Menu
▾
▴
linuxconsole-dev
|
From: Helge H. <hel...@ai...> - 2004-12-11 14:30:01
|
I've used ruby for i386 successfully. But I've got an opteron now, so I want to compile a 64-bit kernel. I followed debian's guide for setting up cross-compiling, but the compile itself ended like this: In file included from arch/x86_64/ia32/ia32_ioctl.c:19: fs/compat_ioctl.c: In function `do_fontx_ioctl': fs/compat_ioctl.c:1616: error: `cfdarg' undeclared (first use in this function) fs/compat_ioctl.c:1616: error: (Each undeclared identifier is reported only once fs/compat_ioctl.c:1616: error: for each function it appears in.) In file included from arch/x86_64/ia32/ia32_ioctl.c:19: fs/compat_ioctl.c: In function `do_kdfontop_ioctl': fs/compat_ioctl.c:1655: warning: unused variable `vt' make[1]: *** [arch/x86_64/ia32/ia32_ioctl.o] Error 1 make: *** [arch/x86_64/ia32] Error 2 I need to have 32-bit emulation so I can switch to a 64-bit kernel first and change over userspace gradually as the next step. Any ideas? Or is 2.6.9 "too old" now? Helge Hafting |
|
From: Zoltan B. <zb...@fr...> - 2004-12-11 15:48:18
|
Helge Hafting =EDrta:
> I've used ruby for i386 successfully. But I've got an opteron
> now, so I want to compile a 64-bit kernel. I followed debian's
> guide for setting up cross-compiling, but the compile itself
> ended like this:
>=20
> In file included from arch/x86_64/ia32/ia32_ioctl.c:19:
> fs/compat_ioctl.c: In function `do_fontx_ioctl':
> fs/compat_ioctl.c:1616: error: `cfdarg' undeclared (first use in this f=
unction)
> fs/compat_ioctl.c:1616: error: (Each undeclared identifier is reported =
only=20
> once
> fs/compat_ioctl.c:1616: error: for each function it appears in.)
> In file included from arch/x86_64/ia32/ia32_ioctl.c:19:
> fs/compat_ioctl.c: In function `do_kdfontop_ioctl':
> fs/compat_ioctl.c:1655: warning: unused variable `vt'
> make[1]: *** [arch/x86_64/ia32/ia32_ioctl.o] Error 1
> make: *** [arch/x86_64/ia32] Error 2
>=20
>=20
> I need to have 32-bit emulation so I can switch to a 64-bit
> kernel first and change over userspace gradually as the next step.
>=20
> Any ideas? =20
> Or is 2.6.9 "too old" now?
>=20
> Helge Hafting
There's a chunk in ruby-2.6.9-20041021, that's wrong.
It causes the compile error in line 1616. Here's the correct chunk.
It's the line after 'case GIO_FONTX:'.
------------------------------------------------------------------
diff -Nurp --exclude-from=3Dexclude linux-2.6.9/fs/compat_ioctl.c=20
r269/fs/compat_ioctl.c
--- linux-2.6.9/fs/compat_ioctl.c 2004-10-19 00:55:27.000000000 +03=
00
+++ r269/fs/compat_ioctl.c 2004-10-20 19:31:58.000000000 +0300
@@ -1609,8 +1611,10 @@ static int do_fontx_ioctl(unsigned int f
get_user(data, &user_cfd->chardata))
return -EFAULT;
op.data =3D compat_ptr(data);
- return con_font_op(fg_console, &op);
+ return con_font_op(vc, &op);
case GIO_FONTX:
+ if (!user_cfd->chardata)
+ return 0;
op.op =3D KD_FONT_OP_GET;
op.flags =3D 0;
op.width =3D 8;
------------------------------------------------------------------
Best regards,
Zolt=E1n B=F6sz=F6rm=E9nyi
|
|
From: Helge H. <hel...@ai...> - 2004-12-12 21:35:05
|
On Sat, Dec 11, 2004 at 04:49:53PM +0100, Zoltan Boszormenyi wrote: > There's a chunk in ruby-2.6.9-20041021, that's wrong. > It causes the compile error in line 1616. Here's the correct chunk. > It's the line after 'case GIO_FONTX:'. > Thanks a lot, this compiled fine! :-) Helge Hafting |
|
From: Zoltan B. <zb...@fr...> - 2004-12-12 22:59:19
|
Helge Hafting =EDrta: > On Sat, Dec 11, 2004 at 04:49:53PM +0100, Zoltan Boszormenyi wrote: >=20 >>There's a chunk in ruby-2.6.9-20041021, that's wrong. >>It causes the compile error in line 1616. Here's the correct chunk. >>It's the line after 'case GIO_FONTX:'. >> >=20 > Thanks a lot, this compiled fine! :-) >=20 > Helge Hafting And I hope it works on your machines, at least it works on mine. ;-) Best regards, Zolt=E1n B=F6sz=F6rm=E9nyi |
|
From: Helge H. <hel...@ai...> - 2004-12-14 21:39:53
|
On Mon, Dec 13, 2004 at 12:00:57AM +0100, Zoltan Boszormenyi wrote: > Helge Hafting írta: > >On Sat, Dec 11, 2004 at 04:49:53PM +0100, Zoltan Boszormenyi wrote: > > > >>There's a chunk in ruby-2.6.9-20041021, that's wrong. > >>It causes the compile error in line 1616. Here's the correct chunk. > >>It's the line after 'case GIO_FONTX:'. > >> > > > >Thanks a lot, this compiled fine! :-) > > > >Helge Hafting > > And I hope it works on your machines, at least it works on mine. ;-) Thanks for all the help so far. The kernel works in the sense that it boots. Unfortunately, 2.6.9 for amd64 has exactly the same problem as 2.6.9 for i386, it hangs the machine when the xserver tries to utilize the radeon 9200SE PCI card. So I tried compiling 2.6.8.1, as that one works for i386 for me. I got this: CC arch/x86_64/kernel/setup.o arch/x86_64/kernel/setup.c: In function `setup_arch': arch/x86_64/kernel/setup.c:582: error: `conswitchp' undeclared (first use in this function) arch/x86_64/kernel/setup.c:582: error: (Each undeclared identifier is reported only once arch/x86_64/kernel/setup.c:582: error: for each function it appears in.) arch/x86_64/kernel/setup.c:582: error: `vga_con' undeclared (first use in this function) make[1]: *** [arch/x86_64/kernel/setup.o] Error 1 make: *** [arch/x86_64/kernel] Error 2 Is there a trick to get this going, or will I have to wait for 2.6.10 and the next ruby release? Helge Hafting |
|
From: Aivils <ai...@un...> - 2004-12-15 07:15:12
|
On Tuesday 14 December 2004 23:47, Helge Hafting wrote:
> On Mon, Dec 13, 2004 at 12:00:57AM +0100, Zoltan Boszormenyi wrote:
> > Helge Hafting =EDrta:
> > >On Sat, Dec 11, 2004 at 04:49:53PM +0100, Zoltan Boszormenyi wrote:
> > >
> > >>There's a chunk in ruby-2.6.9-20041021, that's wrong.
> > >>It causes the compile error in line 1616. Here's the correct chunk.
> > >>It's the line after 'case GIO_FONTX:'.
> > >>
> > >
> > >Thanks a lot, this compiled fine! :-)
> > >
> > >Helge Hafting
> >=20
> > And I hope it works on your machines, at least it works on mine. ;-)
>=20
> Thanks for all the help so far.
> The kernel works in the sense that it boots.
> Unfortunately, 2.6.9 for amd64 has exactly the same problem as
> 2.6.9 for i386, it hangs the machine when the xserver tries to
> utilize the radeon 9200SE PCI card.
>=20
> So I tried compiling 2.6.8.1, as that one works for i386 for me.
> I got this:
>=20
> CC arch/x86_64/kernel/setup.o
> arch/x86_64/kernel/setup.c: In function `setup_arch':
> arch/x86_64/kernel/setup.c:582: error: `conswitchp' undeclared (first use=
in=20
> this function)
> arch/x86_64/kernel/setup.c:582: error: (Each undeclared identifier is rep=
orted=20
> only once
> arch/x86_64/kernel/setup.c:582: error: for each function it appears in.)
> arch/x86_64/kernel/setup.c:582: error: `vga_con' undeclared (first use in=
this=20
> function)
> make[1]: *** [arch/x86_64/kernel/setup.o] Error 1
> make: *** [arch/x86_64/kernel] Error 2
>=20
> Is there a trick to get this going, or will I have to wait
> for 2.6.10 and the next ruby release?
CVS-ed 2.6.9 arch/x86_64/kernel/setup.c and
ruby-2.6.9-20041021.diff.bz2
does not contain these lines :)
You should manualy delete some lines
from arch/x86_64/kernel/setup.c , if You will use versions below 2.6.9=20
diff -Nurp --exclude-from=3Dexclude linux-2.6.9/arch/x86_64/kernel/setup.c =
r269/arch/x86_64/kernel/setup.c
=2D-- linux-2.6.9/arch/x86_64/kernel/setup.c 2004-10-19 00:53:07.00000=
0000 +0300
+++ r269/arch/x86_64/kernel/setup.c 2004-10-21 21:04:27.000000000 +0300
@@ -596,14 +596,6 @@ void __init setup_arch(char **cmdline_p)
#ifdef CONFIG_GART_IOMMU
iommu_hole_init();
#endif
=2D
=2D#ifdef CONFIG_VT
=2D#if defined(CONFIG_VGA_CONSOLE)
=2D conswitchp =3D &vga_con;
=2D#elif defined(CONFIG_DUMMY_CONSOLE)
=2D conswitchp =3D &dummy_con;
=2D#endif
=2D#endif
}
static int __init get_model_name(struct cpuinfo_x86 *c)
Aivils
|
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X