habanalabs: add security violations dump to debugfs

In order to improve driver security debuggability, we add
security violations dump to debugfs.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

Documentation/ABI/testing/debugfs-driver-habanalabs

@@ -182,3 +182,11 @@ KernelVersion:  5.6
 Contact:        oded.gabbay@gmail.com
 Description:    Sets the stop-on_error option for the device engines. Value of
                 "0" is for disable, otherwise enable.
+
+What:           /sys/kernel/debug/habanalabs/hl<n>/dump_security_violations
+Date:           Jan 2021
+KernelVersion:  5.12
+Contact:        oded.gabbay@gmail.com
+Description:    Dumps all security violations to dmesg. This will also ack
+                all security violations meanings those violations will not be
+                dumped next time user calls this API

drivers/misc/habanalabs/common/debugfs.c

@@ -867,6 +867,17 @@ static ssize_t hl_stop_on_err_write(struct file *f, const char __user *buf,
 	return count;
 }
 
+static ssize_t hl_security_violations_read(struct file *f, char __user *buf,
+					size_t count, loff_t *ppos)
+{
+	struct hl_dbg_device_entry *entry = file_inode(f)->i_private;
+	struct hl_device *hdev = entry->hdev;
+
+	hdev->asic_funcs->ack_protection_bits_errors(hdev);
+
+	return 0;
+}
+
 static const struct file_operations hl_data32b_fops = {
 	.owner = THIS_MODULE,
 	.read = hl_data_read32,
@@ -924,6 +935,11 @@ static const struct file_operations hl_stop_on_err_fops = {
 	.write = hl_stop_on_err_write
 };
 
+static const struct file_operations hl_security_violations_fops = {
+	.owner = THIS_MODULE,
+	.read = hl_security_violations_read
+};
+
 static const struct hl_info_list hl_debugfs_list[] = {
 	{"command_buffers", command_buffers_show, NULL},
 	{"command_submission", command_submission_show, NULL},
@@ -1073,6 +1089,12 @@ void hl_debugfs_add_device(struct hl_device *hdev)
 				dev_entry,
 				&hl_stop_on_err_fops);
 
+	debugfs_create_file("dump_security_violations",
+				0644,
+				dev_entry->root,
+				dev_entry,
+				&hl_security_violations_fops);
+
 	for (i = 0, entry = dev_entry->entry_arr ; i < count ; i++, entry++) {
 
 		ent = debugfs_create_file(hl_debugfs_list[i].name,

drivers/misc/habanalabs/common/habanalabs.h

@@ -850,6 +850,7 @@ enum div_select_defs {
  * @collective_wait_create_jobs: allocate collective wait cs jobs
  * @scramble_vaddr: Routine to scramble the virtual address prior of mapping it
  *                  in the MMU.
+ * @ack_protection_bits_errors: ack and dump all security violations
  */
 struct hl_asic_funcs {
 	int (*early_init)(struct hl_device *hdev);
@@ -960,6 +961,7 @@ struct hl_asic_funcs {
 			struct hl_ctx *ctx, struct hl_cs *cs, u32 wait_queue_id,
 			u32 collective_engine_id);
 	u64 (*scramble_vaddr)(struct hl_device *hdev, u64 virt_addr);
+	void (*ack_protection_bits_errors)(struct hl_device *hdev);
 };
 
 

drivers/misc/habanalabs/gaudi/gaudi.c

@@ -8546,7 +8546,8 @@ static const struct hl_asic_funcs gaudi_funcs = {
 	.get_device_time = gaudi_get_device_time,
 	.collective_wait_init_cs = gaudi_collective_wait_init_cs,
 	.collective_wait_create_jobs = gaudi_collective_wait_create_jobs,
-	.scramble_vaddr = hl_mmu_scramble_vaddr
+	.scramble_vaddr = hl_mmu_scramble_vaddr,
+	.ack_protection_bits_errors = gaudi_ack_protection_bits_errors
 };
 
 /**

drivers/misc/habanalabs/gaudi/gaudiP.h

@@ -335,6 +335,7 @@ struct gaudi_device {
 };
 
 void gaudi_init_security(struct hl_device *hdev);
+void gaudi_ack_protection_bits_errors(struct hl_device *hdev);
 void gaudi_add_device_attr(struct hl_device *hdev,
 			struct attribute_group *dev_attr_grp);
 void gaudi_set_pll_profile(struct hl_device *hdev, enum hl_pll_frequency freq);

drivers/misc/habanalabs/gaudi/gaudi_security.c

@@ -13052,3 +13052,8 @@ void gaudi_init_security(struct hl_device *hdev)
 
 	gaudi_init_protection_bits(hdev);
 }
+
+void gaudi_ack_protection_bits_errors(struct hl_device *hdev)
+{
+
+}

drivers/misc/habanalabs/goya/goya.c

@@ -5457,7 +5457,8 @@ static const struct hl_asic_funcs goya_funcs = {
 	.get_device_time = goya_get_device_time,
 	.collective_wait_init_cs = goya_collective_wait_init_cs,
 	.collective_wait_create_jobs = goya_collective_wait_create_jobs,
-	.scramble_vaddr = hl_mmu_scramble_vaddr
+	.scramble_vaddr = hl_mmu_scramble_vaddr,
+	.ack_protection_bits_errors = goya_ack_protection_bits_errors
 };
 
 /*

drivers/misc/habanalabs/goya/goyaP.h

@@ -173,6 +173,7 @@ void goya_init_mme_qmans(struct hl_device *hdev);
 void goya_init_tpc_qmans(struct hl_device *hdev);
 int goya_init_cpu_queues(struct hl_device *hdev);
 void goya_init_security(struct hl_device *hdev);
+void goya_ack_protection_bits_errors(struct hl_device *hdev);
 int goya_late_init(struct hl_device *hdev);
 void goya_late_fini(struct hl_device *hdev);
 

drivers/misc/habanalabs/goya/goya_security.c

@@ -3120,3 +3120,8 @@ void goya_init_security(struct hl_device *hdev)
 
 	goya_init_protection_bits(hdev);
 }
+
+void goya_ack_protection_bits_errors(struct hl_device *hdev)
+{
+
+}