Dear Maxime, Answers inline.
Internal Use - Confidential > -----Original Message----- > From: Maxime Coquelin <[email protected]> > Sent: Monday, January 26, 2026 11:58 AM > To: Bathija, Pravin <[email protected]> > Cc: [email protected]; [email protected] > Subject: Re: [PATCH v6 4/5] vhost_user: Function defs for add/rem mem > regions > > > [EXTERNAL EMAIL] > > On Thu, Jan 22, 2026 at 8:49 AM <[email protected]> wrote: > > > > From: Pravin M Bathija <[email protected]> > > > > These changes cover the function definition for add/remove memory > > region calls which are invoked on receiving vhost user message from > > vhost user front-end (e.g. Qemu). In our case, in addition to testing > > with qemu front-end, the testing has also been performed with libblkio > > front-end and spdk/dpdk back-end. We did I/O using libblkio based > > device driver, to spdk based drives. There are also changes for > > set_mem_table and new definition for get memory slots. Our changes > > optimize the set memory table call to use common support functions. > > Message get memory slots is how the vhost-user front-end queries the > > vhost-user back-end about the number of memory slots available to be > > registered by the back-end. In addition support function to invalidate > > vring is also defined which is used in add/remove memory region functions. > > > > Signed-off-by: Pravin M Bathija <[email protected]> > > --- > > lib/vhost/vhost_user.c | 234 > > +++++++++++++++++++++++++++++++++++------ > > 1 file changed, 202 insertions(+), 32 deletions(-) > > > > diff --git a/lib/vhost/vhost_user.c b/lib/vhost/vhost_user.c index > > e2831b7625..1a2dcfbbca 100644 > > --- a/lib/vhost/vhost_user.c > > +++ b/lib/vhost/vhost_user.c > > @@ -71,6 +71,9 @@ > VHOST_MESSAGE_HANDLER(VHOST_USER_SET_FEATURES, > > vhost_user_set_features, false, t > > VHOST_MESSAGE_HANDLER(VHOST_USER_SET_OWNER, > vhost_user_set_owner, > > false, true) \ VHOST_MESSAGE_HANDLER(VHOST_USER_RESET_OWNER, > > vhost_user_reset_owner, false, false) \ > > VHOST_MESSAGE_HANDLER(VHOST_USER_SET_MEM_TABLE, > > vhost_user_set_mem_table, true, true) \ > > +VHOST_MESSAGE_HANDLER(VHOST_USER_GET_MAX_MEM_SLOTS, > > +vhost_user_get_max_mem_slots, false, false) \ > > +VHOST_MESSAGE_HANDLER(VHOST_USER_ADD_MEM_REG, > vhost_user_add_mem_reg, > > +true, true) \ VHOST_MESSAGE_HANDLER(VHOST_USER_REM_MEM_REG, > > +vhost_user_rem_mem_reg, false, true) \ > > VHOST_MESSAGE_HANDLER(VHOST_USER_SET_LOG_BASE, > > vhost_user_set_log_base, true, true) \ > > VHOST_MESSAGE_HANDLER(VHOST_USER_SET_LOG_FD, > vhost_user_set_log_fd, > > true, true) \ VHOST_MESSAGE_HANDLER(VHOST_USER_SET_VRING_NUM, > vhost_user_set_vring_num, false, true) \ @@ -1520,7 +1523,6 @@ > vhost_user_set_mem_table(struct virtio_net **pdev, > > struct virtio_net *dev = *pdev; > > struct VhostUserMemory *memory = &ctx->msg.payload.memory; > > struct rte_vhost_mem_region *reg; > > - int numa_node = SOCKET_ID_ANY; > > uint64_t mmap_offset; > > uint32_t i; > > bool async_notify = false; > > @@ -1565,39 +1567,13 @@ vhost_user_set_mem_table(struct virtio_net > **pdev, > > if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)) > > vhost_user_iotlb_flush_all(dev); > > > > - free_mem_region(dev); > > + free_all_mem_regions(dev); > > rte_free(dev->mem); > > dev->mem = NULL; > > } > > > > - /* > > - * If VQ 0 has already been allocated, try to allocate on the same > > - * NUMA node. It can be reallocated later in numa_realloc(). > > - */ > > - if (dev->nr_vring > 0) > > - numa_node = dev->virtqueue[0]->numa_node; > > - > > - dev->nr_guest_pages = 0; > > - if (dev->guest_pages == NULL) { > > - dev->max_guest_pages = 8; > > - dev->guest_pages = rte_zmalloc_socket(NULL, > > - dev->max_guest_pages * > > - sizeof(struct guest_page), > > - RTE_CACHE_LINE_SIZE, > > - numa_node); > > - if (dev->guest_pages == NULL) { > > - VHOST_CONFIG_LOG(dev->ifname, ERR, > > - "failed to allocate memory for > > dev->guest_pages"); > > - goto close_msg_fds; > > - } > > - } > > - > > - dev->mem = rte_zmalloc_socket("vhost-mem-table", sizeof(struct > rte_vhost_memory) + > > - sizeof(struct rte_vhost_mem_region) * memory->nregions, 0, > numa_node); > > - if (dev->mem == NULL) { > > - VHOST_CONFIG_LOG(dev->ifname, ERR, "failed to allocate > > memory > for dev->mem"); > > - goto free_guest_pages; > > - } > > + if (vhost_user_initialize_memory(pdev) < 0) > > + goto close_msg_fds; > > > > for (i = 0; i < memory->nregions; i++) { > > reg = &dev->mem->regions[i]; @@ -1661,11 +1637,167 @@ > > vhost_user_set_mem_table(struct virtio_net **pdev, > > return RTE_VHOST_MSG_RESULT_OK; > > > > free_mem_table: > > - free_mem_region(dev); > > + free_all_mem_regions(dev); > > rte_free(dev->mem); > > dev->mem = NULL; > > + rte_free(dev->guest_pages); > > + dev->guest_pages = NULL; > > +close_msg_fds: > > + close_msg_fds(ctx); > > + return RTE_VHOST_MSG_RESULT_ERR; } > > + > > + > > +static int > > +vhost_user_get_max_mem_slots(struct virtio_net **pdev __rte_unused, > > + struct vhu_msg_context *ctx, > > + int main_fd __rte_unused) { > > + uint32_t max_mem_slots = VHOST_MEMORY_MAX_NREGIONS; > > + > > + ctx->msg.payload.u64 = (uint64_t)max_mem_slots; > > + ctx->msg.size = sizeof(ctx->msg.payload.u64); > > + ctx->fd_num = 0; > > + > > + return RTE_VHOST_MSG_RESULT_REPLY; } > > + > > +static void > > +dev_invalidate_vrings(struct virtio_net *dev) { > > + uint32_t i; > > + > > + for (i = 0; i < dev->nr_vring; i++) { > > + struct vhost_virtqueue *vq = dev->virtqueue[i]; > > + > > + if (!vq) > > + continue; > > > > -free_guest_pages: > > + if (vq->desc || vq->avail || vq->used) { > > + /* vhost_user_lock_all_queue_pairs locked all qps */ > > + VHOST_USER_ASSERT_LOCK(dev, vq, > > + VHOST_USER_ADD_MEM_REG); > > + > > + /* > > + * If the memory table got updated, the ring > > addresses > > + * need to be translated again as virtual addresses > > have > > + * changed. > > + */ > > + vring_invalidate(dev, vq); > > + > > + translate_ring_addresses(&dev, &vq); > > + } > > + } > > +} > > + > > +static int > > +vhost_user_add_mem_reg(struct virtio_net **pdev, > > + struct vhu_msg_context *ctx, > > + int main_fd __rte_unused) > > The parameter is marked as unused but is actually used. I have made the corrections in the amended patch-set I will submit following this e-mail. > > > +{ > > + uint32_t i; > > + struct virtio_net *dev = *pdev; > > + struct VhostUserMemoryRegion *region = > > +&ctx->msg.payload.memory_single.region; > > + > > + /* convert first region add to normal memory table set */ > > + if (dev->mem == NULL) { > > + if (vhost_user_initialize_memory(pdev) < 0) > > + goto close_msg_fds; > > + } > > + > > + /* make sure new region will fit */ > > + if (dev->mem->nregions >= VHOST_MEMORY_MAX_NREGIONS) { > > + VHOST_CONFIG_LOG(dev->ifname, ERR, "too many memory > regions already (%u)", > > + > > dev->mem->nregions); > > + goto close_msg_fds; > > + } > > + > > + /* make sure supplied memory fd present */ > > + if (ctx->fd_num != 1) { > > + VHOST_CONFIG_LOG(dev->ifname, ERR, "fd count makes no sense > (%u)", ctx->fd_num); > > + goto close_msg_fds; > > + } > > + > > + /* Make sure no overlap in guest virtual address space */ > > + if (dev->mem != NULL && dev->mem->nregions > 0) { > > + for (i = 0; i < VHOST_MEMORY_MAX_NREGIONS; i++) { > > + struct rte_vhost_mem_region *current_region = > > + &dev->mem->regions[i]; > > + > > + if (current_region->mmap_size == 0) > > + continue; > > + > > + uint64_t current_region_guest_start = > > current_region- > >guest_user_addr; > > + uint64_t current_region_guest_end = > current_region_guest_start > > + + > > current_region->mmap_size - 1; > > + uint64_t proposed_region_guest_start = region- > >userspace_addr; > > + uint64_t proposed_region_guest_end = > proposed_region_guest_start > > + + > > region->memory_size - 1; > > + bool overlap = false; > > + > > + overlap = !((proposed_region_guest_end < > current_region_guest_start) || > > + (proposed_region_guest_start > > > + current_region_guest_end)); > > + > > + if (overlap) { > > + VHOST_CONFIG_LOG(dev->ifname, ERR, > > + "requested memory region overlaps > > with another > region"); > > + VHOST_CONFIG_LOG(dev->ifname, ERR, > > + "\tRequested region address:0x%" > > PRIx64, > > + region->userspace_addr); > > + VHOST_CONFIG_LOG(dev->ifname, ERR, > > + "\tRequested region size:0x%" > > PRIx64, > > + region->memory_size); > > + VHOST_CONFIG_LOG(dev->ifname, ERR, > > + "\tOverlapping region address:0x%" > > PRIx64, > > + current_region->guest_user_addr); > > + VHOST_CONFIG_LOG(dev->ifname, ERR, > > + "\tOverlapping region size:0x%" > > PRIx64, > > + current_region->mmap_size); > > + goto close_msg_fds; > > + } > > + > > + } > > + } > > + > > + /* find a new region and set it like memory table set does */ > > + struct rte_vhost_mem_region *reg = NULL; > > + > > + for (i = 0; i < VHOST_MEMORY_MAX_NREGIONS; i++) { > > + if (dev->mem->regions[i].guest_user_addr == 0) { > > + reg = &dev->mem->regions[i]; > > + break; > > + } > > + } > > + if (reg == NULL) { > > + VHOST_CONFIG_LOG(dev->ifname, ERR, "no free memory region"); > > + goto close_msg_fds; > > + } > > + > > + reg->guest_phys_addr = region->guest_phys_addr; > > + reg->guest_user_addr = region->userspace_addr; > > + reg->size = region->memory_size; > > + reg->fd = ctx->fds[0]; > > + > > + if (vhost_user_mmap_region(dev, reg, region->mmap_offset) < 0) { > > + VHOST_CONFIG_LOG(dev->ifname, ERR, "failed to mmap region"); > > + goto close_msg_fds; > > + } > > + > > + dev->mem->nregions++; > > + > > + if (dev->async_copy && rte_vfio_is_enabled("vfio")) > > + async_dma_map_region(dev, reg, true); > > + > > + if (vhost_user_postcopy_register(dev, main_fd, ctx) < 0) > > Post-copy live-migration is broken. > The assumes the vhost-user message is of type SET_MEM_TABLE, and so > assumes ctx contains a VhostUserMemory payload. > But here ctx contains a VhostUserMemoryRegion payload. > > Given the complexity of the patch, it is worrying that such issues are > overlooked. > What other features could be broken? Although I tested this before I have again tested Post-copy live migration successfully with qemu using the following steps. - Ran dpdk-testpmd while enabling poctcopy migration as option. - Ran two instances of qemu, source and destination. Source was running alpine linux while destination was started with option "-incoming defer" - I did an add memory region at the source VM which invoked vhost_postcopy_register function. After this I did the same add memory region for destination VM. - The "post_copy_register" debug messages which I added in this upcoming patch-set were correctly seen on the testpmd console. - Following this I turned on migrate capabilities on source and destination VMs at qemu prompt. For destination VM I also set the incoming migration to TCP port 4444. - Set the "migrate" destination to host 127.0.0.1 TCP port 4444 and invoked command "migrate_start_postcopy" After the above steps, the Alpine Linux VM memory pages were correctly migrated to destination VM. I would be happy to share console logs of the above test and step by step commands. Source code clarification: In the function vhost_user_add_mem_reg we provide the argument of a single region of type "VhostUserMemoryRegion" in the message payload. This single memory region is then converted to type rte_vhost_mem_region (variable reg line#1761) and then added to the list of memory regions using the call vhost_user_mmap_region(line#1779). By the time, vhost_user_postcopy_register(line#1794) is called, the newl added region is part of the "dev" structure. > > > + goto free_mem_table; > > + > > + dev_invalidate_vrings(dev); > > + dump_guest_pages(dev); > > + > > + return RTE_VHOST_MSG_RESULT_OK; > > + > > +free_mem_table: > > + free_all_mem_regions(dev); > > + rte_free(dev->mem); > > + dev->mem = NULL; > > rte_free(dev->guest_pages); > > dev->guest_pages = NULL; > > close_msg_fds: > > @@ -1673,6 +1805,44 @@ vhost_user_set_mem_table(struct virtio_net > **pdev, > > return RTE_VHOST_MSG_RESULT_ERR; } > > > > +static int > > +vhost_user_rem_mem_reg(struct virtio_net **pdev __rte_unused, > > + struct vhu_msg_context *ctx __rte_unused, > > pdev and ctx are actually used. I have made the corrections in the amended patch-set I will submit following this e-mail. > > > > + int main_fd __rte_unused) { > > + uint32_t i; > > + struct virtio_net *dev = *pdev; > > + struct VhostUserMemoryRegion *region = > > +&ctx->msg.payload.memory_single.region; > > + > > + if (dev->mem != NULL && dev->mem->nregions > 0) { > > + for (i = 0; i < VHOST_MEMORY_MAX_NREGIONS; i++) { > > + struct rte_vhost_mem_region *current_region = > > + &dev->mem->regions[i]; > > + > > + if (current_region->guest_user_addr == 0) > > + continue; > > + > > + /* > > + * According to the vhost-user specification: > > + * The memory region to be removed is identified by > > its guest > address, > > + * user address and size. The mmap offset is > > ignored. > > + */ > > + if (region->userspace_addr == > > current_region->guest_user_addr > > + && region->guest_phys_addr == > > current_region- > >guest_phys_addr > > + && region->memory_size == > > current_region->size) { > > + if (dev->async_copy && > > rte_vfio_is_enabled("vfio")) > > + async_dma_map_region(dev, > > current_region, false); > > + dev_invalidate_vrings(dev); > > + free_mem_region(current_region); > > + dev->mem->nregions--; > > + return RTE_VHOST_MSG_RESULT_OK; > > + } > > + } > > + } > > + > > + VHOST_CONFIG_LOG(dev->ifname, ERR, "failed to find region"); > > + return RTE_VHOST_MSG_RESULT_ERR; } > > + > > static bool > > vq_is_ready(struct virtio_net *dev, struct vhost_virtqueue *vq) { > > -- > > 2.43.0 > >
