Message-ID: <20120216111421.4733.17834.stgit@mars>
Date:	Thu, 16 Feb 2012 16:44:22 +0530
From:	Mahesh J Salgaonkar <mahesh@...ux.vnet.ibm.com>
To:	linuxppc-dev <linuxppc-dev@...abs.org>,
	Linux Kernel <linux-kernel@...r.kernel.org>,
	Benjamin Herrenschmidt <benh@...nel.crashing.org>
Cc:	Amerigo Wang <amwang@...hat.com>,
	Kexec-ml <kexec@...ts.infradead.org>,
	Milton Miller <miltonm@....com>,
	Haren Myneni <hbabu@...ibm.com>,
	Randy Dunlap <rdunlap@...otime.net>,
	Paul Mackerras <paulus@...ba.org>,
	"Eric W. Biederman" <ebiederm@...ssion.com>,
	Ananth Narayan <ananth@...ibm.com>,
	Vivek Goyal <vgoyal@...hat.com>,
	Anton Blanchard <anton@...ba.org>
Subject: [RFC PATCH v7 02/10] fadump: Reserve the memory for firmware assisted
 dump.

From: Mahesh Salgaonkar <mahesh@...ux.vnet.ibm.com>

Reserve the memory during early boot to preserve CPU state data, HPTE region
and RMA (real mode area) region data in case of kernel crash. At the time of
crash, powerpc firmware will store CPU state data, HPTE region data and move
RMA region data to the reserved memory area.

If the firmware-assisted dump fails to reserve the memory, then fallback
to existing kexec-based kdump.

Most of the code implementation to reserve memory has been
adapted from phyp assisted dump implementation written by Linas Vepstas
and Manish Ahuja

This patch also introduces a config option CONFIG_FA_DUMP for firmware
assisted dump feature on Powerpc (ppc64) architecture.

Signed-off-by: Mahesh Salgaonkar <mahesh@...ux.vnet.ibm.com>
---
 arch/powerpc/Kconfig              |   13 ++
 arch/powerpc/include/asm/fadump.h |   71 +++++++++++
 arch/powerpc/kernel/Makefile      |    1 
 arch/powerpc/kernel/fadump.c      |  246 +++++++++++++++++++++++++++++++++++++
 arch/powerpc/kernel/prom.c        |   15 ++
 5 files changed, 345 insertions(+), 1 deletions(-)
 create mode 100644 arch/powerpc/include/asm/fadump.h
 create mode 100644 arch/powerpc/kernel/fadump.c

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 1919634..afa4dab 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -386,6 +386,19 @@ config PHYP_DUMP
 
 	  If unsure, say "N"
 
+config FA_DUMP
+	bool "Firmware-assisted dump"
+	depends on PPC64 && PPC_RTAS && CRASH_DUMP
+	help
+	  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. Firmware-assisted dump
+	  is meant to be a kdump replacement offering robustness and
+	  speed not possible without system firmware assistance.
+
+	  If unsure, say "N"
+
 config IRQ_ALL_CPUS
 	bool "Distribute interrupts on all CPUs by default"
 	depends on SMP && !MV64360
diff --git a/arch/powerpc/include/asm/fadump.h b/arch/powerpc/include/asm/fadump.h
new file mode 100644
index 0000000..7be25d3
--- /dev/null
+++ b/arch/powerpc/include/asm/fadump.h
@@ -0,0 +1,71 @@
+/*
+ * 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@...ux.vnet.ibm.com>
+ */
+
+#ifndef __PPC64_FA_DUMP_H__
+#define __PPC64_FA_DUMP_H__
+
+#ifdef CONFIG_FA_DUMP
+
+/*
+ * The RMA region will be saved for later dumping when kernel crashes.
+ * RMA is Real Mode Area, the first block of logical memory address owned
+ * by logical partition, containing the storage that may be accessed with
+ * translate off.
+ */
+#define RMA_START	0x0
+#define RMA_END		(ppc64_rma_size)
+
+/*
+ * On some Power systems where RMO is 128MB, it still requires minimum of
+ * 256MB for kernel to boot successfully. When kdump infrastructure is
+ * configured to save vmcore over network, we run into OOM issue while
+ * loading modules related to network setup. Hence we need aditional 64M
+ * of memory to avoid OOM issue.
+ */
+#define MIN_BOOT_MEM	(((RMA_END < (0x1UL << 28)) ? (0x1UL << 28) : RMA_END) \
+			+ (0x1UL << 26))
+
+/* 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;
+	/* cmd line option during boot */
+	unsigned long	reserve_bootvar;
+
+	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
diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile
index ee728e4..391bf7e 100644
--- a/arch/powerpc/kernel/Makefile
+++ b/arch/powerpc/kernel/Makefile
@@ -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
diff --git a/arch/powerpc/kernel/fadump.c b/arch/powerpc/kernel/fadump.c
new file mode 100644
index 0000000..deb276a
--- /dev/null
+++ b/arch/powerpc/kernel/fadump.c
@@ -0,0 +1,246 @@
+/*
+ * 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@...ux.vnet.ibm.com>
+ */
+
+#undef DEBUG
+#define pr_fmt(fmt) "fadump: " fmt
+
+#include <linux/string.h>
+#include <linux/memblock.h>
+
+#include <asm/page.h>
+#include <asm/prom.h>
+#include <asm/rtas.h>
+#include <asm/fadump.h>
+
+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)
+{
+	__be32 *sections;
+	int i, num_sections;
+	unsigned long size;
+	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.
+	 * For each dump section type supported, a 32bit cell which defines
+	 * the ID of a supported section followed by two 32 bit cells which
+	 * gives teh size of the section in bytes.
+	 */
+	sections = of_get_flat_dt_prop(node, "ibm,configure-kernel-dump-sizes",
+					&size);
+
+	if (!sections)
+		return 0;
+
+	num_sections = size / (3 * sizeof(u32));
+
+	for (i = 0; i < num_sections; i++, sections += 3) {
+		u32 type = (u32)of_read_number(sections, 1);
+
+		switch (type) {
+		case FADUMP_CPU_STATE_DATA:
+			fw_dump.cpu_state_data_size =
+					of_read_ulong(&sections[1], 2);
+			break;
+		case FADUMP_HPTE_REGION:
+			fw_dump.hpte_region_size =
+					of_read_ulong(&sections[1], 2);
+			break;
+		}
+	}
+	return 1;
+}
+
+/**
+ * fadump_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 fadump_calculate_reserve_size(void)
+{
+	unsigned long size;
+
+	/*
+	 * Check if the size is specified through fadump_reserve_mem= cmdline
+	 * option. If yes, then use that.
+	 */
+	if (fw_dump.reserve_bootvar)
+		return fw_dump.reserve_bootvar;
+
+	/* divide by 20 to get 5% of value */
+	size = memblock_end_of_DRAM() / 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 > MIN_BOOT_MEM ? size : MIN_BOOT_MEM);
+}
+
+/*
+ * Calculate the total memory size required to be reserved for
+ * firmware-assisted dump registration.
+ */
+static unsigned long get_fadump_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_INFO "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 = fadump_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_fadump_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_fadump_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 (strncmp(p, "on", 2) == 0)
+		fw_dump.fadump_enabled = 1;
+	else if (strncmp(p, "off", 3) == 0)
+		fw_dump.fadump_enabled = 0;
+
+	return 0;
+}
+early_param("fadump", early_fadump_param);
+
+/* Look for fadump_reserve_mem= cmdline option */
+static int __init early_fadump_reserve_mem(char *p)
+{
+	if (p)
+		fw_dump.reserve_bootvar = memparse(p, &p);
+	return 0;
+}
+early_param("fadump_reserve_mem", early_fadump_reserve_mem);
diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index abe405d..70222b3 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -55,6 +55,7 @@
 #include <asm/phyp_dump.h>
 #include <asm/kexec.h>
 #include <asm/opal.h>
+#include <asm/fadump.h>
 
 #include <mm/mmu_decl.h>
 
@@ -719,6 +720,11 @@ 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
+
 	/* Pre-initialize the cmd_line with the content of boot_commmand_line,
 	 * which will be empty except when the content of the variable has
 	 * been overriden by a bootloading mechanism. This happens typically
@@ -750,7 +756,14 @@ 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();
 

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists