From patchwork Mon May 23 02:13:56 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jason Wang X-Patchwork-Id: 9131081 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 517B2607D3 for ; Mon, 23 May 2016 02:23:31 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 3D77928117 for ; Mon, 23 May 2016 02:23:31 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 3226E281EB; Mon, 23 May 2016 02:23:31 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.9 required=2.0 tests=BAYES_00,RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id 60A3D28117 for ; Mon, 23 May 2016 02:23:30 +0000 (UTC) Received: from localhost ([::1]:45239 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1b4fWj-00071u-FH for patchwork-qemu-devel@patchwork.kernel.org; Sun, 22 May 2016 22:23:29 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:49912) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1b4fOP-0008OG-Ul for qemu-devel@nongnu.org; Sun, 22 May 2016 22:14:55 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1b4fOO-0002oO-60 for qemu-devel@nongnu.org; Sun, 22 May 2016 22:14:53 -0400 Received: from mx1.redhat.com ([209.132.183.28]:52251) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1b4fON-0002oI-Tf for qemu-devel@nongnu.org; Sun, 22 May 2016 22:14:52 -0400 Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22]) (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 936E085364; Mon, 23 May 2016 02:14:51 +0000 (UTC) Received: from jason-ThinkPad-T450s.redhat.com (vpn1-7-188.pek2.redhat.com [10.72.7.188]) by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id u4N2E4x6000739; Sun, 22 May 2016 22:14:49 -0400 From: Jason Wang To: peter.maydell@linaro.org, qemu-devel@nongnu.org Date: Mon, 23 May 2016 10:13:56 +0800 Message-Id: <1463969642-5908-15-git-send-email-jasowang@redhat.com> In-Reply-To: <1463969642-5908-1-git-send-email-jasowang@redhat.com> References: <1463969642-5908-1-git-send-email-jasowang@redhat.com> X-Scanned-By: MIMEDefang 2.68 on 10.5.11.22 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.25]); Mon, 23 May 2016 02:14:51 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PULL 14/20] vmxnet3: Use pci_dma_* API instead of cpu_physical_memory_* X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Leonid Bloch , Jason Wang , Dmitry Fleytman Errors-To: qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Sender: "Qemu-devel" X-Virus-Scanned: ClamAV using ClamSMTP From: Dmitry Fleytman To make this device and network packets abstractions ready for IOMMU. Reviewed-by: Michael S. Tsirkin Signed-off-by: Dmitry Fleytman Signed-off-by: Leonid Bloch Signed-off-by: Jason Wang --- hw/net/net_tx_pkt.c | 16 +++++++++++----- hw/net/net_tx_pkt.h | 5 +++-- hw/net/vmxnet3.c | 51 ++++++++++++++++++++++++++++++--------------------- 3 files changed, 44 insertions(+), 28 deletions(-) diff --git a/hw/net/net_tx_pkt.c b/hw/net/net_tx_pkt.c index ad2258c..dbcbe23 100644 --- a/hw/net/net_tx_pkt.c +++ b/hw/net/net_tx_pkt.c @@ -20,6 +20,7 @@ #include "net/checksum.h" #include "net/tap.h" #include "net/net.h" +#include "hw/pci/pci.h" enum { NET_TX_PKT_VHDR_FRAG = 0, @@ -30,6 +31,8 @@ enum { /* TX packet private context */ struct NetTxPkt { + PCIDevice *pci_dev; + struct virtio_net_hdr virt_hdr; bool has_virt_hdr; @@ -54,11 +57,13 @@ struct NetTxPkt { bool is_loopback; }; -void net_tx_pkt_init(struct NetTxPkt **pkt, uint32_t max_frags, - bool has_virt_hdr) +void net_tx_pkt_init(struct NetTxPkt **pkt, PCIDevice *pci_dev, + uint32_t max_frags, bool has_virt_hdr) { struct NetTxPkt *p = g_malloc0(sizeof *p); + p->pci_dev = pci_dev; + p->vec = g_malloc((sizeof *p->vec) * (max_frags + NET_TX_PKT_PL_START_FRAG)); @@ -383,7 +388,8 @@ bool net_tx_pkt_add_raw_fragment(struct NetTxPkt *pkt, hwaddr pa, ventry = &pkt->raw[pkt->raw_frags]; mapped_len = len; - ventry->iov_base = cpu_physical_memory_map(pa, &mapped_len, false); + ventry->iov_base = pci_dma_map(pkt->pci_dev, pa, + &mapped_len, DMA_DIRECTION_TO_DEVICE); if ((ventry->iov_base != NULL) && (len == mapped_len)) { ventry->iov_len = mapped_len; @@ -444,8 +450,8 @@ void net_tx_pkt_reset(struct NetTxPkt *pkt) assert(pkt->raw); for (i = 0; i < pkt->raw_frags; i++) { assert(pkt->raw[i].iov_base); - cpu_physical_memory_unmap(pkt->raw[i].iov_base, pkt->raw[i].iov_len, - false, pkt->raw[i].iov_len); + pci_dma_unmap(pkt->pci_dev, pkt->raw[i].iov_base, pkt->raw[i].iov_len, + DMA_DIRECTION_TO_DEVICE, 0); } pkt->raw_frags = 0; diff --git a/hw/net/net_tx_pkt.h b/hw/net/net_tx_pkt.h index e49772d..07b9a20 100644 --- a/hw/net/net_tx_pkt.h +++ b/hw/net/net_tx_pkt.h @@ -31,11 +31,12 @@ struct NetTxPkt; * Init function for tx packet functionality * * @pkt: packet pointer + * @pci_dev: PCI device processing this packet * @max_frags: max tx ip fragments * @has_virt_hdr: device uses virtio header. */ -void net_tx_pkt_init(struct NetTxPkt **pkt, uint32_t max_frags, - bool has_virt_hdr); +void net_tx_pkt_init(struct NetTxPkt **pkt, PCIDevice *pci_dev, + uint32_t max_frags, bool has_virt_hdr); /** * Clean all tx packet resources. diff --git a/hw/net/vmxnet3.c b/hw/net/vmxnet3.c index 33cd07d..16645e6 100644 --- a/hw/net/vmxnet3.c +++ b/hw/net/vmxnet3.c @@ -802,7 +802,9 @@ vmxnet3_pop_rxc_descr(VMXNET3State *s, int qidx, uint32_t *descr_gen) hwaddr daddr = vmxnet3_ring_curr_cell_pa(&s->rxq_descr[qidx].comp_ring); - cpu_physical_memory_read(daddr, &rxcd, sizeof(struct Vmxnet3_RxCompDesc)); + pci_dma_read(PCI_DEVICE(s), daddr, + &rxcd, sizeof(struct Vmxnet3_RxCompDesc)); + ring_gen = vmxnet3_ring_curr_gen(&s->rxq_descr[qidx].comp_ring); if (rxcd.gen != ring_gen) { @@ -1023,10 +1025,11 @@ nocsum: } static void -vmxnet3_physical_memory_writev(const struct iovec *iov, - size_t start_iov_off, - hwaddr target_addr, - size_t bytes_to_copy) +vmxnet3_pci_dma_writev(PCIDevice *pci_dev, + const struct iovec *iov, + size_t start_iov_off, + hwaddr target_addr, + size_t bytes_to_copy) { size_t curr_off = 0; size_t copied = 0; @@ -1036,9 +1039,9 @@ vmxnet3_physical_memory_writev(const struct iovec *iov, size_t chunk_len = MIN((curr_off + iov->iov_len) - start_iov_off, bytes_to_copy); - cpu_physical_memory_write(target_addr + copied, - iov->iov_base + start_iov_off - curr_off, - chunk_len); + pci_dma_write(pci_dev, target_addr + copied, + iov->iov_base + start_iov_off - curr_off, + chunk_len); copied += chunk_len; start_iov_off += chunk_len; @@ -1088,15 +1091,15 @@ vmxnet3_indicate_packet(VMXNET3State *s) } chunk_size = MIN(bytes_left, rxd.len); - vmxnet3_physical_memory_writev(data, bytes_copied, - le64_to_cpu(rxd.addr), chunk_size); + vmxnet3_pci_dma_writev(PCI_DEVICE(s), data, bytes_copied, + le64_to_cpu(rxd.addr), chunk_size); bytes_copied += chunk_size; bytes_left -= chunk_size; vmxnet3_dump_rx_descr(&rxd); if (ready_rxcd_pa != 0) { - cpu_physical_memory_write(ready_rxcd_pa, &rxcd, sizeof(rxcd)); + pci_dma_write(PCI_DEVICE(s), ready_rxcd_pa, &rxcd, sizeof(rxcd)); } memset(&rxcd, 0, sizeof(struct Vmxnet3_RxCompDesc)); @@ -1127,7 +1130,8 @@ vmxnet3_indicate_packet(VMXNET3State *s) if (ready_rxcd_pa != 0) { rxcd.eop = 1; rxcd.err = (bytes_left != 0); - cpu_physical_memory_write(ready_rxcd_pa, &rxcd, sizeof(rxcd)); + + pci_dma_write(PCI_DEVICE(s), ready_rxcd_pa, &rxcd, sizeof(rxcd)); /* Flush RX descriptor changes */ smp_wmb(); @@ -1298,7 +1302,8 @@ static void vmxnet3_update_mcast_filters(VMXNET3State *s) VMXNET3_READ_DRV_SHARED64(s->drv_shmem, devRead.rxFilterConf.mfTablePA); - cpu_physical_memory_read(mcast_list_pa, s->mcast_list, list_bytes); + pci_dma_read(PCI_DEVICE(s), mcast_list_pa, s->mcast_list, list_bytes); + VMW_CFPRN("Current multicast list len is %d:", s->mcast_list_len); for (i = 0; i < s->mcast_list_len; i++) { VMW_CFPRN("\t" MAC_FMT, MAC_ARG(s->mcast_list[i].a)); @@ -1328,15 +1333,17 @@ static void vmxnet3_fill_stats(VMXNET3State *s) return; for (i = 0; i < s->txq_num; i++) { - cpu_physical_memory_write(s->txq_descr[i].tx_stats_pa, - &s->txq_descr[i].txq_stats, - sizeof(s->txq_descr[i].txq_stats)); + pci_dma_write(PCI_DEVICE(s), + s->txq_descr[i].tx_stats_pa, + &s->txq_descr[i].txq_stats, + sizeof(s->txq_descr[i].txq_stats)); } for (i = 0; i < s->rxq_num; i++) { - cpu_physical_memory_write(s->rxq_descr[i].rx_stats_pa, - &s->rxq_descr[i].rxq_stats, - sizeof(s->rxq_descr[i].rxq_stats)); + pci_dma_write(PCI_DEVICE(s), + s->rxq_descr[i].rx_stats_pa, + &s->rxq_descr[i].rxq_stats, + sizeof(s->rxq_descr[i].rxq_stats)); } } @@ -1558,7 +1565,8 @@ static void vmxnet3_activate_device(VMXNET3State *s) /* Preallocate TX packet wrapper */ VMW_CFPRN("Max TX fragments is %u", s->max_tx_frags); - net_tx_pkt_init(&s->tx_pkt, s->max_tx_frags, s->peer_has_vhdr); + net_tx_pkt_init(&s->tx_pkt, PCI_DEVICE(s), + s->max_tx_frags, s->peer_has_vhdr); net_rx_pkt_init(&s->rx_pkt, s->peer_has_vhdr); /* Read rings memory locations for RX queues */ @@ -2536,7 +2544,8 @@ static int vmxnet3_post_load(void *opaque, int version_id) VMXNET3State *s = opaque; PCIDevice *d = PCI_DEVICE(s); - net_tx_pkt_init(&s->tx_pkt, s->max_tx_frags, s->peer_has_vhdr); + net_tx_pkt_init(&s->tx_pkt, PCI_DEVICE(s), + s->max_tx_frags, s->peer_has_vhdr); net_rx_pkt_init(&s->rx_pkt, s->peer_has_vhdr); if (s->msix_used) {