diff options
Diffstat (limited to 'drivers/net/hyperv/netvsc.c')
| -rw-r--r-- | drivers/net/hyperv/netvsc.c | 536 |
1 files changed, 469 insertions, 67 deletions
diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index 41f5cf0bb997..60a4629fe6ba 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -20,8 +20,10 @@ #include <linux/vmalloc.h> #include <linux/rtnetlink.h> #include <linux/prefetch.h> +#include <linux/filter.h> #include <asm/sync_bitops.h> +#include <asm/mshyperv.h> #include "hyperv_net.h" #include "netvsc_trace.h" @@ -30,12 +32,17 @@ * Switch the data path from the synthetic interface to the VF * interface. */ -void netvsc_switch_datapath(struct net_device *ndev, bool vf) +int netvsc_switch_datapath(struct net_device *ndev, bool vf) { struct net_device_context *net_device_ctx = netdev_priv(ndev); struct hv_device *dev = net_device_ctx->device_ctx; struct netvsc_device *nv_dev = rtnl_dereference(net_device_ctx->nvdev); struct nvsp_message *init_pkt = &nv_dev->channel_init_pkt; + int ret, retry = 0; + + /* Block sending traffic to VF if it's about to be gone */ + if (!vf) + net_device_ctx->data_path_is_vf = vf; memset(init_pkt, 0, sizeof(struct nvsp_message)); init_pkt->hdr.msg_type = NVSP_MSG4_TYPE_SWITCH_DATA_PATH; @@ -46,12 +53,41 @@ void netvsc_switch_datapath(struct net_device *ndev, bool vf) init_pkt->msg.v4_msg.active_dp.active_datapath = NVSP_DATAPATH_SYNTHETIC; +again: trace_nvsp_send(ndev, init_pkt); - vmbus_sendpacket(dev->channel, init_pkt, + ret = vmbus_sendpacket(dev->channel, init_pkt, sizeof(struct nvsp_message), - (unsigned long)init_pkt, - VM_PKT_DATA_INBAND, 0); + (unsigned long)init_pkt, VM_PKT_DATA_INBAND, + VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED); + + /* If failed to switch to/from VF, let data_path_is_vf stay false, + * so we use synthetic path to send data. + */ + if (ret) { + if (ret != -EAGAIN) { + netdev_err(ndev, + "Unable to send sw datapath msg, err: %d\n", + ret); + return ret; + } + + if (retry++ < RETRY_MAX) { + usleep_range(RETRY_US_LO, RETRY_US_HI); + goto again; + } else { + netdev_err( + ndev, + "Retry failed to send sw datapath msg, err: %d\n", + ret); + return ret; + } + } + + wait_for_completion(&nv_dev->channel_init_wait); + net_device_ctx->data_path_is_vf = vf; + + return 0; } /* Worker to setup sub channels on initial setup @@ -118,12 +154,16 @@ static void free_netvsc_device(struct rcu_head *head) int i; kfree(nvdev->extension); - vfree(nvdev->recv_buf); - vfree(nvdev->send_buf); - kfree(nvdev->send_section_map); + + 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); for (i = 0; i < VRSS_CHANNEL_MAX; i++) { xdp_rxq_info_unreg(&nvdev->chan_table[i].xdp_rxq); + kfree(nvdev->chan_table[i].recv_buf); vfree(nvdev->chan_table[i].mrc.slots); } @@ -163,7 +203,7 @@ static void netvsc_revoke_recv_buf(struct hv_device *device, ret = vmbus_sendpacket(device->channel, revoke_packet, sizeof(struct nvsp_message), - (unsigned long)revoke_packet, + VMBUS_RQST_ID_NO_RESPONSE, VM_PKT_DATA_INBAND, 0); /* If the failure is because the channel is rescinded; * ignore the failure since we cannot send on a rescinded @@ -213,7 +253,7 @@ static void netvsc_revoke_send_buf(struct hv_device *device, ret = vmbus_sendpacket(device->channel, revoke_packet, sizeof(struct nvsp_message), - (unsigned long)revoke_packet, + VMBUS_RQST_ID_NO_RESPONSE, VM_PKT_DATA_INBAND, 0); /* If the failure is because the channel is rescinded; @@ -242,9 +282,9 @@ static void netvsc_teardown_recv_gpadl(struct hv_device *device, { int ret; - if (net_device->recv_buf_gpadl_handle) { + if (net_device->recv_buf_gpadl_handle.gpadl_handle) { ret = vmbus_teardown_gpadl(device->channel, - net_device->recv_buf_gpadl_handle); + &net_device->recv_buf_gpadl_handle); /* If we failed here, we might as well return and have a leak * rather than continue and a bugchk @@ -254,7 +294,6 @@ static void netvsc_teardown_recv_gpadl(struct hv_device *device, "unable to teardown receive buffer's gpadl\n"); return; } - net_device->recv_buf_gpadl_handle = 0; } } @@ -264,9 +303,9 @@ static void netvsc_teardown_send_gpadl(struct hv_device *device, { int ret; - if (net_device->send_buf_gpadl_handle) { + if (net_device->send_buf_gpadl_handle.gpadl_handle) { ret = vmbus_teardown_gpadl(device->channel, - net_device->send_buf_gpadl_handle); + &net_device->send_buf_gpadl_handle); /* If we failed here, we might as well return and have a leak * rather than continue and a bugchk @@ -276,7 +315,6 @@ static void netvsc_teardown_send_gpadl(struct hv_device *device, "unable to teardown send buffer's gpadl\n"); return; } - net_device->send_buf_gpadl_handle = 0; } } @@ -302,8 +340,7 @@ static int netvsc_init_buf(struct hv_device *device, struct net_device *ndev = hv_get_drvdata(device); struct nvsp_message *init_packet; unsigned int buf_size; - size_t map_words; - int ret = 0; + int i, ret = 0; /* Get receive buffer area. */ buf_size = device_info->recv_sections * device_info->recv_section_size; @@ -344,7 +381,7 @@ static int netvsc_init_buf(struct hv_device *device, memset(init_packet, 0, sizeof(struct nvsp_message)); init_packet->hdr.msg_type = NVSP_MSG1_TYPE_SEND_RECV_BUF; init_packet->msg.v1_msg.send_recv_buf. - gpadl_handle = net_device->recv_buf_gpadl_handle; + gpadl_handle = net_device->recv_buf_gpadl_handle.gpadl_handle; init_packet->msg.v1_msg. send_recv_buf.id = NETVSC_RECEIVE_BUFFER_ID; @@ -388,6 +425,25 @@ static int netvsc_init_buf(struct hv_device *device, net_device->recv_section_size = resp->sections[0].sub_alloc_size; net_device->recv_section_cnt = resp->sections[0].num_sub_allocs; + /* Ensure buffer will not overflow */ + if (net_device->recv_section_size < NETVSC_MTU_MIN || (u64)net_device->recv_section_size * + (u64)net_device->recv_section_cnt > (u64)buf_size) { + netdev_err(ndev, "invalid recv_section_size %u\n", + net_device->recv_section_size); + ret = -EINVAL; + goto cleanup; + } + + for (i = 0; i < VRSS_CHANNEL_MAX; i++) { + struct netvsc_channel *nvchan = &net_device->chan_table[i]; + + nvchan->recv_buf = kzalloc(net_device->recv_section_size, GFP_KERNEL); + if (nvchan->recv_buf == NULL) { + ret = -ENOMEM; + goto cleanup; + } + } + /* Setup receive completion ring. * Add 1 to the recv_section_cnt because at least one entry in a * ring buffer has to be empty. @@ -408,6 +464,7 @@ static int netvsc_init_buf(struct hv_device *device, ret = -ENOMEM; goto cleanup; } + net_device->send_buf_size = buf_size; /* Establish the gpadl handle for this buffer on this * channel. Note: This call uses the vmbus connection rather @@ -427,7 +484,7 @@ static int netvsc_init_buf(struct hv_device *device, memset(init_packet, 0, sizeof(struct nvsp_message)); init_packet->hdr.msg_type = NVSP_MSG1_TYPE_SEND_SEND_BUF; init_packet->msg.v1_msg.send_send_buf.gpadl_handle = - net_device->send_buf_gpadl_handle; + net_device->send_buf_gpadl_handle.gpadl_handle; init_packet->msg.v1_msg.send_send_buf.id = NETVSC_SEND_BUFFER_ID; trace_nvsp_send(ndev, init_packet); @@ -460,6 +517,12 @@ static int netvsc_init_buf(struct hv_device *device, /* Parse the response */ net_device->send_section_size = init_packet->msg. v1_msg.send_send_buf_complete.section_size; + if (net_device->send_section_size < NETVSC_MTU_MIN) { + netdev_err(ndev, "invalid send_section_size %u\n", + net_device->send_section_size); + ret = -EINVAL; + goto cleanup; + } /* Section count is simply the size divided by the section size. */ net_device->send_section_cnt = buf_size / net_device->send_section_size; @@ -468,10 +531,9 @@ static int netvsc_init_buf(struct hv_device *device, net_device->send_section_size, net_device->send_section_cnt); /* Setup state for managing the send buffer. */ - map_words = DIV_ROUND_UP(net_device->send_section_cnt, BITS_PER_LONG); - - net_device->send_section_map = kcalloc(map_words, sizeof(ulong), GFP_KERNEL); - if (net_device->send_section_map == NULL) { + net_device->send_section_map = bitmap_zalloc(net_device->send_section_cnt, + GFP_KERNEL); + if (!net_device->send_section_map) { ret = -ENOMEM; goto cleanup; } @@ -529,7 +591,10 @@ static int negotiate_nvsp_ver(struct hv_device *device, init_packet->msg.v2_msg.send_ndis_config.capability.ieee8021q = 1; if (nvsp_ver >= NVSP_PROTOCOL_VERSION_5) { - init_packet->msg.v2_msg.send_ndis_config.capability.sriov = 1; + if (hv_is_isolation_supported()) + netdev_info(ndev, "SR-IOV not advertised by guests on the host supporting isolation\n"); + else + init_packet->msg.v2_msg.send_ndis_config.capability.sriov = 1; /* Teaming bit is needed to receive link speed updates */ init_packet->msg.v2_msg.send_ndis_config.capability.teaming = 1; @@ -542,7 +607,7 @@ static int negotiate_nvsp_ver(struct hv_device *device, ret = vmbus_sendpacket(device->channel, init_packet, sizeof(struct nvsp_message), - (unsigned long)init_packet, + VMBUS_RQST_ID_NO_RESPONSE, VM_PKT_DATA_INBAND, 0); return ret; @@ -576,6 +641,13 @@ static int netvsc_connect_vsp(struct hv_device *device, goto cleanup; } + if (hv_is_isolation_supported() && net_device->nvsp_version < NVSP_PROTOCOL_VERSION_61) { + netdev_err(ndev, "Invalid NVSP version 0x%x (expected >= 0x%x) from the host supporting isolation\n", + net_device->nvsp_version, NVSP_PROTOCOL_VERSION_61); + ret = -EPROTO; + goto cleanup; + } + pr_debug("Negotiated NVSP version:%x\n", net_device->nvsp_version); /* Send the ndis version */ @@ -599,7 +671,7 @@ static int netvsc_connect_vsp(struct hv_device *device, /* Send the init request */ ret = vmbus_sendpacket(device->channel, init_packet, sizeof(struct nvsp_message), - (unsigned long)init_packet, + VMBUS_RQST_ID_NO_RESPONSE, VM_PKT_DATA_INBAND, 0); if (ret != 0) goto cleanup; @@ -639,7 +711,15 @@ void netvsc_device_remove(struct hv_device *device) /* Disable NAPI and disassociate its context from the device. */ for (i = 0; i < net_device->num_chn; i++) { /* See also vmbus_reset_channel_cb(). */ - napi_disable(&net_device->chan_table[i].napi); + /* only disable enabled NAPI channel */ + if (i < ndev->real_num_rx_queues) { + netif_queue_set_napi(ndev, i, NETDEV_QUEUE_TYPE_TX, + NULL); + netif_queue_set_napi(ndev, i, NETDEV_QUEUE_TYPE_RX, + NULL); + napi_disable(&net_device->chan_table[i].napi); + } + netif_napi_del(&net_device->chan_table[i].napi); } @@ -680,17 +760,26 @@ static void netvsc_send_tx_complete(struct net_device *ndev, const struct vmpacket_descriptor *desc, int budget) { - struct sk_buff *skb = (struct sk_buff *)(unsigned long)desc->trans_id; struct net_device_context *ndev_ctx = netdev_priv(ndev); + struct sk_buff *skb; u16 q_idx = 0; int queue_sends; + u64 cmd_rqst; + + cmd_rqst = channel->request_addr_callback(channel, desc->trans_id); + if (cmd_rqst == VMBUS_RQST_ERROR) { + netdev_err(ndev, "Invalid transaction ID %llx\n", desc->trans_id); + return; + } + + skb = (struct sk_buff *)(unsigned long)cmd_rqst; /* Notify the layer above us */ if (likely(skb)) { - const struct hv_netvsc_packet *packet + struct hv_netvsc_packet *packet = (struct hv_netvsc_packet *)skb->cb; u32 send_index = packet->send_buf_index; - struct netvsc_stats *tx_stats; + struct netvsc_stats_tx *tx_stats; if (send_index != NETVSC_INVALID_INDEX) netvsc_free_send_slot(net_device, send_index); @@ -703,6 +792,7 @@ static void netvsc_send_tx_complete(struct net_device *ndev, tx_stats->bytes += packet->total_bytes; u64_stats_update_end(&tx_stats->syncp); + netvsc_dma_unmap(ndev_ctx->device_ctx, packet); napi_consume_skb(skb, budget); } @@ -730,29 +820,111 @@ static void netvsc_send_completion(struct net_device *ndev, const struct vmpacket_descriptor *desc, int budget) { - const struct nvsp_message *nvsp_packet = hv_pkt_data(desc); + const struct nvsp_message *nvsp_packet; + u32 msglen = hv_pkt_datalen(desc); + struct nvsp_message *pkt_rqst; + u64 cmd_rqst; + u32 status; + + /* First check if this is a VMBUS completion without data payload */ + if (!msglen) { + cmd_rqst = incoming_channel->request_addr_callback(incoming_channel, + desc->trans_id); + if (cmd_rqst == VMBUS_RQST_ERROR) { + netdev_err(ndev, "Invalid transaction ID %llx\n", desc->trans_id); + return; + } + + pkt_rqst = (struct nvsp_message *)(uintptr_t)cmd_rqst; + switch (pkt_rqst->hdr.msg_type) { + case NVSP_MSG4_TYPE_SWITCH_DATA_PATH: + complete(&net_device->channel_init_wait); + break; + default: + netdev_err(ndev, "Unexpected VMBUS completion!!\n"); + } + return; + } + + /* Ensure packet is big enough to read header fields */ + if (msglen < sizeof(struct nvsp_message_header)) { + netdev_err(ndev, "nvsp_message length too small: %u\n", msglen); + return; + } + + nvsp_packet = hv_pkt_data(desc); switch (nvsp_packet->hdr.msg_type) { case NVSP_MSG_TYPE_INIT_COMPLETE: + if (msglen < sizeof(struct nvsp_message_header) + + sizeof(struct nvsp_message_init_complete)) { + netdev_err(ndev, "nvsp_msg length too small: %u\n", + msglen); + return; + } + break; + case NVSP_MSG1_TYPE_SEND_RECV_BUF_COMPLETE: + if (msglen < sizeof(struct nvsp_message_header) + + struct_size_t(struct nvsp_1_message_send_receive_buffer_complete, + sections, 1)) { + netdev_err(ndev, "nvsp_msg1 length too small: %u\n", + msglen); + return; + } + break; + case NVSP_MSG1_TYPE_SEND_SEND_BUF_COMPLETE: + if (msglen < sizeof(struct nvsp_message_header) + + sizeof(struct nvsp_1_message_send_send_buffer_complete)) { + netdev_err(ndev, "nvsp_msg1 length too small: %u\n", + msglen); + return; + } + break; + case NVSP_MSG5_TYPE_SUBCHANNEL: - /* Copy the response back */ - memcpy(&net_device->channel_init_pkt, nvsp_packet, - sizeof(struct nvsp_message)); - complete(&net_device->channel_init_wait); + if (msglen < sizeof(struct nvsp_message_header) + + sizeof(struct nvsp_5_subchannel_complete)) { + netdev_err(ndev, "nvsp_msg5 length too small: %u\n", + msglen); + return; + } break; case NVSP_MSG1_TYPE_SEND_RNDIS_PKT_COMPLETE: + if (msglen < sizeof(struct nvsp_message_header) + + sizeof(struct nvsp_1_message_send_rndis_packet_complete)) { + if (net_ratelimit()) + netdev_err(ndev, "nvsp_rndis_pkt_complete length too small: %u\n", + msglen); + return; + } + + /* If status indicates an error, output a message so we know + * there's a problem. But process the completion anyway so the + * resources are released. + */ + status = nvsp_packet->msg.v1_msg.send_rndis_pkt_complete.status; + if (status != NVSP_STAT_SUCCESS && net_ratelimit()) + netdev_err(ndev, "nvsp_rndis_pkt_complete error status: %x\n", + status); + netvsc_send_tx_complete(ndev, net_device, incoming_channel, desc, budget); - break; + return; default: netdev_err(ndev, "Unknown send completion type %d received!!\n", nvsp_packet->hdr.msg_type); + return; } + + /* Copy the response back */ + memcpy(&net_device->channel_init_pkt, nvsp_packet, + sizeof(struct nvsp_message)); + complete(&net_device->channel_init_wait); } static u32 netvsc_get_next_send_section(struct netvsc_device *net_device) @@ -781,8 +953,7 @@ static void netvsc_copy_to_send_buf(struct netvsc_device *net_device, + pend_size; int i; u32 padding = 0; - u32 page_count = packet->cp_partial ? packet->rmsg_pgcnt : - packet->page_buf_cnt; + u32 page_count = packet->cp_partial ? 1 : packet->page_buf_cnt; u32 remain; /* Add padding */ @@ -794,7 +965,7 @@ static void netvsc_copy_to_send_buf(struct netvsc_device *net_device, } for (i = 0; i < page_count; i++) { - char *src = phys_to_virt(pb[i].pfn << PAGE_SHIFT); + char *src = phys_to_virt(pb[i].pfn << HV_HYP_PAGE_SHIFT); u32 offset = pb[i].offset; u32 len = pb[i].len; @@ -806,6 +977,119 @@ static void netvsc_copy_to_send_buf(struct netvsc_device *net_device, memset(dest, 0, padding); } +void netvsc_dma_unmap(struct hv_device *hv_dev, + struct hv_netvsc_packet *packet) +{ + int i; + + if (!hv_is_isolation_supported()) + return; + + if (!packet->dma_range) + return; + + for (i = 0; i < packet->page_buf_cnt; i++) + dma_unmap_single(&hv_dev->device, packet->dma_range[i].dma, + packet->dma_range[i].mapping_size, + DMA_TO_DEVICE); + + kfree(packet->dma_range); +} + +/* netvsc_dma_map - Map swiotlb bounce buffer with data page of + * packet sent by vmbus_sendpacket_pagebuffer() in the Isolation + * VM. + * + * In isolation VM, netvsc send buffer has been marked visible to + * host and so the data copied to send buffer doesn't need to use + * bounce buffer. The data pages handled by vmbus_sendpacket_pagebuffer() + * may not be copied to send buffer and so these pages need to be + * mapped with swiotlb bounce buffer. netvsc_dma_map() is to do + * that. The pfns in the struct hv_page_buffer need to be converted + * to bounce buffer's pfn. The loop here is necessary because the + * entries in the page buffer array are not necessarily full + * pages of data. Each entry in the array has a separate offset and + * len that may be non-zero, even for entries in the middle of the + * array. And the entries are not physically contiguous. So each + * entry must be individually mapped rather than as a contiguous unit. + * So not use dma_map_sg() here. + */ +static int netvsc_dma_map(struct hv_device *hv_dev, + struct hv_netvsc_packet *packet, + struct hv_page_buffer *pb) +{ + u32 page_count = packet->page_buf_cnt; + dma_addr_t dma; + int i; + + if (!hv_is_isolation_supported()) + return 0; + + packet->dma_range = kcalloc(page_count, + sizeof(*packet->dma_range), + GFP_ATOMIC); + if (!packet->dma_range) + return -ENOMEM; + + for (i = 0; i < page_count; i++) { + char *src = phys_to_virt((pb[i].pfn << HV_HYP_PAGE_SHIFT) + + pb[i].offset); + u32 len = pb[i].len; + + dma = dma_map_single(&hv_dev->device, src, len, + DMA_TO_DEVICE); + if (dma_mapping_error(&hv_dev->device, dma)) { + kfree(packet->dma_range); + return -ENOMEM; + } + + /* pb[].offset and pb[].len are not changed during dma mapping + * and so not reassign. + */ + packet->dma_range[i].dma = dma; + packet->dma_range[i].mapping_size = len; + pb[i].pfn = dma >> HV_HYP_PAGE_SHIFT; + } + + return 0; +} + +/* Build an "array" of mpb entries describing the data to be transferred + * over VMBus. After the desc header fields, each "array" entry is variable + * size, and each entry starts after the end of the previous entry. The + * "offset" and "len" fields for each entry imply the size of the entry. + * + * The pfns are in HV_HYP_PAGE_SIZE, because all communication with Hyper-V + * uses that granularity, even if the system page size of the guest is larger. + * Each entry in the input "pb" array must describe a contiguous range of + * guest physical memory so that the pfns are sequential if the range crosses + * a page boundary. The offset field must be < HV_HYP_PAGE_SIZE. + */ +static inline void netvsc_build_mpb_array(struct hv_page_buffer *pb, + u32 page_buffer_count, + struct vmbus_packet_mpb_array *desc, + u32 *desc_size) +{ + struct hv_mpb_array *mpb_entry = &desc->range; + int i, j; + + for (i = 0; i < page_buffer_count; i++) { + u32 offset = pb[i].offset; + u32 len = pb[i].len; + + mpb_entry->offset = offset; + mpb_entry->len = len; + + for (j = 0; j < HVPFN_UP(offset + len); j++) + mpb_entry->pfn_array[j] = pb[i].pfn + j; + + mpb_entry = (struct hv_mpb_array *)&mpb_entry->pfn_array[j]; + } + + desc->rangecount = page_buffer_count; + *desc_size = (char *)mpb_entry - (char *)desc; +} + static inline int netvsc_send_pkt( struct hv_device *device, struct hv_netvsc_packet *packet, @@ -826,6 +1110,7 @@ static inline int netvsc_send_pkt( int ret; u32 ring_avail = hv_get_avail_to_write_percent(&out_channel->outbound); + memset(&nvmsg, 0, sizeof(struct nvsp_message)); nvmsg.hdr.msg_type = NVSP_MSG1_TYPE_SEND_RNDIS_PKT; if (skb) rpkt->channel_type = 0; /* 0 is RMC_DATA */ @@ -845,14 +1130,28 @@ static inline int netvsc_send_pkt( trace_nvsp_send_pkt(ndev, out_channel, rpkt); + packet->dma_range = NULL; if (packet->page_buf_cnt) { + struct vmbus_channel_packet_page_buffer desc; + u32 desc_size; + if (packet->cp_partial) - pb += packet->rmsg_pgcnt; + pb++; + + ret = netvsc_dma_map(ndev_ctx->device_ctx, packet, pb); + if (ret) { + ret = -EAGAIN; + goto exit; + } - ret = vmbus_sendpacket_pagebuffer(out_channel, - pb, packet->page_buf_cnt, - &nvmsg, sizeof(nvmsg), - req_id); + netvsc_build_mpb_array(pb, packet->page_buf_cnt, + (struct vmbus_packet_mpb_array *)&desc, + &desc_size); + ret = vmbus_sendpacket_mpb_desc(out_channel, + (struct vmbus_packet_mpb_array *)&desc, + desc_size, &nvmsg, sizeof(nvmsg), req_id); + if (ret) + netvsc_dma_unmap(ndev_ctx->device_ctx, packet); } else { ret = vmbus_sendpacket(out_channel, &nvmsg, sizeof(nvmsg), @@ -860,6 +1159,7 @@ static inline int netvsc_send_pkt( VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED); } +exit: if (ret == 0) { atomic_inc_return(&nvchan->queue_sends); @@ -902,6 +1202,26 @@ static inline void move_pkt_msd(struct hv_netvsc_packet **msd_send, } /* RCU already held by caller */ +/* Batching/bouncing logic is designed to attempt to optimize + * performance. + * + * For small, non-LSO packets we copy the packet to a send buffer + * which is pre-registered with the Hyper-V side. This enables the + * hypervisor to avoid remapping the aperture to access the packet + * descriptor and data. + * + * If we already started using a buffer and the netdev is transmitting + * a burst of packets, keep on copying into the buffer until it is + * full or we are done collecting a burst. If there is an existing + * buffer with space for the RNDIS descriptor but not the packet, copy + * the RNDIS descriptor to the buffer, keeping the packet in place. + * + * If we do batching and send more than one packet using a single + * NetVSC message, free the SKBs of the packets copied, except for the + * last packet. This is done to streamline the handling of the case + * where the last packet only had the RNDIS descriptor copied to the + * send buffer, with the data pointers included in the NetVSC message. + */ int netvsc_send(struct net_device *ndev, struct hv_netvsc_packet *packet, struct rndis_message *rndis_msg, @@ -978,7 +1298,7 @@ int netvsc_send(struct net_device *ndev, packet->send_buf_index = section_index; if (packet->cp_partial) { - packet->page_buf_cnt -= packet->rmsg_pgcnt; + packet->page_buf_cnt--; packet->total_data_buflen = msd_len + packet->rmsg_size; } else { packet->page_buf_cnt = 0; @@ -1117,19 +1437,28 @@ static void enq_receive_complete(struct net_device *ndev, static int netvsc_receive(struct net_device *ndev, struct netvsc_device *net_device, struct netvsc_channel *nvchan, - const struct vmpacket_descriptor *desc, - const struct nvsp_message *nvsp) + const struct vmpacket_descriptor *desc) { struct net_device_context *net_device_ctx = netdev_priv(ndev); struct vmbus_channel *channel = nvchan->channel; const struct vmtransfer_page_packet_header *vmxferpage_packet = container_of(desc, const struct vmtransfer_page_packet_header, d); + const struct nvsp_message *nvsp = hv_pkt_data(desc); + u32 msglen = hv_pkt_datalen(desc); u16 q_idx = channel->offermsg.offer.sub_channel_index; char *recv_buf = net_device->recv_buf; u32 status = NVSP_STAT_SUCCESS; int i; int count = 0; + /* Ensure packet is big enough to read header fields */ + if (msglen < sizeof(struct nvsp_message_header)) { + netif_err(net_device_ctx, rx_err, ndev, + "invalid nvsp header, length too small: %u\n", + msglen); + return 0; + } + /* Make sure this is a valid nvsp packet */ if (unlikely(nvsp->hdr.msg_type != NVSP_MSG1_TYPE_SEND_RNDIS_PKT)) { netif_err(net_device_ctx, rx_err, ndev, @@ -1138,6 +1467,14 @@ static int netvsc_receive(struct net_device *ndev, return 0; } + /* Validate xfer page pkt header */ + if ((desc->offset8 << 3) < sizeof(struct vmtransfer_page_packet_header)) { + netif_err(net_device_ctx, rx_err, ndev, + "Invalid xfer page pkt, offset too small: %u\n", + desc->offset8 << 3); + return 0; + } + if (unlikely(vmxferpage_packet->xfer_pageset_id != NETVSC_RECEIVE_BUFFER_ID)) { netif_err(net_device_ctx, rx_err, ndev, "Invalid xfer page set id - expecting %x got %x\n", @@ -1148,6 +1485,14 @@ static int netvsc_receive(struct net_device *ndev, count = vmxferpage_packet->range_cnt; + /* Check count for a valid value */ + if (NETVSC_XFER_HEADER_SIZE(count) > desc->offset8 << 3) { + netif_err(net_device_ctx, rx_err, ndev, + "Range count is not valid: %d\n", + count); + return 0; + } + /* Each range represents 1 RNDIS pkt that contains 1 ethernet frame */ for (i = 0; i < count; i++) { u32 offset = vmxferpage_packet->ranges[i].byte_offset; @@ -1155,7 +1500,8 @@ static int netvsc_receive(struct net_device *ndev, void *data; int ret; - if (unlikely(offset + buflen > net_device->recv_buf_size)) { + if (unlikely(offset > net_device->recv_buf_size || + buflen > net_device->recv_buf_size - offset)) { nvchan->rsc.cnt = 0; status = NVSP_STAT_FAIL; netif_err(net_device_ctx, rx_err, ndev, @@ -1165,6 +1511,19 @@ static int netvsc_receive(struct net_device *ndev, continue; } + /* We're going to copy (sections of) the packet into nvchan->recv_buf; + * make sure that nvchan->recv_buf is large enough to hold the packet. + */ + if (unlikely(buflen > net_device->recv_section_size)) { + nvchan->rsc.cnt = 0; + status = NVSP_STAT_FAIL; + netif_err(net_device_ctx, rx_err, ndev, + "Packet too big: buflen=%u recv_section_size=%u\n", + buflen, net_device->recv_section_size); + + continue; + } + data = recv_buf + offset; nvchan->rsc.is_last = (i == count - 1); @@ -1175,8 +1534,11 @@ static int netvsc_receive(struct net_device *ndev, ret = rndis_filter_receive(ndev, net_device, nvchan, data, buflen); - if (unlikely(ret != NVSP_STAT_SUCCESS)) + if (unlikely(ret != NVSP_STAT_SUCCESS)) { + /* Drop incomplete packet */ + nvchan->rsc.cnt = 0; status = NVSP_STAT_FAIL; + } } enq_receive_complete(ndev, net_device, q_idx, @@ -1194,6 +1556,13 @@ static void netvsc_send_table(struct net_device *ndev, u32 count, offset, *tab; int i; + /* Ensure packet is big enough to read send_table fields */ + if (msglen < sizeof(struct nvsp_message_header) + + sizeof(struct nvsp_5_send_indirect_table)) { + netdev_err(ndev, "nvsp_v5_msg length too small: %u\n", msglen); + return; + } + count = nvmsg->msg.v5_msg.send_table.count; offset = nvmsg->msg.v5_msg.send_table.offset; @@ -1212,7 +1581,7 @@ static void netvsc_send_table(struct net_device *ndev, sizeof(union nvsp_6_message_uber); /* Boundary check for all versions */ - if (offset > msglen - count * sizeof(u32)) { + if (msglen < count * sizeof(u32) || offset > msglen - count * sizeof(u32)) { netdev_err(ndev, "Received send-table offset too big:%u\n", offset); return; @@ -1225,12 +1594,24 @@ static void netvsc_send_table(struct net_device *ndev, } static void netvsc_send_vf(struct net_device *ndev, - const struct nvsp_message *nvmsg) + const struct nvsp_message *nvmsg, + u32 msglen) { struct net_device_context *net_device_ctx = netdev_priv(ndev); + /* Ensure packet is big enough to read its fields */ + if (msglen < sizeof(struct nvsp_message_header) + + sizeof(struct nvsp_4_send_vf_association)) { + netdev_err(ndev, "nvsp_v4_msg length too small: %u\n", msglen); + return; + } + net_device_ctx->vf_alloc = nvmsg->msg.v4_msg.vf_assoc.allocated; net_device_ctx->vf_serial = nvmsg->msg.v4_msg.vf_assoc.serial; + + if (net_device_ctx->vf_alloc) + complete(&net_device_ctx->vf_add); + netdev_info(ndev, "VF slot %u %s\n", net_device_ctx->vf_serial, net_device_ctx->vf_alloc ? "added" : "removed"); @@ -1238,16 +1619,27 @@ static void netvsc_send_vf(struct net_device *ndev, static void netvsc_receive_inband(struct net_device *ndev, struct netvsc_device *nvscdev, - const struct nvsp_message *nvmsg, - u32 msglen) + const struct vmpacket_descriptor *desc) { + const struct nvsp_message *nvmsg = hv_pkt_data(desc); + u32 msglen = hv_pkt_datalen(desc); + + /* Ensure packet is big enough to read header fields */ + if (msglen < sizeof(struct nvsp_message_header)) { + netdev_err(ndev, "inband nvsp_message length too small: %u\n", msglen); + return; + } + switch (nvmsg->hdr.msg_type) { case NVSP_MSG5_TYPE_SEND_INDIRECTION_TABLE: netvsc_send_table(ndev, nvscdev, nvmsg, msglen); break; case NVSP_MSG4_TYPE_SEND_VF_ASSOCIATION: - netvsc_send_vf(ndev, nvmsg); + if (hv_is_isolation_supported()) + netdev_err(ndev, "Ignore VF_ASSOCIATION msg from the host supporting isolation\n"); + else + netvsc_send_vf(ndev, nvmsg, msglen); break; } } @@ -1261,23 +1653,19 @@ static int netvsc_process_raw_pkt(struct hv_device *device, { struct vmbus_channel *channel = nvchan->channel; const struct nvsp_message *nvmsg = hv_pkt_data(desc); - u32 msglen = hv_pkt_datalen(desc); trace_nvsp_recv(ndev, channel, nvmsg); switch (desc->type) { case VM_PKT_COMP: - netvsc_send_completion(ndev, net_device, channel, - desc, budget); + netvsc_send_completion(ndev, net_device, channel, desc, budget); break; case VM_PKT_DATA_USING_XFER_PAGES: - return netvsc_receive(ndev, net_device, nvchan, - desc, nvmsg); - break; + return netvsc_receive(ndev, net_device, nvchan, desc); case VM_PKT_DATA_INBAND: - netvsc_receive_inband(ndev, net_device, nvmsg, msglen); + netvsc_receive_inband(ndev, net_device, desc); break; default: @@ -1315,12 +1703,17 @@ int netvsc_poll(struct napi_struct *napi, int budget) if (!nvchan->desc) nvchan->desc = hv_pkt_iter_first(channel); + nvchan->xdp_flush = false; + while (nvchan->desc && work_done < budget) { work_done += netvsc_process_raw_pkt(device, nvchan, net_device, ndev, nvchan->desc, budget); nvchan->desc = hv_pkt_iter_next(channel, nvchan->desc); } + if (nvchan->xdp_flush) + xdp_do_flush(); + /* Send any pending receive completions */ ret = send_recv_completions(ndev, net_device, nvchan); @@ -1401,7 +1794,7 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device, u64_stats_init(&nvchan->tx_stats.syncp); u64_stats_init(&nvchan->rx_stats.syncp); - ret = xdp_rxq_info_reg(&nvchan->xdp_rxq, ndev, i); + ret = xdp_rxq_info_reg(&nvchan->xdp_rxq, ndev, i, 0); if (ret) { netdev_err(ndev, "xdp_rxq_info_reg fail: %d\n", ret); @@ -1418,10 +1811,19 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device, } /* Enable NAPI handler before init callbacks */ - netif_napi_add(ndev, &net_device->chan_table[0].napi, - netvsc_poll, NAPI_POLL_WEIGHT); + netif_napi_add(ndev, &net_device->chan_table[0].napi, netvsc_poll); + napi_enable(&net_device->chan_table[0].napi); + netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_RX, + &net_device->chan_table[0].napi); + netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_TX, + &net_device->chan_table[0].napi); /* Open the channel */ + device->channel->next_request_id_callback = vmbus_next_request_id; + device->channel->request_addr_callback = vmbus_request_addr; + device->channel->rqstor_size = netvsc_rqstor_size(netvsc_ring_bytes); + device->channel->max_pkt_size = NETVSC_MAX_PKT_SIZE; + ret = vmbus_open(device->channel, netvsc_ring_bytes, netvsc_ring_bytes, NULL, 0, netvsc_channel_cb, net_device->chan_table); @@ -1434,8 +1836,6 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device, /* Channel is opened */ netdev_dbg(ndev, "hv_netvsc channel opened successfully\n"); - napi_enable(&net_device->chan_table[0].napi); - /* Connect with the NetVsp */ ret = netvsc_connect_vsp(device, net_device, device_info); if (ret != 0) { @@ -1453,12 +1853,14 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device, close: RCU_INIT_POINTER(net_device_ctx->nvdev, NULL); - napi_disable(&net_device->chan_table[0].napi); /* Now, we can close the channel safely */ vmbus_close(device->channel); cleanup: + netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_TX, NULL); + netif_queue_set_napi(ndev, 0, NETDEV_QUEUE_TYPE_RX, NULL); + napi_disable(&net_device->chan_table[0].napi); netif_napi_del(&net_device->chan_table[0].napi); cleanup2: |