5.4.287-rc1 review

index 72d62abb6f285..a6ce6b89b271a 100644
@@ -1902,7 +1902,7 @@ static irqreturn_t udc_irq(struct ci_hdrc *ci)
 			}
 		}
 
-		if (USBi_UI  & intr)
+		if ((USBi_UI | USBi_UEI) & intr)
 			isr_tr_complete_handler(ci);
 
 		if ((USBi_SLI & intr) && !(ci->suspended)) {

index 7f4e2c031a9ab..b7ef63614417d 100644
@@ -2856,7 +2856,7 @@ static void __init fixup_device_tree_chrp(void)
 #endif
 
 #if defined(CONFIG_PPC64) && defined(CONFIG_PPC_PMAC)
-static void __init fixup_device_tree_pmac(void)
+static void __init fixup_device_tree_pmac64(void)
 {
 	phandle u3, i2c, mpic;
 	u32 u3_rev;

@@ -2896,7 +2896,31 @@ static void __init fixup_device_tree_pmac(void)
 		     &parent, sizeof(parent));
 }
 #else
-#define fixup_device_tree_pmac()
+#define fixup_device_tree_pmac64()
+#endif
+
+#ifdef CONFIG_PPC_PMAC
+static void __init fixup_device_tree_pmac(void)
+{
+	__be32 val = 1;
+	char type[8];
+	phandle node;
+
+	// Some pmacs are missing #size-cells on escc nodes
+	for (node = 0; prom_next_node(&node); ) {
+		type[0] = '\0';
+		prom_getprop(node, "device_type", type, sizeof(type));
+		if (prom_strcmp(type, "escc"))
+			continue;
+
+		if (prom_getproplen(node, "#size-cells") != PROM_ERROR)
+			continue;
+
+		prom_setprop(node, NULL, "#size-cells", &val, sizeof(val));
+	}
+}
+#else
+static inline void fixup_device_tree_pmac(void) { }
 #endif
 
 #ifdef CONFIG_PPC_EFIKA

@@ -3121,6 +3145,7 @@ static void __init fixup_device_tree(void)
 	fixup_device_tree_maple_memory_controller();
 	fixup_device_tree_chrp();
 	fixup_device_tree_pmac();
+	fixup_device_tree_pmac64();
 	fixup_device_tree_efika();
 	fixup_device_tree_pasemi();
 }

index 36a2eb837371b..6b42ba6705d3f 100644
@@ -186,6 +186,11 @@ void eeprom_93cx6_read(struct eeprom_93cx6 *eeprom, const u8 word,
 	eeprom_93cx6_write_bits(eeprom, command,
 		PCI_EEPROM_WIDTH_OPCODE + eeprom->width);
 
+	if (has_quirk_extra_read_cycle(eeprom)) {
+		eeprom_93cx6_pulse_high(eeprom);
+		eeprom_93cx6_pulse_low(eeprom);
+	}
+
 	/*
 	 * Read the requested 16 bits.
 	 */

@@ -252,6 +257,11 @@ void eeprom_93cx6_readb(struct eeprom_93cx6 *eeprom, const u8 byte,
 	eeprom_93cx6_write_bits(eeprom, command,
 		PCI_EEPROM_WIDTH_OPCODE + eeprom->width + 1);
 
+	if (has_quirk_extra_read_cycle(eeprom)) {
+		eeprom_93cx6_pulse_high(eeprom);
+		eeprom_93cx6_pulse_low(eeprom);
+	}
+
 	/*
 	 * Read the requested 8 bits.
 	 */

index c860c72a921d0..3a485cc0e0fa0 100644
@@ -11,6 +11,8 @@
 	Supported chipsets: 93c46, 93c56 and 93c66.
  */
 
+#include <linux/bits.h>
+
 /*
  * EEPROM operation defines.
  */

@@ -34,6 +36,7 @@
  * @register_write(struct eeprom_93cx6 *eeprom): handler to
  * write to the eeprom register by using all reg_* fields.
  * @width: eeprom width, should be one of the PCI_EEPROM_WIDTH_* defines
+ * @quirks: eeprom or controller quirks
  * @drive_data: Set if we're driving the data line.
  * @reg_data_in: register field to indicate data input
  * @reg_data_out: register field to indicate data output

@@ -50,6 +53,9 @@ struct eeprom_93cx6 {
 	void (*register_write)(struct eeprom_93cx6 *eeprom);
 
 	int width;
+	unsigned int quirks;
+/* Some EEPROMs require an extra clock cycle before reading */
+#define PCI_EEPROM_QUIRK_EXTRA_READ_CYCLE	BIT(0)
 
 	char drive_data;
 	char reg_data_in;

@@ -71,3 +77,8 @@ extern void eeprom_93cx6_wren(struct eeprom_93cx6 *eeprom, bool enable);
 
 extern void eeprom_93cx6_write(struct eeprom_93cx6 *eeprom,
 			       u8 addr, u16 data);
+
+static inline bool has_quirk_extra_read_cycle(struct eeprom_93cx6 *eeprom)
+{
+	return eeprom->quirks & PCI_EEPROM_QUIRK_EXTRA_READ_CYCLE;
+}

index 2370fc31169f5..e63ecb21e0554 100644
@@ -206,7 +206,7 @@ static struct bpf_map *dev_map_alloc(union bpf_attr *attr)
 static void dev_map_free(struct bpf_map *map)
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
-	int i, cpu;
+	int i;
 
 	/* At this point bpf_prog->aux->refcnt == 0 and this map->refcnt == 0,
 	 * so the programs (can be more than one that used this map) were

@@ -226,18 +226,6 @@ static void dev_map_free(struct bpf_map *map)
 	/* Make sure prior __dev_map_entry_free() have completed. */
 	rcu_barrier();
 
-	/* To ensure all pending flush operations have completed wait for flush
-	 * list to empty on _all_ cpus.
-	 * Because the above synchronize_rcu() ensures the map is disconnected
-	 * from the program we can assume no new items will be added.
-	 */
-	for_each_online_cpu(cpu) {
-		struct list_head *flush_list = per_cpu_ptr(dtab->flush_list, cpu);
-
-		while (!list_empty(flush_list))
-			cond_resched();
-	}
-
 	if (dtab->map.map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
 		for (i = 0; i < dtab->n_buckets; i++) {
 			struct bpf_dtab_netdev *dev;

@@ -351,8 +339,7 @@ static int dev_map_hash_get_next_key(struct bpf_map *map, void *key,
 	return -ENOENT;
 }
 
-static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags,
-		       bool in_napi_ctx)
+static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
 {
 	struct bpf_dtab_netdev *obj = bq->obj;
 	struct net_device *dev = obj->dev;

@@ -390,11 +377,7 @@ static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags,
 	for (i = 0; i < bq->count; i++) {
 		struct xdp_frame *xdpf = bq->q[i];
 
-		/* RX path under NAPI protection, can return frames faster */
-		if (likely(in_napi_ctx))
-			xdp_return_frame_rx_napi(xdpf);
-		else
-			xdp_return_frame(xdpf);
+		xdp_return_frame_rx_napi(xdpf);
 		drops++;
 	}
 	goto out;

@@ -415,7 +398,7 @@ void __dev_map_flush(struct bpf_map *map)
 
 	rcu_read_lock();
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node)
-		bq_xmit_all(bq, XDP_XMIT_FLUSH, true);
+		bq_xmit_all(bq, XDP_XMIT_FLUSH);
 	rcu_read_unlock();
 }
 

@@ -446,7 +429,7 @@ static int bq_enqueue(struct bpf_dtab_netdev *obj, struct xdp_frame *xdpf,
 	struct xdp_bulk_queue *bq = this_cpu_ptr(obj->bulkq);
 
 	if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
-		bq_xmit_all(bq, 0, true);
+		bq_xmit_all(bq, 0);
 
 	/* Ingress dev_rx will be the same for all xdp_frame's in
 	 * bulk_queue, because bq stored per-CPU and must be flushed

@@ -515,27 +498,11 @@ static void *dev_map_hash_lookup_elem(struct bpf_map *map, void *key)
 	return dev ? &dev->ifindex : NULL;
 }
 
-static void dev_map_flush_old(struct bpf_dtab_netdev *dev)
-{
-	if (dev->dev->netdev_ops->ndo_xdp_xmit) {
-		struct xdp_bulk_queue *bq;
-		int cpu;
-
-		rcu_read_lock();
-		for_each_online_cpu(cpu) {
-			bq = per_cpu_ptr(dev->bulkq, cpu);
-			bq_xmit_all(bq, XDP_XMIT_FLUSH, false);
-		}
-		rcu_read_unlock();
-	}
-}
-
 static void __dev_map_entry_free(struct rcu_head *rcu)
 {
 	struct bpf_dtab_netdev *dev;
 
 	dev = container_of(rcu, struct bpf_dtab_netdev, rcu);
-	dev_map_flush_old(dev);
 	free_percpu(dev->bulkq);
 	dev_put(dev->dev);
 	kfree(dev);

index e63ecb21e0554..40d4247378592 100644
@@ -206,7 +206,7 @@ static struct bpf_map *dev_map_alloc(union bpf_attr *attr)
 static void dev_map_free(struct bpf_map *map)
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
-	int i;
+	u32 i;
 
 	/* At this point bpf_prog->aux->refcnt == 0 and this map->refcnt == 0,
 	 * so the programs (can be more than one that used this map) were

@@ -512,7 +512,7 @@ static int dev_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
 	struct bpf_dtab_netdev *old_dev;
-	int k = *(u32 *)key;
+	u32 k = *(u32 *)key;
 
 	if (k >= map->max_entries)
 		return -EINVAL;

@@ -535,7 +535,7 @@ static int dev_map_hash_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
 	struct bpf_dtab_netdev *old_dev;
-	int k = *(u32 *)key;
+	u32 k = *(u32 *)key;
 	unsigned long flags;
 	int ret = -ENOENT;
 

@@ -2230,6 +2230,75 @@ static void nfdicf_init(void)
 		file_fail(fold_name);
 }
 
+static void ignore_init(void)
+{
+	FILE *file;
+	unsigned int unichar;
+	unsigned int first;
+	unsigned int last;
+	unsigned int *um;
+	int count;
+	int ret;
+
+	if (verbose > 0)
+		printf("Parsing %s\n", prop_name);
+	file = fopen(prop_name, "r");
+	if (!file)
+		open_fail(prop_name, errno);
+	assert(file);
+	count = 0;
+	while (fgets(line, LINESIZE, file)) {
+		ret = sscanf(line, "%X..%X ; %s # ", &first, &last, buf0);
+		if (ret == 3) {
+			if (strcmp(buf0, "Default_Ignorable_Code_Point"))
+				continue;
+			if (!utf32valid(first) || !utf32valid(last))
+				line_fail(prop_name, line);
+			for (unichar = first; unichar <= last; unichar++) {
+				free(unicode_data[unichar].utf32nfdi);
+				um = malloc(sizeof(unsigned int));
+				*um = 0;
+				unicode_data[unichar].utf32nfdi = um;
+				free(unicode_data[unichar].utf32nfdicf);
+				um = malloc(sizeof(unsigned int));
+				*um = 0;
+				unicode_data[unichar].utf32nfdicf = um;
+				count++;
+			}
+			if (verbose > 1)
+				printf(" %X..%X Default_Ignorable_Code_Point\n",
+					first, last);
+			continue;
+		}
+		ret = sscanf(line, "%X ; %s # ", &unichar, buf0);
+		if (ret == 2) {
+			if (strcmp(buf0, "Default_Ignorable_Code_Point"))
+				continue;
+			if (!utf32valid(unichar))
+				line_fail(prop_name, line);
+			free(unicode_data[unichar].utf32nfdi);
+			um = malloc(sizeof(unsigned int));
+			*um = 0;
+			unicode_data[unichar].utf32nfdi = um;
+			free(unicode_data[unichar].utf32nfdicf);
+			um = malloc(sizeof(unsigned int));
+			*um = 0;
+			unicode_data[unichar].utf32nfdicf = um;
+			if (verbose > 1)
+				printf(" %X Default_Ignorable_Code_Point\n",
+					unichar);
+			count++;
+			continue;
+		}
+	}
+	fclose(file);
+
+	if (verbose > 0)
+		printf("Found %d entries\n", count);
+	if (count == 0)
+		file_fail(prop_name);
+}
+
 static void corrections_init(void)
 {
 	FILE *file;

@@ -3327,6 +3396,7 @@ int main(int argc, char *argv[])
 	ccc_init();
 	nfdi_init();
 	nfdicf_init();
+	ignore_init();
 	corrections_init();
 	hangul_decompose();
 	nfdi_decompose();

@@ -782,11 +782,13 @@ static void pt_buffer_advance(struct pt_
 	buf->cur_idx++;
 
 	if (buf->cur_idx == buf->cur->last) {
-		if (buf->cur == buf->last)
+		if (buf->cur == buf->last) {
 			buf->cur = buf->first;
-		else
+			buf->wrapped = true;
+		} else {
 			buf->cur = list_entry(buf->cur->list.next, struct topa,
 					      list);
+		}
 		buf->cur_idx = 0;
 	}
 }

@@ -800,8 +802,11 @@ static void pt_buffer_advance(struct pt_
 static void pt_update_head(struct pt *pt)
 {
 	struct pt_buffer *buf = perf_get_aux(&pt->handle);
+	bool wrapped = buf->wrapped;
 	u64 topa_idx, base, old;
 
+	buf->wrapped = false;
+
 	/* offset of the first region in this table from the beginning of buf */
 	base = buf->cur->offset + buf->output_off;
 

@@ -814,7 +819,7 @@ static void pt_update_head(struct pt *pt
 	} else {
 		old = (local64_xchg(&buf->head, base) &
 		       ((buf->nr_pages << PAGE_SHIFT) - 1));
-		if (base < old)
+		if (base < old || (base == old && wrapped))
 			base += buf->nr_pages << PAGE_SHIFT;
 
 		local_add(base - old, &buf->data_size);

@@ -64,6 +64,7 @@ struct pt_pmu {
  * @lost:	if data was lost/truncated
  * @head:	logical write offset inside the buffer
  * @snapshot:	if this is for a snapshot/overwrite counter
+ * @wrapped:	buffer advance wrapped back to the first topa table
  * @stop_pos:	STOP topa entry index
  * @intr_pos:	INT topa entry index
  * @stop_te:	STOP topa entry pointer

@@ -80,6 +81,7 @@ struct pt_buffer {
 	local_t			data_size;
 	local64_t		head;
 	bool			snapshot;
+	bool			wrapped;
 	long			stop_pos, intr_pos;
 	struct topa_entry	*stop_te, *intr_te;
 	void			**data_pages;

@@ -2134,7 +2134,6 @@ static int scan_its_table(struct vgic_it
 static int vgic_its_save_ite(struct vgic_its *its, struct its_device *dev,
 			      struct its_ite *ite, gpa_t gpa, int ite_esz)
 {
-	struct kvm *kvm = its->dev->kvm;
 	u32 next_offset;
 	u64 val;
 

@@ -2143,7 +2142,8 @@ static int vgic_its_save_ite(struct vgic
 	       ((u64)ite->irq->intid << KVM_ITS_ITE_PINTID_SHIFT) |
 		ite->collection->collection_id;
 	val = cpu_to_le64(val);
-	return kvm_write_guest_lock(kvm, gpa, &val, ite_esz);
+
+	return vgic_its_write_entry_lock(its, gpa, val, ite_esz);
 }
 
 /**

@@ -2279,7 +2279,6 @@ static int vgic_its_restore_itt(struct v
 static int vgic_its_save_dte(struct vgic_its *its, struct its_device *dev,
 			     gpa_t ptr, int dte_esz)
 {
-	struct kvm *kvm = its->dev->kvm;
 	u64 val, itt_addr_field;
 	u32 next_offset;
 

@@ -2290,7 +2289,8 @@ static int vgic_its_save_dte(struct vgic
 	       (itt_addr_field << KVM_ITS_DTE_ITTADDR_SHIFT) |
 		(dev->num_eventid_bits - 1));
 	val = cpu_to_le64(val);
-	return kvm_write_guest_lock(kvm, ptr, &val, dte_esz);
+
+	return vgic_its_write_entry_lock(its, ptr, val, dte_esz);
 }
 
 /**

@@ -2470,7 +2470,8 @@ static int vgic_its_save_cte(struct vgic
 	       ((u64)collection->target_addr << KVM_ITS_CTE_RDBASE_SHIFT) |
 	       collection->collection_id);
 	val = cpu_to_le64(val);
-	return kvm_write_guest_lock(its->dev->kvm, gpa, &val, esz);
+
+	return vgic_its_write_entry_lock(its, gpa, val, esz);
 }
 
 static int vgic_its_restore_cte(struct vgic_its *its, gpa_t gpa, int esz)

@@ -2481,8 +2482,7 @@ static int vgic_its_restore_cte(struct v
 	u64 val;
 	int ret;
 
-	BUG_ON(esz > sizeof(val));
-	ret = kvm_read_guest_lock(kvm, gpa, &val, esz);
+	ret = vgic_its_read_entry_lock(its, gpa, &val, esz);
 	if (ret)
 		return ret;
 	val = le64_to_cpu(val);

@@ -2516,7 +2516,6 @@ static int vgic_its_save_collection_tabl
 	u64 baser = its->baser_coll_table;
 	gpa_t gpa = GITS_BASER_ADDR_48_to_52(baser);
 	struct its_collection *collection;
-	u64 val;
 	size_t max_size, filled = 0;
 	int ret, cte_esz = abi->cte_esz;
 

@@ -2540,10 +2539,7 @@ static int vgic_its_save_collection_tabl
 	 * table is not fully filled, add a last dummy element
 	 * with valid bit unset
 	 */
-	val = 0;
-	BUG_ON(cte_esz > sizeof(val));
-	ret = kvm_write_guest_lock(its->dev->kvm, gpa, &val, cte_esz);
-	return ret;
+	return vgic_its_write_entry_lock(its, gpa, 0, cte_esz);
 }
 
 /**

@@ -6,6 +6,7 @@
 #define __KVM_ARM_VGIC_NEW_H__
 
 #include <linux/irqchip/arm-gic-common.h>
+#include <asm/kvm_mmu.h>
 
 #define PRODUCT_ID_KVM		0x4b	/* ASCII code K */
 #define IMPLEMENTER_ARM		0x43b

@@ -126,6 +127,29 @@ static inline bool vgic_irq_is_multi_sgi
 	return vgic_irq_get_lr_count(irq) > 1;
 }
 
+static inline int vgic_its_read_entry_lock(struct vgic_its *its, gpa_t eaddr,
+					   u64 *eval, unsigned long esize)
+{
+	struct kvm *kvm = its->dev->kvm;
+
+	if (KVM_BUG_ON(esize != sizeof(*eval), kvm))
+		return -EINVAL;
+
+	return kvm_read_guest_lock(kvm, eaddr, eval, esize);
+
+}
+
+static inline int vgic_its_write_entry_lock(struct vgic_its *its, gpa_t eaddr,
+					    u64 eval, unsigned long esize)
+{
+	struct kvm *kvm = its->dev->kvm;
+
+	if (KVM_BUG_ON(esize != sizeof(eval), kvm))
+		return -EINVAL;
+
+	return kvm_write_guest_lock(kvm, eaddr, &eval, esize);
+}
+
 /*
  * This struct provides an intermediate representation of the fields contained
  * in the GICH_VMCR and ICH_VMCR registers, such that code exporting the GIC

@@ -1181,9 +1181,11 @@ static int vgic_its_cmd_handle_mapd(stru
 	bool valid = its_cmd_get_validbit(its_cmd);
 	u8 num_eventid_bits = its_cmd_get_size(its_cmd);
 	gpa_t itt_addr = its_cmd_get_ittaddr(its_cmd);
+	int dte_esz = vgic_its_get_abi(its)->dte_esz;
 	struct its_device *device;
+	gpa_t gpa;
 
-	if (!vgic_its_check_id(its, its->baser_device_table, device_id, NULL))
+	if (!vgic_its_check_id(its, its->baser_device_table, device_id, &gpa))
 		return E_ITS_MAPD_DEVICE_OOR;
 
 	if (valid && num_eventid_bits > VITS_TYPER_IDBITS)

@@ -1204,7 +1206,7 @@ static int vgic_its_cmd_handle_mapd(stru
 	 * is an error, so we are done in any case.
 	 */
 	if (!valid)
-		return 0;
+		return vgic_its_write_entry_lock(its, gpa, 0, dte_esz);
 
 	device = vgic_its_alloc_device(its, device_id, itt_addr,
 				       num_eventid_bits);

@@ -854,6 +854,9 @@ static int vgic_its_cmd_handle_discard(s
 
 	ite = find_ite(its, device_id, event_id);
 	if (ite && ite->collection) {
+		struct its_device *device = find_its_device(its, device_id);
+		int ite_esz = vgic_its_get_abi(its)->ite_esz;
+		gpa_t gpa = device->itt_addr + ite->event_id * ite_esz;
 		/*
 		 * Though the spec talks about removing the pending state, we
 		 * don't bother here since we clear the ITTE anyway and the

@@ -862,7 +865,8 @@ static int vgic_its_cmd_handle_discard(s
 		vgic_its_invalidate_cache(kvm);
 
 		its_free_ite(kvm, ite);
-		return 0;
+
+		return vgic_its_write_entry_lock(its, gpa, 0, ite_esz);
 	}
 
 	return E_ITS_DISCARD_UNMAPPED_INTERRUPT;

@@ -95,6 +95,9 @@ static int jffs2_rtime_decompress(unsign
 
 		positions[value]=outpos;
 		if (repeat) {
+			if ((outpos + repeat) >= destlen) {
+				return 1;
+			}
 			if (backoffs + repeat >= outpos) {
 				while(repeat) {
 					cpage_out[outpos++] = cpage_out[backoffs++];

@@ -95,7 +95,7 @@ static int jffs2_rtime_decompress(unsign
 
 		positions[value]=outpos;
 		if (repeat) {
-			if ((outpos + repeat) >= destlen) {
+			if ((outpos + repeat) > destlen) {
 				return 1;
 			}
 			if (backoffs + repeat >= outpos) {

@@ -1010,25 +1010,6 @@ static int ocfs2_sync_local_to_main(stru
 		start = bit_off + 1;
 	}
 
-	/* clear the contiguous bits until the end boundary */
-	if (count) {
-		blkno = la_start_blk +
-			ocfs2_clusters_to_blocks(osb->sb,
-					start - count);
-
-		trace_ocfs2_sync_local_to_main_free(
-				count, start - count,
-				(unsigned long long)la_start_blk,
-				(unsigned long long)blkno);
-
-		status = ocfs2_release_clusters(handle,
-				main_bm_inode,
-				main_bm_bh, blkno,
-				count);
-		if (status < 0)
-			mlog_errno(status);
-	}
-
 bail:
 	if (status)
 		mlog_errno(status);

@@ -963,7 +963,7 @@ static void check_section(const char *mo
 		".kprobes.text", ".cpuidle.text", ".noinstr.text"
 #define OTHER_TEXT_SECTIONS ".ref.text", ".head.text", ".spinlock.text", \
 		".fixup", ".entry.text", ".exception.text", ".text.*", \
-		".coldtext"
+		".coldtext", ".irqentry.text"
 
 #define INIT_SECTIONS      ".init.*"
 #define MEM_INIT_SECTIONS  ".meminit.*"

@@ -395,7 +395,7 @@ static ssize_t amdgpu_debugfs_regs_smc_r
 	if (!adev->smc_rreg)
 		return -EOPNOTSUPP;
 
-	if (size > 4096 || size & 0x3 || *pos & 0x3)
+	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 
 	while (size) {

@@ -66,18 +66,26 @@ static void rockchip_pcie_clear_ep_ob_at
 			    ROCKCHIP_PCIE_AT_OB_REGION_CPU_ADDR1(region));
 }
 
+static int rockchip_pcie_ep_ob_atu_num_bits(struct rockchip_pcie *rockchip,
+					    u64 pci_addr, size_t size)
+{
+	int num_pass_bits = fls64(pci_addr ^ (pci_addr + size - 1));
+
+	return clamp(num_pass_bits,
+		     ROCKCHIP_PCIE_AT_MIN_NUM_BITS,
+		     ROCKCHIP_PCIE_AT_MAX_NUM_BITS);
+}
+
 static void rockchip_pcie_prog_ep_ob_atu(struct rockchip_pcie *rockchip, u8 fn,
 					 u32 r, u32 type, u64 cpu_addr,
 					 u64 pci_addr, size_t size)
 {
-	u64 sz = 1ULL << fls64(size - 1);
-	int num_pass_bits = ilog2(sz);
+	int num_pass_bits;
 	u32 addr0, addr1, desc0, desc1;
 	bool is_nor_msg = (type == AXI_WRAPPER_NOR_MSG);
 
-	/* The minimal region size is 1MB */
-	if (num_pass_bits < 8)
-		num_pass_bits = 8;
+	num_pass_bits = rockchip_pcie_ep_ob_atu_num_bits(rockchip,
+							 pci_addr, size);
 
 	cpu_addr -= rockchip->mem_res->start;
 	addr0 = ((is_nor_msg ? 0x10 : (num_pass_bits - 1)) &

@@ -241,6 +241,10 @@
 #define   ROCKCHIP_PCIE_EP_MSI_CTRL_MASK_MSI_CAP	BIT(24)
 #define ROCKCHIP_PCIE_EP_DUMMY_IRQ_ADDR				0x1
 #define ROCKCHIP_PCIE_EP_FUNC_BASE(fn)	(((fn) << 12) & GENMASK(19, 12))
+
+#define ROCKCHIP_PCIE_AT_MIN_NUM_BITS  8
+#define ROCKCHIP_PCIE_AT_MAX_NUM_BITS  20
+
 #define ROCKCHIP_PCIE_AT_IB_EP_FUNC_BAR_ADDR0(fn, bar) \
 	(PCIE_RC_RP_ATS_BASE + 0x0840 + (fn) * 0x0040 + (bar) * 0x0008)
 #define ROCKCHIP_PCIE_AT_IB_EP_FUNC_BAR_ADDR1(fn, bar) \

@@ -21,6 +21,10 @@
 #include "clock.h"
 #include "quirks.h"
 
+/* check whether the descriptor bLength has the minimal length */
+#define DESC_LENGTH_CHECK(p) \
+	 (p->bLength >= sizeof(*p))
+
 static void *find_uac_clock_desc(struct usb_host_interface *iface, int id,
 				 bool (*validator)(void *, int), u8 type)
 {

@@ -38,36 +42,60 @@ static void *find_uac_clock_desc(struct
 static bool validate_clock_source_v2(void *p, int id)
 {
 	struct uac_clock_source_descriptor *cs = p;
+	if (!DESC_LENGTH_CHECK(cs))
+		return false;
 	return cs->bClockID == id;
 }
 
 static bool validate_clock_source_v3(void *p, int id)
 {
 	struct uac3_clock_source_descriptor *cs = p;
+	if (!DESC_LENGTH_CHECK(cs))
+		return false;
 	return cs->bClockID == id;
 }
 
 static bool validate_clock_selector_v2(void *p, int id)
 {
 	struct uac_clock_selector_descriptor *cs = p;
-	return cs->bClockID == id;
+	if (!DESC_LENGTH_CHECK(cs))
+		return false;
+	if (cs->bClockID != id)
+		return false;
+	/* additional length check for baCSourceID array (in bNrInPins size)
+	 * and two more fields (which sizes depend on the protocol)
+	 */
+	return cs->bLength >= sizeof(*cs) + cs->bNrInPins +
+		1 /* bmControls */ + 1 /* iClockSelector */;
 }
 
 static bool validate_clock_selector_v3(void *p, int id)
 {
 	struct uac3_clock_selector_descriptor *cs = p;
-	return cs->bClockID == id;
+	if (!DESC_LENGTH_CHECK(cs))
+		return false;
+	if (cs->bClockID != id)
+		return false;
+	/* additional length check for baCSourceID array (in bNrInPins size)
+	 * and two more fields (which sizes depend on the protocol)
+	 */
+	return cs->bLength >= sizeof(*cs) + cs->bNrInPins +
+		4 /* bmControls */ + 2 /* wCSelectorDescrStr */;
 }
 
 static bool validate_clock_multiplier_v2(void *p, int id)
 {
 	struct uac_clock_multiplier_descriptor *cs = p;
+	if (!DESC_LENGTH_CHECK(cs))
+		return false;
 	return cs->bClockID == id;
 }
 
 static bool validate_clock_multiplier_v3(void *p, int id)
 {
 	struct uac3_clock_multiplier_descriptor *cs = p;
+	if (!DESC_LENGTH_CHECK(cs))
+		return false;
 	return cs->bClockID == id;
 }
 

@@ -210,10 +210,12 @@ static void dev_map_free(struct bpf_map
 
 	/* At this point bpf_prog->aux->refcnt == 0 and this map->refcnt == 0,
 	 * so the programs (can be more than one that used this map) were
-	 * disconnected from events. Wait for outstanding critical sections in
-	 * these programs to complete. The rcu critical section only guarantees
-	 * no further reads against netdev_map. It does __not__ ensure pending
-	 * flush operations (if any) are complete.
+	 * disconnected from events. The following synchronize_rcu() guarantees
+	 * both rcu read critical sections complete and waits for
+	 * preempt-disable regions (NAPI being the relevant context here) so we
+	 * are certain there will be no further reads against the netdev_map and
+	 * all flush operations are complete. Flush operations can only be done
+	 * from NAPI context for this reason.
 	 */
 
 	spin_lock(&dev_map_lock);

@@ -518,12 +520,11 @@ static int dev_map_delete_elem(struct bp
 		return -EINVAL;
 
 	/* Use call_rcu() here to ensure any rcu critical sections have
-	 * completed, but this does not guarantee a flush has happened
-	 * yet. Because driver side rcu_read_lock/unlock only protects the
-	 * running XDP program. However, for pending flush operations the
-	 * dev and ctx are stored in another per cpu map. And additionally,
-	 * the driver tear down ensures all soft irqs are complete before
-	 * removing the net device in the case of dev_put equals zero.
+	 * completed as well as any flush operations because call_rcu
+	 * will wait for preempt-disable region to complete, NAPI in this
+	 * context.  And additionally, the driver tear down ensures all
+	 * soft irqs are complete before removing the net device in the
+	 * case of dev_put equals zero.
 	 */
 	old_dev = xchg(&dtab->netdev_map[k], NULL);
 	if (old_dev)