[PATCH 1/3] habanalabs: skip reading f/w errors on bad status

[PATCH 1/3] habanalabs: skip reading f/w errors on bad status

[Oded Gabbay]

If we read all FF from the boot status register, then something is
totally wrong and there is no point of reading specific errors.

Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/firmware_if.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index 7cf82da67dab..fff29f057b6d 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -850,8 +850,13 @@ int hl_fw_read_preboot_status(struct hl_device *hdev, u32 cpu_boot_status_reg,
 	if (rc) {
 		dev_err(hdev->dev, "Failed to read preboot version\n");
 		detect_cpu_boot_status(hdev, status);
-		fw_read_errors(hdev, boot_err0_reg,
-				cpu_security_boot_status_reg);
+
+		/* If we read all FF, then something is totally wrong, no point
+		 * of reading specific errors
+		 */
+		if (status != -1)
+			fw_read_errors(hdev, boot_err0_reg,
+					cpu_security_boot_status_reg);
 		return -EIO;
 	}
 
-- 
2.25.1

[PATCH 2/3] habanalabs: change error level of security not ready

[Oded Gabbay]

This error indicates a problem in the security initialization inside
the f/w so we need to stop the device loading because it won't be
usable.

Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/firmware_if.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index fff29f057b6d..377a7ca886fe 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -362,12 +362,9 @@ static int fw_read_errors(struct hl_device *hdev, u32 boot_err0_reg,
 	}
 
 	if (err_val & CPU_BOOT_ERR0_SECURITY_NOT_RDY) {
-		dev_warn(hdev->dev,
+		dev_err(hdev->dev,
 			"Device boot warning - security not ready\n");
-		/* This is a warning so we don't want it to disable the
-		 * device
-		 */
-		err_val &= ~CPU_BOOT_ERR0_SECURITY_NOT_RDY;
+		err_exists = true;
 	}
 
 	if (err_val & CPU_BOOT_ERR0_SECURITY_FAIL) {
-- 
2.25.1

[PATCH 3/3] habanalabs: ignore f/w status error

[Oded Gabbay]

In case firmware has a bug and erroneously reports a status error
(e.g. device unusable) during boot, allow the user to tell the driver
to continue the boot regardless of the error status.

This will be done via kernel parameter which exposes a mask. The
user that loads the driver can decide exactly which status error to
ignore and which to take into account. The bitmask is according to
defines in hl_boot_if.h

Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/firmware_if.c    | 3 ++-
 drivers/misc/habanalabs/common/habanalabs.h     | 7 +++++++
 drivers/misc/habanalabs/common/habanalabs_drv.c | 7 +++++++
 3 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index 377a7ca886fe..0713b2c12d54 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -400,7 +400,8 @@ static int fw_read_errors(struct hl_device *hdev, u32 boot_err0_reg,
 		err_exists = true;
 	}
 
-	if (err_exists)
+	if (err_exists && ((err_val & ~CPU_BOOT_ERR0_ENABLED) &
+				lower_32_bits(hdev->boot_error_status_mask)))
 		return -EIO;
 
 	return 0;
diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index 91291a8e201e..6579f8767abd 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -1962,6 +1962,12 @@ struct hl_mmu_funcs {
  * @clock_gating_mask: is clock gating enabled. bitmask that represents the
  *                     different engines. See debugfs-driver-habanalabs for
  *                     details.
+ * @boot_error_status_mask: contains a mask of the device boot error status.
+ *                          Each bit represents a different error, according to
+ *                          the defines in hl_boot_if.h. If the bit is cleared,
+ *                          the error will be ignored by the driver during
+ *                          device initialization. Mainly used to debug and
+ *                          workaround firmware bugs
  * @in_reset: is device in reset flow.
  * @curr_pll_profile: current PLL profile.
  * @card_type: Various ASICs have several card types. This indicates the card
@@ -2077,6 +2083,7 @@ struct hl_device {
 	u64				timeout_jiffies;
 	u64				max_power;
 	u64				clock_gating_mask;
+	u64				boot_error_status_mask;
 	atomic_t			in_reset;
 	enum hl_pll_frequency		curr_pll_profile;
 	enum cpucp_card_types		card_type;
diff --git a/drivers/misc/habanalabs/common/habanalabs_drv.c b/drivers/misc/habanalabs/common/habanalabs_drv.c
index 7135f1e03864..64d1530db985 100644
--- a/drivers/misc/habanalabs/common/habanalabs_drv.c
+++ b/drivers/misc/habanalabs/common/habanalabs_drv.c
@@ -30,6 +30,7 @@ static DEFINE_MUTEX(hl_devs_idr_lock);
 static int timeout_locked = 30;
 static int reset_on_lockup = 1;
 static int memory_scrub = 1;
+static ulong boot_error_status_mask = ULONG_MAX;
 
 module_param(timeout_locked, int, 0444);
 MODULE_PARM_DESC(timeout_locked,
@@ -43,6 +44,10 @@ module_param(memory_scrub, int, 0444);
 MODULE_PARM_DESC(memory_scrub,
 	"Scrub device memory in various states (0 = no, 1 = yes, default yes)");
 
+module_param(boot_error_status_mask, ulong, 0444);
+MODULE_PARM_DESC(boot_error_status_mask,
+	"Mask of the error status during device CPU boot (If bitX is cleared then error X is masked. Default all 1's)");
+
 #define PCI_VENDOR_ID_HABANALABS	0x1da3
 
 #define PCI_IDS_GOYA			0x0001
@@ -319,6 +324,8 @@ int create_hdev(struct hl_device **dev, struct pci_dev *pdev,
 	hdev->major = hl_major;
 	hdev->reset_on_lockup = reset_on_lockup;
 	hdev->memory_scrub = memory_scrub;
+	hdev->boot_error_status_mask = boot_error_status_mask;
+
 	hdev->pldm = 0;
 
 	set_driver_behavior_per_device(hdev);
-- 
2.25.1