Re: [RFC/TileGX 2/6] simplify the handling of skip prologue for plt stub

[Jiong Wang <jiwang@tilera.com>]

On 01/19/2013 01:29 AM, Pedro Alves wrote:
> On 01/18/2013 03:12 PM, Jiong Wang wrote:
>>>> for tilegx,  when skip prologue, if the start_pc is a plt stub address, then
>>>> stop to go further, just return the start_pc.
>>>>
>>>> gdb/ChangeLog:
>>>>
>>>>          * tilegx-tdep.c (tilegx_skip_prologue): simplify the handling for
>>>>          plt stub.
>>> Can you provide an example where this becomes necessary? I don't
>>> see a problem with the patch per se, but I don't remember seeing
>>> other ports doing this...
>>>
>> yes, there are some tricky thing.
>>
>>    for the simple hello.c
>>
>> #include <stdio.h>
>> int main(int argc, char **argv)
>> {
>> printf("hello, %d\n", argc);
>> return 0
>> }
>>
>> without this patch
>> ===
>> -bash-4.1$ gcc -o hello.tile hello.c
>> -bash-4.1$ ./gdb hello.tile
>> (gdb) b printf
>> Error accessing memory address 0x10a40: Success.
>>
>> after this patch
>> ===
>> (gdb) b printf
>> Breakpoint 1 at 0x10a40
>>
>>
>> this is because tilegx skip_prologue will invoke tilegx_analyze_prologue, which
>> will prefetch 32*8 bytes.
>>
>> while for when the address is in plt stub,  you can see it near the eh_frame_hdr section
>>
>>    [14] .plt                         0000000000010a00 000a00 0000a0 28  AX  0   0 64
>>    ...
>>    [16] .eh_frame_hdr     0000000000010ac0 000ac0 000024 00   A  0 0  4
>>    [17] .eh_frame             0000000000010ae8 000ae8 0000b4 00   A 0   0  8
>>
>> the .eh_frame_hdr aligns to 4, there is a hole between .eh_frame_hdr and .eh_frame, and this
>> will cause trouble for section_table_xfer_memory_partial.
>>
>> after fetch memory starting from 0x10ac0 to 0x10ae4, then the memaddr will be 0x10ae4 in section_table_xfer_memory_partial,
>> while this function did not consider this hole situation, so goes to line 666, error occured.
> Right.  It's this issue that needs solving.  Otherwise you're just
> papering over the problem.
Ping,

is this patch OK to commit?

---
Regards,
Jiong
>
>> for other targets, like x86, I have done a brief exploration,  seems they do not have large prefetch window
>
>


-- 
Regards,
Jiong. Wang
Tilera Corporation.