Re: [Qemu-devel] [PATCH v2 6/9] target-i386: Handle I/O breakpoints

[Paolo Bonzini]

On 19/10/2015 19:57, Eduardo Habkost wrote:
> On Mon, Oct 19, 2015 at 07:46:51AM -1000, Richard Henderson wrote:
>> On 10/19/2015 07:30 AM, Eduardo Habkost wrote:
>>>>> +        /* Notice when we should enable calls to bpt_io.  */
>>>>> +        return (hw_breakpoint_enabled(env->dr[7], index)
>>>>> +                ? HF_IOBPT_MASK : 0);
>>> checkpatch.pl error:
>>>
>>>   ERROR: return is not a function, parentheses are not required
>>>   #57: FILE: target-i386/bpt_helper.c:69:
>>>   +        return (hw_breakpoint_enabled(env->dr[7], index)
>>>
>>>   total: 1 errors, 0 warnings, 242 lines checked
>>>
>>> I will fix it in v3.
>>
>> In this case checkpatch is wrong, imo.  The parenthesis are not there to
>> "make return a function", but to make the multi-line expression indent
>> properly.
> 
> I understand if one thinks the expression looks better with the parenthesis,
> but I fail to see why they are needed to indent the expression properly.

Because Emacs indents this:

>     +        return hw_breakpoint_enabled(env->dr[7], index)
>     +               ? HF_IOBPT_MASK : 0;

with the ? under the second "r" of "return", while it indents this as
written:

>     -        return (hw_breakpoint_enabled(env->dr[7], index)
>     -                ? HF_IOBPT_MASK : 0);

Another random example:

static bool sdl_check_format(DisplayChangeListener *dcl,
                             pixman_format_code_t format)
{
    /*
     * We let SDL convert for us a few more formats than,
     * the native ones. Thes are the ones I have tested.
     */
    return (format == PIXMAN_x8r8g8b8 ||
            format == PIXMAN_b8g8r8x8 ||
            format == PIXMAN_x1r5g5b5 ||
            format == PIXMAN_r5g6b5);
}

There's no unanimity though, so your v3 is okay too.

Paolo