diff options
Diffstat (limited to 'sys/kern')
50 files changed, 838 insertions, 760 deletions
diff --git a/sys/kern/imgact_elf.c b/sys/kern/imgact_elf.c index 1bc2491a1a12..779158b41221 100644 --- a/sys/kern/imgact_elf.c +++ b/sys/kern/imgact_elf.c @@ -92,7 +92,7 @@ #define ELF_ABI_ID __CONCAT(elf, __ELF_WORD_SIZE) static int __elfN(check_header)(const Elf_Ehdr *hdr); -static Elf_Brandinfo *__elfN(get_brandinfo)(struct image_params *imgp, +static const Elf_Brandinfo *__elfN(get_brandinfo)(struct image_params *imgp, const char *interp, int32_t *osrel, uint32_t *fctl0); static int __elfN(load_file)(struct proc *p, const char *file, u_long *addr, u_long *entry); @@ -104,7 +104,7 @@ static bool __elfN(freebsd_trans_osrel)(const Elf_Note *note, int32_t *osrel); static bool kfreebsd_trans_osrel(const Elf_Note *note, int32_t *osrel); static bool __elfN(check_note)(struct image_params *imgp, - Elf_Brandnote *checknote, int32_t *osrel, bool *has_fctl0, + const Elf_Brandnote *checknote, int32_t *osrel, bool *has_fctl0, uint32_t *fctl0); static vm_prot_t __elfN(trans_prot)(Elf_Word); static Elf_Word __elfN(untrans_prot)(vm_prot_t); @@ -227,11 +227,11 @@ SYSCTL_BOOL(ELF_NODE_OID, OID_AUTO, allow_wx, CTLFLAG_RWTUN, &__elfN(allow_wx), 0, "Allow pages to be mapped simultaneously writable and executable"); -static Elf_Brandinfo *elf_brand_list[MAX_BRANDS]; +static const Elf_Brandinfo *elf_brand_list[MAX_BRANDS]; #define aligned(a, t) (rounddown2((u_long)(a), sizeof(t)) == (u_long)(a)) -Elf_Brandnote __elfN(freebsd_brandnote) = { +const Elf_Brandnote __elfN(freebsd_brandnote) = { .hdr.n_namesz = sizeof(FREEBSD_ABI_VENDOR), .hdr.n_descsz = sizeof(int32_t), .hdr.n_type = NT_FREEBSD_ABI_TAG, @@ -254,7 +254,7 @@ __elfN(freebsd_trans_osrel)(const Elf_Note *note, int32_t *osrel) static int GNU_KFREEBSD_ABI_DESC = 3; -Elf_Brandnote __elfN(kfreebsd_brandnote) = { +const Elf_Brandnote __elfN(kfreebsd_brandnote) = { .hdr.n_namesz = sizeof(GNU_ABI_VENDOR), .hdr.n_descsz = 16, /* XXX at least 16 */ .hdr.n_type = 1, @@ -286,7 +286,7 @@ kfreebsd_trans_osrel(const Elf_Note *note, int32_t *osrel) } int -__elfN(insert_brand_entry)(Elf_Brandinfo *entry) +__elfN(insert_brand_entry)(const Elf_Brandinfo *entry) { int i; @@ -305,7 +305,7 @@ __elfN(insert_brand_entry)(Elf_Brandinfo *entry) } int -__elfN(remove_brand_entry)(Elf_Brandinfo *entry) +__elfN(remove_brand_entry)(const Elf_Brandinfo *entry) { int i; @@ -321,7 +321,7 @@ __elfN(remove_brand_entry)(Elf_Brandinfo *entry) } bool -__elfN(brand_inuse)(Elf_Brandinfo *entry) +__elfN(brand_inuse)(const Elf_Brandinfo *entry) { struct proc *p; bool rval = false; @@ -338,12 +338,12 @@ __elfN(brand_inuse)(Elf_Brandinfo *entry) return (rval); } -static Elf_Brandinfo * +static const Elf_Brandinfo * __elfN(get_brandinfo)(struct image_params *imgp, const char *interp, int32_t *osrel, uint32_t *fctl0) { const Elf_Ehdr *hdr = (const Elf_Ehdr *)imgp->image_header; - Elf_Brandinfo *bi, *bi_m; + const Elf_Brandinfo *bi, *bi_m; bool ret, has_fctl0; int i, interp_name_len; @@ -492,7 +492,7 @@ __elfN(phdr_in_zero_page)(const Elf_Ehdr *hdr) static int __elfN(check_header)(const Elf_Ehdr *hdr) { - Elf_Brandinfo *bi; + const Elf_Brandinfo *bi; int i; if (!IS_ELF(*hdr) || @@ -1109,7 +1109,7 @@ __CONCAT(exec_, __elfN(imgact))(struct image_params *imgp) struct vmspace *vmspace; vm_map_t map; char *interp; - Elf_Brandinfo *brand_info; + const Elf_Brandinfo *brand_info; struct sysentvec *sv; u_long addr, baddr, entry, proghdr; u_long maxalign, maxsalign, mapsz, maxv, maxv1, anon_loc; @@ -1925,7 +1925,7 @@ __elfN(puthdr)(struct thread *td, void *hdr, size_t hdrsize, int numsegs, Elf_Phdr *phdr; Elf_Shdr *shdr; struct phdr_closure phc; - Elf_Brandinfo *bi; + const Elf_Brandinfo *bi; ehdr = (Elf_Ehdr *)hdr; bi = td->td_proc->p_elf_brandinfo; @@ -2831,7 +2831,7 @@ __elfN(parse_notes)(const struct image_params *imgp, const Elf_Note *checknote, } if ((const char *)note_end - (const char *)note < sizeof(Elf_Note)) { - uprintf("ELF note to short\n"); + uprintf("ELF note too short\n"); goto retf; } if (note->n_namesz != checknote->n_namesz || @@ -2839,9 +2839,9 @@ __elfN(parse_notes)(const struct image_params *imgp, const Elf_Note *checknote, note->n_type != checknote->n_type) goto nextnote; note_name = (const char *)(note + 1); - if (note_name + checknote->n_namesz >= - (const char *)note_end || strncmp(note_vendor, - note_name, checknote->n_namesz) != 0) + if (note_name + roundup2(note->n_namesz, ELF_NOTE_ROUNDSIZE) + + note->n_descsz >= (const char *)note_end || + strncmp(note_vendor, note_name, checknote->n_namesz) != 0) goto nextnote; if (cb(note, cb_arg, &res)) @@ -2861,7 +2861,7 @@ ret: } struct brandnote_cb_arg { - Elf_Brandnote *brandnote; + const Elf_Brandnote *brandnote; int32_t *osrel; }; @@ -2883,7 +2883,7 @@ brandnote_cb(const Elf_Note *note, void *arg0, bool *res) return (true); } -static Elf_Note fctl_note = { +static const Elf_Note fctl_note = { .n_namesz = sizeof(FREEBSD_ABI_VENDOR), .n_descsz = sizeof(uint32_t), .n_type = NT_FREEBSD_FEATURE_CTL, @@ -2918,7 +2918,7 @@ note_fctl_cb(const Elf_Note *note, void *arg0, bool *res) * as for headers. */ static bool -__elfN(check_note)(struct image_params *imgp, Elf_Brandnote *brandnote, +__elfN(check_note)(struct image_params *imgp, const Elf_Brandnote *brandnote, int32_t *osrel, bool *has_fctl0, uint32_t *fctl0) { const Elf_Phdr *phdr; diff --git a/sys/kern/init_main.c b/sys/kern/init_main.c index 87ffdb8dbf07..6612ac685936 100644 --- a/sys/kern/init_main.c +++ b/sys/kern/init_main.c @@ -384,7 +384,7 @@ C_SYSINIT(diagwarn2, SI_SUB_LAST, SI_ORDER_FIFTH, #if __SIZEOF_LONG__ == 4 static const char ilp32_warn[] = - "WARNING: 32-bit kernels are deprecated and may be removed in FreeBSD 15.0.\n"; + "WARNING: 32-bit kernels are deprecated and may be removed in FreeBSD 16.0.\n"; C_SYSINIT(ilp32warn, SI_SUB_COPYRIGHT, SI_ORDER_FIFTH, print_caddr_t, ilp32_warn); C_SYSINIT(ilp32warn2, SI_SUB_LAST, SI_ORDER_FIFTH, diff --git a/sys/kern/init_sysent.c b/sys/kern/init_sysent.c index e42e7dcf8b44..cd305de1ed44 100644 --- a/sys/kern/init_sysent.c +++ b/sys/kern/init_sysent.c @@ -665,4 +665,5 @@ struct sysent sysent[] = { { .sy_narg = AS(setgroups_args), .sy_call = (sy_call_t *)sys_setgroups, .sy_auevent = AUE_SETGROUPS, .sy_flags = 0, .sy_thrcnt = SY_THR_STATIC }, /* 596 = setgroups */ { .sy_narg = AS(jail_attach_jd_args), .sy_call = (sy_call_t *)sys_jail_attach_jd, .sy_auevent = AUE_JAIL_ATTACH, .sy_flags = 0, .sy_thrcnt = SY_THR_STATIC }, /* 597 = jail_attach_jd */ { .sy_narg = AS(jail_remove_jd_args), .sy_call = (sy_call_t *)sys_jail_remove_jd, .sy_auevent = AUE_JAIL_REMOVE, .sy_flags = 0, .sy_thrcnt = SY_THR_STATIC }, /* 598 = jail_remove_jd */ + { .sy_narg = AS(kexec_load_args), .sy_call = (sy_call_t *)sys_kexec_load, .sy_auevent = AUE_NULL, .sy_flags = 0, .sy_thrcnt = SY_THR_STATIC }, /* 599 = kexec_load */ }; diff --git a/sys/kern/kern_boottrace.c b/sys/kern/kern_boottrace.c index 1fa87955a299..c83255bc74ee 100644 --- a/sys/kern/kern_boottrace.c +++ b/sys/kern/kern_boottrace.c @@ -579,7 +579,7 @@ sysctl_boottrace_reset(SYSCTL_HANDLER_ARGS) } static void -boottrace_init(void) +boottrace_init(void *dummy __unused) { if (!boottrace_enabled) diff --git a/sys/kern/kern_conf.c b/sys/kern/kern_conf.c index b891ed84957a..2da51d84ff60 100644 --- a/sys/kern/kern_conf.c +++ b/sys/kern/kern_conf.c @@ -664,7 +664,7 @@ prep_cdevsw(struct cdevsw *devsw, int flags) if ((devsw->d_flags & D_GIANTOK) == 0) { printf( "WARNING: Device \"%s\" is Giant locked and may be " - "deleted before FreeBSD 15.0.\n", + "deleted before FreeBSD 16.0.\n", devsw->d_name == NULL ? "???" : devsw->d_name); } if (devsw->d_gianttrick == NULL) { @@ -1163,6 +1163,9 @@ destroy_devl(struct cdev *dev) devfs_destroy_cdevpriv(p); mtx_lock(&cdevpriv_mtx); } + while (cdp->cdp_fdpriv_dtrc != 0) { + msleep(&cdp->cdp_fdpriv_dtrc, &cdevpriv_mtx, 0, "cdfdpc", 0); + } mtx_unlock(&cdevpriv_mtx); dev_lock(); diff --git a/sys/kern/kern_descrip.c b/sys/kern/kern_descrip.c index 19118eb7f275..a71a601733e5 100644 --- a/sys/kern/kern_descrip.c +++ b/sys/kern/kern_descrip.c @@ -2486,7 +2486,7 @@ fdunshare(struct thread *td) if (refcount_load(&p->p_fd->fd_refcnt) == 1) return; - tmp = fdcopy(p->p_fd); + tmp = fdcopy(p->p_fd, p); fdescfree(td); p->p_fd = tmp; } @@ -2515,14 +2515,17 @@ pdunshare(struct thread *td) * this is to ease callers, not catch errors. */ struct filedesc * -fdcopy(struct filedesc *fdp) +fdcopy(struct filedesc *fdp, struct proc *p1) { struct filedesc *newfdp; struct filedescent *nfde, *ofde; + struct file *fp; int i, lastfile; + bool fork_pass; MPASS(fdp != NULL); + fork_pass = false; newfdp = fdinit(); FILEDESC_SLOCK(fdp); for (;;) { @@ -2533,10 +2536,35 @@ fdcopy(struct filedesc *fdp) fdgrowtable(newfdp, lastfile + 1); FILEDESC_SLOCK(fdp); } - /* copy all passable descriptors (i.e. not kqueue) */ + + /* + * Copy all passable descriptors (i.e. not kqueue), and + * prepare to handle copyable but not passable descriptors + * (kqueues). + * + * The pass to handle copying is performed after all passable + * files are installed into the new file descriptor's table, + * since kqueues need all referenced file descriptors already + * valid, including other kqueues. For the same reason the + * copying is done in two passes by itself, first installing + * not fully initialized ('empty') copyable files into the new + * fd table, and then giving the subsystems a second chance to + * really fill the copied file backing structure with the + * content. + */ newfdp->fd_freefile = fdp->fd_freefile; FILEDESC_FOREACH_FDE(fdp, i, ofde) { - if ((ofde->fde_file->f_ops->fo_flags & DFLAG_PASSABLE) == 0 || + const struct fileops *ops; + + ops = ofde->fde_file->f_ops; + fp = NULL; + if ((ops->fo_flags & DFLAG_FORK) != 0 && + (ofde->fde_flags & UF_FOCLOSE) == 0) { + if (ops->fo_fork(newfdp, ofde->fde_file, &fp, p1, + curthread) != 0) + continue; + fork_pass = true; + } else if ((ops->fo_flags & DFLAG_PASSABLE) == 0 || (ofde->fde_flags & UF_FOCLOSE) != 0 || !fhold(ofde->fde_file)) { if (newfdp->fd_freefile == fdp->fd_freefile) @@ -2545,11 +2573,30 @@ fdcopy(struct filedesc *fdp) } nfde = &newfdp->fd_ofiles[i]; *nfde = *ofde; + if (fp != NULL) + nfde->fde_file = fp; filecaps_copy(&ofde->fde_caps, &nfde->fde_caps, true); fdused_init(newfdp, i); } MPASS(newfdp->fd_freefile != -1); FILEDESC_SUNLOCK(fdp); + + /* + * Now handle copying kqueues, since all fds, including + * kqueues, are in place. + */ + if (__predict_false(fork_pass)) { + FILEDESC_FOREACH_FDE(newfdp, i, nfde) { + const struct fileops *ops; + + ops = nfde->fde_file->f_ops; + if ((ops->fo_flags & DFLAG_FORK) == 0 || + nfde->fde_file == NULL) + continue; + ops->fo_fork(newfdp, NULL, &nfde->fde_file, p1, + curthread); + } + } return (newfdp); } diff --git a/sys/kern/kern_devctl.c b/sys/kern/kern_devctl.c index 7a2818c29b1a..a37cb23efed8 100644 --- a/sys/kern/kern_devctl.c +++ b/sys/kern/kern_devctl.c @@ -130,6 +130,7 @@ static const struct filterops devctl_rfiltops = { .f_isfd = 1, .f_detach = filt_devctl_detach, .f_event = filt_devctl_read, + .f_copy = knote_triv_copy, }; static struct cdev *devctl_dev; @@ -140,7 +141,7 @@ static struct devctlbridge { } devctl_notify_hook = { .send_f = NULL }; static void -devctl_init(void) +devctl_init(void *dummy __unused) { int reserve; uma_zone_t z; diff --git a/sys/kern/kern_event.c b/sys/kern/kern_event.c index 23d8dc9cf54a..1baa24d278bf 100644 --- a/sys/kern/kern_event.c +++ b/sys/kern/kern_event.c @@ -134,6 +134,7 @@ static fo_kqfilter_t kqueue_kqfilter; static fo_stat_t kqueue_stat; static fo_close_t kqueue_close; static fo_fill_kinfo_t kqueue_fill_kinfo; +static fo_fork_t kqueue_fork; static const struct fileops kqueueops = { .fo_read = invfo_rdwr, @@ -148,7 +149,9 @@ static const struct fileops kqueueops = { .fo_chown = invfo_chown, .fo_sendfile = invfo_sendfile, .fo_cmp = file_kcmp_generic, + .fo_fork = kqueue_fork, .fo_fill_kinfo = kqueue_fill_kinfo, + .fo_flags = DFLAG_FORK, }; static int knote_attach(struct knote *kn, struct kqueue *kq); @@ -156,7 +159,7 @@ static void knote_drop(struct knote *kn, struct thread *td); static void knote_drop_detached(struct knote *kn, struct thread *td); static void knote_enqueue(struct knote *kn); static void knote_dequeue(struct knote *kn); -static void knote_init(void); +static void knote_init(void *); static struct knote *knote_alloc(int mflag); static void knote_free(struct knote *kn); @@ -176,6 +179,7 @@ static void filt_timerdetach(struct knote *kn); static void filt_timerstart(struct knote *kn, sbintime_t to); static void filt_timertouch(struct knote *kn, struct kevent *kev, u_long type); +static int filt_timercopy(struct knote *kn, struct proc *p1); static int filt_timervalidate(struct knote *kn, sbintime_t *to); static int filt_timer(struct knote *kn, long hint); static int filt_userattach(struct knote *kn); @@ -187,11 +191,13 @@ static void filt_usertouch(struct knote *kn, struct kevent *kev, static const struct filterops file_filtops = { .f_isfd = 1, .f_attach = filt_fileattach, + .f_copy = knote_triv_copy, }; static const struct filterops kqread_filtops = { .f_isfd = 1, .f_detach = filt_kqdetach, .f_event = filt_kqueue, + .f_copy = knote_triv_copy, }; /* XXX - move to kern_proc.c? */ static const struct filterops proc_filtops = { @@ -199,12 +205,14 @@ static const struct filterops proc_filtops = { .f_attach = filt_procattach, .f_detach = filt_procdetach, .f_event = filt_proc, + .f_copy = knote_triv_copy, }; static const struct filterops jail_filtops = { .f_isfd = 0, .f_attach = filt_jailattach, .f_detach = filt_jaildetach, .f_event = filt_jail, + .f_copy = knote_triv_copy, }; static const struct filterops timer_filtops = { .f_isfd = 0, @@ -212,12 +220,14 @@ static const struct filterops timer_filtops = { .f_detach = filt_timerdetach, .f_event = filt_timer, .f_touch = filt_timertouch, + .f_copy = filt_timercopy, }; static const struct filterops user_filtops = { .f_attach = filt_userattach, .f_detach = filt_userdetach, .f_event = filt_user, .f_touch = filt_usertouch, + .f_copy = knote_triv_copy, }; static uma_zone_t knote_zone; @@ -347,6 +357,7 @@ filt_nullattach(struct knote *kn) static const struct filterops null_filtops = { .f_isfd = 0, .f_attach = filt_nullattach, + .f_copy = knote_triv_copy, }; /* XXX - make SYSINIT to add these, and move into respective modules. */ @@ -940,6 +951,30 @@ filt_timerattach(struct knote *kn) return (0); } +static int +filt_timercopy(struct knote *kn, struct proc *p) +{ + struct kq_timer_cb_data *kc_src, *kc; + + if (atomic_fetchadd_int(&kq_ncallouts, 1) + 1 > kq_calloutmax) { + atomic_subtract_int(&kq_ncallouts, 1); + return (ENOMEM); + } + + kn->kn_status &= ~KN_DETACHED; + kc_src = kn->kn_ptr.p_v; + kn->kn_ptr.p_v = kc = malloc(sizeof(*kc), M_KQUEUE, M_WAITOK); + kc->kn = kn; + kc->p = p; + kc->flags = kc_src->flags & ~KQ_TIMER_CB_ENQUEUED; + kc->next = kc_src->next; + kc->to = kc_src->to; + kc->cpuid = PCPU_GET(cpuid); + callout_init(&kc->c, 1); + kqtimer_sched_callout(kc); + return (0); +} + static void filt_timerstart(struct knote *kn, sbintime_t to) { @@ -1151,7 +1186,7 @@ int sys_kqueue(struct thread *td, struct kqueue_args *uap) { - return (kern_kqueue(td, 0, NULL)); + return (kern_kqueue(td, 0, false, NULL)); } int @@ -1159,55 +1194,76 @@ sys_kqueuex(struct thread *td, struct kqueuex_args *uap) { int flags; - if ((uap->flags & ~(KQUEUE_CLOEXEC)) != 0) + if ((uap->flags & ~(KQUEUE_CLOEXEC | KQUEUE_CPONFORK)) != 0) return (EINVAL); flags = 0; if ((uap->flags & KQUEUE_CLOEXEC) != 0) flags |= O_CLOEXEC; - return (kern_kqueue(td, flags, NULL)); + return (kern_kqueue(td, flags, (uap->flags & KQUEUE_CPONFORK) != 0, + NULL)); } static void -kqueue_init(struct kqueue *kq) +kqueue_init(struct kqueue *kq, bool cponfork) { mtx_init(&kq->kq_lock, "kqueue", NULL, MTX_DEF | MTX_DUPOK); TAILQ_INIT(&kq->kq_head); knlist_init_mtx(&kq->kq_sel.si_note, &kq->kq_lock); TASK_INIT(&kq->kq_task, 0, kqueue_task, kq); + if (cponfork) + kq->kq_state |= KQ_CPONFORK; } -int -kern_kqueue(struct thread *td, int flags, struct filecaps *fcaps) +static int +kern_kqueue_alloc(struct thread *td, struct filedesc *fdp, int *fdip, + struct file **fpp, int flags, struct filecaps *fcaps, bool cponfork, + struct kqueue **kqp) { - struct filedesc *fdp; - struct kqueue *kq; - struct file *fp; struct ucred *cred; - int fd, error; + struct kqueue *kq; + int error; - fdp = td->td_proc->p_fd; cred = td->td_ucred; if (!chgkqcnt(cred->cr_ruidinfo, 1, lim_cur(td, RLIMIT_KQUEUES))) return (ENOMEM); - error = falloc_caps(td, &fp, &fd, flags, fcaps); + error = fdip != NULL ? falloc_caps(td, fpp, fdip, flags, fcaps) : + _falloc_noinstall(td, fpp, 1); if (error != 0) { chgkqcnt(cred->cr_ruidinfo, -1, 0); return (error); } /* An extra reference on `fp' has been held for us by falloc(). */ - kq = malloc(sizeof *kq, M_KQUEUE, M_WAITOK | M_ZERO); - kqueue_init(kq); + kq = malloc(sizeof(*kq), M_KQUEUE, M_WAITOK | M_ZERO); + kqueue_init(kq, cponfork); kq->kq_fdp = fdp; kq->kq_cred = crhold(cred); - FILEDESC_XLOCK(fdp); + if (fdip != NULL) + FILEDESC_XLOCK(fdp); TAILQ_INSERT_HEAD(&fdp->fd_kqlist, kq, kq_list); - FILEDESC_XUNLOCK(fdp); + if (fdip != NULL) + FILEDESC_XUNLOCK(fdp); + + finit(*fpp, FREAD | FWRITE, DTYPE_KQUEUE, kq, &kqueueops); + *kqp = kq; + return (0); +} + +int +kern_kqueue(struct thread *td, int flags, bool cponfork, struct filecaps *fcaps) +{ + struct kqueue *kq; + struct file *fp; + int fd, error; + + error = kern_kqueue_alloc(td, td->td_proc->p_fd, &fd, &fp, flags, + fcaps, cponfork, &kq); + if (error != 0) + return (error); - finit(fp, FREAD | FWRITE, DTYPE_KQUEUE, kq, &kqueueops); fdrop(fp, td); td->td_retval[0] = fd; @@ -1488,7 +1544,7 @@ kern_kevent_anonymous(struct thread *td, int nevents, struct kqueue kq = {}; int error; - kqueue_init(&kq); + kqueue_init(&kq, false); kq.kq_refcnt = 1; error = kqueue_kevent(&kq, td, nevents, nevents, k_ops, NULL); kqueue_drain(&kq, td); @@ -1576,7 +1632,7 @@ kqueue_fo_release(int filt) mtx_lock(&filterops_lock); KASSERT(sysfilt_ops[~filt].for_refcnt > 0, - ("filter object refcount not valid on release")); + ("filter object %d refcount not valid on release", filt)); sysfilt_ops[~filt].for_refcnt--; mtx_unlock(&filterops_lock); } @@ -1855,17 +1911,8 @@ done: } static int -kqueue_acquire(struct file *fp, struct kqueue **kqp) +kqueue_acquire_ref(struct kqueue *kq) { - int error; - struct kqueue *kq; - - error = 0; - - kq = fp->f_data; - if (fp->f_type != DTYPE_KQUEUE || kq == NULL) - return (EINVAL); - *kqp = kq; KQ_LOCK(kq); if ((kq->kq_state & KQ_CLOSING) == KQ_CLOSING) { KQ_UNLOCK(kq); @@ -1873,8 +1920,22 @@ kqueue_acquire(struct file *fp, struct kqueue **kqp) } kq->kq_refcnt++; KQ_UNLOCK(kq); + return (0); +} - return error; +static int +kqueue_acquire(struct file *fp, struct kqueue **kqp) +{ + struct kqueue *kq; + int error; + + kq = fp->f_data; + if (fp->f_type != DTYPE_KQUEUE || kq == NULL) + return (EINVAL); + error = kqueue_acquire_ref(kq); + if (error == 0) + *kqp = kq; + return (error); } static void @@ -2887,7 +2948,7 @@ knote_dequeue(struct knote *kn) } static void -knote_init(void) +knote_init(void *dummy __unused) { knote_zone = uma_zcreate("KNOTE", sizeof(struct knote), NULL, NULL, @@ -2937,6 +2998,152 @@ noacquire: return (error); } +static int +kqueue_fork_alloc(struct filedesc *fdp, struct file *fp, struct file **fp1, + struct thread *td) +{ + struct kqueue *kq, *kq1; + int error; + + MPASS(fp->f_type == DTYPE_KQUEUE); + kq = fp->f_data; + if ((kq->kq_state & KQ_CPONFORK) == 0) + return (EOPNOTSUPP); + error = kqueue_acquire_ref(kq); + if (error != 0) + return (error); + error = kern_kqueue_alloc(td, fdp, NULL, fp1, 0, NULL, true, &kq1); + if (error == 0) { + kq1->kq_forksrc = kq; + (*fp1)->f_flag = fp->f_flag & (FREAD | FWRITE | FEXEC | + O_CLOEXEC | O_CLOFORK); + } else { + kqueue_release(kq, 0); + } + return (error); +} + +static void +kqueue_fork_copy_knote(struct kqueue *kq1, struct knote *kn, struct proc *p1, + struct filedesc *fdp) +{ + struct knote *kn1; + const struct filterops *fop; + int error; + + fop = kn->kn_fop; + if (fop->f_copy == NULL || (fop->f_isfd && + fdp->fd_files->fdt_ofiles[kn->kn_kevent.ident].fde_file == NULL)) + return; + error = kqueue_expand(kq1, fop, kn->kn_kevent.ident, M_WAITOK); + if (error != 0) + return; + + kn1 = knote_alloc(M_WAITOK); + *kn1 = *kn; + kn1->kn_status |= KN_DETACHED; + kn1->kn_status &= ~KN_QUEUED; + kn1->kn_kq = kq1; + error = fop->f_copy(kn1, p1); + if (error != 0) { + knote_free(kn1); + return; + } + (void)kqueue_fo_find(kn->kn_kevent.filter); + if (fop->f_isfd && !fhold(kn1->kn_fp)) { + fop->f_detach(kn1); + kqueue_fo_release(kn->kn_kevent.filter); + knote_free(kn1); + return; + } + if (kn->kn_knlist != NULL) + knlist_add(kn->kn_knlist, kn1, 0); + KQ_LOCK(kq1); + knote_attach(kn1, kq1); + kn1->kn_influx = 0; + if ((kn->kn_status & KN_QUEUED) != 0) + knote_enqueue(kn1); + KQ_UNLOCK(kq1); +} + +static void +kqueue_fork_copy_list(struct klist *knlist, struct knote *marker, + struct kqueue *kq, struct kqueue *kq1, struct proc *p1, + struct filedesc *fdp) +{ + struct knote *kn; + + KQ_OWNED(kq); + kn = SLIST_FIRST(knlist); + while (kn != NULL) { + if ((kn->kn_status & KN_DETACHED) != 0 || + (kn_in_flux(kn) && (kn->kn_status & KN_SCAN) == 0)) { + kn = SLIST_NEXT(kn, kn_link); + continue; + } + kn_enter_flux(kn); + SLIST_INSERT_AFTER(kn, marker, kn_link); + KQ_UNLOCK(kq); + kqueue_fork_copy_knote(kq1, kn, p1, fdp); + KQ_LOCK(kq); + kn_leave_flux(kn); + kn = SLIST_NEXT(marker, kn_link); + /* XXXKIB switch kn_link to LIST? */ + SLIST_REMOVE(knlist, marker, knote, kn_link); + } +} + +static int +kqueue_fork_copy(struct filedesc *fdp, struct file *fp, struct file *fp1, + struct proc *p1, struct thread *td) +{ + struct kqueue *kq, *kq1; + struct knote *marker; + int error, i; + + error = 0; + MPASS(fp == NULL); + MPASS(fp1->f_type == DTYPE_KQUEUE); + + kq1 = fp1->f_data; + kq = kq1->kq_forksrc; + marker = knote_alloc(M_WAITOK); + marker->kn_status = KN_MARKER; + + KQ_LOCK(kq); + for (i = 0; i < kq->kq_knlistsize; i++) { + kqueue_fork_copy_list(&kq->kq_knlist[i], marker, kq, kq1, + p1, fdp); + } + if (kq->kq_knhashmask != 0) { + for (i = 0; i <= kq->kq_knhashmask; i++) { + kqueue_fork_copy_list(&kq->kq_knhash[i], marker, kq, + kq1, p1, fdp); + } + } + kqueue_release(kq, 1); + kq1->kq_forksrc = NULL; + KQ_UNLOCK(kq); + + knote_free(marker); + return (error); +} + +static int +kqueue_fork(struct filedesc *fdp, struct file *fp, struct file **fp1, + struct proc *p1, struct thread *td) +{ + if (*fp1 == NULL) + return (kqueue_fork_alloc(fdp, fp, fp1, td)); + return (kqueue_fork_copy(fdp, fp, *fp1, p1, td)); +} + +int +knote_triv_copy(struct knote *kn __unused, struct proc *p1 __unused) +{ + return (0); +} + struct knote_status_export_bit { int kn_status_bit; int knt_status_bit; diff --git a/sys/kern/kern_exec.c b/sys/kern/kern_exec.c index 0fc2d0e7f1bc..2bdd6faa025a 100644 --- a/sys/kern/kern_exec.c +++ b/sys/kern/kern_exec.c @@ -418,7 +418,7 @@ do_execve(struct thread *td, struct image_args *args, struct mac *mac_p, #endif int error, i, orig_osrel; uint32_t orig_fctl0; - Elf_Brandinfo *orig_brandinfo; + const Elf_Brandinfo *orig_brandinfo; size_t freepath_size; static const char fexecv_proc_title[] = "(fexecv)"; @@ -1314,7 +1314,7 @@ exec_map_stack(struct image_params *imgp) MAP_INHERIT_SHARE | MAP_ACC_NO_CHARGE); } else { sharedpage_addr = sv->sv_shared_page_base; - vm_map_fixed(map, obj, 0, + error = vm_map_fixed(map, obj, 0, sharedpage_addr, sv->sv_shared_page_len, VM_PROT_READ | VM_PROT_EXECUTE, VM_PROT_READ | VM_PROT_EXECUTE, diff --git a/sys/kern/kern_exit.c b/sys/kern/kern_exit.c index ab8ed32ad189..c4b1c8201ff2 100644 --- a/sys/kern/kern_exit.c +++ b/sys/kern/kern_exit.c @@ -807,7 +807,7 @@ kern_abort2(struct thread *td, const char *why, int nargs, void **uargs) } if (nargs > 0) { sbuf_putc(sb, '('); - for (i = 0;i < nargs; i++) + for (i = 0; i < nargs; i++) sbuf_printf(sb, "%s%p", i == 0 ? "" : ", ", uargs[i]); sbuf_putc(sb, ')'); } diff --git a/sys/kern/kern_fork.c b/sys/kern/kern_fork.c index 7f6abae187b3..8b237b6dbd17 100644 --- a/sys/kern/kern_fork.c +++ b/sys/kern/kern_fork.c @@ -423,7 +423,7 @@ do_fork(struct thread *td, struct fork_req *fr, struct proc *p2, struct thread * pd = pdshare(p1->p_pd); else pd = pdcopy(p1->p_pd); - fd = fdcopy(p1->p_fd); + fd = fdcopy(p1->p_fd, p2); fdtol = NULL; } else { if (fr->fr_flags2 & FR2_SHARE_PATHS) diff --git a/sys/kern/kern_jail.c b/sys/kern/kern_jail.c index 3697d95fe0e5..267b60ffb5bc 100644 --- a/sys/kern/kern_jail.c +++ b/sys/kern/kern_jail.c @@ -1088,6 +1088,7 @@ kern_jail_set(struct thread *td, struct uio *optuio, int flags) else { if (!(flags & (JAIL_USE_DESC | JAIL_AT_DESC | JAIL_GET_DESC | JAIL_OWN_DESC))) { + error = EINVAL; vfs_opterror(opts, "unexpected desc"); goto done_errmsg; } @@ -2518,6 +2519,7 @@ kern_jail_get(struct thread *td, struct uio *optuio, int flags) } else if (error == 0) { if (!(flags & (JAIL_USE_DESC | JAIL_AT_DESC | JAIL_GET_DESC | JAIL_OWN_DESC))) { + error = EINVAL; vfs_opterror(opts, "unexpected desc"); goto done; } @@ -2909,12 +2911,6 @@ prison_remove(struct prison *pr) { sx_assert(&allprison_lock, SA_XLOCKED); mtx_assert(&pr->pr_mtx, MA_OWNED); - if (!prison_isalive(pr)) { - /* Silently ignore already-dying prisons. */ - mtx_unlock(&pr->pr_mtx); - sx_xunlock(&allprison_lock); - return; - } prison_deref(pr, PD_KILL | PD_DEREF | PD_LOCKED | PD_LIST_XLOCKED); } @@ -3461,12 +3457,17 @@ prison_deref(struct prison *pr, int flags) /* Kill the prison and its descendents. */ KASSERT(pr != &prison0, ("prison_deref trying to kill prison0")); - if (!(flags & PD_DEREF)) { - prison_hold(pr); - flags |= PD_DEREF; + if (!prison_isalive(pr)) { + /* Silently ignore already-dying prisons. */ + flags &= ~PD_KILL; + } else { + if (!(flags & PD_DEREF)) { + prison_hold(pr); + flags |= PD_DEREF; + } + flags = prison_lock_xlock(pr, flags); + prison_deref_kill(pr, &freeprison); } - flags = prison_lock_xlock(pr, flags); - prison_deref_kill(pr, &freeprison); } if (flags & PD_DEUREF) { /* Drop a user reference. */ diff --git a/sys/kern/kern_jaildesc.c b/sys/kern/kern_jaildesc.c index 3f322b271400..a564393d3366 100644 --- a/sys/kern/kern_jaildesc.c +++ b/sys/kern/kern_jaildesc.c @@ -344,6 +344,7 @@ static const struct filterops jaildesc_kqops = { .f_isfd = 1, .f_detach = jaildesc_kqops_detach, .f_event = jaildesc_kqops_event, + .f_copy = knote_triv_copy, }; static int diff --git a/sys/kern/kern_jailmeta.c b/sys/kern/kern_jailmeta.c index 4e37eccad03a..91bb7155820d 100644 --- a/sys/kern/kern_jailmeta.c +++ b/sys/kern/kern_jailmeta.c @@ -599,22 +599,18 @@ SYSCTL_PROC(_security_jail, OID_AUTO, env, /* Setup and tear down. */ -static int +static void jm_sysinit(void *arg __unused) { meta.osd_slot = osd_jail_register(jm_osd_destructor, meta.methods); env.osd_slot = osd_jail_register(jm_osd_destructor, env.methods); - - return (0); } -static int +static void jm_sysuninit(void *arg __unused) { osd_jail_deregister(meta.osd_slot); osd_jail_deregister(env.osd_slot); - - return (0); } SYSINIT(jailmeta, SI_SUB_DRIVERS, SI_ORDER_ANY, jm_sysinit, NULL); diff --git a/sys/kern/kern_kexec.c b/sys/kern/kern_kexec.c new file mode 100644 index 000000000000..2efea7dcf9a7 --- /dev/null +++ b/sys/kern/kern_kexec.c @@ -0,0 +1,350 @@ +/*- + * SPDX-License-Identifier: BSD-2-Clause + * + * Copyright (c) 2025 Juniper Networks, Inc. + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +#include <sys/param.h> +#include <sys/bus.h> +#include <sys/eventhandler.h> +#include <sys/kernel.h> +#ifdef INTRNG +#include <sys/intr.h> +#endif +#include <sys/kexec.h> +#include <sys/malloc.h> +#include <sys/proc.h> +#include <sys/priv.h> +#include <sys/reboot.h> +#include <sys/rman.h> +#include <sys/rwlock.h> +#include <sys/smp.h> +#include <sys/syscallsubr.h> +#include <sys/sysproto.h> + +#include <vm/vm.h> +#include <vm/pmap.h> +#include <vm/vm_extern.h> +#include <vm/vm_kern.h> +#include <vm/vm_map.h> +#include <vm/vm_object.h> +#include <vm/vm_page.h> +#include <vm/vm_pagequeue.h> +#include <vm/vm_phys.h> +#include <vm/vm_radix.h> + +#include <machine/kexec.h> + +#ifndef KEXEC_MD_PAGES +/* + * Number of MD pages for extra bookkeeping. + * This is a macro because it can be a constant (some architectures make it 0). + * It accepts an argument, which is an array of + * kexec_segment[KEXEC_SEGMENT_MAX]. + */ +#define KEXEC_MD_PAGES(x) 0 +#endif + +/* + * Basic design: + * + * Given an array of "segment descriptors" stage an image to be loaded and + * jumped to at reboot, instead of rebooting via firmware. + * + * Constraints: + * - The segment descriptors' "mem" and "memsz" must each fit within a + * vm_phys_seg segment, which can be obtained via the `vm.phys_segs` sysctl. + * A single segment cannot span multiple vm_phys_seg segments, even if the + * vm_phys_seg segments are adjacent. + * + * Technical details: + * + * Take advantage of the VM subsystem and create a vm_object to hold the staged + * image. When grabbing pages for the object, sort the pages so that if a page + * in the object is located in the physical range of any of the kexec segment + * targets then it gets placed at the pindex corresponding to that physical + * address. This avoids the chance of corruption by writing over the page in + * the final copy, or the need for a copy buffer page. + */ + +static struct kexec_image staged_image; +static vm_offset_t stage_addr; +static vm_object_t kexec_obj; + +static eventhandler_tag kexec_reboot_handler; +static struct mtx kexec_mutex; + +static MALLOC_DEFINE(M_KEXEC, "kexec", "Kexec segments"); + + +static void +kexec_reboot(void *junk __unused, int howto) +{ + if ((howto & RB_KEXEC) == 0 || kexec_obj == NULL) + return; + +#ifdef SMP + cpu_mp_stop(); +#endif /* SMP */ + intr_disable(); + printf("Starting kexec reboot\n"); + + scheduler_stopped = true; + kexec_reboot_md(&staged_image); +} + +MTX_SYSINIT(kexec_mutex, &kexec_mutex, "kexec", MTX_DEF); + +/* Sort the segment list once copied in */ +static int +seg_cmp(const void *seg1, const void *seg2) +{ + const struct kexec_segment *s1, *s2; + + s1 = seg1; + s2 = seg2; + + return ((uintptr_t)s1->mem - (uintptr_t)s2->mem); +} + +static bool +segment_fits(struct kexec_segment *seg) +{ + vm_paddr_t v = (vm_paddr_t)(uintptr_t)seg->mem; + + for (int i = 0; i < vm_phys_nsegs; i++) { + if (v >= vm_phys_segs[i].start && + (v + seg->memsz - 1) <= vm_phys_segs[i].end) + return (true); + } + + return (false); +} + +static vm_paddr_t +pa_for_pindex(struct kexec_segment_stage *segs, int count, vm_pindex_t pind) +{ + for (int i = count; i > 0; --i) { + if (pind >= segs[i - 1].pindex) + return (ptoa(pind - segs[i-1].pindex) + segs[i - 1].target); + } + + panic("No segment for pindex %ju\n", (uintmax_t)pind); +} + +/* + * For now still tied to the system call, so assumes all memory is userspace. + */ +int +kern_kexec_load(struct thread *td, u_long entry, u_long nseg, + struct kexec_segment *seg, u_long flags) +{ + static int kexec_loading; + struct kexec_segment segtmp[KEXEC_SEGMENT_MAX]; + struct kexec_image *new_image_stage = 0; + vm_object_t new_segments = NULL; + uint8_t *buf; + int err = 0; + int i; + const size_t segsize = nseg * sizeof(struct kexec_segment); + vm_page_t *page_list = 0; + vm_size_t image_count, md_pages, page_count, tmpsize; + vm_offset_t segment_va = 0; + /* + * - Do any sanity checking + * - Load the new segments to temporary + * - Remove the old segments + * - Install the new segments + */ + + if (nseg > KEXEC_SEGMENT_MAX) + return (EINVAL); + + if (atomic_cmpset_acq_int(&kexec_loading, false, true) == 0) + return (EBUSY); + + /* Only do error checking if we're installing new segments. */ + if (nseg > 0) { + /* Create the new kexec object before destroying the old one. */ + bzero(&segtmp, sizeof(segtmp)); + err = copyin(seg, segtmp, segsize); + if (err != 0) + goto out; + qsort(segtmp, nseg, sizeof(*segtmp), seg_cmp); + new_image_stage = malloc(sizeof(*new_image_stage), M_TEMP, M_WAITOK | M_ZERO); + /* + * Sanity checking: + * - All segments must not overlap the kernel, so must be fully enclosed + * in a vm_phys_seg (each kexec segment must be in a single + * vm_phys_seg segment, cannot cross even adjacent segments). + */ + image_count = 0; + for (i = 0; i < nseg; i++) { + if (!segment_fits(&segtmp[i]) || + segtmp[i].bufsz > segtmp[i].memsz) { + err = EINVAL; + goto out; + } + new_image_stage->segments[i].pindex = image_count; + new_image_stage->segments[i].target = (vm_offset_t)segtmp[i].mem; + new_image_stage->segments[i].size = segtmp[i].memsz; + image_count += atop(segtmp[i].memsz); + } + md_pages = KEXEC_MD_PAGES(segtmp); + page_count = image_count + md_pages; + new_segments = vm_object_allocate(OBJT_PHYS, page_count); + page_list = malloc(page_count * sizeof(vm_page_t), M_TEMP, M_WAITOK); + + /* + * - Grab all pages for all segments (use pindex to slice it) + * - Walk the list (once) + * - At each pindex, check if the target PA that corresponds + * to that index is in the object. If so, swap the pages. + * - At the end of this the list will be "best" sorted. + */ + vm_page_grab_pages_unlocked(new_segments, 0, + VM_ALLOC_NORMAL | VM_ALLOC_WAITOK | VM_ALLOC_WIRED | VM_ALLOC_NOBUSY | VM_ALLOC_ZERO, + page_list, page_count); + + /* Sort the pages to best match the PA */ + VM_OBJECT_WLOCK(new_segments); + for (i = 0; i < image_count; i++) { + vm_page_t curpg, otherpg, tmp; + vm_pindex_t otheridx; + + curpg = page_list[i]; + otherpg = PHYS_TO_VM_PAGE(pa_for_pindex(new_image_stage->segments, + nseg, curpg->pindex)); + otheridx = otherpg->pindex; + + if (otherpg->object == new_segments) { + /* + * Swap 'curpg' and 'otherpg', since 'otherpg' + * is at the PA 'curpg' covers. + */ + vm_radix_remove(&new_segments->rtree, otheridx); + vm_radix_remove(&new_segments->rtree, i); + otherpg->pindex = i; + curpg->pindex = otheridx; + vm_radix_insert(&new_segments->rtree, curpg); + vm_radix_insert(&new_segments->rtree, otherpg); + tmp = curpg; + page_list[i] = otherpg; + page_list[otheridx] = tmp; + } + } + for (i = 0; i < nseg; i++) { + new_image_stage->segments[i].first_page = + vm_radix_lookup(&new_segments->rtree, + new_image_stage->segments[i].pindex); + } + if (md_pages > 0) + new_image_stage->first_md_page = + vm_radix_lookup(&new_segments->rtree, + page_count - md_pages); + else + new_image_stage->first_md_page = NULL; + VM_OBJECT_WUNLOCK(new_segments); + + /* Map the object to do the copies */ + err = vm_map_find(kernel_map, new_segments, 0, &segment_va, + ptoa(page_count), 0, VMFS_ANY_SPACE, + VM_PROT_RW, VM_PROT_RW, MAP_PREFAULT); + if (err != 0) + goto out; + buf = (void *)segment_va; + new_image_stage->map_addr = segment_va; + new_image_stage->map_size = ptoa(new_segments->size); + new_image_stage->entry = entry; + new_image_stage->map_obj = new_segments; + for (i = 0; i < nseg; i++) { + err = copyin(segtmp[i].buf, buf, segtmp[i].bufsz); + if (err != 0) { + goto out; + } + new_image_stage->segments[i].map_buf = buf; + buf += segtmp[i].bufsz; + tmpsize = segtmp[i].memsz - segtmp[i].bufsz; + if (tmpsize > 0) + memset(buf, 0, tmpsize); + buf += tmpsize; + } + /* What's left are the MD pages, so zero them all out. */ + if (md_pages > 0) + bzero(buf, ptoa(md_pages)); + + cpu_flush_dcache((void *)segment_va, ptoa(page_count)); + if ((err = kexec_load_md(new_image_stage)) != 0) + goto out; + } + if (kexec_obj != NULL) { + vm_object_unwire(kexec_obj, 0, kexec_obj->size, 0); + KASSERT(stage_addr != 0, ("Mapped kexec_obj without address")); + vm_map_remove(kernel_map, stage_addr, stage_addr + kexec_obj->size); + } + kexec_obj = new_segments; + bzero(&staged_image, sizeof(staged_image)); + if (nseg > 0) + memcpy(&staged_image, new_image_stage, sizeof(*new_image_stage)); + + printf("trampoline at %#jx\n", (uintmax_t)staged_image.entry); + if (nseg > 0) { + if (kexec_reboot_handler == NULL) + kexec_reboot_handler = + EVENTHANDLER_REGISTER(shutdown_final, kexec_reboot, NULL, + SHUTDOWN_PRI_DEFAULT - 150); + } else { + if (kexec_reboot_handler != NULL) + EVENTHANDLER_DEREGISTER(shutdown_final, kexec_reboot_handler); + } +out: + /* Clean up the mess if we've gotten far. */ + if (err != 0 && new_segments != NULL) { + vm_object_unwire(new_segments, 0, new_segments->size, 0); + if (segment_va != 0) + vm_map_remove(kernel_map, segment_va, segment_va + kexec_obj->size); + else + vm_object_deallocate(new_segments); + } + atomic_store_rel_int(&kexec_loading, false); + if (new_image_stage != NULL) + free(new_image_stage, M_TEMP); + if (page_list != 0) + free(page_list, M_TEMP); + + return (err); +} + +int +sys_kexec_load(struct thread *td, struct kexec_load_args *uap) +{ + int error; + + // FIXME: Do w need a better privilege check than PRIV_REBOOT here? + error = priv_check(td, PRIV_REBOOT); + if (error != 0) + return (error); + return (kern_kexec_load(td, uap->entry, uap->nseg, uap->segments, uap->flags)); +} diff --git a/sys/kern/kern_linker.c b/sys/kern/kern_linker.c index d566bc01bc5e..e2f63cbc0c5a 100644 --- a/sys/kern/kern_linker.c +++ b/sys/kern/kern_linker.c @@ -435,7 +435,7 @@ linker_file_register_modules(linker_file_t lf) } static void -linker_init_kernel_modules(void) +linker_init_kernel_modules(void *dummy __unused) { sx_xlock(&kld_sx); diff --git a/sys/kern/kern_malloc.c b/sys/kern/kern_malloc.c index 653ce1ee556b..fcbfbe64f854 100644 --- a/sys/kern/kern_malloc.c +++ b/sys/kern/kern_malloc.c @@ -303,7 +303,7 @@ sysctl_vm_malloc_zone_sizes(SYSCTL_HANDLER_ARGS) */ #if MALLOC_DEBUG_MAXZONES > 1 static void -tunable_set_numzones(void) +tunable_set_numzones(void *dummy __unused) { TUNABLE_INT_FETCH("debug.malloc.numzones", @@ -1302,7 +1302,7 @@ mallocinit(void *dummy) #endif align, UMA_ZONE_MALLOC); } - for (;i <= size; i+= KMEM_ZBASE) + for (; i <= size; i+= KMEM_ZBASE) kmemsize[i >> KMEM_ZSHIFT] = indx; } } diff --git a/sys/kern/kern_racct.c b/sys/kern/kern_racct.c index 7351e9cb6313..2aab151aba08 100644 --- a/sys/kern/kern_racct.c +++ b/sys/kern/kern_racct.c @@ -1312,7 +1312,7 @@ static struct kproc_desc racctd_kp = { }; static void -racctd_init(void) +racctd_init(void *dummy __unused) { if (!racct_enable) return; @@ -1322,7 +1322,7 @@ racctd_init(void) SYSINIT(racctd, SI_SUB_RACCTD, SI_ORDER_FIRST, racctd_init, NULL); static void -racct_init(void) +racct_init(void *dummy __unused) { if (!racct_enable) return; diff --git a/sys/kern/kern_rangelock.c b/sys/kern/kern_rangelock.c index 3854ffbeec29..cd66bff62608 100644 --- a/sys/kern/kern_rangelock.c +++ b/sys/kern/kern_rangelock.c @@ -300,7 +300,7 @@ static void rangelock_free_free(struct rl_q_entry *free); static void rangelock_noncheating_destroy(struct rangelock *lock); static void -rangelock_sys_init(void) +rangelock_sys_init(void *dummy __unused) { rl_entry_zone = uma_zcreate("rl_entry", sizeof(struct rl_q_entry), NULL, NULL, NULL, NULL, UMA_ALIGNOF(struct rl_q_entry), diff --git a/sys/kern/kern_rctl.c b/sys/kern/kern_rctl.c index 4232c71f86fb..682ba86d23ff 100644 --- a/sys/kern/kern_rctl.c +++ b/sys/kern/kern_rctl.c @@ -209,7 +209,7 @@ static struct dict actionnames[] = { { "throttle", RCTL_ACTION_THROTTLE }, { NULL, -1 }}; -static void rctl_init(void); +static void rctl_init(void *); SYSINIT(rctl, SI_SUB_RACCT, SI_ORDER_FIRST, rctl_init, NULL); static uma_zone_t rctl_rule_zone; @@ -2175,7 +2175,7 @@ rctl_racct_release(struct racct *racct) } static void -rctl_init(void) +rctl_init(void *dummy __unused) { if (!racct_enable) diff --git a/sys/kern/kern_sharedpage.c b/sys/kern/kern_sharedpage.c index 5b8398caaca9..f48d0e3d616b 100644 --- a/sys/kern/kern_sharedpage.c +++ b/sys/kern/kern_sharedpage.c @@ -130,8 +130,7 @@ shared_page_init(void *dummy __unused) shared_page_mapping = (char *)addr; } -SYSINIT(shp, SI_SUB_EXEC, SI_ORDER_FIRST, (sysinit_cfunc_t)shared_page_init, - NULL); +SYSINIT(shp, SI_SUB_EXEC, SI_ORDER_FIRST, shared_page_init, NULL); /* * Push the timehands update to the shared page. diff --git a/sys/kern/kern_sig.c b/sys/kern/kern_sig.c index 8efc0886988b..a55f3c761449 100644 --- a/sys/kern/kern_sig.c +++ b/sys/kern/kern_sig.c @@ -113,7 +113,7 @@ static int filt_sigattach(struct knote *kn); static void filt_sigdetach(struct knote *kn); static int filt_signal(struct knote *kn, long hint); static struct thread *sigtd(struct proc *p, int sig, bool fast_sigblock); -static void sigqueue_start(void); +static void sigqueue_start(void *); static void sigfastblock_setpend(struct thread *td, bool resched); static void sig_handle_first_stop(struct thread *td, struct proc *p, int sig); @@ -124,6 +124,7 @@ const struct filterops sig_filtops = { .f_attach = filt_sigattach, .f_detach = filt_sigdetach, .f_event = filt_signal, + .f_copy = knote_triv_copy, }; static int kern_forcesigexit = 1; @@ -344,7 +345,7 @@ ast_sigsuspend(struct thread *td, int tda __unused) } static void -sigqueue_start(void) +sigqueue_start(void *dummy __unused) { ksiginfo_zone = uma_zcreate("ksiginfo", sizeof(ksiginfo_t), NULL, NULL, NULL, NULL, UMA_ALIGN_PTR, 0); diff --git a/sys/kern/kern_time.c b/sys/kern/kern_time.c index 2a6f0989f6aa..5b7485c25cd7 100644 --- a/sys/kern/kern_time.c +++ b/sys/kern/kern_time.c @@ -90,7 +90,7 @@ static int user_clock_nanosleep(struct thread *td, clockid_t clock_id, int flags, const struct timespec *ua_rqtp, struct timespec *ua_rmtp); -static void itimer_start(void); +static void itimer_start(void *); static int itimer_init(void *, int, int); static void itimer_fini(void *, int); static void itimer_enter(struct itimer *); @@ -1170,7 +1170,7 @@ eventratecheck(struct timeval *lasttime, int *cureps, int maxeps) } static void -itimer_start(void) +itimer_start(void *dummy __unused) { static const struct kclock rt_clock = { .timer_create = realtimer_create, diff --git a/sys/kern/md4c.c b/sys/kern/md4c.c deleted file mode 100644 index e173e17e3387..000000000000 --- a/sys/kern/md4c.c +++ /dev/null @@ -1,298 +0,0 @@ -/* MD4C.C - RSA Data Security, Inc., MD4 message-digest algorithm - */ - -/*- - SPDX-License-Identifier: RSA-MD - - Copyright (C) 1990-2, RSA Data Security, Inc. All rights reserved. - - License to copy and use this software is granted provided that it - is identified as the "RSA Data Security, Inc. MD4 Message-Digest - Algorithm" in all material mentioning or referencing this software - or this function. - - License is also granted to make and use derivative works provided - that such works are identified as "derived from the RSA Data - Security, Inc. MD4 Message-Digest Algorithm" in all material - mentioning or referencing the derived work. - - RSA Data Security, Inc. makes no representations concerning either - the merchantability of this software or the suitability of this - software for any particular purpose. It is provided "as is" - without express or implied warranty of any kind. - - These notices must be retained in any copies of any part of this - documentation and/or software. - */ - -#include <sys/param.h> -#ifdef _KERNEL -#include <sys/systm.h> -#else -#include <string.h> -#endif -#include <sys/md4.h> - -typedef unsigned char *POINTER; -typedef uint16_t UINT2; -typedef uint32_t UINT4; - -#define PROTO_LIST(list) list - -/* Constants for MD4Transform routine. - */ -#define S11 3 -#define S12 7 -#define S13 11 -#define S14 19 -#define S21 3 -#define S22 5 -#define S23 9 -#define S24 13 -#define S31 3 -#define S32 9 -#define S33 11 -#define S34 15 - -static void MD4Transform PROTO_LIST ((UINT4 [4], const unsigned char [64])); -static void Encode PROTO_LIST - ((unsigned char *, UINT4 *, unsigned int)); -static void Decode PROTO_LIST - ((UINT4 *, const unsigned char *, unsigned int)); - -static unsigned char PADDING[64] = { - 0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 -}; - -/* F, G and H are basic MD4 functions. - */ -#define F(x, y, z) (((x) & (y)) | ((~x) & (z))) -#define G(x, y, z) (((x) & (y)) | ((x) & (z)) | ((y) & (z))) -#define H(x, y, z) ((x) ^ (y) ^ (z)) - -/* ROTATE_LEFT rotates x left n bits. - */ -#define ROTATE_LEFT(x, n) (((x) << (n)) | ((x) >> (32-(n)))) - -/* FF, GG and HH are transformations for rounds 1, 2 and 3 */ -/* Rotation is separate from addition to prevent recomputation */ -#define FF(a, b, c, d, x, s) { \ - (a) += F ((b), (c), (d)) + (x); \ - (a) = ROTATE_LEFT ((a), (s)); \ - } -#define GG(a, b, c, d, x, s) { \ - (a) += G ((b), (c), (d)) + (x) + (UINT4)0x5a827999; \ - (a) = ROTATE_LEFT ((a), (s)); \ - } -#define HH(a, b, c, d, x, s) { \ - (a) += H ((b), (c), (d)) + (x) + (UINT4)0x6ed9eba1; \ - (a) = ROTATE_LEFT ((a), (s)); \ - } - -/* MD4 initialization. Begins an MD4 operation, writing a new context. - */ -void -MD4Init(MD4_CTX *context) -{ - context->count[0] = context->count[1] = 0; - - /* Load magic initialization constants. - */ - context->state[0] = 0x67452301; - context->state[1] = 0xefcdab89; - context->state[2] = 0x98badcfe; - context->state[3] = 0x10325476; -} - -/* MD4 block update operation. Continues an MD4 message-digest - operation, processing another message block, and updating the - context. - */ -void -MD4Update(MD4_CTX *context, const unsigned char *input, - unsigned int inputLen) -{ - unsigned int i, index, partLen; - - /* Compute number of bytes mod 64 */ - index = (unsigned int)((context->count[0] >> 3) & 0x3F); - /* Update number of bits */ - if ((context->count[0] += ((UINT4)inputLen << 3)) - < ((UINT4)inputLen << 3)) - context->count[1]++; - context->count[1] += ((UINT4)inputLen >> 29); - - partLen = 64 - index; - /* Transform as many times as possible. - */ - if (inputLen >= partLen) { - bcopy(input, &context->buffer[index], partLen); - MD4Transform (context->state, context->buffer); - - for (i = partLen; i + 63 < inputLen; i += 64) - MD4Transform (context->state, &input[i]); - - index = 0; - } - else - i = 0; - - /* Buffer remaining input */ - bcopy(&input[i], &context->buffer[index], inputLen-i); -} - -/* MD4 padding. */ -void -MD4Pad(MD4_CTX *context) -{ - unsigned char bits[8]; - unsigned int index, padLen; - - /* Save number of bits */ - Encode (bits, context->count, 8); - - /* Pad out to 56 mod 64. - */ - index = (unsigned int)((context->count[0] >> 3) & 0x3f); - padLen = (index < 56) ? (56 - index) : (120 - index); - MD4Update (context, PADDING, padLen); - - /* Append length (before padding) */ - MD4Update (context, bits, 8); -} - -/* MD4 finalization. Ends an MD4 message-digest operation, writing the - the message digest and zeroizing the context. - */ -void -MD4Final(unsigned char digest[static 16], MD4_CTX *context) -{ - /* Do padding */ - MD4Pad (context); - - /* Store state in digest */ - Encode (digest, context->state, 16); - - /* Zeroize sensitive information. - */ - bzero(context, sizeof (*context)); -} - -/* MD4 basic transformation. Transforms state based on block. - */ -static void -MD4Transform(UINT4 state[4], const unsigned char block[64]) -{ - UINT4 a = state[0], b = state[1], c = state[2], d = state[3], x[16]; - - Decode (x, block, 64); - - /* Round 1 */ - FF (a, b, c, d, x[ 0], S11); /* 1 */ - FF (d, a, b, c, x[ 1], S12); /* 2 */ - FF (c, d, a, b, x[ 2], S13); /* 3 */ - FF (b, c, d, a, x[ 3], S14); /* 4 */ - FF (a, b, c, d, x[ 4], S11); /* 5 */ - FF (d, a, b, c, x[ 5], S12); /* 6 */ - FF (c, d, a, b, x[ 6], S13); /* 7 */ - FF (b, c, d, a, x[ 7], S14); /* 8 */ - FF (a, b, c, d, x[ 8], S11); /* 9 */ - FF (d, a, b, c, x[ 9], S12); /* 10 */ - FF (c, d, a, b, x[10], S13); /* 11 */ - FF (b, c, d, a, x[11], S14); /* 12 */ - FF (a, b, c, d, x[12], S11); /* 13 */ - FF (d, a, b, c, x[13], S12); /* 14 */ - FF (c, d, a, b, x[14], S13); /* 15 */ - FF (b, c, d, a, x[15], S14); /* 16 */ - - /* Round 2 */ - GG (a, b, c, d, x[ 0], S21); /* 17 */ - GG (d, a, b, c, x[ 4], S22); /* 18 */ - GG (c, d, a, b, x[ 8], S23); /* 19 */ - GG (b, c, d, a, x[12], S24); /* 20 */ - GG (a, b, c, d, x[ 1], S21); /* 21 */ - GG (d, a, b, c, x[ 5], S22); /* 22 */ - GG (c, d, a, b, x[ 9], S23); /* 23 */ - GG (b, c, d, a, x[13], S24); /* 24 */ - GG (a, b, c, d, x[ 2], S21); /* 25 */ - GG (d, a, b, c, x[ 6], S22); /* 26 */ - GG (c, d, a, b, x[10], S23); /* 27 */ - GG (b, c, d, a, x[14], S24); /* 28 */ - GG (a, b, c, d, x[ 3], S21); /* 29 */ - GG (d, a, b, c, x[ 7], S22); /* 30 */ - GG (c, d, a, b, x[11], S23); /* 31 */ - GG (b, c, d, a, x[15], S24); /* 32 */ - - /* Round 3 */ - HH (a, b, c, d, x[ 0], S31); /* 33 */ - HH (d, a, b, c, x[ 8], S32); /* 34 */ - HH (c, d, a, b, x[ 4], S33); /* 35 */ - HH (b, c, d, a, x[12], S34); /* 36 */ - HH (a, b, c, d, x[ 2], S31); /* 37 */ - HH (d, a, b, c, x[10], S32); /* 38 */ - HH (c, d, a, b, x[ 6], S33); /* 39 */ - HH (b, c, d, a, x[14], S34); /* 40 */ - HH (a, b, c, d, x[ 1], S31); /* 41 */ - HH (d, a, b, c, x[ 9], S32); /* 42 */ - HH (c, d, a, b, x[ 5], S33); /* 43 */ - HH (b, c, d, a, x[13], S34); /* 44 */ - HH (a, b, c, d, x[ 3], S31); /* 45 */ - HH (d, a, b, c, x[11], S32); /* 46 */ - HH (c, d, a, b, x[ 7], S33); /* 47 */ - HH (b, c, d, a, x[15], S34); /* 48 */ - - state[0] += a; - state[1] += b; - state[2] += c; - state[3] += d; - - /* Zeroize sensitive information. - */ - bzero((POINTER)x, sizeof (x)); -} - -/* Encodes input (UINT4) into output (unsigned char). Assumes len is - a multiple of 4. - */ -static void -Encode(unsigned char *output, UINT4 *input, unsigned int len) -{ - unsigned int i, j; - - for (i = 0, j = 0; j < len; i++, j += 4) { - output[j] = (unsigned char)(input[i] & 0xff); - output[j+1] = (unsigned char)((input[i] >> 8) & 0xff); - output[j+2] = (unsigned char)((input[i] >> 16) & 0xff); - output[j+3] = (unsigned char)((input[i] >> 24) & 0xff); - } -} - -/* Decodes input (unsigned char) into output (UINT4). Assumes len is - a multiple of 4. - */ -static void -Decode(UINT4 *output, const unsigned char *input, unsigned int len) -{ - unsigned int i, j; - - for (i = 0, j = 0; j < len; i++, j += 4) - output[i] = ((UINT4)input[j]) | (((UINT4)input[j+1]) << 8) | - (((UINT4)input[j+2]) << 16) | (((UINT4)input[j+3]) << 24); -} - -#ifdef WEAK_REFS -/* When building libmd, provide weak references. Note: this is not - activated in the context of compiling these sources for internal - use in libcrypt. - */ -#undef MD4Init -__weak_reference(_libmd_MD4Init, MD4Init); -#undef MD4Update -__weak_reference(_libmd_MD4Update, MD4Update); -#undef MD4Pad -__weak_reference(_libmd_MD4Pad, MD4Pad); -#undef MD4Final -__weak_reference(_libmd_MD4Final, MD4Final); -#endif diff --git a/sys/kern/md5c.c b/sys/kern/md5c.c deleted file mode 100644 index 0922d0f8cc61..000000000000 --- a/sys/kern/md5c.c +++ /dev/null @@ -1,341 +0,0 @@ -/*- - * SPDX-License-Identifier: RSA-MD - * - * MD5C.C - RSA Data Security, Inc., MD5 message-digest algorithm - * - * Copyright (C) 1991-2, RSA Data Security, Inc. Created 1991. All - * rights reserved. - * - * License to copy and use this software is granted provided that it - * is identified as the "RSA Data Security, Inc. MD5 Message-Digest - * Algorithm" in all material mentioning or referencing this software - * or this function. - * - * License is also granted to make and use derivative works provided - * that such works are identified as "derived from the RSA Data - * Security, Inc. MD5 Message-Digest Algorithm" in all material - * mentioning or referencing the derived work. - * - * RSA Data Security, Inc. makes no representations concerning either - * the merchantability of this software or the suitability of this - * software for any particular purpose. It is provided "as is" - * without express or implied warranty of any kind. - * - * These notices must be retained in any copies of any part of this - * documentation and/or software. - * - * This code is the same as the code published by RSA Inc. It has been - * edited for clarity and style only. - */ - -#include <sys/types.h> - -#ifdef _KERNEL -#include <sys/systm.h> -#else -#include <string.h> -#endif - -#include <machine/endian.h> -#include <sys/endian.h> -#include <sys/md5.h> - -static void MD5Transform(uint32_t [4], const unsigned char [64]); - -#if (BYTE_ORDER == LITTLE_ENDIAN) -#define Encode memcpy -#define Decode memcpy -#else - -/* - * Encodes input (uint32_t) into output (unsigned char). Assumes len is - * a multiple of 4. - */ - -static void -Encode (unsigned char *output, uint32_t *input, unsigned int len) -{ - unsigned int i; - uint32_t ip; - - for (i = 0; i < len / 4; i++) { - ip = input[i]; - *output++ = ip; - *output++ = ip >> 8; - *output++ = ip >> 16; - *output++ = ip >> 24; - } -} - -/* - * Decodes input (unsigned char) into output (uint32_t). Assumes len is - * a multiple of 4. - */ - -static void -Decode (uint32_t *output, const unsigned char *input, unsigned int len) -{ - unsigned int i; - - for (i = 0; i < len; i += 4) { - *output++ = input[i] | (input[i+1] << 8) | (input[i+2] << 16) | - (input[i+3] << 24); - } -} -#endif - -static unsigned char PADDING[64] = { - 0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 -}; - -/* F, G, H and I are basic MD5 functions. */ -#define F(x, y, z) (((x) & (y)) | ((~x) & (z))) -#define G(x, y, z) (((x) & (z)) | ((y) & (~z))) -#define H(x, y, z) ((x) ^ (y) ^ (z)) -#define I(x, y, z) ((y) ^ ((x) | (~z))) - -/* ROTATE_LEFT rotates x left n bits. */ -#define ROTATE_LEFT(x, n) (((x) << (n)) | ((x) >> (32-(n)))) - -/* - * FF, GG, HH, and II transformations for rounds 1, 2, 3, and 4. - * Rotation is separate from addition to prevent recomputation. - */ -#define FF(a, b, c, d, x, s, ac) { \ - (a) += F ((b), (c), (d)) + (x) + (uint32_t)(ac); \ - (a) = ROTATE_LEFT ((a), (s)); \ - (a) += (b); \ - } -#define GG(a, b, c, d, x, s, ac) { \ - (a) += G ((b), (c), (d)) + (x) + (uint32_t)(ac); \ - (a) = ROTATE_LEFT ((a), (s)); \ - (a) += (b); \ - } -#define HH(a, b, c, d, x, s, ac) { \ - (a) += H ((b), (c), (d)) + (x) + (uint32_t)(ac); \ - (a) = ROTATE_LEFT ((a), (s)); \ - (a) += (b); \ - } -#define II(a, b, c, d, x, s, ac) { \ - (a) += I ((b), (c), (d)) + (x) + (uint32_t)(ac); \ - (a) = ROTATE_LEFT ((a), (s)); \ - (a) += (b); \ - } - -/* MD5 initialization. Begins an MD5 operation, writing a new context. */ - -void -MD5Init(MD5_CTX *context) -{ - - context->count[0] = context->count[1] = 0; - - /* Load magic initialization constants. */ - context->state[0] = 0x67452301; - context->state[1] = 0xefcdab89; - context->state[2] = 0x98badcfe; - context->state[3] = 0x10325476; -} - -/* - * MD5 block update operation. Continues an MD5 message-digest - * operation, processing another message block, and updating the - * context. - */ - -void -MD5Update(MD5_CTX *context, const void *in, unsigned int inputLen) -{ - unsigned int i, index, partLen; - const unsigned char *input = in; - - /* Compute number of bytes mod 64 */ - index = (unsigned int)((context->count[0] >> 3) & 0x3F); - - /* Update number of bits */ - if ((context->count[0] += ((uint32_t)inputLen << 3)) - < ((uint32_t)inputLen << 3)) - context->count[1]++; - context->count[1] += ((uint32_t)inputLen >> 29); - - partLen = 64 - index; - - /* Transform as many times as possible. */ - if (inputLen >= partLen) { - memcpy((void *)&context->buffer[index], (const void *)input, - partLen); - MD5Transform (context->state, context->buffer); - - for (i = partLen; i + 63 < inputLen; i += 64) - MD5Transform (context->state, &input[i]); - - index = 0; - } - else - i = 0; - - /* Buffer remaining input */ - memcpy ((void *)&context->buffer[index], (const void *)&input[i], - inputLen-i); -} - -/* - * MD5 padding. Adds padding followed by original length. - */ - -static void -MD5Pad(MD5_CTX *context) -{ - unsigned char bits[8]; - unsigned int index, padLen; - - /* Save number of bits */ - Encode (bits, context->count, 8); - - /* Pad out to 56 mod 64. */ - index = (unsigned int)((context->count[0] >> 3) & 0x3f); - padLen = (index < 56) ? (56 - index) : (120 - index); - MD5Update (context, PADDING, padLen); - - /* Append length (before padding) */ - MD5Update (context, bits, 8); -} - -/* - * MD5 finalization. Ends an MD5 message-digest operation, writing the - * the message digest and zeroizing the context. - */ - -void -MD5Final(unsigned char digest[static MD5_DIGEST_LENGTH], MD5_CTX *context) -{ - /* Do padding. */ - MD5Pad (context); - - /* Store state in digest */ - Encode (digest, context->state, MD5_DIGEST_LENGTH); - - /* Zeroize sensitive information. */ - explicit_bzero (context, sizeof (*context)); -} - -/* MD5 basic transformation. Transforms state based on block. */ - -static void -MD5Transform(uint32_t state[4], const unsigned char block[64]) -{ - uint32_t a = state[0], b = state[1], c = state[2], d = state[3], x[16]; - - Decode (x, block, 64); - - /* Round 1 */ -#define S11 7 -#define S12 12 -#define S13 17 -#define S14 22 - FF (a, b, c, d, x[ 0], S11, 0xd76aa478); /* 1 */ - FF (d, a, b, c, x[ 1], S12, 0xe8c7b756); /* 2 */ - FF (c, d, a, b, x[ 2], S13, 0x242070db); /* 3 */ - FF (b, c, d, a, x[ 3], S14, 0xc1bdceee); /* 4 */ - FF (a, b, c, d, x[ 4], S11, 0xf57c0faf); /* 5 */ - FF (d, a, b, c, x[ 5], S12, 0x4787c62a); /* 6 */ - FF (c, d, a, b, x[ 6], S13, 0xa8304613); /* 7 */ - FF (b, c, d, a, x[ 7], S14, 0xfd469501); /* 8 */ - FF (a, b, c, d, x[ 8], S11, 0x698098d8); /* 9 */ - FF (d, a, b, c, x[ 9], S12, 0x8b44f7af); /* 10 */ - FF (c, d, a, b, x[10], S13, 0xffff5bb1); /* 11 */ - FF (b, c, d, a, x[11], S14, 0x895cd7be); /* 12 */ - FF (a, b, c, d, x[12], S11, 0x6b901122); /* 13 */ - FF (d, a, b, c, x[13], S12, 0xfd987193); /* 14 */ - FF (c, d, a, b, x[14], S13, 0xa679438e); /* 15 */ - FF (b, c, d, a, x[15], S14, 0x49b40821); /* 16 */ - - /* Round 2 */ -#define S21 5 -#define S22 9 -#define S23 14 -#define S24 20 - GG (a, b, c, d, x[ 1], S21, 0xf61e2562); /* 17 */ - GG (d, a, b, c, x[ 6], S22, 0xc040b340); /* 18 */ - GG (c, d, a, b, x[11], S23, 0x265e5a51); /* 19 */ - GG (b, c, d, a, x[ 0], S24, 0xe9b6c7aa); /* 20 */ - GG (a, b, c, d, x[ 5], S21, 0xd62f105d); /* 21 */ - GG (d, a, b, c, x[10], S22, 0x2441453); /* 22 */ - GG (c, d, a, b, x[15], S23, 0xd8a1e681); /* 23 */ - GG (b, c, d, a, x[ 4], S24, 0xe7d3fbc8); /* 24 */ - GG (a, b, c, d, x[ 9], S21, 0x21e1cde6); /* 25 */ - GG (d, a, b, c, x[14], S22, 0xc33707d6); /* 26 */ - GG (c, d, a, b, x[ 3], S23, 0xf4d50d87); /* 27 */ - GG (b, c, d, a, x[ 8], S24, 0x455a14ed); /* 28 */ - GG (a, b, c, d, x[13], S21, 0xa9e3e905); /* 29 */ - GG (d, a, b, c, x[ 2], S22, 0xfcefa3f8); /* 30 */ - GG (c, d, a, b, x[ 7], S23, 0x676f02d9); /* 31 */ - GG (b, c, d, a, x[12], S24, 0x8d2a4c8a); /* 32 */ - - /* Round 3 */ -#define S31 4 -#define S32 11 -#define S33 16 -#define S34 23 - HH (a, b, c, d, x[ 5], S31, 0xfffa3942); /* 33 */ - HH (d, a, b, c, x[ 8], S32, 0x8771f681); /* 34 */ - HH (c, d, a, b, x[11], S33, 0x6d9d6122); /* 35 */ - HH (b, c, d, a, x[14], S34, 0xfde5380c); /* 36 */ - HH (a, b, c, d, x[ 1], S31, 0xa4beea44); /* 37 */ - HH (d, a, b, c, x[ 4], S32, 0x4bdecfa9); /* 38 */ - HH (c, d, a, b, x[ 7], S33, 0xf6bb4b60); /* 39 */ - HH (b, c, d, a, x[10], S34, 0xbebfbc70); /* 40 */ - HH (a, b, c, d, x[13], S31, 0x289b7ec6); /* 41 */ - HH (d, a, b, c, x[ 0], S32, 0xeaa127fa); /* 42 */ - HH (c, d, a, b, x[ 3], S33, 0xd4ef3085); /* 43 */ - HH (b, c, d, a, x[ 6], S34, 0x4881d05); /* 44 */ - HH (a, b, c, d, x[ 9], S31, 0xd9d4d039); /* 45 */ - HH (d, a, b, c, x[12], S32, 0xe6db99e5); /* 46 */ - HH (c, d, a, b, x[15], S33, 0x1fa27cf8); /* 47 */ - HH (b, c, d, a, x[ 2], S34, 0xc4ac5665); /* 48 */ - - /* Round 4 */ -#define S41 6 -#define S42 10 -#define S43 15 -#define S44 21 - II (a, b, c, d, x[ 0], S41, 0xf4292244); /* 49 */ - II (d, a, b, c, x[ 7], S42, 0x432aff97); /* 50 */ - II (c, d, a, b, x[14], S43, 0xab9423a7); /* 51 */ - II (b, c, d, a, x[ 5], S44, 0xfc93a039); /* 52 */ - II (a, b, c, d, x[12], S41, 0x655b59c3); /* 53 */ - II (d, a, b, c, x[ 3], S42, 0x8f0ccc92); /* 54 */ - II (c, d, a, b, x[10], S43, 0xffeff47d); /* 55 */ - II (b, c, d, a, x[ 1], S44, 0x85845dd1); /* 56 */ - II (a, b, c, d, x[ 8], S41, 0x6fa87e4f); /* 57 */ - II (d, a, b, c, x[15], S42, 0xfe2ce6e0); /* 58 */ - II (c, d, a, b, x[ 6], S43, 0xa3014314); /* 59 */ - II (b, c, d, a, x[13], S44, 0x4e0811a1); /* 60 */ - II (a, b, c, d, x[ 4], S41, 0xf7537e82); /* 61 */ - II (d, a, b, c, x[11], S42, 0xbd3af235); /* 62 */ - II (c, d, a, b, x[ 2], S43, 0x2ad7d2bb); /* 63 */ - II (b, c, d, a, x[ 9], S44, 0xeb86d391); /* 64 */ - - state[0] += a; - state[1] += b; - state[2] += c; - state[3] += d; - - /* Zeroize sensitive information. */ - memset ((void *)x, 0, sizeof (x)); -} - -#ifdef WEAK_REFS -/* When building libmd, provide weak references. Note: this is not - activated in the context of compiling these sources for internal - use in libcrypt. - */ -#undef MD5Init -__weak_reference(_libmd_MD5Init, MD5Init); -#undef MD5Update -__weak_reference(_libmd_MD5Update, MD5Update); -#undef MD5Final -__weak_reference(_libmd_MD5Final, MD5Final); -#endif diff --git a/sys/kern/subr_bus.c b/sys/kern/subr_bus.c index bf5bda7e058d..b84f69cfd03e 100644 --- a/sys/kern/subr_bus.c +++ b/sys/kern/subr_bus.c @@ -4633,7 +4633,7 @@ bus_release_resources(device_t dev, const struct resource_spec *rs, * parent of @p dev. */ struct resource * -bus_alloc_resource(device_t dev, int type, int *rid, rman_res_t start, +(bus_alloc_resource)(device_t dev, int type, int *rid, rman_res_t start, rman_res_t end, rman_res_t count, u_int flags) { struct resource *res; diff --git a/sys/kern/subr_devstat.c b/sys/kern/subr_devstat.c index 07a9cc0f57be..c4d0223d484f 100644 --- a/sys/kern/subr_devstat.c +++ b/sys/kern/subr_devstat.c @@ -415,7 +415,7 @@ sysctl_devstat(SYSCTL_HANDLER_ARGS) if (error != 0) return (error); - for (;nds != NULL;) { + while (nds != NULL) { error = SYSCTL_OUT(req, nds, sizeof(struct devstat)); if (error != 0) return (error); diff --git a/sys/kern/subr_kdb.c b/sys/kern/subr_kdb.c index 56264a96c9fa..909dd10a6e69 100644 --- a/sys/kern/subr_kdb.c +++ b/sys/kern/subr_kdb.c @@ -330,7 +330,7 @@ kdb_reboot(void) #define KEY_CRTLP 16 /* ^P */ #define KEY_CRTLR 18 /* ^R */ -/* States of th KDB "alternate break sequence" detecting state machine. */ +/* States of the KDB "alternate break sequence" detecting state machine. */ enum { KDB_ALT_BREAK_SEEN_NONE, KDB_ALT_BREAK_SEEN_CR, diff --git a/sys/kern/subr_log.c b/sys/kern/subr_log.c index 5380902e602f..aac35a56130e 100644 --- a/sys/kern/subr_log.c +++ b/sys/kern/subr_log.c @@ -79,6 +79,7 @@ static const struct filterops log_read_filterops = { .f_attach = NULL, .f_detach = logkqdetach, .f_event = logkqread, + .f_copy = knote_triv_copy, }; static struct logsoftc { diff --git a/sys/kern/subr_pcpu.c b/sys/kern/subr_pcpu.c index 5c14e15830f4..c9a387a5e87b 100644 --- a/sys/kern/subr_pcpu.c +++ b/sys/kern/subr_pcpu.c @@ -140,7 +140,7 @@ uma_zone_t pcpu_zone_32; uma_zone_t pcpu_zone_64; static void -pcpu_zones_startup(void) +pcpu_zones_startup(void *dummy __unused) { pcpu_zone_4 = uma_zcreate("pcpu-4", 4, diff --git a/sys/kern/subr_prf.c b/sys/kern/subr_prf.c index db0ceb17b9f0..e2070ae3f865 100644 --- a/sys/kern/subr_prf.c +++ b/sys/kern/subr_prf.c @@ -766,7 +766,7 @@ reswitch: switch (ch = (u_char)*fmt++) { PCHAR(hex2ascii(*up & 0x0f)); up++; if (width) - for (q=p;*q;q++) + for (q = p; *q; q++) PCHAR(*q); } break; diff --git a/sys/kern/subr_smp.c b/sys/kern/subr_smp.c index 1f9577fddf9c..9f5106316018 100644 --- a/sys/kern/subr_smp.c +++ b/sys/kern/subr_smp.c @@ -242,7 +242,7 @@ generic_stop_cpus(cpuset_t map, u_int type) KASSERT( type == IPI_STOP || type == IPI_STOP_HARD #if X86 - || type == IPI_SUSPEND + || type == IPI_SUSPEND || type == IPI_OFF #endif , ("%s: invalid stop type", __func__)); @@ -260,7 +260,7 @@ generic_stop_cpus(cpuset_t map, u_int type) * will be lost, violating FreeBSD's assumption of reliable * IPI delivery. */ - if (type == IPI_SUSPEND) + if (type == IPI_SUSPEND || type == IPI_OFF) mtx_lock_spin(&smp_ipi_mtx); #endif @@ -280,7 +280,7 @@ generic_stop_cpus(cpuset_t map, u_int type) #endif #if X86 - if (type == IPI_SUSPEND) + if (type == IPI_SUSPEND || type == IPI_OFF) cpus = &suspended_cpus; else #endif @@ -298,7 +298,7 @@ generic_stop_cpus(cpuset_t map, u_int type) } #if X86 - if (type == IPI_SUSPEND) + if (type == IPI_SUSPEND || type == IPI_OFF) mtx_unlock_spin(&smp_ipi_mtx); #endif @@ -327,6 +327,13 @@ suspend_cpus(cpuset_t map) return (generic_stop_cpus(map, IPI_SUSPEND)); } + +int +offline_cpus(cpuset_t map) +{ + + return (generic_stop_cpus(map, IPI_OFF)); +} #endif /* diff --git a/sys/kern/sys_eventfd.c b/sys/kern/sys_eventfd.c index c2a0f67cae85..04ed107c933d 100644 --- a/sys/kern/sys_eventfd.c +++ b/sys/kern/sys_eventfd.c @@ -85,13 +85,16 @@ static int filt_eventfdwrite(struct knote *kn, long hint); static const struct filterops eventfd_rfiltops = { .f_isfd = 1, .f_detach = filt_eventfddetach, - .f_event = filt_eventfdread + .f_event = filt_eventfdread, + .f_copy = knote_triv_copy, }; + static const struct filterops eventfd_wfiltops = { .f_isfd = 1, .f_detach = filt_eventfddetach, - .f_event = filt_eventfdwrite + .f_event = filt_eventfdwrite, + .f_copy = knote_triv_copy, }; struct eventfd { diff --git a/sys/kern/sys_pipe.c b/sys/kern/sys_pipe.c index 57ebe8dc85f0..6531cea31423 100644 --- a/sys/kern/sys_pipe.c +++ b/sys/kern/sys_pipe.c @@ -181,20 +181,23 @@ static int filt_pipedump(struct proc *p, struct knote *kn, static const struct filterops pipe_nfiltops = { .f_isfd = 1, .f_detach = filt_pipedetach_notsup, - .f_event = filt_pipenotsup + .f_event = filt_pipenotsup, /* no userdump */ + .f_copy = knote_triv_copy, }; static const struct filterops pipe_rfiltops = { .f_isfd = 1, .f_detach = filt_pipedetach, .f_event = filt_piperead, .f_userdump = filt_pipedump, + .f_copy = knote_triv_copy, }; static const struct filterops pipe_wfiltops = { .f_isfd = 1, .f_detach = filt_pipedetach, .f_event = filt_pipewrite, .f_userdump = filt_pipedump, + .f_copy = knote_triv_copy, }; /* diff --git a/sys/kern/sys_procdesc.c b/sys/kern/sys_procdesc.c index acaf1241cb2e..c5db21544b0f 100644 --- a/sys/kern/sys_procdesc.c +++ b/sys/kern/sys_procdesc.c @@ -486,6 +486,7 @@ static const struct filterops procdesc_kqops = { .f_isfd = 1, .f_detach = procdesc_kqops_detach, .f_event = procdesc_kqops_event, + .f_copy = knote_triv_copy, }; static int diff --git a/sys/kern/sys_socket.c b/sys/kern/sys_socket.c index c221106ae067..bc0725230cca 100644 --- a/sys/kern/sys_socket.c +++ b/sys/kern/sys_socket.c @@ -586,7 +586,7 @@ soaio_enqueue(struct task *task) } static void -soaio_init(void) +soaio_init(void *dummy __unused) { soaio_lifetime = AIOD_LIFETIME_DEFAULT; diff --git a/sys/kern/syscalls.c b/sys/kern/syscalls.c index 4cef89cd5219..06a4adc3d8cb 100644 --- a/sys/kern/syscalls.c +++ b/sys/kern/syscalls.c @@ -604,4 +604,5 @@ const char *syscallnames[] = { "setgroups", /* 596 = setgroups */ "jail_attach_jd", /* 597 = jail_attach_jd */ "jail_remove_jd", /* 598 = jail_remove_jd */ + "kexec_load", /* 599 = kexec_load */ }; diff --git a/sys/kern/syscalls.master b/sys/kern/syscalls.master index 967af1f5313c..ea6d2b5aa1ef 100644 --- a/sys/kern/syscalls.master +++ b/sys/kern/syscalls.master @@ -3394,4 +3394,12 @@ ); } +599 AUE_NULL STD { + int kexec_load( + uint64_t entry, + u_long nseg, + _In_reads_(nseg) _Contains_long_ptr_ struct kexec_segment *segments, + u_long flags + ); + } ; vim: syntax=off diff --git a/sys/kern/systrace_args.c b/sys/kern/systrace_args.c index e28fef931ea8..5951cebbe74a 100644 --- a/sys/kern/systrace_args.c +++ b/sys/kern/systrace_args.c @@ -3514,6 +3514,16 @@ systrace_args(int sysnum, void *params, uint64_t *uarg, int *n_args) *n_args = 1; break; } + /* kexec_load */ + case 599: { + struct kexec_load_args *p = params; + uarg[a++] = p->entry; /* uint64_t */ + uarg[a++] = p->nseg; /* u_long */ + uarg[a++] = (intptr_t)p->segments; /* struct kexec_segment * */ + uarg[a++] = p->flags; /* u_long */ + *n_args = 4; + break; + } default: *n_args = 0; break; @@ -9401,6 +9411,25 @@ systrace_entry_setargdesc(int sysnum, int ndx, char *desc, size_t descsz) break; }; break; + /* kexec_load */ + case 599: + switch (ndx) { + case 0: + p = "uint64_t"; + break; + case 1: + p = "u_long"; + break; + case 2: + p = "userland struct kexec_segment *"; + break; + case 3: + p = "u_long"; + break; + default: + break; + }; + break; default: break; }; @@ -11409,6 +11438,11 @@ systrace_return_setargdesc(int sysnum, int ndx, char *desc, size_t descsz) if (ndx == 0 || ndx == 1) p = "int"; break; + /* kexec_load */ + case 599: + if (ndx == 0 || ndx == 1) + p = "int"; + break; default: break; }; diff --git a/sys/kern/tty.c b/sys/kern/tty.c index c8e2c561b7cf..067471eb949a 100644 --- a/sys/kern/tty.c +++ b/sys/kern/tty.c @@ -754,12 +754,14 @@ static const struct filterops tty_kqops_read = { .f_isfd = 1, .f_detach = tty_kqops_read_detach, .f_event = tty_kqops_read_event, + .f_copy = knote_triv_copy, }; static const struct filterops tty_kqops_write = { .f_isfd = 1, .f_detach = tty_kqops_write_detach, .f_event = tty_kqops_write_event, + .f_copy = knote_triv_copy, }; static int diff --git a/sys/kern/tty_pts.c b/sys/kern/tty_pts.c index 1291770a9ccb..2672935c2d89 100644 --- a/sys/kern/tty_pts.c +++ b/sys/kern/tty_pts.c @@ -491,11 +491,13 @@ static const struct filterops pts_kqops_read = { .f_isfd = 1, .f_detach = pts_kqops_read_detach, .f_event = pts_kqops_read_event, + .f_copy = knote_triv_copy, }; static const struct filterops pts_kqops_write = { .f_isfd = 1, .f_detach = pts_kqops_write_detach, .f_event = pts_kqops_write_event, + .f_copy = knote_triv_copy, }; static int diff --git a/sys/kern/uipc_mqueue.c b/sys/kern/uipc_mqueue.c index a8aec397b352..4c1bb1ff228e 100644 --- a/sys/kern/uipc_mqueue.c +++ b/sys/kern/uipc_mqueue.c @@ -281,11 +281,13 @@ static const struct filterops mq_rfiltops = { .f_isfd = 1, .f_detach = filt_mqdetach, .f_event = filt_mqread, + .f_copy = knote_triv_copy, }; static const struct filterops mq_wfiltops = { .f_isfd = 1, .f_detach = filt_mqdetach, .f_event = filt_mqwrite, + .f_copy = knote_triv_copy, }; /* diff --git a/sys/kern/uipc_socket.c b/sys/kern/uipc_socket.c index fe2d8d056062..eb9544628137 100644 --- a/sys/kern/uipc_socket.c +++ b/sys/kern/uipc_socket.c @@ -191,16 +191,19 @@ static const struct filterops soread_filtops = { .f_isfd = 1, .f_detach = filt_sordetach, .f_event = filt_soread, + .f_copy = knote_triv_copy, }; static const struct filterops sowrite_filtops = { .f_isfd = 1, .f_detach = filt_sowdetach, .f_event = filt_sowrite, + .f_copy = knote_triv_copy, }; static const struct filterops soempty_filtops = { .f_isfd = 1, .f_detach = filt_sowdetach, .f_event = filt_soempty, + .f_copy = knote_triv_copy, }; so_gen_t so_gencnt; /* generation count for sockets */ diff --git a/sys/kern/uipc_usrreq.c b/sys/kern/uipc_usrreq.c index 340d84666459..807271488af2 100644 --- a/sys/kern/uipc_usrreq.c +++ b/sys/kern/uipc_usrreq.c @@ -1069,6 +1069,21 @@ uipc_stream_sbspace(struct sockbuf *sb) return (min(space, mbspace)); } +/* + * UNIX version of generic sbwait() for writes. We wait on peer's receive + * buffer, using our timeout. + */ +static int +uipc_stream_sbwait(struct socket *so, sbintime_t timeo) +{ + struct sockbuf *sb = &so->so_rcv; + + SOCK_RECVBUF_LOCK_ASSERT(so); + sb->sb_flags |= SB_WAIT; + return (msleep_sbt(&sb->sb_acc, SOCK_RECVBUF_MTX(so), PSOCK | PCATCH, + "sbwait", timeo, 0, 0)); +} + static int uipc_sosend_stream_or_seqpacket(struct socket *so, struct sockaddr *addr, struct uio *uio0, struct mbuf *m, struct mbuf *c, int flags, @@ -1203,7 +1218,8 @@ restart: error = EWOULDBLOCK; goto out4; } - if ((error = sbwait(so2, SO_RCV)) != 0) { + if ((error = uipc_stream_sbwait(so2, + so->so_snd.sb_timeo)) != 0) { SOCK_RECVBUF_UNLOCK(so2); goto out4; } else @@ -1543,15 +1559,19 @@ restart: mc_init_m(&cmc, control); SOCK_RECVBUF_LOCK(so); - MPASS(!(sb->sb_state & SBS_CANTRCVMORE)); - - if (__predict_false(cmc.mc_len + sb->sb_ccc + - sb->sb_ctl > sb->sb_hiwat)) { + if (__predict_false( + (sb->sb_state & SBS_CANTRCVMORE) || + cmc.mc_len + sb->sb_ccc + sb->sb_ctl > + sb->sb_hiwat)) { /* - * Too bad, while unp_externalize() was - * failing, the other side had filled - * the buffer and we can't prepend data - * back. Losing data! + * While the lock was dropped and we + * were failing in unp_externalize(), + * the peer could has a) disconnected, + * b) filled the buffer so that we + * can't prepend data back. + * These are two edge conditions that + * we just can't handle, so lose the + * data and return the error. */ SOCK_RECVBUF_UNLOCK(so); SOCK_IO_RECV_UNLOCK(so); @@ -1835,11 +1855,13 @@ static const struct filterops uipc_write_filtops = { .f_isfd = 1, .f_detach = uipc_filt_sowdetach, .f_event = uipc_filt_sowrite, + .f_copy = knote_triv_copy, }; static const struct filterops uipc_empty_filtops = { .f_isfd = 1, .f_detach = uipc_filt_sowdetach, .f_event = uipc_filt_soempty, + .f_copy = knote_triv_copy, }; static int @@ -2397,7 +2419,7 @@ uipc_sendfile_wait(struct socket *so, off_t need, int *space) } if (!sockref) soref(so2); - error = sbwait(so2, SO_RCV); + error = uipc_stream_sbwait(so2, so->so_snd.sb_timeo); if (error == 0 && __predict_false(sb->sb_state & SBS_CANTRCVMORE)) error = EPIPE; diff --git a/sys/kern/vfs_aio.c b/sys/kern/vfs_aio.c index e63fa4c01434..60916a9fbd32 100644 --- a/sys/kern/vfs_aio.c +++ b/sys/kern/vfs_aio.c @@ -345,12 +345,14 @@ static const struct filterops aio_filtops = { .f_attach = filt_aioattach, .f_detach = filt_aiodetach, .f_event = filt_aio, + .f_copy = knote_triv_copy, }; static const struct filterops lio_filtops = { .f_isfd = 0, .f_attach = filt_lioattach, .f_detach = filt_liodetach, - .f_event = filt_lio + .f_event = filt_lio, + .f_copy = knote_triv_copy, }; static eventhandler_tag exit_tag, exec_tag; diff --git a/sys/kern/vfs_cluster.c b/sys/kern/vfs_cluster.c index 2e397b8e9e8f..b674313993c4 100644 --- a/sys/kern/vfs_cluster.c +++ b/sys/kern/vfs_cluster.c @@ -260,8 +260,10 @@ cluster_read(struct vnode *vp, u_quad_t filesize, daddr_t lblkno, long size, */ while (lblkno < (origblkno + maxra)) { error = VOP_BMAP(vp, lblkno, NULL, &blkno, &ncontig, NULL); - if (error) + if (error) { + error = 0; break; + } if (blkno == -1) break; diff --git a/sys/kern/vfs_default.c b/sys/kern/vfs_default.c index 05d1120030f3..4eca09aef145 100644 --- a/sys/kern/vfs_default.c +++ b/sys/kern/vfs_default.c @@ -458,6 +458,7 @@ vop_stdpathconf(struct vop_pathconf_args *ap) case _PC_HAS_NAMEDATTR: case _PC_HAS_HIDDENSYSTEM: case _PC_CLONE_BLKSIZE: + case _PC_CASE_INSENSITIVE: *ap->a_retval = 0; return (0); default: diff --git a/sys/kern/vfs_inotify.c b/sys/kern/vfs_inotify.c index b265a5ff3a62..e60d8426ee42 100644 --- a/sys/kern/vfs_inotify.c +++ b/sys/kern/vfs_inotify.c @@ -111,6 +111,7 @@ static const struct filterops inotify_rfiltops = { .f_isfd = 1, .f_detach = filt_inotifydetach, .f_event = filt_inotifyevent, + .f_copy = knote_triv_copy, }; static MALLOC_DEFINE(M_INOTIFY, "inotify", "inotify data structures"); diff --git a/sys/kern/vfs_subr.c b/sys/kern/vfs_subr.c index 73e110c05bc1..58975f7ac932 100644 --- a/sys/kern/vfs_subr.c +++ b/sys/kern/vfs_subr.c @@ -6545,6 +6545,7 @@ const struct filterops fs_filtops = { .f_attach = filt_fsattach, .f_detach = filt_fsdetach, .f_event = filt_fsevent, + .f_copy = knote_triv_copy, }; static int @@ -6624,24 +6625,28 @@ static int filt_vfsvnode(struct knote *kn, long hint); static void filt_vfsdetach(struct knote *kn); static int filt_vfsdump(struct proc *p, struct knote *kn, struct kinfo_knote *kin); +static int filt_vfscopy(struct knote *kn, struct proc *p1); static const struct filterops vfsread_filtops = { .f_isfd = 1, .f_detach = filt_vfsdetach, .f_event = filt_vfsread, .f_userdump = filt_vfsdump, + .f_copy = filt_vfscopy, }; static const struct filterops vfswrite_filtops = { .f_isfd = 1, .f_detach = filt_vfsdetach, .f_event = filt_vfswrite, .f_userdump = filt_vfsdump, + .f_copy = filt_vfscopy, }; static const struct filterops vfsvnode_filtops = { .f_isfd = 1, .f_detach = filt_vfsdetach, .f_event = filt_vfsvnode, .f_userdump = filt_vfsdump, + .f_copy = filt_vfscopy, }; static void @@ -6825,6 +6830,16 @@ filt_vfsdump(struct proc *p, struct knote *kn, struct kinfo_knote *kin) return (0); } +static int +filt_vfscopy(struct knote *kn, struct proc *p1) +{ + struct vnode *vp; + + vp = (struct vnode *)kn->kn_hook; + vhold(vp); + return (0); +} + int vfs_read_dirent(struct vop_readdir_args *ap, struct dirent *dp, off_t off) { diff --git a/sys/kern/vfs_syscalls.c b/sys/kern/vfs_syscalls.c index 9e1275359715..1a739d354f1f 100644 --- a/sys/kern/vfs_syscalls.c +++ b/sys/kern/vfs_syscalls.c @@ -1119,7 +1119,7 @@ flags_to_rights(int flags, cap_rights_t *rightsp) if (flags & O_TRUNC) cap_rights_set_one(rightsp, CAP_FTRUNCATE); - if (flags & (O_SYNC | O_FSYNC)) + if (flags & (O_SYNC | O_FSYNC | O_DSYNC)) cap_rights_set_one(rightsp, CAP_FSYNC); if (flags & (O_EXLOCK | O_SHLOCK)) |