Summary:
LLVM-15 has a warning `-Wunused-but-set-variable` which we treat as an error because it's so often diagnostic of a code issue. Unused variables can compromise readability or, worse, performance.
This diff either (a) removes an unused variable and, possibly, it's associated code, or (b) qualifies the variable with `[[maybe_unused]]`, mostly in cases where the variable _is_ used, but, eg, in an `assert` statement that isn't present in production code.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: danzimm
Differential Revision: D52847958
fbshipit-source-id: b3f98d8e0e4fd0e13191f0c0d907edd7c0e78db0
Summary: reference the new Faiss paper in the docs. Also added Gergely as a co-author of Faiss. Some minor updates.
Reviewed By: mlomeli1
Differential Revision: D52829321
fbshipit-source-id: 0f1845beace6cd88f809ba50b87a8d446475d30b
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: meyering
Differential Revision: D52582914
fbshipit-source-id: edb3983635de4343c4a0b20096e81852cfa058c6
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: meyering
Differential Revision: D52582837
fbshipit-source-id: 275d20f940102030989f9bcfb4884891457b149c
Summary:
This PR introduces the offline IVF (OIVF) framework which contains some tooling to run search using IVFPQ indexes (plus OPQ pretransforms) for large batches of queries using [big_batch_search](https://github.com/mlomeli1/faiss/blob/main/contrib/big_batch_search.py) and GPU faiss. See the [README](36226f5fe8/demos/offline_ivf/README.md) for details about using this framework.
This PR includes the following unit tests, which can be run with the unittest library as so:
````
~/faiss/demos/offline_ivf$ python3 -m unittest tests/test_iterate_input.py -k test_iterate_back
````
In test_offline_ivf:
````
test_consistency_check
test_train_index
test_index_shard_equal_file_sizes
test_index_shard_unequal_file_sizes
test_search
test_evaluate_without_margin
test_evaluate_without_margin_OPQ
````
In test_iterate_input:
````
test_iterate_input_file_larger_than_batch
test_get_vs_iterate
test_iterate_back
````
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/3202
Reviewed By: algoriddle
Differential Revision: D52734222
Pulled By: mlomeli1
fbshipit-source-id: 61fd0084277c1b14bdae1189db8ae43340611e16
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: meyering
Differential Revision: D52582870
fbshipit-source-id: 4c81218e109ff97052036645ab33bc9b17802e91
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: palmje
Differential Revision: D52582927
fbshipit-source-id: be03fb1354898d0637d287f285ea99d096c39008
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: meyering
Differential Revision: D52582813
fbshipit-source-id: 02a742df17614c546e10446bc530835e876942f0
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: meyering
Differential Revision: D52582928
fbshipit-source-id: 57bd901e87cbb8ddfd423c8ae6baefe1048c206f
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: meyering
Differential Revision: D52582916
fbshipit-source-id: 367ea85b569063efa50c82702c3ba5efa2fb92f9
Summary:
LLVM-15 has a warning `-Wunused-but-set-variable` which we treat as an error because it's so often diagnostic of a code issue. Unused variables can compromise readability or, worse, performance.
This diff either (a) removes an unused variable and, possibly, it's associated code, or (b) qualifies the variable with `[[maybe_unused]]`, mostly in cases where the variable _is_ used, but, eg, in an `assert` statement that isn't present in production code.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: meyering
Differential Revision: D52734585
fbshipit-source-id: be5da3411d2965f6cd2516376598576049c03f09
Summary:
LLVM-15 has a warning `-Wunused-but-set-variable` which we treat as an error because it's so often diagnostic of a code issue. Unused variables can compromise readability or, worse, performance.
This diff either (a) removes an unused variable and, possibly, it's associated code, or (b) qualifies the variable with `[[maybe_unused]]`, mostly in cases where the variable _is_ used, but, eg, in an `assert` statement that isn't present in production code.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: meyering
Differential Revision: D52734573
fbshipit-source-id: edf315a4e2e7ddfaf8b879b7f4a3fc3a6c2e607f
Summary:
Diff summary courtesy of Metamate:
This diff bumps the version of libraft to fix nightly issues in the faiss-gpu-raft package. The code changes include updating the libraft version in the meta.yaml file to 24.02. This change is necessary to ensure that the package is compatible with the latest version of libraft.
Reviewed By: mlomeli1
Differential Revision: D52736355
fbshipit-source-id: 7ac2427210c06eb2d1c3ab4a3d37d48494fb039e
Summary:
### Description
We have SIMD support for x86 architecture using AVX2 optimization. But, we don't have a similar optimization for ARM architecture in Scalar Quantizer for the Quantization Type `QT_FP16`. This PR adds SIMD support for ARM using NEON optimization.
### Issues resolved
https://github.com/facebookresearch/faiss/issues/3014
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/3166
Reviewed By: algoriddle, pemazare
Differential Revision: D52510486
Pulled By: mdouze
fbshipit-source-id: 2bb360475a0d9e0816c8e05b44d7da7f2e6b28c5
Summary:
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/3190
This diff adds more result handlers in order to expose them externally.
This enables range search for HSNW and Fast Scan, and nprobe parameter support for FastScan.
Reviewed By: pemazare
Differential Revision: D52547384
fbshipit-source-id: 271da5ffea6411df3d8e50641abade18bd7b774b
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: algoriddle
Differential Revision: D52582910
fbshipit-source-id: 92ee06e819f15c411a3fee75a6c9e17fba96a7f3
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: algoriddle
Differential Revision: D52582823
fbshipit-source-id: 575c286c0445e2388038aac10327d6c1bd63d6df
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: algoriddle
Differential Revision: D52582797
fbshipit-source-id: 4ade95834fee68a4ac86b4d03a00df0760941f4e
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: algoriddle
Differential Revision: D52582890
fbshipit-source-id: ef7e44805fb212c6415030b98f65c00b2fd4a9b7
Summary:
Our upcoming compiler upgrade will require us not to have shadowed variables. Such variables have a _high_ bug rate and reduce readability, so we would like to avoid them even if the compiler was not forcing us to do so.
This codemod attempts to fix an instance of a shadowed variable. Please review with care: if it's failed the result will be a silent bug.
**What's a shadowed variable?**
Shadowed variables are variables in an inner scope with the same name as another variable in an outer scope. Having the same name for both variables might be semantically correct, but it can make the code confusing to read! It can also hide subtle bugs.
This diff fixes such an issue by renaming the variable.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: dmm-fb
Differential Revision: D52582819
fbshipit-source-id: 34c4d5eca31bcc2df5872c9eb3a286335c8ae131
Summary:
`-Wunused-exception-parameter` has identified an unused exception parameter. This diff removes it.
This:
```
try {
...
} catch (exception& e) {
// no use of e
}
```
should instead be written as
```
} catch (exception&) {
```
If the code compiles, this is safe to land.
Reviewed By: dmm-fb
Differential Revision: D51977749
fbshipit-source-id: aab6f8bcb8a0d9273b7894b8e85615d4a31cac86
Summary:
Error: https://www.internalfb.com/sandcastle/workflow/1661828262501424314
Due to a recent change (D52393687) to increase compiler warnings arcoss faiss/PACKAGE aggregator build is failing. The annotation [[fallthrough]]; needs to be used in every [nested]case statement where fallthrough can occur.
WHAT I DIDNT DO
I did NOT EXPORTED TO OSS FAISS
mdouze please do this if it is required. I do not know your policy. Prior to landing i saw the message but was unsure. As well as not having github account.
Reviewed By: r-barnes
Differential Revision: D52420730
fbshipit-source-id: 92ec8382226ea35bdacb668b3c329bd8b41a67fc
Summary:
## Context
Due to a recent change (D52393687) to increase compiler warnings arcoss faiss/PACKAGE aggregator [build](https://www.internalfb.com/chronos/job/gp/9007207455707765) is failing. The annotation [[fallthrough]]; needs to be used in every [nested]case statement where fallthrough can occur.
## COMPILER ERROR
```
[2023-12-22T22:06:48.478-08:00] Action failed: fbcode//faiss:faiss (ovr_config//platform/linux:x86_64-fbcode-platform010-clang15-jemalloc-master#a916d57931e3a679) (cxx_compile impl/LocalSearchQuantizer.cpp)
[CONTEXT] [2023-12-22T22:06:48.478-08:00] Remote command returned non-zero exit code 1
[CONTEXT] [2023-12-22T22:06:48.478-08:00] Remote action, reproduce with: `frecli cas download-action d02e5a31e82ab2a6a2ce8b84dff4d4bce3997ca49783013f860d16dd566edaaf:145`
buck-out/v2/gen/fbcode/a916d57931e3a679/faiss/__faiss__/buck-headers/faiss/utils/hamming_distance/generic-inl.h:278:21: error: unannotated fall-through between switch labels [-Werror,-Wimplicit-fallthrough]
[CONTEXT] case 7:
[CONTEXT] ^
buck-out/v2/gen/fbcode/a916d57931e3a679/faiss/__faiss__/buck-headers/faiss/utils/hamming_distance/generic-inl.h:281:21: error: unannotated fall-through between switch labels [-Werror,-Wimplicit-fallthrough]
[CONTEXT] case 6:
[CONTEXT] ^
buck-out/v2/gen/fbcode/a916d57931e3a679/faiss/__faiss__/buck-headers/faiss/utils/hamming_distance/generic-inl.h:284:21: error: unannotated fall-through between switch labels [-Werror,-Wimplicit-fallthrough]
[CONTEXT] case 5:
```
## WHAT I DID
+1 annotation **[[fallthrough]];** at every case statement that required it.
## WHAT I DIDNT DO
- I did NOT **EXPORTED TO OSS FAISS**
- mdouze please do this if it is required. I do not know your policy. Prior to landing i saw the message but was unsure. As well as not having github account.
Differential Revision: D52396801
fbshipit-source-id: a563068af00135f7fd24de6633007a1957fc6e31
Summary: Revert so that the test data can be replaced with synthetic
Reviewed By: mdouze
Differential Revision: D52388604
fbshipit-source-id: c0037635a4e66c54d42400294d13d9a80610b845
Summary: `-Wunused-function` has identified an unused function. This diff removes it. In many cases these functions have existed for years in an unused state.
Reviewed By: bunnypak
Differential Revision: D52355736
fbshipit-source-id: e315cf420053472218bbf2fdd647ab073c6f6cd6
Summary: `-Wunused-function` has identified an unused function. This diff removes it. In many cases these functions have existed for years in an unused state.
Reviewed By: bunnypak
Differential Revision: D52355753
fbshipit-source-id: 54c8b47224c1dafc4e2464fe4e528c78ecabf0b8
Summary:
This PR introduces the offline IVF (OIVF) framework which contains some tooling to run search using IVFPQ indexes (plus OPQ pretransforms) for large batches of queries using [big_batch_search](https://github.com/mlomeli1/faiss/blob/main/contrib/big_batch_search.py) and GPU faiss. See the [README](https://github.com/mlomeli1/faiss/blob/oivf/demos/offline_ivf/README.md) for details about using this framework.
This PR includes the following unit tests, which can be run with the unittest library as so:
````
~/faiss/demos/offline_ivf$ python3 -m unittest tests/test_iterate_input.py -k test_iterate_back
````
In test_offline_ivf:
````
test_consistency_check
test_train_index
test_index_shard_equal_file_sizes
test_index_shard_unequal_file_sizes
test_search
test_evaluate_without_margin
test_evaluate_without_margin_OPQ
test_evaluate_with_margin
test_split_batch_size_bigger_than_file_sizes
test_split_batch_size_smaller_than_file_sizes
test_split_files_with_corrupted_input_file
````
In test_iterate_input:
````
test_iterate_input_file_larger_than_batch
test_get_vs_iterate
test_iterate_back
````
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/3175
Reviewed By: algoriddle
Differential Revision: D52218447
Pulled By: mlomeli1
fbshipit-source-id: 78b12457c79b02eb2c9ae993560f2e295798e7e5
Summary:
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/3169
InvertedListScanner.keep_max is used in iterate_codes method, this field is not set to true when metric is IP, causing problems when using InvertedListsIterator interface.
Reviewed By: algoriddle
Differential Revision: D52006595
fbshipit-source-id: 6adbe377c0b9f198c9fd8704869133407ef845e5
Summary:
Size of pointer `sub_x` used instead of size of its data. This is likely to lead to a buffer overflow if the user is not lucky enough to be in a x32 bit machine where `sizeof(float*) == sizeof(float)`.
You probably intend to write `sizeof(*sub_x)` or `sizeof(float)` ?
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/3146
Reviewed By: mlomeli1
Differential Revision: D51618892
Pulled By: algoriddle
fbshipit-source-id: 82acd98009515e963c547d5c62946f93ef5561b5
Summary:
`WorkerThread.h` uses `std::function` but it does not have `#include <functional>` and this causes compile errors in some cases. This simple PR fixes this issue.
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/3162
Reviewed By: mlomeli1
Differential Revision: D51987803
Pulled By: algoriddle
fbshipit-source-id: 21db00a7c9836e83401bb2c82e86a49c80d52f88
Summary:
The inconsistency in variable types would generate waaaaaaay too many warnings `comparison of integer expressions of different signedness` during the compilation if all the warnings are made reported
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/3147
Reviewed By: mlomeli1
Differential Revision: D51988074
Pulled By: algoriddle
fbshipit-source-id: 31a57bbe3a14105a968d3f06c0807c2fe60f1a62
Summary:
`-Wextra-semi` or `-Wextra-semi-stmt`
If the code compiles, this is safe to land.
Reviewed By: palmje
Differential Revision: D51995085
fbshipit-source-id: c0a4f134f9361143f337521759793d1cd86e145c
Summary:
`-Wextra-semi` or `-Wextra-semi-stmt`
If the code compiles, this is safe to land.
Reviewed By: palmje
Differential Revision: D51995074
fbshipit-source-id: 4863250ff9f4b87241f6c01cf3b39b6b87172852
Summary:
`-Wunused-exception-parameter` has identified an unused exception parameter. This diff removes it.
This:
```
try {
...
} catch (exception& e) {
// no use of e
}
```
should instead be written as
```
} catch (exception&) {
```
If the code compiles, this is safe to land.
Reviewed By: algoriddle
Differential Revision: D51977717
fbshipit-source-id: 47a308ff3570935829531663252409b4e0cfdec4
Summary:
One of our [C++ 2024 Modern Code Better Engineering Goals](https://fb.workplace.com/groups/fbcode.fyi/posts/6837863056249437) is to reduce or eliminate the use of printf in our codebase, replacing it with safer and more performant alternatives.
This diff replaces `printf` with its preferred alternative: `fmt::print`.
This diff was auto-generated by fbcode/scripts/rbarnes/printf_upgrader.py, please reach out to r-barnes if you see problems.
- If you approve of this diff, please use the "Accept & Ship" button :-)
(11 file modified.)
Reviewed By: bcardosolopes
Differential Revision: D51486397
fbshipit-source-id: 230540b26ec8717d8e8aa4ae14d6a34718a964ff
Summary:
Enables avx512 optimized code (AVX512 subsets F, CD, VL, DQ and BW, which are available for Intel Skylake+ and all AMD Zen4).
Also, introduces `FAISS_OPT_LEVEL` environment variable. Set it to `AVX2`, `AVX512` or empty to pick the appropriate x86_64 instruction set.
Compiled via the following
```
cmake -B build -DCMAKE_BUILD_TYPE=Release -DFAISS_ENABLE_GPU=OFF -DFAISS_OPT_LEVEL=avx512 -DBUILD_TESTING=ON .
make -C build -j 8 faiss_test
make -C build -j 8 swigfaiss
make -C build -j 8 swigfaiss_avx2
make -C build -j 8 swigfaiss_avx512
cd build/faiss/python
python3 setup.py build
python3 setup.py install --force
```
Now, running the following script `1.py`
```
import logging
logging.basicConfig(level=logging.DEBUG)
import faiss
```
produces the following:
```
root@6179abeef23c:~/faiss# LOGLEVEL=DEBUG FAISS_OPT_LEVEL= python3 1.py
DEBUG:faiss.loader:Using as an instruction set.
INFO:faiss.loader:Loading faiss.
INFO:faiss.loader:Successfully loaded faiss.
root@6179abeef23c:~/faiss# LOGLEVEL=DEBUG FAISS_OPT_LEVEL=AVX2 python3 1.py
DEBUG:faiss.loader:Using AVX2 as an instruction set.
INFO:faiss.loader:Loading faiss with AVX2 support.
INFO:faiss.loader:Successfully loaded faiss with AVX2 support.
root@6179abeef23c:~/faiss# LOGLEVEL=DEBUG FAISS_OPT_LEVEL=AVX512 python3 1.py
DEBUG:faiss.loader:Using AVX512 as an instruction set.
INFO:faiss.loader:Loading faiss with AVX512 support.
INFO:faiss.loader:Successfully loaded faiss with AVX512 support.
root@6179abeef23c:~/faiss# LOGLEVEL=DEBUG python3 1.py
DEBUG:faiss.loader:Environment variable FAISS_OPT_LEVEL is not set, so let's pick the instruction set according to the current CPU
INFO:faiss.loader:Loading faiss with AVX512 support.
INFO:faiss.loader:Successfully loaded faiss with AVX512 support.
```
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/3150
Reviewed By: algoriddle
Differential Revision: D51701077
Pulled By: mdouze
fbshipit-source-id: 4db05a287e763ff1ce1f676df7f7402532bf1e9e
Summary:
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/3144
Visualize results of running the benchmark with Pareto optima filtering:
1. per index or across indices
2. for space, time or space & time
3. knn or range search, the latter @ specific precision
Reviewed By: mdouze
Differential Revision: D51552775
fbshipit-source-id: d4f29e3d46ef044e71b54439b3972548c86af5a7
Summary:
1. Support for index construction parameters outside of the factory string (arbitrary depth of quantizers).
2. Refactor that provides an index wrapper which is a prereq for the optimizer, which will generate indices from pre-optimized components (particularly quantizers)
Reviewed By: mdouze
Differential Revision: D51427452
fbshipit-source-id: 014d05dd798d856360f2546963e7cad64c2fcaeb
Summary:
1. Support `search_preassigned` in IVFFastScan
2. `try_extract_index_ivf` to search recursively and support `IndexRefine`
3. `get_InvertedListScanner` to fail where not available
4. Workaround an OpenMP issue with `IndexIVFSpectralHash`
Reviewed By: mdouze
Differential Revision: D51427241
fbshipit-source-id: 365e3f11d24e80f101f986fc358c28dcc00805fa
Summary:
Introduces `FAISS_ALWAYS_INLINE` pragma directive and improves `ScalarQuantizer` performance with it.
Most of performance-critical methods for `ScalarQuantizer` are marked with this new directive, because a compiler (especially, an old one) may be unable to inline it properly. In some of my GCC experiments, such an inlining yields +50% queries per second in a search.
Pull Request resolved: https://github.com/facebookresearch/faiss/pull/3141
Reviewed By: algoriddle
Differential Revision: D51615609
Pulled By: mdouze
fbshipit-source-id: 9c755c3e1a289b5d498306c1b9d6fcc21b0bec28
Summary: It seems that for some build modes, swig chokes on static_assert, so protect this with #idndef SWIG. Let's see what the tests say....
Reviewed By: algoriddle
Differential Revision: D50971042
fbshipit-source-id: 83e2ccb464c0bd024cbf3a494357147d75a76ca2
