I have been trying to understand what was wrong with crafty running on my dad's machine. I have been able to isolate a bit the issue and get a reproduction scenario for the devs to take over, since the ebuild itself seems to be full of extra flags that I know nothing about. On my home system (amd64 but should be irrelevant), I have done: * unpack stage3-amd64-20130130.tar.bz2, mount /proc and /dev and chroot into stage3 * emerge --sync (March 24th-25th) * emerge crafty * start crafty, works as expected. Example: # /usr/games/bin/crafty unable to open book file [./book.bin]. book is disabled unable to open book file [./books.bin]. Crafty v23.4 (1 cpus) White(1): d4 time used: 1.66 time surplus 0.00 time limit 30.00 (+0.00) (3:00) depth time score variation (1) 12 0.10 0.11 1. ... e6 2. Nf3 Nf6 3. Nc3 Nc6 4. e4 Bb4 5. d5 Bxc3+ 6. bxc3 exd5 7. exd5 12-> 0.14 0.11 1. ... e6 2. Nf3 Nf6 3. Nc3 Nc6 4. e4 Bb4 5. d5 Bxc3+ 6. bxc3 exd5 7. exd5 13 0.26 -0.08 1. ... e6 2. Nf3 Nf6 3. Nbd2 Nc6 4. e4 Be7 5. Bd3 d5 6. O-O dxe4 7. Nxe4 Nxd4 13-> 0.29 -0.08 1. ... e6 2. Nf3 Nf6 3. Nbd2 Nc6 4. e4 Be7 5. Bd3 d5 6. O-O dxe4 7. Nxe4 Nxd4 14 0.61 0.17 1. ... e6 2. Nf3 Nf6 3. Nc3 Nc6 4. e3 Bd6 5. Bd3 O-O 6. O-O Nb4 7. Nb5 Nxd3 8. Nxd6 Nxc1 9. Nxc8 14-> 0.85 0.17 1. ... e6 2. Nf3 Nf6 3. Nc3 Nc6 4. e3 Bd6 5. Bd3 O-O 6. O-O Nb4 7. Nb5 Nxd3 8. Nxd6 Nxc1 9. Nxc8 * unpack stage3-i686-20121213.tar.bz2, mount /proc and /dev and chroot into stage3 * emerge --sync (March 24th-25th) * emerge crafty * start crafty, enter some moves (d4, a4, a3, ...). Most of them will be considered invalid and after some attempts (seems to depend on the machine), crafty will detect an inconsistency and exit. Example: # /usr/games/bin/crafty unable to open book file [./book.bin]. book is disabled unable to open book file [./books.bin]. Crafty v23.4 (1 cpus) White(1): d4 Illegal move: d4 White(1): a4 Illegal move: a4 White(1): a3 Illegal move: a3 White(1): a2 Illegal move: a2 White(1): a4 Illegal move: a4 White(1): d4 Illegal move: d4 White(1): d3 Illegal move: d3 White(1): d2 Illegal move: d2 White(1): a5 Illegal move: a5 White(1): a7 time used: 20.83 ERROR! board[17]=0, should be -1 ERROR! board[18]=0, should be -1 ERROR! board[19]=0, should be -1 ERROR! board[20]=0, should be -1 ERROR! board[21]=0, should be -1 ERROR! board[22]=0, should be -1 ERROR! board[23]=0, should be -1 ERROR! board[25]=0, should be -2 ERROR! board[30]=0, should be -2 ERROR! board[26]=0, should be -3 ERROR! board[29]=0, should be -3 ERROR! board[24]=0, should be -4 ERROR! board[31]=0, should be -4 ERROR! board[27]=0, should be -5 ERROR! board[28]=0, should be -6 ERROR! board[16]=0, should be 1 ERROR! bitboards/board[9] don't agree! ERROR! board[9]=1, should be 0 ERROR! hash_key is bad. ERROR! pawn_hash_key is bad. processor id: cpu-0 current move: move= piece=1, from=41, to=48, captured=1, promote=0 +---+---+---+---+---+---+---+---+ 8 |<R>|<N>|<B>|<Q>|<K>|<B>|<N>|<R>| +---+---+---+---+---+---+---+---+ 7 |-P-|<P>|<P>|<P>|<P>|<P>|<P>|<P>| +---+---+---+---+---+---+---+---+ 6 | | . | | . | | . | | . | +---+---+---+---+---+---+---+---+ 5 | . | | . | | . | | . | | +---+---+---+---+---+---+---+---+ 4 | | . | | . | | . | | . | +---+---+---+---+---+---+---+---+ 3 | . | | . | | . | | . | | +---+---+---+---+---+---+---+---+ 2 |-P-|-P-|-P-|-P-|-P-|-P-|-P-|-P-| +---+---+---+---+---+---+---+---+ 1 |-R-|-N-|-B-|-Q-|-K-|-B-|-N-|-R-| +---+---+---+---+---+---+---+---+ a b c d e f g h called from Main(1), ply=0 node=0 active path: Reproducible: Always
I can't reproduce it here, so (reasonable) patches welcome.
Someone on #gentoo-games that I asked could reproduce it. For my information, how did you do your attempt?
I can't reproduce it because I don't have any x86 boxes, not because it works fine for me on x86.
(In reply to comment #3) > I can't reproduce it because I don't have any x86 boxes, not because it > works fine for me on x86. My reproduction (of my dad's x86 problem) uses an x86 chroot as described above. Do you see any issue with this?
It should be fixed. Thanks for report.
(In reply to comment #5) > It should be fixed. Thanks for report. I can confirm that it is fixed. Great job!!