diff --git a/Documentation/ABI/testing/debugfs-driver-habanalabs b/Documentation/ABI/testing/debugfs-driver-habanalabs
index f6d9c2a8d528..2e9ae311e02d 100644
--- a/Documentation/ABI/testing/debugfs-driver-habanalabs
+++ b/Documentation/ABI/testing/debugfs-driver-habanalabs
@@ -16,7 +16,16 @@ Description:    Allow the root user to disable/enable in runtime the clock
                 gating mechanism in Gaudi. Due to how Gaudi is built, the
                 clock gating needs to be disabled in order to access the
                 registers of the TPC and MME engines. This is sometimes needed
-                during debug by the user and hence the user needs this option
+                during debug by the user and hence the user needs this option.
+                The user can supply a bitmask value, each bit represents
+                a different engine to disable/enable its clock gating feature.
+                The bitmask is composed of 20 bits:
+                0  -  7 : DMA channels
+                8  - 11 : MME engines
+                12 - 19 : TPC engines
+                The bit's location of a specific engine can be determined
+                using (1 << GAUDI_ENGINE_ID_*). GAUDI_ENGINE_ID_* values
+                are defined in uapi habanalabs.h file in enum gaudi_engine_id
 
 What:           /sys/kernel/debug/habanalabs/hl<n>/command_buffers
 Date:           Jan 2019
diff --git a/Makefile b/Makefile
index 73b9f02ed49a..5f2a7750248b 100644
--- a/Makefile
+++ b/Makefile
@@ -2,7 +2,7 @@
 VERSION = 5
 PATCHLEVEL = 8
 SUBLEVEL = 0
-EXTRAVERSION = -rc6
+EXTRAVERSION = -rc7
 NAME = Kleptomaniac Octopus
 
 # *DOCUMENTATION*
@@ -571,7 +571,7 @@ ifeq ($(shell $(srctree)/scripts/clang-android.sh $(CC) $(CLANG_FLAGS)), y)
 $(error "Clang with Android --target detected. Did you specify CLANG_TRIPLE?")
 endif
 GCC_TOOLCHAIN_DIR := $(dir $(shell which $(CROSS_COMPILE)elfedit))
-CLANG_FLAGS	+= --prefix=$(GCC_TOOLCHAIN_DIR)
+CLANG_FLAGS	+= --prefix=$(GCC_TOOLCHAIN_DIR)$(notdir $(CROSS_COMPILE))
 GCC_TOOLCHAIN	:= $(realpath $(GCC_TOOLCHAIN_DIR)/..)
 endif
 ifneq ($(GCC_TOOLCHAIN),)
@@ -1763,7 +1763,7 @@ PHONY += descend $(build-dirs)
 descend: $(build-dirs)
 $(build-dirs): prepare
 	$(Q)$(MAKE) $(build)=$@ \
-	single-build=$(if $(filter-out $@/, $(filter $@/%, $(single-no-ko))),1) \
+	single-build=$(if $(filter-out $@/, $(filter $@/%, $(KBUILD_SINGLE_TARGETS))),1) \
 	need-builtin=1 need-modorder=1
 
 clean-dirs := $(addprefix _clean_, $(clean-dirs))
diff --git a/arch/parisc/include/asm/atomic.h b/arch/parisc/include/asm/atomic.h
index 118953d41763..6dd4171c9530 100644
--- a/arch/parisc/include/asm/atomic.h
+++ b/arch/parisc/include/asm/atomic.h
@@ -212,6 +212,8 @@ atomic64_set(atomic64_t *v, s64 i)
 	_atomic_spin_unlock_irqrestore(v, flags);
 }
 
+#define atomic64_set_release(v, i)	atomic64_set((v), (i))
+
 static __inline__ s64
 atomic64_read(const atomic64_t *v)
 {
diff --git a/arch/parisc/include/asm/cmpxchg.h b/arch/parisc/include/asm/cmpxchg.h
index ab5c215cf46c..068958575871 100644
--- a/arch/parisc/include/asm/cmpxchg.h
+++ b/arch/parisc/include/asm/cmpxchg.h
@@ -60,6 +60,7 @@ extern void __cmpxchg_called_with_bad_pointer(void);
 extern unsigned long __cmpxchg_u32(volatile unsigned int *m, unsigned int old,
 				   unsigned int new_);
 extern u64 __cmpxchg_u64(volatile u64 *ptr, u64 old, u64 new_);
+extern u8 __cmpxchg_u8(volatile u8 *ptr, u8 old, u8 new_);
 
 /* don't worry...optimizer will get rid of most of this */
 static inline unsigned long
@@ -71,6 +72,7 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new_, int size)
 #endif
 	case 4: return __cmpxchg_u32((unsigned int *)ptr,
 				     (unsigned int)old, (unsigned int)new_);
+	case 1: return __cmpxchg_u8((u8 *)ptr, (u8)old, (u8)new_);
 	}
 	__cmpxchg_called_with_bad_pointer();
 	return old;
diff --git a/arch/parisc/lib/bitops.c b/arch/parisc/lib/bitops.c
index 70ffbcf889b8..2e4d1f05a926 100644
--- a/arch/parisc/lib/bitops.c
+++ b/arch/parisc/lib/bitops.c
@@ -79,3 +79,15 @@ unsigned long __cmpxchg_u32(volatile unsigned int *ptr, unsigned int old, unsign
 	_atomic_spin_unlock_irqrestore(ptr, flags);
 	return (unsigned long)prev;
 }
+
+u8 __cmpxchg_u8(volatile u8 *ptr, u8 old, u8 new)
+{
+	unsigned long flags;
+	u8 prev;
+
+	_atomic_spin_lock_irqsave(ptr, flags);
+	if ((prev = *ptr) == old)
+		*ptr = new;
+	_atomic_spin_unlock_irqrestore(ptr, flags);
+	return prev;
+}
diff --git a/drivers/android/binder_alloc.c b/drivers/android/binder_alloc.c
index 42c672f1584e..cbe6aa77d50d 100644
--- a/drivers/android/binder_alloc.c
+++ b/drivers/android/binder_alloc.c
@@ -947,7 +947,7 @@ enum lru_status binder_alloc_free_page(struct list_head *item,
 		trace_binder_unmap_user_end(alloc, index);
 	}
 	mmap_read_unlock(mm);
-	mmput(mm);
+	mmput_async(mm);
 
 	trace_binder_unmap_kernel_start(alloc, index);
 
diff --git a/drivers/base/property.c b/drivers/base/property.c
index 1e6d75e65938..d58aa98fe964 100644
--- a/drivers/base/property.c
+++ b/drivers/base/property.c
@@ -721,7 +721,7 @@ struct fwnode_handle *device_get_next_child_node(struct device *dev,
 		return next;
 
 	/* When no more children in primary, continue with secondary */
-	if (!IS_ERR_OR_NULL(fwnode->secondary))
+	if (fwnode && !IS_ERR_OR_NULL(fwnode->secondary))
 		next = fwnode_get_next_child_node(fwnode->secondary, child);
 
 	return next;
diff --git a/drivers/char/mem.c b/drivers/char/mem.c
index 934c92dcb9ab..687d4af6945d 100644
--- a/drivers/char/mem.c
+++ b/drivers/char/mem.c
@@ -814,7 +814,8 @@ static struct inode *devmem_inode;
 #ifdef CONFIG_IO_STRICT_DEVMEM
 void revoke_devmem(struct resource *res)
 {
-	struct inode *inode = READ_ONCE(devmem_inode);
+	/* pairs with smp_store_release() in devmem_init_inode() */
+	struct inode *inode = smp_load_acquire(&devmem_inode);
 
 	/*
 	 * Check that the initialization has completed. Losing the race
@@ -1028,8 +1029,11 @@ static int devmem_init_inode(void)
 		return rc;
 	}
 
-	/* publish /dev/mem initialized */
-	WRITE_ONCE(devmem_inode, inode);
+	/*
+	 * Publish /dev/mem initialized.
+	 * Pairs with smp_load_acquire() in revoke_devmem().
+	 */
+	smp_store_release(&devmem_inode, inode);
 
 	return 0;
 }
diff --git a/drivers/fpga/dfl-afu-main.c b/drivers/fpga/dfl-afu-main.c
index b0c31789a909..3fa2c5992173 100644
--- a/drivers/fpga/dfl-afu-main.c
+++ b/drivers/fpga/dfl-afu-main.c
@@ -83,7 +83,8 @@ int __afu_port_disable(struct platform_device *pdev)
 	 * on this port and minimum soft reset pulse width has elapsed.
 	 * Driver polls port_soft_reset_ack to determine if reset done by HW.
 	 */
-	if (readq_poll_timeout(base + PORT_HDR_CTRL, v, v & PORT_CTRL_SFTRST,
+	if (readq_poll_timeout(base + PORT_HDR_CTRL, v,
+			       v & PORT_CTRL_SFTRST_ACK,
 			       RST_POLL_INVL, RST_POLL_TIMEOUT)) {
 		dev_err(&pdev->dev, "timeout, fail to reset device\n");
 		return -ETIMEDOUT;
diff --git a/drivers/fpga/dfl-pci.c b/drivers/fpga/dfl-pci.c
index 538755062ab7..a78c409bf2c4 100644
--- a/drivers/fpga/dfl-pci.c
+++ b/drivers/fpga/dfl-pci.c
@@ -227,7 +227,6 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
 {
 	struct cci_drvdata *drvdata = pci_get_drvdata(pcidev);
 	struct dfl_fpga_cdev *cdev = drvdata->cdev;
-	int ret = 0;
 
 	if (!num_vfs) {
 		/*
@@ -239,6 +238,8 @@ static int cci_pci_sriov_configure(struct pci_dev *pcidev, int num_vfs)
 		dfl_fpga_cdev_config_ports_pf(cdev);
 
 	} else {
+		int ret;
+
 		/*
 		 * before enable SRIOV, put released ports into VF access mode
 		 * first of all.
diff --git a/drivers/interconnect/core.c b/drivers/interconnect/core.c
index e5f998744501..9e1ab701785c 100644
--- a/drivers/interconnect/core.c
+++ b/drivers/interconnect/core.c
@@ -243,6 +243,7 @@ static int aggregate_requests(struct icc_node *node)
 {
 	struct icc_provider *p = node->provider;
 	struct icc_req *r;
+	u32 avg_bw, peak_bw;
 
 	node->avg_bw = 0;
 	node->peak_bw = 0;
@@ -251,9 +252,14 @@ static int aggregate_requests(struct icc_node *node)
 		p->pre_aggregate(node);
 
 	hlist_for_each_entry(r, &node->req_list, req_node) {
-		if (!r->enabled)
-			continue;
-		p->aggregate(node, r->tag, r->avg_bw, r->peak_bw,
+		if (r->enabled) {
+			avg_bw = r->avg_bw;
+			peak_bw = r->peak_bw;
+		} else {
+			avg_bw = 0;
+			peak_bw = 0;
+		}
+		p->aggregate(node, r->tag, avg_bw, peak_bw,
 			     &node->avg_bw, &node->peak_bw);
 	}
 
diff --git a/drivers/interconnect/qcom/msm8916.c b/drivers/interconnect/qcom/msm8916.c
index e94f3c5228b7..42c6c5581662 100644
--- a/drivers/interconnect/qcom/msm8916.c
+++ b/drivers/interconnect/qcom/msm8916.c
@@ -197,13 +197,13 @@ DEFINE_QNODE(pcnoc_int_0, MSM8916_PNOC_INT_0, 8, -1, -1, MSM8916_PNOC_SNOC_MAS,
 DEFINE_QNODE(pcnoc_int_1, MSM8916_PNOC_INT_1, 8, -1, -1, MSM8916_PNOC_SNOC_MAS);
 DEFINE_QNODE(pcnoc_m_0, MSM8916_PNOC_MAS_0, 8, -1, -1, MSM8916_PNOC_INT_0);
 DEFINE_QNODE(pcnoc_m_1, MSM8916_PNOC_MAS_1, 8, -1, -1, MSM8916_PNOC_SNOC_MAS);
-DEFINE_QNODE(pcnoc_s_0, MSM8916_PNOC_SLV_0, 8, -1, -1, MSM8916_SLAVE_CLK_CTL, MSM8916_SLAVE_TLMM, MSM8916_SLAVE_TCSR, MSM8916_SLAVE_SECURITY, MSM8916_SLAVE_MSS);
-DEFINE_QNODE(pcnoc_s_1, MSM8916_PNOC_SLV_1, 8, -1, -1, MSM8916_SLAVE_IMEM_CFG, MSM8916_SLAVE_CRYPTO_0_CFG, MSM8916_SLAVE_MSG_RAM, MSM8916_SLAVE_PDM, MSM8916_SLAVE_PRNG);
-DEFINE_QNODE(pcnoc_s_2, MSM8916_PNOC_SLV_2, 8, -1, -1, MSM8916_SLAVE_SPDM, MSM8916_SLAVE_BOOT_ROM, MSM8916_SLAVE_BIMC_CFG, MSM8916_SLAVE_PNOC_CFG, MSM8916_SLAVE_PMIC_ARB);
-DEFINE_QNODE(pcnoc_s_3, MSM8916_PNOC_SLV_3, 8, -1, -1, MSM8916_SLAVE_MPM, MSM8916_SLAVE_SNOC_CFG, MSM8916_SLAVE_RBCPR_CFG, MSM8916_SLAVE_QDSS_CFG, MSM8916_SLAVE_DEHR_CFG);
-DEFINE_QNODE(pcnoc_s_4, MSM8916_PNOC_SLV_4, 8, -1, -1, MSM8916_SLAVE_VENUS_CFG, MSM8916_SLAVE_CAMERA_CFG, MSM8916_SLAVE_DISPLAY_CFG);
-DEFINE_QNODE(pcnoc_s_8, MSM8916_PNOC_SLV_8, 8, -1, -1, MSM8916_SLAVE_USB_HS, MSM8916_SLAVE_SDCC_1, MSM8916_SLAVE_BLSP_1);
-DEFINE_QNODE(pcnoc_s_9, MSM8916_PNOC_SLV_9, 8, -1, -1, MSM8916_SLAVE_SDCC_2, MSM8916_SLAVE_LPASS, MSM8916_SLAVE_GRAPHICS_3D_CFG);
+DEFINE_QNODE(pcnoc_s_0, MSM8916_PNOC_SLV_0, 4, -1, -1, MSM8916_SLAVE_CLK_CTL, MSM8916_SLAVE_TLMM, MSM8916_SLAVE_TCSR, MSM8916_SLAVE_SECURITY, MSM8916_SLAVE_MSS);
+DEFINE_QNODE(pcnoc_s_1, MSM8916_PNOC_SLV_1, 4, -1, -1, MSM8916_SLAVE_IMEM_CFG, MSM8916_SLAVE_CRYPTO_0_CFG, MSM8916_SLAVE_MSG_RAM, MSM8916_SLAVE_PDM, MSM8916_SLAVE_PRNG);
+DEFINE_QNODE(pcnoc_s_2, MSM8916_PNOC_SLV_2, 4, -1, -1, MSM8916_SLAVE_SPDM, MSM8916_SLAVE_BOOT_ROM, MSM8916_SLAVE_BIMC_CFG, MSM8916_SLAVE_PNOC_CFG, MSM8916_SLAVE_PMIC_ARB);
+DEFINE_QNODE(pcnoc_s_3, MSM8916_PNOC_SLV_3, 4, -1, -1, MSM8916_SLAVE_MPM, MSM8916_SLAVE_SNOC_CFG, MSM8916_SLAVE_RBCPR_CFG, MSM8916_SLAVE_QDSS_CFG, MSM8916_SLAVE_DEHR_CFG);
+DEFINE_QNODE(pcnoc_s_4, MSM8916_PNOC_SLV_4, 4, -1, -1, MSM8916_SLAVE_VENUS_CFG, MSM8916_SLAVE_CAMERA_CFG, MSM8916_SLAVE_DISPLAY_CFG);
+DEFINE_QNODE(pcnoc_s_8, MSM8916_PNOC_SLV_8, 4, -1, -1, MSM8916_SLAVE_USB_HS, MSM8916_SLAVE_SDCC_1, MSM8916_SLAVE_BLSP_1);
+DEFINE_QNODE(pcnoc_s_9, MSM8916_PNOC_SLV_9, 4, -1, -1, MSM8916_SLAVE_SDCC_2, MSM8916_SLAVE_LPASS, MSM8916_SLAVE_GRAPHICS_3D_CFG);
 DEFINE_QNODE(pcnoc_snoc_mas, MSM8916_PNOC_SNOC_MAS, 8, 29, -1, MSM8916_PNOC_SNOC_SLV);
 DEFINE_QNODE(pcnoc_snoc_slv, MSM8916_PNOC_SNOC_SLV, 8, -1, 45, MSM8916_SNOC_INT_0, MSM8916_SNOC_INT_BIMC, MSM8916_SNOC_INT_1);
 DEFINE_QNODE(qdss_int, MSM8916_SNOC_QDSS_INT, 8, -1, -1, MSM8916_SNOC_INT_0, MSM8916_SNOC_INT_BIMC);
diff --git a/drivers/misc/habanalabs/command_submission.c b/drivers/misc/habanalabs/command_submission.c
index b0f62cbbdc87..f3a8f113865d 100644
--- a/drivers/misc/habanalabs/command_submission.c
+++ b/drivers/misc/habanalabs/command_submission.c
@@ -499,11 +499,19 @@ static int validate_queue_index(struct hl_device *hdev,
 	struct asic_fixed_properties *asic = &hdev->asic_prop;
 	struct hw_queue_properties *hw_queue_prop;
 
+	/* This must be checked here to prevent out-of-bounds access to
+	 * hw_queues_props array
+	 */
+	if (chunk->queue_index >= HL_MAX_QUEUES) {
+		dev_err(hdev->dev, "Queue index %d is invalid\n",
+			chunk->queue_index);
+		return -EINVAL;
+	}
+
 	hw_queue_prop = &asic->hw_queues_props[chunk->queue_index];
 
-	if ((chunk->queue_index >= HL_MAX_QUEUES) ||
-			(hw_queue_prop->type == QUEUE_TYPE_NA)) {
-		dev_err(hdev->dev, "Queue index %d is invalid\n",
+	if (hw_queue_prop->type == QUEUE_TYPE_NA) {
+		dev_err(hdev->dev, "Queue index %d is not applicable\n",
 			chunk->queue_index);
 		return -EINVAL;
 	}
diff --git a/drivers/misc/habanalabs/debugfs.c b/drivers/misc/habanalabs/debugfs.c
index fc4372c18ce2..0bc036e01ee8 100644
--- a/drivers/misc/habanalabs/debugfs.c
+++ b/drivers/misc/habanalabs/debugfs.c
@@ -36,7 +36,7 @@ static int hl_debugfs_i2c_read(struct hl_device *hdev, u8 i2c_bus, u8 i2c_addr,
 	pkt.i2c_reg = i2c_reg;
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-					HL_DEVICE_TIMEOUT_USEC, (long *) val);
+						0, (long *) val);
 
 	if (rc)
 		dev_err(hdev->dev, "Failed to read from I2C, error %d\n", rc);
@@ -63,7 +63,7 @@ static int hl_debugfs_i2c_write(struct hl_device *hdev, u8 i2c_bus, u8 i2c_addr,
 	pkt.value = cpu_to_le64(val);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-					HL_DEVICE_TIMEOUT_USEC, NULL);
+						0, NULL);
 
 	if (rc)
 		dev_err(hdev->dev, "Failed to write to I2C, error %d\n", rc);
@@ -87,7 +87,7 @@ static void hl_debugfs_led_set(struct hl_device *hdev, u8 led, u8 state)
 	pkt.value = cpu_to_le64(state);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-						HL_DEVICE_TIMEOUT_USEC, NULL);
+						0, NULL);
 
 	if (rc)
 		dev_err(hdev->dev, "Failed to set LED %d, error %d\n", led, rc);
@@ -981,7 +981,7 @@ static ssize_t hl_clk_gate_read(struct file *f, char __user *buf,
 	if (*ppos)
 		return 0;
 
-	sprintf(tmp_buf, "%d\n", hdev->clock_gating);
+	sprintf(tmp_buf, "0x%llx\n", hdev->clock_gating_mask);
 	rc = simple_read_from_buffer(buf, strlen(tmp_buf) + 1, ppos, tmp_buf,
 			strlen(tmp_buf) + 1);
 
@@ -993,7 +993,7 @@ static ssize_t hl_clk_gate_write(struct file *f, const char __user *buf,
 {
 	struct hl_dbg_device_entry *entry = file_inode(f)->i_private;
 	struct hl_device *hdev = entry->hdev;
-	u32 value;
+	u64 value;
 	ssize_t rc;
 
 	if (atomic_read(&hdev->in_reset)) {
@@ -1002,19 +1002,12 @@ static ssize_t hl_clk_gate_write(struct file *f, const char __user *buf,
 		return 0;
 	}
 
-	rc = kstrtouint_from_user(buf, count, 10, &value);
+	rc = kstrtoull_from_user(buf, count, 16, &value);
 	if (rc)
 		return rc;
 
-	if (value) {
-		hdev->clock_gating = 1;
-		if (hdev->asic_funcs->enable_clock_gating)
-			hdev->asic_funcs->enable_clock_gating(hdev);
-	} else {
-		if (hdev->asic_funcs->disable_clock_gating)
-			hdev->asic_funcs->disable_clock_gating(hdev);
-		hdev->clock_gating = 0;
-	}
+	hdev->clock_gating_mask = value;
+	hdev->asic_funcs->set_clock_gating(hdev);
 
 	return count;
 }
diff --git a/drivers/misc/habanalabs/device.c b/drivers/misc/habanalabs/device.c
index 2b38a119704c..59608d1bac88 100644
--- a/drivers/misc/habanalabs/device.c
+++ b/drivers/misc/habanalabs/device.c
@@ -608,7 +608,7 @@ int hl_device_set_debug_mode(struct hl_device *hdev, bool enable)
 		hdev->in_debug = 0;
 
 		if (!hdev->hard_reset_pending)
-			hdev->asic_funcs->enable_clock_gating(hdev);
+			hdev->asic_funcs->set_clock_gating(hdev);
 
 		goto out;
 	}
diff --git a/drivers/misc/habanalabs/firmware_if.c b/drivers/misc/habanalabs/firmware_if.c
index baf790cf4b78..d27841cb5bcb 100644
--- a/drivers/misc/habanalabs/firmware_if.c
+++ b/drivers/misc/habanalabs/firmware_if.c
@@ -61,7 +61,7 @@ int hl_fw_send_pci_access_msg(struct hl_device *hdev, u32 opcode)
 	pkt.ctl = cpu_to_le32(opcode << ARMCP_PKT_CTL_OPCODE_SHIFT);
 
 	return hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt,
-				sizeof(pkt), HL_DEVICE_TIMEOUT_USEC, NULL);
+						sizeof(pkt), 0, NULL);
 }
 
 int hl_fw_send_cpu_message(struct hl_device *hdev, u32 hw_queue_id, u32 *msg,
@@ -144,7 +144,7 @@ int hl_fw_unmask_irq(struct hl_device *hdev, u16 event_type)
 	pkt.value = cpu_to_le64(event_type);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-			HL_DEVICE_TIMEOUT_USEC, &result);
+						0, &result);
 
 	if (rc)
 		dev_err(hdev->dev, "failed to unmask RAZWI IRQ %d", event_type);
@@ -183,7 +183,7 @@ int hl_fw_unmask_irq_arr(struct hl_device *hdev, const u32 *irq_arr,
 						ARMCP_PKT_CTL_OPCODE_SHIFT);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) pkt,
-			total_pkt_size, HL_DEVICE_TIMEOUT_USEC, &result);
+						total_pkt_size, 0, &result);
 
 	if (rc)
 		dev_err(hdev->dev, "failed to unmask IRQ array\n");
@@ -204,7 +204,7 @@ int hl_fw_test_cpu_queue(struct hl_device *hdev)
 	test_pkt.value = cpu_to_le64(ARMCP_PACKET_FENCE_VAL);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &test_pkt,
-			sizeof(test_pkt), HL_DEVICE_TIMEOUT_USEC, &result);
+						sizeof(test_pkt), 0, &result);
 
 	if (!rc) {
 		if (result != ARMCP_PACKET_FENCE_VAL)
@@ -248,7 +248,7 @@ int hl_fw_send_heartbeat(struct hl_device *hdev)
 	hb_pkt.value = cpu_to_le64(ARMCP_PACKET_FENCE_VAL);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &hb_pkt,
-			sizeof(hb_pkt), HL_DEVICE_TIMEOUT_USEC, &result);
+						sizeof(hb_pkt), 0, &result);
 
 	if ((rc) || (result != ARMCP_PACKET_FENCE_VAL))
 		rc = -EIO;
diff --git a/drivers/misc/habanalabs/gaudi/gaudi.c b/drivers/misc/habanalabs/gaudi/gaudi.c
index 834470d10b46..637a9d608707 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi.c
@@ -80,6 +80,7 @@
 #define GAUDI_PLDM_QMAN0_TIMEOUT_USEC	(HL_DEVICE_TIMEOUT_USEC * 30)
 #define GAUDI_PLDM_TPC_KERNEL_WAIT_USEC	(HL_DEVICE_TIMEOUT_USEC * 30)
 #define GAUDI_BOOT_FIT_REQ_TIMEOUT_USEC	1000000		/* 1s */
+#define GAUDI_MSG_TO_CPU_TIMEOUT_USEC	4000000		/* 4s */
 
 #define GAUDI_QMAN0_FENCE_VAL		0x72E91AB9
 
@@ -98,6 +99,11 @@
 
 #define GAUDI_ARB_WDT_TIMEOUT		0x1000000
 
+#define GAUDI_CLK_GATE_DEBUGFS_MASK	(\
+		BIT(GAUDI_ENGINE_ID_MME_0) |\
+		BIT(GAUDI_ENGINE_ID_MME_2) |\
+		GENMASK_ULL(GAUDI_ENGINE_ID_TPC_7, GAUDI_ENGINE_ID_TPC_0))
+
 static const char gaudi_irq_name[GAUDI_MSI_ENTRIES][GAUDI_MAX_STRING_LEN] = {
 		"gaudi cq 0_0", "gaudi cq 0_1", "gaudi cq 0_2", "gaudi cq 0_3",
 		"gaudi cq 1_0", "gaudi cq 1_1", "gaudi cq 1_2", "gaudi cq 1_3",
@@ -106,14 +112,14 @@ static const char gaudi_irq_name[GAUDI_MSI_ENTRIES][GAUDI_MAX_STRING_LEN] = {
 };
 
 static const u8 gaudi_dma_assignment[GAUDI_DMA_MAX] = {
-	[GAUDI_PCI_DMA_1] = 0,
-	[GAUDI_PCI_DMA_2] = 1,
-	[GAUDI_PCI_DMA_3] = 5,
-	[GAUDI_HBM_DMA_1] = 2,
-	[GAUDI_HBM_DMA_2] = 3,
-	[GAUDI_HBM_DMA_3] = 4,
-	[GAUDI_HBM_DMA_4] = 6,
-	[GAUDI_HBM_DMA_5] = 7
+	[GAUDI_PCI_DMA_1] = GAUDI_ENGINE_ID_DMA_0,
+	[GAUDI_PCI_DMA_2] = GAUDI_ENGINE_ID_DMA_1,
+	[GAUDI_PCI_DMA_3] = GAUDI_ENGINE_ID_DMA_5,
+	[GAUDI_HBM_DMA_1] = GAUDI_ENGINE_ID_DMA_2,
+	[GAUDI_HBM_DMA_2] = GAUDI_ENGINE_ID_DMA_3,
+	[GAUDI_HBM_DMA_3] = GAUDI_ENGINE_ID_DMA_4,
+	[GAUDI_HBM_DMA_4] = GAUDI_ENGINE_ID_DMA_6,
+	[GAUDI_HBM_DMA_5] = GAUDI_ENGINE_ID_DMA_7
 };
 
 static const u8 gaudi_cq_assignment[NUMBER_OF_CMPLT_QUEUES] = {
@@ -1819,7 +1825,7 @@ static void gaudi_init_golden_registers(struct hl_device *hdev)
 
 	gaudi_init_rate_limiter(hdev);
 
-	gaudi_disable_clock_gating(hdev);
+	hdev->asic_funcs->disable_clock_gating(hdev);
 
 	for (tpc_id = 0, tpc_offset = 0;
 				tpc_id < TPC_NUMBER_OF_ENGINES;
@@ -2531,46 +2537,55 @@ static void gaudi_tpc_stall(struct hl_device *hdev)
 	WREG32(mmTPC7_CFG_TPC_STALL, 1 << TPC0_CFG_TPC_STALL_V_SHIFT);
 }
 
-static void gaudi_enable_clock_gating(struct hl_device *hdev)
+static void gaudi_set_clock_gating(struct hl_device *hdev)
 {
 	struct gaudi_device *gaudi = hdev->asic_specific;
 	u32 qman_offset;
 	int i;
 
-	if (!hdev->clock_gating)
-		return;
-
-	if (gaudi->hw_cap_initialized & HW_CAP_CLK_GATE)
-		return;
-
 	/* In case we are during debug session, don't enable the clock gate
 	 * as it may interfere
 	 */
 	if (hdev->in_debug)
 		return;
 
-	for (i = 0, qman_offset = 0 ; i < PCI_DMA_NUMBER_OF_CHNLS ; i++) {
+	for (i = GAUDI_PCI_DMA_1, qman_offset = 0 ; i < GAUDI_HBM_DMA_1 ; i++) {
+		if (!(hdev->clock_gating_mask &
+					(BIT_ULL(gaudi_dma_assignment[i]))))
+			continue;
+
 		qman_offset = gaudi_dma_assignment[i] * DMA_QMAN_OFFSET;
 		WREG32(mmDMA0_QM_CGM_CFG1 + qman_offset, QMAN_CGM1_PWR_GATE_EN);
 		WREG32(mmDMA0_QM_CGM_CFG + qman_offset,
 				QMAN_UPPER_CP_CGM_PWR_GATE_EN);
 	}
 
-	for (; i < HBM_DMA_NUMBER_OF_CHNLS ; i++) {
+	for (i = GAUDI_HBM_DMA_1 ; i < GAUDI_DMA_MAX ; i++) {
+		if (!(hdev->clock_gating_mask &
+					(BIT_ULL(gaudi_dma_assignment[i]))))
+			continue;
+
 		qman_offset = gaudi_dma_assignment[i] * DMA_QMAN_OFFSET;
 		WREG32(mmDMA0_QM_CGM_CFG1 + qman_offset, QMAN_CGM1_PWR_GATE_EN);
 		WREG32(mmDMA0_QM_CGM_CFG + qman_offset,
 				QMAN_COMMON_CP_CGM_PWR_GATE_EN);
 	}
 
-	WREG32(mmMME0_QM_CGM_CFG1, QMAN_CGM1_PWR_GATE_EN);
-	WREG32(mmMME0_QM_CGM_CFG,
-			QMAN_COMMON_CP_CGM_PWR_GATE_EN);
-	WREG32(mmMME2_QM_CGM_CFG1, QMAN_CGM1_PWR_GATE_EN);
-	WREG32(mmMME2_QM_CGM_CFG,
-			QMAN_COMMON_CP_CGM_PWR_GATE_EN);
+	if (hdev->clock_gating_mask & (BIT_ULL(GAUDI_ENGINE_ID_MME_0))) {
+		WREG32(mmMME0_QM_CGM_CFG1, QMAN_CGM1_PWR_GATE_EN);
+		WREG32(mmMME0_QM_CGM_CFG, QMAN_COMMON_CP_CGM_PWR_GATE_EN);
+	}
+
+	if (hdev->clock_gating_mask & (BIT_ULL(GAUDI_ENGINE_ID_MME_2))) {
+		WREG32(mmMME2_QM_CGM_CFG1, QMAN_CGM1_PWR_GATE_EN);
+		WREG32(mmMME2_QM_CGM_CFG, QMAN_COMMON_CP_CGM_PWR_GATE_EN);
+	}
 
 	for (i = 0, qman_offset = 0 ; i < TPC_NUMBER_OF_ENGINES ; i++) {
+		if (!(hdev->clock_gating_mask &
+					(BIT_ULL(GAUDI_ENGINE_ID_TPC_0 + i))))
+			continue;
+
 		WREG32(mmTPC0_QM_CGM_CFG1 + qman_offset,
 				QMAN_CGM1_PWR_GATE_EN);
 		WREG32(mmTPC0_QM_CGM_CFG + qman_offset,
@@ -2663,7 +2678,7 @@ static void gaudi_halt_engines(struct hl_device *hdev, bool hard_reset)
 	gaudi_stop_hbm_dma_qmans(hdev);
 	gaudi_stop_pci_dma_qmans(hdev);
 
-	gaudi_disable_clock_gating(hdev);
+	hdev->asic_funcs->disable_clock_gating(hdev);
 
 	msleep(wait_timeout_ms);
 
@@ -3003,7 +3018,7 @@ static int gaudi_hw_init(struct hl_device *hdev)
 
 	gaudi_init_tpc_qmans(hdev);
 
-	gaudi_enable_clock_gating(hdev);
+	hdev->asic_funcs->set_clock_gating(hdev);
 
 	gaudi_enable_timestamp(hdev);
 
@@ -3112,7 +3127,9 @@ static void gaudi_hw_fini(struct hl_device *hdev, bool hard_reset)
 					HW_CAP_HBM_DMA | HW_CAP_PLL |
 					HW_CAP_MMU |
 					HW_CAP_SRAM_SCRAMBLER |
-					HW_CAP_HBM_SCRAMBLER);
+					HW_CAP_HBM_SCRAMBLER |
+					HW_CAP_CLK_GATE);
+
 	memset(gaudi->events_stat, 0, sizeof(gaudi->events_stat));
 }
 
@@ -3463,6 +3480,9 @@ static int gaudi_send_cpu_message(struct hl_device *hdev, u32 *msg,
 		return 0;
 	}
 
+	if (!timeout)
+		timeout = GAUDI_MSG_TO_CPU_TIMEOUT_USEC;
+
 	return hl_fw_send_cpu_message(hdev, GAUDI_QUEUE_ID_CPU_PQ, msg, len,
 						timeout, result);
 }
@@ -3865,6 +3885,12 @@ static int gaudi_validate_cb(struct hl_device *hdev,
 			rc = -EPERM;
 			break;
 
+		case PACKET_WREG_BULK:
+			dev_err(hdev->dev,
+				"User not allowed to use WREG_BULK\n");
+			rc = -EPERM;
+			break;
+
 		case PACKET_LOAD_AND_EXE:
 			rc = gaudi_validate_load_and_exe_pkt(hdev, parser,
 				(struct packet_load_and_exe *) user_pkt);
@@ -3880,7 +3906,6 @@ static int gaudi_validate_cb(struct hl_device *hdev,
 			break;
 
 		case PACKET_WREG_32:
-		case PACKET_WREG_BULK:
 		case PACKET_MSG_LONG:
 		case PACKET_MSG_SHORT:
 		case PACKET_REPEAT:
@@ -4521,13 +4546,18 @@ static int gaudi_debugfs_read32(struct hl_device *hdev, u64 addr, u32 *val)
 	int rc = 0;
 
 	if ((addr >= CFG_BASE) && (addr < CFG_BASE + CFG_SIZE)) {
-		if (gaudi->hw_cap_initialized & HW_CAP_CLK_GATE) {
+
+		if ((gaudi->hw_cap_initialized & HW_CAP_CLK_GATE) &&
+				(hdev->clock_gating_mask &
+						GAUDI_CLK_GATE_DEBUGFS_MASK)) {
+
 			dev_err_ratelimited(hdev->dev,
 				"Can't read register - clock gating is enabled!\n");
 			rc = -EFAULT;
 		} else {
 			*val = RREG32(addr - CFG_BASE);
 		}
+
 	} else if ((addr >= SRAM_BASE_ADDR) &&
 			(addr < SRAM_BASE_ADDR + SRAM_BAR_SIZE)) {
 		*val = readl(hdev->pcie_bar[SRAM_BAR_ID] +
@@ -4563,13 +4593,18 @@ static int gaudi_debugfs_write32(struct hl_device *hdev, u64 addr, u32 val)
 	int rc = 0;
 
 	if ((addr >= CFG_BASE) && (addr < CFG_BASE + CFG_SIZE)) {
-		if (gaudi->hw_cap_initialized & HW_CAP_CLK_GATE) {
+
+		if ((gaudi->hw_cap_initialized & HW_CAP_CLK_GATE) &&
+				(hdev->clock_gating_mask &
+						GAUDI_CLK_GATE_DEBUGFS_MASK)) {
+
 			dev_err_ratelimited(hdev->dev,
 				"Can't write register - clock gating is enabled!\n");
 			rc = -EFAULT;
 		} else {
 			WREG32(addr - CFG_BASE, val);
 		}
+
 	} else if ((addr >= SRAM_BASE_ADDR) &&
 			(addr < SRAM_BASE_ADDR + SRAM_BAR_SIZE)) {
 		writel(val, hdev->pcie_bar[SRAM_BAR_ID] +
@@ -4605,7 +4640,11 @@ static int gaudi_debugfs_read64(struct hl_device *hdev, u64 addr, u64 *val)
 	int rc = 0;
 
 	if ((addr >= CFG_BASE) && (addr <= CFG_BASE + CFG_SIZE - sizeof(u64))) {
-		if (gaudi->hw_cap_initialized & HW_CAP_CLK_GATE) {
+
+		if ((gaudi->hw_cap_initialized & HW_CAP_CLK_GATE) &&
+				(hdev->clock_gating_mask &
+						GAUDI_CLK_GATE_DEBUGFS_MASK)) {
+
 			dev_err_ratelimited(hdev->dev,
 				"Can't read register - clock gating is enabled!\n");
 			rc = -EFAULT;
@@ -4615,6 +4654,7 @@ static int gaudi_debugfs_read64(struct hl_device *hdev, u64 addr, u64 *val)
 
 			*val = (((u64) val_h) << 32) | val_l;
 		}
+
 	} else if ((addr >= SRAM_BASE_ADDR) &&
 		   (addr <= SRAM_BASE_ADDR + SRAM_BAR_SIZE - sizeof(u64))) {
 		*val = readq(hdev->pcie_bar[SRAM_BAR_ID] +
@@ -4651,7 +4691,11 @@ static int gaudi_debugfs_write64(struct hl_device *hdev, u64 addr, u64 val)
 	int rc = 0;
 
 	if ((addr >= CFG_BASE) && (addr <= CFG_BASE + CFG_SIZE - sizeof(u64))) {
-		if (gaudi->hw_cap_initialized & HW_CAP_CLK_GATE) {
+
+		if ((gaudi->hw_cap_initialized & HW_CAP_CLK_GATE) &&
+				(hdev->clock_gating_mask &
+						GAUDI_CLK_GATE_DEBUGFS_MASK)) {
+
 			dev_err_ratelimited(hdev->dev,
 				"Can't write register - clock gating is enabled!\n");
 			rc = -EFAULT;
@@ -4660,6 +4704,7 @@ static int gaudi_debugfs_write64(struct hl_device *hdev, u64 addr, u64 val)
 			WREG32(addr + sizeof(u32) - CFG_BASE,
 				upper_32_bits(val));
 		}
+
 	} else if ((addr >= SRAM_BASE_ADDR) &&
 		   (addr <= SRAM_BASE_ADDR + SRAM_BAR_SIZE - sizeof(u64))) {
 		writeq(val, hdev->pcie_bar[SRAM_BAR_ID] +
@@ -4881,7 +4926,7 @@ static void gaudi_mmu_prepare(struct hl_device *hdev, u32 asid)
 	gaudi_mmu_prepare_reg(hdev, mmPSOC_GLOBAL_CONF_TRACE_ARUSER, asid);
 	gaudi_mmu_prepare_reg(hdev, mmPSOC_GLOBAL_CONF_TRACE_AWUSER, asid);
 
-	hdev->asic_funcs->enable_clock_gating(hdev);
+	hdev->asic_funcs->set_clock_gating(hdev);
 
 	mutex_unlock(&gaudi->clk_gate_mutex);
 }
@@ -5262,7 +5307,7 @@ static void gaudi_print_ecc_info_generic(struct hl_device *hdev,
 	}
 
 	if (disable_clock_gating) {
-		hdev->asic_funcs->enable_clock_gating(hdev);
+		hdev->asic_funcs->set_clock_gating(hdev);
 		mutex_unlock(&gaudi->clk_gate_mutex);
 	}
 }
@@ -5749,7 +5794,7 @@ static bool gaudi_tpc_read_interrupts(struct hl_device *hdev, u8 tpc_id,
 	/* Clear interrupts */
 	WREG32(mmTPC0_CFG_TPC_INTR_CAUSE + tpc_offset, 0);
 
-	hdev->asic_funcs->enable_clock_gating(hdev);
+	hdev->asic_funcs->set_clock_gating(hdev);
 
 	mutex_unlock(&gaudi->clk_gate_mutex);
 
@@ -6265,7 +6310,7 @@ static bool gaudi_is_device_idle(struct hl_device *hdev, u32 *mask,
 	if (s)
 		seq_puts(s, "\n");
 
-	hdev->asic_funcs->enable_clock_gating(hdev);
+	hdev->asic_funcs->set_clock_gating(hdev);
 
 	mutex_unlock(&gaudi->clk_gate_mutex);
 
@@ -6366,7 +6411,7 @@ static int gaudi_run_tpc_kernel(struct hl_device *hdev, u64 tpc_kernel,
 		dev_err(hdev->dev,
 			"Timeout while waiting for TPC%d icache prefetch\n",
 			tpc_id);
-		hdev->asic_funcs->enable_clock_gating(hdev);
+		hdev->asic_funcs->set_clock_gating(hdev);
 		mutex_unlock(&gaudi->clk_gate_mutex);
 		return -EIO;
 	}
@@ -6395,7 +6440,7 @@ static int gaudi_run_tpc_kernel(struct hl_device *hdev, u64 tpc_kernel,
 		1000,
 		kernel_timeout);
 
-	hdev->asic_funcs->enable_clock_gating(hdev);
+	hdev->asic_funcs->set_clock_gating(hdev);
 	mutex_unlock(&gaudi->clk_gate_mutex);
 
 	if (rc) {
@@ -6736,7 +6781,7 @@ static const struct hl_asic_funcs gaudi_funcs = {
 	.mmu_invalidate_cache = gaudi_mmu_invalidate_cache,
 	.mmu_invalidate_cache_range = gaudi_mmu_invalidate_cache_range,
 	.send_heartbeat = gaudi_send_heartbeat,
-	.enable_clock_gating = gaudi_enable_clock_gating,
+	.set_clock_gating = gaudi_set_clock_gating,
 	.disable_clock_gating = gaudi_disable_clock_gating,
 	.debug_coresight = gaudi_debug_coresight,
 	.is_device_idle = gaudi_is_device_idle,
diff --git a/drivers/misc/habanalabs/goya/goya.c b/drivers/misc/habanalabs/goya/goya.c
index 0d2952bb58df..88460b2138d8 100644
--- a/drivers/misc/habanalabs/goya/goya.c
+++ b/drivers/misc/habanalabs/goya/goya.c
@@ -88,6 +88,7 @@
 #define GOYA_PLDM_MMU_TIMEOUT_USEC	(MMU_CONFIG_TIMEOUT_USEC * 100)
 #define GOYA_PLDM_QMAN0_TIMEOUT_USEC	(HL_DEVICE_TIMEOUT_USEC * 30)
 #define GOYA_BOOT_FIT_REQ_TIMEOUT_USEC	1000000		/* 1s */
+#define GOYA_MSG_TO_CPU_TIMEOUT_USEC	4000000		/* 4s */
 
 #define GOYA_QMAN0_FENCE_VAL		0xD169B243
 
@@ -2830,6 +2831,9 @@ int goya_send_cpu_message(struct hl_device *hdev, u32 *msg, u16 len,
 		return 0;
 	}
 
+	if (!timeout)
+		timeout = GOYA_MSG_TO_CPU_TIMEOUT_USEC;
+
 	return hl_fw_send_cpu_message(hdev, GOYA_QUEUE_ID_CPU_PQ, msg, len,
 					timeout, result);
 }
@@ -4431,8 +4435,8 @@ static int goya_unmask_irq_arr(struct hl_device *hdev, u32 *irq_arr,
 	pkt->armcp_pkt.ctl = cpu_to_le32(ARMCP_PACKET_UNMASK_RAZWI_IRQ_ARRAY <<
 						ARMCP_PKT_CTL_OPCODE_SHIFT);
 
-	rc = goya_send_cpu_message(hdev, (u32 *) pkt, total_pkt_size,
-			HL_DEVICE_TIMEOUT_USEC, &result);
+	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) pkt,
+						total_pkt_size,	0, &result);
 
 	if (rc)
 		dev_err(hdev->dev, "failed to unmask IRQ array\n");
@@ -4464,8 +4468,8 @@ static int goya_unmask_irq(struct hl_device *hdev, u16 event_type)
 				ARMCP_PKT_CTL_OPCODE_SHIFT);
 	pkt.value = cpu_to_le64(event_type);
 
-	rc = goya_send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-			HL_DEVICE_TIMEOUT_USEC, &result);
+	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
+						0, &result);
 
 	if (rc)
 		dev_err(hdev->dev, "failed to unmask RAZWI IRQ %d", event_type);
@@ -5028,14 +5032,14 @@ int goya_armcp_info_get(struct hl_device *hdev)
 	return 0;
 }
 
-static void goya_enable_clock_gating(struct hl_device *hdev)
+static void goya_set_clock_gating(struct hl_device *hdev)
 {
-
+	/* clock gating not supported in Goya */
 }
 
 static void goya_disable_clock_gating(struct hl_device *hdev)
 {
-
+	/* clock gating not supported in Goya */
 }
 
 static bool goya_is_device_idle(struct hl_device *hdev, u32 *mask,
@@ -5259,7 +5263,7 @@ static const struct hl_asic_funcs goya_funcs = {
 	.mmu_invalidate_cache = goya_mmu_invalidate_cache,
 	.mmu_invalidate_cache_range = goya_mmu_invalidate_cache_range,
 	.send_heartbeat = goya_send_heartbeat,
-	.enable_clock_gating = goya_enable_clock_gating,
+	.set_clock_gating = goya_set_clock_gating,
 	.disable_clock_gating = goya_disable_clock_gating,
 	.debug_coresight = goya_debug_coresight,
 	.is_device_idle = goya_is_device_idle,
diff --git a/drivers/misc/habanalabs/habanalabs.h b/drivers/misc/habanalabs/habanalabs.h
index 1ecdcf8b763a..194d83352696 100644
--- a/drivers/misc/habanalabs/habanalabs.h
+++ b/drivers/misc/habanalabs/habanalabs.h
@@ -578,8 +578,9 @@ enum hl_pll_frequency {
  * @mmu_invalidate_cache_range: flush specific MMU STLB cache lines with
  *                              ASID-VA-size mask.
  * @send_heartbeat: send is-alive packet to ArmCP and verify response.
- * @enable_clock_gating: enable clock gating for reducing power consumption.
- * @disable_clock_gating: disable clock for accessing registers on HBW.
+ * @set_clock_gating: enable/disable clock gating per engine according to
+ *                    clock gating mask in hdev
+ * @disable_clock_gating: disable clock gating completely
  * @debug_coresight: perform certain actions on Coresight for debugging.
  * @is_device_idle: return true if device is idle, false otherwise.
  * @soft_reset_late_init: perform certain actions needed after soft reset.
@@ -587,7 +588,11 @@ enum hl_pll_frequency {
  * @hw_queues_unlock: release H/W queues lock.
  * @get_pci_id: retrieve PCI ID.
  * @get_eeprom_data: retrieve EEPROM data from F/W.
- * @send_cpu_message: send buffer to ArmCP.
+ * @send_cpu_message: send message to F/W. If the message is timedout, the
+ *                    driver will eventually reset the device. The timeout can
+ *                    be determined by the calling function or it can be 0 and
+ *                    then the timeout is the default timeout for the specific
+ *                    ASIC
  * @get_hw_state: retrieve the H/W state
  * @pci_bars_map: Map PCI BARs.
  * @set_dram_bar_base: Set DRAM BAR to map specific device address. Returns
@@ -680,7 +685,7 @@ struct hl_asic_funcs {
 	int (*mmu_invalidate_cache_range)(struct hl_device *hdev, bool is_hard,
 			u32 asid, u64 va, u64 size);
 	int (*send_heartbeat)(struct hl_device *hdev);
-	void (*enable_clock_gating)(struct hl_device *hdev);
+	void (*set_clock_gating)(struct hl_device *hdev);
 	void (*disable_clock_gating)(struct hl_device *hdev);
 	int (*debug_coresight)(struct hl_device *hdev, void *data);
 	bool (*is_device_idle)(struct hl_device *hdev, u32 *mask,
@@ -1398,6 +1403,9 @@ struct hl_device_idle_busy_ts {
  * @max_power: the max power of the device, as configured by the sysadmin. This
  *             value is saved so in case of hard-reset, the driver will restore
  *             this value and update the F/W after the re-initialization
+ * @clock_gating_mask: is clock gating enabled. bitmask that represents the
+ *                     different engines. See debugfs-driver-habanalabs for
+ *                     details.
  * @in_reset: is device in reset flow.
  * @curr_pll_profile: current PLL profile.
  * @cs_active_cnt: number of active command submissions on this device (active
@@ -1425,7 +1433,6 @@ struct hl_device_idle_busy_ts {
  * @init_done: is the initialization of the device done.
  * @mmu_enable: is MMU enabled.
  * @mmu_huge_page_opt: is MMU huge pages optimization enabled.
- * @clock_gating: is clock gating enabled.
  * @device_cpu_disabled: is the device CPU disabled (due to timeouts)
  * @dma_mask: the dma mask that was set for this device
  * @in_debug: is device under debug. This, together with fpriv_list, enforces
@@ -1493,6 +1500,7 @@ struct hl_device {
 	atomic64_t			dram_used_mem;
 	u64				timeout_jiffies;
 	u64				max_power;
+	u64				clock_gating_mask;
 	atomic_t			in_reset;
 	enum hl_pll_frequency		curr_pll_profile;
 	int				cs_active_cnt;
@@ -1514,7 +1522,6 @@ struct hl_device {
 	u8				dram_default_page_mapping;
 	u8				pmmu_huge_range;
 	u8				init_done;
-	u8				clock_gating;
 	u8				device_cpu_disabled;
 	u8				dma_mask;
 	u8				in_debug;
diff --git a/drivers/misc/habanalabs/habanalabs_drv.c b/drivers/misc/habanalabs/habanalabs_drv.c
index 8652c7e5d7f1..22716da9f85f 100644
--- a/drivers/misc/habanalabs/habanalabs_drv.c
+++ b/drivers/misc/habanalabs/habanalabs_drv.c
@@ -232,7 +232,7 @@ static void set_driver_behavior_per_device(struct hl_device *hdev)
 	hdev->fw_loading = 1;
 	hdev->cpu_queues_enable = 1;
 	hdev->heartbeat = 1;
-	hdev->clock_gating = 1;
+	hdev->clock_gating_mask = ULONG_MAX;
 
 	hdev->reset_pcilink = 0;
 	hdev->axi_drain = 0;
diff --git a/drivers/misc/habanalabs/hwmon.c b/drivers/misc/habanalabs/hwmon.c
index 8c6cd77e6af6..b997336fa75f 100644
--- a/drivers/misc/habanalabs/hwmon.c
+++ b/drivers/misc/habanalabs/hwmon.c
@@ -10,7 +10,6 @@
 #include <linux/pci.h>
 #include <linux/hwmon.h>
 
-#define SENSORS_PKT_TIMEOUT		1000000	/* 1s */
 #define HWMON_NR_SENSOR_TYPES		(hwmon_pwm + 1)
 
 int hl_build_hwmon_channel_info(struct hl_device *hdev,
@@ -323,7 +322,7 @@ int hl_get_temperature(struct hl_device *hdev,
 	pkt.type = __cpu_to_le16(attr);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-			SENSORS_PKT_TIMEOUT, value);
+						0, value);
 
 	if (rc) {
 		dev_err(hdev->dev,
@@ -350,7 +349,7 @@ int hl_set_temperature(struct hl_device *hdev,
 	pkt.value = __cpu_to_le64(value);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-						SENSORS_PKT_TIMEOUT, NULL);
+						0, NULL);
 
 	if (rc)
 		dev_err(hdev->dev,
@@ -374,7 +373,7 @@ int hl_get_voltage(struct hl_device *hdev,
 	pkt.type = __cpu_to_le16(attr);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-					SENSORS_PKT_TIMEOUT, value);
+						0, value);
 
 	if (rc) {
 		dev_err(hdev->dev,
@@ -400,7 +399,7 @@ int hl_get_current(struct hl_device *hdev,
 	pkt.type = __cpu_to_le16(attr);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-					SENSORS_PKT_TIMEOUT, value);
+						0, value);
 
 	if (rc) {
 		dev_err(hdev->dev,
@@ -426,7 +425,7 @@ int hl_get_fan_speed(struct hl_device *hdev,
 	pkt.type = __cpu_to_le16(attr);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-					SENSORS_PKT_TIMEOUT, value);
+						0, value);
 
 	if (rc) {
 		dev_err(hdev->dev,
@@ -452,7 +451,7 @@ int hl_get_pwm_info(struct hl_device *hdev,
 	pkt.type = __cpu_to_le16(attr);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-					SENSORS_PKT_TIMEOUT, value);
+						0, value);
 
 	if (rc) {
 		dev_err(hdev->dev,
@@ -479,7 +478,7 @@ void hl_set_pwm_info(struct hl_device *hdev, int sensor_index, u32 attr,
 	pkt.value = cpu_to_le64(value);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-					SENSORS_PKT_TIMEOUT, NULL);
+						0, NULL);
 
 	if (rc)
 		dev_err(hdev->dev,
@@ -502,7 +501,7 @@ int hl_set_voltage(struct hl_device *hdev,
 	pkt.value = __cpu_to_le64(value);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-						SENSORS_PKT_TIMEOUT, NULL);
+						0, NULL);
 
 	if (rc)
 		dev_err(hdev->dev,
@@ -527,7 +526,7 @@ int hl_set_current(struct hl_device *hdev,
 	pkt.value = __cpu_to_le64(value);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-						SENSORS_PKT_TIMEOUT, NULL);
+						0, NULL);
 
 	if (rc)
 		dev_err(hdev->dev,
diff --git a/drivers/misc/habanalabs/sysfs.c b/drivers/misc/habanalabs/sysfs.c
index 5d78d5e1c782..70b6b1863c2e 100644
--- a/drivers/misc/habanalabs/sysfs.c
+++ b/drivers/misc/habanalabs/sysfs.c
@@ -9,9 +9,6 @@
 
 #include <linux/pci.h>
 
-#define SET_CLK_PKT_TIMEOUT	1000000	/* 1s */
-#define SET_PWR_PKT_TIMEOUT	1000000	/* 1s */
-
 long hl_get_frequency(struct hl_device *hdev, u32 pll_index, bool curr)
 {
 	struct armcp_packet pkt;
@@ -29,7 +26,7 @@ long hl_get_frequency(struct hl_device *hdev, u32 pll_index, bool curr)
 	pkt.pll_index = cpu_to_le32(pll_index);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-						SET_CLK_PKT_TIMEOUT, &result);
+						0, &result);
 
 	if (rc) {
 		dev_err(hdev->dev,
@@ -54,7 +51,7 @@ void hl_set_frequency(struct hl_device *hdev, u32 pll_index, u64 freq)
 	pkt.value = cpu_to_le64(freq);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-					SET_CLK_PKT_TIMEOUT, NULL);
+						0, NULL);
 
 	if (rc)
 		dev_err(hdev->dev,
@@ -74,7 +71,7 @@ u64 hl_get_max_power(struct hl_device *hdev)
 				ARMCP_PKT_CTL_OPCODE_SHIFT);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-						SET_PWR_PKT_TIMEOUT, &result);
+						0, &result);
 
 	if (rc) {
 		dev_err(hdev->dev, "Failed to get max power, error %d\n", rc);
@@ -96,7 +93,7 @@ void hl_set_max_power(struct hl_device *hdev, u64 value)
 	pkt.value = cpu_to_le64(value);
 
 	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt),
-					SET_PWR_PKT_TIMEOUT, NULL);
+						0, NULL);
 
 	if (rc)
 		dev_err(hdev->dev, "Failed to set max power, error %d\n", rc);
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 0ba7a65e7c8d..06056e9ec333 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -547,6 +547,15 @@ static void scsi_mq_uninit_cmd(struct scsi_cmnd *cmd)
 	scsi_uninit_cmd(cmd);
 }
 
+static void scsi_run_queue_async(struct scsi_device *sdev)
+{
+	if (scsi_target(sdev)->single_lun ||
+	    !list_empty(&sdev->host->starved_list))
+		kblockd_schedule_work(&sdev->requeue_work);
+	else
+		blk_mq_run_hw_queues(sdev->request_queue, true);
+}
+
 /* Returns false when no more bytes to process, true if there are more */
 static bool scsi_end_request(struct request *req, blk_status_t error,
 		unsigned int bytes)
@@ -591,11 +600,7 @@ static bool scsi_end_request(struct request *req, blk_status_t error,
 
 	__blk_mq_end_request(req, error);
 
-	if (scsi_target(sdev)->single_lun ||
-	    !list_empty(&sdev->host->starved_list))
-		kblockd_schedule_work(&sdev->requeue_work);
-	else
-		blk_mq_run_hw_queues(q, true);
+	scsi_run_queue_async(sdev);
 
 	percpu_ref_put(&q->q_usage_counter);
 	return false;
@@ -1702,6 +1707,7 @@ out_put_budget:
 		 */
 		if (req->rq_flags & RQF_DONTPREP)
 			scsi_mq_uninit_cmd(cmd);
+		scsi_run_queue_async(sdev);
 		break;
 	}
 	return ret;
diff --git a/drivers/staging/comedi/drivers/addi_apci_1032.c b/drivers/staging/comedi/drivers/addi_apci_1032.c
index 560649be9d13..e035c9f757a1 100644
--- a/drivers/staging/comedi/drivers/addi_apci_1032.c
+++ b/drivers/staging/comedi/drivers/addi_apci_1032.c
@@ -106,14 +106,22 @@ static int apci1032_cos_insn_config(struct comedi_device *dev,
 				    unsigned int *data)
 {
 	struct apci1032_private *devpriv = dev->private;
-	unsigned int shift, oldmask;
+	unsigned int shift, oldmask, himask, lomask;
 
 	switch (data[0]) {
 	case INSN_CONFIG_DIGITAL_TRIG:
 		if (data[1] != 0)
 			return -EINVAL;
 		shift = data[3];
-		oldmask = (1U << shift) - 1;
+		if (shift < 32) {
+			oldmask = (1U << shift) - 1;
+			himask = data[4] << shift;
+			lomask = data[5] << shift;
+		} else {
+			oldmask = 0xffffffffu;
+			himask = 0;
+			lomask = 0;
+		}
 		switch (data[2]) {
 		case COMEDI_DIGITAL_TRIG_DISABLE:
 			devpriv->ctrl = 0;
@@ -136,8 +144,8 @@ static int apci1032_cos_insn_config(struct comedi_device *dev,
 				devpriv->mode2 &= oldmask;
 			}
 			/* configure specified channels */
-			devpriv->mode1 |= data[4] << shift;
-			devpriv->mode2 |= data[5] << shift;
+			devpriv->mode1 |= himask;
+			devpriv->mode2 |= lomask;
 			break;
 		case COMEDI_DIGITAL_TRIG_ENABLE_LEVELS:
 			if (devpriv->ctrl != (APCI1032_CTRL_INT_ENA |
@@ -154,8 +162,8 @@ static int apci1032_cos_insn_config(struct comedi_device *dev,
 				devpriv->mode2 &= oldmask;
 			}
 			/* configure specified channels */
-			devpriv->mode1 |= data[4] << shift;
-			devpriv->mode2 |= data[5] << shift;
+			devpriv->mode1 |= himask;
+			devpriv->mode2 |= lomask;
 			break;
 		default:
 			return -EINVAL;
diff --git a/drivers/staging/comedi/drivers/addi_apci_1500.c b/drivers/staging/comedi/drivers/addi_apci_1500.c
index 689acd69a1b9..816dd25b9d0e 100644
--- a/drivers/staging/comedi/drivers/addi_apci_1500.c
+++ b/drivers/staging/comedi/drivers/addi_apci_1500.c
@@ -452,13 +452,14 @@ static int apci1500_di_cfg_trig(struct comedi_device *dev,
 	struct apci1500_private *devpriv = dev->private;
 	unsigned int trig = data[1];
 	unsigned int shift = data[3];
-	unsigned int hi_mask = data[4] << shift;
-	unsigned int lo_mask = data[5] << shift;
-	unsigned int chan_mask = hi_mask | lo_mask;
-	unsigned int old_mask = (1 << shift) - 1;
+	unsigned int hi_mask;
+	unsigned int lo_mask;
+	unsigned int chan_mask;
+	unsigned int old_mask;
 	unsigned int pm;
 	unsigned int pt;
 	unsigned int pp;
+	unsigned int invalid_chan;
 
 	if (trig > 1) {
 		dev_dbg(dev->class_dev,
@@ -466,7 +467,20 @@ static int apci1500_di_cfg_trig(struct comedi_device *dev,
 		return -EINVAL;
 	}
 
-	if (chan_mask > 0xffff) {
+	if (shift <= 16) {
+		hi_mask = data[4] << shift;
+		lo_mask = data[5] << shift;
+		old_mask = (1U << shift) - 1;
+		invalid_chan = (data[4] | data[5]) >> (16 - shift);
+	} else {
+		hi_mask = 0;
+		lo_mask = 0;
+		old_mask = 0xffff;
+		invalid_chan = data[4] | data[5];
+	}
+	chan_mask = hi_mask | lo_mask;
+
+	if (invalid_chan) {
 		dev_dbg(dev->class_dev, "invalid digital trigger channel\n");
 		return -EINVAL;
 	}
diff --git a/drivers/staging/comedi/drivers/addi_apci_1564.c b/drivers/staging/comedi/drivers/addi_apci_1564.c
index 10501fe6bb25..1268ba34be5f 100644
--- a/drivers/staging/comedi/drivers/addi_apci_1564.c
+++ b/drivers/staging/comedi/drivers/addi_apci_1564.c
@@ -331,14 +331,22 @@ static int apci1564_cos_insn_config(struct comedi_device *dev,
 				    unsigned int *data)
 {
 	struct apci1564_private *devpriv = dev->private;
-	unsigned int shift, oldmask;
+	unsigned int shift, oldmask, himask, lomask;
 
 	switch (data[0]) {
 	case INSN_CONFIG_DIGITAL_TRIG:
 		if (data[1] != 0)
 			return -EINVAL;
 		shift = data[3];
-		oldmask = (1U << shift) - 1;
+		if (shift < 32) {
+			oldmask = (1U << shift) - 1;
+			himask = data[4] << shift;
+			lomask = data[5] << shift;
+		} else {
+			oldmask = 0xffffffffu;
+			himask = 0;
+			lomask = 0;
+		}
 		switch (data[2]) {
 		case COMEDI_DIGITAL_TRIG_DISABLE:
 			devpriv->ctrl = 0;
@@ -362,8 +370,8 @@ static int apci1564_cos_insn_config(struct comedi_device *dev,
 				devpriv->mode2 &= oldmask;
 			}
 			/* configure specified channels */
-			devpriv->mode1 |= data[4] << shift;
-			devpriv->mode2 |= data[5] << shift;
+			devpriv->mode1 |= himask;
+			devpriv->mode2 |= lomask;
 			break;
 		case COMEDI_DIGITAL_TRIG_ENABLE_LEVELS:
 			if (devpriv->ctrl != (APCI1564_DI_IRQ_ENA |
@@ -380,8 +388,8 @@ static int apci1564_cos_insn_config(struct comedi_device *dev,
 				devpriv->mode2 &= oldmask;
 			}
 			/* configure specified channels */
-			devpriv->mode1 |= data[4] << shift;
-			devpriv->mode2 |= data[5] << shift;
+			devpriv->mode1 |= himask;
+			devpriv->mode2 |= lomask;
 			break;
 		default:
 			return -EINVAL;
diff --git a/drivers/staging/comedi/drivers/ni_6527.c b/drivers/staging/comedi/drivers/ni_6527.c
index 4d1eccb5041d..4518c2680b7c 100644
--- a/drivers/staging/comedi/drivers/ni_6527.c
+++ b/drivers/staging/comedi/drivers/ni_6527.c
@@ -332,7 +332,7 @@ static int ni6527_intr_insn_config(struct comedi_device *dev,
 		case COMEDI_DIGITAL_TRIG_ENABLE_EDGES:
 			/* check shift amount */
 			shift = data[3];
-			if (shift >= s->n_chan) {
+			if (shift >= 32) {
 				mask = 0;
 				rising = 0;
 				falling = 0;
diff --git a/drivers/staging/wlan-ng/prism2usb.c b/drivers/staging/wlan-ng/prism2usb.c
index 4689b2170e4f..456603fd26c0 100644
--- a/drivers/staging/wlan-ng/prism2usb.c
+++ b/drivers/staging/wlan-ng/prism2usb.c
@@ -61,11 +61,25 @@ static int prism2sta_probe_usb(struct usb_interface *interface,
 			       const struct usb_device_id *id)
 {
 	struct usb_device *dev;
-
+	const struct usb_endpoint_descriptor *epd;
+	const struct usb_host_interface *iface_desc = interface->cur_altsetting;
 	struct wlandevice *wlandev = NULL;
 	struct hfa384x *hw = NULL;
 	int result = 0;
 
+	if (iface_desc->desc.bNumEndpoints != 2) {
+		result = -ENODEV;
+		goto failed;
+	}
+
+	result = -EINVAL;
+	epd = &iface_desc->endpoint[1].desc;
+	if (!usb_endpoint_is_bulk_in(epd))
+		goto failed;
+	epd = &iface_desc->endpoint[2].desc;
+	if (!usb_endpoint_is_bulk_out(epd))
+		goto failed;
+
 	dev = interface_to_usbdev(interface);
 	wlandev = create_wlan();
 	if (!wlandev) {
diff --git a/drivers/tty/serial/8250/8250_core.c b/drivers/tty/serial/8250/8250_core.c
index fc118f649887..cae61d1ebec5 100644
--- a/drivers/tty/serial/8250/8250_core.c
+++ b/drivers/tty/serial/8250/8250_core.c
@@ -524,6 +524,7 @@ static void __init serial8250_isa_init_ports(void)
 		 */
 		up->mcr_mask = ~ALPHA_KLUDGE_MCR;
 		up->mcr_force = ALPHA_KLUDGE_MCR;
+		serial8250_set_defaults(up);
 	}
 
 	/* chain base port ops to support Remote Supervisor Adapter */
@@ -547,7 +548,6 @@ static void __init serial8250_isa_init_ports(void)
 		port->membase  = old_serial_port[i].iomem_base;
 		port->iotype   = old_serial_port[i].io_type;
 		port->regshift = old_serial_port[i].iomem_reg_shift;
-		serial8250_set_defaults(up);
 
 		port->irqflags |= irqflag;
 		if (serial8250_isa_config != NULL)
diff --git a/drivers/tty/serial/8250/8250_exar.c b/drivers/tty/serial/8250/8250_exar.c
index ddb6aeb76dc5..04b9af7ed941 100644
--- a/drivers/tty/serial/8250/8250_exar.c
+++ b/drivers/tty/serial/8250/8250_exar.c
@@ -326,7 +326,17 @@ static void setup_gpio(struct pci_dev *pcidev, u8 __iomem *p)
 	 * devices will export them as GPIOs, so we pre-configure them safely
 	 * as inputs.
 	 */
-	u8 dir = pcidev->vendor == PCI_VENDOR_ID_EXAR ? 0xff : 0x00;
+
+	u8 dir = 0x00;
+
+	if  ((pcidev->vendor == PCI_VENDOR_ID_EXAR) &&
+		(pcidev->subsystem_vendor != PCI_VENDOR_ID_SEALEVEL)) {
+		// Configure GPIO as inputs for Commtech adapters
+		dir = 0xff;
+	} else {
+		// Configure GPIO as outputs for SeaLevel adapters
+		dir = 0x00;
+	}
 
 	writeb(0x00, p + UART_EXAR_MPIOINT_7_0);
 	writeb(0x00, p + UART_EXAR_MPIOLVL_7_0);
diff --git a/drivers/tty/serial/8250/8250_mtk.c b/drivers/tty/serial/8250/8250_mtk.c
index f839380c2f4c..98b8a3e30733 100644
--- a/drivers/tty/serial/8250/8250_mtk.c
+++ b/drivers/tty/serial/8250/8250_mtk.c
@@ -306,8 +306,21 @@ mtk8250_set_termios(struct uart_port *port, struct ktermios *termios,
 	}
 #endif
 
+	/*
+	 * Store the requested baud rate before calling the generic 8250
+	 * set_termios method. Standard 8250 port expects bauds to be
+	 * no higher than (uartclk / 16) so the baud will be clamped if it
+	 * gets out of that bound. Mediatek 8250 port supports speed
+	 * higher than that, therefore we'll get original baud rate back
+	 * after calling the generic set_termios method and recalculate
+	 * the speed later in this method.
+	 */
+	baud = tty_termios_baud_rate(termios);
+
 	serial8250_do_set_termios(port, termios, old);
 
+	tty_termios_encode_baud_rate(termios, baud, baud);
+
 	/*
 	 * Mediatek UARTs use an extra highspeed register (MTK_UART_HIGHS)
 	 *
@@ -339,6 +352,11 @@ mtk8250_set_termios(struct uart_port *port, struct ktermios *termios,
 	 */
 	spin_lock_irqsave(&port->lock, flags);
 
+	/*
+	 * Update the per-port timeout.
+	 */
+	uart_update_timeout(port, termios->c_cflag, baud);
+
 	/* set DLAB we have cval saved in up->lcr from the call to the core */
 	serial_port_out(port, UART_LCR, up->lcr | UART_LCR_DLAB);
 	serial_dl_write(up, quot);
diff --git a/drivers/tty/serial/serial-tegra.c b/drivers/tty/serial/serial-tegra.c
index 8de8bac9c6c7..04d1b0807e66 100644
--- a/drivers/tty/serial/serial-tegra.c
+++ b/drivers/tty/serial/serial-tegra.c
@@ -635,7 +635,7 @@ static void tegra_uart_handle_tx_pio(struct tegra_uart_port *tup)
 }
 
 static void tegra_uart_handle_rx_pio(struct tegra_uart_port *tup,
-		struct tty_port *tty)
+		struct tty_port *port)
 {
 	do {
 		char flag = TTY_NORMAL;
@@ -653,16 +653,18 @@ static void tegra_uart_handle_rx_pio(struct tegra_uart_port *tup,
 		ch = (unsigned char) tegra_uart_read(tup, UART_RX);
 		tup->uport.icount.rx++;
 
-		if (!uart_handle_sysrq_char(&tup->uport, ch) && tty)
-			tty_insert_flip_char(tty, ch, flag);
+		if (uart_handle_sysrq_char(&tup->uport, ch))
+			continue;
 
 		if (tup->uport.ignore_status_mask & UART_LSR_DR)
 			continue;
+
+		tty_insert_flip_char(port, ch, flag);
 	} while (1);
 }
 
 static void tegra_uart_copy_rx_to_tty(struct tegra_uart_port *tup,
-				      struct tty_port *tty,
+				      struct tty_port *port,
 				      unsigned int count)
 {
 	int copied;
@@ -672,17 +674,13 @@ static void tegra_uart_copy_rx_to_tty(struct tegra_uart_port *tup,
 		return;
 
 	tup->uport.icount.rx += count;
-	if (!tty) {
-		dev_err(tup->uport.dev, "No tty port\n");
-		return;
-	}
 
 	if (tup->uport.ignore_status_mask & UART_LSR_DR)
 		return;
 
 	dma_sync_single_for_cpu(tup->uport.dev, tup->rx_dma_buf_phys,
 				count, DMA_FROM_DEVICE);
-	copied = tty_insert_flip_string(tty,
+	copied = tty_insert_flip_string(port,
 			((unsigned char *)(tup->rx_dma_buf_virt)), count);
 	if (copied != count) {
 		WARN_ON(1);
diff --git a/drivers/tty/serial/xilinx_uartps.c b/drivers/tty/serial/xilinx_uartps.c
index 672cfa075e28..2833f1418d6d 100644
--- a/drivers/tty/serial/xilinx_uartps.c
+++ b/drivers/tty/serial/xilinx_uartps.c
@@ -1580,8 +1580,10 @@ static int cdns_uart_probe(struct platform_device *pdev)
 	 * If register_console() don't assign value, then console_port pointer
 	 * is cleanup.
 	 */
-	if (!console_port)
+	if (!console_port) {
+		cdns_uart_console.index = id;
 		console_port = port;
+	}
 #endif
 
 	rc = uart_add_one_port(&cdns_uart_uart_driver, port);
@@ -1594,8 +1596,10 @@ static int cdns_uart_probe(struct platform_device *pdev)
 #ifdef CONFIG_SERIAL_XILINX_PS_UART_CONSOLE
 	/* This is not port which is used for console that's why clean it up */
 	if (console_port == port &&
-	    !(cdns_uart_uart_driver.cons->flags & CON_ENABLED))
+	    !(cdns_uart_uart_driver.cons->flags & CON_ENABLED)) {
 		console_port = NULL;
+		cdns_uart_console.index = -1;
+	}
 #endif
 
 	cdns_uart_data->cts_override = of_property_read_bool(pdev->dev.of_node,
diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c
index 48a8199f7845..42d8c67a481f 100644
--- a/drivers/tty/vt/vt.c
+++ b/drivers/tty/vt/vt.c
@@ -1092,10 +1092,19 @@ static const struct tty_port_operations vc_port_ops = {
 	.destruct = vc_port_destruct,
 };
 
+/*
+ * Change # of rows and columns (0 means unchanged/the size of fg_console)
+ * [this is to be used together with some user program
+ * like resize that changes the hardware videomode]
+ */
+#define VC_MAXCOL (32767)
+#define VC_MAXROW (32767)
+
 int vc_allocate(unsigned int currcons)	/* return 0 on success */
 {
 	struct vt_notifier_param param;
 	struct vc_data *vc;
+	int err;
 
 	WARN_CONSOLE_UNLOCKED();
 
@@ -1125,6 +1134,11 @@ int vc_allocate(unsigned int currcons)	/* return 0 on success */
 	if (!*vc->vc_uni_pagedir_loc)
 		con_set_default_unimap(vc);
 
+	err = -EINVAL;
+	if (vc->vc_cols > VC_MAXCOL || vc->vc_rows > VC_MAXROW ||
+	    vc->vc_screenbuf_size > KMALLOC_MAX_SIZE || !vc->vc_screenbuf_size)
+		goto err_free;
+	err = -ENOMEM;
 	vc->vc_screenbuf = kzalloc(vc->vc_screenbuf_size, GFP_KERNEL);
 	if (!vc->vc_screenbuf)
 		goto err_free;
@@ -1143,7 +1157,7 @@ err_free:
 	visual_deinit(vc);
 	kfree(vc);
 	vc_cons[currcons].d = NULL;
-	return -ENOMEM;
+	return err;
 }
 
 static inline int resize_screen(struct vc_data *vc, int width, int height,
@@ -1158,14 +1172,6 @@ static inline int resize_screen(struct vc_data *vc, int width, int height,
 	return err;
 }
 
-/*
- * Change # of rows and columns (0 means unchanged/the size of fg_console)
- * [this is to be used together with some user program
- * like resize that changes the hardware videomode]
- */
-#define VC_RESIZE_MAXCOL (32767)
-#define VC_RESIZE_MAXROW (32767)
-
 /**
  *	vc_do_resize	-	resizing method for the tty
  *	@tty: tty being resized
@@ -1201,7 +1207,7 @@ static int vc_do_resize(struct tty_struct *tty, struct vc_data *vc,
 	user = vc->vc_resize_user;
 	vc->vc_resize_user = 0;
 
-	if (cols > VC_RESIZE_MAXCOL || lines > VC_RESIZE_MAXROW)
+	if (cols > VC_MAXCOL || lines > VC_MAXROW)
 		return -EINVAL;
 
 	new_cols = (cols ? cols : vc->vc_cols);
@@ -1212,7 +1218,7 @@ static int vc_do_resize(struct tty_struct *tty, struct vc_data *vc,
 	if (new_cols == vc->vc_cols && new_rows == vc->vc_rows)
 		return 0;
 
-	if (new_screen_size > KMALLOC_MAX_SIZE)
+	if (new_screen_size > KMALLOC_MAX_SIZE || !new_screen_size)
 		return -EINVAL;
 	newscreen = kzalloc(new_screen_size, GFP_USER);
 	if (!newscreen)
@@ -3393,6 +3399,7 @@ static int __init con_init(void)
 		INIT_WORK(&vc_cons[currcons].SAK_work, vc_SAK);
 		tty_port_init(&vc->port);
 		visual_init(vc, currcons, 1);
+		/* Assuming vc->vc_{cols,rows,screenbuf_size} are sane here. */
 		vc->vc_screenbuf = kzalloc(vc->vc_screenbuf_size, GFP_NOWAIT);
 		vc_init(vc, vc->vc_rows, vc->vc_cols,
 			currcons || !vc->vc_sw->con_save_screen);
diff --git a/drivers/usb/host/xhci-mtk-sch.c b/drivers/usb/host/xhci-mtk-sch.c
index fea555570ad4..45c54d56ecbd 100644
--- a/drivers/usb/host/xhci-mtk-sch.c
+++ b/drivers/usb/host/xhci-mtk-sch.c
@@ -557,6 +557,10 @@ static bool need_bw_sch(struct usb_host_endpoint *ep,
 	if (is_fs_or_ls(speed) && !has_tt)
 		return false;
 
+	/* skip endpoint with zero maxpkt */
+	if (usb_endpoint_maxp(&ep->desc) == 0)
+		return false;
+
 	return true;
 }
 
diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c
index ef513c2fb843..9234c82e70e4 100644
--- a/drivers/usb/host/xhci-pci.c
+++ b/drivers/usb/host/xhci-pci.c
@@ -265,6 +265,9 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
 	if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA &&
 			pdev->device == 0x1142)
 		xhci->quirks |= XHCI_TRUST_TX_LENGTH;
+	if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA &&
+			pdev->device == 0x2142)
+		xhci->quirks |= XHCI_NO_64BIT_SUPPORT;
 
 	if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA &&
 		pdev->device == PCI_DEVICE_ID_ASMEDIA_1042A_XHCI)
diff --git a/drivers/usb/host/xhci-tegra.c b/drivers/usb/host/xhci-tegra.c
index 2eaf5c0af80c..ee6bf01775bb 100644
--- a/drivers/usb/host/xhci-tegra.c
+++ b/drivers/usb/host/xhci-tegra.c
@@ -856,7 +856,7 @@ static int tegra_xusb_init_context(struct tegra_xusb *tegra)
 	if (!tegra->context.ipfs)
 		return -ENOMEM;
 
-	tegra->context.fpci = devm_kcalloc(tegra->dev, soc->ipfs.num_offsets,
+	tegra->context.fpci = devm_kcalloc(tegra->dev, soc->fpci.num_offsets,
 					   sizeof(u32), GFP_KERNEL);
 	if (!tegra->context.fpci)
 		return -ENOMEM;
diff --git a/drivers/video/fbdev/core/bitblit.c b/drivers/video/fbdev/core/bitblit.c
index ca935c09a261..35ebeeccde4d 100644
--- a/drivers/video/fbdev/core/bitblit.c
+++ b/drivers/video/fbdev/core/bitblit.c
@@ -216,7 +216,7 @@ static void bit_clear_margins(struct vc_data *vc, struct fb_info *info,
 	region.color = color;
 	region.rop = ROP_COPY;
 
-	if (rw && !bottom_only) {
+	if ((int) rw > 0 && !bottom_only) {
 		region.dx = info->var.xoffset + rs;
 		region.dy = 0;
 		region.width = rw;
@@ -224,7 +224,7 @@ static void bit_clear_margins(struct vc_data *vc, struct fb_info *info,
 		info->fbops->fb_fillrect(info, &region);
 	}
 
-	if (bh) {
+	if ((int) bh > 0) {
 		region.dx = info->var.xoffset;
 		region.dy = info->var.yoffset + bs;
 		region.width = rs;
diff --git a/drivers/video/fbdev/core/fbcon_ccw.c b/drivers/video/fbdev/core/fbcon_ccw.c
index dfa9a8aa4509..78f3a5621478 100644
--- a/drivers/video/fbdev/core/fbcon_ccw.c
+++ b/drivers/video/fbdev/core/fbcon_ccw.c
@@ -201,7 +201,7 @@ static void ccw_clear_margins(struct vc_data *vc, struct fb_info *info,
 	region.color = color;
 	region.rop = ROP_COPY;
 
-	if (rw && !bottom_only) {
+	if ((int) rw > 0 && !bottom_only) {
 		region.dx = 0;
 		region.dy = info->var.yoffset;
 		region.height = rw;
@@ -209,7 +209,7 @@ static void ccw_clear_margins(struct vc_data *vc, struct fb_info *info,
 		info->fbops->fb_fillrect(info, &region);
 	}
 
-	if (bh) {
+	if ((int) bh > 0) {
 		region.dx = info->var.xoffset + bs;
 		region.dy = 0;
                 region.height = info->var.yres_virtual;
diff --git a/drivers/video/fbdev/core/fbcon_cw.c b/drivers/video/fbdev/core/fbcon_cw.c
index ce08251bfd38..fd098ff17574 100644
--- a/drivers/video/fbdev/core/fbcon_cw.c
+++ b/drivers/video/fbdev/core/fbcon_cw.c
@@ -184,7 +184,7 @@ static void cw_clear_margins(struct vc_data *vc, struct fb_info *info,
 	region.color = color;
 	region.rop = ROP_COPY;
 
-	if (rw && !bottom_only) {
+	if ((int) rw > 0 && !bottom_only) {
 		region.dx = 0;
 		region.dy = info->var.yoffset + rs;
 		region.height = rw;
@@ -192,7 +192,7 @@ static void cw_clear_margins(struct vc_data *vc, struct fb_info *info,
 		info->fbops->fb_fillrect(info, &region);
 	}
 
-	if (bh) {
+	if ((int) bh > 0) {
 		region.dx = info->var.xoffset;
 		region.dy = info->var.yoffset;
                 region.height = info->var.yres;
diff --git a/drivers/video/fbdev/core/fbcon_ud.c b/drivers/video/fbdev/core/fbcon_ud.c
index 1936afc78fec..e165a3fad29a 100644
--- a/drivers/video/fbdev/core/fbcon_ud.c
+++ b/drivers/video/fbdev/core/fbcon_ud.c
@@ -231,7 +231,7 @@ static void ud_clear_margins(struct vc_data *vc, struct fb_info *info,
 	region.color = color;
 	region.rop = ROP_COPY;
 
-	if (rw && !bottom_only) {
+	if ((int) rw > 0 && !bottom_only) {
 		region.dy = 0;
 		region.dx = info->var.xoffset;
 		region.width  = rw;
@@ -239,7 +239,7 @@ static void ud_clear_margins(struct vc_data *vc, struct fb_info *info,
 		info->fbops->fb_fillrect(info, &region);
 	}
 
-	if (bh) {
+	if ((int) bh > 0) {
 		region.dy = info->var.yoffset;
 		region.dx = info->var.xoffset;
                 region.height  = bh;
diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c
index 6aea65c65745..45f2ab2ec2d4 100644
--- a/scripts/mod/modpost.c
+++ b/scripts/mod/modpost.c
@@ -138,11 +138,19 @@ char *read_text_file(const char *filename)
 
 char *get_line(char **stringp)
 {
+	char *orig = *stringp, *next;
+
 	/* do not return the unwanted extra line at EOF */
-	if (*stringp && **stringp == '\0')
+	if (!orig || *orig == '\0')
 		return NULL;
 
-	return strsep(stringp, "\n");
+	next = strchr(orig, '\n');
+	if (next)
+		*next++ = '\0';
+
+	*stringp = next;
+
+	return orig;
 }
 
 /* A list of all modules we processed */