linux-kernel - [PATCH v1 6/6] bus: mhi: core: Process execution environment changes serially

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <1589934631-22752-7-git-send-email-bbhatt@codeaurora.org>
Date:   Tue, 19 May 2020 17:30:31 -0700
From:   Bhaumik Bhatt <bbhatt@...eaurora.org>
To:     manivannan.sadhasivam@...aro.org
Cc:     linux-arm-msm@...r.kernel.org, hemantk@...eaurora.org,
        jhugo@...eaurora.org, linux-kernel@...r.kernel.org,
        Bhaumik Bhatt <bbhatt@...eaurora.org>
Subject: [PATCH v1 6/6] bus: mhi: core: Process execution environment changes serially

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 or power 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 could replace it with the
execution environment from the BHI EE register. Another example would
be that the device can enter mission mode while the device creation for
SBL is still ongoing leading to multiple attempts at opening the same
channel.

Ensure that EE changes are handled only from appropriate places and
occur one after another.

For RDDM, handle it as a critical event directly from the interrupt
handler and remove RDDM EE change event from the control event ring to
prevent the driver from issuing multiple callbacks. SBL handling
requires no change. For AMSS/mission mode, hold off the update until
the client is notified with a status callback.

To sum it up, ensure client readiness before processing mission mode and
move to an error state as soon as possible to avoid a bad state.

Signed-off-by: Bhaumik Bhatt <bbhatt@...eaurora.org>
---
 drivers/bus/mhi/core/main.c | 12 ++++--------
 drivers/bus/mhi/core/pm.c   | 21 +++++++++++++++++----
 2 files changed, 21 insertions(+), 12 deletions(-)

diff --git a/drivers/bus/mhi/core/main.c b/drivers/bus/mhi/core/main.c
index da32c23..d25f321 100644
--- a/drivers/bus/mhi/core/main.c
+++ b/drivers/bus/mhi/core/main.c
@@ -385,8 +385,7 @@ 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));
@@ -399,7 +398,9 @@ irqreturn_t mhi_intvec_threaded_handler(int irq_number, void *priv)
 			goto exit_intvec;
 		}
 
-		if (mhi_cntrl->ee == MHI_EE_RDDM && mhi_cntrl->ee != ee) {
+		if (ee == MHI_EE_RDDM && mhi_cntrl->ee != ee) {
+			mhi_cntrl->ee = MHI_EE_RDDM;
+
 			/* prevent clients from queueing any more packets */
 			pm_state = mhi_tryset_pm_state(mhi_cntrl,
 						       MHI_PM_SYS_ERR_DETECT);
@@ -794,11 +795,6 @@ int mhi_process_ctrl_ev_ring(struct mhi_controller *mhi_cntrl,
 				st = DEV_ST_TRANSITION_MISSION_MODE;
 				break;
 			case MHI_EE_RDDM:
-				mhi_cntrl->status_cb(mhi_cntrl, MHI_CB_EE_RDDM);
-				write_lock_irq(&mhi_cntrl->pm_lock);
-				mhi_cntrl->ee = event;
-				write_unlock_irq(&mhi_cntrl->pm_lock);
-				wake_up_all(&mhi_cntrl->state_event);
 				break;
 			default:
 				dev_err(dev,
diff --git a/drivers/bus/mhi/core/pm.c b/drivers/bus/mhi/core/pm.c
index 5041df9..4407338 100644
--- a/drivers/bus/mhi/core/pm.c
+++ b/drivers/bus/mhi/core/pm.c
@@ -377,22 +377,35 @@ 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);
 	write_unlock_irq(&mhi_cntrl->pm_lock);
 
-	if (!MHI_IN_MISSION_MODE(mhi_cntrl->ee))
+	if (!MHI_IN_MISSION_MODE(ee)) {
+		dev_err(dev, "Invalid EE for Mission Mode: %s\n",
+			TO_MHI_EXEC_STR(ee));
 		return -EIO;
+	}
 
-	wake_up_all(&mhi_cntrl->state_event);
-
+	/*
+	 * let controller prepare for mission mode before making the execution
+	 * environment change so as to defer core driver activity
+	 */
 	mhi_cntrl->status_cb(mhi_cntrl, MHI_CB_EE_MISSION_MODE);
 
+	/* ready the core driver for mission mode */
+	write_lock_irq(&mhi_cntrl->pm_lock);
+	mhi_cntrl->ee = ee;
+	write_unlock_irq(&mhi_cntrl->pm_lock);
+
+	wake_up_all(&mhi_cntrl->state_event);
+
 	/* Force MHI to be in M0 state before continuing */
 	ret = __mhi_device_get_sync(mhi_cntrl);
 	if (ret)
-- 
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project