github_mirrors/vectorscan
3
0
Fork 1
mirror of https://github.com/VectorCamp/vectorscan.git synced 2025-11-17 01:41:51 +03:00
Code Issues Packages Projects Releases Wiki Activity

Speed up truffle with 256b TBL instructions

Browse Source 
256b wide SVE vectors allow some simplification of truffle.
Up to 40% speedup on graviton3. Going from 12500 MB/s to 17000 MB/s
onhe microbenchmark.
SVE2 also offer this capability for 128b vector with a speedup around
25% compared to normal SVE

Add unit tests and benchmark for this wide variant

Signed-off-by: Yoan Picchi <yoan.picchi@arm.com>
This commit is contained in:
Yoan Picchi
2024-04-23 12:04:40 +00:00
parent 154cb6333b
commit 7054378c93
23 changed files with 1125 additions and 78 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
src/nfa/accelcompile.cpp
View File

@@ -97,11 +97,20 @@ void buildAccelSingle(const AccelInfo &info, AccelAux *aux) {
if (outs <= ACCEL_MAX_STOP_CHAR) {
DEBUG_PRINTF("building Truffle for %zu chars\n", outs);
aux->accel_type = ACCEL_TRUFFLE;
aux->truffle.offset = offset;
truffleBuildMasks(info.single_stops,
reinterpret_cast<u8 *>(&aux->truffle.mask1),
reinterpret_cast<u8 *>(&aux->truffle.mask2));
#if defined(CAN_USE_WIDE_TRUFFLE)
if(CAN_USE_WIDE_TRUFFLE) {
aux->accel_type = ACCEL_TRUFFLE_WIDE;
truffleBuildMasksWide(info.single_stops,
reinterpret_cast<u8 *>(&aux->truffle.mask));
} else
#endif
{
aux->accel_type = ACCEL_TRUFFLE;
truffleBuildMasks(info.single_stops,
reinterpret_cast<u8 *>(&aux->truffle.mask_lo),
reinterpret_cast<u8 *>(&aux->truffle.mask_hi));
}
return;
}