660 lines
17 KiB
C
660 lines
17 KiB
C
// SPDX-License-Identifier: GPL-2.0-or-later
|
|
/* MHI MBIM Network driver - Network/MBIM over MHI bus
|
|
*
|
|
* Copyright (C) 2021 Linaro Ltd <loic.poulain@linaro.org>
|
|
*
|
|
* This driver copy some code from cdc_ncm, which is:
|
|
* Copyright (C) ST-Ericsson 2010-2012
|
|
* and cdc_mbim, which is:
|
|
* Copyright (c) 2012 Smith Micro Software, Inc.
|
|
* Copyright (c) 2012 Bjørn Mork <bjorn@mork.no>
|
|
*
|
|
*/
|
|
|
|
#include <linux/ethtool.h>
|
|
#include <linux/if_arp.h>
|
|
#include <linux/if_vlan.h>
|
|
#include <linux/ip.h>
|
|
#include <linux/mhi.h>
|
|
#include <linux/mii.h>
|
|
#include <linux/mod_devicetable.h>
|
|
#include <linux/module.h>
|
|
#include <linux/netdevice.h>
|
|
#include <linux/skbuff.h>
|
|
#include <linux/u64_stats_sync.h>
|
|
#include <linux/usb.h>
|
|
#include <linux/usb/cdc.h>
|
|
#include <linux/usb/usbnet.h>
|
|
#include <linux/usb/cdc_ncm.h>
|
|
#include <linux/wwan.h>
|
|
|
|
/* 3500 allows to optimize skb allocation, the skbs will basically fit in
|
|
* one 4K page. Large MBIM packets will simply be split over several MHI
|
|
* transfers and chained by the MHI net layer (zerocopy).
|
|
*/
|
|
#define MHI_DEFAULT_MRU 3500
|
|
|
|
#define MHI_MBIM_DEFAULT_MTU 1500
|
|
#define MHI_MAX_BUF_SZ 0xffff
|
|
|
|
#define MBIM_NDP16_SIGN_MASK 0x00ffffff
|
|
|
|
#define MHI_MBIM_LINK_HASH_SIZE 8
|
|
#define LINK_HASH(session) ((session) % MHI_MBIM_LINK_HASH_SIZE)
|
|
|
|
struct mhi_mbim_link {
|
|
struct mhi_mbim_context *mbim;
|
|
struct net_device *ndev;
|
|
unsigned int session;
|
|
|
|
/* stats */
|
|
u64_stats_t rx_packets;
|
|
u64_stats_t rx_bytes;
|
|
u64_stats_t rx_errors;
|
|
u64_stats_t tx_packets;
|
|
u64_stats_t tx_bytes;
|
|
u64_stats_t tx_errors;
|
|
u64_stats_t tx_dropped;
|
|
struct u64_stats_sync tx_syncp;
|
|
struct u64_stats_sync rx_syncp;
|
|
|
|
struct hlist_node hlnode;
|
|
};
|
|
|
|
struct mhi_mbim_context {
|
|
struct mhi_device *mdev;
|
|
struct sk_buff *skbagg_head;
|
|
struct sk_buff *skbagg_tail;
|
|
unsigned int mru;
|
|
u32 rx_queue_sz;
|
|
u16 rx_seq;
|
|
u16 tx_seq;
|
|
struct delayed_work rx_refill;
|
|
spinlock_t tx_lock;
|
|
struct hlist_head link_list[MHI_MBIM_LINK_HASH_SIZE];
|
|
};
|
|
|
|
struct mbim_tx_hdr {
|
|
struct usb_cdc_ncm_nth16 nth16;
|
|
struct usb_cdc_ncm_ndp16 ndp16;
|
|
struct usb_cdc_ncm_dpe16 dpe16[2];
|
|
} __packed;
|
|
|
|
static struct mhi_mbim_link *mhi_mbim_get_link_rcu(struct mhi_mbim_context *mbim,
|
|
unsigned int session)
|
|
{
|
|
struct mhi_mbim_link *link;
|
|
|
|
hlist_for_each_entry_rcu(link, &mbim->link_list[LINK_HASH(session)], hlnode) {
|
|
if (link->session == session)
|
|
return link;
|
|
}
|
|
|
|
return NULL;
|
|
}
|
|
|
|
static struct sk_buff *mbim_tx_fixup(struct sk_buff *skb, unsigned int session,
|
|
u16 tx_seq)
|
|
{
|
|
unsigned int dgram_size = skb->len;
|
|
struct usb_cdc_ncm_nth16 *nth16;
|
|
struct usb_cdc_ncm_ndp16 *ndp16;
|
|
struct mbim_tx_hdr *mbim_hdr;
|
|
|
|
/* Only one NDP is sent, containing the IP packet (no aggregation) */
|
|
|
|
/* Ensure we have enough headroom for crafting MBIM header */
|
|
if (skb_cow_head(skb, sizeof(struct mbim_tx_hdr))) {
|
|
dev_kfree_skb_any(skb);
|
|
return NULL;
|
|
}
|
|
|
|
mbim_hdr = skb_push(skb, sizeof(struct mbim_tx_hdr));
|
|
|
|
/* Fill NTB header */
|
|
nth16 = &mbim_hdr->nth16;
|
|
nth16->dwSignature = cpu_to_le32(USB_CDC_NCM_NTH16_SIGN);
|
|
nth16->wHeaderLength = cpu_to_le16(sizeof(struct usb_cdc_ncm_nth16));
|
|
nth16->wSequence = cpu_to_le16(tx_seq);
|
|
nth16->wBlockLength = cpu_to_le16(skb->len);
|
|
nth16->wNdpIndex = cpu_to_le16(sizeof(struct usb_cdc_ncm_nth16));
|
|
|
|
/* Fill the unique NDP */
|
|
ndp16 = &mbim_hdr->ndp16;
|
|
ndp16->dwSignature = cpu_to_le32(USB_CDC_MBIM_NDP16_IPS_SIGN | (session << 24));
|
|
ndp16->wLength = cpu_to_le16(sizeof(struct usb_cdc_ncm_ndp16)
|
|
+ sizeof(struct usb_cdc_ncm_dpe16) * 2);
|
|
ndp16->wNextNdpIndex = 0;
|
|
|
|
/* Datagram follows the mbim header */
|
|
ndp16->dpe16[0].wDatagramIndex = cpu_to_le16(sizeof(struct mbim_tx_hdr));
|
|
ndp16->dpe16[0].wDatagramLength = cpu_to_le16(dgram_size);
|
|
|
|
/* null termination */
|
|
ndp16->dpe16[1].wDatagramIndex = 0;
|
|
ndp16->dpe16[1].wDatagramLength = 0;
|
|
|
|
return skb;
|
|
}
|
|
|
|
static netdev_tx_t mhi_mbim_ndo_xmit(struct sk_buff *skb, struct net_device *ndev)
|
|
{
|
|
struct mhi_mbim_link *link = wwan_netdev_drvpriv(ndev);
|
|
struct mhi_mbim_context *mbim = link->mbim;
|
|
unsigned long flags;
|
|
int err = -ENOMEM;
|
|
|
|
/* Serialize MHI channel queuing and MBIM seq */
|
|
spin_lock_irqsave(&mbim->tx_lock, flags);
|
|
|
|
skb = mbim_tx_fixup(skb, link->session, mbim->tx_seq);
|
|
if (unlikely(!skb))
|
|
goto exit_unlock;
|
|
|
|
err = mhi_queue_skb(mbim->mdev, DMA_TO_DEVICE, skb, skb->len, MHI_EOT);
|
|
|
|
if (mhi_queue_is_full(mbim->mdev, DMA_TO_DEVICE))
|
|
netif_stop_queue(ndev);
|
|
|
|
if (!err)
|
|
mbim->tx_seq++;
|
|
|
|
exit_unlock:
|
|
spin_unlock_irqrestore(&mbim->tx_lock, flags);
|
|
|
|
if (unlikely(err)) {
|
|
net_err_ratelimited("%s: Failed to queue TX buf (%d)\n",
|
|
ndev->name, err);
|
|
dev_kfree_skb_any(skb);
|
|
goto exit_drop;
|
|
}
|
|
|
|
return NETDEV_TX_OK;
|
|
|
|
exit_drop:
|
|
u64_stats_update_begin(&link->tx_syncp);
|
|
u64_stats_inc(&link->tx_dropped);
|
|
u64_stats_update_end(&link->tx_syncp);
|
|
|
|
return NETDEV_TX_OK;
|
|
}
|
|
|
|
static int mbim_rx_verify_nth16(struct mhi_mbim_context *mbim, struct sk_buff *skb)
|
|
{
|
|
struct usb_cdc_ncm_nth16 *nth16;
|
|
int len;
|
|
|
|
if (skb->len < sizeof(struct usb_cdc_ncm_nth16) +
|
|
sizeof(struct usb_cdc_ncm_ndp16)) {
|
|
net_err_ratelimited("frame too short\n");
|
|
return -EINVAL;
|
|
}
|
|
|
|
nth16 = (struct usb_cdc_ncm_nth16 *)skb->data;
|
|
|
|
if (nth16->dwSignature != cpu_to_le32(USB_CDC_NCM_NTH16_SIGN)) {
|
|
net_err_ratelimited("invalid NTH16 signature <%#010x>\n",
|
|
le32_to_cpu(nth16->dwSignature));
|
|
return -EINVAL;
|
|
}
|
|
|
|
/* No limit on the block length, except the size of the data pkt */
|
|
len = le16_to_cpu(nth16->wBlockLength);
|
|
if (len > skb->len) {
|
|
net_err_ratelimited("NTB does not fit into the skb %u/%u\n",
|
|
len, skb->len);
|
|
return -EINVAL;
|
|
}
|
|
|
|
if (mbim->rx_seq + 1 != le16_to_cpu(nth16->wSequence) &&
|
|
(mbim->rx_seq || le16_to_cpu(nth16->wSequence)) &&
|
|
!(mbim->rx_seq == 0xffff && !le16_to_cpu(nth16->wSequence))) {
|
|
net_err_ratelimited("sequence number glitch prev=%d curr=%d\n",
|
|
mbim->rx_seq, le16_to_cpu(nth16->wSequence));
|
|
}
|
|
mbim->rx_seq = le16_to_cpu(nth16->wSequence);
|
|
|
|
return le16_to_cpu(nth16->wNdpIndex);
|
|
}
|
|
|
|
static int mbim_rx_verify_ndp16(struct sk_buff *skb, struct usb_cdc_ncm_ndp16 *ndp16)
|
|
{
|
|
int ret;
|
|
|
|
if (le16_to_cpu(ndp16->wLength) < USB_CDC_NCM_NDP16_LENGTH_MIN) {
|
|
net_err_ratelimited("invalid DPT16 length <%u>\n",
|
|
le16_to_cpu(ndp16->wLength));
|
|
return -EINVAL;
|
|
}
|
|
|
|
ret = ((le16_to_cpu(ndp16->wLength) - sizeof(struct usb_cdc_ncm_ndp16))
|
|
/ sizeof(struct usb_cdc_ncm_dpe16));
|
|
ret--; /* Last entry is always a NULL terminator */
|
|
|
|
if (sizeof(struct usb_cdc_ncm_ndp16) +
|
|
ret * sizeof(struct usb_cdc_ncm_dpe16) > skb->len) {
|
|
net_err_ratelimited("Invalid nframes = %d\n", ret);
|
|
return -EINVAL;
|
|
}
|
|
|
|
return ret;
|
|
}
|
|
|
|
static void mhi_mbim_rx(struct mhi_mbim_context *mbim, struct sk_buff *skb)
|
|
{
|
|
int ndpoffset;
|
|
|
|
/* Check NTB header and retrieve first NDP offset */
|
|
ndpoffset = mbim_rx_verify_nth16(mbim, skb);
|
|
if (ndpoffset < 0) {
|
|
net_err_ratelimited("mbim: Incorrect NTB header\n");
|
|
goto error;
|
|
}
|
|
|
|
/* Process each NDP */
|
|
while (1) {
|
|
struct usb_cdc_ncm_ndp16 ndp16;
|
|
struct usb_cdc_ncm_dpe16 dpe16;
|
|
struct mhi_mbim_link *link;
|
|
int nframes, n, dpeoffset;
|
|
unsigned int session;
|
|
|
|
if (skb_copy_bits(skb, ndpoffset, &ndp16, sizeof(ndp16))) {
|
|
net_err_ratelimited("mbim: Incorrect NDP offset (%u)\n",
|
|
ndpoffset);
|
|
goto error;
|
|
}
|
|
|
|
/* Check NDP header and retrieve number of datagrams */
|
|
nframes = mbim_rx_verify_ndp16(skb, &ndp16);
|
|
if (nframes < 0) {
|
|
net_err_ratelimited("mbim: Incorrect NDP16\n");
|
|
goto error;
|
|
}
|
|
|
|
/* Only IP data type supported, no DSS in MHI context */
|
|
if ((ndp16.dwSignature & cpu_to_le32(MBIM_NDP16_SIGN_MASK))
|
|
!= cpu_to_le32(USB_CDC_MBIM_NDP16_IPS_SIGN)) {
|
|
net_err_ratelimited("mbim: Unsupported NDP type\n");
|
|
goto next_ndp;
|
|
}
|
|
|
|
session = (le32_to_cpu(ndp16.dwSignature) & ~MBIM_NDP16_SIGN_MASK) >> 24;
|
|
|
|
rcu_read_lock();
|
|
|
|
link = mhi_mbim_get_link_rcu(mbim, session);
|
|
if (!link) {
|
|
net_err_ratelimited("mbim: bad packet session (%u)\n", session);
|
|
goto unlock;
|
|
}
|
|
|
|
/* de-aggregate and deliver IP packets */
|
|
dpeoffset = ndpoffset + sizeof(struct usb_cdc_ncm_ndp16);
|
|
for (n = 0; n < nframes; n++, dpeoffset += sizeof(dpe16)) {
|
|
u16 dgram_offset, dgram_len;
|
|
struct sk_buff *skbn;
|
|
|
|
if (skb_copy_bits(skb, dpeoffset, &dpe16, sizeof(dpe16)))
|
|
break;
|
|
|
|
dgram_offset = le16_to_cpu(dpe16.wDatagramIndex);
|
|
dgram_len = le16_to_cpu(dpe16.wDatagramLength);
|
|
|
|
if (!dgram_offset || !dgram_len)
|
|
break; /* null terminator */
|
|
|
|
skbn = netdev_alloc_skb(link->ndev, dgram_len);
|
|
if (!skbn)
|
|
continue;
|
|
|
|
skb_put(skbn, dgram_len);
|
|
skb_copy_bits(skb, dgram_offset, skbn->data, dgram_len);
|
|
|
|
switch (skbn->data[0] & 0xf0) {
|
|
case 0x40:
|
|
skbn->protocol = htons(ETH_P_IP);
|
|
break;
|
|
case 0x60:
|
|
skbn->protocol = htons(ETH_P_IPV6);
|
|
break;
|
|
default:
|
|
net_err_ratelimited("%s: unknown protocol\n",
|
|
link->ndev->name);
|
|
dev_kfree_skb_any(skbn);
|
|
u64_stats_update_begin(&link->rx_syncp);
|
|
u64_stats_inc(&link->rx_errors);
|
|
u64_stats_update_end(&link->rx_syncp);
|
|
continue;
|
|
}
|
|
|
|
u64_stats_update_begin(&link->rx_syncp);
|
|
u64_stats_inc(&link->rx_packets);
|
|
u64_stats_add(&link->rx_bytes, skbn->len);
|
|
u64_stats_update_end(&link->rx_syncp);
|
|
|
|
netif_rx(skbn);
|
|
}
|
|
unlock:
|
|
rcu_read_unlock();
|
|
next_ndp:
|
|
/* Other NDP to process? */
|
|
ndpoffset = (int)le16_to_cpu(ndp16.wNextNdpIndex);
|
|
if (!ndpoffset)
|
|
break;
|
|
}
|
|
|
|
/* free skb */
|
|
dev_consume_skb_any(skb);
|
|
return;
|
|
error:
|
|
dev_kfree_skb_any(skb);
|
|
}
|
|
|
|
static struct sk_buff *mhi_net_skb_agg(struct mhi_mbim_context *mbim,
|
|
struct sk_buff *skb)
|
|
{
|
|
struct sk_buff *head = mbim->skbagg_head;
|
|
struct sk_buff *tail = mbim->skbagg_tail;
|
|
|
|
/* This is non-paged skb chaining using frag_list */
|
|
if (!head) {
|
|
mbim->skbagg_head = skb;
|
|
return skb;
|
|
}
|
|
|
|
if (!skb_shinfo(head)->frag_list)
|
|
skb_shinfo(head)->frag_list = skb;
|
|
else
|
|
tail->next = skb;
|
|
|
|
head->len += skb->len;
|
|
head->data_len += skb->len;
|
|
head->truesize += skb->truesize;
|
|
|
|
mbim->skbagg_tail = skb;
|
|
|
|
return mbim->skbagg_head;
|
|
}
|
|
|
|
static void mhi_net_rx_refill_work(struct work_struct *work)
|
|
{
|
|
struct mhi_mbim_context *mbim = container_of(work, struct mhi_mbim_context,
|
|
rx_refill.work);
|
|
struct mhi_device *mdev = mbim->mdev;
|
|
int err;
|
|
|
|
while (!mhi_queue_is_full(mdev, DMA_FROM_DEVICE)) {
|
|
struct sk_buff *skb = alloc_skb(mbim->mru, GFP_KERNEL);
|
|
|
|
if (unlikely(!skb))
|
|
break;
|
|
|
|
err = mhi_queue_skb(mdev, DMA_FROM_DEVICE, skb,
|
|
mbim->mru, MHI_EOT);
|
|
if (unlikely(err)) {
|
|
kfree_skb(skb);
|
|
break;
|
|
}
|
|
|
|
/* Do not hog the CPU if rx buffers are consumed faster than
|
|
* queued (unlikely).
|
|
*/
|
|
cond_resched();
|
|
}
|
|
|
|
/* If we're still starved of rx buffers, reschedule later */
|
|
if (mhi_get_free_desc_count(mdev, DMA_FROM_DEVICE) == mbim->rx_queue_sz)
|
|
schedule_delayed_work(&mbim->rx_refill, HZ / 2);
|
|
}
|
|
|
|
static void mhi_mbim_dl_callback(struct mhi_device *mhi_dev,
|
|
struct mhi_result *mhi_res)
|
|
{
|
|
struct mhi_mbim_context *mbim = dev_get_drvdata(&mhi_dev->dev);
|
|
struct sk_buff *skb = mhi_res->buf_addr;
|
|
int free_desc_count;
|
|
|
|
free_desc_count = mhi_get_free_desc_count(mhi_dev, DMA_FROM_DEVICE);
|
|
|
|
if (unlikely(mhi_res->transaction_status)) {
|
|
switch (mhi_res->transaction_status) {
|
|
case -EOVERFLOW:
|
|
/* Packet has been split over multiple transfers */
|
|
skb_put(skb, mhi_res->bytes_xferd);
|
|
mhi_net_skb_agg(mbim, skb);
|
|
break;
|
|
case -ENOTCONN:
|
|
/* MHI layer stopping/resetting the DL channel */
|
|
dev_kfree_skb_any(skb);
|
|
return;
|
|
default:
|
|
/* Unknown error, simply drop */
|
|
dev_kfree_skb_any(skb);
|
|
}
|
|
} else {
|
|
skb_put(skb, mhi_res->bytes_xferd);
|
|
|
|
if (mbim->skbagg_head) {
|
|
/* Aggregate the final fragment */
|
|
skb = mhi_net_skb_agg(mbim, skb);
|
|
mbim->skbagg_head = NULL;
|
|
}
|
|
|
|
mhi_mbim_rx(mbim, skb);
|
|
}
|
|
|
|
/* Refill if RX buffers queue becomes low */
|
|
if (free_desc_count >= mbim->rx_queue_sz / 2)
|
|
schedule_delayed_work(&mbim->rx_refill, 0);
|
|
}
|
|
|
|
static void mhi_mbim_ndo_get_stats64(struct net_device *ndev,
|
|
struct rtnl_link_stats64 *stats)
|
|
{
|
|
struct mhi_mbim_link *link = wwan_netdev_drvpriv(ndev);
|
|
unsigned int start;
|
|
|
|
do {
|
|
start = u64_stats_fetch_begin(&link->rx_syncp);
|
|
stats->rx_packets = u64_stats_read(&link->rx_packets);
|
|
stats->rx_bytes = u64_stats_read(&link->rx_bytes);
|
|
stats->rx_errors = u64_stats_read(&link->rx_errors);
|
|
} while (u64_stats_fetch_retry(&link->rx_syncp, start));
|
|
|
|
do {
|
|
start = u64_stats_fetch_begin(&link->tx_syncp);
|
|
stats->tx_packets = u64_stats_read(&link->tx_packets);
|
|
stats->tx_bytes = u64_stats_read(&link->tx_bytes);
|
|
stats->tx_errors = u64_stats_read(&link->tx_errors);
|
|
stats->tx_dropped = u64_stats_read(&link->tx_dropped);
|
|
} while (u64_stats_fetch_retry(&link->tx_syncp, start));
|
|
}
|
|
|
|
static void mhi_mbim_ul_callback(struct mhi_device *mhi_dev,
|
|
struct mhi_result *mhi_res)
|
|
{
|
|
struct mhi_mbim_context *mbim = dev_get_drvdata(&mhi_dev->dev);
|
|
struct sk_buff *skb = mhi_res->buf_addr;
|
|
struct net_device *ndev = skb->dev;
|
|
struct mhi_mbim_link *link = wwan_netdev_drvpriv(ndev);
|
|
|
|
/* Hardware has consumed the buffer, so free the skb (which is not
|
|
* freed by the MHI stack) and perform accounting.
|
|
*/
|
|
dev_consume_skb_any(skb);
|
|
|
|
u64_stats_update_begin(&link->tx_syncp);
|
|
if (unlikely(mhi_res->transaction_status)) {
|
|
/* MHI layer stopping/resetting the UL channel */
|
|
if (mhi_res->transaction_status == -ENOTCONN) {
|
|
u64_stats_update_end(&link->tx_syncp);
|
|
return;
|
|
}
|
|
|
|
u64_stats_inc(&link->tx_errors);
|
|
} else {
|
|
u64_stats_inc(&link->tx_packets);
|
|
u64_stats_add(&link->tx_bytes, mhi_res->bytes_xferd);
|
|
}
|
|
u64_stats_update_end(&link->tx_syncp);
|
|
|
|
if (netif_queue_stopped(ndev) && !mhi_queue_is_full(mbim->mdev, DMA_TO_DEVICE))
|
|
netif_wake_queue(ndev);
|
|
}
|
|
|
|
static int mhi_mbim_ndo_open(struct net_device *ndev)
|
|
{
|
|
struct mhi_mbim_link *link = wwan_netdev_drvpriv(ndev);
|
|
|
|
/* Feed the MHI rx buffer pool */
|
|
schedule_delayed_work(&link->mbim->rx_refill, 0);
|
|
|
|
/* Carrier is established via out-of-band channel (e.g. qmi) */
|
|
netif_carrier_on(ndev);
|
|
|
|
netif_start_queue(ndev);
|
|
|
|
return 0;
|
|
}
|
|
|
|
static int mhi_mbim_ndo_stop(struct net_device *ndev)
|
|
{
|
|
netif_stop_queue(ndev);
|
|
netif_carrier_off(ndev);
|
|
|
|
return 0;
|
|
}
|
|
|
|
static const struct net_device_ops mhi_mbim_ndo = {
|
|
.ndo_open = mhi_mbim_ndo_open,
|
|
.ndo_stop = mhi_mbim_ndo_stop,
|
|
.ndo_start_xmit = mhi_mbim_ndo_xmit,
|
|
.ndo_get_stats64 = mhi_mbim_ndo_get_stats64,
|
|
};
|
|
|
|
static int mhi_mbim_newlink(void *ctxt, struct net_device *ndev, u32 if_id,
|
|
struct netlink_ext_ack *extack)
|
|
{
|
|
struct mhi_mbim_link *link = wwan_netdev_drvpriv(ndev);
|
|
struct mhi_mbim_context *mbim = ctxt;
|
|
|
|
link->session = if_id;
|
|
link->mbim = mbim;
|
|
link->ndev = ndev;
|
|
u64_stats_init(&link->rx_syncp);
|
|
u64_stats_init(&link->tx_syncp);
|
|
|
|
rcu_read_lock();
|
|
if (mhi_mbim_get_link_rcu(mbim, if_id)) {
|
|
rcu_read_unlock();
|
|
return -EEXIST;
|
|
}
|
|
rcu_read_unlock();
|
|
|
|
/* Already protected by RTNL lock */
|
|
hlist_add_head_rcu(&link->hlnode, &mbim->link_list[LINK_HASH(if_id)]);
|
|
|
|
return register_netdevice(ndev);
|
|
}
|
|
|
|
static void mhi_mbim_dellink(void *ctxt, struct net_device *ndev,
|
|
struct list_head *head)
|
|
{
|
|
struct mhi_mbim_link *link = wwan_netdev_drvpriv(ndev);
|
|
|
|
hlist_del_init_rcu(&link->hlnode);
|
|
synchronize_rcu();
|
|
|
|
unregister_netdevice_queue(ndev, head);
|
|
}
|
|
|
|
static void mhi_mbim_setup(struct net_device *ndev)
|
|
{
|
|
ndev->header_ops = NULL; /* No header */
|
|
ndev->type = ARPHRD_RAWIP;
|
|
ndev->needed_headroom = sizeof(struct mbim_tx_hdr);
|
|
ndev->hard_header_len = 0;
|
|
ndev->addr_len = 0;
|
|
ndev->flags = IFF_POINTOPOINT | IFF_NOARP;
|
|
ndev->netdev_ops = &mhi_mbim_ndo;
|
|
ndev->mtu = MHI_MBIM_DEFAULT_MTU;
|
|
ndev->min_mtu = ETH_MIN_MTU;
|
|
ndev->max_mtu = MHI_MAX_BUF_SZ - ndev->needed_headroom;
|
|
ndev->tx_queue_len = 1000;
|
|
ndev->needs_free_netdev = true;
|
|
}
|
|
|
|
static const struct wwan_ops mhi_mbim_wwan_ops = {
|
|
.priv_size = sizeof(struct mhi_mbim_link),
|
|
.setup = mhi_mbim_setup,
|
|
.newlink = mhi_mbim_newlink,
|
|
.dellink = mhi_mbim_dellink,
|
|
};
|
|
|
|
static int mhi_mbim_probe(struct mhi_device *mhi_dev, const struct mhi_device_id *id)
|
|
{
|
|
struct mhi_controller *cntrl = mhi_dev->mhi_cntrl;
|
|
struct mhi_mbim_context *mbim;
|
|
int err;
|
|
|
|
mbim = devm_kzalloc(&mhi_dev->dev, sizeof(*mbim), GFP_KERNEL);
|
|
if (!mbim)
|
|
return -ENOMEM;
|
|
|
|
spin_lock_init(&mbim->tx_lock);
|
|
dev_set_drvdata(&mhi_dev->dev, mbim);
|
|
mbim->mdev = mhi_dev;
|
|
mbim->mru = mhi_dev->mhi_cntrl->mru ? mhi_dev->mhi_cntrl->mru : MHI_DEFAULT_MRU;
|
|
|
|
INIT_DELAYED_WORK(&mbim->rx_refill, mhi_net_rx_refill_work);
|
|
|
|
/* Start MHI channels */
|
|
err = mhi_prepare_for_transfer(mhi_dev);
|
|
if (err)
|
|
return err;
|
|
|
|
/* Number of transfer descriptors determines size of the queue */
|
|
mbim->rx_queue_sz = mhi_get_free_desc_count(mhi_dev, DMA_FROM_DEVICE);
|
|
|
|
/* Register wwan link ops with MHI controller representing WWAN instance */
|
|
return wwan_register_ops(&cntrl->mhi_dev->dev, &mhi_mbim_wwan_ops, mbim, 0);
|
|
}
|
|
|
|
static void mhi_mbim_remove(struct mhi_device *mhi_dev)
|
|
{
|
|
struct mhi_mbim_context *mbim = dev_get_drvdata(&mhi_dev->dev);
|
|
struct mhi_controller *cntrl = mhi_dev->mhi_cntrl;
|
|
|
|
mhi_unprepare_from_transfer(mhi_dev);
|
|
cancel_delayed_work_sync(&mbim->rx_refill);
|
|
wwan_unregister_ops(&cntrl->mhi_dev->dev);
|
|
kfree_skb(mbim->skbagg_head);
|
|
dev_set_drvdata(&mhi_dev->dev, NULL);
|
|
}
|
|
|
|
static const struct mhi_device_id mhi_mbim_id_table[] = {
|
|
/* Hardware accelerated data PATH (to modem IPA), MBIM protocol */
|
|
{ .chan = "IP_HW0_MBIM", .driver_data = 0 },
|
|
{}
|
|
};
|
|
MODULE_DEVICE_TABLE(mhi, mhi_mbim_id_table);
|
|
|
|
static struct mhi_driver mhi_mbim_driver = {
|
|
.probe = mhi_mbim_probe,
|
|
.remove = mhi_mbim_remove,
|
|
.dl_xfer_cb = mhi_mbim_dl_callback,
|
|
.ul_xfer_cb = mhi_mbim_ul_callback,
|
|
.id_table = mhi_mbim_id_table,
|
|
.driver = {
|
|
.name = "mhi_wwan_mbim",
|
|
.owner = THIS_MODULE,
|
|
},
|
|
};
|
|
|
|
module_mhi_driver(mhi_mbim_driver);
|
|
|
|
MODULE_AUTHOR("Loic Poulain <loic.poulain@linaro.org>");
|
|
MODULE_DESCRIPTION("Network/MBIM over MHI");
|
|
MODULE_LICENSE("GPL v2");
|