[PATCH lib/libX11] Don't use caddr_t casts

[Jon TURNEY]

On 09/03/2012 04:45, Alan Coopersmith wrote:
> On 02/19/12 06:31 AM, Jon TURNEY wrote:
>> On 18/02/2012 19:11, Jamey Sharp wrote:
>>> On Sat, Feb 18, 2012 at 05:21:24PM +0000, Jon TURNEY wrote:
>>>> Casting a (const char *) to (caddr_t) to assign to iovec.io_base
>>>> seems pointless. caddr_t isn't used anywhere else in xcb or libX11
>>>
>>> According to the libxcb git history, I replaced (caddr_t) with (char *)
>>> in 2006 to "help DragonFly and Solaris". I'd be fine with this patch if
>>> it explicitly cast to (char *), which I believe suppresses the constness
>>> warning.
> 
> BTW, Solaris 11 now uses (void *), but Solaris 10 & older used caddr_t there,
> which is why you needed it back then.

This sounds a bit odd to me, that cast would only be needed if caddr_t was
defined as something which was incompatible with having a char * assigned to it?

>> diff --git a/src/xcb_io.c b/src/xcb_io.c
>> index 0af47d8..300ef57 100644
>> --- a/src/xcb_io.c
>> +++ b/src/xcb_io.c
>> @@ -478,9 +478,9 @@ void _XSend(Display *dpy, const char *data, long size)
>>
>>      vec[0].iov_base = dpy->buffer;
>>      vec[0].iov_len = dpy->bufptr - dpy->buffer;
>> -    vec[1].iov_base = (caddr_t) data;
>> +    vec[1].iov_base = (char *)data;
>>      vec[1].iov_len = size;
>> -    vec[2].iov_base = (caddr_t) pad;
>> +    vec[2].iov_base = (char *)pad;
>>      vec[2].iov_len = -size & 3;
>>
>>      for(ext = dpy->flushes; ext; ext = ext->next_flush)
> 
> Reviewed-by: Alan Coopersmith <alan.coopersmith at oracle.com>

Thanks. I've pushed this.
   20adca0..ed00b46  master -> master