Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add AUTHORS #5

Merged
merged 2 commits into from
Jun 23, 2014
Merged

Add AUTHORS #5

merged 2 commits into from
Jun 23, 2014

Conversation

glinscott
Copy link
Contributor

Just a nice way to track contributors

@lucasart
Copy link

it's a bit annoying that people keep changing their names on git. there are lots of duplicates, where the same person appears under different names in the list. is it possible to make git display the login instead ?

anyway, this is a good idea. i'm sure many people have felt that their contribution was not recognized.

@glinscott
Copy link
Contributor Author

I removed some duplicates, maybe in the future we just ask folks what name they would like to be included under?

@zamar
Copy link

zamar commented Jun 23, 2014

I have no objections to this. Feel free to merge.

glinscott added a commit that referenced this pull request Jun 23, 2014
@glinscott
Merge pull request #5 from glinscott/authors
ab58010 
Add AUTHORS
@glinscott glinscott merged commit ab58010 into official-stockfish:master Jun 23, 2014
@glinscott glinscott deleted the authors branch June 23, 2014 16:14
niklasf pushed a commit to niklasf/Stockfish that referenced this pull request Jun 29, 2015
@ddugovic
Merge pull request official-stockfish#5 from official-stockfish/master
c06cbcc 
Retire Search Log
@LouisZulli LouisZulli mentioned this pull request Sep 5, 2015
Closed
@ajithcj ajithcj mentioned this pull request Sep 20, 2015
Closed
@vondele vondele mentioned this pull request Oct 16, 2015
Closed
@mcostalba mcostalba mentioned this pull request Dec 16, 2015
Closed
@ChameleonRed ChameleonRed mentioned this pull request Apr 3, 2016
Closed
vondele referenced this pull request in vondele/Stockfish Jul 26, 2017
@vondele
Avoid a race between uci thread and search threads.
9844c00 
Fixes a race when ucinewgame is issued 'after' a search, as seen for a thread-sanitized binary issuing :

```
go depth 5
[wait for bestmove ..]
ucinewgame
```
leading to

```
WARNING: ThreadSanitizer: data race (pid=5148)
  Write of size 4 at 0x7fbf5dcf6af8 by main thread:
    #0 __gnu_cxx::__enable_if<!std::__is_scalar<Move>::__value, void>::__type std::__fill_a<Move*, Move>(Move*, Move*, Move const&) <null> (stockfish+0x000000462884)
    #1 void std::fill<Move*, Move>(Move*, Move*, Move const&) /usr/include/c++/5/bits/stl_algobase.h:747 (stockfish+0x0000004618cc)
    #2 StatBoards<16, 64, Move>::fill(Move const&) src/movepick.h:36 (stockfish+0x00000046084f)
    #3 Search::clear() src/search.cpp:194 (stockfish+0x000000451732)
    #4 newgame src/uci.cpp:145 (stockfish+0x0000004738a0)
    #5 UCI::loop(int, char**) src/uci.cpp:200 (stockfish+0x000000473d7d)
    #6 main src/main.cpp:47 (stockfish+0x000000433322)

  Previous write of size 4 at 0x7fbf5dcf6af8 by thread T1:
    #0 update_stats src/search.cpp:1417 (stockfish+0x000000453ecb)
    #1 search<(<unnamed>::NodeType)0u> src/search.cpp:1114 (stockfish+0x00000045c4c6)
    #2 search<(<unnamed>::NodeType)1u> src/search.cpp:997 (stockfish+0x000000457658)
    #3 search<(<unnamed>::NodeType)1u> src/search.cpp:1019 (stockfish+0x000000457a50)
    #4 search<(<unnamed>::NodeType)1u> src/search.cpp:1019 (stockfish+0x000000457a50)
    #5 Thread::search() src/search.cpp:402 (stockfish+0x000000452e28)
    #6 MainThread::search() src/search.cpp:264 (stockfish+0x000000451c32)
    #7 Thread::idle_loop() src/thread.cpp:114 (stockfish+0x000000468c90)

```

The solution adopted is to split the uci commands in those that can be issued during search (quit, stop, ponderhit, isready, uci) and those that should not run during search (ucinewgame, setoption, .. ).  Waiting for search to finish was previously done only for a 'go'.

This fixes the above race.

There are two side effects, one possibly negative, depending on how a gui deals with uci (or a user at the cli).
- crashes are avoided (e.g. resizing hash during search).
- as earlier commands can block on the search to be completed, 'stop' can be non-effective. This behavior is already present on master :

```
go depth 20
go depth 5
stop
```

will wait for depth 20 to be completed before stopping depth 5. This behavior now shows for other commands as well. As such some gui testing of this patch would be useful.

No functional change.
vondele referenced this pull request in vondele/Stockfish Jul 27, 2017
@vondele
avoid a race between uci and search threads (II)
d974307 
limits.ponder was used as a signal between uci and search threads, but is not an atomic variable,
leading to the following race as flagged by a sanitized binary.

Expect input:
```
 spawn  ./stockfish
 send "uci\n"
 expect "uciok"
 send "setoption name Ponder value true\n"
 send "go wtime 4000 btime 4000\n"
 expect "bestmove"
 send "position startpos e2e4 d7d5\n"
 send "go wtime 4000 btime 4000 ponder\n"
 sleep 0.01
 send "ponderhit\n"
 expect "bestmove"
 send "quit\n"
 expect eof
```

Race:
```
WARNING: ThreadSanitizer: data race (pid=7191)
  Read of size 4 at 0x0000005c2260 by thread T1:
    #0 MainThread::check_time() src/search.cpp:1488 (stockfish+0x000000454471)
    #1 search<(<unnamed>::NodeType)0u> src/search.cpp:566 (stockfish+0x0000004594e0)
    #2 search<(<unnamed>::NodeType)0u> src/search.cpp:997 (stockfish+0x00000045bfb6)
    #3 search<(<unnamed>::NodeType)0u> src/search.cpp:1006 (stockfish+0x00000045c1a3)
    #4 search<(<unnamed>::NodeType)1u> src/search.cpp:997 (stockfish+0x000000457658)
    #5 Thread::search() src/search.cpp:402 (stockfish+0x000000452e28)
    #6 MainThread::search() src/search.cpp:264 (stockfish+0x000000451c32)
    #7 Thread::idle_loop() src/thread.cpp:114 (stockfish+0x000000468c90)

  Previous write of size 4 at 0x0000005c2260 by main thread:
    #0 UCI::loop(int, char**) src/uci.cpp:193 (stockfish+0x000000473c9b)
    #1 main src/main.cpp:47 (stockfish+0x000000433322)

  Location is global 'Search::Limits' of size 88 at 0x0000005c2220 (stockfish+0x0000005c2260)
```

The fix is to add an atomic bool to the threads structure to signal the ponder status, letting Search::Limits to reflect just what was passed to 'go'.

No functional change.
vondele referenced this pull request in vondele/Stockfish Aug 2, 2017
@vondele
avoid a race between uci and search threads (II)
c7cd9e6 
limits.ponder was used as a signal between uci and search threads, but is not an atomic variable,
leading to the following race as flagged by a sanitized binary.

Expect input:
```
 spawn  ./stockfish
 send "uci\n"
 expect "uciok"
 send "setoption name Ponder value true\n"
 send "go wtime 4000 btime 4000\n"
 expect "bestmove"
 send "position startpos e2e4 d7d5\n"
 send "go wtime 4000 btime 4000 ponder\n"
 sleep 0.01
 send "ponderhit\n"
 expect "bestmove"
 send "quit\n"
 expect eof
```

Race:
```
WARNING: ThreadSanitizer: data race (pid=7191)
  Read of size 4 at 0x0000005c2260 by thread T1:
    #0 MainThread::check_time() src/search.cpp:1488 (stockfish+0x000000454471)
    #1 search<(<unnamed>::NodeType)0u> src/search.cpp:566 (stockfish+0x0000004594e0)
    #2 search<(<unnamed>::NodeType)0u> src/search.cpp:997 (stockfish+0x00000045bfb6)
    #3 search<(<unnamed>::NodeType)0u> src/search.cpp:1006 (stockfish+0x00000045c1a3)
    #4 search<(<unnamed>::NodeType)1u> src/search.cpp:997 (stockfish+0x000000457658)
    #5 Thread::search() src/search.cpp:402 (stockfish+0x000000452e28)
    #6 MainThread::search() src/search.cpp:264 (stockfish+0x000000451c32)
    #7 Thread::idle_loop() src/thread.cpp:114 (stockfish+0x000000468c90)

  Previous write of size 4 at 0x0000005c2260 by main thread:
    #0 UCI::loop(int, char**) src/uci.cpp:193 (stockfish+0x000000473c9b)
    #1 main src/main.cpp:47 (stockfish+0x000000433322)

  Location is global 'Search::Limits' of size 88 at 0x0000005c2220 (stockfish+0x0000005c2260)
```

The fix is to add an atomic bool to the threads structure to signal the ponder status, letting Search::Limits to reflect just what was passed to 'go'.

No functional change.
protonspring pushed a commit to protonspring/Stockfish that referenced this pull request May 28, 2019
edge majority official-stockfish#5
a47a1e9
Lolligerhans pushed a commit to Lolligerhans/Stockfish that referenced this pull request May 30, 2019
@Lolligerhans
edge majority official-stockfish#5
bd030b1
Lolligerhans pushed a commit to Lolligerhans/Stockfish that referenced this pull request Jun 1, 2019
@Lolligerhans
edge majority official-stockfish#5
4c90419
Lolligerhans pushed a commit to Lolligerhans/Stockfish that referenced this pull request Jul 4, 2019
@Lolligerhans
edge majority official-stockfish#5
08423d1
@vondele vondele mentioned this pull request May 20, 2020
Closed
@gvreuls gvreuls mentioned this pull request Jun 3, 2021
Closed
@CavanLeoJacobs CavanLeoJacobs mentioned this pull request Feb 25, 2022
Closed
pb00067 referenced this pull request in pb00068/Stockfish Oct 24, 2023
@pb00068
Don't calculate X-ray attackers if not neccessary. Can this be faster?
02a5e3d 
Fixed logic.
bench: 1241996
locutus2 pushed a commit to locutus2/Stockfish that referenced this pull request Dec 31, 2023
@Vizvezdenec
Merge pull request official-stockfish#5 from Vizvezdenec/revert-4-gua…
23ac0e4 
…rdCorrHist2

Revert "Guard correction history from overflowing staticEval in the future."
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@glinscott @lucasart @zamar