linuxconsole-dev

Starting point for ruby-2.6.7

[Zoltan B. <zb...@fr...>]

Hi Aivil,

if you revert the patches found in the attached tarball
in alphabetical order in Linux-2.6.7 and apply the ruby-2.6.6
patch then only fs/compat_ioctl.c gives three trivial rejects.
The attached compat_ioctl.c is the fixed version.

compat_ioctl.c is not needed on x86, the others compile
without warnings on i686 system.

For others: you can get create the ruby-2.6.6 patch by

diff -ur --unidirectional-new-file /path/to/linux-2.6.6=20
/path/to/linuxconsole/cvs/tree/ruby/ruby-2.6 >ruby-2.6.6.patch

Best regards,
Zolt=E1n B=F6sz=F6rm=E9nyi

Re: Starting point for ruby-2.6.7

[Zoltan B. <zb...@fr...>]

Zoltan Boszormenyi =EDrta:
> ... compile
> without warnings on i686 system.

And works, too. :-) Now compiling it on x86-64...

Re: Starting point for ruby-2.6.7

[Zoltan B. <zb...@fr...>]

Zoltan Boszormenyi =EDrta:
> Zoltan Boszormenyi =EDrta:
>=20
>> ... compile
>> without warnings on i686 system.
>=20
>=20
> And works, too. :-) Now compiling it on x86-64...

And works, compiled both with gcc-3.3.3 and gcc-3.4.0.
Your dumdum module wasn't needed, something was fixed
in the x86-64 port of linux-2.6.7.

Best regards,
Zolt=E1n B=F6sz=F6rm=E9nyi

Re: Starting point for ruby-2.6.7

[Aivils <ai...@un...>]

ruby-2.6.7 on CVS! On public CVS after N hours.
VGA - fixed
fb cursor now blink on each fb console.

On Wednesday 16 June 2004 23:03, Zoltan Boszormenyi wrote:
> if you revert the patches found in the attached tarball
> in alphabetical order in Linux-2.6.7 and apply the ruby-2.6.6

Seems You use another tree of kernel.
Your pathes is not against CVS

> patch then only fs/compat_ioctl.c gives three trivial rejects.
> The attached compat_ioctl.c is the fixed version.

i just use James' s  initial version.

Your semiautomatic updates breaks fbcon functionality.

You can grab ruby from CVS and check out radeonfb.
dual login via text mode console on radeonfb.
You can do it under 64bit arch :)

Aivils

Re: Starting point for ruby-2.6.7

[Zoltan B. <zb...@fr...>]

Aivils =EDrta:
> ruby-2.6.7 on CVS! On public CVS after N hours.
> VGA - fixed
> fb cursor now blink on each fb console.
>=20
> On Wednesday 16 June 2004 23:03, Zoltan Boszormenyi wrote:
>=20
>>if you revert the patches found in the attached tarball
>>in alphabetical order in Linux-2.6.7 and apply the ruby-2.6.6
>=20
>=20
> Seems You use another tree of kernel.
> Your pathes is not against CVS

You meant the ruby CVS? Of course "my patches" weren't against it.
They were downloaded from linus.bkbits.net, they went into mainline
after linux-2.6.6, and I used them with "patch -p1 -R" against vanilla
linux-2.6.7 downloaded from
http://www.kernel.org/pub/linux/kernel/v2.6/linux-2.6.7.tar.bz2.
So what I produced was an almost linux-2.6.7, console behaviour
setting back to 2.6.6 and patched with ruby-2.6.6. Is it clear now?

>>patch then only fs/compat_ioctl.c gives three trivial rejects.
>>The attached compat_ioctl.c is the fixed version.
>=20
> i just use James' s  initial version.

The difference against yours and mine is attached.
What does the removed lines do?

> Your semiautomatic updates breaks fbcon functionality.
>=20
> You can grab ruby from CVS and check out radeonfb.
> dual login via text mode console on radeonfb.
> You can do it under 64bit arch :)

I will try it. :-)

Best regards,
Zolt=E1n B=F6sz=F6rm=E9nyi