ddidderr/zstd-rs
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

feat(compress): move window dictionary invalidation to Rust

Browse Source 
The window preparation callbacks already delegated dictionary-distance policy to
Rust, but C still copied projected limits and cleared loaded-dictionary and
match-state pointers. Add a checked pointer projection so Rust applies both
window-limit and dictionary-validity side effects; C retains private window
layout, pointer-index arithmetic, assertions, and diagnostics.

Test Plan:
- ulimit -v 41943040; CARGO_BUILD_JOBS=1 cargo check --tests (rust)
- ulimit -v 41943040; CARGO_BUILD_JOBS=1 cargo clippy --tests -- -A clippy::manual-bits -D warnings (rust)
- ulimit -v 41943040; make -j1
- ulimit -v 41943040; make -j1 -C tests test
This commit is contained in:
ddidderr
2026-07-22 00:34:27 +02:00
parent e3d1f7057d
commit 791b68cba7
2 changed files with 178 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lib/compress/zstd_compress_internal.h
+24 -13
View File
@@ -696,6 +696,18 @@ typedef char ZSTD_rust_window_dict_state_layout[
&& offsetof(ZSTD_rust_windowDictState, loadedDictEnd) == 2 * sizeof(U32)
&& offsetof(ZSTD_rust_windowDictState, invalidate) == 3 * sizeof(U32)
&& sizeof(ZSTD_rust_windowDictState) == 4 * sizeof(U32)) ? 1 : -1];
typedef struct {
U32* lowLimit;
U32* dictLimit;
U32* loadedDictEnd;
const ZSTD_MatchState_t** dictMatchState;
} ZSTD_rust_windowDictPointers;
typedef char ZSTD_rust_window_dict_pointers_layout[
(offsetof(ZSTD_rust_windowDictPointers, lowLimit) == 0
&& offsetof(ZSTD_rust_windowDictPointers, dictLimit) == sizeof(void*)
&& offsetof(ZSTD_rust_windowDictPointers, loadedDictEnd) == 2 * sizeof(void*)
&& offsetof(ZSTD_rust_windowDictPointers, dictMatchState) == 3 * sizeof(void*)
&& sizeof(ZSTD_rust_windowDictPointers) == 4 * sizeof(void*)) ? 1 : -1];
void ZSTD_rust_windowInit(ZSTD_rust_windowUpdateState* state);
U32 ZSTD_rust_windowIsEmpty(const void* nextSrc, const void* base,
U32 dictLimit, U32 lowLimit);
@@ -704,6 +716,11 @@ ZSTD_rust_windowDictState ZSTD_rust_windowEnforceMaxDist(
U32 blockEndIdx, U32 maxDist, ZSTD_rust_windowDictState state);
U32 ZSTD_rust_windowCheckDictValidity(U32 blockEndIdx, U32 maxDist,
U32 loadedDictEnd, U32 dictLimit);
U32 ZSTD_rust_windowEnforceMaxDistState(
U32 blockEndIdx, U32 maxDist, ZSTD_rust_windowDictPointers* state);
U32 ZSTD_rust_windowCheckDictValidityState(
U32 blockEndIdx, U32 maxDist, U32 loadedDictEnd, U32 dictLimit,
U32* loadedDictEndPtr, const ZSTD_MatchState_t** dictMatchStatePtr);
U32 ZSTD_rust_windowUpdate(ZSTD_rust_windowUpdateState* state,
const void* src, size_t srcSize,
int forceNonContiguous);
@@ -1194,8 +1211,9 @@ ZSTD_window_enforceMaxDist(ZSTD_window_t* window,
U32 const blockEndIdx = (U32)((BYTE const*)blockEnd - window->base);
U32 const loadedDictEnd = (loadedDictEndPtr != NULL) ? *loadedDictEndPtr : 0;
U32 const oldDictLimit = window->dictLimit;
ZSTD_rust_windowDictState state = {
window->lowLimit, window->dictLimit, loadedDictEnd, 0
ZSTD_rust_windowDictPointers state = {
&window->lowLimit, &window->dictLimit,
loadedDictEndPtr, dictMatchStatePtr
};
DEBUGLOG(5, "ZSTD_window_enforceMaxDist: blockEndIdx=%u, maxDist=%u, loadedDictEnd=%u",
(unsigned)blockEndIdx, (unsigned)maxDist, (unsigned)loadedDictEnd);
@@ -1213,17 +1231,11 @@ ZSTD_window_enforceMaxDist(ZSTD_window_t* window,
loadedDictEnd is expressed within the referential of the context,
so it can be directly compared against blockEndIdx.
*/
state = ZSTD_rust_windowEnforceMaxDist(blockEndIdx, maxDist, state);
window->lowLimit = state.lowLimit;
window->dictLimit = state.dictLimit;
if (state.invalidate) {
if (ZSTD_rust_windowEnforceMaxDistState(blockEndIdx, maxDist, &state)) {
if (oldDictLimit < window->lowLimit) {
DEBUGLOG(5, "Update dictLimit to match lowLimit, from %u to %u",
(unsigned)oldDictLimit, (unsigned)window->lowLimit);
}
/* On reaching window size, dictionaries are invalidated */
if (loadedDictEndPtr) *loadedDictEndPtr = state.loadedDictEnd;
if (dictMatchStatePtr) *dictMatchStatePtr = NULL;
}
}
@@ -1248,8 +1260,9 @@ ZSTD_checkDictValidity(const ZSTD_window_t* window,
(unsigned)blockEndIdx, (unsigned)maxDist, (unsigned)loadedDictEnd);
assert(blockEndIdx >= loadedDictEnd);
if (ZSTD_rust_windowCheckDictValidity(
blockEndIdx, maxDist, loadedDictEnd, window->dictLimit)) {
if (ZSTD_rust_windowCheckDictValidityState(
blockEndIdx, maxDist, loadedDictEnd, window->dictLimit,
loadedDictEndPtr, dictMatchStatePtr)) {
/* On reaching window size, dictionaries are invalidated.
* For simplification, if window size is reached anywhere within next block,
* the dictionary is invalidated for the full block.
@@ -1260,8 +1273,6 @@ ZSTD_checkDictValidity(const ZSTD_window_t* window,
* dictMatchState, so setting it to NULL is not a problem.
*/
DEBUGLOG(6, "invalidating dictionary for current block (distance > windowSize)");
*loadedDictEndPtr = 0;
*dictMatchStatePtr = NULL;
} else {
if (*loadedDictEndPtr != 0) {
DEBUGLOG(6, "dictionary considered valid for current block");
rust/src/zstd_compress.rs
+154 -4
View File
@@ -10964,8 +10964,31 @@ const _: () = {
assert!(size_of::<ZSTD_rust_windowDictState>() == 4 * size_of::<u32>());
};
#[no_mangle]
pub extern "C" fn ZSTD_rust_windowEnforceMaxDist(
/// Pointers into the C-owned window and match-state fields used by dictionary
/// invalidation. Rust owns the state transition; C retains the containing
/// private structs and computes source indices against their pointer layout.
#[repr(C)]
pub struct ZSTD_rust_windowDictPointers {
low_limit: *mut c_uint,
dict_limit: *mut c_uint,
loaded_dict_end: *mut c_uint,
dict_match_state: *mut *const c_void,
}
const _: () = {
assert!(offset_of!(ZSTD_rust_windowDictPointers, low_limit) == 0);
assert!(offset_of!(ZSTD_rust_windowDictPointers, dict_limit) == size_of::<usize>());
assert!(
offset_of!(ZSTD_rust_windowDictPointers, loaded_dict_end) == 2 * size_of::<usize>()
);
assert!(
offset_of!(ZSTD_rust_windowDictPointers, dict_match_state) == 3 * size_of::<usize>()
);
assert!(size_of::<ZSTD_rust_windowDictPointers>() == 4 * size_of::<usize>());
};
#[inline]
fn window_enforce_max_dist(
block_end_idx: u32,
max_dist: u32,
mut state: ZSTD_rust_windowDictState,
@@ -10985,6 +11008,25 @@ pub extern "C" fn ZSTD_rust_windowEnforceMaxDist(
state
}
#[inline]
fn window_check_dict_validity(
block_end_idx: u32,
max_dist: u32,
loaded_dict_end: u32,
dict_limit: u32,
) -> bool {
block_end_idx > loaded_dict_end.wrapping_add(max_dist) || loaded_dict_end != dict_limit
}
#[no_mangle]
pub extern "C" fn ZSTD_rust_windowEnforceMaxDist(
block_end_idx: u32,
max_dist: u32,
state: ZSTD_rust_windowDictState,
) -> ZSTD_rust_windowDictState {
window_enforce_max_dist(block_end_idx, max_dist, state)
}
#[no_mangle]
pub extern "C" fn ZSTD_rust_windowCheckDictValidity(
block_end_idx: u32,
@@ -10992,8 +11034,75 @@ pub extern "C" fn ZSTD_rust_windowCheckDictValidity(
loaded_dict_end: u32,
dict_limit: u32,
) -> c_uint {
(block_end_idx > loaded_dict_end.wrapping_add(max_dist) || loaded_dict_end != dict_limit)
as c_uint
window_check_dict_validity(block_end_idx, max_dist, loaded_dict_end, dict_limit) as c_uint
}
/// Apply the window-limit transition to C-owned fields and invalidate any
/// dictionary pointers when the active window can no longer reference them.
#[no_mangle]
pub unsafe extern "C" fn ZSTD_rust_windowEnforceMaxDistState(
block_end_idx: u32,
max_dist: u32,
state: *mut ZSTD_rust_windowDictPointers,
) -> c_uint {
if state.is_null() {
return 0;
}
let state = unsafe { &mut *state };
if state.low_limit.is_null() || state.dict_limit.is_null() {
return 0;
}
let loaded_dict_end = if state.loaded_dict_end.is_null() {
0
} else {
unsafe { *state.loaded_dict_end }
};
let next = window_enforce_max_dist(
block_end_idx,
max_dist,
ZSTD_rust_windowDictState {
lowLimit: unsafe { *state.low_limit },
dictLimit: unsafe { *state.dict_limit },
loadedDictEnd: loaded_dict_end,
invalidate: 0,
},
);
unsafe {
*state.low_limit = next.lowLimit;
*state.dict_limit = next.dictLimit;
if next.invalidate != 0 {
if !state.loaded_dict_end.is_null() {
*state.loaded_dict_end = next.loadedDictEnd;
}
if !state.dict_match_state.is_null() {
*state.dict_match_state = ptr::null();
}
}
}
next.invalidate
}
/// Check dictionary validity and apply the C-owned invalidation side effects.
#[no_mangle]
pub unsafe extern "C" fn ZSTD_rust_windowCheckDictValidityState(
block_end_idx: u32,
max_dist: u32,
loaded_dict_end: u32,
dict_limit: u32,
loaded_dict_end_ptr: *mut c_uint,
dict_match_state_ptr: *mut *const c_void,
) -> c_uint {
if loaded_dict_end_ptr.is_null() || dict_match_state_ptr.is_null() {
return 0;
}
let invalid = window_check_dict_validity(block_end_idx, max_dist, loaded_dict_end, dict_limit);
if invalid {
unsafe {
*loaded_dict_end_ptr = 0;
*dict_match_state_ptr = ptr::null();
}
}
invalid as c_uint
}
/// Update the rolling prefix/ext-dictionary window for one input segment.
@@ -19308,6 +19417,47 @@ mod tests {
assert_eq!(ZSTD_rust_windowCheckDictValidity(70, 64, 8, 10), 1);
}
#[test]
fn window_dictionary_pointer_projection_applies_invalidation_side_effects() {
let mut low_limit = 2;
let mut dict_limit = 10;
let mut loaded_dict_end = 8;
let mut dict_match_state = ptr::dangling::<c_void>();
let mut state = ZSTD_rust_windowDictPointers {
low_limit: &mut low_limit,
dict_limit: &mut dict_limit,
loaded_dict_end: &mut loaded_dict_end,
dict_match_state: &mut dict_match_state,
};
assert_eq!(
unsafe { ZSTD_rust_windowEnforceMaxDistState(100, 64, &mut state) },
1
);
assert_eq!(low_limit, 36);
assert_eq!(dict_limit, 36);
assert_eq!(loaded_dict_end, 0);
assert!(dict_match_state.is_null());
loaded_dict_end = 8;
dict_match_state = ptr::dangling::<c_void>();
assert_eq!(
unsafe {
ZSTD_rust_windowCheckDictValidityState(
70,
64,
loaded_dict_end,
dict_limit,
&mut loaded_dict_end,
&mut dict_match_state,
)
},
1
);
assert_eq!(loaded_dict_end, 0);
assert!(dict_match_state.is_null());
}
#[test]
fn window_update_preserves_contiguous_input_and_clips_overlapping_extdict() {
let storage = [0u8; 64];