Re: kern/12404 (panic: ffs_alloccg: map corrupted)

[Havard Eidnes <he%NetBSD.org@localhost>]

The following reply was made to PR kern/12404; it has been noted by GNATS.

From: Havard Eidnes <he%NetBSD.org@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: kern/12404 (panic: ffs_alloccg: map corrupted)
Date: Fri, 18 Jul 2008 22:59:34 +0200 (CEST)

 Another:
 
 start = 1, len = 23751, fs = /u
 offset=6024 6024
 cg 170
 panic: ffs_alloccg: map corrupted
 fatal breakpoint trap in supervisor mode
 trap type 1 code 0 eip c054d23c cs 8 eflags 246 cr2 bb908000 ilevel 0
 Stopped in pid 27342.1 (nbasn1_compile) at   netbsd:breakpoint+0x4:  popl  %ebp
 db{1}> tra
 breakpoint(c0a67f89,cd97e728,c316e800,c049c585,0,5,0,0,cd97e72c,0) at 
netbsd:breakpoint+0x4
 panic(c0a15eab,aa,1788,c33610d4,cd98c004,f673500,0,d1480000,c3361000,1) at 
netbsd:panic+0x1b8
 ffs_alloccgblk(1ece688,0,2,8000,ffffffff,1,cd97e7e8,0,0,e688) at 
netbsd:ffs_alloccgblk
 ffs_alloccg(d58b6b48,aa,1ece688,0,1000,0,aa,d58b6b48,c3361000,c3361000) at 
netbsd:ffs_alloccg+0x278
 ffs_hashalloc(1ece688,0,1000,c03c49a0,0,0,b700,0,dd38463,0) at 
netbsd:ffs_hashalloc+0x3a
 ffs_alloc(d58b6b48,0,0,1ece688,0,1000,cde45900,cd97e9d4,d786fad8,cd97e98c) at 
netbsd:ffs_alloc+0x22f
 ffs_balloc(d786fad8,0,0,690,cde45900,0,0,c0494bb1,cc402480,1) at 
netbsd:ffs_balloc+0x11e0
 ufs_gop_alloc(d786fad8,0,0,690,0,0,cde45900,0,1c02,0) at 
netbsd:ufs_gop_alloc+0xbe
 ufs_balloc_range(d786fad8,0,0,690,0,cde45900,0,c040817d,0,0) at 
netbsd:ufs_balloc_range+0x26a
 
ffs_write(cd97ec04,0,c081bbc0,d786fad8,2,20002,cd97ec1c,c04e57c8,c081b6c0,d786fad8)
 at netbsd:ffs_write+0x8f1
 VOP_WRITE(d786fad8,cd97ec7c,10,cde45900,ffffffff,ffffffff,0,16,690,bb939000) 
at netbsd:VOP_WRITE+0x6c
 
vn_write(d21bf740,d21bf740,cd97ec7c,cde45900,1,e04dc400,1,ce03fe1c,bb908000,1000)
 at netbsd:vn_write+0xb1
 dofilewrite(6,d21bf740,bb939000,690,d21bf740,1,cd97ed28,0,0,df663380) at 
netbsd:dofilewrite+0x75
 
sys_write(df663380,cd97ed00,cd97ed28,bb908000,bb908000,ce03fe1c,2,6,bb939000,690)
 at netbsd:sys_write+0x6f
 syscall(cd97ed48,b3,ab,1f,1f,bb939000,bbbcf308,bfbfe228,bbbc3118,bbbcf308) at 
netbsd:syscall+0xab
 db{1}> show regi
 ds          0x10
 es          0x10
 fs          0x30
 gs          0x10
 edi         0x2
 esi         0xc0a15eab  copyright+0x3302b
 ebp         0xcd97e6dc
 ebx         0x100
 edx         0x8
 ecx         0
 eax         0x1
 eip         0xc054d23c  breakpoint+0x4
 cs          0x8
 eflags      0x246
 esp         0xcd97e6dc
 ss          0x10
 netbsd:breakpoint+0x4:  popl    %ebp
 db{1}> reboot 4
 
 In all honesty I should perhaps also mention that I've been
 getting some apparently unmotivated "internal compiler errors" (I
 use this host for more-or-less continual rebuilds of -current for
 a number of architectures), which disappear when I rebuild the
 tools and re-build with the same sources.
 
 I have left the machine running memtest86 overnight with no
 issues found, though.
 
 But still, it *may* be that it's not the file system code which
 is at fault here.
 
 Regards,
 
 - Havard