Friday Feb 08, 2008

Good Morning Build 81 - part 2

The responsible engineer for the portfs bug 6659309 sent me some new binaries and now our system is back running build 82 but with a new portfs module. Just to be 100% certain the bug is fixed I ran this D that show that we would have paniced without the patch:

: estale.eu FSS 18 $; pfexec /usr/sbin/dtrace -n 'fbt::port_pfp_setup:entry {>
pfexec /usr/sbin/dtrace -n 'fbt::port_pfp_setup:entry {
	self->pfs = 1;
}
fbt::port_alloc_event_local:return / self->pfs == 1 && arg1 == 0 /  {
	self->pfs = 2;
}
fbt::port_pfp_setup:return /arg1 != 0 && self->pfs == 2/ {
	printf("port_pfp_setup failed %d: We would have crashed!", arg1);
}
fbt::port_pfp_setup:return /self->pfs / {
	self->pfs = 0
}'
dtrace: description 'fbt::port_pfp_setup:entry ' matched 4 probes
CPU     ID                    FUNCTION:NAME
  0  45054            port_pfp_setup:return port_pfp_setup failed 22: We would have crashed!
  0  45054            port_pfp_setup:return port_pfp_setup failed 22: We would have crashed!

So that pretty much tells us the bug is fixed. Well that and the fact that the system is not crashing!


However metacity is. I wonder if the two are related. Time to investigate this core dump:

: estale.eu FSS 11 $; pstack  core-estale-metacity.9140
core 'core-estale-metacity.9140' of 9140:       metacity --sm-save-file 1187603434-7917-1684782535.ms
 c40cab10 _lwp_kill (6, 0, 5, 6, ffffffff, 6) + 8
 c4054b44 abort    (1, 1, 6, c4153940, fbc5c, 0) + 108
 c3e4a550 g_logv   (ba58c, 6, 5, c3ee5404, 4, c3ee3404) + 484
 c3e4a57c g_log    (ba58c, 4, ba598, ba5c4, 1a4, ba7d0) + 1c
 00073f5c meta_window_new_with_attrs (19fa88, 1180001, ba400, ffbfe56c, fb7188, 0) + 3d8
 00073b5c meta_window_new (19fa88, 1180001, 0, daddcafe, 12, fe8) + 78
 00034780 event_callback (ffbfea28, 19fa88, 0, 33400, 14, 1180001) + f74
 00071d64 filter_func (0, ba000, 243068, 1212d0, 3380c, 243068) + 54
 c3f4d308 gdk_event_apply_filters (71d10, 11f1630, 0, 243048, 317c98, ffbfea28) + 24
 c3f4def8 gdk_event_translate (1af020, ffbfea28, ffbfea28, 0, 11f1630, 1a4e3c0) + 44
 c3f4fbc4 _gdk_events_queue (1af020, 0, 5, 0, 1b2228, ffbfea28) + ac
 c3f4fda8 gdk_event_dispatch (197a50, 0, 0, c3f9c8c0, 4cb54, 1af020) + 40
 c3e3ffc4 g_main_dispatch (171f00, 0, 0, 0, c3ee3404, 171f08) + 1e4
 c3e4166c g_main_context_dispatch (1, 1, c3ee53fc, 2, c3ee5400, 171f00) + c8
 c3e41b8c g_main_context_iterate (1, 1, 1, 171f00, 27, 28) + 49c
 c3e42490 g_main_loop_run (113dc0, 0, 0, 141fc8, 113dc8, 1) + 3e4
 000495f4 main     (b2000, b1c00, b1c00, 1, 48c00, 49400) + 4c0
 0002b160 _start   (0, 0, 0, 0, 0, 0) + 108
: estale.eu FSS 12 $; 
: estale.eu FSS 13 $; mdb core-estale-metacity.9140
Loading modules: [ libumem.so.1 libc.so.1 libuutil.so.1 ld.so.1 ]
> ba58c/s
0xba58c:        metacity
> ba598/s
0xba598:        file %s: line %d: assertion failed: (%s)
> ba5c4/s
0xba5c4:        window.c
> ba7d0/s
0xba7d0:        window->screen
> 
About

This is the old blog of Chris Gerhard. It has mostly moved to http://chrisgerhard.wordpress.com

Search

Archives
« April 2014
MonTueWedThuFriSatSun
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
    
       
Today