[PATCH] cxl: Force context lock during EEH flow

[Vaibhav Jain]

During an eeh event when the cxl card is fenced and card sysfs attr
perst_reloads_same_image is set following warning message is seen in the
kernel logs:

 [   60.622727] Adapter context unlocked with 0 active contexts
 [   60.622762] ------------[ cut here ]------------
 [   60.622771] WARNING: CPU: 12 PID: 627 at
 ../drivers/misc/cxl/main.c:325 cxl_adapter_context_unlock+0x60/0x80 [cxl]

Even though this warning is harmless, it clutters the kernel log
during an eeh event. This warning is triggered as the EEH callback
cxl_pci_error_detected doesn't obtain a context-lock before forcibly
detaching all active context and when context-lock is released during
call to cxl_configure_adapter from cxl_pci_slot_reset, a warning in
cxl_adapter_context_unlock is triggered.

To fix this warning and also prevent activation of any context during
eeh the patch introduces a new function cxl_adapter_context_force_lock
that would forcefully acquire the context_lock and warn if any active
contexts exists when this happens. After the EEH flow concludes with
call to cxl_pci_resume the context-lock is released with a call to
cxl_adapter_context_unlock.

Cc: stable at vger.kernel.org
Fixes: 70b565bbdb91("cxl: Prevent adapter reset if an active context exists")
Reported-by: Andrew Donnellan <andrew.donnellan at au1.ibm.com>
Signed-off-by: Vaibhav Jain <vaibhav at linux.vnet.ibm.com>
Reviewed-by: Andrew Donnellan <andrew.donnellan at au1.ibm.com>
---
 drivers/misc/cxl/cxl.h  |  3 +++
 drivers/misc/cxl/main.c | 10 ++++++++++
 drivers/misc/cxl/pci.c  | 18 ++++++++++++++++--
 3 files changed, 29 insertions(+), 2 deletions(-)

diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
index b24d767..34d2ca0 100644
--- a/drivers/misc/cxl/cxl.h
+++ b/drivers/misc/cxl/cxl.h
@@ -970,4 +970,7 @@ int cxl_adapter_context_lock(struct cxl *adapter);
 /* Unlock the contexts-lock if taken. Warn and force unlock otherwise */
 void cxl_adapter_context_unlock(struct cxl *adapter);
 
+/* Force contexts-lock to be taken */
+void cxl_adapter_context_force_lock(struct cxl *adapter);
+
 #endif
diff --git a/drivers/misc/cxl/main.c b/drivers/misc/cxl/main.c
index 62e0dfb..1754a0c 100644
--- a/drivers/misc/cxl/main.c
+++ b/drivers/misc/cxl/main.c
@@ -301,6 +301,16 @@ void cxl_adapter_context_put(struct cxl *adapter)
 	atomic_dec_if_positive(&adapter->contexts_num);
 }
 
+void cxl_adapter_context_force_lock(struct cxl *adapter)
+{
+	int count = atomic_read(&adapter->contexts_num);
+
+	if (count > 0)
+		pr_warn("Forcing context lock with %d active contexts\n",
+			count);
+	atomic_set(&adapter->contexts_num, -1);
+}
+
 int cxl_adapter_context_lock(struct cxl *adapter)
 {
 	int rc;
diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
index 80a87ab..d76fd4d 100644
--- a/drivers/misc/cxl/pci.c
+++ b/drivers/misc/cxl/pci.c
@@ -1487,8 +1487,6 @@ static int cxl_configure_adapter(struct cxl *adapter, struct pci_dev *dev)
 	if ((rc = cxl_native_register_psl_err_irq(adapter)))
 		goto err;
 
-	/* Release the context lock as adapter is configured */
-	cxl_adapter_context_unlock(adapter);
 	return 0;
 
 err:
@@ -1587,6 +1585,9 @@ static struct cxl *cxl_pci_init_adapter(struct pci_dev *dev)
 	if ((rc = cxl_sysfs_adapter_add(adapter)))
 		goto err_put1;
 
+	/* Release the context lock as adapter is configured */
+	cxl_adapter_context_unlock(adapter);
+
 	return adapter;
 
 err_put1:
@@ -1607,6 +1608,9 @@ static void cxl_pci_remove_adapter(struct cxl *adapter)
 {
 	pr_devel("cxl_remove_adapter\n");
 
+	/* Forcibly take the adapter context lock */
+	cxl_adapter_context_force_lock(adapter);
+
 	cxl_sysfs_adapter_remove(adapter);
 	cxl_debugfs_adapter_remove(adapter);
 
@@ -1778,6 +1782,9 @@ static pci_ers_result_t cxl_pci_error_detected(struct pci_dev *pdev,
 	 */
 	schedule();
 
+	/* forcibly take the context lock to prevent new context activation */
+	cxl_adapter_context_force_lock(adapter);
+
 	/* If we're permanently dead, give up. */
 	if (state == pci_channel_io_perm_failure) {
 		/* Tell the AFU drivers; but we don't care what they
@@ -1879,11 +1886,15 @@ static pci_ers_result_t cxl_pci_error_detected(struct pci_dev *pdev,
 		/* Only continue if everyone agrees on NEED_RESET */
 		if (result != PCI_ERS_RESULT_NEED_RESET)
 			return result;
+	}
 
+	for (i = 0; i < adapter->slices; i++) {
+		afu = adapter->afu[i];
 		cxl_context_detach_all(afu);
 		cxl_ops->afu_deactivate_mode(afu, afu->current_mode);
 		pci_deconfigure_afu(afu);
 	}
+
 	cxl_deconfigure_adapter(adapter);
 
 	return result;
@@ -1979,6 +1990,9 @@ static void cxl_pci_resume(struct pci_dev *pdev)
 				afu_dev->driver->err_handler->resume(afu_dev);
 		}
 	}
+
+	/* Unlock context activation for the adapter */
+	cxl_adapter_context_unlock(adapter);
 }
 
 static const struct pci_error_handlers cxl_err_handler = {
-- 
2.9.3