Commit Graph

10 Commits (ce94df4ea8ee1c1d410a09fdd1ce85def2e90bdf)

Author SHA1 Message Date
Alexandr Guzhva 39eab0c0ac Add a missing test file reference to CMakeLists.txt (#2474)
Summary: Pull Request resolved: https://github.com/facebookresearch/faiss/pull/2474

Reviewed By: mdouze

Differential Revision: D39545093

fbshipit-source-id: 93852750649936340a62743b5e7744f5c559e608
2022-09-20 07:14:35 -07:00
wx257osn2@yahoo.co.jp 2952487f0e Fix test_cppcontrib_sa_decode.cpp (#2388)
Summary:
This PR contains below changes:

- Conform C++11
    - [`faiss` is written in C++11](https://github.com/facebookresearch/faiss/blob/main/CONTRIBUTING.md#coding-style), but [`faiss/cppcontrib/SaDecodeKernels-avx2-inl.h`](442d9f4a2d/faiss/cppcontrib/SaDecodeKernels-avx2-inl.h) and [the test](442d9f4a2d/tests/test_cppcontrib_sa_decode.cpp) use some C++17 features. This PR rewrites these codes to make them independent to C++17.
- Enable AVX2 on `faiss_test`
    - Currently `faiss_test` is compiled without `-mavx2` even if `-DFAISS_OPT_LEVEL=avx2` , so **`tests/test_cppcontrib_sa_decode.cpp` hasn't checked `faiss/cppcontrib/SaDecodeKernels-avx2-inl.h` at all** . This PR adds `-mavx2` to `faiss_test` if `-DFAISS_OPT_LEVEL=avx2` , so now `tests/test_cppcontrib_sa_decode.cpp` confirms `faiss/cppcontrib/SaDecodeKernels-avx2-inl.h` if `-DFAISS_OPT_LEVEL=avx2` , and does `faiss/cppcontrib/SaDecodeKernels-inl.h` if not `-DFAISS_OPT_LEVEL=avx2` .

Pull Request resolved: https://github.com/facebookresearch/faiss/pull/2388

Reviewed By: mdouze

Differential Revision: D38005738

Pulled By: alexanderguzhva

fbshipit-source-id: b9319c585c6849e1c7a4782770f2d7ce8c0d8660
2022-07-20 23:46:31 -07:00
Alexandr Guzhva 3986ebffca fast C++ templates for sa_decode (#2354)
Summary:
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/2354

A specialized code that provides 2x-3x faster Index::sa_decode for
* IVF256,PQ[1]x8np
* Residual[1]x8,PQ[2]x8

Reviewed By: mdouze

Differential Revision: D37092134

fbshipit-source-id: d848b6cf1aefa826a5ca01e41935aa5d46f5dcc7
2022-06-16 09:20:19 -07:00
Matthijs Douze b813ba805e Reduce mem usage + improve performance for sequential search imlementation
Summary:
Following up on issue https://github.com/facebookresearch/faiss/issues/2054 it seems that this code crashes Faiss (instead of just leaking memory).

Findings:

- when running in MT mode, each search in an indexflat used as coarse quantizer consumes some memory
- this mem consumption does not appear in single-thread mode or with few threads
- in gdb it appears that even when the nb of queries is 1, each search spawns max_threads threads (80 on the test machine)

This diff:

- adds a C++ test that checks how much mem is used when repeatedly searching a vector
- adjusts the number of search threads to the number of query vectors. This is especially useful for single-vector queries.

Reviewed By: beauby

Differential Revision: D31142383

fbshipit-source-id: 134ddaf141e7c52a854cea398f5dbf89951a7ff8
2021-10-05 15:54:04 -07:00
Lucas Hosseini c65f670523 Add separate targets for libfaiss/libfaiss_avx2. (#1772)
Summary:
This should fix the conda builds.

Pull Request resolved: https://github.com/facebookresearch/faiss/pull/1772

Reviewed By: mdouze

Differential Revision: D27365772

Pulled By: beauby

fbshipit-source-id: 12b9d488d475842030feb1a0452acf26dbe6ac01
2021-03-26 14:28:16 -07:00
Lucas Hosseini b7b261cad1 Move from TravisCI to CircleCI (#1315)
Summary:
Depends on https://github.com/facebookresearch/faiss/issues/1313.

Pull Request resolved: https://github.com/facebookresearch/faiss/pull/1315

Reviewed By: mdouze

Differential Revision: D23148557

Pulled By: beauby

fbshipit-source-id: 0a35f17d22aa04db6bd1c16cfc5ff8eee28f1f74
2020-08-15 04:00:51 -07:00
Lucas Hosseini a8e4c5e2d5 Move build to CMake (#1313)
Summary: Pull Request resolved: https://github.com/facebookresearch/faiss/pull/1313

Reviewed By: mdouze

Differential Revision: D22948267

Pulled By: beauby

fbshipit-source-id: ec16fa0342f37672d46fb7886ecc55c7996011c4
2020-08-14 15:03:10 -07:00
Lucas Hosseini ac7005b6ef Remove CMake. (#645) 2018-12-23 18:45:16 +01:00
Adeykin c056f1d320 mkl support in cmake (#123) 2017-05-31 15:41:42 +02:00
Tianwei Shen 80314d9f07 add initial cmake support (#75)
* add initial cmake support

* update cmake, add cmake instructions to INSTALL

* update findopenmp and INSTALL

* change FindOpenBLAS.cmake to cater for macports

- change cblas.h to openblas_config.h since macports does not ship
cblas.h with openblas.

* revise INSTALL for cmake
2017-05-02 11:04:50 +02:00