Skip to content

Be stricter about sequence and frame header presence #1407

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
iximeow wants to merge 5 commits into
base: main
Choose a base branch
from
Draft

Be stricter about sequence and frame header presence #1407

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
bdbdaaf
Be stricter about sequence and frame header presence
iximeow May 21, 2025
e47fc4f
astounding that this only made issues on i686
iximeow May 26, 2025
ca0f80a
rename everything
iximeow May 26, 2025
c5d58b1
rustfmt
iximeow May 26, 2025
3f73ff2
more strict header typing
iximeow May 26, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions src/cdef_apply.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@ use crate::src::align::AlignedVec64;
use crate::src::cdef::CdefEdgeFlags;
use crate::src::disjoint_mut::DisjointMut;
use crate::src::internal::Rav1dContext;
use crate::src::internal::Rav1dFrameData;
use crate::src::internal::Rav1dFrameDataWithHeaders;
use crate::src::internal::Rav1dTaskContext;
use crate::src::pic_or_buf::PicOrBuf;
use crate::src::strided::Strided as _;
Expand Down Expand Up @@ -134,7 +134,7 @@ fn adjust_strength(strength: u8, var: c_uint) -> c_int {
pub(crate) fn rav1d_cdef_brow<BD: BitDepth>(
c: &Rav1dContext,
tc: &mut Rav1dTaskContext,
f: &Rav1dFrameData,
f: &Rav1dFrameDataWithHeaders,
p: [Rav1dPictureDataComponentOffset; 3],
lflvl_offset: i32,
by_start: c_int,
Expand All @@ -156,7 +156,7 @@ pub(crate) fn rav1d_cdef_brow<BD: BitDepth>(
let mut ptrs = p;
let sbsz = 16;
let sb64w = f.sb128w << 1;
let frame_hdr = &***f.frame_hdr.as_ref().unwrap();
let frame_hdr = &f.frame_hdr;
let damping = frame_hdr.cdef.damping + bitdepth_min_8;
let layout: Rav1dPixelLayout = f.cur.p.layout;
let uv_idx = (Rav1dPixelLayout::I444 as c_uint).wrapping_sub(layout as c_uint) as c_int;
Expand All @@ -167,7 +167,7 @@ pub(crate) fn rav1d_cdef_brow<BD: BitDepth>(
let uv_dir: &[u8; 8] = &UV_DIRS[(layout == Rav1dPixelLayout::I422) as usize];

let have_tt = c.tc.len() > 1;
let sb128 = f.seq_hdr.as_ref().unwrap().sb128;
let sb128 = f.seq_hdr.sb128;
let resize = frame_hdr.size.width[0] != frame_hdr.size.width[1];
let y_stride: ptrdiff_t = BD::pxstride(f.cur.stride[0]);
let uv_stride: ptrdiff_t = BD::pxstride(f.cur.stride[1]);
Expand Down
86 changes: 44 additions & 42 deletions src/decode.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -68,6 +68,7 @@ use crate::src::internal::Rav1dFrameContext;
use crate::src::internal::Rav1dFrameContextFrameThread;
use crate::src::internal::Rav1dFrameContextLf;
use crate::src::internal::Rav1dFrameData;
use crate::src::internal::Rav1dFrameDataWithHeaders;
use crate::src::internal::Rav1dState;
use crate::src::internal::Rav1dTaskContext;
use crate::src::internal::Rav1dTileState;
Expand Down Expand Up @@ -279,7 +280,7 @@ fn read_mv_residual(ts_c: &mut Rav1dTileStateContext, ref_mv: &mut Mv, mv_prec:

fn read_tx_tree(
t: &mut Rav1dTaskContext,
f: &Rav1dFrameData,
f: &Rav1dFrameDataWithHeaders,
ts_c: &mut Rav1dTileStateContext,
from: TxfmSize,
depth: c_int,
Expand Down Expand Up @@ -417,7 +418,7 @@ fn neg_deinterleave(diff: u8, r#ref: SegmentId, max: u8) -> u8 {
}

fn find_matching_ref(
f: &Rav1dFrameData,
f: &Rav1dFrameDataWithHeaders,
t: &Rav1dTaskContext,
intra_edge_flags: EdgeFlags,
bw4: c_int,
Expand Down Expand Up @@ -786,7 +787,7 @@ struct VarTx {
#[inline(never)]
fn read_vartx_tree(
t: &mut Rav1dTaskContext,
f: &Rav1dFrameData,
f: &Rav1dFrameDataWithHeaders,
ts_c: &mut Rav1dTileStateContext,
b: &Av1Block,
bs: BlockSize,
Expand All @@ -800,7 +801,7 @@ fn read_vartx_tree(
// var-tx tree coding
let mut tx_split = [0u16; 2];
let mut max_ytx = dav1d_max_txfm_size_for_bs[bs as usize][0];
let frame_hdr = &***f.frame_hdr.as_ref().unwrap();
let frame_hdr = &f.frame_hdr;
let txfm_mode = frame_hdr.txfm_mode;
let uvtx;
if b.skip == 0 && (frame_hdr.segmentation.lossless[b.seg_id.get()] || max_ytx == TxfmSize::S4x4)
Expand Down Expand Up @@ -1140,14 +1141,14 @@ fn obmc_lowest_px(
fn decode_b(
c: &Rav1dContext,
t: &mut Rav1dTaskContext,
f: &Rav1dFrameData,
f: &Rav1dFrameDataWithHeaders,
pass: &mut FrameThreadPassState,
bl: BlockLevel,
bs: BlockSize,
bp: BlockPartition,
intra_edge_flags: EdgeFlags,
) -> Result<(), ()> {
let seq_hdr = &***f.seq_hdr.as_ref().unwrap();
let seq_hdr = &f.seq_hdr;
use std::fmt;

/// Helper struct for printing a number as a signed hexidecimal value.
Expand Down Expand Up @@ -1191,7 +1192,7 @@ fn decode_b(
let has_chroma = f.cur.p.layout != Rav1dPixelLayout::I400
&& (bw4 > ss_hor || t.b.x & 1 != 0)
&& (bh4 > ss_ver || t.b.y & 1 != 0);
let frame_type = f.frame_hdr.as_ref().unwrap().frame_type;
let frame_type = f.frame_hdr.frame_type;

let FrameThreadPassState::First(ts_c) = pass else {
match &b.ii {
Expand Down Expand Up @@ -1336,7 +1337,7 @@ fn decode_b(

// segment_id (if seg_feature for skip/ref/gmv is enabled)
let mut seg_pred = false;
let frame_hdr: &Rav1dFrameHeader = &f.frame_hdr.as_ref().unwrap();
let frame_hdr = &f.frame_hdr;
if frame_hdr.segmentation.enabled != 0 {
if frame_hdr.segmentation.update_map == 0 {
b.seg_id = f
Expand Down Expand Up @@ -1788,11 +1789,10 @@ fn decode_b(

let mut y_mode = y_mode;
let mut y_angle = y_angle;
let seq_hdr = f.seq_hdr();
if y_mode == DC_PRED
&& pal_sz[0] == 0
&& cmp::max(b_dim[2], b_dim[3]) <= 3
&& seq_hdr.filter_intra != 0
&& f.seq_hdr.filter_intra != 0
{
let is_filter = rav1d_msac_decode_bool_adapt(
&mut ts_c.msac,
Expand Down Expand Up @@ -1882,7 +1882,7 @@ fn decode_b(
}
}

let frame_hdr = f.frame_hdr();
let frame_hdr = &f.frame_hdr;

let tx = if frame_hdr.segmentation.lossless[b.seg_id.get()] {
b.uvtx = TxfmSize::S4x4;
Expand Down Expand Up @@ -1928,7 +1928,7 @@ fn decode_b(
(bd_fn.recon_b_intra)(f, t, Some(ts_c), bs, intra_edge_flags, b, &intra);
}

if f.frame_hdr().loopfilter.level_y != [0, 0] {
if f.frame_hdr.loopfilter.level_y != [0, 0] {
let lflvl = match ts.lflvl.get() {
TileStateRef::Frame => &f.lf.lvl,
TileStateRef::Local => &*ts.lflvlmem.try_read().unwrap(),
Expand Down Expand Up @@ -1965,7 +1965,7 @@ fn decode_b(
} else {
y_mode
};
let is_inter_or_switch = f.frame_hdr().frame_type.is_inter_or_switch();
let is_inter_or_switch = f.frame_hdr.frame_type.is_inter_or_switch();
CaseSet::<32, false>::many(
[(&t.l, t_dim.lh, 1), (ta, t_dim.lw, 0)],
[bh4 as usize, bw4 as usize],
Expand Down Expand Up @@ -2017,7 +2017,7 @@ fn decode_b(
);
}
}
let frame_hdr = f.frame_hdr();
let frame_hdr = &f.frame_hdr;
if frame_hdr.frame_type.is_inter_or_switch() || frame_hdr.allow_intrabc {
splat_intraref(c, t, &f.rf, bs, bw4 as usize, bh4 as usize);
}
Expand Down Expand Up @@ -3060,7 +3060,7 @@ fn decode_b(
(bd_fn.recon_b_inter)(f, t, Some(ts_c), bs, b, &inter)?;
}

let frame_hdr = f.frame_hdr();
let frame_hdr = &f.frame_hdr;
if frame_hdr.loopfilter.level_y != [0, 0] {
let is_globalmv =
(inter_mode == if is_comp { GLOBALMV_GLOBALMV } else { GLOBALMV }) as c_int;
Expand Down Expand Up @@ -3152,7 +3152,7 @@ fn decode_b(
}

// update contexts
let frame_hdr = &***f.frame_hdr.as_ref().unwrap();
let frame_hdr = &f.frame_hdr;
if frame_hdr.segmentation.enabled != 0 && frame_hdr.segmentation.update_map != 0 {
// Need checked casts here because we're using `from_raw_parts_mut` and an overflow would be UB.
let [by, bx, bh4, bw4] = [t.b.y, t.b.x, bh4, bw4].map(|it| usize::try_from(it).unwrap());
Expand Down Expand Up @@ -3430,7 +3430,7 @@ enum FrameThreadPassState<'a> {
fn decode_sb(
c: &Rav1dContext,
t: &mut Rav1dTaskContext,
f: &Rav1dFrameData,
f: &Rav1dFrameDataWithHeaders,
pass: &mut FrameThreadPassState,
bl: BlockLevel,
edge_index: EdgeIndex,
Expand All @@ -3440,7 +3440,7 @@ fn decode_sb(
let have_h_split = f.bw > t.b.x + hsz;
let have_v_split = f.bh > t.b.y + hsz;

let sb128 = f.seq_hdr().sb128 != 0;
let sb128 = f.seq_hdr.sb128 != 0;
let intra_edge = &IntraEdges::DEFAULT;

if !have_h_split && !have_v_split {
Expand All @@ -3458,7 +3458,7 @@ fn decode_sb(
);
}

let frame_hdr = &***f.frame_hdr.as_ref().unwrap();
let frame_hdr = &f.frame_hdr;

let bp;
let mut bx8 = 0;
Expand Down Expand Up @@ -4098,9 +4098,9 @@ fn check_trailing_bits_after_symbol_coder(msac: &MsacContext) -> Result<(), ()>
pub(crate) fn rav1d_decode_tile_sbrow(
c: &Rav1dContext,
t: &mut Rav1dTaskContext,
f: &Rav1dFrameData,
f: &Rav1dFrameDataWithHeaders,
) -> Result<(), ()> {
let seq_hdr = &***f.seq_hdr.as_ref().unwrap();
let seq_hdr = &f.seq_hdr;
let root_bl = if seq_hdr.sb128 != 0 {
BlockLevel::Bl128x128
} else {
Expand All @@ -4110,7 +4110,7 @@ pub(crate) fn rav1d_decode_tile_sbrow(
let sb_step = f.sb_step;
let tile_row = ts.tiling.row;
let tile_col = ts.tiling.col;
let frame_hdr = &***f.frame_hdr.as_ref().unwrap();
let frame_hdr = &f.frame_hdr;
let col_sb_start = frame_hdr.tiling.col_start_sb[tile_col as usize] as c_int;
let col_sb128_start = col_sb_start >> (seq_hdr.sb128 == 0) as c_int;

Expand Down Expand Up @@ -4157,7 +4157,7 @@ pub(crate) fn rav1d_decode_tile_sbrow(
root_bl,
EdgeIndex::root(),
)?;
if t.b.x & 16 != 0 || f.seq_hdr().sb128 != 0 {
if t.b.x & 16 != 0 || f.seq_hdr.sb128 != 0 {
t.a += 1;
}
}
Expand Down Expand Up @@ -4197,7 +4197,7 @@ pub(crate) fn rav1d_decode_tile_sbrow(
let cdef_idx = &cdef_idx[t.cur_sb_cdef_idx..];
cdef_idx[0].set(-1);
}
let frame_hdr = &***f.frame_hdr.as_ref().unwrap();
let frame_hdr = &f.frame_hdr;
// Restoration filter
for p in 0..3 {
if (f.lf.restore_planes.bits() >> p) & 1 == 0 {
Expand Down Expand Up @@ -4272,15 +4272,15 @@ pub(crate) fn rav1d_decode_tile_sbrow(
root_bl,
EdgeIndex::root(),
)?;
if t.b.x & 16 != 0 || f.seq_hdr().sb128 != 0 {
if t.b.x & 16 != 0 || f.seq_hdr.sb128 != 0 {
t.a += 1;
t.lf_mask = t.lf_mask.map(|i| i + 1);
}
}

if f.seq_hdr().ref_frame_mvs != 0
if f.seq_hdr.ref_frame_mvs != 0
&& c.tc.len() > 1
&& f.frame_hdr().frame_type.is_inter_or_switch()
&& f.frame_hdr.frame_type.is_inter_or_switch()
{
c.dsp.refmvs.save_tmvs.call(
&t.rt,
Expand Down Expand Up @@ -4325,7 +4325,7 @@ pub(crate) fn rav1d_decode_tile_sbrow(

if c.strict_std_compliance
&& (t.b.y >> f.sb_shift) + 1
>= f.frame_hdr().tiling.row_start_sb[tile_row as usize + 1].into()
>= f.frame_hdr.tiling.row_start_sb[tile_row as usize + 1].into()
{
return check_trailing_bits_after_symbol_coder(&ts.context.try_lock().unwrap().msac);
}
Expand All @@ -4335,11 +4335,12 @@ pub(crate) fn rav1d_decode_tile_sbrow(
pub(crate) fn rav1d_decode_frame_init(c: &Rav1dContext, fc: &Rav1dFrameContext) -> Rav1dResult {
let mut f = fc.data.try_write().unwrap();
let f = &mut *f;
let f = f.assert_has_headers_mut();

// TODO: Fallible allocation
f.lf.start_of_tile_row.resize(f.sbh as usize, 0);

let frame_hdr = &***f.frame_hdr.as_ref().unwrap();
let frame_hdr = &f.frame_hdr;
let mut sby = 0;
for tile_row in 0..frame_hdr.tiling.rows {
f.lf.start_of_tile_row[sby as usize] = tile_row;
Expand All @@ -4366,7 +4367,7 @@ pub(crate) fn rav1d_decode_frame_init(c: &Rav1dContext, fc: &Rav1dFrameContext)

let num_sb128 = f.sb128w * f.sb128h;
let size_mul = &ss_size_mul[f.cur.p.layout];
let seq_hdr = &***f.seq_hdr.as_ref().unwrap();
let seq_hdr = &f.seq_hdr;
let hbd = (seq_hdr.hbd != 0) as c_int;
if c.fc.len() > 1 {
let mut tile_idx = 0;
Expand Down Expand Up @@ -4644,10 +4645,10 @@ pub(crate) fn rav1d_decode_frame_init(c: &Rav1dContext, fc: &Rav1dFrameContext)
pub(crate) fn rav1d_decode_frame_init_cdf(
c: &Rav1dContext,
fc: &Rav1dFrameContext,
f: &mut Rav1dFrameData,
f: &mut Rav1dFrameDataWithHeaders,
in_cdf: &CdfThreadContext,
) -> Rav1dResult {
let frame_hdr = &***f.frame_hdr.as_ref().unwrap();
let frame_hdr = &f.frame_hdr;

if frame_hdr.refresh_context != 0 {
*f.out_cdf.cdf_write() = rav1d_cdf_thread_copy(in_cdf);
Expand Down Expand Up @@ -4709,7 +4710,7 @@ pub(crate) fn rav1d_decode_frame_init_cdf(
setup_tile(
c,
ts,
&***f.seq_hdr.as_ref().unwrap(),
&f.seq_hdr,
frame_hdr,
f.bitdepth_max,
f.sb_shift,
Expand Down Expand Up @@ -4759,15 +4760,15 @@ pub(crate) fn rav1d_decode_frame_init_cdf(
Ok(())
}

fn rav1d_decode_frame_main(c: &Rav1dContext, f: &mut Rav1dFrameData) -> Rav1dResult {
fn rav1d_decode_frame_main(c: &Rav1dContext, f: &mut Rav1dFrameDataWithHeaders) -> Rav1dResult {
assert!(c.tc.len() == 1);

let Rav1dContextTaskType::Single(t) = &c.tc[0].task else {
panic!("Expected a single-threaded context");
};
let mut t = t.lock();

let frame_hdr = &***f.frame_hdr.as_ref().unwrap();
let frame_hdr = &f.frame_hdr;

for ctx in &mut f.a[..f.sb128w as usize * frame_hdr.tiling.rows as usize] {
reset_context(ctx, frame_hdr.frame_type.is_key_or_intra(), 0);
Expand All @@ -4786,8 +4787,8 @@ fn rav1d_decode_frame_main(c: &Rav1dContext, f: &mut Rav1dFrameData) -> Rav1dRes
let sbh_end = cmp::min(sbh_end.into(), f.sbh);

for sby in sbh_start.into()..sbh_end {
let seq_hdr = &***f.seq_hdr.as_ref().unwrap();
let frame_hdr = &***f.frame_hdr.as_ref().unwrap();
let seq_hdr = &f.seq_hdr;
let frame_hdr = &f.frame_hdr;
t.b.y = sby << 4 + seq_hdr.sb128;
let by_end = t.b.y + f.sb_step >> 1;
if frame_hdr.use_ref_frame_mvs != 0 {
Expand All @@ -4806,7 +4807,7 @@ fn rav1d_decode_frame_main(c: &Rav1dContext, f: &mut Rav1dFrameData) -> Rav1dRes
t.ts = tile_row * cols + col;
rav1d_decode_tile_sbrow(c, &mut t, f).map_err(|()| EINVAL)?;
}
if f.frame_hdr().frame_type.is_inter_or_switch() {
if f.frame_hdr.frame_type.is_inter_or_switch() {
c.dsp
.refmvs
.save_tmvs
Expand Down Expand Up @@ -4884,15 +4885,16 @@ pub(crate) fn rav1d_decode_frame(c: &Rav1dContext, fc: &Rav1dFrameContext) -> Ra
let mut res = rav1d_decode_frame_init(c, fc);
{
// scope ensures f is dropped before rav1d_decode_frame_exit is called
let mut f = fc.data.try_write().unwrap();
let mut f_guard = fc.data.try_write().unwrap();
let mut f = f_guard.assert_has_headers_mut();
if res.is_ok() {
res = rav1d_decode_frame_init_cdf(c, fc, &mut f, &fc.in_cdf());
}
// wait until all threads have completed
if res.is_ok() {
if c.tc.len() > 1 {
res = rav1d_task_create_tile_sbrow(fc, &f, 0, 1);
drop(f); // release the frame data before waiting for the other threads
drop(f_guard); // release the frame data before waiting for the other threads
let mut task_thread_lock = (*fc.task_thread.ttd).lock.lock();
(*fc.task_thread.ttd).cond.notify_one();
if res.is_ok() {
Expand All @@ -4905,8 +4907,8 @@ pub(crate) fn rav1d_decode_frame(c: &Rav1dContext, fc: &Rav1dFrameContext) -> Ra
drop(task_thread_lock);
res = fc.task_thread.retval.try_lock().unwrap().err_or(());
} else {
res = rav1d_decode_frame_main(c, &mut f);
let frame_hdr = &***f.frame_hdr.as_ref().unwrap();
res = rav1d_decode_frame_main(c, f);
let frame_hdr = &f.frame_hdr;
if res.is_ok() && frame_hdr.refresh_context != 0 && fc.task_thread.update_set.get()
{
rav1d_cdf_thread_update(
Expand Down
Loading
Loading