svn commit: r317171 - head/sys/dev/vt

[Jung-uk Kim]

On 05/01/2017 16:55, Ed Maste wrote:
> On 20 April 2017 at 14:06, Jung-uk Kim <jkim at freebsd.org> wrote:
>>
>> Maybe but I haven't seen much improvement in the area for many years.
>> Do you have any reason to believe someone is working on it?
>>
>> Even if we have such backend, I don't expect much difference in drawing
>> small area, i.e., redrawing narrow borders.  On top of that, I don't
>> like "if (vd->vd_driver->vd_drawrect == NULL) return (ENOTSUP)" stuff.
> 
> I don't believe anyone is working on it at the moment, but I expect
> that effort will go into improved VGA / VESA support at some point and
> that might implement an improved drawrect. I agree that there's
> unlikely to be a noticeable difference in the case of occasional
> operations filling a small region (e.g., border clearing after vt
> switch).
> 
> However, the vd_drawrect == NULL case is really no different than
> vd_setpixel == NULL. All vt drivers except ofwfb provide both, and
> ofwfb provides neither.

I can revert it but I don't like the previous implementation.  There
were overlapping regions and off-by-one issues.  Please see the attached
patch, which is against the previous version.

Jung-uk Kim
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vt_core.diff
Type: text/x-patch
Size: 987 bytes
Desc: not available
URL: <http://lists.freebsd.org/pipermail/svn-src-all/attachments/20170501/17a42690/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <http://lists.freebsd.org/pipermail/svn-src-all/attachments/20170501/17a42690/attachment.sig>