]> git.karo-electronics.de Git - karo-tx-linux.git/commitdiff
caif_virtio: Introduce caif over virtio
authorVikram ARV <vikram.arv@stericsson.com>
Tue, 19 Feb 2013 20:45:00 +0000 (07:15 +1030)
committerRusty Russell <rusty@rustcorp.com.au>
Tue, 19 Feb 2013 20:47:29 +0000 (07:17 +1030)
Add the the Virtio shared memory driver for STE Modems.
caif_virtio is implemented utilizing the virtio framework
for data transport and is managed with the remoteproc frameworks.

The Virtio queue is used for transmitting data to the modem, and
the new vringh implementation is receiving data over the vring.

Signed-off-by: Vikram ARV <vikram.arv@stericsson.com>
Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
cc: David S. Miller <davem@davemloft.net>
cc: Ohad Ben-Cohen <ohad@wizery.com>
cc: Rusty Russell <rusty@rustcorp.com.au>
cc: Ido Yariv <ido@wizery.com>
cc: Erwan Yvin <erwan.yvin@stericsson.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
drivers/net/caif/Kconfig
drivers/net/caif/Makefile
drivers/net/caif/caif_virtio.c [new file with mode: 0644]
include/linux/virtio_caif.h [new file with mode: 0644]
include/uapi/linux/virtio_ids.h

index abf4d7a9dcce5183fbc89b7a7e3475b86d219618..a8b67e9be8bea2c896f927c0298d51bf393b9176 100644 (file)
@@ -47,3 +47,11 @@ config CAIF_HSI
        The caif low level driver for CAIF over HSI.
        Be aware that if you enable this then you also need to
        enable a low-level HSI driver.
+
+config CAIF_VIRTIO
+       tristate "CAIF virtio transport driver"
+       depends on CAIF
+       depends on REMOTEPROC
+       default n
+       ---help---
+       The caif driver for CAIF over Virtio.
index 91dff861560f71c8d58010066b614268ce60a935..d9ee26a96c6e19b8699358a79c2a3a6bed325382 100644 (file)
@@ -13,3 +13,6 @@ obj-$(CONFIG_CAIF_SHM) += caif_shm.o
 
 # HSI interface
 obj-$(CONFIG_CAIF_HSI) += caif_hsi.o
+
+# Virtio interface
+obj-$(CONFIG_CAIF_VIRTIO) += caif_virtio.o
diff --git a/drivers/net/caif/caif_virtio.c b/drivers/net/caif/caif_virtio.c
new file mode 100644 (file)
index 0000000..df832e7
--- /dev/null
@@ -0,0 +1,547 @@
+/*
+ * Copyright (C) ST-Ericsson AB 2012
+ * Contact: Sjur Brendeland / sjur.brandeland@stericsson.com
+ * Authors: Vicram Arv / vikram.arv@stericsson.com,
+ *         Dmitry Tarnyagin / dmitry.tarnyagin@stericsson.com
+ *         Sjur Brendeland / sjur.brandeland@stericsson.com
+ * License terms: GNU General Public License (GPL) version 2
+ */
+
+#include <linux/module.h>
+#include <linux/virtio.h>
+#include <linux/virtio_ids.h>
+#include <linux/virtio_config.h>
+#include <linux/dma-mapping.h>
+#include <linux/netdevice.h>
+#include <linux/if_arp.h>
+#include <linux/spinlock.h>
+#include <linux/virtio_caif.h>
+#include <linux/virtio_ring.h>
+#include <linux/vringh.h>
+#include <linux/remoteproc.h>
+#include <net/caif/caif_dev.h>
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Vicram Arv <vikram.arv@stericsson.com>");
+MODULE_AUTHOR("Sjur Brendeland <sjur.brandeland@stericsson.com");
+MODULE_DESCRIPTION("Virtio CAIF Driver");
+
+/* Virtio Ring used in receive direction */
+#define RX_RING_INDEX 0
+
+#define CFV_DEFAULT_QUOTA 32
+
+/* struct cfv_napi_contxt - NAPI context info
+ * @riov: IOV holding data read from the ring. Note that riov may
+ *       still hold data when cfv_rx_poll() returns.
+ * @head: Last descriptor ID we received from vringh_getdesc_kern.
+ *       We use this to put descriptor back on the used ring. USHRT_MAX is
+ *       used to indicate invalid head-id.
+ */
+struct cfv_napi_context {
+       struct vringh_kiov riov;
+       unsigned short head;
+};
+
+/* struct cfv_info - Caif Virtio control structure
+ * @cfdev:     caif common header
+ * @vdev:      Associated virtio device
+ * @vq_rx:     rx/downlink virtqueue
+ * @vq_tx:     tx/uplink virtqueue
+ * @ndev:      associated netdevice
+ * @queued_tx: number of buffers queued in the tx virtqueue
+ * @watermark_tx: indicates number of buffers the tx queue
+ *             should shrink to to unblock datapath
+ * @tx_lock:   protects vq_tx to allow concurrent senders
+ * @tx_hr:     transmit headroom
+ * @rx_hr:     receive headroom
+ * @tx_tr:     transmit tailroom
+ * @rx_tr:     receive tailroom
+ * @mtu:       transmit max size
+ * @mru:       receive max size
+ */
+struct cfv_info {
+       struct caif_dev_common cfdev;
+       struct virtio_device *vdev;
+       struct vringh *vr_rx;
+       struct virtqueue *vq_tx;
+       struct net_device *ndev;
+       unsigned int queued_tx;
+       unsigned int watermark_tx;
+       /* Protect access to vq_tx */
+       spinlock_t tx_lock;
+       struct tasklet_struct tx_release_tasklet;
+       struct napi_struct napi;
+       struct cfv_napi_context ctx;
+       /* Copied from Virtio config space */
+       u16 tx_hr;
+       u16 rx_hr;
+       u16 tx_tr;
+       u16 rx_tr;
+       u32 mtu;
+       u32 mru;
+};
+
+/* struct token_info - maintains Transmit buffer data handle
+ * @size:      size of transmit buffer
+ * @dma_handle: handle to allocated dma device memory area
+ * @vaddr:     virtual address mapping to allocated memory area
+ */
+struct token_info {
+       size_t size;
+       u8 *vaddr;
+       dma_addr_t dma_handle;
+};
+
+/* Default if virtio config space is unavailable */
+#define CFV_DEF_MTU_SIZE 4096
+#define CFV_DEF_HEADROOM 32
+#define CFV_DEF_TAILROOM 32
+
+/* Require IP header to be 4-byte aligned. */
+#define IP_HDR_ALIGN 4
+
+static void cfv_release_cb(struct virtqueue *vq_tx)
+{
+       struct cfv_info *cfv = vq_tx->vdev->priv;
+       tasklet_schedule(&cfv->tx_release_tasklet);
+}
+
+/* This is invoked whenever the remote processor completed processing
+ * a TX msg we just sent it, and the buffer is put back to the used ring.
+ */
+static void cfv_release_used_buf(struct virtqueue *vq_tx)
+{
+       struct cfv_info *cfv = vq_tx->vdev->priv;
+       unsigned long flags;
+
+       BUG_ON(vq_tx != cfv->vq_tx);
+       WARN_ON_ONCE(irqs_disabled());
+
+       for (;;) {
+               unsigned int len;
+               struct token_info *buf_info;
+
+               /* Get used buffer from used ring to recycle used descriptors */
+               spin_lock_irqsave(&cfv->tx_lock, flags);
+               buf_info = virtqueue_get_buf(vq_tx, &len);
+
+               if (!buf_info)
+                       goto out;
+
+               BUG_ON(!cfv->queued_tx);
+               if (--cfv->queued_tx <= cfv->watermark_tx) {
+                       cfv->watermark_tx = 0;
+                       netif_tx_wake_all_queues(cfv->ndev);
+               }
+               spin_unlock_irqrestore(&cfv->tx_lock, flags);
+
+               dma_free_coherent(vq_tx->vdev->dev.parent->parent,
+                                 buf_info->size, buf_info->vaddr,
+                                 buf_info->dma_handle);
+               kfree(buf_info);
+       }
+       return;
+out:
+       spin_unlock_irqrestore(&cfv->tx_lock, flags);
+}
+
+static struct sk_buff *cfv_alloc_and_copy_skb(int *err,
+                                             struct cfv_info *cfv,
+                                             u8 *frm, u32 frm_len)
+{
+       struct sk_buff *skb;
+       u32 cfpkt_len, pad_len;
+
+       *err = 0;
+       /* Verify that packet size with down-link header and mtu size */
+       if (frm_len > cfv->mru || frm_len <= cfv->rx_hr + cfv->rx_tr) {
+               netdev_err(cfv->ndev,
+                          "Invalid frmlen:%u  mtu:%u hr:%d tr:%d\n",
+                          frm_len, cfv->mru,  cfv->rx_hr,
+                          cfv->rx_tr);
+               *err = -EPROTO;
+               return NULL;
+       }
+
+       cfpkt_len = frm_len - (cfv->rx_hr + cfv->rx_tr);
+
+       pad_len = (unsigned long)(frm + cfv->rx_hr) & (IP_HDR_ALIGN - 1);
+
+       skb = netdev_alloc_skb(cfv->ndev, frm_len + pad_len);
+       if (!skb) {
+               *err = -ENOMEM;
+               return NULL;
+       }
+       /* Reserve space for headers. */
+       skb_reserve(skb, cfv->rx_hr + pad_len);
+
+       memcpy(skb_put(skb, cfpkt_len), frm + cfv->rx_hr, cfpkt_len);
+       return skb;
+}
+
+static int cfv_rx_poll(struct napi_struct *napi, int quota)
+{
+       struct cfv_info *cfv = container_of(napi, struct cfv_info, napi);
+       int rxcnt = 0;
+       int err = 0;
+       void *buf;
+       struct sk_buff *skb;
+       struct vringh_kiov *riov = &cfv->ctx.riov;
+
+       do {
+               skb = NULL;
+               if (riov->i == riov->used) {
+                       if (cfv->ctx.head != USHRT_MAX) {
+                               vringh_complete_kern(cfv->vr_rx,
+                                                    cfv->ctx.head,
+                                                    0);
+                               cfv->ctx.head = USHRT_MAX;
+                       }
+
+                       err = vringh_getdesc_kern(
+                               cfv->vr_rx,
+                               riov,
+                               NULL,
+                               &cfv->ctx.head,
+                               GFP_ATOMIC);
+
+                       if (err <= 0)
+                               goto out;
+
+               }
+
+               buf = phys_to_virt((unsigned long) riov->iov[riov->i].iov_base);
+               /* TODO: Add check on valid buffer address */
+
+               skb = cfv_alloc_and_copy_skb(&err, cfv, buf,
+                                            riov->iov[riov->i].iov_len);
+               if (unlikely(err))
+                       goto out;
+
+               /* Push received packet up the stack. */
+               skb->protocol = htons(ETH_P_CAIF);
+               skb_reset_mac_header(skb);
+               skb->dev = cfv->ndev;
+               err = netif_receive_skb(skb);
+               if (unlikely(err)) {
+                       ++cfv->ndev->stats.rx_dropped;
+               } else {
+                       ++cfv->ndev->stats.rx_packets;
+                       cfv->ndev->stats.rx_bytes += skb->len;
+               }
+
+               ++riov->i;
+               ++rxcnt;
+       } while (rxcnt < quota);
+
+       return rxcnt;
+
+out:
+       switch (err) {
+       case 0:
+               /* Empty ring, enable notifications and stop NAPI polling */
+               if (!vringh_notify_enable_kern(cfv->vr_rx))
+                       napi_complete(napi);
+
+               return rxcnt;
+
+       case -ENOMEM:
+               dev_kfree_skb(skb);
+               /* Stop NAPI poll on OOM, we hope to be polled later */
+               napi_complete(napi);
+               vringh_notify_enable_kern(cfv->vr_rx);
+               break;
+
+       default:
+               /* We're doomed, any modem fault is fatal */
+               netdev_warn(cfv->ndev, "Bad ring, disable device\n");
+               cfv->ndev->stats.rx_dropped = riov->used - riov->i;
+               napi_complete(napi);
+               vringh_notify_disable_kern(cfv->vr_rx);
+               netif_carrier_off(cfv->ndev);
+               break;
+       }
+
+       return rxcnt;
+}
+
+static irqreturn_t cfv_recv(struct virtio_device *vdev, struct vringh *vr_rx)
+{
+       struct cfv_info *cfv = vdev->priv;
+
+       vringh_notify_disable_kern(cfv->vr_rx);
+       napi_schedule(&cfv->napi);
+       return IRQ_HANDLED;
+}
+
+static int cfv_netdev_open(struct net_device *netdev)
+{
+       struct cfv_info *cfv = netdev_priv(netdev);
+
+       netif_carrier_on(netdev);
+       napi_enable(&cfv->napi);
+       return 0;
+}
+
+static int cfv_netdev_close(struct net_device *netdev)
+{
+       struct cfv_info *cfv = netdev_priv(netdev);
+
+       netif_carrier_off(netdev);
+       napi_disable(&cfv->napi);
+       return 0;
+}
+
+static struct token_info *cfv_alloc_and_copy_to_dmabuf(struct cfv_info *cfv,
+                                                      struct sk_buff *skb,
+                                                      struct scatterlist *sg)
+{
+       struct caif_payload_info *info = (void *)&skb->cb;
+       struct token_info *buf_info = NULL;
+       u8 pad_len, hdr_ofs;
+
+       if (unlikely(cfv->tx_hr + skb->len + cfv->tx_tr > cfv->mtu)) {
+               netdev_warn(cfv->ndev, "Invalid packet len (%d > %d)\n",
+                           cfv->tx_hr + skb->len + cfv->tx_tr, cfv->mtu);
+               goto err;
+       }
+
+       buf_info = kmalloc(sizeof(struct token_info), GFP_ATOMIC);
+       if (unlikely(!buf_info))
+               goto err;
+
+       /* Make the IP header aligned in tbe buffer */
+       hdr_ofs = cfv->tx_hr + info->hdr_len;
+       pad_len = hdr_ofs & (IP_HDR_ALIGN - 1);
+       buf_info->size = cfv->tx_hr + skb->len + cfv->tx_tr + pad_len;
+
+       if (WARN_ON_ONCE(!cfv->vdev->dev.parent))
+               goto err;
+
+       /* allocate coherent memory for the buffers */
+       buf_info->vaddr =
+               dma_alloc_coherent(cfv->vdev->dev.parent->parent,
+                                  buf_info->size, &buf_info->dma_handle,
+                                  GFP_ATOMIC);
+       if (unlikely(!buf_info->vaddr)) {
+               netdev_warn(cfv->ndev,
+                           "Out of DMA memory (alloc %zu bytes)\n",
+                           buf_info->size);
+               goto err;
+       }
+
+       /* copy skbuf contents to send buffer */
+       skb_copy_bits(skb, 0, buf_info->vaddr + cfv->tx_hr + pad_len, skb->len);
+       sg_init_one(sg, buf_info->vaddr + pad_len,
+                   skb->len + cfv->tx_hr + cfv->rx_hr);
+
+       return buf_info;
+err:
+       kfree(buf_info);
+       return NULL;
+}
+
+/* This is invoked whenever the host processor application has sent
+ * up-link data. Send it in the TX VQ avail ring.
+ *
+ * CAIF Virtio sends does not use linked descriptors in the tx direction.
+ */
+static int cfv_netdev_tx(struct sk_buff *skb, struct net_device *netdev)
+{
+       struct cfv_info *cfv = netdev_priv(netdev);
+       struct token_info *buf_info;
+       struct scatterlist sg;
+       unsigned long flags;
+       int ret;
+
+       buf_info = cfv_alloc_and_copy_to_dmabuf(cfv, skb, &sg);
+
+       spin_lock_irqsave(&cfv->tx_lock, flags);
+       if (unlikely(!buf_info))
+               goto flow_off;
+
+       /* Add buffer to avail ring. Flow control below should ensure
+        * that this always succeedes
+        */
+       ret = virtqueue_add_buf(cfv->vq_tx, &sg, 1, 0,
+                               buf_info, GFP_ATOMIC);
+
+       if (unlikely(WARN_ON(ret < 0))) {
+               kfree(buf_info);
+               goto flow_off;
+       }
+
+
+       /* update netdev statistics */
+       cfv->queued_tx++;
+       cfv->ndev->stats.tx_packets++;
+       cfv->ndev->stats.tx_bytes += skb->len;
+
+       /* tell the remote processor it has a pending message to read */
+       virtqueue_kick(cfv->vq_tx);
+
+       /* Flow-off check takes into account number of cpus to make sure
+        * virtqueue will not be overfilled in any possible smp conditions.
+        *
+        * Flow-on is triggered when sufficient buffers are freed
+        */
+       if (ret <= num_present_cpus()) {
+flow_off:
+               cfv->watermark_tx = cfv->queued_tx >> 1;
+               netif_tx_stop_all_queues(netdev);
+       }
+
+       spin_unlock_irqrestore(&cfv->tx_lock, flags);
+
+       dev_kfree_skb(skb);
+       tasklet_schedule(&cfv->tx_release_tasklet);
+       return NETDEV_TX_OK;
+}
+
+static void cfv_tx_release_tasklet(unsigned long drv)
+{
+       struct cfv_info *cfv = (struct cfv_info *)drv;
+       cfv_release_used_buf(cfv->vq_tx);
+}
+
+static const struct net_device_ops cfv_netdev_ops = {
+       .ndo_open = cfv_netdev_open,
+       .ndo_stop = cfv_netdev_close,
+       .ndo_start_xmit = cfv_netdev_tx,
+};
+
+static void cfv_netdev_setup(struct net_device *netdev)
+{
+       netdev->netdev_ops = &cfv_netdev_ops;
+       netdev->type = ARPHRD_CAIF;
+       netdev->tx_queue_len = 100;
+       netdev->flags = IFF_POINTOPOINT | IFF_NOARP;
+       netdev->mtu = CFV_DEF_MTU_SIZE;
+       netdev->destructor = free_netdev;
+}
+
+static int cfv_probe(struct virtio_device *vdev)
+{
+       vq_callback_t *vq_cbs = cfv_release_cb;
+       const char *names =  "output";
+       const char *cfv_netdev_name = "cfvrt";
+       struct net_device *netdev;
+       struct virtqueue *vqs;
+
+       struct cfv_info *cfv;
+       int err = 0;
+
+       netdev = alloc_netdev(sizeof(struct cfv_info), cfv_netdev_name,
+                             cfv_netdev_setup);
+       if (!netdev)
+               return -ENOMEM;
+
+       cfv = netdev_priv(netdev);
+       cfv->vdev = vdev;
+       cfv->ndev = netdev;
+
+       spin_lock_init(&cfv->tx_lock);
+
+       cfv->vr_rx = rproc_virtio_new_vringh(vdev, RX_RING_INDEX, cfv_recv);
+       if (!cfv->vr_rx)
+               goto free_cfv;
+
+       /* Get the TX (uplink) virtque */
+       err = vdev->config->find_vqs(vdev, 1, &vqs, &vq_cbs, &names);
+       if (err)
+               goto free_cfv;
+
+       cfv->vq_tx = vqs;
+
+#define GET_VIRTIO_CONFIG_OPS(_v, _var, _f) \
+       ((_v)->config->get(_v, offsetof(struct virtio_caif_transf_config, _f), \
+                          &_var, \
+                          FIELD_SIZEOF(struct virtio_caif_transf_config, _f)))
+
+       if (vdev->config->get) {
+               GET_VIRTIO_CONFIG_OPS(vdev, cfv->tx_hr, headroom);
+               GET_VIRTIO_CONFIG_OPS(vdev, cfv->rx_hr, headroom);
+               GET_VIRTIO_CONFIG_OPS(vdev, cfv->tx_tr, tailroom);
+               GET_VIRTIO_CONFIG_OPS(vdev, cfv->rx_tr, tailroom);
+               GET_VIRTIO_CONFIG_OPS(vdev, cfv->mtu, mtu);
+               GET_VIRTIO_CONFIG_OPS(vdev, cfv->mru, mtu);
+       } else {
+               cfv->tx_hr = CFV_DEF_HEADROOM;
+               cfv->rx_hr = CFV_DEF_HEADROOM;
+               cfv->tx_tr = CFV_DEF_TAILROOM;
+               cfv->rx_tr = CFV_DEF_TAILROOM;
+               cfv->mtu = CFV_DEF_MTU_SIZE;
+               cfv->mru = CFV_DEF_MTU_SIZE;
+       }
+
+       netdev->needed_headroom = cfv->tx_hr;
+       netdev->needed_tailroom = cfv->tx_tr;
+
+       /* Subtract needed tailroom from MTU to ensure enough room */
+       netdev->mtu = cfv->mtu - cfv->tx_tr;
+
+       vdev->priv = cfv;
+
+       vringh_kiov_init(&cfv->ctx.riov, NULL, 0);
+       cfv->ctx.head = USHRT_MAX;
+
+       netif_napi_add(netdev, &cfv->napi, cfv_rx_poll, CFV_DEFAULT_QUOTA);
+       tasklet_init(&cfv->tx_release_tasklet,
+                    cfv_tx_release_tasklet,
+                    (unsigned long)cfv);
+
+       netif_carrier_off(netdev);
+
+       /* register Netdev */
+       err = register_netdev(netdev);
+       if (err) {
+               dev_err(&vdev->dev, "Unable to register netdev (%d)\n", err);
+               goto vqs_del;
+       }
+
+       /* tell the remote processor it can start sending messages */
+       rproc_virtio_kick_vringh(vdev, RX_RING_INDEX);
+
+       return 0;
+
+vqs_del:
+       vdev->config->del_vqs(cfv->vdev);
+free_cfv:
+       free_netdev(netdev);
+       return err;
+}
+
+static void cfv_remove(struct virtio_device *vdev)
+{
+       struct cfv_info *cfv = vdev->priv;
+
+       tasklet_kill(&cfv->tx_release_tasklet);
+       vringh_kiov_cleanup(&cfv->ctx.riov);
+       vdev->config->reset(vdev);
+       rproc_virtio_del_vringh(vdev, RX_RING_INDEX);
+       cfv->vr_rx = NULL;
+       vdev->config->del_vqs(cfv->vdev);
+       unregister_netdev(cfv->ndev);
+}
+
+static struct virtio_device_id id_table[] = {
+       { VIRTIO_ID_CAIF, VIRTIO_DEV_ANY_ID },
+       { 0 },
+};
+
+static unsigned int features[] = {
+};
+
+static struct virtio_driver caif_virtio_driver = {
+       .feature_table          = features,
+       .feature_table_size     = ARRAY_SIZE(features),
+       .driver.name            = KBUILD_MODNAME,
+       .driver.owner           = THIS_MODULE,
+       .id_table               = id_table,
+       .probe                  = cfv_probe,
+       .remove                 = cfv_remove,
+};
+
+module_virtio_driver(caif_virtio_driver);
+MODULE_DEVICE_TABLE(virtio, id_table);
diff --git a/include/linux/virtio_caif.h b/include/linux/virtio_caif.h
new file mode 100644 (file)
index 0000000..5d2d312
--- /dev/null
@@ -0,0 +1,24 @@
+/*
+ * Copyright (C) ST-Ericsson AB 2012
+ * Author: Sjur Brændeland <sjur.brandeland@stericsson.com>
+ *
+ * This header is BSD licensed so
+ * anyone can use the definitions to implement compatible remote processors
+ */
+
+#ifndef VIRTIO_CAIF_H
+#define VIRTIO_CAIF_H
+
+#include <linux/types.h>
+struct virtio_caif_transf_config {
+       u16 headroom;
+       u16 tailroom;
+       u32 mtu;
+       u8 reserved[4];
+};
+
+struct virtio_caif_config {
+       struct virtio_caif_transf_config uplink, downlink;
+       u8 reserved[8];
+};
+#endif
index a7630d04029f8860bb3685b1f6863b6e4c8a5465..284fc3a05f7bb5a80c08e6980133c87b6f0c9397 100644 (file)
@@ -38,5 +38,6 @@
 #define VIRTIO_ID_SCSI         8 /* virtio scsi */
 #define VIRTIO_ID_9P           9 /* 9p virtio console */
 #define VIRTIO_ID_RPROC_SERIAL 11 /* virtio remoteproc serial link */
+#define VIRTIO_ID_CAIF        12 /* Virtio caif */
 
 #endif /* _LINUX_VIRTIO_IDS_H */