[RFC v9 0/4] powerpc/hotplug: Update affinity for migrated CPUs

[RFC v9 0/4] powerpc/hotplug: Update affinity for migrated CPUs

[Michael Bringmann]

[Testing delayed due to internal SAN problems.]

The migration of LPARs across Power systems affects many attributes
including that of the associativity of CPUs.  The patches in this
set execute when a system is coming up fresh upon a migration target.
They are intended to,

* Recognize changes to the associativity of CPUs recorded in internal
  data structures when compared to the latest copies in the device tree.
* Generate calls to other code layers to reset the data structures
  related to associativity of the CPUs.
* Re-register the 'changed' entities into the target system.
  Re-registration of CPUs mostly entails acting as if they have been
  newly hot-added into the target system.

Signed-off-by: Michael Bringmann <mwb@linux.vnet.ibm.com>

Michael Bringmann (4):
  hotplug/cpu: Conditionally acquire/release DRC index
  hotplug/cpu: Add operation queuing function
  hotplug/dlpar/cpu: Provide CPU readd operation
  mobility/numa: Ensure numa update does not overlap
---
Changes in RFC:
  -- Restructure and rearrange content of patches to co-locate
     similar or related modifications
  -- Rename pseries_update_drconf_cpu to pseries_update_cpu
  -- Simplify code to update CPU nodes during mobility checks.
     Remove functions to generate extra HP_ELOG messages in favor
     of direct function calls to dlpar_cpu_readd_by_index.
  -- Revise code order in dlpar_cpu_readd_by_index() to present
     more appropriate error codes from underlying layers of the
     implementation.
  -- Add hotplug device lock around all property updates
  -- Schedule all CPU updates as workqueue operations
  -- Rebase to 4.17-rc5 kernel
  -- Various code cleanups and compaction

[RFC v9 1/4] hotplug/cpu: Conditionally acquire/release DRC index

[Michael Bringmann]

[Testing delayed due to internal SAN problems.]

powerpc/cpu: Modify dlpar_cpu_add and dlpar_cpu_remove to allow the
skipping of DRC index acquire or release operations during the CPU
add or remove operations.  This is intended to support subsequent
changes to provide a 'CPU readd' operation.

Signed-off-by: Michael Bringmann <mwb@linux.vnet.ibm.com>
---
Changes in RFC:
  -- Move new validity check added to pseries_smp_notifier
     to another patch
---
 arch/powerpc/platforms/pseries/hotplug-cpu.c |   68 +++++++++++++++-----------
 1 file changed, 39 insertions(+), 29 deletions(-)

diff --git a/arch/powerpc/platforms/pseries/hotplug-cpu.c b/arch/powerpc/platforms/pseries/hotplug-cpu.c
index 6ef77ca..3632db2 100644
--- a/arch/powerpc/platforms/pseries/hotplug-cpu.c
+++ b/arch/powerpc/platforms/pseries/hotplug-cpu.c
@@ -432,7 +432,7 @@ static bool valid_cpu_drc_index(struct device_node *parent, u32 drc_index)
 	return found;
 }
 
-static ssize_t dlpar_cpu_add(u32 drc_index)
+static ssize_t dlpar_cpu_add(u32 drc_index, bool acquire_drc)
 {
 	struct device_node *dn, *parent;
 	int rc, saved_rc;
@@ -457,19 +457,22 @@ static ssize_t dlpar_cpu_add(u32 drc_index)
 		return -EINVAL;
 	}
 
-	rc = dlpar_acquire_drc(drc_index);
-	if (rc) {
-		pr_warn("Failed to acquire DRC, rc: %d, drc index: %x\n",
-			rc, drc_index);
-		of_node_put(parent);
-		return -EINVAL;
+	if (acquire_drc) {
+		rc = dlpar_acquire_drc(drc_index);
+		if (rc) {
+			pr_warn("Failed to acquire DRC, rc: %d, drc index: %x\n",
+				rc, drc_index);
+			of_node_put(parent);
+			return -EINVAL;
+		}
 	}
 
 	dn = dlpar_configure_connector(cpu_to_be32(drc_index), parent);
 	if (!dn) {
 		pr_warn("Failed call to configure-connector, drc index: %x\n",
 			drc_index);
-		dlpar_release_drc(drc_index);
+		if (acquire_drc)
+			dlpar_release_drc(drc_index);
 		of_node_put(parent);
 		return -EINVAL;
 	}
@@ -484,8 +487,9 @@ static ssize_t dlpar_cpu_add(u32 drc_index)
 		pr_warn("Failed to attach node %s, rc: %d, drc index: %x\n",
 			dn->name, rc, drc_index);
 
-		rc = dlpar_release_drc(drc_index);
-		if (!rc)
+		if (acquire_drc)
+			rc = dlpar_release_drc(drc_index);
+		if (!rc || acquire_drc)
 			dlpar_free_cc_nodes(dn);
 
 		return saved_rc;
@@ -498,7 +502,7 @@ static ssize_t dlpar_cpu_add(u32 drc_index)
 			dn->name, rc, drc_index);
 
 		rc = dlpar_detach_node(dn);
-		if (!rc)
+		if (!rc && acquire_drc)
 			dlpar_release_drc(drc_index);
 
 		return saved_rc;
@@ -566,7 +570,8 @@ static int dlpar_offline_cpu(struct device_node *dn)
 
 }
 
-static ssize_t dlpar_cpu_remove(struct device_node *dn, u32 drc_index)
+static ssize_t dlpar_cpu_remove(struct device_node *dn, u32 drc_index,
+				bool release_drc)
 {
 	int rc;
 
@@ -579,12 +584,14 @@ static ssize_t dlpar_cpu_remove(struct device_node *dn, u32 drc_index)
 		return -EINVAL;
 	}
 
-	rc = dlpar_release_drc(drc_index);
-	if (rc) {
-		pr_warn("Failed to release drc (%x) for CPU %s, rc: %d\n",
-			drc_index, dn->name, rc);
-		dlpar_online_cpu(dn);
-		return rc;
+	if (release_drc) {
+		rc = dlpar_release_drc(drc_index);
+		if (rc) {
+			pr_warn("Failed to release drc (%x) for CPU %s, rc: %d\n",
+				drc_index, dn->name, rc);
+			dlpar_online_cpu(dn);
+			return rc;
+		}
 	}
 
 	rc = dlpar_detach_node(dn);
@@ -593,7 +600,10 @@ static ssize_t dlpar_cpu_remove(struct device_node *dn, u32 drc_index)
 
 		pr_warn("Failed to detach CPU %s, rc: %d", dn->name, rc);
 
-		rc = dlpar_acquire_drc(drc_index);
+		if (release_drc)
+			rc = dlpar_acquire_drc(drc_index);
+		else
+			rc = 0;
 		if (!rc)
 			dlpar_online_cpu(dn);
 
@@ -622,7 +632,7 @@ static struct device_node *cpu_drc_index_to_dn(u32 drc_index)
 	return dn;
 }
 
-static int dlpar_cpu_remove_by_index(u32 drc_index)
+static int dlpar_cpu_remove_by_index(u32 drc_index, bool release_drc)
 {
 	struct device_node *dn;
 	int rc;
@@ -634,7 +644,7 @@ static int dlpar_cpu_remove_by_index(u32 drc_index)
 		return -ENODEV;
 	}
 
-	rc = dlpar_cpu_remove(dn, drc_index);
+	rc = dlpar_cpu_remove(dn, drc_index, release_drc);
 	of_node_put(dn);
 	return rc;
 }
@@ -699,7 +709,7 @@ static int dlpar_cpu_remove_by_count(u32 cpus_to_remove)
 	}
 
 	for (i = 0; i < cpus_to_remove; i++) {
-		rc = dlpar_cpu_remove_by_index(cpu_drcs[i]);
+		rc = dlpar_cpu_remove_by_index(cpu_drcs[i], true);
 		if (rc)
 			break;
 
@@ -710,7 +720,7 @@ static int dlpar_cpu_remove_by_count(u32 cpus_to_remove)
 		pr_warn("CPU hot-remove failed, adding back removed CPUs\n");
 
 		for (i = 0; i < cpus_removed; i++)
-			dlpar_cpu_add(cpu_drcs[i]);
+			dlpar_cpu_add(cpu_drcs[i], true);
 
 		rc = -EINVAL;
 	} else {
@@ -780,7 +790,7 @@ static int dlpar_cpu_add_by_count(u32 cpus_to_add)
 	}
 
 	for (i = 0; i < cpus_to_add; i++) {
-		rc = dlpar_cpu_add(cpu_drcs[i]);
+		rc = dlpar_cpu_add(cpu_drcs[i], true);
 		if (rc)
 			break;
 
@@ -791,7 +801,7 @@ static int dlpar_cpu_add_by_count(u32 cpus_to_add)
 		pr_warn("CPU hot-add failed, removing any added CPUs\n");
 
 		for (i = 0; i < cpus_added; i++)
-			dlpar_cpu_remove_by_index(cpu_drcs[i]);
+			dlpar_cpu_remove_by_index(cpu_drcs[i], true);
 
 		rc = -EINVAL;
 	} else {
@@ -817,7 +827,7 @@ int dlpar_cpu(struct pseries_hp_errorlog *hp_elog)
 		if (hp_elog->id_type == PSERIES_HP_ELOG_ID_DRC_COUNT)
 			rc = dlpar_cpu_remove_by_count(count);
 		else if (hp_elog->id_type == PSERIES_HP_ELOG_ID_DRC_INDEX)
-			rc = dlpar_cpu_remove_by_index(drc_index);
+			rc = dlpar_cpu_remove_by_index(drc_index, true);
 		else
 			rc = -EINVAL;
 		break;
@@ -825,7 +835,7 @@ int dlpar_cpu(struct pseries_hp_errorlog *hp_elog)
 		if (hp_elog->id_type == PSERIES_HP_ELOG_ID_DRC_COUNT)
 			rc = dlpar_cpu_add_by_count(count);
 		else if (hp_elog->id_type == PSERIES_HP_ELOG_ID_DRC_INDEX)
-			rc = dlpar_cpu_add(drc_index);
+			rc = dlpar_cpu_add(drc_index, true);
 		else
 			rc = -EINVAL;
 		break;
@@ -850,7 +860,7 @@ static ssize_t dlpar_cpu_probe(const char *buf, size_t count)
 	if (rc)
 		return -EINVAL;
 
-	rc = dlpar_cpu_add(drc_index);
+	rc = dlpar_cpu_add(drc_index, true);
 
 	return rc ? rc : count;
 }
@@ -871,7 +881,7 @@ static ssize_t dlpar_cpu_release(const char *buf, size_t count)
 		return -EINVAL;
 	}
 
-	rc = dlpar_cpu_remove(dn, drc_index);
+	rc = dlpar_cpu_remove(dn, drc_index, true);
 	of_node_put(dn);
 
 	return rc ? rc : count;

[RFC v9 2/4] hotplug/cpu: Add operation queuing function

[Michael Bringmann]

[Testing delayed due to internal SAN problems.]

migration/dlpar: This patch adds function dlpar_queue_action()
which will add information about a CPU/Memory 'readd' operation
according to resource type, action code, and DRC index.  Initial
usage is for the 'readd' CPU and Memory blocks identified as
having changed their associativity during a migration event.

Signed-off-by: Michael Bringmann <mwb@linux.vnet.ibm.com>
---
Changes in RFC:
  -- Correct drc_index for dlpar_queue_action worker invocation
  -- Correct text of notice
  -- Revise queuing model to save up all of the DLPAR actions for
     later execution.
---
 arch/powerpc/platforms/pseries/dlpar.c   |   51 ++++++++++++++++++++++++++++++
 arch/powerpc/platforms/pseries/pseries.h |    2 +
 2 files changed, 53 insertions(+)

diff --git a/arch/powerpc/platforms/pseries/dlpar.c b/arch/powerpc/platforms/pseries/dlpar.c
index a0b20c0..a1300d2 100644
--- a/arch/powerpc/platforms/pseries/dlpar.c
+++ b/arch/powerpc/platforms/pseries/dlpar.c
@@ -407,6 +407,57 @@ void queue_hotplug_event(struct pseries_hp_errorlog *hp_errlog,
 	}
 }
 
+struct DlparWork {
+	struct list_head list;
+	int resource;
+	int action;
+	u32 drc_index;
+} DlparWorkQ;
+
+int dlpar_delayed_queue_action(int resource, int action, u32 drc_index)
+{
+	struct DlparWork *dwq;
+
+	dwq = kmalloc(sizeof(struct DlparWork), GFP_KERNEL);
+	if (!dwq)
+		return -ENOMEM;
+
+	dwq->resource = resource;
+	dwq->action = action;
+	dwq->drc_index = drc_index;
+
+	list_add_tail(&dwq->list, &DlparWorkQ.list);
+
+	return 0;
+}
+
+int dlpar_schedule_delayed_queue(void)
+{
+	struct DlparWork *iter;
+	struct list_head *pos, *q;
+
+	list_for_each_entry(iter, &DlparWorkQ.list, list) {
+	        struct pseries_hp_errorlog hp_elog;
+
+        	hp_elog.resource = iter->resource;
+        	hp_elog.action = iter->action;
+        	hp_elog.id_type = PSERIES_HP_ELOG_ID_DRC_INDEX;
+        	hp_elog._drc_u.drc_index = cpu_to_be32(iter->drc_index);
+
+		handle_dlpar_errorlog(&hp_elog);
+	}
+
+	list_for_each_safe(pos, q, &DlparWorkQ.list) {
+		struct DlparWork *tmp;
+
+		tmp = list_entry(pos, struct DlparWork, list);
+		list_del(pos);
+		kfree(tmp);
+	}
+
+	return 0;
+}
+
 static int dlpar_parse_resource(char **cmd, struct pseries_hp_errorlog *hp_elog)
 {
 	char *arg;
diff --git a/arch/powerpc/platforms/pseries/pseries.h b/arch/powerpc/platforms/pseries/pseries.h
index 60db2ee..de73fda 100644
--- a/arch/powerpc/platforms/pseries/pseries.h
+++ b/arch/powerpc/platforms/pseries/pseries.h
@@ -61,6 +61,8 @@ extern struct device_node *dlpar_configure_connector(__be32,
 
 void queue_hotplug_event(struct pseries_hp_errorlog *hp_errlog,
 			 struct completion *hotplug_done, int *rc);
+int dlpar_delayed_queue_action(int resource, int action, u32 drc_index);
+int dlpar_schedule_delayed_queue(void);
 #ifdef CONFIG_MEMORY_HOTPLUG
 int dlpar_memory(struct pseries_hp_errorlog *hp_elog);
 #else

[RFC v9 3/4] hotplug/dlpar/cpu: Provide CPU readd operation

[Michael Bringmann]

[Testing delayed due to internal SAN problems.]

powerpc/dlpar: Provide hotplug CPU 'readd by index' operation to
support LPAR Post Migration state updates.  When such changes are
invoked by the PowerPC 'mobility' code, they will be queued up so
that modifications to CPU properties will take place after the new
property value is written to the device-tree.

Signed-off-by: Michael Bringmann <mwb@linux.vnet.ibm.com>
---
Changes in RFC:
  -- Add CPU validity check to pseries_smp_notifier
  -- Improve check on 'ibm,associativity' property
---
 arch/powerpc/platforms/pseries/hotplug-cpu.c |   61 ++++++++++++++++++++++++++
 1 file changed, 61 insertions(+)

diff --git a/arch/powerpc/platforms/pseries/hotplug-cpu.c b/arch/powerpc/platforms/pseries/hotplug-cpu.c
index 3632db2..be7ff7c 100644
--- a/arch/powerpc/platforms/pseries/hotplug-cpu.c
+++ b/arch/powerpc/platforms/pseries/hotplug-cpu.c
@@ -305,6 +305,37 @@ static int pseries_add_processor(struct device_node *np)
 	return err;
 }
 
+static int pseries_update_processor(struct of_reconfig_data *pr)
+{
+	int old_entries, new_entries, rc = 0;
+	__be32 *old_assoc, *new_assoc;
+
+	/* We only handle changes due to 'ibm,associativity' property
+	 */
+	old_assoc = pr->old_prop->value;
+	old_entries = be32_to_cpu(*old_assoc++);
+
+	new_assoc = pr->prop->value;
+	new_entries = be32_to_cpu(*new_assoc++);
+
+	if (old_entries == new_entries) {
+		int sz = old_entries * sizeof(int);
+
+		if (memcmp(old_assoc, new_assoc, sz))
+			rc = dlpar_delayed_queue_action(
+					PSERIES_HP_ELOG_RESOURCE_CPU,
+					PSERIES_HP_ELOG_ACTION_READD,
+					pr->dn->phandle);
+	} else {
+		rc = dlpar_delayed_queue_action(
+					PSERIES_HP_ELOG_RESOURCE_CPU,
+					PSERIES_HP_ELOG_ACTION_READD,
+					pr->dn->phandle);
+	}
+
+	return rc;
+}
+
 /*
  * Update the present map for a cpu node which is going away, and set
  * the hard id in the paca(s) to -1 to be consistent with boot time
@@ -649,6 +680,26 @@ static int dlpar_cpu_remove_by_index(u32 drc_index, bool release_drc)
 	return rc;
 }
 
+static int dlpar_cpu_readd_by_index(u32 drc_index)
+{
+	int rc = 0;
+
+	pr_info("Attempting to re-add CPU, drc index %x\n", drc_index);
+
+	rc = dlpar_cpu_remove_by_index(drc_index, false);
+	if (!rc)
+		rc = dlpar_cpu_add(drc_index, false);
+
+	if (rc)
+		pr_info("Failed to update cpu at drc_index %lx\n",
+				(unsigned long int)drc_index);
+	else
+		pr_info("CPU at drc_index %lx was updated\n",
+				(unsigned long int)drc_index);
+
+	return rc;
+}
+
 static int find_dlpar_cpus_to_remove(u32 *cpu_drcs, int cpus_to_remove)
 {
 	struct device_node *dn;
@@ -839,6 +890,9 @@ int dlpar_cpu(struct pseries_hp_errorlog *hp_elog)
 		else
 			rc = -EINVAL;
 		break;
+	case PSERIES_HP_ELOG_ACTION_READD:
+		rc = dlpar_cpu_readd_by_index(drc_index);
+		break;
 	default:
 		pr_err("Invalid action (%d) specified\n", hp_elog->action);
 		rc = -EINVAL;
@@ -895,6 +949,9 @@ static int pseries_smp_notifier(struct notifier_block *nb,
 	struct of_reconfig_data *rd = data;
 	int err = 0;
 
+	if (strcmp(rd->dn->type, "cpu"))
+		return notifier_from_errno(err);
+
 	switch (action) {
 	case OF_RECONFIG_ATTACH_NODE:
 		err = pseries_add_processor(rd->dn);
@@ -902,6 +959,10 @@ static int pseries_smp_notifier(struct notifier_block *nb,
 	case OF_RECONFIG_DETACH_NODE:
 		pseries_remove_processor(rd->dn);
 		break;
+	case OF_RECONFIG_UPDATE_PROPERTY:
+		if (!strcmp(rd->prop->name, "ibm,associativity"))
+			pseries_update_processor(rd);
+		break;
 	}
 	return notifier_from_errno(err);
 }

[RFC v9 4/4] mobility/numa: Ensure numa update does not overlap

[Michael Bringmann]

[Testing delayed due to internal SAN problems.]

mobility/numa: Ensure that numa_update_cpu_topology() can not be
entered multiple times concurrently.  It may be accessed through
many different paths / concurrent work functions, and the lock
ordering may be difficult to ensure otherwise.

Signed-off-by: Michael Bringmann <mwb@linux.vnet.ibm.com>
---
 arch/powerpc/mm/numa.c |    9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index 8802e7d..d4543b3 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -1079,6 +1079,7 @@ struct topology_update_data {
 static int topology_timer_secs = 1;
 static int topology_inited;
 static int topology_update_needed;
+static struct mutex topology_update_lock;
 
 /*
  * Change polling interval for associativity changes.
@@ -1320,6 +1321,11 @@ int numa_update_cpu_topology(bool cpus_locked)
 	if (!updates)
 		return 0;
 
+	if (!mutex_trylock(&topology_update_lock)) {
+		kfree(updates);
+		return 0;
+	}
+
 	cpumask_clear(&updated_cpus);
 
 	for_each_cpu(cpu, &cpu_associativity_changes_mask) {
@@ -1424,6 +1430,7 @@ int numa_update_cpu_topology(bool cpus_locked)
 out:
 	kfree(updates);
 	topology_update_needed = 0;
+	mutex_unlock(&topology_update_lock);
 	return changed;
 }
 
@@ -1598,6 +1605,8 @@ static ssize_t topology_write(struct file *file, const char __user *buf,
 
 static int topology_update_init(void)
 {
+	mutex_init(&topology_update_lock);
+
 	/* Do not poll for changes if disabled at boot */
 	if (topology_updates_enabled)
 		start_topology_update();