[Crash-utility] ARM support for CONFIG_SPARSEMEM: (was Re:DDimage)

[takuo.koguchi.sw at hitachi.com]

Hi Mika,

>Please check your platforms
>arch/arm/mach-<yourmach>/include/mach/memory.h and set both
>_SECTION_SIZE_BITS and _MAX_PHYSMEM_BITS to be the same as defined in
>the header. If crash works with the correct values, we just need to
Both SECTION_SIZE_BITS 28 and MAX_PHYSMEM_BITS 32 are defined correctly in the kernel header and defs.h.
Although, still I got 
>> crash: invalid kernel virtual address: f67f000  type: "page table"

It might come form my vmcore file's incorrectness.
Currently my vmcore contains the following, 
Program Headers:
  Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
  NOTE           0x000074 0x00000000 0x00000000 0x00000 0x00000     0
  LOAD           0x000074 0xc0000000 0x00000000 0x20000000 0x20000000 RWE 0

But actual mapping is,
Virtual 0xc0000000-0xcfffffff corresponds Physical 0x00000000-0x0fffffff
and 
Virtual 0xd0000000-0xdfffffff corresponds Physical 0x40000000-0x4fffffff

Then I tried to prepend the following header instead,
Program Headers:
  Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
  NOTE           0x000094 0x00000000 0x00000000 0x00000 0x00000     0
  LOAD           0x000094 0xc0000000 0x00000000 0x10000000 0x10000000 RWE 0
  LOAD           0x10000094 0xd0000000 0x40000000 0x10000000 0x10000000 RWE 0

This time crash does not recognize the vmcore file.

Perhaps I have to change VTOP/PTOV macros accordingly?

Best Regards,

Takuo 


--------------
>On Wed, May 25, 2011 at 10:11 AM,  <takuo.koguchi.sw at hitachi.com> wrote:
>>>... But I got another error later, which I will report later.
>> Now I set _SECTION_SIZE_BITS 22 and see what happens next.
>> I found that readmem for "page table" failed while excuting init_unwind_tables() in unwind_arm.c and crash exited.
>>
>> crash: invalid kernel virtual address: f67f000  type: "page table"
>>
>> Instead of investigating this issue, I continue by skipping init_unwind_tables for a while.
>> Because crash gives me a prompt now and I am quite interested in whether it gives me some useful information or not.
>>
>
>Please check your platforms
>arch/arm/mach-<yourmach>/include/mach/memory.h and set both
>_SECTION_SIZE_BITS and _MAX_PHYSMEM_BITS to be the same as defined in
>the header. If crash works with the correct values, we just need to
>find out some way of guessing these.
>
>--
>Crash-utility mailing list
>Crash-utility at redhat.com
>https://www.redhat.com/mailman/listinfo/crash-utility
>