[PATCH v2 1/7] powerpc/eeh: Use debugfs_create_u32 for eeh_max_freezes

---
v2: Replaced uint32_t with u32.
---
 arch/powerpc/include/asm/eeh.h |  2 +-
 arch/powerpc/kernel/eeh.c      | 21 +++------------------
 2 files changed, 4 insertions(+), 19 deletions(-)

diff --git a/arch/powerpc/include/asm/eeh.h b/arch/powerpc/include/asm/eeh.h
index 8b596d096ebe..2daecd677855 100644
--- a/arch/powerpc/include/asm/eeh.h
+++ b/arch/powerpc/include/asm/eeh.h

@@ -219,7 +219,7 @@ struct eeh_ops {
 };
 
 extern int eeh_subsystem_flags;
-extern int eeh_max_freezes;
+extern u32 eeh_max_freezes;
 extern struct eeh_ops *eeh_ops;
 extern raw_spinlock_t confirm_error_lock;
 

diff --git a/arch/powerpc/kernel/eeh.c b/arch/powerpc/kernel/eeh.c
index ae05203eb4de..d021f5abeec5 100644
--- a/arch/powerpc/kernel/eeh.c
+++ b/arch/powerpc/kernel/eeh.c

@@ -109,7 +109,7 @@ EXPORT_SYMBOL(eeh_subsystem_flags);
  * frozen count in last hour exceeds this limit, the PE will
  * be forced to be offline permanently.
  */
-int eeh_max_freezes = 5;
+u32 eeh_max_freezes = 5;
 
 /* Platform dependent EEH operations */
 struct eeh_ops *eeh_ops = NULL;

@@ -1796,22 +1796,8 @@ static int eeh_enable_dbgfs_get(void *data, u64 *val)
 	return 0;
 }
 
-static int eeh_freeze_dbgfs_set(void *data, u64 val)
-{
-	eeh_max_freezes = val;
-	return 0;
-}
-
-static int eeh_freeze_dbgfs_get(void *data, u64 *val)
-{
-	*val = eeh_max_freezes;
-	return 0;
-}
-
 DEFINE_DEBUGFS_ATTRIBUTE(eeh_enable_dbgfs_ops, eeh_enable_dbgfs_get,
 			 eeh_enable_dbgfs_set, "0x%llx\n");
-DEFINE_DEBUGFS_ATTRIBUTE(eeh_freeze_dbgfs_ops, eeh_freeze_dbgfs_get,
-			 eeh_freeze_dbgfs_set, "0x%llx\n");
 #endif
 
 static int __init eeh_init_proc(void)

@@ -1822,9 +1808,8 @@ static int __init eeh_init_proc(void)
 		debugfs_create_file_unsafe("eeh_enable", 0600,
 					   powerpc_debugfs_root, NULL,
 					   &eeh_enable_dbgfs_ops);
-		debugfs_create_file_unsafe("eeh_max_freezes", 0600,
-					   powerpc_debugfs_root, NULL,
-					   &eeh_freeze_dbgfs_ops);
+		debugfs_create_u32("eeh_max_freezes", 0600,
+				powerpc_debugfs_root, &eeh_max_freezes);
 #endif
 	}
 

-- 
2.20.1

---
 arch/powerpc/kernel/eeh_cache.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/kernel/eeh_cache.c b/arch/powerpc/kernel/eeh_cache.c
index 201943d54a6e..b2c320e0fcef 100644
--- a/arch/powerpc/kernel/eeh_cache.c
+++ b/arch/powerpc/kernel/eeh_cache.c

@@ -157,10 +157,8 @@ eeh_addr_cache_insert(struct pci_dev *dev, resource_size_t alo,
 	piar->pcidev = dev;
 	piar->flags = flags;
 
-#ifdef DEBUG
 	pr_debug("PIAR: insert range=[%pap:%pap] dev=%s\n",
 		 &alo, &ahi, pci_name(dev));
-#endif
 
 	rb_link_node(&piar->rb_node, parent, p);
 	rb_insert_color(&piar->rb_node, &pci_io_addr_cache_root.rb_root);

@@ -240,6 +238,8 @@ static inline void __eeh_addr_cache_rmv_dev(struct pci_dev *dev)
 		piar = rb_entry(n, struct pci_io_addr_range, rb_node);
 
 		if (piar->pcidev == dev) {
+			pr_debug("PIAR: remove range=[%pap:%pap] dev=%s\n",
+				 &piar->addr_lo, &piar->addr_hi, pci_name(dev));
 			rb_erase(n, &pci_io_addr_cache_root.rb_root);
 			kfree(piar);
 			goto restart;

-- 
2.20.1

---
v2: Added missing #endif
    Replaced while loop with a for
---
 arch/powerpc/include/asm/eeh.h  |  3 +++
 arch/powerpc/kernel/eeh.c       |  1 +
 arch/powerpc/kernel/eeh_cache.c | 30 ++++++++++++++++++++++++++----
 3 files changed, 30 insertions(+), 4 deletions(-)

diff --git a/arch/powerpc/include/asm/eeh.h b/arch/powerpc/include/asm/eeh.h
index 2daecd677855..478f199d5663 100644
--- a/arch/powerpc/include/asm/eeh.h
+++ b/arch/powerpc/include/asm/eeh.h

@@ -460,6 +460,9 @@ static inline void eeh_readsl(const volatile void __iomem *addr, void * buf,
 		eeh_check_failure(addr);
 }
 
+
+void eeh_cache_debugfs_init(void);
+
 #endif /* CONFIG_PPC64 */
 #endif /* __KERNEL__ */
 #endif /* _POWERPC_EEH_H */

diff --git a/arch/powerpc/kernel/eeh.c b/arch/powerpc/kernel/eeh.c
index d021f5abeec5..82d22c671c0e 100644
--- a/arch/powerpc/kernel/eeh.c
+++ b/arch/powerpc/kernel/eeh.c

@@ -1810,6 +1810,7 @@ static int __init eeh_init_proc(void)
 					   &eeh_enable_dbgfs_ops);
 		debugfs_create_u32("eeh_max_freezes", 0600,
 				powerpc_debugfs_root, &eeh_max_freezes);
+		eeh_cache_debugfs_init();
 #endif
 	}
 

diff --git a/arch/powerpc/kernel/eeh_cache.c b/arch/powerpc/kernel/eeh_cache.c
index b2c320e0fcef..5c5697cced41 100644
--- a/arch/powerpc/kernel/eeh_cache.c
+++ b/arch/powerpc/kernel/eeh_cache.c

@@ -26,6 +26,7 @@
 #include <linux/spinlock.h>
 #include <linux/atomic.h>
 #include <asm/pci-bridge.h>
+#include <asm/debugfs.h>
 #include <asm/ppc-pci.h>
 
 

@@ -298,9 +299,30 @@ void eeh_addr_cache_build(void)
 		eeh_addr_cache_insert_dev(dev);
 		eeh_sysfs_add_device(dev);
 	}
+}
 
-#ifdef DEBUG
-	/* Verify tree built up above, echo back the list of addrs. */
-	eeh_addr_cache_print(&pci_io_addr_cache_root);
-#endif
+static int eeh_addr_cache_show(struct seq_file *s, void *v)
+{
+	struct pci_io_addr_range *piar;
+	struct rb_node *n;
+
+	spin_lock(&pci_io_addr_cache_root.piar_lock);
+	for (n = rb_first(&pci_io_addr_cache_root.rb_root); n; n = rb_next(n)) {
+		piar = rb_entry(n, struct pci_io_addr_range, rb_node);
+
+		seq_printf(s, "%s addr range [%pap-%pap]: %s\n",
+		       (piar->flags & IORESOURCE_IO) ? "i/o" : "mem",
+		       &piar->addr_lo, &piar->addr_hi, pci_name(piar->pcidev));
+	}
+	spin_unlock(&pci_io_addr_cache_root.piar_lock);
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(eeh_addr_cache);
+
+void eeh_cache_debugfs_init(void)
+{
+	debugfs_create_file_unsafe("eeh_address_cache", 0400,
+			powerpc_debugfs_root, NULL,
+			&eeh_addr_cache_fops);
 }

-- 
2.20.1

---
 arch/powerpc/kernel/eeh_cache.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/eeh_cache.c b/arch/powerpc/kernel/eeh_cache.c
index 5c5697cced41..9c68f0837385 100644
--- a/arch/powerpc/kernel/eeh_cache.c
+++ b/arch/powerpc/kernel/eeh_cache.c

@@ -114,7 +114,7 @@ static void eeh_addr_cache_print(struct pci_io_addr_cache *cache)
 	while (n) {
 		struct pci_io_addr_range *piar;
 		piar = rb_entry(n, struct pci_io_addr_range, rb_node);
-		pr_debug("PCI: %s addr range %d [%pap-%pap]: %s\n",
+		pr_info("PCI: %s addr range %d [%pap-%pap]: %s\n",
 		       (piar->flags & IORESOURCE_IO) ? "i/o" : "mem", cnt,
 		       &piar->addr_lo, &piar->addr_hi, pci_name(piar->pcidev));
 		cnt++;

-- 
2.20.1

---
v2: Renamed pci_find_hose_for_domain() to
    pci_find_controller_for_domain()
---
 arch/powerpc/include/asm/pci-bridge.h |  2 ++
 arch/powerpc/kernel/pci-common.c      | 11 +++++++++++
 2 files changed, 13 insertions(+)

diff --git a/arch/powerpc/include/asm/pci-bridge.h b/arch/powerpc/include/asm/pci-bridge.h
index aee4fcc24990..feea9534a015 100644
--- a/arch/powerpc/include/asm/pci-bridge.h
+++ b/arch/powerpc/include/asm/pci-bridge.h

@@ -274,6 +274,8 @@ extern int pcibios_map_io_space(struct pci_bus *bus);
 extern struct pci_controller *pci_find_hose_for_OF_device(
 			struct device_node* node);
 
+extern struct pci_controller *pci_find_controller_for_domain(int domain_nr);
+
 /* Fill up host controller resources from the OF node */
 extern void pci_process_bridge_OF_ranges(struct pci_controller *hose,
 			struct device_node *dev, int primary);

diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c
index 88e4f69a09e5..21eeeabb3e0e 100644
--- a/arch/powerpc/kernel/pci-common.c
+++ b/arch/powerpc/kernel/pci-common.c

@@ -357,6 +357,17 @@ struct pci_controller* pci_find_hose_for_OF_device(struct device_node* node)
 	return NULL;
 }
 
+struct pci_controller *pci_find_controller_for_domain(int domain_nr)
+{
+	struct pci_controller *hose;
+
+	list_for_each_entry(hose, &hose_list, list_node)
+		if (hose->global_number == domain_nr)
+			return hose;
+
+	return NULL;
+}
+
 /*
  * Reads the interrupt pin to determine if interrupt is use by card.
  * If the interrupt is used, then gets the interrupt line from the

-- 
2.20.1

Currently when we detect an error we automatically invoke the EEH recovery
handler. This can be annoying when debugging EEH problems, or when working
on EEH itself so this patch adds a debugfs knob that will prevent a
recovery event from being queued up when an issue is detected.

Signed-off-by: Oliver O'Halloran <oohall@gmail.com>
---
 arch/powerpc/include/asm/eeh.h  |  1 +
 arch/powerpc/kernel/eeh.c       | 10 ++++++++++
 arch/powerpc/kernel/eeh_event.c |  9 +++++++++
 3 files changed, 20 insertions(+)

diff --git a/arch/powerpc/include/asm/eeh.h b/arch/powerpc/include/asm/eeh.h
index 478f199d5663..810e05273ad3 100644
--- a/arch/powerpc/include/asm/eeh.h
+++ b/arch/powerpc/include/asm/eeh.h

@@ -220,6 +220,7 @@ struct eeh_ops {
 
 extern int eeh_subsystem_flags;
 extern u32 eeh_max_freezes;
+extern bool eeh_debugfs_no_recover;
 extern struct eeh_ops *eeh_ops;
 extern raw_spinlock_t confirm_error_lock;
 

diff --git a/arch/powerpc/kernel/eeh.c b/arch/powerpc/kernel/eeh.c
index 82d22c671c0e..9f20099ce2d9 100644
--- a/arch/powerpc/kernel/eeh.c
+++ b/arch/powerpc/kernel/eeh.c

@@ -111,6 +111,13 @@ EXPORT_SYMBOL(eeh_subsystem_flags);
  */
 u32 eeh_max_freezes = 5;
 
+/*
+ * Controls whether a recovery event should be scheduled when an
+ * isolated device is discovered. This is only really useful for
+ * debugging problems with the EEH core.
+ */
+bool eeh_debugfs_no_recover;
+
 /* Platform dependent EEH operations */
 struct eeh_ops *eeh_ops = NULL;
 

@@ -1810,6 +1817,9 @@ static int __init eeh_init_proc(void)
 					   &eeh_enable_dbgfs_ops);
 		debugfs_create_u32("eeh_max_freezes", 0600,
 				powerpc_debugfs_root, &eeh_max_freezes);
+		debugfs_create_bool("eeh_disable_recovery", 0600,
+				powerpc_debugfs_root,
+				&eeh_debugfs_no_recover);
 		eeh_cache_debugfs_init();
 #endif
 	}

diff --git a/arch/powerpc/kernel/eeh_event.c b/arch/powerpc/kernel/eeh_event.c
index 227e57f980df..19837798bb1d 100644
--- a/arch/powerpc/kernel/eeh_event.c
+++ b/arch/powerpc/kernel/eeh_event.c

@@ -126,6 +126,15 @@ int eeh_send_failure_event(struct eeh_pe *pe)
 	unsigned long flags;
 	struct eeh_event *event;
 
+	/*
+	 * If we've manually supressed recovery events via debugfs
+	 * then just drop it on the floor.
+	 */
+	if (eeh_debugfs_no_recover) {
+		pr_err("EEH: Event dropped due to no_recover setting\n");
+		return 0;
+	}
+