From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28]) by dpdk.org (Postfix) with ESMTP id 217352951 for ; Tue, 23 Aug 2016 11:58:46 +0200 (CEST) Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 7DF56C057FA9; Tue, 23 Aug 2016 09:58:45 +0000 (UTC) Received: from [10.36.4.245] (vpn1-4-245.ams2.redhat.com [10.36.4.245]) by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id u7N9whiu009471 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Tue, 23 Aug 2016 05:58:44 -0400 To: Yuanhan Liu , dev@dpdk.org References: <1471939839-29778-1-git-send-email-yuanhan.liu@linux.intel.com> <1471939839-29778-3-git-send-email-yuanhan.liu@linux.intel.com> From: Maxime Coquelin Message-ID: <13f37c6e-b389-a758-81cd-861db7337e1f@redhat.com> Date: Tue, 23 Aug 2016 11:58:42 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.2.0 MIME-Version: 1.0 In-Reply-To: <1471939839-29778-3-git-send-email-yuanhan.liu@linux.intel.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit X-Scanned-By: MIMEDefang 2.68 on 10.5.11.23 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.32]); Tue, 23 Aug 2016 09:58:45 +0000 (UTC) Subject: Re: [dpdk-dev] [PATCH 2/6] vhost: get guest/host physical address mappings X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 23 Aug 2016 09:58:46 -0000 On 08/23/2016 10:10 AM, Yuanhan Liu wrote: > So that we can convert a guest physical address to host physical > address, which will be used in later Tx zero copy implementation. > > Signed-off-by: Yuanhan Liu > --- > lib/librte_vhost/vhost.h | 30 +++++++++++++++ > lib/librte_vhost/vhost_user.c | 86 +++++++++++++++++++++++++++++++++++++++++++ > 2 files changed, 116 insertions(+) > > diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h > index df2107b..2d52987 100644 > --- a/lib/librte_vhost/vhost.h > +++ b/lib/librte_vhost/vhost.h > @@ -114,6 +114,12 @@ struct vhost_virtqueue { > #define VIRTIO_F_VERSION_1 32 > #endif > > +struct guest_page { > + uint64_t guest_phys_addr; > + uint64_t host_phys_addr; > + uint64_t size; > +}; > + > /** > * Device structure contains all configuration information relating > * to the device. > @@ -137,6 +143,10 @@ struct virtio_net { > uint64_t log_addr; > struct ether_addr mac; > > + uint32_t nr_guest_pages; > + uint32_t max_guest_pages; > + struct guest_page *guest_pages; > + > } __rte_cache_aligned; > > /** > @@ -217,6 +227,26 @@ gpa_to_vva(struct virtio_net *dev, uint64_t gpa) > return 0; > } > > +/* Convert guest physical address to host physical address */ > +static inline phys_addr_t __attribute__((always_inline)) > +gpa_to_hpa(struct virtio_net *dev, uint64_t gpa, uint64_t size) > +{ > + uint32_t i; > + struct guest_page *page; > + > + for (i = 0; i < dev->nr_guest_pages; i++) { > + page = &dev->guest_pages[i]; > + > + if (gpa >= page->guest_phys_addr && > + gpa + size < page->guest_phys_addr + page->size) { Shouldn't be '<=' here? > + return gpa - page->guest_phys_addr + > + page->host_phys_addr; > + } > + } > + > + return 0; > +} > + > struct virtio_net_device_ops const *notify_ops; > struct virtio_net *get_device(int vid); > > diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c > index d2071fd..045d4f0 100644 > --- a/lib/librte_vhost/vhost_user.c > +++ b/lib/librte_vhost/vhost_user.c > @@ -372,6 +372,81 @@ vhost_user_set_vring_base(struct virtio_net *dev, > return 0; > } > > +static void > +add_one_guest_page(struct virtio_net *dev, uint64_t guest_phys_addr, > + uint64_t host_phys_addr, uint64_t size) > +{ > + struct guest_page *page; > + > + if (dev->nr_guest_pages == dev->max_guest_pages) { > + dev->max_guest_pages *= 2; > + dev->guest_pages = realloc(dev->guest_pages, > + dev->max_guest_pages * sizeof(*page)); Maybe realloc return could be checked? > + } > + > + page = &dev->guest_pages[dev->nr_guest_pages++]; > + page->guest_phys_addr = guest_phys_addr; > + page->host_phys_addr = host_phys_addr; > + page->size = size; > +} > + > +static void > +add_guest_pages(struct virtio_net *dev, struct virtio_memory_region *reg, > + uint64_t page_size) > +{ > + uint64_t reg_size = reg->size; > + uint64_t host_user_addr = reg->host_user_addr; > + uint64_t guest_phys_addr = reg->guest_phys_addr; > + uint64_t host_phys_addr; > + uint64_t size; > + uint32_t pre_read; > + > + pre_read = *((uint32_t *)(uintptr_t)host_user_addr); > + host_phys_addr = rte_mem_virt2phy((void *)(uintptr_t)host_user_addr); > + size = page_size - (guest_phys_addr & (page_size - 1)); > + size = RTE_MIN(size, reg_size); > + > + add_one_guest_page(dev, guest_phys_addr, host_phys_addr, size); > + host_user_addr += size; > + guest_phys_addr += size; > + reg_size -= size; > + > + while (reg_size > 0) { > + pre_read += *((uint32_t *)(uintptr_t)host_user_addr); > + host_phys_addr = rte_mem_virt2phy((void *)(uintptr_t)host_user_addr); > + add_one_guest_page(dev, guest_phys_addr, host_phys_addr, page_size); > + > + host_user_addr += page_size; > + guest_phys_addr += page_size; > + reg_size -= page_size; > + } > + > + /* FIXME */ > + RTE_LOG(INFO, VHOST_CONFIG, ":: %u ::\n", pre_read); For my information, what is the purpose of pre_read? > +} > + > +/* TODO: enable it only in debug mode? */ > +static void > +dump_guest_pages(struct virtio_net *dev) > +{ > + uint32_t i; > + struct guest_page *page; > + > + for (i = 0; i < dev->nr_guest_pages; i++) { > + page = &dev->guest_pages[i]; > + > + RTE_LOG(INFO, VHOST_CONFIG, > + "guest physical page region %u\n" > + "\t guest_phys_addr: %" PRIx64 "\n" > + "\t host_phys_addr : %" PRIx64 "\n" > + "\t size : %" PRIx64 "\n", > + i, > + page->guest_phys_addr, > + page->host_phys_addr, > + page->size); > + } > +} > + > static int > vhost_user_set_mem_table(struct virtio_net *dev, struct VhostUserMsg *pmsg) > { > @@ -396,6 +471,13 @@ vhost_user_set_mem_table(struct virtio_net *dev, struct VhostUserMsg *pmsg) > dev->mem = NULL; > } > > + dev->nr_guest_pages = 0; > + if (!dev->guest_pages) { > + dev->max_guest_pages = 8; > + dev->guest_pages = malloc(dev->max_guest_pages * > + sizeof(struct guest_page)); > + } > + > dev->mem = rte_zmalloc("vhost-mem-table", sizeof(struct virtio_memory) + > sizeof(struct virtio_memory_region) * memory.nregions, 0); > if (dev->mem == NULL) { > @@ -447,6 +529,8 @@ vhost_user_set_mem_table(struct virtio_net *dev, struct VhostUserMsg *pmsg) > reg->mmap_size = mmap_size; > reg->host_user_addr = (uint64_t)(uintptr_t)mmap_addr + mmap_offset; > > + add_guest_pages(dev, reg, alignment); > + > RTE_LOG(INFO, VHOST_CONFIG, > "guest memory region %u, size: 0x%" PRIx64 "\n" > "\t guest physical addr: 0x%" PRIx64 "\n" > @@ -466,6 +550,8 @@ vhost_user_set_mem_table(struct virtio_net *dev, struct VhostUserMsg *pmsg) > mmap_offset); > } > > + dump_guest_pages(dev); > + > return 0; > > err_mmap: >