[PATCH v11 3/9] drm/mediatek: add RDMA fifo size error handle

Hsin-Yi Wang hsinyi at chromium.org
Thu Jan 28 11:12:20 UTC 2021


On Thu, Jan 28, 2021 at 3:52 PM CK Hu <ck.hu at mediatek.com> wrote:
>
> Hi, Hsin-Yi:
>
> On Thu, 2021-01-28 at 15:27 +0800, Hsin-Yi Wang wrote:
> > From: Yongqiang Niu <yongqiang.niu at mediatek.com>
> >
> > This patch add RDMA fifo size error handle
> > rdma fifo size will not always bigger than the calculated threshold
> > if that case happened, we need set fifo size as the threshold
> >
> > Signed-off-by: Yongqiang Niu <yongqiang.niu at mediatek.com>
> > Signed-off-by: Hsin-Yi Wang <hsinyi at chromium.org>
> > ---
> >  drivers/gpu/drm/mediatek/mtk_disp_rdma.c | 4 ++++
> >  1 file changed, 4 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/mediatek/mtk_disp_rdma.c b/drivers/gpu/drm/mediatek/mtk_disp_rdma.c
> > index b84004394970f..04b9542010b00 100644
> > --- a/drivers/gpu/drm/mediatek/mtk_disp_rdma.c
> > +++ b/drivers/gpu/drm/mediatek/mtk_disp_rdma.c
> > @@ -168,6 +168,10 @@ void mtk_rdma_config(struct device *dev, unsigned int width,
> >        * account for blanking, and with a pixel depth of 4 bytes:
> >        */
> >       threshold = width * height * vrefresh * 4 * 7 / 1000000;
> > +
> > +     if (threshold > rdma_fifo_size)
> > +             threshold = rdma_fifo_size;
> > +
>
> Please see the discussion in [1].
>
> [1]
> https://patchwork.kernel.org/project/linux-mediatek/patch/1607591262-21736-6-git-send-email-yongqiang.niu@mediatek.com/
>
> Regards,
> CK
>

Hi CK,

Even if we set threshold to
threshold = RDMA_FIFO_PSEUDO_SIZE(rdma_fifo_size) * width * height *
vrefresh / 2 / MAX_WIDTH / MAX_HEIGHT / MAX_VREFRESH;

I'm not sure what value MAX_WIDTH, MAX_HEIGHT, and MAX_VREFRESH should
set to for each SoC.
Since there's no conclusion yet, I'll drop this patch in the series,
as this seems not an mt8183 specific fix.


> >       reg = RDMA_FIFO_UNDERFLOW_EN |
> >             RDMA_FIFO_PSEUDO_SIZE(rdma_fifo_size) |
> >             RDMA_OUTPUT_VALID_FIFO_THRESHOLD(threshold);
>


More information about the dri-devel mailing list