Re: Path passed to ffs_mount() is garbage?

["Simon 'corecode' Schubert" <corecode@xxxxxxxxxxxx>]

Matthew Dillon wrote:
> :So I take my first foray into the dragonfly kernel and do this to
> :ffs_mount() in /usr/src/sys/vfs/ufs/ffs_vfsops.c: I merely add a
> :
> : kprintf("ffs_mount to %s\n",((path == NULL) ? "(NULL)" : path));
> :
> :as the very first line of code (past the declarations of course).
> :When I test this by starting up my vkernel, if path is non-null, it
> :points to garbage data. 
> :
> :I'm probably being naive here, what am I missing?
>     path is passed from userland.  Its a userland pointer.  It shouldn't
>     be garbage, but you probably need to do a copyin() into a kernel
>     buffer instead of printing it directly.  Actually, I'm a little at 
>     a loss because on the PC architecture the pointer should still be
>     printable (if a bit dangerous due to being a userland address).

yah, but not for vkernels...  the vmspace is different :/  makes it also a bitch to debug

cheers
 simon

--
Serve - BSD     +++  RENT this banner advert  +++    ASCII Ribbon   /"\
Work - Mac      +++  space for low €€€ NOW!1  +++      Campaign     \ /
Party Enjoy Relax   |   http://dragonflybsd.org      Against  HTML   \
Dude 2c 2 the max   !   http://golden-apple.biz       Mail + News   / \

Attachment: signature.asc
Description: OpenPGP digital signature