On Sun, 12 Dec 1999, Takahashi Yoshihiro wrote:
> The 'loader' program is only transplanted for pc98. The boot2
> (sys/boot/pc98/boot2) is the mostly same as the old biosboot
> (sys/pc98/boot/biosboot) .
>
> > boot2 is already capable of loading ELF, right?
>
> No.
I use a modified version of
In article <[EMAIL PROTECTED]>
Marcel Moolenaar <[EMAIL PROTECTED]> writes:
> Since pc98 only has a specialized boot2 and thus shares boot0 and boot1
> with other architectures, it can be assumed that an ELF boot2 doesn't
> need any special hacking in boot0 and/or boot1, right?
No, if HDD is for
Takahashi Yoshihiro wrote:
>
> > BTW: Is there a special reason to have boot2 in aout?
>
> Because nobody transplant from the i386 boot2 :-).
> FreeBSD(98) porting team is always suffering from a shortage of
> workers.
>
> In article <[EMAIL PROTECTED]>
> Mike Smith <[EMAIL PROTECTED]> writes:
In article <[EMAIL PROTECTED]>
Marcel Moolenaar <[EMAIL PROTECTED]> writes:
> Other than that: Go ahead!
Thank you for many comments.
> BTW: Is there a special reason to have boot2 in aout?
Because nobody transplant from the i386 boot2 :-).
FreeBSD(98) porting team is always suffering from a s
> The boot2 for pc98 (sys/boot/pc98/boot2) is a.out program. So, several
> tools for a.out are needed to make boot2.
>
> I make the patch to fix this. Is there any problem?
Only in the longer term, I think. How hard would it be to port the 'new'
i386 boot2?
--
\\ Give a man a fish, and you
Takahashi Yoshihiro wrote:
> +.if ${MACHINE} == "pc98" && ${OBJFORMAT} == "elf"
> +USRDIRS+= usr/libexec/aout
> +.endif
I don't think you need to test for OBJFORMAT. Also, include MACHINE_ARCH
in case of cross-building. eg:
.if ${MACHINE_ARCH} == "i386" && ${MACHINE} == "pc98"
> +.if ${MA
The boot2 for pc98 (sys/boot/pc98/boot2) is a.out program. So, several
tools for a.out are needed to make boot2.
I make the patch to fix this. Is there any problem?
--- Makefile.inc1 1999/12/10 16:13:41 1.102
+++ Makefile.inc1 1999/12/11 10:01:21
@@ -143,6 +143,10 @@
USRDIRS=
