/*      $NetBSD: subr_fault.c,v 1.2 2020/06/30 16:28:17 maxv Exp $      */

/*
* Copyright (c) 2020 The NetBSD Foundation, Inc.
* All rights reserved.
*
* This code is derived from software contributed to The NetBSD Foundation
* by Maxime Villard.
*
* 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 NETBSD FOUNDATION, INC. 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 FOUNDATION 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/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: subr_fault.c,v 1.2 2020/06/30 16:28:17 maxv Exp $");

#include <sys/module.h>
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>

#include <sys/conf.h>
#include <sys/types.h>
#include <sys/specificdata.h>
#include <sys/kmem.h>
#include <sys/atomic.h>
#include <sys/ioccom.h>
#include <sys/lwp.h>
#include <sys/fault.h>

typedef struct {
       volatile bool enabled;
       volatile bool oneshot;
       volatile unsigned long nth;
       volatile unsigned long cnt;
       volatile unsigned long nfaults;
} fault_t;

static fault_t fault_global __cacheline_aligned = {
       .enabled = false,
       .oneshot = false,
       .nth = FAULT_NTH_MIN,
       .cnt = 0,
       .nfaults = 0
};

static kmutex_t fault_global_lock __cacheline_aligned;
static specificdata_key_t fault_lwp_key;

/* -------------------------------------------------------------------------- */

bool
fault_inject(void)
{
       volatile unsigned long cnt;
       fault_t *f;

       if (__predict_false(cold))
               return false;

       if (__predict_false(atomic_load_acquire(&fault_global.enabled))) {
               f = &fault_global;
       } else {
               f = lwp_getspecific(fault_lwp_key);
               if (__predict_true(f == NULL))
                       return false;
               if (__predict_false(!f->enabled))
                       return false;
       }

       if (atomic_load_relaxed(&f->oneshot)) {
               if (__predict_true(atomic_load_relaxed(&f->nfaults) > 0))
                       return false;
       }

       cnt = atomic_inc_ulong_nv(&f->cnt);
       if (__predict_false(cnt % atomic_load_relaxed(&f->nth) == 0)) {
               atomic_inc_ulong(&f->nfaults);
               return true;
       }

       return false;
}

/* -------------------------------------------------------------------------- */

static int
fault_open(dev_t dev, int flag, int mode, struct lwp *l)
{
       return 0;
}

static int
fault_close(dev_t dev, int flag, int mode, struct lwp *l)
{
       return 0;
}

static int
fault_ioc_enable(struct fault_ioc_enable *args)
{
       fault_t *f;

       if (args->mode != FAULT_MODE_NTH_ONESHOT)
               return EINVAL;
       if (args->nth < FAULT_NTH_MIN)
               return EINVAL;

       switch (args->scope) {
       case FAULT_SCOPE_GLOBAL:
               mutex_enter(&fault_global_lock);
               if (fault_global.enabled) {
                       mutex_exit(&fault_global_lock);
                       return EEXIST;
               }
               fault_global.oneshot = true;
               atomic_store_relaxed(&fault_global.nth, args->nth);
               fault_global.cnt = 0;
               fault_global.nfaults = 0;
               atomic_store_release(&fault_global.enabled, true);
               mutex_exit(&fault_global_lock);
               break;
       case FAULT_SCOPE_LWP:
               f = lwp_getspecific(fault_lwp_key);
               if (f != NULL) {
                       if (f->enabled)
                               return EEXIST;
               } else {
                       f = kmem_zalloc(sizeof(*f), KM_SLEEP);
                       lwp_setspecific(fault_lwp_key, f);
               }
               f->oneshot = true;
               atomic_store_relaxed(&f->nth, args->nth);
               f->cnt = 0;
               f->nfaults = 0;
               atomic_store_release(&f->enabled, true);
               break;
       default:
               return EINVAL;
       }

       return 0;
}

static int
fault_ioc_disable(struct fault_ioc_disable *args)
{
       fault_t *f;

       switch (args->scope) {
       case FAULT_SCOPE_GLOBAL:
               mutex_enter(&fault_global_lock);
               if (!fault_global.enabled) {
                       mutex_exit(&fault_global_lock);
                       return ENOENT;
               }
               atomic_store_release(&fault_global.enabled, false);
               mutex_exit(&fault_global_lock);
               break;
       case FAULT_SCOPE_LWP:
               f = lwp_getspecific(fault_lwp_key);
               if (f == NULL)
                       return ENOENT;
               if (!f->enabled)
                       return ENOENT;
               atomic_store_release(&f->enabled, false);
               break;
       default:
               return EINVAL;
       }

       return 0;
}

static int
fault_ioc_getinfo(struct fault_ioc_getinfo *args)
{
       fault_t *f;

       switch (args->scope) {
       case FAULT_SCOPE_GLOBAL:
               args->nfaults = atomic_load_relaxed(&fault_global.nfaults);
               break;
       case FAULT_SCOPE_LWP:
               f = lwp_getspecific(fault_lwp_key);
               if (f == NULL)
                       return ENOENT;
               args->nfaults = atomic_load_relaxed(&f->nfaults);
               break;
       default:
               return EINVAL;
       }

       return 0;
}

static int
fault_ioctl(dev_t dev, u_long cmd, void *addr, int flag, struct lwp *l)
{
       switch (cmd) {
       case FAULT_IOC_ENABLE:
               return fault_ioc_enable(addr);
       case FAULT_IOC_DISABLE:
               return fault_ioc_disable(addr);
       case FAULT_IOC_GETINFO:
               return fault_ioc_getinfo(addr);
       default:
               return EINVAL;
       }
}

const struct cdevsw fault_cdevsw = {
       .d_open = fault_open,
       .d_close = fault_close,
       .d_read = noread,
       .d_write = nowrite,
       .d_ioctl = fault_ioctl,
       .d_stop = nostop,
       .d_tty = notty,
       .d_poll = nopoll,
       .d_mmap = nommap,
       .d_kqfilter = nokqfilter,
       .d_discard = nodiscard,
       .d_flag = D_OTHER | D_MPSAFE
};

/* -------------------------------------------------------------------------- */

MODULE(MODULE_CLASS_MISC, fault, NULL);

static void
fault_lwp_free(void *arg)
{
       fault_t *f = (fault_t *)arg;

       if (f == NULL) {
               return;
       }

       kmem_free(f, sizeof(*f));
}

static void
fault_init(void)
{
       mutex_init(&fault_global_lock, MUTEX_DEFAULT, IPL_NONE);
       lwp_specific_key_create(&fault_lwp_key, fault_lwp_free);
}

static int
fault_modcmd(modcmd_t cmd, void *arg)
{
       switch (cmd) {
       case MODULE_CMD_INIT:
               fault_init();
               return 0;
       case MODULE_CMD_FINI:
               return EINVAL;
       default:
               return ENOTTY;
       }
}