[patch 0/6] is_kdump_kernel() cleanup and related patches

[patch 0/6] is_kdump_kernel() cleanup and related patches

[Simon Horman]

Hi,

this patch series contains all the patches (that I am aware of)
that have recently been posted regarding kdump and in particular
is_kdump_kernel()

For Andrew Morton's benefit I have annotated each patch as to whether it
needs to be included in 2.6.27 or not. My conclusion is that while
I would like to see all of the patches included in 2.6.27, none
of them are particularly dire, and thus they could all wait until
post-2.6.27.

-- 
Horms

[patch 1/6] kdump: Make elfcorehdr_addr independent of CONFIG_PROC_VMCORE

[Simon Horman]

[-- Attachment #1: make-elfcorehdr_addr-independent-of-CONFIG_PROC_VMCORE.patch --]
[-- Type: text/plain, Size: 9809 bytes --]

From: Vivek Goyal <vgoyal@redhat.com>

o elfcorehdr_addr is used by not only the code under CONFIG_PROC_VMCORE but
  also by the code which is not inside CONFIG_PROC_VMCORE. For example,
  is_kdump_kernel() is used by powerpc code to determine if kernel is booting
  after a panic then use previous kernel's TCE table. So even if
  CONFIG_PROC_VMCORE is not set in second kernel, one should be able to
  correctly determine that we are booting after a panic and setup calgary
  iommu accordingly.

o So remove the assumption that elfcorehdr_addr is under CONFIG_PROC_VMCORE.

o Move definition of elfcorehdr_addr to arch dependent crash files.
  (Unfortunately crash dump does not have an arch independent file otherwise
   that would have been the best place).

o kexec.c is not the right place as one can Have CRASH_DUMP enabled in
  second kernel without KEXEC being enabled.

o I don't see sh setup code parsing the command line for elfcorehdr_addr. I 
  am wondering how does vmcore interface work on sh. Anyway, I am atleast
  defining elfcoredhr_addr so that compilation is not broken on sh.

Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Simon Horman <horms@verge.net.au>
Acked-by: Paul Mundt <lethal@linux-sh.org>
---

 arch/ia64/kernel/crash_dump.c    |    4 ++++
 arch/ia64/kernel/setup.c         |    9 ++++++++-
 arch/powerpc/kernel/crash_dump.c |   10 ++++++++--
 arch/sh/kernel/crash_dump.c      |    3 +++
 arch/x86/kernel/crash_dump_32.c  |    3 +++
 arch/x86/kernel/crash_dump_64.c  |    3 +++
 arch/x86/kernel/setup.c          |    8 +++++++-
 fs/proc/vmcore.c                 |    3 ---
 include/linux/crash_dump.h       |   14 ++++++++++----
 9 files changed, 46 insertions(+), 11 deletions(-)

Andrew, this patch is a clean-up and is not required by other patches,
so it could wait until post 2.6.27 if you prefer. However, it is a
valuable cleanup that removes messiness and should remove some of
the confusion around elfcorehdr_addr and is_kdump_kernel(). So I
would prefer it to be included in 2.6.27.

sh setup code has been added in a separate patch that will
be merged by Paul Mundt.

- Simon Horman

diff -puN include/linux/crash_dump.h~remove-elfcore-hdr-addr-definition-vmcore include/linux/crash_dump.h
--- linux-2.6.27-pre-rc1/include/linux/crash_dump.h~remove-elfcore-hdr-addr-definition-vmcore	2008-07-28 12:00:44.000000000 -0400
+++ linux-2.6.27-pre-rc1-root/include/linux/crash_dump.h	2008-07-28 12:00:56.000000000 -0400
@@ -9,11 +9,7 @@
 
 #define ELFCORE_ADDR_MAX	(-1ULL)
 
-#ifdef CONFIG_PROC_VMCORE
 extern unsigned long long elfcorehdr_addr;
-#else
-static const unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX;
-#endif
 
 extern ssize_t copy_oldmem_page(unsigned long, char *, size_t,
 						unsigned long, int);
@@ -28,6 +24,16 @@ extern struct proc_dir_entry *proc_vmcor
 
 #define vmcore_elf_check_arch(x) (elf_check_arch(x) || vmcore_elf_check_arch_cross(x))
 
+/*
+ * is_kdump_kernel() checks whether this kernel is booting after a panic of
+ * previous kernel or not. This is determined by checking if previous kernel
+ * has passed the elf core header address on command line.
+ *
+ * This is not just a test if CONFIG_CRASH_DUMP is enabled or not. It will
+ * return 1 if CONFIG_CRASH_DUMP=y and if kernel is booting after a panic of
+ * previous kernel.
+ */
+
 static inline int is_kdump_kernel(void)
 {
 	return (elfcorehdr_addr != ELFCORE_ADDR_MAX) ? 1 : 0;
diff -puN fs/proc/vmcore.c~remove-elfcore-hdr-addr-definition-vmcore fs/proc/vmcore.c
--- linux-2.6.27-pre-rc1/fs/proc/vmcore.c~remove-elfcore-hdr-addr-definition-vmcore	2008-07-28 09:19:50.000000000 -0400
+++ linux-2.6.27-pre-rc1-root/fs/proc/vmcore.c	2008-07-28 09:20:10.000000000 -0400
@@ -32,9 +32,6 @@ static size_t elfcorebuf_sz;
 /* Total size of vmcore file. */
 static u64 vmcore_size;
 
-/* Stores the physical address of elf header of crash image. */
-unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX;
-
 struct proc_dir_entry *proc_vmcore = NULL;
 
 /* Reads a page from the oldmem device from given offset. */
diff -puN arch/x86/kernel/crash_dump_32.c~remove-elfcore-hdr-addr-definition-vmcore arch/x86/kernel/crash_dump_32.c
--- linux-2.6.27-pre-rc1/arch/x86/kernel/crash_dump_32.c~remove-elfcore-hdr-addr-definition-vmcore	2008-07-29 05:28:26.000000000 -0400
+++ linux-2.6.27-pre-rc1-root/arch/x86/kernel/crash_dump_32.c	2008-07-29 05:28:26.000000000 -0400
@@ -13,6 +13,9 @@
 
 static void *kdump_buf_page;
 
+/* Stores the physical address of elf header of crash image. */
+unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX;
+
 /**
  * copy_oldmem_page - copy one page from "oldmem"
  * @pfn: page frame number to be copied
diff -puN arch/x86/kernel/crash_dump_64.c~remove-elfcore-hdr-addr-definition-vmcore arch/x86/kernel/crash_dump_64.c
--- linux-2.6.27-pre-rc1/arch/x86/kernel/crash_dump_64.c~remove-elfcore-hdr-addr-definition-vmcore	2008-07-29 05:28:26.000000000 -0400
+++ linux-2.6.27-pre-rc1-root/arch/x86/kernel/crash_dump_64.c	2008-07-29 05:28:26.000000000 -0400
@@ -11,6 +11,9 @@
 #include <asm/uaccess.h>
 #include <asm/io.h>
 
+/* Stores the physical address of elf header of crash image. */
+unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX;
+
 /**
  * copy_oldmem_page - copy one page from "oldmem"
  * @pfn: page frame number to be copied
diff -puN arch/x86/kernel/setup.c~remove-elfcore-hdr-addr-definition-vmcore arch/x86/kernel/setup.c
--- linux-2.6.27-pre-rc1/arch/x86/kernel/setup.c~remove-elfcore-hdr-addr-definition-vmcore	2008-07-29 05:28:26.000000000 -0400
+++ linux-2.6.27-pre-rc1-root/arch/x86/kernel/setup.c	2008-07-29 05:28:26.000000000 -0400
@@ -558,7 +558,13 @@ static void __init reserve_standard_io_r
 
 }
 
-#ifdef CONFIG_PROC_VMCORE
+/*
+ * Note: elfcorehdr_addr is not just limited to vmcore. It is also used by
+ * is_kdump_kernel() to determine if we are booting after a panic. Hence
+ * ifdef it under CONFIG_CRASH_DUMP and not CONFIG_PROC_VMCORE.
+ */
+
+#ifdef CONFIG_CRASH_DUMP
 /* elfcorehdr= specifies the location of elf core header
  * stored by the crashed kernel. This option will be passed
  * by kexec loader to the capture kernel.
diff -puN arch/powerpc/kernel/crash_dump.c~remove-elfcore-hdr-addr-definition-vmcore arch/powerpc/kernel/crash_dump.c
--- linux-2.6.27-pre-rc1/arch/powerpc/kernel/crash_dump.c~remove-elfcore-hdr-addr-definition-vmcore	2008-07-29 05:30:33.000000000 -0400
+++ linux-2.6.27-pre-rc1-root/arch/powerpc/kernel/crash_dump.c	2008-07-29 05:30:33.000000000 -0400
@@ -27,6 +27,9 @@
 #define DBG(fmt...)
 #endif
 
+/* Stores the physical address of elf header of crash image. */
+unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX;
+
 void __init reserve_kdump_trampoline(void)
 {
 	lmb_reserve(0, KDUMP_RESERVE_LIMIT);
@@ -66,7 +69,11 @@ void __init setup_kdump_trampoline(void)
 	DBG(" <- setup_kdump_trampoline()\n");
 }
 
-#ifdef CONFIG_PROC_VMCORE
+/*
+ * Note: elfcorehdr_addr is not just limited to vmcore. It is also used by
+ * is_kdump_kernel() to determine if we are booting after a panic. Hence
+ * ifdef it under CONFIG_CRASH_DUMP and not CONFIG_PROC_VMCORE.
+ */
 static int __init parse_elfcorehdr(char *p)
 {
 	if (p)
@@ -75,7 +82,6 @@ static int __init parse_elfcorehdr(char 
 	return 1;
 }
 __setup("elfcorehdr=", parse_elfcorehdr);
-#endif
 
 static int __init parse_savemaxmem(char *p)
 {
diff -puN arch/ia64/kernel/setup.c~remove-elfcore-hdr-addr-definition-vmcore arch/ia64/kernel/setup.c
--- linux-2.6.27-pre-rc1/arch/ia64/kernel/setup.c~remove-elfcore-hdr-addr-definition-vmcore	2008-07-29 05:30:54.000000000 -0400
+++ linux-2.6.27-pre-rc1-root/arch/ia64/kernel/setup.c	2008-07-29 05:30:54.000000000 -0400
@@ -478,7 +478,12 @@ static __init int setup_nomca(char *s)
 }
 early_param("nomca", setup_nomca);
 
-#ifdef CONFIG_PROC_VMCORE
+/*
+ * Note: elfcorehdr_addr is not just limited to vmcore. It is also used by
+ * is_kdump_kernel() to determine if we are booting after a panic. Hence
+ * ifdef it under CONFIG_CRASH_DUMP and not CONFIG_PROC_VMCORE.
+ */
+#ifdef CONFIG_CRASH_DUMP
 /* elfcorehdr= specifies the location of elf core header
  * stored by the crashed kernel.
  */
@@ -491,7 +496,9 @@ static int __init parse_elfcorehdr(char 
 	return 0;
 }
 early_param("elfcorehdr", parse_elfcorehdr);
+#endif
 
+#ifdef CONFIG_PROC_VMCORE
 int __init reserve_elfcorehdr(unsigned long *start, unsigned long *end)
 {
 	unsigned long length;
diff -puN arch/ia64/kernel/crash_dump.c~remove-elfcore-hdr-addr-definition-vmcore arch/ia64/kernel/crash_dump.c
--- linux-2.6.27-pre-rc1/arch/ia64/kernel/crash_dump.c~remove-elfcore-hdr-addr-definition-vmcore	2008-07-29 05:37:10.000000000 -0400
+++ linux-2.6.27-pre-rc1-root/arch/ia64/kernel/crash_dump.c	2008-07-29 05:37:58.000000000 -0400
@@ -8,10 +8,14 @@
 
 #include <linux/errno.h>
 #include <linux/types.h>
+#include <linux/crash_dump.h>
 
 #include <asm/page.h>
 #include <asm/uaccess.h>
 
+/* Stores the physical address of elf header of crash image. */
+unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX;
+
 /**
  * copy_oldmem_page - copy one page from "oldmem"
  * @pfn: page frame number to be copied
diff -puN arch/sh/kernel/crash_dump.c~remove-elfcore-hdr-addr-definition-vmcore arch/sh/kernel/crash_dump.c
--- linux-2.6.27-pre-rc1/arch/sh/kernel/crash_dump.c~remove-elfcore-hdr-addr-definition-vmcore	2008-07-29 05:31:04.000000000 -0400
+++ linux-2.6.27-pre-rc1-root/arch/sh/kernel/crash_dump.c	2008-07-29 05:31:04.000000000 -0400
@@ -10,6 +10,9 @@
 #include <linux/io.h>
 #include <asm/uaccess.h>
 
+/* Stores the physical address of elf header of crash image. */
+unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX;
+
 /**
  * copy_oldmem_page - copy one page from "oldmem"
  * @pfn: page frame number to be copied
_

-- 

-- 
Horms

[patch 2/6] kdump: sh: parse elfcorehdr command line argument

[Simon Horman]

[-- Attachment #1: sh-parse_elfcorehdr.patch --]
[-- Type: text/plain, Size: 1525 bytes --]

A quick cut and paste from other architectures to allow SH
to parse the elfcorehdr command line argument which is required
for both is_kdump_kernel() and vmcore to function.
(the former is as yet unused on SH).

Tested compilation only

Signed-off-by: Simon Horman <horms@verge.net.au>

--- 
Andrew, Paul Mundt has aggreed to apply this patch

Index: linux-2.6/arch/sh/kernel/setup.c
===================================================================
--- linux-2.6.orig/arch/sh/kernel/setup.c	2008-07-30 09:33:45.000000000 +1000
+++ linux-2.6/arch/sh/kernel/setup.c	2008-07-30 09:37:45.000000000 +1000
@@ -25,6 +25,7 @@
 #include <linux/smp.h>
 #include <linux/err.h>
 #include <linux/debugfs.h>
+#include <linux/crash_dump.h>
 #include <asm/uaccess.h>
 #include <asm/io.h>
 #include <asm/page.h>
@@ -286,6 +287,25 @@ static void __init setup_memory(void)
 extern void __init setup_memory(void);
 #endif
 
+/*
+ * Note: elfcorehdr_addr is not just limited to vmcore. It is also used by
+ * is_kdump_kernel() to determine if we are booting after a panic. Hence
+ * ifdef it under CONFIG_CRASH_DUMP and not CONFIG_PROC_VMCORE.
+ */
+#ifdef CONFIG_CRASH_DUMP
+/* elfcorehdr= specifies the location of elf core header
+ * stored by the crashed kernel.
+ */
+static int __init parse_elfcorehdr(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
+	elfcorehdr_addr = memparse(arg, &arg);
+	return 0;
+}
+early_param("elfcorehdr", parse_elfcorehdr);
+#endif
+
 void __init setup_arch(char **cmdline_p)
 {
 	enable_mmu();

-- 

-- 
Horms

[patch 3/6] kdump: update elfcorehdr documentation to reflect supported architectures

[Simon Horman]

[-- Attachment #1: elfcorehdr-doc.patch --]
[-- Type: text/plain, Size: 955 bytes --]

IA64, PPC and SH also support the elfcorehdr command line.

Signed-off-by: Simon Horman <horms@verge.net.au>
Acked-by: Vivek Goyal <vgoyal@redhat.com>

--- 
Andrew, as this is a very minor documentation fix it can wait until
after 2.6.27 if you prefer.

Index: linux-2.6/Documentation/kernel-parameters.txt
===================================================================
--- linux-2.6.orig/Documentation/kernel-parameters.txt	2008-07-30 09:48:40.000000000 +1000
+++ linux-2.6/Documentation/kernel-parameters.txt	2008-07-30 09:48:45.000000000 +1000
@@ -688,7 +688,7 @@ and is between 256 and 4096 characters. 
 			See Documentation/block/as-iosched.txt and
 			Documentation/block/deadline-iosched.txt for details.
 
-	elfcorehdr=	[X86-32, X86_64]
+	elfcorehdr=	[IA64,PPC,SH,X86-32,X86_64]
 			Specifies physical address of start of kernel core
 			image elf header. Generally kexec loader will
 			pass this option to capture kernel.

-- 

-- 
Horms

[patch 4/6] kdump: use is_kdump_kernel() in sba_init()

[Simon Horman]

[-- Attachment #1: ia64-use-is_kdump_kernel.patch --]
[-- Type: text/plain, Size: 1659 bytes --]

o Make use of is_kdump_kernel() rather than checking elfcorehdr_addr directly.

o Remove CONFIG_CRASH_DUMP as is_kdump_kernel() is safe to call anywhere

o Remove CONFIG_PROC_FS as it is bogus, the check
  should occur regardless of if CONFIG_PROC_FS is set or not.

Signed-off-by: Simon Horman <horms@verge.net.au>
Acked-by: Vivek Goyal <vgoyal@redhat.com>

--- 
Andrew, this patch fixes bug in the (unlikely) case where
an ia64 crashdump kernel does not have CONFIG_PROC_FS set.
I think it is worth including in 2.6.27. But breakage cases are
likely to be minimal to non-existent, so I am comfortable
with post 2.6.27 too.

Index: linux-2.6/arch/ia64/hp/common/sba_iommu.c
===================================================================
--- linux-2.6.orig/arch/ia64/hp/common/sba_iommu.c	2008-07-29 10:57:17.000000000 +1000
+++ linux-2.6/arch/ia64/hp/common/sba_iommu.c	2008-07-29 10:57:51.000000000 +1000
@@ -2070,14 +2070,13 @@ sba_init(void)
 	if (!ia64_platform_is("hpzx1") && !ia64_platform_is("hpzx1_swiotlb"))
 		return 0;
 
-#if defined(CONFIG_IA64_GENERIC) && defined(CONFIG_CRASH_DUMP) && \
-        defined(CONFIG_PROC_FS)
+#if defined(CONFIG_IA64_GENERIC)
 	/* If we are booting a kdump kernel, the sba_iommu will
 	 * cause devices that were not shutdown properly to MCA
 	 * as soon as they are turned back on.  Our only option for
 	 * a successful kdump kernel boot is to use the swiotlb.
 	 */
-	if (elfcorehdr_addr < ELFCORE_ADDR_MAX) {
+	if (is_kdump_kernel()) {
 		if (swiotlb_late_init_with_default_size(64 * (1<<20)) != 0)
 			panic("Unable to initialize software I/O TLB:"
 				  " Try machvec=dig boot option");

-- 

-- 
Horms

[patch 5/6] kdump: add is_vmcore_usable() and vmcore_unusable()

[Simon Horman]

[-- Attachment #1: is_vmcore_usable.patch --]
[-- Type: text/plain, Size: 3642 bytes --]

The usage of elfcorehdr_addr has changed recently
such that being set to ELFCORE_ADDR_MAX is used by
is_kdump_kernel() to indicate if the code is executing
in a kernel executed as a crash kernel.

However, arch/ia64/kernel/setup.c:reserve_elfcorehdr will
rest elfcorehdr_addr to ELFCORE_ADDR_MAX on error,
which means any subsequent calls to is_kdump_kernel()
will return 0, even though they should return 1.

Ok, at this point in time there are no subsequent calls,
but I think its fair to say that there is ample scope for error
or at the very least confusion.

This patch add an extra state, ELFCORE_ADDR_ERR, which
indicates that elfcorehdr_addr was passed on the command line,
and thus execution is taking place in a crashdump kernel,
but vmcore can't be used for some reason. This is tested
for using is_vmcore_usable() and set using vmcore_unusable().
A subsequent patch makes use of this new code.

To summarise, the states that elfcorehdr_addr can now be in are as follows:

ELFCORE_ADDR_MAX: not a crashdump kernel
ELFCORE_ADDR_ERR: crashdump kernel but vmcore is unusable
any other value:  crash dump kernel and vmcore is usable

Signed-off-by: Simon Horman <horms@verge.net.au>

--- 

 fs/proc/vmcore.c           |    2 +-
 include/linux/crash_dump.h |   26 ++++++++++++++++++++++++++
 2 files changed, 27 insertions(+), 1 deletion(-)

Wed, 30 Jul 2008 09:10:37 +1000
* Fixed logic bug in vmcore_unusable().
  if (!is_kdump_kernel()) is correct
  if (is_kdump_kernel()) is not

Andrew, this patch can wait until after 2.6.27 as it is (minor)
infrastructure that only has one usage case, and that usage case
is not buggy at this time.

is_vmcore_usable() and vmcore_unusable() are used in a subsequent patch
in this series

Index: linux-2.6/include/linux/crash_dump.h
===================================================================
--- linux-2.6.orig/include/linux/crash_dump.h	2008-07-30 09:05:50.000000000 +1000
+++ linux-2.6/include/linux/crash_dump.h	2008-07-30 09:08:58.000000000 +1000
@@ -8,6 +8,7 @@
 #include <linux/proc_fs.h>
 
 #define ELFCORE_ADDR_MAX	(-1ULL)
+#define ELFCORE_ADDR_ERR	(-2ULL)
 
 extern unsigned long long elfcorehdr_addr;
 
@@ -42,5 +43,30 @@ static inline int is_kdump_kernel(void)
 static inline int is_kdump_kernel(void) { return 0; }
 #endif /* CONFIG_CRASH_DUMP */
 
+#ifdef CONFIG_PROC_VMCORE
+/* is_vmcore_usable() checks if the kernel is booting after a panic and
+ * the vmcore region is usable.
+ *
+ * This makes use of the fact that due to alignment 1 is not
+ * a valid pointer, much in the vain of IS_ERR(), except
+ * dealing directly with an unsigned long long rather than a pointer.
+ */
+
+static inline int is_vmcore_usable(void)
+{
+	return is_kdump_kernel() && elfcorehdr_addr != ELFCORE_ADDR_ERR ? 1 : 0;
+}
+
+/* vmcore_unusable() marks the vmcore as unusable,
+ * without disturbing the logic of is_kdump_kernel()
+ */
+
+static inline void vmcore_unusable(void)
+{
+	if (is_kdump_kernel())
+		elfcorehdr_addr = ELFCORE_ADDR_ERR;
+}
+#endif /* CONFIG_PROC_VMCORE */
+
 extern unsigned long saved_max_pfn;
 #endif /* LINUX_CRASHDUMP_H */
Index: linux-2.6/fs/proc/vmcore.c
===================================================================
--- linux-2.6.orig/fs/proc/vmcore.c	2008-07-30 09:05:50.000000000 +1000
+++ linux-2.6/fs/proc/vmcore.c	2008-07-30 09:07:07.000000000 +1000
@@ -650,7 +650,7 @@ static int __init vmcore_init(void)
 	int rc = 0;
 
 	/* If elfcorehdr= has been passed in cmdline, then capture the dump.*/
-	if (!(elfcorehdr_addr < ELFCORE_ADDR_MAX))
+	if (!(is_vmcore_usable()))
 		return rc;
 	rc = parse_crash_elf_headers();
 	if (rc) {

-- 

-- 
Horms

Re: [patch 5/6] kdump: add is_vmcore_usable() and vmcore_unusable()

[Simon Horman]

The usage of elfcorehdr_addr has changed recently
such that being set to ELFCORE_ADDR_MAX is used by
is_kdump_kernel() to indicate if the code is executing
in a kernel executed as a crash kernel.

However, arch/ia64/kernel/setup.c:reserve_elfcorehdr will
rest elfcorehdr_addr to ELFCORE_ADDR_MAX on error,
which means any subsequent calls to is_kdump_kernel()
will return 0, even though they should return 1.

Ok, at this point in time there are no subsequent calls,
but I think its fair to say that there is ample scope for error
or at the very least confusion.

This patch add an extra state, ELFCORE_ADDR_ERR, which
indicates that elfcorehdr_addr was passed on the command line,
and thus execution is taking place in a crashdump kernel,
but vmcore can't be used for some reason. This is tested
for using is_vmcore_usable() and set using vmcore_unusable().
A subsequent patch makes use of this new code.

To summarise, the states that elfcorehdr_addr can now be in are as follows:

ELFCORE_ADDR_MAX: not a crashdump kernel
ELFCORE_ADDR_ERR: crashdump kernel but vmcore is unusable
any other value:  crash dump kernel and vmcore is usable

Signed-off-by: Simon Horman <horms@verge.net.au>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

--- 

 fs/proc/vmcore.c           |    2 +-
 include/linux/crash_dump.h |   26 ++++++++++++++++++++++++++
 2 files changed, 27 insertions(+), 1 deletion(-)

Wed, 30 Jul 2008 09:10:37 +1000
* Fixed logic bug in vmcore_unusable().
  if (!is_kdump_kernel()) is correct
  if (is_kdump_kernel()) is not

Wed, 06 Aug 2008 09:53:44 +1000
* is_vmcore_usable() and vmcore_unusable may be used when
  CONFIG_CRASH_DUMP is defined but CONFIG_PROC_VMCORE is not.

  This update is needed by
  "kdump, ia64: always reserve elfcore header memory in crash kernel"
  http://lists.infradead.org/pipermail/kexec/2008-August/002294.html

Index: linux-2.6/include/linux/crash_dump.h
===================================================================
--- linux-2.6.orig/include/linux/crash_dump.h	2008-08-05 18:10:21.000000000 +1000
+++ linux-2.6/include/linux/crash_dump.h	2008-08-05 18:11:14.000000000 +1000
@@ -8,6 +8,7 @@
 #include <linux/proc_fs.h>
 
 #define ELFCORE_ADDR_MAX	(-1ULL)
+#define ELFCORE_ADDR_ERR	(-2ULL)
 
 extern unsigned long long elfcorehdr_addr;
 
@@ -38,6 +39,29 @@ static inline int is_kdump_kernel(void)
 {
 	return (elfcorehdr_addr != ELFCORE_ADDR_MAX) ? 1 : 0;
 }
+
+/* is_vmcore_usable() checks if the kernel is booting after a panic and
+ * the vmcore region is usable.
+ *
+ * This makes use of the fact that due to alignment -2ULL is not
+ * a valid pointer, much in the vain of IS_ERR(), except
+ * dealing directly with an unsigned long long rather than a pointer.
+ */
+
+static inline int is_vmcore_usable(void)
+{
+	return is_kdump_kernel() && elfcorehdr_addr != ELFCORE_ADDR_ERR ? 1 : 0;
+}
+
+/* vmcore_unusable() marks the vmcore as unusable,
+ * without disturbing the logic of is_kdump_kernel()
+ */
+
+static inline void vmcore_unusable(void)
+{
+	if (is_kdump_kernel())
+		elfcorehdr_addr = ELFCORE_ADDR_ERR;
+}
 #else /* !CONFIG_CRASH_DUMP */
 static inline int is_kdump_kernel(void) { return 0; }
 #endif /* CONFIG_CRASH_DUMP */
Index: linux-2.6/fs/proc/vmcore.c
===================================================================
--- linux-2.6.orig/fs/proc/vmcore.c	2008-08-05 18:10:21.000000000 +1000
+++ linux-2.6/fs/proc/vmcore.c	2008-08-05 18:10:25.000000000 +1000
@@ -650,7 +650,7 @@ static int __init vmcore_init(void)
 	int rc = 0;
 
 	/* If elfcorehdr= has been passed in cmdline, then capture the dump.*/
-	if (!(elfcorehdr_addr < ELFCORE_ADDR_MAX))
+	if (!(is_vmcore_usable()))
 		return rc;
 	rc = parse_crash_elf_headers();
 	if (rc) {

[patch 6/6] kdump: use is_vmcore_usable() and vmcore_unusable() in reserve_elfcorehdr()

[Simon Horman]

[-- Attachment #1: use-is_vmcore_usable.patch --]
[-- Type: text/plain, Size: 1169 bytes --]

After recent changes setting elfcorehdr_addr to ELFCORE_ADDR_MAX
will cause is_kdump_kernel() to return 0 when it should return 1.
Instead use vmcore_unusable(), which has been added for this purpose.

Signed-off-by: Simon Horman <horms@verge.net.au>

--- 

Andrew, this can wait until post 2.6.27 as although the current
state of affairs is confusing, a bug does not manifest at this time
as there are no callers of is_kdump_kernel on ia64 at this time.

is_vmcore_usable() and vmcore_unusable() are provided by a previous
patch in this series

Index: linux-2.6/arch/ia64/kernel/setup.c
===================================================================
--- linux-2.6.orig/arch/ia64/kernel/setup.c	2008-07-31 09:41:15.000000000 +1000
+++ linux-2.6/arch/ia64/kernel/setup.c	2008-07-31 09:41:36.000000000 +1000
@@ -509,11 +509,11 @@ int __init reserve_elfcorehdr(unsigned l
 	 * to work properly.
 	 */
 
-	if (elfcorehdr_addr >= ELFCORE_ADDR_MAX)
+	if (!is_vmcore_usable())
 		return -EINVAL;
 
 	if ((length = vmcore_find_descriptor_size(elfcorehdr_addr)) == 0) {
-		elfcorehdr_addr = ELFCORE_ADDR_MAX;
+		vmcore_unusable();
 		return -EINVAL;
 	}
 

-- 

-- 
Horms