[PATCH 1/3] UBUNTU: [Upstream] drm/nv04-nv40: Fix up the programmed horizontal sync pulse delay.

[Stefan Bader]

Chris got back to me directly and said that this only were two patches to be
sent. He also wanted to pass them on to stable.
@Chris, this is done by just passing on that patch (with the sha1 reference of
the upstream commit to stable at kernel.org. Usually cc'ing maintainers and other
people on the s-o-b list and asking for inclusion.
This one looks ok to me.

Chris Halse Rogers wrote:
> From: Christopher James Halse Rogers <raof at ubuntu.com>
> 
> The calculated values were a little bit off (~16 clocks), the only
> effect it could have had is a slightly offset image with respect to
> the blob on analog outputs (bug 26790).
> 
> BugLink: http://bugs.launchpad.net/bugs/529130
> 
> Signed-off-by: Francisco Jerez <currojerez at riseup.net>
> Signed-off-by: Christopher James Halse Rogers <raof at ubuntu.com>
Acked-by: Stefan Bader <stefan.bader at canonical.com>
> ---
>  drivers/gpu/drm/nouveau/nv04_crtc.c |    6 +++---
>  1 files changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/gpu/drm/nouveau/nv04_crtc.c b/drivers/gpu/drm/nouveau/nv04_crtc.c
> index d2f143e..9986aba 100644
> --- a/drivers/gpu/drm/nouveau/nv04_crtc.c
> +++ b/drivers/gpu/drm/nouveau/nv04_crtc.c
> @@ -230,9 +230,9 @@ nv_crtc_mode_set_vga(struct drm_crtc *crtc, struct drm_display_mode *mode)
>  	struct drm_framebuffer *fb = crtc->fb;
>  
>  	/* Calculate our timings */
> -	int horizDisplay	= (mode->crtc_hdisplay >> 3) 	- 1;
> -	int horizStart		= (mode->crtc_hsync_start >> 3) 	- 1;
> -	int horizEnd		= (mode->crtc_hsync_end >> 3) 	- 1;
> +	int horizDisplay	= (mode->crtc_hdisplay >> 3)		- 1;
> +	int horizStart		= (mode->crtc_hsync_start >> 3) 	+ 1;
> +	int horizEnd		= (mode->crtc_hsync_end >> 3)		+ 1;
>  	int horizTotal		= (mode->crtc_htotal >> 3)		- 5;
>  	int horizBlankStart	= (mode->crtc_hdisplay >> 3)		- 1;
>  	int horizBlankEnd	= (mode->crtc_htotal >> 3)		- 1;