[PATCH] makedumpfile: fix max_mapnr issue on system has over 44-bit addressing

Ma, Jingbai (Kingboard) kingboard.ma at hp.com
Thu Oct 3 10:37:38 EDT 2013


On 10/2/13 2:20 PM, "Atsushi Kumagai" <kumagai-atsushi at mxc.nes.nec.co.jp>
wrote:


>Hello Jingbai,
>
>I have a additional comment.
>
>(2013/09/25 17:54), Jingbai Ma wrote:
>> On 09/25/2013 04:39 PM, HATAYAMA Daisuke wrote:
>>> (2013/09/25 17:16), Jingbai Ma wrote:
>>> <cut>
>> @@ -790,7 +794,7 @@ get_kdump_compressed_header_info(char *filename)
>>       DEBUG_MSG("  block_size       : %d\n", dh.block_size);
>>       DEBUG_MSG("  sub_hdr_size     : %d\n", dh.sub_hdr_size);
>>       DEBUG_MSG("  bitmap_blocks    : %d\n", dh.bitmap_blocks);
>> -     DEBUG_MSG("  max_mapnr        : 0x%x\n", dh.max_mapnr);
>> +     DEBUG_MSG("  max_mapnr(32bit) : 0x%x\n", dh.max_mapnr);
>>       DEBUG_MSG("  total_ram_blocks : %d\n", dh.total_ram_blocks);
>>       DEBUG_MSG("  device_blocks    : %d\n", dh.device_blocks);
>>       DEBUG_MSG("  written_blocks   : %d\n", dh.written_blocks);
>> @@ -802,6 +806,7 @@ get_kdump_compressed_header_info(char *filename)
>>       DEBUG_MSG("  split            : %d\n", kh.split);
>>       DEBUG_MSG("  start_pfn        : 0x%lx\n", kh.start_pfn);
>>       DEBUG_MSG("  end_pfn          : 0x%lx\n", kh.end_pfn);
>> +     DEBUG_MSG("  max_mapnr(64bit) : 0x%lx\n", kh.max_mapnr);
>
>max_mapnr(64bit) will be included only in header_version 6 and later,
>but your code shows that in any version.
>I think it's better to change the code like below:
>
>diff --git a/makedumpfile.c b/makedumpfile.c
>index 069a903..6c20e88 100644
>--- a/makedumpfile.c
>+++ b/makedumpfile.c
>@@ -847,6 +847,10 @@ get_kdump_compressed_header_info(char *filename)
>  				(unsigned long long)kh.offset_eraseinfo);
>  		DEBUG_MSG("  size_eraseinfo   : 0x%ld\n", kh.size_eraseinfo);
>  	}
>+	if (dh.header_version >= 6) {
>+		/* A dumpfile contains full 64bit max_mapnr. */
>+		DEBUG_MSG("  max_mapnr(64bit) : 0x%llx\n", kh.max_mapnr);
>+	}
>  	return TRUE;
>  error:
>  	free(info->dh_memory);

OK, I will fix it in next version.
Thanks!

>
>
>Thanks
>Atsushi Kumagai

Thanks,
Jingbai Ma




More information about the kexec mailing list