4.9.52-stable review

@@ -702,59 +702,32 @@ found_pool:
 	return task;
 }
 
-/*
- * Create or destroy enough new threads to make the number
- * of threads the given number.  If `pool' is non-NULL, applies
- * only to threads in that pool, otherwise round-robins between
- * all pools.  Caller must ensure that mutual exclusion between this and
- * server startup or shutdown.
- *
- * Destroying threads relies on the service threads filling in
- * rqstp->rq_task, which only the nfs ones do.  Assumes the serv
- * has been created using svc_create_pooled().
- *
- * Based on code that used to be in nfsd_svc() but tweaked
- * to be pool-aware.
- */
-int
-svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
+/* create new threads */
+static int
+svc_start_kthreads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
 {
 	struct svc_rqst	*rqstp;
 	struct task_struct *task;
 	struct svc_pool *chosen_pool;
-	int error = 0;
 	unsigned int state = serv->sv_nrthreads-1;
 	int node;
 
-	if (pool == NULL) {
-		/* The -1 assumes caller has done a svc_get() */
-		nrservs -= (serv->sv_nrthreads-1);
-	} else {
-		spin_lock_bh(&pool->sp_lock);
-		nrservs -= pool->sp_nrthreads;
-		spin_unlock_bh(&pool->sp_lock);
-	}
-
-	/* create new threads */
-	while (nrservs > 0) {
+	do {
 		nrservs--;
 		chosen_pool = choose_pool(serv, pool, &state);
 
 		node = svc_pool_map_get_node(chosen_pool->sp_id);
 		rqstp = svc_prepare_thread(serv, chosen_pool, node);
-		if (IS_ERR(rqstp)) {
-			error = PTR_ERR(rqstp);
-			break;
-		}
+		if (IS_ERR(rqstp))
+			return PTR_ERR(rqstp);
 
 		__module_get(serv->sv_ops->svo_module);
 		task = kthread_create_on_node(serv->sv_ops->svo_function, rqstp,
 					      node, "%s", serv->sv_name);
 		if (IS_ERR(task)) {
-			error = PTR_ERR(task);
 			module_put(serv->sv_ops->svo_module);
 			svc_exit_thread(rqstp);
-			break;
+			return PTR_ERR(task);
 		}
 
 		rqstp->rq_task = task;

@@ -763,15 +736,62 @@ svc_set_num_threads(struct svc_serv *ser
 
 		svc_sock_update_bufs(serv);
 		wake_up_process(task);
-	}
+	} while (nrservs > 0);
+
+	return 0;
+}
+
+
+/* destroy old threads */
+static int
+svc_signal_kthreads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
+{
+	struct task_struct *task;
+	unsigned int state = serv->sv_nrthreads-1;
+
 	/* destroy old threads */
-	while (nrservs < 0 &&
-	       (task = choose_victim(serv, pool, &state)) != NULL) {
+	do {
+		task = choose_victim(serv, pool, &state);
+		if (task == NULL)
+			break;
 		send_sig(SIGINT, task, 1);
 		nrservs++;
+	} while (nrservs < 0);
+
+	return 0;
+}
+
+/*
+ * Create or destroy enough new threads to make the number
+ * of threads the given number.  If `pool' is non-NULL, applies
+ * only to threads in that pool, otherwise round-robins between
+ * all pools.  Caller must ensure that mutual exclusion between this and
+ * server startup or shutdown.
+ *
+ * Destroying threads relies on the service threads filling in
+ * rqstp->rq_task, which only the nfs ones do.  Assumes the serv
+ * has been created using svc_create_pooled().
+ *
+ * Based on code that used to be in nfsd_svc() but tweaked
+ * to be pool-aware.
+ */
+int
+svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
+{
+	if (pool == NULL) {
+		/* The -1 assumes caller has done a svc_get() */
+		nrservs -= (serv->sv_nrthreads-1);
+	} else {
+		spin_lock_bh(&pool->sp_lock);
+		nrservs -= pool->sp_nrthreads;
+		spin_unlock_bh(&pool->sp_lock);
 	}
 
-	return error;
+	if (nrservs > 0)
+		return svc_start_kthreads(serv, pool, nrservs);
+	if (nrservs < 0)
+		return svc_signal_kthreads(serv, pool, nrservs);
+	return 0;
 }
 EXPORT_SYMBOL_GPL(svc_set_num_threads);
 

@@ -75,7 +75,10 @@ nfs4_callback_svc(void *vrqstp)
 
 	set_freezable();
 
-	while (!kthread_should_stop()) {
+	while (!kthread_freezable_should_stop(NULL)) {
+
+		if (signal_pending(current))
+			flush_signals(current);
 		/*
 		 * Listen for a request on the socket
 		 */

@@ -84,6 +87,8 @@ nfs4_callback_svc(void *vrqstp)
 			continue;
 		svc_process(rqstp);
 	}
+	svc_exit_thread(rqstp);
+	module_put_and_exit(0);
 	return 0;
 }
 

@@ -102,9 +107,10 @@ nfs41_callback_svc(void *vrqstp)
 
 	set_freezable();
 
-	while (!kthread_should_stop()) {
-		if (try_to_freeze())
-			continue;
+	while (!kthread_freezable_should_stop(NULL)) {
+
+		if (signal_pending(current))
+			flush_signals(current);
 
 		prepare_to_wait(&serv->sv_cb_waitq, &wq, TASK_INTERRUPTIBLE);
 		spin_lock_bh(&serv->sv_cb_lock);

@@ -120,11 +126,13 @@ nfs41_callback_svc(void *vrqstp)
 				error);
 		} else {
 			spin_unlock_bh(&serv->sv_cb_lock);
-			schedule();
+			if (!kthread_should_stop())
+				schedule();
 			finish_wait(&serv->sv_cb_waitq, &wq);
 		}
-		flush_signals(current);
 	}
+	svc_exit_thread(rqstp);
+	module_put_and_exit(0);
 	return 0;
 }
 

@@ -220,14 +228,14 @@ err_bind:
 static struct svc_serv_ops nfs40_cb_sv_ops = {
 	.svo_function		= nfs4_callback_svc,
 	.svo_enqueue_xprt	= svc_xprt_do_enqueue,
-	.svo_setup		= svc_set_num_threads,
+	.svo_setup		= svc_set_num_threads_sync,
 	.svo_module		= THIS_MODULE,
 };
 #if defined(CONFIG_NFS_V4_1)
 static struct svc_serv_ops nfs41_cb_sv_ops = {
 	.svo_function		= nfs41_callback_svc,
 	.svo_enqueue_xprt	= svc_xprt_do_enqueue,
-	.svo_setup		= svc_set_num_threads,
+	.svo_setup		= svc_set_num_threads_sync,
 	.svo_module		= THIS_MODULE,
 };
 

@@ -470,6 +470,7 @@ void		   svc_pool_map_put(void);
 struct svc_serv *  svc_create_pooled(struct svc_program *, unsigned int,
 			struct svc_serv_ops *);
 int		   svc_set_num_threads(struct svc_serv *, struct svc_pool *, int);
+int		   svc_set_num_threads_sync(struct svc_serv *, struct svc_pool *, int);
 int		   svc_pool_stats_open(struct svc_serv *serv, struct file *file);
 void		   svc_destroy(struct svc_serv *);
 void		   svc_shutdown_net(struct svc_serv *, struct net *);

@@ -795,6 +795,44 @@ svc_set_num_threads(struct svc_serv *ser
 }
 EXPORT_SYMBOL_GPL(svc_set_num_threads);
 
+/* destroy old threads */
+static int
+svc_stop_kthreads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
+{
+	struct task_struct *task;
+	unsigned int state = serv->sv_nrthreads-1;
+
+	/* destroy old threads */
+	do {
+		task = choose_victim(serv, pool, &state);
+		if (task == NULL)
+			break;
+		kthread_stop(task);
+		nrservs++;
+	} while (nrservs < 0);
+	return 0;
+}
+
+int
+svc_set_num_threads_sync(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
+{
+	if (pool == NULL) {
+		/* The -1 assumes caller has done a svc_get() */
+		nrservs -= (serv->sv_nrthreads-1);
+	} else {
+		spin_lock_bh(&pool->sp_lock);
+		nrservs -= pool->sp_nrthreads;
+		spin_unlock_bh(&pool->sp_lock);
+	}
+
+	if (nrservs > 0)
+		return svc_start_kthreads(serv, pool, nrservs);
+	if (nrservs < 0)
+		return svc_stop_kthreads(serv, pool, nrservs);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(svc_set_num_threads_sync);
+
 /*
  * Called from a server thread as it's exiting. Caller must hold the "service
  * mutex" for the service.

@@ -2100,13 +2100,25 @@ static void unreserve_highatomic_pageblo
 				continue;
 
 			/*
-			 * It should never happen but changes to locking could
-			 * inadvertently allow a per-cpu drain to add pages
-			 * to MIGRATE_HIGHATOMIC while unreserving so be safe
-			 * and watch for underflows.
+			 * In page freeing path, migratetype change is racy so
+			 * we can counter several free pages in a pageblock
+			 * in this loop althoug we changed the pageblock type
+			 * from highatomic to ac->migratetype. So we should
+			 * adjust the count once.
 			 */
-			zone->nr_reserved_highatomic -= min(pageblock_nr_pages,
-				zone->nr_reserved_highatomic);
+			if (get_pageblock_migratetype(page) ==
+							MIGRATE_HIGHATOMIC) {
+				/*
+				 * It should never happen but changes to
+				 * locking could inadvertently allow a per-cpu
+				 * drain to add pages to MIGRATE_HIGHATOMIC
+				 * while unreserving so be safe and watch for
+				 * underflows.
+				 */
+				zone->nr_reserved_highatomic -= min(
+						pageblock_nr_pages,
+						zone->nr_reserved_highatomic);
+			}
 
 			/*
 			 * Convert to ac->migratetype and avoid the normal

@@ -61,9 +61,9 @@ struct posix_acl *orangefs_get_acl(struc
 	return acl;
 }
 
-int orangefs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
+static int __orangefs_set_acl(struct inode *inode, struct posix_acl *acl,
+			      int type)
 {
-	struct orangefs_inode_s *orangefs_inode = ORANGEFS_I(inode);
 	int error = 0;
 	void *value = NULL;
 	size_t size = 0;

@@ -72,22 +72,6 @@ int orangefs_set_acl(struct inode *inode
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		name = XATTR_NAME_POSIX_ACL_ACCESS;
-		if (acl) {
-			umode_t mode;
-
-			error = posix_acl_update_mode(inode, &mode, &acl);
-			if (error) {
-				gossip_err("%s: posix_acl_update_mode err: %d\n",
-					   __func__,
-					   error);
-				return error;
-			}
-
-			if (inode->i_mode != mode)
-				SetModeFlag(orangefs_inode);
-			inode->i_mode = mode;
-			mark_inode_dirty_sync(inode);
-		}
 		break;
 	case ACL_TYPE_DEFAULT:
 		name = XATTR_NAME_POSIX_ACL_DEFAULT;

@@ -132,6 +116,29 @@ out:
 	return error;
 }
 
+int orangefs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
+{
+	int error;
+
+	if (type == ACL_TYPE_ACCESS && acl) {
+		umode_t mode;
+
+		error = posix_acl_update_mode(inode, &mode, &acl);
+		if (error) {
+			gossip_err("%s: posix_acl_update_mode err: %d\n",
+				   __func__,
+				   error);
+			return error;
+		}
+
+		if (inode->i_mode != mode)
+			SetModeFlag(ORANGEFS_I(inode));
+		inode->i_mode = mode;
+		mark_inode_dirty_sync(inode);
+	}
+	return __orangefs_set_acl(inode, acl, type);
+}
+
 int orangefs_init_acl(struct inode *inode, struct inode *dir)
 {
 	struct orangefs_inode_s *orangefs_inode = ORANGEFS_I(inode);

@@ -146,13 +153,14 @@ int orangefs_init_acl(struct inode *inod
 		return error;
 
 	if (default_acl) {
-		error = orangefs_set_acl(inode, default_acl, ACL_TYPE_DEFAULT);
+		error = __orangefs_set_acl(inode, default_acl,
+					   ACL_TYPE_DEFAULT);
 		posix_acl_release(default_acl);
 	}
 
 	if (acl) {
 		if (!error)
-			error = orangefs_set_acl(inode, acl, ACL_TYPE_ACCESS);
+			error = __orangefs_set_acl(inode, acl, ACL_TYPE_ACCESS);
 		posix_acl_release(acl);
 	}
 

@@ -551,7 +551,7 @@ int hfi1_make_rc_req(struct rvt_qp *qp,
 		case IB_WR_RDMA_WRITE:
 			if (newreq && !(qp->s_flags & RVT_S_UNLIMITED_CREDIT))
 				qp->s_lsn++;
-			/* FALLTHROUGH */
+			goto no_flow_control;
 		case IB_WR_RDMA_WRITE_WITH_IMM:
 			/* If no credit, return. */
 			if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT) &&

@@ -559,6 +559,7 @@ int hfi1_make_rc_req(struct rvt_qp *qp,
 				qp->s_flags |= RVT_S_WAIT_SSN_CREDIT;
 				goto bail;
 			}
+no_flow_control:
 			put_ib_reth_vaddr(
 				wqe->rdma_wr.remote_addr,
 				&ohdr->u.rc.reth);

@@ -357,7 +357,7 @@ int qib_make_rc_req(struct rvt_qp *qp, u
 		case IB_WR_RDMA_WRITE:
 			if (newreq && !(qp->s_flags & RVT_S_UNLIMITED_CREDIT))
 				qp->s_lsn++;
-			/* FALLTHROUGH */
+			goto no_flow_control;
 		case IB_WR_RDMA_WRITE_WITH_IMM:
 			/* If no credit, return. */
 			if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT) &&

@@ -365,7 +365,7 @@ int qib_make_rc_req(struct rvt_qp *qp, u
 				qp->s_flags |= RVT_S_WAIT_SSN_CREDIT;
 				goto bail;
 			}
-
+no_flow_control:
 			ohdr->u.rc.reth.vaddr =
 				cpu_to_be64(wqe->rdma_wr.remote_addr);
 			ohdr->u.rc.reth.rkey =

@@ -47,6 +47,13 @@ static void sun4i_drv_disable_vblank(str
 	sun4i_tcon_enable_vblank(tcon, false);
 }
 
+static void sun4i_drv_lastclose(struct drm_device *dev)
+{
+	struct sun4i_drv *drv = dev->dev_private;
+
+	drm_fbdev_cma_restore_mode(drv->fbdev);
+}
+
 static const struct file_operations sun4i_drv_fops = {
 	.owner		= THIS_MODULE,
 	.open		= drm_open,

@@ -65,6 +72,7 @@ static struct drm_driver sun4i_drv_drive
 	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME | DRIVER_ATOMIC,
 
 	/* Generic Operations */
+	.lastclose		= sun4i_drv_lastclose,
 	.fops			= &sun4i_drv_fops,
 	.name			= "sun4i-drm",
 	.desc			= "Allwinner sun4i Display Engine",

@@ -446,15 +446,10 @@ static int addr6_resolve(struct sockaddr
 
 	ret = ipv6_stub->ipv6_dst_lookup(addr->net, NULL, &dst, &fl6);
 	if (ret < 0)
-		goto put;
+		return ret;
 
 	rt = (struct rt6_info *)dst;
-	if (ipv6_addr_any(&fl6.saddr)) {
-		ret = ipv6_dev_get_saddr(addr->net, ip6_dst_idev(dst)->dev,
-					 &fl6.daddr, 0, &fl6.saddr);
-		if (ret)
-			goto put;
-
+	if (ipv6_addr_any(&src_in->sin6_addr)) {
 		src_in->sin6_family = AF_INET6;
 		src_in->sin6_addr = fl6.saddr;
 	}

@@ -471,9 +466,6 @@ static int addr6_resolve(struct sockaddr
 
 	*pdst = dst;
 	return 0;
-put:
-	dst_release(dst);
-	return ret;
 }
 #else
 static int addr6_resolve(struct sockaddr_in6 *src_in,

@@ -362,6 +362,30 @@ int tty_insert_flip_string_flags(struct
 EXPORT_SYMBOL(tty_insert_flip_string_flags);
 
 /**
+ *	__tty_insert_flip_char   -	Add one character to the tty buffer
+ *	@port: tty port
+ *	@ch: character
+ *	@flag: flag byte
+ *
+ *	Queue a single byte to the tty buffering, with an optional flag.
+ *	This is the slow path of tty_insert_flip_char.
+ */
+int __tty_insert_flip_char(struct tty_port *port, unsigned char ch, char flag)
+{
+	struct tty_buffer *tb = port->buf.tail;
+	int flags = (flag == TTY_NORMAL) ? TTYB_NORMAL : 0;
+
+	if (!tty_buffer_request_room(port, 1))
+		return 0;
+
+	*flag_buf_ptr(tb, tb->used) = flag;
+	*char_buf_ptr(tb, tb->used++) = ch;
+
+	return 1;
+}
+EXPORT_SYMBOL(__tty_insert_flip_char);
+
+/**
  *	tty_schedule_flip	-	push characters to ldisc
  *	@port: tty port to push from
  *

@@ -12,6 +12,7 @@ extern int tty_prepare_flip_string(struc
 		unsigned char **chars, size_t size);
 extern void tty_flip_buffer_push(struct tty_port *port);
 void tty_schedule_flip(struct tty_port *port);
+int __tty_insert_flip_char(struct tty_port *port, unsigned char ch, char flag);
 
 static inline int tty_insert_flip_char(struct tty_port *port,
 					unsigned char ch, char flag)

@@ -26,7 +27,7 @@ static inline int tty_insert_flip_char(s
 		*char_buf_ptr(tb, tb->used++) = ch;
 		return 1;
 	}
-	return tty_insert_flip_string_flags(port, &ch, &flag, 1);
+	return __tty_insert_flip_char(port, ch, flag);
 }
 
 static inline int tty_insert_flip_string(struct tty_port *port,

@@ -375,10 +375,11 @@ int __tty_insert_flip_char(struct tty_po
 	struct tty_buffer *tb = port->buf.tail;
 	int flags = (flag == TTY_NORMAL) ? TTYB_NORMAL : 0;
 
-	if (!tty_buffer_request_room(port, 1))
+	if (!__tty_buffer_request_room(port, 1, flags))
 		return 0;
 
-	*flag_buf_ptr(tb, tb->used) = flag;
+	if (~tb->flags & TTYB_NORMAL)
+		*flag_buf_ptr(tb, tb->used) = flag;
 	*char_buf_ptr(tb, tb->used++) = ch;
 
 	return 1;

@@ -372,12 +372,13 @@ EXPORT_SYMBOL(tty_insert_flip_string_fla
  */
 int __tty_insert_flip_char(struct tty_port *port, unsigned char ch, char flag)
 {
-	struct tty_buffer *tb = port->buf.tail;
+	struct tty_buffer *tb;
 	int flags = (flag == TTY_NORMAL) ? TTYB_NORMAL : 0;
 
 	if (!__tty_buffer_request_room(port, 1, flags))
 		return 0;
 
+	tb = port->buf.tail;
 	if (~tb->flags & TTYB_NORMAL)
 		*flag_buf_ptr(tb, tb->used) = flag;
 	*char_buf_ptr(tb, tb->used++) = ch;

@@ -32,6 +32,7 @@
 #include <linux/pinctrl/pinconf.h>
 #include <linux/pinctrl/pinconf-generic.h>
 
+#include "core.h"
 #include "pinctrl-utils.h"
 #include "pinctrl-amd.h"
 

@@ -712,6 +713,69 @@ static const struct pinconf_ops amd_pinc
 	.pin_config_group_set = amd_pinconf_group_set,
 };
 
+#ifdef CONFIG_PM_SLEEP
+static bool amd_gpio_should_save(struct amd_gpio *gpio_dev, unsigned int pin)
+{
+	const struct pin_desc *pd = pin_desc_get(gpio_dev->pctrl, pin);
+
+	if (!pd)
+		return false;
+
+	/*
+	 * Only restore the pin if it is actually in use by the kernel (or
+	 * by userspace).
+	 */
+	if (pd->mux_owner || pd->gpio_owner ||
+	    gpiochip_line_is_irq(&gpio_dev->gc, pin))
+		return true;
+
+	return false;
+}
+
+int amd_gpio_suspend(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct amd_gpio *gpio_dev = platform_get_drvdata(pdev);
+	struct pinctrl_desc *desc = gpio_dev->pctrl->desc;
+	int i;
+
+	for (i = 0; i < desc->npins; i++) {
+		int pin = desc->pins[i].number;
+
+		if (!amd_gpio_should_save(gpio_dev, pin))
+			continue;
+
+		gpio_dev->saved_regs[i] = readl(gpio_dev->base + pin*4);
+	}
+
+	return 0;
+}
+
+int amd_gpio_resume(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct amd_gpio *gpio_dev = platform_get_drvdata(pdev);
+	struct pinctrl_desc *desc = gpio_dev->pctrl->desc;
+	int i;
+
+	for (i = 0; i < desc->npins; i++) {
+		int pin = desc->pins[i].number;
+
+		if (!amd_gpio_should_save(gpio_dev, pin))
+			continue;
+
+		writel(gpio_dev->saved_regs[i], gpio_dev->base + pin*4);
+	}
+
+	return 0;
+}
+
+static const struct dev_pm_ops amd_gpio_pm_ops = {
+	SET_LATE_SYSTEM_SLEEP_PM_OPS(amd_gpio_suspend,
+				     amd_gpio_resume)
+};
+#endif
+
 static struct pinctrl_desc amd_pinctrl_desc = {
 	.pins	= kerncz_pins,
 	.npins = ARRAY_SIZE(kerncz_pins),

@@ -751,6 +815,14 @@ static int amd_gpio_probe(struct platfor
 		return -EINVAL;
 	}
 
+#ifdef CONFIG_PM_SLEEP
+	gpio_dev->saved_regs = devm_kcalloc(&pdev->dev, amd_pinctrl_desc.npins,
+					    sizeof(*gpio_dev->saved_regs),
+					    GFP_KERNEL);
+	if (!gpio_dev->saved_regs)
+		return -ENOMEM;
+#endif
+
 	gpio_dev->pdev = pdev;
 	gpio_dev->gc.direction_input	= amd_gpio_direction_input;
 	gpio_dev->gc.direction_output	= amd_gpio_direction_output;

@@ -839,6 +911,9 @@ static struct platform_driver amd_gpio_d
 	.driver		= {
 		.name	= "amd_gpio",
 		.acpi_match_table = ACPI_PTR(amd_gpio_acpi_match),
+#ifdef CONFIG_PM_SLEEP
+		.pm	= &amd_gpio_pm_ops,
+#endif
 	},
 	.probe		= amd_gpio_probe,
 	.remove		= amd_gpio_remove,

@@ -95,6 +95,7 @@ struct amd_gpio {
 	struct gpio_chip        gc;
 	struct resource         *res;
 	struct platform_device  *pdev;
+	u32			*saved_regs;
 };
 
 /*  KERNCZ configuration*/

@@ -905,6 +905,13 @@ static const struct dmi_system_id __init
 		},
 	},
 	{
+		/* Gigabyte P57 - Elantech touchpad */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "GIGABYTE"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "P57"),
+		},
+	},
+	{
 		/* Schenker XMG C504 - Elantech touchpad */
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "XMG"),

@@ -47,14 +47,26 @@ union ieee754dp ieee754dp_fmax(union iee
 	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
 		return ieee754dp_nanxcpt(x);
 
-	/* numbers are preferred to NaNs */
+	/*
+	 * Quiet NaN handling
+	 */
+
+	/*
+	 *    The case of both inputs quiet NaNs
+	 */
+	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
+		return x;
+
+	/*
+	 *    The cases of exactly one input quiet NaN (numbers
+	 *    are here preferred as returned values to NaNs)
+	 */
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):

@@ -147,14 +159,26 @@ union ieee754dp ieee754dp_fmaxa(union ie
 	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
 		return ieee754dp_nanxcpt(x);
 
-	/* numbers are preferred to NaNs */
+	/*
+	 * Quiet NaN handling
+	 */
+
+	/*
+	 *    The case of both inputs quiet NaNs
+	 */
+	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
+		return x;
+
+	/*
+	 *    The cases of exactly one input quiet NaN (numbers
+	 *    are here preferred as returned values to NaNs)
+	 */
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):

@@ -47,14 +47,26 @@ union ieee754dp ieee754dp_fmin(union iee
 	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
 		return ieee754dp_nanxcpt(x);
 
-	/* numbers are preferred to NaNs */
+	/*
+	 * Quiet NaN handling
+	 */
+
+	/*
+	 *    The case of both inputs quiet NaNs
+	 */
+	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
+		return x;
+
+	/*
+	 *    The cases of exactly one input quiet NaN (numbers
+	 *    are here preferred as returned values to NaNs)
+	 */
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):

@@ -147,14 +159,26 @@ union ieee754dp ieee754dp_fmina(union ie
 	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
 		return ieee754dp_nanxcpt(x);
 
-	/* numbers are preferred to NaNs */
+	/*
+	 * Quiet NaN handling
+	 */
+
+	/*
+	 *    The case of both inputs quiet NaNs
+	 */
+	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
+		return x;
+
+	/*
+	 *    The cases of exactly one input quiet NaN (numbers
+	 *    are here preferred as returned values to NaNs)
+	 */
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):

@@ -47,14 +47,26 @@ union ieee754sp ieee754sp_fmax(union iee
 	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
 		return ieee754sp_nanxcpt(x);
 
-	/* numbers are preferred to NaNs */
+	/*
+	 * Quiet NaN handling
+	 */
+
+	/*
+	 *    The case of both inputs quiet NaNs
+	 */
+	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
+		return x;
+
+	/*
+	 *    The cases of exactly one input quiet NaN (numbers
+	 *    are here preferred as returned values to NaNs)
+	 */
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):

@@ -147,14 +159,26 @@ union ieee754sp ieee754sp_fmaxa(union ie
 	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
 		return ieee754sp_nanxcpt(x);
 
-	/* numbers are preferred to NaNs */
+	/*
+	 * Quiet NaN handling
+	 */
+
+	/*
+	 *    The case of both inputs quiet NaNs
+	 */
+	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
+		return x;
+
+	/*
+	 *    The cases of exactly one input quiet NaN (numbers
+	 *    are here preferred as returned values to NaNs)
+	 */
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):

@@ -47,14 +47,26 @@ union ieee754sp ieee754sp_fmin(union iee
 	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
 		return ieee754sp_nanxcpt(x);
 
-	/* numbers are preferred to NaNs */
+	/*
+	 * Quiet NaN handling
+	 */
+
+	/*
+	 *    The case of both inputs quiet NaNs
+	 */
+	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
+		return x;
+
+	/*
+	 *    The cases of exactly one input quiet NaN (numbers
+	 *    are here preferred as returned values to NaNs)
+	 */
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):

@@ -147,14 +159,26 @@ union ieee754sp ieee754sp_fmina(union ie
 	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
 		return ieee754sp_nanxcpt(x);
 
-	/* numbers are preferred to NaNs */
+	/*
+	 * Quiet NaN handling
+	 */
+
+	/*
+	 *    The case of both inputs quiet NaNs
+	 */
+	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
+		return x;
+
+	/*
+	 *    The cases of exactly one input quiet NaN (numbers
+	 *    are here preferred as returned values to NaNs)
+	 */
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):

@@ -92,9 +92,7 @@ union ieee754dp ieee754dp_fmax(union iee
 		return ys ? x : y;
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
-		if (xs == ys)
-			return x;
-		return ieee754dp_zero(1);
+		return ieee754dp_zero(xs & ys);
 
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
 		DPDNORMX;

@@ -204,9 +202,7 @@ union ieee754dp ieee754dp_fmaxa(union ie
 		return y;
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
-		if (xs == ys)
-			return x;
-		return ieee754dp_zero(1);
+		return ieee754dp_zero(xs & ys);
 
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
 		DPDNORMX;

@@ -92,9 +92,7 @@ union ieee754dp ieee754dp_fmin(union iee
 		return ys ? y : x;
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
-		if (xs == ys)
-			return x;
-		return ieee754dp_zero(1);
+		return ieee754dp_zero(xs | ys);
 
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
 		DPDNORMX;

@@ -204,9 +202,7 @@ union ieee754dp ieee754dp_fmina(union ie
 		return y;
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
-		if (xs == ys)
-			return x;
-		return ieee754dp_zero(1);
+		return ieee754dp_zero(xs | ys);
 
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
 		DPDNORMX;

@@ -92,9 +92,7 @@ union ieee754sp ieee754sp_fmax(union iee
 		return ys ? x : y;
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
-		if (xs == ys)
-			return x;
-		return ieee754sp_zero(1);
+		return ieee754sp_zero(xs & ys);
 
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
 		SPDNORMX;

@@ -204,9 +202,7 @@ union ieee754sp ieee754sp_fmaxa(union ie
 		return y;
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
-		if (xs == ys)
-			return x;
-		return ieee754sp_zero(1);
+		return ieee754sp_zero(xs & ys);
 
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
 		SPDNORMX;

@@ -92,9 +92,7 @@ union ieee754sp ieee754sp_fmin(union iee
 		return ys ? y : x;
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
-		if (xs == ys)
-			return x;
-		return ieee754sp_zero(1);
+		return ieee754sp_zero(xs | ys);
 
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
 		SPDNORMX;

@@ -204,9 +202,7 @@ union ieee754sp ieee754sp_fmina(union ie
 		return y;
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
-		if (xs == ys)
-			return x;
-		return ieee754sp_zero(1);
+		return ieee754sp_zero(xs | ys);
 
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
 		SPDNORMX;

@@ -116,16 +116,32 @@ union ieee754dp ieee754dp_fmax(union iee
 	else if (xs < ys)
 		return x;
 
-	/* Compare exponent */
-	if (xe > ye)
-		return x;
-	else if (xe < ye)
-		return y;
+	/* Signs of inputs are equal, let's compare exponents */
+	if (xs == 0) {
+		/* Inputs are both positive */
+		if (xe > ye)
+			return x;
+		else if (xe < ye)
+			return y;
+	} else {
+		/* Inputs are both negative */
+		if (xe > ye)
+			return y;
+		else if (xe < ye)
+			return x;
+	}
 
-	/* Compare mantissa */
+	/* Signs and exponents of inputs are equal, let's compare mantissas */
+	if (xs == 0) {
+		/* Inputs are both positive, with equal signs and exponents */
+		if (xm <= ym)
+			return y;
+		return x;
+	}
+	/* Inputs are both negative, with equal signs and exponents */
 	if (xm <= ym)
-		return y;
-	return x;
+		return x;
+	return y;
 }
 
 union ieee754dp ieee754dp_fmaxa(union ieee754dp x, union ieee754dp y)

@@ -116,16 +116,32 @@ union ieee754dp ieee754dp_fmin(union iee
 	else if (xs < ys)
 		return y;
 
-	/* Compare exponent */
-	if (xe > ye)
-		return y;
-	else if (xe < ye)
-		return x;
+	/* Signs of inputs are the same, let's compare exponents */
+	if (xs == 0) {
+		/* Inputs are both positive */
+		if (xe > ye)
+			return y;
+		else if (xe < ye)
+			return x;
+	} else {
+		/* Inputs are both negative */
+		if (xe > ye)
+			return x;
+		else if (xe < ye)
+			return y;
+	}
 
-	/* Compare mantissa */
+	/* Signs and exponents of inputs are equal, let's compare mantissas */
+	if (xs == 0) {
+		/* Inputs are both positive, with equal signs and exponents */
+		if (xm <= ym)
+			return x;
+		return y;
+	}
+	/* Inputs are both negative, with equal signs and exponents */
 	if (xm <= ym)
-		return x;
-	return y;
+		return y;
+	return x;
 }
 
 union ieee754dp ieee754dp_fmina(union ieee754dp x, union ieee754dp y)

@@ -116,16 +116,32 @@ union ieee754sp ieee754sp_fmax(union iee
 	else if (xs < ys)
 		return x;
 
-	/* Compare exponent */
-	if (xe > ye)
-		return x;
-	else if (xe < ye)
-		return y;
+	/* Signs of inputs are equal, let's compare exponents */
+	if (xs == 0) {
+		/* Inputs are both positive */
+		if (xe > ye)
+			return x;
+		else if (xe < ye)
+			return y;
+	} else {
+		/* Inputs are both negative */
+		if (xe > ye)
+			return y;
+		else if (xe < ye)
+			return x;
+	}
 
-	/* Compare mantissa */
+	/* Signs and exponents of inputs are equal, let's compare mantissas */
+	if (xs == 0) {
+		/* Inputs are both positive, with equal signs and exponents */
+		if (xm <= ym)
+			return y;
+		return x;
+	}
+	/* Inputs are both negative, with equal signs and exponents */
 	if (xm <= ym)
-		return y;
-	return x;
+		return x;
+	return y;
 }
 
 union ieee754sp ieee754sp_fmaxa(union ieee754sp x, union ieee754sp y)

@@ -116,16 +116,32 @@ union ieee754sp ieee754sp_fmin(union iee
 	else if (xs < ys)
 		return y;
 
-	/* Compare exponent */
-	if (xe > ye)
-		return y;
-	else if (xe < ye)
-		return x;
+	/* Signs of inputs are the same, let's compare exponents */
+	if (xs == 0) {
+		/* Inputs are both positive */
+		if (xe > ye)
+			return y;
+		else if (xe < ye)
+			return x;
+	} else {
+		/* Inputs are both negative */
+		if (xe > ye)
+			return x;
+		else if (xe < ye)
+			return y;
+	}
 
-	/* Compare mantissa */
+	/* Signs and exponents of inputs are equal, let's compare mantissas */
+	if (xs == 0) {
+		/* Inputs are both positive, with equal signs and exponents */
+		if (xm <= ym)
+			return x;
+		return y;
+	}
+	/* Inputs are both negative, with equal signs and exponents */
 	if (xm <= ym)
-		return x;
-	return y;
+		return y;
+	return x;
 }
 
 union ieee754sp ieee754sp_fmina(union ieee754sp x, union ieee754sp y)

@@ -243,7 +243,11 @@ union ieee754dp ieee754dp_fmaxa(union ie
 		return y;
 
 	/* Compare mantissa */
-	if (xm <= ym)
+	if (xm < ym)
 		return y;
-	return x;
+	else if (xm > ym)
+		return x;
+	else if (xs == 0)
+		return x;
+	return y;
 }

@@ -243,7 +243,11 @@ union ieee754dp ieee754dp_fmina(union ie
 		return x;
 
 	/* Compare mantissa */
-	if (xm <= ym)
+	if (xm < ym)
+		return x;
+	else if (xm > ym)
+		return y;
+	else if (xs == 1)
 		return x;
 	return y;
 }

@@ -243,7 +243,11 @@ union ieee754sp ieee754sp_fmaxa(union ie
 		return y;
 
 	/* Compare mantissa */
-	if (xm <= ym)
+	if (xm < ym)
 		return y;
-	return x;
+	else if (xm > ym)
+		return x;
+	else if (xs == 0)
+		return x;
+	return y;
 }

@@ -243,7 +243,11 @@ union ieee754sp ieee754sp_fmina(union ie
 		return x;
 
 	/* Compare mantissa */
-	if (xm <= ym)
+	if (xm < ym)
+		return x;
+	else if (xm > ym)
+		return y;
+	else if (xs == 1)
 		return x;
 	return y;
 }

@@ -202,6 +202,9 @@ union ieee754dp ieee754dp_fmaxa(union ie
 	/*
 	 * Infinity and zero handling
 	 */
+	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
+		return ieee754dp_inf(xs & ys);
+
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):

@@ -209,7 +212,6 @@ union ieee754dp ieee754dp_fmaxa(union ie
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):

@@ -202,6 +202,9 @@ union ieee754dp ieee754dp_fmina(union ie
 	/*
 	 * Infinity and zero handling
 	 */
+	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
+		return ieee754dp_inf(xs | ys);
+
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):

@@ -209,7 +212,6 @@ union ieee754dp ieee754dp_fmina(union ie
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):

@@ -202,6 +202,9 @@ union ieee754sp ieee754sp_fmaxa(union ie
 	/*
 	 * Infinity and zero handling
 	 */
+	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
+		return ieee754sp_inf(xs & ys);
+
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):

@@ -209,7 +212,6 @@ union ieee754sp ieee754sp_fmaxa(union ie
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):

@@ -202,6 +202,9 @@ union ieee754sp ieee754sp_fmina(union ie
 	/*
 	 * Infinity and zero handling
 	 */
+	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
+		return ieee754sp_inf(xs | ys);
+
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):

@@ -209,7 +212,6 @@ union ieee754sp ieee754sp_fmina(union ie
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
 		return x;
 
-	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):

@@ -210,14 +210,14 @@ union ieee754dp ieee754dp_fmina(union ie
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
-		return x;
+		return y;
 
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_DNORM):
-		return y;
+		return x;
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
 		return ieee754dp_zero(xs | ys);

@@ -210,14 +210,14 @@ union ieee754sp ieee754sp_fmina(union ie
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
-		return x;
+		return y;
 
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_DNORM):
-		return y;
+		return x;
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
 		return ieee754sp_zero(xs | ys);

@@ -54,7 +54,7 @@ static union ieee754dp _dp_maddf(union i
 		return ieee754dp_nanxcpt(z);
 	case IEEE754_CLASS_DNORM:
 		DPDNORMZ;
-	/* QNAN is handled separately below */
+	/* QNAN and ZERO cases are handled separately below */
 	}
 
 	switch (CLPAIR(xc, yc)) {

@@ -210,6 +210,9 @@ static union ieee754dp _dp_maddf(union i
 	}
 	assert(rm & (DP_HIDDEN_BIT << 3));
 
+	if (zc == IEEE754_CLASS_ZERO)
+		return ieee754dp_format(rs, re, rm);
+
 	/* And now the addition */
 	assert(zm & DP_HIDDEN_BIT);
 

@@ -54,7 +54,7 @@ static union ieee754sp _sp_maddf(union i
 		return ieee754sp_nanxcpt(z);
 	case IEEE754_CLASS_DNORM:
 		SPDNORMZ;
-	/* QNAN is handled separately below */
+	/* QNAN and ZERO cases are handled separately below */
 	}
 
 	switch (CLPAIR(xc, yc)) {

@@ -203,6 +203,9 @@ static union ieee754sp _sp_maddf(union i
 	}
 	assert(rm & (SP_HIDDEN_BIT << 3));
 
+	if (zc == IEEE754_CLASS_ZERO)
+		return ieee754sp_format(rs, re, rm);
+
 	/* And now the addition */
 
 	assert(zm & SP_HIDDEN_BIT);

@@ -48,52 +48,34 @@ static union ieee754dp _dp_maddf(union i
 
 	ieee754_clearcx();
 
-	switch (zc) {
-	case IEEE754_CLASS_SNAN:
-		ieee754_setcx(IEEE754_INVALID_OPERATION);
+	/*
+	 * Handle the cases when at least one of x, y or z is a NaN.
+	 * Order of precedence is sNaN, qNaN and z, x, y.
+	 */
+	if (zc == IEEE754_CLASS_SNAN)
 		return ieee754dp_nanxcpt(z);
-	case IEEE754_CLASS_DNORM:
-		DPDNORMZ;
-	/* QNAN and ZERO cases are handled separately below */
-	}
-
-	switch (CLPAIR(xc, yc)) {
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_SNAN):
-	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_SNAN):
-	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_SNAN):
-	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_SNAN):
-	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_SNAN):
-		return ieee754dp_nanxcpt(y);
-
-	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_SNAN):
-	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_QNAN):
-	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_ZERO):
-	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_NORM):
-	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_DNORM):
-	case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
+	if (xc == IEEE754_CLASS_SNAN)
 		return ieee754dp_nanxcpt(x);
-
-	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
-	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
-	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
-	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
+	if (yc == IEEE754_CLASS_SNAN)
+		return ieee754dp_nanxcpt(y);
+	if (zc == IEEE754_CLASS_QNAN)
+		return z;
+	if (xc == IEEE754_CLASS_QNAN)
+		return x;
+	if (yc == IEEE754_CLASS_QNAN)
 		return y;
 
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):
-	case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_INF):
-		return x;
+	if (zc == IEEE754_CLASS_DNORM)
+		DPDNORMZ;
+	/* ZERO z cases are handled separately below */
 
+	switch (CLPAIR(xc, yc)) {
 
 	/*
 	 * Infinity handling
 	 */
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):
-		if (zc == IEEE754_CLASS_QNAN)
-			return z;
 		ieee754_setcx(IEEE754_INVALID_OPERATION);
 		return ieee754dp_indef();
 

@@ -102,8 +84,6 @@ static union ieee754dp _dp_maddf(union i
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
-		if (zc == IEEE754_CLASS_QNAN)
-			return z;
 		return ieee754dp_inf(xs ^ ys);
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):

@@ -120,25 +100,19 @@ static union ieee754dp _dp_maddf(union i
 		DPDNORMX;
 
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_DNORM):
-		if (zc == IEEE754_CLASS_QNAN)
-			return z;
-		else if (zc == IEEE754_CLASS_INF)
+		if (zc == IEEE754_CLASS_INF)
 			return ieee754dp_inf(zs);
 		DPDNORMY;
 		break;
 
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_NORM):
-		if (zc == IEEE754_CLASS_QNAN)
-			return z;
-		else if (zc == IEEE754_CLASS_INF)
+		if (zc == IEEE754_CLASS_INF)
 			return ieee754dp_inf(zs);
 		DPDNORMX;
 		break;
 
 	case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_NORM):
-		if (zc == IEEE754_CLASS_QNAN)
-			return z;
-		else if (zc == IEEE754_CLASS_INF)
+		if (zc == IEEE754_CLASS_INF)
 			return ieee754dp_inf(zs);
 		/* fall through to real computations */
 	}

105 lines not on this page · raw message

@@ -84,7 +84,27 @@ static union ieee754dp _dp_maddf(union i
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
-		return ieee754dp_inf(xs ^ ys);
+		if ((zc == IEEE754_CLASS_INF) &&
+		    ((!(flags & maddf_negate_product) && (zs != (xs ^ ys))) ||
+		     ((flags & maddf_negate_product) && (zs == (xs ^ ys))))) {
+			/*
+			 * Cases of addition of infinities with opposite signs
+			 * or subtraction of infinities with same signs.
+			 */
+			ieee754_setcx(IEEE754_INVALID_OPERATION);
+			return ieee754dp_indef();
+		}
+		/*
+		 * z is here either not an infinity, or an infinity having the
+		 * same sign as product (x*y) (in case of MADDF.D instruction)
+		 * or product -(x*y) (in MSUBF.D case). The result must be an
+		 * infinity, and its sign is determined only by the value of
+		 * (flags & maddf_negate_product) and the signs of x and y.
+		 */
+		if (flags & maddf_negate_product)
+			return ieee754dp_inf(1 ^ (xs ^ ys));
+		else
+			return ieee754dp_inf(xs ^ ys);
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_NORM):

@@ -85,7 +85,27 @@ static union ieee754sp _sp_maddf(union i
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
-		return ieee754sp_inf(xs ^ ys);
+		if ((zc == IEEE754_CLASS_INF) &&
+		    ((!(flags & maddf_negate_product) && (zs != (xs ^ ys))) ||
+		     ((flags & maddf_negate_product) && (zs == (xs ^ ys))))) {
+			/*
+			 * Cases of addition of infinities with opposite signs
+			 * or subtraction of infinities with same signs.
+			 */
+			ieee754_setcx(IEEE754_INVALID_OPERATION);
+			return ieee754sp_indef();
+		}
+		/*
+		 * z is here either not an infinity, or an infinity having the
+		 * same sign as product (x*y) (in case of MADDF.D instruction)
+		 * or product -(x*y) (in MSUBF.D case). The result must be an
+		 * infinity, and its sign is determined only by the value of
+		 * (flags & maddf_negate_product) and the signs of x and y.
+		 */
+		if (flags & maddf_negate_product)
+			return ieee754sp_inf(1 ^ (xs ^ ys));
+		else
+			return ieee754sp_inf(xs ^ ys);
 
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
 	case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_NORM):

@@ -113,7 +113,23 @@ static union ieee754dp _dp_maddf(union i
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
 		if (zc == IEEE754_CLASS_INF)
 			return ieee754dp_inf(zs);
-		/* Multiplication is 0 so just return z */
+		if (zc == IEEE754_CLASS_ZERO) {
+			/* Handle cases +0 + (-0) and similar ones. */
+			if ((!(flags & maddf_negate_product)
+					&& (zs == (xs ^ ys))) ||
+			    ((flags & maddf_negate_product)
+					&& (zs != (xs ^ ys))))
+				/*
+				 * Cases of addition of zeros of equal signs
+				 * or subtraction of zeroes of opposite signs.
+				 * The sign of the resulting zero is in any
+				 * such case determined only by the sign of z.
+				 */
+				return z;
+
+			return ieee754dp_zero(ieee754_csr.rm == FPU_CSR_RD);
+		}
+		/* x*y is here 0, and z is not 0, so just return z */
 		return z;
 
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):

@@ -114,7 +114,23 @@ static union ieee754sp _sp_maddf(union i
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
 		if (zc == IEEE754_CLASS_INF)
 			return ieee754sp_inf(zs);
-		/* Multiplication is 0 so just return z */
+		if (zc == IEEE754_CLASS_ZERO) {
+			/* Handle cases +0 + (-0) and similar ones. */
+			if ((!(flags & maddf_negate_product)
+					&& (zs == (xs ^ ys))) ||
+			    ((flags & maddf_negate_product)
+					&& (zs != (xs ^ ys))))
+				/*
+				 * Cases of addition of zeros of equal signs
+				 * or subtraction of zeroes of opposite signs.
+				 * The sign of the resulting zero is in any
+				 * such case determined only by the sign of z.
+				 */
+				return z;
+
+			return ieee754sp_zero(ieee754_csr.rm == FPU_CSR_RD);
+		}
+		/* x*y is here 0, and z is not 0, so just return z */
 		return z;
 
 	case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):

@@ -14,9 +14,6 @@
 
 #include "ieee754dp.h"
 
-enum maddf_flags {
-	maddf_negate_product	= 1 << 0,
-};
 
 static union ieee754dp _dp_maddf(union ieee754dp z, union ieee754dp x,
 				 union ieee754dp y, enum maddf_flags flags)

@@ -85,8 +82,8 @@ static union ieee754dp _dp_maddf(union i
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
 		if ((zc == IEEE754_CLASS_INF) &&
-		    ((!(flags & maddf_negate_product) && (zs != (xs ^ ys))) ||
-		     ((flags & maddf_negate_product) && (zs == (xs ^ ys))))) {
+		    ((!(flags & MADDF_NEGATE_PRODUCT) && (zs != (xs ^ ys))) ||
+		     ((flags & MADDF_NEGATE_PRODUCT) && (zs == (xs ^ ys))))) {
 			/*
 			 * Cases of addition of infinities with opposite signs
 			 * or subtraction of infinities with same signs.

@@ -99,9 +96,9 @@ static union ieee754dp _dp_maddf(union i
 		 * same sign as product (x*y) (in case of MADDF.D instruction)
 		 * or product -(x*y) (in MSUBF.D case). The result must be an
 		 * infinity, and its sign is determined only by the value of
-		 * (flags & maddf_negate_product) and the signs of x and y.
+		 * (flags & MADDF_NEGATE_PRODUCT) and the signs of x and y.
 		 */
-		if (flags & maddf_negate_product)
+		if (flags & MADDF_NEGATE_PRODUCT)
 			return ieee754dp_inf(1 ^ (xs ^ ys));
 		else
 			return ieee754dp_inf(xs ^ ys);

@@ -115,9 +112,9 @@ static union ieee754dp _dp_maddf(union i
 			return ieee754dp_inf(zs);
 		if (zc == IEEE754_CLASS_ZERO) {
 			/* Handle cases +0 + (-0) and similar ones. */
-			if ((!(flags & maddf_negate_product)
+			if ((!(flags & MADDF_NEGATE_PRODUCT)
 					&& (zs == (xs ^ ys))) ||
-			    ((flags & maddf_negate_product)
+			    ((flags & MADDF_NEGATE_PRODUCT)
 					&& (zs != (xs ^ ys))))
 				/*
 				 * Cases of addition of zeros of equal signs

@@ -167,7 +164,7 @@ static union ieee754dp _dp_maddf(union i
 
 	re = xe + ye;
 	rs = xs ^ ys;
-	if (flags & maddf_negate_product)
+	if (flags & MADDF_NEGATE_PRODUCT)
 		rs ^= 1;
 
 	/* shunt to top of word */

@@ -291,5 +288,5 @@ union ieee754dp ieee754dp_maddf(union ie
 union ieee754dp ieee754dp_msubf(union ieee754dp z, union ieee754dp x,
 				union ieee754dp y)
 {
-	return _dp_maddf(z, x, y, maddf_negate_product);
+	return _dp_maddf(z, x, y, MADDF_NEGATE_PRODUCT);
 }

@@ -26,6 +26,10 @@
 
 #define CLPAIR(x, y)	((x)*6+(y))
 
+enum maddf_flags {
+	MADDF_NEGATE_PRODUCT	= 1 << 0,
+};
+
 static inline void ieee754_clearcx(void)
 {
 	ieee754_csr.cx = 0;

@@ -14,9 +14,6 @@
 
 #include "ieee754sp.h"
 
-enum maddf_flags {
-	maddf_negate_product	= 1 << 0,
-};
 
 static union ieee754sp _sp_maddf(union ieee754sp z, union ieee754sp x,
 				 union ieee754sp y, enum maddf_flags flags)

@@ -86,8 +83,8 @@ static union ieee754sp _sp_maddf(union i
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
 	case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
 		if ((zc == IEEE754_CLASS_INF) &&
-		    ((!(flags & maddf_negate_product) && (zs != (xs ^ ys))) ||
-		     ((flags & maddf_negate_product) && (zs == (xs ^ ys))))) {
+		    ((!(flags & MADDF_NEGATE_PRODUCT) && (zs != (xs ^ ys))) ||
+		     ((flags & MADDF_NEGATE_PRODUCT) && (zs == (xs ^ ys))))) {
 			/*
 			 * Cases of addition of infinities with opposite signs
 			 * or subtraction of infinities with same signs.

@@ -100,9 +97,9 @@ static union ieee754sp _sp_maddf(union i
 		 * same sign as product (x*y) (in case of MADDF.D instruction)
 		 * or product -(x*y) (in MSUBF.D case). The result must be an
 		 * infinity, and its sign is determined only by the value of
-		 * (flags & maddf_negate_product) and the signs of x and y.
+		 * (flags & MADDF_NEGATE_PRODUCT) and the signs of x and y.
 		 */
-		if (flags & maddf_negate_product)
+		if (flags & MADDF_NEGATE_PRODUCT)
 			return ieee754sp_inf(1 ^ (xs ^ ys));
 		else
 			return ieee754sp_inf(xs ^ ys);

@@ -116,9 +113,9 @@ static union ieee754sp _sp_maddf(union i
 			return ieee754sp_inf(zs);
 		if (zc == IEEE754_CLASS_ZERO) {
 			/* Handle cases +0 + (-0) and similar ones. */
-			if ((!(flags & maddf_negate_product)
+			if ((!(flags & MADDF_NEGATE_PRODUCT)
 					&& (zs == (xs ^ ys))) ||
-			    ((flags & maddf_negate_product)
+			    ((flags & MADDF_NEGATE_PRODUCT)
 					&& (zs != (xs ^ ys))))
 				/*
 				 * Cases of addition of zeros of equal signs

@@ -170,7 +167,7 @@ static union ieee754sp _sp_maddf(union i
 
 	re = xe + ye;
 	rs = xs ^ ys;
-	if (flags & maddf_negate_product)
+	if (flags & MADDF_NEGATE_PRODUCT)
 		rs ^= 1;
 
 	/* shunt to top of word */

@@ -287,5 +284,5 @@ union ieee754sp ieee754sp_maddf(union ie
 union ieee754sp ieee754sp_msubf(union ieee754sp z, union ieee754sp x,
 				union ieee754sp y)
 {
-	return _sp_maddf(z, x, y, maddf_negate_product);
+	return _sp_maddf(z, x, y, MADDF_NEGATE_PRODUCT);
 }

@@ -45,6 +45,10 @@ static inline int ieee754sp_finite(union
 	return SPBEXP(x) != SP_EMAX + 1 + SP_EBIAS;
 }
 
+/* 64 bit right shift with rounding */
+#define XSPSRS64(v, rs)						\
+	(((rs) >= 64) ? ((v) != 0) : ((v) >> (rs)) | ((v) << (64-(rs)) != 0))
+
 /* 3bit extended single precision sticky right shift */
 #define XSPSRS(v, rs)						\
 	((rs > (SP_FBITS+3))?1:((v) >> (rs)) | ((v) << (32-(rs)) != 0))

168 lines not on this page · raw message

@@ -15,18 +15,44 @@
 #include "ieee754dp.h"
 
 
+/* 128 bits shift right logical with rounding. */
+void srl128(u64 *hptr, u64 *lptr, int count)
+{
+	u64 low;
+
+	if (count >= 128) {
+		*lptr = *hptr != 0 || *lptr != 0;
+		*hptr = 0;
+	} else if (count >= 64) {
+		if (count == 64) {
+			*lptr = *hptr | (*lptr != 0);
+		} else {
+			low = *lptr;
+			*lptr = *hptr >> (count - 64);
+			*lptr |= (*hptr << (128 - count)) != 0 || low != 0;
+		}
+		*hptr = 0;
+	} else {
+		low = *lptr;
+		*lptr = low >> count | *hptr << (64 - count);
+		*lptr |= (low << (64 - count)) != 0;
+		*hptr = *hptr >> count;
+	}
+}
+
 static union ieee754dp _dp_maddf(union ieee754dp z, union ieee754dp x,
 				 union ieee754dp y, enum maddf_flags flags)
 {
 	int re;
 	int rs;
-	u64 rm;
 	unsigned lxm;
 	unsigned hxm;
 	unsigned lym;
 	unsigned hym;
 	u64 lrm;
 	u64 hrm;
+	u64 lzm;
+	u64 hzm;
 	u64 t;
 	u64 at;
 	int s;

@@ -172,7 +198,7 @@ static union ieee754dp _dp_maddf(union i
 	ym <<= 64 - (DP_FBITS + 1);
 
 	/*
-	 * Multiply 64 bits xm, ym to give high 64 bits rm with stickness.
+	 * Multiply 64 bits xm and ym to give 128 bits result in hrm:lrm.
 	 */
 
 	/* 32 * 32 => 64 */

@@ -202,81 +228,110 @@ static union ieee754dp _dp_maddf(union i
 
 	hrm = hrm + (t >> 32);
 
-	rm = hrm | (lrm != 0);
-
-	/*
-	 * Sticky shift down to normal rounding precision.
-	 */
-	if ((s64) rm < 0) {
-		rm = (rm >> (64 - (DP_FBITS + 1 + 3))) |
-		     ((rm << (DP_FBITS + 1 + 3)) != 0);
+	/* Put explicit bit at bit 126 if necessary */
+	if ((int64_t)hrm < 0) {
+		lrm = (hrm << 63) | (lrm >> 1);
+		hrm = hrm >> 1;
 		re++;
-	} else {
-		rm = (rm >> (64 - (DP_FBITS + 1 + 3 + 1))) |
-		     ((rm << (DP_FBITS + 1 + 3 + 1)) != 0);
 	}
-	assert(rm & (DP_HIDDEN_BIT << 3));
 
-	if (zc == IEEE754_CLASS_ZERO)
-		return ieee754dp_format(rs, re, rm);
+	assert(hrm & (1 << 62));
 
-	/* And now the addition */
-	assert(zm & DP_HIDDEN_BIT);
+	if (zc == IEEE754_CLASS_ZERO) {
+		/*
+		 * Move explicit bit from bit 126 to bit 55 since the
+		 * ieee754dp_format code expects the mantissa to be
+		 * 56 bits wide (53 + 3 rounding bits).
+		 */
+		srl128(&hrm, &lrm, (126 - 55));
+		return ieee754dp_format(rs, re, lrm);
+	}
 
-	/*
-	 * Provide guard,round and stick bit space.
-	 */
-	zm <<= 3;
+	/* Move explicit bit from bit 52 to bit 126 */
+	lzm = 0;
+	hzm = zm << 10;
+	assert(hzm & (1 << 62));
 
+	/* Make the exponents the same */
 	if (ze > re) {
 		/*
 		 * Have to shift y fraction right to align.
 		 */
 		s = ze - re;
-		rm = XDPSRS(rm, s);
+		srl128(&hrm, &lrm, s);
 		re += s;
 	} else if (re > ze) {
 		/*
 		 * Have to shift x fraction right to align.
 		 */
 		s = re - ze;
-		zm = XDPSRS(zm, s);
+		srl128(&hzm, &lzm, s);
 		ze += s;
 	}
 	assert(ze == re);
 	assert(ze <= DP_EMAX);
 
+	/* Do the addition */
 	if (zs == rs) {
 		/*
-		 * Generate 28 bit result of adding two 27 bit numbers
-		 * leaving result in xm, xs and xe.
+		 * Generate 128 bit result by adding two 127 bit numbers
+		 * leaving result in hzm:lzm, zs and ze.
 		 */
-		zm = zm + rm;
-
-		if (zm >> (DP_FBITS + 1 + 3)) { /* carry out */
-			zm = XDPSRS1(zm);
+		hzm = hzm + hrm + (lzm > (lzm + lrm));
+		lzm = lzm + lrm;
+		if ((int64_t)hzm < 0) {        /* carry out */
+			srl128(&hzm, &lzm, 1);
 			ze++;
 		}
 	} else {
-		if (zm >= rm) {
-			zm = zm - rm;
+		if (hzm > hrm || (hzm == hrm && lzm >= lrm)) {
+			hzm = hzm - hrm - (lzm < lrm);
+			lzm = lzm - lrm;
 		} else {
-			zm = rm - zm;
+			hzm = hrm - hzm - (lrm < lzm);
+			lzm = lrm - lzm;
 			zs = rs;
 		}
-		if (zm == 0)
+		if (lzm == 0 && hzm == 0)
 			return ieee754dp_zero(ieee754_csr.rm == FPU_CSR_RD);
 
 		/*
-		 * Normalize to rounding precision.
+		 * Put explicit bit at bit 126 if necessary.
 		 */
-		while ((zm >> (DP_FBITS + 3)) == 0) {
-			zm <<= 1;
-			ze--;
+		if (hzm == 0) {
+			/* left shift by 63 or 64 bits */
+			if ((int64_t)lzm < 0) {
+				/* MSB of lzm is the explicit bit */
+				hzm = lzm >> 1;
+				lzm = lzm << 63;
+				ze -= 63;
+			} else {
+				hzm = lzm;
+				lzm = 0;
+				ze -= 64;
+			}
+		}
+
+		t = 0;
+		while ((hzm >> (62 - t)) == 0)
+			t++;
+
+		assert(t <= 62);
+		if (t) {
+			hzm = hzm << t | lzm >> (64 - t);
+			lzm = lzm << t;
+			ze -= t;
 		}
 	}
 
-	return ieee754dp_format(zs, ze, zm);
+	/*
+	 * Move explicit bit from bit 126 to bit 55 since the
+	 * ieee754dp_format code expects the mantissa to be
+	 * 56 bits wide (53 + 3 rounding bits).
+	 */
+	srl128(&hzm, &lzm, (126 - 55));
+
+	return ieee754dp_format(zs, ze, lzm);
 }
 
 union ieee754dp ieee754dp_maddf(union ieee754dp z, union ieee754dp x,

@@ -1,8 +1,9 @@
 /*
  * AMD Cryptographic Coprocessor (CCP) AES XTS crypto API support
  *
- * Copyright (C) 2013 Advanced Micro Devices, Inc.
+ * Copyright (C) 2013,2017 Advanced Micro Devices, Inc.
  *
+ * Author: Gary R Hook <[email protected]>
  * Author: Tom Lendacky <[email protected]>
  *
  * This program is free software; you can redistribute it and/or modify

@@ -164,6 +165,7 @@ static int ccp_aes_xts_crypt(struct ablk
 	memset(&rctx->cmd, 0, sizeof(rctx->cmd));
 	INIT_LIST_HEAD(&rctx->cmd.entry);
 	rctx->cmd.engine = CCP_ENGINE_XTS_AES_128;
+	rctx->cmd.u.xts.type = CCP_AES_TYPE_128;
 	rctx->cmd.u.xts.action = (encrypt) ? CCP_AES_ACTION_ENCRYPT
 					   : CCP_AES_ACTION_DECRYPT;
 	rctx->cmd.u.xts.unit_size = unit_size;

@@ -131,6 +131,7 @@ union ccp_function {
 #define	CCP_AES_MODE(p)		((p)->aes.mode)
 #define	CCP_AES_TYPE(p)		((p)->aes.type)
 #define	CCP_XTS_SIZE(p)		((p)->aes_xts.size)
+#define	CCP_XTS_TYPE(p)		((p)->aes_xts.type)
 #define	CCP_XTS_ENCRYPT(p)	((p)->aes_xts.encrypt)
 #define	CCP_SHA_TYPE(p)		((p)->sha.type)
 #define	CCP_RSA_SIZE(p)		((p)->rsa.size)

@@ -318,6 +319,7 @@ static int ccp5_perform_xts_aes(struct c
 	CCP5_CMD_PROT(&desc) = 0;
 
 	function.raw = 0;
+	CCP_XTS_TYPE(&function) = op->u.xts.type;
 	CCP_XTS_ENCRYPT(&function) = op->u.xts.action;
 	CCP_XTS_SIZE(&function) = op->u.xts.unit_size;
 	CCP5_CMD_FUNCTION(&desc) = function.raw;

@@ -187,6 +187,7 @@
 #define CCP_AES_CTX_SB_COUNT		1
 
 #define CCP_XTS_AES_KEY_SB_COUNT	1
+#define CCP5_XTS_AES_KEY_SB_COUNT	2
 #define CCP_XTS_AES_CTX_SB_COUNT	1
 
 #define CCP_SHA_SB_COUNT		1

@@ -472,6 +473,7 @@ struct ccp_aes_op {
 };
 
 struct ccp_xts_aes_op {
+	enum ccp_aes_type type;
 	enum ccp_aes_action action;
 	enum ccp_xts_aes_unit_size unit_size;
 };

70 lines not on this page · raw message

@@ -1,7 +1,7 @@
 /*
  * AMD Cryptographic Coprocessor (CCP) driver
  *
- * Copyright (C) 2013,2016 Advanced Micro Devices, Inc.
+ * Copyright (C) 2013,2017 Advanced Micro Devices, Inc.
  *
  * Author: Tom Lendacky <[email protected]>
  * Author: Gary R Hook <[email protected]>

@@ -222,6 +222,7 @@ enum ccp_xts_aes_unit_size {
  * AES operation the new IV overwrites the old IV.
  */
 struct ccp_xts_aes_engine {
+	enum ccp_aes_type type;
 	enum ccp_aes_action action;
 	enum ccp_xts_aes_unit_size unit_size;
 

@@ -143,8 +143,10 @@ static int skcipher_alloc_sgl(struct soc
 		sg_init_table(sgl->sg, MAX_SGL_ENTS + 1);
 		sgl->cur = 0;
 
-		if (sg)
+		if (sg) {
 			sg_chain(sg, MAX_SGL_ENTS + 1, sgl->sg);
+			sg_unmark_end(sg + (MAX_SGL_ENTS - 1));
+		}
 
 		list_add_tail(&sgl->list, &ctx->tsgl);
 	}

@@ -2372,7 +2372,7 @@ static void ext4_orphan_cleanup(struct s
 #ifdef CONFIG_QUOTA
 	/* Needed for iput() to work correctly and not trash data */
 	sb->s_flags |= MS_ACTIVE;
-	/* Turn on quotas so that they are updated correctly */
+	/* Turn on journaled quotas so that they are updated correctly */
 	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
 		if (EXT4_SB(sb)->s_qf_names[i]) {
 			int ret = ext4_quota_on_mount(sb, i);

@@ -2438,9 +2438,9 @@ static void ext4_orphan_cleanup(struct s
 		ext4_msg(sb, KERN_INFO, "%d truncate%s cleaned up",
 		       PLURAL(nr_truncates));
 #ifdef CONFIG_QUOTA
-	/* Turn quotas off */
+	/* Turn off journaled quotas if they were enabled for orphan cleanup */
 	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
-		if (sb_dqopt(sb)->files[i])
+		if (EXT4_SB(sb)->s_qf_names[i] && sb_dqopt(sb)->files[i])
 			dquot_quota_off(sb, i);
 	}
 #endif

@@ -2334,6 +2334,7 @@ static void ext4_orphan_cleanup(struct s
 	unsigned int s_flags = sb->s_flags;
 	int nr_orphans = 0, nr_truncates = 0;
 #ifdef CONFIG_QUOTA
+	int quota_update = 0;
 	int i;
 #endif
 	if (!es->s_last_orphan) {

@@ -2372,14 +2373,32 @@ static void ext4_orphan_cleanup(struct s
 #ifdef CONFIG_QUOTA
 	/* Needed for iput() to work correctly and not trash data */
 	sb->s_flags |= MS_ACTIVE;
-	/* Turn on journaled quotas so that they are updated correctly */
+
+	/*
+	 * Turn on quotas which were not enabled for read-only mounts if
+	 * filesystem has quota feature, so that they are updated correctly.
+	 */
+	if (ext4_has_feature_quota(sb) && (s_flags & MS_RDONLY)) {
+		int ret = ext4_enable_quotas(sb);
+
+		if (!ret)
+			quota_update = 1;
+		else
+			ext4_msg(sb, KERN_ERR,
+				"Cannot turn on quotas: error %d", ret);
+	}
+
+	/* Turn on journaled quotas used for old sytle */
 	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
 		if (EXT4_SB(sb)->s_qf_names[i]) {
 			int ret = ext4_quota_on_mount(sb, i);
-			if (ret < 0)
+
+			if (!ret)
+				quota_update = 1;
+			else
 				ext4_msg(sb, KERN_ERR,
 					"Cannot turn on journaled "
-					"quota: error %d", ret);
+					"quota: type %d: error %d", i, ret);
 		}
 	}
 #endif

@@ -2438,10 +2457,12 @@ static void ext4_orphan_cleanup(struct s
 		ext4_msg(sb, KERN_INFO, "%d truncate%s cleaned up",
 		       PLURAL(nr_truncates));
 #ifdef CONFIG_QUOTA
-	/* Turn off journaled quotas if they were enabled for orphan cleanup */
-	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
-		if (EXT4_SB(sb)->s_qf_names[i] && sb_dqopt(sb)->files[i])
-			dquot_quota_off(sb, i);
+	/* Turn off quotas if they were enabled for orphan cleanup */
+	if (quota_update) {
+		for (i = 0; i < EXT4_MAXQUOTAS; i++) {
+			if (sb_dqopt(sb)->files[i])
+				dquot_quota_off(sb, i);
+		}
 	}
 #endif
 	sb->s_flags = s_flags; /* Restore MS_RDONLY status */

@@ -5365,6 +5386,9 @@ static int ext4_enable_quotas(struct sup
 				DQUOT_USAGE_ENABLED |
 				(quota_mopt[type] ? DQUOT_LIMITS_ENABLED : 0));
 			if (err) {
+				for (type--; type >= 0; type--)
+					dquot_quota_off(sb, type);
+
 				ext4_warning(sb,
 					"Failed to enable quota tracking "
 					"(type=%d, err=%d). Please run "