[RFC PATCH 00/10] fadump: Firmware-assisted dump support for Powerpc.

@@ -0,0 +1,231 @@
+
+                   Firmware-Assisted Dump
+                   ------------------------
+                       July 2011
+
+The goal of firmware-assisted dump is to enable the dump of
+a crashed system, and to do so from a fully-reset system, and
+to minimize the total elapsed time until the system is back
+in production use.
+
+As compared to kdump or other strategies, firmware-assisted
+dump offers several strong, practical advantages:
+
+-- Unlike kdump, the system has been reset, and loaded
+   with a fresh copy of the kernel.  In particular,
+   PCI and I/O devices have been reinitialized and are
+   in a clean, consistent state.
+-- Once the dump is copied out, the memory that held the dump
+   is immediately available to the running kernel. A further
+   reboot isn't required.
+
+The above can only be accomplished by coordination with,
+and assistance from the Power firmware. The procedure is
+as follows:
+
+-- The first kernel registers the sections of memory with the
+   Power firmware for dump preservation during OS initialization.
+   This registered sections of memory is reserved by the first
+   kernel during early boot.
+
+-- When a system crashes, the Power firmware will save
+   the low memory (boot memory of size larger of 5% of system RAM
+   or 256MB) of RAM to a previously registered save region. It
+   will also save system registers, and hardware PTE's.
+
+   NOTE: The term 'boot memory' means size of the low memory chunk
+         that is required for a kernel to boot successfully when
+         booted with restricted memory.
+
+-- After the low memory (boot memory) area has been saved, the
+   firmware will reset PCI and other hardware state.  It will
+   *not* clear the RAM. It will then launch the bootloader, as
+   normal.
+
+-- The freshly booted kernel will notice that there is a new
+   node (ibm,dump-kernel) in the device tree, indicating that
+   there is crash data available from a previous boot. During
+   the early boot OS will reserve rest of the memory above
+   boot memory size effectively booting with restricted memory
+   size. This will make sure that the second kernel will not
+   touch any of the dump memory area.
+
+-- Userspace tools will read /proc/vmcore to obtain the contents
+   of memory, which holds the previous crashed kernel dump in ELF
+   format. The userspace tools may copy this info to disk, or
+   network, nas, san, iscsi, etc. as desired.
+
+-- Once the userspace tool is done saving dump, it will echo
+   '1' to /sys/kernel/fadump_release_mem to release the reserved
+   memory back to general use, except the memory required for
+   next firmware-assisted dump registration.
+
+   e.g.
+     # echo 1 > /sys/kernel/fadump_release_mem
+
+Please note that the firmware-assisted dump feature
+is only available on Power6 and above systems with recent
+firmware versions.
+
+Implementation details:
+----------------------
+
+During boot, a check is made to see if firmware supports
+this feature on that particular machine. If it does, then
+we check to see if an active dump is waiting for us. If yes
+then everything but boot memory size of RAM is reserved during
+early boot (See Fig. 2). This area is released once we collect a
+dump from user land scripts (kdump scripts) that are run. If
+there is dump data, then the /sys/kernel/fadump_release_mem
+file is created, and the reserved memory is held.
+
+If there is no waiting dump data, then only the memory required
+to hold CPU state, HPTE region, boot memory dump and elfcore
+header, is reserved at the top of memory (see Fig. 1). This area
+is *not* released: this region will be kept permanently reserved,
+so that it can act as a receptacle for a copy of the boot memory
+content in addition to CPU state and HPTE region, in the case a
+crash does occur.
+
+  o Memory Reservation during first kernel
+
+  Low memory                                        Top of memory
+  0      boot memory size                                       |
+  |           |                       |<--Reserved dump area -->|
+  V           V                       |   Permanent Reservation V
+  +-----------+----------/ /----------+---+----+-----------+----+
+  |           |                       |CPU|HPTE|  DUMP     |ELF |
+  +-----------+----------/ /----------+---+----+-----------+----+
+        |                                           ^
+        |                                           |
+        \                                           /
+         -------------------------------------------
+          Boot memory content gets transferred to
+          reserved area by firmware at the time of
+          crash
+                   Fig. 1
+
+  o Memory Reservation during second kernel after crash
+
+  Low memory                                        Top of memory
+  0      boot memory size                                       |
+  |           |<------------- Reserved dump area ----------- -->|
+  V           V                                                 V
+  +-----------+----------/ /----------+---+----+-----------+----+
+  |           |                       |CPU|HPTE|  DUMP     |ELF |
+  +-----------+----------/ /----------+---+----+-----------+----+
+        |                                                    |
+        V                                                    V
+   Used by second                                    /proc/vmcore
+   kernel to boot
+                   Fig. 2
+
+Currently the dump will be copied from /proc/vmcore to a
+a new file upon user intervention. The dump data available through
+/proc/vmcore will be in ELF format. Hence the existing kdump
+infrastructure (kdump scripts) to save the dump works fine
+with minor modifications. The kdump script requires following
+modifications:
+-- During service kdump start if /proc/vmcore entry is not present,
+   look for the existence of /sys/kernel/fadump_enabled and read
+   value exported by it. If value is set to '1' then print
+   success otherwise fallback to existing kexec based kdump.
+
+-- During service kdump start if /proc/vmcore entry is present,
+   execute the existing routine to save the dump. Once the dump
+   is saved, echo 1 > /sys/kernel/fadump_release_mem (if the
+   file exists) to release the reserved memory for general use
+   and continue without rebooting. At this point the memory
+   reservation map will look like as shown in Fig. 1. If the file
+   /sys/kernel/fadump_release_mem is not present then follow
+   the existing routine to reboot into new kernel.
+
+The tools to examine the dump will be same as the ones
+used for kdump.
+
+How to enable firmware-assisted dump (fadump):
+-------------------------------------
+
+1. Set config option CONFIG_FA_DUMP=y and build kernel.
+2. Boot into linux kernel with 'fadump=1' kernel cmdline option.
+
+NOTE: If firmware-assisted dump fails to reserve memory then it will
+   fallback to existing kdump mechanism if 'crashkernel=' option
+   is set at kernel cmdline.
+
+Sysfs files:
+------------
+
+Firmware-assisted dump feature uses sysfs file system to hold
+the control files as well as the files to display memory reserved
+region.
+
+Here is the list of files under kernel sysfs:
+
+ /sys/kernel/fadump_enabled
+
+    This is used to display the fadump status.
+    0 = fadump is disabled
+    1 = fadump is enabled
+
+ /sys/kernel/fadump_region
+
+    This file shows the reserved memory regions if fadump is
+    enabled otherwise this file is empty. The output format
+    is:
+    <region>: [<start>-<end>] <reserved-size> bytes, Dumped: <dump-size>
+
+    e.g.
+    Contents when fadump is registered during first kernel
+
+    # cat /sys/kernel/fadump_region
+    CPU : [0x0000006ffb0000-0x0000006fff001f] 0x40020 bytes, Dumped: 0x0
+    HPTE: [0x0000006fff0020-0x0000006fff101f] 0x1000 bytes, Dumped: 0x0
+    DUMP: [0x0000006fff1020-0x0000007fff101f] 0x10000000 bytes, Dumped: 0x0
+
+    Contents when fadump is active during second kernel
+
+    # cat /sys/kernel/fadump_region
+    CPU : [0x0000006ffb0000-0x0000006fff001f] 0x40020 bytes, Dumped: 0x40020
+    HPTE: [0x0000006fff0020-0x0000006fff101f] 0x1000 bytes, Dumped: 0x1000
+    DUMP: [0x0000006fff1020-0x0000007fff101f] 0x10000000 bytes, Dumped: 0x10000000
+        : [0x00000010000000-0x0000006ffaffff] 0x5ffb0000 bytes, Dumped: 0x5ffb0000
+
+ /sys/kernel/fadump_release_mem
+
+    This file is available only when fadump is active during
+    second kernel. This is used to release the reserved memory
+    region that are held for saving crash dump. To release the
+    reserved memory echo 1 to it:
+
+    echo 1  > /sys/kernel/fadump_release_mem
+
+    After echo 1, the content of the /sys/kernel/fadump_region
+    file will change to reflect the new memory reservations.
+
+TODO:
+-----
+ o Need to come up with the better approach to find out more
+   accurate boot memory size that is required for a kernel to
+   boot successfully when booted with restricted memory.
+ o The fadump implementation introduces a fadump crash info structure
+   in the scratch area before the ELF core header. The idea of introducing
+   this structure is to pass some important crash info data to the second
+   kernel which will help second kernel to populate ELF core header with
+   correct data before it gets exported through /proc/vmcore. The current
+   design implementation does not address a possibility of introducing
+   additional fields (in future) to this structure without affecting
+   compatibility. Need to come up with the better approach to address this.
+   The possible approaches are:
+	1. Introduce version field for version tracking, bump up the version
+	whenever a new field is added to the structure in future. The version
+	field can be used to find out what fields are valid for the current
+	version of the structure.
+	2. Reserve the area of predefined size (say PAGE_SIZE) for this
+	structure and have unused area as reserved (initialized to zero)
+	for future field additions.
+   The advantage of approach 1 over 2 is we don't need to reserve extra space.
+---
+Author: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
+This document is based on the original documentation written for phyp
+assisted dump by Linas Vepstas and Manish Ahuja.

@@ -0,0 +1,56 @@
+/*
+ * Firmware Assisted dump header file.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright 2011 IBM Corporation
+ * Author: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
+ */
+
+#ifndef __PPC64_FA_DUMP_H__
+#define __PPC64_FA_DUMP_H__
+
+#ifdef CONFIG_FA_DUMP
+
+/*
+ * The RMR region will be saved for later dumping when kernel crashes.
+ * Set this to 256MB.
+ */
+#define RMR_START	0x0
+#define RMR_END		(0x1UL << 28)	/* 256 MB */
+
+/* Firmware provided dump sections */
+#define FADUMP_CPU_STATE_DATA	0x0001
+#define FADUMP_HPTE_REGION	0x0002
+#define FADUMP_REAL_MODE_REGION	0x0011
+
+struct fw_dump {
+	unsigned long	cpu_state_data_size;
+	unsigned long	hpte_region_size;
+	unsigned long	boot_memory_size;
+	unsigned long	reserve_dump_area_start;
+	unsigned long	reserve_dump_area_size;
+	int		ibm_configure_kernel_dump;
+
+	unsigned long	fadump_enabled:1;
+	unsigned long	fadump_supported:1;
+	unsigned long	dump_active:1;
+};
+
+extern int early_init_dt_scan_fw_dump(unsigned long node,
+		const char *uname, int depth, void *data);
+extern int fadump_reserve_mem(void);
+#endif
+#endif

@@ -60,6 +60,7 @@ obj-$(CONFIG_IBMVIO)		+= vio.o
 obj-$(CONFIG_IBMEBUS)           += ibmebus.o
 obj-$(CONFIG_GENERIC_TBSYNC)	+= smp-tbsync.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump.o
+obj-$(CONFIG_FA_DUMP)		+= fadump.o
 ifeq ($(CONFIG_PPC32),y)
 obj-$(CONFIG_E500)		+= idle_e500.o
 endif

@@ -0,0 +1,240 @@
+/*
+ * Firmware Assisted dump: A robust mechanism to get reliable kernel crash
+ * dump with assistance from firmware. This approach does not use kexec,
+ * instead firmware assists in booting the kdump kernel while preserving
+ * memory contents. The most of the code implementation has been adapted
+ * from phyp assisted dump implementation written by Linas Vepstas and
+ * Manish Ahuja
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright 2011 IBM Corporation
+ * Author: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
+ */
+
+#undef DEBUG
+
+#include <linux/string.h>
+#include <linux/memblock.h>
+
+#include <asm/page.h>
+#include <asm/prom.h>
+#include <asm/rtas.h>
+#include <asm/fadump.h>
+
+#ifdef DEBUG
+#define PREFIX		"fadump: "
+#define DBG(fmt...)	printk(KERN_ERR PREFIX fmt)
+#else
+#define DBG(fmt...)
+#endif
+
+/*
+ * The RTAS property "ibm,configure-kernel-dump-sizes" returns dump
+ * sizes for two of the firmware provided dump sections (cpu state data
+ * and hpte region).
+ */
+#define FW_DUMP_NUM_SECTIONS	2
+
+struct dump_section {
+	u32		dump_section;
+	unsigned long	section_size;
+} __packed;
+
+/* Global variable to hold firmware assisted dump configuration info. */
+static struct fw_dump fw_dump;
+
+/* Scan the Firmware Assisted dump configuration details. */
+int __init early_init_dt_scan_fw_dump(unsigned long node,
+			const char *uname, int depth, void *data)
+{
+	const struct dump_section *sections;
+	int i;
+	const int *token;
+
+	if (depth != 1 || strcmp(uname, "rtas") != 0)
+		return 0;
+
+	/*
+	 * Check if Firmware Assisted dump is supported. if yes, check
+	 * if dump has been initiated on last reboot.
+	 */
+	token = of_get_flat_dt_prop(node, "ibm,configure-kernel-dump", NULL);
+	if (!token)
+		return 0;
+
+	fw_dump.fadump_supported = 1;
+	fw_dump.ibm_configure_kernel_dump = *token;
+
+	/*
+	 * The 'ibm,kernel-dump' rtas node is present only if there is
+	 * dump data waiting for us.
+	 */
+	if (of_get_flat_dt_prop(node, "ibm,kernel-dump", NULL))
+		fw_dump.dump_active = 1;
+
+	/* Get the sizes required to store dump data for the firmware provided
+	 * dump sections.
+	 */
+	sections = of_get_flat_dt_prop(node, "ibm,configure-kernel-dump-sizes",
+					NULL);
+
+	if (!sections)
+		return 0;
+
+	for (i = 0; i < FW_DUMP_NUM_SECTIONS; i++) {
+		switch (sections[i].dump_section) {
+		case FADUMP_CPU_STATE_DATA:
+			fw_dump.cpu_state_data_size = sections[i].section_size;
+			break;
+		case FADUMP_HPTE_REGION:
+			fw_dump.hpte_region_size = sections[i].section_size;
+			break;
+		}
+	}
+	return 1;
+}
+
+/**
+ * calculate_reserve_size() - reserve variable boot area 5% of System RAM
+ *
+ * Function to find the largest memory size we need to reserve during early
+ * boot process. This will be the size of the memory that is required for a
+ * kernel to boot successfully.
+ *
+ * This function has been taken from phyp-assisted dump feature implementation.
+ *
+ * returns larger of 256MB or 5% rounded down to multiples of 256MB.
+ *
+ * TODO: Come up with better approach to find out more accurate memory size
+ * that is required for a kernel to boot successfully.
+ *
+ */
+static inline unsigned long calculate_reserve_size(void)
+{
+	unsigned long size;
+
+	/* divide by 20 to get 5% of value */
+	size = memblock_end_of_DRAM();
+	do_div(size, 20);
+
+	/* round it down in multiples of 256 */
+	size = size & ~0x0FFFFFFFUL;
+
+	/* Truncate to memory_limit. We don't want to over reserve the memory.*/
+	if (memory_limit && size > memory_limit)
+		size = memory_limit;
+
+	return (size > RMR_END ? size : RMR_END);
+}
+
+/*
+ * Calculate the total memory size required to be reserved for
+ * firmware-assisted dump registration.
+ */
+static unsigned long get_dump_area_size(void)
+{
+	unsigned long size = 0;
+
+	size += fw_dump.cpu_state_data_size;
+	size += fw_dump.hpte_region_size;
+	size += fw_dump.boot_memory_size;
+
+	size = PAGE_ALIGN(size);
+	return size;
+}
+
+int __init fadump_reserve_mem(void)
+{
+	unsigned long base, size, memory_boundary;
+
+	if (!fw_dump.fadump_enabled)
+		return 0;
+
+	if (!fw_dump.fadump_supported) {
+		printk(KERN_ERR "Firmware-assisted dump is not supported on"
+				" this hardware\n");
+		fw_dump.fadump_enabled = 0;
+		return 0;
+	}
+	/* Initialize boot memory size */
+	fw_dump.boot_memory_size = calculate_reserve_size();
+
+	/*
+	 * Calculate the memory boundary.
+	 * If memory_limit is less than actual memory boundary then reserve
+	 * the memory for fadump beyond the memory_limit and adjust the
+	 * memory_limit accordingly, so that the running kernel can run with
+	 * specified memory_limit.
+	 */
+	if (memory_limit && memory_limit < memblock_end_of_DRAM()) {
+		size = get_dump_area_size();
+		if ((memory_limit + size) < memblock_end_of_DRAM())
+			memory_limit += size;
+		else
+			memory_limit = memblock_end_of_DRAM();
+		printk(KERN_INFO "Adjusted memory_limit for firmware-assisted"
+				" dump, now %#016llx\n",
+				(unsigned long long)memory_limit);
+	}
+	if (memory_limit)
+		memory_boundary = memory_limit;
+	else
+		memory_boundary = memblock_end_of_DRAM();
+
+	if (fw_dump.dump_active) {
+		printk(KERN_INFO "Firmware-assisted dump is active.\n");
+		/*
+		 * If last boot has crashed then reserve all the memory
+		 * above boot_memory_size so that we don't touch it until
+		 * dump is written to disk by userspace tool. This memory
+		 * will be released for general use once the dump is saved.
+		 */
+		base = fw_dump.boot_memory_size;
+		size = memory_boundary - base;
+		memblock_reserve(base, size);
+		printk(KERN_INFO "Reserved %ldMB of memory at %ldMB "
+				"for saving crash dump\n",
+				(unsigned long)(size >> 20),
+				(unsigned long)(base >> 20));
+	} else {
+		/* Reserve the memory at the top of memory. */
+		size = get_dump_area_size();
+		base = memory_boundary - size;
+		memblock_reserve(base, size);
+		printk(KERN_INFO "Reserved %ldMB of memory at %ldMB "
+				"for firmware-assisted dump\n",
+				(unsigned long)(size >> 20),
+				(unsigned long)(base >> 20));
+	}
+	fw_dump.reserve_dump_area_start = base;
+	fw_dump.reserve_dump_area_size = size;
+	return 1;
+}
+
+/* Look for fadump= cmdline option. */
+static int __init early_fadump_param(char *p)
+{
+	if (!p)
+		return 1;
+
+	if (p[0] == '1')
+		fw_dump.fadump_enabled = 1;
+	else if (p[0] == '0')
+		fw_dump.fadump_enabled = 0;
+
+	return 0;
+}
+early_param("fadump", early_fadump_param);

@@ -54,6 +54,7 @@
 #include <asm/pci-bridge.h>
 #include <asm/phyp_dump.h>
 #include <asm/kexec.h>
+#include <asm/fadump.h>
 #include <mm/mmu_decl.h>
 
 #ifdef DEBUG

@@ -711,6 +712,11 @@ void __init early_init_devtree(void *params)
 	of_scan_flat_dt(early_init_dt_scan_phyp_dump, NULL);
 #endif
 
+#ifdef CONFIG_FA_DUMP
+	/* scan tree to see if dump is active during last boot */
+	of_scan_flat_dt(early_init_dt_scan_fw_dump, NULL);
+#endif
+
 	/* Retrieve various informations from the /chosen node of the
 	 * device-tree, including the platform type, initrd location and
 	 * size, TCE reserve, and more ...

@@ -734,7 +740,14 @@ void __init early_init_devtree(void *params)
 	if (PHYSICAL_START > MEMORY_START)
 		memblock_reserve(MEMORY_START, 0x8000);
 	reserve_kdump_trampoline();
-	reserve_crashkernel();
+#ifdef CONFIG_FA_DUMP
+	/*
+	 * If we fail to reserve memory for firmware-assisted dump then
+	 * fallback to kexec based kdump.
+	 */
+	if (fadump_reserve_mem() == 0)
+#endif
+		reserve_crashkernel();
 	early_reserve_mem();
 	phyp_dump_reserve_mem();
 

@@ -36,6 +36,58 @@
 #define FADUMP_HPTE_REGION	0x0002
 #define FADUMP_REAL_MODE_REGION	0x0011
 
+/* Dump request flag */
+#define FADUMP_REQUEST_FLAG	0x00000001
+
+/* FAD commands */
+#define FADUMP_REGISTER	1
+#define FADUMP_UNREGISTER	2
+#define FADUMP_INVALIDATE	3
+
+/* Kernel Dump section info */
+struct fadump_section {
+	u32	request_flag;
+	u16	source_data_type;
+	u16	error_flags;
+	u64	source_address;
+	u64	source_len;
+	u64	bytes_dumped;
+	u64	destination_address;
+};
+
+/* ibm,configure-kernel-dump header. */
+struct fadump_section_header {
+	u32	dump_format_version;
+	u16	dump_num_sections;
+	u16	dump_status_flag;
+	u32	offset_first_dump_section;
+
+	/* Fields for disk dump option. */
+	u32	dd_block_size;
+	u64	dd_block_offset;
+	u64	dd_num_blocks;
+	u32	dd_offset_disk_path;
+
+	/* Maximum time allowed to prevent an automatic dump-reboot. */
+	u32	max_time_auto;
+};
+
+/*
+ * Firmware Assisted dump memory structure. This structure is required for
+ * registering future kernel dump with power firmware through rtas call.
+ *
+ * No disk dump option. Hence disk dump path string section is not included.
+ */
+struct fadump_mem_struct {
+	struct fadump_section_header	header;
+
+	/* Kernel dump sections */
+	struct fadump_section		cpu_state_data;
+	struct fadump_section		hpte_region;
+	struct fadump_section		rmr_region;
+};
+
+/* Firmware-assisted dump configuration details. */
 struct fw_dump {
 	unsigned long	cpu_state_data_size;
 	unsigned long	hpte_region_size;

@@ -47,10 +99,13 @@ struct fw_dump {
 	unsigned long	fadump_enabled:1;
 	unsigned long	fadump_supported:1;
 	unsigned long	dump_active:1;
+	unsigned long	dump_registered:1;
 };
 
 extern int early_init_dt_scan_fw_dump(unsigned long node,
 		const char *uname, int depth, void *data);
 extern int fadump_reserve_mem(void);
+extern int setup_fadump(void);
+extern int is_fadump_active(void);
 #endif
 #endif

@@ -28,6 +28,7 @@
 
 #include <linux/string.h>
 #include <linux/memblock.h>
+#include <linux/delay.h>
 
 #include <asm/page.h>
 #include <asm/prom.h>

@@ -55,6 +56,8 @@ struct dump_section {
 
 /* Global variable to hold firmware assisted dump configuration info. */
 static struct fw_dump fw_dump;
+static struct fadump_mem_struct fdm;
+static const struct fadump_mem_struct *fdm_active;
 
 /* Scan the Firmware Assisted dump configuration details. */
 int __init early_init_dt_scan_fw_dump(unsigned long node,

@@ -82,7 +85,8 @@ int __init early_init_dt_scan_fw_dump(unsigned long node,
 	 * The 'ibm,kernel-dump' rtas node is present only if there is
 	 * dump data waiting for us.
 	 */
-	if (of_get_flat_dt_prop(node, "ibm,kernel-dump", NULL))
+	fdm_active = of_get_flat_dt_prop(node, "ibm,kernel-dump", NULL);
+	if (fdm_active)
 		fw_dump.dump_active = 1;
 
 	/* Get the sizes required to store dump data for the firmware provided

@@ -107,6 +111,163 @@ int __init early_init_dt_scan_fw_dump(unsigned long node,
 	return 1;
 }
 
+int is_fadump_active(void)
+{
+	return fw_dump.dump_active;
+}
+
+/* Print firmware assisted dump configurations for debugging purpose. */
+static void fadump_show_config(void)
+{
+	DBG("Support for firmware-assisted dump (fadump): %s\n",
+			(fw_dump.fadump_supported ? "present" : "no support"));
+
+	if (!fw_dump.fadump_supported)
+		return;
+
+	DBG("Fadump enabled    : %s\n",
+				(fw_dump.fadump_enabled ? "yes" : "no"));
+	DBG("Dump Active       : %s\n", (fw_dump.dump_active ? "yes" : "no"));
+	DBG("Dump section sizes:\n");
+	DBG("	CPU state data size: %lx\n", fw_dump.cpu_state_data_size);
+	DBG("	HPTE region size   : %lx\n", fw_dump.hpte_region_size);
+	DBG("Boot memory size  : %lx\n", fw_dump.boot_memory_size);
+	DBG("Reserve area start: %lx\n", fw_dump.reserve_dump_area_start);
+	DBG("Reserve area size : %lx\n", fw_dump.reserve_dump_area_size);
+}
+
+static void show_fadump_mem_struct(const struct fadump_mem_struct *fdm)
+{
+	if (!fdm)
+		return;
+
+	DBG("--------Firmware-assisted dump memory structure---------\n");
+	DBG("header.dump_format_version        : 0x%08x\n",
+					fdm->header.dump_format_version);
+	DBG("header.dump_num_sections          : %d\n",
+					fdm->header.dump_num_sections);
+	DBG("header.dump_status_flag           : 0x%04x\n",
+					fdm->header.dump_status_flag);
+	DBG("header.offset_first_dump_section  : 0x%x\n",
+					fdm->header.offset_first_dump_section);
+
+	DBG("header.dd_block_size              : %d\n",
+					fdm->header.dd_block_size);
+	DBG("header.dd_block_offset            : 0x%Lx\n",
+					fdm->header.dd_block_offset);
+	DBG("header.dd_num_blocks              : %Lx\n",
+					fdm->header.dd_num_blocks);
+	DBG("header.dd_offset_disk_path        : 0x%x\n",
+					fdm->header.dd_offset_disk_path);
+
+	DBG("header.max_time_auto              : %d\n",
+					fdm->header.max_time_auto);
+
+	/* Kernel dump sections */
+	DBG("cpu_state_data.request_flag       : 0x%08x\n",
+					fdm->cpu_state_data.request_flag);
+	DBG("cpu_state_data.source_data_type   : 0x%04x\n",
+					fdm->cpu_state_data.source_data_type);
+	DBG("cpu_state_data.error_flags        : 0x%04x\n",
+					fdm->cpu_state_data.error_flags);
+	DBG("cpu_state_data.source_address     : 0x%016Lx\n",
+					fdm->cpu_state_data.source_address);
+	DBG("cpu_state_data.source_len         : 0x%Lx\n",
+					fdm->cpu_state_data.source_len);
+	DBG("cpu_state_data.bytes_dumped       : 0x%Lx\n",
+					fdm->cpu_state_data.bytes_dumped);
+	DBG("cpu_state_data.destination_address: 0x%016Lx\n",
+				fdm->cpu_state_data.destination_address);
+
+	DBG("hpte_region.request_flag          : 0x%08x\n",
+					fdm->hpte_region.request_flag);
+	DBG("hpte_region.source_data_type      : 0x%04x\n",
+					fdm->hpte_region.source_data_type);
+	DBG("hpte_region.error_flags           : 0x%04x\n",
+					fdm->hpte_region.error_flags);
+	DBG("hpte_region.source_address        : 0x%016Lx\n",
+					fdm->hpte_region.source_address);
+	DBG("hpte_region.source_len            : 0x%Lx\n",
+					fdm->hpte_region.source_len);
+	DBG("hpte_region.bytes_dumped          : 0x%Lx\n",
+					fdm->hpte_region.bytes_dumped);
+	DBG("hpte_region.destination_address   : 0x%016Lx\n",
+				fdm->hpte_region.destination_address);
+
+	DBG("rmr_region.request_flag           : 0x%08x\n",
+					fdm->rmr_region.request_flag);
+	DBG("rmr_region.source_data_type       : 0x%04x\n",
+					fdm->rmr_region.source_data_type);
+	DBG("rmr_region.error_flags            : 0x%04x\n",
+					fdm->rmr_region.error_flags);
+	DBG("rmr_region.source_address         : 0x%016Lx\n",
+					fdm->rmr_region.source_address);
+	DBG("rmr_region.source_len             : 0x%Lx\n",
+					fdm->rmr_region.source_len);
+	DBG("rmr_region.bytes_dumped           : 0x%Lx\n",
+					fdm->rmr_region.bytes_dumped);
+	DBG("rmr_region.destination_address    : 0x%016Lx\n",
+				fdm->rmr_region.destination_address);
+
+	DBG("--------Firmware-assisted dump memory structure---------\n");
+}
+
+static unsigned long init_fadump_mem_struct(struct fadump_mem_struct *fdm,
+				unsigned long addr)
+{
+	if (!fdm)
+		return 0;
+
+	memset(fdm, 0, sizeof(struct fadump_mem_struct));
+	addr = addr & PAGE_MASK;
+
+	fdm->header.dump_format_version = 0x00000001;
+	fdm->header.dump_num_sections = 3;
+	fdm->header.dump_status_flag = 0;
+	fdm->header.offset_first_dump_section =
+		(u32)offsetof(struct fadump_mem_struct, cpu_state_data);
+
+	/*
+	 * Fields for disk dump option.
+	 * We are not using disk dump option, hence set these fields to 0.
+	 */
+	fdm->header.dd_block_size = 0;
+	fdm->header.dd_block_offset = 0;
+	fdm->header.dd_num_blocks = 0;
+	fdm->header.dd_offset_disk_path = 0;
+
+	/* set 0 to disable an automatic dump-reboot. */
+	fdm->header.max_time_auto = 0;
+
+	/* Kernel dump sections */
+	/* cpu state data section. */
+	fdm->cpu_state_data.request_flag = FADUMP_REQUEST_FLAG;
+	fdm->cpu_state_data.source_data_type = FADUMP_CPU_STATE_DATA;
+	fdm->cpu_state_data.source_address = 0;
+	fdm->cpu_state_data.source_len = fw_dump.cpu_state_data_size;
+	fdm->cpu_state_data.destination_address = addr;
+	addr += fw_dump.cpu_state_data_size;
+
+	/* hpte region section */
+	fdm->hpte_region.request_flag = FADUMP_REQUEST_FLAG;
+	fdm->hpte_region.source_data_type = FADUMP_HPTE_REGION;
+	fdm->hpte_region.source_address = 0;
+	fdm->hpte_region.source_len = fw_dump.hpte_region_size;
+	fdm->hpte_region.destination_address = addr;
+	addr += fw_dump.hpte_region_size;
+
+	/* RMR region section */
+	fdm->rmr_region.request_flag = FADUMP_REQUEST_FLAG;
+	fdm->rmr_region.source_data_type = FADUMP_REAL_MODE_REGION;
+	fdm->rmr_region.source_address = RMR_START;
+	fdm->rmr_region.source_len = fw_dump.boot_memory_size;
+	fdm->rmr_region.destination_address = addr;
+	addr += fw_dump.boot_memory_size;
+
+	show_fadump_mem_struct(fdm);
+	return addr;
+}
+
 /**
  * calculate_reserve_size() - reserve variable boot area 5% of System RAM
  *

@@ -169,8 +330,15 @@ int __init fadump_reserve_mem(void)
 		fw_dump.fadump_enabled = 0;
 		return 0;
 	}
-	/* Initialize boot memory size */
-	fw_dump.boot_memory_size = calculate_reserve_size();
+	/*
+	 * Initialize boot memory size
+	 * If dump is active then we have already calculated the size during
+	 * first kernel.
+	 */
+	if (fdm_active)
+		fw_dump.boot_memory_size = fdm_active->rmr_region.source_len;
+	else
+		fw_dump.boot_memory_size = calculate_reserve_size();
 
 	/*
 	 * Calculate the memory boundary.

@@ -238,3 +406,165 @@ static int __init early_fadump_param(char *p)
 	return 0;
 }
 early_param("fadump", early_fadump_param);
+
+static void register_fw_dump(struct fadump_mem_struct *fdm)
+{
+	int rc;
+	unsigned int wait_time;
+
+	DBG("Registering for firmware-assisted kernel dump...\n");
+
+	/* TODO: Add upper time limit for the delay */
+	do {
+		rc = rtas_call(fw_dump.ibm_configure_kernel_dump, 3, 1, NULL,
+			FADUMP_REGISTER, fdm,
+			sizeof(struct fadump_mem_struct));
+
+		wait_time = rtas_busy_delay_time(rc);
+		if (wait_time)
+			mdelay(wait_time);
+
+	} while (wait_time);
+
+	switch (rc) {
+	case -1:
+		printk(KERN_ERR "Failed to register firmware-assisted kernel"
+			" dump. Hardware Error(%d).\n", rc);
+		break;
+	case -3:
+		printk(KERN_ERR "Failed to register firmware-assisted kernel"
+			" dump. Parameter Error(%d).\n", rc);
+		break;
+	case -9:
+		printk(KERN_ERR "firmware-assisted kernel dump is already "
+			" registered.");
+		fw_dump.dump_registered = 1;
+		break;
+	case 0:
+		printk(KERN_INFO "firmware-assisted kernel dump registration"
+			" is successful\n");
+		fw_dump.dump_registered = 1;
+		break;
+	}
+}
+
+static void register_fadump(void)
+{
+	/*
+	 * If no memory is reserved then we can not register for firmware-
+	 * assisted dump.
+	 */
+	if (!fw_dump.reserve_dump_area_size)
+		return;
+
+	/* Initialize the kernel dump memory structure for FAD registration. */
+	init_fadump_mem_struct(&fdm, fw_dump.reserve_dump_area_start);
+
+	/* register the future kernel dump with firmware. */
+	register_fw_dump(&fdm);
+}
+
+static ssize_t fadump_enabled_show(struct kobject *kobj,
+					struct kobj_attribute *attr,
+					char *buf)
+{
+	return sprintf(buf, "%d\n", fw_dump.fadump_enabled);
+}
+
+static ssize_t fadump_region_show(struct kobject *kobj,
+					struct kobj_attribute *attr,
+					char *buf)
+{
+	const struct fadump_mem_struct *fdm_ptr;
+	ssize_t n = 0;
+
+	if (!fw_dump.fadump_enabled)
+		return n;
+
+	if (fdm_active)
+		fdm_ptr = fdm_active;
+	else
+		fdm_ptr = &fdm;
+
+	n += sprintf(buf,
+			"CPU : [%#016llx-%#016llx] %#llx bytes, "
+			"Dumped: %#llx\n",
+			fdm_ptr->cpu_state_data.destination_address,
+			fdm_ptr->cpu_state_data.destination_address +
+			fdm_ptr->cpu_state_data.source_len - 1,
+			fdm_ptr->cpu_state_data.source_len,
+			fdm_ptr->cpu_state_data.bytes_dumped);
+	n += sprintf(buf + n,
+			"HPTE: [%#016llx-%#016llx] %#llx bytes, "
+			"Dumped: %#llx\n",
+			fdm_ptr->hpte_region.destination_address,
+			fdm_ptr->hpte_region.destination_address +
+			fdm_ptr->hpte_region.source_len - 1,
+			fdm_ptr->hpte_region.source_len,
+			fdm_ptr->hpte_region.bytes_dumped);
+	n += sprintf(buf + n,
+			"DUMP: [%#016llx-%#016llx] %#llx bytes, "
+			"Dumped: %#llx\n",
+			fdm_ptr->rmr_region.destination_address,
+			fdm_ptr->rmr_region.destination_address +
+			fdm_ptr->rmr_region.source_len - 1,
+			fdm_ptr->rmr_region.source_len,
+			fdm_ptr->rmr_region.bytes_dumped);
+
+	if (!fdm_active ||
+		(fw_dump.reserve_dump_area_start ==
+		fdm_ptr->cpu_state_data.destination_address))
+		return n;
+
+	/* Dump is active. Show reserved memory region. */
+	n += sprintf(buf + n,
+			"    : [%#016llx-%#016llx] %#llx bytes, "
+			"Dumped: %#llx\n",
+			(unsigned long long)fw_dump.reserve_dump_area_start,
+			fdm_ptr->cpu_state_data.destination_address - 1,
+			fdm_ptr->cpu_state_data.destination_address -
+			fw_dump.reserve_dump_area_start,
+			fdm_ptr->cpu_state_data.destination_address -
+			fw_dump.reserve_dump_area_start);
+	return n;
+}
+
+static struct kobj_attribute fadump_attr = __ATTR(fadump_enabled,
+						0444, fadump_enabled_show,
+						NULL);
+static struct kobj_attribute fadump_region_attr = __ATTR(fadump_region,
+						0444, fadump_region_show, NULL);
+
+static int fadump_init_sysfs(void)
+{
+	int rc = 0;
+
+	rc = sysfs_create_file(kernel_kobj, &fadump_attr.attr);
+	if (rc)
+		printk(KERN_ERR "fadump: unable to create sysfs file"
+			" (%d)\n", rc);
+
+	rc = sysfs_create_file(kernel_kobj, &fadump_region_attr.attr);
+	if (rc)
+		printk(KERN_ERR "fadump: unable to create sysfs file"
+			" (%d)\n", rc);
+	return rc;
+}
+subsys_initcall(fadump_init_sysfs);
+
+/*
+ * Prepare for firmware-assisted dump.
+ */
+int __init setup_fadump(void)
+{
+	if (!fw_dump.fadump_supported) {
+		printk(KERN_ERR "Firmware-assisted dump is not supported on"
+			" this hardware\n");
+		return 0;
+	}
+
+	fadump_show_config();
+	register_fadump();
+
+	return 1;
+}

@@ -63,6 +63,7 @@
 #include <asm/kexec.h>
 #include <asm/mmu_context.h>
 #include <asm/code-patching.h>
+#include <asm/fadump.h>
 
 #include "setup.h"
 

@@ -371,6 +372,13 @@ void __init setup_system(void)
 	rtas_initialize();
 #endif /* CONFIG_PPC_RTAS */
 
+#ifdef CONFIG_FA_DUMP
+	/*
+	 * Setup Firmware-assisted dump.
+	 */
+	setup_fadump();
+#endif
+
 	/*
 	 * Check if we have an initrd provided via the device-tree
 	 */

@@ -54,6 +54,7 @@
 #include <asm/spu.h>
 #include <asm/udbg.h>
 #include <asm/code-patching.h>
+#include <asm/fadump.h>
 
 #ifdef DEBUG
 #define DBG(fmt...) udbg_printf(fmt)

@@ -627,6 +628,16 @@ static void __init htab_initialize(void)
 		/* Using a hypervisor which owns the htab */
 		htab_address = NULL;
 		_SDR1 = 0; 
+#ifdef CONFIG_FA_DUMP
+		/*
+		 * If firmware assisted dump is active firmware preserves
+		 * the contents of htab along with entire partition memory.
+		 * Clear the htab if firmware assisted dump is active so
+		 * that we dont end up using old mappings.
+		 */
+		if (is_fadump_active() && ppc_md.hpte_clear_all)
+			ppc_md.hpte_clear_all();
+#endif
 	} else {
 		/* Find storage for the HPT.  Must be contiguous in
 		 * the absolute address space. On cell we want it to be

@@ -44,6 +44,9 @@
 #define FADUMP_UNREGISTER	2
 #define FADUMP_INVALIDATE	3
 
+/* Dump status flag */
+#define FADUMP_ERROR_FLAG	0x2000
+
 /* Kernel Dump section info */
 struct fadump_section {
 	u32	request_flag;

@@ -94,6 +97,7 @@ struct fw_dump {
 	unsigned long	boot_memory_size;
 	unsigned long	reserve_dump_area_start;
 	unsigned long	reserve_dump_area_size;
+	unsigned long	fadumphdr_addr;
 	int		ibm_configure_kernel_dump;
 
 	unsigned long	fadump_enabled:1;

@@ -102,6 +106,39 @@ struct fw_dump {
 	unsigned long	dump_registered:1;
 };
 
+/*
+ * Copy the ascii values for first 8 characters from a string into u64
+ * variable at their respective indexes.
+ * e.g.
+ *  The string "FADMPINF" will be converted into 0x4641444d50494e46
+ */
+static inline u64 str_to_u64(const char *str)
+{
+	u64 val = 0;
+	int i;
+
+	for (i = 0; i < sizeof(val); i++)
+		val = (*str) ? (val << 8) | *str++ : val << 8;
+	return val;
+}
+#define STR_TO_HEX(x)	str_to_u64(x)
+
+#define FADUMP_CRASH_INFO_MAGIC		STR_TO_HEX("FADMPINF")
+
+/* fadump crash info structure */
+struct fadump_crash_info_header {
+	u64		magic_number;
+	u64		elfcorehdr_addr;
+};
+
+/* Crash memory ranges */
+#define INIT_CRASHMEM_RANGES	(INIT_MEMBLOCK_REGIONS + 2)
+
+struct fad_crash_memory_ranges {
+	unsigned long long	base;
+	unsigned long long	size;
+};
+
 extern int early_init_dt_scan_fw_dump(unsigned long node,
 		const char *uname, int depth, void *data);
 extern int fadump_reserve_mem(void);