#153 vmhgfs kernel module panics on FreeBSD

[Mikhail T.]

The problem reported in bug 3584833 suggests, the module has not been tested on FreeBSD in a while... Indeed, after fixing the build, and trying to use the module I get the following kernel panic (this is FreeBSD/x86 8.3, 32-bit):

Fatal trap 12: page fault while in kernel mode
cpuid = 1; apic id = 01
fault virtual address = 0x4
...
#6 0xc07c0d2c in calltrap () at /mi/src/sys/i386/i386/exception.s:168
#7 0xc83add06 in DblLnkLst_Link (l1=0xc6ff3644, l2=0xc7560000)
at dbllnklst.h:153
#8 0xc83adcdb in DblLnkLst_LinkLast (head=0xc6ff3644, l=0xc7560000)
at dbllnklst.h:286
#9 0xc83adcab in HgfsKReq_AllocateRequest (container=0xc6ff3640,
errorRet=0xe9461994) at request.c:418
#10 0xc83b2da6 in HgfsStatfsInt (vp=0xc831ca78, stat=0xc727ecf4)
at vfsopscommon.c:77
#11 0xc83aa44b in HgfsVfsMount (mp=0xc727ec94) at vfsops.c:258
#12 0xc0670732 in vfs_donmount (td=0xc78825c0, fsflags=0, fsoptions=0xc728bd00)
at /mi/src/sys/kern/vfs_mount.c:1011
#13 0xc0671011 in nmount (td=0xc78825c0, uap=0xe9461cec)
at /mi/src/sys/kern/vfs_mount.c:447
...

Examining frame #7 reveals:

(kgdb) p *l1
$1 = {prev = 0x0, next = 0xc7560000}
(kgdb) p *l2
$2 = {prev = 0x0, next = 0x0}

And then:
(kgdb) up
#8 0xc83adcdb in DblLnkLst_LinkLast (head=0xc6ff3644, l=0xc7560000)
at dbllnklst.h:286
286 in dbllnklst.h
(kgdb) p *head
$3 = {prev = 0x0, next = 0xc7560000}
(kgdb) p *l
$4 = {prev = 0x0, next = 0x0}

(It is possible, the problem reported in bug 3580633 is related.)

I have the core-dump with debug symbols, so, if anybody cares for more variables, etc. let me know.

[Mikhail T.]

I should add, that I experienced this problem with 9.2.0-799703.

[Mikhail T.]

FWIW, I just tried the 9.2.2-893683 and got the exact same panic.

[Douglas Carmichael]

I've experienced the same problem on FreeBSD 9.1-RELEASE with 9.2.3-1031360:

Fatal trap 12: page fault while in kernel mode
cpuid = 1; apic id = 02
fault virtual address = 0x2a0
fault code = supervisor read data, page not present
instruction pointer = 0x20:0xffffffff80e29487
stack pointer = 0x28:0xffffff82332e06f0
frame pointer = 0x28:0xffffff82332e0770
code segment = base 0x0, limit 0xfffff, type 0x1b
= DPL 0, pres 1, long 1, def32 0, gran 1
processor eflags = interrupt enabled, resume, IOPL = 0
current process = 934 (mount)
trap number = 12
panic: page fault
cpuid = 1

[Harry Weppner]

Hi,

a page fault still occurs when attempting to mount -t vmhgfs ... using version 9.4.0.25793 (build-1280544) and FreeBSD 10.0-BETA4 FreeBSD 10.0-BETA4 #0 f166130(stable/10), cf. attached screenshot.

This ticket has been open for a while now... Who's the owner of this issue?

Thanks & cheerio, Harry.

[Steve]

Hi all,

Sorry for the bad news from VMware on this but these vmhgfs bugs are not going to be addressed. There have not been any resources allocated to this feature in FreeBSD for quite some time. Furthermore, in subsequent updates this feature will not be distributed for this platform.

Steve