summaryrefslogtreecommitdiffstats
path: root/kernel/drivers/net/wireless/ath/wil6210/txrx.c
diff options
context:
space:
mode:
authorJosé Pekkarinen <jose.pekkarinen@nokia.com>2016-04-11 10:41:07 +0300
committerJosé Pekkarinen <jose.pekkarinen@nokia.com>2016-04-13 08:17:18 +0300
commite09b41010ba33a20a87472ee821fa407a5b8da36 (patch)
treed10dc367189862e7ca5c592f033dc3726e1df4e3 /kernel/drivers/net/wireless/ath/wil6210/txrx.c
parentf93b97fd65072de626c074dbe099a1fff05ce060 (diff)
These changes are the raw update to linux-4.4.6-rt14. Kernel sources
are taken from kernel.org, and rt patch from the rt wiki download page. During the rebasing, the following patch collided: Force tick interrupt and get rid of softirq magic(I70131fb85). Collisions have been removed because its logic was found on the source already. Change-Id: I7f57a4081d9deaa0d9ccfc41a6c8daccdee3b769 Signed-off-by: José Pekkarinen <jose.pekkarinen@nokia.com>
Diffstat (limited to 'kernel/drivers/net/wireless/ath/wil6210/txrx.c')
-rw-r--r--kernel/drivers/net/wireless/ath/wil6210/txrx.c503
1 files changed, 450 insertions, 53 deletions
diff --git a/kernel/drivers/net/wireless/ath/wil6210/txrx.c b/kernel/drivers/net/wireless/ath/wil6210/txrx.c
index e8bd512d8..3bc9bc0ef 100644
--- a/kernel/drivers/net/wireless/ath/wil6210/txrx.c
+++ b/kernel/drivers/net/wireless/ath/wil6210/txrx.c
@@ -1,5 +1,5 @@
/*
- * Copyright (c) 2012-2014 Qualcomm Atheros, Inc.
+ * Copyright (c) 2012-2015 Qualcomm Atheros, Inc.
*
* Permission to use, copy, modify, and/or distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
@@ -236,7 +236,7 @@ static int wil_vring_alloc_skb(struct wil6210_priv *wil, struct vring *vring,
return -ENOMEM;
}
- d->dma.d0 = BIT(9) | RX_DMA_D0_CMD_DMA_IT;
+ d->dma.d0 = RX_DMA_D0_CMD_DMA_RT | RX_DMA_D0_CMD_DMA_IT;
wil_desc_addr_set(&d->dma.addr, pa);
/* ip_length don't care */
/* b11 don't care */
@@ -358,6 +358,13 @@ static void wil_rx_add_radiotap_header(struct wil6210_priv *wil,
}
}
+/* similar to ieee80211_ version, but FC contain only 1-st byte */
+static inline int wil_is_back_req(u8 fc)
+{
+ return (fc & (IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+ (IEEE80211_FTYPE_CTL | IEEE80211_STYPE_BACK_REQ);
+}
+
/**
* reap 1 frame from @swhead
*
@@ -379,14 +386,16 @@ static struct sk_buff *wil_vring_reap_rx(struct wil6210_priv *wil,
u16 dmalen;
u8 ftype;
int cid;
- int i = (int)vring->swhead;
+ int i;
struct wil_net_stats *stats;
BUILD_BUG_ON(sizeof(struct vring_rx_desc) > sizeof(skb->cb));
+again:
if (unlikely(wil_vring_is_empty(vring)))
return NULL;
+ i = (int)vring->swhead;
_d = &vring->va[i].rx;
if (unlikely(!(_d->dma.status & RX_DMA_STATUS_DU))) {
/* it is not error, we just reached end of Rx done area */
@@ -398,7 +407,7 @@ static struct sk_buff *wil_vring_reap_rx(struct wil6210_priv *wil,
wil_vring_advance_head(vring, 1);
if (!skb) {
wil_err(wil, "No Rx skb at [%d]\n", i);
- return NULL;
+ goto again;
}
d = wil_skb_rxdesc(skb);
*d = *_d;
@@ -409,13 +418,17 @@ static struct sk_buff *wil_vring_reap_rx(struct wil6210_priv *wil,
trace_wil6210_rx(i, d);
wil_dbg_txrx(wil, "Rx[%3d] : %d bytes\n", i, dmalen);
- wil_hex_dump_txrx("Rx ", DUMP_PREFIX_NONE, 32, 4,
+ wil_hex_dump_txrx("RxD ", DUMP_PREFIX_NONE, 32, 4,
(const void *)d, sizeof(*d), false);
+ cid = wil_rxdesc_cid(d);
+ stats = &wil->sta[cid].stats;
+
if (unlikely(dmalen > sz)) {
wil_err(wil, "Rx size too large: %d bytes!\n", dmalen);
+ stats->rx_large_frame++;
kfree_skb(skb);
- return NULL;
+ goto again;
}
skb_trim(skb, dmalen);
@@ -424,9 +437,9 @@ static struct sk_buff *wil_vring_reap_rx(struct wil6210_priv *wil,
wil_hex_dump_txrx("Rx ", DUMP_PREFIX_OFFSET, 16, 1,
skb->data, skb_headlen(skb), false);
- cid = wil_rxdesc_cid(d);
- stats = &wil->sta[cid].stats;
stats->last_mcs_rx = wil_rxdesc_mcs(d);
+ if (stats->last_mcs_rx < ARRAY_SIZE(stats->rx_per_mcs))
+ stats->rx_per_mcs[stats->last_mcs_rx]++;
/* use radiotap header only if required */
if (ndev->type == ARPHRD_IEEE80211_RADIOTAP)
@@ -435,24 +448,47 @@ static struct sk_buff *wil_vring_reap_rx(struct wil6210_priv *wil,
/* no extra checks if in sniffer mode */
if (ndev->type != ARPHRD_ETHER)
return skb;
- /*
- * Non-data frames may be delivered through Rx DMA channel (ex: BAR)
+ /* Non-data frames may be delivered through Rx DMA channel (ex: BAR)
* Driver should recognize it by frame type, that is found
* in Rx descriptor. If type is not data, it is 802.11 frame as is
*/
ftype = wil_rxdesc_ftype(d) << 2;
if (unlikely(ftype != IEEE80211_FTYPE_DATA)) {
- wil_dbg_txrx(wil, "Non-data frame ftype 0x%08x\n", ftype);
- /* TODO: process it */
+ u8 fc1 = wil_rxdesc_fc1(d);
+ int mid = wil_rxdesc_mid(d);
+ int tid = wil_rxdesc_tid(d);
+ u16 seq = wil_rxdesc_seq(d);
+
+ wil_dbg_txrx(wil,
+ "Non-data frame FC[7:0] 0x%02x MID %d CID %d TID %d Seq 0x%03x\n",
+ fc1, mid, cid, tid, seq);
+ stats->rx_non_data_frame++;
+ if (wil_is_back_req(fc1)) {
+ wil_dbg_txrx(wil,
+ "BAR: MID %d CID %d TID %d Seq 0x%03x\n",
+ mid, cid, tid, seq);
+ wil_rx_bar(wil, cid, tid, seq);
+ } else {
+ /* print again all info. One can enable only this
+ * without overhead for printing every Rx frame
+ */
+ wil_dbg_txrx(wil,
+ "Unhandled non-data frame FC[7:0] 0x%02x MID %d CID %d TID %d Seq 0x%03x\n",
+ fc1, mid, cid, tid, seq);
+ wil_hex_dump_txrx("RxD ", DUMP_PREFIX_NONE, 32, 4,
+ (const void *)d, sizeof(*d), false);
+ wil_hex_dump_txrx("Rx ", DUMP_PREFIX_OFFSET, 16, 1,
+ skb->data, skb_headlen(skb), false);
+ }
kfree_skb(skb);
- return NULL;
+ goto again;
}
if (unlikely(skb->len < ETH_HLEN + snaplen)) {
wil_err(wil, "Short frame, len = %d\n", skb->len);
- /* TODO: process it (i.e. BAR) */
+ stats->rx_short_frame++;
kfree_skb(skb);
- return NULL;
+ goto again;
}
/* L4 IDENT is on when HW calculated checksum, check status
@@ -507,7 +543,7 @@ static int wil_rx_refill(struct wil6210_priv *wil, int count)
break;
}
}
- iowrite32(v->swtail, wil->csr + HOSTADDR(v->hwtail));
+ wil_w(wil, v->hwtail, v->swtail);
return rc;
}
@@ -539,6 +575,14 @@ void wil_netif_rx_any(struct sk_buff *skb, struct net_device *ndev)
[GRO_DROP] = "GRO_DROP",
};
+ if (ndev->features & NETIF_F_RXHASH)
+ /* fake L4 to ensure it won't be re-calculated later
+ * set hash to any non-zero value to activate rps
+ * mechanism, core will be chosen according
+ * to user-level rps configuration.
+ */
+ skb_set_hash(skb, 1, PKT_HASH_TYPE_L4);
+
skb_orphan(skb);
if (wdev->iftype == NL80211_IFTYPE_AP && !wil->ap_isolate) {
@@ -724,6 +768,8 @@ int wil_vring_init_tx(struct wil6210_priv *wil, int id, int size,
cmd.vring_cfg.tx_sw_ring.ring_mem_base = cpu_to_le64(vring->pa);
+ if (!wil->privacy)
+ txdata->dot1x_open = true;
rc = wmi_call(wil, WMI_VRING_CFG_CMDID, &cmd, sizeof(cmd),
WMI_VRING_CFG_DONE_EVENTID, &reply, sizeof(reply), 100);
if (rc)
@@ -738,11 +784,13 @@ int wil_vring_init_tx(struct wil6210_priv *wil, int id, int size,
vring->hwtail = le32_to_cpu(reply.cmd.tx_vring_tail_ptr);
txdata->enabled = 1;
- if (wil->sta[cid].data_port_open && (agg_wsize >= 0))
+ if (txdata->dot1x_open && (agg_wsize >= 0))
wil_addba_tx_request(wil, id, agg_wsize);
return 0;
out_free:
+ txdata->dot1x_open = false;
+ txdata->enabled = 0;
wil_vring_free(wil, vring, 1);
out:
@@ -792,6 +840,8 @@ int wil_vring_init_bcast(struct wil6210_priv *wil, int id, int size)
cmd.vring_cfg.tx_sw_ring.ring_mem_base = cpu_to_le64(vring->pa);
+ if (!wil->privacy)
+ txdata->dot1x_open = true;
rc = wmi_call(wil, WMI_BCAST_VRING_CFG_CMDID, &cmd, sizeof(cmd),
WMI_VRING_CFG_DONE_EVENTID, &reply, sizeof(reply), 100);
if (rc)
@@ -809,6 +859,8 @@ int wil_vring_init_bcast(struct wil6210_priv *wil, int id, int size)
return 0;
out_free:
+ txdata->enabled = 0;
+ txdata->dot1x_open = false;
wil_vring_free(wil, vring, 1);
out:
@@ -828,6 +880,7 @@ void wil_vring_fini_tx(struct wil6210_priv *wil, int id)
wil_dbg_misc(wil, "%s() id=%d\n", __func__, id);
spin_lock_bh(&txdata->lock);
+ txdata->dot1x_open = false;
txdata->enabled = 0; /* no Tx can be in progress or start anew */
spin_unlock_bh(&txdata->lock);
/* make sure NAPI won't touch this vring */
@@ -848,12 +901,11 @@ static struct vring *wil_find_tx_ucast(struct wil6210_priv *wil,
if (cid < 0)
return NULL;
- if (!wil->sta[cid].data_port_open &&
- (skb->protocol != cpu_to_be16(ETH_P_PAE)))
- return NULL;
-
/* TODO: fix for multiple TID */
for (i = 0; i < ARRAY_SIZE(wil->vring2cid_tid); i++) {
+ if (!wil->vring_tx_data[i].dot1x_open &&
+ (skb->protocol != cpu_to_be16(ETH_P_PAE)))
+ continue;
if (wil->vring2cid_tid[i][0] == cid) {
struct vring *v = &wil->vring_tx[i];
@@ -883,7 +935,7 @@ static struct vring *wil_find_tx_vring_sta(struct wil6210_priv *wil,
/* In the STA mode, it is expected to have only 1 VRING
* for the AP we connected to.
- * find 1-st vring and see whether it is eligible for data
+ * find 1-st vring eligible for this skb and use it.
*/
for (i = 0; i < WIL6210_MAX_TX_RINGS; i++) {
v = &wil->vring_tx[i];
@@ -894,9 +946,9 @@ static struct vring *wil_find_tx_vring_sta(struct wil6210_priv *wil,
if (cid >= WIL6210_MAX_CID) /* skip BCAST */
continue;
- if (!wil->sta[cid].data_port_open &&
+ if (!wil->vring_tx_data[i].dot1x_open &&
(skb->protocol != cpu_to_be16(ETH_P_PAE)))
- break;
+ continue;
wil_dbg_txrx(wil, "Tx -> ring %d\n", i);
@@ -918,7 +970,6 @@ static struct vring *wil_find_tx_vring_sta(struct wil6210_priv *wil,
* in all cases override dest address to unicast peer's address
* Use old strategy when new is not supported yet:
* - for PBSS
- * - for secure link
*/
static struct vring *wil_find_tx_bcast_1(struct wil6210_priv *wil,
struct sk_buff *skb)
@@ -931,6 +982,9 @@ static struct vring *wil_find_tx_bcast_1(struct wil6210_priv *wil,
v = &wil->vring_tx[i];
if (!v->va)
return NULL;
+ if (!wil->vring_tx_data[i].dot1x_open &&
+ (skb->protocol != cpu_to_be16(ETH_P_PAE)))
+ return NULL;
return v;
}
@@ -963,7 +1017,8 @@ static struct vring *wil_find_tx_bcast_2(struct wil6210_priv *wil,
cid = wil->vring2cid_tid[i][0];
if (cid >= WIL6210_MAX_CID) /* skip BCAST */
continue;
- if (!wil->sta[cid].data_port_open)
+ if (!wil->vring_tx_data[i].dot1x_open &&
+ (skb->protocol != cpu_to_be16(ETH_P_PAE)))
continue;
/* don't Tx back to source when re-routing Rx->Tx at the AP */
@@ -989,7 +1044,8 @@ found:
cid = wil->vring2cid_tid[i][0];
if (cid >= WIL6210_MAX_CID) /* skip BCAST */
continue;
- if (!wil->sta[cid].data_port_open)
+ if (!wil->vring_tx_data[i].dot1x_open &&
+ (skb->protocol != cpu_to_be16(ETH_P_PAE)))
continue;
if (0 == memcmp(wil->sta[cid].addr, src, ETH_ALEN))
@@ -1016,9 +1072,6 @@ static struct vring *wil_find_tx_bcast(struct wil6210_priv *wil,
if (wdev->iftype != NL80211_IFTYPE_AP)
return wil_find_tx_bcast_2(wil, skb);
- if (wil->privacy)
- return wil_find_tx_bcast_2(wil, skb);
-
return wil_find_tx_bcast_1(wil, skb);
}
@@ -1047,14 +1100,52 @@ static int wil_tx_desc_map(struct vring_tx_desc *d, dma_addr_t pa, u32 len,
static inline
void wil_tx_desc_set_nr_frags(struct vring_tx_desc *d, int nr_frags)
{
- d->mac.d[2] |= ((nr_frags + 1) <<
- MAC_CFG_DESC_TX_2_NUM_OF_DESCRIPTORS_POS);
+ d->mac.d[2] |= (nr_frags << MAC_CFG_DESC_TX_2_NUM_OF_DESCRIPTORS_POS);
}
-static int wil_tx_desc_offload_cksum_set(struct wil6210_priv *wil,
- struct vring_tx_desc *d,
- struct sk_buff *skb)
+/**
+ * Sets the descriptor @d up for csum and/or TSO offloading. The corresponding
+ * @skb is used to obtain the protocol and headers length.
+ * @tso_desc_type is a descriptor type for TSO: 0 - a header, 1 - first data,
+ * 2 - middle, 3 - last descriptor.
+ */
+
+static void wil_tx_desc_offload_setup_tso(struct vring_tx_desc *d,
+ struct sk_buff *skb,
+ int tso_desc_type, bool is_ipv4,
+ int tcp_hdr_len, int skb_net_hdr_len)
{
+ d->dma.b11 = ETH_HLEN; /* MAC header length */
+ d->dma.b11 |= is_ipv4 << DMA_CFG_DESC_TX_OFFLOAD_CFG_L3T_IPV4_POS;
+
+ d->dma.d0 |= (2 << DMA_CFG_DESC_TX_0_L4_TYPE_POS);
+ /* L4 header len: TCP header length */
+ d->dma.d0 |= (tcp_hdr_len & DMA_CFG_DESC_TX_0_L4_LENGTH_MSK);
+
+ /* Setup TSO: bit and desc type */
+ d->dma.d0 |= (BIT(DMA_CFG_DESC_TX_0_TCP_SEG_EN_POS)) |
+ (tso_desc_type << DMA_CFG_DESC_TX_0_SEGMENT_BUF_DETAILS_POS);
+ d->dma.d0 |= (is_ipv4 << DMA_CFG_DESC_TX_0_IPV4_CHECKSUM_EN_POS);
+
+ d->dma.ip_length = skb_net_hdr_len;
+ /* Enable TCP/UDP checksum */
+ d->dma.d0 |= BIT(DMA_CFG_DESC_TX_0_TCP_UDP_CHECKSUM_EN_POS);
+ /* Calculate pseudo-header */
+ d->dma.d0 |= BIT(DMA_CFG_DESC_TX_0_PSEUDO_HEADER_CALC_EN_POS);
+}
+
+/**
+ * Sets the descriptor @d up for csum. The corresponding
+ * @skb is used to obtain the protocol and headers length.
+ * Returns the protocol: 0 - not TCP, 1 - TCPv4, 2 - TCPv6.
+ * Note, if d==NULL, the function only returns the protocol result.
+ *
+ * It is very similar to previous wil_tx_desc_offload_setup_tso. This
+ * is "if unrolling" to optimize the critical path.
+ */
+
+static int wil_tx_desc_offload_setup(struct vring_tx_desc *d,
+ struct sk_buff *skb){
int protocol;
if (skb->ip_summed != CHECKSUM_PARTIAL)
@@ -1099,6 +1190,305 @@ static int wil_tx_desc_offload_cksum_set(struct wil6210_priv *wil,
return 0;
}
+static inline void wil_tx_last_desc(struct vring_tx_desc *d)
+{
+ d->dma.d0 |= BIT(DMA_CFG_DESC_TX_0_CMD_EOP_POS) |
+ BIT(DMA_CFG_DESC_TX_0_CMD_MARK_WB_POS) |
+ BIT(DMA_CFG_DESC_TX_0_CMD_DMA_IT_POS);
+}
+
+static inline void wil_set_tx_desc_last_tso(volatile struct vring_tx_desc *d)
+{
+ d->dma.d0 |= wil_tso_type_lst <<
+ DMA_CFG_DESC_TX_0_SEGMENT_BUF_DETAILS_POS;
+}
+
+static int __wil_tx_vring_tso(struct wil6210_priv *wil, struct vring *vring,
+ struct sk_buff *skb)
+{
+ struct device *dev = wil_to_dev(wil);
+
+ /* point to descriptors in shared memory */
+ volatile struct vring_tx_desc *_desc = NULL, *_hdr_desc,
+ *_first_desc = NULL;
+
+ /* pointers to shadow descriptors */
+ struct vring_tx_desc desc_mem, hdr_desc_mem, first_desc_mem,
+ *d = &hdr_desc_mem, *hdr_desc = &hdr_desc_mem,
+ *first_desc = &first_desc_mem;
+
+ /* pointer to shadow descriptors' context */
+ struct wil_ctx *hdr_ctx, *first_ctx = NULL;
+
+ int descs_used = 0; /* total number of used descriptors */
+ int sg_desc_cnt = 0; /* number of descriptors for current mss*/
+
+ u32 swhead = vring->swhead;
+ int used, avail = wil_vring_avail_tx(vring);
+ int nr_frags = skb_shinfo(skb)->nr_frags;
+ int min_desc_required = nr_frags + 1;
+ int mss = skb_shinfo(skb)->gso_size; /* payload size w/o headers */
+ int f, len, hdrlen, headlen;
+ int vring_index = vring - wil->vring_tx;
+ struct vring_tx_data *txdata = &wil->vring_tx_data[vring_index];
+ uint i = swhead;
+ dma_addr_t pa;
+ const skb_frag_t *frag = NULL;
+ int rem_data = mss;
+ int lenmss;
+ int hdr_compensation_need = true;
+ int desc_tso_type = wil_tso_type_first;
+ bool is_ipv4;
+ int tcp_hdr_len;
+ int skb_net_hdr_len;
+ int gso_type;
+ int rc = -EINVAL;
+
+ wil_dbg_txrx(wil, "%s() %d bytes to vring %d\n",
+ __func__, skb->len, vring_index);
+
+ if (unlikely(!txdata->enabled))
+ return -EINVAL;
+
+ /* A typical page 4K is 3-4 payloads, we assume each fragment
+ * is a full payload, that's how min_desc_required has been
+ * calculated. In real we might need more or less descriptors,
+ * this is the initial check only.
+ */
+ if (unlikely(avail < min_desc_required)) {
+ wil_err_ratelimited(wil,
+ "TSO: Tx ring[%2d] full. No space for %d fragments\n",
+ vring_index, min_desc_required);
+ return -ENOMEM;
+ }
+
+ /* Header Length = MAC header len + IP header len + TCP header len*/
+ hdrlen = ETH_HLEN +
+ (int)skb_network_header_len(skb) +
+ tcp_hdrlen(skb);
+
+ gso_type = skb_shinfo(skb)->gso_type & (SKB_GSO_TCPV6 | SKB_GSO_TCPV4);
+ switch (gso_type) {
+ case SKB_GSO_TCPV4:
+ /* TCP v4, zero out the IP length and IPv4 checksum fields
+ * as required by the offloading doc
+ */
+ ip_hdr(skb)->tot_len = 0;
+ ip_hdr(skb)->check = 0;
+ is_ipv4 = true;
+ break;
+ case SKB_GSO_TCPV6:
+ /* TCP v6, zero out the payload length */
+ ipv6_hdr(skb)->payload_len = 0;
+ is_ipv4 = false;
+ break;
+ default:
+ /* other than TCPv4 or TCPv6 types are not supported for TSO.
+ * It is also illegal for both to be set simultaneously
+ */
+ return -EINVAL;
+ }
+
+ if (skb->ip_summed != CHECKSUM_PARTIAL)
+ return -EINVAL;
+
+ /* tcp header length and skb network header length are fixed for all
+ * packet's descriptors - read then once here
+ */
+ tcp_hdr_len = tcp_hdrlen(skb);
+ skb_net_hdr_len = skb_network_header_len(skb);
+
+ _hdr_desc = &vring->va[i].tx;
+
+ pa = dma_map_single(dev, skb->data, hdrlen, DMA_TO_DEVICE);
+ if (unlikely(dma_mapping_error(dev, pa))) {
+ wil_err(wil, "TSO: Skb head DMA map error\n");
+ goto err_exit;
+ }
+
+ wil_tx_desc_map(hdr_desc, pa, hdrlen, vring_index);
+ wil_tx_desc_offload_setup_tso(hdr_desc, skb, wil_tso_type_hdr, is_ipv4,
+ tcp_hdr_len, skb_net_hdr_len);
+ wil_tx_last_desc(hdr_desc);
+
+ vring->ctx[i].mapped_as = wil_mapped_as_single;
+ hdr_ctx = &vring->ctx[i];
+
+ descs_used++;
+ headlen = skb_headlen(skb) - hdrlen;
+
+ for (f = headlen ? -1 : 0; f < nr_frags; f++) {
+ if (headlen) {
+ len = headlen;
+ wil_dbg_txrx(wil, "TSO: process skb head, len %u\n",
+ len);
+ } else {
+ frag = &skb_shinfo(skb)->frags[f];
+ len = frag->size;
+ wil_dbg_txrx(wil, "TSO: frag[%d]: len %u\n", f, len);
+ }
+
+ while (len) {
+ wil_dbg_txrx(wil,
+ "TSO: len %d, rem_data %d, descs_used %d\n",
+ len, rem_data, descs_used);
+
+ if (descs_used == avail) {
+ wil_err_ratelimited(wil, "TSO: ring overflow\n");
+ rc = -ENOMEM;
+ goto mem_error;
+ }
+
+ lenmss = min_t(int, rem_data, len);
+ i = (swhead + descs_used) % vring->size;
+ wil_dbg_txrx(wil, "TSO: lenmss %d, i %d\n", lenmss, i);
+
+ if (!headlen) {
+ pa = skb_frag_dma_map(dev, frag,
+ frag->size - len, lenmss,
+ DMA_TO_DEVICE);
+ vring->ctx[i].mapped_as = wil_mapped_as_page;
+ } else {
+ pa = dma_map_single(dev,
+ skb->data +
+ skb_headlen(skb) - headlen,
+ lenmss,
+ DMA_TO_DEVICE);
+ vring->ctx[i].mapped_as = wil_mapped_as_single;
+ headlen -= lenmss;
+ }
+
+ if (unlikely(dma_mapping_error(dev, pa))) {
+ wil_err(wil, "TSO: DMA map page error\n");
+ goto mem_error;
+ }
+
+ _desc = &vring->va[i].tx;
+
+ if (!_first_desc) {
+ _first_desc = _desc;
+ first_ctx = &vring->ctx[i];
+ d = first_desc;
+ } else {
+ d = &desc_mem;
+ }
+
+ wil_tx_desc_map(d, pa, lenmss, vring_index);
+ wil_tx_desc_offload_setup_tso(d, skb, desc_tso_type,
+ is_ipv4, tcp_hdr_len,
+ skb_net_hdr_len);
+
+ /* use tso_type_first only once */
+ desc_tso_type = wil_tso_type_mid;
+
+ descs_used++; /* desc used so far */
+ sg_desc_cnt++; /* desc used for this segment */
+ len -= lenmss;
+ rem_data -= lenmss;
+
+ wil_dbg_txrx(wil,
+ "TSO: len %d, rem_data %d, descs_used %d, sg_desc_cnt %d,\n",
+ len, rem_data, descs_used, sg_desc_cnt);
+
+ /* Close the segment if reached mss size or last frag*/
+ if (rem_data == 0 || (f == nr_frags - 1 && len == 0)) {
+ if (hdr_compensation_need) {
+ /* first segment include hdr desc for
+ * release
+ */
+ hdr_ctx->nr_frags = sg_desc_cnt;
+ wil_tx_desc_set_nr_frags(first_desc,
+ sg_desc_cnt +
+ 1);
+ hdr_compensation_need = false;
+ } else {
+ wil_tx_desc_set_nr_frags(first_desc,
+ sg_desc_cnt);
+ }
+ first_ctx->nr_frags = sg_desc_cnt - 1;
+
+ wil_tx_last_desc(d);
+
+ /* first descriptor may also be the last
+ * for this mss - make sure not to copy
+ * it twice
+ */
+ if (first_desc != d)
+ *_first_desc = *first_desc;
+
+ /*last descriptor will be copied at the end
+ * of this TS processing
+ */
+ if (f < nr_frags - 1 || len > 0)
+ *_desc = *d;
+
+ rem_data = mss;
+ _first_desc = NULL;
+ sg_desc_cnt = 0;
+ } else if (first_desc != d) /* update mid descriptor */
+ *_desc = *d;
+ }
+ }
+
+ /* first descriptor may also be the last.
+ * in this case d pointer is invalid
+ */
+ if (_first_desc == _desc)
+ d = first_desc;
+
+ /* Last data descriptor */
+ wil_set_tx_desc_last_tso(d);
+ *_desc = *d;
+
+ /* Fill the total number of descriptors in first desc (hdr)*/
+ wil_tx_desc_set_nr_frags(hdr_desc, descs_used);
+ *_hdr_desc = *hdr_desc;
+
+ /* hold reference to skb
+ * to prevent skb release before accounting
+ * in case of immediate "tx done"
+ */
+ vring->ctx[i].skb = skb_get(skb);
+
+ /* performance monitoring */
+ used = wil_vring_used_tx(vring);
+ if (wil_val_in_range(vring_idle_trsh,
+ used, used + descs_used)) {
+ txdata->idle += get_cycles() - txdata->last_idle;
+ wil_dbg_txrx(wil, "Ring[%2d] not idle %d -> %d\n",
+ vring_index, used, used + descs_used);
+ }
+
+ /* advance swhead */
+ wil_vring_advance_head(vring, descs_used);
+ wil_dbg_txrx(wil, "TSO: Tx swhead %d -> %d\n", swhead, vring->swhead);
+
+ /* make sure all writes to descriptors (shared memory) are done before
+ * committing them to HW
+ */
+ wmb();
+
+ wil_w(wil, vring->hwtail, vring->swhead);
+ return 0;
+
+mem_error:
+ while (descs_used > 0) {
+ struct wil_ctx *ctx;
+
+ i = (swhead + descs_used) % vring->size;
+ d = (struct vring_tx_desc *)&vring->va[i].tx;
+ _desc = &vring->va[i].tx;
+ *d = *_desc;
+ _desc->dma.status = TX_DMA_STATUS_DU;
+ ctx = &vring->ctx[i];
+ wil_txdesc_unmap(dev, d, ctx);
+ memset(ctx, 0, sizeof(*ctx));
+ descs_used--;
+ }
+err_exit:
+ return rc;
+}
+
static int __wil_tx_vring(struct wil6210_priv *wil, struct vring *vring,
struct sk_buff *skb)
{
@@ -1117,7 +1507,8 @@ static int __wil_tx_vring(struct wil6210_priv *wil, struct vring *vring,
bool mcast = (vring_index == wil->bcast_vring);
uint len = skb_headlen(skb);
- wil_dbg_txrx(wil, "%s()\n", __func__);
+ wil_dbg_txrx(wil, "%s() %d bytes to vring %d\n",
+ __func__, skb->len, vring_index);
if (unlikely(!txdata->enabled))
return -EINVAL;
@@ -1144,23 +1535,18 @@ static int __wil_tx_vring(struct wil6210_priv *wil, struct vring *vring,
wil_tx_desc_map(d, pa, len, vring_index);
if (unlikely(mcast)) {
d->mac.d[0] |= BIT(MAC_CFG_DESC_TX_0_MCS_EN_POS); /* MCS 0 */
- if (unlikely(len > WIL_BCAST_MCS0_LIMIT)) {
- /* set MCS 1 */
+ if (unlikely(len > WIL_BCAST_MCS0_LIMIT)) /* set MCS 1 */
d->mac.d[0] |= (1 << MAC_CFG_DESC_TX_0_MCS_INDEX_POS);
- /* packet mode 2 */
- d->mac.d[1] |= BIT(MAC_CFG_DESC_TX_1_PKT_MODE_EN_POS) |
- (2 << MAC_CFG_DESC_TX_1_PKT_MODE_POS);
- }
}
/* Process TCP/UDP checksum offloading */
- if (unlikely(wil_tx_desc_offload_cksum_set(wil, d, skb))) {
+ if (unlikely(wil_tx_desc_offload_setup(d, skb))) {
wil_err(wil, "Tx[%2d] Failed to set cksum, drop packet\n",
vring_index);
goto dma_error;
}
vring->ctx[i].nr_frags = nr_frags;
- wil_tx_desc_set_nr_frags(d, nr_frags);
+ wil_tx_desc_set_nr_frags(d, nr_frags + 1);
/* middle segments */
for (; f < nr_frags; f++) {
@@ -1176,15 +1562,18 @@ static int __wil_tx_vring(struct wil6210_priv *wil, struct vring *vring,
_d = &vring->va[i].tx;
pa = skb_frag_dma_map(dev, frag, 0, skb_frag_size(frag),
DMA_TO_DEVICE);
- if (unlikely(dma_mapping_error(dev, pa)))
+ if (unlikely(dma_mapping_error(dev, pa))) {
+ wil_err(wil, "Tx[%2d] failed to map fragment\n",
+ vring_index);
goto dma_error;
+ }
vring->ctx[i].mapped_as = wil_mapped_as_page;
wil_tx_desc_map(d, pa, len, vring_index);
/* no need to check return code -
* if it succeeded for 1-st descriptor,
* it will succeed here too
*/
- wil_tx_desc_offload_cksum_set(wil, d, skb);
+ wil_tx_desc_offload_setup(d, skb);
}
/* for the last seg only */
d->dma.d0 |= BIT(DMA_CFG_DESC_TX_0_CMD_EOP_POS);
@@ -1215,7 +1604,13 @@ static int __wil_tx_vring(struct wil6210_priv *wil, struct vring *vring,
wil_dbg_txrx(wil, "Tx[%2d] swhead %d -> %d\n", vring_index, swhead,
vring->swhead);
trace_wil6210_tx(vring_index, swhead, skb->len, nr_frags);
- iowrite32(vring->swhead, wil->csr + HOSTADDR(vring->hwtail));
+
+ /* make sure all writes to descriptors (shared memory) are done before
+ * committing them to HW
+ */
+ wmb();
+
+ wil_w(wil, vring->hwtail, vring->swhead);
return 0;
dma_error:
@@ -1231,9 +1626,6 @@ static int __wil_tx_vring(struct wil6210_priv *wil, struct vring *vring,
_d->dma.status = TX_DMA_STATUS_DU;
wil_txdesc_unmap(dev, d, ctx);
- if (ctx->skb)
- dev_kfree_skb_any(ctx->skb);
-
memset(ctx, 0, sizeof(*ctx));
}
@@ -1248,8 +1640,12 @@ static int wil_tx_vring(struct wil6210_priv *wil, struct vring *vring,
int rc;
spin_lock(&txdata->lock);
- rc = __wil_tx_vring(wil, vring, skb);
+
+ rc = (skb_is_gso(skb) ? __wil_tx_vring_tso : __wil_tx_vring)
+ (wil, vring, skb);
+
spin_unlock(&txdata->lock);
+
return rc;
}
@@ -1271,7 +1667,7 @@ netdev_tx_t wil_start_xmit(struct sk_buff *skb, struct net_device *ndev)
goto drop;
}
if (unlikely(!test_bit(wil_status_fwconnected, wil->status))) {
- wil_err(wil, "FW not connected\n");
+ wil_err_ratelimited(wil, "FW not connected\n");
goto drop;
}
if (unlikely(wil->wdev->iftype == NL80211_IFTYPE_MONITOR)) {
@@ -1376,7 +1772,8 @@ int wil_tx_complete(struct wil6210_priv *wil, int ringid)
struct wil_ctx *ctx = &vring->ctx[vring->swtail];
/**
* For the fragmented skb, HW will set DU bit only for the
- * last fragment. look for it
+ * last fragment. look for it.
+ * In TSO the first DU will include hdr desc
*/
int lf = (vring->swtail + ctx->nr_frags) % vring->size;
/* TODO: check we are not past head */