Hi, On Tue 15 Oct 19, 19:16, Jernej Škrabec wrote: > Please understand that I was working on this on and off for almost half a year > and checked many times all register values. At one point I tried libvdpau- > sunxi which has no problem with sample video. Still, all relevant register > values were the same. In a desperate attempt, I tried with HW header parsing > which magically solved the issue. After that, I reused values provided in > controls and then finally I made minimal solution as suggested in this patch. Okay thanks for the details. I think I've delayed this for far too long already so I think we should get it in without further delay. The patch apparently no longer applies on top of media/master, but feel free to send out a rebased series with: Acked-by: Paul Kocialkowski Let's leave out 2/3 though, I think I will submit a series adding the flag as indication for the per-slice value in the uAPI and use it in cedrus. Cheers, Paul > > > > I could try and have a look if you have an available sample for testing the > > erroneous case! > > Of course: http://jernej.libreelec.tv/videos/h264/test.mkv > > > > > Another minor thing: do you have some idea of whether the udelay call adds > > significant delay in the process? > > I didn't notice any issue with it. Do you have any better idea? I just didn't > want to make empty loop and udelay is the shortest delay that is provided by > the kernel API. > > Best regards, > Jernej > > > > > Cheers and thanks for the patch! > > > > Paul > > > > > Signed-off-by: Jernej Skrabec > > > --- > > > > > > .../staging/media/sunxi/cedrus/cedrus_h264.c | 30 +++++++++++++++++-- > > > .../staging/media/sunxi/cedrus/cedrus_regs.h | 3 ++ > > > 2 files changed, 30 insertions(+), 3 deletions(-) > > > > > > diff --git a/drivers/staging/media/sunxi/cedrus/cedrus_h264.c > > > b/drivers/staging/media/sunxi/cedrus/cedrus_h264.c index > > > d6a782703c9b..bd848146eada 100644 > > > --- a/drivers/staging/media/sunxi/cedrus/cedrus_h264.c > > > +++ b/drivers/staging/media/sunxi/cedrus/cedrus_h264.c > > > @@ -6,6 +6,7 @@ > > > > > > * Copyright (c) 2018 Bootlin > > > */ > > > > > > +#include > > > > > > #include > > > > > > #include > > > > > > @@ -289,6 +290,28 @@ static void cedrus_write_pred_weight_table(struct > > > cedrus_ctx *ctx,> > > > } > > > > > > } > > > > > > +/* > > > + * It turns out that using VE_H264_VLD_OFFSET to skip bits is not > > > reliable. In + * rare cases frame is not decoded correctly. However, > > > setting offset to 0 and + * skipping appropriate amount of bits with > > > flush bits trigger always works. + */ > > > +static void cedrus_skip_bits(struct cedrus_dev *dev, int num) > > > +{ > > > + int count = 0; > > > + > > > + while (count < num) { > > > + int tmp = min(num - count, 32); > > > > > > + > > > + cedrus_write(dev, VE_H264_TRIGGER_TYPE, > > > + VE_H264_TRIGGER_TYPE_FLUSH_BITS | > > > + VE_H264_TRIGGER_TYPE_N_BITS(tmp)); > > > + while (cedrus_read(dev, VE_H264_STATUS) & > VE_H264_STATUS_VLD_BUSY) > > > + udelay(1); > > > + > > > + count += tmp; > > > + } > > > +} > > > + > > > > > > static void cedrus_set_params(struct cedrus_ctx *ctx, > > > > > > struct cedrus_run *run) > > > > > > { > > > > > > @@ -299,12 +322,11 @@ static void cedrus_set_params(struct cedrus_ctx > > > *ctx, > > > > > > struct vb2_buffer *src_buf = &run->src->vb2_buf; > > > struct cedrus_dev *dev = ctx->dev; > > > dma_addr_t src_buf_addr; > > > > > > - u32 offset = slice->header_bit_size; > > > - u32 len = (slice->size * 8) - offset; > > > + u32 len = slice->size * 8; > > > > > > u32 reg; > > > > > > cedrus_write(dev, VE_H264_VLD_LEN, len); > > > > > > - cedrus_write(dev, VE_H264_VLD_OFFSET, offset); > > > + cedrus_write(dev, VE_H264_VLD_OFFSET, 0); > > > > > > src_buf_addr = vb2_dma_contig_plane_dma_addr(src_buf, 0); > > > cedrus_write(dev, VE_H264_VLD_END, > > > > > > @@ -323,6 +345,8 @@ static void cedrus_set_params(struct cedrus_ctx *ctx, > > > > > > cedrus_write(dev, VE_H264_TRIGGER_TYPE, > > > > > > VE_H264_TRIGGER_TYPE_INIT_SWDEC); > > > > > > + cedrus_skip_bits(dev, slice->header_bit_size); > > > + > > > > > > if (((pps->flags & V4L2_H264_PPS_FLAG_WEIGHTED_PRED) && > > > > > > (slice->slice_type == V4L2_H264_SLICE_TYPE_P || > > > > > > slice->slice_type == V4L2_H264_SLICE_TYPE_SP)) || > > > > > > diff --git a/drivers/staging/media/sunxi/cedrus/cedrus_regs.h > > > b/drivers/staging/media/sunxi/cedrus/cedrus_regs.h index > > > 3329f9aaf975..b52926a54025 100644 > > > --- a/drivers/staging/media/sunxi/cedrus/cedrus_regs.h > > > +++ b/drivers/staging/media/sunxi/cedrus/cedrus_regs.h > > > @@ -538,13 +538,16 @@ > > > > > > > VE_H264_CTRL_SLICE_DECODE_INT) > > > > > > #define VE_H264_TRIGGER_TYPE 0x224 > > > > > > +#define VE_H264_TRIGGER_TYPE_N_BITS(x) (((x) & 0x3f) << 8) > > > > > > #define VE_H264_TRIGGER_TYPE_AVC_SLICE_DECODE (8 << 0) > > > #define VE_H264_TRIGGER_TYPE_INIT_SWDEC (7 << 0) > > > > > > +#define VE_H264_TRIGGER_TYPE_FLUSH_BITS (3 << 0) > > > > > > #define VE_H264_STATUS 0x228 > > > #define VE_H264_STATUS_VLD_DATA_REQ_INT > VE_H264_CTRL_VLD_DATA_REQ_INT > > > #define VE_H264_STATUS_DECODE_ERR_INT > VE_H264_CTRL_DECODE_ERR_INT > > > #define VE_H264_STATUS_SLICE_DECODE_INT > VE_H264_CTRL_SLICE_DECODE_INT > > > > > > +#define VE_H264_STATUS_VLD_BUSY BIT(8) > > > > > > #define VE_H264_STATUS_INT_MASK > VE_H264_CTRL_INT_MASK > > > >