[v15,15/20] arm64: kdump: reserve memory for crash dump kernel

Commit Message

Geoff Levand March 14, 2016, 5:48 p.m. UTC

From: AKASHI Takahiro <[e-mail]>

On the startup of primary kernel, the memory region used by crash dump
kernel must be specified by "crashkernel=" kernel parameter.
reserve_crashkernel() will allocate and reserve the region for later use.

User space tools, like kexec-tools, will be able to find that region marked
as "Crash kernel" in /proc/iomem.

Signed-off-by: AKASHI Takahiro <[e-mail]>
Signed-off-by: Mark Salter <[e-mail]>
Signed-off-by: Pratyush Anand <[e-mail]>
---
 arch/arm64/kernel/setup.c |  7 +++++-
 arch/arm64/mm/init.c      | 60 +++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 66 insertions(+), 1 deletion(-)

Comments

James Morse March 18, 2016, 6:08 p.m. UTC | #1

Hi!

On 14/03/16 17:48, Geoff Levand wrote:
> From: AKASHI Takahiro <[e-mail]>
> 
> On the startup of primary kernel, the memory region used by crash dump
> kernel must be specified by "crashkernel=" kernel parameter.
> reserve_crashkernel() will allocate and reserve the region for later use.
> 
> User space tools, like kexec-tools, will be able to find that region marked
> as "Crash kernel" in /proc/iomem.

[NB: Re-ordered diff hunks ]

> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
> index 7802f21..4edf181 100644
> --- a/arch/arm64/mm/init.c
> +++ b/arch/arm64/mm/init.c
> @@ -171,6 +229,8 @@ void __init arm64_memblock_init(void)
>  		memblock_reserve(__virt_to_phys(initrd_start), initrd_end - initrd_start);
>  #endif
>
> +	reserve_crashkernel();
> +
>  	early_init_fdt_scan_reserved_mem();
>
>  	/* 4GB maximum for 32-bit only capable devices */
>


This is 'nit' territory, but if you were to make this:
>	if (IS_ENABLED(CONFIG_KEXEC_CORE))
>		reserve_crashkernel();

then the #ifdefs around reserve_crashkernel() can go, and the compiler will work
out that this static function can be optimised out. It also means the
compiler performs its checks, even if CONFIG_KEXEC_CORE isn't selected. The same
trick can be applied in patch 18 (around reserve_elfcorehdr()).


> @@ -66,6 +67,63 @@ static int __init early_initrd(char *p)
>  early_param("initrd", early_initrd);
>  #endif
>  
> +#ifdef CONFIG_KEXEC_CORE
> +/*
> + * reserve_crashkernel() - reserves memory for crash kernel
> + *
> + * This function reserves memory area given in "crashkernel=" kernel command
> + * line parameter. The memory reserved is used by dump capture kernel when
> + * primary kernel is crashing.
> + */
> +static void __init reserve_crashkernel(void)
> +{
> +	unsigned long long crash_size = 0, crash_base = 0;
> +	int ret;
> +
> +	ret = parse_crashkernel(boot_command_line, memblock_phys_mem_size(),
> +				&crash_size, &crash_base);
> +	if (ret)
> +		return;
> +
> +	if (crash_base == 0) {
> +		crash_base = memblock_find_in_range(0,
> +				MEMBLOCK_ALLOC_ACCESSIBLE, crash_size, 1 << 21);
> +		if (crash_base == 0) {
> +			pr_warn("Unable to allocate crashkernel (size:%llx)\n",
> +				crash_size);
> +			return;
> +		}
> +		memblock_reserve(crash_base, crash_size);
> +
> +	} else {
> +		/* User specifies base address explicitly. */
> +		if (!memblock_is_region_memory(crash_base, crash_size) ||
> +			memblock_is_region_reserved(crash_base, crash_size)) {
> +			pr_warn("crashkernel has wrong address or size\n");
> +			return;
> +		}
> +
> +		if (crash_base & ((1 << 21) - 1)) {
> +			pr_warn("crashkernel base address is not 2MB aligned\n");
> +			return;
> +		}
> +
> +		memblock_reserve(crash_base, crash_size);
> +	}
> +
> +	pr_info("Reserving %lldMB of memory at %lldMB for crashkernel\n",
> +		crash_size >> 20, crash_base >> 20);
> +
> +	crashk_res.start = crash_base;
> +	crashk_res.end = crash_base + crash_size - 1;
> +}
> +#else
> +static void __init reserve_crashkernel(void)
> +{
> +	;
> +}
> +#endif /* CONFIG_KEXEC_CORE */
> +
>  /*
>   * Return the maximum physical address for ZONE_DMA (DMA_BIT_MASK(32)). It
>   * currently assumes that for memory starting above 4G, 32-bit devices will


Thanks,

James

AKASHI Takahiro March 31, 2016, 7:19 a.m. UTC | #2

On Fri, Mar 18, 2016 at 06:08:30PM +0000, James Morse wrote:
> Hi!
> 
> On 14/03/16 17:48, Geoff Levand wrote:
> > From: AKASHI Takahiro <[e-mail]>
> > 
> > On the startup of primary kernel, the memory region used by crash dump
> > kernel must be specified by "crashkernel=" kernel parameter.
> > reserve_crashkernel() will allocate and reserve the region for later use.
> > 
> > User space tools, like kexec-tools, will be able to find that region marked
> > as "Crash kernel" in /proc/iomem.
> 
> [NB: Re-ordered diff hunks ]
> 
> > diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
> > index 7802f21..4edf181 100644
> > --- a/arch/arm64/mm/init.c
> > +++ b/arch/arm64/mm/init.c
> > @@ -171,6 +229,8 @@ void __init arm64_memblock_init(void)
> >  		memblock_reserve(__virt_to_phys(initrd_start), initrd_end - initrd_start);
> >  #endif
> >
> > +	reserve_crashkernel();
> > +
> >  	early_init_fdt_scan_reserved_mem();
> >
> >  	/* 4GB maximum for 32-bit only capable devices */
> >
> 
> 
> This is 'nit' territory, but if you were to make this:
> >	if (IS_ENABLED(CONFIG_KEXEC_CORE))
> >		reserve_crashkernel();
> 
> then the #ifdefs around reserve_crashkernel() can go, and the compiler will work
> out that this static function can be optimised out. It also means the
> compiler performs its checks, even if CONFIG_KEXEC_CORE isn't selected. The same
> trick can be applied in patch 18 (around reserve_elfcorehdr()).

It will also make the code more understandable.

Thanks,
-Takahiro AKASHI

> > @@ -66,6 +67,63 @@ static int __init early_initrd(char *p)
> >  early_param("initrd", early_initrd);
> >  #endif
> >  
> > +#ifdef CONFIG_KEXEC_CORE
> > +/*
> > + * reserve_crashkernel() - reserves memory for crash kernel
> > + *
> > + * This function reserves memory area given in "crashkernel=" kernel command
> > + * line parameter. The memory reserved is used by dump capture kernel when
> > + * primary kernel is crashing.
> > + */
> > +static void __init reserve_crashkernel(void)
> > +{
> > +	unsigned long long crash_size = 0, crash_base = 0;
> > +	int ret;
> > +
> > +	ret = parse_crashkernel(boot_command_line, memblock_phys_mem_size(),
> > +				&crash_size, &crash_base);
> > +	if (ret)
> > +		return;
> > +
> > +	if (crash_base == 0) {
> > +		crash_base = memblock_find_in_range(0,
> > +				MEMBLOCK_ALLOC_ACCESSIBLE, crash_size, 1 << 21);
> > +		if (crash_base == 0) {
> > +			pr_warn("Unable to allocate crashkernel (size:%llx)\n",
> > +				crash_size);
> > +			return;
> > +		}
> > +		memblock_reserve(crash_base, crash_size);
> > +
> > +	} else {
> > +		/* User specifies base address explicitly. */
> > +		if (!memblock_is_region_memory(crash_base, crash_size) ||
> > +			memblock_is_region_reserved(crash_base, crash_size)) {
> > +			pr_warn("crashkernel has wrong address or size\n");
> > +			return;
> > +		}
> > +
> > +		if (crash_base & ((1 << 21) - 1)) {
> > +			pr_warn("crashkernel base address is not 2MB aligned\n");
> > +			return;
> > +		}
> > +
> > +		memblock_reserve(crash_base, crash_size);
> > +	}
> > +
> > +	pr_info("Reserving %lldMB of memory at %lldMB for crashkernel\n",
> > +		crash_size >> 20, crash_base >> 20);
> > +
> > +	crashk_res.start = crash_base;
> > +	crashk_res.end = crash_base + crash_size - 1;
> > +}
> > +#else
> > +static void __init reserve_crashkernel(void)
> > +{
> > +	;
> > +}
> > +#endif /* CONFIG_KEXEC_CORE */
> > +
> >  /*
> >   * Return the maximum physical address for ZONE_DMA (DMA_BIT_MASK(32)). It
> >   * currently assumes that for memory starting above 4G, 32-bit devices will
> 
> 
> Thanks,
> 
> James
>

AKASHI Takahiro April 1, 2016, 6:16 a.m. UTC | #3

On 03/31/2016 02:19 PM, AKASHI Takahiro wrote:
> On Fri, Mar 18, 2016 at 06:08:30PM +0000, James Morse wrote:
>> Hi!
>>
>> On 14/03/16 17:48, Geoff Levand wrote:
>>> From: AKASHI Takahiro <[e-mail]>
>>>
>>> On the startup of primary kernel, the memory region used by crash dump
>>> kernel must be specified by "crashkernel=" kernel parameter.
>>> reserve_crashkernel() will allocate and reserve the region for later use.
>>>
>>> User space tools, like kexec-tools, will be able to find that region marked
>>> as "Crash kernel" in /proc/iomem.
>>
>> [NB: Re-ordered diff hunks ]
>>
>>> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
>>> index 7802f21..4edf181 100644
>>> --- a/arch/arm64/mm/init.c
>>> +++ b/arch/arm64/mm/init.c
>>> @@ -171,6 +229,8 @@ void __init arm64_memblock_init(void)
>>>   		memblock_reserve(__virt_to_phys(initrd_start), initrd_end - initrd_start);
>>>   #endif
>>>
>>> +	reserve_crashkernel();
>>> +
>>>   	early_init_fdt_scan_reserved_mem();
>>>
>>>   	/* 4GB maximum for 32-bit only capable devices */
>>>
>>
>>
>> This is 'nit' territory, but if you were to make this:
>>> 	if (IS_ENABLED(CONFIG_KEXEC_CORE))
>>> 		reserve_crashkernel();
>>
>> then the #ifdefs around reserve_crashkernel() can go, and the compiler will work
>> out that this static function can be optimised out. It also means the
>> compiler performs its checks, even if CONFIG_KEXEC_CORE isn't selected. The same
>> trick can be applied in patch 18 (around reserve_elfcorehdr()).
> 
> It will also make the code more understandable.

I tried this change, but
unfortunately, preserve_crashkernel() calls an external function, parse_crashkernel(),
which is defined only if CONFIG_KEXEC_CORE. So we will see a compiler error in !CONFIG_KEXEC_CORE
if we take #ifdef out around perserve_crashkernel().

-Takahiro AKASHI


> Thanks,
> -Takahiro AKASHI
> 
>>> @@ -66,6 +67,63 @@ static int __init early_initrd(char *p)
>>>   early_param("initrd", early_initrd);
>>>   #endif
>>>   
>>> +#ifdef CONFIG_KEXEC_CORE
>>> +/*
>>> + * reserve_crashkernel() - reserves memory for crash kernel
>>> + *
>>> + * This function reserves memory area given in "crashkernel=" kernel command
>>> + * line parameter. The memory reserved is used by dump capture kernel when
>>> + * primary kernel is crashing.
>>> + */
>>> +static void __init reserve_crashkernel(void)
>>> +{
>>> +	unsigned long long crash_size = 0, crash_base = 0;
>>> +	int ret;
>>> +
>>> +	ret = parse_crashkernel(boot_command_line, memblock_phys_mem_size(),
>>> +				&crash_size, &crash_base);
>>> +	if (ret)
>>> +		return;
>>> +
>>> +	if (crash_base == 0) {
>>> +		crash_base = memblock_find_in_range(0,
>>> +				MEMBLOCK_ALLOC_ACCESSIBLE, crash_size, 1 << 21);
>>> +		if (crash_base == 0) {
>>> +			pr_warn("Unable to allocate crashkernel (size:%llx)\n",
>>> +				crash_size);
>>> +			return;
>>> +		}
>>> +		memblock_reserve(crash_base, crash_size);
>>> +
>>> +	} else {
>>> +		/* User specifies base address explicitly. */
>>> +		if (!memblock_is_region_memory(crash_base, crash_size) ||
>>> +			memblock_is_region_reserved(crash_base, crash_size)) {
>>> +			pr_warn("crashkernel has wrong address or size\n");
>>> +			return;
>>> +		}
>>> +
>>> +		if (crash_base & ((1 << 21) - 1)) {
>>> +			pr_warn("crashkernel base address is not 2MB aligned\n");
>>> +			return;
>>> +		}
>>> +
>>> +		memblock_reserve(crash_base, crash_size);
>>> +	}
>>> +
>>> +	pr_info("Reserving %lldMB of memory at %lldMB for crashkernel\n",
>>> +		crash_size >> 20, crash_base >> 20);
>>> +
>>> +	crashk_res.start = crash_base;
>>> +	crashk_res.end = crash_base + crash_size - 1;
>>> +}
>>> +#else
>>> +static void __init reserve_crashkernel(void)
>>> +{
>>> +	;
>>> +}
>>> +#endif /* CONFIG_KEXEC_CORE */
>>> +
>>>   /*
>>>    * Return the maximum physical address for ZONE_DMA (DMA_BIT_MASK(32)). It
>>>    * currently assumes that for memory starting above 4G, 32-bit devices will
>>
>>
>> Thanks,
>>
>> James
>>

Patch

diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c
index 8119479..293cee2 100644
--- a/arch/arm64/kernel/setup.c
+++ b/arch/arm64/kernel/setup.c
@@ -31,7 +31,6 @@ 
 #include <linux/screen_info.h>
 #include <linux/init.h>
 #include <linux/kexec.h>
-#include <linux/crash_dump.h>
 #include <linux/root_dev.h>
 #include <linux/cpu.h>
 #include <linux/interrupt.h>
@@ -221,6 +220,12 @@  static void __init request_standard_resources(void)
 		    kernel_data.end <= res->end)
 			request_resource(res, &kernel_data);
 	}
+
+#ifdef CONFIG_KEXEC_CORE
+	/* User space tools will find "Crash kernel" region in /proc/iomem. */
+	if (crashk_res.end)
+		insert_resource(&iomem_resource, &crashk_res);
+#endif
 }
 
 #ifdef CONFIG_BLK_DEV_INITRD
diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index 7802f21..4edf181 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -34,6 +34,7 @@ 
 #include <linux/dma-contiguous.h>
 #include <linux/efi.h>
 #include <linux/swiotlb.h>
+#include <linux/kexec.h>
 
 #include <asm/fixmap.h>
 #include <asm/memory.h>
@@ -66,6 +67,63 @@  static int __init early_initrd(char *p)
 early_param("initrd", early_initrd);
 #endif
 
+#ifdef CONFIG_KEXEC_CORE
+/*
+ * reserve_crashkernel() - reserves memory for crash kernel
+ *
+ * This function reserves memory area given in "crashkernel=" kernel command
+ * line parameter. The memory reserved is used by dump capture kernel when
+ * primary kernel is crashing.
+ */
+static void __init reserve_crashkernel(void)
+{
+	unsigned long long crash_size = 0, crash_base = 0;
+	int ret;
+
+	ret = parse_crashkernel(boot_command_line, memblock_phys_mem_size(),
+				&crash_size, &crash_base);
+	if (ret)
+		return;
+
+	if (crash_base == 0) {
+		crash_base = memblock_find_in_range(0,
+				MEMBLOCK_ALLOC_ACCESSIBLE, crash_size, 1 << 21);
+		if (crash_base == 0) {
+			pr_warn("Unable to allocate crashkernel (size:%llx)\n",
+				crash_size);
+			return;
+		}
+		memblock_reserve(crash_base, crash_size);
+
+	} else {
+		/* User specifies base address explicitly. */
+		if (!memblock_is_region_memory(crash_base, crash_size) ||
+			memblock_is_region_reserved(crash_base, crash_size)) {
+			pr_warn("crashkernel has wrong address or size\n");
+			return;
+		}
+
+		if (crash_base & ((1 << 21) - 1)) {
+			pr_warn("crashkernel base address is not 2MB aligned\n");
+			return;
+		}
+
+		memblock_reserve(crash_base, crash_size);
+	}
+
+	pr_info("Reserving %lldMB of memory at %lldMB for crashkernel\n",
+		crash_size >> 20, crash_base >> 20);
+
+	crashk_res.start = crash_base;
+	crashk_res.end = crash_base + crash_size - 1;
+}
+#else
+static void __init reserve_crashkernel(void)
+{
+	;
+}
+#endif /* CONFIG_KEXEC_CORE */
+
 /*
  * Return the maximum physical address for ZONE_DMA (DMA_BIT_MASK(32)). It
  * currently assumes that for memory starting above 4G, 32-bit devices will
@@ -171,6 +229,8 @@  void __init arm64_memblock_init(void)
 		memblock_reserve(__virt_to_phys(initrd_start), initrd_end - initrd_start);
 #endif
 
+	reserve_crashkernel();
+
 	early_init_fdt_scan_reserved_mem();
 
 	/* 4GB maximum for 32-bit only capable devices */