Allow for NUMA memory replication for NNUE weights. Bind threads to ensure execution on a specific NUMA node.
This patch introduces NUMA memory replication, currently only utilized for the NNUE weights. Along with it comes all machinery required to identify NUMA nodes and bind threads to specific processors/nodes. It also comes with small changes to Thread and ThreadPool to allow easier execution of custom functions on the designated thread. Old thread binding (WinProcGroup) machinery is removed because it's incompatible with this patch. Small changes to unrelated parts of the code were made to ensure correctness, like some classes being made unmovable, raw pointers replaced with unique_ptr. etc.
Windows 7 and Windows 10 is partially supported. Windows 11 is fully supported. Linux is fully supported, with explicit exclusion of Android. No additional dependencies.
-----------------
A new UCI option `NumaPolicy` is introduced. It can take the following values:
```
system - gathers NUMA node information from the system (lscpu or windows api), for each threads binds it to a single NUMA node
none - assumes there is 1 NUMA node, never binds threads
auto - this is the default value, depends on the number of set threads and NUMA nodes, will only enable binding on multinode systems and when the number of threads reaches a threshold (dependent on node size and count)
[[custom]] -
// ':'-separated numa nodes
// ','-separated cpu indices
// supports "first-last" range syntax for cpu indices,
for example '0-15,32-47:16-31,48-63'
```
Setting `NumaPolicy` forces recreation of the threads in the ThreadPool, which in turn forces the recreation of the TT.
The threads are distributed among NUMA nodes in a round-robin fashion based on fill percentage (i.e. it will strive to fill all NUMA nodes evenly). Threads are bound to NUMA nodes, not specific processors, because that's our only requirement and the OS can schedule them better.
Special care is made that maximum memory usage on systems that do not require memory replication stays as previously, that is, unnecessary copies are avoided.
On linux the process' processor affinity is respected. This means that if you for example use taskset to restrict Stockfish to a single NUMA node then the `system` and `auto` settings will only see a single NUMA node (more precisely, the processors included in the current affinity mask) and act accordingly.
-----------------
We can't ensure that a memory allocation takes place on a given NUMA node without using libnuma on linux, or using appropriate custom allocators on windows (https://learn.microsoft.com/en-us/windows/win32/memory/allocating-memory-from-a-numa-node), so to avoid complications the current implementation relies on first-touch policy. Due to this we also rely on the memory allocator to give us a new chunk of untouched memory from the system. This appears to work reliably on linux, but results may vary.
MacOS is not supported, because AFAIK it's not affected, and implementation would be problematic anyway.
Windows is supported since Windows 7 (https://learn.microsoft.com/en-us/windows/win32/api/processtopologyapi/nf-processtopologyapi-setthreadgroupaffinity). Until Windows 11/Server 2022 NUMA nodes are split such that they cannot span processor groups. This is because before Windows 11/Server 2022 it's not possible to set thread affinity spanning processor groups. The splitting is done manually in some cases (required after Windows 10 Build 20348). Since Windows 11/Server 2022 we can set affinites spanning processor group so this splitting is not done, so the behaviour is pretty much like on linux.
Linux is supported, **without** libnuma requirement. `lscpu` is expected.
-----------------
Passed 60+1 @ 256t 16000MB hash: https://tests.stockfishchess.org/tests/view/6654e443a86388d5e27db0d8
```
LLR: 2.95 (-2.94,2.94) <0.00,10.00>
Total: 278 W: 110 L: 29 D: 139
Ptnml(0-2): 0, 1, 56, 82, 0
```
Passed SMP STC: https://tests.stockfishchess.org/tests/view/6654fc74a86388d5e27db1cd
```
LLR: 2.95 (-2.94,2.94) <-1.75,0.25>
Total: 67152 W: 17354 L: 17177 D: 32621
Ptnml(0-2): 64, 7428, 18408, 7619, 57
```
Passed STC: https://tests.stockfishchess.org/tests/view/6654fb27a86388d5e27db15c
```
LLR: 2.94 (-2.94,2.94) <-1.75,0.25>
Total: 131648 W: 34155 L: 34045 D: 63448
Ptnml(0-2): 426, 13878, 37096, 14008, 416
```
fixes#5253
closes https://github.com/official-stockfish/Stockfish/pull/5285
No functional change
This speedup was first inspired by a comment by @AndyGrant on my recent
PR "If mullo_epi16 would preserve the signedness, then this could be
used to remove 50% of the max operations during the halfkp-pairwise
mat-mul relu deal."
That got me thinking, because although mullo_epi16 did not preserve the
signedness, mulhi_epi16 did, and so we could shift left and then use
mulhi_epi16, instead of shifting right after the mullo.
However, due to some issues with shifting into the sign bit, the FT
weights and biases had to be multiplied by 2 for the optimisation to
work.
Speedup on "Arch=x86-64-bmi2 COMP=clang", courtesy of @Torom
Result of 50 runs
base (...es/stockfish) = 962946 +/- 1202
test (...ise-max-less) = 979696 +/- 1084
diff = +16750 +/- 1794
speedup = +0.0174
P(speedup > 0) = 1.0000
CPU: 4 x Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz
Hyperthreading: on
Also a speedup on "COMP=gcc", courtesy of Torom once again
Result of 50 runs
base (...tockfish_gcc) = 966033 +/- 1574
test (...max-less_gcc) = 983319 +/- 1513
diff = +17286 +/- 2515
speedup = +0.0179
P(speedup > 0) = 1.0000
CPU: 4 x Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz
Hyperthreading: on
Passed STC:
LLR: 2.96 (-2.94,2.94) <0.00,2.00>
Total: 67712 W: 17715 L: 17358 D: 32639
Ptnml(0-2): 225, 7472, 18140, 7759, 260
https://tests.stockfishchess.org/tests/view/664c1d75830eb9f886616906
closes https://github.com/official-stockfish/Stockfish/pull/5282
No functional change
Removes some max calls
Some speedup stats, courtesy of @AndyGrant (albeit measured in an alternate implementation)
Dev 749240 nps
Base 748495 nps
Gain 0.100%
289936 games
STC:
LLR: 2.94 (-2.94,2.94) <-1.75,0.25>
Total: 203040 W: 52213 L: 52179 D: 98648
Ptnml(0-2): 480, 20722, 59139, 20642, 537
https://tests.stockfishchess.org/tests/view/664805fe6dcff0d1d6b05f2ccloses#5261
No functional change
Stockfish appears to take too much time on the first move of a game and
then not enough on moves 2,3,4... Probably caused by most of the factors
that increase time usually applying on the first move.
Attempts to give more time to the subsequent moves have not worked so
far, but this change to simply reduce first move time by 5% worked.
STC 10+0.1 :
LLR: 2.96 (-2.94,2.94) <0.00,2.00>
Total: 78496 W: 20516 L: 20135 D: 37845
Ptnml(0-2): 340, 8859, 20456, 9266, 327
https://tests.stockfishchess.org/tests/view/663d47bf507ebe1c0e9200ba
LTC 60+0.6 :
LLR: 2.95 (-2.94,2.94) <0.50,2.50>
Total: 94872 W: 24179 L: 23751 D: 46942
Ptnml(0-2): 61, 9743, 27405, 10161, 66
https://tests.stockfishchess.org/tests/view/663e779cbb28828150dd9089
closes https://github.com/official-stockfish/Stockfish/pull/5235
Bench: 1876282
Adds size in memory as well as layer sizes as in
info string NNUE evaluation using nn-ae6a388e4a1a.nnue (132MiB, (22528, 3072, 15, 32, 1))
info string NNUE evaluation using nn-baff1ede1f90.nnue (6MiB, (22528, 128, 15, 32, 1))
For example, the size in MiB is useful to keep the fishtest memory sizes up-to-date,
the L1-L3 sizes give a useful hint about the architecture used.
closes https://github.com/official-stockfish/Stockfish/pull/5193
No functional change
For each thread persist an accumulator cache for the network, where each
cache contains multiple entries for each of the possible king squares.
When the accumulator needs to be refreshed, the cached entry is used to more
efficiently update the accumulator, instead of rebuilding it from scratch.
This idea, was first described by Luecx (author of Koivisto) and
is commonly referred to as "Finny Tables".
When the accumulator needs to be refreshed, instead of filling it with
biases and adding every piece from scratch, we...
1. Take the `AccumulatorRefreshEntry` associated with the new king bucket
2. Calculate the features to activate and deactivate (from differences
between bitboards in the entry and bitboards of the actual position)
3. Apply the updates on the refresh entry
4. Copy the content of the refresh entry accumulator to the accumulator
we were refreshing
5. Copy the bitboards from the position to the refresh entry, to match
the newly updated accumulator
Results at STC:
https://tests.stockfishchess.org/tests/view/662301573fe04ce4cefc1386
(first version)
https://tests.stockfishchess.org/tests/view/6627fa063fe04ce4cefc6560
(final)
Non-Regression between first and final:
https://tests.stockfishchess.org/tests/view/662801e33fe04ce4cefc660a
STC SMP:
https://tests.stockfishchess.org/tests/view/662808133fe04ce4cefc667c
closes https://github.com/official-stockfish/Stockfish/pull/5183
No functional change
This is another refactor which aims to decouple uci from stockfish. A new engine
class manages all engine related logic and uci is a "small" wrapper around it.
In the future we should also try to remove the need for the Position object in
the uci and replace the options with an actual options struct instead of using a
map. Also convert the std::string's in the Info structs a string_view.
closes#5147
No functional change
This PR proposes to change the parameter dependence of Stockfish's
internal WDL model from full move counter to material count. In addition
it ensures that an evaluation of 100 centipawns always corresponds to a
50% win probability at fishtest LTC, whereas for master this holds only
at move number 32. See also
https://github.com/official-stockfish/Stockfish/pull/4920 and the
discussion therein.
The new model was fitted based on about 340M positions extracted from
5.6M fishtest LTC games from the last three weeks, involving SF versions
from e67cc979fd (SF 16.1) to current
master.
The involved commands are for
[WDL_model](https://github.com/official-stockfish/WDL_model) are:
```
./updateWDL.sh --firstrev e67cc979fd
python scoreWDL.py updateWDL.json --plot save --pgnName update_material.png --momType "material" --momTarget 58 --materialMin 10 --modelFitting optimizeProbability
```
The anchor `58` for the material count value was chosen to be as close
as possible to the observed average material count of fishtest LTC games
at move 32 (`43`), while not changing the value of
`NormalizeToPawnValue` compared to the move-based WDL model by more than
1.
The patch only affects the displayed cp and wdl values.
closes https://github.com/official-stockfish/Stockfish/pull/5121
No functional change
Reported by @Torom over discord.
> dev build fails on Raspberry Pi 5 with clang
```
clang++ -o stockfish benchmark.o bitboard.o evaluate.o main.o misc.o movegen.o movepick.o position.o search.o thread.o timeman.o tt.o uci.o ucioption.o tune.o tbprobe.o nnue_misc.o half_ka_v2_hm.o network.o -fprofile-instr-generate -latomic -lpthread -Wall -Wcast-qual -fno-exceptions -std=c++17 -fprofile-instr-generate -pedantic -Wextra -Wshadow -Wmissing-prototypes -Wconditional-uninitialized -DUSE_PTHREADS -DNDEBUG -O3 -funroll-loops -DIS_64BIT -DUSE_POPCNT -DUSE_NEON=8 -march=armv8.2-a+dotprod -DUSE_NEON_DOTPROD -DGIT_SHA=627974c9 -DGIT_DATE=20240312 -DARCH=armv8-dotprod -flto=full
/tmp/lto-llvm-e9300e.o: in function `_GLOBAL__sub_I_network.cpp':
ld-temp.o:(.text.startup+0x704c): relocation truncated to fit: R_AARCH64_LDST64_ABS_LO12_NC against symbol `gEmbeddedNNUEBigEnd' defined in .rodata section in /tmp/lto-llvm-e9300e.o
/usr/bin/ld: ld-temp.o:(.text.startup+0x704c): warning: one possible cause of this error is that the symbol is being referenced in the indicated code as if it had a larger alignment than was declared where it was defined
ld-temp.o:(.text.startup+0x7068): relocation truncated to fit: R_AARCH64_LDST64_ABS_LO12_NC against symbol `gEmbeddedNNUESmallEnd' defined in .rodata section in /tmp/lto-llvm-e9300e.o
/usr/bin/ld: ld-temp.o:(.text.startup+0x7068): warning: one possible cause of this error is that the symbol is being referenced in the indicated code as if it had a larger alignment than was declared where it was defined
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: *** [Makefile:1051: stockfish] Error 1
make[2]: Leaving directory '/home/torsten/chess/Stockfish_master/src'
make[1]: *** [Makefile:1058: clang-profile-make] Error 2
make[1]: Leaving directory '/home/torsten/chess/Stockfish_master/src'
make: *** [Makefile:886: profile-build] Error 2
```
closes https://github.com/official-stockfish/Stockfish/pull/5106
No functional change
- fix naming convention for `workingDirectory`
- use type alias for `EvalFiles` everywhere
- move `ponderMode` into `LimitsType`
- move limits parsing into standalone static function
closes https://github.com/official-stockfish/Stockfish/pull/5098
No functional change
This aims to remove some of the annoying global structure which Stockfish has.
Overall there is no major elo regression to be expected.
Non regression SMP STC (paused, early version):
https://tests.stockfishchess.org/tests/view/65983d7979aa8af82b9608f1
LLR: 0.23 (-2.94,2.94) <-1.75,0.25>
Total: 76232 W: 19035 L: 19096 D: 38101
Ptnml(0-2): 92, 8735, 20515, 8690, 84
Non regression STC (early version):
https://tests.stockfishchess.org/tests/view/6595b3a479aa8af82b95da7f
LLR: 2.93 (-2.94,2.94) <-1.75,0.25>
Total: 185344 W: 47027 L: 46972 D: 91345
Ptnml(0-2): 571, 21285, 48943, 21264, 609
Non regression SMP STC:
https://tests.stockfishchess.org/tests/view/65a0715c79aa8af82b96b7e4
LLR: 2.94 (-2.94,2.94) <-1.75,0.25>
Total: 142936 W: 35761 L: 35662 D: 71513
Ptnml(0-2): 209, 16400, 38135, 16531, 193
These global structures/variables add hidden dependencies and allow data
to be mutable from where it shouldn't it be (i.e. options). They also
prevent Stockfish from internal selfplay, which would be a nice thing to
be able to do, i.e. instantiate two Stockfish instances and let them
play against each other. It will also allow us to make Stockfish a
library, which can be easier used on other platforms.
For consistency with the old search code, `thisThread` has been kept,
even though it is not strictly necessary anymore. This the first major
refactor of this kind (in recent time), and future changes are required,
to achieve the previously described goals. This includes cleaning up the
dependencies, transforming the network to be self contained and coming
up with a plan to deal with proper tablebase memory management (see
comments for more information on this).
The removal of these global structures has been discussed in parts with
Vondele and Sopel.
closes https://github.com/official-stockfish/Stockfish/pull/4968
No functional change
Created by training an L1-128 net from scratch with a wider range of
evals in the training data and wld-fen-skipping disabled during
training. The differences in this training data compared to the first
dual nnue PR are:
- removal of all positions with 3 pieces
- when piece count >= 16, keep positions with simple eval above 750
- when piece count < 16, remove positions with simple eval above 3000
The asymmetric data filtering was meant to flatten the training data
piece count distribution, which was previously heavily skewed towards
positions with low piece counts.
Additionally, the simple eval range where the smallnet is used was
widened to cover more positions previously evaluated by the big net and
simple eval.
```yaml
experiment-name: 128--S1-hse-S7-v4-S3-v1-no-wld-skip
training-dataset:
- /data/hse/S3/leela96-filt-v2.min.high-simple-eval-1k.binpack
- /data/hse/S3/dfrc99-16tb7p-eval-filt-v2.min.high-simple-eval-1k.binpack
- /data/hse/S3/test80-apr2022-16tb7p.min.high-simple-eval-1k.binpack
- /data/hse/S7/test60-2020-2tb7p.v6-3072.high-simple-eval-v4.binpack
- /data/hse/S7/test60-novdec2021-12tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-v4.binpack
- /data/hse/S7/test77-nov2021-2tb7p.v6-3072.min.high-simple-eval-v4.binpack
- /data/hse/S7/test77-dec2021-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-v4.binpack
- /data/hse/S7/test77-jan2022-2tb7p.high-simple-eval-v4.binpack
- /data/hse/S7/test78-jantomay2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-v4.binpack
- /data/hse/S7/test78-juntosep2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-v4.binpack
- /data/hse/S7/test79-apr2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-v4.binpack
- /data/hse/S7/test79-may2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-v4.binpack
- /data/hse/S7/test80-may2022-16tb7p.high-simple-eval-v4.binpack
- /data/hse/S7/test80-jun2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-v4.binpack
- /data/hse/S7/test80-jul2022-16tb7p.v6-dd.min.high-simple-eval-v4.binpack
- /data/hse/S7/test80-aug2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-v4.binpack
- /data/hse/S7/test80-sep2022-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-v4.binpack
- /data/hse/S7/test80-oct2022-16tb7p.v6-dd.high-simple-eval-v4.binpack
- /data/hse/S7/test80-nov2022-16tb7p-v6-dd.min.high-simple-eval-v4.binpack
- /data/hse/S7/test80-jan2023-3of3-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-v4.binpack
- /data/hse/S7/test80-feb2023-16tb7p-filter-v6-dd.min-mar2023.unmin.high-simple-eval-v4.binpack
- /data/hse/S7/test80-mar2023-2tb7p.v6-sk16.min.high-simple-eval-v4.binpack
- /data/hse/S7/test80-apr2023-2tb7p-filter-v6-sk16.min.high-simple-eval-v4.binpack
- /data/hse/S7/test80-may2023-2tb7p.v6.min.high-simple-eval-v4.binpack
- /data/hse/S7/test80-jun2023-2tb7p.v6-3072.min.high-simple-eval-v4.binpack
- /data/hse/S7/test80-jul2023-2tb7p.v6-3072.min.high-simple-eval-v4.binpack
- /data/hse/S7/test80-aug2023-2tb7p.v6.min.high-simple-eval-v4.binpack
- /data/hse/S7/test80-sep2023-2tb7p.high-simple-eval-v4.binpack
- /data/hse/S7/test80-oct2023-2tb7p.high-simple-eval-v4.binpack
wld-fen-skipping: False
start-from-engine-test-net: False
nnue-pytorch-branch: linrock/nnue-pytorch/L1-128
engine-test-branch: linrock/Stockfish/L1-128-nolazy
engine-base-branch: linrock/Stockfish/L1-128
num-epochs: 500
start-lambda: 1.0
end-lambda: 1.0
```
Experiment yaml configs converted to easy_train.sh commands with:
https://github.com/linrock/nnue-tools/blob/4339954/yaml_easy_train.py
Binpacks interleaved at training time with:
https://github.com/official-stockfish/nnue-pytorch/pull/259
FT weights permuted with 10k positions from fishpack32.binpack with:
https://github.com/official-stockfish/nnue-pytorch/pull/254
Data filtered for high simple eval positions (v4) with:
https://github.com/linrock/Stockfish/blob/b9c8440/src/tools/transform.cpp#L640-L675
Training data can be found at:
https://robotmoon.com/nnue-training-data/
Local elo at 25k nodes per move of
L1-128 smallnet (nnue-only eval) vs. L1-128 trained on standard S1 data:
nn-epoch319.nnue : -241.7 +/- 3.2
Passed STC vs. 36db936:
https://tests.stockfishchess.org/tests/view/6576b3484d789acf40aabbfe
LLR: 2.94 (-2.94,2.94) <0.00,2.00>
Total: 21920 W: 5680 L: 5381 D: 10859
Ptnml(0-2): 82, 2488, 5520, 2789, 81
Passed LTC vs. DualNNUE #4915:
https://tests.stockfishchess.org/tests/view/65775c034d789acf40aac7e3
LLR: 2.95 (-2.94,2.94) <0.50,2.50>
Total: 147606 W: 36619 L: 36063 D: 74924
Ptnml(0-2): 98, 16591, 39891, 17103, 120
closes https://github.com/official-stockfish/Stockfish/pull/4919
Bench: 1438336
Remove a redundant int cast in the calculation of fwdOut. The variable
OutputType is already defined as std::int32_t, which is an integer type, making
the cast unnecessary.
closes https://github.com/official-stockfish/Stockfish/pull/4961
No functional change
The primary rationale behind this lies in the fact that enums were not
originally designed to be employed in the manner we currently utilize them.
The Value enum was used like a type alias throughout the code and was often
misused. Furthermore, changing the underlying size of the enum to int16_t broke
everything, mostly because of the operator overloads for the Value enum, were
causing data to be truncated. Since Value is now a type alias, the operator
overloads are no longer required.
Passed Non-Regression STC:
https://tests.stockfishchess.org/tests/view/6593b8bb79aa8af82b95b401
LLR: 2.95 (-2.94,2.94) <-1.75,0.25>
Total: 235296 W: 59919 L: 59917 D: 115460
Ptnml(0-2): 743, 27085, 62054, 26959, 807
closes https://github.com/official-stockfish/Stockfish/pull/4960
No functional change
- remove the blank line between the declaration of the function and it's
comment, leads to better IDE support when hovering over a function to see it's
description
- remove the unnecessary duplication of the function name in the functions
description
- slightly refactored code for lsb, msb in bitboard.h There are still a few
things we can be improved later on, move the description of a function where
it was declared (instead of implemented) and add descriptions to functions
which are behind macros ifdefs
closes https://github.com/official-stockfish/Stockfish/pull/4840
No functional change
This introduces clang-format to enforce a consistent code style for Stockfish.
Having a documented and consistent style across the code will make contributing easier
for new developers, and will make larger changes to the codebase easier to make.
To facilitate formatting, this PR includes a Makefile target (`make format`) to format the code,
this requires clang-format (version 17 currently) to be installed locally.
Installing clang-format is straightforward on most OS and distros
(e.g. with https://apt.llvm.org/, brew install clang-format, etc), as this is part of quite commonly
used suite of tools and compilers (llvm / clang).
Additionally, a CI action is present that will verify if the code requires formatting,
and comment on the PR as needed. Initially, correct formatting is not required, it will be
done by maintainers as part of the merge or in later commits, but obviously this is encouraged.
fixes https://github.com/official-stockfish/Stockfish/issues/3608
closes https://github.com/official-stockfish/Stockfish/pull/4790
Co-Authored-By: Joost VandeVondele <Joost.VandeVondele@gmail.com>
