Merge tag 'for-linus-5.6b-rc5-tag' of git://git.kernel.org/pub/scm/li…

…nux/kernel/git/xen/tip

Pull xen fixes from Juergen Gross:
 "Four fixes and a small cleanup patch:

   - two fixes by Dongli Zhang fixing races in the xenbus driver

   - two fixes by me fixing issues introduced in 5.6

   - a small cleanup by Gustavo Silva replacing a zero-length array with
     a flexible-array"

* tag 'for-linus-5.6b-rc5-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen/blkfront: fix ring info addressing
  xen/xenbus: fix locking
  xenbus: req->err should be updated before req->state
  xenbus: req->body should be updated before req->state
  xen: Replace zero-length array with flexible-array member

drivers/block/xen-blkfront.c

@@ -213,6 +213,7 @@ struct blkfront_info
 	struct blk_mq_tag_set tag_set;
 	struct blkfront_ring_info *rinfo;
 	unsigned int nr_rings;
+	unsigned int rinfo_size;
 	/* Save uncomplete reqs and bios for migration. */
 	struct list_head requests;
 	struct bio_list bio_list;
@@ -259,6 +260,18 @@ static int blkfront_setup_indirect(struct blkfront_ring_info *rinfo);
 static void blkfront_gather_backend_features(struct blkfront_info *info);
 static int negotiate_mq(struct blkfront_info *info);
 
+#define for_each_rinfo(info, ptr, idx)				\
+	for ((ptr) = (info)->rinfo, (idx) = 0;			\
+	     (idx) < (info)->nr_rings;				\
+	     (idx)++, (ptr) = (void *)(ptr) + (info)->rinfo_size)
+
+static inline struct blkfront_ring_info *
+get_rinfo(const struct blkfront_info *info, unsigned int i)
+{
+	BUG_ON(i >= info->nr_rings);
+	return (void *)info->rinfo + i * info->rinfo_size;
+}
+
 static int get_id_from_freelist(struct blkfront_ring_info *rinfo)
 {
 	unsigned long free = rinfo->shadow_free;
@@ -883,8 +896,7 @@ static blk_status_t blkif_queue_rq(struct blk_mq_hw_ctx *hctx,
 	struct blkfront_info *info = hctx->queue->queuedata;
 	struct blkfront_ring_info *rinfo = NULL;
 
-	BUG_ON(info->nr_rings <= qid);
-	rinfo = &info->rinfo[qid];
+	rinfo = get_rinfo(info, qid);
 	blk_mq_start_request(qd->rq);
 	spin_lock_irqsave(&rinfo->ring_lock, flags);
 	if (RING_FULL(&rinfo->ring))
@@ -1181,16 +1193,15 @@ static int xlvbd_alloc_gendisk(blkif_sector_t capacity,
 static void xlvbd_release_gendisk(struct blkfront_info *info)
 {
 	unsigned int minor, nr_minors, i;
+	struct blkfront_ring_info *rinfo;
 
 	if (info->rq == NULL)
 		return;
 
 	/* No more blkif_request(). */
 	blk_mq_stop_hw_queues(info->rq);
 
-	for (i = 0; i < info->nr_rings; i++) {
-		struct blkfront_ring_info *rinfo = &info->rinfo[i];
-
+	for_each_rinfo(info, rinfo, i) {
 		/* No more gnttab callback work. */
 		gnttab_cancel_free_callback(&rinfo->callback);
 
@@ -1339,6 +1350,7 @@ static void blkif_free_ring(struct blkfront_ring_info *rinfo)
 static void blkif_free(struct blkfront_info *info, int suspend)
 {
 	unsigned int i;
+	struct blkfront_ring_info *rinfo;
 
 	/* Prevent new requests being issued until we fix things up. */
 	info->connected = suspend ?
@@ -1347,8 +1359,8 @@ static void blkif_free(struct blkfront_info *info, int suspend)
 	if (info->rq)
 		blk_mq_stop_hw_queues(info->rq);
 
-	for (i = 0; i < info->nr_rings; i++)
-		blkif_free_ring(&info->rinfo[i]);
+	for_each_rinfo(info, rinfo, i)
+		blkif_free_ring(rinfo);
 
 	kvfree(info->rinfo);
 	info->rinfo = NULL;
@@ -1775,6 +1787,7 @@ static int talk_to_blkback(struct xenbus_device *dev,
 	int err;
 	unsigned int i, max_page_order;
 	unsigned int ring_page_order;
+	struct blkfront_ring_info *rinfo;
 
 	if (!info)
 		return -ENODEV;
@@ -1788,9 +1801,7 @@ static int talk_to_blkback(struct xenbus_device *dev,
 	if (err)
 		goto destroy_blkring;
 
-	for (i = 0; i < info->nr_rings; i++) {
-		struct blkfront_ring_info *rinfo = &info->rinfo[i];
-
+	for_each_rinfo(info, rinfo, i) {
 		/* Create shared ring, alloc event channel. */
 		err = setup_blkring(dev, rinfo);
 		if (err)
@@ -1815,7 +1826,7 @@ static int talk_to_blkback(struct xenbus_device *dev,
 
 	/* We already got the number of queues/rings in _probe */
 	if (info->nr_rings == 1) {
-		err = write_per_ring_nodes(xbt, &info->rinfo[0], dev->nodename);
+		err = write_per_ring_nodes(xbt, info->rinfo, dev->nodename);
 		if (err)
 			goto destroy_blkring;
 	} else {
@@ -1837,10 +1848,10 @@ static int talk_to_blkback(struct xenbus_device *dev,
 			goto abort_transaction;
 		}
 
-		for (i = 0; i < info->nr_rings; i++) {
+		for_each_rinfo(info, rinfo, i) {
 			memset(path, 0, pathsize);
 			snprintf(path, pathsize, "%s/queue-%u", dev->nodename, i);
-			err = write_per_ring_nodes(xbt, &info->rinfo[i], path);
+			err = write_per_ring_nodes(xbt, rinfo, path);
 			if (err) {
 				kfree(path);
 				goto destroy_blkring;
@@ -1868,9 +1879,8 @@ static int talk_to_blkback(struct xenbus_device *dev,
 		goto destroy_blkring;
 	}
 
-	for (i = 0; i < info->nr_rings; i++) {
+	for_each_rinfo(info, rinfo, i) {
 		unsigned int j;
-		struct blkfront_ring_info *rinfo = &info->rinfo[i];
 
 		for (j = 0; j < BLK_RING_SIZE(info); j++)
 			rinfo->shadow[j].req.u.rw.id = j + 1;
@@ -1900,6 +1910,7 @@ static int negotiate_mq(struct blkfront_info *info)
 {
 	unsigned int backend_max_queues;
 	unsigned int i;
+	struct blkfront_ring_info *rinfo;
 
 	BUG_ON(info->nr_rings);
 
@@ -1911,20 +1922,16 @@ static int negotiate_mq(struct blkfront_info *info)
 	if (!info->nr_rings)
 		info->nr_rings = 1;
 
-	info->rinfo = kvcalloc(info->nr_rings,
-			       struct_size(info->rinfo, shadow,
-					   BLK_RING_SIZE(info)),
-			       GFP_KERNEL);
+	info->rinfo_size = struct_size(info->rinfo, shadow,
+				       BLK_RING_SIZE(info));
+	info->rinfo = kvcalloc(info->nr_rings, info->rinfo_size, GFP_KERNEL);
 	if (!info->rinfo) {
 		xenbus_dev_fatal(info->xbdev, -ENOMEM, "allocating ring_info structure");
 		info->nr_rings = 0;
 		return -ENOMEM;
 	}
 
-	for (i = 0; i < info->nr_rings; i++) {
-		struct blkfront_ring_info *rinfo;
-
-		rinfo = &info->rinfo[i];
+	for_each_rinfo(info, rinfo, i) {
 		INIT_LIST_HEAD(&rinfo->indirect_pages);
 		INIT_LIST_HEAD(&rinfo->grants);
 		rinfo->dev_info = info;
@@ -2017,16 +2024,15 @@ static int blkif_recover(struct blkfront_info *info)
 	int rc;
 	struct bio *bio;
 	unsigned int segs;
+	struct blkfront_ring_info *rinfo;
 
 	blkfront_gather_backend_features(info);
 	/* Reset limits changed by blk_mq_update_nr_hw_queues(). */
 	blkif_set_queue_limits(info);
 	segs = info->max_indirect_segments ? : BLKIF_MAX_SEGMENTS_PER_REQUEST;
 	blk_queue_max_segments(info->rq, segs / GRANTS_PER_PSEG);
 
-	for (r_index = 0; r_index < info->nr_rings; r_index++) {
-		struct blkfront_ring_info *rinfo = &info->rinfo[r_index];
-
+	for_each_rinfo(info, rinfo, r_index) {
 		rc = blkfront_setup_indirect(rinfo);
 		if (rc)
 			return rc;
@@ -2036,10 +2042,7 @@ static int blkif_recover(struct blkfront_info *info)
 	/* Now safe for us to use the shared ring */
 	info->connected = BLKIF_STATE_CONNECTED;
 
-	for (r_index = 0; r_index < info->nr_rings; r_index++) {
-		struct blkfront_ring_info *rinfo;
-
-		rinfo = &info->rinfo[r_index];
+	for_each_rinfo(info, rinfo, r_index) {
 		/* Kick any other new requests queued since we resumed */
 		kick_pending_request_queues(rinfo);
 	}
@@ -2072,13 +2075,13 @@ static int blkfront_resume(struct xenbus_device *dev)
 	struct blkfront_info *info = dev_get_drvdata(&dev->dev);
 	int err = 0;
 	unsigned int i, j;
+	struct blkfront_ring_info *rinfo;
 
 	dev_dbg(&dev->dev, "blkfront_resume: %s\n", dev->nodename);
 
 	bio_list_init(&info->bio_list);
 	INIT_LIST_HEAD(&info->requests);
-	for (i = 0; i < info->nr_rings; i++) {
-		struct blkfront_ring_info *rinfo = &info->rinfo[i];
+	for_each_rinfo(info, rinfo, i) {
 		struct bio_list merge_bio;
 		struct blk_shadow *shadow = rinfo->shadow;
 
@@ -2337,6 +2340,7 @@ static void blkfront_connect(struct blkfront_info *info)
 	unsigned int binfo;
 	char *envp[] = { "RESIZE=1", NULL };
 	int err, i;
+	struct blkfront_ring_info *rinfo;
 
 	switch (info->connected) {
 	case BLKIF_STATE_CONNECTED:
@@ -2394,8 +2398,8 @@ static void blkfront_connect(struct blkfront_info *info)
 						    "physical-sector-size",
 						    sector_size);
 	blkfront_gather_backend_features(info);
-	for (i = 0; i < info->nr_rings; i++) {
-		err = blkfront_setup_indirect(&info->rinfo[i]);
+	for_each_rinfo(info, rinfo, i) {
+		err = blkfront_setup_indirect(rinfo);
 		if (err) {
 			xenbus_dev_fatal(info->xbdev, err, "setup_indirect at %s",
 					 info->xbdev->otherend);
@@ -2416,8 +2420,8 @@ static void blkfront_connect(struct blkfront_info *info)
 
 	/* Kick pending requests. */
 	info->connected = BLKIF_STATE_CONNECTED;
-	for (i = 0; i < info->nr_rings; i++)
-		kick_pending_request_queues(&info->rinfo[i]);
+	for_each_rinfo(info, rinfo, i)
+		kick_pending_request_queues(rinfo);
 
 	device_add_disk(&info->xbdev->dev, info->gd, NULL);
 
@@ -2652,9 +2656,9 @@ static void purge_persistent_grants(struct blkfront_info *info)
 {
 	unsigned int i;
 	unsigned long flags;
+	struct blkfront_ring_info *rinfo;
 
-	for (i = 0; i < info->nr_rings; i++) {
-		struct blkfront_ring_info *rinfo = &info->rinfo[i];
+	for_each_rinfo(info, rinfo, i) {
 		struct grant *gnt_list_entry, *tmp;
 
 		spin_lock_irqsave(&rinfo->ring_lock, flags);

drivers/xen/xen-pciback/pciback.h

@@ -52,7 +52,7 @@ struct xen_pcibk_dev_data {
 	unsigned int ack_intr:1; /* .. and ACK-ing */
 	unsigned long handled;
 	unsigned int irq; /* Saved in case device transitions to MSI/MSI-X */
-	char irq_name[0]; /* xen-pcibk[000:04:00.0] */
+	char irq_name[]; /* xen-pcibk[000:04:00.0] */
 };
 
 /* Used by XenBus and xen_pcibk_ops.c */

drivers/xen/xenbus/xenbus_comms.c

@@ -313,6 +313,8 @@ static int process_msg(void)
 			req->msg.type = state.msg.type;
 			req->msg.len = state.msg.len;
 			req->body = state.body;
+			/* write body, then update state */
+			virt_wmb();
 			req->state = xb_req_state_got_reply;
 			req->cb(req);
 		} else
@@ -395,6 +397,8 @@ static int process_writes(void)
 	if (state.req->state == xb_req_state_aborted)
 		kfree(state.req);
 	else {
+		/* write err, then update state */
+		virt_wmb();
 		state.req->state = xb_req_state_got_reply;
 		wake_up(&state.req->wq);
 	}

drivers/xen/xenbus/xenbus_probe.c

@@ -239,9 +239,9 @@ int xenbus_dev_probe(struct device *_dev)
 		goto fail;
 	}
 
-	spin_lock(&dev->reclaim_lock);
+	down(&dev->reclaim_sem);
 	err = drv->probe(dev, id);
-	spin_unlock(&dev->reclaim_lock);
+	up(&dev->reclaim_sem);
 	if (err)
 		goto fail_put;
 
@@ -271,9 +271,9 @@ int xenbus_dev_remove(struct device *_dev)
 	free_otherend_watch(dev);
 
 	if (drv->remove) {
-		spin_lock(&dev->reclaim_lock);
+		down(&dev->reclaim_sem);
 		drv->remove(dev);
-		spin_unlock(&dev->reclaim_lock);
+		up(&dev->reclaim_sem);
 	}
 
 	module_put(drv->driver.owner);
@@ -473,7 +473,7 @@ int xenbus_probe_node(struct xen_bus_type *bus,
 		goto fail;
 
 	dev_set_name(&xendev->dev, "%s", devname);
-	spin_lock_init(&xendev->reclaim_lock);
+	sema_init(&xendev->reclaim_sem, 1);
 
 	/* Register with generic device framework. */
 	err = device_register(&xendev->dev);

drivers/xen/xenbus/xenbus_probe_backend.c

@@ -45,6 +45,7 @@
 #include <linux/mm.h>
 #include <linux/notifier.h>
 #include <linux/export.h>
+#include <linux/semaphore.h>
 
 #include <asm/page.h>
 #include <asm/pgtable.h>
@@ -257,10 +258,10 @@ static int backend_reclaim_memory(struct device *dev, void *data)
 	drv = to_xenbus_driver(dev->driver);
 	if (drv && drv->reclaim_memory) {
 		xdev = to_xenbus_device(dev);
-		if (!spin_trylock(&xdev->reclaim_lock))
+		if (down_trylock(&xdev->reclaim_sem))
 			return 0;
 		drv->reclaim_memory(xdev);
-		spin_unlock(&xdev->reclaim_lock);
+		up(&xdev->reclaim_sem);
 	}
 	return 0;
 }

drivers/xen/xenbus/xenbus_xs.c

@@ -191,8 +191,11 @@ static bool xenbus_ok(void)
 
 static bool test_reply(struct xb_req_data *req)
 {
-	if (req->state == xb_req_state_got_reply || !xenbus_ok())
+	if (req->state == xb_req_state_got_reply || !xenbus_ok()) {
+		/* read req->state before all other fields */
+		virt_rmb();
 		return true;
+	}
 
 	/* Make sure to reread req->state each time. */
 	barrier();
@@ -202,7 +205,7 @@ static bool test_reply(struct xb_req_data *req)
 
 static void *read_reply(struct xb_req_data *req)
 {
-	while (req->state != xb_req_state_got_reply) {
+	do {
 		wait_event(req->wq, test_reply(req));
 
 		if (!xenbus_ok())
@@ -216,7 +219,7 @@ static void *read_reply(struct xb_req_data *req)
 		if (req->err)
 			return ERR_PTR(req->err);
 
-	}
+	} while (req->state != xb_req_state_got_reply);
 
 	return req->body;
 }

include/xen/interface/io/tpmif.h

@@ -46,7 +46,7 @@ struct vtpm_shared_page {
 	uint8_t pad;
 
 	uint8_t nr_extra_pages;  /* extra pages for long packets; may be zero */
-	uint32_t extra_pages[0]; /* grant IDs; length in nr_extra_pages */
+	uint32_t extra_pages[]; /* grant IDs; length in nr_extra_pages */
 };
 
 #endif

include/xen/xenbus.h

@@ -42,6 +42,7 @@
 #include <linux/completion.h>
 #include <linux/init.h>
 #include <linux/slab.h>
+#include <linux/semaphore.h>
 #include <xen/interface/xen.h>
 #include <xen/interface/grant_table.h>
 #include <xen/interface/io/xenbus.h>
@@ -76,7 +77,7 @@ struct xenbus_device {
 	enum xenbus_state state;
 	struct completion down;
 	struct work_struct work;
-	spinlock_t reclaim_lock;
+	struct semaphore reclaim_sem;
 };
 
 static inline struct xenbus_device *to_xenbus_device(struct device *dev)