Merge tag 'hyperv-fixes-signed-20240411' of...

Merge tag 'hyperv-fixes-signed-20240411' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux

Pull hyperv fixes from Wei Liu:

 - Some cosmetic changes (Erni Sri Satya Vennela, Li Zhijian)

 - Introduce hv_numa_node_to_pxm_info() (Nuno Das Neves)

 - Fix KVP daemon to handle IPv4 and IPv6 combination for keyfile format
   (Shradha Gupta)

 - Avoid freeing decrypted memory in a confidential VM (Rick Edgecombe
   and Michael Kelley)

* tag 'hyperv-fixes-signed-20240411' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux:
  Drivers: hv: vmbus: Don't free ring buffers that couldn't be re-encrypted
  uio_hv_generic: Don't free decrypted memory
  hv_netvsc: Don't free decrypted memory
  Drivers: hv: vmbus: Track decrypted status in vmbus_gpadl
  Drivers: hv: vmbus: Leak pages if set_memory_encrypted() fails
  hv/hv_kvp_daemon: Handle IPv4 and Ipv6 combination for keyfile format
  hv: vmbus: Convert sprintf() family to sysfs_emit() family
  mshyperv: Introduce hv_numa_node_to_pxm_info()
  x86/hyperv: Cosmetic changes for hv_apic.c

arch/x86/hyperv/hv_apic.c

@@ -132,7 +132,7 @@ static bool __send_ipi_mask_ex(const struct cpumask *mask, int vector,
 	if (!cpumask_equal(mask, cpu_present_mask) || exclude_self) {
 		ipi_arg->vp_set.format = HV_GENERIC_SET_SPARSE_4K;
 
-		nr_bank = cpumask_to_vpset_skip(&(ipi_arg->vp_set), mask,
+		nr_bank = cpumask_to_vpset_skip(&ipi_arg->vp_set, mask,
 						exclude_self ? cpu_is_self : NULL);
 
 		/*
@@ -181,7 +181,7 @@ static bool __send_ipi_mask(const struct cpumask *mask, int vector,
 			return false;
 	}
 
-	if ((vector < HV_IPI_LOW_VECTOR) || (vector > HV_IPI_HIGH_VECTOR))
+	if (vector < HV_IPI_LOW_VECTOR || vector > HV_IPI_HIGH_VECTOR)
 		return false;
 
 	/*
@@ -241,7 +241,7 @@ static bool __send_ipi_one(int cpu, int vector)
 			return false;
 	}
 
-	if ((vector < HV_IPI_LOW_VECTOR) || (vector > HV_IPI_HIGH_VECTOR))
+	if (vector < HV_IPI_LOW_VECTOR || vector > HV_IPI_HIGH_VECTOR)
 		return false;
 
 	if (vp >= 64)

arch/x86/hyperv/hv_proc.c

@@ -3,7 +3,6 @@
 #include <linux/vmalloc.h>
 #include <linux/mm.h>
 #include <linux/clockchips.h>
-#include <linux/acpi.h>
 #include <linux/hyperv.h>
 #include <linux/slab.h>
 #include <linux/cpuhotplug.h>
@@ -116,12 +115,11 @@ int hv_call_deposit_pages(int node, u64 partition_id, u32 num_pages)
 
 int hv_call_add_logical_proc(int node, u32 lp_index, u32 apic_id)
 {
-	struct hv_add_logical_processor_in *input;
-	struct hv_add_logical_processor_out *output;
+	struct hv_input_add_logical_processor *input;
+	struct hv_output_add_logical_processor *output;
 	u64 status;
 	unsigned long flags;
 	int ret = HV_STATUS_SUCCESS;
-	int pxm = node_to_pxm(node);
 
 	/*
 	 * When adding a logical processor, the hypervisor may return
@@ -137,11 +135,7 @@ int hv_call_add_logical_proc(int node, u32 lp_index, u32 apic_id)
 
 		input->lp_index = lp_index;
 		input->apic_id = apic_id;
-		input->flags = 0;
-		input->proximity_domain_info.domain_id = pxm;
-		input->proximity_domain_info.flags.reserved = 0;
-		input->proximity_domain_info.flags.proximity_info_valid = 1;
-		input->proximity_domain_info.flags.proximity_preferred = 1;
+		input->proximity_domain_info = hv_numa_node_to_pxm_info(node);
 		status = hv_do_hypercall(HVCALL_ADD_LOGICAL_PROCESSOR,
 					 input, output);
 		local_irq_restore(flags);
@@ -166,7 +160,6 @@ int hv_call_create_vp(int node, u64 partition_id, u32 vp_index, u32 flags)
 	u64 status;
 	unsigned long irq_flags;
 	int ret = HV_STATUS_SUCCESS;
-	int pxm = node_to_pxm(node);
 
 	/* Root VPs don't seem to need pages deposited */
 	if (partition_id != hv_current_partition_id) {
@@ -185,14 +178,7 @@ int hv_call_create_vp(int node, u64 partition_id, u32 vp_index, u32 flags)
 		input->vp_index = vp_index;
 		input->flags = flags;
 		input->subnode_type = HvSubnodeAny;
-		if (node != NUMA_NO_NODE) {
-			input->proximity_domain_info.domain_id = pxm;
-			input->proximity_domain_info.flags.reserved = 0;
-			input->proximity_domain_info.flags.proximity_info_valid = 1;
-			input->proximity_domain_info.flags.proximity_preferred = 1;
-		} else {
-			input->proximity_domain_info.as_uint64 = 0;
-		}
+		input->proximity_domain_info = hv_numa_node_to_pxm_info(node);
 		status = hv_do_hypercall(HVCALL_CREATE_VP, input, NULL);
 		local_irq_restore(irq_flags);
 

drivers/hv/channel.c

@@ -153,6 +153,8 @@ void vmbus_free_ring(struct vmbus_channel *channel)
 	hv_ringbuffer_cleanup(&channel->inbound);
 
 	if (channel->ringbuffer_page) {
+		/* In a CoCo VM leak the memory if it didn't get re-encrypted */
+		if (!channel->ringbuffer_gpadlhandle.decrypted)
 			__free_pages(channel->ringbuffer_page,
 			     get_order(channel->ringbuffer_pagecount
 				       << PAGE_SHIFT));
@@ -436,9 +438,18 @@ static int __vmbus_establish_gpadl(struct vmbus_channel *channel,
 		(atomic_inc_return(&vmbus_connection.next_gpadl_handle) - 1);
 
 	ret = create_gpadl_header(type, kbuffer, size, send_offset, &msginfo);
-	if (ret)
+	if (ret) {
+		gpadl->decrypted = false;
 		return ret;
+	}
 
+	/*
+	 * Set the "decrypted" flag to true for the set_memory_decrypted()
+	 * success case. In the failure case, the encryption state of the
+	 * memory is unknown. Leave "decrypted" as true to ensure the
+	 * memory will be leaked instead of going back on the free list.
+	 */
+	gpadl->decrypted = true;
 	ret = set_memory_decrypted((unsigned long)kbuffer,
 				   PFN_UP(size));
 	if (ret) {
@@ -527,9 +538,15 @@ static int __vmbus_establish_gpadl(struct vmbus_channel *channel,
 
 	kfree(msginfo);
 
-	if (ret)
-		set_memory_encrypted((unsigned long)kbuffer,
-				     PFN_UP(size));
+	if (ret) {
+		/*
+		 * If set_memory_encrypted() fails, the decrypted flag is
+		 * left as true so the memory is leaked instead of being
+		 * put back on the free list.
+		 */
+		if (!set_memory_encrypted((unsigned long)kbuffer, PFN_UP(size)))
+			gpadl->decrypted = false;
+	}
 
 	return ret;
 }
@@ -850,6 +867,8 @@ int vmbus_teardown_gpadl(struct vmbus_channel *channel, struct vmbus_gpadl *gpad
 	if (ret)
 		pr_warn("Fail to set mem host visibility in GPADL teardown %d.\n", ret);
 
+	gpadl->decrypted = ret;
+
 	return ret;
 }
 EXPORT_SYMBOL_GPL(vmbus_teardown_gpadl);

drivers/hv/connection.c

@@ -237,8 +237,17 @@ int vmbus_connect(void)
 				vmbus_connection.monitor_pages[0], 1);
 	ret |= set_memory_decrypted((unsigned long)
 				vmbus_connection.monitor_pages[1], 1);
-	if (ret)
+	if (ret) {
+		/*
+		 * If set_memory_decrypted() fails, the encryption state
+		 * of the memory is unknown. So leak the memory instead
+		 * of risking returning decrypted memory to the free list.
+		 * For simplicity, always handle both pages the same.
+		 */
+		vmbus_connection.monitor_pages[0] = NULL;
+		vmbus_connection.monitor_pages[1] = NULL;
 		goto cleanup;
+	}
 
 	/*
 	 * Set_memory_decrypted() will change the memory contents if
@@ -337,13 +346,19 @@ void vmbus_disconnect(void)
 		vmbus_connection.int_page = NULL;
 	}
 
-	set_memory_encrypted((unsigned long)vmbus_connection.monitor_pages[0], 1);
-	set_memory_encrypted((unsigned long)vmbus_connection.monitor_pages[1], 1);
-
+	if (vmbus_connection.monitor_pages[0]) {
+		if (!set_memory_encrypted(
+			(unsigned long)vmbus_connection.monitor_pages[0], 1))
 			hv_free_hyperv_page(vmbus_connection.monitor_pages[0]);
-	hv_free_hyperv_page(vmbus_connection.monitor_pages[1]);
 		vmbus_connection.monitor_pages[0] = NULL;
+	}
+
+	if (vmbus_connection.monitor_pages[1]) {
+		if (!set_memory_encrypted(
+			(unsigned long)vmbus_connection.monitor_pages[1], 1))
+			hv_free_hyperv_page(vmbus_connection.monitor_pages[1]);
 		vmbus_connection.monitor_pages[1] = NULL;
+	}
 }
 
 /*

drivers/hv/vmbus_drv.c

@@ -131,7 +131,7 @@ static ssize_t id_show(struct device *dev, struct device_attribute *dev_attr,
 
 	if (!hv_dev->channel)
 		return -ENODEV;
-	return sprintf(buf, "%d\n", hv_dev->channel->offermsg.child_relid);
+	return sysfs_emit(buf, "%d\n", hv_dev->channel->offermsg.child_relid);
 }
 static DEVICE_ATTR_RO(id);
 
@@ -142,7 +142,7 @@ static ssize_t state_show(struct device *dev, struct device_attribute *dev_attr,
 
 	if (!hv_dev->channel)
 		return -ENODEV;
-	return sprintf(buf, "%d\n", hv_dev->channel->state);
+	return sysfs_emit(buf, "%d\n", hv_dev->channel->state);
 }
 static DEVICE_ATTR_RO(state);
 
@@ -153,7 +153,7 @@ static ssize_t monitor_id_show(struct device *dev,
 
 	if (!hv_dev->channel)
 		return -ENODEV;
-	return sprintf(buf, "%d\n", hv_dev->channel->offermsg.monitorid);
+	return sysfs_emit(buf, "%d\n", hv_dev->channel->offermsg.monitorid);
 }
 static DEVICE_ATTR_RO(monitor_id);
 
@@ -164,7 +164,7 @@ static ssize_t class_id_show(struct device *dev,
 
 	if (!hv_dev->channel)
 		return -ENODEV;
-	return sprintf(buf, "{%pUl}\n",
+	return sysfs_emit(buf, "{%pUl}\n",
 			  &hv_dev->channel->offermsg.offer.if_type);
 }
 static DEVICE_ATTR_RO(class_id);
@@ -176,7 +176,7 @@ static ssize_t device_id_show(struct device *dev,
 
 	if (!hv_dev->channel)
 		return -ENODEV;
-	return sprintf(buf, "{%pUl}\n",
+	return sysfs_emit(buf, "{%pUl}\n",
 			  &hv_dev->channel->offermsg.offer.if_instance);
 }
 static DEVICE_ATTR_RO(device_id);
@@ -186,7 +186,7 @@ static ssize_t modalias_show(struct device *dev,
 {
 	struct hv_device *hv_dev = device_to_hv_device(dev);
 
-	return sprintf(buf, "vmbus:%*phN\n", UUID_SIZE, &hv_dev->dev_type);
+	return sysfs_emit(buf, "vmbus:%*phN\n", UUID_SIZE, &hv_dev->dev_type);
 }
 static DEVICE_ATTR_RO(modalias);
 
@@ -199,7 +199,7 @@ static ssize_t numa_node_show(struct device *dev,
 	if (!hv_dev->channel)
 		return -ENODEV;
 
-	return sprintf(buf, "%d\n", cpu_to_node(hv_dev->channel->target_cpu));
+	return sysfs_emit(buf, "%d\n", cpu_to_node(hv_dev->channel->target_cpu));
 }
 static DEVICE_ATTR_RO(numa_node);
 #endif
@@ -212,8 +212,7 @@ static ssize_t server_monitor_pending_show(struct device *dev,
 
 	if (!hv_dev->channel)
 		return -ENODEV;
-	return sprintf(buf, "%d\n",
-		       channel_pending(hv_dev->channel,
+	return sysfs_emit(buf, "%d\n", channel_pending(hv_dev->channel,
 			  vmbus_connection.monitor_pages[0]));
 }
 static DEVICE_ATTR_RO(server_monitor_pending);
@@ -226,8 +225,7 @@ static ssize_t client_monitor_pending_show(struct device *dev,
 
 	if (!hv_dev->channel)
 		return -ENODEV;
-	return sprintf(buf, "%d\n",
-		       channel_pending(hv_dev->channel,
+	return sysfs_emit(buf, "%d\n", channel_pending(hv_dev->channel,
 			  vmbus_connection.monitor_pages[1]));
 }
 static DEVICE_ATTR_RO(client_monitor_pending);
@@ -240,8 +238,7 @@ static ssize_t server_monitor_latency_show(struct device *dev,
 
 	if (!hv_dev->channel)
 		return -ENODEV;
-	return sprintf(buf, "%d\n",
-		       channel_latency(hv_dev->channel,
+	return sysfs_emit(buf, "%d\n", channel_latency(hv_dev->channel,
 			  vmbus_connection.monitor_pages[0]));
 }
 static DEVICE_ATTR_RO(server_monitor_latency);
@@ -254,8 +251,7 @@ static ssize_t client_monitor_latency_show(struct device *dev,
 
 	if (!hv_dev->channel)
 		return -ENODEV;
-	return sprintf(buf, "%d\n",
-		       channel_latency(hv_dev->channel,
+	return sysfs_emit(buf, "%d\n", channel_latency(hv_dev->channel,
 			  vmbus_connection.monitor_pages[1]));
 }
 static DEVICE_ATTR_RO(client_monitor_latency);
@@ -268,8 +264,7 @@ static ssize_t server_monitor_conn_id_show(struct device *dev,
 
 	if (!hv_dev->channel)
 		return -ENODEV;
-	return sprintf(buf, "%d\n",
-		       channel_conn_id(hv_dev->channel,
+	return sysfs_emit(buf, "%d\n", channel_conn_id(hv_dev->channel,
 			  vmbus_connection.monitor_pages[0]));
 }
 static DEVICE_ATTR_RO(server_monitor_conn_id);
@@ -282,8 +277,7 @@ static ssize_t client_monitor_conn_id_show(struct device *dev,
 
 	if (!hv_dev->channel)
 		return -ENODEV;
-	return sprintf(buf, "%d\n",
-		       channel_conn_id(hv_dev->channel,
+	return sysfs_emit(buf, "%d\n", channel_conn_id(hv_dev->channel,
 			  vmbus_connection.monitor_pages[1]));
 }
 static DEVICE_ATTR_RO(client_monitor_conn_id);
@@ -303,7 +297,7 @@ static ssize_t out_intr_mask_show(struct device *dev,
 	if (ret < 0)
 		return ret;
 
-	return sprintf(buf, "%d\n", outbound.current_interrupt_mask);
+	return sysfs_emit(buf, "%d\n", outbound.current_interrupt_mask);
 }
 static DEVICE_ATTR_RO(out_intr_mask);
 
@@ -321,7 +315,7 @@ static ssize_t out_read_index_show(struct device *dev,
 					  &outbound);
 	if (ret < 0)
 		return ret;
-	return sprintf(buf, "%d\n", outbound.current_read_index);
+	return sysfs_emit(buf, "%d\n", outbound.current_read_index);
 }
 static DEVICE_ATTR_RO(out_read_index);
 
@@ -340,7 +334,7 @@ static ssize_t out_write_index_show(struct device *dev,
 					  &outbound);
 	if (ret < 0)
 		return ret;
-	return sprintf(buf, "%d\n", outbound.current_write_index);
+	return sysfs_emit(buf, "%d\n", outbound.current_write_index);
 }
 static DEVICE_ATTR_RO(out_write_index);
 
@@ -359,7 +353,7 @@ static ssize_t out_read_bytes_avail_show(struct device *dev,
 					  &outbound);
 	if (ret < 0)
 		return ret;
-	return sprintf(buf, "%d\n", outbound.bytes_avail_toread);
+	return sysfs_emit(buf, "%d\n", outbound.bytes_avail_toread);
 }
 static DEVICE_ATTR_RO(out_read_bytes_avail);
 
@@ -378,7 +372,7 @@ static ssize_t out_write_bytes_avail_show(struct device *dev,
 					  &outbound);
 	if (ret < 0)
 		return ret;
-	return sprintf(buf, "%d\n", outbound.bytes_avail_towrite);
+	return sysfs_emit(buf, "%d\n", outbound.bytes_avail_towrite);
 }
 static DEVICE_ATTR_RO(out_write_bytes_avail);
 
@@ -396,7 +390,7 @@ static ssize_t in_intr_mask_show(struct device *dev,
 	if (ret < 0)
 		return ret;
 
-	return sprintf(buf, "%d\n", inbound.current_interrupt_mask);
+	return sysfs_emit(buf, "%d\n", inbound.current_interrupt_mask);
 }
 static DEVICE_ATTR_RO(in_intr_mask);
 
@@ -414,7 +408,7 @@ static ssize_t in_read_index_show(struct device *dev,
 	if (ret < 0)
 		return ret;
 
-	return sprintf(buf, "%d\n", inbound.current_read_index);
+	return sysfs_emit(buf, "%d\n", inbound.current_read_index);
 }
 static DEVICE_ATTR_RO(in_read_index);
 
@@ -432,7 +426,7 @@ static ssize_t in_write_index_show(struct device *dev,
 	if (ret < 0)
 		return ret;
 
-	return sprintf(buf, "%d\n", inbound.current_write_index);
+	return sysfs_emit(buf, "%d\n", inbound.current_write_index);
 }
 static DEVICE_ATTR_RO(in_write_index);
 
@@ -451,7 +445,7 @@ static ssize_t in_read_bytes_avail_show(struct device *dev,
 	if (ret < 0)
 		return ret;
 
-	return sprintf(buf, "%d\n", inbound.bytes_avail_toread);
+	return sysfs_emit(buf, "%d\n", inbound.bytes_avail_toread);
 }
 static DEVICE_ATTR_RO(in_read_bytes_avail);
 
@@ -470,7 +464,7 @@ static ssize_t in_write_bytes_avail_show(struct device *dev,
 	if (ret < 0)
 		return ret;
 
-	return sprintf(buf, "%d\n", inbound.bytes_avail_towrite);
+	return sysfs_emit(buf, "%d\n", inbound.bytes_avail_towrite);
 }
 static DEVICE_ATTR_RO(in_write_bytes_avail);
 
@@ -480,7 +474,7 @@ static ssize_t channel_vp_mapping_show(struct device *dev,
 {
 	struct hv_device *hv_dev = device_to_hv_device(dev);
 	struct vmbus_channel *channel = hv_dev->channel, *cur_sc;
-	int buf_size = PAGE_SIZE, n_written, tot_written;
+	int n_written;
 	struct list_head *cur;
 
 	if (!channel)
@@ -488,25 +482,21 @@ static ssize_t channel_vp_mapping_show(struct device *dev,
 
 	mutex_lock(&vmbus_connection.channel_mutex);
 
-	tot_written = snprintf(buf, buf_size, "%u:%u\n",
-		channel->offermsg.child_relid, channel->target_cpu);
+	n_written = sysfs_emit(buf, "%u:%u\n",
+			       channel->offermsg.child_relid,
+			       channel->target_cpu);
 
 	list_for_each(cur, &channel->sc_list) {
-		if (tot_written >= buf_size - 1)
-			break;
 
 		cur_sc = list_entry(cur, struct vmbus_channel, sc_list);
-		n_written = scnprintf(buf + tot_written,
-				     buf_size - tot_written,
-				     "%u:%u\n",
+		n_written += sysfs_emit_at(buf, n_written, "%u:%u\n",
 					  cur_sc->offermsg.child_relid,
 					  cur_sc->target_cpu);
-		tot_written += n_written;
 	}
 
 	mutex_unlock(&vmbus_connection.channel_mutex);
 
-	return tot_written;
+	return n_written;
 }
 static DEVICE_ATTR_RO(channel_vp_mapping);
 
@@ -516,7 +506,7 @@ static ssize_t vendor_show(struct device *dev,
 {
 	struct hv_device *hv_dev = device_to_hv_device(dev);
 
-	return sprintf(buf, "0x%x\n", hv_dev->vendor_id);
+	return sysfs_emit(buf, "0x%x\n", hv_dev->vendor_id);
 }
 static DEVICE_ATTR_RO(vendor);
 
@@ -526,7 +516,7 @@ static ssize_t device_show(struct device *dev,
 {
 	struct hv_device *hv_dev = device_to_hv_device(dev);
 
-	return sprintf(buf, "0x%x\n", hv_dev->device_id);
+	return sysfs_emit(buf, "0x%x\n", hv_dev->device_id);
 }
 static DEVICE_ATTR_RO(device);
 
@@ -551,7 +541,7 @@ static ssize_t driver_override_show(struct device *dev,
 	ssize_t len;
 
 	device_lock(dev);
-	len = snprintf(buf, PAGE_SIZE, "%s\n", hv_dev->driver_override);
+	len = sysfs_emit(buf, "%s\n", hv_dev->driver_override);
 	device_unlock(dev);
 
 	return len;

drivers/net/hyperv/netvsc.c

@@ -154,7 +154,10 @@ static void free_netvsc_device(struct rcu_head *head)
 	int i;
 
 	kfree(nvdev->extension);
+
+	if (!nvdev->recv_buf_gpadl_handle.decrypted)
 		vfree(nvdev->recv_buf);
+	if (!nvdev->send_buf_gpadl_handle.decrypted)
 		vfree(nvdev->send_buf);
 	bitmap_free(nvdev->send_section_map);
 

drivers/uio/uio_hv_generic.c

@@ -181,11 +181,13 @@ hv_uio_cleanup(struct hv_device *dev, struct hv_uio_private_data *pdata)
 {
 	if (pdata->send_gpadl.gpadl_handle) {
 		vmbus_teardown_gpadl(dev->channel, &pdata->send_gpadl);
+		if (!pdata->send_gpadl.decrypted)
 			vfree(pdata->send_buf);
 	}
 
 	if (pdata->recv_gpadl.gpadl_handle) {
 		vmbus_teardown_gpadl(dev->channel, &pdata->recv_gpadl);
+		if (!pdata->recv_gpadl.decrypted)
 			vfree(pdata->recv_buf);
 	}
 }
@@ -295,6 +297,7 @@ hv_uio_probe(struct hv_device *dev,
 	ret = vmbus_establish_gpadl(channel, pdata->recv_buf,
 				    RECV_BUFFER_SIZE, &pdata->recv_gpadl);
 	if (ret) {
+		if (!pdata->recv_gpadl.decrypted)
 			vfree(pdata->recv_buf);
 		goto fail_close;
 	}
@@ -317,6 +320,7 @@ hv_uio_probe(struct hv_device *dev,
 	ret = vmbus_establish_gpadl(channel, pdata->send_buf,
 				    SEND_BUFFER_SIZE, &pdata->send_gpadl);
 	if (ret) {
+		if (!pdata->send_gpadl.decrypted)
 			vfree(pdata->send_buf);
 		goto fail_close;
 	}

include/asm-generic/hyperv-tlfs.h

@@ -512,13 +512,9 @@ struct hv_proximity_domain_flags {
 	u32 proximity_info_valid : 1;
 } __packed;
 
-/* Not a union in windows but useful for zeroing */
-union hv_proximity_domain_info {
-	struct {
+struct hv_proximity_domain_info {
 	u32 domain_id;
 	struct hv_proximity_domain_flags flags;
-	};
-	u64 as_uint64;
 } __packed;
 
 struct hv_lp_startup_status {
@@ -532,14 +528,13 @@ struct hv_lp_startup_status {
 } __packed;
 
 /* HvAddLogicalProcessor hypercall */
-struct hv_add_logical_processor_in {
+struct hv_input_add_logical_processor {
 	u32 lp_index;
 	u32 apic_id;
-	union hv_proximity_domain_info proximity_domain_info;
-	u64 flags;
+	struct hv_proximity_domain_info proximity_domain_info;
 } __packed;
 
-struct hv_add_logical_processor_out {
+struct hv_output_add_logical_processor {
 	struct hv_lp_startup_status startup_status;
 } __packed;
 
@@ -560,7 +555,7 @@ struct hv_create_vp {
 	u8 padding[3];
 	u8 subnode_type;
 	u64 subnode_id;
-	union hv_proximity_domain_info proximity_domain_info;
+	struct hv_proximity_domain_info proximity_domain_info;
 	u64 flags;
 } __packed;
 

include/asm-generic/mshyperv.h

@@ -21,6 +21,7 @@
 #include <linux/types.h>
 #include <linux/atomic.h>
 #include <linux/bitops.h>
+#include <acpi/acpi_numa.h>
 #include <linux/cpumask.h>
 #include <linux/nmi.h>
 #include <asm/ptrace.h>
@@ -67,6 +68,19 @@ extern u64 hv_do_fast_hypercall8(u16 control, u64 input8);
 bool hv_isolation_type_snp(void);
 bool hv_isolation_type_tdx(void);
 
+static inline struct hv_proximity_domain_info hv_numa_node_to_pxm_info(int node)
+{
+	struct hv_proximity_domain_info pxm_info = {};
+
+	if (node != NUMA_NO_NODE) {
+		pxm_info.domain_id = node_to_pxm(node);
+		pxm_info.flags.proximity_info_valid = 1;
+		pxm_info.flags.proximity_preferred = 1;
+	}
+
+	return pxm_info;
+}
+
 /* Helper functions that provide a consistent pattern for checking Hyper-V hypercall status. */
 static inline int hv_result(u64 status)
 {

include/linux/hyperv.h

@@ -832,6 +832,7 @@ struct vmbus_gpadl {
 	u32 gpadl_handle;
 	u32 size;
 	void *buffer;
+	bool decrypted;
 };
 
 struct vmbus_channel {

tools/hv/hv_kvp_daemon.c

@@ -76,6 +76,12 @@ enum {
 	DNS
 };
 
+enum {
+	IPV4 = 1,
+	IPV6,
+	IP_TYPE_MAX
+};
+
 static int in_hand_shake;
 
 static char *os_name = "";
@@ -102,6 +108,11 @@ static struct utsname uts_buf;
 
 #define MAX_FILE_NAME 100
 #define ENTRIES_PER_BLOCK 50
+/*
+ * Change this entry if the number of addresses increases in future
+ */
+#define MAX_IP_ENTRIES 64
+#define OUTSTR_BUF_SIZE ((INET6_ADDRSTRLEN + 1) * MAX_IP_ENTRIES)
 
 struct kvp_record {
 	char key[HV_KVP_EXCHANGE_MAX_KEY_SIZE];
@@ -1171,6 +1182,18 @@ static int process_ip_string(FILE *f, char *ip_string, int type)
 	return 0;
 }
 
+int ip_version_check(const char *input_addr)
+{
+	struct in6_addr addr;
+
+	if (inet_pton(AF_INET, input_addr, &addr))
+		return IPV4;
+	else if (inet_pton(AF_INET6, input_addr, &addr))
+		return IPV6;
+
+	return -EINVAL;
+}
+
 /*
  * Only IPv4 subnet strings needs to be converted to plen
  * For IPv6 the subnet is already privided in plen format
@@ -1197,14 +1220,75 @@ static int kvp_subnet_to_plen(char *subnet_addr_str)
 	return plen;
 }
 
+static int process_dns_gateway_nm(FILE *f, char *ip_string, int type,
+				  int ip_sec)
+{
+	char addr[INET6_ADDRSTRLEN], *output_str;
+	int ip_offset = 0, error = 0, ip_ver;
+	char *param_name;
+
+	if (type == DNS)
+		param_name = "dns";
+	else if (type == GATEWAY)
+		param_name = "gateway";
+	else
+		return -EINVAL;
+
+	output_str = (char *)calloc(OUTSTR_BUF_SIZE, sizeof(char));
+	if (!output_str)
+		return -ENOMEM;
+
+	while (1) {
+		memset(addr, 0, sizeof(addr));
+
+		if (!parse_ip_val_buffer(ip_string, &ip_offset, addr,
+					 (MAX_IP_ADDR_SIZE * 2)))
+			break;
+
+		ip_ver = ip_version_check(addr);
+		if (ip_ver < 0)
+			continue;
+
+		if ((ip_ver == IPV4 && ip_sec == IPV4) ||
+		    (ip_ver == IPV6 && ip_sec == IPV6)) {
+			/*
+			 * do a bound check to avoid out-of bound writes
+			 */
+			if ((OUTSTR_BUF_SIZE - strlen(output_str)) >
+			    (strlen(addr) + 1)) {
+				strncat(output_str, addr,
+					OUTSTR_BUF_SIZE -
+					strlen(output_str) - 1);
+				strncat(output_str, ",",
+					OUTSTR_BUF_SIZE -
+					strlen(output_str) - 1);
+			}
+		} else {
+			continue;
+		}
+	}
+
+	if (strlen(output_str)) {
+		/*
+		 * This is to get rid of that extra comma character
+		 * in the end of the string
+		 */
+		output_str[strlen(output_str) - 1] = '\0';
+		error = fprintf(f, "%s=%s\n", param_name, output_str);
+	}
+
+	free(output_str);
+	return error;
+}
+
 static int process_ip_string_nm(FILE *f, char *ip_string, char *subnet,
-				int is_ipv6)
+				int ip_sec)
 {
 	char addr[INET6_ADDRSTRLEN];
 	char subnet_addr[INET6_ADDRSTRLEN];
-	int error, i = 0;
+	int error = 0, i = 0;
 	int ip_offset = 0, subnet_offset = 0;
-	int plen;
+	int plen, ip_ver;
 
 	memset(addr, 0, sizeof(addr));
 	memset(subnet_addr, 0, sizeof(subnet_addr));
@@ -1216,10 +1300,16 @@ static int process_ip_string_nm(FILE *f, char *ip_string, char *subnet,
 						       subnet_addr,
 						       (MAX_IP_ADDR_SIZE *
 							2))) {
-		if (!is_ipv6)
+		ip_ver = ip_version_check(addr);
+		if (ip_ver < 0)
+			continue;
+
+		if (ip_ver == IPV4 && ip_sec == IPV4)
 			plen = kvp_subnet_to_plen((char *)subnet_addr);
-		else
+		else if (ip_ver == IPV6 && ip_sec == IPV6)
 			plen = atoi(subnet_addr);
+		else
+			continue;
 
 		if (plen < 0)
 			return plen;
@@ -1233,17 +1323,16 @@ static int process_ip_string_nm(FILE *f, char *ip_string, char *subnet,
 		memset(subnet_addr, 0, sizeof(subnet_addr));
 	}
 
-	return 0;
+	return error;
 }
 
 static int kvp_set_ip_info(char *if_name, struct hv_kvp_ipaddr_value *new_val)
 {
-	int error = 0;
+	int error = 0, ip_ver;
 	char if_filename[PATH_MAX];
 	char nm_filename[PATH_MAX];
 	FILE *ifcfg_file, *nmfile;
 	char cmd[PATH_MAX];
-	int is_ipv6 = 0;
 	char *mac_addr;
 	int str_len;
 
@@ -1421,52 +1510,94 @@ static int kvp_set_ip_info(char *if_name, struct hv_kvp_ipaddr_value *new_val)
 	if (error)
 		goto setval_error;
 
-	if (new_val->addr_family & ADDR_FAMILY_IPV6) {
-		error = fprintf(nmfile, "\n[ipv6]\n");
+	/*
+	 * Now we populate the keyfile format
+	 *
+	 * The keyfile format expects the IPv6 and IPv4 configuration in
+	 * different sections. Therefore we iterate through the list twice,
+	 * once to populate the IPv4 section and the next time for IPv6
+	 */
+	ip_ver = IPV4;
+	do {
+		if (ip_ver == IPV4) {
+			error = fprintf(nmfile, "\n[ipv4]\n");
 			if (error < 0)
 				goto setval_error;
-		is_ipv6 = 1;
 		} else {
-		error = fprintf(nmfile, "\n[ipv4]\n");
+			error = fprintf(nmfile, "\n[ipv6]\n");
 			if (error < 0)
 				goto setval_error;
 		}
 
 		/*
-	 * Now we populate the keyfile format
+		 * Write the configuration for ipaddress, netmask, gateway and
+		 * name services
 		 */
+		error = process_ip_string_nm(nmfile, (char *)new_val->ip_addr,
+					     (char *)new_val->sub_net,
+					     ip_ver);
+		if (error < 0)
+			goto setval_error;
 
+		/*
+		 * As dhcp_enabled is only valid for ipv4, we do not set dhcp
+		 * methods for ipv6 based on dhcp_enabled flag.
+		 *
+		 * For ipv4, set method to manual only when dhcp_enabled is
+		 * false and specific ipv4 addresses are configured. If neither
+		 * dhcp_enabled is true and no ipv4 addresses are configured,
+		 * set method to 'disabled'.
+		 *
+		 * For ipv6, set method to manual when we configure ipv6
+		 * addresses. Otherwise set method to 'auto' so that SLAAC from
+		 * RA may be used.
+		 */
+		if (ip_ver == IPV4) {
 			if (new_val->dhcp_enabled) {
-		error = kvp_write_file(nmfile, "method", "", "auto");
+				error = kvp_write_file(nmfile, "method", "",
+						       "auto");
+				if (error < 0)
+					goto setval_error;
+			} else if (error) {
+				error = kvp_write_file(nmfile, "method", "",
+						       "manual");
 				if (error < 0)
 					goto setval_error;
 			} else {
-		error = kvp_write_file(nmfile, "method", "", "manual");
+				error = kvp_write_file(nmfile, "method", "",
+						       "disabled");
 				if (error < 0)
 					goto setval_error;
 			}
-
-	/*
-	 * Write the configuration for ipaddress, netmask, gateway and
-	 * name services
-	 */
-	error = process_ip_string_nm(nmfile, (char *)new_val->ip_addr,
-				     (char *)new_val->sub_net, is_ipv6);
+		} else if (ip_ver == IPV6) {
+			if (error) {
+				error = kvp_write_file(nmfile, "method", "",
+						       "manual");
 				if (error < 0)
 					goto setval_error;
-
-	/* we do not want ipv4 addresses in ipv6 section and vice versa */
-	if (is_ipv6 != is_ipv4((char *)new_val->gate_way)) {
-		error = fprintf(nmfile, "gateway=%s\n", (char *)new_val->gate_way);
+			} else {
+				error = kvp_write_file(nmfile, "method", "",
+						       "auto");
 				if (error < 0)
 					goto setval_error;
 			}
+		}
 
-	if (is_ipv6 != is_ipv4((char *)new_val->dns_addr)) {
-		error = fprintf(nmfile, "dns=%s\n", (char *)new_val->dns_addr);
+		error = process_dns_gateway_nm(nmfile,
+					       (char *)new_val->gate_way,
+					       GATEWAY, ip_ver);
 		if (error < 0)
 			goto setval_error;
-	}
+
+		error = process_dns_gateway_nm(nmfile,
+					       (char *)new_val->dns_addr, DNS,
+					       ip_ver);
+		if (error < 0)
+			goto setval_error;
+
+		ip_ver++;
+	} while (ip_ver < IP_TYPE_MAX);
+
 	fclose(nmfile);
 	fclose(ifcfg_file);