Commit Graph

875 Commits

Author SHA1 Message Date
Andrew Kane
35ab919bf5 Switched to statically-allocated IndexAmRoutine for Postgres 19 [skip ci] 2026-01-21 16:40:35 -08:00
Andrew Kane
aca89aec73 Switched to scanner_isspace for Postgres 17+ - resolves #942 2026-01-05 12:54:04 -08:00
Andrew Kane
3d6dd20469 Removed unneeded check 2025-12-11 16:28:45 -08:00
Andrew Kane
c711da411c Improved includes for indexes 2025-12-11 15:35:37 -08:00
Andrew Kane
3975e212cc Improved includes for ivfkmeans [skip ci] 2025-12-11 13:26:36 -08:00
Andrew Kane
2b741ef068 Improved includes for types 2025-12-11 00:35:18 -08:00
Andrew Kane
f85a7d3fd0 Keep clang-tidy happy [skip ci] 2025-12-10 20:03:26 -08:00
Andrew Kane
5c25f81789 Fixed inconsistent parameter names [skip ci] 2025-12-10 18:49:00 -08:00
Andrew Kane
d72ee71f23 Fixed compilation errors with Postgres 19 2025-12-10 14:59:29 -08:00
Andrew Kane
3ebb9a506e Added varatt headers for Postgres 16+ 2025-10-22 11:56:05 -07:00
Andrew Kane
727d4836d2 Fixed Index Searches in EXPLAIN output for Postgres 18 2025-10-22 11:47:11 -07:00
Andrew Kane
d823c44591 Removed unused parameter [skip ci] 2025-09-27 15:58:18 -07:00
Andrew Kane
8f4aa0751c Removed unused parameters [skip ci] 2025-09-27 15:51:32 -07:00
Andrew Kane
778dacf20c Version bump to 0.8.1 [skip ci] 2025-09-04 17:51:09 -07:00
Andrew Kane
3f687687ee Fixed compilation error with Postgres 19 2025-09-04 15:58:23 -07:00
Julien Rouhaud
dd3a1e9137 Use NIL for empty lists (#890)
Postgres standard way to check for list emptiness is to compare a pointer to
NIL rather than NULL.
2025-08-23 03:31:22 -07:00
Andrew Kane
ea4746f6c0 Updated comment [skip ci] 2025-08-19 14:26:36 -07:00
Andrew Kane
6aec80ccdb Improved performance of binary_quantize function for halfvec 2025-08-19 13:47:11 -07:00
Luca Giacchino
0c9070ba82 Add autovectorized implementation of binary quantize 2025-08-19 13:18:50 -07:00
Andrew Kane
30382418da Ran pgindent [skip ci] 2025-08-19 12:44:54 -07:00
Andrew Kane
e29fc3aa1a Switched to PG_MODULE_MAGIC_EXT for Postgres 18+ [skip ci] 2025-07-31 19:59:27 -07:00
Andrew Kane
5b8b68ba1d Use consistent style [skip ci] 2025-07-27 18:19:28 -07:00
Andrew Kane
e6bad96a03 Ran pgindent [skip ci] 2025-06-18 20:07:46 -07:00
Jon Daniel
3a49d141b3 Vectorizing vector_concat for improved performance (#861)
* Vectorizing vector_concat for improved performance

On an ARM chip this should generate SIMD instructions to copy the two
incoming vectors to the new vector as opposed to doing it all in
software.

* Moving declarations to above CheckDim

* Removing const from dims

* Formatting
2025-06-18 20:06:32 -07:00
Andrew Kane
ce09c9a27a Improved variable names [skip ci] 2025-06-18 17:32:37 -07:00
Andrew Kane
8ad680f009 Ran pgindent [skip ci] 2025-06-18 16:11:15 -07:00
Jon Daniel
fe697e8788 vectorize: optimize VectorSumCenter and HalfvecSumCenter (#860)
* vectorize: optimize VectorSumCenter and HalfvecSumCenter

The functions VectorSumCenter and HalfvecSumCenter were not being
vectorized by the compiler. A few slight changes will allow these
optimizations to take place and get a performance boost by utilizing
SIMD instructions.

This optimization helps improve performance of vector operations in IVF
index building and updating.

* Removing const, commenting that it is only vectoirzed on ARM
2025-06-18 16:09:43 -07:00
Andrew Kane
bf28ed8176 Set random seed for IVFFLAT_BENCH [skip ci] 2025-06-18 10:57:18 -07:00
Andrew Kane
e575866297 Revert "Fixed warnings with Postgres 18 [skip ci]"
This reverts commit 32e95a8598.
2025-04-05 12:56:00 -07:00
Andrew Kane
35f4f7fc80 Improved warning check [skip ci] 2025-04-05 12:38:30 -07:00
Andrew Kane
32e95a8598 Fixed warnings with Postgres 18 [skip ci] 2025-04-05 12:13:38 -07:00
Andrew Kane
a03dc5b7d0 Added fields to IndexAmRoutine for Postgres 18 [skip ci] 2025-04-05 11:31:57 -07:00
Andrew Kane
78ed8f1157 Fixed compilation error with Postgres 18 - fixes #779 2025-02-17 16:38:14 -08:00
Andrew Kane
2fe560dc58 Fixed extra-semi warnings [skip ci] 2025-01-18 13:08:11 -08:00
Andrew Kane
2c53c30415 Fixed compilation error with Postgres 18 2025-01-10 08:50:51 -08:00
Andrew Kane
cd218aae5a Removed unneeded code 2024-10-30 13:05:10 -07:00
Andrew Kane
a2a0b377f0 Removed memory limit debug message from HNSW index scans (EXPLAIN ANALYZE can be used instead) 2024-10-28 13:46:49 -07:00
Andrew Kane
2f770307b8 Removed unused variable [skip ci] 2024-10-28 13:38:42 -07:00
Andrew Kane
c04e16ff5b Removed debug message from IVFFlat index scans [skip ci] 2024-10-28 13:36:57 -07:00
Andrew Kane
c530a3c490 Updated comment [skip ci] 2024-10-28 00:56:10 -07:00
Andrew Kane
04d5e934a1 Scan 2000 more tuples with default work_mem 2024-10-27 22:23:28 -07:00
Andrew Kane
b163b5b196 Moved code [skip ci] 2024-10-27 22:21:43 -07:00
Andrew Kane
2db1b19644 Use greater than [skip ci] 2024-10-27 21:07:22 -07:00
Andrew Kane
305d62146e Updated comment [skip ci] 2024-10-27 21:05:32 -07:00
Andrew Kane
f9d627c9a9 Updated default value of hnsw.scan_mem_multiplier [skip ci] 2024-10-27 21:05:04 -07:00
Andrew Kane
00492d7e57 Ensure max memory fits into Size for HNSW index scans 2024-10-27 14:21:15 -07:00
Andrew Kane
857d716d9e Renamed iterative_search to iterative_scan 2024-10-27 14:02:22 -07:00
Andrew Kane
c5dd2af750 Added comments [skip ci] 2024-10-25 21:39:03 -07:00
Andrew Kane
78b877bdaf Revert "Renamed iterative_search to iterative_scan"
This reverts commit 7043cce893.
2024-10-24 20:32:07 -07:00
Andrew Kane
7043cce893 Renamed iterative_search to iterative_scan 2024-10-24 20:31:43 -07:00