Index: vendor/illumos/dist/lib/libzpool/common/kernel.c =================================================================== --- vendor/illumos/dist/lib/libzpool/common/kernel.c (revision 297504) +++ vendor/illumos/dist/lib/libzpool/common/kernel.c (revision 297505) @@ -1,1154 +1,1157 @@ /* * 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 (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. * Copyright (c) 2012, 2015 by Delphix. All rights reserved. * Copyright (c) 2013, Joyent, Inc. All rights reserved. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include /* * Emulation of kernel services in userland. */ int aok; uint64_t physmem; vnode_t *rootdir = (vnode_t *)0xabcd1234; char hw_serial[HW_HOSTID_LEN]; kmutex_t cpu_lock; vmem_t *zio_arena = NULL; /* If set, all blocks read will be copied to the specified directory. */ char *vn_dumpdir = NULL; struct utsname utsname = { "userland", "libzpool", "1", "1", "na" }; /* this only exists to have its address taken */ struct proc p0; /* * ========================================================================= * 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(const char *module, int instance, const char *name, const 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) {} /*ARGSUSED*/ void kstat_waitq_enter(kstat_io_t *kiop) {} /*ARGSUSED*/ void kstat_waitq_exit(kstat_io_t *kiop) {} /*ARGSUSED*/ void kstat_runq_enter(kstat_io_t *kiop) {} /*ARGSUSED*/ void kstat_runq_exit(kstat_io_t *kiop) {} /*ARGSUSED*/ void kstat_waitq_to_runq(kstat_io_t *kiop) {} /*ARGSUSED*/ void kstat_runq_back_to_waitq(kstat_io_t *kiop) {} /* * ========================================================================= * mutexes * ========================================================================= */ void zmutex_init(kmutex_t *mp) { mp->m_owner = NULL; mp->initialized = B_TRUE; (void) _mutex_init(&mp->m_lock, USYNC_THREAD, NULL); } void zmutex_destroy(kmutex_t *mp) { ASSERT(mp->initialized == B_TRUE); ASSERT(mp->m_owner == NULL); (void) _mutex_destroy(&(mp)->m_lock); mp->m_owner = (void *)-1UL; mp->initialized = B_FALSE; } void mutex_enter(kmutex_t *mp) { ASSERT(mp->initialized == B_TRUE); 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->initialized == B_TRUE); ASSERT(mp->m_owner != (void *)-1UL); if (0 == mutex_trylock(&mp->m_lock)) { ASSERT(mp->m_owner == NULL); mp->m_owner = curthread; return (1); } else { return (0); } } void mutex_exit(kmutex_t *mp) { ASSERT(mp->initialized == B_TRUE); ASSERT(mutex_owner(mp) == curthread); mp->m_owner = NULL; VERIFY(mutex_unlock(&mp->m_lock) == 0); } void * mutex_owner(kmutex_t *mp) { ASSERT(mp->initialized == B_TRUE); 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->initialized = B_TRUE; } void rw_destroy(krwlock_t *rwlp) { rwlock_destroy(&rwlp->rw_lock); rwlp->rw_owner = (void *)-1UL; rwlp->initialized = B_FALSE; } void rw_enter(krwlock_t *rwlp, krw_t rw) { ASSERT(!RW_LOCK_HELD(rwlp)); ASSERT(rwlp->initialized == B_TRUE); ASSERT(rwlp->rw_owner != (void *)-1UL); ASSERT(rwlp->rw_owner != curthread); if (rw == RW_WRITER) VERIFY(rw_wrlock(&rwlp->rw_lock) == 0); else VERIFY(rw_rdlock(&rwlp->rw_lock) == 0); rwlp->rw_owner = curthread; } void rw_exit(krwlock_t *rwlp) { ASSERT(rwlp->initialized == B_TRUE); ASSERT(rwlp->rw_owner != (void *)-1UL); rwlp->rw_owner = NULL; VERIFY(rw_unlock(&rwlp->rw_lock) == 0); } int rw_tryenter(krwlock_t *rwlp, krw_t rw) { int rv; ASSERT(rwlp->initialized == B_TRUE); ASSERT(rwlp->rw_owner != (void *)-1UL); if (rw == RW_WRITER) rv = rw_trywrlock(&rwlp->rw_lock); else rv = rw_tryrdlock(&rwlp->rw_lock); if (rv == 0) { rwlp->rw_owner = curthread; return (1); } return (0); } /*ARGSUSED*/ int rw_tryupgrade(krwlock_t *rwlp) { ASSERT(rwlp->initialized == B_TRUE); ASSERT(rwlp->rw_owner != (void *)-1UL); return (0); } /* * ========================================================================= * condition variables * ========================================================================= */ /*ARGSUSED*/ void cv_init(kcondvar_t *cv, char *name, int type, void *arg) { VERIFY(cond_init(cv, type, 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; timestruc_t ts; clock_t delta; top: delta = abstime - ddi_get_lbolt(); if (delta <= 0) return (-1); ts.tv_sec = delta / hz; ts.tv_nsec = (delta % hz) * (NANOSEC / hz); ASSERT(mutex_owner(mp) == curthread); mp->m_owner = NULL; error = cond_reltimedwait(cv, &mp->m_lock, &ts); mp->m_owner = curthread; if (error == ETIME) return (-1); if (error == EINTR) goto top; ASSERT(error == 0); return (1); } /*ARGSUSED*/ clock_t cv_timedwait_hires(kcondvar_t *cv, kmutex_t *mp, hrtime_t tim, hrtime_t res, int flag) { int error; timestruc_t ts; hrtime_t delta; - ASSERT(flag == 0); + ASSERT(flag == 0 || flag == CALLOUT_FLAG_ABSOLUTE); top: - delta = tim - gethrtime(); + delta = tim; + if (flag & CALLOUT_FLAG_ABSOLUTE) + delta -= gethrtime(); + if (delta <= 0) return (-1); ts.tv_sec = delta / NANOSEC; ts.tv_nsec = delta % NANOSEC; ASSERT(mutex_owner(mp) == curthread); mp->m_owner = NULL; error = cond_reltimedwait(cv, &mp->m_lock, &ts); mp->m_owner = curthread; if (error == ETIME) return (-1); if (error == EINTR) goto top; 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; int dump_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 (vn_dumpdir != NULL) { char dumppath[MAXPATHLEN]; (void) snprintf(dumppath, sizeof (dumppath), "%s/%s", vn_dumpdir, basename(realpath)); dump_fd = open64(dumppath, O_CREAT | O_WRONLY, 0666); if (dump_fd == -1) return (errno); } else { dump_fd = -1; } 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); vp->v_dump_fd = dump_fd; return (0); } /*ARGSUSED*/ int vn_openat(char *path, int x1, int flags, int mode, vnode_t **vpp, int x2, int x3, vnode_t *startvp, int fd) { char *realpath = umem_alloc(strlen(path) + 2, UMEM_NOFAIL); int ret; ASSERT(startvp == rootdir); (void) sprintf(realpath, "/%s", path); /* fd ignored for now, need if want to simulate nbmand support */ 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); if (vp->v_dump_fd != -1) { int status = pwrite64(vp->v_dump_fd, addr, iolen, offset); ASSERT(status != -1); } } else { /* * To simulate partial disk writes, we split writes into two * system calls so that the process can be killed in between. */ int sectors = len >> SPA_MINBLOCKSHIFT; split = (sectors > 0 ? rand() % sectors : 0) << SPA_MINBLOCKSHIFT; 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); if (vp->v_dump_fd != -1) close(vp->v_dump_fd); spa_strfree(vp->v_path); umem_free(vp, sizeof (vnode_t)); } /* * At a minimum we need to update the size since vdev_reopen() * will no longer call vn_openat(). */ int fop_getattr(vnode_t *vp, vattr_t *vap) { struct stat64 st; if (fstat64(vp->v_fd, &st) == -1) { close(vp->v_fd); return (errno); } vap->va_size = st.st_size; return (0); } #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; if (dprintf_string != NULL) zfs_flags |= ZFS_DEBUG_DPRINTF; } /* * ========================================================================= * 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 (dprintf_find_string("cpu")) (void) printf("%u ", getcpuid()); 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, -1) != 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)); } /* * Find highest one bit set. * Returns bit number + 1 of highest bit that is set, otherwise returns 0. */ int highbit64(uint64_t i) { int h = 1; if (i == 0) return (0); if (i & 0xffffffff00000000ULL) { h += 32; i >>= 32; } 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); } static int random_fd = -1, urandom_fd = -1; static int random_get_bytes_common(uint8_t *ptr, size_t len, int fd) { size_t resid = len; ssize_t bytes; ASSERT(fd != -1); while (resid != 0) { bytes = read(fd, ptr, resid); ASSERT3S(bytes, >=, 0); ptr += bytes; resid -= bytes; } return (0); } int random_get_bytes(uint8_t *ptr, size_t len) { return (random_get_bytes_common(ptr, len, random_fd)); } int random_get_pseudo_bytes(uint8_t *ptr, size_t len) { return (random_get_bytes_common(ptr, len, urandom_fd)); } int ddi_strtoul(const char *hw_serial, char **nptr, int base, unsigned long *result) { char *end; *result = strtoul(hw_serial, &end, base); if (*result == 0) return (errno); return (0); } int ddi_strtoull(const char *str, char **nptr, int base, u_longlong_t *result) { char *end; *result = strtoull(str, &end, base); if (*result == 0) return (errno); return (0); } /* ARGSUSED */ cyclic_id_t cyclic_add(cyc_handler_t *hdlr, cyc_time_t *when) { return (1); } /* ARGSUSED */ void cyclic_remove(cyclic_id_t id) { } /* ARGSUSED */ int cyclic_reprogram(cyclic_id_t id, hrtime_t expiration) { return (1); } /* * ========================================================================= * 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) { extern uint_t rrw_tsd_key; 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)); (void) snprintf(hw_serial, sizeof (hw_serial), "%ld", (mode & FWRITE) ? gethostid() : 0); VERIFY((random_fd = open("/dev/random", O_RDONLY)) != -1); VERIFY((urandom_fd = open("/dev/urandom", O_RDONLY)) != -1); system_taskq_init(); mutex_init(&cpu_lock, NULL, MUTEX_DEFAULT, NULL); spa_init(mode); tsd_create(&rrw_tsd_key, rrw_tsd_destroy); } void kernel_fini(void) { spa_fini(); system_taskq_fini(); close(random_fd); close(urandom_fd); random_fd = -1; urandom_fd = -1; } 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); } uid_t crgetuid(cred_t *cr) { return (0); } uid_t crgetruid(cred_t *cr) { return (0); } gid_t crgetgid(cred_t *cr) { return (0); } int crgetngroups(cred_t *cr) { return (0); } gid_t * crgetgroups(cred_t *cr) { return (NULL); } int zfs_secpolicy_snapshot_perms(const char *name, cred_t *cr) { return (0); } int zfs_secpolicy_rename_perms(const char *from, const char *to, cred_t *cr) { return (0); } int zfs_secpolicy_destroy_perms(const char *name, cred_t *cr) { return (0); } ksiddomain_t * ksid_lookupdomain(const char *dom) { ksiddomain_t *kd; kd = umem_zalloc(sizeof (ksiddomain_t), UMEM_NOFAIL); kd->kd_name = spa_strdup(dom); return (kd); } void ksiddomain_rele(ksiddomain_t *ksid) { spa_strfree(ksid->kd_name); umem_free(ksid, sizeof (ksiddomain_t)); } /* * Do not change the length of the returned string; it must be freed * with strfree(). */ char * kmem_asprintf(const char *fmt, ...) { int size; va_list adx; char *buf; va_start(adx, fmt); size = vsnprintf(NULL, 0, fmt, adx) + 1; va_end(adx); buf = kmem_alloc(size, KM_SLEEP); va_start(adx, fmt); size = vsnprintf(buf, size, fmt, adx); va_end(adx); return (buf); } /* ARGSUSED */ int zfs_onexit_fd_hold(int fd, minor_t *minorp) { *minorp = 0; return (0); } /* ARGSUSED */ void zfs_onexit_fd_rele(int fd) { } /* ARGSUSED */ int zfs_onexit_add_cb(minor_t minor, void (*func)(void *), void *data, uint64_t *action_handle) { return (0); } /* ARGSUSED */ int zfs_onexit_del_cb(minor_t minor, uint64_t action_handle, boolean_t fire) { return (0); } /* ARGSUSED */ int zfs_onexit_cb_data(minor_t minor, uint64_t action_handle, void **data) { return (0); } void bioinit(buf_t *bp) { bzero(bp, sizeof (buf_t)); } void biodone(buf_t *bp) { if (bp->b_iodone != NULL) { (*(bp->b_iodone))(bp); return; } ASSERT((bp->b_flags & B_DONE) == 0); bp->b_flags |= B_DONE; } void bioerror(buf_t *bp, int error) { ASSERT(bp != NULL); ASSERT(error >= 0); if (error != 0) { bp->b_flags |= B_ERROR; } else { bp->b_flags &= ~B_ERROR; } bp->b_error = error; } int geterror(struct buf *bp) { int error = 0; if (bp->b_flags & B_ERROR) { error = bp->b_error; if (!error) error = EIO; } return (error); } Index: vendor/illumos/dist/lib/libzpool/common/sys/zfs_context.h =================================================================== --- vendor/illumos/dist/lib/libzpool/common/sys/zfs_context.h (revision 297504) +++ vendor/illumos/dist/lib/libzpool/common/sys/zfs_context.h (revision 297505) @@ -1,692 +1,693 @@ /* * 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 (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. * Copyright 2011 Nexenta Systems, Inc. All rights reserved. - * Copyright (c) 2012, 2014 by Delphix. All rights reserved. + * Copyright (c) 2012, 2015 by Delphix. All rights reserved. * Copyright (c) 2012, Joyent, Inc. All rights reserved. */ #ifndef _SYS_ZFS_CONTEXT_H #define _SYS_ZFS_CONTEXT_H #ifdef __cplusplus extern "C" { #endif #define _SYS_MUTEX_H #define _SYS_RWLOCK_H #define _SYS_CONDVAR_H #define _SYS_SYSTM_H #define _SYS_T_LOCK_H #define _SYS_VNODE_H #define _SYS_VFS_H #define _SYS_SUNDDI_H #define _SYS_CALLB_H #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include "zfs.h" /* * 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 */ #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); #define fm_panic panic extern int aok; /* * DTrace SDT probes have different signatures in userland than they do in * the kernel. If they're being used in kernel code, re-define them out of * existence for their counterparts in libzpool. * * Here's an example of how to use the set-error probes in userland: * zfs$target:::set-error /arg0 == EBUSY/ {stack();} * * Here's an example of how to use DTRACE_PROBE probes in userland: * If there is a probe declared as follows: * DTRACE_PROBE2(zfs__probe_name, uint64_t, blkid, dnode_t *, dn); * Then you can use it as follows: * zfs$target:::probe2 /copyinstr(arg0) == "zfs__probe_name"/ * {printf("%u %p\n", arg1, arg2);} */ #ifdef DTRACE_PROBE #undef DTRACE_PROBE #endif /* DTRACE_PROBE */ #define DTRACE_PROBE(a) \ ZFS_PROBE0(#a) #ifdef DTRACE_PROBE1 #undef DTRACE_PROBE1 #endif /* DTRACE_PROBE1 */ #define DTRACE_PROBE1(a, b, c) \ ZFS_PROBE1(#a, (unsigned long)c) #ifdef DTRACE_PROBE2 #undef DTRACE_PROBE2 #endif /* DTRACE_PROBE2 */ #define DTRACE_PROBE2(a, b, c, d, e) \ ZFS_PROBE2(#a, (unsigned long)c, (unsigned long)e) #ifdef DTRACE_PROBE3 #undef DTRACE_PROBE3 #endif /* DTRACE_PROBE3 */ #define DTRACE_PROBE3(a, b, c, d, e, f, g) \ ZFS_PROBE3(#a, (unsigned long)c, (unsigned long)e, (unsigned long)g) #ifdef DTRACE_PROBE4 #undef DTRACE_PROBE4 #endif /* DTRACE_PROBE4 */ #define DTRACE_PROBE4(a, b, c, d, e, f, g, h, i) \ ZFS_PROBE4(#a, (unsigned long)c, (unsigned long)e, (unsigned long)g, \ (unsigned long)i) /* * We use the comma operator so that this macro can be used without much * additional code. For example, "return (EINVAL);" becomes * "return (SET_ERROR(EINVAL));". Note that the argument will be evaluated * twice, so it should not have side effects (e.g. something like: * "return (SET_ERROR(log_error(EINVAL, info)));" would log the error twice). */ #define SET_ERROR(err) (ZFS_SET_ERROR(err), err) /* * Threads */ #define curthread ((void *)(uintptr_t)thr_self()) #define kpreempt(x) yield() 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) #define thread_join(t) panic("libzpool cannot join threads") #define newproc(f, a, cid, pri, ctp, pid) (ENOSYS) /* in libzpool, p0 exists only to have its address taken */ struct proc { uintptr_t this_is_never_used_dont_dereference_it; }; extern struct proc p0; #define curproc (&p0) #define PS_NONE -1 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; boolean_t initialized; mutex_t m_lock; } kmutex_t; #define MUTEX_DEFAULT USYNC_THREAD #undef MUTEX_HELD #undef MUTEX_NOT_HELD #define MUTEX_HELD(m) _mutex_held(&(m)->m_lock) #define MUTEX_NOT_HELD(m) (!MUTEX_HELD(m)) /* * 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 { void *rw_owner; boolean_t initialized; 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 #define RW_READ_HELD(x) _rw_read_held(&(x)->rw_lock) #undef RW_WRITE_HELD #define RW_WRITE_HELD(x) _rw_write_held(&(x)->rw_lock) #undef RW_LOCK_HELD #define RW_LOCK_HELD(x) (RW_READ_HELD(x) || RW_WRITE_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); #define rw_downgrade(rwlp) do { } while (0) extern uid_t crgetuid(cred_t *cr); extern uid_t crgetruid(cred_t *cr); extern gid_t crgetgid(cred_t *cr); extern int crgetngroups(cred_t *cr); extern gid_t *crgetgroups(cred_t *cr); /* * Condition variables */ typedef cond_t kcondvar_t; #define CV_DEFAULT USYNC_THREAD +#define CALLOUT_FLAG_ABSOLUTE 0x2 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 clock_t cv_timedwait_hires(kcondvar_t *cvp, kmutex_t *mp, hrtime_t tim, hrtime_t res, int flag); extern void cv_signal(kcondvar_t *cv); extern void cv_broadcast(kcondvar_t *cv); /* * Thread-specific data */ #define tsd_get(k) pthread_getspecific(k) #define tsd_set(k, v) pthread_setspecific(k, v) #define tsd_create(kp, d) pthread_key_create(kp, d) #define tsd_destroy(kp) /* nothing */ /* * kstat creation, installation and deletion */ extern kstat_t *kstat_create(const char *, int, const char *, const char *, uchar_t, ulong_t, uchar_t); extern void kstat_install(kstat_t *); extern void kstat_delete(kstat_t *); extern void kstat_waitq_enter(kstat_io_t *); extern void kstat_waitq_exit(kstat_io_t *); extern void kstat_runq_enter(kstat_io_t *); extern void kstat_runq_exit(kstat_io_t *); extern void kstat_waitq_to_runq(kstat_io_t *); extern void kstat_runq_back_to_waitq(kstat_io_t *); /* * Kernel memory */ #define KM_SLEEP UMEM_NOFAIL #define KM_PUSHPAGE KM_SLEEP #define KM_NOSLEEP UMEM_DEFAULT #define KMC_NODEBUG UMC_NODEBUG #define KMC_NOTOUCH 0 /* not needed for userland caches */ #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) /* nothing */ #define kmem_cache_set_move(_c, _cb) /* nothing */ #define vmem_qcache_reap(_v) /* nothing */ #define POINTER_INVALIDATE(_pp) /* nothing */ #define POINTER_IS_VALID(_p) 0 extern vmem_t *zio_arena; typedef umem_cache_t kmem_cache_t; typedef enum kmem_cbrc { KMEM_CBRC_YES, KMEM_CBRC_NO, KMEM_CBRC_LATER, KMEM_CBRC_DONT_NEED, KMEM_CBRC_DONT_KNOW } kmem_cbrc_t; /* * Task queues */ typedef struct taskq taskq_t; typedef uintptr_t taskqid_t; typedef void (task_func_t)(void *); typedef struct taskq_ent { struct taskq_ent *tqent_next; struct taskq_ent *tqent_prev; task_func_t *tqent_func; void *tqent_arg; uintptr_t tqent_flags; } taskq_ent_t; #define TQENT_FLAG_PREALLOC 0x1 /* taskq_dispatch_ent used */ #define TASKQ_PREPOPULATE 0x0001 #define TASKQ_CPR_SAFE 0x0002 /* Use CPR safe protocol */ #define TASKQ_DYNAMIC 0x0004 /* Use dynamic thread scheduling */ #define TASKQ_THREADS_CPU_PCT 0x0008 /* Scale # threads by # cpus */ #define TASKQ_DC_BATCH 0x0010 /* Mark threads as batch */ #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 */ #define TQ_FRONT 0x08 /* Queue in front */ extern taskq_t *system_taskq; extern taskq_t *taskq_create(const char *, int, pri_t, int, int, uint_t); #define taskq_create_proc(a, b, c, d, e, p, f) \ (taskq_create(a, b, c, d, e, f)) #define taskq_create_sysdc(a, b, d, e, p, dc, f) \ (taskq_create(a, b, maxclsyspri, d, e, f)) extern taskqid_t taskq_dispatch(taskq_t *, task_func_t, void *, uint_t); extern void taskq_dispatch_ent(taskq_t *, task_func_t, void *, uint_t, taskq_ent_t *); extern void taskq_destroy(taskq_t *); extern void taskq_wait(taskq_t *); extern int taskq_member(taskq_t *, void *); extern void system_taskq_init(void); extern void system_taskq_fini(void); #define XVA_MAPSIZE 3 #define XVA_MAGIC 0x78766174 /* * vnodes */ typedef struct vnode { uint64_t v_size; int v_fd; char *v_path; int v_dump_fd; } vnode_t; extern char *vn_dumpdir; #define AV_SCANSTAMP_SZ 32 /* length of anti-virus scanstamp */ typedef struct xoptattr { timestruc_t xoa_createtime; /* Create time of file */ uint8_t xoa_archive; uint8_t xoa_system; uint8_t xoa_readonly; uint8_t xoa_hidden; uint8_t xoa_nounlink; uint8_t xoa_immutable; uint8_t xoa_appendonly; uint8_t xoa_nodump; uint8_t xoa_settable; uint8_t xoa_opaque; uint8_t xoa_av_quarantined; uint8_t xoa_av_modified; uint8_t xoa_av_scanstamp[AV_SCANSTAMP_SZ]; uint8_t xoa_reparse; uint8_t xoa_offline; uint8_t xoa_sparse; } xoptattr_t; typedef struct vattr { uint_t va_mask; /* bit-mask of attributes */ u_offset_t va_size; /* file size in bytes */ } vattr_t; typedef struct xvattr { vattr_t xva_vattr; /* Embedded vattr structure */ uint32_t xva_magic; /* Magic Number */ uint32_t xva_mapsize; /* Size of attr bitmap (32-bit words) */ uint32_t *xva_rtnattrmapp; /* Ptr to xva_rtnattrmap[] */ uint32_t xva_reqattrmap[XVA_MAPSIZE]; /* Requested attrs */ uint32_t xva_rtnattrmap[XVA_MAPSIZE]; /* Returned attrs */ xoptattr_t xva_xoptattrs; /* Optional attributes */ } xvattr_t; typedef struct vsecattr { uint_t vsa_mask; /* See below */ int vsa_aclcnt; /* ACL entry count */ void *vsa_aclentp; /* pointer to ACL entries */ int vsa_dfaclcnt; /* default ACL entry count */ void *vsa_dfaclentp; /* pointer to default ACL entries */ size_t vsa_aclentsz; /* ACE size in bytes of vsa_aclentp */ } vsecattr_t; #define AT_TYPE 0x00001 #define AT_MODE 0x00002 #define AT_UID 0x00004 #define AT_GID 0x00008 #define AT_FSID 0x00010 #define AT_NODEID 0x00020 #define AT_NLINK 0x00040 #define AT_SIZE 0x00080 #define AT_ATIME 0x00100 #define AT_MTIME 0x00200 #define AT_CTIME 0x00400 #define AT_RDEV 0x00800 #define AT_BLKSIZE 0x01000 #define AT_NBLOCKS 0x02000 #define AT_SEQ 0x08000 #define AT_XVATTR 0x10000 #define CRCREAT 0 extern int fop_getattr(vnode_t *vp, vattr_t *vap); #define VOP_CLOSE(vp, f, c, o, cr, ct) 0 #define VOP_PUTPAGE(vp, of, sz, fl, cr, ct) 0 #define VOP_GETATTR(vp, vap, fl, cr, ct) fop_getattr((vp), (vap)); #define VOP_FSYNC(vp, f, cr, ct) 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, int fd); 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 /* for FREAD, FWRITE, etc */ /* * Random stuff */ #define ddi_get_lbolt() (gethrtime() >> 23) #define ddi_get_lbolt64() (gethrtime() >> 23) #define hz 119 /* frequency when using gethrtime() >> 23 for lbolt */ extern void delay(clock_t ticks); #define SEC_TO_TICK(sec) ((sec) * hz) #define NSEC_TO_TICK(usec) ((usec) / (NANOSEC / hz)) #define gethrestime_sec() time(NULL) #define gethrestime(t) \ do {\ (t)->tv_sec = gethrestime_sec();\ (t)->tv_nsec = 0;\ } while (0); #define max_ncpus 64 #define boot_ncpus (sysconf(_SC_NPROCESSORS_ONLN)) #define minclsyspri 60 #define maxclsyspri 99 #define CPU_SEQID (thr_self() & (max_ncpus - 1)) #define kcred NULL #define CRED() NULL #define ptob(x) ((x) * PAGESIZE) extern uint64_t physmem; extern int highbit64(uint64_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) extern char *kmem_asprintf(const char *fmt, ...); #define strfree(str) kmem_free((str), strlen(str) + 1) /* * Hostname information */ extern char hw_serial[]; /* for userland-emulated hostid access */ extern int ddi_strtoul(const char *str, char **nptr, int base, unsigned long *result); extern int ddi_strtoull(const char *str, char **nptr, int base, u_longlong_t *result); /* ZFS Boot Related stuff. */ struct _buf { intptr_t _fd; }; struct bootstat { uint64_t st_size; }; typedef struct ace_object { uid_t a_who; uint32_t a_access_mask; uint16_t a_flags; uint16_t a_type; uint8_t a_obj_type[16]; uint8_t a_inherit_obj_type[16]; } ace_object_t; #define ACE_ACCESS_ALLOWED_OBJECT_ACE_TYPE 0x05 #define ACE_ACCESS_DENIED_OBJECT_ACE_TYPE 0x06 #define ACE_SYSTEM_AUDIT_OBJECT_ACE_TYPE 0x07 #define ACE_SYSTEM_ALARM_OBJECT_ACE_TYPE 0x08 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); extern int zfs_secpolicy_snapshot_perms(const char *name, cred_t *cr); extern int zfs_secpolicy_rename_perms(const char *from, const char *to, cred_t *cr); extern int zfs_secpolicy_destroy_perms(const char *name, cred_t *cr); extern zoneid_t getzoneid(void); /* SID stuff */ typedef struct ksiddomain { uint_t kd_ref; uint_t kd_len; char *kd_name; } ksiddomain_t; ksiddomain_t *ksid_lookupdomain(const char *); void ksiddomain_rele(ksiddomain_t *); #define DDI_SLEEP KM_SLEEP #define ddi_log_sysevent(_a, _b, _c, _d, _e, _f, _g) \ sysevent_post_event(_c, _d, _b, "libzpool", _e, _f) /* * Cyclic information */ extern kmutex_t cpu_lock; typedef uintptr_t cyclic_id_t; typedef uint16_t cyc_level_t; typedef void (*cyc_func_t)(void *); #define CY_LOW_LEVEL 0 #define CY_INFINITY INT64_MAX #define CYCLIC_NONE ((cyclic_id_t)0) typedef struct cyc_time { hrtime_t cyt_when; hrtime_t cyt_interval; } cyc_time_t; typedef struct cyc_handler { cyc_func_t cyh_func; void *cyh_arg; cyc_level_t cyh_level; } cyc_handler_t; extern cyclic_id_t cyclic_add(cyc_handler_t *, cyc_time_t *); extern void cyclic_remove(cyclic_id_t); extern int cyclic_reprogram(cyclic_id_t, hrtime_t); /* * Buf structure */ #define B_BUSY 0x0001 #define B_DONE 0x0002 #define B_ERROR 0x0004 #define B_READ 0x0040 /* read when I/O occurs */ #define B_WRITE 0x0100 /* non-read pseudo-flag */ typedef struct buf { int b_flags; size_t b_bcount; union { caddr_t b_addr; } b_un; lldaddr_t _b_blkno; #define b_lblkno _b_blkno._f size_t b_resid; size_t b_bufsize; int (*b_iodone)(struct buf *); int b_error; void *b_private; } buf_t; extern void bioinit(buf_t *); extern void biodone(buf_t *); extern void bioerror(buf_t *, int); extern int geterror(buf_t *); #ifdef __cplusplus } #endif #endif /* _SYS_ZFS_CONTEXT_H */