On Fri, 22 Aug 2003, Patrick Welche wrote:

> Somewhat of a cross port question.. I just plugged an adfs disk from an
> Acorn RiscPC into a 1.6R/i386 computer:
>
> # mount -t filecore /dev/wd1d /mnt
> # ls /mnt
>                      !Music F      f      ionsF  n      nsF
> # ls -lR /mnt | more
> ls: : No such file or directory
> ls: : No such file or directory
> ls: : No such file or directory
> ls: F: No such file or directory
> ls: f: No such file or directory
> ls: ionsF: No such file or directory
>
> uvm_fault(0xc02ab4c0, 0xc3e1b000, 0, 1) -> e
> kernel: page fault trap, code=0
> Stopped in pid 12486.1 (ls) at  filecore_map+0x246:     movl    0(%edi,%eax,4),%
> edx
> db> bt
> filecore_map(c0548600,620816,0,0,c6d260c4) at filecore_map+0x246
> filecore_dbread(c6d2607c,c6d48d64,1a,0,c6d48eb4) at filecore_dbread+0x34
> filecore_readdir(c6d48eb4,20001,c6d48f78,c01f97bc,0) at filecore_readdir+0xcb
> vn_readdir(c6ce496c,805a000,0,1000,c6d48f38) at vn_readdir+0xac
> sys_getdents(c6ac9e80,c6d48f80,c6d48f78,c022a91c,0) at sys_getdents+0x53
> syscall_plain(1f,1f,1f,1f,804e340) at syscall_plain+0xab
> db>
>
> Anyone interested?

	Curious... I assume the filesystem is OK under NetBSD/arm32.
	I wonder if it could be related to different default alignment
	of any of the structs?

	As a quick test you could sprinkle __attribute__((__packed__))
	after all the struct definitions in sys/fs/filecorefs/*.h
-- 
		David/absolute          -- www.netbsd.org: No hype required --