diff options
-rw-r--r-- | drivers/media/platform/vicodec/codec-fwht.c | 83 | ||||
-rw-r--r-- | drivers/media/platform/vicodec/codec-fwht.h | 11 | ||||
-rw-r--r-- | drivers/media/platform/vicodec/codec-v4l2-fwht.c | 255 | ||||
-rw-r--r-- | drivers/media/platform/vicodec/codec-v4l2-fwht.h | 1 | ||||
-rw-r--r-- | drivers/media/platform/vicodec/vicodec-core.c | 40 |
5 files changed, 122 insertions, 268 deletions
diff --git a/drivers/media/platform/vicodec/codec-fwht.c b/drivers/media/platform/vicodec/codec-fwht.c index cf469a1191aa..5d3e844d2802 100644 --- a/drivers/media/platform/vicodec/codec-fwht.c +++ b/drivers/media/platform/vicodec/codec-fwht.c @@ -639,12 +639,13 @@ decide_blocktype(const u8 *cur, const u8 *reference, s16 *deltablock, return vari <= vard ? IBLOCK : PBLOCK; } -static void fill_decoder_block(u8 *dst, const s16 *input, int stride) +static void fill_decoder_block(u8 *dst, const s16 *input, int stride, + unsigned int dst_step) { int i, j; for (i = 0; i < 8; i++) { - for (j = 0; j < 8; j++, input++, dst++) { + for (j = 0; j < 8; j++, input++, dst += dst_step) { if (*input < 0) *dst = 0; else if (*input > 255) @@ -652,17 +653,19 @@ static void fill_decoder_block(u8 *dst, const s16 *input, int stride) else *dst = *input; } - dst += stride - 8; + dst += stride - (8 * dst_step); } } -static void add_deltas(s16 *deltas, const u8 *ref, int stride) +static void add_deltas(s16 *deltas, const u8 *ref, int stride, + unsigned int ref_step) { int k, l; for (k = 0; k < 8; k++) { for (l = 0; l < 8; l++) { - *deltas += *ref++; + *deltas += *ref; + ref += ref_step; /* * Due to quantizing, it might possible that the * decoded coefficients are slightly out of range @@ -673,7 +676,7 @@ static void add_deltas(s16 *deltas, const u8 *ref, int stride) *deltas = 255; deltas++; } - ref += stride - 8; + ref += stride - (8 * ref_step); } } @@ -718,8 +721,8 @@ static u32 encode_plane(u8 *input, u8 *refp, __be16 **rlco, __be16 *rlco_max, ifwht(cf->de_coeffs, cf->de_fwht, blocktype); if (blocktype == PBLOCK) - add_deltas(cf->de_fwht, refp, 8); - fill_decoder_block(refp, cf->de_fwht, 8); + add_deltas(cf->de_fwht, refp, 8, 1); + fill_decoder_block(refp, cf->de_fwht, 8, 1); } input += 8 * input_step; @@ -828,8 +831,10 @@ u32 fwht_encode_frame(struct fwht_raw_frame *frm, return encoding; } -static bool decode_plane(struct fwht_cframe *cf, const __be16 **rlco, u8 *ref, - u32 height, u32 width, u32 coded_width, +static bool decode_plane(struct fwht_cframe *cf, const __be16 **rlco, + u32 height, u32 width, const u8 *ref, u32 ref_stride, + unsigned int ref_step, u8 *dst, + unsigned int dst_stride, unsigned int dst_step, bool uncompressed, const __be16 *end_of_rlco_buf) { unsigned int copies = 0; @@ -841,10 +846,15 @@ static bool decode_plane(struct fwht_cframe *cf, const __be16 **rlco, u8 *ref, height = round_up(height, 8); if (uncompressed) { + int i; + if (end_of_rlco_buf + 1 < *rlco + width * height / 2) return false; - memcpy(ref, *rlco, width * height); - *rlco += width * height / 2; + for (i = 0; i < height; i++) { + memcpy(dst, *rlco, width); + dst += dst_stride; + *rlco += width / 2; + } return true; } @@ -856,15 +866,17 @@ static bool decode_plane(struct fwht_cframe *cf, const __be16 **rlco, u8 *ref, */ for (j = 0; j < height / 8; j++) { for (i = 0; i < width / 8; i++) { - u8 *refp = ref + j * 8 * coded_width + i * 8; + const u8 *refp = ref + j * 8 * ref_stride + + i * 8 * ref_step; + u8 *dstp = dst + j * 8 * dst_stride + i * 8 * dst_step; if (copies) { memcpy(cf->de_fwht, copy, sizeof(copy)); if (stat & PFRAME_BIT) add_deltas(cf->de_fwht, refp, - coded_width); - fill_decoder_block(refp, cf->de_fwht, - coded_width); + ref_stride, ref_step); + fill_decoder_block(dstp, cf->de_fwht, + dst_stride, dst_step); copies--; continue; } @@ -884,23 +896,29 @@ static bool decode_plane(struct fwht_cframe *cf, const __be16 **rlco, u8 *ref, if (copies) memcpy(copy, cf->de_fwht, sizeof(copy)); if (stat & PFRAME_BIT) - add_deltas(cf->de_fwht, refp, coded_width); - fill_decoder_block(refp, cf->de_fwht, coded_width); + add_deltas(cf->de_fwht, refp, + ref_stride, ref_step); + fill_decoder_block(dstp, cf->de_fwht, dst_stride, + dst_step); } } return true; } -bool fwht_decode_frame(struct fwht_cframe *cf, struct fwht_raw_frame *ref, - u32 hdr_flags, unsigned int components_num, - unsigned int width, unsigned int height, - unsigned int coded_width) +bool fwht_decode_frame(struct fwht_cframe *cf, u32 hdr_flags, + unsigned int components_num, unsigned int width, + unsigned int height, const struct fwht_raw_frame *ref, + unsigned int ref_stride, unsigned int ref_chroma_stride, + struct fwht_raw_frame *dst, unsigned int dst_stride, + unsigned int dst_chroma_stride) { const __be16 *rlco = cf->rlc_data; const __be16 *end_of_rlco_buf = cf->rlc_data + (cf->size / sizeof(*rlco)) - 1; - if (!decode_plane(cf, &rlco, ref->luma, height, width, coded_width, + if (!decode_plane(cf, &rlco, height, width, ref->luma, ref_stride, + ref->luma_alpha_step, dst->luma, dst_stride, + dst->luma_alpha_step, hdr_flags & FWHT_FL_LUMA_IS_UNCOMPRESSED, end_of_rlco_buf)) return false; @@ -908,27 +926,30 @@ bool fwht_decode_frame(struct fwht_cframe *cf, struct fwht_raw_frame *ref, if (components_num >= 3) { u32 h = height; u32 w = width; - u32 c = coded_width; if (!(hdr_flags & FWHT_FL_CHROMA_FULL_HEIGHT)) h /= 2; - if (!(hdr_flags & FWHT_FL_CHROMA_FULL_WIDTH)) { + if (!(hdr_flags & FWHT_FL_CHROMA_FULL_WIDTH)) w /= 2; - c /= 2; - } - if (!decode_plane(cf, &rlco, ref->cb, h, w, c, + + if (!decode_plane(cf, &rlco, h, w, ref->cb, ref_chroma_stride, + ref->chroma_step, dst->cb, dst_chroma_stride, + dst->chroma_step, hdr_flags & FWHT_FL_CB_IS_UNCOMPRESSED, end_of_rlco_buf)) return false; - if (!decode_plane(cf, &rlco, ref->cr, h, w, c, + if (!decode_plane(cf, &rlco, h, w, ref->cr, ref_chroma_stride, + ref->chroma_step, dst->cr, dst_chroma_stride, + dst->chroma_step, hdr_flags & FWHT_FL_CR_IS_UNCOMPRESSED, end_of_rlco_buf)) return false; } if (components_num == 4) - if (!decode_plane(cf, &rlco, ref->alpha, height, width, - coded_width, + if (!decode_plane(cf, &rlco, height, width, ref->alpha, ref_stride, + ref->luma_alpha_step, dst->alpha, dst_stride, + dst->luma_alpha_step, hdr_flags & FWHT_FL_ALPHA_IS_UNCOMPRESSED, end_of_rlco_buf)) return false; diff --git a/drivers/media/platform/vicodec/codec-fwht.h b/drivers/media/platform/vicodec/codec-fwht.h index 8f0b790839f8..b6fec2b1cbca 100644 --- a/drivers/media/platform/vicodec/codec-fwht.h +++ b/drivers/media/platform/vicodec/codec-fwht.h @@ -141,9 +141,10 @@ u32 fwht_encode_frame(struct fwht_raw_frame *frm, bool is_intra, bool next_is_intra, unsigned int width, unsigned int height, unsigned int stride, unsigned int chroma_stride); -bool fwht_decode_frame(struct fwht_cframe *cf, struct fwht_raw_frame *ref, - u32 hdr_flags, unsigned int components_num, - unsigned int width, unsigned int height, - unsigned int coded_width); - +bool fwht_decode_frame(struct fwht_cframe *cf, u32 hdr_flags, + unsigned int components_num, unsigned int width, + unsigned int height, const struct fwht_raw_frame *ref, + unsigned int ref_stride, unsigned int ref_chroma_stride, + struct fwht_raw_frame *dst, unsigned int dst_stride, + unsigned int dst_chroma_stride); #endif diff --git a/drivers/media/platform/vicodec/codec-v4l2-fwht.c b/drivers/media/platform/vicodec/codec-v4l2-fwht.c index 515b3115b3c6..f15d76fae45c 100644 --- a/drivers/media/platform/vicodec/codec-v4l2-fwht.c +++ b/drivers/media/platform/vicodec/codec-v4l2-fwht.c @@ -248,14 +248,17 @@ int v4l2_fwht_encode(struct v4l2_fwht_state *state, u8 *p_in, u8 *p_out) int v4l2_fwht_decode(struct v4l2_fwht_state *state, u8 *p_in, u8 *p_out) { - unsigned int i, j, k; u32 flags; struct fwht_cframe cf; - u8 *p, *ref_p; unsigned int components_num = 3; unsigned int version; const struct v4l2_fwht_pixfmt_info *info; unsigned int hdr_width_div, hdr_height_div; + struct fwht_raw_frame dst_rf; + unsigned int dst_chroma_stride = state->stride; + unsigned int ref_chroma_stride = state->ref_stride; + unsigned int dst_size = state->stride * state->coded_height; + unsigned int ref_size; if (!state->info) return -EINVAL; @@ -303,241 +306,29 @@ int v4l2_fwht_decode(struct v4l2_fwht_state *state, u8 *p_in, u8 *p_out) hdr_height_div != info->height_div) return -EINVAL; - if (!fwht_decode_frame(&cf, &state->ref_frame, flags, components_num, - state->visible_width, state->visible_height, - state->coded_width)) + if (prepare_raw_frame(&dst_rf, info, p_out, dst_size)) return -EINVAL; + if (info->planes_num == 3) { + dst_chroma_stride /= 2; + ref_chroma_stride /= 2; + } + if (info->id == V4L2_PIX_FMT_NV24 || + info->id == V4L2_PIX_FMT_NV42) { + dst_chroma_stride *= 2; + ref_chroma_stride *= 2; + } - /* - * TODO - handle the case where the compressed stream encodes a - * different format than the requested decoded format. - */ - switch (state->info->id) { - case V4L2_PIX_FMT_GREY: - ref_p = state->ref_frame.luma; - for (i = 0; i < state->coded_height; i++) { - memcpy(p_out, ref_p, state->visible_width); - p_out += state->stride; - ref_p += state->coded_width; - } - break; - case V4L2_PIX_FMT_YUV420: - case V4L2_PIX_FMT_YUV422P: - ref_p = state->ref_frame.luma; - for (i = 0; i < state->coded_height; i++) { - memcpy(p_out, ref_p, state->visible_width); - p_out += state->stride; - ref_p += state->coded_width; - } - - ref_p = state->ref_frame.cb; - for (i = 0; i < state->coded_height / 2; i++) { - memcpy(p_out, ref_p, state->visible_width / 2); - p_out += state->stride / 2; - ref_p += state->coded_width / 2; - } - ref_p = state->ref_frame.cr; - for (i = 0; i < state->coded_height / 2; i++) { - memcpy(p_out, ref_p, state->visible_width / 2); - p_out += state->stride / 2; - ref_p += state->coded_width / 2; - } - break; - case V4L2_PIX_FMT_YVU420: - ref_p = state->ref_frame.luma; - for (i = 0; i < state->coded_height; i++) { - memcpy(p_out, ref_p, state->visible_width); - p_out += state->stride; - ref_p += state->coded_width; - } - ref_p = state->ref_frame.cr; - for (i = 0; i < state->coded_height / 2; i++) { - memcpy(p_out, ref_p, state->visible_width / 2); - p_out += state->stride / 2; - ref_p += state->coded_width / 2; - } - ref_p = state->ref_frame.cb; - for (i = 0; i < state->coded_height / 2; i++) { - memcpy(p_out, ref_p, state->visible_width / 2); - p_out += state->stride / 2; - ref_p += state->coded_width / 2; - } - break; - case V4L2_PIX_FMT_NV12: - case V4L2_PIX_FMT_NV16: - case V4L2_PIX_FMT_NV24: - ref_p = state->ref_frame.luma; - for (i = 0; i < state->coded_height; i++) { - memcpy(p_out, ref_p, state->visible_width); - p_out += state->stride; - ref_p += state->coded_width; - } + ref_size = state->ref_stride * state->coded_height; - k = 0; - for (i = 0; i < state->coded_height / 2; i++) { - for (j = 0, p = p_out; j < state->coded_width / 2; j++) { - *p++ = state->ref_frame.cb[k]; - *p++ = state->ref_frame.cr[k]; - k++; - } - p_out += state->stride; - } - break; - case V4L2_PIX_FMT_NV21: - case V4L2_PIX_FMT_NV61: - case V4L2_PIX_FMT_NV42: - ref_p = state->ref_frame.luma; - for (i = 0; i < state->coded_height; i++) { - memcpy(p_out, ref_p, state->visible_width); - p_out += state->stride; - ref_p += state->coded_width; - } + if (prepare_raw_frame(&state->ref_frame, info, state->ref_frame.buf, + ref_size)) + return -EINVAL; - k = 0; - for (i = 0; i < state->coded_height / 2; i++) { - for (j = 0, p = p_out; j < state->coded_width / 2; j++) { - *p++ = state->ref_frame.cr[k]; - *p++ = state->ref_frame.cb[k]; - k++; - } - p_out += state->stride; - } - break; - case V4L2_PIX_FMT_YUYV: - k = 0; - for (i = 0; i < state->coded_height; i++) { - for (j = 0, p = p_out; j < state->coded_width / 2; j++) { - *p++ = state->ref_frame.luma[k]; - *p++ = state->ref_frame.cb[k / 2]; - *p++ = state->ref_frame.luma[k + 1]; - *p++ = state->ref_frame.cr[k / 2]; - k += 2; - } - p_out += state->stride; - } - break; - case V4L2_PIX_FMT_YVYU: - k = 0; - for (i = 0; i < state->coded_height; i++) { - for (j = 0, p = p_out; j < state->coded_width / 2; j++) { - *p++ = state->ref_frame.luma[k]; - *p++ = state->ref_frame.cr[k / 2]; - *p++ = state->ref_frame.luma[k + 1]; - *p++ = state->ref_frame.cb[k / 2]; - k += 2; - } - p_out += state->stride; - } - break; - case V4L2_PIX_FMT_UYVY: - k = 0; - for (i = 0; i < state->coded_height; i++) { - for (j = 0, p = p_out; j < state->coded_width / 2; j++) { - *p++ = state->ref_frame.cb[k / 2]; - *p++ = state->ref_frame.luma[k]; - *p++ = state->ref_frame.cr[k / 2]; - *p++ = state->ref_frame.luma[k + 1]; - k += 2; - } - p_out += state->stride; - } - break; - case V4L2_PIX_FMT_VYUY: - k = 0; - for (i = 0; i < state->coded_height; i++) { - for (j = 0, p = p_out; j < state->coded_width / 2; j++) { - *p++ = state->ref_frame.cr[k / 2]; - *p++ = state->ref_frame.luma[k]; - *p++ = state->ref_frame.cb[k / 2]; - *p++ = state->ref_frame.luma[k + 1]; - k += 2; - } - p_out += state->stride; - } - break; - case V4L2_PIX_FMT_RGB24: - case V4L2_PIX_FMT_HSV24: - k = 0; - for (i = 0; i < state->coded_height; i++) { - for (j = 0, p = p_out; j < state->coded_width; j++) { - *p++ = state->ref_frame.cr[k]; - *p++ = state->ref_frame.luma[k]; - *p++ = state->ref_frame.cb[k]; - k++; - } - p_out += state->stride; - } - break; - case V4L2_PIX_FMT_BGR24: - k = 0; - for (i = 0; i < state->coded_height; i++) { - for (j = 0, p = p_out; j < state->coded_width; j++) { - *p++ = state->ref_frame.cb[k]; - *p++ = state->ref_frame.luma[k]; - *p++ = state->ref_frame.cr[k]; - k++; - } - p_out += state->stride; - } - break; - case V4L2_PIX_FMT_RGB32: - case V4L2_PIX_FMT_XRGB32: - case V4L2_PIX_FMT_HSV32: - k = 0; - for (i = 0; i < state->coded_height; i++) { - for (j = 0, p = p_out; j < state->coded_width; j++) { - *p++ = 0; - *p++ = state->ref_frame.cr[k]; - *p++ = state->ref_frame.luma[k]; - *p++ = state->ref_frame.cb[k]; - k++; - } - p_out += state->stride; - } - break; - case V4L2_PIX_FMT_BGR32: - case V4L2_PIX_FMT_XBGR32: - k = 0; - for (i = 0; i < state->coded_height; i++) { - for (j = 0, p = p_out; j < state->coded_width; j++) { - *p++ = state->ref_frame.cb[k]; - *p++ = state->ref_frame.luma[k]; - *p++ = state->ref_frame.cr[k]; - *p++ = 0; - k++; - } - p_out += state->stride; - } - break; - case V4L2_PIX_FMT_ARGB32: - k = 0; - for (i = 0; i < state->coded_height; i++) { - for (j = 0, p = p_out; j < state->coded_width; j++) { - *p++ = state->ref_frame.alpha[k]; - *p++ = state->ref_frame.cr[k]; - *p++ = state->ref_frame.luma[k]; - *p++ = state->ref_frame.cb[k]; - k++; - } - p_out += state->stride; - } - break; - case V4L2_PIX_FMT_ABGR32: - k = 0; - for (i = 0; i < state->coded_height; i++) { - for (j = 0, p = p_out; j < state->coded_width; j++) { - *p++ = state->ref_frame.cb[k]; - *p++ = state->ref_frame.luma[k]; - *p++ = state->ref_frame.cr[k]; - *p++ = state->ref_frame.alpha[k]; - k++; - } - p_out += state->stride; - } - break; - default: + if (!fwht_decode_frame(&cf, flags, components_num, + state->visible_width, state->visible_height, + &state->ref_frame, state->ref_stride, ref_chroma_stride, + &dst_rf, state->stride, dst_chroma_stride)) return -EINVAL; - } return 0; } diff --git a/drivers/media/platform/vicodec/codec-v4l2-fwht.h b/drivers/media/platform/vicodec/codec-v4l2-fwht.h index aa6fa90a48be..53eba97ebc83 100644 --- a/drivers/media/platform/vicodec/codec-v4l2-fwht.h +++ b/drivers/media/platform/vicodec/codec-v4l2-fwht.h @@ -30,6 +30,7 @@ struct v4l2_fwht_state { unsigned int coded_width; unsigned int coded_height; unsigned int stride; + unsigned int ref_stride; unsigned int gop_size; unsigned int gop_cnt; u16 i_frame_qp; diff --git a/drivers/media/platform/vicodec/vicodec-core.c b/drivers/media/platform/vicodec/vicodec-core.c index 467352d71674..970b307b6ffa 100644 --- a/drivers/media/platform/vicodec/vicodec-core.c +++ b/drivers/media/platform/vicodec/vicodec-core.c @@ -153,6 +153,43 @@ static struct vicodec_q_data *get_q_data(struct vicodec_ctx *ctx, return NULL; } +static void copy_cap_to_ref(const u8 *cap, const struct v4l2_fwht_pixfmt_info *info, + struct v4l2_fwht_state *state) +{ + int plane_idx; + u8 *p_ref = state->ref_frame.buf; + unsigned int cap_stride = state->stride; + unsigned int ref_stride = state->ref_stride; + + for (plane_idx = 0; plane_idx < info->planes_num; plane_idx++) { + int i; + unsigned int h_div = (plane_idx == 1 || plane_idx == 2) ? + info->height_div : 1; + const u8 *row_cap = cap; + u8 *row_ref = p_ref; + + if (info->planes_num == 3 && plane_idx == 1) { + cap_stride /= 2; + ref_stride /= 2; + } + + if (plane_idx == 1 && + (info->id == V4L2_PIX_FMT_NV24 || + info->id == V4L2_PIX_FMT_NV42)) { + cap_stride *= 2; + ref_stride *= 2; + } + + for (i = 0; i < state->visible_height / h_div; i++) { + memcpy(row_ref, row_cap, ref_stride); + row_ref += ref_stride; + row_cap += cap_stride; + } + cap += cap_stride * (state->coded_height / h_div); + p_ref += ref_stride * (state->coded_height / h_div); + } +} + static int device_process(struct vicodec_ctx *ctx, struct vb2_v4l2_buffer *src_vb, struct vb2_v4l2_buffer *dst_vb) @@ -194,6 +231,8 @@ static int device_process(struct vicodec_ctx *ctx, ret = v4l2_fwht_decode(state, p_src, p_dst); if (ret < 0) return ret; + copy_cap_to_ref(p_dst, ctx->state.info, &ctx->state); + vb2_set_plane_payload(&dst_vb->vb2_buf, 0, q_dst->sizeimage); } return 0; @@ -1366,6 +1405,7 @@ static int vicodec_start_streaming(struct vb2_queue *q, state->stride = q_data->coded_width * info->bytesperline_mult; + state->ref_stride = q_data->coded_width * info->luma_alpha_step; state->ref_frame.buf = kvmalloc(total_planes_size, GFP_KERNEL); state->ref_frame.luma = state->ref_frame.buf; ctx->comp_max_size = total_planes_size; |