The following handshake mechanism needs be followed after firmware download is completed to bring the firmware to running state. After firmware fragments of Operational image are downloaded and secure sends result of the image succeeds, 1. Driver sends HCI Intel reset with boot option #1 to switch FW image. 2. FW sends Alive GP[0] MSIx 3. Driver enables data path (doorbell 0x460 for RBDs, etc...) 4. Driver gets Bootup event from firmware 5. Driver performs D0 entry to device (WRITE to IPC_Sleep_Control =0x0) 6. FW sends Alive GP[0] MSIx 7. Device host interface is fully set for BT protocol stack operation. 8. Driver may optionally get debug event with ID 0x97 which can be dropped For Intermediate loadger image, all the above steps are applicable expcept #5 and #6. On HCI_OP_RESET, firmware raises alive interrupt. Driver needs to wait for it before passing control over to bluetooth stack. Co-developed-by: Devegowda Chandrashekar <chandrashekar.devegowda@intel.com> Signed-off-by: Devegowda Chandrashekar <chandrashekar.devegowda@intel.com> Signed-off-by: Kiran K <kiran.k@intel.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
1624 lines
43 KiB
C
1624 lines
43 KiB
C
// SPDX-License-Identifier: GPL-2.0-or-later
|
|
/*
|
|
*
|
|
* Bluetooth support for Intel PCIe devices
|
|
*
|
|
* Copyright (C) 2024 Intel Corporation
|
|
*/
|
|
|
|
#include <linux/kernel.h>
|
|
#include <linux/module.h>
|
|
#include <linux/firmware.h>
|
|
#include <linux/pci.h>
|
|
#include <linux/wait.h>
|
|
#include <linux/delay.h>
|
|
#include <linux/interrupt.h>
|
|
|
|
#include <linux/unaligned.h>
|
|
|
|
#include <net/bluetooth/bluetooth.h>
|
|
#include <net/bluetooth/hci_core.h>
|
|
|
|
#include "btintel.h"
|
|
#include "btintel_pcie.h"
|
|
|
|
#define VERSION "0.1"
|
|
|
|
#define BTINTEL_PCI_DEVICE(dev, subdev) \
|
|
.vendor = PCI_VENDOR_ID_INTEL, \
|
|
.device = (dev), \
|
|
.subvendor = PCI_ANY_ID, \
|
|
.subdevice = (subdev), \
|
|
.driver_data = 0
|
|
|
|
#define POLL_INTERVAL_US 10
|
|
|
|
/* Intel Bluetooth PCIe device id table */
|
|
static const struct pci_device_id btintel_pcie_table[] = {
|
|
{ BTINTEL_PCI_DEVICE(0xA876, PCI_ANY_ID) },
|
|
{ 0 }
|
|
};
|
|
MODULE_DEVICE_TABLE(pci, btintel_pcie_table);
|
|
|
|
/* Intel PCIe uses 4 bytes of HCI type instead of 1 byte BT SIG HCI type */
|
|
#define BTINTEL_PCIE_HCI_TYPE_LEN 4
|
|
#define BTINTEL_PCIE_HCI_CMD_PKT 0x00000001
|
|
#define BTINTEL_PCIE_HCI_ACL_PKT 0x00000002
|
|
#define BTINTEL_PCIE_HCI_SCO_PKT 0x00000003
|
|
#define BTINTEL_PCIE_HCI_EVT_PKT 0x00000004
|
|
#define BTINTEL_PCIE_HCI_ISO_PKT 0x00000005
|
|
|
|
/* Alive interrupt context */
|
|
enum {
|
|
BTINTEL_PCIE_ROM,
|
|
BTINTEL_PCIE_FW_DL,
|
|
BTINTEL_PCIE_HCI_RESET,
|
|
BTINTEL_PCIE_INTEL_HCI_RESET1,
|
|
BTINTEL_PCIE_INTEL_HCI_RESET2,
|
|
BTINTEL_PCIE_D0,
|
|
BTINTEL_PCIE_D3
|
|
};
|
|
|
|
static inline void ipc_print_ia_ring(struct hci_dev *hdev, struct ia *ia,
|
|
u16 queue_num)
|
|
{
|
|
bt_dev_dbg(hdev, "IA: %s: tr-h:%02u tr-t:%02u cr-h:%02u cr-t:%02u",
|
|
queue_num == BTINTEL_PCIE_TXQ_NUM ? "TXQ" : "RXQ",
|
|
ia->tr_hia[queue_num], ia->tr_tia[queue_num],
|
|
ia->cr_hia[queue_num], ia->cr_tia[queue_num]);
|
|
}
|
|
|
|
static inline void ipc_print_urbd1(struct hci_dev *hdev, struct urbd1 *urbd1,
|
|
u16 index)
|
|
{
|
|
bt_dev_dbg(hdev, "RXQ:urbd1(%u) frbd_tag:%u status: 0x%x fixed:0x%x",
|
|
index, urbd1->frbd_tag, urbd1->status, urbd1->fixed);
|
|
}
|
|
|
|
static int btintel_pcie_poll_bit(struct btintel_pcie_data *data, u32 offset,
|
|
u32 bits, u32 mask, int timeout_us)
|
|
{
|
|
int t = 0;
|
|
u32 reg;
|
|
|
|
do {
|
|
reg = btintel_pcie_rd_reg32(data, offset);
|
|
|
|
if ((reg & mask) == (bits & mask))
|
|
return t;
|
|
udelay(POLL_INTERVAL_US);
|
|
t += POLL_INTERVAL_US;
|
|
} while (t < timeout_us);
|
|
|
|
return -ETIMEDOUT;
|
|
}
|
|
|
|
static struct btintel_pcie_data *btintel_pcie_get_data(struct msix_entry *entry)
|
|
{
|
|
u8 queue = entry->entry;
|
|
struct msix_entry *entries = entry - queue;
|
|
|
|
return container_of(entries, struct btintel_pcie_data, msix_entries[0]);
|
|
}
|
|
|
|
/* Set the doorbell for TXQ to notify the device that @index (actually index-1)
|
|
* of the TFD is updated and ready to transmit.
|
|
*/
|
|
static void btintel_pcie_set_tx_db(struct btintel_pcie_data *data, u16 index)
|
|
{
|
|
u32 val;
|
|
|
|
val = index;
|
|
val |= (BTINTEL_PCIE_TX_DB_VEC << 16);
|
|
|
|
btintel_pcie_wr_reg32(data, BTINTEL_PCIE_CSR_HBUS_TARG_WRPTR, val);
|
|
}
|
|
|
|
/* Copy the data to next(@tfd_index) data buffer and update the TFD(transfer
|
|
* descriptor) with the data length and the DMA address of the data buffer.
|
|
*/
|
|
static void btintel_pcie_prepare_tx(struct txq *txq, u16 tfd_index,
|
|
struct sk_buff *skb)
|
|
{
|
|
struct data_buf *buf;
|
|
struct tfd *tfd;
|
|
|
|
tfd = &txq->tfds[tfd_index];
|
|
memset(tfd, 0, sizeof(*tfd));
|
|
|
|
buf = &txq->bufs[tfd_index];
|
|
|
|
tfd->size = skb->len;
|
|
tfd->addr = buf->data_p_addr;
|
|
|
|
/* Copy the outgoing data to DMA buffer */
|
|
memcpy(buf->data, skb->data, tfd->size);
|
|
}
|
|
|
|
static int btintel_pcie_send_sync(struct btintel_pcie_data *data,
|
|
struct sk_buff *skb)
|
|
{
|
|
int ret;
|
|
u16 tfd_index;
|
|
struct txq *txq = &data->txq;
|
|
|
|
tfd_index = data->ia.tr_hia[BTINTEL_PCIE_TXQ_NUM];
|
|
|
|
if (tfd_index > txq->count)
|
|
return -ERANGE;
|
|
|
|
/* Prepare for TX. It updates the TFD with the length of data and
|
|
* address of the DMA buffer, and copy the data to the DMA buffer
|
|
*/
|
|
btintel_pcie_prepare_tx(txq, tfd_index, skb);
|
|
|
|
tfd_index = (tfd_index + 1) % txq->count;
|
|
data->ia.tr_hia[BTINTEL_PCIE_TXQ_NUM] = tfd_index;
|
|
|
|
/* Arm wait event condition */
|
|
data->tx_wait_done = false;
|
|
|
|
/* Set the doorbell to notify the device */
|
|
btintel_pcie_set_tx_db(data, tfd_index);
|
|
|
|
/* Wait for the complete interrupt - URBD0 */
|
|
ret = wait_event_timeout(data->tx_wait_q, data->tx_wait_done,
|
|
msecs_to_jiffies(BTINTEL_PCIE_TX_WAIT_TIMEOUT_MS));
|
|
if (!ret)
|
|
return -ETIME;
|
|
|
|
return 0;
|
|
}
|
|
|
|
/* Set the doorbell for RXQ to notify the device that @index (actually index-1)
|
|
* is available to receive the data
|
|
*/
|
|
static void btintel_pcie_set_rx_db(struct btintel_pcie_data *data, u16 index)
|
|
{
|
|
u32 val;
|
|
|
|
val = index;
|
|
val |= (BTINTEL_PCIE_RX_DB_VEC << 16);
|
|
|
|
btintel_pcie_wr_reg32(data, BTINTEL_PCIE_CSR_HBUS_TARG_WRPTR, val);
|
|
}
|
|
|
|
/* Update the FRBD (free buffer descriptor) with the @frbd_index and the
|
|
* DMA address of the free buffer.
|
|
*/
|
|
static void btintel_pcie_prepare_rx(struct rxq *rxq, u16 frbd_index)
|
|
{
|
|
struct data_buf *buf;
|
|
struct frbd *frbd;
|
|
|
|
/* Get the buffer of the FRBD for DMA */
|
|
buf = &rxq->bufs[frbd_index];
|
|
|
|
frbd = &rxq->frbds[frbd_index];
|
|
memset(frbd, 0, sizeof(*frbd));
|
|
|
|
/* Update FRBD */
|
|
frbd->tag = frbd_index;
|
|
frbd->addr = buf->data_p_addr;
|
|
}
|
|
|
|
static int btintel_pcie_submit_rx(struct btintel_pcie_data *data)
|
|
{
|
|
u16 frbd_index;
|
|
struct rxq *rxq = &data->rxq;
|
|
|
|
frbd_index = data->ia.tr_hia[BTINTEL_PCIE_RXQ_NUM];
|
|
|
|
if (frbd_index > rxq->count)
|
|
return -ERANGE;
|
|
|
|
/* Prepare for RX submit. It updates the FRBD with the address of DMA
|
|
* buffer
|
|
*/
|
|
btintel_pcie_prepare_rx(rxq, frbd_index);
|
|
|
|
frbd_index = (frbd_index + 1) % rxq->count;
|
|
data->ia.tr_hia[BTINTEL_PCIE_RXQ_NUM] = frbd_index;
|
|
ipc_print_ia_ring(data->hdev, &data->ia, BTINTEL_PCIE_RXQ_NUM);
|
|
|
|
/* Set the doorbell to notify the device */
|
|
btintel_pcie_set_rx_db(data, frbd_index);
|
|
|
|
return 0;
|
|
}
|
|
|
|
static int btintel_pcie_start_rx(struct btintel_pcie_data *data)
|
|
{
|
|
int i, ret;
|
|
|
|
for (i = 0; i < BTINTEL_PCIE_RX_MAX_QUEUE; i++) {
|
|
ret = btintel_pcie_submit_rx(data);
|
|
if (ret)
|
|
return ret;
|
|
}
|
|
|
|
return 0;
|
|
}
|
|
|
|
static void btintel_pcie_reset_ia(struct btintel_pcie_data *data)
|
|
{
|
|
memset(data->ia.tr_hia, 0, sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES);
|
|
memset(data->ia.tr_tia, 0, sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES);
|
|
memset(data->ia.cr_hia, 0, sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES);
|
|
memset(data->ia.cr_tia, 0, sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES);
|
|
}
|
|
|
|
static void btintel_pcie_reset_bt(struct btintel_pcie_data *data)
|
|
{
|
|
btintel_pcie_wr_reg32(data, BTINTEL_PCIE_CSR_FUNC_CTRL_REG,
|
|
BTINTEL_PCIE_CSR_FUNC_CTRL_SW_RESET);
|
|
}
|
|
|
|
/* This function enables BT function by setting BTINTEL_PCIE_CSR_FUNC_CTRL_MAC_INIT bit in
|
|
* BTINTEL_PCIE_CSR_FUNC_CTRL_REG register and wait for MSI-X with
|
|
* BTINTEL_PCIE_MSIX_HW_INT_CAUSES_GP0.
|
|
* Then the host reads firmware version from BTINTEL_CSR_F2D_MBX and the boot stage
|
|
* from BTINTEL_PCIE_CSR_BOOT_STAGE_REG.
|
|
*/
|
|
static int btintel_pcie_enable_bt(struct btintel_pcie_data *data)
|
|
{
|
|
int err;
|
|
|
|
data->gp0_received = false;
|
|
|
|
/* Update the DMA address of CI struct to CSR */
|
|
btintel_pcie_wr_reg32(data, BTINTEL_PCIE_CSR_CI_ADDR_LSB_REG,
|
|
data->ci_p_addr & 0xffffffff);
|
|
btintel_pcie_wr_reg32(data, BTINTEL_PCIE_CSR_CI_ADDR_MSB_REG,
|
|
(u64)data->ci_p_addr >> 32);
|
|
|
|
/* Reset the cached value of boot stage. it is updated by the MSI-X
|
|
* gp0 interrupt handler.
|
|
*/
|
|
data->boot_stage_cache = 0x0;
|
|
|
|
/* Set MAC_INIT bit to start primary bootloader */
|
|
btintel_pcie_rd_reg32(data, BTINTEL_PCIE_CSR_FUNC_CTRL_REG);
|
|
|
|
btintel_pcie_set_reg_bits(data, BTINTEL_PCIE_CSR_FUNC_CTRL_REG,
|
|
BTINTEL_PCIE_CSR_FUNC_CTRL_MAC_INIT);
|
|
|
|
/* Wait until MAC_ACCESS is granted */
|
|
err = btintel_pcie_poll_bit(data, BTINTEL_PCIE_CSR_FUNC_CTRL_REG,
|
|
BTINTEL_PCIE_CSR_FUNC_CTRL_MAC_ACCESS_STS,
|
|
BTINTEL_PCIE_CSR_FUNC_CTRL_MAC_ACCESS_STS,
|
|
BTINTEL_DEFAULT_MAC_ACCESS_TIMEOUT_US);
|
|
if (err < 0)
|
|
return -ENODEV;
|
|
|
|
/* MAC is ready. Enable BT FUNC */
|
|
btintel_pcie_set_reg_bits(data, BTINTEL_PCIE_CSR_FUNC_CTRL_REG,
|
|
BTINTEL_PCIE_CSR_FUNC_CTRL_FUNC_ENA |
|
|
BTINTEL_PCIE_CSR_FUNC_CTRL_FUNC_INIT);
|
|
|
|
btintel_pcie_rd_reg32(data, BTINTEL_PCIE_CSR_FUNC_CTRL_REG);
|
|
|
|
/* wait for interrupt from the device after booting up to primary
|
|
* bootloader.
|
|
*/
|
|
data->alive_intr_ctxt = BTINTEL_PCIE_ROM;
|
|
err = wait_event_timeout(data->gp0_wait_q, data->gp0_received,
|
|
msecs_to_jiffies(BTINTEL_DEFAULT_INTR_TIMEOUT_MS));
|
|
if (!err)
|
|
return -ETIME;
|
|
|
|
/* Check cached boot stage is BTINTEL_PCIE_CSR_BOOT_STAGE_ROM(BIT(0)) */
|
|
if (~data->boot_stage_cache & BTINTEL_PCIE_CSR_BOOT_STAGE_ROM)
|
|
return -ENODEV;
|
|
|
|
return 0;
|
|
}
|
|
|
|
/* BIT(0) - ROM, BIT(1) - IML and BIT(3) - OP
|
|
* Sometimes during firmware image switching from ROM to IML or IML to OP image,
|
|
* the previous image bit is not cleared by firmware when alive interrupt is
|
|
* received. Driver needs to take care of these sticky bits when deciding the
|
|
* current image running on controller.
|
|
* Ex: 0x10 and 0x11 - both represents that controller is running IML
|
|
*/
|
|
static inline bool btintel_pcie_in_rom(struct btintel_pcie_data *data)
|
|
{
|
|
return data->boot_stage_cache & BTINTEL_PCIE_CSR_BOOT_STAGE_ROM &&
|
|
!(data->boot_stage_cache & BTINTEL_PCIE_CSR_BOOT_STAGE_IML) &&
|
|
!(data->boot_stage_cache & BTINTEL_PCIE_CSR_BOOT_STAGE_OPFW);
|
|
}
|
|
|
|
static inline bool btintel_pcie_in_op(struct btintel_pcie_data *data)
|
|
{
|
|
return data->boot_stage_cache & BTINTEL_PCIE_CSR_BOOT_STAGE_OPFW;
|
|
}
|
|
|
|
static inline bool btintel_pcie_in_iml(struct btintel_pcie_data *data)
|
|
{
|
|
return data->boot_stage_cache & BTINTEL_PCIE_CSR_BOOT_STAGE_IML &&
|
|
!(data->boot_stage_cache & BTINTEL_PCIE_CSR_BOOT_STAGE_OPFW);
|
|
}
|
|
|
|
static inline bool btintel_pcie_in_d3(struct btintel_pcie_data *data)
|
|
{
|
|
return data->boot_stage_cache & BTINTEL_PCIE_CSR_BOOT_STAGE_D3_STATE_READY;
|
|
}
|
|
|
|
static inline bool btintel_pcie_in_d0(struct btintel_pcie_data *data)
|
|
{
|
|
return !(data->boot_stage_cache & BTINTEL_PCIE_CSR_BOOT_STAGE_D3_STATE_READY);
|
|
}
|
|
|
|
static void btintel_pcie_wr_sleep_cntrl(struct btintel_pcie_data *data,
|
|
u32 dxstate)
|
|
{
|
|
bt_dev_dbg(data->hdev, "writing sleep_ctl_reg: 0x%8.8x", dxstate);
|
|
btintel_pcie_wr_reg32(data, BTINTEL_PCIE_CSR_IPC_SLEEP_CTL_REG, dxstate);
|
|
}
|
|
|
|
static inline char *btintel_pcie_alivectxt_state2str(u32 alive_intr_ctxt)
|
|
{
|
|
switch (alive_intr_ctxt) {
|
|
case BTINTEL_PCIE_ROM:
|
|
return "rom";
|
|
case BTINTEL_PCIE_FW_DL:
|
|
return "fw_dl";
|
|
case BTINTEL_PCIE_D0:
|
|
return "d0";
|
|
case BTINTEL_PCIE_D3:
|
|
return "d3";
|
|
case BTINTEL_PCIE_HCI_RESET:
|
|
return "hci_reset";
|
|
case BTINTEL_PCIE_INTEL_HCI_RESET1:
|
|
return "intel_reset1";
|
|
case BTINTEL_PCIE_INTEL_HCI_RESET2:
|
|
return "intel_reset2";
|
|
default:
|
|
return "unknown";
|
|
}
|
|
return "null";
|
|
}
|
|
|
|
/* This function handles the MSI-X interrupt for gp0 cause (bit 0 in
|
|
* BTINTEL_PCIE_CSR_MSIX_HW_INT_CAUSES) which is sent for boot stage and image response.
|
|
*/
|
|
static void btintel_pcie_msix_gp0_handler(struct btintel_pcie_data *data)
|
|
{
|
|
bool submit_rx, signal_waitq;
|
|
u32 reg, old_ctxt;
|
|
|
|
/* This interrupt is for three different causes and it is not easy to
|
|
* know what causes the interrupt. So, it compares each register value
|
|
* with cached value and update it before it wake up the queue.
|
|
*/
|
|
reg = btintel_pcie_rd_reg32(data, BTINTEL_PCIE_CSR_BOOT_STAGE_REG);
|
|
if (reg != data->boot_stage_cache)
|
|
data->boot_stage_cache = reg;
|
|
|
|
bt_dev_dbg(data->hdev, "Alive context: %s old_boot_stage: 0x%8.8x new_boot_stage: 0x%8.8x",
|
|
btintel_pcie_alivectxt_state2str(data->alive_intr_ctxt),
|
|
data->boot_stage_cache, reg);
|
|
reg = btintel_pcie_rd_reg32(data, BTINTEL_PCIE_CSR_IMG_RESPONSE_REG);
|
|
if (reg != data->img_resp_cache)
|
|
data->img_resp_cache = reg;
|
|
|
|
data->gp0_received = true;
|
|
|
|
old_ctxt = data->alive_intr_ctxt;
|
|
submit_rx = false;
|
|
signal_waitq = false;
|
|
|
|
switch (data->alive_intr_ctxt) {
|
|
case BTINTEL_PCIE_ROM:
|
|
data->alive_intr_ctxt = BTINTEL_PCIE_FW_DL;
|
|
signal_waitq = true;
|
|
break;
|
|
case BTINTEL_PCIE_FW_DL:
|
|
/* Error case is already handled. Ideally control shall not
|
|
* reach here
|
|
*/
|
|
break;
|
|
case BTINTEL_PCIE_INTEL_HCI_RESET1:
|
|
if (btintel_pcie_in_op(data)) {
|
|
submit_rx = true;
|
|
break;
|
|
}
|
|
|
|
if (btintel_pcie_in_iml(data)) {
|
|
submit_rx = true;
|
|
data->alive_intr_ctxt = BTINTEL_PCIE_FW_DL;
|
|
break;
|
|
}
|
|
break;
|
|
case BTINTEL_PCIE_INTEL_HCI_RESET2:
|
|
if (btintel_test_and_clear_flag(data->hdev, INTEL_WAIT_FOR_D0)) {
|
|
btintel_wake_up_flag(data->hdev, INTEL_WAIT_FOR_D0);
|
|
data->alive_intr_ctxt = BTINTEL_PCIE_D0;
|
|
}
|
|
break;
|
|
case BTINTEL_PCIE_D0:
|
|
if (btintel_pcie_in_d3(data)) {
|
|
data->alive_intr_ctxt = BTINTEL_PCIE_D3;
|
|
signal_waitq = true;
|
|
break;
|
|
}
|
|
break;
|
|
case BTINTEL_PCIE_D3:
|
|
if (btintel_pcie_in_d0(data)) {
|
|
data->alive_intr_ctxt = BTINTEL_PCIE_D0;
|
|
submit_rx = true;
|
|
signal_waitq = true;
|
|
break;
|
|
}
|
|
break;
|
|
case BTINTEL_PCIE_HCI_RESET:
|
|
data->alive_intr_ctxt = BTINTEL_PCIE_D0;
|
|
submit_rx = true;
|
|
signal_waitq = true;
|
|
break;
|
|
default:
|
|
bt_dev_err(data->hdev, "Unknown state: 0x%2.2x",
|
|
data->alive_intr_ctxt);
|
|
break;
|
|
}
|
|
|
|
if (submit_rx) {
|
|
btintel_pcie_reset_ia(data);
|
|
btintel_pcie_start_rx(data);
|
|
}
|
|
|
|
if (signal_waitq) {
|
|
bt_dev_dbg(data->hdev, "wake up gp0 wait_q");
|
|
wake_up(&data->gp0_wait_q);
|
|
}
|
|
|
|
if (old_ctxt != data->alive_intr_ctxt)
|
|
bt_dev_dbg(data->hdev, "alive context changed: %s -> %s",
|
|
btintel_pcie_alivectxt_state2str(old_ctxt),
|
|
btintel_pcie_alivectxt_state2str(data->alive_intr_ctxt));
|
|
}
|
|
|
|
/* This function handles the MSX-X interrupt for rx queue 0 which is for TX
|
|
*/
|
|
static void btintel_pcie_msix_tx_handle(struct btintel_pcie_data *data)
|
|
{
|
|
u16 cr_tia, cr_hia;
|
|
struct txq *txq;
|
|
struct urbd0 *urbd0;
|
|
|
|
cr_tia = data->ia.cr_tia[BTINTEL_PCIE_TXQ_NUM];
|
|
cr_hia = data->ia.cr_hia[BTINTEL_PCIE_TXQ_NUM];
|
|
|
|
if (cr_tia == cr_hia)
|
|
return;
|
|
|
|
txq = &data->txq;
|
|
|
|
while (cr_tia != cr_hia) {
|
|
data->tx_wait_done = true;
|
|
wake_up(&data->tx_wait_q);
|
|
|
|
urbd0 = &txq->urbd0s[cr_tia];
|
|
|
|
if (urbd0->tfd_index > txq->count)
|
|
return;
|
|
|
|
cr_tia = (cr_tia + 1) % txq->count;
|
|
data->ia.cr_tia[BTINTEL_PCIE_TXQ_NUM] = cr_tia;
|
|
ipc_print_ia_ring(data->hdev, &data->ia, BTINTEL_PCIE_TXQ_NUM);
|
|
}
|
|
}
|
|
|
|
static int btintel_pcie_recv_event(struct hci_dev *hdev, struct sk_buff *skb)
|
|
{
|
|
struct hci_event_hdr *hdr = (void *)skb->data;
|
|
const char diagnostics_hdr[] = { 0x87, 0x80, 0x03 };
|
|
struct btintel_pcie_data *data = hci_get_drvdata(hdev);
|
|
|
|
if (skb->len > HCI_EVENT_HDR_SIZE && hdr->evt == 0xff &&
|
|
hdr->plen > 0) {
|
|
const void *ptr = skb->data + HCI_EVENT_HDR_SIZE + 1;
|
|
unsigned int len = skb->len - HCI_EVENT_HDR_SIZE - 1;
|
|
|
|
if (btintel_test_flag(hdev, INTEL_BOOTLOADER)) {
|
|
switch (skb->data[2]) {
|
|
case 0x02:
|
|
/* When switching to the operational firmware
|
|
* the device sends a vendor specific event
|
|
* indicating that the bootup completed.
|
|
*/
|
|
btintel_bootup(hdev, ptr, len);
|
|
|
|
/* If bootup event is from operational image,
|
|
* driver needs to write sleep control register to
|
|
* move into D0 state
|
|
*/
|
|
if (btintel_pcie_in_op(data)) {
|
|
btintel_pcie_wr_sleep_cntrl(data, BTINTEL_PCIE_STATE_D0);
|
|
data->alive_intr_ctxt = BTINTEL_PCIE_INTEL_HCI_RESET2;
|
|
break;
|
|
}
|
|
|
|
if (btintel_pcie_in_iml(data)) {
|
|
/* In case of IML, there is no concept
|
|
* of D0 transition. Just mimic as if
|
|
* IML moved to D0 by clearing INTEL_WAIT_FOR_D0
|
|
* bit and waking up the task waiting on
|
|
* INTEL_WAIT_FOR_D0. This is required
|
|
* as intel_boot() is common function for
|
|
* both IML and OP image loading.
|
|
*/
|
|
if (btintel_test_and_clear_flag(data->hdev,
|
|
INTEL_WAIT_FOR_D0))
|
|
btintel_wake_up_flag(data->hdev,
|
|
INTEL_WAIT_FOR_D0);
|
|
}
|
|
break;
|
|
case 0x06:
|
|
/* When the firmware loading completes the
|
|
* device sends out a vendor specific event
|
|
* indicating the result of the firmware
|
|
* loading.
|
|
*/
|
|
btintel_secure_send_result(hdev, ptr, len);
|
|
break;
|
|
}
|
|
}
|
|
|
|
/* Handle all diagnostics events separately. May still call
|
|
* hci_recv_frame.
|
|
*/
|
|
if (len >= sizeof(diagnostics_hdr) &&
|
|
memcmp(&skb->data[2], diagnostics_hdr,
|
|
sizeof(diagnostics_hdr)) == 0) {
|
|
return btintel_diagnostics(hdev, skb);
|
|
}
|
|
|
|
/* This is a debug event that comes from IML and OP image when it
|
|
* starts execution. There is no need pass this event to stack.
|
|
*/
|
|
if (skb->data[2] == 0x97)
|
|
return 0;
|
|
}
|
|
|
|
return hci_recv_frame(hdev, skb);
|
|
}
|
|
/* Process the received rx data
|
|
* It check the frame header to identify the data type and create skb
|
|
* and calling HCI API
|
|
*/
|
|
static int btintel_pcie_recv_frame(struct btintel_pcie_data *data,
|
|
struct sk_buff *skb)
|
|
{
|
|
int ret;
|
|
u8 pkt_type;
|
|
u16 plen;
|
|
u32 pcie_pkt_type;
|
|
struct sk_buff *new_skb;
|
|
void *pdata;
|
|
struct hci_dev *hdev = data->hdev;
|
|
|
|
spin_lock(&data->hci_rx_lock);
|
|
|
|
/* The first 4 bytes indicates the Intel PCIe specific packet type */
|
|
pdata = skb_pull_data(skb, BTINTEL_PCIE_HCI_TYPE_LEN);
|
|
if (!pdata) {
|
|
bt_dev_err(hdev, "Corrupted packet received");
|
|
ret = -EILSEQ;
|
|
goto exit_error;
|
|
}
|
|
|
|
pcie_pkt_type = get_unaligned_le32(pdata);
|
|
|
|
switch (pcie_pkt_type) {
|
|
case BTINTEL_PCIE_HCI_ACL_PKT:
|
|
if (skb->len >= HCI_ACL_HDR_SIZE) {
|
|
plen = HCI_ACL_HDR_SIZE + __le16_to_cpu(hci_acl_hdr(skb)->dlen);
|
|
pkt_type = HCI_ACLDATA_PKT;
|
|
} else {
|
|
bt_dev_err(hdev, "ACL packet is too short");
|
|
ret = -EILSEQ;
|
|
goto exit_error;
|
|
}
|
|
break;
|
|
|
|
case BTINTEL_PCIE_HCI_SCO_PKT:
|
|
if (skb->len >= HCI_SCO_HDR_SIZE) {
|
|
plen = HCI_SCO_HDR_SIZE + hci_sco_hdr(skb)->dlen;
|
|
pkt_type = HCI_SCODATA_PKT;
|
|
} else {
|
|
bt_dev_err(hdev, "SCO packet is too short");
|
|
ret = -EILSEQ;
|
|
goto exit_error;
|
|
}
|
|
break;
|
|
|
|
case BTINTEL_PCIE_HCI_EVT_PKT:
|
|
if (skb->len >= HCI_EVENT_HDR_SIZE) {
|
|
plen = HCI_EVENT_HDR_SIZE + hci_event_hdr(skb)->plen;
|
|
pkt_type = HCI_EVENT_PKT;
|
|
} else {
|
|
bt_dev_err(hdev, "Event packet is too short");
|
|
ret = -EILSEQ;
|
|
goto exit_error;
|
|
}
|
|
break;
|
|
|
|
case BTINTEL_PCIE_HCI_ISO_PKT:
|
|
if (skb->len >= HCI_ISO_HDR_SIZE) {
|
|
plen = HCI_ISO_HDR_SIZE + __le16_to_cpu(hci_iso_hdr(skb)->dlen);
|
|
pkt_type = HCI_ISODATA_PKT;
|
|
} else {
|
|
bt_dev_err(hdev, "ISO packet is too short");
|
|
ret = -EILSEQ;
|
|
goto exit_error;
|
|
}
|
|
break;
|
|
|
|
default:
|
|
bt_dev_err(hdev, "Invalid packet type received: 0x%4.4x",
|
|
pcie_pkt_type);
|
|
ret = -EINVAL;
|
|
goto exit_error;
|
|
}
|
|
|
|
if (skb->len < plen) {
|
|
bt_dev_err(hdev, "Received corrupted packet. type: 0x%2.2x",
|
|
pkt_type);
|
|
ret = -EILSEQ;
|
|
goto exit_error;
|
|
}
|
|
|
|
bt_dev_dbg(hdev, "pkt_type: 0x%2.2x len: %u", pkt_type, plen);
|
|
|
|
new_skb = bt_skb_alloc(plen, GFP_ATOMIC);
|
|
if (!new_skb) {
|
|
bt_dev_err(hdev, "Failed to allocate memory for skb of len: %u",
|
|
skb->len);
|
|
ret = -ENOMEM;
|
|
goto exit_error;
|
|
}
|
|
|
|
hci_skb_pkt_type(new_skb) = pkt_type;
|
|
skb_put_data(new_skb, skb->data, plen);
|
|
hdev->stat.byte_rx += plen;
|
|
|
|
if (pcie_pkt_type == BTINTEL_PCIE_HCI_EVT_PKT)
|
|
ret = btintel_pcie_recv_event(hdev, new_skb);
|
|
else
|
|
ret = hci_recv_frame(hdev, new_skb);
|
|
|
|
exit_error:
|
|
if (ret)
|
|
hdev->stat.err_rx++;
|
|
|
|
spin_unlock(&data->hci_rx_lock);
|
|
|
|
return ret;
|
|
}
|
|
|
|
static void btintel_pcie_rx_work(struct work_struct *work)
|
|
{
|
|
struct btintel_pcie_data *data = container_of(work,
|
|
struct btintel_pcie_data, rx_work);
|
|
struct sk_buff *skb;
|
|
int err;
|
|
struct hci_dev *hdev = data->hdev;
|
|
|
|
/* Process the sk_buf in queue and send to the HCI layer */
|
|
while ((skb = skb_dequeue(&data->rx_skb_q))) {
|
|
err = btintel_pcie_recv_frame(data, skb);
|
|
if (err)
|
|
bt_dev_err(hdev, "Failed to send received frame: %d",
|
|
err);
|
|
kfree_skb(skb);
|
|
}
|
|
}
|
|
|
|
/* create sk_buff with data and save it to queue and start RX work */
|
|
static int btintel_pcie_submit_rx_work(struct btintel_pcie_data *data, u8 status,
|
|
void *buf)
|
|
{
|
|
int ret, len;
|
|
struct rfh_hdr *rfh_hdr;
|
|
struct sk_buff *skb;
|
|
|
|
rfh_hdr = buf;
|
|
|
|
len = rfh_hdr->packet_len;
|
|
if (len <= 0) {
|
|
ret = -EINVAL;
|
|
goto resubmit;
|
|
}
|
|
|
|
/* Remove RFH header */
|
|
buf += sizeof(*rfh_hdr);
|
|
|
|
skb = alloc_skb(len, GFP_ATOMIC);
|
|
if (!skb) {
|
|
ret = -ENOMEM;
|
|
goto resubmit;
|
|
}
|
|
|
|
skb_put_data(skb, buf, len);
|
|
skb_queue_tail(&data->rx_skb_q, skb);
|
|
queue_work(data->workqueue, &data->rx_work);
|
|
|
|
resubmit:
|
|
ret = btintel_pcie_submit_rx(data);
|
|
|
|
return ret;
|
|
}
|
|
|
|
/* Handles the MSI-X interrupt for rx queue 1 which is for RX */
|
|
static void btintel_pcie_msix_rx_handle(struct btintel_pcie_data *data)
|
|
{
|
|
u16 cr_hia, cr_tia;
|
|
struct rxq *rxq;
|
|
struct urbd1 *urbd1;
|
|
struct data_buf *buf;
|
|
int ret;
|
|
struct hci_dev *hdev = data->hdev;
|
|
|
|
cr_hia = data->ia.cr_hia[BTINTEL_PCIE_RXQ_NUM];
|
|
cr_tia = data->ia.cr_tia[BTINTEL_PCIE_RXQ_NUM];
|
|
|
|
bt_dev_dbg(hdev, "RXQ: cr_hia: %u cr_tia: %u", cr_hia, cr_tia);
|
|
|
|
/* Check CR_TIA and CR_HIA for change */
|
|
if (cr_tia == cr_hia) {
|
|
bt_dev_warn(hdev, "RXQ: no new CD found");
|
|
return;
|
|
}
|
|
|
|
rxq = &data->rxq;
|
|
|
|
/* The firmware sends multiple CD in a single MSI-X and it needs to
|
|
* process all received CDs in this interrupt.
|
|
*/
|
|
while (cr_tia != cr_hia) {
|
|
urbd1 = &rxq->urbd1s[cr_tia];
|
|
ipc_print_urbd1(data->hdev, urbd1, cr_tia);
|
|
|
|
buf = &rxq->bufs[urbd1->frbd_tag];
|
|
if (!buf) {
|
|
bt_dev_err(hdev, "RXQ: failed to get the DMA buffer for %d",
|
|
urbd1->frbd_tag);
|
|
return;
|
|
}
|
|
|
|
ret = btintel_pcie_submit_rx_work(data, urbd1->status,
|
|
buf->data);
|
|
if (ret) {
|
|
bt_dev_err(hdev, "RXQ: failed to submit rx request");
|
|
return;
|
|
}
|
|
|
|
cr_tia = (cr_tia + 1) % rxq->count;
|
|
data->ia.cr_tia[BTINTEL_PCIE_RXQ_NUM] = cr_tia;
|
|
ipc_print_ia_ring(data->hdev, &data->ia, BTINTEL_PCIE_RXQ_NUM);
|
|
}
|
|
}
|
|
|
|
static irqreturn_t btintel_pcie_msix_isr(int irq, void *data)
|
|
{
|
|
return IRQ_WAKE_THREAD;
|
|
}
|
|
|
|
static irqreturn_t btintel_pcie_irq_msix_handler(int irq, void *dev_id)
|
|
{
|
|
struct msix_entry *entry = dev_id;
|
|
struct btintel_pcie_data *data = btintel_pcie_get_data(entry);
|
|
u32 intr_fh, intr_hw;
|
|
|
|
spin_lock(&data->irq_lock);
|
|
intr_fh = btintel_pcie_rd_reg32(data, BTINTEL_PCIE_CSR_MSIX_FH_INT_CAUSES);
|
|
intr_hw = btintel_pcie_rd_reg32(data, BTINTEL_PCIE_CSR_MSIX_HW_INT_CAUSES);
|
|
|
|
/* Clear causes registers to avoid being handling the same cause */
|
|
btintel_pcie_wr_reg32(data, BTINTEL_PCIE_CSR_MSIX_FH_INT_CAUSES, intr_fh);
|
|
btintel_pcie_wr_reg32(data, BTINTEL_PCIE_CSR_MSIX_HW_INT_CAUSES, intr_hw);
|
|
spin_unlock(&data->irq_lock);
|
|
|
|
if (unlikely(!(intr_fh | intr_hw))) {
|
|
/* Ignore interrupt, inta == 0 */
|
|
return IRQ_NONE;
|
|
}
|
|
|
|
/* This interrupt is triggered by the firmware after updating
|
|
* boot_stage register and image_response register
|
|
*/
|
|
if (intr_hw & BTINTEL_PCIE_MSIX_HW_INT_CAUSES_GP0)
|
|
btintel_pcie_msix_gp0_handler(data);
|
|
|
|
/* For TX */
|
|
if (intr_fh & BTINTEL_PCIE_MSIX_FH_INT_CAUSES_0)
|
|
btintel_pcie_msix_tx_handle(data);
|
|
|
|
/* For RX */
|
|
if (intr_fh & BTINTEL_PCIE_MSIX_FH_INT_CAUSES_1)
|
|
btintel_pcie_msix_rx_handle(data);
|
|
|
|
/*
|
|
* Before sending the interrupt the HW disables it to prevent a nested
|
|
* interrupt. This is done by writing 1 to the corresponding bit in
|
|
* the mask register. After handling the interrupt, it should be
|
|
* re-enabled by clearing this bit. This register is defined as write 1
|
|
* clear (W1C) register, meaning that it's cleared by writing 1
|
|
* to the bit.
|
|
*/
|
|
btintel_pcie_wr_reg32(data, BTINTEL_PCIE_CSR_MSIX_AUTOMASK_ST,
|
|
BIT(entry->entry));
|
|
|
|
return IRQ_HANDLED;
|
|
}
|
|
|
|
/* This function requests the irq for MSI-X and registers the handlers per irq.
|
|
* Currently, it requests only 1 irq for all interrupt causes.
|
|
*/
|
|
static int btintel_pcie_setup_irq(struct btintel_pcie_data *data)
|
|
{
|
|
int err;
|
|
int num_irqs, i;
|
|
|
|
for (i = 0; i < BTINTEL_PCIE_MSIX_VEC_MAX; i++)
|
|
data->msix_entries[i].entry = i;
|
|
|
|
num_irqs = pci_alloc_irq_vectors(data->pdev, BTINTEL_PCIE_MSIX_VEC_MIN,
|
|
BTINTEL_PCIE_MSIX_VEC_MAX, PCI_IRQ_MSIX);
|
|
if (num_irqs < 0)
|
|
return num_irqs;
|
|
|
|
data->alloc_vecs = num_irqs;
|
|
data->msix_enabled = 1;
|
|
data->def_irq = 0;
|
|
|
|
/* setup irq handler */
|
|
for (i = 0; i < data->alloc_vecs; i++) {
|
|
struct msix_entry *msix_entry;
|
|
|
|
msix_entry = &data->msix_entries[i];
|
|
msix_entry->vector = pci_irq_vector(data->pdev, i);
|
|
|
|
err = devm_request_threaded_irq(&data->pdev->dev,
|
|
msix_entry->vector,
|
|
btintel_pcie_msix_isr,
|
|
btintel_pcie_irq_msix_handler,
|
|
IRQF_SHARED,
|
|
KBUILD_MODNAME,
|
|
msix_entry);
|
|
if (err) {
|
|
pci_free_irq_vectors(data->pdev);
|
|
data->alloc_vecs = 0;
|
|
return err;
|
|
}
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
struct btintel_pcie_causes_list {
|
|
u32 cause;
|
|
u32 mask_reg;
|
|
u8 cause_num;
|
|
};
|
|
|
|
static struct btintel_pcie_causes_list causes_list[] = {
|
|
{ BTINTEL_PCIE_MSIX_FH_INT_CAUSES_0, BTINTEL_PCIE_CSR_MSIX_FH_INT_MASK, 0x00 },
|
|
{ BTINTEL_PCIE_MSIX_FH_INT_CAUSES_1, BTINTEL_PCIE_CSR_MSIX_FH_INT_MASK, 0x01 },
|
|
{ BTINTEL_PCIE_MSIX_HW_INT_CAUSES_GP0, BTINTEL_PCIE_CSR_MSIX_HW_INT_MASK, 0x20 },
|
|
};
|
|
|
|
/* This function configures the interrupt masks for both HW_INT_CAUSES and
|
|
* FH_INT_CAUSES which are meaningful to us.
|
|
*
|
|
* After resetting BT function via PCIE FLR or FUNC_CTRL reset, the driver
|
|
* need to call this function again to configure since the masks
|
|
* are reset to 0xFFFFFFFF after reset.
|
|
*/
|
|
static void btintel_pcie_config_msix(struct btintel_pcie_data *data)
|
|
{
|
|
int i;
|
|
int val = data->def_irq | BTINTEL_PCIE_MSIX_NON_AUTO_CLEAR_CAUSE;
|
|
|
|
/* Set Non Auto Clear Cause */
|
|
for (i = 0; i < ARRAY_SIZE(causes_list); i++) {
|
|
btintel_pcie_wr_reg8(data,
|
|
BTINTEL_PCIE_CSR_MSIX_IVAR(causes_list[i].cause_num),
|
|
val);
|
|
btintel_pcie_clr_reg_bits(data,
|
|
causes_list[i].mask_reg,
|
|
causes_list[i].cause);
|
|
}
|
|
|
|
/* Save the initial interrupt mask */
|
|
data->fh_init_mask = ~btintel_pcie_rd_reg32(data, BTINTEL_PCIE_CSR_MSIX_FH_INT_MASK);
|
|
data->hw_init_mask = ~btintel_pcie_rd_reg32(data, BTINTEL_PCIE_CSR_MSIX_HW_INT_MASK);
|
|
}
|
|
|
|
static int btintel_pcie_config_pcie(struct pci_dev *pdev,
|
|
struct btintel_pcie_data *data)
|
|
{
|
|
int err;
|
|
|
|
err = pcim_enable_device(pdev);
|
|
if (err)
|
|
return err;
|
|
|
|
pci_set_master(pdev);
|
|
|
|
err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
|
|
if (err) {
|
|
err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
|
|
if (err)
|
|
return err;
|
|
}
|
|
|
|
err = pcim_iomap_regions(pdev, BIT(0), KBUILD_MODNAME);
|
|
if (err)
|
|
return err;
|
|
|
|
data->base_addr = pcim_iomap_table(pdev)[0];
|
|
if (!data->base_addr)
|
|
return -ENODEV;
|
|
|
|
err = btintel_pcie_setup_irq(data);
|
|
if (err)
|
|
return err;
|
|
|
|
/* Configure MSI-X with causes list */
|
|
btintel_pcie_config_msix(data);
|
|
|
|
return 0;
|
|
}
|
|
|
|
static void btintel_pcie_init_ci(struct btintel_pcie_data *data,
|
|
struct ctx_info *ci)
|
|
{
|
|
ci->version = 0x1;
|
|
ci->size = sizeof(*ci);
|
|
ci->config = 0x0000;
|
|
ci->addr_cr_hia = data->ia.cr_hia_p_addr;
|
|
ci->addr_tr_tia = data->ia.tr_tia_p_addr;
|
|
ci->addr_cr_tia = data->ia.cr_tia_p_addr;
|
|
ci->addr_tr_hia = data->ia.tr_hia_p_addr;
|
|
ci->num_cr_ia = BTINTEL_PCIE_NUM_QUEUES;
|
|
ci->num_tr_ia = BTINTEL_PCIE_NUM_QUEUES;
|
|
ci->addr_urbdq0 = data->txq.urbd0s_p_addr;
|
|
ci->addr_tfdq = data->txq.tfds_p_addr;
|
|
ci->num_tfdq = data->txq.count;
|
|
ci->num_urbdq0 = data->txq.count;
|
|
ci->tfdq_db_vec = BTINTEL_PCIE_TXQ_NUM;
|
|
ci->urbdq0_db_vec = BTINTEL_PCIE_TXQ_NUM;
|
|
ci->rbd_size = BTINTEL_PCIE_RBD_SIZE_4K;
|
|
ci->addr_frbdq = data->rxq.frbds_p_addr;
|
|
ci->num_frbdq = data->rxq.count;
|
|
ci->frbdq_db_vec = BTINTEL_PCIE_RXQ_NUM;
|
|
ci->addr_urbdq1 = data->rxq.urbd1s_p_addr;
|
|
ci->num_urbdq1 = data->rxq.count;
|
|
ci->urbdq_db_vec = BTINTEL_PCIE_RXQ_NUM;
|
|
}
|
|
|
|
static void btintel_pcie_free_txq_bufs(struct btintel_pcie_data *data,
|
|
struct txq *txq)
|
|
{
|
|
/* Free data buffers first */
|
|
dma_free_coherent(&data->pdev->dev, txq->count * BTINTEL_PCIE_BUFFER_SIZE,
|
|
txq->buf_v_addr, txq->buf_p_addr);
|
|
kfree(txq->bufs);
|
|
}
|
|
|
|
static int btintel_pcie_setup_txq_bufs(struct btintel_pcie_data *data,
|
|
struct txq *txq)
|
|
{
|
|
int i;
|
|
struct data_buf *buf;
|
|
|
|
/* Allocate the same number of buffers as the descriptor */
|
|
txq->bufs = kmalloc_array(txq->count, sizeof(*buf), GFP_KERNEL);
|
|
if (!txq->bufs)
|
|
return -ENOMEM;
|
|
|
|
/* Allocate full chunk of data buffer for DMA first and do indexing and
|
|
* initialization next, so it can be freed easily
|
|
*/
|
|
txq->buf_v_addr = dma_alloc_coherent(&data->pdev->dev,
|
|
txq->count * BTINTEL_PCIE_BUFFER_SIZE,
|
|
&txq->buf_p_addr,
|
|
GFP_KERNEL | __GFP_NOWARN);
|
|
if (!txq->buf_v_addr) {
|
|
kfree(txq->bufs);
|
|
return -ENOMEM;
|
|
}
|
|
|
|
/* Setup the allocated DMA buffer to bufs. Each data_buf should
|
|
* have virtual address and physical address
|
|
*/
|
|
for (i = 0; i < txq->count; i++) {
|
|
buf = &txq->bufs[i];
|
|
buf->data_p_addr = txq->buf_p_addr + (i * BTINTEL_PCIE_BUFFER_SIZE);
|
|
buf->data = txq->buf_v_addr + (i * BTINTEL_PCIE_BUFFER_SIZE);
|
|
}
|
|
|
|
return 0;
|
|
}
|
|
|
|
static void btintel_pcie_free_rxq_bufs(struct btintel_pcie_data *data,
|
|
struct rxq *rxq)
|
|
{
|
|
/* Free data buffers first */
|
|
dma_free_coherent(&data->pdev->dev, rxq->count * BTINTEL_PCIE_BUFFER_SIZE,
|
|
rxq->buf_v_addr, rxq->buf_p_addr);
|
|
kfree(rxq->bufs);
|
|
}
|
|
|
|
static int btintel_pcie_setup_rxq_bufs(struct btintel_pcie_data *data,
|
|
struct rxq *rxq)
|
|
{
|
|
int i;
|
|
struct data_buf *buf;
|
|
|
|
/* Allocate the same number of buffers as the descriptor */
|
|
rxq->bufs = kmalloc_array(rxq->count, sizeof(*buf), GFP_KERNEL);
|
|
if (!rxq->bufs)
|
|
return -ENOMEM;
|
|
|
|
/* Allocate full chunk of data buffer for DMA first and do indexing and
|
|
* initialization next, so it can be freed easily
|
|
*/
|
|
rxq->buf_v_addr = dma_alloc_coherent(&data->pdev->dev,
|
|
rxq->count * BTINTEL_PCIE_BUFFER_SIZE,
|
|
&rxq->buf_p_addr,
|
|
GFP_KERNEL | __GFP_NOWARN);
|
|
if (!rxq->buf_v_addr) {
|
|
kfree(rxq->bufs);
|
|
return -ENOMEM;
|
|
}
|
|
|
|
/* Setup the allocated DMA buffer to bufs. Each data_buf should
|
|
* have virtual address and physical address
|
|
*/
|
|
for (i = 0; i < rxq->count; i++) {
|
|
buf = &rxq->bufs[i];
|
|
buf->data_p_addr = rxq->buf_p_addr + (i * BTINTEL_PCIE_BUFFER_SIZE);
|
|
buf->data = rxq->buf_v_addr + (i * BTINTEL_PCIE_BUFFER_SIZE);
|
|
}
|
|
|
|
return 0;
|
|
}
|
|
|
|
static void btintel_pcie_setup_ia(struct btintel_pcie_data *data,
|
|
dma_addr_t p_addr, void *v_addr,
|
|
struct ia *ia)
|
|
{
|
|
/* TR Head Index Array */
|
|
ia->tr_hia_p_addr = p_addr;
|
|
ia->tr_hia = v_addr;
|
|
|
|
/* TR Tail Index Array */
|
|
ia->tr_tia_p_addr = p_addr + sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES;
|
|
ia->tr_tia = v_addr + sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES;
|
|
|
|
/* CR Head index Array */
|
|
ia->cr_hia_p_addr = p_addr + (sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES * 2);
|
|
ia->cr_hia = v_addr + (sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES * 2);
|
|
|
|
/* CR Tail Index Array */
|
|
ia->cr_tia_p_addr = p_addr + (sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES * 3);
|
|
ia->cr_tia = v_addr + (sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES * 3);
|
|
}
|
|
|
|
static void btintel_pcie_free(struct btintel_pcie_data *data)
|
|
{
|
|
btintel_pcie_free_rxq_bufs(data, &data->rxq);
|
|
btintel_pcie_free_txq_bufs(data, &data->txq);
|
|
|
|
dma_pool_free(data->dma_pool, data->dma_v_addr, data->dma_p_addr);
|
|
dma_pool_destroy(data->dma_pool);
|
|
}
|
|
|
|
/* Allocate tx and rx queues, any related data structures and buffers.
|
|
*/
|
|
static int btintel_pcie_alloc(struct btintel_pcie_data *data)
|
|
{
|
|
int err = 0;
|
|
size_t total;
|
|
dma_addr_t p_addr;
|
|
void *v_addr;
|
|
|
|
/* Allocate the chunk of DMA memory for descriptors, index array, and
|
|
* context information, instead of allocating individually.
|
|
* The DMA memory for data buffer is allocated while setting up the
|
|
* each queue.
|
|
*
|
|
* Total size is sum of the following
|
|
* + size of TFD * Number of descriptors in queue
|
|
* + size of URBD0 * Number of descriptors in queue
|
|
* + size of FRBD * Number of descriptors in queue
|
|
* + size of URBD1 * Number of descriptors in queue
|
|
* + size of index * Number of queues(2) * type of index array(4)
|
|
* + size of context information
|
|
*/
|
|
total = (sizeof(struct tfd) + sizeof(struct urbd0) + sizeof(struct frbd)
|
|
+ sizeof(struct urbd1)) * BTINTEL_DESCS_COUNT;
|
|
|
|
/* Add the sum of size of index array and size of ci struct */
|
|
total += (sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES * 4) + sizeof(struct ctx_info);
|
|
|
|
/* Allocate DMA Pool */
|
|
data->dma_pool = dma_pool_create(KBUILD_MODNAME, &data->pdev->dev,
|
|
total, BTINTEL_PCIE_DMA_POOL_ALIGNMENT, 0);
|
|
if (!data->dma_pool) {
|
|
err = -ENOMEM;
|
|
goto exit_error;
|
|
}
|
|
|
|
v_addr = dma_pool_zalloc(data->dma_pool, GFP_KERNEL | __GFP_NOWARN,
|
|
&p_addr);
|
|
if (!v_addr) {
|
|
dma_pool_destroy(data->dma_pool);
|
|
err = -ENOMEM;
|
|
goto exit_error;
|
|
}
|
|
|
|
data->dma_p_addr = p_addr;
|
|
data->dma_v_addr = v_addr;
|
|
|
|
/* Setup descriptor count */
|
|
data->txq.count = BTINTEL_DESCS_COUNT;
|
|
data->rxq.count = BTINTEL_DESCS_COUNT;
|
|
|
|
/* Setup tfds */
|
|
data->txq.tfds_p_addr = p_addr;
|
|
data->txq.tfds = v_addr;
|
|
|
|
p_addr += (sizeof(struct tfd) * BTINTEL_DESCS_COUNT);
|
|
v_addr += (sizeof(struct tfd) * BTINTEL_DESCS_COUNT);
|
|
|
|
/* Setup urbd0 */
|
|
data->txq.urbd0s_p_addr = p_addr;
|
|
data->txq.urbd0s = v_addr;
|
|
|
|
p_addr += (sizeof(struct urbd0) * BTINTEL_DESCS_COUNT);
|
|
v_addr += (sizeof(struct urbd0) * BTINTEL_DESCS_COUNT);
|
|
|
|
/* Setup FRBD*/
|
|
data->rxq.frbds_p_addr = p_addr;
|
|
data->rxq.frbds = v_addr;
|
|
|
|
p_addr += (sizeof(struct frbd) * BTINTEL_DESCS_COUNT);
|
|
v_addr += (sizeof(struct frbd) * BTINTEL_DESCS_COUNT);
|
|
|
|
/* Setup urbd1 */
|
|
data->rxq.urbd1s_p_addr = p_addr;
|
|
data->rxq.urbd1s = v_addr;
|
|
|
|
p_addr += (sizeof(struct urbd1) * BTINTEL_DESCS_COUNT);
|
|
v_addr += (sizeof(struct urbd1) * BTINTEL_DESCS_COUNT);
|
|
|
|
/* Setup data buffers for txq */
|
|
err = btintel_pcie_setup_txq_bufs(data, &data->txq);
|
|
if (err)
|
|
goto exit_error_pool;
|
|
|
|
/* Setup data buffers for rxq */
|
|
err = btintel_pcie_setup_rxq_bufs(data, &data->rxq);
|
|
if (err)
|
|
goto exit_error_txq;
|
|
|
|
/* Setup Index Array */
|
|
btintel_pcie_setup_ia(data, p_addr, v_addr, &data->ia);
|
|
|
|
/* Setup Context Information */
|
|
p_addr += sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES * 4;
|
|
v_addr += sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES * 4;
|
|
|
|
data->ci = v_addr;
|
|
data->ci_p_addr = p_addr;
|
|
|
|
/* Initialize the CI */
|
|
btintel_pcie_init_ci(data, data->ci);
|
|
|
|
return 0;
|
|
|
|
exit_error_txq:
|
|
btintel_pcie_free_txq_bufs(data, &data->txq);
|
|
exit_error_pool:
|
|
dma_pool_free(data->dma_pool, data->dma_v_addr, data->dma_p_addr);
|
|
dma_pool_destroy(data->dma_pool);
|
|
exit_error:
|
|
return err;
|
|
}
|
|
|
|
static int btintel_pcie_open(struct hci_dev *hdev)
|
|
{
|
|
bt_dev_dbg(hdev, "");
|
|
|
|
return 0;
|
|
}
|
|
|
|
static int btintel_pcie_close(struct hci_dev *hdev)
|
|
{
|
|
bt_dev_dbg(hdev, "");
|
|
|
|
return 0;
|
|
}
|
|
|
|
static int btintel_pcie_inject_cmd_complete(struct hci_dev *hdev, __u16 opcode)
|
|
{
|
|
struct sk_buff *skb;
|
|
struct hci_event_hdr *hdr;
|
|
struct hci_ev_cmd_complete *evt;
|
|
|
|
skb = bt_skb_alloc(sizeof(*hdr) + sizeof(*evt) + 1, GFP_KERNEL);
|
|
if (!skb)
|
|
return -ENOMEM;
|
|
|
|
hdr = (struct hci_event_hdr *)skb_put(skb, sizeof(*hdr));
|
|
hdr->evt = HCI_EV_CMD_COMPLETE;
|
|
hdr->plen = sizeof(*evt) + 1;
|
|
|
|
evt = (struct hci_ev_cmd_complete *)skb_put(skb, sizeof(*evt));
|
|
evt->ncmd = 0x01;
|
|
evt->opcode = cpu_to_le16(opcode);
|
|
|
|
*(u8 *)skb_put(skb, 1) = 0x00;
|
|
|
|
hci_skb_pkt_type(skb) = HCI_EVENT_PKT;
|
|
|
|
return hci_recv_frame(hdev, skb);
|
|
}
|
|
|
|
static int btintel_pcie_send_frame(struct hci_dev *hdev,
|
|
struct sk_buff *skb)
|
|
{
|
|
struct btintel_pcie_data *data = hci_get_drvdata(hdev);
|
|
struct hci_command_hdr *cmd;
|
|
__u16 opcode = ~0;
|
|
int ret;
|
|
u32 type;
|
|
u32 old_ctxt;
|
|
|
|
/* Due to the fw limitation, the type header of the packet should be
|
|
* 4 bytes unlike 1 byte for UART. In UART, the firmware can read
|
|
* the first byte to get the packet type and redirect the rest of data
|
|
* packet to the right handler.
|
|
*
|
|
* But for PCIe, THF(Transfer Flow Handler) fetches the 4 bytes of data
|
|
* from DMA memory and by the time it reads the first 4 bytes, it has
|
|
* already consumed some part of packet. Thus the packet type indicator
|
|
* for iBT PCIe is 4 bytes.
|
|
*
|
|
* Luckily, when HCI core creates the skb, it allocates 8 bytes of
|
|
* head room for profile and driver use, and before sending the data
|
|
* to the device, append the iBT PCIe packet type in the front.
|
|
*/
|
|
switch (hci_skb_pkt_type(skb)) {
|
|
case HCI_COMMAND_PKT:
|
|
type = BTINTEL_PCIE_HCI_CMD_PKT;
|
|
cmd = (void *)skb->data;
|
|
opcode = le16_to_cpu(cmd->opcode);
|
|
if (btintel_test_flag(hdev, INTEL_BOOTLOADER)) {
|
|
struct hci_command_hdr *cmd = (void *)skb->data;
|
|
__u16 opcode = le16_to_cpu(cmd->opcode);
|
|
|
|
/* When the 0xfc01 command is issued to boot into
|
|
* the operational firmware, it will actually not
|
|
* send a command complete event. To keep the flow
|
|
* control working inject that event here.
|
|
*/
|
|
if (opcode == 0xfc01)
|
|
btintel_pcie_inject_cmd_complete(hdev, opcode);
|
|
}
|
|
hdev->stat.cmd_tx++;
|
|
break;
|
|
case HCI_ACLDATA_PKT:
|
|
type = BTINTEL_PCIE_HCI_ACL_PKT;
|
|
hdev->stat.acl_tx++;
|
|
break;
|
|
case HCI_SCODATA_PKT:
|
|
type = BTINTEL_PCIE_HCI_SCO_PKT;
|
|
hdev->stat.sco_tx++;
|
|
break;
|
|
case HCI_ISODATA_PKT:
|
|
type = BTINTEL_PCIE_HCI_ISO_PKT;
|
|
break;
|
|
default:
|
|
bt_dev_err(hdev, "Unknown HCI packet type");
|
|
return -EILSEQ;
|
|
}
|
|
memcpy(skb_push(skb, BTINTEL_PCIE_HCI_TYPE_LEN), &type,
|
|
BTINTEL_PCIE_HCI_TYPE_LEN);
|
|
|
|
ret = btintel_pcie_send_sync(data, skb);
|
|
if (ret) {
|
|
hdev->stat.err_tx++;
|
|
bt_dev_err(hdev, "Failed to send frame (%d)", ret);
|
|
goto exit_error;
|
|
}
|
|
|
|
if (type == BTINTEL_PCIE_HCI_CMD_PKT &&
|
|
(opcode == HCI_OP_RESET || opcode == 0xfc01)) {
|
|
old_ctxt = data->alive_intr_ctxt;
|
|
data->alive_intr_ctxt =
|
|
(opcode == 0xfc01 ? BTINTEL_PCIE_INTEL_HCI_RESET1 :
|
|
BTINTEL_PCIE_HCI_RESET);
|
|
bt_dev_dbg(data->hdev, "sent cmd: 0x%4.4x alive context changed: %s -> %s",
|
|
opcode, btintel_pcie_alivectxt_state2str(old_ctxt),
|
|
btintel_pcie_alivectxt_state2str(data->alive_intr_ctxt));
|
|
if (opcode == HCI_OP_RESET) {
|
|
data->gp0_received = false;
|
|
ret = wait_event_timeout(data->gp0_wait_q,
|
|
data->gp0_received,
|
|
msecs_to_jiffies(BTINTEL_DEFAULT_INTR_TIMEOUT_MS));
|
|
if (!ret) {
|
|
hdev->stat.err_tx++;
|
|
bt_dev_err(hdev, "No alive interrupt received for %s",
|
|
btintel_pcie_alivectxt_state2str(data->alive_intr_ctxt));
|
|
ret = -ETIME;
|
|
goto exit_error;
|
|
}
|
|
}
|
|
}
|
|
hdev->stat.byte_tx += skb->len;
|
|
kfree_skb(skb);
|
|
|
|
exit_error:
|
|
return ret;
|
|
}
|
|
|
|
static void btintel_pcie_release_hdev(struct btintel_pcie_data *data)
|
|
{
|
|
struct hci_dev *hdev;
|
|
|
|
hdev = data->hdev;
|
|
hci_unregister_dev(hdev);
|
|
hci_free_dev(hdev);
|
|
data->hdev = NULL;
|
|
}
|
|
|
|
static int btintel_pcie_setup(struct hci_dev *hdev)
|
|
{
|
|
const u8 param[1] = { 0xFF };
|
|
struct intel_version_tlv ver_tlv;
|
|
struct sk_buff *skb;
|
|
int err;
|
|
|
|
BT_DBG("%s", hdev->name);
|
|
|
|
skb = __hci_cmd_sync(hdev, 0xfc05, 1, param, HCI_CMD_TIMEOUT);
|
|
if (IS_ERR(skb)) {
|
|
bt_dev_err(hdev, "Reading Intel version command failed (%ld)",
|
|
PTR_ERR(skb));
|
|
return PTR_ERR(skb);
|
|
}
|
|
|
|
/* Check the status */
|
|
if (skb->data[0]) {
|
|
bt_dev_err(hdev, "Intel Read Version command failed (%02x)",
|
|
skb->data[0]);
|
|
err = -EIO;
|
|
goto exit_error;
|
|
}
|
|
|
|
/* Apply the common HCI quirks for Intel device */
|
|
set_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks);
|
|
set_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, &hdev->quirks);
|
|
set_bit(HCI_QUIRK_NON_PERSISTENT_DIAG, &hdev->quirks);
|
|
|
|
/* Set up the quality report callback for Intel devices */
|
|
hdev->set_quality_report = btintel_set_quality_report;
|
|
|
|
memset(&ver_tlv, 0, sizeof(ver_tlv));
|
|
/* For TLV type device, parse the tlv data */
|
|
err = btintel_parse_version_tlv(hdev, &ver_tlv, skb);
|
|
if (err) {
|
|
bt_dev_err(hdev, "Failed to parse TLV version information");
|
|
goto exit_error;
|
|
}
|
|
|
|
switch (INTEL_HW_PLATFORM(ver_tlv.cnvi_bt)) {
|
|
case 0x37:
|
|
break;
|
|
default:
|
|
bt_dev_err(hdev, "Unsupported Intel hardware platform (0x%2x)",
|
|
INTEL_HW_PLATFORM(ver_tlv.cnvi_bt));
|
|
err = -EINVAL;
|
|
goto exit_error;
|
|
}
|
|
|
|
/* Check for supported iBT hardware variants of this firmware
|
|
* loading method.
|
|
*
|
|
* This check has been put in place to ensure correct forward
|
|
* compatibility options when newer hardware variants come
|
|
* along.
|
|
*/
|
|
switch (INTEL_HW_VARIANT(ver_tlv.cnvi_bt)) {
|
|
case 0x1e: /* BzrI */
|
|
/* Display version information of TLV type */
|
|
btintel_version_info_tlv(hdev, &ver_tlv);
|
|
|
|
/* Apply the device specific HCI quirks for TLV based devices
|
|
*
|
|
* All TLV based devices support WBS
|
|
*/
|
|
set_bit(HCI_QUIRK_WIDEBAND_SPEECH_SUPPORTED, &hdev->quirks);
|
|
|
|
/* Setup MSFT Extension support */
|
|
btintel_set_msft_opcode(hdev,
|
|
INTEL_HW_VARIANT(ver_tlv.cnvi_bt));
|
|
|
|
err = btintel_bootloader_setup_tlv(hdev, &ver_tlv);
|
|
if (err)
|
|
goto exit_error;
|
|
break;
|
|
default:
|
|
bt_dev_err(hdev, "Unsupported Intel hw variant (%u)",
|
|
INTEL_HW_VARIANT(ver_tlv.cnvi_bt));
|
|
err = -EINVAL;
|
|
goto exit_error;
|
|
break;
|
|
}
|
|
|
|
btintel_print_fseq_info(hdev);
|
|
exit_error:
|
|
kfree_skb(skb);
|
|
|
|
return err;
|
|
}
|
|
|
|
static int btintel_pcie_setup_hdev(struct btintel_pcie_data *data)
|
|
{
|
|
int err;
|
|
struct hci_dev *hdev;
|
|
|
|
hdev = hci_alloc_dev_priv(sizeof(struct btintel_data));
|
|
if (!hdev)
|
|
return -ENOMEM;
|
|
|
|
hdev->bus = HCI_PCI;
|
|
hci_set_drvdata(hdev, data);
|
|
|
|
data->hdev = hdev;
|
|
SET_HCIDEV_DEV(hdev, &data->pdev->dev);
|
|
|
|
hdev->manufacturer = 2;
|
|
hdev->open = btintel_pcie_open;
|
|
hdev->close = btintel_pcie_close;
|
|
hdev->send = btintel_pcie_send_frame;
|
|
hdev->setup = btintel_pcie_setup;
|
|
hdev->shutdown = btintel_shutdown_combined;
|
|
hdev->hw_error = btintel_hw_error;
|
|
hdev->set_diag = btintel_set_diag;
|
|
hdev->set_bdaddr = btintel_set_bdaddr;
|
|
|
|
err = hci_register_dev(hdev);
|
|
if (err < 0) {
|
|
BT_ERR("Failed to register to hdev (%d)", err);
|
|
goto exit_error;
|
|
}
|
|
|
|
return 0;
|
|
|
|
exit_error:
|
|
hci_free_dev(hdev);
|
|
return err;
|
|
}
|
|
|
|
static int btintel_pcie_probe(struct pci_dev *pdev,
|
|
const struct pci_device_id *ent)
|
|
{
|
|
int err;
|
|
struct btintel_pcie_data *data;
|
|
|
|
if (!pdev)
|
|
return -ENODEV;
|
|
|
|
data = devm_kzalloc(&pdev->dev, sizeof(*data), GFP_KERNEL);
|
|
if (!data)
|
|
return -ENOMEM;
|
|
|
|
data->pdev = pdev;
|
|
|
|
spin_lock_init(&data->irq_lock);
|
|
spin_lock_init(&data->hci_rx_lock);
|
|
|
|
init_waitqueue_head(&data->gp0_wait_q);
|
|
data->gp0_received = false;
|
|
|
|
init_waitqueue_head(&data->tx_wait_q);
|
|
data->tx_wait_done = false;
|
|
|
|
data->workqueue = alloc_ordered_workqueue(KBUILD_MODNAME, WQ_HIGHPRI);
|
|
if (!data->workqueue)
|
|
return -ENOMEM;
|
|
|
|
skb_queue_head_init(&data->rx_skb_q);
|
|
INIT_WORK(&data->rx_work, btintel_pcie_rx_work);
|
|
|
|
data->boot_stage_cache = 0x00;
|
|
data->img_resp_cache = 0x00;
|
|
|
|
err = btintel_pcie_config_pcie(pdev, data);
|
|
if (err)
|
|
goto exit_error;
|
|
|
|
pci_set_drvdata(pdev, data);
|
|
|
|
err = btintel_pcie_alloc(data);
|
|
if (err)
|
|
goto exit_error;
|
|
|
|
err = btintel_pcie_enable_bt(data);
|
|
if (err)
|
|
goto exit_error;
|
|
|
|
/* CNV information (CNVi and CNVr) is in CSR */
|
|
data->cnvi = btintel_pcie_rd_reg32(data, BTINTEL_PCIE_CSR_HW_REV_REG);
|
|
|
|
data->cnvr = btintel_pcie_rd_reg32(data, BTINTEL_PCIE_CSR_RF_ID_REG);
|
|
|
|
err = btintel_pcie_start_rx(data);
|
|
if (err)
|
|
goto exit_error;
|
|
|
|
err = btintel_pcie_setup_hdev(data);
|
|
if (err)
|
|
goto exit_error;
|
|
|
|
bt_dev_dbg(data->hdev, "cnvi: 0x%8.8x cnvr: 0x%8.8x", data->cnvi,
|
|
data->cnvr);
|
|
return 0;
|
|
|
|
exit_error:
|
|
/* reset device before exit */
|
|
btintel_pcie_reset_bt(data);
|
|
|
|
pci_clear_master(pdev);
|
|
|
|
pci_set_drvdata(pdev, NULL);
|
|
|
|
return err;
|
|
}
|
|
|
|
static void btintel_pcie_remove(struct pci_dev *pdev)
|
|
{
|
|
struct btintel_pcie_data *data;
|
|
|
|
data = pci_get_drvdata(pdev);
|
|
|
|
btintel_pcie_reset_bt(data);
|
|
for (int i = 0; i < data->alloc_vecs; i++) {
|
|
struct msix_entry *msix_entry;
|
|
|
|
msix_entry = &data->msix_entries[i];
|
|
free_irq(msix_entry->vector, msix_entry);
|
|
}
|
|
|
|
pci_free_irq_vectors(pdev);
|
|
|
|
btintel_pcie_release_hdev(data);
|
|
|
|
flush_work(&data->rx_work);
|
|
|
|
destroy_workqueue(data->workqueue);
|
|
|
|
btintel_pcie_free(data);
|
|
|
|
pci_clear_master(pdev);
|
|
|
|
pci_set_drvdata(pdev, NULL);
|
|
}
|
|
|
|
static struct pci_driver btintel_pcie_driver = {
|
|
.name = KBUILD_MODNAME,
|
|
.id_table = btintel_pcie_table,
|
|
.probe = btintel_pcie_probe,
|
|
.remove = btintel_pcie_remove,
|
|
};
|
|
module_pci_driver(btintel_pcie_driver);
|
|
|
|
MODULE_AUTHOR("Tedd Ho-Jeong An <tedd.an@intel.com>");
|
|
MODULE_DESCRIPTION("Intel Bluetooth PCIe transport driver ver " VERSION);
|
|
MODULE_VERSION(VERSION);
|
|
MODULE_LICENSE("GPL");
|