[PATCH v1 0/2] Serialize execution environment changes for MHI

[PATCH v1 0/2] Serialize execution environment changes for MHI

[Bhaumik Bhatt]

During full boot chain firmware download, the PM state worker downloads the AMSS
image after waiting for the SBL execution environment change in PBL mode itself.
Since getting rid of the firmware load worker thread, this design needs to
change and MHI host must download the AMSS image from the SBL mode of PM state
worker thread instead.

Also ensure that EE changes are handled only from appropriate places and occur
one after another and handle only PBL or RDDM EE changes as critical events
directly from the interrupt handler and the status callback is given to the
controller drivers promptly.

Bhaumik Bhatt (2):
  bus: mhi: core: Download AMSS image from appropriate function
  bus: mhi: core: Process execution environment changes serially

 drivers/bus/mhi/core/boot.c     | 44 ++++++++++++++++++++---------------------
 drivers/bus/mhi/core/internal.h |  1 +
 drivers/bus/mhi/core/main.c     | 14 +++++++------
 drivers/bus/mhi/core/pm.c       | 14 +++++++++----
 4 files changed, 40 insertions(+), 33 deletions(-)

-- 
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

[PATCH v1 1/2] bus: mhi: core: Download AMSS image from appropriate function

[Bhaumik Bhatt]

During full boot chain firmware download, the PM state worker
downloads the AMSS image after waiting for the SBL execution
environment change in PBL mode itself. Since getting rid of the
firmware load worker thread, this design needs to change and MHI
host must download the AMSS image from the SBL mode of PM state
worker thread instead. Since the full boot chain firmware
download is associated with a synchronous power up and has MHI
host waiting for a transition to mission mode with a timeout, we
can skip creating any devices (or probing any client drivers) in
SBL mode transition and proceed directly with the AMSS image
download.

This means that if MHI host driver is not responsible for the
AMSS image download or the controller plans to have client
drivers opening any SBL channels, for example, to download images
or monitor debug logs for memory allocations or power management,
the device can be powered up asynchronously.

Signed-off-by: Bhaumik Bhatt <bbhatt@codeaurora.org>
---
 drivers/bus/mhi/core/boot.c     | 44 ++++++++++++++++++++---------------------
 drivers/bus/mhi/core/internal.h |  1 +
 drivers/bus/mhi/core/pm.c       |  8 ++++++--
 3 files changed, 28 insertions(+), 25 deletions(-)

diff --git a/drivers/bus/mhi/core/boot.c b/drivers/bus/mhi/core/boot.c
index c2546bf..b91c7b3 100644
--- a/drivers/bus/mhi/core/boot.c
+++ b/drivers/bus/mhi/core/boot.c
@@ -389,7 +389,6 @@ static void mhi_firmware_copy(struct mhi_controller *mhi_cntrl,
 void mhi_fw_load_handler(struct mhi_controller *mhi_cntrl)
 {
 	const struct firmware *firmware = NULL;
-	struct image_info *image_info;
 	struct device *dev = &mhi_cntrl->mhi_dev->dev;
 	const char *fw_name;
 	void *buf;
@@ -500,28 +499,6 @@ void mhi_fw_load_handler(struct mhi_controller *mhi_cntrl)
 		goto error_ready_state;
 	}
 
-	/* Wait for the SBL event */
-	ret = wait_event_timeout(mhi_cntrl->state_event,
-				 mhi_cntrl->ee == MHI_EE_SBL ||
-				 MHI_PM_IN_ERROR_STATE(mhi_cntrl->pm_state),
-				 msecs_to_jiffies(mhi_cntrl->timeout_ms));
-
-	if (!ret || MHI_PM_IN_ERROR_STATE(mhi_cntrl->pm_state)) {
-		dev_err(dev, "MHI did not enter SBL\n");
-		goto error_ready_state;
-	}
-
-	/* Start full firmware image download */
-	image_info = mhi_cntrl->fbc_image;
-	ret = mhi_fw_load_bhie(mhi_cntrl,
-			       /* Vector table is the last entry */
-			       &image_info->mhi_buf[image_info->entries - 1]);
-	if (ret) {
-		dev_err(dev, "MHI did not load image over BHIe, ret: %d\n",
-			ret);
-		goto error_fw_load;
-	}
-
 	return;
 
 error_ready_state:
@@ -532,3 +509,24 @@ void mhi_fw_load_handler(struct mhi_controller *mhi_cntrl)
 	mhi_cntrl->pm_state = MHI_PM_FW_DL_ERR;
 	wake_up_all(&mhi_cntrl->state_event);
 }
+
+int mhi_download_amss_image(struct mhi_controller *mhi_cntrl)
+{
+	struct image_info *image_info = mhi_cntrl->fbc_image;
+	struct device *dev = &mhi_cntrl->mhi_dev->dev;
+	int ret;
+
+	if (!image_info)
+		return -EIO;
+
+	ret = mhi_fw_load_bhie(mhi_cntrl,
+			       /* Vector table is the last entry */
+			       &image_info->mhi_buf[image_info->entries - 1]);
+	if (ret) {
+		dev_err(dev, "MHI did not load AMSS, ret:%d\n", ret);
+		mhi_cntrl->pm_state = MHI_PM_FW_DL_ERR;
+		wake_up_all(&mhi_cntrl->state_event);
+	}
+
+	return ret;
+}
diff --git a/drivers/bus/mhi/core/internal.h b/drivers/bus/mhi/core/internal.h
index 6f80ec3..6f37439 100644
--- a/drivers/bus/mhi/core/internal.h
+++ b/drivers/bus/mhi/core/internal.h
@@ -619,6 +619,7 @@ int mhi_pm_m3_transition(struct mhi_controller *mhi_cntrl);
 int __mhi_device_get_sync(struct mhi_controller *mhi_cntrl);
 int mhi_send_cmd(struct mhi_controller *mhi_cntrl, struct mhi_chan *mhi_chan,
 		 enum mhi_cmd_type cmd);
+int mhi_download_amss_image(struct mhi_controller *mhi_cntrl);
 static inline bool mhi_is_active(struct mhi_controller *mhi_cntrl)
 {
 	return (mhi_cntrl->dev_state >= MHI_STATE_M0 &&
diff --git a/drivers/bus/mhi/core/pm.c b/drivers/bus/mhi/core/pm.c
index a671f58..aa156b9 100644
--- a/drivers/bus/mhi/core/pm.c
+++ b/drivers/bus/mhi/core/pm.c
@@ -752,9 +752,13 @@ void mhi_pm_st_worker(struct work_struct *work)
 			/*
 			 * The MHI devices are only created when the client
 			 * device switches its Execution Environment (EE) to
-			 * either SBL or AMSS states
+			 * either SBL or AMSS states. If FBC download is
+			 * expected, we do not create devices in SBL.
 			 */
-			mhi_create_devices(mhi_cntrl);
+			if (mhi_cntrl->fbc_download)
+				mhi_download_amss_image(mhi_cntrl);
+			else
+				mhi_create_devices(mhi_cntrl);
 			break;
 		case DEV_ST_TRANSITION_MISSION_MODE:
 			mhi_pm_mission_mode_transition(mhi_cntrl);
-- 
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

[PATCH v1 2/2] bus: mhi: core: Process execution environment changes serially

[Bhaumik Bhatt]

In current design, whenever the BHI interrupt is fired, the execution
environment is updated. This can cause race conditions and impede any
ongoing power up/down processing. For example, if a power down is in
progress and the host has updated the execution environment to a
local "disabled" state, any BHI interrupt firing later could replace
it with the value from the BHI EE register. Another example would be
that the device can enter mission mode while device creation for SBL
is still going on, leading to multiple attempts at opening the same
channel.

Ensure that EE changes are handled only from appropriate places and
occur one after another and handle only PBL or RDDM EE changes as
critical events directly from the interrupt handler. This also makes
sure that we use the correct execution environment to notify the
controller driver when the device resets to one of the PBL execution
environments.

Signed-off-by: Bhaumik Bhatt <bbhatt@codeaurora.org>
---
 drivers/bus/mhi/core/main.c | 14 ++++++++------
 drivers/bus/mhi/core/pm.c   |  6 ++++--
 2 files changed, 12 insertions(+), 8 deletions(-)

diff --git a/drivers/bus/mhi/core/main.c b/drivers/bus/mhi/core/main.c
index 4eb93d8..cd712f2 100644
--- a/drivers/bus/mhi/core/main.c
+++ b/drivers/bus/mhi/core/main.c
@@ -377,7 +377,7 @@ irqreturn_t mhi_intvec_threaded_handler(int irq_number, void *priv)
 	struct device *dev = &mhi_cntrl->mhi_dev->dev;
 	enum mhi_state state = MHI_STATE_MAX;
 	enum mhi_pm_state pm_state = 0;
-	enum mhi_ee_type ee = 0;
+	enum mhi_ee_type ee = MHI_EE_MAX;
 
 	write_lock_irq(&mhi_cntrl->pm_lock);
 	if (!MHI_REG_ACCESS_VALID(mhi_cntrl->pm_state)) {
@@ -386,8 +386,7 @@ irqreturn_t mhi_intvec_threaded_handler(int irq_number, void *priv)
 	}
 
 	state = mhi_get_mhi_state(mhi_cntrl);
-	ee = mhi_cntrl->ee;
-	mhi_cntrl->ee = mhi_get_exec_env(mhi_cntrl);
+	ee = mhi_get_exec_env(mhi_cntrl);
 	dev_dbg(dev, "local ee:%s device ee:%s dev_state:%s\n",
 		TO_MHI_EXEC_STR(mhi_cntrl->ee), TO_MHI_EXEC_STR(ee),
 		TO_MHI_STATE_STR(state));
@@ -405,8 +404,9 @@ irqreturn_t mhi_intvec_threaded_handler(int irq_number, void *priv)
 		if (!mhi_is_active(mhi_cntrl))
 			goto exit_intvec;
 
-		if (mhi_cntrl->ee == MHI_EE_RDDM && mhi_cntrl->ee != ee) {
+		if (ee == MHI_EE_RDDM && mhi_cntrl->ee != MHI_EE_RDDM) {
 			mhi_cntrl->status_cb(mhi_cntrl, MHI_CB_EE_RDDM);
+			mhi_cntrl->ee = ee;
 			wake_up_all(&mhi_cntrl->state_event);
 		}
 		goto exit_intvec;
@@ -416,10 +416,12 @@ irqreturn_t mhi_intvec_threaded_handler(int irq_number, void *priv)
 		wake_up_all(&mhi_cntrl->state_event);
 
 		/* For fatal errors, we let controller decide next step */
-		if (MHI_IN_PBL(ee))
+		if (MHI_IN_PBL(ee)) {
 			mhi_cntrl->status_cb(mhi_cntrl, MHI_CB_FATAL_ERROR);
-		else
+			mhi_cntrl->ee = ee;
+		} else {
 			mhi_pm_sys_err_handler(mhi_cntrl);
+		}
 	}
 
 exit_intvec:
diff --git a/drivers/bus/mhi/core/pm.c b/drivers/bus/mhi/core/pm.c
index aa156b9..c1b18d6 100644
--- a/drivers/bus/mhi/core/pm.c
+++ b/drivers/bus/mhi/core/pm.c
@@ -377,20 +377,22 @@ static int mhi_pm_mission_mode_transition(struct mhi_controller *mhi_cntrl)
 {
 	struct mhi_event *mhi_event;
 	struct device *dev = &mhi_cntrl->mhi_dev->dev;
+	enum mhi_ee_type ee = MHI_EE_MAX;
 	int i, ret;
 
 	dev_dbg(dev, "Processing Mission Mode transition\n");
 
 	write_lock_irq(&mhi_cntrl->pm_lock);
 	if (MHI_REG_ACCESS_VALID(mhi_cntrl->pm_state))
-		mhi_cntrl->ee = mhi_get_exec_env(mhi_cntrl);
+		ee = mhi_get_exec_env(mhi_cntrl);
 
-	if (!MHI_IN_MISSION_MODE(mhi_cntrl->ee)) {
+	if (!MHI_IN_MISSION_MODE(ee)) {
 		mhi_cntrl->pm_state = MHI_PM_LD_ERR_FATAL_DETECT;
 		write_unlock_irq(&mhi_cntrl->pm_lock);
 		wake_up_all(&mhi_cntrl->state_event);
 		return -EIO;
 	}
+	mhi_cntrl->ee = ee;
 	write_unlock_irq(&mhi_cntrl->pm_lock);
 
 	wake_up_all(&mhi_cntrl->state_event);
-- 
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

RE: [PATCH v1 2/2] bus: mhi: core: Process execution environment changes serially

[Carl Yin(殷张成)]

Hi bbhat:

On November 17, 2020 10:58 AM, bbhatt wrote:
> In current design, whenever the BHI interrupt is fired, the execution
> environment is updated. This can cause race conditions and impede any ongoing
> power up/down processing. For example, if a power down is in progress and the
> host has updated the execution environment to a local "disabled" state, any BHI
> interrupt firing later could replace it with the value from the BHI EE register.
> Another example would be that the device can enter mission mode while device
> creation for SBL is still going on, leading to multiple attempts at opening the same
> channel.
> 
> Ensure that EE changes are handled only from appropriate places and occur one
> after another and handle only PBL or RDDM EE changes as critical events directly
> from the interrupt handler. This also makes sure that we use the correct
> execution environment to notify the controller driver when the device resets to
> one of the PBL execution environments.
> 
> Signed-off-by: Bhaumik Bhatt <bbhatt@codeaurora.org>
> ---
>  drivers/bus/mhi/core/main.c | 14 ++++++++------
>  drivers/bus/mhi/core/pm.c   |  6 ++++--
>  2 files changed, 12 insertions(+), 8 deletions(-)
> 
> diff --git a/drivers/bus/mhi/core/main.c b/drivers/bus/mhi/core/main.c index
> 4eb93d8..cd712f2 100644
> --- a/drivers/bus/mhi/core/main.c
> +++ b/drivers/bus/mhi/core/main.c
> @@ -377,7 +377,7 @@ irqreturn_t mhi_intvec_threaded_handler(int
> irq_number, void *priv)
>  	struct device *dev = &mhi_cntrl->mhi_dev->dev;
>  	enum mhi_state state = MHI_STATE_MAX;
>  	enum mhi_pm_state pm_state = 0;
> -	enum mhi_ee_type ee = 0;
> +	enum mhi_ee_type ee = MHI_EE_MAX;
> 
>  	write_lock_irq(&mhi_cntrl->pm_lock);
>  	if (!MHI_REG_ACCESS_VALID(mhi_cntrl->pm_state)) { @@ -386,8 +386,7
> @@ irqreturn_t mhi_intvec_threaded_handler(int irq_number, void *priv)
>  	}
> 
>  	state = mhi_get_mhi_state(mhi_cntrl);
> -	ee = mhi_cntrl->ee;
> -	mhi_cntrl->ee = mhi_get_exec_env(mhi_cntrl);
> +	ee = mhi_get_exec_env(mhi_cntrl);
>  	dev_dbg(dev, "local ee:%s device ee:%s dev_state:%s\n",
>  		TO_MHI_EXEC_STR(mhi_cntrl->ee), TO_MHI_EXEC_STR(ee),
>  		TO_MHI_STATE_STR(state));
> @@ -405,8 +404,9 @@ irqreturn_t mhi_intvec_threaded_handler(int
> irq_number, void *priv)
>  		if (!mhi_is_active(mhi_cntrl))
>  			goto exit_intvec;
> 
> -		if (mhi_cntrl->ee == MHI_EE_RDDM && mhi_cntrl->ee != ee) {
> +		if (ee == MHI_EE_RDDM && mhi_cntrl->ee != MHI_EE_RDDM) {
>  			mhi_cntrl->status_cb(mhi_cntrl, MHI_CB_EE_RDDM);
> +			mhi_cntrl->ee = ee;
>  			wake_up_all(&mhi_cntrl->state_event);
>  		}
>  		goto exit_intvec;
> @@ -416,10 +416,12 @@ irqreturn_t mhi_intvec_threaded_handler(int
> irq_number, void *priv)
>  		wake_up_all(&mhi_cntrl->state_event);
> 
>  		/* For fatal errors, we let controller decide next step */
> -		if (MHI_IN_PBL(ee))
> +		if (MHI_IN_PBL(ee)) {
>  			mhi_cntrl->status_cb(mhi_cntrl, MHI_CB_FATAL_ERROR);
> -		else
> +			mhi_cntrl->ee = ee;
> +		} else {
>  			mhi_pm_sys_err_handler(mhi_cntrl);
> +		}
>  	}
> 
[carl.yin] I guess this patch can fix another bug (maybe is not bug)
For hw event with brst mode enabled. If M0 event come before AMSS event, as next log:
mhi_pm_m0_transition() will call mhi_ring_cmd_db(), but ring->wp not correctly set
until the later mhi_pm_mission_mode_transition() set ring->wp.

[ 1550.901882] mhi 0000:03:00.0: Requested to power ON
[ 1550.902065] mhi 0000:03:00.0: Power on setup success
[ 1550.902256] mhi 0000:03:00.0: Handling state transition: PBL
[ 1551.751325] pci 0000:01:00.0: PCI bridge to [bus 02]
[ 1556.113314] mhi 0000:03:00.0: Device in READY State
[ 1556.113319] mhi 0000:03:00.0: Initializing MHI registers
[ 1559.391644] mhi 0000:03:00.0: local ee:AMSS device ee:PASS THRU dev_state:READY
[ 1559.410969] mhi 0000:03:00.0: State change event to state: M0
[ 1559.414807] mhi 0000:03:00.0: Received EE event: AMSS
[ 1559.414860] mhi 0000:03:00.0: Handling state transition: MISSION_MODE
[ 1559.414863] mhi 0000:03:00.0: Processing Mission Mode transition

>  exit_intvec:
> diff --git a/drivers/bus/mhi/core/pm.c b/drivers/bus/mhi/core/pm.c index
> aa156b9..c1b18d6 100644
> --- a/drivers/bus/mhi/core/pm.c
> +++ b/drivers/bus/mhi/core/pm.c
> @@ -377,20 +377,22 @@ static int mhi_pm_mission_mode_transition(struct
> mhi_controller *mhi_cntrl)  {
>  	struct mhi_event *mhi_event;
>  	struct device *dev = &mhi_cntrl->mhi_dev->dev;
> +	enum mhi_ee_type ee = MHI_EE_MAX;
>  	int i, ret;
> 
>  	dev_dbg(dev, "Processing Mission Mode transition\n");
> 
>  	write_lock_irq(&mhi_cntrl->pm_lock);
>  	if (MHI_REG_ACCESS_VALID(mhi_cntrl->pm_state))
> -		mhi_cntrl->ee = mhi_get_exec_env(mhi_cntrl);
> +		ee = mhi_get_exec_env(mhi_cntrl);
> 
> -	if (!MHI_IN_MISSION_MODE(mhi_cntrl->ee)) {
> +	if (!MHI_IN_MISSION_MODE(ee)) {
>  		mhi_cntrl->pm_state = MHI_PM_LD_ERR_FATAL_DETECT;
>  		write_unlock_irq(&mhi_cntrl->pm_lock);
>  		wake_up_all(&mhi_cntrl->state_event);
>  		return -EIO;
>  	}
> +	mhi_cntrl->ee = ee;
>  	write_unlock_irq(&mhi_cntrl->pm_lock);
> 
>  	wake_up_all(&mhi_cntrl->state_event);
> --
> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, a
> Linux Foundation Collaborative Project