An updated fuse module is available for test on :
https://jp-andre.pagesperso-orange.fr/fuse-kernel-1.4AR.1.pkg.zip
This is an old format package (building current format is well
over my capacity). After unzipping, you can install it by
pkgadd -d fuse-kernel-1.4AR.1.pkg all
(you should uninstall the current version first, but this would
lead to uninstalling other packages...).
Please note : I just recompiled it (no source change), but I
cannot test it with the kernel I am using (from april 2020),
so be sure to have a way back.
Jean-Pierre
Jean-Pierre André wrote on 9/15/20 12:17 PM:
Actually the change appears to be in "struct thread".
Old code disassembled :
:000063F2 E800000000 call 0000063F7H (threadp)
:000063F7 488B8090010000 mov rax,qword 0190H[rax]
:000063FE 4883BD10FFFFFF00 cmp qword -240[rbp],0
:00006406 488B80B0000000 mov rax,qword 176[rax]
:0000640D 8B4004 mov eax,dword 4[rax]
New code recompiled :
969 000a97 e800000000 call near threadp
970 000a9c 488b8098010000 mov rax,qword ptr 408[rax]
971 000aa3 488b80b0000000 mov rax,qword ptr 176[rax]
972 000aaa 488b5dc0 mov rbx,qword ptr -64[rbp]
973 000aae 8b4004 mov eax,dword ptr 4[rax]
The field t_procp was at offset 400, it is now at 408.
I will make a new binary module for testing (unless a recompiled
package is available in the meantime).
Jean-Pierre
Jean-Pierre André wrote on 9/15/20 9:52 AM:
The bad dereferencing occurs while getting the current process id :
fhp->process_id = curproc->p_pidp->pid_id;
With curproc->p_pidp found to be null (based on fuse 1.4-2020.0.1.0)
My first bet is that the system "struct proc" has been changed without
the fuse module being recompiled (or you did not install the recompiled
one).
Jean-Pierre
Stephan Althaus wrote on 9/14/20 9:54 PM:
Hello all!
I am using the fuse module "exfat" for about half a year now and it
works well.
When i updated OI on 2020-09-10, the system panics when mounting an
exfat fs.
What recent change could have lead to this behaviour?
I did set "set hires_tick=0" in /etc/system as a shot in the dark,
it does not help.
When mounting an exfat file system, i get a system panic and an
instant automatic reboot.
Any hints are welcome how to identify the root of the fault.
Thanks,
Stephan
*$ fmdump -Vp -u fb5fcce4-0a78-65e6-be82-f41d72c5d705*
TIME UUID SUNW-MSG-ID
Sep 12 2020 12:04:00.880479000 fb5fcce4-0a78-65e6-be82-f41d72c5d705
SUNOS-8000-KL
TIME CLASS ENA
Sep 12 12:03:50.3209 ireport.os.sunos.panic.dump_pending_on_device
0x0000000000000000
nvlist version: 0
version = 0x0
class = list.suspect
uuid = fb5fcce4-0a78-65e6-be82-f41d72c5d705
code = SUNOS-8000-KL
diag-time = 1599905040 827325
de = fmd:///module/software-diagnosis
fault-list-sz = 0x1
fault-list = (array of embedded nvlists)
(start fault-list[0])
nvlist version: 0
version = 0x0
class = defect.sunos.kernel.panic
certainty = 0x64
asru =
sw:///:path=/var/crash/dell/.fb5fcce4-0a78-65e6-be82-f41d72c5d705
resource =
sw:///:path=/var/crash/dell/.fb5fcce4-0a78-65e6-be82-f41d72c5d705
savecore-succcess = 0
os-instance-uuid = fb5fcce4-0a78-65e6-be82-f41d72c5d705
* panicstr = BAD TRAP: type=e (#pf Page fault)
rp=fffffe003fc20850 addr=4 occurred in module "fuse" due to a NULL
pointer dereference**
** panicstack = unix:die+c6 () | unix:trap+1169 () |
unix:cmntrap+e9 () | fuse:get_filehandle+52d () | fuse:fuse_open+6e
() | genunix:fop_open+96 () | genunix:vn_openat+203 () |
genunix:copen+4a9 () | genunix:openat+29 () |
unix:brand_sys_syscall+1fe () | **
* crashtime = 1599904948
panic-time = 12 September 2020 at 12:02:28 CEST CEST
(end fault-list[0])
fault-status = 0x1
severity = Major
__ttl = 0x1
__tod = 0x5f5c9d10 0x347b0b18
_______________________________________________
openindiana-discuss mailing list
[email protected]
https://openindiana.org/mailman/listinfo/openindiana-discuss
