From: Mark Chen mark-yw.chen@mediatek.com
[ Upstream commit 2fc967cc0dadad6735448cfbcbc77fe0ea30203d ]
According to the MCU firmware behavior, as the driver is aware of the notification of the interrupt source FW_MAILBOX_INT that shows the MCU completed delivered a core dump piece to the host, the driver must acknowledge the MCU with the register PH2DSM0R bit PH2DSM0R_DRIVER_OWN to notify the MCU to handle the next core dump piece.
Fixes: db57b625912a ("Bluetooth: btmtksdio: add support of processing firmware coredump and log") Co-developed-by: Sean Wang sean.wang@mediatek.com Signed-off-by: Sean Wang sean.wang@mediatek.com Signed-off-by: Mark Chen mark-yw.chen@mediatek.com Signed-off-by: Marcel Holtmann marcel@holtmann.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/bluetooth/btmtksdio.c | 10 ++++++++++ 1 file changed, 10 insertions(+)
diff --git a/drivers/bluetooth/btmtksdio.c b/drivers/bluetooth/btmtksdio.c index b5ea8d3bffaa..d4e2541a4873 100644 --- a/drivers/bluetooth/btmtksdio.c +++ b/drivers/bluetooth/btmtksdio.c @@ -87,8 +87,12 @@ MODULE_DEVICE_TABLE(sdio, btmtksdio_table); #define RX_DONE_INT BIT(1) #define TX_EMPTY BIT(2) #define TX_FIFO_OVERFLOW BIT(8) +#define FW_MAILBOX_INT BIT(15) #define RX_PKT_LEN GENMASK(31, 16)
+#define MTK_REG_PH2DSM0R 0xc4 +#define PH2DSM0R_DRIVER_OWN BIT(0) + #define MTK_REG_CTDR 0x18
#define MTK_REG_CRDR 0x1c @@ -481,6 +485,12 @@ static void btmtksdio_txrx_work(struct work_struct *work) */ sdio_writel(bdev->func, int_status, MTK_REG_CHISR, NULL);
+ if ((int_status & FW_MAILBOX_INT) && + bdev->data->chipid == 0x7921) { + sdio_writel(bdev->func, PH2DSM0R_DRIVER_OWN, + MTK_REG_PH2DSM0R, 0); + } + if (int_status & FW_OWN_BACK_INT) bt_dev_dbg(bdev->hdev, "Get fw own back");