OpenCloudOS-Kernel/drivers/net/wireless/ath/wil6210/main.c

1968 lines
51 KiB
C
Raw Normal View History

/*
* Copyright (c) 2012-2017 Qualcomm Atheros, Inc.
* Copyright (c) 2018, The Linux Foundation. All rights reserved.
*
* Permission to use, copy, modify, and/or distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
#include <linux/moduleparam.h>
#include <linux/if_arp.h>
#include <linux/etherdevice.h>
#include <linux/rtnetlink.h>
#include "wil6210.h"
#include "txrx.h"
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
#include "txrx_edma.h"
#include "wmi.h"
#include "boot_loader.h"
#define WAIT_FOR_HALP_VOTE_MS 100
#define WAIT_FOR_SCAN_ABORT_MS 1000
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
#define WIL_DEFAULT_NUM_RX_STATUS_RINGS 1
#define WIL_BOARD_FILE_MAX_NAMELEN 128
bool debug_fw; /* = false; */
module_param(debug_fw, bool, 0444);
MODULE_PARM_DESC(debug_fw, " do not perform card reset. For FW debug");
static u8 oob_mode;
module_param(oob_mode, byte, 0444);
MODULE_PARM_DESC(oob_mode,
" enable out of the box (OOB) mode in FW, for diagnostics and certification");
bool no_fw_recovery;
module_param(no_fw_recovery, bool, 0644);
MODULE_PARM_DESC(no_fw_recovery, " disable automatic FW error recovery");
/* if not set via modparam, will be set to default value of 1/8 of
* rx ring size during init flow
*/
unsigned short rx_ring_overflow_thrsh = WIL6210_RX_HIGH_TRSH_INIT;
module_param(rx_ring_overflow_thrsh, ushort, 0444);
MODULE_PARM_DESC(rx_ring_overflow_thrsh,
" RX ring overflow threshold in descriptors.");
/* We allow allocation of more than 1 page buffers to support large packets.
* It is suboptimal behavior performance wise in case MTU above page size.
*/
unsigned int mtu_max = TXRX_BUF_LEN_DEFAULT - WIL_MAX_MPDU_OVERHEAD;
static int mtu_max_set(const char *val, const struct kernel_param *kp)
{
int ret;
/* sets mtu_max directly. no need to restore it in case of
* illegal value since we assume this will fail insmod
*/
ret = param_set_uint(val, kp);
if (ret)
return ret;
if (mtu_max < 68 || mtu_max > WIL_MAX_ETH_MTU)
ret = -EINVAL;
return ret;
}
static const struct kernel_param_ops mtu_max_ops = {
.set = mtu_max_set,
.get = param_get_uint,
};
module_param_cb(mtu_max, &mtu_max_ops, &mtu_max, 0444);
MODULE_PARM_DESC(mtu_max, " Max MTU value.");
static uint rx_ring_order;
static uint tx_ring_order = WIL_TX_RING_SIZE_ORDER_DEFAULT;
static uint bcast_ring_order = WIL_BCAST_RING_SIZE_ORDER_DEFAULT;
static int ring_order_set(const char *val, const struct kernel_param *kp)
{
int ret;
uint x;
ret = kstrtouint(val, 0, &x);
if (ret)
return ret;
if ((x < WIL_RING_SIZE_ORDER_MIN) || (x > WIL_RING_SIZE_ORDER_MAX))
return -EINVAL;
*((uint *)kp->arg) = x;
return 0;
}
static const struct kernel_param_ops ring_order_ops = {
.set = ring_order_set,
.get = param_get_uint,
};
module_param_cb(rx_ring_order, &ring_order_ops, &rx_ring_order, 0444);
MODULE_PARM_DESC(rx_ring_order, " Rx ring order; size = 1 << order");
module_param_cb(tx_ring_order, &ring_order_ops, &tx_ring_order, 0444);
MODULE_PARM_DESC(tx_ring_order, " Tx ring order; size = 1 << order");
module_param_cb(bcast_ring_order, &ring_order_ops, &bcast_ring_order, 0444);
MODULE_PARM_DESC(bcast_ring_order, " Bcast ring order; size = 1 << order");
enum {
WIL_BOOT_ERR,
WIL_BOOT_VANILLA,
WIL_BOOT_PRODUCTION,
WIL_BOOT_DEVELOPMENT,
};
enum {
WIL_SIG_STATUS_VANILLA = 0x0,
WIL_SIG_STATUS_DEVELOPMENT = 0x1,
WIL_SIG_STATUS_PRODUCTION = 0x2,
WIL_SIG_STATUS_CORRUPTED_PRODUCTION = 0x3,
};
#define RST_DELAY (20) /* msec, for loop in @wil_wait_device_ready */
#define RST_COUNT (1 + 1000/RST_DELAY) /* round up to be above 1 sec total */
#define PMU_READY_DELAY_MS (4) /* ms, for sleep in @wil_wait_device_ready */
#define OTP_HW_DELAY (200) /* usec, loop in @wil_wait_device_ready_talyn_mb */
/* round up to be above 2 ms total */
#define OTP_HW_COUNT (1 + 2000 / OTP_HW_DELAY)
/*
* Due to a hardware issue,
* one has to read/write to/from NIC in 32-bit chunks;
* regular memcpy_fromio and siblings will
* not work on 64-bit platform - it uses 64-bit transactions
*
* Force 32-bit transactions to enable NIC on 64-bit platforms
*
* To avoid byte swap on big endian host, __raw_{read|write}l
* should be used - {read|write}l would swap bytes to provide
* little endian on PCI value in host endianness.
*/
void wil_memcpy_fromio_32(void *dst, const volatile void __iomem *src,
size_t count)
{
u32 *d = dst;
const volatile u32 __iomem *s = src;
for (; count >= 4; count -= 4)
*d++ = __raw_readl(s++);
if (unlikely(count)) {
/* count can be 1..3 */
u32 tmp = __raw_readl(s);
memcpy(d, &tmp, count);
}
}
void wil_memcpy_toio_32(volatile void __iomem *dst, const void *src,
size_t count)
{
volatile u32 __iomem *d = dst;
const u32 *s = src;
for (; count >= 4; count -= 4)
__raw_writel(*s++, d++);
if (unlikely(count)) {
/* count can be 1..3 */
u32 tmp = 0;
memcpy(&tmp, s, count);
__raw_writel(tmp, d);
}
}
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
static void wil_ring_fini_tx(struct wil6210_priv *wil, int id)
{
struct wil_ring *ring = &wil->ring_tx[id];
struct wil_ring_tx_data *txdata = &wil->ring_tx_data[id];
lockdep_assert_held(&wil->mutex);
if (!ring->va)
return;
wil_dbg_misc(wil, "vring_fini_tx: id=%d\n", id);
spin_lock_bh(&txdata->lock);
txdata->dot1x_open = false;
txdata->mid = U8_MAX;
txdata->enabled = 0; /* no Tx can be in progress or start anew */
spin_unlock_bh(&txdata->lock);
/* napi_synchronize waits for completion of the current NAPI but will
* not prevent the next NAPI run.
* Add a memory barrier to guarantee that txdata->enabled is zeroed
* before napi_synchronize so that the next scheduled NAPI will not
* handle this vring
*/
wmb();
/* make sure NAPI won't touch this vring */
if (test_bit(wil_status_napi_en, wil->status))
napi_synchronize(&wil->napi_tx);
wil->txrx_ops.ring_fini_tx(wil, ring);
}
static bool wil_vif_is_connected(struct wil6210_priv *wil, u8 mid)
{
int i;
for (i = 0; i < WIL6210_MAX_CID; i++) {
if (wil->sta[i].mid == mid &&
wil->sta[i].status == wil_sta_connected)
return true;
}
return false;
}
static void wil_disconnect_cid_complete(struct wil6210_vif *vif, int cid,
u16 reason_code)
__acquires(&sta->tid_rx_lock) __releases(&sta->tid_rx_lock)
{
uint i;
struct wil6210_priv *wil = vif_to_wil(vif);
struct net_device *ndev = vif_to_ndev(vif);
struct wireless_dev *wdev = vif_to_wdev(vif);
struct wil_sta_info *sta = &wil->sta[cid];
int min_ring_id = wil_get_min_tx_ring_id(wil);
might_sleep();
wil_dbg_misc(wil,
"disconnect_cid_complete: CID %d, MID %d, status %d\n",
cid, sta->mid, sta->status);
/* inform upper layers */
if (sta->status != wil_sta_unused) {
if (vif->mid != sta->mid) {
wil_err(wil, "STA MID mismatch with VIF MID(%d)\n",
vif->mid);
}
switch (wdev->iftype) {
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_P2P_GO:
/* AP-like interface */
cfg80211_del_sta(ndev, sta->addr, GFP_KERNEL);
break;
default:
break;
}
sta->status = wil_sta_unused;
sta->mid = U8_MAX;
}
/* reorder buffers */
for (i = 0; i < WIL_STA_TID_NUM; i++) {
struct wil_tid_ampdu_rx *r;
spin_lock_bh(&sta->tid_rx_lock);
r = sta->tid_rx[i];
sta->tid_rx[i] = NULL;
wil_tid_ampdu_rx_free(wil, r);
spin_unlock_bh(&sta->tid_rx_lock);
}
/* crypto context */
memset(sta->tid_crypto_rx, 0, sizeof(sta->tid_crypto_rx));
memset(&sta->group_crypto_rx, 0, sizeof(sta->group_crypto_rx));
/* release vrings */
for (i = min_ring_id; i < ARRAY_SIZE(wil->ring_tx); i++) {
if (wil->ring2cid_tid[i][0] == cid)
wil_ring_fini_tx(wil, i);
}
/* statistics */
memset(&sta->stats, 0, sizeof(sta->stats));
sta->stats.tx_latency_min_us = U32_MAX;
}
static void _wil6210_disconnect_complete(struct wil6210_vif *vif,
const u8 *bssid, u16 reason_code)
{
struct wil6210_priv *wil = vif_to_wil(vif);
int cid = -ENOENT;
struct net_device *ndev;
struct wireless_dev *wdev;
ndev = vif_to_ndev(vif);
wdev = vif_to_wdev(vif);
might_sleep();
wil_info(wil, "disconnect_complete: bssid=%pM, reason=%d\n",
bssid, reason_code);
/* Cases are:
* - disconnect single STA, still connected
* - disconnect single STA, already disconnected
* - disconnect all
*
* For "disconnect all", there are 3 options:
* - bssid == NULL
* - bssid is broadcast address (ff:ff:ff:ff:ff:ff)
* - bssid is our MAC address
*/
if (bssid && !is_broadcast_ether_addr(bssid) &&
!ether_addr_equal_unaligned(ndev->dev_addr, bssid)) {
cid = wil_find_cid(wil, vif->mid, bssid);
wil_dbg_misc(wil,
"Disconnect complete %pM, CID=%d, reason=%d\n",
bssid, cid, reason_code);
if (cid >= 0) /* disconnect 1 peer */
wil_disconnect_cid_complete(vif, cid, reason_code);
} else { /* all */
wil_dbg_misc(wil, "Disconnect complete all\n");
for (cid = 0; cid < WIL6210_MAX_CID; cid++)
wil_disconnect_cid_complete(vif, cid, reason_code);
}
/* link state */
switch (wdev->iftype) {
case NL80211_IFTYPE_STATION:
case NL80211_IFTYPE_P2P_CLIENT:
wil_bcast_fini(vif);
wil_update_net_queues_bh(wil, vif, NULL, true);
netif_carrier_off(ndev);
if (!wil_has_other_active_ifaces(wil, ndev, false, true))
wil6210_bus_request(wil, WIL_DEFAULT_BUS_REQUEST_KBPS);
if (test_and_clear_bit(wil_vif_fwconnected, vif->status)) {
atomic_dec(&wil->connected_vifs);
cfg80211_disconnected(ndev, reason_code,
NULL, 0,
vif->locally_generated_disc,
GFP_KERNEL);
vif->locally_generated_disc = false;
} else if (test_bit(wil_vif_fwconnecting, vif->status)) {
cfg80211_connect_result(ndev, bssid, NULL, 0, NULL, 0,
WLAN_STATUS_UNSPECIFIED_FAILURE,
GFP_KERNEL);
vif->bss = NULL;
}
clear_bit(wil_vif_fwconnecting, vif->status);
clear_bit(wil_vif_ft_roam, vif->status);
break;
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_P2P_GO:
if (!wil_vif_is_connected(wil, vif->mid)) {
wil_update_net_queues_bh(wil, vif, NULL, true);
if (test_and_clear_bit(wil_vif_fwconnected,
vif->status))
atomic_dec(&wil->connected_vifs);
} else {
wil_update_net_queues_bh(wil, vif, NULL, false);
}
break;
default:
break;
}
}
static int wil_disconnect_cid(struct wil6210_vif *vif, int cid,
u16 reason_code)
{
struct wil6210_priv *wil = vif_to_wil(vif);
struct wireless_dev *wdev = vif_to_wdev(vif);
struct wil_sta_info *sta = &wil->sta[cid];
bool del_sta = false;
might_sleep();
wil_dbg_misc(wil, "disconnect_cid: CID %d, MID %d, status %d\n",
cid, sta->mid, sta->status);
if (sta->status == wil_sta_unused)
return 0;
if (vif->mid != sta->mid) {
wil_err(wil, "STA MID mismatch with VIF MID(%d)\n", vif->mid);
return -EINVAL;
}
/* inform lower layers */
if (wdev->iftype == NL80211_IFTYPE_AP && disable_ap_sme)
del_sta = true;
/* disconnect by sending command disconnect/del_sta and wait
* synchronously for WMI_DISCONNECT_EVENTID event.
*/
return wmi_disconnect_sta(vif, sta->addr, reason_code, del_sta);
}
static void _wil6210_disconnect(struct wil6210_vif *vif, const u8 *bssid,
u16 reason_code)
{
struct wil6210_priv *wil;
struct net_device *ndev;
int cid = -ENOENT;
if (unlikely(!vif))
return;
wil = vif_to_wil(vif);
ndev = vif_to_ndev(vif);
might_sleep();
wil_info(wil, "disconnect bssid=%pM, reason=%d\n", bssid, reason_code);
/* Cases are:
* - disconnect single STA, still connected
* - disconnect single STA, already disconnected
* - disconnect all
*
* For "disconnect all", there are 3 options:
* - bssid == NULL
* - bssid is broadcast address (ff:ff:ff:ff:ff:ff)
* - bssid is our MAC address
*/
if (bssid && !is_broadcast_ether_addr(bssid) &&
!ether_addr_equal_unaligned(ndev->dev_addr, bssid)) {
cid = wil_find_cid(wil, vif->mid, bssid);
wil_dbg_misc(wil, "Disconnect %pM, CID=%d, reason=%d\n",
bssid, cid, reason_code);
if (cid >= 0) /* disconnect 1 peer */
wil_disconnect_cid(vif, cid, reason_code);
} else { /* all */
wil_dbg_misc(wil, "Disconnect all\n");
for (cid = 0; cid < WIL6210_MAX_CID; cid++)
wil_disconnect_cid(vif, cid, reason_code);
}
/* call event handler manually after processing wmi_call,
* to avoid deadlock - disconnect event handler acquires
* wil->mutex while it is already held here
*/
_wil6210_disconnect_complete(vif, bssid, reason_code);
}
void wil_disconnect_worker(struct work_struct *work)
{
struct wil6210_vif *vif = container_of(work,
struct wil6210_vif, disconnect_worker);
struct wil6210_priv *wil = vif_to_wil(vif);
struct net_device *ndev = vif_to_ndev(vif);
int rc;
struct {
struct wmi_cmd_hdr wmi;
struct wmi_disconnect_event evt;
} __packed reply;
if (test_bit(wil_vif_fwconnected, vif->status))
/* connect succeeded after all */
return;
if (!test_bit(wil_vif_fwconnecting, vif->status))
/* already disconnected */
return;
memset(&reply, 0, sizeof(reply));
rc = wmi_call(wil, WMI_DISCONNECT_CMDID, vif->mid, NULL, 0,
WMI_DISCONNECT_EVENTID, &reply, sizeof(reply),
WIL6210_DISCONNECT_TO_MS);
if (rc) {
wil_err(wil, "disconnect error %d\n", rc);
return;
}
wil_update_net_queues_bh(wil, vif, NULL, true);
netif_carrier_off(ndev);
cfg80211_connect_result(ndev, NULL, NULL, 0, NULL, 0,
WLAN_STATUS_UNSPECIFIED_FAILURE, GFP_KERNEL);
clear_bit(wil_vif_fwconnecting, vif->status);
}
static int wil_wait_for_recovery(struct wil6210_priv *wil)
{
if (wait_event_interruptible(wil->wq, wil->recovery_state !=
fw_recovery_pending)) {
wil_err(wil, "Interrupt, canceling recovery\n");
return -ERESTARTSYS;
}
if (wil->recovery_state != fw_recovery_running) {
wil_info(wil, "Recovery cancelled\n");
return -EINTR;
}
wil_info(wil, "Proceed with recovery\n");
return 0;
}
void wil_set_recovery_state(struct wil6210_priv *wil, int state)
{
wil_dbg_misc(wil, "set_recovery_state: %d -> %d\n",
wil->recovery_state, state);
wil->recovery_state = state;
wake_up_interruptible(&wil->wq);
}
bool wil_is_recovery_blocked(struct wil6210_priv *wil)
{
return no_fw_recovery && (wil->recovery_state == fw_recovery_pending);
}
static void wil_fw_error_worker(struct work_struct *work)
{
struct wil6210_priv *wil = container_of(work, struct wil6210_priv,
fw_error_worker);
struct net_device *ndev = wil->main_ndev;
struct wireless_dev *wdev;
wil_dbg_misc(wil, "fw error worker\n");
if (!ndev || !(ndev->flags & IFF_UP)) {
wil_info(wil, "No recovery - interface is down\n");
return;
}
wdev = ndev->ieee80211_ptr;
/* increment @recovery_count if less then WIL6210_FW_RECOVERY_TO
* passed since last recovery attempt
*/
if (time_is_after_jiffies(wil->last_fw_recovery +
WIL6210_FW_RECOVERY_TO))
wil->recovery_count++;
else
wil->recovery_count = 1; /* fw was alive for a long time */
if (wil->recovery_count > WIL6210_FW_RECOVERY_RETRIES) {
wil_err(wil, "too many recovery attempts (%d), giving up\n",
wil->recovery_count);
return;
}
wil->last_fw_recovery = jiffies;
wil_info(wil, "fw error recovery requested (try %d)...\n",
wil->recovery_count);
if (!no_fw_recovery)
wil->recovery_state = fw_recovery_running;
if (wil_wait_for_recovery(wil) != 0)
return;
rtnl_lock();
mutex_lock(&wil->mutex);
/* Needs adaptation for multiple VIFs
* need to go over all VIFs and consider the appropriate
* recovery because each one can have different iftype.
*/
switch (wdev->iftype) {
case NL80211_IFTYPE_STATION:
case NL80211_IFTYPE_P2P_CLIENT:
case NL80211_IFTYPE_MONITOR:
/* silent recovery, upper layers will see disconnect */
wil6210: fix for oops while stopping interface When interface stopped while running intensive Rx traffic, the following oops observed: [89846.734683] Call trace: [89846.737117] [<ffffffc00083aa64>] dev_gro_receive+0xac/0x358 [89846.742674] [<ffffffc00083ae94>] napi_gro_receive+0x24/0xa4 [89846.748251] [<ffffffbffc1c2f88>] $x+0xec/0x1f8 [wil6210] wil_netif_rx_any [89846.753547] [<ffffffbffc1c4830>] $x+0x34/0x54 [wil6210] wil_release_reorder_frame [89846.758755] [<ffffffbffc1c48ac>] wil_release_reorder_frames+0x5c/0x78 [wil6210] [89846.766044] [<ffffffbffc1c4bf8>] wil_tid_ampdu_rx_free+0x20/0x48 [wil6210] [89846.772901] [<ffffffbffc1bedc8>] $x+0x190/0x1e8 [wil6210] [89846.778285] [<ffffffbffc1c0ed4>] wmi_event_worker+0x230/0x2f8 [wil6210] [89846.784865] [<ffffffc0000b0bc8>] process_one_work+0x278/0x3fc [89846.790591] [<ffffffc0000b1218>] worker_thread+0x200/0x330 [89846.796060] [<ffffffc0000b6664>] kthread+0xac/0xb8 [89846.800836] Code: b940c661 f9406a62 8b010041 f9400026 (f8636882) [89846.807008] ---[ end trace d6fdc17cd27d18f6 ]--- Reason is the following: when removing Rx vring (wil_netdev_ops.ndo_stop -> wil_stop -> wil_down -> __wil_down -> wil_rx_fini), Rx interrupt occurs. It trigger Rx NAPI, calling wil_rx_handle() that reaps (already cleaned) buffer, causing skb referring to garbage memory being set into reorder buffer. Then, network stack trying to access this buffer and fails. Prevent Rx NAPI from being scheduled if device going to stop. Bit wil_status_napi_en reflects NAPI enablement state, check it when triggering Rx NAPI. Testing shows that check for wil_status_napi_en sometimes gets negative, and new error message get printed - in this case kernel oops would be observed. Original oops is no more reproducible. This change requires also changes in the AP flows. Properly enable/disable NAPI for the AP. Make sure Rx VRING is disabled when resetting target. For this, promote __wil_up() and __wil_down() to the module scope, and use it in the relevant flows. Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2014-09-10 21:34:50 +08:00
__wil_down(wil);
__wil_up(wil);
break;
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_P2P_GO:
if (no_fw_recovery) /* upper layers do recovery */
break;
/* silent recovery, upper layers will see disconnect */
__wil_down(wil);
__wil_up(wil);
mutex_unlock(&wil->mutex);
wil_cfg80211_ap_recovery(wil);
mutex_lock(&wil->mutex);
wil_info(wil, "... completed\n");
break;
default:
wil_err(wil, "No recovery - unknown interface type %d\n",
wdev->iftype);
break;
}
mutex_unlock(&wil->mutex);
rtnl_unlock();
}
static int wil_find_free_ring(struct wil6210_priv *wil)
{
int i;
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
int min_ring_id = wil_get_min_tx_ring_id(wil);
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
for (i = min_ring_id; i < WIL6210_MAX_TX_RINGS; i++) {
if (!wil->ring_tx[i].va)
return i;
}
return -EINVAL;
}
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
int wil_ring_init_tx(struct wil6210_vif *vif, int cid)
{
struct wil6210_priv *wil = vif_to_wil(vif);
int rc = -EINVAL, ringid;
if (cid < 0) {
wil_err(wil, "No connection pending\n");
goto out;
}
ringid = wil_find_free_ring(wil);
if (ringid < 0) {
wil_err(wil, "No free vring found\n");
goto out;
}
wil_dbg_wmi(wil, "Configure for connection CID %d MID %d ring %d\n",
cid, vif->mid, ringid);
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
rc = wil->txrx_ops.ring_init_tx(vif, ringid, 1 << tx_ring_order,
cid, 0);
if (rc)
wil_err(wil, "init TX for CID %d MID %d vring %d failed\n",
cid, vif->mid, ringid);
out:
return rc;
}
int wil_bcast_init(struct wil6210_vif *vif)
{
struct wil6210_priv *wil = vif_to_wil(vif);
int ri = vif->bcast_ring, rc;
if (ri >= 0 && wil->ring_tx[ri].va)
return 0;
ri = wil_find_free_ring(wil);
if (ri < 0)
return ri;
vif->bcast_ring = ri;
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
rc = wil->txrx_ops.ring_init_bcast(vif, ri, 1 << bcast_ring_order);
if (rc)
vif->bcast_ring = -1;
return rc;
}
void wil_bcast_fini(struct wil6210_vif *vif)
{
struct wil6210_priv *wil = vif_to_wil(vif);
int ri = vif->bcast_ring;
if (ri < 0)
return;
vif->bcast_ring = -1;
wil_ring_fini_tx(wil, ri);
}
void wil_bcast_fini_all(struct wil6210_priv *wil)
{
int i;
struct wil6210_vif *vif;
for (i = 0; i < wil->max_vifs; i++) {
vif = wil->vifs[i];
if (vif)
wil_bcast_fini(vif);
}
}
int wil_priv_init(struct wil6210_priv *wil)
{
uint i;
wil_dbg_misc(wil, "priv_init\n");
memset(wil->sta, 0, sizeof(wil->sta));
for (i = 0; i < WIL6210_MAX_CID; i++) {
spin_lock_init(&wil->sta[i].tid_rx_lock);
wil->sta[i].mid = U8_MAX;
}
for (i = 0; i < WIL6210_MAX_TX_RINGS; i++) {
spin_lock_init(&wil->ring_tx_data[i].lock);
wil->ring2cid_tid[i][0] = WIL6210_MAX_CID;
}
mutex_init(&wil->mutex);
mutex_init(&wil->vif_mutex);
mutex_init(&wil->wmi_mutex);
mutex_init(&wil->halp.lock);
init_completion(&wil->wmi_ready);
init_completion(&wil->wmi_call);
init_completion(&wil->halp.comp);
INIT_WORK(&wil->wmi_event_worker, wmi_event_worker);
INIT_WORK(&wil->fw_error_worker, wil_fw_error_worker);
INIT_LIST_HEAD(&wil->pending_wmi_ev);
spin_lock_init(&wil->wmi_ev_lock);
spin_lock_init(&wil->net_queue_lock);
init_waitqueue_head(&wil->wq);
wil6210: ADDBA/DELBA flows Introduce BACK establishment procedures; decision logic is not implemented yet; debugfs entry 'addba' used to manually trigger addba/delba for ringid 0. debugfs usage: to establish BACK with agg_wsize 16: echo 16 > /sys/kernel/debug/ieee80211/phy0/wil6210/addba to delete BACK: echo 0 > /sys/kernel/debug/ieee80211/phy0/wil6210/addba to change agg_wsize, one need to delete BACK and establish it anew ADDBA flow for: - originator Tx side (initiator) sends WMI_VRING_BA_EN_CMDID providing agg_wsize and timeout parameters. Eventually, it gets event confirming BACK agreement - WMI_BA_STATUS_EVENTID with negotiated parameters. On this event, update Tx vring data (struct vring_tx_data) and display BACK parameters on debugfs - recipient Rx side (recipient) firmware informs driver about ADDBA with WMI_RCP_ADDBA_REQ_EVENTID, driver process it in service work queue wq_service. It adjusts parameters and sends response with WMI_RCP_ADDBA_RESP_CMDID, and final confirmation provided by firmware with WMI_ADDBA_RESP_SENT_EVENTID. In case of success, driver updates Rx BACK reorder buffer. policy for BACK parameters: - aggregation size (agg_wsize * MPDUsize)) to not exceed 64Kbytes DELBA flow for: - originator driver decides to terminate BACK, it sends WMI_VRING_BA_DIS_CMDID and updates struct vring_tx_data associated with vring; ignore WMI_DELBA_EVENTID. - recipient firmware informs driver with WMI_DELBA_EVENTID, driver deletes correspondent reorder buffer ADDBA request processing requires sending WMI command, therefore it is processed in work queue context. Same work queue used as for connect, it get renamed to wq_service Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2014-12-23 15:47:03 +08:00
wil->wmi_wq = create_singlethread_workqueue(WIL_NAME "_wmi");
if (!wil->wmi_wq)
return -EAGAIN;
wil6210: ADDBA/DELBA flows Introduce BACK establishment procedures; decision logic is not implemented yet; debugfs entry 'addba' used to manually trigger addba/delba for ringid 0. debugfs usage: to establish BACK with agg_wsize 16: echo 16 > /sys/kernel/debug/ieee80211/phy0/wil6210/addba to delete BACK: echo 0 > /sys/kernel/debug/ieee80211/phy0/wil6210/addba to change agg_wsize, one need to delete BACK and establish it anew ADDBA flow for: - originator Tx side (initiator) sends WMI_VRING_BA_EN_CMDID providing agg_wsize and timeout parameters. Eventually, it gets event confirming BACK agreement - WMI_BA_STATUS_EVENTID with negotiated parameters. On this event, update Tx vring data (struct vring_tx_data) and display BACK parameters on debugfs - recipient Rx side (recipient) firmware informs driver about ADDBA with WMI_RCP_ADDBA_REQ_EVENTID, driver process it in service work queue wq_service. It adjusts parameters and sends response with WMI_RCP_ADDBA_RESP_CMDID, and final confirmation provided by firmware with WMI_ADDBA_RESP_SENT_EVENTID. In case of success, driver updates Rx BACK reorder buffer. policy for BACK parameters: - aggregation size (agg_wsize * MPDUsize)) to not exceed 64Kbytes DELBA flow for: - originator driver decides to terminate BACK, it sends WMI_VRING_BA_DIS_CMDID and updates struct vring_tx_data associated with vring; ignore WMI_DELBA_EVENTID. - recipient firmware informs driver with WMI_DELBA_EVENTID, driver deletes correspondent reorder buffer ADDBA request processing requires sending WMI command, therefore it is processed in work queue context. Same work queue used as for connect, it get renamed to wq_service Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2014-12-23 15:47:03 +08:00
wil->wq_service = create_singlethread_workqueue(WIL_NAME "_service");
if (!wil->wq_service)
goto out_wmi_wq;
wil->last_fw_recovery = jiffies;
wil->tx_interframe_timeout = WIL6210_ITR_TX_INTERFRAME_TIMEOUT_DEFAULT;
wil->rx_interframe_timeout = WIL6210_ITR_RX_INTERFRAME_TIMEOUT_DEFAULT;
wil->tx_max_burst_duration = WIL6210_ITR_TX_MAX_BURST_DURATION_DEFAULT;
wil->rx_max_burst_duration = WIL6210_ITR_RX_MAX_BURST_DURATION_DEFAULT;
if (rx_ring_overflow_thrsh == WIL6210_RX_HIGH_TRSH_INIT)
rx_ring_overflow_thrsh = WIL6210_RX_HIGH_TRSH_DEFAULT;
wil->ps_profile = WMI_PS_PROFILE_TYPE_DEFAULT;
wil->wakeup_trigger = WMI_WAKEUP_TRIGGER_UCAST |
WMI_WAKEUP_TRIGGER_BCAST;
memset(&wil->suspend_stats, 0, sizeof(wil->suspend_stats));
wil->ring_idle_trsh = 16;
wil->reply_mid = U8_MAX;
wil->max_vifs = 1;
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
/* edma configuration can be updated via debugfs before allocation */
wil->num_rx_status_rings = WIL_DEFAULT_NUM_RX_STATUS_RINGS;
wil->tx_status_ring_order = WIL_TX_SRING_SIZE_ORDER_DEFAULT;
/* Rx status ring size should be bigger than the number of RX buffers
* in order to prevent backpressure on the status ring, which may
* cause HW freeze.
*/
wil->rx_status_ring_order = WIL_RX_SRING_SIZE_ORDER_DEFAULT;
/* Number of RX buffer IDs should be bigger than the RX descriptor
* ring size as in HW reorder flow, the HW can consume additional
* buffers before releasing the previous ones.
*/
wil->rx_buff_id_count = WIL_RX_BUFF_ARR_SIZE_DEFAULT;
wil->amsdu_en = 1;
return 0;
wil6210: ADDBA/DELBA flows Introduce BACK establishment procedures; decision logic is not implemented yet; debugfs entry 'addba' used to manually trigger addba/delba for ringid 0. debugfs usage: to establish BACK with agg_wsize 16: echo 16 > /sys/kernel/debug/ieee80211/phy0/wil6210/addba to delete BACK: echo 0 > /sys/kernel/debug/ieee80211/phy0/wil6210/addba to change agg_wsize, one need to delete BACK and establish it anew ADDBA flow for: - originator Tx side (initiator) sends WMI_VRING_BA_EN_CMDID providing agg_wsize and timeout parameters. Eventually, it gets event confirming BACK agreement - WMI_BA_STATUS_EVENTID with negotiated parameters. On this event, update Tx vring data (struct vring_tx_data) and display BACK parameters on debugfs - recipient Rx side (recipient) firmware informs driver about ADDBA with WMI_RCP_ADDBA_REQ_EVENTID, driver process it in service work queue wq_service. It adjusts parameters and sends response with WMI_RCP_ADDBA_RESP_CMDID, and final confirmation provided by firmware with WMI_ADDBA_RESP_SENT_EVENTID. In case of success, driver updates Rx BACK reorder buffer. policy for BACK parameters: - aggregation size (agg_wsize * MPDUsize)) to not exceed 64Kbytes DELBA flow for: - originator driver decides to terminate BACK, it sends WMI_VRING_BA_DIS_CMDID and updates struct vring_tx_data associated with vring; ignore WMI_DELBA_EVENTID. - recipient firmware informs driver with WMI_DELBA_EVENTID, driver deletes correspondent reorder buffer ADDBA request processing requires sending WMI command, therefore it is processed in work queue context. Same work queue used as for connect, it get renamed to wq_service Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2014-12-23 15:47:03 +08:00
out_wmi_wq:
destroy_workqueue(wil->wmi_wq);
return -EAGAIN;
}
void wil6210_bus_request(struct wil6210_priv *wil, u32 kbps)
{
if (wil->platform_ops.bus_request) {
wil->bus_request_kbps = kbps;
wil->platform_ops.bus_request(wil->platform_handle, kbps);
}
}
/**
* wil6210_disconnect - disconnect one connection
* @vif: virtual interface context
* @bssid: peer to disconnect, NULL to disconnect all
* @reason_code: Reason code for the Disassociation frame
*
* Disconnect and release associated resources. Issue WMI
* command(s) to trigger MAC disconnect. When command was issued
* successfully, call the wil6210_disconnect_complete function
* to handle the event synchronously
*/
void wil6210_disconnect(struct wil6210_vif *vif, const u8 *bssid,
u16 reason_code)
{
struct wil6210_priv *wil = vif_to_wil(vif);
wil_dbg_misc(wil, "disconnecting\n");
del_timer_sync(&vif->connect_timer);
_wil6210_disconnect(vif, bssid, reason_code);
}
/**
* wil6210_disconnect_complete - handle disconnect event
* @vif: virtual interface context
* @bssid: peer to disconnect, NULL to disconnect all
* @reason_code: Reason code for the Disassociation frame
*
* Release associated resources and indicate upper layers the
* connection is terminated.
*/
void wil6210_disconnect_complete(struct wil6210_vif *vif, const u8 *bssid,
u16 reason_code)
{
struct wil6210_priv *wil = vif_to_wil(vif);
wil_dbg_misc(wil, "got disconnect\n");
del_timer_sync(&vif->connect_timer);
_wil6210_disconnect_complete(vif, bssid, reason_code);
}
void wil_priv_deinit(struct wil6210_priv *wil)
{
wil_dbg_misc(wil, "priv_deinit\n");
wil_set_recovery_state(wil, fw_recovery_idle);
cancel_work_sync(&wil->fw_error_worker);
wmi_event_flush(wil);
wil6210: ADDBA/DELBA flows Introduce BACK establishment procedures; decision logic is not implemented yet; debugfs entry 'addba' used to manually trigger addba/delba for ringid 0. debugfs usage: to establish BACK with agg_wsize 16: echo 16 > /sys/kernel/debug/ieee80211/phy0/wil6210/addba to delete BACK: echo 0 > /sys/kernel/debug/ieee80211/phy0/wil6210/addba to change agg_wsize, one need to delete BACK and establish it anew ADDBA flow for: - originator Tx side (initiator) sends WMI_VRING_BA_EN_CMDID providing agg_wsize and timeout parameters. Eventually, it gets event confirming BACK agreement - WMI_BA_STATUS_EVENTID with negotiated parameters. On this event, update Tx vring data (struct vring_tx_data) and display BACK parameters on debugfs - recipient Rx side (recipient) firmware informs driver about ADDBA with WMI_RCP_ADDBA_REQ_EVENTID, driver process it in service work queue wq_service. It adjusts parameters and sends response with WMI_RCP_ADDBA_RESP_CMDID, and final confirmation provided by firmware with WMI_ADDBA_RESP_SENT_EVENTID. In case of success, driver updates Rx BACK reorder buffer. policy for BACK parameters: - aggregation size (agg_wsize * MPDUsize)) to not exceed 64Kbytes DELBA flow for: - originator driver decides to terminate BACK, it sends WMI_VRING_BA_DIS_CMDID and updates struct vring_tx_data associated with vring; ignore WMI_DELBA_EVENTID. - recipient firmware informs driver with WMI_DELBA_EVENTID, driver deletes correspondent reorder buffer ADDBA request processing requires sending WMI command, therefore it is processed in work queue context. Same work queue used as for connect, it get renamed to wq_service Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2014-12-23 15:47:03 +08:00
destroy_workqueue(wil->wq_service);
destroy_workqueue(wil->wmi_wq);
}
static void wil_shutdown_bl(struct wil6210_priv *wil)
{
u32 val;
wil_s(wil, RGF_USER_BL +
offsetof(struct bl_dedicated_registers_v1,
bl_shutdown_handshake), BL_SHUTDOWN_HS_GRTD);
usleep_range(100, 150);
val = wil_r(wil, RGF_USER_BL +
offsetof(struct bl_dedicated_registers_v1,
bl_shutdown_handshake));
if (val & BL_SHUTDOWN_HS_RTD) {
wil_dbg_misc(wil, "BL is ready for halt\n");
return;
}
wil_err(wil, "BL did not report ready for halt\n");
}
/* this format is used by ARC embedded CPU for instruction memory */
static inline u32 ARC_me_imm32(u32 d)
{
return ((d & 0xffff0000) >> 16) | ((d & 0x0000ffff) << 16);
}
/* defines access to interrupt vectors for wil_freeze_bl */
#define ARC_IRQ_VECTOR_OFFSET(N) ((N) * 8)
/* ARC long jump instruction */
#define ARC_JAL_INST (0x20200f80)
static void wil_freeze_bl(struct wil6210_priv *wil)
{
u32 jal, upc, saved;
u32 ivt3 = ARC_IRQ_VECTOR_OFFSET(3);
jal = wil_r(wil, wil->iccm_base + ivt3);
if (jal != ARC_me_imm32(ARC_JAL_INST)) {
wil_dbg_misc(wil, "invalid IVT entry found, skipping\n");
return;
}
/* prevent the target from entering deep sleep
* and disabling memory access
*/
saved = wil_r(wil, RGF_USER_USAGE_8);
wil_w(wil, RGF_USER_USAGE_8, saved | BIT_USER_PREVENT_DEEP_SLEEP);
usleep_range(20, 25); /* let the BL process the bit */
/* redirect to endless loop in the INT_L1 context and let it trap */
wil_w(wil, wil->iccm_base + ivt3 + 4, ARC_me_imm32(ivt3));
usleep_range(20, 25); /* let the BL get into the trap */
/* verify the BL is frozen */
upc = wil_r(wil, RGF_USER_CPU_PC);
if (upc < ivt3 || (upc > (ivt3 + 8)))
wil_dbg_misc(wil, "BL freeze failed, PC=0x%08X\n", upc);
wil_w(wil, RGF_USER_USAGE_8, saved);
}
static void wil_bl_prepare_halt(struct wil6210_priv *wil)
{
u32 tmp, ver;
/* before halting device CPU driver must make sure BL is not accessing
* host memory. This is done differently depending on BL version:
* 1. For very old BL versions the procedure is skipped
* (not supported).
* 2. For old BL version we use a special trick to freeze the BL
* 3. For new BL versions we shutdown the BL using handshake procedure.
*/
tmp = wil_r(wil, RGF_USER_BL +
offsetof(struct bl_dedicated_registers_v0,
boot_loader_struct_version));
if (!tmp) {
wil_dbg_misc(wil, "old BL, skipping halt preparation\n");
return;
}
tmp = wil_r(wil, RGF_USER_BL +
offsetof(struct bl_dedicated_registers_v1,
bl_shutdown_handshake));
ver = BL_SHUTDOWN_HS_PROT_VER(tmp);
if (ver > 0)
wil_shutdown_bl(wil);
else
wil_freeze_bl(wil);
}
static inline void wil_halt_cpu(struct wil6210_priv *wil)
{
if (wil->hw_version >= HW_VER_TALYN_MB) {
wil_w(wil, RGF_USER_USER_CPU_0_TALYN_MB,
BIT_USER_USER_CPU_MAN_RST);
wil_w(wil, RGF_USER_MAC_CPU_0_TALYN_MB,
BIT_USER_MAC_CPU_MAN_RST);
} else {
wil_w(wil, RGF_USER_USER_CPU_0, BIT_USER_USER_CPU_MAN_RST);
wil_w(wil, RGF_USER_MAC_CPU_0, BIT_USER_MAC_CPU_MAN_RST);
}
}
static inline void wil_release_cpu(struct wil6210_priv *wil)
{
/* Start CPU */
if (wil->hw_version >= HW_VER_TALYN_MB)
wil_w(wil, RGF_USER_USER_CPU_0_TALYN_MB, 1);
else
wil_w(wil, RGF_USER_USER_CPU_0, 1);
}
static void wil_set_oob_mode(struct wil6210_priv *wil, u8 mode)
{
wil_info(wil, "oob_mode to %d\n", mode);
switch (mode) {
case 0:
wil_c(wil, RGF_USER_USAGE_6, BIT_USER_OOB_MODE |
BIT_USER_OOB_R2_MODE);
break;
case 1:
wil_c(wil, RGF_USER_USAGE_6, BIT_USER_OOB_R2_MODE);
wil_s(wil, RGF_USER_USAGE_6, BIT_USER_OOB_MODE);
break;
case 2:
wil_c(wil, RGF_USER_USAGE_6, BIT_USER_OOB_MODE);
wil_s(wil, RGF_USER_USAGE_6, BIT_USER_OOB_R2_MODE);
break;
default:
wil_err(wil, "invalid oob_mode: %d\n", mode);
}
}
static int wil_wait_device_ready(struct wil6210_priv *wil, int no_flash)
{
int delay = 0;
u32 x, x1 = 0;
/* wait until device ready. */
if (no_flash) {
msleep(PMU_READY_DELAY_MS);
wil_dbg_misc(wil, "Reset completed\n");
} else {
do {
msleep(RST_DELAY);
x = wil_r(wil, RGF_USER_BL +
offsetof(struct bl_dedicated_registers_v0,
boot_loader_ready));
if (x1 != x) {
wil_dbg_misc(wil, "BL.ready 0x%08x => 0x%08x\n",
x1, x);
x1 = x;
}
if (delay++ > RST_COUNT) {
wil_err(wil, "Reset not completed, bl.ready 0x%08x\n",
x);
return -ETIME;
}
} while (x != BL_READY);
wil_dbg_misc(wil, "Reset completed in %d ms\n",
delay * RST_DELAY);
}
return 0;
}
static int wil_wait_device_ready_talyn_mb(struct wil6210_priv *wil)
{
u32 otp_hw;
u8 signature_status;
bool otp_signature_err;
bool hw_section_done;
u32 otp_qc_secured;
int delay = 0;
/* Wait for OTP signature test to complete */
usleep_range(2000, 2200);
wil->boot_config = WIL_BOOT_ERR;
/* Poll until OTP signature status is valid.
* In vanilla and development modes, when signature test is complete
* HW sets BIT_OTP_SIGNATURE_ERR_TALYN_MB.
* In production mode BIT_OTP_SIGNATURE_ERR_TALYN_MB remains 0, poll
* for signature status change to 2 or 3.
*/
do {
otp_hw = wil_r(wil, RGF_USER_OTP_HW_RD_MACHINE_1);
signature_status = WIL_GET_BITS(otp_hw, 8, 9);
otp_signature_err = otp_hw & BIT_OTP_SIGNATURE_ERR_TALYN_MB;
if (otp_signature_err &&
signature_status == WIL_SIG_STATUS_VANILLA) {
wil->boot_config = WIL_BOOT_VANILLA;
break;
}
if (otp_signature_err &&
signature_status == WIL_SIG_STATUS_DEVELOPMENT) {
wil->boot_config = WIL_BOOT_DEVELOPMENT;
break;
}
if (!otp_signature_err &&
signature_status == WIL_SIG_STATUS_PRODUCTION) {
wil->boot_config = WIL_BOOT_PRODUCTION;
break;
}
if (!otp_signature_err &&
signature_status ==
WIL_SIG_STATUS_CORRUPTED_PRODUCTION) {
/* Unrecognized OTP signature found. Possibly a
* corrupted production signature, access control
* is applied as in production mode, therefore
* do not fail
*/
wil->boot_config = WIL_BOOT_PRODUCTION;
break;
}
if (delay++ > OTP_HW_COUNT)
break;
usleep_range(OTP_HW_DELAY, OTP_HW_DELAY + 10);
} while (!otp_signature_err && signature_status == 0);
if (wil->boot_config == WIL_BOOT_ERR) {
wil_err(wil,
"invalid boot config, signature_status %d otp_signature_err %d\n",
signature_status, otp_signature_err);
return -ETIME;
}
wil_dbg_misc(wil,
"signature test done in %d usec, otp_hw 0x%x, boot_config %d\n",
delay * OTP_HW_DELAY, otp_hw, wil->boot_config);
if (wil->boot_config == WIL_BOOT_VANILLA)
/* Assuming not SPI boot (currently not supported) */
goto out;
hw_section_done = otp_hw & BIT_OTP_HW_SECTION_DONE_TALYN_MB;
delay = 0;
while (!hw_section_done) {
msleep(RST_DELAY);
otp_hw = wil_r(wil, RGF_USER_OTP_HW_RD_MACHINE_1);
hw_section_done = otp_hw & BIT_OTP_HW_SECTION_DONE_TALYN_MB;
if (delay++ > RST_COUNT) {
wil_err(wil, "TO waiting for hw_section_done\n");
return -ETIME;
}
}
wil_dbg_misc(wil, "HW section done in %d ms\n", delay * RST_DELAY);
otp_qc_secured = wil_r(wil, RGF_OTP_QC_SECURED);
wil->secured_boot = otp_qc_secured & BIT_BOOT_FROM_ROM ? 1 : 0;
wil_dbg_misc(wil, "secured boot is %sabled\n",
wil->secured_boot ? "en" : "dis");
out:
wil_dbg_misc(wil, "Reset completed\n");
return 0;
}
static int wil_target_reset(struct wil6210_priv *wil, int no_flash)
{
u32 x;
int rc;
wil_dbg_misc(wil, "Resetting \"%s\"...\n", wil->hw_name);
if (wil->hw_version < HW_VER_TALYN) {
/* Clear MAC link up */
wil_s(wil, RGF_HP_CTRL, BIT(15));
wil_s(wil, RGF_USER_CLKS_CTL_SW_RST_MASK_0,
BIT_HPAL_PERST_FROM_PAD);
wil_s(wil, RGF_USER_CLKS_CTL_SW_RST_MASK_0, BIT_CAR_PERST_RST);
}
wil_halt_cpu(wil);
if (!no_flash) {
/* clear all boot loader "ready" bits */
wil_w(wil, RGF_USER_BL +
offsetof(struct bl_dedicated_registers_v0,
boot_loader_ready), 0);
/* this should be safe to write even with old BLs */
wil_w(wil, RGF_USER_BL +
offsetof(struct bl_dedicated_registers_v1,
bl_shutdown_handshake), 0);
}
/* Clear Fw Download notification */
wil_c(wil, RGF_USER_USAGE_6, BIT(0));
wil_s(wil, RGF_CAF_OSC_CONTROL, BIT_CAF_OSC_XTAL_EN);
/* XTAL stabilization should take about 3ms */
usleep_range(5000, 7000);
x = wil_r(wil, RGF_CAF_PLL_LOCK_STATUS);
if (!(x & BIT_CAF_OSC_DIG_XTAL_STABLE)) {
wil_err(wil, "Xtal stabilization timeout\n"
"RGF_CAF_PLL_LOCK_STATUS = 0x%08x\n", x);
return -ETIME;
}
/* switch 10k to XTAL*/
wil_c(wil, RGF_USER_SPARROW_M_4, BIT_SPARROW_M_4_SEL_SLEEP_OR_REF);
/* 40 MHz */
wil_c(wil, RGF_USER_CLKS_CTL_0, BIT_USER_CLKS_CAR_AHB_SW_SEL);
wil_w(wil, RGF_USER_CLKS_CTL_EXT_SW_RST_VEC_0, 0x3ff81f);
wil_w(wil, RGF_USER_CLKS_CTL_EXT_SW_RST_VEC_1, 0xf);
if (wil->hw_version >= HW_VER_TALYN_MB) {
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_2, 0x7e000000);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_1, 0x0000003f);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_3, 0xc00000f0);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_0, 0xffe7fe00);
} else {
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_2, 0xfe000000);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_1, 0x0000003f);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_3, 0x000000f0);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_0, 0xffe7fe00);
}
wil_w(wil, RGF_USER_CLKS_CTL_EXT_SW_RST_VEC_0, 0x0);
wil_w(wil, RGF_USER_CLKS_CTL_EXT_SW_RST_VEC_1, 0x0);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_3, 0);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_2, 0);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_1, 0);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_0, 0);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_3, 0x00000003);
/* reset A2 PCIE AHB */
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_2, 0x00008000);
wil_w(wil, RGF_USER_CLKS_CTL_SW_RST_VEC_0, 0);
if (wil->hw_version == HW_VER_TALYN_MB)
rc = wil_wait_device_ready_talyn_mb(wil);
else
rc = wil_wait_device_ready(wil, no_flash);
if (rc)
return rc;
wil_c(wil, RGF_USER_CLKS_CTL_0, BIT_USER_CLKS_RST_PWGD);
/* enable fix for HW bug related to the SA/DA swap in AP Rx */
wil_s(wil, RGF_DMA_OFUL_NID_0, BIT_DMA_OFUL_NID_0_RX_EXT_TR_EN |
BIT_DMA_OFUL_NID_0_RX_EXT_A3_SRC);
if (wil->hw_version < HW_VER_TALYN_MB && no_flash) {
/* Reset OTP HW vectors to fit 40MHz */
wil_w(wil, RGF_USER_XPM_IFC_RD_TIME1, 0x60001);
wil_w(wil, RGF_USER_XPM_IFC_RD_TIME2, 0x20027);
wil_w(wil, RGF_USER_XPM_IFC_RD_TIME3, 0x1);
wil_w(wil, RGF_USER_XPM_IFC_RD_TIME4, 0x20027);
wil_w(wil, RGF_USER_XPM_IFC_RD_TIME5, 0x30003);
wil_w(wil, RGF_USER_XPM_IFC_RD_TIME6, 0x20002);
wil_w(wil, RGF_USER_XPM_IFC_RD_TIME7, 0x60001);
wil_w(wil, RGF_USER_XPM_IFC_RD_TIME8, 0x60001);
wil_w(wil, RGF_USER_XPM_IFC_RD_TIME9, 0x60001);
wil_w(wil, RGF_USER_XPM_IFC_RD_TIME10, 0x60001);
wil_w(wil, RGF_USER_XPM_RD_DOUT_SAMPLE_TIME, 0x57);
}
return 0;
}
static void wil_collect_fw_info(struct wil6210_priv *wil)
{
struct wiphy *wiphy = wil_to_wiphy(wil);
u8 retry_short;
int rc;
wil_refresh_fw_capabilities(wil);
rc = wmi_get_mgmt_retry(wil, &retry_short);
if (!rc) {
wiphy->retry_short = retry_short;
wil_dbg_misc(wil, "FW retry_short: %d\n", retry_short);
}
}
void wil_refresh_fw_capabilities(struct wil6210_priv *wil)
{
struct wiphy *wiphy = wil_to_wiphy(wil);
int features;
wil->keep_radio_on_during_sleep =
test_bit(WIL_PLATFORM_CAPA_RADIO_ON_IN_SUSPEND,
wil->platform_capa) &&
test_bit(WMI_FW_CAPABILITY_D3_SUSPEND, wil->fw_capabilities);
wil_info(wil, "keep_radio_on_during_sleep (%d)\n",
wil->keep_radio_on_during_sleep);
if (test_bit(WMI_FW_CAPABILITY_RSSI_REPORTING, wil->fw_capabilities))
wiphy->signal_type = CFG80211_SIGNAL_TYPE_MBM;
else
wiphy->signal_type = CFG80211_SIGNAL_TYPE_UNSPEC;
if (test_bit(WMI_FW_CAPABILITY_PNO, wil->fw_capabilities)) {
wiphy->max_sched_scan_reqs = 1;
wiphy->max_sched_scan_ssids = WMI_MAX_PNO_SSID_NUM;
wiphy->max_match_sets = WMI_MAX_PNO_SSID_NUM;
wiphy->max_sched_scan_ie_len = WMI_MAX_IE_LEN;
wiphy->max_sched_scan_plans = WMI_MAX_PLANS_NUM;
}
if (test_bit(WMI_FW_CAPABILITY_TX_REQ_EXT, wil->fw_capabilities))
wiphy->flags |= WIPHY_FLAG_OFFCHAN_TX;
if (wil->platform_ops.set_features) {
features = (test_bit(WMI_FW_CAPABILITY_REF_CLOCK_CONTROL,
wil->fw_capabilities) &&
test_bit(WIL_PLATFORM_CAPA_EXT_CLK,
wil->platform_capa)) ?
BIT(WIL_PLATFORM_FEATURE_FW_EXT_CLK_CONTROL) : 0;
if (wil->n_msi == 3)
features |= BIT(WIL_PLATFORM_FEATURE_TRIPLE_MSI);
wil->platform_ops.set_features(wil->platform_handle, features);
}
if (test_bit(WMI_FW_CAPABILITY_BACK_WIN_SIZE_64,
wil->fw_capabilities)) {
wil->max_agg_wsize = WIL_MAX_AGG_WSIZE_64;
wil->max_ampdu_size = WIL_MAX_AMPDU_SIZE_128;
} else {
wil->max_agg_wsize = WIL_MAX_AGG_WSIZE;
wil->max_ampdu_size = WIL_MAX_AMPDU_SIZE;
}
update_supported_bands(wil);
}
void wil_mbox_ring_le2cpus(struct wil6210_mbox_ring *r)
{
le32_to_cpus(&r->base);
le16_to_cpus(&r->entry_size);
le16_to_cpus(&r->size);
le32_to_cpus(&r->tail);
le32_to_cpus(&r->head);
}
/* construct actual board file name to use */
void wil_get_board_file(struct wil6210_priv *wil, char *buf, size_t len)
{
const char *board_file;
const char *wil_talyn_fw_name = ftm_mode ? WIL_FW_NAME_FTM_TALYN :
WIL_FW_NAME_TALYN;
if (wil->board_file) {
board_file = wil->board_file;
} else {
/* If specific FW file is used for Talyn,
* use specific board file
*/
if (strcmp(wil->wil_fw_name, wil_talyn_fw_name) == 0)
board_file = WIL_BRD_NAME_TALYN;
else
board_file = WIL_BOARD_FILE_NAME;
}
strlcpy(buf, board_file, len);
}
static int wil_get_bl_info(struct wil6210_priv *wil)
{
struct net_device *ndev = wil->main_ndev;
struct wiphy *wiphy = wil_to_wiphy(wil);
union {
struct bl_dedicated_registers_v0 bl0;
struct bl_dedicated_registers_v1 bl1;
} bl;
u32 bl_ver;
u8 *mac;
u16 rf_status;
wil_memcpy_fromio_32(&bl, wil->csr + HOSTADDR(RGF_USER_BL),
sizeof(bl));
bl_ver = le32_to_cpu(bl.bl0.boot_loader_struct_version);
mac = bl.bl0.mac_address;
if (bl_ver == 0) {
le32_to_cpus(&bl.bl0.rf_type);
le32_to_cpus(&bl.bl0.baseband_type);
rf_status = 0; /* actually, unknown */
wil_info(wil,
"Boot Loader struct v%d: MAC = %pM RF = 0x%08x bband = 0x%08x\n",
bl_ver, mac,
bl.bl0.rf_type, bl.bl0.baseband_type);
wil_info(wil, "Boot Loader build unknown for struct v0\n");
} else {
le16_to_cpus(&bl.bl1.rf_type);
rf_status = le16_to_cpu(bl.bl1.rf_status);
le32_to_cpus(&bl.bl1.baseband_type);
le16_to_cpus(&bl.bl1.bl_version_subminor);
le16_to_cpus(&bl.bl1.bl_version_build);
wil_info(wil,
"Boot Loader struct v%d: MAC = %pM RF = 0x%04x (status 0x%04x) bband = 0x%08x\n",
bl_ver, mac,
bl.bl1.rf_type, rf_status,
bl.bl1.baseband_type);
wil_info(wil, "Boot Loader build %d.%d.%d.%d\n",
bl.bl1.bl_version_major, bl.bl1.bl_version_minor,
bl.bl1.bl_version_subminor, bl.bl1.bl_version_build);
}
if (!is_valid_ether_addr(mac)) {
wil_err(wil, "BL: Invalid MAC %pM\n", mac);
return -EINVAL;
}
ether_addr_copy(ndev->perm_addr, mac);
ether_addr_copy(wiphy->perm_addr, mac);
if (!is_valid_ether_addr(ndev->dev_addr))
ether_addr_copy(ndev->dev_addr, mac);
if (rf_status) {/* bad RF cable? */
wil_err(wil, "RF communication error 0x%04x",
rf_status);
return -EAGAIN;
}
return 0;
}
static void wil_bl_crash_info(struct wil6210_priv *wil, bool is_err)
{
u32 bl_assert_code, bl_assert_blink, bl_magic_number;
u32 bl_ver = wil_r(wil, RGF_USER_BL +
offsetof(struct bl_dedicated_registers_v0,
boot_loader_struct_version));
if (bl_ver < 2)
return;
bl_assert_code = wil_r(wil, RGF_USER_BL +
offsetof(struct bl_dedicated_registers_v1,
bl_assert_code));
bl_assert_blink = wil_r(wil, RGF_USER_BL +
offsetof(struct bl_dedicated_registers_v1,
bl_assert_blink));
bl_magic_number = wil_r(wil, RGF_USER_BL +
offsetof(struct bl_dedicated_registers_v1,
bl_magic_number));
if (is_err) {
wil_err(wil,
"BL assert code 0x%08x blink 0x%08x magic 0x%08x\n",
bl_assert_code, bl_assert_blink, bl_magic_number);
} else {
wil_dbg_misc(wil,
"BL assert code 0x%08x blink 0x%08x magic 0x%08x\n",
bl_assert_code, bl_assert_blink, bl_magic_number);
}
}
static int wil_get_otp_info(struct wil6210_priv *wil)
{
struct net_device *ndev = wil->main_ndev;
struct wiphy *wiphy = wil_to_wiphy(wil);
u8 mac[8];
int mac_addr;
if (wil->hw_version >= HW_VER_TALYN_MB)
mac_addr = RGF_OTP_MAC_TALYN_MB;
else
mac_addr = RGF_OTP_MAC;
wil_memcpy_fromio_32(mac, wil->csr + HOSTADDR(mac_addr),
sizeof(mac));
if (!is_valid_ether_addr(mac)) {
wil_err(wil, "Invalid MAC %pM\n", mac);
return -EINVAL;
}
ether_addr_copy(ndev->perm_addr, mac);
ether_addr_copy(wiphy->perm_addr, mac);
if (!is_valid_ether_addr(ndev->dev_addr))
ether_addr_copy(ndev->dev_addr, mac);
return 0;
}
static int wil_wait_for_fw_ready(struct wil6210_priv *wil)
{
ulong to = msecs_to_jiffies(2000);
ulong left = wait_for_completion_timeout(&wil->wmi_ready, to);
if (0 == left) {
wil_err(wil, "Firmware not ready\n");
return -ETIME;
} else {
wil_info(wil, "FW ready after %d ms. HW version 0x%08x\n",
jiffies_to_msecs(to-left), wil->hw_version);
}
return 0;
}
void wil_abort_scan(struct wil6210_vif *vif, bool sync)
{
struct wil6210_priv *wil = vif_to_wil(vif);
int rc;
struct cfg80211_scan_info info = {
.aborted = true,
};
lockdep_assert_held(&wil->vif_mutex);
if (!vif->scan_request)
return;
wil_dbg_misc(wil, "Abort scan_request 0x%p\n", vif->scan_request);
del_timer_sync(&vif->scan_timer);
mutex_unlock(&wil->vif_mutex);
rc = wmi_abort_scan(vif);
if (!rc && sync)
wait_event_interruptible_timeout(wil->wq, !vif->scan_request,
msecs_to_jiffies(
WAIT_FOR_SCAN_ABORT_MS));
mutex_lock(&wil->vif_mutex);
if (vif->scan_request) {
cfg80211_scan_done(vif->scan_request, &info);
vif->scan_request = NULL;
}
}
void wil_abort_scan_all_vifs(struct wil6210_priv *wil, bool sync)
{
int i;
lockdep_assert_held(&wil->vif_mutex);
for (i = 0; i < wil->max_vifs; i++) {
struct wil6210_vif *vif = wil->vifs[i];
if (vif)
wil_abort_scan(vif, sync);
}
}
int wil_ps_update(struct wil6210_priv *wil, enum wmi_ps_profile_type ps_profile)
{
int rc;
if (!test_bit(WMI_FW_CAPABILITY_PS_CONFIG, wil->fw_capabilities)) {
wil_err(wil, "set_power_mgmt not supported\n");
return -EOPNOTSUPP;
}
rc = wmi_ps_dev_profile_cfg(wil, ps_profile);
if (rc)
wil_err(wil, "wmi_ps_dev_profile_cfg failed (%d)\n", rc);
else
wil->ps_profile = ps_profile;
return rc;
}
static void wil_pre_fw_config(struct wil6210_priv *wil)
{
/* Mark FW as loaded from host */
wil_s(wil, RGF_USER_USAGE_6, 1);
/* clear any interrupts which on-card-firmware
* may have set
*/
wil6210_clear_irq(wil);
/* CAF_ICR - clear and mask */
/* it is W1C, clear by writing back same value */
if (wil->hw_version < HW_VER_TALYN_MB) {
wil_s(wil, RGF_CAF_ICR + offsetof(struct RGF_ICR, ICR), 0);
wil_w(wil, RGF_CAF_ICR + offsetof(struct RGF_ICR, IMV), ~0);
} else {
wil_s(wil,
RGF_CAF_ICR_TALYN_MB + offsetof(struct RGF_ICR, ICR), 0);
wil_w(wil, RGF_CAF_ICR_TALYN_MB +
offsetof(struct RGF_ICR, IMV), ~0);
}
/* clear PAL_UNIT_ICR (potential D0->D3 leftover)
* In Talyn-MB host cannot access this register due to
* access control, hence PAL_UNIT_ICR is cleared by the FW
*/
if (wil->hw_version < HW_VER_TALYN_MB)
wil_s(wil, RGF_PAL_UNIT_ICR + offsetof(struct RGF_ICR, ICR),
0);
if (wil->fw_calib_result > 0) {
__le32 val = cpu_to_le32(wil->fw_calib_result |
(CALIB_RESULT_SIGNATURE << 8));
wil_w(wil, RGF_USER_FW_CALIB_RESULT, (u32 __force)val);
}
}
static int wil_restore_vifs(struct wil6210_priv *wil)
{
struct wil6210_vif *vif;
struct net_device *ndev;
struct wireless_dev *wdev;
int i, rc;
for (i = 0; i < wil->max_vifs; i++) {
vif = wil->vifs[i];
if (!vif)
continue;
vif->ap_isolate = 0;
if (vif->mid) {
ndev = vif_to_ndev(vif);
wdev = vif_to_wdev(vif);
rc = wmi_port_allocate(wil, vif->mid, ndev->dev_addr,
wdev->iftype);
if (rc) {
wil_err(wil, "fail to restore VIF %d type %d, rc %d\n",
i, wdev->iftype, rc);
return rc;
}
}
}
return 0;
}
/*
* We reset all the structures, and we reset the UMAC.
* After calling this routine, you're expected to reload
* the firmware.
*/
int wil_reset(struct wil6210_priv *wil, bool load_fw)
{
int rc, i;
unsigned long status_flags = BIT(wil_status_resetting);
int no_flash;
struct wil6210_vif *vif;
wil_dbg_misc(wil, "reset\n");
WARN_ON(!mutex_is_locked(&wil->mutex));
WARN_ON(test_bit(wil_status_napi_en, wil->status));
if (debug_fw) {
static const u8 mac[ETH_ALEN] = {
0x00, 0xde, 0xad, 0x12, 0x34, 0x56,
};
struct net_device *ndev = wil->main_ndev;
ether_addr_copy(ndev->perm_addr, mac);
ether_addr_copy(ndev->dev_addr, ndev->perm_addr);
return 0;
}
if (wil->hw_version == HW_VER_UNKNOWN)
return -ENODEV;
if (test_bit(WIL_PLATFORM_CAPA_T_PWR_ON_0, wil->platform_capa)) {
wil_dbg_misc(wil, "Notify FW to set T_POWER_ON=0\n");
wil_s(wil, RGF_USER_USAGE_8, BIT_USER_SUPPORT_T_POWER_ON_0);
}
if (test_bit(WIL_PLATFORM_CAPA_EXT_CLK, wil->platform_capa)) {
wil_dbg_misc(wil, "Notify FW on ext clock configuration\n");
wil_s(wil, RGF_USER_USAGE_8, BIT_USER_EXT_CLK);
}
if (wil->platform_ops.notify) {
rc = wil->platform_ops.notify(wil->platform_handle,
WIL_PLATFORM_EVT_PRE_RESET);
if (rc)
wil_err(wil, "PRE_RESET platform notify failed, rc %d\n",
rc);
}
set_bit(wil_status_resetting, wil->status);
if (test_bit(wil_status_collecting_dumps, wil->status)) {
/* Device collects crash dump, cancel the reset.
* following crash dump collection, reset would take place.
*/
wil_dbg_misc(wil, "reject reset while collecting crash dump\n");
rc = -EBUSY;
goto out;
}
mutex_lock(&wil->vif_mutex);
wil_abort_scan_all_vifs(wil, false);
mutex_unlock(&wil->vif_mutex);
for (i = 0; i < wil->max_vifs; i++) {
vif = wil->vifs[i];
if (vif) {
cancel_work_sync(&vif->disconnect_worker);
wil6210_disconnect(vif, NULL,
WLAN_REASON_DEAUTH_LEAVING);
}
}
wil_bcast_fini_all(wil);
/* Disable device led before reset*/
wmi_led_cfg(wil, false);
/* prevent NAPI from being scheduled and prevent wmi commands */
mutex_lock(&wil->wmi_mutex);
if (test_bit(wil_status_suspending, wil->status))
status_flags |= BIT(wil_status_suspending);
bitmap_and(wil->status, wil->status, &status_flags,
wil_status_last);
wil_dbg_misc(wil, "wil->status (0x%lx)\n", *wil->status);
mutex_unlock(&wil->wmi_mutex);
wil_mask_irq(wil);
wmi_event_flush(wil);
wil6210: ADDBA/DELBA flows Introduce BACK establishment procedures; decision logic is not implemented yet; debugfs entry 'addba' used to manually trigger addba/delba for ringid 0. debugfs usage: to establish BACK with agg_wsize 16: echo 16 > /sys/kernel/debug/ieee80211/phy0/wil6210/addba to delete BACK: echo 0 > /sys/kernel/debug/ieee80211/phy0/wil6210/addba to change agg_wsize, one need to delete BACK and establish it anew ADDBA flow for: - originator Tx side (initiator) sends WMI_VRING_BA_EN_CMDID providing agg_wsize and timeout parameters. Eventually, it gets event confirming BACK agreement - WMI_BA_STATUS_EVENTID with negotiated parameters. On this event, update Tx vring data (struct vring_tx_data) and display BACK parameters on debugfs - recipient Rx side (recipient) firmware informs driver about ADDBA with WMI_RCP_ADDBA_REQ_EVENTID, driver process it in service work queue wq_service. It adjusts parameters and sends response with WMI_RCP_ADDBA_RESP_CMDID, and final confirmation provided by firmware with WMI_ADDBA_RESP_SENT_EVENTID. In case of success, driver updates Rx BACK reorder buffer. policy for BACK parameters: - aggregation size (agg_wsize * MPDUsize)) to not exceed 64Kbytes DELBA flow for: - originator driver decides to terminate BACK, it sends WMI_VRING_BA_DIS_CMDID and updates struct vring_tx_data associated with vring; ignore WMI_DELBA_EVENTID. - recipient firmware informs driver with WMI_DELBA_EVENTID, driver deletes correspondent reorder buffer ADDBA request processing requires sending WMI command, therefore it is processed in work queue context. Same work queue used as for connect, it get renamed to wq_service Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2014-12-23 15:47:03 +08:00
flush_workqueue(wil->wq_service);
flush_workqueue(wil->wmi_wq);
no_flash = test_bit(hw_capa_no_flash, wil->hw_capa);
if (!no_flash)
wil_bl_crash_info(wil, false);
wil_disable_irq(wil);
rc = wil_target_reset(wil, no_flash);
wil6210_clear_irq(wil);
wil_enable_irq(wil);
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
wil->txrx_ops.rx_fini(wil);
wil->txrx_ops.tx_fini(wil);
if (rc) {
if (!no_flash)
wil_bl_crash_info(wil, true);
goto out;
}
if (no_flash) {
rc = wil_get_otp_info(wil);
} else {
rc = wil_get_bl_info(wil);
if (rc == -EAGAIN && !load_fw)
/* ignore RF error if not going up */
rc = 0;
}
if (rc)
goto out;
wil_set_oob_mode(wil, oob_mode);
if (load_fw) {
char board_file[WIL_BOARD_FILE_MAX_NAMELEN];
if (wil->secured_boot) {
wil_err(wil, "secured boot is not supported\n");
return -ENOTSUPP;
}
board_file[0] = '\0';
wil_get_board_file(wil, board_file, sizeof(board_file));
wil_info(wil, "Use firmware <%s> + board <%s>\n",
wil->wil_fw_name, board_file);
if (!no_flash)
wil_bl_prepare_halt(wil);
wil_halt_cpu(wil);
memset(wil->fw_version, 0, sizeof(wil->fw_version));
/* Loading f/w from the file */
rc = wil_request_firmware(wil, wil->wil_fw_name, true);
if (rc)
goto out;
if (wil->brd_file_addr)
rc = wil_request_board(wil, board_file);
else
rc = wil_request_firmware(wil, board_file, true);
if (rc)
goto out;
wil_pre_fw_config(wil);
wil_release_cpu(wil);
}
/* init after reset */
reinit_completion(&wil->wmi_ready);
reinit_completion(&wil->wmi_call);
reinit_completion(&wil->halp.comp);
clear_bit(wil_status_resetting, wil->status);
if (load_fw) {
wil_unmask_irq(wil);
/* we just started MAC, wait for FW ready */
rc = wil_wait_for_fw_ready(wil);
if (rc)
return rc;
/* check FW is responsive */
rc = wmi_echo(wil);
if (rc) {
wil_err(wil, "wmi_echo failed, rc %d\n", rc);
return rc;
}
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
wil->txrx_ops.configure_interrupt_moderation(wil);
/* Enable OFU rdy valid bug fix, to prevent hang in oful34_rx
* while there is back-pressure from Host during RX
*/
if (wil->hw_version >= HW_VER_TALYN_MB)
wil_s(wil, RGF_DMA_MISC_CTL,
BIT_OFUL34_RDY_VALID_BUG_FIX_EN);
rc = wil_restore_vifs(wil);
if (rc) {
wil_err(wil, "failed to restore vifs, rc %d\n", rc);
return rc;
}
wil_collect_fw_info(wil);
if (wil->ps_profile != WMI_PS_PROFILE_TYPE_DEFAULT)
wil_ps_update(wil, wil->ps_profile);
if (wil->platform_ops.notify) {
rc = wil->platform_ops.notify(wil->platform_handle,
WIL_PLATFORM_EVT_FW_RDY);
if (rc) {
wil_err(wil, "FW_RDY notify failed, rc %d\n",
rc);
rc = 0;
}
}
}
return rc;
out:
clear_bit(wil_status_resetting, wil->status);
return rc;
}
void wil_fw_error_recovery(struct wil6210_priv *wil)
{
wil_dbg_misc(wil, "starting fw error recovery\n");
if (test_bit(wil_status_resetting, wil->status)) {
wil_info(wil, "Reset already in progress\n");
return;
}
wil->recovery_state = fw_recovery_pending;
schedule_work(&wil->fw_error_worker);
}
wil6210: fix for oops while stopping interface When interface stopped while running intensive Rx traffic, the following oops observed: [89846.734683] Call trace: [89846.737117] [<ffffffc00083aa64>] dev_gro_receive+0xac/0x358 [89846.742674] [<ffffffc00083ae94>] napi_gro_receive+0x24/0xa4 [89846.748251] [<ffffffbffc1c2f88>] $x+0xec/0x1f8 [wil6210] wil_netif_rx_any [89846.753547] [<ffffffbffc1c4830>] $x+0x34/0x54 [wil6210] wil_release_reorder_frame [89846.758755] [<ffffffbffc1c48ac>] wil_release_reorder_frames+0x5c/0x78 [wil6210] [89846.766044] [<ffffffbffc1c4bf8>] wil_tid_ampdu_rx_free+0x20/0x48 [wil6210] [89846.772901] [<ffffffbffc1bedc8>] $x+0x190/0x1e8 [wil6210] [89846.778285] [<ffffffbffc1c0ed4>] wmi_event_worker+0x230/0x2f8 [wil6210] [89846.784865] [<ffffffc0000b0bc8>] process_one_work+0x278/0x3fc [89846.790591] [<ffffffc0000b1218>] worker_thread+0x200/0x330 [89846.796060] [<ffffffc0000b6664>] kthread+0xac/0xb8 [89846.800836] Code: b940c661 f9406a62 8b010041 f9400026 (f8636882) [89846.807008] ---[ end trace d6fdc17cd27d18f6 ]--- Reason is the following: when removing Rx vring (wil_netdev_ops.ndo_stop -> wil_stop -> wil_down -> __wil_down -> wil_rx_fini), Rx interrupt occurs. It trigger Rx NAPI, calling wil_rx_handle() that reaps (already cleaned) buffer, causing skb referring to garbage memory being set into reorder buffer. Then, network stack trying to access this buffer and fails. Prevent Rx NAPI from being scheduled if device going to stop. Bit wil_status_napi_en reflects NAPI enablement state, check it when triggering Rx NAPI. Testing shows that check for wil_status_napi_en sometimes gets negative, and new error message get printed - in this case kernel oops would be observed. Original oops is no more reproducible. This change requires also changes in the AP flows. Properly enable/disable NAPI for the AP. Make sure Rx VRING is disabled when resetting target. For this, promote __wil_up() and __wil_down() to the module scope, and use it in the relevant flows. Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2014-09-10 21:34:50 +08:00
int __wil_up(struct wil6210_priv *wil)
{
struct net_device *ndev = wil->main_ndev;
struct wireless_dev *wdev = ndev->ieee80211_ptr;
int rc;
WARN_ON(!mutex_is_locked(&wil->mutex));
rc = wil_reset(wil, true);
if (rc)
return rc;
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
/* Rx RING. After MAC and beacon */
if (rx_ring_order == 0)
rx_ring_order = wil->hw_version < HW_VER_TALYN_MB ?
WIL_RX_RING_SIZE_ORDER_DEFAULT :
WIL_RX_RING_SIZE_ORDER_TALYN_DEFAULT;
rc = wil->txrx_ops.rx_init(wil, rx_ring_order);
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
if (rc)
return rc;
rc = wil->txrx_ops.tx_init(wil);
if (rc)
return rc;
switch (wdev->iftype) {
case NL80211_IFTYPE_STATION:
wil_dbg_misc(wil, "type: STATION\n");
ndev->type = ARPHRD_ETHER;
break;
case NL80211_IFTYPE_AP:
wil_dbg_misc(wil, "type: AP\n");
ndev->type = ARPHRD_ETHER;
break;
case NL80211_IFTYPE_P2P_CLIENT:
wil_dbg_misc(wil, "type: P2P_CLIENT\n");
ndev->type = ARPHRD_ETHER;
break;
case NL80211_IFTYPE_P2P_GO:
wil_dbg_misc(wil, "type: P2P_GO\n");
ndev->type = ARPHRD_ETHER;
break;
case NL80211_IFTYPE_MONITOR:
wil_dbg_misc(wil, "type: Monitor\n");
ndev->type = ARPHRD_IEEE80211_RADIOTAP;
/* ARPHRD_IEEE80211 or ARPHRD_IEEE80211_RADIOTAP ? */
break;
default:
return -EOPNOTSUPP;
}
/* MAC address - pre-requisite for other commands */
wmi_set_mac_address(wil, ndev->dev_addr);
wil6210: fix for oops while stopping interface When interface stopped while running intensive Rx traffic, the following oops observed: [89846.734683] Call trace: [89846.737117] [<ffffffc00083aa64>] dev_gro_receive+0xac/0x358 [89846.742674] [<ffffffc00083ae94>] napi_gro_receive+0x24/0xa4 [89846.748251] [<ffffffbffc1c2f88>] $x+0xec/0x1f8 [wil6210] wil_netif_rx_any [89846.753547] [<ffffffbffc1c4830>] $x+0x34/0x54 [wil6210] wil_release_reorder_frame [89846.758755] [<ffffffbffc1c48ac>] wil_release_reorder_frames+0x5c/0x78 [wil6210] [89846.766044] [<ffffffbffc1c4bf8>] wil_tid_ampdu_rx_free+0x20/0x48 [wil6210] [89846.772901] [<ffffffbffc1bedc8>] $x+0x190/0x1e8 [wil6210] [89846.778285] [<ffffffbffc1c0ed4>] wmi_event_worker+0x230/0x2f8 [wil6210] [89846.784865] [<ffffffc0000b0bc8>] process_one_work+0x278/0x3fc [89846.790591] [<ffffffc0000b1218>] worker_thread+0x200/0x330 [89846.796060] [<ffffffc0000b6664>] kthread+0xac/0xb8 [89846.800836] Code: b940c661 f9406a62 8b010041 f9400026 (f8636882) [89846.807008] ---[ end trace d6fdc17cd27d18f6 ]--- Reason is the following: when removing Rx vring (wil_netdev_ops.ndo_stop -> wil_stop -> wil_down -> __wil_down -> wil_rx_fini), Rx interrupt occurs. It trigger Rx NAPI, calling wil_rx_handle() that reaps (already cleaned) buffer, causing skb referring to garbage memory being set into reorder buffer. Then, network stack trying to access this buffer and fails. Prevent Rx NAPI from being scheduled if device going to stop. Bit wil_status_napi_en reflects NAPI enablement state, check it when triggering Rx NAPI. Testing shows that check for wil_status_napi_en sometimes gets negative, and new error message get printed - in this case kernel oops would be observed. Original oops is no more reproducible. This change requires also changes in the AP flows. Properly enable/disable NAPI for the AP. Make sure Rx VRING is disabled when resetting target. For this, promote __wil_up() and __wil_down() to the module scope, and use it in the relevant flows. Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2014-09-10 21:34:50 +08:00
wil_dbg_misc(wil, "NAPI enable\n");
napi_enable(&wil->napi_rx);
napi_enable(&wil->napi_tx);
set_bit(wil_status_napi_en, wil->status);
wil6210_bus_request(wil, WIL_DEFAULT_BUS_REQUEST_KBPS);
return 0;
}
int wil_up(struct wil6210_priv *wil)
{
int rc;
wil_dbg_misc(wil, "up\n");
mutex_lock(&wil->mutex);
rc = __wil_up(wil);
mutex_unlock(&wil->mutex);
return rc;
}
wil6210: fix for oops while stopping interface When interface stopped while running intensive Rx traffic, the following oops observed: [89846.734683] Call trace: [89846.737117] [<ffffffc00083aa64>] dev_gro_receive+0xac/0x358 [89846.742674] [<ffffffc00083ae94>] napi_gro_receive+0x24/0xa4 [89846.748251] [<ffffffbffc1c2f88>] $x+0xec/0x1f8 [wil6210] wil_netif_rx_any [89846.753547] [<ffffffbffc1c4830>] $x+0x34/0x54 [wil6210] wil_release_reorder_frame [89846.758755] [<ffffffbffc1c48ac>] wil_release_reorder_frames+0x5c/0x78 [wil6210] [89846.766044] [<ffffffbffc1c4bf8>] wil_tid_ampdu_rx_free+0x20/0x48 [wil6210] [89846.772901] [<ffffffbffc1bedc8>] $x+0x190/0x1e8 [wil6210] [89846.778285] [<ffffffbffc1c0ed4>] wmi_event_worker+0x230/0x2f8 [wil6210] [89846.784865] [<ffffffc0000b0bc8>] process_one_work+0x278/0x3fc [89846.790591] [<ffffffc0000b1218>] worker_thread+0x200/0x330 [89846.796060] [<ffffffc0000b6664>] kthread+0xac/0xb8 [89846.800836] Code: b940c661 f9406a62 8b010041 f9400026 (f8636882) [89846.807008] ---[ end trace d6fdc17cd27d18f6 ]--- Reason is the following: when removing Rx vring (wil_netdev_ops.ndo_stop -> wil_stop -> wil_down -> __wil_down -> wil_rx_fini), Rx interrupt occurs. It trigger Rx NAPI, calling wil_rx_handle() that reaps (already cleaned) buffer, causing skb referring to garbage memory being set into reorder buffer. Then, network stack trying to access this buffer and fails. Prevent Rx NAPI from being scheduled if device going to stop. Bit wil_status_napi_en reflects NAPI enablement state, check it when triggering Rx NAPI. Testing shows that check for wil_status_napi_en sometimes gets negative, and new error message get printed - in this case kernel oops would be observed. Original oops is no more reproducible. This change requires also changes in the AP flows. Properly enable/disable NAPI for the AP. Make sure Rx VRING is disabled when resetting target. For this, promote __wil_up() and __wil_down() to the module scope, and use it in the relevant flows. Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2014-09-10 21:34:50 +08:00
int __wil_down(struct wil6210_priv *wil)
{
WARN_ON(!mutex_is_locked(&wil->mutex));
set_bit(wil_status_resetting, wil->status);
wil6210_bus_request(wil, 0);
wil6210: fix for oops while stopping interface When interface stopped while running intensive Rx traffic, the following oops observed: [89846.734683] Call trace: [89846.737117] [<ffffffc00083aa64>] dev_gro_receive+0xac/0x358 [89846.742674] [<ffffffc00083ae94>] napi_gro_receive+0x24/0xa4 [89846.748251] [<ffffffbffc1c2f88>] $x+0xec/0x1f8 [wil6210] wil_netif_rx_any [89846.753547] [<ffffffbffc1c4830>] $x+0x34/0x54 [wil6210] wil_release_reorder_frame [89846.758755] [<ffffffbffc1c48ac>] wil_release_reorder_frames+0x5c/0x78 [wil6210] [89846.766044] [<ffffffbffc1c4bf8>] wil_tid_ampdu_rx_free+0x20/0x48 [wil6210] [89846.772901] [<ffffffbffc1bedc8>] $x+0x190/0x1e8 [wil6210] [89846.778285] [<ffffffbffc1c0ed4>] wmi_event_worker+0x230/0x2f8 [wil6210] [89846.784865] [<ffffffc0000b0bc8>] process_one_work+0x278/0x3fc [89846.790591] [<ffffffc0000b1218>] worker_thread+0x200/0x330 [89846.796060] [<ffffffc0000b6664>] kthread+0xac/0xb8 [89846.800836] Code: b940c661 f9406a62 8b010041 f9400026 (f8636882) [89846.807008] ---[ end trace d6fdc17cd27d18f6 ]--- Reason is the following: when removing Rx vring (wil_netdev_ops.ndo_stop -> wil_stop -> wil_down -> __wil_down -> wil_rx_fini), Rx interrupt occurs. It trigger Rx NAPI, calling wil_rx_handle() that reaps (already cleaned) buffer, causing skb referring to garbage memory being set into reorder buffer. Then, network stack trying to access this buffer and fails. Prevent Rx NAPI from being scheduled if device going to stop. Bit wil_status_napi_en reflects NAPI enablement state, check it when triggering Rx NAPI. Testing shows that check for wil_status_napi_en sometimes gets negative, and new error message get printed - in this case kernel oops would be observed. Original oops is no more reproducible. This change requires also changes in the AP flows. Properly enable/disable NAPI for the AP. Make sure Rx VRING is disabled when resetting target. For this, promote __wil_up() and __wil_down() to the module scope, and use it in the relevant flows. Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2014-09-10 21:34:50 +08:00
wil_disable_irq(wil);
if (test_and_clear_bit(wil_status_napi_en, wil->status)) {
wil6210: fix for oops while stopping interface When interface stopped while running intensive Rx traffic, the following oops observed: [89846.734683] Call trace: [89846.737117] [<ffffffc00083aa64>] dev_gro_receive+0xac/0x358 [89846.742674] [<ffffffc00083ae94>] napi_gro_receive+0x24/0xa4 [89846.748251] [<ffffffbffc1c2f88>] $x+0xec/0x1f8 [wil6210] wil_netif_rx_any [89846.753547] [<ffffffbffc1c4830>] $x+0x34/0x54 [wil6210] wil_release_reorder_frame [89846.758755] [<ffffffbffc1c48ac>] wil_release_reorder_frames+0x5c/0x78 [wil6210] [89846.766044] [<ffffffbffc1c4bf8>] wil_tid_ampdu_rx_free+0x20/0x48 [wil6210] [89846.772901] [<ffffffbffc1bedc8>] $x+0x190/0x1e8 [wil6210] [89846.778285] [<ffffffbffc1c0ed4>] wmi_event_worker+0x230/0x2f8 [wil6210] [89846.784865] [<ffffffc0000b0bc8>] process_one_work+0x278/0x3fc [89846.790591] [<ffffffc0000b1218>] worker_thread+0x200/0x330 [89846.796060] [<ffffffc0000b6664>] kthread+0xac/0xb8 [89846.800836] Code: b940c661 f9406a62 8b010041 f9400026 (f8636882) [89846.807008] ---[ end trace d6fdc17cd27d18f6 ]--- Reason is the following: when removing Rx vring (wil_netdev_ops.ndo_stop -> wil_stop -> wil_down -> __wil_down -> wil_rx_fini), Rx interrupt occurs. It trigger Rx NAPI, calling wil_rx_handle() that reaps (already cleaned) buffer, causing skb referring to garbage memory being set into reorder buffer. Then, network stack trying to access this buffer and fails. Prevent Rx NAPI from being scheduled if device going to stop. Bit wil_status_napi_en reflects NAPI enablement state, check it when triggering Rx NAPI. Testing shows that check for wil_status_napi_en sometimes gets negative, and new error message get printed - in this case kernel oops would be observed. Original oops is no more reproducible. This change requires also changes in the AP flows. Properly enable/disable NAPI for the AP. Make sure Rx VRING is disabled when resetting target. For this, promote __wil_up() and __wil_down() to the module scope, and use it in the relevant flows. Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2014-09-10 21:34:50 +08:00
napi_disable(&wil->napi_rx);
napi_disable(&wil->napi_tx);
wil_dbg_misc(wil, "NAPI disable\n");
}
wil_enable_irq(wil);
mutex_lock(&wil->vif_mutex);
wil_p2p_stop_radio_operations(wil);
wil_abort_scan_all_vifs(wil, false);
mutex_unlock(&wil->vif_mutex);
return wil_reset(wil, false);
}
int wil_down(struct wil6210_priv *wil)
{
int rc;
wil_dbg_misc(wil, "down\n");
wil_set_recovery_state(wil, fw_recovery_idle);
mutex_lock(&wil->mutex);
rc = __wil_down(wil);
mutex_unlock(&wil->mutex);
return rc;
}
int wil_find_cid(struct wil6210_priv *wil, u8 mid, const u8 *mac)
{
int i;
int rc = -ENOENT;
for (i = 0; i < ARRAY_SIZE(wil->sta); i++) {
if (wil->sta[i].mid == mid &&
wil->sta[i].status != wil_sta_unused &&
ether_addr_equal(wil->sta[i].addr, mac)) {
rc = i;
break;
}
}
return rc;
}
void wil_halp_vote(struct wil6210_priv *wil)
{
unsigned long rc;
unsigned long to_jiffies = msecs_to_jiffies(WAIT_FOR_HALP_VOTE_MS);
mutex_lock(&wil->halp.lock);
wil_dbg_irq(wil, "halp_vote: start, HALP ref_cnt (%d)\n",
wil->halp.ref_cnt);
if (++wil->halp.ref_cnt == 1) {
reinit_completion(&wil->halp.comp);
wil6210_set_halp(wil);
rc = wait_for_completion_timeout(&wil->halp.comp, to_jiffies);
if (!rc) {
wil_err(wil, "HALP vote timed out\n");
/* Mask HALP as done in case the interrupt is raised */
wil6210_mask_halp(wil);
} else {
wil_dbg_irq(wil,
"halp_vote: HALP vote completed after %d ms\n",
jiffies_to_msecs(to_jiffies - rc));
}
}
wil_dbg_irq(wil, "halp_vote: end, HALP ref_cnt (%d)\n",
wil->halp.ref_cnt);
mutex_unlock(&wil->halp.lock);
}
void wil_halp_unvote(struct wil6210_priv *wil)
{
WARN_ON(wil->halp.ref_cnt == 0);
mutex_lock(&wil->halp.lock);
wil_dbg_irq(wil, "halp_unvote: start, HALP ref_cnt (%d)\n",
wil->halp.ref_cnt);
if (--wil->halp.ref_cnt == 0) {
wil6210_clear_halp(wil);
wil_dbg_irq(wil, "HALP unvote\n");
}
wil_dbg_irq(wil, "halp_unvote:end, HALP ref_cnt (%d)\n",
wil->halp.ref_cnt);
mutex_unlock(&wil->halp.lock);
}
wil6210: initialize TX and RX enhanced DMA rings Enhanced DMA design includes the following rings: - Single RX descriptor ring is used for all VIFs - Multiple RX status rings are supported, to allow RSS - TX descriptor ring is allocated per connection - A single TX status ring is used for all TX descriptor rings This patch initializes and frees the above descriptor and status rings. The RX SKBs are handled by a new entity of RX buffers manager, which handles RX buffers, each one points to an allocated SKB. During Rx completion processing, the driver extracts a buffer ID which is used as an index to the buffers array. After the SKB is freed the buffer is moved from the 'active' list to the 'free' list, indicating it can be used for another descriptor. During Rx refill, SKBs are allocated and attached to 'free' buffers. Those buffers are attached to new descriptors and moved to the 'active' list. New debugfs entries were added to allow edma configuration: Run the following command to configure the number of status rings: echo NUM_OF_STATUS_RINGS > num_rx_status_rings Run the following command to use extended RX status message for additional debug fields from HW: echo 0 > compressed_rx_status Run the following command to control the size of the TX status ring: echo TX_STATUS_RING_ORDER > tx_status_ring_order The status ring size will be 1 << tx_status_ring_order Run the following command to control the size of the RX status ring: echo RX_STATUS_RING_ORDER > rx_status_ring_order Due to HW constrains RX sring order should be bigger than RX ring order The status ring size will be 1 << rx_status_ring_order Run the following command to change the number of RX buffer IDs: echo RX_BUFF_ID_COUNT > rx_buff_id_count Signed-off-by: Gidon Studinski <gidons@codeaurora.org> Signed-off-by: Maya Erez <merez@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-06-29 21:28:23 +08:00
void wil_init_txrx_ops(struct wil6210_priv *wil)
{
if (wil->use_enhanced_dma_hw)
wil_init_txrx_ops_edma(wil);
else
wil_init_txrx_ops_legacy_dma(wil);
}