Amit Kucheria | 1 Apr 03:41 2009

Re: [PATCH 1/1] UBUNTU: build CRAMFS into kernel to support mounting CRAMFS initrd on iop32x machine

On Tue, Mar 31, 2009 at 11:22 PM,  <bryan.wu <at> canonical.com> wrote:
> From: Bryan Wu <bryan.wu <at> canonical.com>
>
> Bug: #349104
>
> Currently, iop32x built CRAMFS as a module. The kernel for iop32x can not mount CRAMFS
> initrd as rootfs. This patch enables the option of the iop32x kernel config and removes
> CRAMFS module from iop32x kernel module list.

Hi Bryan,

Why do we need to support a cramfs-based initrd?

Regards,
Amit

Steve Conklin | 1 Apr 15:15 2009

Re: [PATCH] lpia configuration changes to allow d-i to run

On 03/31/2009 04:47 PM, sonicmctails <at> gmail.com wrote:
> From: Michael Casadevall <mcasadevall <at> ubuntu.com>
> 
> The attached patch changes the configuration of the lpia kernel to allow
> d-i to both successfully detect installation media, and being able to auto-
> partition using the ext2-4 file systems.
> 
> Michael Casadevall (1):
>   UBUNTU: [lpia] Change ATA, SCSI, SD, ext2-4 modules into compiled-in
>     components
> 
>  debian/abi/2.6.28-11.37/modules.ignore |   62 ++++++++++++++++
>  debian/config/lpia/config              |  122 ++++++++++++++++----------------
>  2 files changed, 123 insertions(+), 61 deletions(-)
>  create mode 100644 debian/abi/2.6.28-11.37/modules.ignore
> 
> 

These look pretty sane, but I'd like to know why the debian installer fails if
these drivers aren't built in. Does it not provide a initramfs? I'm just curious,
can you explain it?

Thanks,

Steve

Amit Kucheria | 1 Apr 15:28 2009

Re: [PATCH] lpia configuration changes to allow d-i to run

On Tue, Mar 31, 2009 at 05:47:19PM -0400, sonicmctails <at> gmail.com wrote:
> From: Michael Casadevall <mcasadevall <at> ubuntu.com>
> 
> The attached patch changes the configuration of the lpia kernel to allow
> d-i to both successfully detect installation media, and being able to auto-
> partition using the ext2-4 file systems.
> 
> Michael Casadevall (1):
>   UBUNTU: [lpia] Change ATA, SCSI, SD, ext2-4 modules into compiled-in
>     components
> 
>  debian/abi/2.6.28-11.37/modules.ignore |   62 ++++++++++++++++
>  debian/config/lpia/config              |  122 ++++++++++++++++----------------
>  2 files changed, 123 insertions(+), 61 deletions(-)
>  create mode 100644 debian/abi/2.6.28-11.37/modules.ignore

Umm.. missing patch?

--

-- 
----------------------------------------------------------------------
Amit Kucheria, Kernel Engineer || amit <at> canonical.com
----------------------------------------------------------------------

Michael Casadevall | 1 Apr 17:56 2009
Picon

Re: [PATCH] lpia configuration changes to allow d-i to run

On 04/01/2009 09:15 AM, Steve Conklin wrote:
> On 03/31/2009 04:47 PM, sonicmctails <at> gmail.com wrote:
>> From: Michael Casadevall<mcasadevall <at> ubuntu.com>
>>
>> The attached patch changes the configuration of the lpia kernel to allow
>> d-i to both successfully detect installation media, and being able to auto-
>> partition using the ext2-4 file systems.
>>
>> Michael Casadevall (1):
>>    UBUNTU: [lpia] Change ATA, SCSI, SD, ext2-4 modules into compiled-in
>>      components
>>
>>   debian/abi/2.6.28-11.37/modules.ignore |   62 ++++++++++++++++
>>   debian/config/lpia/config              |  122 ++++++++++++++++----------------
>>   2 files changed, 123 insertions(+), 61 deletions(-)
>>   create mode 100644 debian/abi/2.6.28-11.37/modules.ignore
>>
>>
>
> These look pretty sane, but I'd like to know why the debian installer fails if
> these drivers aren't built in. Does it not provide a initramfs? I'm just curious,
> can you explain it?
>
d-i does provide an initrd, and can load kernel modules (it for instance loads 
modules for all filesystems in d-i/modules/fs-core-modules). The installer 
loads up with a base RAMdisk which contains the installer, and a few kernel 
modules like usb_storage. It them attempts to mount the installation media on 
/cdrom, and then load any udebs it can find on said media. For the first bit 
to work, d-i would have to be told to load that udeb into the initial ramdisk 
so it could find the CD (the storage-core-modules are currently loaded onto 
(Continue reading)

Tim Gardner | 1 Apr 20:16 2009

Re: [PATCH] Adding linux-restricted modules for server

Stefan Bader wrote:
> Bug: #220781
> 
> There are requests to have a dependency on lrm for server as well.
> 
> Signed-off-by: Stefan Bader <stefan.bader <at> canonical.com>
> ---
>  meta-source/debian/control.d/server |   11 ++++++++++-
>  1 files changed, 10 insertions(+), 1 deletions(-)
> 
> diff --git a/meta-source/debian/control.d/server b/meta-source/debian/control.d/server
> index 312e9ef..9249fed 100644
> --- a/meta-source/debian/control.d/server
> +++ b/meta-source/debian/control.d/server
>  <at>  <at>  -16,11 +16,20  <at>  <at>  Description: Linux kernel image on Server Equipment.
>   This package will always depend on the latest kernel image available
>   for Server Equipment.
>  
> +Package: linux-restricted-modules-server
> +Architecture: i386 amd64
> +Section: restricted/metapackages
> +Priority: optional
> +Depends: linux-restricted-modules-${kernel-abi-version}-server
> +Description: Restricted Linux modules for server kernels
> + This package will always depend on the latest restricted modules available
> + for server kernels.
> +
>  Package: linux-server
>  Architecture: i386 amd64
>  Section: metapackages
(Continue reading)

Andy Whitcroft | 1 Apr 20:40 2009

[APPLIED] [PATCH 0/1] [Jaunty] LP#319825 -- Acer Aspire One wireless

Applied to Jaunty.

-apw

Stefan Bader | 1 Apr 21:39 2009

Re: [Intrepid, Jaunty] SRU: Fix bluetooth hci timout

Stefan Bader wrote:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/268502
> 
> SRU justification:
> 
> Impact: A series of bluetooth dongles [1] do not work properly with the new
>         btusb driver which replaced usb_hci. This has been fixed upstream
>         (2.6.29-rc3). The required changes are all local to the btusb driver.
> 
> Fix: The following three patches are required to fix this (two as prerequisite
>      and one for the actual fix). The fix deviates from upstream by the change
>      of a version number and the missing resume part which does not exist for
>      Intrepid nor Jaunty.
> 
> Test: Fix has been tested by bug reporters.
> 
> [1] Bluetooth doesn't work for at least the following devices:
>     ID 1131:1001 Integrated System Solution Corp. KY-BT100 Bluetooth Adapter
>     ID 1131:1004 Integrated System Solution Corp. Bluetooth Device
>     ID 2001:f111 D-Link Corp. [hex] DBT-122 Bluetooth adapter
>     ID 0a5c:200a Broadcom Corp. Bluetooth dongle
> 
Applied to Intrepid and Jaunty

--

-- 

When all other means of communication fail, try words!

Steve Conklin | 2 Apr 00:58 2009

Re: [PATCH] UBUNTU: [lpia] Change ATA, SCSI, SD, ext2-4 modules into compiled-in components

On 03/31/2009 04:47 PM, sonicmctails <at> gmail.com wrote:
> From: Michael Casadevall <mcasadevall <at> ubuntu.com>
> 
> debian-installer requires that modules needed to see the boot device (i.e.
> CDROM or USB stick), and the installation media be available in the kernel,
> and not as modules. All ext versions also must be compiled in. This change
> puts lpia in line with other architectures.
> 
> Bug: #347458
> 
> Signed-off-by: Michael Casadevall <mcasadevall <at> ubuntu.com>
> ---
>  debian/abi/2.6.28-11.37/modules.ignore |   62 ++++++++++++++++
>  debian/config/lpia/config              |  122 ++++++++++++++++----------------
>  2 files changed, 123 insertions(+), 61 deletions(-)
>  create mode 100644 debian/abi/2.6.28-11.37/modules.ignore
> 
> diff --git a/debian/abi/2.6.28-11.37/modules.ignore b/debian/abi/2.6.28-11.37/modules.ignore
> new file mode 100644
> index 0000000..09e0b68
> --- /dev/null
> +++ b/debian/abi/2.6.28-11.37/modules.ignore
>  <at>  <at>  -0,0 +1,62  <at>  <at> 
> +ata_generic
> +ata_piix
> +cdrom
> +crc16
> +ext2
> +ext3
> +ext4
(Continue reading)

realKano@directbox.com | 2 Apr 02:58 2009

missing dvb-s2 firmware for 2.6.28: dvb-fe-cx24116.fw

Hauppauge DVB-S2 PCI cards directly work with 2.6.28+, but the firmware is missing, thats the way to get it:

wget -N http://www.hauppauge.de/software/mce/88x_2_122_26109_WHQL.zip
unzip -oj 88x_2_122_26109_WHQL.zip Driver88/hcw88bda.sys
dd if=hcw88bda.sys of=dvb-fe-cx24116.fw skip=81768 bs=1 count=32522
echo "9950fe612d47217e6068f7141de225b0 *dvb-fe-cx24116.fw"|md5sum -c

Please add it.

Tim Gardner | 2 Apr 01:14 2009

Re: [PATCH] UBUNTU: SAUCE: PM: Increase TEST_SUSPEND_SECONDS to avoidfalse kernel oops on resume

Tim Gardner wrote:
> TJ wrote:
>> On Mon, 2009-03-23 at 09:51 -0500, Mario_Limonciello <at> Dell.com wrote:
>>> Matter of fact the change for the 5 seconds for sata links has landed
>>> upstream now:
>>>
>>> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commi
>>> t;h=e7d3ef13d52a126438f687a1a32da65ff926ed57
>> Looking at this I think we'd be better off doing a cherry-pick of
>> e7d3ef13d52a126438f687a1a32da65ff926ed57 and setting
>> TEST_SUSPEND_SECONDS to 10.
>>
>> That would mean the SATA link timeouts wouldn't show up and the upstream
>> patch deals with the situation Stefan and Andy talked of: SATA links
>> getting stuck in soft/hard reset delays.
>>
>> 10 seconds for 'everything else' would make sense in the context of the
>> other reasons for longer resume times - mostly a by-product of more than
>> the average number of devices to resume.
>>
>>
> 
> Jaunty is already at a 5 second timeout;
> 'b65db6fd5d341d27f6d3f62c2b111ca0df0c6dee UBUNTU: SAUCE: Increase
> ATA_TMOUT_PMP_SRST_WAIT to 5 seconds.'
> 
> However, increasing TEST_SUSPEND_SECONDS a bit seems reasonable.
> 
> Andy - what think you?
> 
(Continue reading)


Gmane