[Intel-gfx] [PATCH 33/37] drm/i915: double-write DDI translation table
Rodrigo Vivi
rodrigo.vivi at gmail.com
Mon Mar 26 19:51:12 CEST 2012
Reviewed-by: Rodrigo Vivi <rodrigo.vivi at gmail.com>
On Wed, Mar 21, 2012 at 10:10 PM, Eugeni Dodonov
<eugeni.dodonov at intel.com> wrote:
> Some double-buffered registers need to be written twice.
>
> Note that it is being sent as a separate patch because sometimes these
> registers do work when written only once. But double-writing on my machine
> ensured that they work more often.
>
> Signed-off-by: Eugeni Dodonov <eugeni.dodonov at intel.com>
> ---
> drivers/gpu/drm/i915/intel_display.c | 12 ++++++++----
> 1 file changed, 8 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> index 04625d5..8a839be 100644
> --- a/drivers/gpu/drm/i915/intel_display.c
> +++ b/drivers/gpu/drm/i915/intel_display.c
> @@ -2769,12 +2769,16 @@ static void hsw_fdi_link_train(struct drm_crtc *crtc)
> struct drm_i915_private *dev_priv = dev->dev_private;
> struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
> int pipe = intel_crtc->pipe;
> - u32 reg, temp, i;
> + u32 reg, temp, i, j;
>
> /* Prior to enabling DDI, configure buffer translation with FDI values */
> - for (i=0, reg=DDI_BUF_TRANS_E; i < ARRAY_SIZE(hsw_ddi_translations_fdi); i++) {
> - I915_WRITE(reg, hsw_ddi_translations_fdi[i]);
> - reg += 4;
> + /* Those registers seem to be double-buffered - at least, the hw team writes them twice. */
> + for (j=0; j < 2; j++) {
> + for (i=0, reg=DDI_BUF_TRANS_E; i < ARRAY_SIZE(hsw_ddi_translations_fdi); i++) {
> + I915_WRITE(reg, hsw_ddi_translations_fdi[i]);
> + reg += 4;
> + }
> + udelay(20);
> }
>
> /* Configure CPU PLL, wait for warmup */
> --
> 1.7.9.2
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx
--
Rodrigo Vivi
Blog: http://blog.vivi.eng.br
GPG: 0x905BE242 @ wwwkeys.pgp.net
More information about the Intel-gfx
mailing list