[Freeciv-Dev] [bug #16592] Freeciv 2.2.2 crash for i don't know reason, maybe for stealth fighter issue.

2010-09-04 Thread pepeto

Update of bug #16592 (project freeciv):

  Status:  Ready For Test = Fixed  
 Open/Closed:Open = Closed 


___

Reply to this item at:

  http://gna.org/bugs/?16592

___
  Message posté via/par Gna!
  http://gna.org/


___
Freeciv-dev mailing list
Freeciv-dev@gna.org
https://mail.gna.org/listinfo/freeciv-dev


[Freeciv-Dev] [bug #16592] Freeciv 2.2.2 crash for i don't know reason, maybe for stealth fighter issue.

2010-09-02 Thread pepeto

Update of bug #16592 (project freeciv):

 Assigned to:None = pepeto 
 Planned Release: = 2.2.3, 2.3.0   

___

Follow-up Comment #8:

Attached a very different patch for trunk. The client in trunk would assume
that tile fogging only affect V_MAIN units, which is very true, especially if
custom ruleset defines units, cities or bases with bigger invisible vision
range than main one.


(file #10135)
___

Additional Item Attachment:

File name: trunk_V_INVIS_and_fog.diff Size:2 KB


___

Reply to this item at:

  http://gna.org/bugs/?16592

___
  Message posté via/par Gna!
  http://gna.org/


___
Freeciv-dev mailing list
Freeciv-dev@gna.org
https://mail.gna.org/listinfo/freeciv-dev


[Freeciv-Dev] [bug #16592] Freeciv 2.2.2 crash for i don't know reason, maybe for stealth fighter issue.

2010-08-30 Thread pepeto

Update of bug #16592 (project freeciv):

  Status:None = Confirmed  

___

Follow-up Comment #1:

Backtrace:

(gdb) bt
#0  0x729b5a75 in *__GI_raise (sig=value optimised out)
at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
pid = value optimised out
selftid = value optimised out
#1  0x729b95c0 in *__GI_abort () at abort.c:92
act = {__sigaction_handler = {sa_handler = 0x5346d8, 
sa_sigaction = 0x5346d8}, sa_mask = {__val = {140737264790063, 
  140737488345808, 2543, 140737488346048, 140737263866086, 
  206158430232, 140737488346064, 140737488345840,
140737263773432, 
  206158430256, 140737488346088, 22824880, 1926656, 30419216, 
  8390250522101773151, 140737488348691}}, sa_flags = -223573286,

  sa_restorer = 0x533aa0}
sigs = {__val = {32, 0 repeats 15 times}}
#2  0x729ae941 in *__GI___assert_fail (
assertion=0x5346d8 unit_list_size(ptile-units) == 0, 
file=value optimised out, line=2543, 
function=0x534ef0 handle_tile_info) at assert.c:81
buf = 0x15c47b0 freeciv-gtk2:
packhand.c :2543 : handle_tile_info:  L'assertion
« unit_list_size(ptile-units) == 0 » a échoué.\n
#3  0x0043c0e8 in handle_tile_info (packet=0x1ca47d0)
at packhand.c:2543
new_known = TILE_KNOWN_UNSEEN
old_known = TILE_KNOWN_SEEN
tile_changed = false
powner = 0x0
presource = 0x1b37a40
pterrain = 0x1
ptile = 0x1e1a4b0
pnation = 0x0
__PRETTY_FUNCTION__ = handle_tile_info
#4  0x0044038f in client_handle_packet (type=value optimised out, 
packet=0x0) at packhand_gen.c:62
No locals.
#5  0x00417c1d in client_packet_input (packet=value optimised out,

type=14) at client_main.c:529
No locals.
#6  0x0041d149 in input_from_server (fd=value optimised out)
at clinet.c:383
result = true
packet = 0x0
type = PACKET_TILE_INFO
__PRETTY_FUNCTION__ = input_from_server
#7  0x74f27d3f in gdk_io_invoke (source=value optimised out, 
condition=value optimised out, data=value optimised out)
at /build/buildd/gtk+2.0-2.20.1/gdk/gdkevents.c:1082
closure = 0x13f48e0
gdk_cond = GDK_INPUT_READ
#8  0x72d438c2 in g_main_dispatch (context=0x91bcf0)
at /build/buildd/glib2.0-2.24.1/glib/gmain.c:1960
dispatch = 0x72d85e10 g_io_unix_dispatch
user_data = 0x13f48e0
callback = 0x74f27cf0 gdk_io_invoke
cb_funcs = 0x72fe1610
cb_data = 0x13fc530
current_source_link = {data = 0x13fc4b0, next = 0x0}
source = 0x13fc4b0
current = 0x13e0a20
i = 2
#9  IA__g_main_context_dispatch (context=0x91bcf0)
at /build/buildd/glib2.0-2.24.1/glib/gmain.c:2513
No locals.
#10 0x72d47748 in g_main_context_iterate (context=0x91bcf0, 
block=value optimised out, dispatch=value optimised out, 
self=value optimised out)
at /build/buildd/glib2.0-2.24.1/glib/gmain.c:2591
max_priority = 0
timeout = 0
some_ready = 1
nfds = 4
allocated_nfds = -220903856
fds = value optimised out
__PRETTY_FUNCTION__ = g_main_context_iterate
#11 0x72d47c55 in IA__g_main_loop_run (loop=0xfa3b00)
at /build/buildd/glib2.0-2.24.1/glib/gmain.c:2799
self = 0x9823a0
__PRETTY_FUNCTION__ = IA__g_main_loop_run
#12 0x752eabb7 in IA__gtk_main ()
at /build/buildd/gtk+2.0-2.20.1/gtk/gtkmain.c:1219
tmp_list = 0x927740
functions = 0x0
init = 0xa2d190
loop = value optimised out
#13 0x00417059 in ui_main (argc=1, argv=0x7fffe348)
at gui_main.c:1615
home = value optimised out
sig = value optimised out
#14 0x00417f9a in client_main (argc=value optimised out, 
argv=0x7fffe348) at client_main.c:481
i = 1
loglevel = 2
ui_options = value optimised out
ui_separator = 64
option = value optimised out
user_tileset = false
#15 0x729a0c4d in __libc_start_main (main=value optimised out, 
argc=value optimised out, ubp_av=value optimised out, 
init=value optimised out, fini=value optimised out, 
rtld_fini=value optimised out, stack_end=0x7fffe338)
at libc-start.c:226
result = value optimised out
unwind_buf = {cancel_jmp_buf = {{jmp_buf = {0, -727124314455238109, 
4271136, 140737488347968, 0, 0, 727124314305370659, 
727148261492169251}, mask_was_saved = 0}}, priv = {pad =
{0x0, 
  0x0, 0x52da40, 0x7fffe348}, data = {prev = 0x0, 
  cleanup = 0x0, canceltype = 5429824}}}
not_first_call = value optimised out
#16 0x00412c49 in _start ()
No symbol table info available.



[Freeciv-Dev] [bug #16592] Freeciv 2.2.2 crash for i don't know reason, maybe for stealth fighter issue.

2010-08-30 Thread pepeto

Update of bug #16592 (project freeciv):

Category:  client-gtk-2.0 = general

___

Follow-up Comment #2:

French removes shared vision at the turn change. It seems the units in the
V_INVIS vision layer are not properly removed for the client. Sounds like an
important server bug, but the client shouldn't assert there anyway.


___

Reply to this item at:

  http://gna.org/bugs/?16592

___
  Message posté via/par Gna!
  http://gna.org/


___
Freeciv-dev mailing list
Freeciv-dev@gna.org
https://mail.gna.org/listinfo/freeciv-dev


[Freeciv-Dev] [bug #16592] Freeciv 2.2.2 crash for i don't know reason, maybe for stealth fighter issue.

2010-08-30 Thread Marko Lindqvist

Follow-up Comment #3, bug #16592 (project freeciv):

That client side assert has revealed at least 5 server side bugs over the
years, so it is very useful if not elegant.

___

Reply to this item at:

  http://gna.org/bugs/?16592

___
  Message sent via/by Gna!
  http://gna.org/


___
Freeciv-dev mailing list
Freeciv-dev@gna.org
https://mail.gna.org/listinfo/freeciv-dev


[Freeciv-Dev] [bug #16592] Freeciv 2.2.2 crash for i don't know reason, maybe for stealth fighter issue.

2010-08-30 Thread pepeto

Follow-up Comment #4, bug #16592 (project freeciv):

I would suggest to assert there only if 2.2 is configured with --enable-debug
is enabled. Else, it will crash any normal client which connect an old server.
For 2.3, using fc_assert() seems good to prevent the client to crash in
classic configurations.


___

Reply to this item at:

  http://gna.org/bugs/?16592

___
  Message posté via/par Gna!
  http://gna.org/


___
Freeciv-dev mailing list
Freeciv-dev@gna.org
https://mail.gna.org/listinfo/freeciv-dev


[Freeciv-Dev] [bug #16592] Freeciv 2.2.2 crash for i don't know reason, maybe for stealth fighter issue.

2010-08-30 Thread pepeto

Follow-up Comment #5, bug #16592 (project freeciv):

Reason of the problem:
remove_shared_vision() uses vision_layer_iterate macro which iterates V_MAIN
then V_INVIS. When V_MAIN tile seen count is 0, the tile is fogged, whereas
it didn't handle the V_INVIS units yet.

It should iterate the vision layers in the reverse order like commented in
vision_clear_sight(). I will check if there are other way to reproduce
similar bugs.


___

Reply to this item at:

  http://gna.org/bugs/?16592

___
  Message posté via/par Gna!
  http://gna.org/


___
Freeciv-dev mailing list
Freeciv-dev@gna.org
https://mail.gna.org/listinfo/freeciv-dev


[Freeciv-Dev] [bug #16592] Freeciv 2.2.2 crash for i don't know reason, maybe for stealth fighter issue.

2010-08-30 Thread pepeto

Update of bug #16592 (project freeciv):

  Status:   Confirmed = Ready For Test 

___

Follow-up Comment #6:

Fix attached:
* Always remove invisible units before fogging the tile. It was possible to
happen a bit everywhere, notably with non-default rulesets. However, when the
invisible layer radius is bigger than the main layer, the result is still
buggy (impossible to fix with the current implementation).
* Added vision_layer_reverse_iterate macro.
* Removed assertion for non debug clients.

Testing would be very appreciated.


(file #10095)
___

Additional Item Attachment:

File name: S2_2_vision_layers.diffSize:18 KB


___

Reply to this item at:

  http://gna.org/bugs/?16592

___
  Message posté via/par Gna!
  http://gna.org/


___
Freeciv-dev mailing list
Freeciv-dev@gna.org
https://mail.gna.org/listinfo/freeciv-dev