diff options
author | pjd <pjd@FreeBSD.org> | 2007-04-06 01:09:06 +0000 |
---|---|---|
committer | pjd <pjd@FreeBSD.org> | 2007-04-06 01:09:06 +0000 |
commit | 3b005d330261f33318ca1ee3fef1940237fd788b (patch) | |
tree | 3061c8734d9ce560165e672836837a0f411a83c9 /cddl/contrib/opensolaris/lib/libzpool | |
parent | 3be454b8211f48e634e6587f53807d3b5013e973 (diff) | |
download | FreeBSD-src-3b005d330261f33318ca1ee3fef1940237fd788b.zip FreeBSD-src-3b005d330261f33318ca1ee3fef1940237fd788b.tar.gz |
Please welcome ZFS - The last word in file systems.
ZFS file system was ported from OpenSolaris operating system. The code in under
CDDL license.
I'd like to thank all SUN developers that created this great piece of software.
Supported by: Wheel LTD (http://www.wheel.pl/)
Supported by: The FreeBSD Foundation (http://www.freebsdfoundation.org/)
Supported by: Sentex (http://www.sentex.net/)
Diffstat (limited to 'cddl/contrib/opensolaris/lib/libzpool')
-rw-r--r-- | cddl/contrib/opensolaris/lib/libzpool/common/kernel.c | 830 | ||||
-rw-r--r-- | cddl/contrib/opensolaris/lib/libzpool/common/sys/zfs_context.h | 497 | ||||
-rw-r--r-- | cddl/contrib/opensolaris/lib/libzpool/common/taskq.c | 250 | ||||
-rw-r--r-- | cddl/contrib/opensolaris/lib/libzpool/common/util.c | 144 |
4 files changed, 1721 insertions, 0 deletions
diff --git a/cddl/contrib/opensolaris/lib/libzpool/common/kernel.c b/cddl/contrib/opensolaris/lib/libzpool/common/kernel.c new file mode 100644 index 0000000..086f33e --- /dev/null +++ b/cddl/contrib/opensolaris/lib/libzpool/common/kernel.c @@ -0,0 +1,830 @@ +/* + * CDDL HEADER START + * + * The contents of this file are subject to the terms of the + * Common Development and Distribution License (the "License"). + * You may not use this file except in compliance with the License. + * + * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE + * or http://www.opensolaris.org/os/licensing. + * See the License for the specific language governing permissions + * and limitations under the License. + * + * When distributing Covered Code, include this CDDL HEADER in each + * file and include the License file at usr/src/OPENSOLARIS.LICENSE. + * If applicable, add the following below this CDDL HEADER, with the + * fields enclosed by brackets "[]" replaced with your own identifying + * information: Portions Copyright [yyyy] [name of copyright owner] + * + * CDDL HEADER END + */ +/* + * Copyright 2007 Sun Microsystems, Inc. All rights reserved. + * Use is subject to license terms. + */ + +#pragma ident "%Z%%M% %I% %E% SMI" + +#include <assert.h> +#include <fcntl.h> +#include <poll.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <zlib.h> +#include <sys/spa.h> +#include <sys/stat.h> +#include <sys/processor.h> +#include <sys/zfs_context.h> +#include <sys/zmod.h> + +/* + * Emulation of kernel services in userland. + */ + +int hz = 119; /* frequency when using gethrtime() >> 23 for lbolt */ +uint64_t physmem; +vnode_t *rootdir = (vnode_t *)0xabcd1234; + +/* + * ========================================================================= + * threads + * ========================================================================= + */ +/*ARGSUSED*/ +kthread_t * +zk_thread_create(void (*func)(), void *arg) +{ + thread_t tid; + + VERIFY(thr_create(0, 0, (void *(*)(void *))func, arg, THR_DETACHED, + &tid) == 0); + + return ((void *)(uintptr_t)tid); +} + +/* + * ========================================================================= + * kstats + * ========================================================================= + */ +/*ARGSUSED*/ +kstat_t * +kstat_create(char *module, int instance, char *name, char *class, + uchar_t type, ulong_t ndata, uchar_t ks_flag) +{ + return (NULL); +} + +/*ARGSUSED*/ +void +kstat_install(kstat_t *ksp) +{} + +/*ARGSUSED*/ +void +kstat_delete(kstat_t *ksp) +{} + +/* + * ========================================================================= + * mutexes + * ========================================================================= + */ +void +zmutex_init(kmutex_t *mp) +{ + mp->m_owner = NULL; + (void) _mutex_init(&mp->m_lock, USYNC_THREAD, NULL); +} + +void +zmutex_destroy(kmutex_t *mp) +{ + ASSERT(mp->m_owner == NULL); + (void) _mutex_destroy(&(mp)->m_lock); + mp->m_owner = (void *)-1UL; +} + +void +mutex_enter(kmutex_t *mp) +{ + ASSERT(mp->m_owner != (void *)-1UL); + ASSERT(mp->m_owner != curthread); + VERIFY(mutex_lock(&mp->m_lock) == 0); + ASSERT(mp->m_owner == NULL); + mp->m_owner = curthread; +} + +int +mutex_tryenter(kmutex_t *mp) +{ + ASSERT(mp->m_owner != (void *)-1UL); + if (mutex_trylock(&mp->m_lock) == 0) { + ASSERT(mp->m_owner == NULL); + mp->m_owner = curthread; + return (1); + } else { + return (0); + } +} + +void +mutex_exit(kmutex_t *mp) +{ + ASSERT(mp->m_owner == curthread); + mp->m_owner = NULL; + VERIFY(mutex_unlock(&mp->m_lock) == 0); +} + +void * +mutex_owner(kmutex_t *mp) +{ + return (mp->m_owner); +} + +/* + * ========================================================================= + * rwlocks + * ========================================================================= + */ +/*ARGSUSED*/ +void +rw_init(krwlock_t *rwlp, char *name, int type, void *arg) +{ + rwlock_init(&rwlp->rw_lock, USYNC_THREAD, NULL); + rwlp->rw_owner = NULL; + rwlp->rw_count = 0; +} + +void +rw_destroy(krwlock_t *rwlp) +{ + rwlock_destroy(&rwlp->rw_lock); + rwlp->rw_owner = (void *)-1UL; + rwlp->rw_count = -2; +} + +void +rw_enter(krwlock_t *rwlp, krw_t rw) +{ + //ASSERT(!RW_LOCK_HELD(rwlp)); + ASSERT(rwlp->rw_owner != (void *)-1UL); + ASSERT(rwlp->rw_owner != curthread); + + if (rw == RW_READER) { + (void) rw_rdlock(&rwlp->rw_lock); + ASSERT(rwlp->rw_count >= 0); + atomic_add_int(&rwlp->rw_count, 1); + } else { + (void) rw_wrlock(&rwlp->rw_lock); + ASSERT(rwlp->rw_count == 0); + rwlp->rw_count = -1; + rwlp->rw_owner = curthread; + } +} + +void +rw_exit(krwlock_t *rwlp) +{ + ASSERT(rwlp->rw_owner != (void *)-1UL); + + if (rwlp->rw_owner == curthread) { + /* Write locked. */ + ASSERT(rwlp->rw_count == -1); + rwlp->rw_count = 0; + rwlp->rw_owner = NULL; + } else { + /* Read locked. */ + ASSERT(rwlp->rw_count > 0); + atomic_add_int(&rwlp->rw_count, -1); + } + (void) rw_unlock(&rwlp->rw_lock); +} + +int +rw_tryenter(krwlock_t *rwlp, krw_t rw) +{ + int rv; + + ASSERT(rwlp->rw_owner != (void *)-1UL); + ASSERT(rwlp->rw_owner != curthread); + + if (rw == RW_READER) + rv = rw_tryrdlock(&rwlp->rw_lock); + else + rv = rw_trywrlock(&rwlp->rw_lock); + + if (rv == 0) { + ASSERT(rwlp->rw_owner == NULL); + if (rw == RW_READER) { + ASSERT(rwlp->rw_count >= 0); + atomic_add_int(&rwlp->rw_count, 1); + } else { + ASSERT(rwlp->rw_count == 0); + rwlp->rw_count = -1; + rwlp->rw_owner = curthread; + } + return (1); + } + + return (0); +} + +/*ARGSUSED*/ +int +rw_tryupgrade(krwlock_t *rwlp) +{ + ASSERT(rwlp->rw_owner != (void *)-1UL); + + return (0); +} + +int +rw_lock_held(krwlock_t *rwlp) +{ + + return (rwlp->rw_count != 0); +} + +/* + * ========================================================================= + * condition variables + * ========================================================================= + */ +/*ARGSUSED*/ +void +cv_init(kcondvar_t *cv, char *name, int type, void *arg) +{ + VERIFY(cond_init(cv, name, NULL) == 0); +} + +void +cv_destroy(kcondvar_t *cv) +{ + VERIFY(cond_destroy(cv) == 0); +} + +void +cv_wait(kcondvar_t *cv, kmutex_t *mp) +{ + ASSERT(mutex_owner(mp) == curthread); + mp->m_owner = NULL; + int ret = cond_wait(cv, &mp->m_lock); + VERIFY(ret == 0 || ret == EINTR); + mp->m_owner = curthread; +} + +clock_t +cv_timedwait(kcondvar_t *cv, kmutex_t *mp, clock_t abstime) +{ + int error; + struct timespec ts; + struct timeval tv; + clock_t delta; + + ASSERT(abstime > 0); +top: + delta = abstime - lbolt; + if (delta <= 0) + return (-1); + + if (gettimeofday(&tv, NULL) != 0) + assert(!"gettimeofday() failed"); + + ts.tv_sec = tv.tv_sec + delta / hz; + ts.tv_nsec = tv.tv_usec * 1000 + (delta % hz) * (NANOSEC / hz); + ASSERT(ts.tv_nsec >= 0); + + if(ts.tv_nsec >= NANOSEC) { + ts.tv_sec++; + ts.tv_nsec -= NANOSEC; + } + + ASSERT(mutex_owner(mp) == curthread); + mp->m_owner = NULL; + error = pthread_cond_timedwait(cv, &mp->m_lock, &ts); + mp->m_owner = curthread; + + if (error == EINTR) + goto top; + + if (error == ETIMEDOUT) + return (-1); + + ASSERT(error == 0); + + return (1); +} + +void +cv_signal(kcondvar_t *cv) +{ + VERIFY(cond_signal(cv) == 0); +} + +void +cv_broadcast(kcondvar_t *cv) +{ + VERIFY(cond_broadcast(cv) == 0); +} + +/* + * ========================================================================= + * vnode operations + * ========================================================================= + */ +/* + * Note: for the xxxat() versions of these functions, we assume that the + * starting vp is always rootdir (which is true for spa_directory.c, the only + * ZFS consumer of these interfaces). We assert this is true, and then emulate + * them by adding '/' in front of the path. + */ + +/*ARGSUSED*/ +int +vn_open(char *path, int x1, int flags, int mode, vnode_t **vpp, int x2, int x3) +{ + int fd; + vnode_t *vp; + int old_umask; + char realpath[MAXPATHLEN]; + struct stat64 st; + + /* + * If we're accessing a real disk from userland, we need to use + * the character interface to avoid caching. This is particularly + * important if we're trying to look at a real in-kernel storage + * pool from userland, e.g. via zdb, because otherwise we won't + * see the changes occurring under the segmap cache. + * On the other hand, the stupid character device returns zero + * for its size. So -- gag -- we open the block device to get + * its size, and remember it for subsequent VOP_GETATTR(). + */ + if (strncmp(path, "/dev/", 5) == 0) { + char *dsk; + fd = open64(path, O_RDONLY); + if (fd == -1) + return (errno); + if (fstat64(fd, &st) == -1) { + close(fd); + return (errno); + } + close(fd); + (void) sprintf(realpath, "%s", path); + dsk = strstr(path, "/dsk/"); + if (dsk != NULL) + (void) sprintf(realpath + (dsk - path) + 1, "r%s", + dsk + 1); + } else { + (void) sprintf(realpath, "%s", path); + if (!(flags & FCREAT) && stat64(realpath, &st) == -1) + return (errno); + } + + if (flags & FCREAT) + old_umask = umask(0); + + /* + * The construct 'flags - FREAD' conveniently maps combinations of + * FREAD and FWRITE to the corresponding O_RDONLY, O_WRONLY, and O_RDWR. + */ + fd = open64(realpath, flags - FREAD, mode); + + if (flags & FCREAT) + (void) umask(old_umask); + + if (fd == -1) + return (errno); + + if (fstat64(fd, &st) == -1) { + close(fd); + return (errno); + } + + (void) fcntl(fd, F_SETFD, FD_CLOEXEC); + + *vpp = vp = umem_zalloc(sizeof (vnode_t), UMEM_NOFAIL); + + vp->v_fd = fd; + vp->v_size = st.st_size; + vp->v_path = spa_strdup(path); + + return (0); +} + +int +vn_openat(char *path, int x1, int flags, int mode, vnode_t **vpp, int x2, + int x3, vnode_t *startvp) +{ + char *realpath = umem_alloc(strlen(path) + 2, UMEM_NOFAIL); + int ret; + + ASSERT(startvp == rootdir); + (void) sprintf(realpath, "/%s", path); + + ret = vn_open(realpath, x1, flags, mode, vpp, x2, x3); + + umem_free(realpath, strlen(path) + 2); + + return (ret); +} + +/*ARGSUSED*/ +int +vn_rdwr(int uio, vnode_t *vp, void *addr, ssize_t len, offset_t offset, + int x1, int x2, rlim64_t x3, void *x4, ssize_t *residp) +{ + ssize_t iolen, split; + + if (uio == UIO_READ) { + iolen = pread64(vp->v_fd, addr, len, offset); + } else { + /* + * To simulate partial disk writes, we split writes into two + * system calls so that the process can be killed in between. + */ + split = (len > 0 ? rand() % len : 0); + iolen = pwrite64(vp->v_fd, addr, split, offset); + iolen += pwrite64(vp->v_fd, (char *)addr + split, + len - split, offset + split); + } + + if (iolen == -1) + return (errno); + if (residp) + *residp = len - iolen; + else if (iolen != len) + return (EIO); + return (0); +} + +void +vn_close(vnode_t *vp) +{ + close(vp->v_fd); + spa_strfree(vp->v_path); + umem_free(vp, sizeof (vnode_t)); +} + +#ifdef ZFS_DEBUG + +/* + * ========================================================================= + * Figure out which debugging statements to print + * ========================================================================= + */ + +static char *dprintf_string; +static int dprintf_print_all; + +int +dprintf_find_string(const char *string) +{ + char *tmp_str = dprintf_string; + int len = strlen(string); + + /* + * Find out if this is a string we want to print. + * String format: file1.c,function_name1,file2.c,file3.c + */ + + while (tmp_str != NULL) { + if (strncmp(tmp_str, string, len) == 0 && + (tmp_str[len] == ',' || tmp_str[len] == '\0')) + return (1); + tmp_str = strchr(tmp_str, ','); + if (tmp_str != NULL) + tmp_str++; /* Get rid of , */ + } + return (0); +} + +void +dprintf_setup(int *argc, char **argv) +{ + int i, j; + + /* + * Debugging can be specified two ways: by setting the + * environment variable ZFS_DEBUG, or by including a + * "debug=..." argument on the command line. The command + * line setting overrides the environment variable. + */ + + for (i = 1; i < *argc; i++) { + int len = strlen("debug="); + /* First look for a command line argument */ + if (strncmp("debug=", argv[i], len) == 0) { + dprintf_string = argv[i] + len; + /* Remove from args */ + for (j = i; j < *argc; j++) + argv[j] = argv[j+1]; + argv[j] = NULL; + (*argc)--; + } + } + + if (dprintf_string == NULL) { + /* Look for ZFS_DEBUG environment variable */ + dprintf_string = getenv("ZFS_DEBUG"); + } + + /* + * Are we just turning on all debugging? + */ + if (dprintf_find_string("on")) + dprintf_print_all = 1; +} + +/* + * ========================================================================= + * debug printfs + * ========================================================================= + */ +void +__dprintf(const char *file, const char *func, int line, const char *fmt, ...) +{ + const char *newfile; + va_list adx; + + /* + * Get rid of annoying "../common/" prefix to filename. + */ + newfile = strrchr(file, '/'); + if (newfile != NULL) { + newfile = newfile + 1; /* Get rid of leading / */ + } else { + newfile = file; + } + + if (dprintf_print_all || + dprintf_find_string(newfile) || + dprintf_find_string(func)) { + /* Print out just the function name if requested */ + flockfile(stdout); + if (dprintf_find_string("pid")) + (void) printf("%d ", getpid()); + if (dprintf_find_string("tid")) + (void) printf("%u ", thr_self()); +#if 0 + if (dprintf_find_string("cpu")) + (void) printf("%u ", getcpuid()); +#endif + if (dprintf_find_string("time")) + (void) printf("%llu ", gethrtime()); + if (dprintf_find_string("long")) + (void) printf("%s, line %d: ", newfile, line); + (void) printf("%s: ", func); + va_start(adx, fmt); + (void) vprintf(fmt, adx); + va_end(adx); + funlockfile(stdout); + } +} + +#endif /* ZFS_DEBUG */ + +/* + * ========================================================================= + * cmn_err() and panic() + * ========================================================================= + */ +static char ce_prefix[CE_IGNORE][10] = { "", "NOTICE: ", "WARNING: ", "" }; +static char ce_suffix[CE_IGNORE][2] = { "", "\n", "\n", "" }; + +void +vpanic(const char *fmt, va_list adx) +{ + (void) fprintf(stderr, "error: "); + (void) vfprintf(stderr, fmt, adx); + (void) fprintf(stderr, "\n"); + + abort(); /* think of it as a "user-level crash dump" */ +} + +void +panic(const char *fmt, ...) +{ + va_list adx; + + va_start(adx, fmt); + vpanic(fmt, adx); + va_end(adx); +} + +void +vcmn_err(int ce, const char *fmt, va_list adx) +{ + if (ce == CE_PANIC) + vpanic(fmt, adx); + if (ce != CE_NOTE) { /* suppress noise in userland stress testing */ + (void) fprintf(stderr, "%s", ce_prefix[ce]); + (void) vfprintf(stderr, fmt, adx); + (void) fprintf(stderr, "%s", ce_suffix[ce]); + } +} + +/*PRINTFLIKE2*/ +void +cmn_err(int ce, const char *fmt, ...) +{ + va_list adx; + + va_start(adx, fmt); + vcmn_err(ce, fmt, adx); + va_end(adx); +} + +/* + * ========================================================================= + * kobj interfaces + * ========================================================================= + */ +struct _buf * +kobj_open_file(char *name) +{ + struct _buf *file; + vnode_t *vp; + + /* set vp as the _fd field of the file */ + if (vn_openat(name, UIO_SYSSPACE, FREAD, 0, &vp, 0, 0, rootdir) != 0) + return ((void *)-1UL); + + file = umem_zalloc(sizeof (struct _buf), UMEM_NOFAIL); + file->_fd = (intptr_t)vp; + return (file); +} + +int +kobj_read_file(struct _buf *file, char *buf, unsigned size, unsigned off) +{ + ssize_t resid; + + vn_rdwr(UIO_READ, (vnode_t *)file->_fd, buf, size, (offset_t)off, + UIO_SYSSPACE, 0, 0, 0, &resid); + + return (size - resid); +} + +void +kobj_close_file(struct _buf *file) +{ + vn_close((vnode_t *)file->_fd); + umem_free(file, sizeof (struct _buf)); +} + +int +kobj_get_filesize(struct _buf *file, uint64_t *size) +{ + struct stat64 st; + vnode_t *vp = (vnode_t *)file->_fd; + + if (fstat64(vp->v_fd, &st) == -1) { + vn_close(vp); + return (errno); + } + *size = st.st_size; + return (0); +} + +/* + * ========================================================================= + * misc routines + * ========================================================================= + */ + +void +delay(clock_t ticks) +{ + poll(0, 0, ticks * (1000 / hz)); +} + +#if 0 +/* + * Find highest one bit set. + * Returns bit number + 1 of highest bit that is set, otherwise returns 0. + * High order bit is 31 (or 63 in _LP64 kernel). + */ +int +highbit(ulong_t i) +{ + register int h = 1; + + if (i == 0) + return (0); +#ifdef _LP64 + if (i & 0xffffffff00000000ul) { + h += 32; i >>= 32; + } +#endif + if (i & 0xffff0000) { + h += 16; i >>= 16; + } + if (i & 0xff00) { + h += 8; i >>= 8; + } + if (i & 0xf0) { + h += 4; i >>= 4; + } + if (i & 0xc) { + h += 2; i >>= 2; + } + if (i & 0x2) { + h += 1; + } + return (h); +} +#endif + +static int +random_get_bytes_common(uint8_t *ptr, size_t len, char *devname) +{ + int fd = open(devname, O_RDONLY); + size_t resid = len; + ssize_t bytes; + + ASSERT(fd != -1); + + while (resid != 0) { + bytes = read(fd, ptr, resid); + ASSERT(bytes >= 0); + ptr += bytes; + resid -= bytes; + } + + close(fd); + + return (0); +} + +int +random_get_bytes(uint8_t *ptr, size_t len) +{ + return (random_get_bytes_common(ptr, len, "/dev/random")); +} + +int +random_get_pseudo_bytes(uint8_t *ptr, size_t len) +{ + return (random_get_bytes_common(ptr, len, "/dev/urandom")); +} + +/* + * ========================================================================= + * kernel emulation setup & teardown + * ========================================================================= + */ +static int +umem_out_of_memory(void) +{ + char errmsg[] = "out of memory -- generating core dump\n"; + + write(fileno(stderr), errmsg, sizeof (errmsg)); + abort(); + return (0); +} + +void +kernel_init(int mode) +{ + umem_nofail_callback(umem_out_of_memory); + + physmem = sysconf(_SC_PHYS_PAGES); + + dprintf("physmem = %llu pages (%.2f GB)\n", physmem, + (double)physmem * sysconf(_SC_PAGE_SIZE) / (1ULL << 30)); + + spa_init(mode); +} + +void +kernel_fini(void) +{ + spa_fini(); +} + +int +z_uncompress(void *dst, size_t *dstlen, const void *src, size_t srclen) +{ + int ret; + uLongf len = *dstlen; + + if ((ret = uncompress(dst, &len, src, srclen)) == Z_OK) + *dstlen = (size_t)len; + + return (ret); +} + +int +z_compress_level(void *dst, size_t *dstlen, const void *src, size_t srclen, + int level) +{ + int ret; + uLongf len = *dstlen; + + if ((ret = compress2(dst, &len, src, srclen, level)) == Z_OK) + *dstlen = (size_t)len; + + return (ret); +} diff --git a/cddl/contrib/opensolaris/lib/libzpool/common/sys/zfs_context.h b/cddl/contrib/opensolaris/lib/libzpool/common/sys/zfs_context.h new file mode 100644 index 0000000..6b9783f --- /dev/null +++ b/cddl/contrib/opensolaris/lib/libzpool/common/sys/zfs_context.h @@ -0,0 +1,497 @@ +/* + * CDDL HEADER START + * + * The contents of this file are subject to the terms of the + * Common Development and Distribution License (the "License"). + * You may not use this file except in compliance with the License. + * + * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE + * or http://www.opensolaris.org/os/licensing. + * See the License for the specific language governing permissions + * and limitations under the License. + * + * When distributing Covered Code, include this CDDL HEADER in each + * file and include the License file at usr/src/OPENSOLARIS.LICENSE. + * If applicable, add the following below this CDDL HEADER, with the + * fields enclosed by brackets "[]" replaced with your own identifying + * information: Portions Copyright [yyyy] [name of copyright owner] + * + * CDDL HEADER END + */ +/* + * Copyright 2007 Sun Microsystems, Inc. All rights reserved. + * Use is subject to license terms. + */ + +#ifndef _SYS_ZFS_CONTEXT_H +#define _SYS_ZFS_CONTEXT_H + +#pragma ident "%Z%%M% %I% %E% SMI" + +#ifdef __cplusplus +extern "C" { +#endif + +#define _SYS_MUTEX_H +#define _SYS_RWLOCK_H +#define _SYS_CONDVAR_H +#define _SYS_SYSTM_H +#define _SYS_DEBUG_H +#define _SYS_T_LOCK_H +#define _SYS_VNODE_H +#define _SYS_VFS_H +#define _SYS_SUNDDI_H +#define _SYS_CALLB_H +#define _SYS_SCHED_H_ + +#include <solaris.h> +#include <stdio.h> +#include <stdlib.h> +#include <stddef.h> +#include <stdarg.h> +#include <fcntl.h> +#include <unistd.h> +#include <errno.h> +#include <string.h> +#include <strings.h> +#include <thread.h> +#include <assert.h> +#include <limits.h> +#include <dirent.h> +#include <time.h> +#include <math.h> +#include <umem.h> +#include <vmem.h> +#include <fsshare.h> +#include <sys/note.h> +#include <sys/types.h> +#include <sys/atomic.h> +#include <sys/sysmacros.h> +#include <sys/bitmap.h> +#include <sys/resource.h> +#include <sys/byteorder.h> +#include <sys/list.h> +#include <sys/time.h> +#include <sys/uio.h> +#include <sys/mntent.h> +#include <sys/mnttab.h> +#include <sys/zfs_debug.h> +#include <sys/debug.h> +#include <sys/sdt.h> +#include <sys/kstat.h> +#include <sys/kernel.h> +#include <machine/atomic.h> + +#define ZFS_EXPORTS_PATH "/etc/zfs/exports" + +/* + * Debugging + */ + +/* + * Note that we are not using the debugging levels. + */ + +#define CE_CONT 0 /* continuation */ +#define CE_NOTE 1 /* notice */ +#define CE_WARN 2 /* warning */ +#define CE_PANIC 3 /* panic */ +#define CE_IGNORE 4 /* print nothing */ + +/* + * ZFS debugging + */ + +#define ZFS_LOG(...) do { } while (0) + +typedef u_longlong_t rlim64_t; +#define RLIM64_INFINITY ((rlim64_t)-3) + +#ifdef ZFS_DEBUG +extern void dprintf_setup(int *argc, char **argv); +#endif /* ZFS_DEBUG */ + +extern void cmn_err(int, const char *, ...); +extern void vcmn_err(int, const char *, __va_list); +extern void panic(const char *, ...); +extern void vpanic(const char *, __va_list); + +/* This definition is copied from assert.h. */ +#if defined(__STDC__) +#if __STDC_VERSION__ - 0 >= 199901L +#define verify(EX) (void)((EX) || \ + (__assert_c99(#EX, __FILE__, __LINE__, __func__), 0)) +#else +#define verify(EX) (void)((EX) || (__assert(#EX, __FILE__, __LINE__), 0)) +#endif /* __STDC_VERSION__ - 0 >= 199901L */ +#else +#define verify(EX) (void)((EX) || (_assert("EX", __FILE__, __LINE__), 0)) +#endif /* __STDC__ */ + + +#define VERIFY verify +#define ASSERT assert + +extern void __assert(const char *, const char *, int); + +#ifdef lint +#define VERIFY3_IMPL(x, y, z, t) if (x == z) ((void)0) +#else +/* BEGIN CSTYLED */ +#define VERIFY3_IMPL(LEFT, OP, RIGHT, TYPE) do { \ + const TYPE __left = (TYPE)(LEFT); \ + const TYPE __right = (TYPE)(RIGHT); \ + if (!(__left OP __right)) { \ + char *__buf = alloca(256); \ + (void) snprintf(__buf, 256, "%s %s %s (0x%llx %s 0x%llx)", \ + #LEFT, #OP, #RIGHT, \ + (u_longlong_t)__left, #OP, (u_longlong_t)__right); \ + __assert(__buf, __FILE__, __LINE__); \ + } \ +_NOTE(CONSTCOND) } while (0) +/* END CSTYLED */ +#endif /* lint */ + +#define VERIFY3S(x, y, z) VERIFY3_IMPL(x, y, z, int64_t) +#define VERIFY3U(x, y, z) VERIFY3_IMPL(x, y, z, uint64_t) +#define VERIFY3P(x, y, z) VERIFY3_IMPL(x, y, z, uintptr_t) + +#ifdef NDEBUG +#define ASSERT3S(x, y, z) ((void)0) +#define ASSERT3U(x, y, z) ((void)0) +#define ASSERT3P(x, y, z) ((void)0) +#else +#define ASSERT3S(x, y, z) VERIFY3S(x, y, z) +#define ASSERT3U(x, y, z) VERIFY3U(x, y, z) +#define ASSERT3P(x, y, z) VERIFY3P(x, y, z) +#endif + +/* + * Dtrace SDT probes have different signatures in userland than they do in + * kernel. If they're being used in kernel code, re-define them out of + * existence for their counterparts in libzpool. + */ + +#ifdef DTRACE_PROBE1 +#undef DTRACE_PROBE1 +#define DTRACE_PROBE1(a, b, c) ((void)0) +#endif /* DTRACE_PROBE1 */ + +#ifdef DTRACE_PROBE2 +#undef DTRACE_PROBE2 +#define DTRACE_PROBE2(a, b, c, d, e) ((void)0) +#endif /* DTRACE_PROBE2 */ + +#ifdef DTRACE_PROBE3 +#undef DTRACE_PROBE3 +#define DTRACE_PROBE3(a, b, c, d, e, f, g) ((void)0) +#endif /* DTRACE_PROBE3 */ + +#ifdef DTRACE_PROBE4 +#undef DTRACE_PROBE4 +#define DTRACE_PROBE4(a, b, c, d, e, f, g, h, i) ((void)0) +#endif /* DTRACE_PROBE4 */ + +/* + * Threads + */ +#define curthread ((void *)(uintptr_t)thr_self()) + +typedef struct kthread kthread_t; + +#define thread_create(stk, stksize, func, arg, len, pp, state, pri) \ + zk_thread_create(func, arg) +#define thread_exit() thr_exit(NULL) + +extern kthread_t *zk_thread_create(void (*func)(), void *arg); + +#define issig(why) (FALSE) +#define ISSIG(thr, why) (FALSE) + +/* + * Mutexes + */ +typedef struct kmutex { + void *m_owner; + mutex_t m_lock; +} kmutex_t; + +#define MUTEX_DEFAULT USYNC_THREAD +#undef MUTEX_HELD +#define MUTEX_HELD(m) ((m)->m_owner == curthread) + +/* + * Argh -- we have to get cheesy here because the kernel and userland + * have different signatures for the same routine. + */ +//extern int _mutex_init(mutex_t *mp, int type, void *arg); +//extern int _mutex_destroy(mutex_t *mp); + +#define mutex_init(mp, b, c, d) zmutex_init((kmutex_t *)(mp)) +#define mutex_destroy(mp) zmutex_destroy((kmutex_t *)(mp)) + +extern void zmutex_init(kmutex_t *mp); +extern void zmutex_destroy(kmutex_t *mp); +extern void mutex_enter(kmutex_t *mp); +extern void mutex_exit(kmutex_t *mp); +extern int mutex_tryenter(kmutex_t *mp); +extern void *mutex_owner(kmutex_t *mp); + +/* + * RW locks + */ +typedef struct krwlock { + int rw_count; + void *rw_owner; + rwlock_t rw_lock; +} krwlock_t; + +typedef int krw_t; + +#define RW_READER 0 +#define RW_WRITER 1 +#define RW_DEFAULT USYNC_THREAD + +#undef RW_READ_HELD + +#undef RW_WRITE_HELD +#define RW_WRITE_HELD(x) ((x)->rw_owner == curthread) +#define RW_LOCK_HELD(x) rw_lock_held(x) + +extern void rw_init(krwlock_t *rwlp, char *name, int type, void *arg); +extern void rw_destroy(krwlock_t *rwlp); +extern void rw_enter(krwlock_t *rwlp, krw_t rw); +extern int rw_tryenter(krwlock_t *rwlp, krw_t rw); +extern int rw_tryupgrade(krwlock_t *rwlp); +extern void rw_exit(krwlock_t *rwlp); +extern int rw_lock_held(krwlock_t *rwlp); +#define rw_downgrade(rwlp) do { } while (0) + +/* + * Condition variables + */ +typedef cond_t kcondvar_t; + +#define CV_DEFAULT USYNC_THREAD + +extern void cv_init(kcondvar_t *cv, char *name, int type, void *arg); +extern void cv_destroy(kcondvar_t *cv); +extern void cv_wait(kcondvar_t *cv, kmutex_t *mp); +extern clock_t cv_timedwait(kcondvar_t *cv, kmutex_t *mp, clock_t abstime); +extern void cv_signal(kcondvar_t *cv); +extern void cv_broadcast(kcondvar_t *cv); + +/* + * Kernel memory + */ +#define KM_SLEEP UMEM_NOFAIL +#define KM_NOSLEEP UMEM_DEFAULT +#define KMC_NODEBUG UMC_NODEBUG +#define kmem_alloc(_s, _f) umem_alloc(_s, _f) +#define kmem_zalloc(_s, _f) umem_zalloc(_s, _f) +#define kmem_free(_b, _s) umem_free(_b, _s) +#define kmem_cache_create(_a, _b, _c, _d, _e, _f, _g, _h, _i) \ + umem_cache_create(_a, _b, _c, _d, _e, _f, _g, _h, _i) +#define kmem_cache_destroy(_c) umem_cache_destroy(_c) +#define kmem_cache_alloc(_c, _f) umem_cache_alloc(_c, _f) +#define kmem_cache_free(_c, _b) umem_cache_free(_c, _b) +#define kmem_debugging() 0 +#define kmem_cache_reap_now(c) + +typedef umem_cache_t kmem_cache_t; + +/* + * Task queues + */ +typedef struct taskq taskq_t; +typedef uintptr_t taskqid_t; +typedef void (task_func_t)(void *); + +#define TASKQ_PREPOPULATE 0x0001 +#define TASKQ_CPR_SAFE 0x0002 /* Use CPR safe protocol */ +#define TASKQ_DYNAMIC 0x0004 /* Use dynamic thread scheduling */ + +#define TQ_SLEEP KM_SLEEP /* Can block for memory */ +#define TQ_NOSLEEP KM_NOSLEEP /* cannot block for memory; may fail */ +#define TQ_NOQUEUE 0x02 /* Do not enqueue if can't dispatch */ + +extern taskq_t *taskq_create(const char *, int, pri_t, int, int, uint_t); +extern taskqid_t taskq_dispatch(taskq_t *, task_func_t, void *, uint_t); +extern void taskq_destroy(taskq_t *); +extern void taskq_wait(taskq_t *); +extern int taskq_member(taskq_t *, void *); + +/* + * vnodes + */ +typedef struct vnode { + uint64_t v_size; + int v_fd; + char *v_path; +} vnode_t; + +typedef struct vattr { + uint_t va_mask; /* bit-mask of attributes */ + u_offset_t va_size; /* file size in bytes */ +} vattr_t; + +#define AT_TYPE 0x0001 +#define AT_MODE 0x0002 +#define AT_UID 0x0004 +#define AT_GID 0x0008 +#define AT_FSID 0x0010 +#define AT_NODEID 0x0020 +#define AT_NLINK 0x0040 +#define AT_SIZE 0x0080 +#define AT_ATIME 0x0100 +#define AT_MTIME 0x0200 +#define AT_CTIME 0x0400 +#define AT_RDEV 0x0800 +#define AT_BLKSIZE 0x1000 +#define AT_NBLOCKS 0x2000 +#define AT_SEQ 0x8000 + +#define CRCREAT 0 + +#define VOP_CLOSE(vp, f, c, o, cr) 0 +#define VOP_PUTPAGE(vp, of, sz, fl, cr) 0 +#define VOP_GETATTR(vp, vap, fl, cr) ((vap)->va_size = (vp)->v_size, 0) + +#define VOP_FSYNC(vp, f, cr) fsync((vp)->v_fd) + +#define VN_RELE(vp) vn_close(vp) + +extern int vn_open(char *path, int x1, int oflags, int mode, vnode_t **vpp, + int x2, int x3); +extern int vn_openat(char *path, int x1, int oflags, int mode, vnode_t **vpp, + int x2, int x3, vnode_t *vp); +extern int vn_rdwr(int uio, vnode_t *vp, void *addr, ssize_t len, + offset_t offset, int x1, int x2, rlim64_t x3, void *x4, ssize_t *residp); +extern void vn_close(vnode_t *vp); + +#define vn_remove(path, x1, x2) remove(path) +#define vn_rename(from, to, seg) rename((from), (to)) +#define vn_is_readonly(vp) B_FALSE + +extern vnode_t *rootdir; + +#include <sys/file.h> /* for FREAD, FWRITE, etc */ +#define FTRUNC O_TRUNC + +/* + * Random stuff + */ +#define lbolt (gethrtime() >> 23) +#define lbolt64 (gethrtime() >> 23) +//#define hz 119 /* frequency when using gethrtime() >> 23 for lbolt */ + +extern void delay(clock_t ticks); + +#define gethrestime_sec() time(NULL) + +#define max_ncpus 64 + +#define minclsyspri 60 +#define maxclsyspri 99 + +#define CPU_SEQID (thr_self() & (max_ncpus - 1)) + +#define kcred NULL +#define CRED() NULL + +extern uint64_t physmem; + +extern int highbit(ulong_t i); +extern int random_get_bytes(uint8_t *ptr, size_t len); +extern int random_get_pseudo_bytes(uint8_t *ptr, size_t len); + +extern void kernel_init(int); +extern void kernel_fini(void); + +struct spa; +extern void nicenum(uint64_t num, char *buf); +extern void show_pool_stats(struct spa *); + +typedef struct callb_cpr { + kmutex_t *cc_lockp; +} callb_cpr_t; + +#define CALLB_CPR_INIT(cp, lockp, func, name) { \ + (cp)->cc_lockp = lockp; \ +} + +#define CALLB_CPR_SAFE_BEGIN(cp) { \ + ASSERT(MUTEX_HELD((cp)->cc_lockp)); \ +} + +#define CALLB_CPR_SAFE_END(cp, lockp) { \ + ASSERT(MUTEX_HELD((cp)->cc_lockp)); \ +} + +#define CALLB_CPR_EXIT(cp) { \ + ASSERT(MUTEX_HELD((cp)->cc_lockp)); \ + mutex_exit((cp)->cc_lockp); \ +} + +#define zone_dataset_visible(x, y) (1) +#define INGLOBALZONE(z) (1) + +#ifdef __cplusplus +} +#endif + +/* ZFS Boot Related stuff. */ + +struct _buf { + intptr_t _fd; +}; + +struct bootstat { + uint64_t st_size; +}; + +extern struct _buf *kobj_open_file(char *name); +extern int kobj_read_file(struct _buf *file, char *buf, unsigned size, + unsigned off); +extern void kobj_close_file(struct _buf *file); +extern int kobj_get_filesize(struct _buf *file, uint64_t *size); +/* Random compatibility stuff. */ +#define lbolt (gethrtime() >> 23) +#define lbolt64 (gethrtime() >> 23) + +extern int hz; +extern uint64_t physmem; + +#define gethrestime_sec() time(NULL) + +#define open64(...) open(__VA_ARGS__) +#define pread64(d, p, n, o) pread(d, p, n, o) +#define pwrite64(d, p, n, o) pwrite(d, p, n, o) +#define readdir64(d) readdir(d) +#define SIGPENDING(td) (0) +#define root_mount_wait() do { } while (0) + +struct file { + void *dummy; +}; + +#define FCREAT O_CREAT +#define FOFFMAX 0x0 + +#define SX_SYSINIT(name, lock, desc) + +#define SYSCTL_DECL(...) +#define SYSCTL_INT(...) +#define SYSCTL_NODE(...) +#ifdef TUNABLE_INT +#undef TUNABLE_INT +#endif +#define TUNABLE_INT(...) + +/* Errors */ + +#ifndef ERESTART +#define ERESTART (-1) +#endif + +#endif /* _SYS_ZFS_CONTEXT_H */ diff --git a/cddl/contrib/opensolaris/lib/libzpool/common/taskq.c b/cddl/contrib/opensolaris/lib/libzpool/common/taskq.c new file mode 100644 index 0000000..f7b6571 --- /dev/null +++ b/cddl/contrib/opensolaris/lib/libzpool/common/taskq.c @@ -0,0 +1,250 @@ +/* + * CDDL HEADER START + * + * The contents of this file are subject to the terms of the + * Common Development and Distribution License, Version 1.0 only + * (the "License"). You may not use this file except in compliance + * with the License. + * + * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE + * or http://www.opensolaris.org/os/licensing. + * See the License for the specific language governing permissions + * and limitations under the License. + * + * When distributing Covered Code, include this CDDL HEADER in each + * file and include the License file at usr/src/OPENSOLARIS.LICENSE. + * If applicable, add the following below this CDDL HEADER, with the + * fields enclosed by brackets "[]" replaced with your own identifying + * information: Portions Copyright [yyyy] [name of copyright owner] + * + * CDDL HEADER END + */ +/* + * Copyright 2005 Sun Microsystems, Inc. All rights reserved. + * Use is subject to license terms. + */ + +#pragma ident "%Z%%M% %I% %E% SMI" + +#include <sys/zfs_context.h> + +int taskq_now; + +typedef struct task { + struct task *task_next; + struct task *task_prev; + task_func_t *task_func; + void *task_arg; +} task_t; + +#define TASKQ_ACTIVE 0x00010000 + +struct taskq { + kmutex_t tq_lock; + krwlock_t tq_threadlock; + kcondvar_t tq_dispatch_cv; + kcondvar_t tq_wait_cv; + thread_t *tq_threadlist; + int tq_flags; + int tq_active; + int tq_nthreads; + int tq_nalloc; + int tq_minalloc; + int tq_maxalloc; + task_t *tq_freelist; + task_t tq_task; +}; + +static task_t * +task_alloc(taskq_t *tq, int tqflags) +{ + task_t *t; + + if ((t = tq->tq_freelist) != NULL && tq->tq_nalloc >= tq->tq_minalloc) { + tq->tq_freelist = t->task_next; + } else { + mutex_exit(&tq->tq_lock); + if (tq->tq_nalloc >= tq->tq_maxalloc) { + if (!(tqflags & KM_SLEEP)) { + mutex_enter(&tq->tq_lock); + return (NULL); + } + /* + * We don't want to exceed tq_maxalloc, but we can't + * wait for other tasks to complete (and thus free up + * task structures) without risking deadlock with + * the caller. So, we just delay for one second + * to throttle the allocation rate. + */ + delay(hz); + } + t = kmem_alloc(sizeof (task_t), tqflags); + mutex_enter(&tq->tq_lock); + if (t != NULL) + tq->tq_nalloc++; + } + return (t); +} + +static void +task_free(taskq_t *tq, task_t *t) +{ + if (tq->tq_nalloc <= tq->tq_minalloc) { + t->task_next = tq->tq_freelist; + tq->tq_freelist = t; + } else { + tq->tq_nalloc--; + mutex_exit(&tq->tq_lock); + kmem_free(t, sizeof (task_t)); + mutex_enter(&tq->tq_lock); + } +} + +taskqid_t +taskq_dispatch(taskq_t *tq, task_func_t func, void *arg, uint_t tqflags) +{ + task_t *t; + + if (taskq_now) { + func(arg); + return (1); + } + + mutex_enter(&tq->tq_lock); + ASSERT(tq->tq_flags & TASKQ_ACTIVE); + if ((t = task_alloc(tq, tqflags)) == NULL) { + mutex_exit(&tq->tq_lock); + return (0); + } + t->task_next = &tq->tq_task; + t->task_prev = tq->tq_task.task_prev; + t->task_next->task_prev = t; + t->task_prev->task_next = t; + t->task_func = func; + t->task_arg = arg; + cv_signal(&tq->tq_dispatch_cv); + mutex_exit(&tq->tq_lock); + return (1); +} + +void +taskq_wait(taskq_t *tq) +{ + mutex_enter(&tq->tq_lock); + while (tq->tq_task.task_next != &tq->tq_task || tq->tq_active != 0) + cv_wait(&tq->tq_wait_cv, &tq->tq_lock); + mutex_exit(&tq->tq_lock); +} + +static void * +taskq_thread(void *arg) +{ + taskq_t *tq = arg; + task_t *t; + + mutex_enter(&tq->tq_lock); + while (tq->tq_flags & TASKQ_ACTIVE) { + if ((t = tq->tq_task.task_next) == &tq->tq_task) { + if (--tq->tq_active == 0) + cv_broadcast(&tq->tq_wait_cv); + cv_wait(&tq->tq_dispatch_cv, &tq->tq_lock); + tq->tq_active++; + continue; + } + t->task_prev->task_next = t->task_next; + t->task_next->task_prev = t->task_prev; + mutex_exit(&tq->tq_lock); + + rw_enter(&tq->tq_threadlock, RW_READER); + t->task_func(t->task_arg); + rw_exit(&tq->tq_threadlock); + + mutex_enter(&tq->tq_lock); + task_free(tq, t); + } + tq->tq_nthreads--; + cv_broadcast(&tq->tq_wait_cv); + mutex_exit(&tq->tq_lock); + return (NULL); +} + +/*ARGSUSED*/ +taskq_t * +taskq_create(const char *name, int nthreads, pri_t pri, + int minalloc, int maxalloc, uint_t flags) +{ + taskq_t *tq = kmem_zalloc(sizeof (taskq_t), KM_SLEEP); + int t; + + rw_init(&tq->tq_threadlock, NULL, RW_DEFAULT, NULL); + tq->tq_flags = flags | TASKQ_ACTIVE; + tq->tq_active = nthreads; + tq->tq_nthreads = nthreads; + tq->tq_minalloc = minalloc; + tq->tq_maxalloc = maxalloc; + tq->tq_task.task_next = &tq->tq_task; + tq->tq_task.task_prev = &tq->tq_task; + tq->tq_threadlist = kmem_alloc(nthreads * sizeof (thread_t), KM_SLEEP); + + if (flags & TASKQ_PREPOPULATE) { + mutex_enter(&tq->tq_lock); + while (minalloc-- > 0) + task_free(tq, task_alloc(tq, KM_SLEEP)); + mutex_exit(&tq->tq_lock); + } + + for (t = 0; t < nthreads; t++) + (void) thr_create(0, 0, taskq_thread, + tq, THR_BOUND, &tq->tq_threadlist[t]); + + return (tq); +} + +void +taskq_destroy(taskq_t *tq) +{ + int t; + int nthreads = tq->tq_nthreads; + + taskq_wait(tq); + + mutex_enter(&tq->tq_lock); + + tq->tq_flags &= ~TASKQ_ACTIVE; + cv_broadcast(&tq->tq_dispatch_cv); + + while (tq->tq_nthreads != 0) + cv_wait(&tq->tq_wait_cv, &tq->tq_lock); + + tq->tq_minalloc = 0; + while (tq->tq_nalloc != 0) { + ASSERT(tq->tq_freelist != NULL); + task_free(tq, task_alloc(tq, KM_SLEEP)); + } + + mutex_exit(&tq->tq_lock); + + for (t = 0; t < nthreads; t++) + (void) thr_join(tq->tq_threadlist[t], NULL, NULL); + + kmem_free(tq->tq_threadlist, nthreads * sizeof (thread_t)); + + rw_destroy(&tq->tq_threadlock); + + kmem_free(tq, sizeof (taskq_t)); +} + +int +taskq_member(taskq_t *tq, void *t) +{ + int i; + + if (taskq_now) + return (1); + + for (i = 0; i < tq->tq_nthreads; i++) + if (tq->tq_threadlist[i] == (thread_t)(uintptr_t)t) + return (1); + + return (0); +} diff --git a/cddl/contrib/opensolaris/lib/libzpool/common/util.c b/cddl/contrib/opensolaris/lib/libzpool/common/util.c new file mode 100644 index 0000000..df49adb --- /dev/null +++ b/cddl/contrib/opensolaris/lib/libzpool/common/util.c @@ -0,0 +1,144 @@ +/* + * CDDL HEADER START + * + * The contents of this file are subject to the terms of the + * Common Development and Distribution License (the "License"). + * You may not use this file except in compliance with the License. + * + * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE + * or http://www.opensolaris.org/os/licensing. + * See the License for the specific language governing permissions + * and limitations under the License. + * + * When distributing Covered Code, include this CDDL HEADER in each + * file and include the License file at usr/src/OPENSOLARIS.LICENSE. + * If applicable, add the following below this CDDL HEADER, with the + * fields enclosed by brackets "[]" replaced with your own identifying + * information: Portions Copyright [yyyy] [name of copyright owner] + * + * CDDL HEADER END + */ +/* + * Copyright 2006 Sun Microsystems, Inc. All rights reserved. + * Use is subject to license terms. + */ + +#pragma ident "%Z%%M% %I% %E% SMI" + +#include <assert.h> +#include <sys/zfs_context.h> +#include <sys/avl.h> +#include <string.h> +#include <stdio.h> +#include <stdlib.h> +#include <sys/spa.h> +#include <sys/fs/zfs.h> +#include <sys/refcount.h> + +/* + * Routines needed by more than one client of libzpool. + */ + +void +nicenum(uint64_t num, char *buf) +{ + uint64_t n = num; + int index = 0; + char u; + + while (n >= 1024) { + n = (n + (1024 / 2)) / 1024; /* Round up or down */ + index++; + } + + u = " KMGTPE"[index]; + + if (index == 0) { + (void) sprintf(buf, "%llu", (u_longlong_t)n); + } else if (n < 10 && (num & (num - 1)) != 0) { + (void) sprintf(buf, "%.2f%c", + (double)num / (1ULL << 10 * index), u); + } else if (n < 100 && (num & (num - 1)) != 0) { + (void) sprintf(buf, "%.1f%c", + (double)num / (1ULL << 10 * index), u); + } else { + (void) sprintf(buf, "%llu%c", (u_longlong_t)n, u); + } +} + +static void +show_vdev_stats(const char *desc, nvlist_t *nv, int indent) +{ + nvlist_t **child; + uint_t c, children; + vdev_stat_t *vs; + uint64_t sec; + char used[6], avail[6]; + char rops[6], wops[6], rbytes[6], wbytes[6], rerr[6], werr[6], cerr[6]; + + if (indent == 0) { + (void) printf(" " + " capacity operations bandwidth ---- errors ----\n"); + (void) printf("description " + "used avail read write read write read write cksum\n"); + } + + VERIFY(nvlist_lookup_uint64_array(nv, ZPOOL_CONFIG_STATS, + (uint64_t **)&vs, &c) == 0); + + sec = MAX(1, vs->vs_timestamp / NANOSEC); + + nicenum(vs->vs_alloc, used); + nicenum(vs->vs_space - vs->vs_alloc, avail); + nicenum(vs->vs_ops[ZIO_TYPE_READ] / sec, rops); + nicenum(vs->vs_ops[ZIO_TYPE_WRITE] / sec, wops); + nicenum(vs->vs_bytes[ZIO_TYPE_READ] / sec, rbytes); + nicenum(vs->vs_bytes[ZIO_TYPE_WRITE] / sec, wbytes); + nicenum(vs->vs_read_errors, rerr); + nicenum(vs->vs_write_errors, werr); + nicenum(vs->vs_checksum_errors, cerr); + + (void) printf("%*s%*s%*s%*s %5s %5s %5s %5s %5s %5s %5s\n", + indent, "", + indent - 19 - (vs->vs_space ? 0 : 12), desc, + vs->vs_space ? 6 : 0, vs->vs_space ? used : "", + vs->vs_space ? 6 : 0, vs->vs_space ? avail : "", + rops, wops, rbytes, wbytes, rerr, werr, cerr); + + if (nvlist_lookup_nvlist_array(nv, ZPOOL_CONFIG_CHILDREN, + &child, &children) != 0) + return; + + for (c = 0; c < children; c++) { + nvlist_t *cnv = child[c]; + char *cname, *tname; + uint64_t np; + if (nvlist_lookup_string(cnv, ZPOOL_CONFIG_PATH, &cname) && + nvlist_lookup_string(cnv, ZPOOL_CONFIG_TYPE, &cname)) + cname = "<unknown>"; + tname = calloc(1, strlen(cname) + 2); + (void) strcpy(tname, cname); + if (nvlist_lookup_uint64(cnv, ZPOOL_CONFIG_NPARITY, &np) == 0) + tname[strlen(tname)] = '0' + np; + show_vdev_stats(tname, cnv, indent + 2); + free(tname); + } +} + +void +show_pool_stats(spa_t *spa) +{ + nvlist_t *config, *nvroot; + char *name; + + spa_config_enter(spa, RW_READER, FTAG); + config = spa_config_generate(spa, NULL, -1ULL, B_TRUE); + spa_config_exit(spa, FTAG); + + VERIFY(nvlist_lookup_nvlist(config, ZPOOL_CONFIG_VDEV_TREE, + &nvroot) == 0); + VERIFY(nvlist_lookup_string(config, ZPOOL_CONFIG_POOL_NAME, + &name) == 0); + + show_vdev_stats(name, nvroot, 0); +} |