slab_alloc(size_t size, gfp_t flags) +{ + int idx; + + idx = JBD_SLAB_INDEX(size); + BUG_ON(jbd_slab[idx] == NULL); + return kmem_cache_alloc(jbd_slab[idx], flags | __GFP_NOFAIL); +} + +void jbd_slab_free(void *ptr, size_t size) +{ + int idx; + + idx = JBD_SLAB_INDEX(size); + BUG_ON(jbd_slab[idx] == NULL); + kmem_cache_free(jbd_slab[idx], ptr); +} + +/* * Journal_head storage management */ static kmem_cache_t *journal_head_cache; @@ -1799,13 +1878,13 @@ static void __journal_remove_journal_head(struct buffer_head *bh) printk(KERN_WARNING "%s: freeing " "b_frozen_data\n", __FUNCTION__); - kfree(jh->b_frozen_data); + jbd_slab_free(jh->b_frozen_data, bh->b_size); } if (jh->b_committed_data) { printk(KERN_WARNING "%s: freeing " "b_committed_data\n", __FUNCTION__); - kfree(jh->b_committed_data); + jbd_slab_free(jh->b_committed_data, bh->b_size); } bh->b_private = NULL; jh->b_bh = NULL; /* debug, really */ @@ -1961,6 +2040,7 @@ static void journal_destroy_caches(void) journal_destroy_revoke_caches(); journal_destroy_journal_head_cache(); journal_destroy_handle_cache(); + journal_destroy_jbd_slabs(); } static int __init journal_init(void) diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index 508b2ea91f4..f5169a96260 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -666,8 +666,9 @@ repeat: if (!frozen_buffer) { JBUFFER_TRACE(jh, "allocate memory for buffer"); jbd_unlock_bh_state(bh); - frozen_buffer = jbd_kmalloc(jh2bh(jh)->b_size, - GFP_NOFS); + frozen_buffer = + jbd_slab_alloc(jh2bh(jh)->b_size, + GFP_NOFS); if (!frozen_buffer) { printk(KERN_EMERG "%s: OOM for frozen_buffer\n", @@ -726,7 +727,7 @@ done: out: if (unlikely(frozen_buffer)) /* It's usually NULL */ - kfree(frozen_buffer); + jbd_slab_free(frozen_buffer, bh->b_size); JBUFFER_TRACE(jh, "exit"); return error; @@ -879,7 +880,7 @@ int journal_get_undo_access(handle_t *handle, struct buffer_head *bh) repeat: if (!jh->b_committed_data) { - committed_data = jbd_kmalloc(jh2bh(jh)->b_size, GFP_NOFS); + committed_data = jbd_slab_alloc(jh2bh(jh)->b_size, GFP_NOFS); if (!committed_data) { printk(KERN_EMERG "%s: No memory for committed data\n", __FUNCTION__); @@ -906,7 +907,7 @@ repeat: out: journal_put_journal_head(jh); if (unlikely(committed_data)) - kfree(committed_data); + jbd_slab_free(committed_data, bh->b_size); return err; } diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c index 9c2077e7e08..0ae3cd10702 100644 --- a/fs/jffs2/acl.c +++ b/fs/jffs2/acl.c @@ -345,10 +345,8 @@ int jffs2_init_acl(struct inode *inode, struct inode *dir) return rc; } -void jffs2_clear_acl(struct inode *inode) +void jffs2_clear_acl(struct jffs2_inode_info *f) { - struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode); - if (f->i_acl_access && f->i_acl_access != JFFS2_ACL_NOT_CACHED) { posix_acl_release(f->i_acl_access); f->i_acl_access = JFFS2_ACL_NOT_CACHED; diff --git a/fs/jffs2/acl.h b/fs/jffs2/acl.h index 8893bd1a6ba..fa327dbd317 100644 --- a/fs/jffs2/acl.h +++ b/fs/jffs2/acl.h @@ -30,7 +30,7 @@ struct jffs2_acl_header { extern int jffs2_permission(struct inode *, int, struct nameidata *); extern int jffs2_acl_chmod(struct inode *); extern int jffs2_init_acl(struct inode *, struct inode *); -extern void jffs2_clear_acl(struct inode *); +extern void jffs2_clear_acl(struct jffs2_inode_info *); extern struct xattr_handler jffs2_acl_access_xattr_handler; extern struct xattr_handler jffs2_acl_default_xattr_handler; @@ -40,6 +40,6 @@ extern struct xattr_handler jffs2_acl_default_xattr_handler; #define jffs2_permission NULL #define jffs2_acl_chmod(inode) (0) #define jffs2_init_acl(inode,dir) (0) -#define jffs2_clear_acl(inode) +#define jffs2_clear_acl(f) #endif /* CONFIG_JFFS2_FS_POSIX_ACL */ diff --git a/fs/jffs2/malloc.c b/fs/jffs2/malloc.c index 8310c95478e..33f29100501 100644 --- a/fs/jffs2/malloc.c +++ b/fs/jffs2/malloc.c @@ -190,7 +190,7 @@ void jffs2_free_tmp_dnode_info(struct jffs2_tmp_dnode_info *x) kmem_cache_free(tmp_dnode_info_slab, x); } -struct jffs2_raw_node_ref *jffs2_alloc_refblock(void) +static struct jffs2_raw_node_ref *jffs2_alloc_refblock(void) { struct jffs2_raw_node_ref *ret; diff --git a/fs/jffs2/nodelist.h b/fs/jffs2/nodelist.h index f752baa8d39..cae92c14116 100644 --- a/fs/jffs2/nodelist.h +++ b/fs/jffs2/nodelist.h @@ -426,8 +426,6 @@ char *jffs2_getlink(struct jffs2_sb_info *c, struct jffs2_inode_info *f); /* scan.c */ int jffs2_scan_medium(struct jffs2_sb_info *c); void jffs2_rotate_lists(struct jffs2_sb_info *c); -int jffs2_fill_scan_buf(struct jffs2_sb_info *c, void *buf, - uint32_t ofs, uint32_t len); struct jffs2_inode_cache *jffs2_scan_make_ino_cache(struct jffs2_sb_info *c, uint32_t ino); int jffs2_scan_classify_jeb(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb); int jffs2_scan_dirty_space(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, uint32_t size); diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c index cc1899268c4..266423b2709 100644 --- a/fs/jffs2/readinode.c +++ b/fs/jffs2/readinode.c @@ -968,6 +968,7 @@ void jffs2_do_clear_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f) struct jffs2_full_dirent *fd, *fds; int deleted; + jffs2_clear_acl(f); jffs2_xattr_delete_inode(c, f->inocache); down(&f->sem); deleted = f->inocache && !f->inocache->nlink; diff --git a/fs/jffs2/scan.c b/fs/jffs2/scan.c index 2bfdc33752d..e2413466ddd 100644 --- a/fs/jffs2/scan.c +++ b/fs/jffs2/scan.c @@ -274,8 +274,8 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) return ret; } -int jffs2_fill_scan_buf (struct jffs2_sb_info *c, void *buf, - uint32_t ofs, uint32_t len) +static int jffs2_fill_scan_buf(struct jffs2_sb_info *c, void *buf, + uint32_t ofs, uint32_t len) { int ret; size_t retlen; diff --git a/fs/jffs2/summary.c b/fs/jffs2/summary.c index c19bd476e8e..e52cef526d9 100644 --- a/fs/jffs2/summary.c +++ b/fs/jffs2/summary.c @@ -252,6 +252,11 @@ int jffs2_sum_add_kvec(struct jffs2_sb_info *c, const struct kvec *invecs, union jffs2_node_union *node; struct jffs2_eraseblock *jeb; + if (c->summary->sum_size == JFFS2_SUMMARY_NOSUM_SIZE) { + dbg_summary("Summary is disabled for this jeb! Skipping summary info!\n"); + return 0; + } + node = invecs[0].iov_base; jeb = &c->blocks[ofs / c->sector_size]; ofs -= jeb->offset; diff --git a/fs/jffs2/xattr.c b/fs/jffs2/xattr.c index 18e66dbf23b..25bc1ae0864 100644 --- a/fs/jffs2/xattr.c +++ b/fs/jffs2/xattr.c @@ -50,9 +50,10 @@ * is used to write xdatum to medium. xd->version will be incremented. * create_xattr_datum(c, xprefix, xname, xvalue, xsize) * is used to create new xdatum and write to medium. - * delete_xattr_datum(c, xd) - * is used to delete a xdatum. It marks xd JFFS2_XFLAGS_DEAD, and allows - * GC to reclaim those physical nodes. + * unrefer_xattr_datum(c, xd) + * is used to delete a xdatum. When nobody refers this xdatum, JFFS2_XFLAGS_DEAD + * is set on xd->flags and chained xattr_dead_list or release it immediately. + * In the first case, the garbage collector release it later. * -------------------------------------------------- */ static uint32_t xattr_datum_hashkey(int xprefix, const char *xname, const char *xvalue, int xsize) { @@ -394,22 +395,24 @@ static struct jffs2_xattr_datum *create_xattr_datum(struct jffs2_sb_info *c, return xd; } -static void delete_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) +static void unrefer_xattr_datum(struct jffs2_sb_info *c, struct jffs2_xattr_datum *xd) { /* must be called under down_write(xattr_sem) */ - BUG_ON(atomic_read(&xd->refcnt)); + if (atomic_dec_and_lock(&xd->refcnt, &c->erase_completion_lock)) { + uint32_t xid = xd->xid, version = xd->version; - unload_xattr_datum(c, xd); - xd->flags |= JFFS2_XFLAGS_DEAD; - spin_lock(&c->erase_completion_lock); - if (xd->node == (void *)xd) { - BUG_ON(!(xd->flags & JFFS2_XFLAGS_INVALID)); - jffs2_free_xattr_datum(xd); - } else { - list_add(&xd->xindex, &c->xattr_dead_list); + unload_xattr_datum(c, xd); + xd->flags |= JFFS2_XFLAGS_DEAD; + if (xd->node == (void *)xd) { + BUG_ON(!(xd->flags & JFFS2_XFLAGS_INVALID)); + jffs2_free_xattr_datum(xd); + } else { + list_add(&xd->xindex, &c->xattr_dead_list); + } + spin_unlock(&c->erase_completion_lock); + + dbg_xattr("xdatum(xid=%u, version=%u) was removed.\n", xid, version); } - spin_unlock(&c->erase_completion_lock); - dbg_xattr("xdatum(xid=%u, version=%u) was removed.\n", xd->xid, xd->version); } /* -------- xref related functions ------------------ @@ -580,8 +583,7 @@ static void delete_xattr_ref(struct jffs2_sb_info *c, struct jffs2_xattr_ref *re dbg_xattr("xref(ino=%u, xid=%u, xseqno=%u) was removed.\n", ref->ino, ref->xid, ref->xseqno); - if (atomic_dec_and_test(&xd->refcnt)) - delete_xattr_datum(c, xd); + unrefer_xattr_datum(c, xd); } void jffs2_xattr_delete_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic) @@ -1119,8 +1121,7 @@ int do_jffs2_setxattr(struct inode *inode, int xprefix, const char *xname, ref->next = c->xref_dead_list; c->xref_dead_list = ref; spin_unlock(&c->erase_completion_lock); - if (atomic_dec_and_test(&xd->refcnt)) - delete_xattr_datum(c, xd); + unrefer_xattr_datum(c, xd); } else { ref->ic = ic; ref->xd = xd; @@ -1156,8 +1157,7 @@ int do_jffs2_setxattr(struct inode *inode, int xprefix, const char *xname, down_write(&c->xattr_sem); if (rc) { JFFS2_WARNING("jffs2_reserve_space()=%d, request=%u\n", rc, request); - if (atomic_dec_and_test(&xd->refcnt)) - delete_xattr_datum(c, xd); + unrefer_xattr_datum(c, xd); up_write(&c->xattr_sem); return rc; } @@ -1170,8 +1170,7 @@ int do_jffs2_setxattr(struct inode *inode, int xprefix, const char *xname, ic->xref = ref; } rc = PTR_ERR(newref); - if (atomic_dec_and_test(&xd->refcnt)) - delete_xattr_datum(c, xd); + unrefer_xattr_datum(c, xd); } else if (ref) { delete_xattr_ref(c, ref); } diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c index 43e3f566aad..a223cf4faa9 100644 --- a/fs/jfs/inode.c +++ b/fs/jfs/inode.c @@ -168,16 +168,15 @@ void jfs_dirty_inode(struct inode *inode) set_cflag(COMMIT_Dirty, inode); } -static int -jfs_get_blocks(struct inode *ip, sector_t lblock, unsigned long max_blocks, - struct buffer_head *bh_result, int create) +int jfs_get_block(struct inode *ip, sector_t lblock, + struct buffer_head *bh_result, int create) { s64 lblock64 = lblock; int rc = 0; xad_t xad; s64 xaddr; int xflag; - s32 xlen = max_blocks; + s32 xlen = bh_result->b_size >> ip->i_blkbits; /* * Take appropriate lock on inode @@ -188,7 +187,7 @@ jfs_get_blocks(struct inode *ip, sector_t lblock, unsigned long max_blocks, IREAD_LOCK(ip); if (((lblock64 << ip->i_sb->s_blocksize_bits) < ip->i_size) && - (!xtLookup(ip, lblock64, max_blocks, &xflag, &xaddr, &xlen, 0)) && + (!xtLookup(ip, lblock64, xlen, &xflag, &xaddr, &xlen, 0)) && xaddr) { if (xflag & XAD_NOTRECORDED) { if (!create) @@ -255,13 +254,6 @@ jfs_get_blocks(struct inode *ip, sector_t lblock, unsigned long max_blocks, return rc; } -static int jfs_get_block(struct inode *ip, sector_t lblock, - struct buffer_head *bh_result, int create) -{ - return jfs_get_blocks(ip, lblock, bh_result->b_size >> ip->i_blkbits, - bh_result, create); -} - static int jfs_writepage(struct page *page, struct writeback_control *wbc) { return nobh_writepage(page, jfs_get_block, wbc); diff --git a/fs/jfs/jfs_inode.h b/fs/jfs/jfs_inode.h index b5c7da6190d..1fc48df670c 100644 --- a/fs/jfs/jfs_inode.h +++ b/fs/jfs/jfs_inode.h @@ -32,6 +32,7 @@ extern void jfs_truncate_nolock(struct inode *, loff_t); extern void jfs_free_zero_link(struct inode *); extern struct dentry *jfs_get_parent(struct dentry *dentry); extern void jfs_set_inode_flags(struct inode *); +extern int jfs_get_block(struct inode *, sector_t, struct buffer_head *, int); extern const struct address_space_operations jfs_aops; extern struct inode_operations jfs_dir_inode_operations; diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c index 10c46231ce1..efbb586bed4 100644 --- a/fs/jfs/jfs_txnmgr.c +++ b/fs/jfs/jfs_txnmgr.c @@ -2944,7 +2944,7 @@ int jfs_sync(void *arg) * Inode is being freed */ list_del_init(&jfs_ip->anon_inode_list); - } else if (! !mutex_trylock(&jfs_ip->commit_mutex)) { + } else if (mutex_trylock(&jfs_ip->commit_mutex)) { /* * inode will be removed from anonymous list * when it is committed diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index 09ea03f6227..295268ad231 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -165,8 +165,8 @@ static int jfs_create(struct inode *dip, struct dentry *dentry, int mode, out3: txEnd(tid); - mutex_unlock(&JFS_IP(dip)->commit_mutex); mutex_unlock(&JFS_IP(ip)->commit_mutex); + mutex_unlock(&JFS_IP(dip)->commit_mutex); if (rc) { free_ea_wmap(ip); ip->i_nlink = 0; @@ -300,8 +300,8 @@ static int jfs_mkdir(struct inode *dip, struct dentry *dentry, int mode) out3: txEnd(tid); - mutex_unlock(&JFS_IP(dip)->commit_mutex); mutex_unlock(&JFS_IP(ip)->commit_mutex); + mutex_unlock(&JFS_IP(dip)->commit_mutex); if (rc) { free_ea_wmap(ip); ip->i_nlink = 0; @@ -384,8 +384,8 @@ static int jfs_rmdir(struct inode *dip, struct dentry *dentry) if (rc == -EIO) txAbort(tid, 1); txEnd(tid); - mutex_unlock(&JFS_IP(dip)->commit_mutex); mutex_unlock(&JFS_IP(ip)->commit_mutex); + mutex_unlock(&JFS_IP(dip)->commit_mutex); goto out2; } @@ -422,8 +422,8 @@ static int jfs_rmdir(struct inode *dip, struct dentry *dentry) txEnd(tid); - mutex_unlock(&JFS_IP(dip)->commit_mutex); mutex_unlock(&JFS_IP(ip)->commit_mutex); + mutex_unlock(&JFS_IP(dip)->commit_mutex); /* * Truncating the directory index table is not guaranteed. It @@ -503,8 +503,8 @@ static int jfs_unlink(struct inode *dip, struct dentry *dentry) if (rc == -EIO) txAbort(tid, 1); /* Marks FS Dirty */ txEnd(tid); - mutex_unlock(&JFS_IP(dip)->commit_mutex); mutex_unlock(&JFS_IP(ip)->commit_mutex); + mutex_unlock(&JFS_IP(dip)->commit_mutex); IWRITE_UNLOCK(ip); goto out1; } @@ -527,8 +527,8 @@ static int jfs_unlink(struct inode *dip, struct dentry *dentry) if ((new_size = commitZeroLink(tid, ip)) < 0) { txAbort(tid, 1); /* Marks FS Dirty */ txEnd(tid); - mutex_unlock(&JFS_IP(dip)->commit_mutex); mutex_unlock(&JFS_IP(ip)->commit_mutex); + mutex_unlock(&JFS_IP(dip)->commit_mutex); IWRITE_UNLOCK(ip); rc = new_size; goto out1; @@ -556,9 +556,8 @@ static int jfs_unlink(struct inode *dip, struct dentry *dentry) txEnd(tid); - mutex_unlock(&JFS_IP(dip)->commit_mutex); mutex_unlock(&JFS_IP(ip)->commit_mutex); - + mutex_unlock(&JFS_IP(dip)->commit_mutex); while (new_size && (rc == 0)) { tid = txBegin(dip->i_sb, 0); @@ -847,8 +846,8 @@ static int jfs_link(struct dentry *old_dentry, out: txEnd(tid); - mutex_unlock(&JFS_IP(dir)->commit_mutex); mutex_unlock(&JFS_IP(ip)->commit_mutex); + mutex_unlock(&JFS_IP(dir)->commit_mutex); jfs_info("jfs_link: rc:%d", rc); return rc; @@ -1037,8 +1036,8 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry, out3: txEnd(tid); - mutex_unlock(&JFS_IP(dip)->commit_mutex); mutex_unlock(&JFS_IP(ip)->commit_mutex); + mutex_unlock(&JFS_IP(dip)->commit_mutex); if (rc) { free_ea_wmap(ip); ip->i_nlink = 0; @@ -1160,10 +1159,11 @@ static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (S_ISDIR(new_ip->i_mode)) { new_ip->i_nlink--; if (new_ip->i_nlink) { - mutex_unlock(&JFS_IP(new_dir)->commit_mutex); - mutex_unlock(&JFS_IP(old_ip)->commit_mutex); + mutex_unlock(&JFS_IP(new_ip)->commit_mutex); if (old_dir != new_dir) mutex_unlock(&JFS_IP(old_dir)->commit_mutex); + mutex_unlock(&JFS_IP(old_ip)->commit_mutex); + mutex_unlock(&JFS_IP(new_dir)->commit_mutex); if (!S_ISDIR(old_ip->i_mode) && new_ip) IWRITE_UNLOCK(new_ip); jfs_error(new_ip->i_sb, @@ -1281,13 +1281,12 @@ static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry, out4: txEnd(tid); - - mutex_unlock(&JFS_IP(new_dir)->commit_mutex); - mutex_unlock(&JFS_IP(old_ip)->commit_mutex); - if (old_dir != new_dir) - mutex_unlock(&JFS_IP(old_dir)->commit_mutex); if (new_ip) mutex_unlock(&JFS_IP(new_ip)->commit_mutex); + if (old_dir != new_dir) + mutex_unlock(&JFS_IP(old_dir)->commit_mutex); + mutex_unlock(&JFS_IP(old_ip)->commit_mutex); + mutex_unlock(&JFS_IP(new_dir)->commit_mutex); while (new_size && (rc == 0)) { tid = txBegin(new_ip->i_sb, 0); diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 4f6cfebc82d..143bcd1d5ea 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -26,6 +26,7 @@ #include <linux/moduleparam.h> #include <linux/kthread.h> #include <linux/posix_acl.h> +#include <linux/buffer_head.h> #include <asm/uaccess.h> #include <linux/seq_file.h> @@ -298,7 +299,7 @@ static int parse_options(char *options, struct super_block *sb, s64 *newLVSize, break; } -#if defined(CONFIG_QUOTA) +#ifdef CONFIG_QUOTA case Opt_quota: case Opt_usrquota: *flag |= JFS_USRQUOTA; @@ -597,7 +598,7 @@ static int jfs_show_options(struct seq_file *seq, struct vfsmount *vfs) if (sbi->flag & JFS_NOINTEGRITY) seq_puts(seq, ",nointegrity"); -#if defined(CONFIG_QUOTA) +#ifdef CONFIG_QUOTA if (sbi->flag & JFS_USRQUOTA) seq_puts(seq, ",usrquota"); @@ -608,6 +609,113 @@ static int jfs_show_options(struct seq_file *seq, struct vfsmount *vfs) return 0; } +#ifdef CONFIG_QUOTA + +/* Read data from quotafile - avoid pagecache and such because we cannot afford + * acquiring the locks... As quota files are never truncated and quota code + * itself serializes the operations (and noone else should touch the files) + * we don't have to be afraid of races */ +static ssize_t jfs_quota_read(struct super_block *sb, int type, char *data, + size_t len, loff_t off) +{ + struct inode *inode = sb_dqopt(sb)->files[type]; + sector_t blk = off >> sb->s_blocksize_bits; + int err = 0; + int offset = off & (sb->s_blocksize - 1); + int tocopy; + size_t toread; + struct buffer_head tmp_bh; + struct buffer_head *bh; + loff_t i_size = i_size_read(inode); + + if (off > i_size) + return 0; + if (off+len > i_size) + len = i_size-off; + toread = len; + while (toread > 0) { + tocopy = sb->s_blocksize - offset < toread ? + sb->s_blocksize - offset : toread; + + tmp_bh.b_state = 0; + tmp_bh.b_size = 1 << inode->i_blkbits; + err = jfs_get_block(inode, blk, &tmp_bh, 0); + if (err) + return err; + if (!buffer_mapped(&tmp_bh)) /* A hole? */ + memset(data, 0, tocopy); + else { + bh = sb_bread(sb, tmp_bh.b_blocknr); + if (!bh) + return -EIO; + memcpy(data, bh->b_data+offset, tocopy); + brelse(bh); + } + offset = 0; + toread -= tocopy; + data += tocopy; + blk++; + } + return len; +} + +/* Write to quotafile */ +static ssize_t jfs_quota_write(struct super_block *sb, int type, + const char *data, size_t len, loff_t off) +{ + struct inode *inode = sb_dqopt(sb)->files[type]; + sector_t blk = off >> sb->s_blocksize_bits; + int err = 0; + int offset = off & (sb->s_blocksize - 1); + int tocopy; + size_t towrite = len; + struct buffer_head tmp_bh; + struct buffer_head *bh; + + mutex_lock(&inode->i_mutex); + while (towrite > 0) { + tocopy = sb->s_blocksize - offset < towrite ? + sb->s_blocksize - offset : towrite; + + tmp_bh.b_state = 0; + tmp_bh.b_size = 1 << inode->i_blkbits; + err = jfs_get_block(inode, blk, &tmp_bh, 1); + if (err) + goto out; + if (offset || tocopy != sb->s_blocksize) + bh = sb_bread(sb, tmp_bh.b_blocknr); + else + bh = sb_getblk(sb, tmp_bh.b_blocknr); + if (!bh) { + err = -EIO; + goto out; + } + lock_buffer(bh); + memcpy(bh->b_data+offset, data, tocopy); + flush_dcache_page(bh->b_page); + set_buffer_uptodate(bh); + mark_buffer_dirty(bh); + unlock_buffer(bh); + brelse(bh); + offset = 0; + towrite -= tocopy; + data += tocopy; + blk++; + } +out: + if (len == towrite) + return err; + if (inode->i_size < off+len-towrite) + i_size_write(inode, off+len-towrite); + inode->i_version++; + inode->i_mtime = inode->i_ctime = CURRENT_TIME; + mark_inode_dirty(inode); + mutex_unlock(&inode->i_mutex); + return len - towrite; +} + +#endif + static struct super_operations jfs_super_operations = { .alloc_inode = jfs_alloc_inode, .destroy_inode = jfs_destroy_inode, @@ -621,7 +729,11 @@ static struct super_operations jfs_super_operations = { .unlockfs = jfs_unlockfs, .statfs = jfs_statfs, .remount_fs = jfs_remount, - .show_options = jfs_show_options + .show_options = jfs_show_options, +#ifdef CONFIG_QUOTA + .quota_read = jfs_quota_read, + .quota_write = jfs_quota_write, +#endif }; static struct export_operations jfs_export_operations = { diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index 5980c45998c..89ba0df14c2 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -454,7 +454,7 @@ static void nlmclnt_locks_init_private(struct file_lock *fl, struct nlm_host *ho fl->fl_ops = &nlmclnt_lock_ops; } -static void do_vfs_lock(struct file_lock *fl) +static int do_vfs_lock(struct file_lock *fl) { int res = 0; switch (fl->fl_flags & (FL_POSIX|FL_FLOCK)) { @@ -467,9 +467,7 @@ static void do_vfs_lock(struct file_lock *fl) default: BUG(); } - if (res < 0) - printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", - __FUNCTION__); + return res; } /* @@ -498,6 +496,7 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl) struct nlm_host *host = req->a_host; struct nlm_res *resp = &req->a_res; struct nlm_wait *block = NULL; + unsigned char fl_flags = fl->fl_flags; int status = -ENOLCK; if (!host->h_monitored && nsm_monitor(host) < 0) { @@ -505,6 +504,10 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl) host->h_name); goto out; } + fl->fl_flags |= FL_ACCESS; + status = do_vfs_lock(fl); + if (status < 0) + goto out; block = nlmclnt_prepare_block(host, fl); again: @@ -539,9 +542,10 @@ again: up_read(&host->h_rwsem); goto again; } - fl->fl_flags |= FL_SLEEP; /* Ensure the resulting lock will get added to granted list */ - do_vfs_lock(fl); + fl->fl_flags = fl_flags | FL_SLEEP; + if (do_vfs_lock(fl) < 0) + printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", __FUNCTION__); up_read(&host->h_rwsem); } status = nlm_stat_to_errno(resp->status); @@ -552,6 +556,7 @@ out_unblock: nlmclnt_cancel(host, req->a_args.block, fl); out: nlm_release_call(req); + fl->fl_flags = fl_flags; return status; } @@ -606,15 +611,19 @@ nlmclnt_unlock(struct nlm_rqst *req, struct file_lock *fl) { struct nlm_host *host = req->a_host; struct nlm_res *resp = &req->a_res; - int status; + int status = 0; /* * Note: the server is supposed to either grant us the unlock * request, or to deny it with NLM_LCK_DENIED_GRACE_PERIOD. In either * case, we want to unlock. */ + fl->fl_flags |= FL_EXISTS; down_read(&host->h_rwsem); - do_vfs_lock(fl); + if (do_vfs_lock(fl) == -ENOENT) { + up_read(&host->h_rwsem); + goto out; + } up_read(&host->h_rwsem); if (req->a_flags & RPC_TASK_ASYNC) @@ -624,7 +633,6 @@ nlmclnt_unlock(struct nlm_rqst *req, struct file_lock *fl) if (status < 0) goto out; - status = 0; if (resp->status == NLM_LCK_GRANTED) goto out; diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index baf5ae51348..c9d419703cf 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -638,9 +638,6 @@ static void nlmsvc_grant_callback(struct rpc_task *task, void *data) if (task->tk_status < 0) { /* RPC error: Re-insert for retransmission */ timeout = 10 * HZ; - } else if (block->b_done) { - /* Block already removed, kill it for real */ - timeout = 0; } else { /* Call was successful, now wait for client callback */ timeout = 60 * HZ; @@ -709,13 +706,10 @@ nlmsvc_retry_blocked(void) break; if (time_after(block->b_when,jiffies)) break; - dprintk("nlmsvc_retry_blocked(%p, when=%ld, done=%d)\n", - block, block->b_when, block->b_done); + dprintk("nlmsvc_retry_blocked(%p, when=%ld)\n", + block, block->b_when); kref_get(&block->b_count); - if (block->b_done) - nlmsvc_unlink_block(block); - else - nlmsvc_grant_blocked(block); + nlmsvc_grant_blocked(block); nlmsvc_release_block(block); } diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index 2a4df9b3779..01b4db9e546 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -237,19 +237,22 @@ static int nlm_traverse_files(struct nlm_host *host, int action) { struct nlm_file *file, **fp; - int i; + int i, ret = 0; mutex_lock(&nlm_file_mutex); for (i = 0; i < FILE_NRHASH; i++) { fp = nlm_files + i; while ((file = *fp) != NULL) { + file->f_count++; + mutex_unlock(&nlm_file_mutex); + /* Traverse locks, blocks and shares of this file * and update file->f_locks count */ - if (nlm_inspect_file(host, file, action)) { - mutex_unlock(&nlm_file_mutex); - return 1; - } + if (nlm_inspect_file(host, file, action)) + ret = 1; + mutex_lock(&nlm_file_mutex); + file->f_count--; /* No more references to this file. Let go of it. */ if (!file->f_blocks && !file->f_locks && !file->f_shares && !file->f_count) { @@ -262,7 +265,7 @@ nlm_traverse_files(struct nlm_host *host, int action) } } mutex_unlock(&nlm_file_mutex); - return 0; + return ret; } /* diff --git a/fs/locks.c b/fs/locks.c index 1ad29c9b625..d7c53392cac 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -725,6 +725,10 @@ next_task: /* Try to create a FLOCK lock on filp. We always insert new FLOCK locks * at the head of the list, but that's secret knowledge known only to * flock_lock_file and posix_lock_file. + * + * Note that if called with an FL_EXISTS argument, the caller may determine + * whether or not a lock was successfully freed by testing the return + * value for -ENOENT. */ static int flock_lock_file(struct file *filp, struct file_lock *request) { @@ -735,6 +739,8 @@ static int flock_lock_file(struct file *filp, struct file_lock *request) int found = 0; lock_kernel(); + if (request->fl_flags & FL_ACCESS) + goto find_conflict; for_each_lock(inode, before) { struct file_lock *fl = *before; if (IS_POSIX(fl)) @@ -750,8 +756,11 @@ static int flock_lock_file(struct file *filp, struct file_lock *request) break; } - if (request->fl_type == F_UNLCK) + if (request->fl_type == F_UNLCK) { + if ((request->fl_flags & FL_EXISTS) && !found) + error = -ENOENT; goto out; + } error = -ENOMEM; new_fl = locks_alloc_lock(); @@ -764,6 +773,7 @@ static int flock_lock_file(struct file *filp, struct file_lock *request) if (found) cond_resched(); +find_conflict: for_each_lock(inode, before) { struct file_lock *fl = *before; if (IS_POSIX(fl)) @@ -777,6 +787,8 @@ static int flock_lock_file(struct file *filp, struct file_lock *request) locks_insert_block(fl, request); goto out; } + if (request->fl_flags & FL_ACCESS) + goto out; locks_copy_lock(new_fl, request); locks_insert_lock(&inode->i_flock, new_fl); new_fl = NULL; @@ -948,8 +960,11 @@ static int __posix_lock_file_conf(struct inode *inode, struct file_lock *request error = 0; if (!added) { - if (request->fl_type == F_UNLCK) + if (request->fl_type == F_UNLCK) { + if (request->fl_flags & FL_EXISTS) + error = -ENOENT; goto out; + } if (!new_fl) { error = -ENOLCK; @@ -996,6 +1011,10 @@ static int __posix_lock_file_conf(struct inode *inode, struct file_lock *request * Add a POSIX style lock to a file. * We merge adjacent & overlapping locks whenever possible. * POSIX locks are sorted by owner task, then by starting address + * + * Note that if called with an FL_EXISTS argument, the caller may determine + * whether or not a lock was successfully freed by testing the return + * value for -ENOENT. */ int posix_lock_file(struct file *filp, struct file_lock *fl) { @@ -1402,8 +1421,9 @@ static int __setlease(struct file *filp, long arg, struct file_lock **flp) if (!leases_enable) goto out; - error = lease_alloc(filp, arg, &fl); - if (error) + error = -ENOMEM; + fl = locks_alloc_lock(); + if (fl == NULL) goto out; locks_copy_lock(fl, lease); @@ -1411,6 +1431,7 @@ static int __setlease(struct file *filp, long arg, struct file_lock **flp) locks_insert_lock(before, fl); *flp = fl; + error = 0; out: return error; } diff --git a/fs/minix/inode.c b/fs/minix/inode.c index 9ea91c5eeb7..330ff9fc7cf 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -204,6 +204,8 @@ static int minix_fill_super(struct super_block *s, void *data, int silent) /* * Allocate the buffer map to keep the superblock small. */ + if (sbi->s_imap_blocks == 0 || sbi->s_zmap_blocks == 0) + goto out_illegal_sb; i = (sbi->s_imap_blocks + sbi->s_zmap_blocks) * sizeof(bh); map = kmalloc(i, GFP_KERNEL); if (!map) @@ -263,7 +265,7 @@ out_no_root: out_no_bitmap: printk("MINIX-fs: bad superblock or unable to read bitmaps\n"); - out_freemap: +out_freemap: for (i = 0; i < sbi->s_imap_blocks; i++) brelse(sbi->s_imap[i]); for (i = 0; i < sbi->s_zmap_blocks; i++) @@ -276,11 +278,16 @@ out_no_map: printk("MINIX-fs: can't allocate map\n"); goto out_release; +out_illegal_sb: + if (!silent) + printk("MINIX-fs: bad superblock\n"); + goto out_release; + out_no_fs: if (!silent) printk("VFS: Can't find a Minix or Minix V2 filesystem " "on device %s\n", s->s_id); - out_release: +out_release: brelse(bh); goto out; @@ -290,7 +297,7 @@ out_bad_hblock: out_bad_sb: printk("MINIX-fs: unable to read superblock\n"); - out: +out: s->s_fs_info = NULL; kfree(sbi); return -EINVAL; diff --git a/fs/namei.c b/fs/namei.c index c784e8bb57a..432d6bc6fab 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -159,7 +159,7 @@ char * getname(const char __user * filename) #ifdef CONFIG_AUDITSYSCALL void putname(const char *name) { - if (unlikely(current->audit_context)) + if (unlikely(!audit_dummy_context())) audit_putname(name); else __putname(name); @@ -227,10 +227,10 @@ int generic_permission(struct inode *inode, int mask, int permission(struct inode *inode, int mask, struct nameidata *nd) { + umode_t mode = inode->i_mode; int retval, submask; if (mask & MAY_WRITE) { - umode_t mode = inode->i_mode; /* * Nobody gets write access to a read-only fs. @@ -247,6 +247,13 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) } + /* + * MAY_EXEC on regular files requires special handling: We override + * filesystem execute permissions if the mode bits aren't set. + */ + if ((mask & MAY_EXEC) && S_ISREG(mode) && !(mode & S_IXUGO)) + return -EACCES; + /* Ordinary permission routines do not understand MAY_APPEND. */ submask = mask & ~MAY_APPEND; if (inode->i_op && inode->i_op->permission) @@ -1125,7 +1132,7 @@ static int fastcall do_path_lookup(int dfd, const char *name, retval = link_path_walk(name, nd); out: if (likely(retval == 0)) { - if (unlikely(current->audit_context && nd && nd->dentry && + if (unlikely(!audit_dummy_context() && nd && nd->dentry && nd->dentry->d_inode)) audit_inode(name, nd->dentry->d_inode); } @@ -1357,7 +1364,7 @@ static int may_delete(struct inode *dir,struct dentry *victim,int isdir) return -ENOENT; BUG_ON(victim->d_parent->d_inode != dir); - audit_inode_child(victim->d_name.name, victim->d_inode, dir->i_ino); + audit_inode_child(victim->d_name.name, victim->d_inode, dir); error = permission(dir,MAY_WRITE | MAY_EXEC, NULL); if (error) @@ -1423,7 +1430,7 @@ struct dentry *lock_rename(struct dentry *p1, struct dentry *p2) struct dentry *p; if (p1 == p2) { - mutex_lock(&p1->d_inode->i_mutex); + mutex_lock_nested(&p1->d_inode->i_mutex, I_MUTEX_PARENT); return NULL; } @@ -1431,22 +1438,22 @@ struct dentry *lock_rename(struct dentry *p1, struct dentry *p2) for (p = p1; p->d_parent != p; p = p->d_parent) { if (p->d_parent == p2) { - mutex_lock(&p2->d_inode->i_mutex); - mutex_lock(&p1->d_inode->i_mutex); + mutex_lock_nested(&p2->d_inode->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&p1->d_inode->i_mutex, I_MUTEX_CHILD); return p; } } for (p = p2; p->d_parent != p; p = p->d_parent) { if (p->d_parent == p1) { - mutex_lock(&p1->d_inode->i_mutex); - mutex_lock(&p2->d_inode->i_mutex); + mutex_lock_nested(&p1->d_inode->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&p2->d_inode->i_mutex, I_MUTEX_CHILD); return p; } } - mutex_lock(&p1->d_inode->i_mutex); - mutex_lock(&p2->d_inode->i_mutex); + mutex_lock_nested(&p1->d_inode->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&p2->d_inode->i_mutex, I_MUTEX_CHILD); return NULL; } @@ -1659,6 +1666,7 @@ do_last: * It already exists. */ mutex_unlock(&dir->d_inode->i_mutex); + audit_inode_update(path.dentry->d_inode); error = -EEXIST; if (flag & O_EXCL) @@ -1669,6 +1677,7 @@ do_last: if (flag & O_NOFOLLOW) goto exit_dput; } + error = -ENOENT; if (!path.dentry->d_inode) goto exit_dput; @@ -1712,8 +1721,14 @@ do_link: if (error) goto exit_dput; error = __do_follow_link(&path, nd); - if (error) + if (error) { + /* Does someone understand code flow here? Or it is only + * me so stupid? Anathema to whoever designed this non-sense + * with "intent.open". + */ + release_open_intent(nd); return error; + } nd->flags &= ~LOOKUP_PARENT; if (nd->last_type == LAST_BIND) goto ok; @@ -1751,7 +1766,7 @@ struct dentry *lookup_create(struct nameidata *nd, int is_dir) { struct dentry *dentry = ERR_PTR(-EEXIST); - mutex_lock(&nd->dentry->d_inode->i_mutex); + mutex_lock_nested(&nd->dentry->d_inode->i_mutex, I_MUTEX_PARENT); /* * Yucky last component or no last component at all? * (foo/., foo/.., /////) @@ -1759,6 +1774,8 @@ struct dentry *lookup_create(struct nameidata *nd, int is_dir) if (nd->last_type != LAST_NORM) goto fail; nd->flags &= ~LOOKUP_PARENT; + nd->flags |= LOOKUP_CREATE; + nd->intent.open.flags = O_EXCL; /* * Do the final lookup. @@ -2008,7 +2025,7 @@ static long do_rmdir(int dfd, const char __user *pathname) error = -EBUSY; goto exit1; } - mutex_lock(&nd.dentry->d_inode->i_mutex); + mutex_lock_nested(&nd.dentry->d_inode->i_mutex, I_MUTEX_PARENT); dentry = lookup_hash(&nd); error = PTR_ERR(dentry); if (!IS_ERR(dentry)) { @@ -2082,7 +2099,7 @@ static long do_unlinkat(int dfd, const char __user *pathname) error = -EISDIR; if (nd.last_type != LAST_NORM) goto exit1; - mutex_lock(&nd.dentry->d_inode->i_mutex); + mutex_lock_nested(&nd.dentry->d_inode->i_mutex, I_MUTEX_PARENT); dentry = lookup_hash(&nd); error = PTR_ERR(dentry); if (!IS_ERR(dentry)) { diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 3ddda6f7ecc..e7ffb4deb3e 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -690,7 +690,9 @@ int nfs_lookup_verify_inode(struct inode *inode, struct nameidata *nd) goto out_force; /* This is an open(2) */ if (nfs_lookup_check_intent(nd, LOOKUP_OPEN) != 0 && - !(server->flags & NFS_MOUNT_NOCTO)) + !(server->flags & NFS_MOUNT_NOCTO) && + (S_ISREG(inode->i_mode) || + S_ISDIR(inode->i_mode))) goto out_force; } return nfs_revalidate_inode(server, inode); diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 4cdd1b499e3..76ca1cbc38f 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -67,25 +67,19 @@ struct nfs_direct_req { struct kref kref; /* release manager */ /* I/O parameters */ - struct list_head list, /* nfs_read/write_data structs */ - rewrite_list; /* saved nfs_write_data structs */ struct nfs_open_context *ctx; /* file open context info */ struct kiocb * iocb; /* controlling i/o request */ struct inode * inode; /* target file of i/o */ - unsigned long user_addr; /* location of user's buffer */ - size_t user_count; /* total bytes to move */ - loff_t pos; /* starting offset in file */ - struct page ** pages; /* pages in our buffer */ - unsigned int npages; /* count of pages */ /* completion state */ + atomic_t io_count; /* i/os we're waiting for */ spinlock_t lock; /* protect completion state */ - int outstanding; /* i/os we're waiting for */ ssize_t count, /* bytes actually processed */ error; /* any reported error */ struct completion completion; /* wait for i/o completion */ /* commit state */ + struct list_head rewrite_list; /* saved nfs_write_data structs */ struct nfs_write_data * commit_data; /* special write_data for commits */ int flags; #define NFS_ODIRECT_DO_COMMIT (1) /* an unstable reply was received */ @@ -93,8 +87,18 @@ struct nfs_direct_req { struct nfs_writeverf verf; /* unstable write verifier */ }; -static void nfs_direct_write_schedule(struct nfs_direct_req *dreq, int sync); static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode *inode); +static const struct rpc_call_ops nfs_write_direct_ops; + +static inline void get_dreq(struct nfs_direct_req *dreq) +{ + atomic_inc(&dreq->io_count); +} + +static inline int put_dreq(struct nfs_direct_req *dreq) +{ + return atomic_dec_and_test(&dreq->io_count); +} /** * nfs_direct_IO - NFS address space operation for direct I/O @@ -118,50 +122,21 @@ ssize_t nfs_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_ return -EINVAL; } -static void nfs_free_user_pages(struct page **pages, int npages, int do_dirty) +static void nfs_direct_dirty_pages(struct page **pages, int npages) { int i; for (i = 0; i < npages; i++) { struct page *page = pages[i]; - if (do_dirty && !PageCompound(page)) + if (!PageCompound(page)) set_page_dirty_lock(page); - page_cache_release(page); } - kfree(pages); } -static inline int nfs_get_user_pages(int rw, unsigned long user_addr, size_t size, struct page ***pages) +static void nfs_direct_release_pages(struct page **pages, int npages) { - int result = -ENOMEM; - unsigned long page_count; - size_t array_size; - - page_count = (user_addr + size + PAGE_SIZE - 1) >> PAGE_SHIFT; - page_count -= user_addr >> PAGE_SHIFT; - - array_size = (page_count * sizeof(struct page *)); - *pages = kmalloc(array_size, GFP_KERNEL); - if (*pages) { - down_read(¤t->mm->mmap_sem); - result = get_user_pages(current, current->mm, user_addr, - page_count, (rw == READ), 0, - *pages, NULL); - up_read(¤t->mm->mmap_sem); - if (result != page_count) { - /* - * If we got fewer pages than expected from - * get_user_pages(), the user buffer runs off the - * end of a mapping; return EFAULT. - */ - if (result >= 0) { - nfs_free_user_pages(*pages, result, 0); - result = -EFAULT; - } else - kfree(*pages); - *pages = NULL; - } - } - return result; + int i; + for (i = 0; i < npages; i++) + page_cache_release(pages[i]); } static inline struct nfs_direct_req *nfs_direct_req_alloc(void) @@ -173,13 +148,13 @@ static inline struct nfs_direct_req *nfs_direct_req_alloc(void) return NULL; kref_init(&dreq->kref); + kref_get(&dreq->kref); init_completion(&dreq->completion); - INIT_LIST_HEAD(&dreq->list); INIT_LIST_HEAD(&dreq->rewrite_list); dreq->iocb = NULL; dreq->ctx = NULL; spin_lock_init(&dreq->lock); - dreq->outstanding = 0; + atomic_set(&dreq->io_count, 0); dreq->count = 0; dreq->error = 0; dreq->flags = 0; @@ -220,18 +195,11 @@ out: } /* - * We must hold a reference to all the pages in this direct read request - * until the RPCs complete. This could be long *after* we are woken up in - * nfs_direct_wait (for instance, if someone hits ^C on a slow server). - * - * In addition, synchronous I/O uses a stack-allocated iocb. Thus we - * can't trust the iocb is still valid here if this is a synchronous - * request. If the waiter is woken prematurely, the iocb is long gone. + * Synchronous I/O uses a stack-allocated iocb. Thus we can't trust + * the iocb is still valid here if this is a synchronous request. */ static void nfs_direct_complete(struct nfs_direct_req *dreq) { - nfs_free_user_pages(dreq->pages, dreq->npages, 1); - if (dreq->iocb) { long res = (long) dreq->error; if (!res) @@ -244,48 +212,10 @@ static void nfs_direct_complete(struct nfs_direct_req *dreq) } /* - * Note we also set the number of requests we have in the dreq when we are - * done. This prevents races with I/O completion so we will always wait - * until all requests have been dispatched and completed. + * We must hold a reference to all the pages in this direct read request + * until the RPCs complete. This could be long *after* we are woken up in + * nfs_direct_wait (for instance, if someone hits ^C on a slow server). */ -static struct nfs_direct_req *nfs_direct_read_alloc(size_t nbytes, size_t rsize) -{ - struct list_head *list; - struct nfs_direct_req *dreq; - unsigned int rpages = (rsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - - dreq = nfs_direct_req_alloc(); - if (!dreq) - return NULL; - - list = &dreq->list; - for(;;) { - struct nfs_read_data *data = nfs_readdata_alloc(rpages); - - if (unlikely(!data)) { - while (!list_empty(list)) { - data = list_entry(list->next, - struct nfs_read_data, pages); - list_del(&data->pages); - nfs_readdata_free(data); - } - kref_put(&dreq->kref, nfs_direct_req_release); - return NULL; - } - - INIT_LIST_HEAD(&data->pages); - list_add(&data->pages, list); - - data->req = (struct nfs_page *) dreq; - dreq->outstanding++; - if (nbytes <= rsize) - break; - nbytes -= rsize; - } - kref_get(&dreq->kref); - return dreq; -} - static void nfs_direct_read_result(struct rpc_task *task, void *calldata) { struct nfs_read_data *data = calldata; @@ -294,6 +224,9 @@ static void nfs_direct_read_result(struct rpc_task *task, void *calldata) if (nfs_readpage_result(task, data) != 0) return; + nfs_direct_dirty_pages(data->pagevec, data->npages); + nfs_direct_release_pages(data->pagevec, data->npages); + spin_lock(&dreq->lock); if (likely(task->tk_status >= 0)) @@ -301,13 +234,10 @@ static void nfs_direct_read_result(struct rpc_task *task, void *calldata) else dreq->error = task->tk_status; - if (--dreq->outstanding) { - spin_unlock(&dreq->lock); - return; - } - spin_unlock(&dreq->lock); - nfs_direct_complete(dreq); + + if (put_dreq(dreq)) + nfs_direct_complete(dreq); } static const struct rpc_call_ops nfs_read_direct_ops = { @@ -316,41 +246,56 @@ static const struct rpc_call_ops nfs_read_direct_ops = { }; /* - * For each nfs_read_data struct that was allocated on the list, dispatch - * an NFS READ operation + * For each rsize'd chunk of the user's buffer, dispatch an NFS READ + * operation. If nfs_readdata_alloc() or get_user_pages() fails, + * bail and stop sending more reads. Read length accounting is + * handled automatically by nfs_direct_read_result(). Otherwise, if + * no requests have been sent, just return an error. */ -static void nfs_direct_read_schedule(struct nfs_direct_req *dreq) +static ssize_t nfs_direct_read_schedule(struct nfs_direct_req *dreq, unsigned long user_addr, size_t count, loff_t pos) { struct nfs_open_context *ctx = dreq->ctx; struct inode *inode = ctx->dentry->d_inode; - struct list_head *list = &dreq->list; - struct page **pages = dreq->pages; - size_t count = dreq->user_count; - loff_t pos = dreq->pos; size_t rsize = NFS_SERVER(inode)->rsize; - unsigned int curpage, pgbase; + unsigned int pgbase; + int result; + ssize_t started = 0; + + get_dreq(dreq); - curpage = 0; - pgbase = dreq->user_addr & ~PAGE_MASK; do { struct nfs_read_data *data; size_t bytes; - bytes = rsize; - if (count < rsize) - bytes = count; + pgbase = user_addr & ~PAGE_MASK; + bytes = min(rsize,count); - BUG_ON(list_empty(list)); - data = list_entry(list->next, struct nfs_read_data, pages); - list_del_init(&data->pages); + result = -ENOMEM; + data = nfs_readdata_alloc(pgbase + bytes); + if (unlikely(!data)) + break; + + down_read(¤t->mm->mmap_sem); + result = get_user_pages(current, current->mm, user_addr, + data->npages, 1, 0, data->pagevec, NULL); + up_read(¤t->mm->mmap_sem); + if (unlikely(result < data->npages)) { + if (result > 0) + nfs_direct_release_pages(data->pagevec, result); + nfs_readdata_release(data); + break; + } + + get_dreq(dreq); + data->req = (struct nfs_page *) dreq; data->inode = inode; data->cred = ctx->cred; data->args.fh = NFS_FH(inode); data->args.context = ctx; data->args.offset = pos; data->args.pgbase = pgbase; - data->args.pages = &pages[curpage]; + data->args.pages = data->pagevec; data->args.count = bytes; data->res.fattr = &data->fattr; data->res.eof = 0; @@ -373,33 +318,37 @@ static void nfs_direct_read_schedule(struct nfs_direct_req *dreq) bytes, (unsigned long long)data->args.offset); + started += bytes; + user_addr += bytes; pos += bytes; + /* FIXME: Remove this unnecessary math from final patch */ pgbase += bytes; - curpage += pgbase >> PAGE_SHIFT; pgbase &= ~PAGE_MASK; + BUG_ON(pgbase != (user_addr & ~PAGE_MASK)); count -= bytes; } while (count != 0); - BUG_ON(!list_empty(list)); + + if (put_dreq(dreq)) + nfs_direct_complete(dreq); + + if (started) + return 0; + return result < 0 ? (ssize_t) result : -EFAULT; } -static ssize_t nfs_direct_read(struct kiocb *iocb, unsigned long user_addr, size_t count, loff_t pos, struct page **pages, unsigned int nr_pages) +static ssize_t nfs_direct_read(struct kiocb *iocb, unsigned long user_addr, size_t count, loff_t pos) { - ssize_t result; + ssize_t result = 0; sigset_t oldset; struct inode *inode = iocb->ki_filp->f_mapping->host; struct rpc_clnt *clnt = NFS_CLIENT(inode); struct nfs_direct_req *dreq; - dreq = nfs_direct_read_alloc(count, NFS_SERVER(inode)->rsize); + dreq = nfs_direct_req_alloc(); if (!dreq) return -ENOMEM; - dreq->user_addr = user_addr; - dreq->user_count = count; - dreq->pos = pos; - dreq->pages = pages; - dreq->npages = nr_pages; dreq->inode = inode; dreq->ctx = get_nfs_open_context((struct nfs_open_context *)iocb->ki_filp->private_data); if (!is_sync_kiocb(iocb)) @@ -407,8 +356,9 @@ static ssize_t nfs_direct_read(struct kiocb *iocb, unsigned long user_addr, size nfs_add_stats(inode, NFSIOS_DIRECTREADBYTES, count); rpc_clnt_sigmask(clnt, &oldset); - nfs_direct_read_schedule(dreq); - result = nfs_direct_wait(dreq); + result = nfs_direct_read_schedule(dreq, user_addr, count, pos); + if (!result) + result = nfs_direct_wait(dreq); rpc_clnt_sigunmask(clnt, &oldset); return result; @@ -416,10 +366,10 @@ static ssize_t nfs_direct_read(struct kiocb *iocb, unsigned long user_addr, size static void nfs_direct_free_writedata(struct nfs_direct_req *dreq) { - list_splice_init(&dreq->rewrite_list, &dreq->list); - while (!list_empty(&dreq->list)) { - struct nfs_write_data *data = list_entry(dreq->list.next, struct nfs_write_data, pages); + while (!list_empty(&dreq->rewrite_list)) { + struct nfs_write_data *data = list_entry(dreq->rewrite_list.next, struct nfs_write_data, pages); list_del(&data->pages); + nfs_direct_release_pages(data->pagevec, data->npages); nfs_writedata_release(data); } } @@ -427,14 +377,51 @@ static void nfs_direct_free_writedata(struct nfs_direct_req *dreq) #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) { - struct list_head *pos; + struct inode *inode = dreq->inode; + struct list_head *p; + struct nfs_write_data *data; - list_splice_init(&dreq->rewrite_list, &dreq->list); - list_for_each(pos, &dreq->list) - dreq->outstanding++; dreq->count = 0; + get_dreq(dreq); + + list_for_each(p, &dreq->rewrite_list) { + data = list_entry(p, struct nfs_write_data, pages); + + get_dreq(dreq); - nfs_direct_write_schedule(dreq, FLUSH_STABLE); + /* + * Reset data->res. + */ + nfs_fattr_init(&data->fattr); + data->res.count = data->args.count; + memset(&data->verf, 0, sizeof(data->verf)); + + /* + * Reuse data->task; data->args should not have changed + * since the original request was sent. + */ + rpc_init_task(&data->task, NFS_CLIENT(inode), RPC_TASK_ASYNC, + &nfs_write_direct_ops, data); + NFS_PROTO(inode)->write_setup(data, FLUSH_STABLE); + + data->task.tk_priority = RPC_PRIORITY_NORMAL; + data->task.tk_cookie = (unsigned long) inode; + + /* + * We're called via an RPC callback, so BKL is already held. + */ + rpc_execute(&data->task); + + dprintk("NFS: %5u rescheduled direct write call (req %s/%Ld, %u bytes @ offset %Lu)\n", + data->task.tk_pid, + inode->i_sb->s_id, + (long long)NFS_FILEID(inode), + data->args.count, + (unsigned long long)data->args.offset); + } + + if (put_dreq(dreq)) + nfs_direct_write_complete(dreq, inode); } static void nfs_direct_commit_result(struct rpc_task *task, void *calldata) @@ -471,8 +458,8 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq) data->cred = dreq->ctx->cred; data->args.fh = NFS_FH(data->inode); - data->args.offset = dreq->pos; - data->args.count = dreq->user_count; + data->args.offset = 0; + data->args.count = 0; data->res.count = 0; data->res.fattr = &data->fattr; data->res.verf = &data->verf; @@ -516,7 +503,7 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode static void nfs_alloc_commit_data(struct nfs_direct_req *dreq) { - dreq->commit_data = nfs_commit_alloc(0); + dreq->commit_data = nfs_commit_alloc(); if (dreq->commit_data != NULL) dreq->commit_data->req = (struct nfs_page *) dreq; } @@ -534,47 +521,6 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode } #endif -static struct nfs_direct_req *nfs_direct_write_alloc(size_t nbytes, size_t wsize) -{ - struct list_head *list; - struct nfs_direct_req *dreq; - unsigned int wpages = (wsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - - dreq = nfs_direct_req_alloc(); - if (!dreq) - return NULL; - - list = &dreq->list; - for(;;) { - struct nfs_write_data *data = nfs_writedata_alloc(wpages); - - if (unlikely(!data)) { - while (!list_empty(list)) { - data = list_entry(list->next, - struct nfs_write_data, pages); - list_del(&data->pages); - nfs_writedata_free(data); - } - kref_put(&dreq->kref, nfs_direct_req_release); - return NULL; - } - - INIT_LIST_HEAD(&data->pages); - list_add(&data->pages, list); - - data->req = (struct nfs_page *) dreq; - dreq->outstanding++; - if (nbytes <= wsize) - break; - nbytes -= wsize; - } - - nfs_alloc_commit_data(dreq); - - kref_get(&dreq->kref); - return dreq; -} - static void nfs_direct_write_result(struct rpc_task *task, void *calldata) { struct nfs_write_data *data = calldata; @@ -604,8 +550,6 @@ static void nfs_direct_write_result(struct rpc_task *task, void *calldata) } } } - /* In case we have to resend */ - data->args.stable = NFS_FILE_SYNC; spin_unlock(&dreq->lock); } @@ -619,14 +563,8 @@ static void nfs_direct_write_release(void *calldata) struct nfs_write_data *data = calldata; struct nfs_direct_req *dreq = (struct nfs_direct_req *) data->req; - spin_lock(&dreq->lock); - if (--dreq->outstanding) { - spin_unlock(&dreq->lock); - return; - } - spin_unlock(&dreq->lock); - - nfs_direct_write_complete(dreq, data->inode); + if (put_dreq(dreq)) + nfs_direct_write_complete(dreq, data->inode); } static const struct rpc_call_ops nfs_write_direct_ops = { @@ -635,41 +573,58 @@ static const struct rpc_call_ops nfs_write_direct_ops = { }; /* - * For each nfs_write_data struct that was allocated on the list, dispatch - * an NFS WRITE operation + * For each wsize'd chunk of the user's buffer, dispatch an NFS WRITE + * operation. If nfs_writedata_alloc() or get_user_pages() fails, + * bail and stop sending more writes. Write length accounting is + * handled automatically by nfs_direct_write_result(). Otherwise, if + * no requests have been sent, just return an error. */ -static void nfs_direct_write_schedule(struct nfs_direct_req *dreq, int sync) +static ssize_t nfs_direct_write_schedule(struct nfs_direct_req *dreq, unsigned long user_addr, size_t count, loff_t pos, int sync) { struct nfs_open_context *ctx = dreq->ctx; struct inode *inode = ctx->dentry->d_inode; - struct list_head *list = &dreq->list; - struct page **pages = dreq->pages; - size_t count = dreq->user_count; - loff_t pos = dreq->pos; size_t wsize = NFS_SERVER(inode)->wsize; - unsigned int curpage, pgbase; + unsigned int pgbase; + int result; + ssize_t started = 0; + + get_dreq(dreq); - curpage = 0; - pgbase = dreq->user_addr & ~PAGE_MASK; do { struct nfs_write_data *data; size_t bytes; - bytes = wsize; - if (count < wsize) - bytes = count; + pgbase = user_addr & ~PAGE_MASK; + bytes = min(wsize,count); + + result = -ENOMEM; + data = nfs_writedata_alloc(pgbase + bytes); + if (unlikely(!data)) + break; + + down_read(¤t->mm->mmap_sem); + result = get_user_pages(current, current->mm, user_addr, + data->npages, 0, 0, data->pagevec, NULL); + up_read(¤t->mm->mmap_sem); + if (unlikely(result < data->npages)) { + if (result > 0) + nfs_direct_release_pages(data->pagevec, result); + nfs_writedata_release(data); + break; + } + + get_dreq(dreq); - BUG_ON(list_empty(list)); - data = list_entry(list->next, struct nfs_write_data, pages); list_move_tail(&data->pages, &dreq->rewrite_list); + data->req = (struct nfs_page *) dreq; data->inode = inode; data->cred = ctx->cred; data->args.fh = NFS_FH(inode); data->args.context = ctx; data->args.offset = pos; data->args.pgbase = pgbase; - data->args.pages = &pages[curpage]; + data->args.pages = data->pagevec; data->args.count = bytes; data->res.fattr = &data->fattr; data->res.count = bytes; @@ -693,19 +648,29 @@ static void nfs_direct_write_schedule(struct nfs_direct_req *dreq, int sync) bytes, (unsigned long long)data->args.offset); + started += bytes; + user_addr += bytes; pos += bytes; + + /* FIXME: Remove this useless math from the final patch */ pgbase += bytes; - curpage += pgbase >> PAGE_SHIFT; pgbase &= ~PAGE_MASK; + BUG_ON(pgbase != (user_addr & ~PAGE_MASK)); count -= bytes; } while (count != 0); - BUG_ON(!list_empty(list)); + + if (put_dreq(dreq)) + nfs_direct_write_complete(dreq, inode); + + if (started) + return 0; + return result < 0 ? (ssize_t) result : -EFAULT; } -static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, size_t count, loff_t pos, struct page **pages, int nr_pages) +static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, size_t count, loff_t pos) { - ssize_t result; + ssize_t result = 0; sigset_t oldset; struct inode *inode = iocb->ki_filp->f_mapping->host; struct rpc_clnt *clnt = NFS_CLIENT(inode); @@ -713,17 +678,14 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, siz size_t wsize = NFS_SERVER(inode)->wsize; int sync = 0; - dreq = nfs_direct_write_alloc(count, wsize); + dreq = nfs_direct_req_alloc(); if (!dreq) return -ENOMEM; + nfs_alloc_commit_data(dreq); + if (dreq->commit_data == NULL || count < wsize) sync = FLUSH_STABLE; - dreq->user_addr = user_addr; - dreq->user_count = count; - dreq->pos = pos; - dreq->pages = pages; - dreq->npages = nr_pages; dreq->inode = inode; dreq->ctx = get_nfs_open_context((struct nfs_open_context *)iocb->ki_filp->private_data); if (!is_sync_kiocb(iocb)) @@ -734,8 +696,9 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, siz nfs_begin_data_update(inode); rpc_clnt_sigmask(clnt, &oldset); - nfs_direct_write_schedule(dreq, sync); - result = nfs_direct_wait(dreq); + result = nfs_direct_write_schedule(dreq, user_addr, count, pos, sync); + if (!result) + result = nfs_direct_wait(dreq); rpc_clnt_sigunmask(clnt, &oldset); return result; @@ -765,8 +728,6 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, siz ssize_t nfs_file_direct_read(struct kiocb *iocb, char __user *buf, size_t count, loff_t pos) { ssize_t retval = -EINVAL; - int page_count; - struct page **pages; struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; @@ -788,14 +749,7 @@ ssize_t nfs_file_direct_read(struct kiocb *iocb, char __user *buf, size_t count, if (retval) goto out; - retval = nfs_get_user_pages(READ, (unsigned long) buf, - count, &pages); - if (retval < 0) - goto out; - page_count = retval; - - retval = nfs_direct_read(iocb, (unsigned long) buf, count, pos, - pages, page_count); + retval = nfs_direct_read(iocb, (unsigned long) buf, count, pos); if (retval > 0) iocb->ki_pos = pos + retval; @@ -831,8 +785,6 @@ out: ssize_t nfs_file_direct_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t pos) { ssize_t retval; - int page_count; - struct page **pages; struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; @@ -860,14 +812,7 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, const char __user *buf, size_t if (retval) goto out; - retval = nfs_get_user_pages(WRITE, (unsigned long) buf, - count, &pages); - if (retval < 0) - goto out; - page_count = retval; - - retval = nfs_direct_write(iocb, (unsigned long) buf, count, - pos, pages, page_count); + retval = nfs_direct_write(iocb, (unsigned long) buf, count, pos); /* * XXX: nfs_end_data_update() already ensures this file's diff --git a/fs/nfs/file.c b/fs/nfs/file.c index cc2b874ad5a..48e892880d5 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -312,7 +312,13 @@ static void nfs_invalidate_page(struct page *page, unsigned long offset) static int nfs_release_page(struct page *page, gfp_t gfp) { - return !nfs_wb_page(page->mapping->host, page); + if (gfp & __GFP_FS) + return !nfs_wb_page(page->mapping->host, page); + else + /* + * Avoid deadlock on nfs_wait_on_request(). + */ + return 0; } const struct address_space_operations nfs_file_aops = { diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c index b81e7ed3c90..07a5dd57646 100644 --- a/fs/nfs/idmap.c +++ b/fs/nfs/idmap.c @@ -130,9 +130,7 @@ nfs_idmap_delete(struct nfs4_client *clp) if (!idmap) return; - dput(idmap->idmap_dentry); - idmap->idmap_dentry = NULL; - rpc_unlink(idmap->idmap_path); + rpc_unlink(idmap->idmap_dentry); clp->cl_idmap = NULL; kfree(idmap); } diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index 19b98ca468e..86b3169c8ca 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -51,7 +51,7 @@ char *nfs_path(const char *base, const struct dentry *dentry, namelen = dentry->d_name.len; buflen -= namelen + 1; if (buflen < 0) - goto Elong; + goto Elong_unlock; end -= namelen; memcpy(end, dentry->d_name.name, namelen); *--end = '/'; @@ -68,6 +68,8 @@ char *nfs_path(const char *base, const struct dentry *dentry, end -= namelen; memcpy(end, base, namelen); return end; +Elong_unlock: + spin_unlock(&dcache_lock); Elong: return ERR_PTR(-ENAMETOOLONG); } diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index b4916b09219..153898e1331 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2668,7 +2668,7 @@ out: nfs4_set_cached_acl(inode, acl); } -static inline ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen) +static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen) { struct page *pages[NFS4ACL_MAXPAGES]; struct nfs_getaclargs args = { @@ -2721,6 +2721,19 @@ out_free: return ret; } +static ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen) +{ + struct nfs4_exception exception = { }; + ssize_t ret; + do { + ret = __nfs4_get_acl_uncached(inode, buf, buflen); + if (ret >= 0) + break; + ret = nfs4_handle_exception(NFS_SERVER(inode), ret, &exception); + } while (exception.retry); + return ret; +} + static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen) { struct nfs_server *server = NFS_SERVER(inode); @@ -2737,7 +2750,7 @@ static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen) return nfs4_get_acl_uncached(inode, buf, buflen); } -static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen) +static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen) { struct nfs_server *server = NFS_SERVER(inode); struct page *pages[NFS4ACL_MAXPAGES]; @@ -2763,6 +2776,18 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen return ret; } +static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen) +{ + struct nfs4_exception exception = { }; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(inode), + __nfs4_proc_set_acl(inode, buf, buflen), + &exception); + } while (exception.retry); + return err; +} + static int nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server) { @@ -3144,9 +3169,6 @@ static int do_vfs_lock(struct file *file, struct file_lock *fl) default: BUG(); } - if (res < 0) - printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", - __FUNCTION__); return res; } @@ -3258,8 +3280,6 @@ static struct rpc_task *nfs4_do_unlck(struct file_lock *fl, return ERR_PTR(-ENOMEM); } - /* Unlock _before_ we do the RPC call */ - do_vfs_lock(fl->fl_file, fl); return rpc_run_task(NFS_CLIENT(lsp->ls_state->inode), RPC_TASK_ASYNC, &nfs4_locku_ops, data); } @@ -3270,30 +3290,28 @@ static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock * struct rpc_task *task; int status = 0; - /* Is this a delegated lock? */ - if (test_bit(NFS_DELEGATED_STATE, &state->flags)) - goto out_unlock; - /* Is this open_owner holding any locks on the server? */ - if (test_bit(LK_STATE_IN_USE, &state->flags) == 0) - goto out_unlock; - status = nfs4_set_lock_state(state, request); + /* Unlock _before_ we do the RPC call */ + request->fl_flags |= FL_EXISTS; + if (do_vfs_lock(request->fl_file, request) == -ENOENT) + goto out; if (status != 0) - goto out_unlock; + goto out; + /* Is this a delegated lock? */ + if (test_bit(NFS_DELEGATED_STATE, &state->flags)) + goto out; lsp = request->fl_u.nfs4_fl.owner; - status = -ENOMEM; seqid = nfs_alloc_seqid(&lsp->ls_seqid); + status = -ENOMEM; if (seqid == NULL) - goto out_unlock; + goto out; task = nfs4_do_unlck(request, request->fl_file->private_data, lsp, seqid); status = PTR_ERR(task); if (IS_ERR(task)) - goto out_unlock; + goto out; status = nfs4_wait_for_completion_rpc_task(task); rpc_release_task(task); - return status; -out_unlock: - do_vfs_lock(request->fl_file, request); +out: return status; } @@ -3461,10 +3479,10 @@ static int nfs4_lock_reclaim(struct nfs4_state *state, struct file_lock *request struct nfs4_exception exception = { }; int err; - /* Cache the lock if possible... */ - if (test_bit(NFS_DELEGATED_STATE, &state->flags)) - return 0; do { + /* Cache the lock if possible... */ + if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0) + return 0; err = _nfs4_do_setlk(state, F_SETLK, request, 1); if (err != -NFS4ERR_DELAY) break; @@ -3483,6 +3501,8 @@ static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request if (err != 0) return err; do { + if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0) + return 0; err = _nfs4_do_setlk(state, F_SETLK, request, 0); if (err != -NFS4ERR_DELAY) break; @@ -3494,29 +3514,42 @@ static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *request) { struct nfs4_client *clp = state->owner->so_client; + unsigned char fl_flags = request->fl_flags; int status; /* Is this a delegated open? */ - if (NFS_I(state->inode)->delegation_state != 0) { - /* Yes: cache locks! */ - status = do_vfs_lock(request->fl_file, request); - /* ...but avoid races with delegation recall... */ - if (status < 0 || test_bit(NFS_DELEGATED_STATE, &state->flags)) - return status; - } - down_read(&clp->cl_sem); status = nfs4_set_lock_state(state, request); if (status != 0) goto out; + request->fl_flags |= FL_ACCESS; + status = do_vfs_lock(request->fl_file, request); + if (status < 0) + goto out; + down_read(&clp->cl_sem); + if (test_bit(NFS_DELEGATED_STATE, &state->flags)) { + struct nfs_inode *nfsi = NFS_I(state->inode); + /* Yes: cache locks! */ + down_read(&nfsi->rwsem); + /* ...but avoid races with delegation recall... */ + if (test_bit(NFS_DELEGATED_STATE, &state->flags)) { + request->fl_flags = fl_flags & ~FL_SLEEP; + status = do_vfs_lock(request->fl_file, request); + up_read(&nfsi->rwsem); + goto out_unlock; + } + up_read(&nfsi->rwsem); + } status = _nfs4_do_setlk(state, cmd, request, 0); if (status != 0) - goto out; + goto out_unlock; /* Note: we always want to sleep here! */ - request->fl_flags |= FL_SLEEP; + request->fl_flags = fl_flags | FL_SLEEP; if (do_vfs_lock(request->fl_file, request) < 0) printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", __FUNCTION__); -out: +out_unlock: up_read(&clp->cl_sem); +out: + request->fl_flags = fl_flags; return status; } diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 1750d996f49..730ec8fb31c 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -3355,7 +3355,7 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n struct kvec *iov = rcvbuf->head; unsigned int nr, pglen = rcvbuf->page_len; uint32_t *end, *entry, *p, *kaddr; - uint32_t len, attrlen; + uint32_t len, attrlen, xlen; int hdrlen, recvd, status; status = decode_op_hdr(xdr, OP_READDIR); @@ -3377,10 +3377,10 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n BUG_ON(pglen + readdir->pgbase > PAGE_CACHE_SIZE); kaddr = p = (uint32_t *) kmap_atomic(page, KM_USER0); - end = (uint32_t *) ((char *)p + pglen + readdir->pgbase); + end = p + ((pglen + readdir->pgbase) >> 2); entry = p; for (nr = 0; *p++; nr++) { - if (p + 3 > end) + if (end - p < 3) goto short_pkt; dprintk("cookie = %Lu, ", *((unsigned long long *)p)); p += 2; /* cookie */ @@ -3389,18 +3389,19 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n printk(KERN_WARNING "NFS: giant filename in readdir (len 0x%x)\n", len); goto err_unmap; } - dprintk("filename = %*s\n", len, (char *)p); - p += XDR_QUADLEN(len); - if (p + 1 > end) + xlen = XDR_QUADLEN(len); + if (end - p < xlen + 1) goto short_pkt; + dprintk("filename = %*s\n", len, (char *)p); + p += xlen; len = ntohl(*p++); /* bitmap length */ - p += len; - if (p + 1 > end) + if (end - p < len + 1) goto short_pkt; + p += len; attrlen = XDR_QUADLEN(ntohl(*p++)); - p += attrlen; /* attributes */ - if (p + 2 > end) + if (end - p < attrlen + 2) goto short_pkt; + p += attrlen; /* attributes */ entry = p; } if (!nr && (entry[0] != 0 || entry[1] == 0)) diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 52bf634260a..7a9ee00e0c6 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -43,13 +43,15 @@ static mempool_t *nfs_rdata_mempool; #define MIN_POOL_READ (32) -struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount) +struct nfs_read_data *nfs_readdata_alloc(size_t len) { + unsigned int pagecount = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; struct nfs_read_data *p = mempool_alloc(nfs_rdata_mempool, SLAB_NOFS); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); + p->npages = pagecount; if (pagecount <= ARRAY_SIZE(p->page_array)) p->pagevec = p->page_array; else { @@ -63,7 +65,7 @@ struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount) return p; } -void nfs_readdata_free(struct nfs_read_data *p) +static void nfs_readdata_free(struct nfs_read_data *p) { if (p && (p->pagevec != &p->page_array[0])) kfree(p->pagevec); @@ -116,10 +118,17 @@ static void nfs_readpage_truncate_uninitialised_page(struct nfs_read_data *data) pages = &data->args.pages[base >> PAGE_CACHE_SHIFT]; base &= ~PAGE_CACHE_MASK; pglen = PAGE_CACHE_SIZE - base; - if (pglen < remainder) + for (;;) { + if (remainder <= pglen) { + memclear_highpage_flush(*pages, base, remainder); + break; + } memclear_highpage_flush(*pages, base, pglen); - else - memclear_highpage_flush(*pages, base, remainder); + pages++; + remainder -= pglen; + pglen = PAGE_CACHE_SIZE; + base = 0; + } } /* @@ -133,7 +142,7 @@ static int nfs_readpage_sync(struct nfs_open_context *ctx, struct inode *inode, int result; struct nfs_read_data *rdata; - rdata = nfs_readdata_alloc(1); + rdata = nfs_readdata_alloc(count); if (!rdata) return -ENOMEM; @@ -329,25 +338,25 @@ static int nfs_pagein_multi(struct list_head *head, struct inode *inode) struct nfs_page *req = nfs_list_entry(head->next); struct page *page = req->wb_page; struct nfs_read_data *data; - unsigned int rsize = NFS_SERVER(inode)->rsize; - unsigned int nbytes, offset; + size_t rsize = NFS_SERVER(inode)->rsize, nbytes; + unsigned int offset; int requests = 0; LIST_HEAD(list); nfs_list_remove_request(req); nbytes = req->wb_bytes; - for(;;) { - data = nfs_readdata_alloc(1); + do { + size_t len = min(nbytes,rsize); + + data = nfs_readdata_alloc(len); if (!data) goto out_bad; INIT_LIST_HEAD(&data->pages); list_add(&data->pages, &list); requests++; - if (nbytes <= rsize) - break; - nbytes -= rsize; - } + nbytes -= len; + } while(nbytes != 0); atomic_set(&req->wb_complete, requests); ClearPageError(page); @@ -395,7 +404,7 @@ static int nfs_pagein_one(struct list_head *head, struct inode *inode) if (NFS_SERVER(inode)->rsize < PAGE_CACHE_SIZE) return nfs_pagein_multi(head, inode); - data = nfs_readdata_alloc(NFS_SERVER(inode)->rpages); + data = nfs_readdata_alloc(NFS_SERVER(inode)->rsize); if (!data) goto out_bad; @@ -476,6 +485,8 @@ static void nfs_readpage_set_pages_uptodate(struct nfs_read_data *data) unsigned int base = data->args.pgbase; struct page **pages; + if (data->res.eof) + count = data->args.count; if (unlikely(count == 0)) return; pages = &data->args.pages[base >> PAGE_CACHE_SHIFT]; @@ -483,11 +494,7 @@ static void nfs_readpage_set_pages_uptodate(struct nfs_read_data *data) count += base; for (;count >= PAGE_CACHE_SIZE; count -= PAGE_CACHE_SIZE, pages++) SetPageUptodate(*pages); - /* - * Was this an eof or a short read? If the latter, don't mark the page - * as uptodate yet. - */ - if (count > 0 && (data->res.eof || data->args.count == data->res.count)) + if (count != 0) SetPageUptodate(*pages); } @@ -502,6 +509,8 @@ static void nfs_readpage_set_pages_error(struct nfs_read_data *data) count += base; for (;count >= PAGE_CACHE_SIZE; count -= PAGE_CACHE_SIZE, pages++) SetPageError(*pages); + if (count != 0) + SetPageError(*pages); } /* diff --git a/fs/nfs/write.c b/fs/nfs/write.c index bca5734ca9f..8ab3cf10d79 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -90,22 +90,13 @@ static mempool_t *nfs_commit_mempool; static DECLARE_WAIT_QUEUE_HEAD(nfs_write_congestion); -struct nfs_write_data *nfs_commit_alloc(unsigned int pagecount) +struct nfs_write_data *nfs_commit_alloc(void) { struct nfs_write_data *p = mempool_alloc(nfs_commit_mempool, SLAB_NOFS); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); - if (pagecount <= ARRAY_SIZE(p->page_array)) - p->pagevec = p->page_array; - else { - p->pagevec = kcalloc(pagecount, sizeof(struct page *), GFP_NOFS); - if (!p->pagevec) { - mempool_free(p, nfs_commit_mempool); - p = NULL; - } - } } return p; } @@ -117,13 +108,15 @@ void nfs_commit_free(struct nfs_write_data *p) mempool_free(p, nfs_commit_mempool); } -struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount) +struct nfs_write_data *nfs_writedata_alloc(size_t len) { + unsigned int pagecount = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; struct nfs_write_data *p = mempool_alloc(nfs_wdata_mempool, SLAB_NOFS); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); + p->npages = pagecount; if (pagecount <= ARRAY_SIZE(p->page_array)) p->pagevec = p->page_array; else { @@ -137,7 +130,7 @@ struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount) return p; } -void nfs_writedata_free(struct nfs_write_data *p) +static void nfs_writedata_free(struct nfs_write_data *p) { if (p && (p->pagevec != &p->page_array[0])) kfree(p->pagevec); @@ -208,7 +201,7 @@ static int nfs_writepage_sync(struct nfs_open_context *ctx, struct inode *inode, int result, written = 0; struct nfs_write_data *wdata; - wdata = nfs_writedata_alloc(1); + wdata = nfs_writedata_alloc(wsize); if (!wdata) return -ENOMEM; @@ -578,7 +571,7 @@ static int nfs_wait_on_requests(struct inode *inode, unsigned long idx_start, un return ret; } -static void nfs_cancel_requests(struct list_head *head) +static void nfs_cancel_dirty_list(struct list_head *head) { struct nfs_page *req; while(!list_empty(head)) { @@ -589,6 +582,19 @@ static void nfs_cancel_requests(struct list_head *head) } } +static void nfs_cancel_commit_list(struct list_head *head) +{ + struct nfs_page *req; + + while(!list_empty(head)) { + req = nfs_list_entry(head->next); + nfs_list_remove_request(req); + nfs_inode_remove_request(req); + nfs_clear_page_writeback(req); + dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS); + } +} + /* * nfs_scan_dirty - Scan an inode for dirty requests * @inode: NFS inode to scan @@ -986,24 +992,24 @@ static int nfs_flush_multi(struct inode *inode, struct list_head *head, int how) struct nfs_page *req = nfs_list_entry(head->next); struct page *page = req->wb_page; struct nfs_write_data *data; - unsigned int wsize = NFS_SERVER(inode)->wsize; - unsigned int nbytes, offset; + size_t wsize = NFS_SERVER(inode)->wsize, nbytes; + unsigned int offset; int requests = 0; LIST_HEAD(list); nfs_list_remove_request(req); nbytes = req->wb_bytes; - for (;;) { - data = nfs_writedata_alloc(1); + do { + size_t len = min(nbytes, wsize); + + data = nfs_writedata_alloc(len); if (!data) goto out_bad; list_add(&data->pages, &list); requests++; - if (nbytes <= wsize) - break; - nbytes -= wsize; - } + nbytes -= len; + } while (nbytes != 0); atomic_set(&req->wb_complete, requests); ClearPageError(page); @@ -1057,7 +1063,7 @@ static int nfs_flush_one(struct inode *inode, struct list_head *head, int how) struct nfs_write_data *data; unsigned int count; - data = nfs_writedata_alloc(NFS_SERVER(inode)->wpages); + data = nfs_writedata_alloc(NFS_SERVER(inode)->wsize); if (!data) goto out_bad; @@ -1365,7 +1371,7 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how) struct nfs_write_data *data; struct nfs_page *req; - data = nfs_commit_alloc(NFS_SERVER(inode)->wpages); + data = nfs_commit_alloc(); if (!data) goto out_bad; @@ -1381,6 +1387,7 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how) nfs_list_remove_request(req); nfs_mark_request_commit(req); nfs_clear_page_writeback(req); + dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS); } return -ENOMEM; } @@ -1499,7 +1506,7 @@ int nfs_sync_inode_wait(struct inode *inode, unsigned long idx_start, if (pages != 0) { spin_unlock(&nfsi->req_lock); if (how & FLUSH_INVALIDATE) - nfs_cancel_requests(&head); + nfs_cancel_dirty_list(&head); else ret = nfs_flush_list(inode, &head, pages, how); spin_lock(&nfsi->req_lock); @@ -1512,7 +1519,7 @@ int nfs_sync_inode_wait(struct inode *inode, unsigned long idx_start, break; if (how & FLUSH_INVALIDATE) { spin_unlock(&nfsi->req_lock); - nfs_cancel_requests(&head); + nfs_cancel_commit_list(&head); spin_lock(&nfsi->req_lock); continue; } diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index b0e095ea0c0..ee4eff27aed 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -721,6 +721,12 @@ nfsd4_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) return nfs_ok; } +static inline void nfsd4_increment_op_stats(u32 opnum) +{ + if (opnum >= FIRST_NFS4_OP && opnum <= LAST_NFS4_OP) + nfsdstats.nfs4_opcount[opnum]++; +} + /* * COMPOUND call. @@ -930,6 +936,8 @@ encode_op: /* XXX Ugh, we need to get rid of this kind of special case: */ if (op->opnum == OP_READ && op->u.read.rd_filp) fput(op->u.read.rd_filp); + + nfsd4_increment_op_stats(op->opnum); } out: diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index ecc439d2565..501d8388453 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -187,6 +187,11 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) goto out; } + /* Set user creds for this exportpoint */ + error = nfserrno(nfsd_setuser(rqstp, exp)); + if (error) + goto out; + /* * Look up the dentry using the NFS file handle. */ @@ -241,16 +246,17 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) dprintk("nfsd: fh_verify - just checking\n"); dentry = fhp->fh_dentry; exp = fhp->fh_export; + /* Set user creds for this exportpoint; necessary even + * in the "just checking" case because this may be a + * filehandle that was created by fh_compose, and that + * is about to be used in another nfsv4 compound + * operation */ + error = nfserrno(nfsd_setuser(rqstp, exp)); + if (error) + goto out; } cache_get(&exp->h); - /* Set user creds for this exportpoint; necessary even in the "just - * checking" case because this may be a filehandle that was created by - * fh_compose, and that is about to be used in another nfsv4 compound - * operation */ - error = nfserrno(nfsd_setuser(rqstp, exp)); - if (error) - goto out; error = nfsd_mode_check(rqstp, dentry->d_inode->i_mode, type); if (error) diff --git a/fs/nfsd/stats.c b/fs/nfsd/stats.c index 57265d56380..71944cddf68 100644 --- a/fs/nfsd/stats.c +++ b/fs/nfsd/stats.c @@ -72,6 +72,16 @@ static int nfsd_proc_show(struct seq_file *seq, void *v) /* show my rpc info */ svc_seq_show(seq, &nfsd_svcstats); +#ifdef CONFIG_NFSD_V4 + /* Show count for individual nfsv4 operations */ + /* Writing operation numbers 0 1 2 also for maintaining uniformity */ + seq_printf(seq,"proc4ops %u", LAST_NFS4_OP + 1); + for (i = 0; i <= LAST_NFS4_OP; i++) + seq_printf(seq, " %u", nfsdstats.nfs4_opcount[i]); + + seq_putc(seq, '\n'); +#endif + return 0; } diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index 4c86b7e1d1e..d313f356e66 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -367,6 +367,12 @@ static void ntfs_destroy_extent_inode(ntfs_inode *ni) kmem_cache_free(ntfs_inode_cache, ni); } +/* + * The attribute runlist lock has separate locking rules from the + * normal runlist lock, so split the two lock-classes: + */ +static struct lock_class_key attr_list_rl_lock_class; + /** * __ntfs_init_inode - initialize ntfs specific part of an inode * @sb: super block of mounted volume @@ -394,6 +400,8 @@ void __ntfs_init_inode(struct super_block *sb, ntfs_inode *ni) ni->attr_list_size = 0; ni->attr_list = NULL; ntfs_init_runlist(&ni->attr_list_rl); + lockdep_set_class(&ni->attr_list_rl.lock, + &attr_list_rl_lock_class); ni->itype.index.bmp_ino = NULL; ni->itype.index.block_size = 0; ni->itype.index.vcn_size = 0; @@ -405,6 +413,13 @@ void __ntfs_init_inode(struct super_block *sb, ntfs_inode *ni) ni->ext.base_ntfs_ino = NULL; } +/* + * Extent inodes get MFT-mapped in a nested way, while the base inode + * is still mapped. Teach this nesting to the lock validator by creating + * a separate class for nested inode's mrec_lock's: + */ +static struct lock_class_key extent_inode_mrec_lock_key; + inline ntfs_inode *ntfs_new_extent_inode(struct super_block *sb, unsigned long mft_no) { @@ -413,6 +428,7 @@ inline ntfs_inode *ntfs_new_extent_inode(struct super_block *sb, ntfs_debug("Entering."); if (likely(ni != NULL)) { __ntfs_init_inode(sb, ni); + lockdep_set_class(&ni->mrec_lock, &extent_inode_mrec_lock_key); ni->mft_no = mft_no; ni->type = AT_UNUSED; ni->name = NULL; @@ -1722,6 +1738,15 @@ err_out: return err; } +/* + * The MFT inode has special locking, so teach the lock validator + * about this by splitting off the locking rules of the MFT from + * the locking rules of other inodes. The MFT inode can never be + * accessed from the VFS side (or even internally), only by the + * map_mft functions. + */ +static struct lock_class_key mft_ni_runlist_lock_key, mft_ni_mrec_lock_key; + /** * ntfs_read_inode_mount - special read_inode for mount time use only * @vi: inode to read @@ -2148,6 +2173,14 @@ int ntfs_read_inode_mount(struct inode *vi) ntfs_attr_put_search_ctx(ctx); ntfs_debug("Done."); ntfs_free(m); + + /* + * Split the locking rules of the MFT inode from the + * locking rules of other inodes: + */ + lockdep_set_class(&ni->runlist.lock, &mft_ni_runlist_lock_key); + lockdep_set_class(&ni->mrec_lock, &mft_ni_mrec_lock_key); + return 0; em_put_err_out: diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index 0e14acea3f8..74e0ee8fce7 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -1724,6 +1724,14 @@ upcase_failed: return FALSE; } +/* + * The lcn and mft bitmap inodes are NTFS-internal inodes with + * their own special locking rules: + */ +static struct lock_class_key + lcnbmp_runlist_lock_key, lcnbmp_mrec_lock_key, + mftbmp_runlist_lock_key, mftbmp_mrec_lock_key; + /** * load_system_files - open the system files using normal functions * @vol: ntfs super block describing device whose system files to load @@ -1780,6 +1788,10 @@ static BOOL load_system_files(ntfs_volume *vol) ntfs_error(sb, "Failed to load $MFT/$BITMAP attribute."); goto iput_mirr_err_out; } + lockdep_set_class(&NTFS_I(vol->mftbmp_ino)->runlist.lock, + &mftbmp_runlist_lock_key); + lockdep_set_class(&NTFS_I(vol->mftbmp_ino)->mrec_lock, + &mftbmp_mrec_lock_key); /* Read upcase table and setup @vol->upcase and @vol->upcase_len. */ if (!load_and_init_upcase(vol)) goto iput_mftbmp_err_out; @@ -1802,6 +1814,11 @@ static BOOL load_system_files(ntfs_volume *vol) iput(vol->lcnbmp_ino); goto bitmap_failed; } + lockdep_set_class(&NTFS_I(vol->lcnbmp_ino)->runlist.lock, + &lcnbmp_runlist_lock_key); + lockdep_set_class(&NTFS_I(vol->lcnbmp_ino)->mrec_lock, + &lcnbmp_mrec_lock_key); + NInoSetSparseDisabled(NTFS_I(vol->lcnbmp_ino)); if ((vol->nr_clusters + 7) >> 3 > i_size_read(vol->lcnbmp_ino)) { iput(vol->lcnbmp_ino); @@ -2743,6 +2760,17 @@ static int ntfs_fill_super(struct super_block *sb, void *opt, const int silent) struct inode *tmp_ino; int blocksize, result; + /* + * We do a pretty difficult piece of bootstrap by reading the + * MFT (and other metadata) from disk into memory. We'll only + * release this metadata during umount, so the locking patterns + * observed during bootstrap do not count. So turn off the + * observation of locking patterns (strictly for this context + * only) while mounting NTFS. [The validator is still active + * otherwise, even for this context: it will for example record + * lock class registrations.] + */ + lockdep_off(); ntfs_debug("Entering."); #ifndef NTFS_RW sb->s_flags |= MS_RDONLY; @@ -2754,6 +2782,7 @@ static int ntfs_fill_super(struct super_block *sb, void *opt, const int silent) if (!silent) ntfs_error(sb, "Allocation of NTFS volume structure " "failed. Aborting mount..."); + lockdep_on(); return -ENOMEM; } /* Initialize ntfs_volume structure. */ @@ -2940,6 +2969,7 @@ static int ntfs_fill_super(struct super_block *sb, void *opt, const int silent) mutex_unlock(&ntfs_lock); sb->s_export_op = &ntfs_export_ops; lock_kernel(); + lockdep_on(); return 0; } ntfs_error(sb, "Failed to allocate root directory."); @@ -3059,6 +3089,7 @@ err_out_now: sb->s_fs_info = NULL; kfree(vol); ntfs_debug("Failed, returning -EINVAL."); + lockdep_on(); return -EINVAL; } diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 1b8346dd057..9503240ef0e 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -2375,7 +2375,6 @@ leave: mlog(0, "returning %d\n", ret); return ret; } -EXPORT_SYMBOL_GPL(dlm_migrate_lockres); int dlm_lock_basts_flushed(struct dlm_ctxt *dlm, struct dlm_lock *lock) { diff --git a/fs/ocfs2/dlm/dlmunlock.c b/fs/ocfs2/dlm/dlmunlock.c index b0c3134f4f7..37be4b2e0d4 100644 --- a/fs/ocfs2/dlm/dlmunlock.c +++ b/fs/ocfs2/dlm/dlmunlock.c @@ -155,7 +155,7 @@ static enum dlm_status dlmunlock_common(struct dlm_ctxt *dlm, else status = dlm_get_unlock_actions(dlm, res, lock, lksb, &actions); - if (status != DLM_NORMAL) + if (status != DLM_NORMAL && (status != DLM_CANCELGRANT || !master_node)) goto leave; /* By now this has been masked out of cancel requests. */ @@ -183,8 +183,7 @@ static enum dlm_status dlmunlock_common(struct dlm_ctxt *dlm, spin_lock(&lock->spinlock); /* if the master told us the lock was already granted, * let the ast handle all of these actions */ - if (status == DLM_NORMAL && - lksb->status == DLM_CANCELGRANT) { + if (status == DLM_CANCELGRANT) { actions &= ~(DLM_UNLOCK_REMOVE_LOCK| DLM_UNLOCK_REGRANT_LOCK| DLM_UNLOCK_CLEAR_CONVERT_TYPE); @@ -349,14 +348,9 @@ static enum dlm_status dlm_send_remote_unlock_request(struct dlm_ctxt *dlm, vec, veclen, owner, &status); if (tmpret >= 0) { // successfully sent and received - if (status == DLM_CANCELGRANT) - ret = DLM_NORMAL; - else if (status == DLM_FORWARD) { + if (status == DLM_FORWARD) mlog(0, "master was in-progress. retry\n"); - ret = DLM_FORWARD; - } else - ret = status; - lksb->status = status; + ret = status; } else { mlog_errno(tmpret); if (dlm_is_host_down(tmpret)) { @@ -372,7 +366,6 @@ static enum dlm_status dlm_send_remote_unlock_request(struct dlm_ctxt *dlm, /* something bad. this will BUG in ocfs2 */ ret = dlm_err_to_dlm_status(tmpret); } - lksb->status = ret; } return ret; @@ -483,6 +476,10 @@ int dlm_unlock_lock_handler(struct o2net_msg *msg, u32 len, void *data) /* lock was found on queue */ lksb = lock->lksb; + if (flags & (LKM_VALBLK|LKM_PUT_LVB) && + lock->ml.type != LKM_EXMODE) + flags &= ~(LKM_VALBLK|LKM_PUT_LVB); + /* unlockast only called on originating node */ if (flags & LKM_PUT_LVB) { lksb->flags |= DLM_LKSB_PUT_LVB; @@ -507,11 +504,8 @@ not_found: "cookie=%u:%llu\n", dlm_get_lock_cookie_node(unlock->cookie), dlm_get_lock_cookie_seq(unlock->cookie)); - else { - /* send the lksb->status back to the other node */ - status = lksb->status; + else dlm_lock_put(lock); - } leave: if (res) @@ -533,26 +527,22 @@ static enum dlm_status dlm_get_cancel_actions(struct dlm_ctxt *dlm, if (dlm_lock_on_list(&res->blocked, lock)) { /* cancel this outright */ - lksb->status = DLM_NORMAL; status = DLM_NORMAL; *actions = (DLM_UNLOCK_CALL_AST | DLM_UNLOCK_REMOVE_LOCK); } else if (dlm_lock_on_list(&res->converting, lock)) { /* cancel the request, put back on granted */ - lksb->status = DLM_NORMAL; status = DLM_NORMAL; *actions = (DLM_UNLOCK_CALL_AST | DLM_UNLOCK_REMOVE_LOCK | DLM_UNLOCK_REGRANT_LOCK | DLM_UNLOCK_CLEAR_CONVERT_TYPE); } else if (dlm_lock_on_list(&res->granted, lock)) { - /* too late, already granted. DLM_CANCELGRANT */ - lksb->status = DLM_CANCELGRANT; - status = DLM_NORMAL; + /* too late, already granted. */ + status = DLM_CANCELGRANT; *actions = DLM_UNLOCK_CALL_AST; } else { mlog(ML_ERROR, "lock to cancel is not on any list!\n"); - lksb->status = DLM_IVLOCKID; status = DLM_IVLOCKID; *actions = 0; } @@ -569,13 +559,11 @@ static enum dlm_status dlm_get_unlock_actions(struct dlm_ctxt *dlm, /* unlock request */ if (!dlm_lock_on_list(&res->granted, lock)) { - lksb->status = DLM_DENIED; status = DLM_DENIED; dlm_error(status); *actions = 0; } else { /* unlock granted lock */ - lksb->status = DLM_NORMAL; status = DLM_NORMAL; *actions = (DLM_UNLOCK_FREE_LOCK | DLM_UNLOCK_CALL_AST | @@ -632,6 +620,8 @@ retry: spin_lock(&res->spinlock); is_master = (res->owner == dlm->node_num); + if (flags & LKM_VALBLK && lock->ml.type != LKM_EXMODE) + flags &= ~LKM_VALBLK; spin_unlock(&res->spinlock); if (is_master) { @@ -665,7 +655,7 @@ retry: } if (call_ast) { - mlog(0, "calling unlockast(%p, %d)\n", data, lksb->status); + mlog(0, "calling unlockast(%p, %d)\n", data, status); if (is_master) { /* it is possible that there is one last bast * pending. make sure it is flushed, then @@ -677,9 +667,12 @@ retry: wait_event(dlm->ast_wq, dlm_lock_basts_flushed(dlm, lock)); } - (*unlockast)(data, lksb->status); + (*unlockast)(data, status); } + if (status == DLM_CANCELGRANT) + status = DLM_NORMAL; + if (status == DLM_NORMAL) { mlog(0, "kicking the thread\n"); dlm_kick_thread(dlm, res); diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index 0d1973ea32b..1f17a4d0828 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -840,6 +840,12 @@ static int ocfs2_local_alloc_new_window(struct ocfs2_super *osb, mlog(0, "Allocating %u clusters for a new window.\n", ocfs2_local_alloc_window_bits(osb)); + + /* Instruct the allocation code to try the most recently used + * cluster group. We'll re-record the group used this pass + * below. */ + ac->ac_last_group = osb->la_last_gd; + /* we used the generic suballoc reserve function, but we set * everything up nicely, so there's no reason why we can't use * the more specific cluster api to claim bits. */ @@ -852,6 +858,8 @@ static int ocfs2_local_alloc_new_window(struct ocfs2_super *osb, goto bail; } + osb->la_last_gd = ac->ac_last_group; + la->la_bm_off = cpu_to_le32(cluster_off); alloc->id1.bitmap1.i_total = cpu_to_le32(cluster_count); /* just in case... In the future when we find space ourselves, diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index cd4a6f253d1..0462a7f4e21 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -197,7 +197,6 @@ struct ocfs2_super struct ocfs2_node_map recovery_map; struct ocfs2_node_map umount_map; - u32 num_clusters; u64 root_blkno; u64 system_dir_blkno; u64 bitmap_blkno; @@ -237,6 +236,7 @@ struct ocfs2_super enum ocfs2_local_alloc_state local_alloc_state; struct buffer_head *local_alloc_bh; + u64 la_last_gd; /* Next two fields are for local node slot recovery during * mount. */ diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index 195523090c8..9d91e66f51a 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -70,12 +70,6 @@ static int ocfs2_block_group_search(struct inode *inode, struct buffer_head *group_bh, u32 bits_wanted, u32 min_bits, u16 *bit_off, u16 *bits_found); -static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, - u32 bits_wanted, - u32 min_bits, - u16 *bit_off, - unsigned int *num_bits, - u64 *bg_blkno); static int ocfs2_claim_suballoc_bits(struct ocfs2_super *osb, struct ocfs2_alloc_context *ac, u32 bits_wanted, @@ -85,11 +79,6 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_super *osb, u64 *bg_blkno); static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh, int nr); -static int ocfs2_block_group_find_clear_bits(struct ocfs2_super *osb, - struct buffer_head *bg_bh, - unsigned int bits_wanted, - u16 *bit_off, - u16 *bits_found); static inline int ocfs2_block_group_set_bits(struct ocfs2_journal_handle *handle, struct inode *alloc_inode, struct ocfs2_group_desc *bg, @@ -143,6 +132,64 @@ static u32 ocfs2_bits_per_group(struct ocfs2_chain_list *cl) return (u32)le16_to_cpu(cl->cl_cpg) * (u32)le16_to_cpu(cl->cl_bpc); } +/* somewhat more expensive than our other checks, so use sparingly. */ +static int ocfs2_check_group_descriptor(struct super_block *sb, + struct ocfs2_dinode *di, + struct ocfs2_group_desc *gd) +{ + unsigned int max_bits; + + if (!OCFS2_IS_VALID_GROUP_DESC(gd)) { + OCFS2_RO_ON_INVALID_GROUP_DESC(sb, gd); + return -EIO; + } + + if (di->i_blkno != gd->bg_parent_dinode) { + ocfs2_error(sb, "Group descriptor # %llu has bad parent " + "pointer (%llu, expected %llu)", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + (unsigned long long)le64_to_cpu(gd->bg_parent_dinode), + (unsigned long long)le64_to_cpu(di->i_blkno)); + return -EIO; + } + + max_bits = le16_to_cpu(di->id2.i_chain.cl_cpg) * le16_to_cpu(di->id2.i_chain.cl_bpc); + if (le16_to_cpu(gd->bg_bits) > max_bits) { + ocfs2_error(sb, "Group descriptor # %llu has bit count of %u", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + le16_to_cpu(gd->bg_bits)); + return -EIO; + } + + if (le16_to_cpu(gd->bg_chain) >= + le16_to_cpu(di->id2.i_chain.cl_next_free_rec)) { + ocfs2_error(sb, "Group descriptor # %llu has bad chain %u", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + le16_to_cpu(gd->bg_chain)); + return -EIO; + } + + if (le16_to_cpu(gd->bg_free_bits_count) > le16_to_cpu(gd->bg_bits)) { + ocfs2_error(sb, "Group descriptor # %llu has bit count %u but " + "claims that %u are free", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + le16_to_cpu(gd->bg_bits), + le16_to_cpu(gd->bg_free_bits_count)); + return -EIO; + } + + if (le16_to_cpu(gd->bg_bits) > (8 * le16_to_cpu(gd->bg_size))) { + ocfs2_error(sb, "Group descriptor # %llu has bit count %u but " + "max bitmap bits of %u", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + le16_to_cpu(gd->bg_bits), + 8 * le16_to_cpu(gd->bg_size)); + return -EIO; + } + + return 0; +} + static int ocfs2_block_group_fill(struct ocfs2_journal_handle *handle, struct inode *alloc_inode, struct buffer_head *bg_bh, @@ -663,6 +710,7 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh, static int ocfs2_block_group_find_clear_bits(struct ocfs2_super *osb, struct buffer_head *bg_bh, unsigned int bits_wanted, + unsigned int total_bits, u16 *bit_off, u16 *bits_found) { @@ -679,10 +727,8 @@ static int ocfs2_block_group_find_clear_bits(struct ocfs2_super *osb, found = start = best_offset = best_size = 0; bitmap = bg->bg_bitmap; - while((offset = ocfs2_find_next_zero_bit(bitmap, - le16_to_cpu(bg->bg_bits), - start)) != -1) { - if (offset == le16_to_cpu(bg->bg_bits)) + while((offset = ocfs2_find_next_zero_bit(bitmap, total_bits, start)) != -1) { + if (offset == total_bits) break; if (!ocfs2_test_bg_bit_allocatable(bg_bh, offset)) { @@ -911,14 +957,35 @@ static int ocfs2_cluster_group_search(struct inode *inode, { int search = -ENOSPC; int ret; - struct ocfs2_group_desc *bg = (struct ocfs2_group_desc *) group_bh->b_data; + struct ocfs2_group_desc *gd = (struct ocfs2_group_desc *) group_bh->b_data; u16 tmp_off, tmp_found; + unsigned int max_bits, gd_cluster_off; BUG_ON(!ocfs2_is_cluster_bitmap(inode)); - if (bg->bg_free_bits_count) { + if (gd->bg_free_bits_count) { + max_bits = le16_to_cpu(gd->bg_bits); + + /* Tail groups in cluster bitmaps which aren't cpg + * aligned are prone to partial extention by a failed + * fs resize. If the file system resize never got to + * update the dinode cluster count, then we don't want + * to trust any clusters past it, regardless of what + * the group descriptor says. */ + gd_cluster_off = ocfs2_blocks_to_clusters(inode->i_sb, + le64_to_cpu(gd->bg_blkno)); + if ((gd_cluster_off + max_bits) > + OCFS2_I(inode)->ip_clusters) { + max_bits = OCFS2_I(inode)->ip_clusters - gd_cluster_off; + mlog(0, "Desc %llu, bg_bits %u, clusters %u, use %u\n", + (unsigned long long)le64_to_cpu(gd->bg_blkno), + le16_to_cpu(gd->bg_bits), + OCFS2_I(inode)->ip_clusters, max_bits); + } + ret = ocfs2_block_group_find_clear_bits(OCFS2_SB(inode->i_sb), group_bh, bits_wanted, + max_bits, &tmp_off, &tmp_found); if (ret) return ret; @@ -951,17 +1018,109 @@ static int ocfs2_block_group_search(struct inode *inode, if (bg->bg_free_bits_count) ret = ocfs2_block_group_find_clear_bits(OCFS2_SB(inode->i_sb), group_bh, bits_wanted, + le16_to_cpu(bg->bg_bits), bit_off, bits_found); return ret; } +static int ocfs2_alloc_dinode_update_counts(struct inode *inode, + struct ocfs2_journal_handle *handle, + struct buffer_head *di_bh, + u32 num_bits, + u16 chain) +{ + int ret; + u32 tmp_used; + struct ocfs2_dinode *di = (struct ocfs2_dinode *) di_bh->b_data; + struct ocfs2_chain_list *cl = (struct ocfs2_chain_list *) &di->id2.i_chain; + + ret = ocfs2_journal_access(handle, inode, di_bh, + OCFS2_JOURNAL_ACCESS_WRITE); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + + tmp_used = le32_to_cpu(di->id1.bitmap1.i_used); + di->id1.bitmap1.i_used = cpu_to_le32(num_bits + tmp_used); + le32_add_cpu(&cl->cl_recs[chain].c_free, -num_bits); + + ret = ocfs2_journal_dirty(handle, di_bh); + if (ret < 0) + mlog_errno(ret); + +out: + return ret; +} + +static int ocfs2_search_one_group(struct ocfs2_alloc_context *ac, + u32 bits_wanted, + u32 min_bits, + u16 *bit_off, + unsigned int *num_bits, + u64 gd_blkno, + u16 *bits_left) +{ + int ret; + u16 found; + struct buffer_head *group_bh = NULL; + struct ocfs2_group_desc *gd; + struct inode *alloc_inode = ac->ac_inode; + struct ocfs2_journal_handle *handle = ac->ac_handle; + + ret = ocfs2_read_block(OCFS2_SB(alloc_inode->i_sb), gd_blkno, + &group_bh, OCFS2_BH_CACHED, alloc_inode); + if (ret < 0) { + mlog_errno(ret); + return ret; + } + + gd = (struct ocfs2_group_desc *) group_bh->b_data; + if (!OCFS2_IS_VALID_GROUP_DESC(gd)) { + OCFS2_RO_ON_INVALID_GROUP_DESC(alloc_inode->i_sb, gd); + ret = -EIO; + goto out; + } + + ret = ac->ac_group_search(alloc_inode, group_bh, bits_wanted, min_bits, + bit_off, &found); + if (ret < 0) { + if (ret != -ENOSPC) + mlog_errno(ret); + goto out; + } + + *num_bits = found; + + ret = ocfs2_alloc_dinode_update_counts(alloc_inode, handle, ac->ac_bh, + *num_bits, + le16_to_cpu(gd->bg_chain)); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + + ret = ocfs2_block_group_set_bits(handle, alloc_inode, gd, group_bh, + *bit_off, *num_bits); + if (ret < 0) + mlog_errno(ret); + + *bits_left = le16_to_cpu(gd->bg_free_bits_count); + +out: + brelse(group_bh); + + return ret; +} + static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, u32 bits_wanted, u32 min_bits, u16 *bit_off, unsigned int *num_bits, - u64 *bg_blkno) + u64 *bg_blkno, + u16 *bits_left) { int status; u16 chain, tmp_bits; @@ -988,9 +1147,9 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, goto bail; } bg = (struct ocfs2_group_desc *) group_bh->b_data; - if (!OCFS2_IS_VALID_GROUP_DESC(bg)) { - OCFS2_RO_ON_INVALID_GROUP_DESC(alloc_inode->i_sb, bg); - status = -EIO; + status = ocfs2_check_group_descriptor(alloc_inode->i_sb, fe, bg); + if (status) { + mlog_errno(status); goto bail; } @@ -1018,9 +1177,9 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, goto bail; } bg = (struct ocfs2_group_desc *) group_bh->b_data; - if (!OCFS2_IS_VALID_GROUP_DESC(bg)) { - OCFS2_RO_ON_INVALID_GROUP_DESC(alloc_inode->i_sb, bg); - status = -EIO; + status = ocfs2_check_group_descriptor(alloc_inode->i_sb, fe, bg); + if (status) { + mlog_errno(status); goto bail; } } @@ -1099,6 +1258,7 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, (unsigned long long)fe->i_blkno); *bg_blkno = le64_to_cpu(bg->bg_blkno); + *bits_left = le16_to_cpu(bg->bg_free_bits_count); bail: if (group_bh) brelse(group_bh); @@ -1120,6 +1280,8 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_super *osb, { int status; u16 victim, i; + u16 bits_left = 0; + u64 hint_blkno = ac->ac_last_group; struct ocfs2_chain_list *cl; struct ocfs2_dinode *fe; @@ -1146,6 +1308,28 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_super *osb, goto bail; } + if (hint_blkno) { + /* Attempt to short-circuit the usual search mechanism + * by jumping straight to the most recently used + * allocation group. This helps us mantain some + * contiguousness across allocations. */ + status = ocfs2_search_one_group(ac, bits_wanted, min_bits, + bit_off, num_bits, + hint_blkno, &bits_left); + if (!status) { + /* Be careful to update *bg_blkno here as the + * caller is expecting it to be filled in, and + * ocfs2_search_one_group() won't do that for + * us. */ + *bg_blkno = hint_blkno; + goto set_hint; + } + if (status < 0 && status != -ENOSPC) { + mlog_errno(status); + goto bail; + } + } + cl = (struct ocfs2_chain_list *) &fe->id2.i_chain; victim = ocfs2_find_victim_chain(cl); @@ -1153,9 +1337,9 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_super *osb, ac->ac_allow_chain_relink = 1; status = ocfs2_search_chain(ac, bits_wanted, min_bits, bit_off, - num_bits, bg_blkno); + num_bits, bg_blkno, &bits_left); if (!status) - goto bail; + goto set_hint; if (status < 0 && status != -ENOSPC) { mlog_errno(status); goto bail; @@ -1177,8 +1361,8 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_super *osb, ac->ac_chain = i; status = ocfs2_search_chain(ac, bits_wanted, min_bits, - bit_off, num_bits, - bg_blkno); + bit_off, num_bits, bg_blkno, + &bits_left); if (!status) break; if (status < 0 && status != -ENOSPC) { @@ -1186,8 +1370,19 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_super *osb, goto bail; } } -bail: +set_hint: + if (status != -ENOSPC) { + /* If the next search of this group is not likely to + * yield a suitable extent, then we reset the last + * group hint so as to not waste a disk read */ + if (bits_left < min_bits) + ac->ac_last_group = 0; + else + ac->ac_last_group = *bg_blkno; + } + +bail: mlog_exit(status); return status; } @@ -1341,7 +1536,7 @@ int ocfs2_claim_clusters(struct ocfs2_super *osb, { int status; unsigned int bits_wanted = ac->ac_bits_wanted - ac->ac_bits_given; - u64 bg_blkno; + u64 bg_blkno = 0; u16 bg_bit_off; mlog_entry_void(); @@ -1494,9 +1689,9 @@ static int ocfs2_free_suballoc_bits(struct ocfs2_journal_handle *handle, } group = (struct ocfs2_group_desc *) group_bh->b_data; - if (!OCFS2_IS_VALID_GROUP_DESC(group)) { - OCFS2_RO_ON_INVALID_GROUP_DESC(alloc_inode->i_sb, group); - status = -EIO; + status = ocfs2_check_group_descriptor(alloc_inode->i_sb, fe, group); + if (status) { + mlog_errno(status); goto bail; } BUG_ON((count + start_bit) > le16_to_cpu(group->bg_bits)); diff --git a/fs/ocfs2/suballoc.h b/fs/ocfs2/suballoc.h index a76c82a7cea..c787838d105 100644 --- a/fs/ocfs2/suballoc.h +++ b/fs/ocfs2/suballoc.h @@ -49,6 +49,8 @@ struct ocfs2_alloc_context { u16 ac_chain; int ac_allow_chain_relink; group_search_t *ac_group_search; + + u64 ac_last_group; }; void ocfs2_free_alloc_context(struct ocfs2_alloc_context *ac); diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 382706a67ff..d17e33e66a1 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1442,8 +1442,13 @@ static int ocfs2_initialize_super(struct super_block *sb, osb->bitmap_blkno = OCFS2_I(inode)->ip_blkno; + /* We don't have a cluster lock on the bitmap here because + * we're only interested in static information and the extra + * complexity at mount time isn't worht it. Don't pass the + * inode in to the read function though as we don't want it to + * be put in the cache. */ status = ocfs2_read_block(osb, osb->bitmap_blkno, &bitmap_bh, 0, - inode); + NULL); iput(inode); if (status < 0) { mlog_errno(status); @@ -1452,7 +1457,6 @@ static int ocfs2_initialize_super(struct super_block *sb, di = (struct ocfs2_dinode *) bitmap_bh->b_data; osb->bitmap_cpg = le16_to_cpu(di->id2.i_chain.cl_cpg); - osb->num_clusters = le32_to_cpu(di->id1.bitmap1.i_total); brelse(bitmap_bh); mlog(0, "cluster bitmap inode: %llu, clusters per group: %u\n", (unsigned long long)osb->bitmap_blkno, osb->bitmap_cpg); diff --git a/fs/partitions/Kconfig b/fs/partitions/Kconfig index c9a47809928..e478f194183 100644 --- a/fs/partitions/Kconfig +++ b/fs/partitions/Kconfig @@ -99,7 +99,7 @@ config IBM_PARTITION config MAC_PARTITION bool "Macintosh partition map support" if PARTITION_ADVANCED - default y if MAC + default y if (MAC || PPC_PMAC) help Say Y here if you would like to use hard disks under Linux which were partitioned on a Macintosh. diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 839634026eb..51c6a748df4 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -339,6 +339,7 @@ void add_partition(struct gendisk *disk, int part, sector_t start, sector_t len) p->start_sect = start; p->nr_sects = len; p->partno = part; + p->policy = disk->policy; if (isdigit(disk->kobj.name[strlen(disk->kobj.name)-1])) snprintf(p->kobj.name,KOBJ_NAME_LEN,"%sp%d",disk->kobj.name,part); diff --git a/fs/partitions/sun.c b/fs/partitions/sun.c index abe91ca03ed..0a5927c806c 100644 --- a/fs/partitions/sun.c +++ b/fs/partitions/sun.c @@ -74,7 +74,7 @@ int sun_partition(struct parsed_partitions *state, struct block_device *bdev) spc = be16_to_cpu(label->ntrks) * be16_to_cpu(label->nsect); for (i = 0; i < 8; i++, p++) { unsigned long st_sector; - int num_sectors; + unsigned int num_sectors; st_sector = be32_to_cpu(p->start_cylinder) * spc; num_sectors = be32_to_cpu(p->num_sectors); diff --git a/fs/proc/array.c b/fs/proc/array.c index 7495d3e2077..0b615d62a15 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -74,6 +74,7 @@ #include <linux/times.h> #include <linux/cpuset.h> #include <linux/rcupdate.h> +#include <linux/delayacct.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -411,7 +412,7 @@ static int do_task_stat(struct task_struct *task, char * buffer, int whole) res = sprintf(buffer,"%d (%s) %c %d %d %d %d %d %lu %lu \ %lu %lu %lu %lu %lu %ld %ld %ld %ld %d 0 %llu %lu %ld %lu %lu %lu %lu %lu \ -%lu %lu %lu %lu %lu %lu %lu %lu %d %d %lu %lu\n", +%lu %lu %lu %lu %lu %lu %lu %lu %d %d %lu %lu %llu\n", task->pid, tcomm, state, @@ -455,7 +456,8 @@ static int do_task_stat(struct task_struct *task, char * buffer, int whole) task->exit_signal, task_cpu(task), task->rt_priority, - task->policy); + task->policy, + (unsigned long long)delayacct_blkio_ticks(task)); if(mm) mmput(mm); return res; diff --git a/fs/proc/base.c b/fs/proc/base.c index 243a94af042..fe8d55fb17c 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -551,6 +551,27 @@ static int proc_fd_access_allowed(struct inode *inode) return allowed; } +static int proc_setattr(struct dentry *dentry, struct iattr *attr) +{ + int error; + struct inode *inode = dentry->d_inode; + + if (attr->ia_valid & ATTR_MODE) + return -EPERM; + + error = inode_change_ok(inode, attr); + if (!error) { + error = security_inode_setattr(dentry, attr); + if (!error) + error = inode_setattr(inode, attr); + } + return error; +} + +static struct inode_operations proc_def_inode_operations = { + .setattr = proc_setattr, +}; + extern struct seq_operations mounts_op; struct proc_mounts { struct seq_file m; @@ -1111,7 +1132,8 @@ out: static struct inode_operations proc_pid_link_inode_operations = { .readlink = proc_pid_readlink, - .follow_link = proc_pid_follow_link + .follow_link = proc_pid_follow_link, + .setattr = proc_setattr, }; static int proc_readfd(struct file * filp, void * dirent, filldir_t filldir) @@ -1285,6 +1307,7 @@ static struct inode *proc_pid_make_inode(struct super_block * sb, struct task_st ei = PROC_I(inode); inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; inode->i_ino = fake_ino(task->pid, ino); + inode->i_op = &proc_def_inode_operations; /* * grab the reference to task. @@ -1339,6 +1362,7 @@ static int pid_revalidate(struct dentry *dentry, struct nameidata *nd) inode->i_uid = 0; inode->i_gid = 0; } + inode->i_mode &= ~(S_ISUID | S_ISGID); security_task_to_inode(task, inode); put_task_struct(task); return 1; @@ -1389,6 +1413,7 @@ static int tid_fd_revalidate(struct dentry *dentry, struct nameidata *nd) inode->i_uid = 0; inode->i_gid = 0; } + inode->i_mode &= ~(S_ISUID | S_ISGID); security_task_to_inode(task, inode); put_task_struct(task); return 1; @@ -1527,11 +1552,13 @@ static struct file_operations proc_task_operations = { */ static struct inode_operations proc_fd_inode_operations = { .lookup = proc_lookupfd, + .setattr = proc_setattr, }; static struct inode_operations proc_task_inode_operations = { .lookup = proc_task_lookup, .getattr = proc_task_getattr, + .setattr = proc_setattr, }; #ifdef CONFIG_SECURITY @@ -1845,11 +1872,13 @@ static struct file_operations proc_tid_base_operations = { static struct inode_operations proc_tgid_base_inode_operations = { .lookup = proc_tgid_base_lookup, .getattr = pid_getattr, + .setattr = proc_setattr, }; static struct inode_operations proc_tid_base_inode_operations = { .lookup = proc_tid_base_lookup, .getattr = pid_getattr, + .setattr = proc_setattr, }; #ifdef CONFIG_SECURITY @@ -1892,11 +1921,13 @@ static struct dentry *proc_tid_attr_lookup(struct inode *dir, static struct inode_operations proc_tgid_attr_inode_operations = { .lookup = proc_tgid_attr_lookup, .getattr = pid_getattr, + .setattr = proc_setattr, }; static struct inode_operations proc_tid_attr_inode_operations = { .lookup = proc_tid_attr_lookup, .getattr = pid_getattr, + .setattr = proc_setattr, }; #endif diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 6dcef089e18..49dfb2ab783 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -192,7 +192,7 @@ int proc_fill_super(struct super_block *s, void *data, int silent) { struct inode * root_inode; - s->s_flags |= MS_NODIRATIME; + s->s_flags |= MS_NODIRATIME | MS_NOSUID | MS_NOEXEC; s->s_blocksize = 1024; s->s_blocksize_bits = 10; s->s_magic = PROC_SUPER_MAGIC; diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c index 036d14d8362..6a984f64edd 100644 --- a/fs/proc/kcore.c +++ b/fs/proc/kcore.c @@ -42,8 +42,6 @@ const struct file_operations proc_kcore_operations = { #define kc_offset_to_vaddr(o) ((o) + PAGE_OFFSET) #endif -#define roundup(x, y) ((((x)+((y)-1))/(y))*(y)) - /* An ELF note in memory */ struct memelfnote { @@ -384,7 +382,7 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos) */ if (n) { if (clear_user(buffer + tsz - n, - tsz - n)) + n)) return -EFAULT; } } else { diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index 9f2cfc30f9c..94215622544 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -169,7 +169,7 @@ static int meminfo_read_proc(char *page, char **start, off_t off, "Mapped: %8lu kB\n" "Slab: %8lu kB\n" "PageTables: %8lu kB\n" - "NFS Unstable: %8lu kB\n" + "NFS_Unstable: %8lu kB\n" "Bounce: %8lu kB\n" "CommitLimit: %8lu kB\n" "Committed_AS: %8lu kB\n" diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c index af69f28277b..4616ed50ffc 100644 --- a/fs/proc/task_nommu.c +++ b/fs/proc/task_nommu.c @@ -107,7 +107,7 @@ int proc_exe_link(struct inode *inode, struct dentry **dentry, struct vfsmount * { struct vm_list_struct *vml; struct vm_area_struct *vma; - struct task_struct *task = proc_task(inode); + struct task_struct *task = get_proc_task(inode); struct mm_struct *mm = get_task_mm(task); int result = -ENOENT; diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c index 99fffc9e1bf..677139b48e0 100644 --- a/fs/ramfs/file-nommu.c +++ b/fs/ramfs/file-nommu.c @@ -283,9 +283,9 @@ unsigned long ramfs_nommu_get_unmapped_area(struct file *file, /*****************************************************************************/ /* - * set up a mapping + * set up a mapping for shared memory segments */ int ramfs_nommu_mmap(struct file *file, struct vm_area_struct *vma) { - return 0; + return vma->vm_flags & VM_SHARED ? 0 : -ENOSYS; } diff --git a/fs/read_write.c b/fs/read_write.c index 5bc0e9234f9..d4cb3183c99 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -436,7 +436,7 @@ unsigned long iov_shorten(struct iovec *iov, unsigned long nr_segs, size_t to) return seg; } -EXPORT_SYMBOL(iov_shorten); +EXPORT_UNUSED_SYMBOL(iov_shorten); /* June 2006 */ /* A write operation does a read from user space and vice versa */ #define vrfy_dir(type) ((type) == READ ? VERIFY_WRITE : VERIFY_READ) diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index 752cea12e30..1627edd5081 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -48,8 +48,8 @@ static int reiserfs_file_release(struct inode *inode, struct file *filp) return 0; } - reiserfs_write_lock(inode->i_sb); mutex_lock(&inode->i_mutex); + reiserfs_write_lock(inode->i_sb); /* freeing preallocation only involves relogging blocks that * are already in the current transaction. preallocation gets * freed at the end of each transaction, so it is impossible for @@ -860,8 +860,12 @@ static int reiserfs_submit_file_region_for_write(struct reiserfs_transaction_han // this sets the proper flags for O_SYNC to trigger a commit mark_inode_dirty(inode); reiserfs_write_unlock(inode->i_sb); - } else + } else { + reiserfs_write_lock(inode->i_sb); + reiserfs_update_inode_transaction(inode); mark_inode_dirty(inode); + reiserfs_write_unlock(inode->i_sb); + } sd_update = 1; } diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 12dfdcfbee3..52f1e213654 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -39,14 +39,10 @@ void reiserfs_delete_inode(struct inode *inode) /* The = 0 happens when we abort creating a new inode for some reason like lack of space.. */ if (!(inode->i_state & I_NEW) && INODE_PKEY(inode)->k_objectid != 0) { /* also handles bad_inode case */ - mutex_lock(&inode->i_mutex); - reiserfs_delete_xattrs(inode); - if (journal_begin(&th, inode->i_sb, jbegin_count)) { - mutex_unlock(&inode->i_mutex); + if (journal_begin(&th, inode->i_sb, jbegin_count)) goto out; - } reiserfs_update_inode_transaction(inode); err = reiserfs_delete_object(&th, inode); @@ -57,12 +53,8 @@ void reiserfs_delete_inode(struct inode *inode) if (!err) DQUOT_FREE_INODE(inode); - if (journal_end(&th, inode->i_sb, jbegin_count)) { - mutex_unlock(&inode->i_mutex); + if (journal_end(&th, inode->i_sb, jbegin_count)) goto out; - } - - mutex_unlock(&inode->i_mutex); /* check return value from reiserfs_delete_object after * ending the transaction @@ -2348,6 +2340,7 @@ static int reiserfs_write_full_page(struct page *page, unsigned long end_index = inode->i_size >> PAGE_CACHE_SHIFT; int error = 0; unsigned long block; + sector_t last_block; struct buffer_head *head, *bh; int partial = 0; int nr = 0; @@ -2395,10 +2388,19 @@ static int reiserfs_write_full_page(struct page *page, } bh = head; block = page->index << (PAGE_CACHE_SHIFT - s->s_blocksize_bits); + last_block = (i_size_read(inode) - 1) >> inode->i_blkbits; /* first map all the buffers, logging any direct items we find */ do { - if ((checked || buffer_dirty(bh)) && (!buffer_mapped(bh) || - (buffer_mapped(bh) + if (block > last_block) { + /* + * This can happen when the block size is less than + * the page size. The corresponding bytes in the page + * were zero filled above + */ + clear_buffer_dirty(bh); + set_buffer_uptodate(bh); + } else if ((checked || buffer_dirty(bh)) && + (!buffer_mapped(bh) || (buffer_mapped(bh) && bh->b_blocknr == 0))) { /* not mapped yet, or it points to a direct item, search diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c index 745c8810089..a986b5e1e28 100644 --- a/fs/reiserfs/ioctl.c +++ b/fs/reiserfs/ioctl.c @@ -116,12 +116,12 @@ static int reiserfs_unpack(struct inode *inode, struct file *filp) if (REISERFS_I(inode)->i_flags & i_nopack_mask) { return 0; } - reiserfs_write_lock(inode->i_sb); /* we need to make sure nobody is changing the file size beneath ** us */ mutex_lock(&inode->i_mutex); + reiserfs_write_lock(inode->i_sb); write_from = inode->i_size & (blocksize - 1); /* if we are on a block boundary, we are already unpacked. */ diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c index 5d8a8cfebc7..c533ec1bcae 100644 --- a/fs/reiserfs/procfs.c +++ b/fs/reiserfs/procfs.c @@ -492,9 +492,17 @@ static void add_file(struct super_block *sb, char *name, int reiserfs_proc_info_init(struct super_block *sb) { + char b[BDEVNAME_SIZE]; + char *s; + + /* Some block devices use /'s */ + strlcpy(b, reiserfs_bdevname(sb), BDEVNAME_SIZE); + s = strchr(b, '/'); + if (s) + *s = '!'; + spin_lock_init(&__PINFO(sb).lock); - REISERFS_SB(sb)->procdir = - proc_mkdir(reiserfs_bdevname(sb), proc_info_root); + REISERFS_SB(sb)->procdir = proc_mkdir(b, proc_info_root); if (REISERFS_SB(sb)->procdir) { REISERFS_SB(sb)->procdir->owner = THIS_MODULE; REISERFS_SB(sb)->procdir->data = sb; @@ -508,13 +516,22 @@ int reiserfs_proc_info_init(struct super_block *sb) return 0; } reiserfs_warning(sb, "reiserfs: cannot create /proc/%s/%s", - proc_info_root_name, reiserfs_bdevname(sb)); + proc_info_root_name, b); return 1; } int reiserfs_proc_info_done(struct super_block *sb) { struct proc_dir_entry *de = REISERFS_SB(sb)->procdir; + char b[BDEVNAME_SIZE]; + char *s; + + /* Some block devices use /'s */ + strlcpy(b, reiserfs_bdevname(sb), BDEVNAME_SIZE); + s = strchr(b, '/'); + if (s) + *s = '!'; + if (de) { remove_proc_entry("journal", de); remove_proc_entry("oidmap", de); @@ -528,7 +545,7 @@ int reiserfs_proc_info_done(struct super_block *sb) __PINFO(sb).exiting = 1; spin_unlock(&__PINFO(sb).lock); if (proc_info_root) { - remove_proc_entry(reiserfs_bdevname(sb), proc_info_root); + remove_proc_entry(b, proc_info_root); REISERFS_SB(sb)->procdir = NULL; } return 0; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 28eb3c88603..5567328f104 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -2203,7 +2203,7 @@ static ssize_t reiserfs_quota_write(struct super_block *sb, int type, size_t towrite = len; struct buffer_head tmp_bh, *bh; - mutex_lock(&inode->i_mutex); + mutex_lock_nested(&inode->i_mutex, I_MUTEX_QUOTA); while (towrite > 0) { tocopy = sb->s_blocksize - offset < towrite ? sb->s_blocksize - offset : towrite; diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 39fedaa88a0..d935fb9394e 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -424,7 +424,7 @@ int xattr_readdir(struct file *file, filldir_t filler, void *buf) int res = -ENOTDIR; if (!file->f_op || !file->f_op->readdir) goto out; - mutex_lock(&inode->i_mutex); + mutex_lock_nested(&inode->i_mutex, I_MUTEX_XATTR); // down(&inode->i_zombie); res = -ENOENT; if (!IS_DEADDIR(inode)) { diff --git a/fs/splice.c b/fs/splice.c index 05fd2787be9..684bca3d3a1 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1307,6 +1307,85 @@ asmlinkage long sys_splice(int fd_in, loff_t __user *off_in, } /* + * Make sure there's data to read. Wait for input if we can, otherwise + * return an appropriate error. + */ +static int link_ipipe_prep(struct pipe_inode_info *pipe, unsigned int flags) +{ + int ret; + + /* + * Check ->nrbufs without the inode lock first. This function + * is speculative anyways, so missing one is ok. + */ + if (pipe->nrbufs) + return 0; + + ret = 0; + mutex_lock(&pipe->inode->i_mutex); + + while (!pipe->nrbufs) { + if (signal_pending(current)) { + ret = -ERESTARTSYS; + break; + } + if (!pipe->writers) + break; + if (!pipe->waiting_writers) { + if (flags & SPLICE_F_NONBLOCK) { + ret = -EAGAIN; + break; + } + } + pipe_wait(pipe); + } + + mutex_unlock(&pipe->inode->i_mutex); + return ret; +} + +/* + * Make sure there's writeable room. Wait for room if we can, otherwise + * return an appropriate error. + */ +static int link_opipe_prep(struct pipe_inode_info *pipe, unsigned int flags) +{ + int ret; + + /* + * Check ->nrbufs without the inode lock first. This function + * is speculative anyways, so missing one is ok. + */ + if (pipe->nrbufs < PIPE_BUFFERS) + return 0; + + ret = 0; + mutex_lock(&pipe->inode->i_mutex); + + while (pipe->nrbufs >= PIPE_BUFFERS) { + if (!pipe->readers) { + send_sig(SIGPIPE, current, 0); + ret = -EPIPE; + break; + } + if (flags & SPLICE_F_NONBLOCK) { + ret = -EAGAIN; + break; + } + if (signal_pending(current)) { + ret = -ERESTARTSYS; + break; + } + pipe->waiting_writers++; + pipe_wait(pipe); + pipe->waiting_writers--; + } + + mutex_unlock(&pipe->inode->i_mutex); + return ret; +} + +/* * Link contents of ipipe to opipe. */ static int link_pipe(struct pipe_inode_info *ipipe, @@ -1314,9 +1393,7 @@ static int link_pipe(struct pipe_inode_info *ipipe, size_t len, unsigned int flags) { struct pipe_buffer *ibuf, *obuf; - int ret, do_wakeup, i, ipipe_first; - - ret = do_wakeup = ipipe_first = 0; + int ret = 0, i = 0, nbuf; /* * Potential ABBA deadlock, work around it by ordering lock @@ -1324,126 +1401,62 @@ static int link_pipe(struct pipe_inode_info *ipipe, * could deadlock (one doing tee from A -> B, the other from B -> A). */ if (ipipe->inode < opipe->inode) { - ipipe_first = 1; - mutex_lock(&ipipe->inode->i_mutex); - mutex_lock(&opipe->inode->i_mutex); + mutex_lock_nested(&ipipe->inode->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&opipe->inode->i_mutex, I_MUTEX_CHILD); } else { - mutex_lock(&opipe->inode->i_mutex); - mutex_lock(&ipipe->inode->i_mutex); + mutex_lock_nested(&opipe->inode->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&ipipe->inode->i_mutex, I_MUTEX_CHILD); } - for (i = 0;; i++) { + do { if (!opipe->readers) { send_sig(SIGPIPE, current, 0); if (!ret) ret = -EPIPE; break; } - if (ipipe->nrbufs - i) { - ibuf = ipipe->bufs + ((ipipe->curbuf + i) & (PIPE_BUFFERS - 1)); - /* - * If we have room, fill this buffer - */ - if (opipe->nrbufs < PIPE_BUFFERS) { - int nbuf = (opipe->curbuf + opipe->nrbufs) & (PIPE_BUFFERS - 1); - - /* - * Get a reference to this pipe buffer, - * so we can copy the contents over. - */ - ibuf->ops->get(ipipe, ibuf); - - obuf = opipe->bufs + nbuf; - *obuf = *ibuf; - - /* - * Don't inherit the gift flag, we need to - * prevent multiple steals of this page. - */ - obuf->flags &= ~PIPE_BUF_FLAG_GIFT; - - if (obuf->len > len) - obuf->len = len; - - opipe->nrbufs++; - do_wakeup = 1; - ret += obuf->len; - len -= obuf->len; - - if (!len) - break; - if (opipe->nrbufs < PIPE_BUFFERS) - continue; - } - - /* - * We have input available, but no output room. - * If we already copied data, return that. If we - * need to drop the opipe lock, it must be ordered - * last to avoid deadlocks. - */ - if ((flags & SPLICE_F_NONBLOCK) || !ipipe_first) { - if (!ret) - ret = -EAGAIN; - break; - } - if (signal_pending(current)) { - if (!ret) - ret = -ERESTARTSYS; - break; - } - if (do_wakeup) { - smp_mb(); - if (waitqueue_active(&opipe->wait)) - wake_up_interruptible(&opipe->wait); - kill_fasync(&opipe->fasync_readers, SIGIO, POLL_IN); - do_wakeup = 0; - } + /* + * If we have iterated all input buffers or ran out of + * output room, break. + */ + if (i >= ipipe->nrbufs || opipe->nrbufs >= PIPE_BUFFERS) + break; - opipe->waiting_writers++; - pipe_wait(opipe); - opipe->waiting_writers--; - continue; - } + ibuf = ipipe->bufs + ((ipipe->curbuf + i) & (PIPE_BUFFERS - 1)); + nbuf = (opipe->curbuf + opipe->nrbufs) & (PIPE_BUFFERS - 1); /* - * No input buffers, do the usual checks for available - * writers and blocking and wait if necessary + * Get a reference to this pipe buffer, + * so we can copy the contents over. */ - if (!ipipe->writers) - break; - if (!ipipe->waiting_writers) { - if (ret) - break; - } + ibuf->ops->get(ipipe, ibuf); + + obuf = opipe->bufs + nbuf; + *obuf = *ibuf; + /* - * pipe_wait() drops the ipipe mutex. To avoid deadlocks - * with another process, we can only safely do that if - * the ipipe lock is ordered last. + * Don't inherit the gift flag, we need to + * prevent multiple steals of this page. */ - if ((flags & SPLICE_F_NONBLOCK) || ipipe_first) { - if (!ret) - ret = -EAGAIN; - break; - } - if (signal_pending(current)) { - if (!ret) - ret = -ERESTARTSYS; - break; - } + obuf->flags &= ~PIPE_BUF_FLAG_GIFT; - if (waitqueue_active(&ipipe->wait)) - wake_up_interruptible_sync(&ipipe->wait); - kill_fasync(&ipipe->fasync_writers, SIGIO, POLL_OUT); + if (obuf->len > len) + obuf->len = len; - pipe_wait(ipipe); - } + opipe->nrbufs++; + ret += obuf->len; + len -= obuf->len; + i++; + } while (len); mutex_unlock(&ipipe->inode->i_mutex); mutex_unlock(&opipe->inode->i_mutex); - if (do_wakeup) { + /* + * If we put data in the output pipe, wakeup any potential readers. + */ + if (ret > 0) { smp_mb(); if (waitqueue_active(&opipe->wait)) wake_up_interruptible(&opipe->wait); @@ -1464,14 +1477,29 @@ static long do_tee(struct file *in, struct file *out, size_t len, { struct pipe_inode_info *ipipe = in->f_dentry->d_inode->i_pipe; struct pipe_inode_info *opipe = out->f_dentry->d_inode->i_pipe; + int ret = -EINVAL; /* - * Link ipipe to the two output pipes, consuming as we go along. + * Duplicate the contents of ipipe to opipe without actually + * copying the data. */ - if (ipipe && opipe) - return link_pipe(ipipe, opipe, len, flags); + if (ipipe && opipe && ipipe != opipe) { + /* + * Keep going, unless we encounter an error. The ipipe/opipe + * ordering doesn't really matter. + */ + ret = link_ipipe_prep(ipipe, flags); + if (!ret) { + ret = link_opipe_prep(opipe, flags); + if (!ret) { + ret = link_pipe(ipipe, opipe, len, flags); + if (!ret && (flags & SPLICE_F_NONBLOCK)) + ret = -EAGAIN; + } + } + } - return -EINVAL; + return ret; } asmlinkage long sys_tee(int fdin, int fdout, size_t len, unsigned int flags) diff --git a/fs/super.c b/fs/super.c index 9b780c42d84..5c4c94d5495 100644 --- a/fs/super.c +++ b/fs/super.c @@ -49,11 +49,12 @@ DEFINE_SPINLOCK(sb_lock); /** * alloc_super - create new superblock + * @type: filesystem type superblock should belong to * * Allocates and initializes a new &struct super_block. alloc_super() * returns a pointer new superblock or %NULL if allocation had failed. */ -static struct super_block *alloc_super(void) +static struct super_block *alloc_super(struct file_system_type *type) { struct super_block *s = kzalloc(sizeof(struct super_block), GFP_USER); static struct super_operations default_op; @@ -72,6 +73,13 @@ static struct super_block *alloc_super(void) INIT_LIST_HEAD(&s->s_inodes); init_rwsem(&s->s_umount); mutex_init(&s->s_lock); + lockdep_set_class(&s->s_umount, &type->s_umount_key); + /* + * The locking rules for s_lock are up to the + * filesystem. For example ext3fs has different + * lock ordering than usbfs: + */ + lockdep_set_class(&s->s_lock, &type->s_lock_key); down_write(&s->s_umount); s->s_count = S_BIAS; atomic_set(&s->s_active, 1); @@ -295,7 +303,7 @@ retry: } if (!s) { spin_unlock(&sb_lock); - s = alloc_super(); + s = alloc_super(type); if (!s) return ERR_PTR(-ENOMEM); goto retry; diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index 5e0e31cc46f..9889e54e1f1 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -109,6 +109,17 @@ static inline void set_inode_attr(struct inode * inode, struct iattr * iattr) inode->i_ctime = iattr->ia_ctime; } + +/* + * sysfs has a different i_mutex lock order behavior for i_mutex than other + * filesystems; sysfs i_mutex is called in many places with subsystem locks + * held. At the same time, many of the VFS locking rules do not apply to + * sysfs at all (cross directory rename for example). To untangle this mess + * (which gives false positives in lockdep), we're giving sysfs inodes their + * own class for i_mutex. + */ +static struct lock_class_key sysfs_inode_imutex_key; + struct inode * sysfs_new_inode(mode_t mode, struct sysfs_dirent * sd) { struct inode * inode = new_inode(sysfs_sb); @@ -118,6 +129,7 @@ struct inode * sysfs_new_inode(mode_t mode, struct sysfs_dirent * sd) inode->i_mapping->a_ops = &sysfs_aops; inode->i_mapping->backing_dev_info = &sysfs_backing_dev_info; inode->i_op = &sysfs_inode_operations; + lockdep_set_class(&inode->i_mutex, &sysfs_inode_imutex_key); if (sd->s_iattr) { /* sysfs_dirent has non-default attributes diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c index 3873c672cb4..33323473e3c 100644 --- a/fs/udf/ialloc.c +++ b/fs/udf/ialloc.c @@ -75,6 +75,12 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err) } *err = -ENOSPC; + UDF_I_UNIQUE(inode) = 0; + UDF_I_LENEXTENTS(inode) = 0; + UDF_I_NEXT_ALLOC_BLOCK(inode) = 0; + UDF_I_NEXT_ALLOC_GOAL(inode) = 0; + UDF_I_STRAT4096(inode) = 0; + block = udf_new_block(dir->i_sb, NULL, UDF_I_LOCATION(dir).partitionReferenceNum, start, err); if (*err) @@ -84,11 +90,6 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err) } mutex_lock(&sbi->s_alloc_mutex); - UDF_I_UNIQUE(inode) = 0; - UDF_I_LENEXTENTS(inode) = 0; - UDF_I_NEXT_ALLOC_BLOCK(inode) = 0; - UDF_I_NEXT_ALLOC_GOAL(inode) = 0; - UDF_I_STRAT4096(inode) = 0; if (UDF_SB_LVIDBH(sb)) { struct logicalVolHeaderDesc *lvhd; diff --git a/fs/udf/super.c b/fs/udf/super.c index 4df822c881b..fcce1a21a51 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -115,6 +115,13 @@ static struct inode *udf_alloc_inode(struct super_block *sb) ei = (struct udf_inode_info *)kmem_cache_alloc(udf_inode_cachep, SLAB_KERNEL); if (!ei) return NULL; + + ei->i_unique = 0; + ei->i_lenExtents = 0; + ei->i_next_alloc_block = 0; + ei->i_next_alloc_goal = 0; + ei->i_strat4096 = 0; + return &ei->vfs_inode; } @@ -1652,7 +1659,7 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) iput(inode); goto error_out; } - sb->s_maxbytes = MAX_LFS_FILESIZE; + sb->s_maxbytes = 1<<30; return 0; error_out: diff --git a/fs/udf/truncate.c b/fs/udf/truncate.c index e1b0e8cfecb..0abd66ce36e 100644 --- a/fs/udf/truncate.c +++ b/fs/udf/truncate.c @@ -239,37 +239,51 @@ void udf_truncate_extents(struct inode * inode) { if (offset) { - extoffset -= adsize; - etype = udf_next_aext(inode, &bloc, &extoffset, &eloc, &elen, &bh, 1); - if (etype == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) - { - extoffset -= adsize; - elen = EXT_NOT_RECORDED_NOT_ALLOCATED | (elen + offset); - udf_write_aext(inode, bloc, &extoffset, eloc, elen, bh, 0); + /* + * OK, there is not extent covering inode->i_size and + * no extent above inode->i_size => truncate is + * extending the file by 'offset'. + */ + if ((!bh && extoffset == udf_file_entry_alloc_offset(inode)) || + (bh && extoffset == sizeof(struct allocExtDesc))) { + /* File has no extents at all! */ + memset(&eloc, 0x00, sizeof(kernel_lb_addr)); + elen = EXT_NOT_RECORDED_NOT_ALLOCATED | offset; + udf_add_aext(inode, &bloc, &extoffset, eloc, elen, &bh, 1); } - else if (etype == (EXT_NOT_RECORDED_ALLOCATED >> 30)) - { - kernel_lb_addr neloc = { 0, 0 }; + else { extoffset -= adsize; - nelen = EXT_NOT_RECORDED_NOT_ALLOCATED | - ((elen + offset + inode->i_sb->s_blocksize - 1) & - ~(inode->i_sb->s_blocksize - 1)); - udf_write_aext(inode, bloc, &extoffset, neloc, nelen, bh, 1); - udf_add_aext(inode, &bloc, &extoffset, eloc, (etype << 30) | elen, &bh, 1); - } - else - { - if (elen & (inode->i_sb->s_blocksize - 1)) + etype = udf_next_aext(inode, &bloc, &extoffset, &eloc, &elen, &bh, 1); + if (etype == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) + { + extoffset -= adsize; + elen = EXT_NOT_RECORDED_NOT_ALLOCATED | (elen + offset); + udf_write_aext(inode, bloc, &extoffset, eloc, elen, bh, 0); + } + else if (etype == (EXT_NOT_RECORDED_ALLOCATED >> 30)) { + kernel_lb_addr neloc = { 0, 0 }; extoffset -= adsize; - elen = EXT_RECORDED_ALLOCATED | - ((elen + inode->i_sb->s_blocksize - 1) & + nelen = EXT_NOT_RECORDED_NOT_ALLOCATED | + ((elen + offset + inode->i_sb->s_blocksize - 1) & ~(inode->i_sb->s_blocksize - 1)); - udf_write_aext(inode, bloc, &extoffset, eloc, elen, bh, 1); + udf_write_aext(inode, bloc, &extoffset, neloc, nelen, bh, 1); + udf_add_aext(inode, &bloc, &extoffset, eloc, (etype << 30) | elen, &bh, 1); + } + else + { + if (elen & (inode->i_sb->s_blocksize - 1)) + { + extoffset -= adsize; + elen = EXT_RECORDED_ALLOCATED | + ((elen + inode->i_sb->s_blocksize - 1) & + ~(inode->i_sb->s_blocksize - 1)); + udf_write_aext(inode, bloc, &extoffset, eloc, elen, bh, 1); + } + memset(&eloc, 0x00, sizeof(kernel_lb_addr)); + elen = EXT_NOT_RECORDED_NOT_ALLOCATED | offset; + udf_add_aext(inode, &bloc, &extoffset, eloc, elen, &bh, 1); } - memset(&eloc, 0x00, sizeof(kernel_lb_addr)); - elen = EXT_NOT_RECORDED_NOT_ALLOCATED | offset; - udf_add_aext(inode, &bloc, &extoffset, eloc, elen, &bh, 1); } } } diff --git a/fs/ufs/balloc.c b/fs/ufs/balloc.c index b01804baa12..b8238147577 100644 --- a/fs/ufs/balloc.c +++ b/fs/ufs/balloc.c @@ -248,7 +248,7 @@ static void ufs_change_blocknr(struct inode *inode, unsigned int baseblk, if (likely(cur_index != index)) { page = ufs_get_locked_page(mapping, index); - if (IS_ERR(page)) + if (!page || IS_ERR(page)) /* it was truncated or EIO */ continue; } else page = locked_page; diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c index e7c8615beb6..30c6e8a9446 100644 --- a/fs/ufs/inode.c +++ b/fs/ufs/inode.c @@ -169,18 +169,20 @@ static void ufs_clear_frag(struct inode *inode, struct buffer_head *bh) static struct buffer_head * ufs_clear_frags(struct inode *inode, sector_t beg, - unsigned int n) + unsigned int n, sector_t want) { - struct buffer_head *res, *bh; + struct buffer_head *res = NULL, *bh; sector_t end = beg + n; - res = sb_getblk(inode->i_sb, beg); - ufs_clear_frag(inode, res); - for (++beg; beg < end; ++beg) { + for (; beg < end; ++beg) { bh = sb_getblk(inode->i_sb, beg); ufs_clear_frag(inode, bh); - brelse(bh); + if (want != beg) + brelse(bh); + else + res = bh; } + BUG_ON(!res); return res; } @@ -265,7 +267,9 @@ repeat: lastfrag = ufsi->i_lastfrag; } - goal = fs32_to_cpu(sb, ufsi->i_u1.i_data[lastblock]) + uspi->s_fpb; + tmp = fs32_to_cpu(sb, ufsi->i_u1.i_data[lastblock]); + if (tmp) + goal = tmp + uspi->s_fpb; tmp = ufs_new_fragments (inode, p, fragment - blockoff, goal, required + blockoff, err, locked_page); @@ -277,13 +281,15 @@ repeat: tmp = ufs_new_fragments(inode, p, fragment - (blockoff - lastblockoff), fs32_to_cpu(sb, *p), required + (blockoff - lastblockoff), err, locked_page); - } + } else /* (lastblock > block) */ { /* * We will allocate new block before last allocated block */ - else /* (lastblock > block) */ { - if (lastblock && (tmp = fs32_to_cpu(sb, ufsi->i_u1.i_data[lastblock-1]))) - goal = tmp + uspi->s_fpb; + if (block) { + tmp = fs32_to_cpu(sb, ufsi->i_u1.i_data[block-1]); + if (tmp) + goal = tmp + uspi->s_fpb; + } tmp = ufs_new_fragments(inode, p, fragment - blockoff, goal, uspi->s_fpb, err, locked_page); } @@ -296,7 +302,7 @@ repeat: } if (!phys) { - result = ufs_clear_frags(inode, tmp + blockoff, required); + result = ufs_clear_frags(inode, tmp, required, tmp + blockoff); } else { *phys = tmp + blockoff; result = NULL; @@ -383,7 +389,7 @@ repeat: } } - if (block && (tmp = fs32_to_cpu(sb, ((__fs32*)bh->b_data)[block-1]) + uspi->s_fpb)) + if (block && (tmp = fs32_to_cpu(sb, ((__fs32*)bh->b_data)[block-1]))) goal = tmp + uspi->s_fpb; else goal = bh->b_blocknr + uspi->s_fpb; @@ -397,7 +403,8 @@ repeat: if (!phys) { - result = ufs_clear_frags(inode, tmp + blockoff, uspi->s_fpb); + result = ufs_clear_frags(inode, tmp, uspi->s_fpb, + tmp + blockoff); } else { *phys = tmp + blockoff; *new = 1; diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c index abd5f23a426..d344b411e26 100644 --- a/fs/ufs/namei.c +++ b/fs/ufs/namei.c @@ -129,7 +129,7 @@ static int ufs_symlink (struct inode * dir, struct dentry * dentry, struct inode * inode; if (l > sb->s_blocksize) - goto out; + goto out_notlocked; lock_kernel(); inode = ufs_new_inode(dir, S_IFLNK | S_IRWXUGO); @@ -155,6 +155,7 @@ static int ufs_symlink (struct inode * dir, struct dentry * dentry, err = ufs_add_nondir(dentry, inode); out: unlock_kernel(); +out_notlocked: return err; out_fail: diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 19a99726e58..992ee0b87cc 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1326,7 +1326,7 @@ static ssize_t ufs_quota_write(struct super_block *sb, int type, size_t towrite = len; struct buffer_head *bh; - mutex_lock(&inode->i_mutex); + mutex_lock_nested(&inode->i_mutex, I_MUTEX_QUOTA); while (towrite > 0) { tocopy = sb->s_blocksize - offset < towrite ? sb->s_blocksize - offset : towrite; diff --git a/fs/ufs/truncate.c b/fs/ufs/truncate.c index c9b55872079..ea11d04c41a 100644 --- a/fs/ufs/truncate.c +++ b/fs/ufs/truncate.c @@ -375,17 +375,15 @@ static int ufs_alloc_lastblock(struct inode *inode) int err = 0; struct address_space *mapping = inode->i_mapping; struct ufs_sb_private_info *uspi = UFS_SB(inode->i_sb)->s_uspi; - struct ufs_inode_info *ufsi = UFS_I(inode); unsigned lastfrag, i, end; struct page *lastpage; struct buffer_head *bh; lastfrag = (i_size_read(inode) + uspi->s_fsize - 1) >> uspi->s_fshift; - if (!lastfrag) { - ufsi->i_lastfrag = 0; + if (!lastfrag) goto out; - } + lastfrag--; lastpage = ufs_get_locked_page(mapping, lastfrag >> @@ -400,25 +398,25 @@ static int ufs_alloc_lastblock(struct inode *inode) for (i = 0; i < end; ++i) bh = bh->b_this_page; - if (!buffer_mapped(bh)) { - err = ufs_getfrag_block(inode, lastfrag, bh, 1); - - if (unlikely(err)) - goto out_unlock; - - if (buffer_new(bh)) { - clear_buffer_new(bh); - unmap_underlying_metadata(bh->b_bdev, - bh->b_blocknr); - /* - * we do not zeroize fragment, because of - * if it maped to hole, it already contains zeroes - */ - set_buffer_uptodate(bh); - mark_buffer_dirty(bh); - set_page_dirty(lastpage); - } + + err = ufs_getfrag_block(inode, lastfrag, bh, 1); + + if (unlikely(err)) + goto out_unlock; + + if (buffer_new(bh)) { + clear_buffer_new(bh); + unmap_underlying_metadata(bh->b_bdev, + bh->b_blocknr); + /* + * we do not zeroize fragment, because of + * if it maped to hole, it already contains zeroes + */ + set_buffer_uptodate(bh); + mark_buffer_dirty(bh); + set_page_dirty(lastpage); } + out_unlock: ufs_put_locked_page(lastpage); out: @@ -440,23 +438,11 @@ int ufs_truncate(struct inode *inode, loff_t old_i_size) if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) return -EPERM; - if (inode->i_size > old_i_size) { - /* - * if we expand file we should care about - * allocation of block for last byte first of all - */ - err = ufs_alloc_lastblock(inode); + err = ufs_alloc_lastblock(inode); - if (err) { - i_size_write(inode, old_i_size); - goto out; - } - /* - * go away, because of we expand file, and we do not - * need free blocks, and zeroizes page - */ - lock_kernel(); - goto almost_end; + if (err) { + i_size_write(inode, old_i_size); + goto out; } block_truncate_page(inode->i_mapping, inode->i_size, ufs_getfrag_block); @@ -477,21 +463,8 @@ int ufs_truncate(struct inode *inode, loff_t old_i_size) yield(); } - if (inode->i_size < old_i_size) { - /* - * now we should have enough space - * to allocate block for last byte - */ - err = ufs_alloc_lastblock(inode); - if (err) - /* - * looks like all the same - we have no space, - * but we truncate file already - */ - inode->i_size = (ufsi->i_lastfrag - 1) * uspi->s_fsize; - } -almost_end: inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; + ufsi->i_lastfrag = DIRECT_FRAGMENT; unlock_kernel(); mark_inode_dirty(inode); out: diff --git a/fs/ufs/util.c b/fs/ufs/util.c index 337cf2c46d1..22f820a9b15 100644 --- a/fs/ufs/util.c +++ b/fs/ufs/util.c @@ -251,12 +251,12 @@ struct page *ufs_get_locked_page(struct address_space *mapping, { struct page *page; -try_again: page = find_lock_page(mapping, index); if (!page) { page = read_cache_page(mapping, index, (filler_t*)mapping->a_ops->readpage, NULL); + if (IS_ERR(page)) { printk(KERN_ERR "ufs_change_blocknr: " "read_cache_page error: ino %lu, index: %lu\n", @@ -266,6 +266,14 @@ try_again: lock_page(page); + if (unlikely(page->mapping == NULL)) { + /* Truncate got there first */ + unlock_page(page); + page_cache_release(page); + page = NULL; + goto out; + } + if (!PageUptodate(page) || PageError(page)) { unlock_page(page); page_cache_release(page); @@ -275,15 +283,8 @@ try_again: mapping->host->i_ino, index); page = ERR_PTR(-EIO); - goto out; } } - - if (unlikely(!page->mapping || !page_has_buffers(page))) { - unlock_page(page); - page_cache_release(page); - goto try_again;/*we really need these buffers*/ - } out: return page; } diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index c40f81ba9b1..34dcb43a783 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -1390,11 +1390,19 @@ xfs_vm_direct_IO( iocb->private = xfs_alloc_ioend(inode, IOMAP_UNWRITTEN); - ret = blockdev_direct_IO_own_locking(rw, iocb, inode, - iomap.iomap_target->bt_bdev, - iov, offset, nr_segs, - xfs_get_blocks_direct, - xfs_end_io_direct); + if (rw == WRITE) { + ret = blockdev_direct_IO_own_locking(rw, iocb, inode, + iomap.iomap_target->bt_bdev, + iov, offset, nr_segs, + xfs_get_blocks_direct, + xfs_end_io_direct); + } else { + ret = blockdev_direct_IO_no_locking(rw, iocb, inode, + iomap.iomap_target->bt_bdev, + iov, offset, nr_segs, + xfs_get_blocks_direct, + xfs_end_io_direct); + } if (unlikely(ret <= 0 && iocb->private)) xfs_destroy_ioend(iocb->private); diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h index ceda3a2859d..7858703ed84 100644 --- a/fs/xfs/linux-2.6/xfs_buf.h +++ b/fs/xfs/linux-2.6/xfs_buf.h @@ -246,8 +246,8 @@ extern void xfs_buf_trace(xfs_buf_t *, char *, void *, void *); #define BUF_BUSY XBF_DONT_BLOCK #define XFS_BUF_BFLAGS(bp) ((bp)->b_flags) -#define XFS_BUF_ZEROFLAGS(bp) \ - ((bp)->b_flags &= ~(XBF_READ|XBF_WRITE|XBF_ASYNC|XBF_DELWRI)) +#define XFS_BUF_ZEROFLAGS(bp) ((bp)->b_flags &= \ + ~(XBF_READ|XBF_WRITE|XBF_ASYNC|XBF_DELWRI|XBF_ORDERED)) #define XFS_BUF_STALE(bp) ((bp)->b_flags |= XFS_B_STALE) #define XFS_BUF_UNSTALE(bp) ((bp)->b_flags &= ~XFS_B_STALE) diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h index 8c021dc57d1..a13f75c1a93 100644 --- a/fs/xfs/linux-2.6/xfs_linux.h +++ b/fs/xfs/linux-2.6/xfs_linux.h @@ -215,7 +215,6 @@ BUFFER_FNS(PrivateStart, unwritten); #define MIN(a,b) (min(a,b)) #define MAX(a,b) (max(a,b)) #define howmany(x, y) (((x)+((y)-1))/(y)) -#define roundup(x, y) ((((x)+((y)-1))/(y))*(y)) /* * Various platform dependent calls that don't fit anywhere else diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c index 5d9cfd91ad0..ee788b1cb36 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ b/fs/xfs/linux-2.6/xfs_lrw.c @@ -264,7 +264,9 @@ xfs_read( dmflags, &locktype); if (ret) { xfs_iunlock(ip, XFS_IOLOCK_SHARED); - goto unlock_mutex; + if (unlikely(ioflags & IO_ISDIRECT)) + mutex_unlock(&inode->i_mutex); + return ret; } } @@ -272,6 +274,9 @@ xfs_read( bhv_vop_flushinval_pages(vp, ctooff(offtoct(*offset)), -1, FI_REMAPF_LOCKED); + if (unlikely(ioflags & IO_ISDIRECT)) + mutex_unlock(&inode->i_mutex); + xfs_rw_enter_trace(XFS_READ_ENTER, &ip->i_iocore, (void *)iovp, segs, *offset, ioflags); ret = __generic_file_aio_read(iocb, iovp, segs, offset); @@ -281,10 +286,6 @@ xfs_read( XFS_STATS_ADD(xs_read_bytes, ret); xfs_iunlock(ip, XFS_IOLOCK_SHARED); - -unlock_mutex: - if (unlikely(ioflags & IO_ISDIRECT)) - mutex_unlock(&inode->i_mutex); return ret; } @@ -390,6 +391,8 @@ xfs_splice_write( xfs_inode_t *ip = XFS_BHVTOI(bdp); xfs_mount_t *mp = ip->i_mount; ssize_t ret; + struct inode *inode = outfilp->f_mapping->host; + xfs_fsize_t isize; XFS_STATS_INC(xs_write_calls); if (XFS_FORCED_SHUTDOWN(ip->i_mount)) @@ -416,6 +419,20 @@ xfs_splice_write( if (ret > 0) XFS_STATS_ADD(xs_write_bytes, ret); + isize = i_size_read(inode); + if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize)) + *ppos = isize; + + if (*ppos > ip->i_d.di_size) { + xfs_ilock(ip, XFS_ILOCK_EXCL); + if (*ppos > ip->i_d.di_size) { + ip->i_d.di_size = *ppos; + i_size_write(inode, *ppos); + ip->i_update_core = 1; + ip->i_update_size = 1; + } + xfs_iunlock(ip, XFS_ILOCK_EXCL); + } xfs_iunlock(ip, XFS_IOLOCK_EXCL); return ret; } diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 9bdef9d5190..4754f342a5d 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -314,6 +314,13 @@ xfs_mountfs_check_barriers(xfs_mount_t *mp) return; } + if (xfs_readonly_buftarg(mp->m_ddev_targp)) { + xfs_fs_cmn_err(CE_NOTE, mp, + "Disabling barriers, underlying device is readonly"); + mp->m_flags &= ~XFS_MOUNT_BARRIER; + return; + } + error = xfs_barrier_test(mp); if (error) { xfs_fs_cmn_err(CE_NOTE, mp, diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c index e95e99f7168..db8872be8c8 100644 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ b/fs/xfs/quota/xfs_qm_bhv.c @@ -203,7 +203,7 @@ xfs_qm_statvfs( if (error || !vnode) return error; - mp = XFS_BHVTOM(bhv); + mp = xfs_vfstom(bhvtovfs(bhv)); ip = xfs_vtoi(vnode); if (!(ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)) @@ -217,17 +217,24 @@ xfs_qm_statvfs( return 0; dp = &dqp->q_core; - limit = dp->d_blk_softlimit ? dp->d_blk_softlimit : dp->d_blk_hardlimit; + limit = dp->d_blk_softlimit ? + be64_to_cpu(dp->d_blk_softlimit) : + be64_to_cpu(dp->d_blk_hardlimit); if (limit && statp->f_blocks > limit) { statp->f_blocks = limit; - statp->f_bfree = (statp->f_blocks > dp->d_bcount) ? - (statp->f_blocks - dp->d_bcount) : 0; + statp->f_bfree = + (statp->f_blocks > be64_to_cpu(dp->d_bcount)) ? + (statp->f_blocks - be64_to_cpu(dp->d_bcount)) : 0; } - limit = dp->d_ino_softlimit ? dp->d_ino_softlimit : dp->d_ino_hardlimit; + + limit = dp->d_ino_softlimit ? + be64_to_cpu(dp->d_ino_softlimit) : + be64_to_cpu(dp->d_ino_hardlimit); if (limit && statp->f_files > limit) { statp->f_files = limit; - statp->f_ffree = (statp->f_files > dp->d_icount) ? - (statp->f_ffree - dp->d_icount) : 0; + statp->f_ffree = + (statp->f_files > be64_to_cpu(dp->d_icount)) ? + (statp->f_ffree - be64_to_cpu(dp->d_icount)) : 0; } xfs_qm_dqput(dqp); diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c index eef6763f3a6..d2bbcd882a6 100644 --- a/fs/xfs/xfs_alloc.c +++ b/fs/xfs/xfs_alloc.c @@ -1835,40 +1835,47 @@ xfs_alloc_fix_freelist( &agbp))) return error; if (!pag->pagf_init) { + ASSERT(flags & XFS_ALLOC_FLAG_TRYLOCK); + ASSERT(!(flags & XFS_ALLOC_FLAG_FREEING)); args->agbp = NULL; return 0; } } else agbp = NULL; - /* If this is a metadata preferred pag and we are user data + /* + * If this is a metadata preferred pag and we are user data * then try somewhere else if we are not being asked to * try harder at this point */ - if (pag->pagf_metadata && args->userdata && flags) { + if (pag->pagf_metadata && args->userdata && + (flags & XFS_ALLOC_FLAG_TRYLOCK)) { + ASSERT(!(flags & XFS_ALLOC_FLAG_FREEING)); args->agbp = NULL; return 0; } - need = XFS_MIN_FREELIST_PAG(pag, mp); - delta = need > pag->pagf_flcount ? need - pag->pagf_flcount : 0; - /* - * If it looks like there isn't a long enough extent, or enough - * total blocks, reject it. - */ - longest = (pag->pagf_longest > delta) ? - (pag->pagf_longest - delta) : - (pag->pagf_flcount > 0 || pag->pagf_longest > 0); - if (args->minlen + args->alignment + args->minalignslop - 1 > longest || - (!(flags & XFS_ALLOC_FLAG_FREEING) && - (int)(pag->pagf_freeblks + pag->pagf_flcount - - need - args->total) < - (int)args->minleft)) { - if (agbp) - xfs_trans_brelse(tp, agbp); - args->agbp = NULL; - return 0; + if (!(flags & XFS_ALLOC_FLAG_FREEING)) { + need = XFS_MIN_FREELIST_PAG(pag, mp); + delta = need > pag->pagf_flcount ? need - pag->pagf_flcount : 0; + /* + * If it looks like there isn't a long enough extent, or enough + * total blocks, reject it. + */ + longest = (pag->pagf_longest > delta) ? + (pag->pagf_longest - delta) : + (pag->pagf_flcount > 0 || pag->pagf_longest > 0); + if ((args->minlen + args->alignment + args->minalignslop - 1) > + longest || + ((int)(pag->pagf_freeblks + pag->pagf_flcount - + need - args->total) < (int)args->minleft)) { + if (agbp) + xfs_trans_brelse(tp, agbp); + args->agbp = NULL; + return 0; + } } + /* * Get the a.g. freespace buffer. * Can fail if we're not blocking on locks, and it's held. @@ -1878,6 +1885,8 @@ xfs_alloc_fix_freelist( &agbp))) return error; if (agbp == NULL) { + ASSERT(flags & XFS_ALLOC_FLAG_TRYLOCK); + ASSERT(!(flags & XFS_ALLOC_FLAG_FREEING)); args->agbp = NULL; return 0; } @@ -1887,22 +1896,24 @@ xfs_alloc_fix_freelist( */ agf = XFS_BUF_TO_AGF(agbp); need = XFS_MIN_FREELIST(agf, mp); - delta = need > be32_to_cpu(agf->agf_flcount) ? - (need - be32_to_cpu(agf->agf_flcount)) : 0; /* * If there isn't enough total or single-extent, reject it. */ - longest = be32_to_cpu(agf->agf_longest); - longest = (longest > delta) ? (longest - delta) : - (be32_to_cpu(agf->agf_flcount) > 0 || longest > 0); - if (args->minlen + args->alignment + args->minalignslop - 1 > longest || - (!(flags & XFS_ALLOC_FLAG_FREEING) && - (int)(be32_to_cpu(agf->agf_freeblks) + - be32_to_cpu(agf->agf_flcount) - need - args->total) < - (int)args->minleft)) { - xfs_trans_brelse(tp, agbp); - args->agbp = NULL; - return 0; + if (!(flags & XFS_ALLOC_FLAG_FREEING)) { + delta = need > be32_to_cpu(agf->agf_flcount) ? + (need - be32_to_cpu(agf->agf_flcount)) : 0; + longest = be32_to_cpu(agf->agf_longest); + longest = (longest > delta) ? (longest - delta) : + (be32_to_cpu(agf->agf_flcount) > 0 || longest > 0); + if ((args->minlen + args->alignment + args->minalignslop - 1) > + longest || + ((int)(be32_to_cpu(agf->agf_freeblks) + + be32_to_cpu(agf->agf_flcount) - need - args->total) < + (int)args->minleft)) { + xfs_trans_brelse(tp, agbp); + args->agbp = NULL; + return 0; + } } /* * Make the freelist shorter if it's too long. @@ -1950,12 +1961,11 @@ xfs_alloc_fix_freelist( * on a completely full ag. */ if (targs.agbno == NULLAGBLOCK) { - if (!(flags & XFS_ALLOC_FLAG_FREEING)) { - xfs_trans_brelse(tp, agflbp); - args->agbp = NULL; - return 0; - } - break; + if (flags & XFS_ALLOC_FLAG_FREEING) + break; + xfs_trans_brelse(tp, agflbp); + args->agbp = NULL; + return 0; } /* * Put each allocated block on the list. @@ -2442,31 +2452,26 @@ xfs_free_extent( xfs_fsblock_t bno, /* starting block number of extent */ xfs_extlen_t len) /* length of extent */ { -#ifdef DEBUG - xfs_agf_t *agf; /* a.g. freespace header */ -#endif - xfs_alloc_arg_t args; /* allocation argument structure */ + xfs_alloc_arg_t args; int error; ASSERT(len != 0); + memset(&args, 0, sizeof(xfs_alloc_arg_t)); args.tp = tp; args.mp = tp->t_mountp; args.agno = XFS_FSB_TO_AGNO(args.mp, bno); ASSERT(args.agno < args.mp->m_sb.sb_agcount); args.agbno = XFS_FSB_TO_AGBNO(args.mp, bno); - args.alignment = 1; - args.minlen = args.minleft = args.minalignslop = 0; down_read(&args.mp->m_peraglock); args.pag = &args.mp->m_perag[args.agno]; if ((error = xfs_alloc_fix_freelist(&args, XFS_ALLOC_FLAG_FREEING))) goto error0; #ifdef DEBUG ASSERT(args.agbp != NULL); - agf = XFS_BUF_TO_AGF(args.agbp); - ASSERT(args.agbno + len <= be32_to_cpu(agf->agf_length)); + ASSERT((args.agbno + len) <= + be32_to_cpu(XFS_BUF_TO_AGF(args.agbp)->agf_length)); #endif - error = xfs_free_ag_extent(tp, args.agbp, args.agno, args.agbno, - len, 0); + error = xfs_free_ag_extent(tp, args.agbp, args.agno, args.agbno, len, 0); error0: up_read(&args.mp->m_peraglock); return error; diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h index 650591f999a..5a4256120cc 100644 --- a/fs/xfs/xfs_alloc.h +++ b/fs/xfs/xfs_alloc.h @@ -44,6 +44,26 @@ typedef enum xfs_alloctype #define XFS_ALLOC_FLAG_FREEING 0x00000002 /* indicate caller is freeing extents*/ /* + * In order to avoid ENOSPC-related deadlock caused by + * out-of-order locking of AGF buffer (PV 947395), we place + * constraints on the relationship among actual allocations for + * data blocks, freelist blocks, and potential file data bmap + * btree blocks. However, these restrictions may result in no + * actual space allocated for a delayed extent, for example, a data + * block in a certain AG is allocated but there is no additional + * block for the additional bmap btree block due to a split of the + * bmap btree of the file. The result of this may lead to an + * infinite loop in xfssyncd when the file gets flushed to disk and + * all delayed extents need to be actually allocated. To get around + * this, we explicitly set aside a few blocks which will not be + * reserved in delayed allocation. Considering the minimum number of + * needed freelist blocks is 4 fsbs _per AG_, a potential split of file's bmap + * btree requires 1 fsb, so we set the number of set-aside blocks + * to 4 + 4*agcount. + */ +#define XFS_ALLOC_SET_ASIDE(mp) (4 + ((mp)->m_sb.sb_agcount * 4)) + +/* * Argument structure for xfs_alloc routines. * This is turned into a structure to avoid having 20 arguments passed * down several levels of the stack. diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 3a613753906..bf46fae303a 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -4993,7 +4993,7 @@ xfs_bmapi( bma.firstblock = *firstblock; bma.alen = alen; bma.off = aoff; - bma.conv = (flags & XFS_BMAPI_CONVERT); + bma.conv = !!(flags & XFS_BMAPI_CONVERT); bma.wasdel = wasdelay; bma.minlen = minlen; bma.low = flist->xbf_low; diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 077629bab53..c064e72ada9 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -462,7 +462,7 @@ xfs_fs_counts( xfs_icsb_sync_counters_lazy(mp); s = XFS_SB_LOCK(mp); - cnt->freedata = mp->m_sb.sb_fdblocks; + cnt->freedata = mp->m_sb.sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); cnt->freertx = mp->m_sb.sb_frextents; cnt->freeino = mp->m_sb.sb_ifree; cnt->allocino = mp->m_sb.sb_icount; @@ -519,15 +519,19 @@ xfs_reserve_blocks( } mp->m_resblks = request; } else { + __int64_t free; + + free = mp->m_sb.sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); delta = request - mp->m_resblks; - lcounter = mp->m_sb.sb_fdblocks - delta; + lcounter = free - delta; if (lcounter < 0) { /* We can't satisfy the request, just get what we can */ - mp->m_resblks += mp->m_sb.sb_fdblocks; - mp->m_resblks_avail += mp->m_sb.sb_fdblocks; - mp->m_sb.sb_fdblocks = 0; + mp->m_resblks += free; + mp->m_resblks_avail += free; + mp->m_sb.sb_fdblocks = XFS_ALLOC_SET_ASIDE(mp); } else { - mp->m_sb.sb_fdblocks = lcounter; + mp->m_sb.sb_fdblocks = + lcounter + XFS_ALLOC_SET_ASIDE(mp); mp->m_resblks = request; mp->m_resblks_avail += delta; } diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 86c1bf0bba9..1f8ecff8553 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -334,10 +334,9 @@ xfs_itobp( #if !defined(__KERNEL__) ni = 0; #elif defined(DEBUG) - ni = (imap_flags & XFS_IMAP_BULKSTAT) ? 0 : - (BBTOB(imap.im_len) >> mp->m_sb.sb_inodelog); + ni = BBTOB(imap.im_len) >> mp->m_sb.sb_inodelog; #else /* usual case */ - ni = (imap_flags & XFS_IMAP_BULKSTAT) ? 0 : 1; + ni = 1; #endif for (i = 0; i < ni; i++) { @@ -348,11 +347,15 @@ xfs_itobp( (i << mp->m_sb.sb_inodelog)); di_ok = INT_GET(dip->di_core.di_magic, ARCH_CONVERT) == XFS_DINODE_MAGIC && XFS_DINODE_GOOD_VERSION(INT_GET(dip->di_core.di_version, ARCH_CONVERT)); - if (unlikely(XFS_TEST_ERROR(!di_ok, mp, XFS_ERRTAG_ITOBP_INOTOBP, - XFS_RANDOM_ITOBP_INOTOBP))) { + if (unlikely(XFS_TEST_ERROR(!di_ok, mp, + XFS_ERRTAG_ITOBP_INOTOBP, + XFS_RANDOM_ITOBP_INOTOBP))) { + if (imap_flags & XFS_IMAP_BULKSTAT) { + xfs_trans_brelse(tp, bp); + return XFS_ERROR(EINVAL); + } #ifdef DEBUG - if (!(imap_flags & XFS_IMAP_BULKSTAT)) - cmn_err(CE_ALERT, + cmn_err(CE_ALERT, "Device %s - bad inode magic/vsn " "daddr %lld #%d (magic=%x)", XFS_BUFTARG_NAME(mp->m_ddev_targp), diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index e730328636c..21ac1a67e3e 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1413,7 +1413,7 @@ xlog_sync(xlog_t *log, ops = iclog->ic_header.h_num_logops; INT_SET(iclog->ic_header.h_num_logops, ARCH_CONVERT, ops); - bp = iclog->ic_bp; + bp = iclog->ic_bp; ASSERT(XFS_BUF_FSPRIVATE2(bp, unsigned long) == (unsigned long)1); XFS_BUF_SET_FSPRIVATE2(bp, (unsigned long)2); XFS_BUF_SET_ADDR(bp, BLOCK_LSN(INT_GET(iclog->ic_header.h_lsn, ARCH_CONVERT))); @@ -1430,15 +1430,14 @@ xlog_sync(xlog_t *log, } XFS_BUF_SET_PTR(bp, (xfs_caddr_t) &(iclog->ic_header), count); XFS_BUF_SET_FSPRIVATE(bp, iclog); /* save for later */ + XFS_BUF_ZEROFLAGS(bp); XFS_BUF_BUSY(bp); XFS_BUF_ASYNC(bp); /* * Do an ordered write for the log block. - * - * It may not be needed to flush the first split block in the log wrap - * case, but do it anyways to be safe -AK + * Its unnecessary to flush the first split block in the log wrap case. */ - if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) + if (!split && (log->l_mp->m_flags & XFS_MOUNT_BARRIER)) XFS_BUF_ORDERED(bp); ASSERT(XFS_BUF_ADDR(bp) <= log->l_logBBsize-1); @@ -1460,7 +1459,7 @@ xlog_sync(xlog_t *log, return error; } if (split) { - bp = iclog->ic_log->l_xbuf; + bp = iclog->ic_log->l_xbuf; ASSERT(XFS_BUF_FSPRIVATE2(bp, unsigned long) == (unsigned long)1); XFS_BUF_SET_FSPRIVATE2(bp, (unsigned long)2); @@ -1468,6 +1467,7 @@ xlog_sync(xlog_t *log, XFS_BUF_SET_PTR(bp, (xfs_caddr_t)((__psint_t)&(iclog->ic_header)+ (__psint_t)count), split); XFS_BUF_SET_FSPRIVATE(bp, iclog); + XFS_BUF_ZEROFLAGS(bp); XFS_BUF_BUSY(bp); XFS_BUF_ASYNC(bp); if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 4be5c0b2d29..9dfae18d995 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1243,24 +1243,6 @@ xfs_mod_sb(xfs_trans_t *tp, __int64_t fields) xfs_trans_log_buf(tp, bp, first, last); } -/* - * In order to avoid ENOSPC-related deadlock caused by - * out-of-order locking of AGF buffer (PV 947395), we place - * constraints on the relationship among actual allocations for - * data blocks, freelist blocks, and potential file data bmap - * btree blocks. However, these restrictions may result in no - * actual space allocated for a delayed extent, for example, a data - * block in a certain AG is allocated but there is no additional - * block for the additional bmap btree block due to a split of the - * bmap btree of the file. The result of this may lead to an - * infinite loop in xfssyncd when the file gets flushed to disk and - * all delayed extents need to be actually allocated. To get around - * this, we explicitly set aside a few blocks which will not be - * reserved in delayed allocation. Considering the minimum number of - * needed freelist blocks is 4 fsbs, a potential split of file's bmap - * btree requires 1 fsb, so we set the number of set-aside blocks to 8. -*/ -#define SET_ASIDE_BLOCKS 8 /* * xfs_mod_incore_sb_unlocked() is a utility routine common used to apply @@ -1306,7 +1288,8 @@ xfs_mod_incore_sb_unlocked(xfs_mount_t *mp, xfs_sb_field_t field, return 0; case XFS_SBS_FDBLOCKS: - lcounter = (long long)mp->m_sb.sb_fdblocks - SET_ASIDE_BLOCKS; + lcounter = (long long) + mp->m_sb.sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); res_used = (long long)(mp->m_resblks - mp->m_resblks_avail); if (delta > 0) { /* Putting blocks back */ @@ -1340,7 +1323,7 @@ xfs_mod_incore_sb_unlocked(xfs_mount_t *mp, xfs_sb_field_t field, } } - mp->m_sb.sb_fdblocks = lcounter + SET_ASIDE_BLOCKS; + mp->m_sb.sb_fdblocks = lcounter + XFS_ALLOC_SET_ASIDE(mp); return 0; case XFS_SBS_FREXTENTS: lcounter = (long long)mp->m_sb.sb_frextents; @@ -2021,7 +2004,8 @@ xfs_icsb_sync_counters_lazy( * when we get near ENOSPC. */ #define XFS_ICSB_INO_CNTR_REENABLE 64 -#define XFS_ICSB_FDBLK_CNTR_REENABLE 512 +#define XFS_ICSB_FDBLK_CNTR_REENABLE(mp) \ + (512 + XFS_ALLOC_SET_ASIDE(mp)) STATIC void xfs_icsb_balance_counter( xfs_mount_t *mp, @@ -2055,7 +2039,7 @@ xfs_icsb_balance_counter( case XFS_SBS_FDBLOCKS: count = mp->m_sb.sb_fdblocks; resid = do_div(count, weight); - if (count < XFS_ICSB_FDBLK_CNTR_REENABLE) + if (count < XFS_ICSB_FDBLK_CNTR_REENABLE(mp)) goto out; break; default: @@ -2110,11 +2094,11 @@ again: case XFS_SBS_FDBLOCKS: BUG_ON((mp->m_resblks - mp->m_resblks_avail) != 0); - lcounter = icsbp->icsb_fdblocks; + lcounter = icsbp->icsb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); lcounter += delta; if (unlikely(lcounter < 0)) goto slow_path; - icsbp->icsb_fdblocks = lcounter; + icsbp->icsb_fdblocks = lcounter + XFS_ALLOC_SET_ASIDE(mp); break; default: BUG(); diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c index 6c96391f3f1..a34796e57af 100644 --- a/fs/xfs/xfs_vfsops.c +++ b/fs/xfs/xfs_vfsops.c @@ -515,7 +515,7 @@ xfs_mount( if (error) goto error2; - if ((mp->m_flags & XFS_MOUNT_BARRIER) && !(vfsp->vfs_flag & VFS_RDONLY)) + if (mp->m_flags & XFS_MOUNT_BARRIER) xfs_mountfs_check_barriers(mp); error = XFS_IOINIT(vfsp, args, flags); @@ -811,7 +811,8 @@ xfs_statvfs( statp->f_bsize = sbp->sb_blocksize; lsize = sbp->sb_logstart ? sbp->sb_logblocks : 0; statp->f_blocks = sbp->sb_dblocks - lsize; - statp->f_bfree = statp->f_bavail = sbp->sb_fdblocks; + statp->f_bfree = statp->f_bavail = + sbp->sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); fakeinos = statp->f_bfree << sbp->sb_inopblog; #if XFS_BIG_INUMS fakeinos += mp->m_inoadd; |