[PATCH v14 08/11] arm64: kdump: reimplement crashkernel=X

chenzhou chenzhou10 at huawei.com
Fri Feb 26 05:43:07 EST 2021



On 2021/2/26 18:31, chenzhou wrote:
>
> On 2021/2/25 0:04, Catalin Marinas wrote:
>> On Sat, Jan 30, 2021 at 03:10:22PM +0800, Chen Zhou wrote:
>>> There are following issues in arm64 kdump:
>>> 1. We use crashkernel=X to reserve crashkernel below 4G, which
>>> will fail when there is no enough low memory.
>>> 2. If reserving crashkernel above 4G, in this case, crash dump
>>> kernel will boot failure because there is no low memory available
>>> for allocation.
>>>
>>> To solve these issues, change the behavior of crashkernel=X and
>>> introduce crashkernel=X,[high,low]. crashkernel=X tries low allocation
>>> in DMA zone, and fall back to high allocation if it fails.
>>> We can also use "crashkernel=X,high" to select a region above DMA zone,
>>> which also tries to allocate at least 256M in DMA zone automatically.
>>> "crashkernel=Y,low" can be used to allocate specified size low memory.
>>>
>>> Another minor change, there may be two regions reserved for crash
>>> dump kernel, in order to distinct from the high region and make no
>>> effect to the use of existing kexec-tools, rename the low region as
>>> "Crash kernel (low)".
>> I think we discussed this but I don't remember the conclusion. Is this
>> only renamed conditionally so that we don't break current kexec-tools?
> Yes.
>> IOW, assuming that the full crashkernel region is reserved below 4GB,
>> does the "(low)" suffix still appear or it's only if a high region is
>> additionally reserved?
> Suffix "low" only appear if a high region is additionally reserved.
>>> diff --git a/arch/arm64/include/asm/kexec.h b/arch/arm64/include/asm/kexec.h
>>> index 3f6ecae0bc68..f0caed0cb5e1 100644
>>> --- a/arch/arm64/include/asm/kexec.h
>>> +++ b/arch/arm64/include/asm/kexec.h
>>> @@ -96,6 +96,10 @@ static inline void crash_prepare_suspend(void) {}
>>>  static inline void crash_post_resume(void) {}
>>>  #endif
>>>  
>>> +#ifdef CONFIG_KEXEC_CORE
>>> +extern void __init reserve_crashkernel(void);
>>> +#endif
>> Why not have this in some generic header?
>>
>>> diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c
>>> index c18aacde8bb0..69c592c546de 100644
>>> --- a/arch/arm64/kernel/setup.c
>>> +++ b/arch/arm64/kernel/setup.c
>>> @@ -238,7 +238,18 @@ static void __init request_standard_resources(void)
>>>  		    kernel_data.end <= res->end)
>>>  			request_resource(res, &kernel_data);
>>>  #ifdef CONFIG_KEXEC_CORE
>>> -		/* Userspace will find "Crash kernel" region in /proc/iomem. */
>>> +		/*
>>> +		 * Userspace will find "Crash kernel" or "Crash kernel (low)"
>>> +		 * region in /proc/iomem.
>>> +		 * In order to distinct from the high region and make no effect
>>> +		 * to the use of existing kexec-tools, rename the low region as
>>> +		 * "Crash kernel (low)".
>>> +		 */
>>> +		if (crashk_low_res.end && crashk_low_res.start >= res->start &&
>>> +				crashk_low_res.end <= res->end) {
>>> +			crashk_low_res.name = "Crash kernel (low)";
>>> +			request_resource(res, &crashk_low_res);
>>> +		}
>>>  		if (crashk_res.end && crashk_res.start >= res->start &&
>>>  		    crashk_res.end <= res->end)
>>>  			request_resource(res, &crashk_res);
>> My reading of the new generic reserve_crashkernel() is that
>> crashk_low_res will only be populated if crask_res is above 4GB. If
>> that's correct, I'm fine with the renaming here since current systems
>> would not get a renamed low reservation (as long as they don't change
>> the kernel cmdline).
>>
>>> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
>>> index 912f64f505f7..d20f5c444ebf 100644
>>> --- a/arch/arm64/mm/init.c
>>> +++ b/arch/arm64/mm/init.c
>>> @@ -35,6 +35,7 @@
>>>  #include <asm/fixmap.h>
>>>  #include <asm/kasan.h>
>>>  #include <asm/kernel-pgtable.h>
>>> +#include <asm/kexec.h>
>>>  #include <asm/memory.h>
>>>  #include <asm/numa.h>
>>>  #include <asm/sections.h>
>>> @@ -61,66 +62,11 @@ EXPORT_SYMBOL(memstart_addr);
>>>   */
>>>  phys_addr_t arm64_dma_phys_limit __ro_after_init;
>>>  
>>> -#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.
>>> - */
>>> +#ifndef CONFIG_KEXEC_CORE
>>>  static void __init reserve_crashkernel(void)
>>>  {
>> [...]
>>>  }
>>> +#endif
>> Can we not have the dummy reserve_crashkernel() in the generic code as
>> well and avoid the #ifndef here?
> You mean put the dummy reserve_crashkernel() and the relate function declaration in some generic header?
>  
> Baoquan also mentioned about this.
> Now all the arch that support kdump have the dummy reserve_crashkernel() and
> function declaration, such as arm/arm64/ppc/s390..
>
> But currently different arch may have different CONFIG and different function declaration about this,
> for example,
>
> for s390,
> static void __init reserve_crashkernel(void)
> {                  
> #ifdef CONFIG_CRASH_DUMP
> ...
> #endif        
> }
>
> for ppc,
> #ifdef CONFIG_KEXEC_CORE
> extern void reserve_crashkernel(void);
> #else
> static inline void reserve_crashkernel(void) { ; }
> #endif
>
> If we move these to generic header we need think about:
> 1. the related config in different arch
> 2. function declaration(static/non static)
>
> As Baoquan said in patch 9, how about leave with it for now and i try to solve this later?
>
>>>  #ifdef CONFIG_CRASH_DUMP
>>>  static int __init early_init_dt_scan_elfcorehdr(unsigned long node,
>>> @@ -446,6 +392,14 @@ void __init bootmem_init(void)
>>>  	 * reserved, so do it here.
>>>  	 */
>>>  	reserve_crashkernel();
>>> +#ifdef CONFIG_KEXEC_CORE
>>> +	/*
>>> +	 * The low region is intended to be used for crash dump kernel devices,
>>> +	 * just mark the low region as "nomap" simply.
>>> +	 */
>>> +	if (crashk_low_res.end)
>>> +		memblock_mark_nomap(crashk_low_res.start, resource_size(&crashk_low_res));
>>> +#endif
>> Do we do something similar for crashk_res?
> Not. In the primary kernel(production kernel), we need to use crashk_res memory for crash kernel
> elf core header, initrd...

Sorry, missed one comma after crash kernel.
Not. In the primary kernel(production kernel), we need to use crashk_res memory for crash kernel,
elf core header, initrd and so on.


>
> Different with this, the crashk_low_res is only for crash dump kernel devices.
>> Also, I can see we call crash_exclude_mem_range() only for crashk_res.
>> Do we need to do this for crashk_low_res as well?
> You are right, i missed about this. Will do in next version.
>
> Thanks,
> Chen Zhou




More information about the kexec mailing list