diff options
Diffstat (limited to 'fs')
55 files changed, 1580 insertions, 656 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/binfmt_elf.c b/fs/binfmt_elf.c index f42e64210ee..672a3b90bc5 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1185,8 +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/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/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/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/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/inode.c b/fs/ext3/inode.c index f804d5e9d60..c5ee9f0691e 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1158,7 +1158,7 @@ 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 +1244,7 @@ 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 +1494,7 @@ 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 +2402,15 @@ 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/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/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/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/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/locks.c b/fs/locks.c index 1ad29c9b625..b0b41a64e10 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) { diff --git a/fs/namei.c b/fs/namei.c index c9750d755af..55a131230f9 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); @@ -1125,7 +1125,7 @@ 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 +1357,7 @@ 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) @@ -1659,6 +1659,7 @@ 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 +1670,7 @@ do_last: if (flag & O_NOFOLLOW) goto exit_dput; } + error = -ENOENT; if (!path.dentry->d_inode) goto exit_dput; @@ -1712,8 +1714,14 @@ 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; 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..fecd3b095de 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 +87,37 @@ 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); +} + +/* + * "size" is never larger than rsize or wsize. + */ +static inline int nfs_direct_count_pages(unsigned long user_addr, size_t size) +{ + int page_count; + + page_count = (user_addr + size + PAGE_SIZE - 1) >> PAGE_SHIFT; + page_count -= user_addr >> PAGE_SHIFT; + BUG_ON(page_count < 0); + + return page_count; +} + +static inline unsigned int nfs_max_pages(unsigned int size) +{ + return (size + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; +} /** * nfs_direct_IO - NFS address space operation for direct I/O @@ -118,50 +141,21 @@ 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 +167,13 @@ 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 +214,11 @@ 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 +231,10 @@ 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 +243,9 @@ 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 +253,10 @@ 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 +265,60 @@ 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 rpages = nfs_max_pages(rsize); + unsigned int pgbase; + int result; + ssize_t started = 0; + + get_dreq(dreq); - curpage = 0; - pgbase = dreq->user_addr & ~PAGE_MASK; + pgbase = user_addr & ~PAGE_MASK; do { struct nfs_read_data *data; size_t bytes; + result = -ENOMEM; + data = nfs_readdata_alloc(rpages); + if (unlikely(!data)) + break; + bytes = rsize; if (count < rsize) bytes = count; - BUG_ON(list_empty(list)); - data = list_entry(list->next, struct nfs_read_data, pages); - list_del_init(&data->pages); + data->npages = nfs_direct_count_pages(user_addr, bytes); + 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 +341,35 @@ 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; pgbase += bytes; - curpage += pgbase >> PAGE_SHIFT; pgbase &= ~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 +377,9 @@ 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 +387,10 @@ 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 +398,51 @@ 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); + + /* + * 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); + } - nfs_direct_write_schedule(dreq, FLUSH_STABLE); + if (put_dreq(dreq)) + nfs_direct_write_complete(dreq, inode); } static void nfs_direct_commit_result(struct rpc_task *task, void *calldata) @@ -471,8 +479,8 @@ 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; @@ -534,47 +542,6 @@ 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 +571,6 @@ 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 +584,8 @@ 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 +594,62 @@ 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 wpages = nfs_max_pages(wsize); + unsigned int pgbase; + int result; + ssize_t started = 0; - curpage = 0; - pgbase = dreq->user_addr & ~PAGE_MASK; + get_dreq(dreq); + + pgbase = user_addr & ~PAGE_MASK; do { struct nfs_write_data *data; size_t bytes; + result = -ENOMEM; + data = nfs_writedata_alloc(wpages); + if (unlikely(!data)) + break; + bytes = wsize; if (count < wsize) bytes = count; - BUG_ON(list_empty(list)); - data = list_entry(list->next, struct nfs_write_data, pages); + data->npages = nfs_direct_count_pages(user_addr, bytes); + 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); + 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 +673,26 @@ 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; pgbase += bytes; - curpage += pgbase >> PAGE_SHIFT; pgbase &= ~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 +700,14 @@ 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 +718,9 @@ 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 +750,6 @@ 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 +771,7 @@ 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 +807,6 @@ 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 +834,7 @@ 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/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..e6ee97f19d8 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -3144,9 +3144,6 @@ 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 +3255,6 @@ 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 +3265,28 @@ 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 +3454,10 @@ 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 +3476,8 @@ 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 +3489,42 @@ 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/read.c b/fs/nfs/read.c index 52bf634260a..65c0c5b3235 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -63,7 +63,7 @@ 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); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index bca5734ca9f..50774991f8d 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -137,7 +137,7 @@ 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); @@ -578,7 +578,7 @@ 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 +589,19 @@ 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 @@ -1381,6 +1394,7 @@ 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 +1513,7 @@ 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 +1526,7 @@ 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/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/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/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..f318b58510f 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -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/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/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/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/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/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_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..f137856c326 100644 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ b/fs/xfs/quota/xfs_qm_bhv.c @@ -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_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_vfsops.c b/fs/xfs/xfs_vfsops.c index 6c96391f3f1..b427d220a16 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); |