linuxconsole-dev

Patches to revert in 2.6.7-rc1 or -rc2

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

Hi,

I searched through http://linus.bkbits.net:8080/linux-2.5
for the patches that are needed to be reverted so ruby-2.6.6
patch apply cleanly. These patches should be applied with
"patch -R -p1" in alphabetical order.

I want to try newer kernels/fixes but I also need
the multihead capability.

I also needed two additional patches to compile
this beast on x86_64, one for fs/compat_ioctl.c.
Aivil, there is a missing semicolon and a new line
there, look into line 1722 in your CVS!
The other patch is against arch/x86_64/kernel/setup.c ,
it's basically the same seven lines deleted as in
arch/i386/kernel/setup.c. No testing were done on x86_64
besides compilation.

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

1 success, 1 failure, was: Re: Patches to revert in 2.6.7-rc1 or -rc2

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

Hi,

finally I was able to test 2.6.7-rc1-mm1-ruby on my machine
with both i386 and x86_64 systems. The i386 version works
as expected but the x86_64 does not: the machine does not
set up vc17 although I appended a "dumbcon=3D1" parameter to
the boot prompt.

# uname -a
Linux wl-193.226.227-37-szolnok.dunaweb.hu 2.6.7-rc1-mm1-ruby #1 Sun May=20
30 23:09:52 CEST 2004 x86_64 x86_64 x86_64 GNU/Linux
# cat /proc/cmdline
ro root=3D/dev/hda6 dumbcon=3D1 console=3Dtty0

The config is otherwise the same: :0 would be the PCI Radeon 7000,
:1 would be (it does not start - no vc17) the AGP Radeon 9200/SE.

I appended the dmesg and the .configs for both the i386 and the
x86_64 kernels, in case you can find something.

I will try again with 2.6.7-rc1-ruby (without the -mm1) soon
but the same kernel works on i386... Somehow the dummycon
driver seems to not initialize.

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

Re: 1 success, 1 failure, was: Re: Patches to revert in 2.6.7-rc1 or -rc2

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

Hi,

I forgot to tell that the x86_64 system is Fedora Core 2
and I recompiled its xorg-x11-6.7.0-2.src.rpm with your
080_pci_isolate_device_feature.diff patch added.
The man page parts did not apply since the manpage file
is not XF86Config.man anymore. Cutting that out the real
code applied cleanly and works.
The system works nicely otherwise but I still miss vc17. :-)

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

Zoltan Boszormenyi =EDrta:
> Hi,
>=20
> finally I was able to test 2.6.7-rc1-mm1-ruby on my machine
> with both i386 and x86_64 systems. The i386 version works
> as expected but the x86_64 does not: the machine does not
> set up vc17 although I appended a "dumbcon=3D1" parameter to
> the boot prompt.

Re: Patches to revert in 2.6.7-rc1 or -rc2

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

On Sunday 30 May 2004 23:01, Zoltan Boszormenyi wrote:
> Hi,
> 
> I searched through http://linus.bkbits.net:8080/linux-2.5
> for the patches that are needed to be reverted so ruby-2.6.6
> patch apply cleanly. These patches should be applied with
> "patch -R -p1" in alphabetical order.

i have limted coding power, but RC* and PRE* changes
any hour. current linuxconsole.sf.net stay sticky with releases

> I want to try newer kernels/fixes but I also need
> the multihead capability.
> 
> I also needed two additional patches to compile
> this beast on x86_64, one for fs/compat_ioctl.c.
> Aivil, there is a missing semicolon and a new line
> there, look into line 1722 in your CVS!

applied.

> The other patch is against arch/x86_64/kernel/setup.c ,
> it's basically the same seven lines deleted as in
> arch/i386/kernel/setup.c. No testing were done on x86_64
> besides compilation.

we with for results.

Aivils

Re: Patches to revert in 2.6.7-rc1 or -rc2

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

Aivils =EDrta:
> i have limted coding power, but RC* and PRE* changes
> any hour. current linuxconsole.sf.net stay sticky with releases

Tell me I am not useful to you finding the changesets
that cause rejects when patching with ruby... ;-)
I know the bkbits.net tree is changing, that's why I do it.
There are fixes in -rc and -mm patchsets for bugs that may bite.

>>Aivil, there is a missing semicolon and a new line
>>there, look into line 1722 in your CVS!
>=20
> applied.

Thanks, I saw it.

>>The other patch is against arch/x86_64/kernel/setup.c ,
>>it's basically the same seven lines deleted as in
>>arch/i386/kernel/setup.c. No testing were done on x86_64
>>besides compilation.
>=20
> we with for results.
>=20
> Aivils

I keep you posted. :-)

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