Re: [PATCH] makedumpfile: clear mem_map if not in pt_loads

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 




On 04/14/2020 03:09 AM, HAGIO KAZUHITO(萩尾 一仁) wrote:
> Hi Pingfan,
> 
>> -----Original Message-----
>> The crashed kernel passes usable RAM info through pt_loads[], but the
>> current code ignore this, and parse all sections below info->max_mapnr.
>> Refer to code in get_mm_sparsemem()
>>   num_section = divideup(info->max_mapnr, PAGES_PER_SECTION());
>>
>> These ranges can include unwanted memory e.g. reserved-memory for crash
>> kernel, PMEM for fs-dax. Excluding them by setting mem_maps[section_nr]=NULL.
> 
> hmm, I think that bitmap1 corresponds to pt_loads ranges, so even without
> the patch, only the ranges in pt_loads can be included in a dumpfile.
> Am I missing anything?
> 
Check the code, and I think you are right.

Thanks,
Pingfan
> Thanks,
> Kazu
> 
>>
>> Signed-off-by: Pingfan Liu <piliu@xxxxxxxxxx>
>> ---
>>  elf_info.c     | 12 ++++++++++++
>>  elf_info.h     |  1 +
>>  makedumpfile.c |  5 +++++
>>  3 files changed, 18 insertions(+)
>>
>> diff --git a/elf_info.c b/elf_info.c
>> index 7d72742..297c3e4 100644
>> --- a/elf_info.c
>> +++ b/elf_info.c
>> @@ -1302,3 +1302,15 @@ get_max_file_offset(void)
>>  {
>>  	return max_file_offset;
>>  }
>> +
>> +int phyaddr_in_pt_loads(unsigned long long phys_start)
>> +{
>> +	 struct pt_load_segment *pls;
>> +
>> +	for (int i = 0; i < num_pt_loads; i++) {
>> +		pls = &pt_loads[i];
>> +		if (phys_start >= pls->phys_start && phys_start < pls->phys_end)
>> +			return TRUE;
>> +	}
>> +	return FALSE;
>> +}
>> diff --git a/elf_info.h b/elf_info.h
>> index 934b608..f5bb07c 100644
>> --- a/elf_info.h
>> +++ b/elf_info.h
>> @@ -90,6 +90,7 @@ void get_eraseinfo(off_t *offset, unsigned long *size);
>>  void set_eraseinfo(off_t offset, unsigned long size);
>>
>>  off_t get_max_file_offset(void);
>> +int phyaddr_in_pt_loads(unsigned long long phys_start);
>>
>>  #endif  /* ELF_INFO_H */
>>
>> diff --git a/makedumpfile.c b/makedumpfile.c
>> index f5860a1..c0fa075 100644
>> --- a/makedumpfile.c
>> +++ b/makedumpfile.c
>> @@ -3556,8 +3556,13 @@ get_mm_sparsemem(void)
>>  		goto out;
>>  	}
>>  	for (section_nr = 0; section_nr < num_section; section_nr++) {
>> +		unsigned long phys_start;
>> +
>>  		pfn_start = section_nr * PAGES_PER_SECTION();
>>  		pfn_end   = pfn_start + PAGES_PER_SECTION();
>> +		phys_start = pfn_start << PAGESHIFT();
>> +		if (!phyaddr_in_pt_loads(phys_start))
>> +			mem_maps[section_nr] = (unsigned long)NULL;
>>  		if (info->max_mapnr < pfn_end)
>>  			pfn_end = info->max_mapnr;
>>  		dump_mem_map(pfn_start, pfn_end, mem_maps[section_nr], section_nr);
>> --
>> 2.7.5
> 


_______________________________________________
kexec mailing list
kexec@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/kexec



[Index of Archives]     [LM Sensors]     [Linux Sound]     [ALSA Users]     [ALSA Devel]     [Linux Audio Users]     [Linux Media]     [Kernel]     [Gimp]     [Yosemite News]     [Linux Media]

  Powered by Linux