On Sat, Dec 31, 2011 at 03:20:10AM +0000, Peter Wemm wrote: > The following reply was made to PR amd64/163710; it has been noted by GNATS. > > From: Peter Wemm <[email protected]> > To: Russell Cattelan <[email protected]> > Cc: [email protected] > Subject: Re: amd64/163710: setjump in userboot.so causes stack corruption > Date: Fri, 30 Dec 2011 19:16:10 -0800 > > On Fri, Dec 30, 2011 at 12:24 PM, Russell Cattelan <[email protected]> w= > rote: > > On 12/30/11 1:33 AM, Peter Wemm wrote: > >> On Thu, Dec 29, 2011 at 7:16 PM, Russell Cattelan > >> <[email protected]> wrote: > >> > >>>> Description: > >>> For some reason the forth interpreter is built and linked as 32bit even > >>> on amd64. > >> That's the catch. We use the same 32 bit loader on i386 and amd64. > >> The common loader understands both kernel formats. =A0This unfortunately > >> has meant that the libstand and sys/boot environment has had to be 32 > >> bit. > >> > > Yes this is bit of an odd situation. > > the loader is linked 32bit but userboot.so / libstand is built 64bit > > but pulls in the 32bit ficl lib. > > The 64bit libstand provides the setjmp call that needs 96 bytes > > of space but since ficl is built with the 32bit jump_buf (48 byte ) > > structure > > things go sideways. > > We link the loader with the 32 bit libstand though. It cannot work > any other way. BTX is 32 bit only. userboot.so is 64bit, and, as I already pointed the originator, it uses the separate ficl build, sys/boot/userboot/ficl.
pgpo8KjLnP2uZ.pgp
Description: PGP signature
