Bugs and complaints on current master

[Nick]

Another crash from verbose mode object drop code:
[New Thread 8384.0x26a0]

Program received signal SIGSEGV, Segmentation fault.
object_is_ignored (obj=0x31fa494) at obj-ignore.c:584
584 if (obj->known->notice & OBJ_NOTICE_IGNORE)
(gdb) where
#0 object_is_ignored (obj=0x31fa494) at obj-ignore.c:584
#1 0x00459b3a in drop_near (c=0x32514a4, dropped=dropped@entry=0xb0f05c,
chance=chance@entry=0, y=57, x=139, verbose=verbose@entry=true)
at obj-pile.c:957
#2 0x0043c832 in monster_death (mon=mon@entry=0x7aafda8,

It is (again) dereferencing **dropped when that is already deleted/merged.

Hm. Looks like *dropped was probably OK but (*dropped)->known was probably already freed.

Do you know what the monster was - a mimic maybe?

[Pete Mack]

Another crash from verbose mode object drop code:
[New Thread 8384.0x26a0]

Program received signal SIGSEGV, Segmentation fault.
object_is_ignored (obj=0x31fa494) at obj-ignore.c:584
584 if (obj->known->notice & OBJ_NOTICE_IGNORE)
(gdb) where
#0 object_is_ignored (obj=0x31fa494) at obj-ignore.c:584
#1 0x00459b3a in drop_near (c=0x32514a4, dropped=dropped@entry=0xb0f05c,
chance=chance@entry=0, y=57, x=139, verbose=verbose@entry=true)
at obj-pile.c:957
#2 0x0043c832 in monster_death (mon=mon@entry=0x7aafda8,

It is (again) dereferencing **dropped when that is already deleted/merged.

[Nick]

I'm not sure what happened here....I was playing this character (yesterday) on one mac, moved the save file from the angband save file to my dropbox to play while I traveled, but opening the save file crashes angband.

Any thoughts?

Thanks, that was a considerable (if easily fixed) bug in saving and loading of traps, which would have occurred any time you saved and reloaded twice on the same level.

Attached is a fixed savefile - it is not quite as you saw it last, every trap on the current level is a pit...

I have pushed a build which fixes this problem, it should be up on the nightlies page before too long.

[Pete Mack]

Getting crashes on loading saved game--fails when scanning loaded trap data, which appears hopelessly corrupt.

[StMicah]

I'm not sure what happened here....I was playing this character (yesterday) on one mac, moved the save file from the angband save file to my dropbox to play while I traveled, but opening the save file crashes angband.

Any thoughts?

[Nick]

Look, it's not Nowhere Town.

I'm not done yet

[Derakon]

OK, thanks. I fail at town again

Look, it's not Nowhere Town. You're getting better at this.

[Nick]

I suspect this is not supposed to happen! Just started a new game and can't make it to town.

OK, thanks. I fail at town again

[StMicah]

I suspect this is not supposed to happen! Just started a new game and can't make it to town.

[kandrc]

And when I try to recover from that one, my @ is in a different location from where it was when I crashed, and I get the following:

Program received signal SIGABRT, Aborted.
0x00007ffff6d7f979 in raise () from /lib64/libc.so.6
(gdb) bt
#0  0x00007ffff6d7f979 in raise () from /lib64/libc.so.6
#1  0x00007ffff6d81088 in abort () from /lib64/libc.so.6
#2  0x00007ffff6d78966 in __assert_fail_base () from /lib64/libc.so.6
#3  0x00007ffff6d78a12 in __assert_fail () from /lib64/libc.so.6
#4  0x0000000000407791 in square_istrap (c=0x894fa8, y=19, x=38) at cave-square.c:464
#5  0x000000000049f735 in square_trap_flag (c=0x894fa8, y=19, x=38, flag=3) at trap.c:90
#6  0x00000000004084d5 in square_isvisibletrap (c=0x894fa8, y=19, x=38) at cave-square.c:705
#7  0x00000000004084fd in square_issecrettrap (c=0x894fa8, y=19, x=38) at cave-square.c:712
#8  0x000000000048e31b in search (p=0x9bb9e8) at player-util.c:943
#9  0x0000000000487106 in notice_stuff (p=0x9bb9e8) at player-calcs.c:2278
#10 0x000000000041c094 in on_new_level () at game-world.c:716
#11 0x00000000004ad45a in start_game (new_game=false) at ui-game.c:417
#12 0x00000000004ad475 in play_game (new_game=false) at ui-game.c:426
#13 0x00000000004e583d in main (argc=1, argv=0x7fffffffdc48) at main.c:524
(gdb)

EDIT:
Looking at it more closely, I'm being placed on the permanent wall in a small (labyrinth) level. Going to try to edit my position in the debugger to save the game. Stairs don't work. Loaded into a bad place. If I can get him moved, I'll try to recall and see if that salvages things.

Rogue @ just found an early ESP on a *Slay Animal*. I'd rather not abandon.

EDIT 2:
Ah, this did the trick:

Breakpoint 1, start_game (new_game=false) at ui-game.c:415
(gdb) print character_dungeon
$1 = true
(gdb) call character_dungeon = 0
$2 = false
(gdb) cont
Continuing.

[kandrc]

Here's another segfault in 4.0.3-603-g81b3e23:

Program received signal SIGSEGV, Segmentation fault.
0x00000000004dd857 in flag_on_dbg (flags=0x411 <Address 0x411 out of bounds>, size=3, flag=1, fi=0x4eb64d "c->squares[y][x].info", fl=0x4eb641 "SQUARE_MARK") at z-bitflag.c:226
226             if (flags[flag_offset] & flag_binary) return false;
(gdb) bt
#0  0x00000000004dd857 in flag_on_dbg (flags=0x411 <Address 0x411 out of bounds>, size=3, flag=1, fi=0x4eb64d "c->squares[y][x].info", fl=0x4eb641 "SQUARE_MARK")
    at z-bitflag.c:226
#1  0x00000000004099ad in square_mark (c=0xb19578, y=0, x=55) at cave-square.c:1098
#2  0x0000000000405568 in wiz_light (c=0xbb7a18, full=false) at cave-map.c:446
#3  0x0000000000420cd0 in labyrinth_gen (p=0x9bbc28) at gen-cave.c:887
#4  0x000000000041e2ed in cave_generate (c=0x74b480 <cave>, p=0x9bbc28) at generate.c:864
#5  0x000000000041c476 in run_game_loop () at game-world.c:840
#6  0x00000000004ad48f in play_game (new_game=false) at ui-game.c:433
#7  0x00000000004e583d in main (argc=1, argv=0x7fffffffdc48) at main.c:524
(gdb)

[Nick]

I hit an infinite loop in scatter(). Specifically, it's never passing the test for square_in_bounds_fully(), as ny is being generated in the range [-2, 0].

Here's some detail. I don't know this code well enough to see what the issue is exactly, but I suspect the parameters coming into scatter() are bad.

Thanks, that's really helpful.

[kandrc]

I hit an infinite loop in scatter(). Specifically, it's never passing the test for square_in_bounds_fully(), as ny is being generated in the range [-2, 0].

Here's some detail. I don't know this code well enough to see what the issue is exactly, but I suspect the parameters coming into scatter() are bad.

(gdb) bt
#0  square_in_bounds_fully (c=0xd9e558, y=-1, x=74) at cave-square.c:764
#1  0x00000000004044db in scatter (c=0xd9e558, yp=0x7fffffffd760, xp=0x7fffffffd75c, y=-1, x=74, d=1, need_los=false) at cave.c:297
#2  0x00000000004301a1 in vault_monsters (c=0xd9e558, y1=-1, x1=74, depth=18, num=1) at gen-util.c:731
#3  0x0000000000429b33 in build_room_template (c=0xd9e558, y0=4, x0=74, ymax=7, xmax=28, doors=1, 
    data=0x9da588 "%%%####%%#%%####%%#%%####%%%%...##...+...##..9+...##...%%##....#####....#####....##%  ##..##   ##..##   ##..##  %##....#####....#####....##%%...##...+9..##...+...##...%%%%####%%#%%####%%#%%####%%%", tval=0) at gen-room.c:1009
#4  0x0000000000429df5 in build_room_template_type (c=0xd9e558, y0=56, x0=172, typ=1, rating=1) at gen-room.c:1074
#5  0x000000000042d7c4 in build_template (c=0xd9e558, y0=56, x0=172, rating=1) at gen-room.c:2384
#6  0x000000000042eef7 in room_build (c=0xd9e558, by0=0, bx0=0, profile=..., finds_own_space=true) at gen-room.c:2990
#7  0x0000000000422d08 in modified_chunk (depth=17, height=56, width=172) at gen-cave.c:1688
#8  0x000000000042327c in modified_gen (p=0x9bbbf8) at gen-cave.c:1791
#9  0x000000000041e2ed in cave_generate (c=0x74b480 <cave>, p=0x9bbbf8) at generate.c:864
#10 0x000000000041c476 in run_game_loop () at game-world.c:840
#11 0x00000000004ad48f in play_game (new_game=false) at ui-game.c:433
#12 0x00000000004e583d in main (argc=1, argv=0x7fffffffdc48) at main.c:524
(gdb) print *c
$24 = {
  name = 0x0, 
  created_at = 43089, 
  depth = 17, 
  feeling = 0 '\000', 
  obj_rating = 65, 
  mon_rating = 242, 
  good_item = false, 
  height = 56, 
  width = 172, 
  feeling_squares = 0, 
  feat_count = 0xc6e938, 
  squares = 0xc6ea48, 
  objects = 0xd64db8, 
  obj_max = 127, 
  monsters = 0xda3648, 
  mon_max = 4, 
  mon_cnt = 3, 
  mon_current = -1
}
(gdb) up
#1  0x00000000004044db in scatter (c=0xd9e558, yp=0x7fffffffd760, xp=0x7fffffffd75c, y=-1, x=74, d=1, need_los=false) at cave.c:297
297                     if (!square_in_bounds_fully(c, ny, nx)) continue;
(gdb) print y
$25 = -1
(gdb) print x
$26 = 74
(gdb) print d
$27 = 1
(gdb) down
#0  square_in_bounds_fully (c=0xd9e558, y=-1, x=74) at cave-square.c:764
764             assert(c);
(gdb) list
759             return x >= 0 && x < c->width && y >= 0 && y < c->height;
760     }
761
762     bool square_in_bounds_fully(struct chunk *c, int y, int x)
763     {
764             assert(c);
765             return x > 0 && x < c->width - 1 && y > 0 && y < c->height - 1;
766     }
767
768
(gdb) print *yp
No symbol "yp" in current context.
(gdb) up
#1  0x00000000004044db in scatter (c=0xd9e558, yp=0x7fffffffd760, xp=0x7fffffffd75c, y=-1, x=74, d=1, need_los=false) at cave.c:297
297                     if (!square_in_bounds_fully(c, ny, nx)) continue;
(gdb) print *yp
$28 = 77
(gdb) print *xp
$29 = 2
(gdb)

[kandrc]

Not sure is this one should be considered a bug or not, nor am I certain of historical behavior, but staff of sleep monsters wakes sleeping monsters that make the saving throw.

[Derakon]

Always available? Unless this has been changed when I wasn't watching, the only always-available scrolls are phase door and word of recall. If mapping scrolls are indeed always available (or we're willing to make them so) then I agree that this is an adequate solution. Otherwise, we probably do need to bring stair detection back.

They shouldn't be always available; magic mapping is a powerful ability and the fact that holy casters are the only ones with unlimited access to it is a powerful defining feature of the class. I'd say Sense Surroundings and Clairvoyance are their most distinctive spells after Healing.