linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
* [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation
@ 2015-02-16  3:45 Gavin Shan
  2015-02-16  3:45 ` [PATCH 01/10] powerpc/powernv: Shorten EEH function names Gavin Shan
                   ` (9 more replies)
  0 siblings, 10 replies; 11+ messages in thread
From: Gavin Shan @ 2015-02-16  3:45 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Gavin Shan

The EEH was implemented for PowerNV platform with 2 layers, which are
supported by "struct eeh_ops" and "struct pnv_eeh_ops" separately. The
latter, called chip layer, was introduced for EEH guest support initially.
However, the EEH guest support was implemented depending on "struct eeh_ops"
completely. So it's uncessary to maintain 2 layers for EEH implementation
on PowerNV platform, as Ben pointed out.

The series of patches is pretty much mechanically to move all functions
in eeh-ioda.c to eeh-powernv.c, which enables us to remove the chip
layer, equivalently to remove "struct pnv_eeh_ops" and eeh-ioda.c, to
have simplifed EEH implementation on PowerNV platform.

Gavin Shan (10):
  powerpc/powernv: Shorten EEH function names
  powerpc/powernv: Drop PHB operation err_inject()
  powerpc/powernv: Drop PHB operation post_init()
  powerpc/powernv: Drop PHB operation get_log()
  powerpc/powernv: Drop PHB operation configure_bridge()
  powerpc/powernv: Drop PHB operation set_option()
  powerpc/powernv: Drop PHB operation get_state()
  powerpc/powernv: Drop PHB operation next_error()
  powerpc/powernv: Drop PHB operation reset()
  powerpc/powernv: Remove unused file

 arch/powerpc/platforms/powernv/Makefile      |    2 +-
 arch/powerpc/platforms/powernv/eeh-ioda.c    | 1149 --------------------------
 arch/powerpc/platforms/powernv/eeh-powernv.c | 1146 ++++++++++++++++++++++---
 arch/powerpc/platforms/powernv/pci-ioda.c    |    7 +-
 arch/powerpc/platforms/powernv/pci.h         |   25 +-
 5 files changed, 1041 insertions(+), 1288 deletions(-)
 delete mode 100644 arch/powerpc/platforms/powernv/eeh-ioda.c

-- 
1.8.3.2

^ permalink raw reply	[flat|nested] 11+ messages in thread

* [PATCH 01/10] powerpc/powernv: Shorten EEH function names
  2015-02-16  3:45 [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation Gavin Shan
@ 2015-02-16  3:45 ` Gavin Shan
  2015-02-16  3:45 ` [PATCH 02/10] powerpc/powernv: Drop PHB operation err_inject() Gavin Shan
                   ` (8 subsequent siblings)
  9 siblings, 0 replies; 11+ messages in thread
From: Gavin Shan @ 2015-02-16  3:45 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Gavin Shan

The patch shortens names of EEH functions in powernv-eeh.c and no
logic change introduced by this patch.

Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
---
 arch/powerpc/platforms/powernv/eeh-powernv.c | 104 +++++++++++++--------------
 1 file changed, 52 insertions(+), 52 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index e261869..f562dd1 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -39,11 +39,11 @@
 #include "pci.h"
 
 /**
- * powernv_eeh_init - EEH platform dependent initialization
+ * pnv_eeh_init - EEH platform dependent initialization
  *
  * EEH platform dependent initialization on powernv
  */
-static int powernv_eeh_init(void)
+static int pnv_eeh_init(void)
 {
 	struct pci_controller *hose;
 	struct pnv_phb *phb;
@@ -86,14 +86,14 @@ static int powernv_eeh_init(void)
 }
 
 /**
- * powernv_eeh_post_init - EEH platform dependent post initialization
+ * pnv_eeh_post_init - EEH platform dependent post initialization
  *
  * EEH platform dependent post initialization on powernv. When
  * the function is called, the EEH PEs and devices should have
  * been built. If the I/O cache staff has been built, EEH is
  * ready to supply service.
  */
-static int powernv_eeh_post_init(void)
+static int pnv_eeh_post_init(void)
 {
 	struct pci_controller *hose;
 	struct pnv_phb *phb;
@@ -113,7 +113,7 @@ static int powernv_eeh_post_init(void)
 }
 
 /**
- * powernv_eeh_dev_probe - Do probe on PCI device
+ * pnv_eeh_dev_probe - Do probe on PCI device
  * @dev: PCI device
  * @flag: unused
  *
@@ -129,7 +129,7 @@ static int powernv_eeh_post_init(void)
  * was possiblly triggered by EEH core, the binding between EEH device
  * and the PCI device isn't built yet.
  */
-static int powernv_eeh_dev_probe(struct pci_dev *dev, void *flag)
+static int pnv_eeh_dev_probe(struct pci_dev *dev, void *flag)
 {
 	struct pci_controller *hose = pci_bus_to_host(dev->bus);
 	struct pnv_phb *phb = hose->private_data;
@@ -221,7 +221,7 @@ static int powernv_eeh_dev_probe(struct pci_dev *dev, void *flag)
 }
 
 /**
- * powernv_eeh_set_option - Initialize EEH or MMIO/DMA reenable
+ * pnv_eeh_set_option - Initialize EEH or MMIO/DMA reenable
  * @pe: EEH PE
  * @option: operation to be issued
  *
@@ -229,7 +229,7 @@ static int powernv_eeh_dev_probe(struct pci_dev *dev, void *flag)
  * Currently, following options are support according to PAPR:
  * Enable EEH, Disable EEH, Enable MMIO and Enable DMA
  */
-static int powernv_eeh_set_option(struct eeh_pe *pe, int option)
+static int pnv_eeh_set_option(struct eeh_pe *pe, int option)
 {
 	struct pci_controller *hose = pe->phb;
 	struct pnv_phb *phb = hose->private_data;
@@ -246,19 +246,19 @@ static int powernv_eeh_set_option(struct eeh_pe *pe, int option)
 }
 
 /**
- * powernv_eeh_get_pe_addr - Retrieve PE address
+ * pnv_eeh_get_pe_addr - Retrieve PE address
  * @pe: EEH PE
  *
  * Retrieve the PE address according to the given tranditional
  * PCI BDF (Bus/Device/Function) address.
  */
-static int powernv_eeh_get_pe_addr(struct eeh_pe *pe)
+static int pnv_eeh_get_pe_addr(struct eeh_pe *pe)
 {
 	return pe->addr;
 }
 
 /**
- * powernv_eeh_get_state - Retrieve PE state
+ * pnv_eeh_get_state - Retrieve PE state
  * @pe: EEH PE
  * @delay: delay while PE state is temporarily unavailable
  *
@@ -267,7 +267,7 @@ static int powernv_eeh_get_pe_addr(struct eeh_pe *pe)
  * we prefer passing down to hardware implementation to handle
  * it.
  */
-static int powernv_eeh_get_state(struct eeh_pe *pe, int *delay)
+static int pnv_eeh_get_state(struct eeh_pe *pe, int *delay)
 {
 	struct pci_controller *hose = pe->phb;
 	struct pnv_phb *phb = hose->private_data;
@@ -292,13 +292,13 @@ static int powernv_eeh_get_state(struct eeh_pe *pe, int *delay)
 }
 
 /**
- * powernv_eeh_reset - Reset the specified PE
+ * pnv_eeh_reset - Reset the specified PE
  * @pe: EEH PE
  * @option: reset option
  *
  * Reset the specified PE
  */
-static int powernv_eeh_reset(struct eeh_pe *pe, int option)
+static int pnv_eeh_reset(struct eeh_pe *pe, int option)
 {
 	struct pci_controller *hose = pe->phb;
 	struct pnv_phb *phb = hose->private_data;
@@ -311,20 +311,20 @@ static int powernv_eeh_reset(struct eeh_pe *pe, int option)
 }
 
 /**
- * powernv_eeh_wait_state - Wait for PE state
+ * pnv_eeh_wait_state - Wait for PE state
  * @pe: EEH PE
  * @max_wait: maximal period in microsecond
  *
  * Wait for the state of associated PE. It might take some time
  * to retrieve the PE's state.
  */
-static int powernv_eeh_wait_state(struct eeh_pe *pe, int max_wait)
+static int pnv_eeh_wait_state(struct eeh_pe *pe, int max_wait)
 {
 	int ret;
 	int mwait;
 
 	while (1) {
-		ret = powernv_eeh_get_state(pe, &mwait);
+		ret = pnv_eeh_get_state(pe, &mwait);
 
 		/*
 		 * If the PE's state is temporarily unavailable,
@@ -348,7 +348,7 @@ static int powernv_eeh_wait_state(struct eeh_pe *pe, int max_wait)
 }
 
 /**
- * powernv_eeh_get_log - Retrieve error log
+ * pnv_eeh_get_log - Retrieve error log
  * @pe: EEH PE
  * @severity: temporary or permanent error log
  * @drv_log: driver log to be combined with retrieved error log
@@ -356,8 +356,8 @@ static int powernv_eeh_wait_state(struct eeh_pe *pe, int max_wait)
  *
  * Retrieve the temporary or permanent error from the PE.
  */
-static int powernv_eeh_get_log(struct eeh_pe *pe, int severity,
-			       char *drv_log, unsigned long len)
+static int pnv_eeh_get_log(struct eeh_pe *pe, int severity,
+			   char *drv_log, unsigned long len)
 {
 	struct pci_controller *hose = pe->phb;
 	struct pnv_phb *phb = hose->private_data;
@@ -370,14 +370,14 @@ static int powernv_eeh_get_log(struct eeh_pe *pe, int severity,
 }
 
 /**
- * powernv_eeh_configure_bridge - Configure PCI bridges in the indicated PE
+ * pnv_eeh_configure_bridge - Configure PCI bridges in the indicated PE
  * @pe: EEH PE
  *
  * The function will be called to reconfigure the bridges included
  * in the specified PE so that the mulfunctional PE would be recovered
  * again.
  */
-static int powernv_eeh_configure_bridge(struct eeh_pe *pe)
+static int pnv_eeh_configure_bridge(struct eeh_pe *pe)
 {
 	struct pci_controller *hose = pe->phb;
 	struct pnv_phb *phb = hose->private_data;
@@ -390,7 +390,7 @@ static int powernv_eeh_configure_bridge(struct eeh_pe *pe)
 }
 
 /**
- * powernv_pe_err_inject - Inject specified error to the indicated PE
+ * pnv_pe_err_inject - Inject specified error to the indicated PE
  * @pe: the indicated PE
  * @type: error type
  * @func: specific error type
@@ -401,8 +401,8 @@ static int powernv_eeh_configure_bridge(struct eeh_pe *pe)
  * determined by @type and @func, to the indicated PE for
  * testing purpose.
  */
-static int powernv_eeh_err_inject(struct eeh_pe *pe, int type, int func,
-				  unsigned long addr, unsigned long mask)
+static int pnv_eeh_err_inject(struct eeh_pe *pe, int type, int func,
+			      unsigned long addr, unsigned long mask)
 {
 	struct pci_controller *hose = pe->phb;
 	struct pnv_phb *phb = hose->private_data;
@@ -414,7 +414,7 @@ static int powernv_eeh_err_inject(struct eeh_pe *pe, int type, int func,
 	return ret;
 }
 
-static inline bool powernv_eeh_cfg_blocked(struct device_node *dn)
+static inline bool pnv_eeh_cfg_blocked(struct device_node *dn)
 {
 	struct eeh_dev *edev = of_node_to_eeh_dev(dn);
 
@@ -427,10 +427,10 @@ static inline bool powernv_eeh_cfg_blocked(struct device_node *dn)
 	return false;
 }
 
-static int powernv_eeh_read_config(struct device_node *dn,
-				   int where, int size, u32 *val)
+static int pnv_eeh_read_config(struct device_node *dn,
+			       int where, int size, u32 *val)
 {
-	if (powernv_eeh_cfg_blocked(dn)) {
+	if (pnv_eeh_cfg_blocked(dn)) {
 		*val = 0xFFFFFFFF;
 		return PCIBIOS_SET_FAILED;
 	}
@@ -438,22 +438,22 @@ static int powernv_eeh_read_config(struct device_node *dn,
 	return pnv_pci_cfg_read(dn, where, size, val);
 }
 
-static int powernv_eeh_write_config(struct device_node *dn,
-				    int where, int size, u32 val)
+static int pnv_eeh_write_config(struct device_node *dn,
+				int where, int size, u32 val)
 {
-	if (powernv_eeh_cfg_blocked(dn))
+	if (pnv_eeh_cfg_blocked(dn))
 		return PCIBIOS_SET_FAILED;
 
 	return pnv_pci_cfg_write(dn, where, size, val);
 }
 
 /**
- * powernv_eeh_next_error - Retrieve next EEH error to handle
+ * pnv_eeh_next_error - Retrieve next EEH error to handle
  * @pe: Affected PE
  *
  * Using OPAL API, to retrieve next EEH error for EEH core to handle
  */
-static int powernv_eeh_next_error(struct eeh_pe **pe)
+static int pnv_eeh_next_error(struct eeh_pe **pe)
 {
 	struct pci_controller *hose;
 	struct pnv_phb *phb = NULL;
@@ -469,7 +469,7 @@ static int powernv_eeh_next_error(struct eeh_pe **pe)
 	return -EEXIST;
 }
 
-static int powernv_eeh_restore_config(struct device_node *dn)
+static int pnv_eeh_restore_config(struct device_node *dn)
 {
 	struct eeh_dev *edev = of_node_to_eeh_dev(dn);
 	struct pnv_phb *phb;
@@ -490,24 +490,24 @@ static int powernv_eeh_restore_config(struct device_node *dn)
 	return 0;
 }
 
-static struct eeh_ops powernv_eeh_ops = {
+static struct eeh_ops pnv_eeh_ops = {
 	.name                   = "powernv",
-	.init                   = powernv_eeh_init,
-	.post_init              = powernv_eeh_post_init,
+	.init                   = pnv_eeh_init,
+	.post_init              = pnv_eeh_post_init,
 	.of_probe               = NULL,
-	.dev_probe              = powernv_eeh_dev_probe,
-	.set_option             = powernv_eeh_set_option,
-	.get_pe_addr            = powernv_eeh_get_pe_addr,
-	.get_state              = powernv_eeh_get_state,
-	.reset                  = powernv_eeh_reset,
-	.wait_state             = powernv_eeh_wait_state,
-	.get_log                = powernv_eeh_get_log,
-	.configure_bridge       = powernv_eeh_configure_bridge,
-	.err_inject		= powernv_eeh_err_inject,
-	.read_config            = powernv_eeh_read_config,
-	.write_config           = powernv_eeh_write_config,
-	.next_error		= powernv_eeh_next_error,
-	.restore_config		= powernv_eeh_restore_config
+	.dev_probe              = pnv_eeh_dev_probe,
+	.set_option             = pnv_eeh_set_option,
+	.get_pe_addr            = pnv_eeh_get_pe_addr,
+	.get_state              = pnv_eeh_get_state,
+	.reset                  = pnv_eeh_reset,
+	.wait_state             = pnv_eeh_wait_state,
+	.get_log                = pnv_eeh_get_log,
+	.configure_bridge       = pnv_eeh_configure_bridge,
+	.err_inject		= pnv_eeh_err_inject,
+	.read_config            = pnv_eeh_read_config,
+	.write_config           = pnv_eeh_write_config,
+	.next_error		= pnv_eeh_next_error,
+	.restore_config		= pnv_eeh_restore_config
 };
 
 /**
@@ -521,7 +521,7 @@ static int __init eeh_powernv_init(void)
 	int ret = -EINVAL;
 
 	eeh_set_pe_aux_size(PNV_PCI_DIAG_BUF_SIZE);
-	ret = eeh_ops_register(&powernv_eeh_ops);
+	ret = eeh_ops_register(&pnv_eeh_ops);
 	if (!ret)
 		pr_info("EEH: PowerNV platform initialized\n");
 	else
-- 
1.8.3.2

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 02/10] powerpc/powernv: Drop PHB operation err_inject()
  2015-02-16  3:45 [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation Gavin Shan
  2015-02-16  3:45 ` [PATCH 01/10] powerpc/powernv: Shorten EEH function names Gavin Shan
@ 2015-02-16  3:45 ` Gavin Shan
  2015-02-16  3:45 ` [PATCH 03/10] powerpc/powernv: Drop PHB operation post_init() Gavin Shan
                   ` (7 subsequent siblings)
  9 siblings, 0 replies; 11+ messages in thread
From: Gavin Shan @ 2015-02-16  3:45 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Gavin Shan

The patch drops PHB EEH operation err_inject() and merge its logic
to eeh_ops::err_inject().

Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
---
 arch/powerpc/platforms/powernv/eeh-ioda.c    | 49 ++--------------------------
 arch/powerpc/platforms/powernv/eeh-powernv.c | 38 ++++++++++++++++++---
 arch/powerpc/platforms/powernv/pci.h         |  2 --
 3 files changed, 36 insertions(+), 53 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c
index 2809c98..dd154cf 100644
--- a/arch/powerpc/platforms/powernv/eeh-ioda.c
+++ b/arch/powerpc/platforms/powernv/eeh-ioda.c
@@ -70,7 +70,6 @@ static ssize_t ioda_eeh_ei_write(struct file *filp,
 				 size_t count, loff_t *ppos)
 {
 	struct pci_controller *hose = filp->private_data;
-	struct pnv_phb *phb = hose->private_data;
 	struct eeh_dev *edev;
 	struct eeh_pe *pe;
 	int pe_no, type, func;
@@ -78,7 +77,7 @@ static ssize_t ioda_eeh_ei_write(struct file *filp,
 	char buf[50];
 	int ret;
 
-	if (!phb->eeh_ops || !phb->eeh_ops->err_inject)
+	if (!eeh_ops || !eeh_ops->err_inject)
 		return -ENXIO;
 
 	ret = simple_write_to_buffer(buf, sizeof(buf), ppos, user_buf, count);
@@ -103,7 +102,7 @@ static ssize_t ioda_eeh_ei_write(struct file *filp,
 		return -ENODEV;
 
 	/* Do error injection */
-	ret = phb->eeh_ops->err_inject(pe, type, func, addr, mask);
+	ret = eeh_ops->err_inject(pe, type, func, addr, mask);
 	return ret < 0 ? ret : count;
 }
 
@@ -756,49 +755,6 @@ static int ioda_eeh_configure_bridge(struct eeh_pe *pe)
 	return 0;
 }
 
-static int ioda_eeh_err_inject(struct eeh_pe *pe, int type, int func,
-			       unsigned long addr, unsigned long mask)
-{
-	struct pci_controller *hose = pe->phb;
-	struct pnv_phb *phb = hose->private_data;
-	s64 ret;
-
-	/* Sanity check on error type */
-	if (type != OPAL_ERR_INJECT_TYPE_IOA_BUS_ERR &&
-	    type != OPAL_ERR_INJECT_TYPE_IOA_BUS_ERR64) {
-		pr_warn("%s: Invalid error type %d\n",
-			__func__, type);
-		return -ERANGE;
-	}
-
-	if (func < OPAL_ERR_INJECT_FUNC_IOA_LD_MEM_ADDR ||
-	    func > OPAL_ERR_INJECT_FUNC_IOA_DMA_WR_TARGET) {
-		pr_warn("%s: Invalid error function %d\n",
-			__func__, func);
-		return -ERANGE;
-	}
-
-	/* Firmware supports error injection ? */
-	if (!opal_check_token(OPAL_PCI_ERR_INJECT)) {
-		pr_warn("%s: Firmware doesn't support error injection\n",
-			__func__);
-		return -ENXIO;
-	}
-
-	/* Do error injection */
-	ret = opal_pci_err_inject(phb->opal_id, pe->addr,
-				  type, func, addr, mask);
-	if (ret != OPAL_SUCCESS) {
-		pr_warn("%s: Failure %lld injecting error "
-			"%d-%d to PHB#%x-PE#%x\n",
-			__func__, ret, type, func,
-			hose->global_number, pe->addr);
-		return -EIO;
-	}
-
-	return 0;
-}
-
 static void ioda_eeh_hub_diag_common(struct OpalIoP7IOCErrorData *data)
 {
 	/* GEM */
@@ -1144,6 +1100,5 @@ struct pnv_eeh_ops ioda_eeh_ops = {
 	.reset			= ioda_eeh_reset,
 	.get_log		= ioda_eeh_get_log,
 	.configure_bridge	= ioda_eeh_configure_bridge,
-	.err_inject		= ioda_eeh_err_inject,
 	.next_error		= ioda_eeh_next_error
 };
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index f562dd1..df33daa 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -406,12 +406,42 @@ static int pnv_eeh_err_inject(struct eeh_pe *pe, int type, int func,
 {
 	struct pci_controller *hose = pe->phb;
 	struct pnv_phb *phb = hose->private_data;
-	int ret = -EEXIST;
+	s64 rc;
+
+	/* Sanity check on error type */
+	if (type != OPAL_ERR_INJECT_TYPE_IOA_BUS_ERR &&
+	    type != OPAL_ERR_INJECT_TYPE_IOA_BUS_ERR64) {
+		pr_warn("%s: Invalid error type %d\n",
+			__func__, type);
+		return -ERANGE;
+	}
 
-	if (phb->eeh_ops && phb->eeh_ops->err_inject)
-		ret = phb->eeh_ops->err_inject(pe, type, func, addr, mask);
+	if (func < OPAL_ERR_INJECT_FUNC_IOA_LD_MEM_ADDR ||
+	    func > OPAL_ERR_INJECT_FUNC_IOA_DMA_WR_TARGET) {
+		pr_warn("%s: Invalid error function %d\n",
+			__func__, func);
+		return -ERANGE;
+	}
 
-	return ret;
+	/* Firmware supports error injection ? */
+	if (!opal_check_token(OPAL_PCI_ERR_INJECT)) {
+		pr_warn("%s: Firmware doesn't support error injection\n",
+			__func__);
+		return -ENXIO;
+	}
+
+	/* Do error injection */
+	rc = opal_pci_err_inject(phb->opal_id, pe->addr,
+				 type, func, addr, mask);
+	if (rc != OPAL_SUCCESS) {
+		pr_warn("%s: Failure %lld injecting error "
+			"%d-%d to PHB#%x-PE#%x\n",
+			__func__, rc, type, func,
+			hose->global_number, pe->addr);
+		return -EIO;
+	}
+
+	return 0;
 }
 
 static inline bool pnv_eeh_cfg_blocked(struct device_node *dn)
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index 6c02ff8..a9f2362 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -85,8 +85,6 @@ struct pnv_eeh_ops {
 	int (*get_log)(struct eeh_pe *pe, int severity,
 		       char *drv_log, unsigned long len);
 	int (*configure_bridge)(struct eeh_pe *pe);
-	int (*err_inject)(struct eeh_pe *pe, int type, int func,
-			  unsigned long addr, unsigned long mask);
 	int (*next_error)(struct eeh_pe **pe);
 };
 #endif /* CONFIG_EEH */
-- 
1.8.3.2

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 03/10] powerpc/powernv: Drop PHB operation post_init()
  2015-02-16  3:45 [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation Gavin Shan
  2015-02-16  3:45 ` [PATCH 01/10] powerpc/powernv: Shorten EEH function names Gavin Shan
  2015-02-16  3:45 ` [PATCH 02/10] powerpc/powernv: Drop PHB operation err_inject() Gavin Shan
@ 2015-02-16  3:45 ` Gavin Shan
  2015-02-16  3:45 ` [PATCH 04/10] powerpc/powernv: Drop PHB operation get_log() Gavin Shan
                   ` (6 subsequent siblings)
  9 siblings, 0 replies; 11+ messages in thread
From: Gavin Shan @ 2015-02-16  3:45 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Gavin Shan

The patch drops PHB EEH operation post_init() and merge its logic
to eeh_ops::post_init().

Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
---
 arch/powerpc/platforms/powernv/eeh-ioda.c    | 193 ---------------------------
 arch/powerpc/platforms/powernv/eeh-powernv.c | 184 ++++++++++++++++++++++++-
 arch/powerpc/platforms/powernv/pci.h         |   1 -
 3 files changed, 179 insertions(+), 199 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c
index dd154cf..bd509ad 100644
--- a/arch/powerpc/platforms/powernv/eeh-ioda.c
+++ b/arch/powerpc/platforms/powernv/eeh-ioda.c
@@ -34,198 +34,6 @@
 #include "powernv.h"
 #include "pci.h"
 
-static int ioda_eeh_nb_init = 0;
-
-static int ioda_eeh_event(struct notifier_block *nb,
-			  unsigned long events, void *change)
-{
-	uint64_t changed_evts = (uint64_t)change;
-
-	/*
-	 * We simply send special EEH event if EEH has
-	 * been enabled, or clear pending events in
-	 * case that we enable EEH soon
-	 */
-	if (!(changed_evts & OPAL_EVENT_PCI_ERROR) ||
-	    !(events & OPAL_EVENT_PCI_ERROR))
-		return 0;
-
-	if (eeh_enabled())
-		eeh_send_failure_event(NULL);
-	else
-		opal_notifier_update_evt(OPAL_EVENT_PCI_ERROR, 0x0ul);
-
-	return 0;
-}
-
-static struct notifier_block ioda_eeh_nb = {
-	.notifier_call	= ioda_eeh_event,
-	.next		= NULL,
-	.priority	= 0
-};
-
-#ifdef CONFIG_DEBUG_FS
-static ssize_t ioda_eeh_ei_write(struct file *filp,
-				 const char __user *user_buf,
-				 size_t count, loff_t *ppos)
-{
-	struct pci_controller *hose = filp->private_data;
-	struct eeh_dev *edev;
-	struct eeh_pe *pe;
-	int pe_no, type, func;
-	unsigned long addr, mask;
-	char buf[50];
-	int ret;
-
-	if (!eeh_ops || !eeh_ops->err_inject)
-		return -ENXIO;
-
-	ret = simple_write_to_buffer(buf, sizeof(buf), ppos, user_buf, count);
-	if (!ret)
-		return -EFAULT;
-
-	/* Retrieve parameters */
-	ret = sscanf(buf, "%x:%x:%x:%lx:%lx",
-		     &pe_no, &type, &func, &addr, &mask);
-	if (ret != 5)
-		return -EINVAL;
-
-	/* Retrieve PE */
-	edev = kzalloc(sizeof(*edev), GFP_KERNEL);
-	if (!edev)
-		return -ENOMEM;
-	edev->phb = hose;
-	edev->pe_config_addr = pe_no;
-	pe = eeh_pe_get(edev);
-	kfree(edev);
-	if (!pe)
-		return -ENODEV;
-
-	/* Do error injection */
-	ret = eeh_ops->err_inject(pe, type, func, addr, mask);
-	return ret < 0 ? ret : count;
-}
-
-static const struct file_operations ioda_eeh_ei_fops = {
-	.open   = simple_open,
-	.llseek = no_llseek,
-	.write  = ioda_eeh_ei_write,
-};
-
-static int ioda_eeh_dbgfs_set(void *data, int offset, u64 val)
-{
-	struct pci_controller *hose = data;
-	struct pnv_phb *phb = hose->private_data;
-
-	out_be64(phb->regs + offset, val);
-	return 0;
-}
-
-static int ioda_eeh_dbgfs_get(void *data, int offset, u64 *val)
-{
-	struct pci_controller *hose = data;
-	struct pnv_phb *phb = hose->private_data;
-
-	*val = in_be64(phb->regs + offset);
-	return 0;
-}
-
-static int ioda_eeh_outb_dbgfs_set(void *data, u64 val)
-{
-	return ioda_eeh_dbgfs_set(data, 0xD10, val);
-}
-
-static int ioda_eeh_outb_dbgfs_get(void *data, u64 *val)
-{
-	return ioda_eeh_dbgfs_get(data, 0xD10, val);
-}
-
-static int ioda_eeh_inbA_dbgfs_set(void *data, u64 val)
-{
-	return ioda_eeh_dbgfs_set(data, 0xD90, val);
-}
-
-static int ioda_eeh_inbA_dbgfs_get(void *data, u64 *val)
-{
-	return ioda_eeh_dbgfs_get(data, 0xD90, val);
-}
-
-static int ioda_eeh_inbB_dbgfs_set(void *data, u64 val)
-{
-	return ioda_eeh_dbgfs_set(data, 0xE10, val);
-}
-
-static int ioda_eeh_inbB_dbgfs_get(void *data, u64 *val)
-{
-	return ioda_eeh_dbgfs_get(data, 0xE10, val);
-}
-
-DEFINE_SIMPLE_ATTRIBUTE(ioda_eeh_outb_dbgfs_ops, ioda_eeh_outb_dbgfs_get,
-			ioda_eeh_outb_dbgfs_set, "0x%llx\n");
-DEFINE_SIMPLE_ATTRIBUTE(ioda_eeh_inbA_dbgfs_ops, ioda_eeh_inbA_dbgfs_get,
-			ioda_eeh_inbA_dbgfs_set, "0x%llx\n");
-DEFINE_SIMPLE_ATTRIBUTE(ioda_eeh_inbB_dbgfs_ops, ioda_eeh_inbB_dbgfs_get,
-			ioda_eeh_inbB_dbgfs_set, "0x%llx\n");
-#endif /* CONFIG_DEBUG_FS */
-
-
-/**
- * ioda_eeh_post_init - Chip dependent post initialization
- * @hose: PCI controller
- *
- * The function will be called after eeh PEs and devices
- * have been built. That means the EEH is ready to supply
- * service with I/O cache.
- */
-static int ioda_eeh_post_init(struct pci_controller *hose)
-{
-	struct pnv_phb *phb = hose->private_data;
-	int ret;
-
-	/* Register OPAL event notifier */
-	if (!ioda_eeh_nb_init) {
-		ret = opal_notifier_register(&ioda_eeh_nb);
-		if (ret) {
-			pr_err("%s: Can't register OPAL event notifier (%d)\n",
-			       __func__, ret);
-			return ret;
-		}
-
-		ioda_eeh_nb_init = 1;
-	}
-
-#ifdef CONFIG_DEBUG_FS
-	if (!phb->has_dbgfs && phb->dbgfs) {
-		phb->has_dbgfs = 1;
-
-		debugfs_create_file("err_injct", 0200,
-				    phb->dbgfs, hose,
-				    &ioda_eeh_ei_fops);
-
-		debugfs_create_file("err_injct_outbound", 0600,
-				    phb->dbgfs, hose,
-				    &ioda_eeh_outb_dbgfs_ops);
-		debugfs_create_file("err_injct_inboundA", 0600,
-				    phb->dbgfs, hose,
-				    &ioda_eeh_inbA_dbgfs_ops);
-		debugfs_create_file("err_injct_inboundB", 0600,
-				    phb->dbgfs, hose,
-				    &ioda_eeh_inbB_dbgfs_ops);
-	}
-#endif
-
-	/* If EEH is enabled, we're going to rely on that.
-	 * Otherwise, we restore to conventional mechanism
-	 * to clear frozen PE during PCI config access.
-	 */
-	if (eeh_enabled())
-		phb->flags |= PNV_PHB_FLAG_EEH;
-	else
-		phb->flags &= ~PNV_PHB_FLAG_EEH;
-
-	return 0;
-}
-
 /**
  * ioda_eeh_set_option - Set EEH operation or I/O setting
  * @pe: EEH PE
@@ -1094,7 +902,6 @@ static int ioda_eeh_next_error(struct eeh_pe **pe)
 }
 
 struct pnv_eeh_ops ioda_eeh_ops = {
-	.post_init		= ioda_eeh_post_init,
 	.set_option		= ioda_eeh_set_option,
 	.get_state		= ioda_eeh_get_state,
 	.reset			= ioda_eeh_reset,
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index df33daa..641ba33 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -12,6 +12,7 @@
  */
 
 #include <linux/atomic.h>
+#include <linux/debugfs.h>
 #include <linux/delay.h>
 #include <linux/export.h>
 #include <linux/init.h>
@@ -38,6 +39,8 @@
 #include "powernv.h"
 #include "pci.h"
 
+static bool pnv_eeh_nb_init = false;
+
 /**
  * pnv_eeh_init - EEH platform dependent initialization
  *
@@ -85,6 +88,139 @@ static int pnv_eeh_init(void)
 	return 0;
 }
 
+static int pnv_eeh_event(struct notifier_block *nb,
+			 unsigned long events, void *change)
+{
+	uint64_t changed_evts = (uint64_t)change;
+
+	/*
+	 * We simply send special EEH event if EEH has
+	 * been enabled, or clear pending events in
+	 * case that we enable EEH soon
+	 */
+	if (!(changed_evts & OPAL_EVENT_PCI_ERROR) ||
+	    !(events & OPAL_EVENT_PCI_ERROR))
+		return 0;
+
+	if (eeh_enabled())
+		eeh_send_failure_event(NULL);
+	else
+		opal_notifier_update_evt(OPAL_EVENT_PCI_ERROR, 0x0ul);
+
+	return 0;
+}
+
+static struct notifier_block pnv_eeh_nb = {
+	.notifier_call	= pnv_eeh_event,
+	.next		= NULL,
+	.priority	= 0
+};
+
+#ifdef CONFIG_DEBUG_FS
+static ssize_t pnv_eeh_ei_write(struct file *filp,
+				const char __user *user_buf,
+				size_t count, loff_t *ppos)
+{
+	struct pci_controller *hose = filp->private_data;
+	struct eeh_dev *edev;
+	struct eeh_pe *pe;
+	int pe_no, type, func;
+	unsigned long addr, mask;
+	char buf[50];
+	int ret;
+
+	if (!eeh_ops || !eeh_ops->err_inject)
+		return -ENXIO;
+
+	/* Copy over argument buffer */
+	ret = simple_write_to_buffer(buf, sizeof(buf), ppos, user_buf, count);
+	if (!ret)
+		return -EFAULT;
+
+	/* Retrieve parameters */
+	ret = sscanf(buf, "%x:%x:%x:%lx:%lx",
+		     &pe_no, &type, &func, &addr, &mask);
+	if (ret != 5)
+		return -EINVAL;
+
+	/* Retrieve PE */
+	edev = kzalloc(sizeof(*edev), GFP_KERNEL);
+	if (!edev)
+		return -ENOMEM;
+	edev->phb = hose;
+	edev->pe_config_addr = pe_no;
+	pe = eeh_pe_get(edev);
+	kfree(edev);
+	if (!pe)
+		return -ENODEV;
+
+	/* Do error injection */
+	ret = eeh_ops->err_inject(pe, type, func, addr, mask);
+	return ret < 0 ? ret : count;
+}
+
+static const struct file_operations pnv_eeh_ei_fops = {
+	.open	= simple_open,
+	.llseek	= no_llseek,
+	.write	= pnv_eeh_ei_write,
+};
+
+static int pnv_eeh_dbgfs_set(void *data, int offset, u64 val)
+{
+	struct pci_controller *hose = data;
+	struct pnv_phb *phb = hose->private_data;
+
+	out_be64(phb->regs + offset, val);
+	return 0;
+}
+
+static int pnv_eeh_dbgfs_get(void *data, int offset, u64 *val)
+{
+	struct pci_controller *hose = data;
+	struct pnv_phb *phb = hose->private_data;
+
+	*val = in_be64(phb->regs + offset);
+	return 0;
+}
+
+static int pnv_eeh_outb_dbgfs_set(void *data, u64 val)
+{
+	return pnv_eeh_dbgfs_set(data, 0xD10, val);
+}
+
+static int pnv_eeh_outb_dbgfs_get(void *data, u64 *val)
+{
+	return pnv_eeh_dbgfs_get(data, 0xD10, val);
+}
+
+static int pnv_eeh_inbA_dbgfs_set(void *data, u64 val)
+{
+	return pnv_eeh_dbgfs_set(data, 0xD90, val);
+}
+
+static int pnv_eeh_inbA_dbgfs_get(void *data, u64 *val)
+{
+	return pnv_eeh_dbgfs_get(data, 0xD90, val);
+}
+
+static int pnv_eeh_inbB_dbgfs_set(void *data, u64 val)
+{
+	return pnv_eeh_dbgfs_set(data, 0xE10, val);
+}
+
+static int pnv_eeh_inbB_dbgfs_get(void *data, u64 *val)
+{
+	return pnv_eeh_dbgfs_get(data, 0xE10, val);
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(pnv_eeh_outb_dbgfs_ops, pnv_eeh_outb_dbgfs_get,
+			pnv_eeh_outb_dbgfs_set, "0x%llx\n");
+DEFINE_SIMPLE_ATTRIBUTE(pnv_eeh_inbA_dbgfs_ops, pnv_eeh_inbA_dbgfs_get,
+			pnv_eeh_inbA_dbgfs_set, "0x%llx\n");
+DEFINE_SIMPLE_ATTRIBUTE(pnv_eeh_inbB_dbgfs_ops, pnv_eeh_inbB_dbgfs_get,
+			pnv_eeh_inbB_dbgfs_set, "0x%llx\n");
+#endif /* CONFIG_DEBUG_FS */
+
 /**
  * pnv_eeh_post_init - EEH platform dependent post initialization
  *
@@ -99,16 +235,54 @@ static int pnv_eeh_post_init(void)
 	struct pnv_phb *phb;
 	int ret = 0;
 
+	/* Register OPAL event notifier */
+	if (!pnv_eeh_nb_init) {
+		ret = opal_notifier_register(&pnv_eeh_nb);
+		if (ret) {
+			pr_warn("%s: Can't register OPAL event notifier (%d)\n",
+				__func__, ret);
+			return ret;
+		}
+
+		pnv_eeh_nb_init = true;
+	}
+
 	list_for_each_entry(hose, &hose_list, list_node) {
 		phb = hose->private_data;
 
-		if (phb->eeh_ops && phb->eeh_ops->post_init) {
-			ret = phb->eeh_ops->post_init(hose);
-			if (ret)
-				break;
-		}
+		/*
+		 * If EEH is enabled, we're going to rely on that.
+		 * Otherwise, we restore to conventional mechanism
+		 * to clear frozen PE during PCI config access.
+		 */
+		if (eeh_enabled())
+			phb->flags |= PNV_PHB_FLAG_EEH;
+		else
+			phb->flags &= ~PNV_PHB_FLAG_EEH;
+
+		/* Create debugfs entries */
+#ifdef CONFIG_DEBUG_FS
+		if (phb->has_dbgfs || !phb->dbgfs)
+			continue;
+
+		phb->has_dbgfs = 1;
+		debugfs_create_file("err_injct", 0200,
+				    phb->dbgfs, hose,
+				    &pnv_eeh_ei_fops);
+
+		debugfs_create_file("err_injct_outbound", 0600,
+				    phb->dbgfs, hose,
+				    &pnv_eeh_outb_dbgfs_ops);
+		debugfs_create_file("err_injct_inboundA", 0600,
+				    phb->dbgfs, hose,
+				    &pnv_eeh_inbA_dbgfs_ops);
+		debugfs_create_file("err_injct_inboundB", 0600,
+				    phb->dbgfs, hose,
+				    &pnv_eeh_inbB_dbgfs_ops);
+#endif /* CONFIG_DEBUG_FS */
 	}
 
+
 	return ret;
 }
 
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index a9f2362..c7e047f 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -78,7 +78,6 @@ struct pnv_ioda_pe {
 /* IOC dependent EEH operations */
 #ifdef CONFIG_EEH
 struct pnv_eeh_ops {
-	int (*post_init)(struct pci_controller *hose);
 	int (*set_option)(struct eeh_pe *pe, int option);
 	int (*get_state)(struct eeh_pe *pe);
 	int (*reset)(struct eeh_pe *pe, int option);
-- 
1.8.3.2

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 04/10] powerpc/powernv: Drop PHB operation get_log()
  2015-02-16  3:45 [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation Gavin Shan
                   ` (2 preceding siblings ...)
  2015-02-16  3:45 ` [PATCH 03/10] powerpc/powernv: Drop PHB operation post_init() Gavin Shan
@ 2015-02-16  3:45 ` Gavin Shan
  2015-02-16  3:45 ` [PATCH 05/10] powerpc/powernv: Drop PHB operation configure_bridge() Gavin Shan
                   ` (5 subsequent siblings)
  9 siblings, 0 replies; 11+ messages in thread
From: Gavin Shan @ 2015-02-16  3:45 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Gavin Shan

The patch drops PHB operation get_log() and merges its logic to
eeh_ops::get_log().

Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
---
 arch/powerpc/platforms/powernv/eeh-ioda.c    | 20 --------------------
 arch/powerpc/platforms/powernv/eeh-powernv.c | 10 +++-------
 arch/powerpc/platforms/powernv/pci.h         |  2 --
 3 files changed, 3 insertions(+), 29 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c
index bd509ad..7eb6e72 100644
--- a/arch/powerpc/platforms/powernv/eeh-ioda.c
+++ b/arch/powerpc/platforms/powernv/eeh-ioda.c
@@ -531,25 +531,6 @@ static int ioda_eeh_reset(struct eeh_pe *pe, int option)
 }
 
 /**
- * ioda_eeh_get_log - Retrieve error log
- * @pe: frozen PE
- * @severity: permanent or temporary error
- * @drv_log: device driver log
- * @len: length of device driver log
- *
- * Retrieve error log, which contains log from device driver
- * and firmware.
- */
-static int ioda_eeh_get_log(struct eeh_pe *pe, int severity,
-			    char *drv_log, unsigned long len)
-{
-	if (!eeh_has_flag(EEH_EARLY_DUMP_LOG))
-		pnv_pci_dump_phb_diag_data(pe->phb, pe->data);
-
-	return 0;
-}
-
-/**
  * ioda_eeh_configure_bridge - Configure the PCI bridges for the indicated PE
  * @pe: EEH PE
  *
@@ -905,7 +886,6 @@ struct pnv_eeh_ops ioda_eeh_ops = {
 	.set_option		= ioda_eeh_set_option,
 	.get_state		= ioda_eeh_get_state,
 	.reset			= ioda_eeh_reset,
-	.get_log		= ioda_eeh_get_log,
 	.configure_bridge	= ioda_eeh_configure_bridge,
 	.next_error		= ioda_eeh_next_error
 };
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index 641ba33..465deb5 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -533,14 +533,10 @@ static int pnv_eeh_wait_state(struct eeh_pe *pe, int max_wait)
 static int pnv_eeh_get_log(struct eeh_pe *pe, int severity,
 			   char *drv_log, unsigned long len)
 {
-	struct pci_controller *hose = pe->phb;
-	struct pnv_phb *phb = hose->private_data;
-	int ret = -EEXIST;
+	if (!eeh_has_flag(EEH_EARLY_DUMP_LOG))
+		pnv_pci_dump_phb_diag_data(pe->phb, pe->data);
 
-	if (phb->eeh_ops && phb->eeh_ops->get_log)
-		ret = phb->eeh_ops->get_log(pe, severity, drv_log, len);
-
-	return ret;
+	return 0;
 }
 
 /**
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index c7e047f..1e7a623 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -81,8 +81,6 @@ struct pnv_eeh_ops {
 	int (*set_option)(struct eeh_pe *pe, int option);
 	int (*get_state)(struct eeh_pe *pe);
 	int (*reset)(struct eeh_pe *pe, int option);
-	int (*get_log)(struct eeh_pe *pe, int severity,
-		       char *drv_log, unsigned long len);
 	int (*configure_bridge)(struct eeh_pe *pe);
 	int (*next_error)(struct eeh_pe **pe);
 };
-- 
1.8.3.2

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 05/10] powerpc/powernv: Drop PHB operation configure_bridge()
  2015-02-16  3:45 [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation Gavin Shan
                   ` (3 preceding siblings ...)
  2015-02-16  3:45 ` [PATCH 04/10] powerpc/powernv: Drop PHB operation get_log() Gavin Shan
@ 2015-02-16  3:45 ` Gavin Shan
  2015-02-16  3:45 ` [PATCH 06/10] powerpc/powernv: Drop PHB operation set_option() Gavin Shan
                   ` (4 subsequent siblings)
  9 siblings, 0 replies; 11+ messages in thread
From: Gavin Shan @ 2015-02-16  3:45 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Gavin Shan

The patch drops PHB EEH operation configure_bridge() and merges
its logic to eeh_ops::configure_bridge().

Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
---
 arch/powerpc/platforms/powernv/eeh-ioda.c    | 15 ---------------
 arch/powerpc/platforms/powernv/eeh-powernv.c |  9 +--------
 arch/powerpc/platforms/powernv/pci.h         |  1 -
 3 files changed, 1 insertion(+), 24 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c
index 7eb6e72..6459516 100644
--- a/arch/powerpc/platforms/powernv/eeh-ioda.c
+++ b/arch/powerpc/platforms/powernv/eeh-ioda.c
@@ -530,20 +530,6 @@ static int ioda_eeh_reset(struct eeh_pe *pe, int option)
 	return ret;
 }
 
-/**
- * ioda_eeh_configure_bridge - Configure the PCI bridges for the indicated PE
- * @pe: EEH PE
- *
- * For particular PE, it might have included PCI bridges. In order
- * to make the PE work properly, those PCI bridges should be configured
- * correctly. However, we need do nothing on P7IOC since the reset
- * function will do everything that should be covered by the function.
- */
-static int ioda_eeh_configure_bridge(struct eeh_pe *pe)
-{
-	return 0;
-}
-
 static void ioda_eeh_hub_diag_common(struct OpalIoP7IOCErrorData *data)
 {
 	/* GEM */
@@ -886,6 +872,5 @@ struct pnv_eeh_ops ioda_eeh_ops = {
 	.set_option		= ioda_eeh_set_option,
 	.get_state		= ioda_eeh_get_state,
 	.reset			= ioda_eeh_reset,
-	.configure_bridge	= ioda_eeh_configure_bridge,
 	.next_error		= ioda_eeh_next_error
 };
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index 465deb5..a7087f4 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -549,14 +549,7 @@ static int pnv_eeh_get_log(struct eeh_pe *pe, int severity,
  */
 static int pnv_eeh_configure_bridge(struct eeh_pe *pe)
 {
-	struct pci_controller *hose = pe->phb;
-	struct pnv_phb *phb = hose->private_data;
-	int ret = 0;
-
-	if (phb->eeh_ops && phb->eeh_ops->configure_bridge)
-		ret = phb->eeh_ops->configure_bridge(pe);
-
-	return ret;
+	return 0;
 }
 
 /**
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index 1e7a623..d8808ca 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -81,7 +81,6 @@ struct pnv_eeh_ops {
 	int (*set_option)(struct eeh_pe *pe, int option);
 	int (*get_state)(struct eeh_pe *pe);
 	int (*reset)(struct eeh_pe *pe, int option);
-	int (*configure_bridge)(struct eeh_pe *pe);
 	int (*next_error)(struct eeh_pe **pe);
 };
 #endif /* CONFIG_EEH */
-- 
1.8.3.2

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 06/10] powerpc/powernv: Drop PHB operation set_option()
  2015-02-16  3:45 [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation Gavin Shan
                   ` (4 preceding siblings ...)
  2015-02-16  3:45 ` [PATCH 05/10] powerpc/powernv: Drop PHB operation configure_bridge() Gavin Shan
@ 2015-02-16  3:45 ` Gavin Shan
  2015-02-16  3:45 ` [PATCH 07/10] powerpc/powernv: Drop PHB operation get_state() Gavin Shan
                   ` (3 subsequent siblings)
  9 siblings, 0 replies; 11+ messages in thread
From: Gavin Shan @ 2015-02-16  3:45 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Gavin Shan

The patch drops PHB EEH operation set_option() and merges its
logic to eeh_ops::set_option().

Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
---
 arch/powerpc/platforms/powernv/eeh-ioda.c    | 84 ----------------------------
 arch/powerpc/platforms/powernv/eeh-powernv.c | 61 +++++++++++++++++---
 arch/powerpc/platforms/powernv/pci.h         |  1 -
 3 files changed, 54 insertions(+), 92 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c
index 6459516..349c083 100644
--- a/arch/powerpc/platforms/powernv/eeh-ioda.c
+++ b/arch/powerpc/platforms/powernv/eeh-ioda.c
@@ -34,89 +34,6 @@
 #include "powernv.h"
 #include "pci.h"
 
-/**
- * ioda_eeh_set_option - Set EEH operation or I/O setting
- * @pe: EEH PE
- * @option: options
- *
- * Enable or disable EEH option for the indicated PE. The
- * function also can be used to enable I/O or DMA for the
- * PE.
- */
-static int ioda_eeh_set_option(struct eeh_pe *pe, int option)
-{
-	struct pci_controller *hose = pe->phb;
-	struct pnv_phb *phb = hose->private_data;
-	bool freeze_pe = false;
-	int enable, ret = 0;
-	s64 rc;
-
-	/* Check on PE number */
-	if (pe->addr < 0 || pe->addr >= phb->ioda.total_pe) {
-		pr_err("%s: PE address %x out of range [0, %x] "
-		       "on PHB#%x\n",
-			__func__, pe->addr, phb->ioda.total_pe,
-			hose->global_number);
-		return -EINVAL;
-	}
-
-	switch (option) {
-	case EEH_OPT_DISABLE:
-		return -EPERM;
-	case EEH_OPT_ENABLE:
-		return 0;
-	case EEH_OPT_THAW_MMIO:
-		enable = OPAL_EEH_ACTION_CLEAR_FREEZE_MMIO;
-		break;
-	case EEH_OPT_THAW_DMA:
-		enable = OPAL_EEH_ACTION_CLEAR_FREEZE_DMA;
-		break;
-	case EEH_OPT_FREEZE_PE:
-		freeze_pe = true;
-		enable = OPAL_EEH_ACTION_SET_FREEZE_ALL;
-		break;
-	default:
-		pr_warn("%s: Invalid option %d\n",
-			__func__, option);
-		return -EINVAL;
-	}
-
-	/* If PHB supports compound PE, to handle it */
-	if (freeze_pe) {
-		if (phb->freeze_pe) {
-			phb->freeze_pe(phb, pe->addr);
-		} else {
-			rc = opal_pci_eeh_freeze_set(phb->opal_id,
-						     pe->addr,
-						     enable);
-			if (rc != OPAL_SUCCESS) {
-				pr_warn("%s: Failure %lld freezing "
-					"PHB#%x-PE#%x\n",
-					__func__, rc,
-					phb->hose->global_number, pe->addr);
-				ret = -EIO;
-			}
-		}
-	} else {
-		if (phb->unfreeze_pe) {
-			ret = phb->unfreeze_pe(phb, pe->addr, enable);
-		} else {
-			rc = opal_pci_eeh_freeze_clear(phb->opal_id,
-						       pe->addr,
-						       enable);
-			if (rc != OPAL_SUCCESS) {
-				pr_warn("%s: Failure %lld enable %d "
-					"for PHB#%x-PE#%x\n",
-					__func__, rc, option,
-					phb->hose->global_number, pe->addr);
-				ret = -EIO;
-			}
-		}
-	}
-
-	return ret;
-}
-
 static void ioda_eeh_phb_diag(struct eeh_pe *pe)
 {
 	struct pnv_phb *phb = pe->phb->private_data;
@@ -869,7 +786,6 @@ static int ioda_eeh_next_error(struct eeh_pe **pe)
 }
 
 struct pnv_eeh_ops ioda_eeh_ops = {
-	.set_option		= ioda_eeh_set_option,
 	.get_state		= ioda_eeh_get_state,
 	.reset			= ioda_eeh_reset,
 	.next_error		= ioda_eeh_next_error
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index a7087f4..2429a23 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -407,14 +407,61 @@ static int pnv_eeh_set_option(struct eeh_pe *pe, int option)
 {
 	struct pci_controller *hose = pe->phb;
 	struct pnv_phb *phb = hose->private_data;
-	int ret = -EEXIST;
+	bool freeze_pe = false;
+	int opt, ret = 0;
+	s64 rc;
 
-	/*
-	 * What we need do is pass it down for hardware
-	 * implementation to handle it.
-	 */
-	if (phb->eeh_ops && phb->eeh_ops->set_option)
-		ret = phb->eeh_ops->set_option(pe, option);
+	/* Sanity check on option */
+	switch (option) {
+	case EEH_OPT_DISABLE:
+		return -EPERM;
+	case EEH_OPT_ENABLE:
+		return 0;
+	case EEH_OPT_THAW_MMIO:
+		opt = OPAL_EEH_ACTION_CLEAR_FREEZE_MMIO;
+		break;
+	case EEH_OPT_THAW_DMA:
+		opt = OPAL_EEH_ACTION_CLEAR_FREEZE_DMA;
+		break;
+	case EEH_OPT_FREEZE_PE:
+		freeze_pe = true;
+		opt = OPAL_EEH_ACTION_SET_FREEZE_ALL;
+		break;
+	default:
+		pr_warn("%s: Invalid option %d\n", __func__, option);
+		return -EINVAL;
+	}
+
+	/* If PHB supports compound PE, to handle it */
+	if (freeze_pe) {
+		if (phb->freeze_pe) {
+			phb->freeze_pe(phb, pe->addr);
+		} else {
+			rc = opal_pci_eeh_freeze_set(phb->opal_id,
+						     pe->addr, opt);
+			if (rc != OPAL_SUCCESS) {
+				pr_warn("%s: Failure %lld freezing "
+					"PHB#%x-PE#%x\n",
+					__func__, rc,
+					phb->hose->global_number, pe->addr);
+				ret = -EIO;
+			}
+		}
+	} else {
+		if (phb->unfreeze_pe) {
+			ret = phb->unfreeze_pe(phb, pe->addr, opt);
+		} else {
+			rc = opal_pci_eeh_freeze_clear(phb->opal_id,
+						       pe->addr, opt);
+			if (rc != OPAL_SUCCESS) {
+				pr_warn("%s: Failure %lld enable %d "
+					"for PHB#%x-PE#%x\n",
+					__func__, rc, option,
+					phb->hose->global_number, pe->addr);
+				ret = -EIO;
+			}
+		}
+	}
 
 	return ret;
 }
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index d8808ca..8043dee 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -78,7 +78,6 @@ struct pnv_ioda_pe {
 /* IOC dependent EEH operations */
 #ifdef CONFIG_EEH
 struct pnv_eeh_ops {
-	int (*set_option)(struct eeh_pe *pe, int option);
 	int (*get_state)(struct eeh_pe *pe);
 	int (*reset)(struct eeh_pe *pe, int option);
 	int (*next_error)(struct eeh_pe **pe);
-- 
1.8.3.2

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 07/10] powerpc/powernv: Drop PHB operation get_state()
  2015-02-16  3:45 [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation Gavin Shan
                   ` (5 preceding siblings ...)
  2015-02-16  3:45 ` [PATCH 06/10] powerpc/powernv: Drop PHB operation set_option() Gavin Shan
@ 2015-02-16  3:45 ` Gavin Shan
  2015-02-16  3:45 ` [PATCH 08/10] powerpc/powernv: Drop PHB operation next_error() Gavin Shan
                   ` (2 subsequent siblings)
  9 siblings, 0 replies; 11+ messages in thread
From: Gavin Shan @ 2015-02-16  3:45 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Gavin Shan

The patch drops PHB EEH operation get_state() and merges its logic
to eeh_ops::get_state().

Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
---
 arch/powerpc/platforms/powernv/eeh-ioda.c    | 170 +-----------------------
 arch/powerpc/platforms/powernv/eeh-powernv.c | 185 ++++++++++++++++++++++++---
 arch/powerpc/platforms/powernv/pci.h         |   1 -
 3 files changed, 170 insertions(+), 186 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c
index 349c083..dc34c36 100644
--- a/arch/powerpc/platforms/powernv/eeh-ioda.c
+++ b/arch/powerpc/platforms/powernv/eeh-ioda.c
@@ -46,173 +46,6 @@ static void ioda_eeh_phb_diag(struct eeh_pe *pe)
 			__func__, pe->phb->global_number, rc);
 }
 
-static int ioda_eeh_get_phb_state(struct eeh_pe *pe)
-{
-	struct pnv_phb *phb = pe->phb->private_data;
-	u8 fstate;
-	__be16 pcierr;
-	s64 rc;
-	int result = 0;
-
-	rc = opal_pci_eeh_freeze_status(phb->opal_id,
-					pe->addr,
-					&fstate,
-					&pcierr,
-					NULL);
-	if (rc != OPAL_SUCCESS) {
-		pr_warn("%s: Failure %lld getting PHB#%x state\n",
-			__func__, rc, phb->hose->global_number);
-		return EEH_STATE_NOT_SUPPORT;
-	}
-
-	/*
-	 * Check PHB state. If the PHB is frozen for the
-	 * first time, to dump the PHB diag-data.
-	 */
-	if (be16_to_cpu(pcierr) != OPAL_EEH_PHB_ERROR) {
-		result = (EEH_STATE_MMIO_ACTIVE  |
-			  EEH_STATE_DMA_ACTIVE   |
-			  EEH_STATE_MMIO_ENABLED |
-			  EEH_STATE_DMA_ENABLED);
-	} else if (!(pe->state & EEH_PE_ISOLATED)) {
-		eeh_pe_state_mark(pe, EEH_PE_ISOLATED);
-		ioda_eeh_phb_diag(pe);
-
-		if (eeh_has_flag(EEH_EARLY_DUMP_LOG))
-			pnv_pci_dump_phb_diag_data(pe->phb, pe->data);
-	}
-
-	return result;
-}
-
-static int ioda_eeh_get_pe_state(struct eeh_pe *pe)
-{
-	struct pnv_phb *phb = pe->phb->private_data;
-	u8 fstate;
-	__be16 pcierr;
-	s64 rc;
-	int result;
-
-	/*
-	 * We don't clobber hardware frozen state until PE
-	 * reset is completed. In order to keep EEH core
-	 * moving forward, we have to return operational
-	 * state during PE reset.
-	 */
-	if (pe->state & EEH_PE_RESET) {
-		result = (EEH_STATE_MMIO_ACTIVE  |
-			  EEH_STATE_DMA_ACTIVE   |
-			  EEH_STATE_MMIO_ENABLED |
-			  EEH_STATE_DMA_ENABLED);
-		return result;
-	}
-
-	/*
-	 * Fetch PE state from hardware. If the PHB
-	 * supports compound PE, let it handle that.
-	 */
-	if (phb->get_pe_state) {
-		fstate = phb->get_pe_state(phb, pe->addr);
-	} else {
-		rc = opal_pci_eeh_freeze_status(phb->opal_id,
-						pe->addr,
-						&fstate,
-						&pcierr,
-						NULL);
-		if (rc != OPAL_SUCCESS) {
-			pr_warn("%s: Failure %lld getting PHB#%x-PE%x state\n",
-				__func__, rc, phb->hose->global_number, pe->addr);
-			return EEH_STATE_NOT_SUPPORT;
-		}
-	}
-
-	/* Figure out state */
-	switch (fstate) {
-	case OPAL_EEH_STOPPED_NOT_FROZEN:
-		result = (EEH_STATE_MMIO_ACTIVE  |
-			  EEH_STATE_DMA_ACTIVE   |
-			  EEH_STATE_MMIO_ENABLED |
-			  EEH_STATE_DMA_ENABLED);
-		break;
-	case OPAL_EEH_STOPPED_MMIO_FREEZE:
-		result = (EEH_STATE_DMA_ACTIVE |
-			  EEH_STATE_DMA_ENABLED);
-		break;
-	case OPAL_EEH_STOPPED_DMA_FREEZE:
-		result = (EEH_STATE_MMIO_ACTIVE |
-			  EEH_STATE_MMIO_ENABLED);
-		break;
-	case OPAL_EEH_STOPPED_MMIO_DMA_FREEZE:
-		result = 0;
-		break;
-	case OPAL_EEH_STOPPED_RESET:
-		result = EEH_STATE_RESET_ACTIVE;
-		break;
-	case OPAL_EEH_STOPPED_TEMP_UNAVAIL:
-		result = EEH_STATE_UNAVAILABLE;
-		break;
-	case OPAL_EEH_STOPPED_PERM_UNAVAIL:
-		result = EEH_STATE_NOT_SUPPORT;
-		break;
-	default:
-		result = EEH_STATE_NOT_SUPPORT;
-		pr_warn("%s: Invalid PHB#%x-PE#%x state %x\n",
-			__func__, phb->hose->global_number,
-			pe->addr, fstate);
-	}
-
-	/*
-	 * If PHB supports compound PE, to freeze all
-	 * slave PEs for consistency.
-	 *
-	 * If the PE is switching to frozen state for the
-	 * first time, to dump the PHB diag-data.
-	 */
-	if (!(result & EEH_STATE_NOT_SUPPORT) &&
-	    !(result & EEH_STATE_UNAVAILABLE) &&
-	    !(result & EEH_STATE_MMIO_ACTIVE) &&
-	    !(result & EEH_STATE_DMA_ACTIVE)  &&
-	    !(pe->state & EEH_PE_ISOLATED)) {
-		if (phb->freeze_pe)
-			phb->freeze_pe(phb, pe->addr);
-
-		eeh_pe_state_mark(pe, EEH_PE_ISOLATED);
-		ioda_eeh_phb_diag(pe);
-
-		if (eeh_has_flag(EEH_EARLY_DUMP_LOG))
-			pnv_pci_dump_phb_diag_data(pe->phb, pe->data);
-	}
-
-	return result;
-}
-
-/**
- * ioda_eeh_get_state - Retrieve the state of PE
- * @pe: EEH PE
- *
- * The PE's state should be retrieved from the PEEV, PEST
- * IODA tables. Since the OPAL has exported the function
- * to do it, it'd better to use that.
- */
-static int ioda_eeh_get_state(struct eeh_pe *pe)
-{
-	struct pnv_phb *phb = pe->phb->private_data;
-
-	/* Sanity check on PE number. PHB PE should have 0 */
-	if (pe->addr < 0 ||
-	    pe->addr >= phb->ioda.total_pe) {
-		pr_warn("%s: PHB#%x-PE#%x out of range [0, %x]\n",
-			__func__, phb->hose->global_number,
-			pe->addr, phb->ioda.total_pe);
-		return EEH_STATE_NOT_SUPPORT;
-	}
-
-	if (pe->type & EEH_PE_PHB)
-		return ioda_eeh_get_phb_state(pe);
-
-	return ioda_eeh_get_pe_state(pe);
-}
-
 static s64 ioda_eeh_phb_poll(struct pnv_phb *phb)
 {
 	s64 rc = OPAL_HARDWARE;
@@ -759,7 +592,7 @@ static int ioda_eeh_next_error(struct eeh_pe **pe)
 					break;
 
 				/* Frozen parent PE ? */
-				state = ioda_eeh_get_state(parent_pe);
+				state = eeh_ops->get_state(parent_pe, NULL);
 				if (state > 0 &&
 				    (state & active_flags) != active_flags)
 					*pe = parent_pe;
@@ -786,7 +619,6 @@ static int ioda_eeh_next_error(struct eeh_pe **pe)
 }
 
 struct pnv_eeh_ops ioda_eeh_ops = {
-	.get_state		= ioda_eeh_get_state,
 	.reset			= ioda_eeh_reset,
 	.next_error		= ioda_eeh_next_error
 };
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index 2429a23..127ef0c 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -478,6 +478,159 @@ static int pnv_eeh_get_pe_addr(struct eeh_pe *pe)
 	return pe->addr;
 }
 
+static void pnv_eeh_get_phb_diag(struct eeh_pe *pe)
+{
+	struct pnv_phb *phb = pe->phb->private_data;
+	s64 rc;
+
+	rc = opal_pci_get_phb_diag_data2(phb->opal_id, pe->data,
+					 PNV_PCI_DIAG_BUF_SIZE);
+	if (rc != OPAL_SUCCESS)
+		pr_warn("%s: Failure %lld getting PHB#%x diag-data\n",
+			__func__, rc, pe->phb->global_number);
+}
+
+static int pnv_eeh_get_phb_state(struct eeh_pe *pe)
+{
+	struct pnv_phb *phb = pe->phb->private_data;
+	u8 fstate;
+	__be16 pcierr;
+	s64 rc;
+	int result = 0;
+
+	rc = opal_pci_eeh_freeze_status(phb->opal_id,
+					pe->addr,
+					&fstate,
+					&pcierr,
+					NULL);
+	if (rc != OPAL_SUCCESS) {
+		pr_warn("%s: Failure %lld getting PHB#%x state\n",
+			__func__, rc, phb->hose->global_number);
+		return EEH_STATE_NOT_SUPPORT;
+	}
+
+	/*
+	 * Check PHB state. If the PHB is frozen for the
+	 * first time, to dump the PHB diag-data.
+	 */
+	if (be16_to_cpu(pcierr) != OPAL_EEH_PHB_ERROR) {
+		result = (EEH_STATE_MMIO_ACTIVE  |
+			  EEH_STATE_DMA_ACTIVE   |
+			  EEH_STATE_MMIO_ENABLED |
+			  EEH_STATE_DMA_ENABLED);
+	} else if (!(pe->state & EEH_PE_ISOLATED)) {
+		eeh_pe_state_mark(pe, EEH_PE_ISOLATED);
+		pnv_eeh_get_phb_diag(pe);
+
+		if (eeh_has_flag(EEH_EARLY_DUMP_LOG))
+			pnv_pci_dump_phb_diag_data(pe->phb, pe->data);
+	}
+
+	return result;
+}
+
+static int pnv_eeh_get_pe_state(struct eeh_pe *pe)
+{
+	struct pnv_phb *phb = pe->phb->private_data;
+	u8 fstate;
+	__be16 pcierr;
+	s64 rc;
+	int result;
+
+	/*
+	 * We don't clobber hardware frozen state until PE
+	 * reset is completed. In order to keep EEH core
+	 * moving forward, we have to return operational
+	 * state during PE reset.
+	 */
+	if (pe->state & EEH_PE_RESET) {
+		result = (EEH_STATE_MMIO_ACTIVE  |
+			  EEH_STATE_DMA_ACTIVE   |
+			  EEH_STATE_MMIO_ENABLED |
+			  EEH_STATE_DMA_ENABLED);
+		return result;
+	}
+
+	/*
+	 * Fetch PE state from hardware. If the PHB
+	 * supports compound PE, let it handle that.
+	 */
+	if (phb->get_pe_state) {
+		fstate = phb->get_pe_state(phb, pe->addr);
+	} else {
+		rc = opal_pci_eeh_freeze_status(phb->opal_id,
+						pe->addr,
+						&fstate,
+						&pcierr,
+						NULL);
+		if (rc != OPAL_SUCCESS) {
+			pr_warn("%s: Failure %lld getting PHB#%x-PE%x state\n",
+				__func__, rc, phb->hose->global_number,
+				pe->addr);
+			return EEH_STATE_NOT_SUPPORT;
+		}
+	}
+
+	/* Figure out state */
+	switch (fstate) {
+	case OPAL_EEH_STOPPED_NOT_FROZEN:
+		result = (EEH_STATE_MMIO_ACTIVE  |
+			  EEH_STATE_DMA_ACTIVE   |
+			  EEH_STATE_MMIO_ENABLED |
+			  EEH_STATE_DMA_ENABLED);
+		break;
+	case OPAL_EEH_STOPPED_MMIO_FREEZE:
+		result = (EEH_STATE_DMA_ACTIVE |
+			  EEH_STATE_DMA_ENABLED);
+		break;
+	case OPAL_EEH_STOPPED_DMA_FREEZE:
+		result = (EEH_STATE_MMIO_ACTIVE |
+			  EEH_STATE_MMIO_ENABLED);
+		break;
+	case OPAL_EEH_STOPPED_MMIO_DMA_FREEZE:
+		result = 0;
+		break;
+	case OPAL_EEH_STOPPED_RESET:
+		result = EEH_STATE_RESET_ACTIVE;
+		break;
+	case OPAL_EEH_STOPPED_TEMP_UNAVAIL:
+		result = EEH_STATE_UNAVAILABLE;
+		break;
+	case OPAL_EEH_STOPPED_PERM_UNAVAIL:
+		result = EEH_STATE_NOT_SUPPORT;
+		break;
+	default:
+		result = EEH_STATE_NOT_SUPPORT;
+		pr_warn("%s: Invalid PHB#%x-PE#%x state %x\n",
+			__func__, phb->hose->global_number,
+			pe->addr, fstate);
+	}
+
+	/*
+	 * If PHB supports compound PE, to freeze all
+	 * slave PEs for consistency.
+	 *
+	 * If the PE is switching to frozen state for the
+	 * first time, to dump the PHB diag-data.
+	 */
+	if (!(result & EEH_STATE_NOT_SUPPORT) &&
+	    !(result & EEH_STATE_UNAVAILABLE) &&
+	    !(result & EEH_STATE_MMIO_ACTIVE) &&
+	    !(result & EEH_STATE_DMA_ACTIVE)  &&
+	    !(pe->state & EEH_PE_ISOLATED)) {
+		if (phb->freeze_pe)
+			phb->freeze_pe(phb, pe->addr);
+
+		eeh_pe_state_mark(pe, EEH_PE_ISOLATED);
+		pnv_eeh_get_phb_diag(pe);
+
+		if (eeh_has_flag(EEH_EARLY_DUMP_LOG))
+			pnv_pci_dump_phb_diag_data(pe->phb, pe->data);
+	}
+
+	return result;
+}
+
 /**
  * pnv_eeh_get_state - Retrieve PE state
  * @pe: EEH PE
@@ -490,24 +643,24 @@ static int pnv_eeh_get_pe_addr(struct eeh_pe *pe)
  */
 static int pnv_eeh_get_state(struct eeh_pe *pe, int *delay)
 {
-	struct pci_controller *hose = pe->phb;
-	struct pnv_phb *phb = hose->private_data;
-	int ret = EEH_STATE_NOT_SUPPORT;
+	int ret;
 
-	if (phb->eeh_ops && phb->eeh_ops->get_state) {
-		ret = phb->eeh_ops->get_state(pe);
+	if (pe->type & EEH_PE_PHB)
+		ret = pnv_eeh_get_phb_state(pe);
+	else
+		ret = pnv_eeh_get_pe_state(pe);
 
-		/*
-		 * If the PE state is temporarily unavailable,
-		 * to inform the EEH core delay for default
-		 * period (1 second)
-		 */
-		if (delay) {
-			*delay = 0;
-			if (ret & EEH_STATE_UNAVAILABLE)
-				*delay = 1000;
-		}
-	}
+	if (!delay)
+		return ret;
+
+	/*
+	 * If the PE state is temporarily unavailable,
+	 * to inform the EEH core delay for default
+	 * period (1 second)
+	 */
+	*delay = 0;
+	if (ret & EEH_STATE_UNAVAILABLE)
+		*delay = 1000;
 
 	return ret;
 }
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index 8043dee..773a026 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -78,7 +78,6 @@ struct pnv_ioda_pe {
 /* IOC dependent EEH operations */
 #ifdef CONFIG_EEH
 struct pnv_eeh_ops {
-	int (*get_state)(struct eeh_pe *pe);
 	int (*reset)(struct eeh_pe *pe, int option);
 	int (*next_error)(struct eeh_pe **pe);
 };
-- 
1.8.3.2

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 08/10] powerpc/powernv: Drop PHB operation next_error()
  2015-02-16  3:45 [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation Gavin Shan
                   ` (6 preceding siblings ...)
  2015-02-16  3:45 ` [PATCH 07/10] powerpc/powernv: Drop PHB operation get_state() Gavin Shan
@ 2015-02-16  3:45 ` Gavin Shan
  2015-02-16  3:45 ` [PATCH 09/10] powerpc/powernv: Drop PHB operation reset() Gavin Shan
  2015-02-16  3:45 ` [PATCH 10/10] powerpc/powernv: Remove unused file Gavin Shan
  9 siblings, 0 replies; 11+ messages in thread
From: Gavin Shan @ 2015-02-16  3:45 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Gavin Shan

The patch drops PHB EEH operation next_error() and merges its
logic to eeh_ops::next_error().

Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
---
 arch/powerpc/platforms/powernv/eeh-ioda.c    | 351 ---------------------------
 arch/powerpc/platforms/powernv/eeh-powernv.c | 334 ++++++++++++++++++++++++-
 arch/powerpc/platforms/powernv/pci.h         |   1 -
 3 files changed, 327 insertions(+), 359 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c
index dc34c36..94d94b4 100644
--- a/arch/powerpc/platforms/powernv/eeh-ioda.c
+++ b/arch/powerpc/platforms/powernv/eeh-ioda.c
@@ -34,18 +34,6 @@
 #include "powernv.h"
 #include "pci.h"
 
-static void ioda_eeh_phb_diag(struct eeh_pe *pe)
-{
-	struct pnv_phb *phb = pe->phb->private_data;
-	long rc;
-
-	rc = opal_pci_get_phb_diag_data2(phb->opal_id, pe->data,
-					 PNV_PCI_DIAG_BUF_SIZE);
-	if (rc != OPAL_SUCCESS)
-		pr_warn("%s: Failed to get diag-data for PHB#%x (%ld)\n",
-			__func__, pe->phb->global_number, rc);
-}
-
 static s64 ioda_eeh_phb_poll(struct pnv_phb *phb)
 {
 	s64 rc = OPAL_HARDWARE;
@@ -280,345 +268,6 @@ static int ioda_eeh_reset(struct eeh_pe *pe, int option)
 	return ret;
 }
 
-static void ioda_eeh_hub_diag_common(struct OpalIoP7IOCErrorData *data)
-{
-	/* GEM */
-	if (data->gemXfir || data->gemRfir ||
-	    data->gemRirqfir || data->gemMask || data->gemRwof)
-		pr_info("  GEM: %016llx %016llx %016llx %016llx %016llx\n",
-			be64_to_cpu(data->gemXfir),
-			be64_to_cpu(data->gemRfir),
-			be64_to_cpu(data->gemRirqfir),
-			be64_to_cpu(data->gemMask),
-			be64_to_cpu(data->gemRwof));
-
-	/* LEM */
-	if (data->lemFir || data->lemErrMask ||
-	    data->lemAction0 || data->lemAction1 || data->lemWof)
-		pr_info("  LEM: %016llx %016llx %016llx %016llx %016llx\n",
-			be64_to_cpu(data->lemFir),
-			be64_to_cpu(data->lemErrMask),
-			be64_to_cpu(data->lemAction0),
-			be64_to_cpu(data->lemAction1),
-			be64_to_cpu(data->lemWof));
-}
-
-static void ioda_eeh_hub_diag(struct pci_controller *hose)
-{
-	struct pnv_phb *phb = hose->private_data;
-	struct OpalIoP7IOCErrorData *data = &phb->diag.hub_diag;
-	long rc;
-
-	rc = opal_pci_get_hub_diag_data(phb->hub_id, data, sizeof(*data));
-	if (rc != OPAL_SUCCESS) {
-		pr_warn("%s: Failed to get HUB#%llx diag-data (%ld)\n",
-			__func__, phb->hub_id, rc);
-		return;
-	}
-
-	switch (data->type) {
-	case OPAL_P7IOC_DIAG_TYPE_RGC:
-		pr_info("P7IOC diag-data for RGC\n\n");
-		ioda_eeh_hub_diag_common(data);
-		if (data->rgc.rgcStatus || data->rgc.rgcLdcp)
-			pr_info("  RGC: %016llx %016llx\n",
-				be64_to_cpu(data->rgc.rgcStatus),
-				be64_to_cpu(data->rgc.rgcLdcp));
-		break;
-	case OPAL_P7IOC_DIAG_TYPE_BI:
-		pr_info("P7IOC diag-data for BI %s\n\n",
-			data->bi.biDownbound ? "Downbound" : "Upbound");
-		ioda_eeh_hub_diag_common(data);
-		if (data->bi.biLdcp0 || data->bi.biLdcp1 ||
-		    data->bi.biLdcp2 || data->bi.biFenceStatus)
-			pr_info("  BI:  %016llx %016llx %016llx %016llx\n",
-				be64_to_cpu(data->bi.biLdcp0),
-				be64_to_cpu(data->bi.biLdcp1),
-				be64_to_cpu(data->bi.biLdcp2),
-				be64_to_cpu(data->bi.biFenceStatus));
-		break;
-	case OPAL_P7IOC_DIAG_TYPE_CI:
-		pr_info("P7IOC diag-data for CI Port %d\n\n",
-			data->ci.ciPort);
-		ioda_eeh_hub_diag_common(data);
-		if (data->ci.ciPortStatus || data->ci.ciPortLdcp)
-			pr_info("  CI:  %016llx %016llx\n",
-				be64_to_cpu(data->ci.ciPortStatus),
-				be64_to_cpu(data->ci.ciPortLdcp));
-		break;
-	case OPAL_P7IOC_DIAG_TYPE_MISC:
-		pr_info("P7IOC diag-data for MISC\n\n");
-		ioda_eeh_hub_diag_common(data);
-		break;
-	case OPAL_P7IOC_DIAG_TYPE_I2C:
-		pr_info("P7IOC diag-data for I2C\n\n");
-		ioda_eeh_hub_diag_common(data);
-		break;
-	default:
-		pr_warn("%s: Invalid type of HUB#%llx diag-data (%d)\n",
-			__func__, phb->hub_id, data->type);
-	}
-}
-
-static int ioda_eeh_get_pe(struct pci_controller *hose,
-			   u16 pe_no, struct eeh_pe **pe)
-{
-	struct pnv_phb *phb = hose->private_data;
-	struct pnv_ioda_pe *pnv_pe;
-	struct eeh_pe *dev_pe;
-	struct eeh_dev edev;
-
-	/*
-	 * If PHB supports compound PE, to fetch
-	 * the master PE because slave PE is invisible
-	 * to EEH core.
-	 */
-	pnv_pe = &phb->ioda.pe_array[pe_no];
-	if (pnv_pe->flags & PNV_IODA_PE_SLAVE) {
-		pnv_pe = pnv_pe->master;
-		WARN_ON(!pnv_pe ||
-			!(pnv_pe->flags & PNV_IODA_PE_MASTER));
-		pe_no = pnv_pe->pe_number;
-	}
-
-	/* Find the PE according to PE# */
-	memset(&edev, 0, sizeof(struct eeh_dev));
-	edev.phb = hose;
-	edev.pe_config_addr = pe_no;
-	dev_pe = eeh_pe_get(&edev);
-	if (!dev_pe)
-		return -EEXIST;
-
-	/* Freeze the (compound) PE */
-	*pe = dev_pe;
-	if (!(dev_pe->state & EEH_PE_ISOLATED))
-		phb->freeze_pe(phb, pe_no);
-
-	/*
-	 * At this point, we're sure the (compound) PE should
-	 * have been frozen. However, we still need poke until
-	 * hitting the frozen PE on top level.
-	 */
-	dev_pe = dev_pe->parent;
-	while (dev_pe && !(dev_pe->type & EEH_PE_PHB)) {
-		int ret;
-		int active_flags = (EEH_STATE_MMIO_ACTIVE |
-				    EEH_STATE_DMA_ACTIVE);
-
-		ret = eeh_ops->get_state(dev_pe, NULL);
-		if (ret <= 0 || (ret & active_flags) == active_flags) {
-			dev_pe = dev_pe->parent;
-			continue;
-		}
-
-		/* Frozen parent PE */
-		*pe = dev_pe;
-		if (!(dev_pe->state & EEH_PE_ISOLATED))
-			phb->freeze_pe(phb, dev_pe->addr);
-
-		/* Next one */
-		dev_pe = dev_pe->parent;
-	}
-
-	return 0;
-}
-
-/**
- * ioda_eeh_next_error - Retrieve next error for EEH core to handle
- * @pe: The affected PE
- *
- * The function is expected to be called by EEH core while it gets
- * special EEH event (without binding PE). The function calls to
- * OPAL APIs for next error to handle. The informational error is
- * handled internally by platform. However, the dead IOC, dead PHB,
- * fenced PHB and frozen PE should be handled by EEH core eventually.
- */
-static int ioda_eeh_next_error(struct eeh_pe **pe)
-{
-	struct pci_controller *hose;
-	struct pnv_phb *phb;
-	struct eeh_pe *phb_pe, *parent_pe;
-	__be64 frozen_pe_no;
-	__be16 err_type, severity;
-	int active_flags = (EEH_STATE_MMIO_ACTIVE | EEH_STATE_DMA_ACTIVE);
-	long rc;
-	int state, ret = EEH_NEXT_ERR_NONE;
-
-	/*
-	 * While running here, it's safe to purge the event queue.
-	 * And we should keep the cached OPAL notifier event sychronized
-	 * between the kernel and firmware.
-	 */
-	eeh_remove_event(NULL, false);
-	opal_notifier_update_evt(OPAL_EVENT_PCI_ERROR, 0x0ul);
-
-	list_for_each_entry(hose, &hose_list, list_node) {
-		/*
-		 * If the subordinate PCI buses of the PHB has been
-		 * removed or is exactly under error recovery, we
-		 * needn't take care of it any more.
-		 */
-		phb = hose->private_data;
-		phb_pe = eeh_phb_pe_get(hose);
-		if (!phb_pe || (phb_pe->state & EEH_PE_ISOLATED))
-			continue;
-
-		rc = opal_pci_next_error(phb->opal_id,
-				&frozen_pe_no, &err_type, &severity);
-
-		/* If OPAL API returns error, we needn't proceed */
-		if (rc != OPAL_SUCCESS) {
-			pr_devel("%s: Invalid return value on "
-				 "PHB#%x (0x%lx) from opal_pci_next_error",
-				 __func__, hose->global_number, rc);
-			continue;
-		}
-
-		/* If the PHB doesn't have error, stop processing */
-		if (be16_to_cpu(err_type) == OPAL_EEH_NO_ERROR ||
-		    be16_to_cpu(severity) == OPAL_EEH_SEV_NO_ERROR) {
-			pr_devel("%s: No error found on PHB#%x\n",
-				 __func__, hose->global_number);
-			continue;
-		}
-
-		/*
-		 * Processing the error. We're expecting the error with
-		 * highest priority reported upon multiple errors on the
-		 * specific PHB.
-		 */
-		pr_devel("%s: Error (%d, %d, %llu) on PHB#%x\n",
-			 __func__, be16_to_cpu(err_type), be16_to_cpu(severity),
-			 be64_to_cpu(frozen_pe_no), hose->global_number);
-		switch (be16_to_cpu(err_type)) {
-		case OPAL_EEH_IOC_ERROR:
-			if (be16_to_cpu(severity) == OPAL_EEH_SEV_IOC_DEAD) {
-				pr_err("EEH: dead IOC detected\n");
-				ret = EEH_NEXT_ERR_DEAD_IOC;
-			} else if (be16_to_cpu(severity) == OPAL_EEH_SEV_INF) {
-				pr_info("EEH: IOC informative error "
-					"detected\n");
-				ioda_eeh_hub_diag(hose);
-				ret = EEH_NEXT_ERR_NONE;
-			}
-
-			break;
-		case OPAL_EEH_PHB_ERROR:
-			if (be16_to_cpu(severity) == OPAL_EEH_SEV_PHB_DEAD) {
-				*pe = phb_pe;
-				pr_err("EEH: dead PHB#%x detected, "
-				       "location: %s\n",
-				       hose->global_number,
-				       eeh_pe_loc_get(phb_pe));
-				ret = EEH_NEXT_ERR_DEAD_PHB;
-			} else if (be16_to_cpu(severity) ==
-						OPAL_EEH_SEV_PHB_FENCED) {
-				*pe = phb_pe;
-				pr_err("EEH: Fenced PHB#%x detected, "
-				       "location: %s\n",
-				       hose->global_number,
-				       eeh_pe_loc_get(phb_pe));
-				ret = EEH_NEXT_ERR_FENCED_PHB;
-			} else if (be16_to_cpu(severity) == OPAL_EEH_SEV_INF) {
-				pr_info("EEH: PHB#%x informative error "
-					"detected, location: %s\n",
-					hose->global_number,
-					eeh_pe_loc_get(phb_pe));
-				ioda_eeh_phb_diag(phb_pe);
-				pnv_pci_dump_phb_diag_data(hose, phb_pe->data);
-				ret = EEH_NEXT_ERR_NONE;
-			}
-
-			break;
-		case OPAL_EEH_PE_ERROR:
-			/*
-			 * If we can't find the corresponding PE, we
-			 * just try to unfreeze.
-			 */
-			if (ioda_eeh_get_pe(hose,
-					    be64_to_cpu(frozen_pe_no), pe)) {
-				/* Try best to clear it */
-				pr_info("EEH: Clear non-existing PHB#%x-PE#%llx\n",
-					hose->global_number, frozen_pe_no);
-				pr_info("EEH: PHB location: %s\n",
-					eeh_pe_loc_get(phb_pe));
-				opal_pci_eeh_freeze_clear(phb->opal_id, frozen_pe_no,
-					OPAL_EEH_ACTION_CLEAR_FREEZE_ALL);
-				ret = EEH_NEXT_ERR_NONE;
-			} else if ((*pe)->state & EEH_PE_ISOLATED ||
-				   eeh_pe_passed(*pe)) {
-				ret = EEH_NEXT_ERR_NONE;
-			} else {
-				pr_err("EEH: Frozen PE#%x on PHB#%x detected\n",
-					(*pe)->addr, (*pe)->phb->global_number);
-				pr_err("EEH: PE location: %s, PHB location: %s\n",
-					eeh_pe_loc_get(*pe), eeh_pe_loc_get(phb_pe));
-				ret = EEH_NEXT_ERR_FROZEN_PE;
-			}
-
-			break;
-		default:
-			pr_warn("%s: Unexpected error type %d\n",
-				__func__, be16_to_cpu(err_type));
-		}
-
-		/*
-		 * EEH core will try recover from fenced PHB or
-		 * frozen PE. In the time for frozen PE, EEH core
-		 * enable IO path for that before collecting logs,
-		 * but it ruins the site. So we have to dump the
-		 * log in advance here.
-		 */
-		if ((ret == EEH_NEXT_ERR_FROZEN_PE  ||
-		    ret == EEH_NEXT_ERR_FENCED_PHB) &&
-		    !((*pe)->state & EEH_PE_ISOLATED)) {
-			eeh_pe_state_mark(*pe, EEH_PE_ISOLATED);
-			ioda_eeh_phb_diag(*pe);
-
-			if (eeh_has_flag(EEH_EARLY_DUMP_LOG))
-				pnv_pci_dump_phb_diag_data((*pe)->phb,
-							   (*pe)->data);
-		}
-
-		/*
-		 * We probably have the frozen parent PE out there and
-		 * we need have to handle frozen parent PE firstly.
-		 */
-		if (ret == EEH_NEXT_ERR_FROZEN_PE) {
-			parent_pe = (*pe)->parent;
-			while (parent_pe) {
-				/* Hit the ceiling ? */
-				if (parent_pe->type & EEH_PE_PHB)
-					break;
-
-				/* Frozen parent PE ? */
-				state = eeh_ops->get_state(parent_pe, NULL);
-				if (state > 0 &&
-				    (state & active_flags) != active_flags)
-					*pe = parent_pe;
-
-				/* Next parent level */
-				parent_pe = parent_pe->parent;
-			}
-
-			/* We possibly migrate to another PE */
-			eeh_pe_state_mark(*pe, EEH_PE_ISOLATED);
-		}
-
-		/*
-		 * If we have no errors on the specific PHB or only
-		 * informative error there, we continue poking it.
-		 * Otherwise, we need actions to be taken by upper
-		 * layer.
-		 */
-		if (ret > EEH_NEXT_ERR_INF)
-			break;
-	}
-
-	return ret;
-}
-
 struct pnv_eeh_ops ioda_eeh_ops = {
 	.reset			= ioda_eeh_reset,
-	.next_error		= ioda_eeh_next_error
 };
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index 127ef0c..e51ac2d 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -840,26 +840,346 @@ static int pnv_eeh_write_config(struct device_node *dn,
 	return pnv_pci_cfg_write(dn, where, size, val);
 }
 
+static void pnv_eeh_dump_hub_diag_common(struct OpalIoP7IOCErrorData *data)
+{
+	/* GEM */
+	if (data->gemXfir || data->gemRfir ||
+	    data->gemRirqfir || data->gemMask || data->gemRwof)
+		pr_info("  GEM: %016llx %016llx %016llx %016llx %016llx\n",
+			be64_to_cpu(data->gemXfir),
+			be64_to_cpu(data->gemRfir),
+			be64_to_cpu(data->gemRirqfir),
+			be64_to_cpu(data->gemMask),
+			be64_to_cpu(data->gemRwof));
+
+	/* LEM */
+	if (data->lemFir || data->lemErrMask ||
+	    data->lemAction0 || data->lemAction1 || data->lemWof)
+		pr_info("  LEM: %016llx %016llx %016llx %016llx %016llx\n",
+			be64_to_cpu(data->lemFir),
+			be64_to_cpu(data->lemErrMask),
+			be64_to_cpu(data->lemAction0),
+			be64_to_cpu(data->lemAction1),
+			be64_to_cpu(data->lemWof));
+}
+
+static void pnv_eeh_get_and_dump_hub_diag(struct pci_controller *hose)
+{
+	struct pnv_phb *phb = hose->private_data;
+	struct OpalIoP7IOCErrorData *data = &phb->diag.hub_diag;
+	long rc;
+
+	rc = opal_pci_get_hub_diag_data(phb->hub_id, data, sizeof(*data));
+	if (rc != OPAL_SUCCESS) {
+		pr_warn("%s: Failed to get HUB#%llx diag-data (%ld)\n",
+			__func__, phb->hub_id, rc);
+		return;
+	}
+
+	switch (data->type) {
+	case OPAL_P7IOC_DIAG_TYPE_RGC:
+		pr_info("P7IOC diag-data for RGC\n\n");
+		pnv_eeh_dump_hub_diag_common(data);
+		if (data->rgc.rgcStatus || data->rgc.rgcLdcp)
+			pr_info("  RGC: %016llx %016llx\n",
+				be64_to_cpu(data->rgc.rgcStatus),
+				be64_to_cpu(data->rgc.rgcLdcp));
+		break;
+	case OPAL_P7IOC_DIAG_TYPE_BI:
+		pr_info("P7IOC diag-data for BI %s\n\n",
+			data->bi.biDownbound ? "Downbound" : "Upbound");
+		pnv_eeh_dump_hub_diag_common(data);
+		if (data->bi.biLdcp0 || data->bi.biLdcp1 ||
+		    data->bi.biLdcp2 || data->bi.biFenceStatus)
+			pr_info("  BI:  %016llx %016llx %016llx %016llx\n",
+				be64_to_cpu(data->bi.biLdcp0),
+				be64_to_cpu(data->bi.biLdcp1),
+				be64_to_cpu(data->bi.biLdcp2),
+				be64_to_cpu(data->bi.biFenceStatus));
+		break;
+	case OPAL_P7IOC_DIAG_TYPE_CI:
+		pr_info("P7IOC diag-data for CI Port %d\n\n",
+			data->ci.ciPort);
+		pnv_eeh_dump_hub_diag_common(data);
+		if (data->ci.ciPortStatus || data->ci.ciPortLdcp)
+			pr_info("  CI:  %016llx %016llx\n",
+				be64_to_cpu(data->ci.ciPortStatus),
+				be64_to_cpu(data->ci.ciPortLdcp));
+		break;
+	case OPAL_P7IOC_DIAG_TYPE_MISC:
+		pr_info("P7IOC diag-data for MISC\n\n");
+		pnv_eeh_dump_hub_diag_common(data);
+		break;
+	case OPAL_P7IOC_DIAG_TYPE_I2C:
+		pr_info("P7IOC diag-data for I2C\n\n");
+		pnv_eeh_dump_hub_diag_common(data);
+		break;
+	default:
+		pr_warn("%s: Invalid type of HUB#%llx diag-data (%d)\n",
+			__func__, phb->hub_id, data->type);
+	}
+}
+
+static int pnv_eeh_get_pe(struct pci_controller *hose,
+			  u16 pe_no, struct eeh_pe **pe)
+{
+	struct pnv_phb *phb = hose->private_data;
+	struct pnv_ioda_pe *pnv_pe;
+	struct eeh_pe *dev_pe;
+	struct eeh_dev edev;
+
+	/*
+	 * If PHB supports compound PE, to fetch
+	 * the master PE because slave PE is invisible
+	 * to EEH core.
+	 */
+	pnv_pe = &phb->ioda.pe_array[pe_no];
+	if (pnv_pe->flags & PNV_IODA_PE_SLAVE) {
+		pnv_pe = pnv_pe->master;
+		WARN_ON(!pnv_pe ||
+			!(pnv_pe->flags & PNV_IODA_PE_MASTER));
+		pe_no = pnv_pe->pe_number;
+	}
+
+	/* Find the PE according to PE# */
+	memset(&edev, 0, sizeof(struct eeh_dev));
+	edev.phb = hose;
+	edev.pe_config_addr = pe_no;
+	dev_pe = eeh_pe_get(&edev);
+	if (!dev_pe)
+		return -EEXIST;
+
+	/* Freeze the (compound) PE */
+	*pe = dev_pe;
+	if (!(dev_pe->state & EEH_PE_ISOLATED))
+		phb->freeze_pe(phb, pe_no);
+
+	/*
+	 * At this point, we're sure the (compound) PE should
+	 * have been frozen. However, we still need poke until
+	 * hitting the frozen PE on top level.
+	 */
+	dev_pe = dev_pe->parent;
+	while (dev_pe && !(dev_pe->type & EEH_PE_PHB)) {
+		int ret;
+		int active_flags = (EEH_STATE_MMIO_ACTIVE |
+				    EEH_STATE_DMA_ACTIVE);
+
+		ret = eeh_ops->get_state(dev_pe, NULL);
+		if (ret <= 0 || (ret & active_flags) == active_flags) {
+			dev_pe = dev_pe->parent;
+			continue;
+		}
+
+		/* Frozen parent PE */
+		*pe = dev_pe;
+		if (!(dev_pe->state & EEH_PE_ISOLATED))
+			phb->freeze_pe(phb, dev_pe->addr);
+
+		/* Next one */
+		dev_pe = dev_pe->parent;
+	}
+
+	return 0;
+}
+
 /**
  * pnv_eeh_next_error - Retrieve next EEH error to handle
  * @pe: Affected PE
  *
- * Using OPAL API, to retrieve next EEH error for EEH core to handle
+ * The function is expected to be called by EEH core while it gets
+ * special EEH event (without binding PE). The function calls to
+ * OPAL APIs for next error to handle. The informational error is
+ * handled internally by platform. However, the dead IOC, dead PHB,
+ * fenced PHB and frozen PE should be handled by EEH core eventually.
  */
 static int pnv_eeh_next_error(struct eeh_pe **pe)
 {
 	struct pci_controller *hose;
-	struct pnv_phb *phb = NULL;
+	struct pnv_phb *phb;
+	struct eeh_pe *phb_pe, *parent_pe;
+	__be64 frozen_pe_no;
+	__be16 err_type, severity;
+	int active_flags = (EEH_STATE_MMIO_ACTIVE | EEH_STATE_DMA_ACTIVE);
+	long rc;
+	int state, ret = EEH_NEXT_ERR_NONE;
+
+	/*
+	 * While running here, it's safe to purge the event queue.
+	 * And we should keep the cached OPAL notifier event sychronized
+	 * between the kernel and firmware.
+	 */
+	eeh_remove_event(NULL, false);
+	opal_notifier_update_evt(OPAL_EVENT_PCI_ERROR, 0x0ul);
 
 	list_for_each_entry(hose, &hose_list, list_node) {
+		/*
+		 * If the subordinate PCI buses of the PHB has been
+		 * removed or is exactly under error recovery, we
+		 * needn't take care of it any more.
+		 */
 		phb = hose->private_data;
-		break;
-	}
+		phb_pe = eeh_phb_pe_get(hose);
+		if (!phb_pe || (phb_pe->state & EEH_PE_ISOLATED))
+			continue;
+
+		rc = opal_pci_next_error(phb->opal_id,
+					 &frozen_pe_no, &err_type, &severity);
+		if (rc != OPAL_SUCCESS) {
+			pr_devel("%s: Invalid return value on "
+				 "PHB#%x (0x%lx) from opal_pci_next_error",
+				 __func__, hose->global_number, rc);
+			continue;
+		}
+
+		/* If the PHB doesn't have error, stop processing */
+		if (be16_to_cpu(err_type) == OPAL_EEH_NO_ERROR ||
+		    be16_to_cpu(severity) == OPAL_EEH_SEV_NO_ERROR) {
+			pr_devel("%s: No error found on PHB#%x\n",
+				 __func__, hose->global_number);
+			continue;
+		}
 
-	if (phb && phb->eeh_ops->next_error)
-		return phb->eeh_ops->next_error(pe);
+		/*
+		 * Processing the error. We're expecting the error with
+		 * highest priority reported upon multiple errors on the
+		 * specific PHB.
+		 */
+		pr_devel("%s: Error (%d, %d, %llu) on PHB#%x\n",
+			__func__, be16_to_cpu(err_type),
+			be16_to_cpu(severity), be64_to_cpu(frozen_pe_no),
+			hose->global_number);
+		switch (be16_to_cpu(err_type)) {
+		case OPAL_EEH_IOC_ERROR:
+			if (be16_to_cpu(severity) == OPAL_EEH_SEV_IOC_DEAD) {
+				pr_err("EEH: dead IOC detected\n");
+				ret = EEH_NEXT_ERR_DEAD_IOC;
+			} else if (be16_to_cpu(severity) == OPAL_EEH_SEV_INF) {
+				pr_info("EEH: IOC informative error "
+					"detected\n");
+				pnv_eeh_get_and_dump_hub_diag(hose);
+				ret = EEH_NEXT_ERR_NONE;
+			}
+
+			break;
+		case OPAL_EEH_PHB_ERROR:
+			if (be16_to_cpu(severity) == OPAL_EEH_SEV_PHB_DEAD) {
+				*pe = phb_pe;
+				pr_err("EEH: dead PHB#%x detected, "
+				       "location: %s\n",
+					hose->global_number,
+					eeh_pe_loc_get(phb_pe));
+				ret = EEH_NEXT_ERR_DEAD_PHB;
+			} else if (be16_to_cpu(severity) ==
+				   OPAL_EEH_SEV_PHB_FENCED) {
+				*pe = phb_pe;
+				pr_err("EEH: Fenced PHB#%x detected, "
+				       "location: %s\n",
+					hose->global_number,
+					eeh_pe_loc_get(phb_pe));
+				ret = EEH_NEXT_ERR_FENCED_PHB;
+			} else if (be16_to_cpu(severity) == OPAL_EEH_SEV_INF) {
+				pr_info("EEH: PHB#%x informative error "
+					"detected, location: %s\n",
+					hose->global_number,
+					eeh_pe_loc_get(phb_pe));
+				pnv_eeh_get_phb_diag(phb_pe);
+				pnv_pci_dump_phb_diag_data(hose, phb_pe->data);
+				ret = EEH_NEXT_ERR_NONE;
+			}
 
-	return -EEXIST;
+			break;
+		case OPAL_EEH_PE_ERROR:
+			/*
+			 * If we can't find the corresponding PE, we
+			 * just try to unfreeze.
+			 */
+			if (pnv_eeh_get_pe(hose,
+				be64_to_cpu(frozen_pe_no), pe)) {
+				/* Try best to clear it */
+				pr_info("EEH: Clear non-existing PHB#%x-PE#%llx\n",
+					hose->global_number, frozen_pe_no);
+				pr_info("EEH: PHB location: %s\n",
+					eeh_pe_loc_get(phb_pe));
+				opal_pci_eeh_freeze_clear(phb->opal_id,
+					frozen_pe_no,
+					OPAL_EEH_ACTION_CLEAR_FREEZE_ALL);
+				ret = EEH_NEXT_ERR_NONE;
+			} else if ((*pe)->state & EEH_PE_ISOLATED ||
+				   eeh_pe_passed(*pe)) {
+				ret = EEH_NEXT_ERR_NONE;
+			} else {
+				pr_err("EEH: Frozen PE#%x "
+				       "on PHB#%x detected\n",
+				       (*pe)->addr,
+					(*pe)->phb->global_number);
+				pr_err("EEH: PE location: %s, "
+				       "PHB location: %s\n",
+				       eeh_pe_loc_get(*pe),
+				       eeh_pe_loc_get(phb_pe));
+				ret = EEH_NEXT_ERR_FROZEN_PE;
+			}
+
+			break;
+		default:
+			pr_warn("%s: Unexpected error type %d\n",
+				__func__, be16_to_cpu(err_type));
+		}
+
+		/*
+		 * EEH core will try recover from fenced PHB or
+		 * frozen PE. In the time for frozen PE, EEH core
+		 * enable IO path for that before collecting logs,
+		 * but it ruins the site. So we have to dump the
+		 * log in advance here.
+		 */
+		if ((ret == EEH_NEXT_ERR_FROZEN_PE  ||
+		    ret == EEH_NEXT_ERR_FENCED_PHB) &&
+		    !((*pe)->state & EEH_PE_ISOLATED)) {
+			eeh_pe_state_mark(*pe, EEH_PE_ISOLATED);
+			pnv_eeh_get_phb_diag(*pe);
+
+			if (eeh_has_flag(EEH_EARLY_DUMP_LOG))
+				pnv_pci_dump_phb_diag_data((*pe)->phb,
+							   (*pe)->data);
+		}
+
+		/*
+		 * We probably have the frozen parent PE out there and
+		 * we need have to handle frozen parent PE firstly.
+		 */
+		if (ret == EEH_NEXT_ERR_FROZEN_PE) {
+			parent_pe = (*pe)->parent;
+			while (parent_pe) {
+				/* Hit the ceiling ? */
+				if (parent_pe->type & EEH_PE_PHB)
+					break;
+
+				/* Frozen parent PE ? */
+				state = eeh_ops->get_state(parent_pe, NULL);
+				if (state > 0 &&
+				    (state & active_flags) != active_flags)
+					*pe = parent_pe;
+
+				/* Next parent level */
+				parent_pe = parent_pe->parent;
+			}
+
+			/* We possibly migrate to another PE */
+			eeh_pe_state_mark(*pe, EEH_PE_ISOLATED);
+		}
+
+		/*
+		 * If we have no errors on the specific PHB or only
+		 * informative error there, we continue poking it.
+		 * Otherwise, we need actions to be taken by upper
+		 * layer.
+		 */
+		if (ret > EEH_NEXT_ERR_INF)
+			break;
+	}
+
+	return ret;
 }
 
 static int pnv_eeh_restore_config(struct device_node *dn)
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index 773a026..5275d89 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -79,7 +79,6 @@ struct pnv_ioda_pe {
 #ifdef CONFIG_EEH
 struct pnv_eeh_ops {
 	int (*reset)(struct eeh_pe *pe, int option);
-	int (*next_error)(struct eeh_pe **pe);
 };
 #endif /* CONFIG_EEH */
 
-- 
1.8.3.2

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 09/10] powerpc/powernv: Drop PHB operation reset()
  2015-02-16  3:45 [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation Gavin Shan
                   ` (7 preceding siblings ...)
  2015-02-16  3:45 ` [PATCH 08/10] powerpc/powernv: Drop PHB operation next_error() Gavin Shan
@ 2015-02-16  3:45 ` Gavin Shan
  2015-02-16  3:45 ` [PATCH 10/10] powerpc/powernv: Remove unused file Gavin Shan
  9 siblings, 0 replies; 11+ messages in thread
From: Gavin Shan @ 2015-02-16  3:45 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Gavin Shan

The patch drops PHB EEH operation reset() and merges its logic to
eeh_ops::reset().

Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
---
 arch/powerpc/platforms/powernv/eeh-ioda.c    | 235 ---------------------------
 arch/powerpc/platforms/powernv/eeh-powernv.c | 225 ++++++++++++++++++++++++-
 arch/powerpc/platforms/powernv/pci-ioda.c    |   4 +-
 arch/powerpc/platforms/powernv/pci.h         |   3 +-
 4 files changed, 223 insertions(+), 244 deletions(-)

diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c
index 94d94b4..9fcfc45 100644
--- a/arch/powerpc/platforms/powernv/eeh-ioda.c
+++ b/arch/powerpc/platforms/powernv/eeh-ioda.c
@@ -34,240 +34,5 @@
 #include "powernv.h"
 #include "pci.h"
 
-static s64 ioda_eeh_phb_poll(struct pnv_phb *phb)
-{
-	s64 rc = OPAL_HARDWARE;
-
-	while (1) {
-		rc = opal_pci_poll(phb->opal_id);
-		if (rc <= 0)
-			break;
-
-		if (system_state < SYSTEM_RUNNING)
-			udelay(1000 * rc);
-		else
-			msleep(rc);
-	}
-
-	return rc;
-}
-
-int ioda_eeh_phb_reset(struct pci_controller *hose, int option)
-{
-	struct pnv_phb *phb = hose->private_data;
-	s64 rc = OPAL_HARDWARE;
-
-	pr_debug("%s: Reset PHB#%x, option=%d\n",
-		 __func__, hose->global_number, option);
-
-	/* Issue PHB complete reset request */
-	if (option == EEH_RESET_FUNDAMENTAL ||
-	    option == EEH_RESET_HOT)
-		rc = opal_pci_reset(phb->opal_id,
-				OPAL_RESET_PHB_COMPLETE,
-				OPAL_ASSERT_RESET);
-	else if (option == EEH_RESET_DEACTIVATE)
-		rc = opal_pci_reset(phb->opal_id,
-				OPAL_RESET_PHB_COMPLETE,
-				OPAL_DEASSERT_RESET);
-	if (rc < 0)
-		goto out;
-
-	/*
-	 * Poll state of the PHB until the request is done
-	 * successfully. The PHB reset is usually PHB complete
-	 * reset followed by hot reset on root bus. So we also
-	 * need the PCI bus settlement delay.
-	 */
-	rc = ioda_eeh_phb_poll(phb);
-	if (option == EEH_RESET_DEACTIVATE) {
-		if (system_state < SYSTEM_RUNNING)
-			udelay(1000 * EEH_PE_RST_SETTLE_TIME);
-		else
-			msleep(EEH_PE_RST_SETTLE_TIME);
-	}
-out:
-	if (rc != OPAL_SUCCESS)
-		return -EIO;
-
-	return 0;
-}
-
-static int ioda_eeh_root_reset(struct pci_controller *hose, int option)
-{
-	struct pnv_phb *phb = hose->private_data;
-	s64 rc = OPAL_SUCCESS;
-
-	pr_debug("%s: Reset PHB#%x, option=%d\n",
-		 __func__, hose->global_number, option);
-
-	/*
-	 * During the reset deassert time, we needn't care
-	 * the reset scope because the firmware does nothing
-	 * for fundamental or hot reset during deassert phase.
-	 */
-	if (option == EEH_RESET_FUNDAMENTAL)
-		rc = opal_pci_reset(phb->opal_id,
-				OPAL_RESET_PCI_FUNDAMENTAL,
-				OPAL_ASSERT_RESET);
-	else if (option == EEH_RESET_HOT)
-		rc = opal_pci_reset(phb->opal_id,
-				OPAL_RESET_PCI_HOT,
-				OPAL_ASSERT_RESET);
-	else if (option == EEH_RESET_DEACTIVATE)
-		rc = opal_pci_reset(phb->opal_id,
-				OPAL_RESET_PCI_HOT,
-				OPAL_DEASSERT_RESET);
-	if (rc < 0)
-		goto out;
-
-	/* Poll state of the PHB until the request is done */
-	rc = ioda_eeh_phb_poll(phb);
-	if (option == EEH_RESET_DEACTIVATE)
-		msleep(EEH_PE_RST_SETTLE_TIME);
-out:
-	if (rc != OPAL_SUCCESS)
-		return -EIO;
-
-	return 0;
-}
-
-static int ioda_eeh_bridge_reset(struct pci_dev *dev, int option)
-
-{
-	struct device_node *dn = pci_device_to_OF_node(dev);
-	struct eeh_dev *edev = of_node_to_eeh_dev(dn);
-	int aer = edev ? edev->aer_cap : 0;
-	u32 ctrl;
-
-	pr_debug("%s: Reset PCI bus %04x:%02x with option %d\n",
-		 __func__, pci_domain_nr(dev->bus),
-		 dev->bus->number, option);
-
-	switch (option) {
-	case EEH_RESET_FUNDAMENTAL:
-	case EEH_RESET_HOT:
-		/* Don't report linkDown event */
-		if (aer) {
-			eeh_ops->read_config(dn, aer + PCI_ERR_UNCOR_MASK,
-					     4, &ctrl);
-			ctrl |= PCI_ERR_UNC_SURPDN;
-                        eeh_ops->write_config(dn, aer + PCI_ERR_UNCOR_MASK,
-					      4, ctrl);
-                }
-
-		eeh_ops->read_config(dn, PCI_BRIDGE_CONTROL, 2, &ctrl);
-		ctrl |= PCI_BRIDGE_CTL_BUS_RESET;
-		eeh_ops->write_config(dn, PCI_BRIDGE_CONTROL, 2, ctrl);
-		msleep(EEH_PE_RST_HOLD_TIME);
-
-		break;
-	case EEH_RESET_DEACTIVATE:
-		eeh_ops->read_config(dn, PCI_BRIDGE_CONTROL, 2, &ctrl);
-		ctrl &= ~PCI_BRIDGE_CTL_BUS_RESET;
-		eeh_ops->write_config(dn, PCI_BRIDGE_CONTROL, 2, ctrl);
-		msleep(EEH_PE_RST_SETTLE_TIME);
-
-		/* Continue reporting linkDown event */
-		if (aer) {
-			eeh_ops->read_config(dn, aer + PCI_ERR_UNCOR_MASK,
-					     4, &ctrl);
-			ctrl &= ~PCI_ERR_UNC_SURPDN;
-			eeh_ops->write_config(dn, aer + PCI_ERR_UNCOR_MASK,
-					      4, ctrl);
-		}
-
-		break;
-	}
-
-	return 0;
-}
-
-void pnv_pci_reset_secondary_bus(struct pci_dev *dev)
-{
-	struct pci_controller *hose;
-
-	if (pci_is_root_bus(dev->bus)) {
-		hose = pci_bus_to_host(dev->bus);
-		ioda_eeh_root_reset(hose, EEH_RESET_HOT);
-		ioda_eeh_root_reset(hose, EEH_RESET_DEACTIVATE);
-	} else {
-		ioda_eeh_bridge_reset(dev, EEH_RESET_HOT);
-		ioda_eeh_bridge_reset(dev, EEH_RESET_DEACTIVATE);
-	}
-}
-
-/**
- * ioda_eeh_reset - Reset the indicated PE
- * @pe: EEH PE
- * @option: reset option
- *
- * Do reset on the indicated PE. For PCI bus sensitive PE,
- * we need to reset the parent p2p bridge. The PHB has to
- * be reinitialized if the p2p bridge is root bridge. For
- * PCI device sensitive PE, we will try to reset the device
- * through FLR. For now, we don't have OPAL APIs to do HARD
- * reset yet, so all reset would be SOFT (HOT) reset.
- */
-static int ioda_eeh_reset(struct eeh_pe *pe, int option)
-{
-	struct pci_controller *hose = pe->phb;
-	struct pci_bus *bus;
-	int ret;
-
-	/*
-	 * For PHB reset, we always have complete reset. For those PEs whose
-	 * primary bus derived from root complex (root bus) or root port
-	 * (usually bus#1), we apply hot or fundamental reset on the root port.
-	 * For other PEs, we always have hot reset on the PE primary bus.
-	 *
-	 * Here, we have different design to pHyp, which always clear the
-	 * frozen state during PE reset. However, the good idea here from
-	 * benh is to keep frozen state before we get PE reset done completely
-	 * (until BAR restore). With the frozen state, HW drops illegal IO
-	 * or MMIO access, which can incur recrusive frozen PE during PE
-	 * reset. The side effect is that EEH core has to clear the frozen
-	 * state explicitly after BAR restore.
-	 */
-	if (pe->type & EEH_PE_PHB) {
-		ret = ioda_eeh_phb_reset(hose, option);
-	} else {
-		struct pnv_phb *phb;
-		s64 rc;
-
-		/*
-		 * The frozen PE might be caused by PAPR error injection
-		 * registers, which are expected to be cleared after hitting
-		 * frozen PE as stated in the hardware spec. Unfortunately,
-		 * that's not true on P7IOC. So we have to clear it manually
-		 * to avoid recursive EEH errors during recovery.
-		 */
-		phb = hose->private_data;
-		if (phb->model == PNV_PHB_MODEL_P7IOC &&
-		    (option == EEH_RESET_HOT ||
-		    option == EEH_RESET_FUNDAMENTAL)) {
-			rc = opal_pci_reset(phb->opal_id,
-					    OPAL_RESET_PHB_ERROR,
-					    OPAL_ASSERT_RESET);
-			if (rc != OPAL_SUCCESS) {
-				pr_warn("%s: Failure %lld clearing "
-					"error injection registers\n",
-					__func__, rc);
-				return -EIO;
-			}
-		}
-
-		bus = eeh_pe_bus_get(pe);
-		if (pci_is_root_bus(bus) ||
-		    pci_is_root_bus(bus->parent))
-			ret = ioda_eeh_root_reset(hose, option);
-		else
-			ret = ioda_eeh_bridge_reset(bus->self, option);
-	}
-
-	return ret;
-}
-
 struct pnv_eeh_ops ioda_eeh_ops = {
-	.reset			= ioda_eeh_reset,
 };
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index e51ac2d..ede6906 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -665,21 +665,236 @@ static int pnv_eeh_get_state(struct eeh_pe *pe, int *delay)
 	return ret;
 }
 
+static s64 pnv_eeh_phb_poll(struct pnv_phb *phb)
+{
+	s64 rc = OPAL_HARDWARE;
+
+	while (1) {
+		rc = opal_pci_poll(phb->opal_id);
+		if (rc <= 0)
+			break;
+
+		if (system_state < SYSTEM_RUNNING)
+			udelay(1000 * rc);
+		else
+			msleep(rc);
+	}
+
+	return rc;
+}
+
+int pnv_eeh_phb_reset(struct pci_controller *hose, int option)
+{
+	struct pnv_phb *phb = hose->private_data;
+	s64 rc = OPAL_HARDWARE;
+
+	pr_debug("%s: Reset PHB#%x, option=%d\n",
+		 __func__, hose->global_number, option);
+
+	/* Issue PHB complete reset request */
+	if (option == EEH_RESET_FUNDAMENTAL ||
+	    option == EEH_RESET_HOT)
+		rc = opal_pci_reset(phb->opal_id,
+				    OPAL_RESET_PHB_COMPLETE,
+				    OPAL_ASSERT_RESET);
+	else if (option == EEH_RESET_DEACTIVATE)
+		rc = opal_pci_reset(phb->opal_id,
+				    OPAL_RESET_PHB_COMPLETE,
+				    OPAL_DEASSERT_RESET);
+	if (rc < 0)
+		goto out;
+
+	/*
+	 * Poll state of the PHB until the request is done
+	 * successfully. The PHB reset is usually PHB complete
+	 * reset followed by hot reset on root bus. So we also
+	 * need the PCI bus settlement delay.
+	 */
+	rc = pnv_eeh_phb_poll(phb);
+	if (option == EEH_RESET_DEACTIVATE) {
+		if (system_state < SYSTEM_RUNNING)
+			udelay(1000 * EEH_PE_RST_SETTLE_TIME);
+		else
+			msleep(EEH_PE_RST_SETTLE_TIME);
+	}
+out:
+	if (rc != OPAL_SUCCESS)
+		return -EIO;
+
+	return 0;
+}
+
+static int pnv_eeh_root_reset(struct pci_controller *hose, int option)
+{
+	struct pnv_phb *phb = hose->private_data;
+	s64 rc = OPAL_HARDWARE;
+
+	pr_debug("%s: Reset PHB#%x, option=%d\n",
+		 __func__, hose->global_number, option);
+
+	/*
+	 * During the reset deassert time, we needn't care
+	 * the reset scope because the firmware does nothing
+	 * for fundamental or hot reset during deassert phase.
+	 */
+	if (option == EEH_RESET_FUNDAMENTAL)
+		rc = opal_pci_reset(phb->opal_id,
+				    OPAL_RESET_PCI_FUNDAMENTAL,
+				    OPAL_ASSERT_RESET);
+	else if (option == EEH_RESET_HOT)
+		rc = opal_pci_reset(phb->opal_id,
+				    OPAL_RESET_PCI_HOT,
+				    OPAL_ASSERT_RESET);
+	else if (option == EEH_RESET_DEACTIVATE)
+		rc = opal_pci_reset(phb->opal_id,
+				    OPAL_RESET_PCI_HOT,
+				    OPAL_DEASSERT_RESET);
+	if (rc < 0)
+		goto out;
+
+	/* Poll state of the PHB until the request is done */
+	rc = pnv_eeh_phb_poll(phb);
+	if (option == EEH_RESET_DEACTIVATE)
+		msleep(EEH_PE_RST_SETTLE_TIME);
+out:
+	if (rc != OPAL_SUCCESS)
+		return -EIO;
+
+	return 0;
+}
+
+static int pnv_eeh_bridge_reset(struct pci_dev *dev, int option)
+{
+	struct device_node *dn = pci_device_to_OF_node(dev);
+	struct eeh_dev *edev = of_node_to_eeh_dev(dn);
+	int aer = edev ? edev->aer_cap : 0;
+	u32 ctrl;
+
+	pr_debug("%s: Reset PCI bus %04x:%02x with option %d\n",
+		 __func__, pci_domain_nr(dev->bus),
+		 dev->bus->number, option);
+
+	switch (option) {
+	case EEH_RESET_FUNDAMENTAL:
+	case EEH_RESET_HOT:
+		/* Don't report linkDown event */
+		if (aer) {
+			eeh_ops->read_config(dn, aer + PCI_ERR_UNCOR_MASK,
+					     4, &ctrl);
+			ctrl |= PCI_ERR_UNC_SURPDN;
+			eeh_ops->write_config(dn, aer + PCI_ERR_UNCOR_MASK,
+					      4, ctrl);
+		}
+
+		eeh_ops->read_config(dn, PCI_BRIDGE_CONTROL, 2, &ctrl);
+		ctrl |= PCI_BRIDGE_CTL_BUS_RESET;
+		eeh_ops->write_config(dn, PCI_BRIDGE_CONTROL, 2, ctrl);
+
+		msleep(EEH_PE_RST_HOLD_TIME);
+		break;
+	case EEH_RESET_DEACTIVATE:
+		eeh_ops->read_config(dn, PCI_BRIDGE_CONTROL, 2, &ctrl);
+		ctrl &= ~PCI_BRIDGE_CTL_BUS_RESET;
+		eeh_ops->write_config(dn, PCI_BRIDGE_CONTROL, 2, ctrl);
+
+		msleep(EEH_PE_RST_SETTLE_TIME);
+
+		/* Continue reporting linkDown event */
+		if (aer) {
+			eeh_ops->read_config(dn, aer + PCI_ERR_UNCOR_MASK,
+					     4, &ctrl);
+			ctrl &= ~PCI_ERR_UNC_SURPDN;
+			eeh_ops->write_config(dn, aer + PCI_ERR_UNCOR_MASK,
+					      4, ctrl);
+		}
+
+		break;
+	}
+
+	return 0;
+}
+
+void pnv_pci_reset_secondary_bus(struct pci_dev *dev)
+{
+	struct pci_controller *hose;
+
+	if (pci_is_root_bus(dev->bus)) {
+		hose = pci_bus_to_host(dev->bus);
+		pnv_eeh_root_reset(hose, EEH_RESET_HOT);
+		pnv_eeh_root_reset(hose, EEH_RESET_DEACTIVATE);
+	} else {
+		pnv_eeh_bridge_reset(dev, EEH_RESET_HOT);
+		pnv_eeh_bridge_reset(dev, EEH_RESET_DEACTIVATE);
+	}
+}
+
 /**
  * pnv_eeh_reset - Reset the specified PE
  * @pe: EEH PE
  * @option: reset option
  *
- * Reset the specified PE
+ * Do reset on the indicated PE. For PCI bus sensitive PE,
+ * we need to reset the parent p2p bridge. The PHB has to
+ * be reinitialized if the p2p bridge is root bridge. For
+ * PCI device sensitive PE, we will try to reset the device
+ * through FLR. For now, we don't have OPAL APIs to do HARD
+ * reset yet, so all reset would be SOFT (HOT) reset.
  */
 static int pnv_eeh_reset(struct eeh_pe *pe, int option)
 {
 	struct pci_controller *hose = pe->phb;
-	struct pnv_phb *phb = hose->private_data;
-	int ret = -EEXIST;
+	struct pci_bus *bus;
+	int ret;
+
+	/*
+	 * For PHB reset, we always have complete reset. For those PEs whose
+	 * primary bus derived from root complex (root bus) or root port
+	 * (usually bus#1), we apply hot or fundamental reset on the root port.
+	 * For other PEs, we always have hot reset on the PE primary bus.
+	 *
+	 * Here, we have different design to pHyp, which always clear the
+	 * frozen state during PE reset. However, the good idea here from
+	 * benh is to keep frozen state before we get PE reset done completely
+	 * (until BAR restore). With the frozen state, HW drops illegal IO
+	 * or MMIO access, which can incur recrusive frozen PE during PE
+	 * reset. The side effect is that EEH core has to clear the frozen
+	 * state explicitly after BAR restore.
+	 */
+	if (pe->type & EEH_PE_PHB) {
+		ret = pnv_eeh_phb_reset(hose, option);
+	} else {
+		struct pnv_phb *phb;
+		s64 rc;
 
-	if (phb->eeh_ops && phb->eeh_ops->reset)
-		ret = phb->eeh_ops->reset(pe, option);
+		/*
+		 * The frozen PE might be caused by PAPR error injection
+		 * registers, which are expected to be cleared after hitting
+		 * frozen PE as stated in the hardware spec. Unfortunately,
+		 * that's not true on P7IOC. So we have to clear it manually
+		 * to avoid recursive EEH errors during recovery.
+		 */
+		phb = hose->private_data;
+		if (phb->model == PNV_PHB_MODEL_P7IOC &&
+		    (option == EEH_RESET_HOT ||
+		    option == EEH_RESET_FUNDAMENTAL)) {
+			rc = opal_pci_reset(phb->opal_id,
+					    OPAL_RESET_PHB_ERROR,
+					    OPAL_ASSERT_RESET);
+			if (rc != OPAL_SUCCESS) {
+				pr_warn("%s: Failure %lld clearing "
+					"error injection registers\n",
+					__func__, rc);
+				return -EIO;
+			}
+		}
+
+		bus = eeh_pe_bus_get(pe);
+		if (pci_is_root_bus(bus) ||
+			pci_is_root_bus(bus->parent))
+			ret = pnv_eeh_root_reset(hose, option);
+		else
+			ret = pnv_eeh_bridge_reset(bus->self, option);
+	}
 
 	return ret;
 }
diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c
index 6c9ff2b..4e21596 100644
--- a/arch/powerpc/platforms/powernv/pci-ioda.c
+++ b/arch/powerpc/platforms/powernv/pci-ioda.c
@@ -2121,8 +2121,8 @@ static void __init pnv_pci_init_ioda_phb(struct device_node *np,
 	 */
 	if (is_kdump_kernel()) {
 		pr_info("  Issue PHB reset ...\n");
-		ioda_eeh_phb_reset(hose, EEH_RESET_FUNDAMENTAL);
-		ioda_eeh_phb_reset(hose, EEH_RESET_DEACTIVATE);
+		pnv_eeh_phb_reset(hose, EEH_RESET_FUNDAMENTAL);
+		pnv_eeh_phb_reset(hose, EEH_RESET_DEACTIVATE);
 	}
 
 	/* Remove M64 resource if we can't configure it successfully */
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index 5275d89..39f2ca3 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -78,7 +78,6 @@ struct pnv_ioda_pe {
 /* IOC dependent EEH operations */
 #ifdef CONFIG_EEH
 struct pnv_eeh_ops {
-	int (*reset)(struct eeh_pe *pe, int option);
 };
 #endif /* CONFIG_EEH */
 
@@ -223,6 +222,6 @@ extern void pnv_pci_init_ioda2_phb(struct device_node *np);
 extern void pnv_pci_ioda_tce_invalidate(struct iommu_table *tbl,
 					__be64 *startp, __be64 *endp, bool rm);
 extern void pnv_pci_reset_secondary_bus(struct pci_dev *dev);
-extern int ioda_eeh_phb_reset(struct pci_controller *hose, int option);
+extern int pnv_eeh_phb_reset(struct pci_controller *hose, int option);
 
 #endif /* __POWERNV_PCI_H */
-- 
1.8.3.2

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH 10/10] powerpc/powernv: Remove unused file
  2015-02-16  3:45 [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation Gavin Shan
                   ` (8 preceding siblings ...)
  2015-02-16  3:45 ` [PATCH 09/10] powerpc/powernv: Drop PHB operation reset() Gavin Shan
@ 2015-02-16  3:45 ` Gavin Shan
  9 siblings, 0 replies; 11+ messages in thread
From: Gavin Shan @ 2015-02-16  3:45 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Gavin Shan

The patch removes unused file eeh-ioda.c and updates makefile
accordingly. Besides, the definition of "struct pnv_eeh_ops" and
the instances are all removed. Until now, the chip layer of EEH
implementation for PowerNV platform is removed completely.

Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
---
 arch/powerpc/platforms/powernv/Makefile   |  2 +-
 arch/powerpc/platforms/powernv/eeh-ioda.c | 38 -------------------------------
 arch/powerpc/platforms/powernv/pci-ioda.c |  3 ---
 arch/powerpc/platforms/powernv/pci.h      | 13 -----------
 4 files changed, 1 insertion(+), 55 deletions(-)
 delete mode 100644 arch/powerpc/platforms/powernv/eeh-ioda.c

diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms/powernv/Makefile
index 6f3c5d3..33e44f3 100644
--- a/arch/powerpc/platforms/powernv/Makefile
+++ b/arch/powerpc/platforms/powernv/Makefile
@@ -5,7 +5,7 @@ obj-y			+= opal-msglog.o opal-hmi.o opal-power.o
 
 obj-$(CONFIG_SMP)	+= smp.o subcore.o subcore-asm.o
 obj-$(CONFIG_PCI)	+= pci.o pci-p5ioc2.o pci-ioda.o
-obj-$(CONFIG_EEH)	+= eeh-ioda.o eeh-powernv.o
+obj-$(CONFIG_EEH)	+= eeh-powernv.o
 obj-$(CONFIG_PPC_SCOM)	+= opal-xscom.o
 obj-$(CONFIG_MEMORY_FAILURE)	+= opal-memory-errors.o
 obj-$(CONFIG_TRACEPOINTS)	+= opal-tracepoints.o
diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c
deleted file mode 100644
index 9fcfc45..0000000
--- a/arch/powerpc/platforms/powernv/eeh-ioda.c
+++ /dev/null
@@ -1,38 +0,0 @@
-/*
- * The file intends to implement the functions needed by EEH, which is
- * built on IODA compliant chip. Actually, lots of functions related
- * to EEH would be built based on the OPAL APIs.
- *
- * Copyright Benjamin Herrenschmidt & Gavin Shan, IBM Corporation 2013.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
-
-#include <linux/debugfs.h>
-#include <linux/delay.h>
-#include <linux/io.h>
-#include <linux/irq.h>
-#include <linux/kernel.h>
-#include <linux/msi.h>
-#include <linux/notifier.h>
-#include <linux/pci.h>
-#include <linux/string.h>
-
-#include <asm/eeh.h>
-#include <asm/eeh_event.h>
-#include <asm/io.h>
-#include <asm/iommu.h>
-#include <asm/msi_bitmap.h>
-#include <asm/opal.h>
-#include <asm/pci-bridge.h>
-#include <asm/ppc-pci.h>
-#include <asm/tce.h>
-
-#include "powernv.h"
-#include "pci.h"
-
-struct pnv_eeh_ops ioda_eeh_ops = {
-};
diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c
index 4e21596..26fe099 100644
--- a/arch/powerpc/platforms/powernv/pci-ioda.c
+++ b/arch/powerpc/platforms/powernv/pci-ioda.c
@@ -2078,9 +2078,6 @@ static void __init pnv_pci_init_ioda_phb(struct device_node *np,
 	phb->get_pe_state = pnv_ioda_get_pe_state;
 	phb->freeze_pe = pnv_ioda_freeze_pe;
 	phb->unfreeze_pe = pnv_ioda_unfreeze_pe;
-#ifdef CONFIG_EEH
-	phb->eeh_ops = &ioda_eeh_ops;
-#endif
 
 	/* Setup RID -> PE mapping function */
 	phb->bdfn_to_pe = pnv_ioda_bdfn_to_pe;
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index 39f2ca3..18ae927 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -75,12 +75,6 @@ struct pnv_ioda_pe {
 	struct list_head	list;
 };
 
-/* IOC dependent EEH operations */
-#ifdef CONFIG_EEH
-struct pnv_eeh_ops {
-};
-#endif /* CONFIG_EEH */
-
 #define PNV_PHB_FLAG_EEH	(1 << 0)
 
 struct pnv_phb {
@@ -94,10 +88,6 @@ struct pnv_phb {
 	int			initialized;
 	spinlock_t		lock;
 
-#ifdef CONFIG_EEH
-	struct pnv_eeh_ops	*eeh_ops;
-#endif
-
 #ifdef CONFIG_DEBUG_FS
 	int			has_dbgfs;
 	struct dentry		*dbgfs;
@@ -203,9 +193,6 @@ struct pnv_phb {
 };
 
 extern struct pci_ops pnv_pci_ops;
-#ifdef CONFIG_EEH
-extern struct pnv_eeh_ops ioda_eeh_ops;
-#endif
 
 void pnv_pci_dump_phb_diag_data(struct pci_controller *hose,
 				unsigned char *log_buff);
-- 
1.8.3.2

^ permalink raw reply related	[flat|nested] 11+ messages in thread

end of thread, other threads:[~2015-02-16  3:47 UTC | newest]

Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-02-16  3:45 [PATCH 00/10] powerpc/powernv: Simplify EEH Implementation Gavin Shan
2015-02-16  3:45 ` [PATCH 01/10] powerpc/powernv: Shorten EEH function names Gavin Shan
2015-02-16  3:45 ` [PATCH 02/10] powerpc/powernv: Drop PHB operation err_inject() Gavin Shan
2015-02-16  3:45 ` [PATCH 03/10] powerpc/powernv: Drop PHB operation post_init() Gavin Shan
2015-02-16  3:45 ` [PATCH 04/10] powerpc/powernv: Drop PHB operation get_log() Gavin Shan
2015-02-16  3:45 ` [PATCH 05/10] powerpc/powernv: Drop PHB operation configure_bridge() Gavin Shan
2015-02-16  3:45 ` [PATCH 06/10] powerpc/powernv: Drop PHB operation set_option() Gavin Shan
2015-02-16  3:45 ` [PATCH 07/10] powerpc/powernv: Drop PHB operation get_state() Gavin Shan
2015-02-16  3:45 ` [PATCH 08/10] powerpc/powernv: Drop PHB operation next_error() Gavin Shan
2015-02-16  3:45 ` [PATCH 09/10] powerpc/powernv: Drop PHB operation reset() Gavin Shan
2015-02-16  3:45 ` [PATCH 10/10] powerpc/powernv: Remove unused file Gavin Shan

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).