powerpc/powernv Platform dump interface

This enables support for userspace to fetch and initiate FSP and
Platform dumps from the service processor (via firmware) through sysfs.

Based on original patch from Vasant Hegde <hegdevasant@linux.vnet.ibm.com>

Flow:
  - We register for OPAL notification events.
  - OPAL sends new dump available notification.
  - We make information on dump available via sysfs
  - Userspace requests dump contents
  - We retrieve the dump via OPAL interface
  - User copies the dump data
  - userspace sends ack for dump
  - We send ACK to OPAL.

sysfs files:
  - We add the /sys/firmware/opal/dump directory
  - echoing 1 (well, anything, but in future we may support
    different dump types) to /sys/firmware/opal/dump/initiate_dump
    will initiate a dump.
  - Each dump that we've been notified of gets a directory
    in /sys/firmware/opal/dump/ with a name of the dump type and ID (in hex,
    as this is what's used elsewhere to identify the dump).
  - Each dump has files: id, type, dump and acknowledge
    dump is binary and is the dump itself.
    echoing 'ack' to acknowledge (currently any string will do) will
    acknowledge the dump and it will soon after disappear from sysfs.

OPAL APIs:
  - opal_dump_init()
  - opal_dump_info()
  - opal_dump_read()
  - opal_dump_ack()
  - opal_dump_resend_notification()

Currently we are only ever notified for one dump at a time (until
the user explicitly acks the current dump, then we get a notification
of the next dump), but this kernel code should "just work" when OPAL
starts notifying us of all the dumps present.

Signed-off-by: Stewart Smith <stewart@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>

Documentation/ABI/stable/sysfs-firmware-opal-dump

@@ -0,0 +1,41 @@
+What:		/sys/firmware/opal/dump
+Date:		Feb 2014
+Contact:	Stewart Smith <stewart@linux.vnet.ibm.com>
+Description:
+		This directory exposes interfaces for interacting with
+		the FSP and platform dumps through OPAL firmware interface.
+
+		This is only for the powerpc/powernv platform.
+
+		initiate_dump:	When '1' is written to it,
+				we will initiate a dump.
+				Read this file for supported commands.
+
+		0xXX-0xYYYY:	A directory for dump of type 0xXX and
+				id 0xYYYY (in hex). The name of this
+				directory should not be relied upon to
+				be in this format, only that it's unique
+				among all dumps. For determining the type
+				and ID of the dump, use the id and type files.
+				Do not rely on any particular size of dump
+				type or dump id.
+
+		Each dump has the following files:
+		id:		An ASCII representation of the dump ID
+				in hex (e.g. '0x01')
+		type:		An ASCII representation of the type of
+				dump in the format "0x%x %s" with the ID
+				in hex and a description of the dump type
+				(or 'unknown').
+				Type '0xffffffff unknown' is used when
+				we could not get the type from firmware.
+				e.g. '0x02 System/Platform Dump'
+		dump:		A binary file containing the dump.
+				The size of the dump is the size of this file.
+		acknowledge:	When 'ack' is written to this, we will
+				acknowledge that we've retrieved the
+				dump to the service processor. It will
+				then remove it, making the dump
+				inaccessible.
+				Reading this file will get a list of
+				supported actions.

arch/powerpc/include/asm/opal.h

@@ -159,9 +159,15 @@ extern int opal_enter_rtas(struct rtas_args *args,
 #define OPAL_FLASH_VALIDATE			76
 #define OPAL_FLASH_MANAGE			77
 #define OPAL_FLASH_UPDATE			78
+#define OPAL_DUMP_INIT				81
+#define OPAL_DUMP_INFO				82
+#define OPAL_DUMP_READ				83
+#define OPAL_DUMP_ACK				84
 #define OPAL_GET_MSG				85
 #define OPAL_CHECK_ASYNC_COMPLETION		86
+#define OPAL_DUMP_RESEND			91
 #define OPAL_SYNC_HOST_REBOOT			87
+#define OPAL_DUMP_INFO2				94
 
 #ifndef __ASSEMBLY__
 
@@ -242,6 +248,7 @@ enum OpalPendingState {
 	OPAL_EVENT_EPOW			= 0x80,
 	OPAL_EVENT_LED_STATUS		= 0x100,
 	OPAL_EVENT_PCI_ERROR		= 0x200,
+	OPAL_EVENT_DUMP_AVAIL		= 0x400,
 	OPAL_EVENT_MSG_PENDING		= 0x800,
 };
 
@@ -838,6 +845,12 @@ void opal_resend_pending_logs(void);
 int64_t opal_validate_flash(uint64_t buffer, uint32_t *size, uint32_t *result);
 int64_t opal_manage_flash(uint8_t op);
 int64_t opal_update_flash(uint64_t blk_list);
+int64_t opal_dump_init(uint8_t dump_type);
+int64_t opal_dump_info(uint32_t *dump_id, uint32_t *dump_size);
+int64_t opal_dump_info2(uint32_t *dump_id, uint32_t *dump_size, uint32_t *dump_type);
+int64_t opal_dump_read(uint32_t dump_id, uint64_t buffer);
+int64_t opal_dump_ack(uint32_t dump_id);
+int64_t opal_dump_resend_notification(void);
 
 int64_t opal_get_msg(uint64_t buffer, size_t size);
 int64_t opal_check_completion(uint64_t buffer, size_t size, uint64_t token);
@@ -876,6 +889,7 @@ extern unsigned long opal_get_boot_time(void);
 extern void opal_nvram_init(void);
 extern void opal_flash_init(void);
 extern int opal_elog_init(void);
+extern void opal_platform_dump_init(void);
 
 extern int opal_machine_check(struct pt_regs *regs);
 extern bool opal_mce_check_early_recovery(struct pt_regs *regs);

arch/powerpc/platforms/powernv/Makefile

@@ -1,6 +1,6 @@
 obj-y			+= setup.o opal-takeover.o opal-wrappers.o opal.o
 obj-y			+= opal-rtc.o opal-nvram.o opal-lpc.o opal-flash.o
-obj-y			+= rng.o opal-elog.o
+obj-y			+= rng.o opal-elog.o opal-dump.o
 
 obj-$(CONFIG_SMP)	+= smp.o
 obj-$(CONFIG_PCI)	+= pci.o pci-p5ioc2.o pci-ioda.o

arch/powerpc/platforms/powernv/opal-dump.c

@@ -0,0 +1,525 @@
+/*
+ * PowerNV OPAL Dump Interface
+ *
+ * Copyright 2013,2014 IBM Corp.
+ *
+ * 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/kobject.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/pagemap.h>
+#include <linux/delay.h>
+
+#include <asm/opal.h>
+
+#define DUMP_TYPE_FSP	0x01
+
+struct dump_obj {
+	struct kobject  kobj;
+	struct bin_attribute dump_attr;
+	uint32_t	id;  /* becomes object name */
+	uint32_t	type;
+	uint32_t	size;
+	char		*buffer;
+};
+#define to_dump_obj(x) container_of(x, struct dump_obj, kobj)
+
+struct dump_attribute {
+	struct attribute attr;
+	ssize_t (*show)(struct dump_obj *dump, struct dump_attribute *attr,
+			char *buf);
+	ssize_t (*store)(struct dump_obj *dump, struct dump_attribute *attr,
+			 const char *buf, size_t count);
+};
+#define to_dump_attr(x) container_of(x, struct dump_attribute, attr)
+
+static ssize_t dump_id_show(struct dump_obj *dump_obj,
+			    struct dump_attribute *attr,
+			    char *buf)
+{
+	return sprintf(buf, "0x%x\n", dump_obj->id);
+}
+
+static const char* dump_type_to_string(uint32_t type)
+{
+	switch (type) {
+	case 0x01: return "SP Dump";
+	case 0x02: return "System/Platform Dump";
+	case 0x03: return "SMA Dump";
+	default: return "unknown";
+	}
+}
+
+static ssize_t dump_type_show(struct dump_obj *dump_obj,
+			      struct dump_attribute *attr,
+			      char *buf)
+{
+	
+	return sprintf(buf, "0x%x %s\n", dump_obj->type,
+		       dump_type_to_string(dump_obj->type));
+}
+
+static ssize_t dump_ack_show(struct dump_obj *dump_obj,
+			     struct dump_attribute *attr,
+			     char *buf)
+{
+	return sprintf(buf, "ack - acknowledge dump\n");
+}
+
+/*
+ * Send acknowledgement to OPAL
+ */
+static int64_t dump_send_ack(uint32_t dump_id)
+{
+	int rc;
+
+	rc = opal_dump_ack(dump_id);
+	if (rc)
+		pr_warn("%s: Failed to send ack to Dump ID 0x%x (%d)\n",
+			__func__, dump_id, rc);
+	return rc;
+}
+
+static void delay_release_kobj(void *kobj)
+{
+	kobject_put((struct kobject *)kobj);
+}
+
+static ssize_t dump_ack_store(struct dump_obj *dump_obj,
+			      struct dump_attribute *attr,
+			      const char *buf,
+			      size_t count)
+{
+	dump_send_ack(dump_obj->id);
+	sysfs_schedule_callback(&dump_obj->kobj, delay_release_kobj,
+				&dump_obj->kobj, THIS_MODULE);
+	return count;
+}
+
+/* Attributes of a dump
+ * The binary attribute of the dump itself is dynamic
+ * due to the dynamic size of the dump
+ */
+static struct dump_attribute id_attribute =
+	__ATTR(id, 0666, dump_id_show, NULL);
+static struct dump_attribute type_attribute =
+	__ATTR(type, 0666, dump_type_show, NULL);
+static struct dump_attribute ack_attribute =
+	__ATTR(acknowledge, 0660, dump_ack_show, dump_ack_store);
+
+static ssize_t init_dump_show(struct dump_obj *dump_obj,
+			      struct dump_attribute *attr,
+			      char *buf)
+{
+	return sprintf(buf, "1 - initiate dump\n");
+}
+
+static int64_t dump_fips_init(uint8_t type)
+{
+	int rc;
+
+	rc = opal_dump_init(type);
+	if (rc)
+		pr_warn("%s: Failed to initiate FipS dump (%d)\n",
+			__func__, rc);
+	return rc;
+}
+
+static ssize_t init_dump_store(struct dump_obj *dump_obj,
+			       struct dump_attribute *attr,
+			       const char *buf,
+			       size_t count)
+{
+	dump_fips_init(DUMP_TYPE_FSP);
+	pr_info("%s: Initiated FSP dump\n", __func__);
+	return count;
+}
+
+static struct dump_attribute initiate_attribute =
+	__ATTR(initiate_dump, 0600, init_dump_show, init_dump_store);
+
+static struct attribute *initiate_attrs[] = {
+	&initiate_attribute.attr,
+	NULL,
+};
+
+static struct attribute_group initiate_attr_group = {
+	.attrs = initiate_attrs,
+};
+
+static struct kset *dump_kset;
+
+static ssize_t dump_attr_show(struct kobject *kobj,
+			      struct attribute *attr,
+			      char *buf)
+{
+	struct dump_attribute *attribute;
+	struct dump_obj *dump;
+
+	attribute = to_dump_attr(attr);
+	dump = to_dump_obj(kobj);
+
+	if (!attribute->show)
+		return -EIO;
+
+	return attribute->show(dump, attribute, buf);
+}
+
+static ssize_t dump_attr_store(struct kobject *kobj,
+			       struct attribute *attr,
+			       const char *buf, size_t len)
+{
+	struct dump_attribute *attribute;
+	struct dump_obj *dump;
+
+	attribute = to_dump_attr(attr);
+	dump = to_dump_obj(kobj);
+
+	if (!attribute->store)
+		return -EIO;
+
+	return attribute->store(dump, attribute, buf, len);
+}
+
+static const struct sysfs_ops dump_sysfs_ops = {
+	.show = dump_attr_show,
+	.store = dump_attr_store,
+};
+
+static void dump_release(struct kobject *kobj)
+{
+	struct dump_obj *dump;
+
+	dump = to_dump_obj(kobj);
+	vfree(dump->buffer);
+	kfree(dump);
+}
+
+static struct attribute *dump_default_attrs[] = {
+	&id_attribute.attr,
+	&type_attribute.attr,
+	&ack_attribute.attr,
+	NULL,
+};
+
+static struct kobj_type dump_ktype = {
+	.sysfs_ops = &dump_sysfs_ops,
+	.release = &dump_release,
+	.default_attrs = dump_default_attrs,
+};
+
+static void free_dump_sg_list(struct opal_sg_list *list)
+{
+	struct opal_sg_list *sg1;
+	while (list) {
+		sg1 = list->next;
+		kfree(list);
+		list = sg1;
+	}
+	list = NULL;
+}
+
+static struct opal_sg_list *dump_data_to_sglist(struct dump_obj *dump)
+{
+	struct opal_sg_list *sg1, *list = NULL;
+	void *addr;
+	int64_t size;
+
+	addr = dump->buffer;
+	size = dump->size;
+
+	sg1 = kzalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!sg1)
+		goto nomem;
+
+	list = sg1;
+	sg1->num_entries = 0;
+	while (size > 0) {
+		/* Translate virtual address to physical address */
+		sg1->entry[sg1->num_entries].data =
+			(void *)(vmalloc_to_pfn(addr) << PAGE_SHIFT);
+
+		if (size > PAGE_SIZE)
+			sg1->entry[sg1->num_entries].length = PAGE_SIZE;
+		else
+			sg1->entry[sg1->num_entries].length = size;
+
+		sg1->num_entries++;
+		if (sg1->num_entries >= SG_ENTRIES_PER_NODE) {
+			sg1->next = kzalloc(PAGE_SIZE, GFP_KERNEL);
+			if (!sg1->next)
+				goto nomem;
+
+			sg1 = sg1->next;
+			sg1->num_entries = 0;
+		}
+		addr += PAGE_SIZE;
+		size -= PAGE_SIZE;
+	}
+	return list;
+
+nomem:
+	pr_err("%s : Failed to allocate memory\n", __func__);
+	free_dump_sg_list(list);
+	return NULL;
+}
+
+static void sglist_to_phy_addr(struct opal_sg_list *list)
+{
+	struct opal_sg_list *sg, *next;
+
+	for (sg = list; sg; sg = next) {
+		next = sg->next;
+		/* Don't translate NULL pointer for last entry */
+		if (sg->next)
+			sg->next = (struct opal_sg_list *)__pa(sg->next);
+		else
+			sg->next = NULL;
+
+		/* Convert num_entries to length */
+		sg->num_entries =
+			sg->num_entries * sizeof(struct opal_sg_entry) + 16;
+	}
+}
+
+static int64_t dump_read_info(uint32_t *id, uint32_t *size, uint32_t *type)
+{
+	int rc;
+	*type = 0xffffffff;
+
+	rc = opal_dump_info2(id, size, type);
+
+	if (rc == OPAL_PARAMETER)
+		rc = opal_dump_info(id, size);
+
+	if (rc)
+		pr_warn("%s: Failed to get dump info (%d)\n",
+			__func__, rc);
+	return rc;
+}
+
+static int64_t dump_read_data(struct dump_obj *dump)
+{
+	struct opal_sg_list *list;
+	uint64_t addr;
+	int64_t rc;
+
+	/* Allocate memory */
+	dump->buffer = vzalloc(PAGE_ALIGN(dump->size));
+	if (!dump->buffer) {
+		pr_err("%s : Failed to allocate memory\n", __func__);
+		rc = -ENOMEM;
+		goto out;
+	}
+
+	/* Generate SG list */
+	list = dump_data_to_sglist(dump);
+	if (!list) {
+		rc = -ENOMEM;
+		goto out;
+	}
+
+	/* Translate sg list addr to real address */
+	sglist_to_phy_addr(list);
+
+	/* First entry address */
+	addr = __pa(list);
+
+	/* Fetch data */
+	rc = OPAL_BUSY_EVENT;
+	while (rc == OPAL_BUSY || rc == OPAL_BUSY_EVENT) {
+		rc = opal_dump_read(dump->id, addr);
+		if (rc == OPAL_BUSY_EVENT) {
+			opal_poll_events(NULL);
+			msleep(20);
+		}
+	}
+
+	if (rc != OPAL_SUCCESS && rc != OPAL_PARTIAL)
+		pr_warn("%s: Extract dump failed for ID 0x%x\n",
+			__func__, dump->id);
+
+	/* Free SG list */
+	free_dump_sg_list(list);
+
+out:
+	return rc;
+}
+
+static ssize_t dump_attr_read(struct file *filep, struct kobject *kobj,
+			      struct bin_attribute *bin_attr,
+			      char *buffer, loff_t pos, size_t count)
+{
+	ssize_t rc;
+
+	struct dump_obj *dump = to_dump_obj(kobj);
+
+	if (!dump->buffer) {
+		rc = dump_read_data(dump);
+
+		if (rc != OPAL_SUCCESS && rc != OPAL_PARTIAL) {
+			vfree(dump->buffer);
+			dump->buffer = NULL;
+
+			return -EIO;
+		}
+		if (rc == OPAL_PARTIAL) {
+			/* On a partial read, we just return EIO
+			 * and rely on userspace to ask us to try
+			 * again.
+			 */
+			pr_info("%s: Platform dump partially read.ID = 0x%x\n",
+				__func__, dump->id);
+			return -EIO;
+		}
+	}
+
+	memcpy(buffer, dump->buffer + pos, count);
+
+	/* You may think we could free the dump buffer now and retrieve
+	 * it again later if needed, but due to current firmware limitation,
+	 * that's not the case. So, once read into userspace once,
+	 * we keep the dump around until it's acknowledged by userspace.
+	 */
+
+	return count;
+}
+
+static struct dump_obj *create_dump_obj(uint32_t id, size_t size,
+					uint32_t type)
+{
+	struct dump_obj *dump;
+	int rc;
+
+	dump = kzalloc(sizeof(*dump), GFP_KERNEL);
+	if (!dump)
+		return NULL;
+
+	dump->kobj.kset = dump_kset;
+
+	kobject_init(&dump->kobj, &dump_ktype);
+
+	sysfs_bin_attr_init(&dump->dump_attr);
+
+	dump->dump_attr.attr.name = "dump";
+	dump->dump_attr.attr.mode = 0400;
+	dump->dump_attr.size = size;
+	dump->dump_attr.read = dump_attr_read;
+
+	dump->id = id;
+	dump->size = size;
+	dump->type = type;
+
+	rc = kobject_add(&dump->kobj, NULL, "0x%x-0x%x", type, id);
+	if (rc) {
+		kobject_put(&dump->kobj);
+		return NULL;
+	}
+
+	rc = sysfs_create_bin_file(&dump->kobj, &dump->dump_attr);
+	if (rc) {
+		kobject_put(&dump->kobj);
+		return NULL;
+	}
+
+	pr_info("%s: New platform dump. ID = 0x%x Size %u\n",
+		__func__, dump->id, dump->size);
+
+	kobject_uevent(&dump->kobj, KOBJ_ADD);
+
+	return dump;
+}
+
+static int process_dump(void)
+{
+	int rc;
+	uint32_t dump_id, dump_size, dump_type;
+	struct dump_obj *dump;
+	char name[22];
+
+	rc = dump_read_info(&dump_id, &dump_size, &dump_type);
+	if (rc != OPAL_SUCCESS)
+		return rc;
+
+	sprintf(name, "0x%x-0x%x", dump_type, dump_id);
+
+	/* we may get notified twice, let's handle
+	 * that gracefully and not create two conflicting
+	 * entries.
+	 */
+	if (kset_find_obj(dump_kset, name))
+		return 0;
+
+	dump = create_dump_obj(dump_id, dump_size, dump_type);
+	if (!dump)
+		return -1;
+
+	return 0;
+}
+
+static void dump_work_fn(struct work_struct *work)
+{
+	process_dump();
+}
+
+static DECLARE_WORK(dump_work, dump_work_fn);
+
+static void schedule_process_dump(void)
+{
+	schedule_work(&dump_work);
+}
+
+/*
+ * New dump available notification
+ *
+ * Once we get notification, we add sysfs entries for it.
+ * We only fetch the dump on demand, and create sysfs asynchronously.
+ */
+static int dump_event(struct notifier_block *nb,
+		      unsigned long events, void *change)
+{
+	if (events & OPAL_EVENT_DUMP_AVAIL)
+		schedule_process_dump();
+
+	return 0;
+}
+
+static struct notifier_block dump_nb = {
+	.notifier_call  = dump_event,
+	.next           = NULL,
+	.priority       = 0
+};
+
+void __init opal_platform_dump_init(void)
+{
+	int rc;
+
+	dump_kset = kset_create_and_add("dump", NULL, opal_kobj);
+	if (!dump_kset) {
+		pr_warn("%s: Failed to create dump kset\n", __func__);
+		return;
+	}
+
+	rc = sysfs_create_group(&dump_kset->kobj, &initiate_attr_group);
+	if (rc) {
+		pr_warn("%s: Failed to create initiate dump attr group\n",
+			__func__);
+		kobject_put(&dump_kset->kobj);
+		return;
+	}
+
+	rc = opal_notifier_register(&dump_nb);
+	if (rc) {
+		pr_warn("%s: Can't register OPAL event notifier (%d)\n",
+			__func__, rc);
+		return;
+	}
+
+	opal_dump_resend_notification();
+}

arch/powerpc/platforms/powernv/opal-wrappers.S

@@ -131,6 +131,12 @@ OPAL_CALL(opal_write_elog,			OPAL_ELOG_WRITE);
 OPAL_CALL(opal_validate_flash,			OPAL_FLASH_VALIDATE);
 OPAL_CALL(opal_manage_flash,			OPAL_FLASH_MANAGE);
 OPAL_CALL(opal_update_flash,			OPAL_FLASH_UPDATE);
+OPAL_CALL(opal_dump_init,			OPAL_DUMP_INIT);
+OPAL_CALL(opal_dump_info,			OPAL_DUMP_INFO);
+OPAL_CALL(opal_dump_info2,			OPAL_DUMP_INFO2);
+OPAL_CALL(opal_dump_read,			OPAL_DUMP_READ);
+OPAL_CALL(opal_dump_ack,			OPAL_DUMP_ACK);
 OPAL_CALL(opal_get_msg,				OPAL_GET_MSG);
 OPAL_CALL(opal_check_completion,		OPAL_CHECK_ASYNC_COMPLETION);
+OPAL_CALL(opal_dump_resend_notification,	OPAL_DUMP_RESEND);
 OPAL_CALL(opal_sync_host_reboot,		OPAL_SYNC_HOST_REBOOT);

arch/powerpc/platforms/powernv/opal.c

@@ -570,6 +570,8 @@ static int __init opal_init(void)
 		rc = opal_elog_init();
 		/* Setup code update interface */
 		opal_flash_init();
+		/* Setup platform dump extract interface */
+		opal_platform_dump_init();
 	}
 
 	return 0;