crypto: qat - enable detection of accelerators hang

Enable the detection of hangs by setting watchdog timers (WDTs) on generations that supports that feature. The default timeout value comes from HW specs. WTDs are reset each time an accelerator wins arbitration and is able to send/read a command to/from an accelerator. The value has added significant margin to make sure there are no spurious timeouts. The scope of watchdog is per QAT device. If a timeout is detected, the firmware resets the accelerator and returns a response descriptor with an appropriate error code. Signed-off-by: Wojciech Ziemba <[email protected]> Reviewed-by: Giovanni Cabiddu <[email protected]> Signed-off-by: Herbert Xu <[email protected]>
zandrey · Apr 16, 2021 · 1c4d9d5 · 1c4d9d5
1 parent 7e958d3
commit 1c4d9d5
Show file tree

Hide file tree

Showing 9 changed files with 99 additions and 1 deletion.
diff --git a/drivers/crypto/qat/qat_4xxx/adf_4xxx_hw_data.c b/drivers/crypto/qat/qat_4xxx/adf_4xxx_hw_data.c
@@ -224,6 +224,7 @@ void adf_init_hw_data_4xxx(struct adf_hw_device_data *hw_data)
 	hw_data->uof_get_name = uof_get_name;
 	hw_data->uof_get_ae_mask = uof_get_ae_mask;
 	hw_data->set_msix_rttable = set_msix_default_rttable;
+	hw_data->set_ssm_wdtimer = adf_gen4_set_ssm_wdtimer;
 
 	adf_gen4_init_hw_csr_ops(&hw_data->csr_ops);
 }

diff --git a/drivers/crypto/qat/qat_c3xxx/adf_c3xxx_hw_data.c b/drivers/crypto/qat/qat_c3xxx/adf_c3xxx_hw_data.c
@@ -212,6 +212,7 @@ void adf_init_hw_data_c3xxx(struct adf_hw_device_data *hw_data)
 	hw_data->enable_vf2pf_comms = adf_pf_enable_vf2pf_comms;
 	hw_data->reset_device = adf_reset_flr;
 	hw_data->min_iov_compat_ver = ADF_PFVF_COMPATIBILITY_VERSION;
+	hw_data->set_ssm_wdtimer = adf_gen2_set_ssm_wdtimer;
 	adf_gen2_init_hw_csr_ops(&hw_data->csr_ops);
 }
 

diff --git a/drivers/crypto/qat/qat_c62x/adf_c62x_hw_data.c b/drivers/crypto/qat/qat_c62x/adf_c62x_hw_data.c
@@ -214,6 +214,7 @@ void adf_init_hw_data_c62x(struct adf_hw_device_data *hw_data)
 	hw_data->enable_vf2pf_comms = adf_pf_enable_vf2pf_comms;
 	hw_data->reset_device = adf_reset_flr;
 	hw_data->min_iov_compat_ver = ADF_PFVF_COMPATIBILITY_VERSION;
+	hw_data->set_ssm_wdtimer = adf_gen2_set_ssm_wdtimer;
 	adf_gen2_init_hw_csr_ops(&hw_data->csr_ops);
 }
 

diff --git a/drivers/crypto/qat/qat_common/adf_accel_devices.h b/drivers/crypto/qat/qat_common/adf_accel_devices.h
@@ -173,6 +173,7 @@ struct adf_hw_device_data {
 	void (*configure_iov_threads)(struct adf_accel_dev *accel_dev,
 				      bool enable);
 	void (*enable_ints)(struct adf_accel_dev *accel_dev);
+	void (*set_ssm_wdtimer)(struct adf_accel_dev *accel_dev);
 	int (*enable_vf2pf_comms)(struct adf_accel_dev *accel_dev);
 	void (*reset_device)(struct adf_accel_dev *accel_dev);
 	void (*set_msix_rttable)(struct adf_accel_dev *accel_dev);

diff --git a/drivers/crypto/qat/qat_common/adf_gen2_hw_data.c b/drivers/crypto/qat/qat_common/adf_gen2_hw_data.c
@@ -179,3 +179,28 @@ u32 adf_gen2_get_accel_cap(struct adf_accel_dev *accel_dev)
 	return capabilities;
 }
 EXPORT_SYMBOL_GPL(adf_gen2_get_accel_cap);
+
+void adf_gen2_set_ssm_wdtimer(struct adf_accel_dev *accel_dev)
+{
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	u32 timer_val_pke = ADF_SSM_WDT_PKE_DEFAULT_VALUE;
+	u32 timer_val = ADF_SSM_WDT_DEFAULT_VALUE;
+	unsigned long accel_mask = hw_data->accel_mask;
+	void __iomem *pmisc_addr;
+	struct adf_bar *pmisc;
+	int pmisc_id;
+	u32 i = 0;
+
+	pmisc_id = hw_data->get_misc_bar_id(hw_data);
+	pmisc = &GET_BARS(accel_dev)[pmisc_id];
+	pmisc_addr = pmisc->virt_addr;
+
+	/* Configures WDT timers */
+	for_each_set_bit(i, &accel_mask, hw_data->num_accel) {
+		/* Enable WDT for sym and dc */
+		ADF_CSR_WR(pmisc_addr, ADF_SSMWDT(i), timer_val);
+		/* Enable WDT for pke */
+		ADF_CSR_WR(pmisc_addr, ADF_SSMWDTPKE(i), timer_val_pke);
+	}
+}
+EXPORT_SYMBOL_GPL(adf_gen2_set_ssm_wdtimer);
diff --git a/drivers/crypto/qat/qat_common/adf_gen2_hw_data.h b/drivers/crypto/qat/qat_common/adf_gen2_hw_data.h
@@ -113,11 +113,24 @@ do { \
 /* Power gating */
 #define ADF_POWERGATE_PKE		BIT(24)
 
+/* WDT timers
+ *
+ * Timeout is in cycles. Clock speed may vary across products but this
+ * value should be a few milli-seconds.
+ */
+#define ADF_SSM_WDT_DEFAULT_VALUE	0x200000
+#define ADF_SSM_WDT_PKE_DEFAULT_VALUE	0x2000000
+#define ADF_SSMWDT_OFFSET		0x54
+#define ADF_SSMWDTPKE_OFFSET		0x58
+#define ADF_SSMWDT(i)		(ADF_SSMWDT_OFFSET + ((i) * 0x4000))
+#define ADF_SSMWDTPKE(i)	(ADF_SSMWDTPKE_OFFSET + ((i) * 0x4000))
+
 void adf_gen2_cfg_iov_thds(struct adf_accel_dev *accel_dev, bool enable,
 			   int num_a_regs, int num_b_regs);
 void adf_gen2_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops);
 void adf_gen2_get_admin_info(struct admin_info *admin_csrs_info);
 void adf_gen2_get_arb_info(struct arb_info *arb_info);
 u32 adf_gen2_get_accel_cap(struct adf_accel_dev *accel_dev);
+void adf_gen2_set_ssm_wdtimer(struct adf_accel_dev *accel_dev);
 
 #endif
diff --git a/drivers/crypto/qat/qat_common/adf_gen4_hw_data.c b/drivers/crypto/qat/qat_common/adf_gen4_hw_data.c
@@ -99,3 +99,43 @@ void adf_gen4_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops)
 	csr_ops->write_csr_ring_srv_arb_en = write_csr_ring_srv_arb_en;
 }
 EXPORT_SYMBOL_GPL(adf_gen4_init_hw_csr_ops);
+
+static inline void adf_gen4_unpack_ssm_wdtimer(u64 value, u32 *upper,
+					       u32 *lower)
+{
+	*lower = lower_32_bits(value);
+	*upper = upper_32_bits(value);
+}
+
+void adf_gen4_set_ssm_wdtimer(struct adf_accel_dev *accel_dev)
+{
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	u64 timer_val_pke = ADF_SSM_WDT_PKE_DEFAULT_VALUE;
+	u64 timer_val = ADF_SSM_WDT_DEFAULT_VALUE;
+	u32 ssm_wdt_pke_high = 0;
+	u32 ssm_wdt_pke_low = 0;
+	u32 ssm_wdt_high = 0;
+	u32 ssm_wdt_low = 0;
+	void __iomem *pmisc_addr;
+	struct adf_bar *pmisc;
+	int pmisc_id;
+
+	pmisc_id = hw_data->get_misc_bar_id(hw_data);
+	pmisc = &GET_BARS(accel_dev)[pmisc_id];
+	pmisc_addr = pmisc->virt_addr;
+
+	/* Convert 64bit WDT timer value into 32bit values for
+	 * mmio write to 32bit CSRs.
+	 */
+	adf_gen4_unpack_ssm_wdtimer(timer_val, &ssm_wdt_high, &ssm_wdt_low);
+	adf_gen4_unpack_ssm_wdtimer(timer_val_pke, &ssm_wdt_pke_high,
+				    &ssm_wdt_pke_low);
+
+	/* Enable WDT for sym and dc */
+	ADF_CSR_WR(pmisc_addr, ADF_SSMWDTL_OFFSET, ssm_wdt_low);
+	ADF_CSR_WR(pmisc_addr, ADF_SSMWDTH_OFFSET, ssm_wdt_high);
+	/* Enable WDT for pke */
+	ADF_CSR_WR(pmisc_addr, ADF_SSMWDTPKEL_OFFSET, ssm_wdt_pke_low);
+	ADF_CSR_WR(pmisc_addr, ADF_SSMWDTPKEH_OFFSET, ssm_wdt_pke_high);
+}
+EXPORT_SYMBOL_GPL(adf_gen4_set_ssm_wdtimer);
diff --git a/drivers/crypto/qat/qat_common/adf_gen4_hw_data.h b/drivers/crypto/qat/qat_common/adf_gen4_hw_data.h
@@ -94,6 +94,18 @@ do { \
 		   ADF_RING_BUNDLE_SIZE * (bank) + \
 		   ADF_RING_CSR_RING_SRV_ARB_EN, (value))
 
-void adf_gen4_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops);
+/* WDT timers
+ *
+ * Timeout is in cycles. Clock speed may vary across products but this
+ * value should be a few milli-seconds.
+ */
+#define ADF_SSM_WDT_DEFAULT_VALUE	0x200000
+#define ADF_SSM_WDT_PKE_DEFAULT_VALUE	0x8000000
+#define ADF_SSMWDTL_OFFSET		0x54
+#define ADF_SSMWDTH_OFFSET		0x5C
+#define ADF_SSMWDTPKEL_OFFSET		0x58
+#define ADF_SSMWDTPKEH_OFFSET		0x60
 
+void adf_gen4_set_ssm_wdtimer(struct adf_accel_dev *accel_dev);
+void adf_gen4_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops);
 #endif
diff --git a/drivers/crypto/qat/qat_common/adf_init.c b/drivers/crypto/qat/qat_common/adf_init.c
@@ -162,6 +162,10 @@ int adf_dev_start(struct adf_accel_dev *accel_dev)
 		return -EFAULT;
 	}
 
+	/* Set ssm watch dog timer */
+	if (hw_data->set_ssm_wdtimer)
+		hw_data->set_ssm_wdtimer(accel_dev);
+
 	list_for_each(list_itr, &service_table) {
 		service = list_entry(list_itr, struct service_hndl, list);
 		if (service->event_hld(accel_dev, ADF_EVENT_START)) {