Dear Wolfgang,

2011/10/7 Wolfgang Denk <w...@denx.de>:
> Dear Macpaul Lin,
>
> In message <1317896723-9284-1-git-send-email-macp...@andestech.com> you wrote:
>> Add generic header files support for nds32 architecture.
>> Cache, ptregs, data type and other definitions are included.
>>
>> Signed-off-by: Macpaul Lin <macp...@andestech.com>
...
>> Changes for v15:
>>   - u-boot.h: fix for new image.h according to Mike's contribution.

I believe you must be exhausted for reviewing these patches again and again.

However, we've discussed before, to support a new architecture, there
are some definitions
and codes which the checkpatch rules cannot be adapted.
For example, some typedefs definitions, special asm syntax, and those
special cases described
in Linux documents, and u-boot's typedef global structures.
ex: http://www.mail-archive.com/u-boot@lists.denx.de/msg52083.html

We've also discussed that Linux documents said checkpatch is just a
reference rule,
you still need to have human examination.
I guess you just forget that because the patch "works" is too heavy
and patch check
automation didn't help on special case much.

When we discuss about those was about NDS32 patch v7, and now is patch v15.
We've to a lot of fix, includes relocation.

So, could you please tell me now, should I resend the patch v16 or you
still have my v15
patches and kindly tell me which part need to be fixed? Thanks!

>
> Checkpatch says:
>
> total: 8 errors, 58 warnings, 1447 lines checked
>
> Please clean up and resubmit.  Thanks.
>
> Best regards,
>
> Wolfgang Denk
>

Note:
The 8 errors is due to the new assembly march.h added.
Only 2 new cases introduced.

ERROR: space prohibited before open square bracket '['
#1032: FILE: arch/nds32/include/asm/macro.h:66:
+       lwi     $r5, [$r4]
This is to load a 32bit from the address x, while the value of
address x is stored in $r4, and load it into $r5

ERROR: spaces required around that ':' (ctx:VxE)
#1055: FILE: arch/nds32/include/asm/macro.h:89:
+1:
  ^
The original form is like this:

.macro  wait_timer, time
        li      $r4, time
1: <--------Checkpatch indicates error here. However this should be
checkpatch's fault.
        nop
        addi    $r4, $r4, -1
        bnez    $r4, 1b
.endm


Thanks!

-- 
Best regards,
Macpaul Lin
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot

Reply via email to