github_mirrors/vectorscan
3
0
Fork 1
mirror of https://github.com/VectorCamp/vectorscan.git synced 2025-11-15 17:02:14 +03:00
Code Issues Packages Projects Releases Wiki Activity

Fix double shufti's vector end false positive (#325)

Browse Source 
* Add regression test for double shufti

It tests for false positive at vector edges.

Signed-off-by: Yoan Picchi <yoan.picchi@arm.com>

* Fix double shufti reporting false positives

Double shufti used to offset one vector, resulting in losing one character
at the end of every vector. This was replaced by a magic value indicating a
match. This meant that if the first char of a pattern fell on the last char of
a vector, double shufti would assume the second character is present and
report a match.
This patch fixes it by keeping the previous vector and feeding its data to the
new one when we shift it, preventing any loss of data.

Signed-off-by: Yoan Picchi <yoan.picchi@arm.com>

* vshl() will call the correct implementation

* implement missing vshr_512_imm(), simplifies caller x86 code

* Fix x86 case, use alignr instead

* it's the reverse, the avx512 alignr is incorrect, need to fix

* Make shufti's OR reduce size agnostic

Signed-off-by: Yoan Picchi <yoan.picchi@arm.com>

* Fix test's array size

Signed-off-by: Yoan Picchi <yoan.picchi@arm.com>

* Fix AVX2/AVX512 alignr implementations and unit tests

* Fix Power VSX alignr

---------

Signed-off-by: Yoan Picchi <yoan.picchi@arm.com>
Co-authored-by: Konstantinos Margaritis <konstantinos@vectorcamp.gr>
This commit is contained in:
ypicchi-arm
2025-06-11 16:55:10 +01:00
committed by GitHub
parent 8a44a6eda2
commit ca70a3d9be
9 changed files with 240 additions and 148 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
unit/internal/supervector.cpp
View File

@@ -697,7 +697,6 @@ TEST(SuperVectorUtilsTest,RShift128_256c){
}
}
/*Define ALIGNR256 macro*/
#define TEST_ALIGNR256(v1, v2, buf, l) { \
auto v_aligned = v2.alignr(v1, l); \
@@ -706,6 +705,7 @@ TEST(SuperVectorUtilsTest,RShift128_256c){
} \
}
TEST(SuperVectorUtilsTest,Alignr256c){
u8 vec[64];
for (int i=0; i<64; i++) {
@@ -713,7 +713,7 @@ TEST(SuperVectorUtilsTest,Alignr256c){
}
auto SP1 = SuperVector<32>::loadu(vec);
auto SP2 = SuperVector<32>::loadu(vec+32);
for(int j=0; j<32; j++) {
for(size_t j=0; j<32; j++) {
TEST_ALIGNR256(SP1, SP2, vec, j);
}
}
@@ -1045,10 +1045,9 @@ TEST(SuperVectorUtilsTest,LShift128_512c){
}
}
/*Define ALIGNR512 macro*/
#define TEST_ALIGNR512(v1, v2, buf, l) { \
auto v_aligned = v1.alignr(v2, l); \
auto v_aligned = v2.alignr(v1, l); \
for (size_t i=0; i<64; i++) { \
ASSERT_EQ(v_aligned.u.u8[i], vec[i + l]); \
} \
@@ -1061,7 +1060,7 @@ TEST(SuperVectorUtilsTest,Alignr512c){
}
auto SP1 = SuperVector<64>::loadu(vec);
auto SP2 = SuperVector<64>::loadu(vec+64);
for(int j=0; j<64; j++){
for(size_t j=0; j<64; j++){
TEST_ALIGNR512(SP1, SP2, vec, j);
}
}