Re: [PATCH 1/2] microblaze: add stack unwinder

[Michal Simek <monstr@monstr.eu>]

Steven J. Magnani wrote:
> On Wed, 2010-04-14 at 18:45 +0200, Michal Simek wrote:
>>> diff -uprN a/arch/microblaze/kernel/traps.c
>> b/arch/microblaze/kernel/traps.c
>>> --- a/arch/microblaze/kernel/traps.c  2010-04-09 21:52:36.000000000
>>> +++ b/arch/microblaze/kernel/traps.c  2010-04-12 22:16:01.000000000 
> [snip]
>>>  
>>> -     if (!stack)
>>> -             stack = (unsigned long *)&stack;
>>> +     if (fp == 0) {
>>> +             if (task)
>>> +                     fp = ((struct thread_info *)
>>> +                             (task->stack))->cpu_context.r1;
>>> +             else {
>>> +                     /* Pick up caller of dump_stack() */
>>> +                     fp = (__u32)&sp - 8;
>>> +             }
>>> +     }
>> just coding style.
>>
>>         if (fp == 0)
>>                 if (task)
>>                         fp = ((struct thread_info *)
>>                                 (task->stack))->cpu_context.r1;
>>                 else
>>                         /* Pick up caller of dump_stack() */
>>                         fp = (__u32)&sp - 8;
> 
> Do you feel strongly about this? I try to always use braces on if/else
> clauses that have more than one line. I've found that the extra
> characters are well worth the savings in debugging time when someone
> tries to extend the clause and forgets to add the braces.

It will work. I like brackets too. I agree that can save a lot of time 
with debugging.

I just didn't like that inconsistency in if (task) part.

If is if (task)  { ... } else {...} then I am ok with it if 
checkpatch.pl doesn't report it as warning.


Michal



-- 
Michal Simek, Ing. (M.Eng)
w: www.monstr.eu p: +42-0-721842854
Maintainer of Linux kernel 2.6 Microblaze Linux - http://www.monstr.eu/fdt/
Microblaze U-BOOT custodian