Bug#815324: linux: Please add support for Novena

[Vagrant Cascadian]

Source: linux
Version: 4.4.2-2
Severity: wishlist
Tags: patch

Please add the following patches taken from 4.5-rc4 to add the .dtb
file for the Novena.

Tested MMC, SATA, USB, ethernet, and serial console. Haven't had luck
with HDMI or LCD; that may require further work upstream.

Thanks!

live well,
  vagrant


diff --git 
a/debian/patches/features/arm/novena/0008-ARM-dts-imx6q-add-Novena-board.patch 
b/debian/patches/features/arm/novena/0008-ARM-dts-imx6q-add-Novena-board.patch
new file mode 100644
index 000..fd5b81e
--- /dev/null
+++ 
b/debian/patches/features/arm/novena/0008-ARM-dts-imx6q-add-Novena-board.patch
@@ -0,0 +1,825 @@
+From 70a8c03bd9eea54fcd2616302403b80c20729db9 Mon Sep 17 00:00:00 2001
+From: Sean Cross 
+Date: Fri, 18 Dec 2015 06:29:50 +0100
+Subject: [PATCH 8/9] ARM: dts: imx6q: add Novena board
+
+Novena is an open-hardware laptop/desktop/bare board.
+
+See http://www.kosagi.com/w/index.php?title=Novena_Main_Page
+
+Signed-off-by: Sean Cross 
+Signed-off-by: Marek Vasut 
+Cc: Lucas Stach 
+Reviewed-by: Fabio Estevam 
+Signed-off-by: Shawn Guo 
+---
+ arch/arm/boot/dts/Makefile |   1 +
+ arch/arm/boot/dts/imx6q-novena.dts | 785 +
+ 2 files changed, 786 insertions(+)
+
+diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile
+index ad3e25d..f0573a8 100644
+--- a/arch/arm/boot/dts/Makefile
 b/arch/arm/boot/dts/Makefile
+@@ -326,6 +326,7 @@ dtb-$(CONFIG_SOC_IMX6Q) += \
+   imx6q-hummingboard.dtb \
+   imx6q-nitrogen6x.dtb \
+   imx6q-nitrogen6_max.dtb \
++  imx6q-novena.dtb \
+   imx6q-phytec-pbab01.dtb \
+   imx6q-rex-pro.dtb \
+   imx6q-sabreauto.dtb \
+diff --git a/arch/arm/boot/dts/imx6q-novena.dts 
b/arch/arm/boot/dts/imx6q-novena.dts
+new file mode 100644
+index 000..5acd0c6
+--- /dev/null
 b/arch/arm/boot/dts/imx6q-novena.dts
+@@ -0,0 +1,785 @@
++/*
++ * Copyright 2015 Sutajio Ko-Usagi PTE LTD
++ *
++ * This file is dual-licensed: you can use it either under the terms
++ * of the GPL or the X11 license, at your option. Note that this dual
++ * licensing only applies to this file, and not this project as a
++ * whole.
++ *
++ *  a) This file is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License as
++ * published by the Free Software Foundation; either version 2 of
++ * the License, or (at your option) any later version.
++ *
++ * This file is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public
++ * License along with this file; if not, write to the Free
++ * Software Foundation, Inc., 51 Franklin St, Fifth Floor, Boston,
++ * MA 02110-1301 USA
++ *
++ * Or, alternatively,
++ *
++ *  b) Permission is hereby granted, free of charge, to any person
++ * obtaining a copy of this software and associated documentation
++ * files (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use,
++ * copy, modify, merge, publish, distribute, sublicense, and/or
++ * sell copies of the Software, and to permit persons to whom the
++ * Software is furnished to do so, subject to the following
++ * conditions:
++ *
++ * The above copyright notice and this permission notice shall be
++ * included in all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
++ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
++ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
++ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
++ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
++ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
++ * OTHER DEALINGS IN THE SOFTWARE.
++ *
++ */
++
++/dts-v1/;
++#include "imx6q.dtsi"
++#include 
++#include 
++
++/ {
++  model = "Kosagi Novena Dual/Quad";
++  compatible = "kosagi,imx6q-novena", "fsl,imx6q";
++
++  chosen {
++  stdout-path = 
++  };
++
++  backlight: backlight {
++  compatible = "pwm-backlight";
++  pwms = < 0 1000>;
++  pinctrl-names = "default";
++  pinctrl-0 = <_backlight_novena>;
++  power-supply = <_lvds_lcd>;
++  brightness-levels = <0 3 6 12 16 24 32 48 64 96 128 192 255>;
++  default-brightness-level = <12>;
++  };
++
++  gpio-keys {
++  compatible = 

Bug#813881: linux-image-4.3.0-1-armmp install wrong dtb on Wandboard Quad Rev B1

[Vagrant Cascadian]

On 2016-02-06, Heinrich Schuchardt wrote:
> Booting with u-boot-imx requires imx6q-wandboard-revb1.dtb.
> linux-image-4.3.0-1-armmp installs imx6q-wandboard.dtb
> leaving me with a system that will not boot.
>
> With imx6q-wandboard-revb1.dtb the system boots.

As you've noted, flash-kernel has no way of distinguishing which variant
to support, though u-boot does. I've sometimes wondered weather u-boot
should pass a boot argument for which .dtb to use...

To make matters worse, in older versions of the linux kernel, such as
the 3.16.x in jessie, imx6q-wandboard.dtb may actually be for wandboard
revb variants, not for the revc variants (although I have one of each,
both running the same .dtb in jessie without obvious problem, though I
don't make use of the wifi or bluetooth on either).

When the revc was added, backwards compatibility was broken by renaming
the revb .dtb file instead of keeping it and introducing the revc in a
new .dtb... kind of hard to fix correctly now...

Adding support for flash-kernel to copy multiple, or even optionally all
.dtb files could at least work around the issue.


live well,
  vagrant


signature.asc
Description: PGP signature

Bug#814061: linux: kernel panic on wandboard solo

[Vagrant Cascadian]

Source: linux
Version: 4.3.3-5
Severity: normal

I tried booting debian-installer daily (20160207) and stretch (20160106)
images on a wandboard solo rev c1, but both resulted in a kernel
panic. It loaded the correct imx6dl-wandboard.dtb file (and manually
forcing to try the revb1 variant didn't change anything I could see).

Booting debian-installer from jessie worked fine.

Log attached is from booting the 20160207 debian-installer image, with
4.3.3-7.


live well,
  vagrant



wb-solo-kernel-panic
Description: Binary data


signature.asc
Description: PGP signature

Bug#810048: linux: Add support for USB/FAN on BeagleBoard-X15

[Vagrant Cascadian]

Source: linux
Version: 4.4~rc8-1~exp1
Severity: normal
Tags: patch

Thanks for enabling DRA7XX support!

Adding the following configuration options make USB and FAN support
work on the BeagleBone-X15:

diff --git a/debian/config/armhf/config.armmp b/debian/config/armhf/config.armmp
index bef9ace..1d19120 100644
--- a/debian/config/armhf/config.armmp
+++ b/debian/config/armhf/config.armmp
@@ -1156,3 +1156,11 @@ CONFIG_SND_SUN4I_CODEC=m
 ##
  CONFIG_SND_SOC_TEGRA=m

+#drivers/usb/dwc3/Kconfig:
+CONFIG_USB_DWC3_DUAL_ROLE=y
+
+#drivers/extcon/Kconfig:
+CONFIG_EXTCON_USB_GPIO=m
+
+#drivers/hwmon/Kconfig:
+CONFIG_SENSORS_GPIO_FAN=m


live well,
  vagrant


signature.asc
Description: PGP signature

Bug#809083: linux: armhf/armmp: Please enable support for Rockchip platforms

[Vagrant Cascadian]

On 2016-01-01, Ian Campbell wrote:
> On Sat, 2015-12-26 at 18:26 -0800, Vagrant Cascadian wrote:
>> +#drivers/crypto/Kconfig:
>> +CONFIG_CRYPTO_DEV_ROCKCHIP=m
>
> This one didn't seem to exist anywhere in the mainline tree.

Apparently, it's only in linux-next so far.


> The rest looks good to me, I intend to apply once I get a chance to
> build test.

Great, thanks!


live well,
  vagrant


signature.asc
Description: PGP signature

Bug#807624: armhf: Please enable support for DRA7XX systems (Beagle-x15)

[Vagrant Cascadian]

On 2016-01-01, Ian Campbell wrote:
> On Tue, 2015-12-15 at 16:04 -0800, Vagrant Cascadian wrote:
>> Would you consider enabling it even though there are some notable
>> features not yet working?
>
> Seems like there is a sufficiently useful set of stuff which is
> working, so I think we should.
>
> The final change to the generated .config with your one liner is below,
> I think we should probably enable CONFIG_DRA752_THERMAL
> and CONFIG_VIDEO_TI_VPE? The former seems obvious so I'll include it,
> not so sure about the other option, will leave it for now.

Ben also suggested CONFIG_VIDEO_TI_VPE, and at first I didn't see it
load, but on my last build attempt I did. I don't know how I would
actually test it, however.  Same for CONFIG_SND_DAVINCI_SOC_MCASP and
CONFIG_SPI_TI_QSPI, so I would suggest adding:

CONFIG_SND_DAVINCI_SOC_MCASP=m
CONFIG_VIDEO_TI_VPE=m
CONFIG_SPI_TI_QSPI=m
CONFIG_PCI_DRA7XX=y
CONFIG_DRA752_THERMAL=y
CONFIG_SOC_DRA7XX=y


live well,
  vagrant


signature.asc
Description: PGP signature

Bug#809083: linux: armhf/armmp: Please enable support for Rockchip platforms

[Vagrant Cascadian]

Source: linux
Version: 4.4~rc6-1~exp1
Severity: wishlist
Tags: patch

I've tested the following options on a Firefly-rk3288, which boots and
supports USB, MMC, eMMC and ethernet. Tested against 4.4~rc6-1~exp1,
though I suspect most if not all options are also available on a 4.3.x
kernel.

Please consider adding the options below to the armmp kernels.

Thanks!


live well,
  vagrant

diff --git a/debian/config/armhf/config.armmp b/debian/config/armhf/config.armmp
index 025e388..8c1bad7 100644
--- a/debian/config/armhf/config.armmp
+++ b/debian/config/armhf/config.armmp
@@ -1046,3 +1046,54 @@ CONFIG_SND_OMAP_SOC_OMAP3_PANDORA=m
 ##
 CONFIG_SND_SOC_TEGRA=m
 
+## file: arch/arm/mach-rockchip/Kconfig
+CONFIG_ARCH_ROCKCHIP=y
+#drivers/crypto/Kconfig:
+CONFIG_CRYPTO_DEV_ROCKCHIP=m
+#drivers/gpu/drm/rockchip/Kconfig:
+CONFIG_DRM_ROCKCHIP=m
+#drivers/gpu/drm/rockchip/Kconfig:
+CONFIG_ROCKCHIP_DW_HDMI=m
+#drivers/iio/adc/Kconfig:
+CONFIG_ROCKCHIP_SARADC=m
+#drivers/iommu/Kconfig:
+CONFIG_ROCKCHIP_IOMMU=y
+#drivers/mmc/host/Kconfig:
+CONFIG_MMC_DW_ROCKCHIP=m
+#drivers/net/ethernet/arc/Kconfig:
+CONFIG_EMAC_ROCKCHIP=m
+#drivers/net/ethernet/stmicro/stmmac/Kconfig:
+CONFIG_DWMAC_ROCKCHIP=m
+#drivers/nvmem/Kconfig:
+CONFIG_ROCKCHIP_EFUSE=m
+#drivers/phy/Kconfig:
+CONFIG_PHY_ROCKCHIP_USB=m
+#drivers/power/avs/Kconfig:
+CONFIG_ROCKCHIP_IODOMAIN=m
+#drivers/pwm/Kconfig:
+CONFIG_PWM_ROCKCHIP=m
+#drivers/spi/Kconfig:
+CONFIG_SPI_ROCKCHIP=m
+#drivers/thermal/Kconfig:
+CONFIG_ROCKCHIP_THERMAL=m
+#sound/soc/rockchip/Kconfig:
+CONFIG_SND_SOC_ROCKCHIP=m
+#sound/soc/rockchip/Kconfig:
+CONFIG_SND_SOC_ROCKCHIP_I2S=m
+#sound/soc/rockchip/Kconfig:
+CONFIG_SND_SOC_ROCKCHIP_SPDIF=m
+#sound/soc/rockchip/Kconfig:
+CONFIG_SND_SOC_ROCKCHIP_MAX98090=m
+#sound/soc/rockchip/Kconfig:
+CONFIG_SND_SOC_ROCKCHIP_RT5645=m
+
+#drivers/i2c/busses/Kconfig:
+CONFIG_I2C_RK3X=m
+
+#drivers/usb/dwc2/Kconfig:
+CONFIG_USB_DWC2=m
+
+# drivers/regulator/Kconfig:
+CONFIG_REGULATOR_ACT8865=m
+CONFIG_REGULATOR_FAN53555=m
+CONFIG_REGULATOR_TPS65910=m


signature.asc
Description: PGP signature

Bug#807624: armhf: Please enable support for DRA7XX systems (Beagle-x15)

[Vagrant Cascadian]

On 2015-12-10, Ben Hutchings wrote:
> On Thu, 2015-12-10 at 16:09 -0800, Vagrant Cascadian wrote:
>> Please enable support for DRA7XX systems, such as the Beagle-x15.
>> 
>> I was able to boot a Beagle-x15 system by applying the following
>> patch
>> to the debian packaging for linux 4.3-1~exp1:
>> 
>> diff --git a/debian/config/armhf/config.armmp
>> b/debian/config/armhf/config.armmp
>> index 285f9c3..457ca53 100644
>> --- a/debian/config/armhf/config.armmp
>> +++ b/debian/config/armhf/config.armmp
>> @@ -69,6 +69,7 @@ CONFIG_ARCH_OMAP3=y
>>  CONFIG_ARCH_OMAP4=y
>>  CONFIG_SOC_OMAP5=y
>>  CONFIG_SOC_AM33XX=y
>> +CONFIG_SOC_DRA7XX=y
>>  CONFIG_ARCH_OMAP2PLUS_TYPICAL=y
>>  CONFIG_SOC_OMAP3430=y
>>  CONFIG_SOC_TI81XX=y
>
> How about additional device drivers?  I found these which seem to be
> useful for DRA7xx chips but aren't enabled yet:
>
> CONFIG_SND_DAVINCI_SOC_MCASP
> CONFIG_VIDEO_TI_VPE
> CONFIG_SPI_TI_QSPI
> CONFIG_PCI_DRA7XX

With those enabled, there were no additional modules automatically
loaded and no PCI devices listed. So it doesn't seem to change anything
for the beagle-x15.

USB didn't work for me; all I saw were the USB3 and USB2 hub devices,
nothing plugged into them.

I haven't yet tested eSATA or video/framebuffer output yet. Hope to try
that later today. I'll also test a build with the 4.4.x kernel, to see
if updates to any of the drivers improve the situation.

live well,
  vagrant


signature.asc
Description: PGP signature

Bug#807624: armhf: Please enable support for DRA7XX systems (Beagle-x15)

[Vagrant Cascadian]

Control: tags 807624 -moreinfo

On 2015-12-15, Vagrant Cascadian wrote:
> On 2015-12-10, Ben Hutchings wrote:
>> On Thu, 2015-12-10 at 16:09 -0800, Vagrant Cascadian wrote:
>>> Please enable support for DRA7XX systems, such as the Beagle-x15.
...
>> How about additional device drivers?  I found these which seem to be
>> useful for DRA7xx chips but aren't enabled yet:
>>
>> CONFIG_SND_DAVINCI_SOC_MCASP
>> CONFIG_VIDEO_TI_VPE
>> CONFIG_SPI_TI_QSPI
>> CONFIG_PCI_DRA7XX
>
> With those enabled, there were no additional modules automatically
> loaded and no PCI devices listed. So it doesn't seem to change anything
> for the beagle-x15.
>
> USB didn't work for me; all I saw were the USB3 and USB2 hub devices,
> nothing plugged into them.
>
> I haven't yet tested eSATA or video/framebuffer output yet. Hope to try
> that later today. I'll also test a build with the 4.4.x kernel, to see
> if updates to any of the drivers improve the situation.

With 4.4.x and with CONFIG_SOC_DRA7XX=y added, I get:

Works:
frambuffer video
serial console
both CPUs detected
all RAM detected
eMMC works
microSD works

Doesn't work:
eSATA
USB

Haven't tested:
audio

dmesg and lsmod output attached if that would be helpful.


Would you consider enabling it even though there are some notable
features not yet working?


live well,
  vagrant



dmesg
Description: Binary data


lsmod
Description: Binary data


signature.asc
Description: PGP signature

Bug#807624: armhf: Please enable support for DRA7XX systems (Beagle-x15)

[Vagrant Cascadian]

Source: linux
Version: 4.3-1~exp1
Severity: wishlist
Tags: patch

Please enable support for DRA7XX systems, such as the Beagle-x15.

I was able to boot a Beagle-x15 system by applying the following patch
to the debian packaging for linux 4.3-1~exp1:

diff --git a/debian/config/armhf/config.armmp b/debian/config/armhf/config.armmp
index 285f9c3..457ca53 100644
--- a/debian/config/armhf/config.armmp
+++ b/debian/config/armhf/config.armmp
@@ -69,6 +69,7 @@ CONFIG_ARCH_OMAP3=y
 CONFIG_ARCH_OMAP4=y
 CONFIG_SOC_OMAP5=y
 CONFIG_SOC_AM33XX=y
+CONFIG_SOC_DRA7XX=y
 CONFIG_ARCH_OMAP2PLUS_TYPICAL=y
 CONFIG_SOC_OMAP3430=y
 CONFIG_SOC_TI81XX=y


live well,
  vagrant


signature.asc
Description: PGP signature

Bug#807624: armhf: Please enable support for DRA7XX systems (Beagle-x15)

[Vagrant Cascadian]

On 2015-12-10, Ben Hutchings wrote:
>> Please enable support for DRA7XX systems, such as the Beagle-x15.
...
> How about additional device drivers?  I found these which seem to be
> useful for DRA7xx chips but aren't enabled yet:
>
> CONFIG_SND_DAVINCI_SOC_MCASP
> CONFIG_VIDEO_TI_VPE
> CONFIG_SPI_TI_QSPI
> CONFIG_PCI_DRA7XX

I'm not sure which of those are relevent for Beagle-x15, but will give
them a try with another kernel build over the weekend.

Even without those device drivers, my quick test worked with serial
console, ethernet and microSD.


live well,
  vagrant


signature.asc
Description: PGP signature

Bug#806284: armhf/armmp: Please enable CONFIG_REGULATOR_PFUZE100 module

[Vagrant Cascadian]

Source: linux
Version: 4.2.6-1
Severity: wishlist
Tags: patch

Please enable CONFIG_REGULATOR_PFUZE100=m in the armmp* kernel
flavors.

This is needed for USB support on some platforms, such as Novena.

diff --git a/debian/config/armhf/config.armmp b/debian/config/armhf/config.armmp
index b1a7daf..7a30ec9 100644
--- a/debian/config/armhf/config.armmp
+++ b/debian/config/armhf/config.armmp
@@ -739,6 +739,7 @@ CONFIG_REGULATOR_MC13783=m
 CONFIG_REGULATOR_MC13892=m
 CONFIG_REGULATOR_PALMAS=y
 CONFIG_REGULATOR_PBIAS=m
+CONFIG_REGULATOR_PFUZE100=m
 CONFIG_REGULATOR_S2MPA01=m
 CONFIG_REGULATOR_S2MPS11=m
 CONFIG_REGULATOR_S5M8767=m


Thanks!


live well,
  vagrant


signature.asc
Description: PGP signature

Bug#805856: initramfs-tools: on cubox generated boot.scr has bad syntax

[Vagrant Cascadian]

Control: reassign -1 flash-kernel
Control: tags -1 pending

On 2015-11-23, Rick Thomas wrote:
> The /boot/boot.scr file generated by (I assume) "update-initramfs -u" 
> contains 
> if -z "${prefix}" ; then
> which should be
> if test -z "${prefix}" ; then

This is actually a bug in flash-kernel, as well as a couple other
similar syntax errors in the wandboard and uboot-generic scripts.

Sorry about that!

Thanks for the report, I've fixed in flash-kernel git.

live well,
  vagrant


signature.asc
Description: PGP signature

Bug#804850: linux: Please enable support for Odroid-XU4

[Vagrant Cascadian]

Source: linux
Version: 4.3-1~exp1
Severity: wishlist
Tags: patch

Please enable support for the Odroid-XU4, by applying the attached
patches from mainline linux, which adds the .dtb file
exynos5422-odroidxu4.dtb.

Tested to work using a custom build of 4.3-1~exp1.


live well,
  vagrant

From 0fb033bb3a384fdf1c08bcfaf429ca6431de91c7 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski 
Date: Mon, 17 Aug 2015 08:55:51 +0900
Subject: [PATCH 1/3] ARM: dts: Fix power off method for
 exynos5422-odroidxu3-common

The Odroid XU3 family boards have ACOKB pin of PMIC grounded, instead of
pulled up as usual. This means that PMIC must manually set PWRHOLD field
in its CTRL1 register to low before initiating power down.

This fixes Odroid XU3 powering off:
[   25.966053] reboot: Power down
[   25.967679] Power down.
[   26.070174] Power down failed, please power off system manually.

Signed-off-by: Krzysztof Kozlowski 
Reported-by: Anand Moon 
Reviewed-by: Javier Martinez Canillas 
Signed-off-by: Kukjin Kim 
---
 arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi b/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi
index 3b43e57..90d298d 100644
--- a/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi
+++ b/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi
@@ -160,6 +160,7 @@
 		s2mps11,buck2-ramp-enable = <1>;
 		s2mps11,buck3-ramp-enable = <1>;
 		s2mps11,buck4-ramp-enable = <1>;
+		samsung,s2mps11-acokb-ground;
 
 		interrupt-parent = <>;
 		interrupts = <4 IRQ_TYPE_EDGE_FALLING>;
-- 
2.1.4

From 8cec734904f55d5551ac97fa9910752ef521a673 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski 
Date: Wed, 16 Sep 2015 08:37:07 +0900
Subject: [PATCH 2/3] ARM: dts: Split audio configuration to separate
 exynos5422-odroidxu3-audio

The Odroid XU4 board does not have audio codec so before adding DTS
for new board split the audio codec to separate DTSI file. Include
the audio codec DTSI in Odroid XU3 and XU3-Lite boards.

Signed-off-by: Krzysztof Kozlowski 
Reviewed-by: Javier Martinez Canillas 
Signed-off-by: Kukjin Kim 
---
 arch/arm/boot/dts/exynos5422-odroidxu3-audio.dtsi  | 61 ++
 arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi | 47 -
 arch/arm/boot/dts/exynos5422-odroidxu3-lite.dts|  1 +
 arch/arm/boot/dts/exynos5422-odroidxu3.dts |  1 +
 4 files changed, 63 insertions(+), 47 deletions(-)
 create mode 100644 arch/arm/boot/dts/exynos5422-odroidxu3-audio.dtsi

diff --git a/arch/arm/boot/dts/exynos5422-odroidxu3-audio.dtsi b/arch/arm/boot/dts/exynos5422-odroidxu3-audio.dtsi
new file mode 100644
index 000..9493923
--- /dev/null
+++ b/arch/arm/boot/dts/exynos5422-odroidxu3-audio.dtsi
@@ -0,0 +1,61 @@
+/*
+ * Hardkernel Odroid XU3 Audio Codec device tree source
+ *
+ * Copyright (c) 2015 Krzysztof Kozlowski
+ * Copyright (c) 2014 Collabora Ltd.
+ * Copyright (c) 2013 Samsung Electronics Co., Ltd.
+ *		http://www.samsung.com
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+/ {
+	sound: sound {
+		compatible = "simple-audio-card";
+
+		simple-audio-card,name = "Odroid-XU3";
+		simple-audio-card,widgets =
+			"Headphone", "Headphone Jack",
+			"Speakers", "Speakers";
+		simple-audio-card,routing =
+			"Headphone Jack", "HPL",
+			"Headphone Jack", "HPR",
+			"Headphone Jack", "MICBIAS",
+			"IN1", "Headphone Jack",
+			"Speakers", "SPKL",
+			"Speakers", "SPKR";
+
+		simple-audio-card,format = "i2s";
+		simple-audio-card,bitclock-master = <_codec>;
+		simple-audio-card,frame-master = <_codec>;
+
+		simple-audio-card,cpu {
+			sound-dai = < 0>;
+			system-clock-frequency = <1920>;
+		};
+
+		link0_codec: simple-audio-card,codec {
+			sound-dai = <>;
+			clocks = < CLK_I2S_CDCLK>;
+		};
+	};
+};
+
+_5 {
+	status = "okay";
+	max98090: max98090@10 {
+		compatible = "maxim,max98090";
+		reg = <0x10>;
+		interrupt-parent = <>;
+		interrupts = <2 0>;
+		clocks = < CLK_I2S_CDCLK>;
+		clock-names = "mclk";
+		#sound-dai-cells = <0>;
+	};
+};
+
+ {
+	status = "okay";
+};
diff --git a/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi b/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi
index 90d298d..2f0fb86 100644
--- a/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi
+++ b/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi
@@ -80,36 +80,6 @@
 		};
 	};
 
-	sound: sound {
-		compatible = "simple-audio-card";
-
-		simple-audio-card,name = "Odroid-XU3";
-		simple-audio-card,widgets =
-			"Headphone", "Headphone Jack",
-			"Speakers", "Speakers";
-		simple-audio-card,routing =
-			"Headphone Jack", "HPL",
-			"Headphone Jack", 

Bug#762634: initramfs-tools: [armhf] mounting rootfs on USB disk fails / some USB host controller drivers missing in initramfs

[Vagrant Cascadian]

On 2014-09-30, Ben Hutchings wrote:
> On Tue, 2014-09-30 at 08:19 +0100, Ian Campbell wrote:
>> On Fri, 2014-09-26 at 00:08 +0100, Ben Hutchings wrote:
>> > However, at the moment initramfs-tools won't include PHY drivers even in
>> > that configuration.
>> 
>> I spent some time last week hunting for a sysfs link between a device
>> and the phys which it is using, without success. Do you have any ideas?
>
> I suspect they're not visible there yet.
>
> I think you could include all PHY drivers (drivers/phy and
> drivers/usb/phy) when MODULES=most, and only the currently loaded
> drivers if MODULES=dep.
>
> USB non-generic PHY drivers don't appear in the device model at all
> (ugh!) so in the MODULES=dep case you may have to bodge it by checking
> for modules with names beginning with "phy-" (check both /sys/module
> and /lib/modules/$(uname -r)/modules.builtin).

Would definitely like to see this, with a recent install on a Wandboard
Dual with a USB2 sata disk for the rootfs. It installed fine with
jessie's debian-installer, but failed on initial boot.

I worked around it by adding to /etc/initramfs-tools/modules:

  ci_hdrc_imx
  phy_mxs_usb

I haven't yet verified if only adding "phy_mxs_usb" instead of both will
work.

Had a similar problem with an Odroid-XU4 install (which isn't yet
supported by debian-installer), and worked around it similarly, although
haven't narrowed down exactly which modules are needed, though I
suspect one or both of:

  phy_exynos_usb2
  phy_exynos5_usbdrd


Something that pulls in all the phy-* modules would likely fix the issue
in a generic way, rather than playing whack-a-mole with various phy
types.


live well,
  vagrant


signature.asc
Description: PGP signature

Bug#786925: linux: overlay filesystem incompatible with NFS

[Vagrant Cascadian]

On 2015-08-11, Vagrant Cascadian vagr...@debian.org wrote:
 Control: found 4.1.3-1

 On 2015-05-26, Vagrant Cascadian wrote:
 LTSP has been using aufs to provide a writeable tmpfs layer on top of
 NFS since wheezy.

 With the switch to the overlay filesystem (after adjusting LTSP's
 initramfs-tools hooks to support overlay), it appears to fail due to an
 incompatibility with NFS. 

 I haven't confirmed, but this may be fixed in mainline, not sure how
 difficult it would be to backport to 4.1.x:

The attached 3 patches from mainline (one of them refreshed to apply to
4.1.3) allow it to mount with NFS as a lowerdir, and it mostly works,
although with some odd behavior when writing to files:

  # echo ltsp95  /etc/hostname
  bash: /etc/hostname: Permission denied
  # touch /etc/hostname
  # echo ltsp95  /etc/hostname
  # cat /etc/hostname
  ltsp95

Something seems a bit odd with the copyup behavior...

live well,
  vagrant

From a6f15d9a756571babbb2b2cd4fdd1b64a5de232b Mon Sep 17 00:00:00 2001
From: Miklos Szeredi mszer...@suse.cz
Date: Mon, 22 Jun 2015 13:53:48 +0200
Subject: [PATCH 1/2] ovl: don't traverse automount points

NFS and other distributed filesystems may place automount points in the
tree.  Previoulsy overlayfs refused to mount such filesystems types (based
on the existence of the .d_automount callback), even if the actual export
didn't have any automount points.

It cannot be determined in advance whether the filesystem has automount
points or not.  The solution is to allow fs with .d_automount but refuse to
traverse any automount points encountered.

Signed-off-by: Miklos Szeredi mszer...@suse.cz
---
 fs/overlayfs/super.c | 8 ++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c
index bf8537c..de9d2ee 100644
--- a/fs/overlayfs/super.c
+++ b/fs/overlayfs/super.c
@@ -303,6 +303,10 @@ static inline struct dentry *ovl_lookup_real(struct dentry *dir,
 	} else if (!dentry-d_inode) {
 		dput(dentry);
 		dentry = NULL;
+	} else if (dentry-d_flags  DCACHE_MANAGED_DENTRY) {
+		dput(dentry);
+		/* Don't support traversing automounts */
+		dentry = ERR_PTR(-EREMOTE);
 	}
 	return dentry;
 }
@@ -700,12 +704,12 @@ static bool ovl_is_allowed_fs_type(struct dentry *root)
 
 	/*
 	 * We don't support:
-	 *  - automount filesystems
+	 *  - autofs
 	 *  - filesystems with revalidate (FIXME for lower layer)
 	 *  - filesystems with case insensitive names
 	 */
 	if (dop 
-	(dop-d_manage || dop-d_automount ||
+	(dop-d_manage ||
 	 dop-d_revalidate || dop-d_weak_revalidate ||
 	 dop-d_compare || dop-d_hash)) {
 		return false;
-- 
2.1.4

From 7c03b5d45b8eebf025053d8fe887cc262ba6 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi mszer...@suse.cz
Date: Mon, 22 Jun 2015 13:53:48 +0200
Subject: [PATCH 2/2] ovl: allow distributed fs as lower layer

Allow filesystems with .d_revalidate as lower layer(s), but not as upper
layer.

For local filesystems the rule was that modifications on the layers
directly while being part of the overlay results in undefined behavior.

This can easily be extended to distributed filesystems: we assume the tree
used as lower layer is static, which means -d_revalidate() should always
return 1.  If that is not the case, return -ESTALE, don't try to work
around the modification.

Signed-off-by: Miklos Szeredi mszer...@suse.cz
---
 fs/overlayfs/super.c | 113 +++
 1 file changed, 88 insertions(+), 25 deletions(-)

diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c
index de9d2ee..8a08c58 100644
--- a/fs/overlayfs/super.c
+++ b/fs/overlayfs/super.c
@@ -273,10 +273,57 @@ static void ovl_dentry_release(struct dentry *dentry)
 	}
 }
 
+static int ovl_dentry_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	struct ovl_entry *oe = dentry-d_fsdata;
+	unsigned int i;
+	int ret = 1;
+
+	for (i = 0; i  oe-numlower; i++) {
+		struct dentry *d = oe-lowerstack[i].dentry;
+
+		if (d-d_flags  DCACHE_OP_REVALIDATE) {
+			ret = d-d_op-d_revalidate(d, flags);
+			if (ret  0)
+return ret;
+			if (!ret) {
+if (!(flags  LOOKUP_RCU))
+	d_invalidate(d);
+return -ESTALE;
+			}
+		}
+	}
+	return 1;
+}
+
+static int ovl_dentry_weak_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	struct ovl_entry *oe = dentry-d_fsdata;
+	unsigned int i;
+	int ret = 1;
+
+	for (i = 0; i  oe-numlower; i++) {
+		struct dentry *d = oe-lowerstack[i].dentry;
+
+		if (d-d_flags  DCACHE_OP_WEAK_REVALIDATE) {
+			ret = d-d_op-d_weak_revalidate(d, flags);
+			if (ret = 0)
+break;
+		}
+	}
+	return ret;
+}
+
 static const struct dentry_operations ovl_dentry_operations = {
 	.d_release = ovl_dentry_release,
 };
 
+static const struct dentry_operations ovl_reval_dentry_operations = {
+	.d_release = ovl_dentry_release,
+	.d_revalidate = ovl_dentry_revalidate,
+	.d_weak_revalidate = ovl_dentry_weak_revalidate,
+};
+
 static struct ovl_entry *ovl_alloc_entry(unsigned int numlower

Bug#468114: Loopback file system support.

[Vagrant Cascadian]

On 2015-08-11, Vagrant Cascadian wrote:
 On 2015-08-11, Vagrant Cascadian wrote:
 I've taken a quick stab as refreshing this, though I haven't yet tested
 it. I did move mounting of the loopback to mount_loop_root in functions,
 as I would like to eventually support loopback mounted files from NFS.

 Made some small changes, updated and *tested* this time! Patch below...

And added NFS support! Full patch below...

live well,
  vagrant

diff --git a/init b/init
index abf7f25..2760bcb 100755
--- a/init
+++ b/init
@@ -98,6 +98,15 @@ for x in $(cat /proc/cmdline); do
;;
esac
;;
+   loop=*)
+   LOOP=${x#loop=}
+   ;;
+   loopflags=*)
+   LOOPFLAGS=-o ${x#loopflags=}
+   ;;
+   loopfstype=*)
+   LOOPFSTYPE=${x#loopfstype=}
+   ;;
nfsroot=*)
NFSROOT=${x#nfsroot=}
;;
diff --git a/initramfs-tools.8 b/initramfs-tools.8
index ea8c098..ce8e830 100644
--- a/initramfs-tools.8
+++ b/initramfs-tools.8
@@ -42,6 +42,19 @@ The default is 180 seconds.
 set the file system mount option string.
 
 .TP
+\fB\fI loop
+path within the original root file system to loop-mount and use as the
+real root file system.
+
+.TP
+\fB\fI loopflags
+set the loop file system mount option string, if applicable.
+
+.TP
+\fB\fI loopfstype
+set the loop file system type, if applicable.
+
+.TP
 \fB\fI nfsroot
 can be either auto to try to get the relevant information from DHCP or a
 string of the form NFSSERVER:NFSPATH or NFSSERVER:NFSPATH:NFSOPTS.
diff --git a/scripts/functions b/scripts/functions
index 8c1bb1f..2ed3ce3 100644
--- a/scripts/functions
+++ b/scripts/functions
@@ -426,6 +426,42 @@ mountfs()
${type}_mount_fs $1
 }
 
+# Mount a loopback device, which is present on the mounted filesystem.
+mount_loop_root()
+{
+   mkdir -p /host
+   mount -o move ${rootmnt} /host
+   loopfile=/host/${LOOP#/}
+
+   while [ ! -e $loopfile ]; do
+   panic ALERT! $loopfile does not exist.  Dropping to a shell!
+   done
+
+   if [ ${readonly} = y ]; then
+   roflag=-r
+   else
+   roflag=-w
+   fi
+
+   # Get the loop filesystem type if not set
+   if [ -z ${LOOPFSTYPE} ]; then
+   FSTYPE=$(get_fstype $loopfile)
+   else
+   FSTYPE=${LOOPFSTYPE}
+   fi
+
+   # FIXME This has no error checking
+   modprobe loop
+   modprobe ${FSTYPE}
+
+   # FIXME This has no error checking
+   mount ${roflag} -o loop -t ${FSTYPE} ${LOOPFLAGS} $loopfile ${rootmnt}
+
+   if [ -d ${rootmnt}/host ]; then
+   mount -o move /host ${rootmnt}/host
+   fi
+}
+
 # Mount the root file system.  It should be overridden by all
 # boot scripts.
 mountroot()
diff --git a/scripts/local b/scripts/local
index f6424f0..072013e 100644
--- a/scripts/local
+++ b/scripts/local
@@ -135,7 +135,8 @@ local_mount_root()
 
ROOT=$(resolve_device $ROOT)
 
-   if [ ${readonly} = y ]; then
+   if [ ${readonly} = y ]  \
+  ([ -z $LOOP ] || [ ${FSTYPE#ntfs} = $FSTYPE ]); then
roflag=-r
else
roflag=-w
@@ -153,6 +154,10 @@ local_mount_root()
else
mount ${roflag} ${ROOTFLAGS} ${ROOT} ${rootmnt}
fi
+
+   if [ ${LOOP} ]; then
+   mount_loop_root
+   fi
 }
 
 local_mount_fs()
diff --git a/scripts/nfs b/scripts/nfs
index 1c29850..d382413 100644
--- a/scripts/nfs
+++ b/scripts/nfs
@@ -72,6 +72,10 @@ nfs_mount_root_impl()
fi
 
nfsmount -o nolock ${roflag} ${NFSOPTS} ${NFSROOT} ${rootmnt}
+
+   if [ ${LOOP} ]; then
+   mount_loop_root
+   fi
 }
 
 # NFS root mounting


signature.asc
Description: PGP signature

Bug#786925: linux: overlay filesystem incompatible with NFS

[Vagrant Cascadian]

Control: found 4.1.3-1

On 2015-05-26, Vagrant Cascadian wrote:
 LTSP has been using aufs to provide a writeable tmpfs layer on top of
 NFS since wheezy.

 With the switch to the overlay filesystem (after adjusting LTSP's
 initramfs-tools hooks to support overlay), it appears to fail due to an
 incompatibility with NFS. 

I haven't confirmed, but this may be fixed in mainline, not sure how
difficult it would be to backport to 4.1.x:

commit 7c03b5d45b8eebf025053d8fe887cc262ba6
Author: Miklos Szeredi mszer...@suse.cz
Date:   Mon Jun 22 13:53:48 2015 +0200

ovl: allow distributed fs as lower layer

Allow filesystems with .d_revalidate as lower layer(s), but not as upper
layer.

For local filesystems the rule was that modifications on the layers
directly while being part of the overlay results in undefined behavior.

This can easily be extended to distributed filesystems: we assume the tree
used as lower layer is static, which means -d_revalidate() should always
return 1.  If that is not the case, return -ESTALE, don't try to work
around the modification.

Signed-off-by: Miklos Szeredi mszer...@suse.cz

diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c
index de9d2ee..8a08c58 100644
--- a/fs/overlayfs/super.c
+++ b/fs/overlayfs/super.c
@@ -273,10 +273,57 @@ static void ovl_dentry_release(struct dentry *dentry)
}
 }
 
+static int ovl_dentry_revalidate(struct dentry *dentry, unsigned int flags)
+{
+   struct ovl_entry *oe = dentry-d_fsdata;
+   unsigned int i;
+   int ret = 1;
+
+   for (i = 0; i  oe-numlower; i++) {
+   struct dentry *d = oe-lowerstack[i].dentry;
+
+   if (d-d_flags  DCACHE_OP_REVALIDATE) {
+   ret = d-d_op-d_revalidate(d, flags);
+   if (ret  0)
+   return ret;
+   if (!ret) {
+   if (!(flags  LOOKUP_RCU))
+   d_invalidate(d);
+   return -ESTALE;
+   }
+   }
+   }
+   return 1;
+}
+
+static int ovl_dentry_weak_revalidate(struct dentry *dentry, unsigned int 
flags)
+{
+   struct ovl_entry *oe = dentry-d_fsdata;
+   unsigned int i;
+   int ret = 1;
+
+   for (i = 0; i  oe-numlower; i++) {
+   struct dentry *d = oe-lowerstack[i].dentry;
+
+   if (d-d_flags  DCACHE_OP_WEAK_REVALIDATE) {
+   ret = d-d_op-d_weak_revalidate(d, flags);
+   if (ret = 0)
+   break;
+   }
+   }
+   return ret;
+}
+
 static const struct dentry_operations ovl_dentry_operations = {
.d_release = ovl_dentry_release,
 };
 
+static const struct dentry_operations ovl_reval_dentry_operations = {
+   .d_release = ovl_dentry_release,
+   .d_revalidate = ovl_dentry_revalidate,
+   .d_weak_revalidate = ovl_dentry_weak_revalidate,
+};
+
 static struct ovl_entry *ovl_alloc_entry(unsigned int numlower)
 {
size_t size = offsetof(struct ovl_entry, lowerstack[numlower]);
@@ -288,6 +335,20 @@ static struct ovl_entry *ovl_alloc_entry(unsigned int 
numlower)
return oe;
 }
 
+static bool ovl_dentry_remote(struct dentry *dentry)
+{
+   return dentry-d_flags 
+   (DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE);
+}
+
+static bool ovl_dentry_weird(struct dentry *dentry)
+{
+   return dentry-d_flags  (DCACHE_NEED_AUTOMOUNT |
+ DCACHE_MANAGE_TRANSIT |
+ DCACHE_OP_HASH |
+ DCACHE_OP_COMPARE);
+}
+
 static inline struct dentry *ovl_lookup_real(struct dentry *dir,
 struct qstr *name)
 {
@@ -303,9 +364,9 @@ static inline struct dentry *ovl_lookup_real(struct dentry 
*dir,
} else if (!dentry-d_inode) {
dput(dentry);
dentry = NULL;
-   } else if (dentry-d_flags  DCACHE_MANAGED_DENTRY) {
+   } else if (ovl_dentry_weird(dentry)) {
dput(dentry);
-   /* Don't support traversing automounts */
+   /* Don't support traversing automounts and other weirdness */
dentry = ERR_PTR(-EREMOTE);
}
return dentry;
@@ -354,6 +415,11 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry 
*dentry,
goto out;
 
if (this) {
+   if (unlikely(ovl_dentry_remote(this))) {
+   dput(this);
+   err = -EREMOTE;
+   goto out;
+   }
if (ovl_is_whiteout(this)) {
dput(this);
this = NULL;
@@ -698,25 +764,6 @@ static void ovl_unescape(char *s)
}
 }
 
-static

Bug#468114: Loopback file system support.

[Vagrant Cascadian]

I've taken a quick stab as refreshing this, though I haven't yet tested
it. I did move mounting of the loopback to mount_loop_root in functions,
as I would like to eventually support loopback mounted files from NFS.

live well,
  vagrant

commit aab4c43f484fa44b6f35e9bf0b358a70b165bcbd
Author: Vagrant Cascadian vagr...@debian.org
Date:   Tue Aug 11 16:17:09 2015 +0200

Refreshed patch from Luke Yelavich to add loopback support to
initramfs-tools.

diff --git a/init b/init
index abf7f25..2760bcb 100755
--- a/init
+++ b/init
@@ -98,6 +98,15 @@ for x in $(cat /proc/cmdline); do
;;
esac
;;
+   loop=*)
+   LOOP=${x#loop=}
+   ;;
+   loopflags=*)
+   LOOPFLAGS=-o ${x#loopflags=}
+   ;;
+   loopfstype=*)
+   LOOPFSTYPE=${x#loopfstype=}
+   ;;
nfsroot=*)
NFSROOT=${x#nfsroot=}
;;
diff --git a/initramfs-tools.8 b/initramfs-tools.8
index ea8c098..ce8e830 100644
--- a/initramfs-tools.8
+++ b/initramfs-tools.8
@@ -42,6 +42,19 @@ The default is 180 seconds.
 set the file system mount option string.
 
 .TP
+\fB\fI loop
+path within the original root file system to loop-mount and use as the
+real root file system.
+
+.TP
+\fB\fI loopflags
+set the loop file system mount option string, if applicable.
+
+.TP
+\fB\fI loopfstype
+set the loop file system type, if applicable.
+
+.TP
 \fB\fI nfsroot
 can be either auto to try to get the relevant information from DHCP or a
 string of the form NFSSERVER:NFSPATH or NFSSERVER:NFSPATH:NFSOPTS.
diff --git a/scripts/functions b/scripts/functions
index 8c1bb1f..99de1aa 100644
--- a/scripts/functions
+++ b/scripts/functions
@@ -426,6 +426,36 @@ mountfs()
${type}_mount_fs $1
 }
 
+# Mount a loopback device, which is present on the mounted filesystem.
+mount_loop_root()
+{
+   if [ $LOOP ]; then
+   mkdir -p /host
+   mount -o move ${rootmnt} /host
+
+   while [ ! -e /host/${LOOP#/} ]; do
+   panic ALERT!  /host/${LOOP#/} does not exist.  
Dropping to a shell!
+   done
+
+   if [ ${readonly} = y ]; then
+   roflag=-r
+   else
+   roflag=-w
+   fi
+
+   # FIXME This has no error checking
+   modprobe loop
+   modprobe ${FSTYPE}
+
+   # FIXME This has no error checking
+   mount ${roflag} -o loop -t ${FSTYPE} ${LOOPFLAGS} 
/host/${LOOP#/} ${rootmnt}
+
+   if [ -d ${rootmnt}/host ]; then
+   mount -o move /host ${rootmnt}/host
+   fi
+   fi
+}
+
 # Mount the root file system.  It should be overridden by all
 # boot scripts.
 mountroot()
diff --git a/scripts/local b/scripts/local
index f6424f0..94d6ddc 100644
--- a/scripts/local
+++ b/scripts/local
@@ -135,7 +135,8 @@ local_mount_root()
 
ROOT=$(resolve_device $ROOT)
 
-   if [ ${readonly} = y ]; then
+   if [ ${readonly} = y ]  \
+  ([ -z $LOOP ] || [ ${FSTYPE#ntfs} = $FSTYPE ]); then
roflag=-r
else
roflag=-w
@@ -153,6 +154,8 @@ local_mount_root()
else
mount ${roflag} ${ROOTFLAGS} ${ROOT} ${rootmnt}
fi
+
+   mount_loop_root
 }
 
 local_mount_fs()


signature.asc
Description: PGP signature

Bug#468114: Loopback file system support.

[Vagrant Cascadian]

On 2015-08-11, Vagrant Cascadian wrote:
 I've taken a quick stab as refreshing this, though I haven't yet tested
 it. I did move mounting of the loopback to mount_loop_root in functions,
 as I would like to eventually support loopback mounted files from NFS.

Made some small changes, updated and *tested* this time! Patch below...

live well,
  vagrant

diff --git a/init b/init
index abf7f25..2760bcb 100755
--- a/init
+++ b/init
@@ -98,6 +98,15 @@ for x in $(cat /proc/cmdline); do
;;
esac
;;
+   loop=*)
+   LOOP=${x#loop=}
+   ;;
+   loopflags=*)
+   LOOPFLAGS=-o ${x#loopflags=}
+   ;;
+   loopfstype=*)
+   LOOPFSTYPE=${x#loopfstype=}
+   ;;
nfsroot=*)
NFSROOT=${x#nfsroot=}
;;
diff --git a/initramfs-tools.8 b/initramfs-tools.8
index ea8c098..ce8e830 100644
--- a/initramfs-tools.8
+++ b/initramfs-tools.8
@@ -42,6 +42,19 @@ The default is 180 seconds.
 set the file system mount option string.
 
 .TP
+\fB\fI loop
+path within the original root file system to loop-mount and use as the
+real root file system.
+
+.TP
+\fB\fI loopflags
+set the loop file system mount option string, if applicable.
+
+.TP
+\fB\fI loopfstype
+set the loop file system type, if applicable.
+
+.TP
 \fB\fI nfsroot
 can be either auto to try to get the relevant information from DHCP or a
 string of the form NFSSERVER:NFSPATH or NFSSERVER:NFSPATH:NFSOPTS.
diff --git a/scripts/functions b/scripts/functions
index 8c1bb1f..2ed3ce3 100644
--- a/scripts/functions
+++ b/scripts/functions
@@ -426,6 +426,42 @@ mountfs()
${type}_mount_fs $1
 }
 
+# Mount a loopback device, which is present on the mounted filesystem.
+mount_loop_root()
+{
+   mkdir -p /host
+   mount -o move ${rootmnt} /host
+   loopfile=/host/${LOOP#/}
+
+   while [ ! -e $loopfile ]; do
+   panic ALERT! $loopfile does not exist.  Dropping to a shell!
+   done
+
+   if [ ${readonly} = y ]; then
+   roflag=-r
+   else
+   roflag=-w
+   fi
+
+   # Get the loop filesystem type if not set
+   if [ -z ${LOOPFSTYPE} ]; then
+   FSTYPE=$(get_fstype $loopfile)
+   else
+   FSTYPE=${LOOPFSTYPE}
+   fi
+
+   # FIXME This has no error checking
+   modprobe loop
+   modprobe ${FSTYPE}
+
+   # FIXME This has no error checking
+   mount ${roflag} -o loop -t ${FSTYPE} ${LOOPFLAGS} $loopfile ${rootmnt}
+
+   if [ -d ${rootmnt}/host ]; then
+   mount -o move /host ${rootmnt}/host
+   fi
+}
+
 # Mount the root file system.  It should be overridden by all
 # boot scripts.
 mountroot()
diff --git a/scripts/local b/scripts/local
index f6424f0..072013e 100644
--- a/scripts/local
+++ b/scripts/local
@@ -135,7 +135,8 @@ local_mount_root()
 
ROOT=$(resolve_device $ROOT)
 
-   if [ ${readonly} = y ]; then
+   if [ ${readonly} = y ]  \
+  ([ -z $LOOP ] || [ ${FSTYPE#ntfs} = $FSTYPE ]); then
roflag=-r
else
roflag=-w
@@ -153,6 +154,10 @@ local_mount_root()
else
mount ${roflag} ${ROOTFLAGS} ${ROOT} ${rootmnt}
fi
+
+   if [ ${LOOP} ]; then
+   mount_loop_root
+   fi
 }
 
 local_mount_fs()



signature.asc
Description: PGP signature

Bug#787029: linux: [armhf] Enable PCIe support for IMX6 boards

[Vagrant Cascadian]

Package: linux
Version: 4.0.2-1
Severity: wishlist
Tags: patch

Please enable PCIe support for IMX6 boards, such as the
hummingboard-i2ex.

diff --git a/config/armhf/config.armmp b/config/armhf/config.armmp
index cd0b3aa..1c5b7a2 100644
--- a/config/armhf/config.armmp
+++ b/config/armhf/config.armmp
@@ -642,6 +642,7 @@ CONFIG_WLCORE_SDIO=m
 ## file: drivers/pci/host/Kconfig
 ##
 CONFIG_PCI_MVEBU=y
+CONFIG_PCI_IMX6=y
 CONFIG_PCI_TEGRA=y
 
 ##


live well,
  vagrant


signature.asc
Description: PGP signature

Bug#786925: linux: overlay filesystem incompatible with NFS

[Vagrant Cascadian]

Package: src:linux
Version: 4.0.2-1
Severity: normal
Control: affects -1 ltsp-client-core

LTSP has been using aufs to provide a writeable tmpfs layer on top of
NFS since wheezy.

With the switch to the overlay filesystem (after adjusting LTSP's
initramfs-tools hooks to support overlay), it appears to fail due to an
incompatibility with NFS. The following is from a failed initramfs-tools
boot running the commands manually:

  BusyBox v1.22.1 (Debian 1:1.22.0-15) built-in shell (ash)
  Enter 'help' for a list of built-in commands.
  
  sh: can't access tty; job control turned off
  / # mount
  rootfs on / type rootfs (rw)
  sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
  proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
  udev on /dev type devtmpfs (rw,relatime,size=10240k,nr_inodes=61069,mode=755)
  devpts on /dev/pts type devpts 
(rw,nosuid,noexec,relatime,gid=5,mode=620,ptmxmod
  e=000)
  tmpfs on /run type tmpfs (rw,nosuid,relatime,size=101168k,mode=755)
  192.168.67.1:/opt/ltsp/amd64 on /rofs type nfs 
(ro,relatime,vers=3,rsize=524288,
  
wsize=524288,namlen=255,hard,nolock,proto=tcp,port=2049,timeo=7,retrans=10,sec=s
  ys,local_lock=all,addr=192.168.67.1)
  tmpfs on /cow type tmpfs (rw,relatime,mode=755)
  / # mount -t overlay -o upperdir=/cow/up,lowerdir=/rofs,workdir=/cow/work 
overla
  y /root
  [   49.944625] overlayfs: filesystem on '/rofs' not supported
  mount: mounting overlay on /root failed:  Invalid argument


It works fine when using a loopback mounted squashfs image over NFS, or
with NBD+squashfs, but mounting NFS as a lowerdir directly appears to
fail.


This may be the same issue as mentioned here:

  https://lkml.org/lkml/2014/10/30/242

Though that hasn't received much response...


live well,
  vagrant


-- Package-specific info:
** Kernel log: boot messages should be attached

** Model information
not available

** PCI devices:
not available

** USB devices:
not available


-- System Information:
Debian Release: stretch/sid
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 3.16.0-4-amd64
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: unable to detect

Versions of packages linux-image-4.0.0-1-amd64 depends on:
ii  debconf [debconf-2.0]   1.5.56
ii  initramfs-tools [linux-initramfs-tool]  0.120
ii  kmod20-1
ii  linux-base  3.5

Versions of packages linux-image-4.0.0-1-amd64 recommends:
ii  firmware-linux-free  3.3
ii  irqbalance   1.0.6-3

Versions of packages linux-image-4.0.0-1-amd64 suggests:
pn  debian-kernel-handbook none
pn  grub-pc | grub-efi | extlinux  none
pn  linux-doc-4.0  none

Versions of packages linux-image-4.0.0-1-amd64 is related to:
pn  firmware-atherosnone
pn  firmware-bnx2   none
pn  firmware-bnx2x  none
pn  firmware-brcm80211  none
pn  firmware-intelwimax none
pn  firmware-ipw2x00none
pn  firmware-ivtv   none
pn  firmware-iwlwifinone
pn  firmware-libertas   none
pn  firmware-linux  none
pn  firmware-linux-nonfree  none
pn  firmware-myricomnone
pn  firmware-netxen none
pn  firmware-qlogic none
pn  firmware-ralink none
pn  firmware-realteknone
pn  xen-hypervisor  none

-- debconf information excluded


signature.asc
Description: PGP signature

Bug#783320: linux: FTBFS on armhf, imx-hdmi module renamed to dw_hdmi-imx

[Vagrant Cascadian]

Source: linux
Version: 4.0-1~exp1
Severity: important
Tags: Patch

From buildd log:

  kernel-wedge copy-modules 4.0.0-trunk armmp 4.0.0-trunk-armmp
  missing module imx-hdmi
  command exited with status 1
  make[2]: *** [install-udeb_armhf] Error 2

diff --git a/installer/armhf/modules/armhf-armmp/fb-modules 
b/installer/armhf/modules/armhf-armmp/fb-modules
index 03d6305..849ebe2 100644
--- a/installer/armhf/modules/armhf-armmp/fb-modules
+++ b/installer/armhf/modules/armhf-armmp/fb-modules
@@ -1,3 +1,3 @@
 imx-ipuv3-crtc
-imx-hdmi
+dw_hdmi-imx
 tegra-drm


live well,
  vagrant


signature.asc
Description: PGP signature

Bug#778783: src:linux: [armhf] Enable CONFIG_BACKLIGHT_GPIO for armmp

[Vagrant Cascadian]

Package: src:linux
Version: 3.16.7-ckt4-3
Severity: wishlist
Tags: patch

Please consider enabling CONFIG_BACKLIGHT_GPIO as a module in the
armmp kernel.

This module enables support for the BeagleBone Black with the LCD7
cape (although still requires a custom .dtb). It may also be useful to
other ARM boards.

diff --git a/config/armhf/config.armmp b/config/armhf/config.armmp
index 62e512f..03302a0 100644
--- a/config/armhf/config.armmp
+++ b/config/armhf/config.armmp
@@ -832,6 +832,11 @@ CONFIG_TWL6030_USB=m
 CONFIG_USB_MXS_PHY=m
 
 ##
+## file: drivers/video/backlight/Kconfig
+##
+CONFIG_BACKLIGHT_GPIO=m
+
+##
 ## file: drivers/video/console/Kconfig
 ##
 CONFIG_FRAMEBUFFER_CONSOLE=y


Thanks!


live well,
  vagrant


signature.asc
Description: PGP signature

Re: [Raspbian-devel] Proper way for vendors to build deb packages of kernels.

[Vagrant Cascadian]

On 2014-12-07, Ian Campbell wrote:
 On Sun, 2014-12-07 at 11:15 +0100, Uwe Kleine-König wrote:
 On Sat, Dec 06, 2014 at 12:24:55AM +, peter green wrote:
... 
  Currently i'm aware of three ways of building deb packaged kernels.
  
  1: modify the Debian linux source package
  2: use make deb-pkg
  3: use make-kpkg
 
  Option 1 can be made to work and probablly gives the closest
  experiance to kernels actually from Debian but it's a PITA,
  especially if there are a large number of changes from the source
  tree Debian uses. We do produce kernel package from a mashup of the
  Debian linux source package and the raspberry pi foundation's git
  tree but they are a pain to update and so tend to be updated far
  less frequently than the raspberry pi foundation's kernels.

 Once you worked a bit with the kernel source package it's quite nice.
 OK, it's a big and complicated package because it generates a big amount
 of architecture specific packages, but for the complexity it handles
 it's really nice (IMHO).
 
 I'd go for this option,

 FWIW so would I. I'd expect 90% of the work it be just replacing
 debian/patches and debian/config with Raspbian specific stuff (where
 debian/patches might be nearly empty or one big megapatch resynching
 with the Pi-foundations releases, depending on how that upstream works).

Last year I helped with the initial work to get that going for the 3.2
kernel, based off of the kernel source package in wheezy. But even
though the patchset from the raspberry pi foundation side of things for
3.2.x wasn't changing much (if at all), the security and point release
updates requires a fair bit of work to keep in sync.

I'm not sure how ugly the diff is with newer series kernels, I'd like to
think a platform as mature as the raspberry pi would be focused on
resolving mainline support by now. Some patches have definitely made
their way to mainline, identifying which are left and targeting them,
and/or building kernels without *all* the features needed might help
reduce the maintenance burden.

The only long-term sanity-saving approach really seems to be
mainlining the needed patches...


live well,
  vagrant


signature.asc
Description: PGP signature

Bug#770635: linux: [armhf/armmp] Add udeb modules to support video and keyboard for imx6

[Vagrant Cascadian]

Package: linux
Version: 3.16.7-2
Severity: normal
Tags: d-i patch

Debian-installer hd-media image on armhf doesn't work with the hdmi
output and usb keyboard, though these work fine from an installed
system.

I believe adding the following modules to the fb-modules and
usb-modules udebs should allow this to work for hd-media (and possibly
netboot as well).

I've tested it to work by manually appending these modules to the
hd-media initrd, and it seemed to work on both a wandboard quad and
cubox-i4pro.

I've tried creating an installer image with my custom kernel and udebs
to test with greater assurance, using the instructions found at:

  https://wiki.debian.org/DebianInstaller/Modify/CustomKernel

Unfortunately debian-installer seemed to build with the udebs
currently shipped in Jessie rather than the locally build udebs,
though am unsure if there's an easier way to force it to use the udebs
From localudebs.  I think it will just work once these modules are
actually in sid/jessie...


diff --git a/installer/armhf/modules/armhf-armmp/fb-modules 
b/installer/armhf/modules/armhf-armmp/fb-modules
new file mode 100644
index 000..29bdd07
--- /dev/null
+++ b/installer/armhf/modules/armhf-armmp/fb-modules
@@ -0,0 +1,2 @@
+imx-ipuv3-crtc
+imx-hdmi
diff --git a/installer/armhf/modules/armhf-armmp/usb-modules 
b/installer/armhf/modules/armhf-armmp/usb-modules
index c0f4e1b..38f91fe 100644
--- a/installer/armhf/modules/armhf-armmp/usb-modules
+++ b/installer/armhf/modules/armhf-armmp/usb-modules
@@ -5,3 +5,4 @@ ohci-exynos
 ehci-exynos
 phy-exynos-usb2
 ci_hdrc_imx
+phy-mxs-usb


Thanks for your work!

live well,
  vagrant


signature.asc
Description: PGP signature

Bug#770635: linux: [armhf/armmp] Add udeb modules to support video and keyboard for imx6

[Vagrant Cascadian]

On 2014-11-22, Vagrant Cascadian wrote:
 Debian-installer hd-media image on armhf doesn't work with the hdmi
 output and usb keyboard, though these work fine from an installed
 system.

 I believe adding the following modules to the fb-modules and
 usb-modules udebs should allow this to work for hd-media (and possibly
 netboot as well).
...
 Unfortunately debian-installer seemed to build with the udebs
 currently shipped in Jessie

I figured out a fix for that (was a corner case of localudebs + custom
sources.list)... and it works!

Please consider applying the patch, which enables console installs on
wandboard, cubox-i and probably several other imx6 based devices.

live well,
  vagrant


signature.asc
Description: PGP signature

Bug#761614: /usr/bin/linux-version: sorts -trunk as newer than -1 version

[Vagrant Cascadian]

Package: linux-base
Version: 3.5
Severity: normal
File: /usr/bin/linux-version

It seems like the -trunk versions, typically from experimental, are
sorted higher than the -1 version, which is often after being uploaded
to unstable, and often newer than the -trunk versions...

  $ linux-version sort 3.16-trunk-armmp 3.16-rc6-armmp 3.16-1-armmp 3.16-2-armmp
  3.16-rc6-armmp
  3.16-1-armmp
  3.16-2-armmp
  3.16-trunk-armmp

Seems like trunk should be between the rc versions and numbered
versions... unless I'm misunderstanding something.

Thanks!

live well,
  vagrant


pgpR4K3QV9I0v.pgp
Description: PGP signature

Bug#756810: linux-image-3.16-rc6-armmp: HDMI on wandboard/cubox-i no longer works

[Vagrant Cascadian]

On 2014-08-11, Vagrant Cascadian wrote:
 I think this is caused by dropping CONFIG_DRM_IMX_IPUV3_CORE=m, which
 was moved out of staging into drivers/gpu/ipu-v3/Kconfig.

And the configuration option was also renamed to
CONFIG_IMX_IPUV3_CORE...

I've tested that the following patch on a Cubox-i works:

diff --git a/config/armhf/config.armmp b/config/armhf/config.armmp
index 2e9343e..8c2ce84 100644
--- a/config/armhf/config.armmp
+++ b/config/armhf/config.armmp
@@ -191,6 +191,11 @@ CONFIG_DRM_I2C_NXP_TDA998X=m
 CONFIG_DRM_TILCDC=m

 ##
+## file: drivers/gpu/ipu-v3/Kconfig
+##
+CONFIG_IMX_IPUV3_CORE=m
+
+##
 ## file: drivers/hwspinlock/Kconfig
 ##
 CONFIG_HWSPINLOCK_OMAP=m


live well,
  vagrant


pgp6ibGM9q8nW.pgp
Description: PGP signature

Bug#756810: linux-image-3.16-rc6-armmp: HDMI on wandboard/cubox-i no longer works

[Vagrant Cascadian]

Control: tags -1 patch

On 2014-08-01, Vagrant Cascadian wrote:
 With linux-image-3.15-trunk-armmp 3.15.5-1~exp1 which included patches applied
 from upstream, HDMI video output worked for me on both the Cubox i4pro and
 Wandboard Quad. The console framebuffer showed output, X with fbdev worked...

 Now with 3.16~rc6-1~exp1, which shold have the applied patches, there's no
 output. Perhaps other additional patches caused issues.

I think this is caused by dropping CONFIG_DRM_IMX_IPUV3_CORE=m, which
was moved out of staging into drivers/gpu/ipu-v3/Kconfig.

The following patch will hopefully address this:

diff --git a/config/armhf/config.armmp b/config/armhf/config.armmp
index 2e9343e..d0a6d40 100644
--- a/config/armhf/config.armmp
+++ b/config/armhf/config.armmp
@@ -191,6 +191,11 @@ CONFIG_DRM_I2C_NXP_TDA998X=m
 CONFIG_DRM_TILCDC=m

 ##
+## file: drivers/gpu/ipu-v3/Kconfig
+##
+CONFIG_DRM_IMX_IPUV3_CORE=m
+
+##
 ## file: drivers/hwspinlock/Kconfig
 ##
 CONFIG_HWSPINLOCK_OMAP=m


Currently building a kernel to test the fix...

live well,
  vagrant


pgpW8Ib4sUqxm.pgp
Description: PGP signature

Bug#756810: linux-image-3.16-rc6-armmp: HDMI on wandboard/cubox-i no longer works

[Vagrant Cascadian]

Package: src:linux
Version: 3.16~rc6-1~exp1
Severity: normal

With linux-image-3.15-trunk-armmp 3.15.5-1~exp1 which included patches applied
from upstream, HDMI video output worked for me on both the Cubox i4pro and
Wandboard Quad. The console framebuffer showed output, X with fbdev worked...

Now with 3.16~rc6-1~exp1, which shold have the applied patches, there's no
output. Perhaps other additional patches caused issues.

I also tested with 3.16~rc5-1~exp1, no luck.

hrm.

Thanks for working on Debian kernel packaging!

live well,
  vagrant

-- Package-specific info:
** Version:
Linux version 3.16-rc6-armmp (debian-kernel@lists.debian.org) (gcc version 
4.8.3 (Debian 4.8.3-5) ) #1 SMP Debian 3.16~rc6-1~exp1 (2014-07-21)

** Command line:
console=ttymxc0,115200

** Tainted: C (1024)
 * Module from drivers/staging has been loaded.

** Kernel log:
[1.384242] mmc0: new high speed SDHC card at address 1234
[1.390289] mmcblk0: mmc0:1234 SA04G 3.63 GiB 
[1.398676]  mmcblk0: p1 p2 p3  p5 
[1.802878] PM: Starting manual resume from disk
[1.807583] PM: Hibernation image partition 179:5 present
[1.807592] PM: Looking for hibernation image.
[1.809096] PM: Image not found (code -22)
[1.809106] PM: Hibernation image not present or could not be loaded.
[1.860617] EXT4-fs (mmcblk0p2): mounted filesystem with ordered data mode. 
Opts: (null)
[3.002967] systemd-udevd[260]: starting version 208
[3.354410] Registered IR keymap rc-empty
[3.359051] input: gpio_ir_recv as /devices/soc0/ir-receiver/rc/rc0/input0
[3.366521] rc0: gpio_ir_recv as /devices/soc0/ir-receiver/rc/rc0
[3.374681] imx2-wdt 20bc000.wdog: timeout 60 sec (nowayout=0)
[3.398170] i2c i2c-1: IMX I2C adapter registered
[3.413338] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
[3.453949] [drm] Initialized drm 1.1.0 20060810
[3.459654] SCSI subsystem initialized
[3.484164] i2c i2c-2: IMX I2C adapter registered
[3.511686] libata version 3.00 loaded.
[3.542674] ahci-imx 220.sata: SSS flag set, parallel bus scan disabled
[3.549529] imxdrm: module is from the staging directory, the quality is 
unknown, you have been warned.
[3.549534] imxdrm: module is from the staging directory, the quality is 
unknown, you have been warned.
[3.556139] IR Sony protocol handler initialized
[3.556148] IR SANYO protocol handler initialized
[3.562674] lirc_dev: IR Remote Control driver registered, major 243 
[3.568302] rc rc0: lirc_dev: driver ir-lirc-codec (gpio-rc-recv) registered 
at minor = 0
[3.568306] IR LIRC bridge handler initialized
[3.574839] IR NEC protocol handler initialized
[3.579150] IR JVC protocol handler initialized
[3.584020] input: MCE IR Keyboard/Mouse (gpio-rc-recv) as 
/devices/virtual/input/input1
[3.584623] IR MCE Keyboard/mouse protocol handler initialized
[3.587476] imx_hdmi: module is from the staging directory, the quality is 
unknown, you have been warned.
[3.592797] ci_hdrc ci_hdrc.0: ChipIdea HDRC found, lpm: 0; cap: f00c6100 
op: f00c6140
[3.597022] ci_hdrc ci_hdrc.0: It is OTG capable controller
[3.597522] ci_hdrc ci_hdrc.0: EHCI Host Controller
[3.597781] ci_hdrc ci_hdrc.0: new USB bus registered, assigned bus number 1
[3.606796] ci_hdrc ci_hdrc.0: USB 2.0 started, EHCI 1.00
[3.607067] usb usb1: New USB device found, idVendor=1d6b, idProduct=0002
[3.607074] usb usb1: New USB device strings: Mfr=3, Product=2, 
SerialNumber=1
[3.607082] usb usb1: Product: EHCI Host Controller
[3.607087] usb usb1: Manufacturer: Linux 3.16-rc6-armmp ehci_hcd
[3.607092] usb usb1: SerialNumber: ci_hdrc.0
[3.607779] hub 1-0:1.0: USB hub found
[3.607818] hub 1-0:1.0: 1 port detected
[3.608715] ci_hdrc ci_hdrc.1: ChipIdea HDRC found, lpm: 0; cap: f00ee300 
op: f00ee340
[3.611702] ci_hdrc ci_hdrc.1: doesn't support gadget
[3.611712] ci_hdrc ci_hdrc.1: EHCI Host Controller
[3.611956] ci_hdrc ci_hdrc.1: new USB bus registered, assigned bus number 2
[3.622801] ci_hdrc ci_hdrc.1: USB 2.0 started, EHCI 1.00
[3.623048] usb usb2: New USB device found, idVendor=1d6b, idProduct=0002
[3.623056] usb usb2: New USB device strings: Mfr=3, Product=2, 
SerialNumber=1
[3.623062] usb usb2: Product: EHCI Host Controller
[3.623069] usb usb2: Manufacturer: Linux 3.16-rc6-armmp ehci_hcd
[3.623074] usb usb2: SerialNumber: ci_hdrc.1
[3.623758] hub 2-0:1.0: USB hub found
[3.623801] hub 2-0:1.0: 1 port detected
[3.651967] imx-spdif sound-spdif: ASoC: CPU DAI (null) not registered
[3.651998] imx-spdif sound-spdif: snd_soc_register_card failed: -517
[3.652015] platform sound-spdif: Driver imx-spdif requests probe deferral
[3.666015] imx-spdif sound-spdif: snd-soc-dummy-dai - 2004000.spdif 
mapping ok
[3.770603] ahci-imx 220.sata: AHCI 0001.0300 32 slots 1 ports 3 Gbps 
0x1 impl platform mode
[3.771291] IR RC5(x) protocol handler initialized

Bug#754491: armmp: Include net/mmc modules in udebs for BeagleBone Black

[Vagrant Cascadian]

Source: linux
Version: 3.14.9-1
Severity: normal
Tags: patch

Please enable the following modules in the armmp udebs, which are needed to use
debian-installer the BeagleBone Black:

diff --git a/installer/armhf/modules/armhf-armmp/mmc-modules 
b/installer/armhf/modules/armhf-armmp/mmc-modules
index 70e14e8..c6e6e9b 100644
--- a/installer/armhf/modules/armhf-armmp/mmc-modules
+++ b/installer/armhf/modules/armhf-armmp/mmc-modules
@@ -1,3 +1,4 @@
 #include mmc-modules
 sdhci-esdhc-imx
 mmci
+omap-hsmmc
diff --git a/installer/armhf/modules/armhf-armmp/nic-modules 
b/installer/armhf/modules/armhf-armmp/nic-modules
index 3b3f35b..5c22356 100644
--- a/installer/armhf/modules/armhf-armmp/nic-modules
+++ b/installer/armhf/modules/armhf-armmp/nic-modules
@@ -5,3 +5,6 @@ smsc911x
 stmmac
 sun4i-emac
 xgmac
+ti_cpsw
+davinci_mdio
+davinci_cpdma


live well,
  vagrant


signature.asc
Description: Digital signature

Bug#750406: Enable IMX DRM/HDMI in Wandboard and Cubox-i DTS's

[Vagrant Cascadian]

On Tue, Jun 03, 2014 at 08:35:17AM +0100, Ian Campbell wrote:
  I tested both patches with a local build of 3.15~rc7-1~exp2 on a wandboard 
  quad
  and a cubox i4pro, and they seem to work for a fbdev console and X at 
  16-bit,
  at least.
  
  If that's considered sufficiently upstream, It would be nice to apply them.
 
 I think that should be fine, I've copied submit@ to create new bug for
 this bit (original, closed, bugreport on bcc).

Just re-tested with a build of 3.15~rc8, still seems to work on Wandboard Quad
and Cubox i4pro. 

The upstream commit also modifies the Hummingboard dts, but I
was not able to test that so left it out of the patch (Suppose I should
rename the patch appropriately, too, then. hrm):

commit ac8aba3531d89aecde1a5f635634e7b87c3615ba
Author: Vagrant Cascadian vagr...@debian.org
Date:   Thu Jun 5 12:24:44 2014 -0700

Add patches to enable HDMI on wandboard/cubox-i.

diff --git 
a/patches/features/arm/ARM-dts-imx6qdl-wandboard-Add-HDMI-support.patch 
b/patches/features/arm/ARM-dts-imx6qdl-wandboard-Add-HDMI-support.patch
new file mode 100644
index 000..28f41c5
--- /dev/null
+++ b/patches/features/arm/ARM-dts-imx6qdl-wandboard-Add-HDMI-support.patch
@@ -0,0 +1,51 @@
+From fed687c526a4e93a9605f779537bf654cda1a36f Mon Sep 17 00:00:00 2001
+From: Fabio Estevam fabio.este...@freescale.com
+Date: Tue, 22 Apr 2014 11:26:22 -0300
+Subject: [PATCH 1/2] ARM: dts: imx6qdl-wandboard: Add HDMI support
+
+Signed-off-by: Fabio Estevam fabio.este...@freescale.com
+Signed-off-by: Shawn Guo shawn@freescale.com
+---
+ arch/arm/boot/dts/imx6qdl-wandboard.dtsi | 19 +++
+ 1 file changed, 19 insertions(+)
+
+diff --git a/arch/arm/boot/dts/imx6qdl-wandboard.dtsi 
b/arch/arm/boot/dts/imx6qdl-wandboard.dtsi
+index bdfdf89..5c6f10c 100644
+--- a/arch/arm/boot/dts/imx6qdl-wandboard.dtsi
 b/arch/arm/boot/dts/imx6qdl-wandboard.dtsi
+@@ -62,6 +62,18 @@
+   status = okay;
+ };
+ 
++hdmi {
++  ddc-i2c-bus = i2c1;
++  status = okay;
++};
++
++i2c1 {
++  clock-frequency = 10;
++  pinctrl-names = default;
++  pinctrl-0 = pinctrl_i2c1;
++  status = okay;
++};
++
+ i2c2 {
+   clock-frequency = 10;
+   pinctrl-names = default;
+@@ -127,6 +139,13 @@
+   ;
+   };
+ 
++  pinctrl_i2c1: i2c1grp {
++  fsl,pins = 
++  MX6QDL_PAD_EIM_D21__I2C1_SCL
0x4001b8b1
++  MX6QDL_PAD_EIM_D28__I2C1_SDA
0x4001b8b1
++  ;
++  };
++
+   pinctrl_i2c2: i2c2grp {
+   fsl,pins = 
+   MX6QDL_PAD_KEY_COL3__I2C2_SCL   
0x4001b8b1
+-- 
+2.0.0.rc2
+
diff --git 
a/patches/features/arm/ARM-imx-add-HDMI-support-for-SolidRun-HummingBoard-a.patch
 
b/patches/features/arm/ARM-imx-add-HDMI-support-for-SolidRun-HummingBoard-a.patch
new file mode 100644
index 000..1ad986b
--- /dev/null
+++ 
b/patches/features/arm/ARM-imx-add-HDMI-support-for-SolidRun-HummingBoard-a.patch
@@ -0,0 +1,62 @@
+From f34d0d5dd86e794477c8d23622111f9709085041 Mon Sep 17 00:00:00 2001
+From: Russell King rmk+ker...@arm.linux.org.uk
+Date: Fri, 18 Apr 2014 11:01:27 +0100
+Subject: [PATCH 2/2] ARM: imx: add HDMI support for SolidRun HummingBoard and
+ Cubox-i
+
+Add the HDMI DT configuration for the SolidRun HummingBoard and Cubox-i.
+
+Signed-off-by: Russell King rmk+ker...@arm.linux.org.uk
+Signed-off-by: Shawn Guo shawn@freescale.com
+---
+ arch/arm/boot/dts/imx6qdl-cubox-i.dtsi | 27 +++
+ 1 file changed, 27 insertions(+)
+
+diff --git a/arch/arm/boot/dts/imx6qdl-cubox-i.dtsi 
b/arch/arm/boot/dts/imx6qdl-cubox-i.dtsi
+index c2a2488..25da82a 100644
+--- a/arch/arm/boot/dts/imx6qdl-cubox-i.dtsi
 b/arch/arm/boot/dts/imx6qdl-cubox-i.dtsi
+@@ -55,6 +55,20 @@
+   };
+ };
+ 
++hdmi {
++  pinctrl-names = default;
++  pinctrl-0 = pinctrl_cubox_i_hdmi;
++  ddc-i2c-bus = i2c2;
++  status = okay;
++};
++
++i2c2 {
++  clock-frequency = 10;
++  pinctrl-names = default;
++  pinctrl-0 = pinctrl_cubox_i_i2c2;
++  status = okay;
++};
++
+ i2c3 {
+   pinctrl-names = default;
+   pinctrl-0 = pinctrl_cubox_i_i2c3;
+@@ -69,6 +83,19 @@
+ 
+ iomuxc {
+   cubox_i {
++  pinctrl_cubox_i_hdmi: cubox-i-hdmi {
++  fsl,pins = 
++  MX6QDL_PAD_KEY_ROW2__HDMI_TX_CEC_LINE 0x1f8b0
++  ;
++  };
++
++  pinctrl_cubox_i_i2c2: cubox-i-i2c2 {
++  fsl,pins = 
++  MX6QDL_PAD_KEY_COL3__I2C2_SCL 0x4001b8b1
++  MX6QDL_PAD_KEY_ROW3__I2C2_SDA 0x4001b8b1
++  ;
++  };
++
+   pinctrl_cubox_i_i2c3: cubox-i-i2c3 {
+   fsl,pins = 
+   MX6QDL_PAD_EIM_D17__I2C3_SCL

Bug#748406: linux: [armhf/armmp] Include IMX DRM/HDMI drivers

[Vagrant Cascadian]

On Mon, May 26, 2014 at 03:25:28PM +0100, Ian Campbell wrote:
 On Fri, 2014-05-16 at 15:26 -0700, Vagrant Cascadian wrote:
  The patch for wandboard was from linux-next, the cubox-i/hummingboard patch
  from: 
  

  http://www.home.arm.linux.org.uk/~rmk/cubox/hummingboard-cubox-i-v3.15-rc1/0214-ARM-imx-add-HDMI-support-for-SolidRun-HummingBoard-a.patch
 
 I'm not sure, but I think our policy of only taking changes which are
 already would/should extend to DTBs too. Is this stuff destined to go
 upstream?

There are now dts patches for both wandboard and cubox-i on linus's next
branch, fwiw:

  git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git


commit f34d0d5dd86e794477c8d23622111f9709085041
Author: Russell King rmk+ker...@arm.linux.org.uk
Date:   Fri Apr 18 11:01:27 2014 +0100

ARM: imx: add HDMI support for SolidRun HummingBoard and Cubox-i

Add the HDMI DT configuration for the SolidRun HummingBoard and Cubox-i.

Signed-off-by: Russell King rmk+ker...@arm.linux.org.uk
Signed-off-by: Shawn Guo shawn@freescale.com

 arch/arm/boot/dts/imx6qdl-cubox-i.dtsi | 27 +++
 1 file changed, 27 insertions(+)

commit fed687c526a4e93a9605f779537bf654cda1a36f
Author: Fabio Estevam fabio.este...@freescale.com
Date:   Tue Apr 22 11:26:22 2014 -0300

ARM: dts: imx6qdl-wandboard: Add HDMI support

Signed-off-by: Fabio Estevam fabio.este...@freescale.com
Signed-off-by: Shawn Guo shawn@freescale.com

 arch/arm/boot/dts/imx6qdl-wandboard.dtsi | 19 +++
 1 file changed, 19 insertions(+)


I tested both patches with a local build of 3.15~rc7-1~exp2 on a wandboard quad
and a cubox i4pro, and they seem to work for a fbdev console and X at 16-bit,
at least.

If that's considered sufficiently upstream, It would be nice to apply them.

Thanks!


live well,
  vagrant


signature.asc
Description: Digital signature

Bug#748890: linux: [armhf/armmp] Support SPDIF audio on Wandboard and Cubox-i

[Vagrant Cascadian]

Source: linux
Version: 3.14.4-1
Severity: wishlist
Tags: patch

Both the Wandboard and Cubox-i support SPDIF audio, and it's enabled in the
corresponding .dts files. The armmp kernel just needs the kernel module
enabled:

diff --git a/config/armhf/config.armmp b/config/armhf/config.armmp
index f9d65d9..6142e60 100644
--- a/config/armhf/config.armmp
+++ b/config/armhf/config.armmp
@@ -785,6 +785,7 @@ CONFIG_SND_IMX_SOC=m
 CONFIG_SND_SOC_EUKREA_TLV320=m
 CONFIG_SND_SOC_IMX_SGTL5000=m
 CONFIG_SND_SOC_IMX_MC13783=m
+CONFIG_SND_SOC_IMX_SPDIF=m
 
 ##
 ## file: sound/soc/omap/Kconfig


Thanks!


live well,
  vagrant


signature.asc
Description: Digital signature

Bug#748406: linux: [armhf/armmp] Include IMX DRM/HDMI drivers

[Vagrant Cascadian]

Source: linux
Version: 3.15~rc5-1~exp1
Severity: wishlist
Tags: patch

Please apply the following patches, which enable HDMI video output on the
cubox-i (tested) and untested on wandboard and hummingboard. 

The patch for wandboard was from linux-next, the cubox-i/hummingboard patch
from: 

  
http://www.home.arm.linux.org.uk/~rmk/cubox/hummingboard-cubox-i-v3.15-rc1/0214-ARM-imx-add-HDMI-support-for-SolidRun-HummingBoard-a.patch

At the very least, please consider the config.armmp changes, as this would
allow users to use a custom .dtb file without having to rebuild the kernel.

I tried patches for 3.14, but the diff was considerably larger. So I've tested
using local builds of 3.15~rc5 with the debian packaging from svn.

Thanks!

live well,
  vagrant

diff --git a/config/armhf/config.armmp b/config/armhf/config.armmp
index f9d65d9..b692d4b 100644
--- a/config/armhf/config.armmp
+++ b/config/armhf/config.armmp
@@ -209,6 +209,16 @@ CONFIG_DRM_TILCDC=m
 CONFIG_HWSPINLOCK_OMAP=m
 
 ##
+## file: drivers/staging/imx-drm/Kconfig
+##
+CONFIG_DRM_IMX=m
+CONFIG_DRM_IMX_FB_HELPER=m
+CONFIG_DRM_IMX_LDB=m
+CONFIG_DRM_IMX_HDMI=m
+CONFIG_DRM_IMX_IPUV3_CORE=m
+CONFIG_DRM_IMX_IPUV3=m
+
+##
 ## file: drivers/i2c/Kconfig
 ##
 CONFIG_I2C_CHARDEV=m
diff --git a/patches/0001-ARM-dts-imx6qdl-wandboard-Add-HDMI-support.patch 
b/patches/0001-ARM-dts-imx6qdl-wandboard-Add-HDMI-support.patch
new file mode 100644
index 000..a67c1ed
--- /dev/null
+++ b/patches/0001-ARM-dts-imx6qdl-wandboard-Add-HDMI-support.patch
@@ -0,0 +1,51 @@
+From 840e684cc847e847672a0abefd38875f682ad94d Mon Sep 17 00:00:00 2001
+From: Fabio Estevam fabio.este...@freescale.com
+Date: Tue, 22 Apr 2014 11:26:22 -0300
+Subject: [PATCH] ARM: dts: imx6qdl-wandboard: Add HDMI support
+
+Signed-off-by: Fabio Estevam fabio.este...@freescale.com
+Signed-off-by: Shawn Guo shawn@freescale.com
+---
+ arch/arm/boot/dts/imx6qdl-wandboard.dtsi | 19 +++
+ 1 file changed, 19 insertions(+)
+
+diff --git a/arch/arm/boot/dts/imx6qdl-wandboard.dtsi 
b/arch/arm/boot/dts/imx6qdl-wandboard.dtsi
+index bdfdf89..5c6f10c 100644
+--- a/arch/arm/boot/dts/imx6qdl-wandboard.dtsi
 b/arch/arm/boot/dts/imx6qdl-wandboard.dtsi
+@@ -62,6 +62,18 @@
+   status = okay;
+ };
+ 
++hdmi {
++  ddc-i2c-bus = i2c1;
++  status = okay;
++};
++
++i2c1 {
++  clock-frequency = 10;
++  pinctrl-names = default;
++  pinctrl-0 = pinctrl_i2c1;
++  status = okay;
++};
++
+ i2c2 {
+   clock-frequency = 10;
+   pinctrl-names = default;
+@@ -127,6 +139,13 @@
+   ;
+   };
+ 
++  pinctrl_i2c1: i2c1grp {
++  fsl,pins = 
++  MX6QDL_PAD_EIM_D21__I2C1_SCL
0x4001b8b1
++  MX6QDL_PAD_EIM_D28__I2C1_SDA
0x4001b8b1
++  ;
++  };
++
+   pinctrl_i2c2: i2c2grp {
+   fsl,pins = 
+   MX6QDL_PAD_KEY_COL3__I2C2_SCL   
0x4001b8b1
+-- 
+2.0.0.rc0
+
diff --git 
a/patches/0214-ARM-imx-add-HDMI-support-for-SolidRun-HummingBoard-a.patch 
b/patches/0214-ARM-imx-add-HDMI-support-for-SolidRun-HummingBoard-a.patch
new file mode 100644
index 000..94ac895
--- /dev/null
+++ b/patches/0214-ARM-imx-add-HDMI-support-for-SolidRun-HummingBoard-a.patch
@@ -0,0 +1,121 @@
+From: Russell King rmk+ker...@arm.linux.org.uk
+Subject: [PATCH 214/230] ARM: imx: add HDMI support for SolidRun HummingBoard
+ and Cubox-i
+
+Add the HDMI DT configuration for the SolidRun HummingBoard and Cubox-i.
+
+Signed-off-by: Russell King rmk+ker...@arm.linux.org.uk
+---
+ arch/arm/boot/dts/imx6dl-hummingboard.dts | 28 
+ arch/arm/boot/dts/imx6qdl-cubox-i.dtsi| 28 
+ 2 files changed, 56 insertions(+)
+
+diff --git a/arch/arm/boot/dts/imx6dl-hummingboard.dts 
b/arch/arm/boot/dts/imx6dl-hummingboard.dts
+index 5bfae54fb780..d3095f58bf7b 100644
+--- a/arch/arm/boot/dts/imx6dl-hummingboard.dts
 b/arch/arm/boot/dts/imx6dl-hummingboard.dts
+@@ -67,6 +67,14 @@
+   status = okay;
+ };
+ 
++hdmi {
++  pinctrl-names = default;
++  pinctrl-0 = pinctrl_hummingboard_hdmi;
++  ddc-i2c-bus = i2c2;
++  status = okay;
++  crtcs = ipu1 0;
++};
++
+ i2c1 {
+   pinctrl-names = default;
+   pinctrl-0 = pinctrl_hummingboard_i2c1;
+@@ -82,6 +90,13 @@
+*/
+ };
+ 
++i2c2 {
++  clock-frequency = 10;
++  pinctrl-names = default;
++  pinctrl-0 = pinctrl_hummingboard_i2c2;
++  status = okay;
++};
++
+ iomuxc {
+   hummingboard {
+   pinctrl_hummingboard_flexcan1: hummingboard-flexcan1 {
+@@ -97,6 +112,12 @@
+   ;
+   };
+ 
++  pinctrl_hummingboard_hdmi: hummingboard-hdmi {
++  fsl,pins = 
++  MX6QDL_PAD_KEY_ROW2__HDMI_TX_CEC_LINE 0x1f8b0
++  ;
++

Bug#747762: linux: [armhf/armmp] Support IR receiver for cubox-i

[Vagrant Cascadian]

Source: linux
Version: 3.14.2-1
Severity: wishlist
Tags: patch

The Cubox-i on the armmp kernel has an IR port that can be enabled by applying
the following patch:

diff --git a/config/armhf/config.armmp b/config/armhf/config.armmp
index c43f74e..38578d3 100644
--- a/config/armhf/config.armmp
+++ b/config/armhf/config.armmp
@@ -281,6 +281,11 @@ CONFIG_RADIO_SI4713=m
 CONFIG_RADIO_WL128X=m

 ##
+## file: drivers/media/rc/Kconfig
+##
+CONFIG_IR_GPIO_CIR=m
+
+##
 ## file: drivers/mfd/Kconfig
 ##
 CONFIG_MFD_DA9052_SPI=y


Thanks for considering it!


live well,
  vagrant


signature.asc
Description: Digital signature

Bug#747364: linux: [armhf/armmp] Enable support for BeagleBone Black

[Vagrant Cascadian]

On Fri, May 09, 2014 at 03:17:28AM +0100, Ben Hutchings wrote:
 On Wed, 2014-05-07 at 18:10 -0700, Vagrant Cascadian wrote:
  On Wed, May 07, 2014 at 01:46:53PM -0700, Vagrant Cascadian wrote:
  USB support seems flaky, at best... not sure if there are other options that
  could improve that.
 
 I think that's because the MUSB driver can only be built with support
 for one type of DMA controller, and so doesn't support DMA at all in
 multiplatform kernels.

hrm. That doesn't sound promising...


   CONFIG_DMA_OMAP=y
  +CONFIG_TI_CPPI41=y
 
 Why not =m?

Updated to use =m.


  +CONFIG_I2C_CHARDEV=y
 
 This symbol seems to be enabled in most configurations that have I2C,
 but often as a module. 

Updated to use =m.


  +CONFIG_SERIAL_OF_PLATFORM=y

Removed.

Also removed the various DEVFREQ entries; there were no enabled drivers, just
governors...


diff --git a/config/armhf/config.armmp b/config/armhf/config.armmp
index c43f74e..d9b2308 100644
--- a/config/armhf/config.armmp
+++ b/config/armhf/config.armmp
@@ -136,6 +136,12 @@ CONFIG_OMAP_OCP2SCP=m
 CONFIG_OMAP_INTERCONNECT=m
 
 ##
+## file: drivers/char/hw_random/Kconfig
+##
+CONFIG_HW_RANDOM=m
+CONFIG_HW_RANDOM_OMAP=m
+
+##
 ## file: drivers/clk/Kconfig
 ##
 CONFIG_CLK_TWL6040=m
@@ -170,6 +176,8 @@ CONFIG_IMX_SDMA=y
 CONFIG_IMX_DMA=y
 CONFIG_MXS_DMA=y
 CONFIG_DMA_OMAP=y
+CONFIG_TI_CPPI41=m
+CONFIG_TI_EDMA=y
 
 ##
 ## file: drivers/gpio/Kconfig
@@ -178,6 +186,22 @@ CONFIG_GPIO_SYSFS=y
 CONFIG_GPIO_DA9052=m
 CONFIG_GPIO_TWL4030=y
 CONFIG_GPIO_TWL6040=y
+CONFIG_GPIO_GENERIC_PLATFORM=m
+
+##
+## file: drivers/gpu/drm/Kconfig
+##
+CONFIG_DRM=m
+
+##
+## file: drivers/gpu/drm/i2c/Kconfig
+##
+CONFIG_DRM_I2C_NXP_TDA998X=m
+
+##
+## file: drivers/gpu/drm/tilcdc/Kconfig
+##
+CONFIG_DRM_TILCDC=m
 
 ##
 ## file: drivers/hwspinlock/Kconfig
@@ -185,6 +209,11 @@ CONFIG_GPIO_TWL6040=y
 CONFIG_HWSPINLOCK_OMAP=m
 
 ##
+## file: drivers/i2c/Kconfig
+##
+CONFIG_I2C_CHARDEV=m
+
+##
 ## file: drivers/i2c/busses/Kconfig
 ##
 CONFIG_I2C_GPIO=y
@@ -199,6 +228,11 @@ CONFIG_I2C_VERSATILE=m
 CONFIG_IIO=m
 
 ##
+## file: drivers/iio/adc/Kconfig
+##
+CONFIG_TI_AM335X_ADC=m
+
+##
 ## file: drivers/iio/light/Kconfig
 ##
 CONFIG_SENSORS_TSL2563=m
@@ -229,6 +263,7 @@ CONFIG_SERIO_AMBAKMI=m
 ##
 CONFIG_INPUT_TOUCHSCREEN=y
 CONFIG_TOUCHSCREEN_TSC2005=m
+CONFIG_TOUCHSCREEN_TI_AM335X_TSC=m
 
 ##
 ## file: drivers/iommu/Kconfig
@@ -491,6 +526,11 @@ CONFIG_WLCORE_SDIO=m
 CONFIG_PHY_SUN4I_USB=m
 
 ##
+## file: drivers/pinctrl/Kconfig
+##
+CONFIG_PINCTRL_SINGLE=y
+
+##
 ## file: drivers/pinctrl/vt8500/Kconfig
 ##
 CONFIG_PINCTRL_WM8850=y
@@ -592,6 +632,8 @@ CONFIG_SERIAL_VT8500_CONSOLE=y
 CONFIG_SERIAL_ARC=y
 CONFIG_SERIAL_ARC_CONSOLE=y
 CONFIG_SERIAL_ARC_NR_PORTS=1
+CONFIG_SERIAL_OMAP=y
+CONFIG_SERIAL_OMAP_CONSOLE=y
 
 ##
 ## file: drivers/tty/serial/8250/Kconfig
@@ -663,6 +705,8 @@ CONFIG_OMAP_USB3=m
 CONFIG_TWL4030_USB=m
 CONFIG_TWL6030_USB=m
 CONFIG_USB_MXS_PHY=m
+CONFIG_AM335X_CONTROL_USB=m
+CONFIG_AM335X_PHY_USB=m
 
 ##
 ## file: drivers/video/Kconfig


live well,
  vagrant


signature.asc
Description: Digital signature

Bug#747364: linux: [armhf/armmp] Enable support for BeagleBone Black

[Vagrant Cascadian]

Source: linux
Version: 3.14.2-1
Severity: wishlist
Tags: patch

The following patch enables configuration options needed for use with
BeagleBone Black.

Ethernet, serial console, MMC support work. Video may require some additional
patches from upstream...

Same patch also works with 3.15-rc4-1~exp1 in svn.

Please consider applying it.

diff --git a/config/armhf/config.armmp b/config/armhf/config.armmp
index c43f74e..d28511d 100644
--- a/config/armhf/config.armmp
+++ b/config/armhf/config.armmp
@@ -136,6 +136,12 @@ CONFIG_OMAP_OCP2SCP=m
 CONFIG_OMAP_INTERCONNECT=m
 
 ##
+## file: drivers/char/hw_random/Kconfig
+##
+CONFIG_HW_RANDOM=m
+CONFIG_HW_RANDOM_OMAP=m
+
+##
 ## file: drivers/clk/Kconfig
 ##
 CONFIG_CLK_TWL6040=m
@@ -153,6 +159,15 @@ CONFIG_GENERIC_CPUFREQ_CPU0=m
 CONFIG_ARM_IMX6Q_CPUFREQ=m
 
 ##
+## file: drivers/devfreq/Kconfig
+##
+CONFIG_PM_DEVFREQ=y
+CONFIG_DEVFREQ_GOV_SIMPLE_ONDEMAND=m
+CONFIG_DEVFREQ_GOV_PERFORMANCE=m
+CONFIG_DEVFREQ_GOV_POWERSAVE=m
+CONFIG_DEVFREQ_GOV_USERSPACE=m
+
+##
 ## file: drivers/crypto/Kconfig
 ##
 CONFIG_CRYPTO_DEV_MV_CESA=m
@@ -170,6 +185,8 @@ CONFIG_IMX_SDMA=y
 CONFIG_IMX_DMA=y
 CONFIG_MXS_DMA=y
 CONFIG_DMA_OMAP=y
+CONFIG_TI_CPPI41=y
+CONFIG_TI_EDMA=y
 
 ##
 ## file: drivers/gpio/Kconfig
@@ -178,6 +195,7 @@ CONFIG_GPIO_SYSFS=y
 CONFIG_GPIO_DA9052=m
 CONFIG_GPIO_TWL4030=y
 CONFIG_GPIO_TWL6040=y
+CONFIG_GPIO_GENERIC_PLATFORM=m
 
 ##
 ## file: drivers/hwspinlock/Kconfig
@@ -185,6 +203,11 @@ CONFIG_GPIO_TWL6040=y
 CONFIG_HWSPINLOCK_OMAP=m
 
 ##
+## file: drivers/i2c/Kconfig
+##
+CONFIG_I2C_CHARDEV=y
+
+##
 ## file: drivers/i2c/busses/Kconfig
 ##
 CONFIG_I2C_GPIO=y
@@ -199,6 +222,11 @@ CONFIG_I2C_VERSATILE=m
 CONFIG_IIO=m
 
 ##
+## file: drivers/iio/adc/Kconfig
+##
+CONFIG_TI_AM335X_ADC=m
+
+##
 ## file: drivers/iio/light/Kconfig
 ##
 CONFIG_SENSORS_TSL2563=m
@@ -229,6 +257,7 @@ CONFIG_SERIO_AMBAKMI=m
 ##
 CONFIG_INPUT_TOUCHSCREEN=y
 CONFIG_TOUCHSCREEN_TSC2005=m
+CONFIG_TOUCHSCREEN_TI_AM335X_TSC=m
 
 ##
 ## file: drivers/iommu/Kconfig
@@ -491,6 +520,11 @@ CONFIG_WLCORE_SDIO=m
 CONFIG_PHY_SUN4I_USB=m
 
 ##
+## file: drivers/pinctrl/Kconfig
+##
+CONFIG_PINCTRL_SINGLE=y
+
+##
 ## file: drivers/pinctrl/vt8500/Kconfig
 ##
 CONFIG_PINCTRL_WM8850=y
@@ -592,6 +626,9 @@ CONFIG_SERIAL_VT8500_CONSOLE=y
 CONFIG_SERIAL_ARC=y
 CONFIG_SERIAL_ARC_CONSOLE=y
 CONFIG_SERIAL_ARC_NR_PORTS=1
+CONFIG_SERIAL_OF_PLATFORM=y
+CONFIG_SERIAL_OMAP=y
+CONFIG_SERIAL_OMAP_CONSOLE=y
 
 ##
 ## file: drivers/tty/serial/8250/Kconfig
@@ -663,6 +700,8 @@ CONFIG_OMAP_USB3=m
 CONFIG_TWL4030_USB=m
 CONFIG_TWL6030_USB=m
 CONFIG_USB_MXS_PHY=m
+CONFIG_AM335X_CONTROL_USB=m
+CONFIG_AM335X_PHY_USB=m
 
 ##
 ## file: drivers/video/Kconfig
@@ -691,6 +730,7 @@ CONFIG_OMAP2_DSS=m
 CONFIG_OMAP2_DSS_VENC=y
 CONFIG_OMAP4_DSS_HDMI=y
 CONFIG_OMAP2_DSS_SDI=y
+CONFIG_OMAP2_DSS_DSI=y
 
 ##
 ## file: drivers/video/omap2/omapfb/Kconfig


live well,
  vagrant


signature.asc
Description: Digital signature

Bug#747364: linux: [armhf/armmp] Enable support for BeagleBone Black

[Vagrant Cascadian]

On Wed, May 07, 2014 at 01:46:53PM -0700, Vagrant Cascadian wrote:
 The following patch enables configuration options needed for use with
 BeagleBone Black.
 
 Ethernet, serial console, MMC support work. Video may require some additional
 patches from upstream...

Updated patch with support for video output as modules:
  CONFIG_DRM_I2C_NXP_TDA998X=m
  CONFIG_DRM_TILCDC=m

Thanks to Robert C. Nelson for the additional options!

USB support seems flaky, at best... not sure if there are other options that
could improve that.

live well,
  vagrant

diff --git a/config/armhf/config.armmp b/config/armhf/config.armmp
index c43f74e..0ab26fc 100644
--- a/config/armhf/config.armmp
+++ b/config/armhf/config.armmp
@@ -136,6 +136,12 @@ CONFIG_OMAP_OCP2SCP=m
 CONFIG_OMAP_INTERCONNECT=m
 
 ##
+## file: drivers/char/hw_random/Kconfig
+##
+CONFIG_HW_RANDOM=m
+CONFIG_HW_RANDOM_OMAP=m
+
+##
 ## file: drivers/clk/Kconfig
 ##
 CONFIG_CLK_TWL6040=m
@@ -153,6 +159,15 @@ CONFIG_GENERIC_CPUFREQ_CPU0=m
 CONFIG_ARM_IMX6Q_CPUFREQ=m
 
 ##
+## file: drivers/devfreq/Kconfig
+##
+CONFIG_PM_DEVFREQ=y
+CONFIG_DEVFREQ_GOV_SIMPLE_ONDEMAND=m
+CONFIG_DEVFREQ_GOV_PERFORMANCE=m
+CONFIG_DEVFREQ_GOV_POWERSAVE=m
+CONFIG_DEVFREQ_GOV_USERSPACE=m
+
+##
 ## file: drivers/crypto/Kconfig
 ##
 CONFIG_CRYPTO_DEV_MV_CESA=m
@@ -170,6 +185,8 @@ CONFIG_IMX_SDMA=y
 CONFIG_IMX_DMA=y
 CONFIG_MXS_DMA=y
 CONFIG_DMA_OMAP=y
+CONFIG_TI_CPPI41=y
+CONFIG_TI_EDMA=y
 
 ##
 ## file: drivers/gpio/Kconfig
@@ -178,6 +195,22 @@ CONFIG_GPIO_SYSFS=y
 CONFIG_GPIO_DA9052=m
 CONFIG_GPIO_TWL4030=y
 CONFIG_GPIO_TWL6040=y
+CONFIG_GPIO_GENERIC_PLATFORM=m
+
+##
+## file: drivers/gpu/drm/Kconfig
+##
+CONFIG_DRM=m
+
+##
+## file: drivers/gpu/drm/i2c/Kconfig
+##
+CONFIG_DRM_I2C_NXP_TDA998X=m
+
+##
+## file: drivers/gpu/drm/tilcdc/Kconfig
+##
+CONFIG_DRM_TILCDC=m
 
 ##
 ## file: drivers/hwspinlock/Kconfig
@@ -185,6 +218,11 @@ CONFIG_GPIO_TWL6040=y
 CONFIG_HWSPINLOCK_OMAP=m
 
 ##
+## file: drivers/i2c/Kconfig
+##
+CONFIG_I2C_CHARDEV=y
+
+##
 ## file: drivers/i2c/busses/Kconfig
 ##
 CONFIG_I2C_GPIO=y
@@ -199,6 +237,11 @@ CONFIG_I2C_VERSATILE=m
 CONFIG_IIO=m
 
 ##
+## file: drivers/iio/adc/Kconfig
+##
+CONFIG_TI_AM335X_ADC=m
+
+##
 ## file: drivers/iio/light/Kconfig
 ##
 CONFIG_SENSORS_TSL2563=m
@@ -229,6 +272,7 @@ CONFIG_SERIO_AMBAKMI=m
 ##
 CONFIG_INPUT_TOUCHSCREEN=y
 CONFIG_TOUCHSCREEN_TSC2005=m
+CONFIG_TOUCHSCREEN_TI_AM335X_TSC=m
 
 ##
 ## file: drivers/iommu/Kconfig
@@ -491,6 +535,11 @@ CONFIG_WLCORE_SDIO=m
 CONFIG_PHY_SUN4I_USB=m
 
 ##
+## file: drivers/pinctrl/Kconfig
+##
+CONFIG_PINCTRL_SINGLE=y
+
+##
 ## file: drivers/pinctrl/vt8500/Kconfig
 ##
 CONFIG_PINCTRL_WM8850=y
@@ -592,6 +641,9 @@ CONFIG_SERIAL_VT8500_CONSOLE=y
 CONFIG_SERIAL_ARC=y
 CONFIG_SERIAL_ARC_CONSOLE=y
 CONFIG_SERIAL_ARC_NR_PORTS=1
+CONFIG_SERIAL_OF_PLATFORM=y
+CONFIG_SERIAL_OMAP=y
+CONFIG_SERIAL_OMAP_CONSOLE=y
 
 ##
 ## file: drivers/tty/serial/8250/Kconfig
@@ -663,6 +715,8 @@ CONFIG_OMAP_USB3=m
 CONFIG_TWL4030_USB=m
 CONFIG_TWL6030_USB=m
 CONFIG_USB_MXS_PHY=m
+CONFIG_AM335X_CONTROL_USB=m
+CONFIG_AM335X_PHY_USB=m
 
 ##
 ## file: drivers/video/Kconfig
@@ -691,6 +745,7 @@ CONFIG_OMAP2_DSS=m
 CONFIG_OMAP2_DSS_VENC=y
 CONFIG_OMAP4_DSS_HDMI=y
 CONFIG_OMAP2_DSS_SDI=y
+CONFIG_OMAP2_DSS_DSI=y
 
 ##
 ## file: drivers/video/omap2/omapfb/Kconfig


signature.asc
Description: Digital signature

Bug#746420: linux: FTBFS on armhf: sun4i-a10-a1000.dtb

[Vagrant Cascadian]

Source: linux
Version: 3.14.2-1
Severity: normal
Tags: patch

When attempting to build on armhf, the build fails with:

  DTC arch/arm/boot/dts/socfpga_vt.dtb
  DTC arch/arm/boot/dts/sun4i-a10-a1000.dtb
ERROR (phandle_references): Reference to non-existent node or label usb_clk

ERROR (phandle_references): Reference to non-existent node or label usb_clk

ERROR (phandle_references): Reference to non-existent node or label usb_clk

ERROR (phandle_references): Reference to non-existent node or label usb_clk

ERROR (phandle_references): Reference to non-existent node or label usb_clk

ERROR: Input tree has errors, aborting (use -f to force output)
make[6]: *** [arch/arm/boot/dts/sun4i-a10-a1000.dtb] Error 2
make[5]: *** [dtbs] Error 2
make[5]: *** Waiting for unfinished jobs
  CC  init/calibrate.o
  CC  init/init_task.o
  CC  init/version.o
  LD  init/mounts.o
  LD  init/built-in.o
make[4]: *** [sub-make] Error 2
make[3]: *** [all] Error 2
make[3]: Leaving directory `/«PKGBUILDDIR»/debian/build/build_armhf_none_armmp'
make[2]: *** [debian/stamps/build_armhf_none_armmp_plain] Error 2
make[1]: *** [build-arch_armhf_none_armmp_real] Error 2
make[2]: Leaving directory `/«PKGBUILDDIR»'
make[1]: Leaving directory `/«PKGBUILDDIR»'
make: *** [build-arch] Error 2
dpkg-buildpackage: error: debian/rules build-arch gave error exit status 2


It seems to be missing the following commit from mainline, which applies
cleanly, and fixes the FTBFS:

commit 0076c8bdb417a96209e28755cf5156bd4439737b
Author: Roman Byshko rbys...@gmail.com
Date:   Fri Feb 7 16:21:51 2014 +0100

ARM: sun4i: dt: Add bindings for USB clocks

Signed-off-by: Roman Byshko rbys...@gmail.com
Signed-off-by: Hans de Goede hdego...@redhat.com
Signed-off-by: Maxime Ripard maxime.rip...@free-electrons.com

diff --git a/arch/arm/boot/dts/sun4i-a10.dtsi b/arch/arm/boot/dts/sun4i-a10.dtsi
index 26cf191..b268f68 100644
--- a/arch/arm/boot/dts/sun4i-a10.dtsi
+++ b/arch/arm/boot/dts/sun4i-a10.dtsi
@@ -315,6 +315,15 @@
clock-output-names = ir1;
};

+   usb_clk: clk@01c200cc {
+   #clock-cells = 1;
+   #reset-cells = 1;
+   compatible = allwinner,sun4i-a10-usb-clk;
+   reg = 0x01c200cc 0x4;
+   clocks = pll6 1;
+   clock-output-names = usb_ohci0, usb_ohci1, 
usb_phy;
+   };
+
spi3_clk: clk@01c200d4 {
#clock-cells = 0;
compatible = allwinner,sun4i-mod0-clk;


live well,
  vagrant


signature.asc
Description: Digital signature

Re: Arm64 port live on debian-ports

[Vagrant Cascadian]

On Sun, Apr 20, 2014 at 04:28:45PM +0100, Ian Campbell wrote:
 On Sun, 2014-04-20 at 16:13 +0100, Ben Hutchings wrote:
  Given that, it seems like a good time to add arm64 to src:linux with a
  configuration that will run on at least a typical QEMU ARM64 emulation.
 
 AIUI qemu 2.0 only does qemu-aarch64-user, with the system emulation
 portion slated to be merged shortly[0].

Not sure if it works, but qemu-system-aarch64 is in the 2.0 packages in sid:

  
https://packages.debian.org/search?suite=sidsearchon=contentskeywords=qemu-system-aarch64


live well,
  vagrant


signature.asc
Description: Digital signature

Bug#741142: linux: FTBFS on armhf (experimental)

[Vagrant Cascadian]

On Sun, Mar 09, 2014 at 03:58:29AM +, Ben Hutchings wrote:
 On Sat, 2014-03-08 at 18:22 -0800, Vagrant Cascadian wrote:
  Source: linux
  Version: 3.14~rc5-1~exp1
  Severity: important
  
  I tried building the armhf packages locally, to test some new features in 
  3.14,
  but they failed to build from source for me:
 
 Does this fix it?
 
 Ben.
 
 diff --git a/drivers/net/ethernet/brocade/bna/bfa_ioc.c 
 b/drivers/net/ethernet/brocade/bna/bfa_ioc.c
 index 1803c39..354ae97 100644
 --- a/drivers/net/ethernet/brocade/bna/bfa_ioc.c
 +++ b/drivers/net/ethernet/brocade/bna/bfa_ioc.c
 @@ -1704,7 +1704,7 @@ bfa_flash_sem_get(void __iomem *bar)
   while (!bfa_raw_sem_get(bar)) {
   if (--n = 0)
   return BFA_STATUS_BADFLASH;
 - udelay(1);
 + mdelay(10);
   }
   return BFA_STATUS_OK;
  }
 diff --git a/drivers/scsi/bfa/bfa_ioc.c b/drivers/scsi/bfa/bfa_ioc.c
 index 65180e1..50c75e1 100644
 --- a/drivers/scsi/bfa/bfa_ioc.c
 +++ b/drivers/scsi/bfa/bfa_ioc.c
 @@ -7006,7 +7006,7 @@ bfa_flash_sem_get(void __iomem *bar)
   while (!bfa_raw_sem_get(bar)) {
   if (--n = 0)
   return BFA_STATUS_BADFLASH;
 - udelay(1);
 + mdelay(10);
   }
   return BFA_STATUS_OK;
  }
 ---

With the above patch applied, it builds. Thanks!

live well,
  vagrant


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20140309082843.GM4707@prl

Bug#734430: linux-image-3.12-1-armmp: Enable BRCMFMAC wireless drivers

[Vagrant Cascadian]

Package: linux-image-3.12-1-armmp
Version: 3.12.6-2
Severity: wishlist
Tags: patch

Please enable the Broadcom IEEE802.11n embedded FullMAC WLAN driver, which is
used in the wandboard quad.

--- a/config/armhf/config.armmp
+++ b/config/armhf/config.armmp
@@ -451,6 +451,12 @@ CONFIG_SKFP=m
 CONFIG_MDIO_SUN4I=y

 ##
+## file: drivers/net/wireless/brcm80211/Kconfig
+##
+CONFIG_BRCMFMAC=m
+CONFIG_BRCMFMAC_SDIO=m
+
+##
 ## file: drivers/net/wireless/ti/Kconfig
 ##
 CONFIG_WL_TI=m


Thanks!

live well,
  vagrant


signature.asc
Description: Digital signature

Bug#565696: initramfs-tools: Support for AoE

[Vagrant Cascadian]

On Thu, Feb 25, 2010 at 06:22:47PM +0100, maximilian attems wrote:
  Basically, the idea is to make it possible to do diskless virtual
  machines with a minimum of fuss.  I can't think of any easy way to make
  all the moving parts fit together.

We've been experimenting with AoE for use with LTSP, and Alkis wrote a
simple initramfs-tools local-top script which loads a the aoe module when
root=/dev/etherd/e*, and then the local script can mount the device like any
other block device:

  
https://bazaar.launchpad.net/~ltsp-upstream/ltsp/ltsp-trunk/view/head:/client/Debian/share/initramfs-tools/scripts/local-top/aoe_ltsp

There's nothing really ltsp-specific about it, though.

It would seem more appropriate to be in initramfs-tools directly, as it doesn't
require any userspace utilities; there aren't really any existing AoE-related
packages that would make sense to add a single script to.

Interestingly enough, you don't even need IP networking for a remote root
filesystem!

 
 well i could ship it in examples section then AoE people would
 just have to copy over or you could add it to a special aoe package.
 
 iscsi or such seems in wider use, thus higher priority,
 don't know if exerternal package does provide some boot scripts.

iSCSI seems a bit more complicated to set up. That said, I don't see support or
non-support for iSCSI as a reason to exclude AoE.


live well,
  vagrant


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131230013624.ge32...@cascadia.debian.net

Re: linux-image-3.11-2-armmp_3.11.8-1_armhf not include AHCI_IMX

[Vagrant Cascadian]

On Tue, Dec 17, 2013 at 09:58:17AM +, Ian Campbell wrote:
 On Tue, 2013-12-17 at 13:31 +0800, Niew, Sh. wrote:
  On Dec 13, 2013, at 6:34 PM, Ian Campbell i...@hellion.org.uk wrote:
   On Fri, 2013-12-06 at 13:54 +0800, Niew, Sh. wrote:
   [...]
   Is it ahci_imx need ahci_platform compile with yes?
   
   IMHO it should be loaded automatically as required. This is probably
   something which you need to take up with the upstream developers I'm
   afraid.
  
  Before reaching to the upstream, i get the v3.13-rc4 a shot
  And the issue is eliminate, but i not include every part as like the debian 
  config included.
 
 So to make sure I understand what you are saying: You built upstream
 v3.13-rc4 with a config which was different to the Debian one and it
 worked fine?
 
 If so then either something is fixed between 3.12.3 and 3.13-rc4 or
 there is something in the config which matters. My money would be on a
 code change, but perhaps you could try with a config which is mostly the
 same as the Debian one? (i.e. start from the Debian config and run make
 oldconfig).

FWIW: SATA is working fine for me on the wandboard quad using the 3.12.3-1~exp1
armmp kernel from experimental(thanks for that!), which is imx6q (and Niew, sh.
was talking about a sabrelite board?).

Maybe changes to the relevent dts file in 3.13* fixed it?

Maybe loading an outdated dts file at boot?


live well,
  vagrant


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131217191136.gq27...@cascadia.debian.net

Bug#676604: nfs-kernel-server: fails to start if exports are only defined in /etc/exports.d/*.exports

[Vagrant Cascadian]

tags 676604 patch
thanks

On Fri, Jun 08, 2012 at 12:06:15AM -0700, Vagrant Cascadian wrote:
 After upgrading nfs-kernel-server, it fails to start, complaining that there 
 are no exports.
 
   service nfs-kernel-server start
   Not starting NFS kernel daemon: no exports. ... (warning).
 
 I do not have any changes to /etc/exports, it is true, but do have exports 
 defined in /etc/exports.d/*.exports, and these are valid exports.
 
 By commenting out the check for an empty /etc/exports (introduced to resolve 
 #673798) in /etc/init.d/nfs-kernel-server, nfsd starts and all is well again:
 
   if [ -f /etc/exports ] #  grep -q '^[[:space:]]*[^#]*/' /etc/exports
 
 This check should probably be either be reverted, or it should at least 
 iterate 
 it through all the /etc/exports.d/*.exports files before deciding there are 
 no 
 exports.

patch implementing the latter follows:

--- a/init.d/nfs-kernel-server
+++ b/init.d/nfs-kernel-server
@@ -60,7 +60,13 @@ do_mount() {
 # See how we were called.
 case $1 in
   start)
-   if [ -f /etc/exports ]  grep -q '^[[:space:]]*[^#]*/' /etc/exports
+   export_files=/etc/exports
+   for file in /etc/exports.d/*.exports ; do
+   if [ -f $file ]; then
+   export_files=$export_files $file
+   fi
+   done
+   if [ -f /etc/exports ]  grep -q '^[[:space:]]*[^#]*/' $export_files
then
do_modprobe nfsd


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20120627210147.GM21245@talon.fglan

Bug#677157: initramfs-tools: root=/dev/nfs mostly useless due to BOOT=local being defined in initramfs.conf

[Vagrant Cascadian]

Package: initramfs-tools
Version: 0.106
Severity: normal
Tags: patch


Specifying root=/dev/nfs doesn't really do anything without also either 
specifying boot=nfs or changing BOOT= in initramfs.conf:

  /dev/nfs)
[ -z ${BOOT} ]  BOOT=nfs
;;

At which point, you could just as well not specify root=/dev/nfs at all.

This breaks the historical method for network booting with nfsroot with the 
kernel arguments:

  ip=dhcp root=/dev/nfs

I guess the check to make sure BOOT wasn't already set was changed in order to 
fix putting boot= anywhere in the boot commandline:

  http://bugs.debian.org/453294

But it seems to me the correct thing to do in that case would be to specify 
boot=live after root=/dev/nfs... since the order of the argument processing 
matters, rather than preferring the result of one argument over another.

Alternately/Additionally, why is BOOT=local defined in initramfs.conf at all? 
What about only setting BOOT=local if $BOOT is undefined, after all the 
argument processing is done:

diff --git a/conf/initramfs.conf b/conf/initramfs.conf
index abc231b..995d455 100644
--- a/conf/initramfs.conf
+++ b/conf/initramfs.conf
@@ -53,7 +53,7 @@ COMPRESS=gzip
 # nfs - Boot using an NFS drive as the root of the drive.
 #

-BOOT=local
+#BOOT=local

 #
 # DEVICE: ...
diff --git a/init b/init
index cbd22e2..7e54a73 100755
--- a/init
+++ b/init
@@ -186,6 +186,11 @@ for x in $(cat /proc/cmdline); do
esac
 done

+if [ -z ${BOOT} ]; then
+   # Default to BOOT=local if no boot script defined.
+   BOOT=local
+fi
+
 if [ -n ${noresume} ]; then
export noresume
unset resume


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20120611222455.GV2630@talon.fglan

Bug#676604: nfs-kernel-server: fails to start if exports are only defined in /etc/exports.d/*.exports

[Vagrant Cascadian]

Package: nfs-kernel-server
Version: 1:1.2.6-2
Severity: important

After upgrading nfs-kernel-server, it fails to start, complaining that there 
are no exports.

  service nfs-kernel-server start
  Not starting NFS kernel daemon: no exports. ... (warning).

I do not have any changes to /etc/exports, it is true, but do have exports 
defined in /etc/exports.d/*.exports, and these are valid exports.

By commenting out the check for an empty /etc/exports (introduced to resolve 
#673798) in /etc/init.d/nfs-kernel-server, nfsd starts and all is well again:

if [ -f /etc/exports ] #  grep -q '^[[:space:]]*[^#]*/' /etc/exports

This check should probably be either be reverted, or it should at least iterate 
it through all the /etc/exports.d/*.exports files before deciding there are no 
exports.


live well,
  vagrant


-- Package-specific info:
-- rpcinfo --
   program vers proto   port  service
104   tcp111  portmapper
103   tcp111  portmapper
102   tcp111  portmapper
104   udp111  portmapper
103   udp111  portmapper
102   udp111  portmapper
1000241   udp  53540  status
1000241   tcp  37861  status
132   tcp   2049  nfs
133   tcp   2049  nfs
134   tcp   2049  nfs
1002272   tcp   2049
1002273   tcp   2049
132   udp   2049  nfs
133   udp   2049  nfs
134   udp   2049  nfs
1002272   udp   2049
1002273   udp   2049
1000211   udp  36032  nlockmgr
1000213   udp  36032  nlockmgr
1000214   udp  36032  nlockmgr
1000211   tcp  49636  nlockmgr
1000213   tcp  49636  nlockmgr
1000214   tcp  49636  nlockmgr
151   udp  38872  mountd
151   tcp  56753  mountd
152   udp  42405  mountd
152   tcp  46676  mountd
153   udp  40428  mountd
153   tcp  33127  mountd
-- /etc/default/nfs-kernel-server --
RPCNFSDCOUNT=8
RPCNFSDPRIORITY=0
RPCMOUNTDOPTS=--manage-gids
NEED_SVCGSSD=
RPCSVCGSSDOPTS=
-- /etc/exports --

-- System Information:
Debian Release: wheezy/sid
  APT prefers stable-updates
  APT policy: (500, 'stable-updates'), (500, 'testing'), (500, 'stable'), (120, 
'unstable'), (110, 'experimental')
Architecture: i386 (i686)

Kernel: Linux 3.2.0-2-686-pae (SMP w/2 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages nfs-kernel-server depends on:
ii  libblkid1   2.20.1-5
ii  libc6   2.13-32
ii  libtirpc1   0.2.2-5
ii  libwrap07.6.q-23
ii  lsb-base4.1+Debian4
ii  nfs-common  1:1.2.6-2
ii  ucf 3.0025+nmu3

nfs-kernel-server recommends no packages.

nfs-kernel-server suggests no packages.

-- Configuration Files:
/etc/init.d/nfs-kernel-server changed:
DESC=NFS kernel daemon
PREFIX=/usr
[ -x $PREFIX/sbin/rpc.nfsd] || exit 0
[ -x $PREFIX/sbin/rpc.mountd  ] || exit 0
[ -x $PREFIX/sbin/exportfs] || exit 0
DEFAULTFILE=/etc/default/nfs-kernel-server
RPCNFSDCOUNT=8
RPCNFSDPRIORITY=0
RPCMOUNTDOPTS=
NEED_SVCGSSD=no
RPCSVCGSSDOPTS=
PROCNFSD_MOUNTPOINT=/proc/fs/nfsd
if [ -f $DEFAULTFILE ]; then
. $DEFAULTFILE
fi
. /lib/lsb/init-functions
do_modprobe() {
if [ -x /sbin/modprobe -a -f /proc/modules ]
then
modprobe -q $1 || true
fi
}
do_mount() {
if ! grep -E -qs $1\$ /proc/filesystems
then
return 1
fi
if ! mountpoint -q $2
then
mount -t $1 $1 $2
return
fi
return 0
}
case $1 in
  start)
if [ -f /etc/exports ] #  grep -q '^[[:space:]]*[^#]*/' /etc/exports
then
do_modprobe nfsd
# See if our running kernel supports the NFS kernel server
if ! grep -E -qs [[:space:]]nfsd\$ /proc/filesystems; then
log_warning_msg Not starting $DESC: no support in 
current kernel.
exit 0
fi

do_mount nfsd $PROCNFSD_MOUNTPOINT || NEED_SVCGSSD=no
log_begin_msg Exporting directories for $DESC...
$PREFIX/sbin/exportfs -r
RET=$?
if [ $RET != 0 ]; then
log_end_msg $RET
exit $RET
fi
log_end_msg 0
log_daemon_msg Starting $DESC
log_progress_msg nfsd

# See if rpcbind is running
/usr/sbin/rpcinfo -p /dev/null 21
RET=$?
if [ $RET != 0 ]; then
echo
log_warning_msg Not starting: portmapper is not running
exit 0
fi
start-stop-daemon --start --oknodo --quiet \
--nicelevel $RPCNFSDPRIORITY \
--exec $PREFIX/sbin/rpc.nfsd -- $RPCNFSDCOUNT

Bug#670462: linux-2.6: Support new armhf kernel variant for Versatile Express (vexpress)

[Vagrant Cascadian]

On Mon, May 07, 2012 at 08:38:27AM +0200, Arnaud Patard wrote:
 Vagrant Cascadian vagr...@freegeek.org writes:
  only in patch2:
  unchanged:
  --- linux-2.6-3.2.16.orig/debian/config/armhf/config.vexpress
  +++ linux-2.6-3.2.16/debian/config/armhf/config.vexpress
  @@ -0,0 +1,54 @@
  +##
  +## file: arch/arm/Kconfig
  +##
  +## choice: ARM system type
  +CONFIG_ARCH_VEXPRESS=y
  +## end choice
  +
  +## arch/arm/mach-vexpress/Kconfig
  +CONFIG_ARCH_VEXPRESS_CA9X4=y
  +
  +CONFIG_LOG_BUF_SHIFT=14
 
 any reason for this change ?

It was in the default vexpress config shipped upstream in
arch/arm/configs/vexpress_defconfig. it is also present in a couple of the
debian/config/armel/ configs. Honestly, no idea what it actually does, so I'm
not attached to keeping it if it still works without it.


  +CONFIG_OPROFILE=m
  +
  +CONFIG_SMP=y
  +CONFIG_VMSPLIT_2G=y
  +CONFIG_HOTPLUG_CPU=y
  +
  +CONFIG_ZBOOT_ROM_TEXT=0x0
  +CONFIG_ZBOOT_ROM_BSS=0x0
  +
  +CONFIG_NEON=y
  +
  +
  +CONFIG_MTD=m
 
 hm. why is it still enabled ? from what I understand, qemu vexpress doesn't
 simulate the flash peripheral. If you want to support something else
 than qemu, you're missing some drivers I guess. If you want to support
 only qemu vexpress emulation, maybe it worthes to be mentionned in the
 description ?

Again, it was part of the vexpress default config shipped in the upstream 
kernel (although i switched it from enabled to a module).

 
  +CONFIG_MTD_CONCAT=m
  +CONFIG_MTD_PARTITIONS=m
  +CONFIG_MTD_CMDLINE_PARTS=m
  +CONFIG_MTD_CHAR=m
  +CONFIG_MTD_BLOCK=m
  +CONFIG_MTD_CFI=m
  +CONFIG_MTD_CFI_INTELEXT=m
  +CONFIG_MTD_CFI_AMDSTD=m
  +CONFIG_MTD_ARM_INTEGRATOR=m
 
 fwiw, doesn't exist anymore. physmap should be used now.

Same story there, too.

 
 Nevertheless, I've built tested this configuration and it worked in Qemu
 as expected. I'll add that once we agree on the configuration.

Great! Thanks for taking the time to review it.

I'll defer to others on matters of why to enable or disble particular options; 
I just grabbed the default config, tested that it works in qemu, and then 
trimmed out what seemed to be duplicated in other parts of debian/config/.


 Also, do you intend to add support for it into the debian-installer too ?

I think that would be usefuul, and can put some time into that. Though it 
seemed like the first steep would be to get a working kernel shipped.


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20120507162822.GE5888@talon.fglan

Bug#670462: linux-2.6: Support new armhf kernel variant for Versatile Express (vexpress)

[Vagrant Cascadian]

The following trimmed down debian/config/armhf/config.vexpress still worked 
for me in qemu (built against 3.2.16-1). It also trimmed out the extraneous 
Supported text in the defines file. The control file can be generated, as I 
understand it, so didn't include in this diff.

diff -u linux-2.6-3.2.16/debian/config/armhf/defines 
linux-2.6-3.2.16/debian/config/armhf/defines
--- linux-2.6-3.2.16/debian/config/armhf/defines
+++ linux-2.6-3.2.16/debian/config/armhf/defines
@@ -2,6 +2,7 @@
 flavours:
  mx5
  omap
+ vexpress
 kernel-arch: arm
 
 [image]
@@ -28,0 +30,8 @@
+
+[vexpress_build]
+image-file: arch/arm/boot/zImage
+
+[vexpress_description]
+hardware: ARM Ltd. Versatile Express
+hardware-long: ARM Ltd. Versatile Express family of processors
+

only in patch2:
unchanged:
--- linux-2.6-3.2.16.orig/debian/config/armhf/config.vexpress
+++ linux-2.6-3.2.16/debian/config/armhf/config.vexpress
@@ -0,0 +1,54 @@
+##
+## file: arch/arm/Kconfig
+##
+## choice: ARM system type
+CONFIG_ARCH_VEXPRESS=y
+## end choice
+
+## arch/arm/mach-vexpress/Kconfig
+CONFIG_ARCH_VEXPRESS_CA9X4=y
+
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_OPROFILE=m
+
+CONFIG_SMP=y
+CONFIG_VMSPLIT_2G=y
+CONFIG_HOTPLUG_CPU=y
+
+CONFIG_ZBOOT_ROM_TEXT=0x0
+CONFIG_ZBOOT_ROM_BSS=0x0
+
+CONFIG_NEON=y
+
+
+CONFIG_MTD=m
+CONFIG_MTD_CONCAT=m
+CONFIG_MTD_PARTITIONS=m
+CONFIG_MTD_CMDLINE_PARTS=m
+CONFIG_MTD_CHAR=m
+CONFIG_MTD_BLOCK=m
+CONFIG_MTD_CFI=m
+CONFIG_MTD_CFI_INTELEXT=m
+CONFIG_MTD_CFI_AMDSTD=m
+CONFIG_MTD_ARM_INTEGRATOR=m
+CONFIG_MISC_DEVICES=m
+
+CONFIG_NET_ETHERNET=m
+CONFIG_SMSC911X=m
+
+CONFIG_SERIO_AMBAKMI=y
+CONFIG_SERIAL_AMBA_PL011=y
+CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
+
+CONFIG_FB=y
+CONFIG_FB_ARMCLCD=y
+CONFIG_FRAMEBUFFER_CONSOLE=y
+
+CONFIG_SND_ARMAACI=m
+
+CONFIG_USB=m
+CONFIG_USB_ISP1760_HCD=m
+CONFIG_MMC=y
+CONFIG_MMC_ARMMMCI=y
+
+CONFIG_RTC_DRV_PL031=y



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20120502065556.GB5888@talon.fglan

Bug#670462: linux-2.6: Support new armhf kernel variant for Versatile Express (vexpress)

[Vagrant Cascadian]

Package: linux-2.6
Version: 3.2.15-1
Severity: wishlist
Tags: patch

Please consider a new kernel variant to support Versatile Express, an 
armhf capable platform that it is also supported by the qemu version currently 
in wheezy/sid (1.0.1+dfsg-1).

This would allow more people to test the armhf port without needing armhf 
capable hardware.

For some information about qemu support:

  https://wiki.linaro.org/PeterMaydell/QemuVersatileExpress

I've tested that the following configuration at least boots to a Debian armhf 
LTSP environment in QEMU. qemu-system-arm -M vexpress-a9 -kernel ... worked 
like a charm.

It essentially grabs the vexpress default configuration from the upstream 
kernel, perhaps this could/should be trimmed down...


live well,
  vagrant


diff -urN ../l/linux-2.6-3.2.15/debian/changelog debian/changelog
--- ../l/linux-2.6-3.2.15/debian/changelog  2012-04-25 12:52:15.0 
-0700
+++ debian/changelog2012-04-25 13:13:40.0 -0700
@@ -1,3 +1,10 @@
+linux-2.6 (3.2.15-2) UNRELEASED; urgency=low
+
+  [ Vagrant Cascadian ]
+  * [armhf/vexpress] New kernel variant to support Versatile Express.
+
+ -- Vagrant Cascadian vagr...@debian.org  Wed, 25 Apr 2012 13:12:06 -0700
+
 linux-2.6 (3.2.15-1) unstable; urgency=high
 
   * New upstream stable update:
diff -urN ../l/linux-2.6-3.2.15/debian/config/armhf/config.vexpress 
debian/config/armhf/config.vexpress
--- ../l/linux-2.6-3.2.15/debian/config/armhf/config.vexpress   1969-12-31 
16:00:00.0 -0800
+++ debian/config/armhf/config.vexpress 2012-04-23 20:02:13.0 -0700
@@ -0,0 +1,151 @@
+##
+## file: arch/arm/Kconfig
+##
+## choice: ARM system type
+CONFIG_ARCH_VEXPRESS=y
+## end choice
+
+## arch/arm/mach-vexpress/Kconfig
+CONFIG_ARCH_VEXPRESS_CA9X4=y
+
+# from vexpress_defconfig
+CONFIG_EXPERIMENTAL=y
+# CONFIG_LOCALVERSION_AUTO is not set
+CONFIG_SYSVIPC=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_CGROUPS=y
+CONFIG_CPUSETS=y
+# CONFIG_UTS_NS is not set
+# CONFIG_IPC_NS is not set
+# CONFIG_USER_NS is not set
+# CONFIG_PID_NS is not set
+# CONFIG_NET_NS is not set
+CONFIG_BLK_DEV_INITRD=y
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
+CONFIG_PROFILING=y
+CONFIG_OPROFILE=y
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_LBDAF is not set
+# CONFIG_BLK_DEV_BSG is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+CONFIG_ARCH_VEXPRESS=y
+CONFIG_ARCH_VEXPRESS_CA9X4=y
+# CONFIG_SWP_EMULATE is not set
+CONFIG_SMP=y
+CONFIG_VMSPLIT_2G=y
+CONFIG_HOTPLUG_CPU=y
+CONFIG_AEABI=y
+CONFIG_ZBOOT_ROM_TEXT=0x0
+CONFIG_ZBOOT_ROM_BSS=0x0
+CONFIG_CMDLINE=root=/dev/nfs nfsroot=10.1.69.3:/work/nfsroot ip=dhcp 
console=ttyAMA0 mem=128M
+CONFIG_VFP=y
+CONFIG_NEON=y
+# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_INET=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+# CONFIG_INET_LRO is not set
+# CONFIG_IPV6 is not set
+# CONFIG_WIRELESS is not set
+CONFIG_UEVENT_HELPER_PATH=/sbin/hotplug
+CONFIG_MTD=y
+CONFIG_MTD_CONCAT=y
+CONFIG_MTD_PARTITIONS=y
+CONFIG_MTD_CMDLINE_PARTS=y
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLOCK=y
+CONFIG_MTD_CFI=y
+CONFIG_MTD_CFI_INTELEXT=y
+CONFIG_MTD_CFI_AMDSTD=y
+CONFIG_MTD_ARM_INTEGRATOR=y
+CONFIG_MISC_DEVICES=y
+# CONFIG_SCSI_PROC_FS is not set
+CONFIG_BLK_DEV_SD=y
+# CONFIG_SCSI_LOWLEVEL is not set
+CONFIG_ATA=y
+# CONFIG_SATA_PMP is not set
+CONFIG_NETDEVICES=y
+CONFIG_NET_ETHERNET=y
+CONFIG_SMSC911X=y
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_1 is not set
+# CONFIG_WLAN is not set
+CONFIG_INPUT_EVDEV=y
+# CONFIG_SERIO_SERPORT is not set
+CONFIG_SERIO_AMBAKMI=y
+CONFIG_SERIAL_AMBA_PL011=y
+CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
+CONFIG_LEGACY_PTY_COUNT=16
+# CONFIG_HW_RANDOM is not set
+# CONFIG_HWMON is not set
+CONFIG_FB=y
+CONFIG_FB_ARMCLCD=y
+CONFIG_FRAMEBUFFER_CONSOLE=y
+CONFIG_LOGO=y
+# CONFIG_LOGO_LINUX_MONO is not set
+# CONFIG_LOGO_LINUX_VGA16 is not set
+CONFIG_SOUND=y
+CONFIG_SND=y
+CONFIG_SND_MIXER_OSS=y
+CONFIG_SND_PCM_OSS=y
+# CONFIG_SND_DRIVERS is not set
+CONFIG_SND_ARMAACI=y
+CONFIG_HID_DRAGONRISE=y
+CONFIG_HID_GYRATION=y
+CONFIG_HID_TWINHAN=y
+CONFIG_HID_NTRIG=y
+CONFIG_HID_PANTHERLORD=y
+CONFIG_HID_PETALYNX=y
+CONFIG_HID_SAMSUNG=y
+CONFIG_HID_SONY=y
+CONFIG_HID_SUNPLUS=y
+CONFIG_HID_GREENASIA=y
+CONFIG_HID_SMARTJOYPLUS=y
+CONFIG_HID_TOPSEED=y
+CONFIG_HID_THRUSTMASTER=y
+CONFIG_HID_ZEROPLUS=y
+CONFIG_USB=y
+CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
+# CONFIG_USB_DEVICE_CLASS is not set
+CONFIG_USB_MON=y
+CONFIG_USB_ISP1760_HCD=y
+CONFIG_USB_STORAGE=y
+CONFIG_MMC=y
+CONFIG_MMC_ARMMMCI=y
+CONFIG_RTC_CLASS=y
+CONFIG_RTC_DRV_PL031=y
+CONFIG_EXT2_FS=y
+CONFIG_EXT3_FS=y
+# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
+# CONFIG_EXT3_FS_XATTR is not set
+CONFIG_VFAT_FS=y
+CONFIG_TMPFS=y
+CONFIG_JFFS2_FS=y
+CONFIG_CRAMFS=y
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+CONFIG_ROOT_NFS=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+CONFIG_NLS_CODEPAGE_437=y
+CONFIG_NLS_ISO8859_1=y

Bug#660297: initramfs-tools: busybox's switch_root doesn't handle /proc or /sys moving

[Vagrant Cascadian]

Package: initramfs-tools
Version: 0.100
Severity: normal
Tags: patch

Greetings.

the change to using switch_root leaves /sys and /proc in an unmounted state.
busybox's switch_root doesn't appear to handle the /sys and /proc moving the
way that util-linux's switch_root does.

either busybox's switch_root should be fixed, initramfs-tools uses util-linux's
switch_root, or manually handle mounting the dirs, just like how it's done with
run-init. the following patch implements the last approach.

live well,
  vagrant


commit 705c5e3529536862135d90e1d710d02d750765d7
Author: Vagrant Cascadian vagr...@freegeek.org
Date:   Fri Feb 17 16:11:32 2012 -0800

Work around busybox's switch_root, which doesn't handle moving /proc and 
/sys
into the root before switching, by moving the /proc and /sys directories
directly.

Signed-off-by: Vagrant Cascadian vagr...@debian.org

diff --git a/init b/init
index 472b279..242b29d 100755
--- a/init
+++ b/init
@@ -310,13 +310,14 @@ unset readonly
 unset resume
 unset resume_offset
 
+# Move virtual filesystems over to the real filesystem
+mount -n -o move /sys ${rootmnt}/sys
+mount -n -o move /proc ${rootmnt}/proc
+
 # Chain to real filesystem
 if command -v switch_root /dev/null 21; then
exec switch_root ${rootmnt} ${init} $@ ${rootmnt}/dev/console 
${rootmnt}/dev/console
 elif command -v run-init /dev/null 21; then
-   # Move virtual filesystems over to the real filesystem
-   mount -n -o move /sys ${rootmnt}/sys
-   mount -n -o move /proc ${rootmnt}/proc
 
exec run-init ${rootmnt} ${init} $@ ${rootmnt}/dev/console 
${rootmnt}/dev/console
 fi



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20120218002313.GU3372@talon.fglan

Bug#645012: 2.6.32-38: breaks DRM/KMS with nouveau X.org driver

[Vagrant Cascadian]

fixed 645012 3.1.1-1
fixed 645012 3.1.4-1
thanks

i've got a machine with a similar card that exhibits the problem, and can now 
run tests for more information, if needed.

tested with the versions from sid and wheezy, and those seem to work fine.


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20111211020318.GE26987@talon.fglan

Bug#645012: linux-image-2.6.32-5-686: breaks DRM/KMS with nouveau X.org driver

[Vagrant Cascadian]

Package: linux-2.6
Version: 2.6.32-38
Severity: important

The stable release update unfortunately broke X.org with the nouveau driver 
with:

01:00.0 VGA compatible controller [0300]: nVidia Corporation NV6 [Vanta/Vanta 
LT] [10de:002c] (rev 15) (prog-if 00 [VGA controller])


X.org responds excruciatingly slowly to the point of being un-useable.

Downgrading to 2.6.32-35squeeze2 from security.debian.org fixes the problem.


live well,
  vagrant


-- Package-specific info:
** Version:
Linux version 2.6.32-5-686 (Debian 2.6.32-35squeeze2) (da...@debian.org) (gcc 
version 4.3.5 (Debian 4.3.5-4) ) #1 SMP Fri Sep 9 20:51:05 UTC 2011

** Command line:
BOOT_IMAGE=/boot/vmlinuz-2.6.32-5-686 
root=UUID=ec01d157-3803-48e9-9190-db68b61fc7f3 ro quiet

** Not tainted

** Kernel log:
[2.199199] sd 0:0:0:0: [sda] Attached SCSI disk
[2.216153] usb 1-1: new low speed USB device using uhci_hcd and address 2
[2.243541] sd 0:0:0:0: Attached scsi generic sg0 type 0
[2.245151] sr 1:0:0:0: Attached scsi generic sg1 type 5
[2.402847] usb 1-1: New USB device found, idVendor=047d, idProduct=102d
[2.402856] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[2.402863] usb 1-1: Product: Kensington USB/PS2 Wheel Mouse
[2.402868] usb 1-1: Manufacturer: Kensington 
[2.403114] usb 1-1: configuration #1 chosen from 1 choice
[2.435452] usbcore: registered new interface driver hiddev
[2.449871] input: Kensington  Kensington USB/PS2 Wheel Mouse as 
/devices/pci:00/:00:1f.2/usb1/1-1/1-1:1.0/input/input1
[2.450389] generic-usb 0003:047D:102D.0001: input,hidraw0: USB HID v1.10 
Mouse [Kensington  Kensington USB/PS2 Wheel Mouse] on 
usb-:00:1f.2-1/input0
[2.450473] usbcore: registered new interface driver usbhid
[2.451075] usbhid: v2.6:USB HID core driver
[2.687249] PM: Starting manual resume from disk
[2.687260] PM: Resume from partition 8:5
[2.687263] PM: Checking hibernation image.
[2.709812] PM: Error -22 checking image file
[2.709818] PM: Resume from disk failed.
[2.727716] PM: Marking nosave pages: 0009f000 - 0010
[2.727729] PM: Basic memory bitmaps created
[2.735480] PM: Basic memory bitmaps freed
[2.889220] kjournald starting.  Commit interval 5 seconds
[2.889251] EXT3-fs: mounted filesystem with ordered data mode.
[5.910304] udev[267]: starting version 164
[6.813797] processor LNXCPU:00: registered as cooling_device0
[6.817413] input: Power Button as 
/devices/LNXSYSTM:00/LNXSYBUS:00/PNP0C0C:00/input/input2
[6.817445] ACPI: Power Button [PBTN]
[6.818838] input: Power Button as 
/devices/LNXSYSTM:00/LNXPWRBN:00/input/input3
[6.818862] ACPI: Power Button [PWRF]
[6.902668] input: PC Speaker as /devices/platform/pcspkr/input/input4
[7.131769] parport_pc 00:07: reported by Plug and Play ACPI
[7.131807] parport0: PC-style at 0x378, irq 7 [PCSPP,TRISTATE,EPP]
[7.304675] pci_hotplug: PCI Hot Plug PCI Core version: 0.5
[7.396517] intel_rng: Firmware space is locked read-only. If you can't or
[7.396522] intel_rng: don't want to disable this in firmware setup, and if
[7.396525] intel_rng: you are certain that your system has a functional
[7.396528] intel_rng: RNG, try using the 'no_fwh_detect' option.
[7.414295] shpchp: Standard Hot Plug PCI Controller Driver version: 0.4
[8.842252] [drm] Initialized drm 1.1.0 20060810
[9.353634] nouveau :01:00.0: PCI INT A - GSI 18 (level, low) - IRQ 18
[9.369677] [drm] nouveau :01:00.0: Detected an NV 0 generation card 
(0x20154000)
[9.369849] [drm] nouveau :01:00.0: Attempting to load BIOS image from 
PROM
[9.487922] [drm] nouveau :01:00.0: ... appears to be valid
[9.488282] [drm] nouveau :01:00.0: BMP BIOS found
[9.488287] [drm] nouveau :01:00.0: BMP version 5.17
[9.488294] [drm] nouveau :01:00.0: Bios version 03.05.00.10
[9.488298] [drm] nouveau :01:00.0: Assuming a CRT output exists
[9.488304] [drm] nouveau :01:00.0: Probing TV encoders on I2C bus: 1
[9.497074] [drm] nouveau :01:00.0: No TV encoders found.
[9.497331] [drm] nouveau :01:00.0: Parsing VBIOS init table 0 at offset 
0x0198
[9.497377] [drm] nouveau :01:00.0: Parsing VBIOS init table 1 at offset 
0x02A1
[9.510212] [TTM] Zone  kernel: Available graphics memory: 257310 kiB.
[9.510239] [drm] nouveau :01:00.0: 16 MiB VRAM
[9.510377] agpgart-intel :00:00.0: AGP 2.0 bridge
[9.510404] agpgart-intel :00:00.0: putting AGP V2 device into 4x mode
[9.510434] nouveau :01:00.0: putting AGP V2 device into 4x mode
[9.510447] [drm] nouveau :01:00.0: 64 MiB GART (aperture)
[9.510767] [drm] nouveau :01:00.0: Allocating FIFO number 0
[9.511298] [drm] nouveau :01:00.0: nouveau_channel_alloc: initialised 
FIFO 0
[9.511386] [drm] nouveau :01:00.0: Saving VGA fonts
[9.570480] [drm] nouveau 

Bug#631762: linux-image-2.6.39-2-kirkwood: please enable CONFIG_SND_KIRKWOOD_SOC_T5325

[Vagrant Cascadian]

Package: linux-image-2.6.39-2-kirkwood
Version: 2.6.39-2
Severity: wishlist
Tags: patch

Please consider enabling sound support for the HP t5325 thin clients:

--- config.kirkwood 2011-06-26 14:14:32.0 -0700
+++ config.kirkwood.hpt5325 2011-06-26 14:19:39.0 -0700
@@ -634,6 +634,7 @@
 ##
 CONFIG_SND_KIRKWOOD_SOC=m
 CONFIG_SND_KIRKWOOD_SOC_OPENRD=m
+CONFIG_SND_KIRKWOOD_SOC_T5325=m

 ##
 ## file: unknown


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110626212202.GF10841@talon.fglan

Bug#628641: linux-image-2.6.32-5-686: crashes when closing laptop lid (dell inspiron 300m)

[Vagrant Cascadian]

On Tue, May 31, 2011 at 01:26:26AM +0100, Ben Hutchings wrote:
 On Mon, 2011-05-30 at 16:56 -0700, Vagrant Cascadian wrote:
  after upgrading the kernel, my laptop seems to crash immediately after 
  closing
  the laptop screen. 
...
  so far, it seems to be completely repeatable.
 
 Try booting without VirtualBox modules.  (I don't think it would cause
 this, but I really don't trust it.)

no luck.


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110613170258.GK21010@talon.fglan

Bug#628641: linux-image-2.6.32-5-686: crashes when closing laptop lid (dell inspiron 300m)

[Vagrant Cascadian]

On Mon, Jun 13, 2011 at 05:41:07PM -0400, Ben Hutchings wrote:
 On Mon, 2011-06-13 at 10:02 -0700, Vagrant Cascadian wrote:
  On Tue, May 31, 2011 at 01:26:26AM +0100, Ben Hutchings wrote:
   On Mon, 2011-05-30 at 16:56 -0700, Vagrant Cascadian wrote:
after upgrading the kernel, my laptop seems to crash immediately after 
closing
the laptop screen. 

 Please switch to a text console and suspend by running:
 
 echo mem  /sys/power/state

this worked without problem. i routinely invoke /usr/sbin/pm-suspend-hybrid
manually, and that works fine, though i'm not entirely sure what exactly that
calls.

i should probably be clear that previous to upgrading the kernel, all closing
the lid did was turn off the screen; the computer continued to run.
downgrading to the older kernel continues to behave that way.

(this is a different machine and different symptoms than what i reported in
#595187)
 

 If this produces an 'oops' message then send that.  If the system just
 hangs then try a partial suspend/resume as explained in
 http://www.kernel.org/doc/Documentation/power/basic-pm-debugging.txt
 (but don't bother with s2ram as that should not be needed any more).

given all that, does it make sense to try troubleshooting suspend/resume 
further?


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110613235753.GO21010@talon.fglan

Bug#630113: linux-image-2.6.39-2-kirkwood: doesn't boot on HP t5325 thin client

[Vagrant Cascadian]

Package: linux-image-2.6.39-2-kirkwood
Version: 2.6.39-2
Severity: normal

when trying to boot on an HP t5325, i get the following error:

  Error: unrecognized/unsupported machine ID (r1 = 0x020f).

  Available machine support:

  ID (hex)NAME
  0690Marvell DB-88F6281-BP Development Board
  0691Marvell RD-88F6192-NAS Development Board
  0692Marvell RD-88F6281 Reference Board
  078cMarvell 88F6281 GTW GE Board
  0a76Marvell eSATA SheevaPlug Reference Board
  0831Marvell SheevaPlug Reference Board
  0a63Marvell GuruPlug Reference Board
  0bb6Seagate FreeAgent DockStar
  085bQNAP TS-119/TS-219
  09c6QNAP TS-41x
  0b44Marvell OpenRD Ultimate Board
  0939Marvell OpenRD Client Board
  0915Marvell OpenRD Base Board
  0b1eHP t5325 Thin Client

  Please check your kernel config and/or bootloader.

i also get the same error with the 2.6.32 or 2.6.38 kernels in debian.


i'm using the following uboot environment variables/settings to boot it over
the network:

nfsbootargs=setenv bootargs console=ttyS0,115200 boot=nfs
tftpload=bootp ; tftpboot 0x80 /ltsp/armel/uImage ; tftpboot 0x110 
/ltsp/armel/uInitrd ; bootm 0x80 0x110
nfsboot=${nfsbootargs} ; run tftpload
run nfsboot

these uboot values work with the 2.6.22.18-2-armel kernel that came with the
thin client.


what other information would be useful?

thanks!


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110611052242.GG21010@talon.fglan

Bug#628641: linux-image-2.6.32-5-686: crashes when closing laptop lid (dell inspiron 300m)

[Vagrant Cascadian]

Package: linux-2.6
Version: 2.6.32-34squeeze1
Severity: normal

after upgrading the kernel, my laptop seems to crash immediately after closing
the laptop screen. i've seen it happen both when running console (screen goes
black and stays black) or when running X (screen freezes, but whatever content
was there remains visible).

previous versions (including 2.6.32-31, just confirmed) worked fine, in that
closing the lid didn't really do anything.

so far, it seems to be completely repeatable.

live well,
  vagrant


-- Package-specific info:
** Version:
Linux version 2.6.32-5-686 (Debian 2.6.32-34squeeze1) (da...@debian.org) (gcc 
version 4.3.5 (Debian 4.3.5-4) ) #1 SMP Wed May 18 07:08:50 UTC 2011

** Command line:
BOOT_IMAGE=/vmlinuz-2.6.32-5-686 root=/dev/mapper/lilwalrus-rt ro keyboard=us 
quiet

** Not tainted

** Kernel log:
[   14.514303] nsc-ircc, chip-init
[   14.514315] nsc-ircc, Found chip at base=0x02e
[   14.514339] nsc-ircc, driver loaded (Dag Brattli)
[   14.514544] nsc_ircc_open(), can't get iobase of 0x2f8
[   14.514571] nsc-ircc, Found chip at base=0x02e
[   14.514595] nsc-ircc, driver loaded (Dag Brattli)
[   14.514601] nsc_ircc_open(), can't get iobase of 0x2f8
[   14.515534] nsc-ircc 00:06: disabled
[   14.614073] lib80211: common routines for IEEE802.11 drivers
[   14.614079] lib80211_crypt: registered algorithm 'NULL'
[   15.013847] [drm] Initialized drm 1.1.0 20060810
[   15.038928] ieee80211: 802.11 data/management/control stack, git-1.1.13
[   15.038935] ieee80211: Copyright (C) 2004-2005 Intel Corporation 
jketr...@linux.intel.com
[   15.240345] ipw2100: Intel(R) PRO/Wireless 2100 Network Driver, git-1.2.2
[   15.240351] ipw2100: Copyright(c) 2003-2006 Intel Corporation
[   15.240803] ACPI: PCI Interrupt Link [LNKE] enabled at IRQ 5
[   15.240808] PCI: setting IRQ 5 as level-triggered
[   15.240816] ipw2100 :02:04.0: PCI INT A - Link[LNKE] - GSI 5 (level, 
low) - IRQ 5
[   15.241488] ipw2100: Detected Intel PRO/Wireless 2100 Network Connection
[   15.241513] ipw2100 :02:04.0: firmware: requesting ipw2100-1.3.fw
[   15.795771] i915 :00:02.0: PCI INT A - Link[LNKA] - GSI 10 (level, 
low) - IRQ 10
[   15.795781] i915 :00:02.0: setting latency timer to 64
[   15.801373] [drm] set up 7M of stolen space
[   15.950305] [drm] initialized overlay support
[   16.212227] yenta_cardbus :02:03.0: CardBus bridge found [1028:0150]
[   16.340844] yenta_cardbus :02:03.0: ISA IRQ mask 0x0018, PCI irq 10
[   16.340851] yenta_cardbus :02:03.0: Socket status: 3006
[   16.340858] pci_bus :02: Raising subordinate bus# of parent bus (#02) 
from #02 to #06
[   16.340872] yenta_cardbus :02:03.0: pcmcia: parent PCI bridge I/O 
window: 0x3000 - 0x3fff
[   16.340878] pcmcia_socket pcmcia_socket0: cs: IO port probe 0x3000-0x3fff: 
clean.
[   16.341260] yenta_cardbus :02:03.0: pcmcia: parent PCI bridge Memory 
window: 0xe020 - 0xe02f
[   16.341265] yenta_cardbus :02:03.0: pcmcia: parent PCI bridge Memory 
window: 0x4800 - 0x4fff
[   16.352161] yenta_cardbus :02:03.1: CardBus bridge found [1028:0150]
[   16.752833] yenta_cardbus :02:03.1: ISA IRQ mask 0x0018, PCI irq 10
[   16.752839] yenta_cardbus :02:03.1: Socket status: 3006
[   16.752845] pci_bus :02: Raising subordinate bus# of parent bus (#02) 
from #06 to #0a
[   16.752856] yenta_cardbus :02:03.1: pcmcia: parent PCI bridge I/O 
window: 0x3000 - 0x3fff
[   16.752862] pcmcia_socket pcmcia_socket1: cs: IO port probe 0x3000-0x3fff: 
clean.
[   16.753241] yenta_cardbus :02:03.1: pcmcia: parent PCI bridge Memory 
window: 0xe020 - 0xe02f
[   16.753246] yenta_cardbus :02:03.1: pcmcia: parent PCI bridge Memory 
window: 0x4800 - 0x4fff
[   16.932969] Console: switching to colour frame buffer device 128x48
[   16.941850] fb0: inteldrmfb frame buffer device
[   16.941853] registered panic notifier
[   16.943520] [drm] Initialized i915 1.6.0 20080730 for :00:02.0 on minor 0
[   16.943596] Intel ICH Modem :00:1f.6: PCI INT B - Link[LNKB] - GSI 10 
(level, low) - IRQ 10
[   16.943622] Intel ICH Modem :00:1f.6: setting latency timer to 64
[   16.948818] Intel ICH :00:1f.5: PCI INT B - Link[LNKB] - GSI 10 
(level, low) - IRQ 10
[   16.948856] Intel ICH :00:1f.5: setting latency timer to 64
[   17.537475] pcmcia_socket pcmcia_socket0: cs: IO port probe 0x100-0x3af: 
clean.
[   17.538347] pcmcia_socket pcmcia_socket0: cs: IO port probe 0x3e0-0x4ff: 
excluding 0x4d0-0x4d7
[   17.538744] pcmcia_socket pcmcia_socket0: cs: IO port probe 0x820-0x8ff: 
clean.
[   17.539077] pcmcia_socket pcmcia_socket0: cs: IO port probe 0xc00-0xcf7: 
clean.
[   17.539586] pcmcia_socket pcmcia_socket0: cs: IO port probe 0xa00-0xaff: 
clean.
[   17.540188] pcmcia_socket pcmcia_socket1: cs: IO port probe 0x100-0x3af: 
clean.
[   17.541053] pcmcia_socket pcmcia_socket1: cs: IO port probe 0x3e0-0x4ff: 
excluding 0x4d0-0x4d7
[   17.541451] pcmcia_socket pcmcia_socket1: 

Bug#625804: rtc/mc13xxx: don't call rtc_device_register with the lock held

[Vagrant Cascadian]

On Fri, May 06, 2011 at 11:57:47AM +0200, Uwe Kleine-König wrote:
 diff --git a/drivers/rtc/rtc-mc13xxx.c b/drivers/rtc/rtc-mc13xxx.c
 index c5ac037..a1a278b 100644
 --- a/drivers/rtc/rtc-mc13xxx.c
 +++ b/drivers/rtc/rtc-mc13xxx.c
 @@ -349,11 +349,15 @@ static int __devinit mc13xxx_rtc_probe(struct 
 platform_device *pdev)
   if (ret)
   goto err_alarm_irq_request;
  
 + mc13xxx_unlock(mc13xxx);
 +
   priv-rtc = rtc_device_register(pdev-name,
   pdev-dev, mc13xxx_rtc_ops, THIS_MODULE);
   if (IS_ERR(priv-rtc)) {
   ret = PTR_ERR(priv-rtc);
  
 + mc13xxx_lock(mc13xxx);
 +
   mc13xxx_irq_free(mc13xxx, MC13XXX_IRQ_TODA, priv);
  err_alarm_irq_request:
  
 @@ -365,12 +369,12 @@ err_reset_irq_status:
   mc13xxx_irq_free(mc13xxx, MC13XXX_IRQ_RTCRST, priv);
  err_reset_irq_request:
  
 + mc13xxx_unlock(mc13xxx);
 +
   platform_set_drvdata(pdev, NULL);
   kfree(priv);
   }
  
 - mc13xxx_unlock(mc13xxx);
 -
   return ret;
  }
  
 

  http://patchwork.ozlabs.org/patch/94354/

thanks! i just tested and confirmed that this patch also resolves the issue.


live well,
  vagrant



--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110507201648.GD9959@talon.fglan

Bug#625804: linux-2.6: lockups in rtc on armhf

[Vagrant Cascadian]

Package: linux-2.6
Version: 2.6.39~rc5-1~experimental.1
Severity: normal

i was experiencing lockups on an armhf kernel built from the experimental 
branch:

[2.644589] usb 2-1.1: SerialNumber: 1.0
[  240.664217] INFO: task swapper:1 blocked for more than 120 seconds.
[  240.670509] echo 0  /proc/sys/kernel/hung_task_timeout_secs disables this 
message.
[  240.678374] swapper D c0285240 0 1  0 0x
[  240.684824] [c0285240] (schedule+0x4bc/0x510) from [c0285ca0] 
(__mutex_lock_slowpath+0x7c/0xa0)
[  240.693914] [c0285ca0] (__mutex_lock_slowpath+0x7c/0xa0) from [c01e2bb8] 
(mc13xxx_rtc_read_time+0x18/0xe4)
[  240.703967] [c01e2bb8] (mc13xxx_rtc_read_time+0x18/0xe4) from [c01e0974] 
(__rtc_read_time+0x4c/0x50)

the following patch by Arnaud Patard seemed to resolve the problem:

  http://patchwork.ozlabs.org/patch/89644/

Index: imx-test/drivers/rtc/rtc-mc13xxx.c
===
--- imx-test.orig/drivers/rtc/rtc-mc13xxx.c
+++ imx-test/drivers/rtc/rtc-mc13xxx.c
@@ -349,8 +349,13 @@  static int __devinit mc13xxx_rtc_probe(s
if (ret)
goto err_alarm_irq_request;

+   mc13xxx_unlock(mc13xxx);
+
priv-rtc = rtc_device_register(pdev-name,
pdev-dev, mc13xxx_rtc_ops, THIS_MODULE);
+
+   mc13xxx_lock(mc13xxx);
+
if (IS_ERR(priv-rtc)) {
ret = PTR_ERR(priv-rtc);


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110506051246.GC10524@talon.fglan

Bug#621032: linux-2.6: please add armhf support

[Vagrant Cascadian]

ok, one last patch trimmed even more from the last, tested to work on the 
efikamx smarttop with ethernet, mmc and serial console.

live well,
  vagrant

 debian/config/armhf/config  |   58 
 debian/config/armhf/config.mx5  |  253 
 debian/config/armhf/defines |   17 +
 linux-2.6-2.6.38/debian/config/defines  |1
 linux-2.6-2.6.38/debian/rules.real  |3
 linux-2.6-2.6.38/debian/templates/control.main.in   |2
 linux-2.6-2.6.38/debian/templates/control.source.in |2
 7 files changed, 333 insertions(+), 3 deletions(-)

diff -u linux-2.6-2.6.38/debian/rules.real linux-2.6-2.6.38/debian/rules.real
--- linux-2.6-2.6.38/debian/rules.real
+++ linux-2.6-2.6.38/debian/rules.real
@@ -39,7 +39,7 @@
 #
 binary-arch-arch: install-headers_$(ARCH)
 binary-arch-arch: install-libc-dev_$(ARCH)
-ifneq ($(filter alpha amd64 armel hppa i386 powerpc ppc64 s390 sh4 sparc 
sparc64,$(ARCH)),)
+ifneq ($(filter alpha amd64 armel armhf hppa i386 powerpc ppc64 s390 sh4 sparc 
sparc64,$(ARCH)),)
   binary-arch-arch: install-tools_$(ARCH)
 endif
 binary-arch-featureset: install-headers_$(ARCH)_$(FEATURESET)
@@ -364,6 +364,7 @@
 
 install-image_arm_$(FEATURESET)_$(FLAVOUR)_plain_image \
 install-image_armel_$(FEATURESET)_$(FLAVOUR)_plain_image \
+install-image_armhf_$(FEATURESET)_$(FLAVOUR)_plain_image \
 install-image_sparc_$(FEATURESET)_$(FLAVOUR)_plain_image \
 install-image_sparc64_$(FEATURESET)_$(FLAVOUR)_plain_image \
 install-image_sh4_$(FEATURESET)_$(FLAVOUR)_plain_image:

diff -u linux-2.6-2.6.38/debian/templates/control.source.in 
linux-2.6-2.6.38/debian/templates/control.source.in
--- linux-2.6-2.6.38/debian/templates/control.source.in
+++ linux-2.6-2.6.38/debian/templates/control.source.in
@@ -3,7 +3,7 @@
 Maintainer: Debian Kernel Team debian-kernel@lists.debian.org
 Uploaders: Bastian Blank wa...@debian.org, Frederik Schüler 
f...@debian.org, maximilian attems m...@debian.org, Ben Hutchings 
b...@decadent.org.uk
 Standards-Version: 3.9.1
-Build-Depends: debhelper ( 7), cpio, module-init-tools, python, lzma 
[armel], libelf-dev, libdw-dev, binutils-dev, asciidoc, xmlto, libperl-dev, 
python-dev, libnewt-dev, python-support
+Build-Depends: debhelper ( 7), cpio, module-init-tools, python, lzma [armel 
armhf], libelf-dev, libdw-dev, binutils-dev, asciidoc, xmlto, libperl-dev, 
python-dev, libnewt-dev, python-support
 Build-Depends-Indep: bzip2
 Vcs-Svn: svn://svn.debian.org/svn/kernel/dists/sid/linux-2.6/
 Vcs-Browser: http://svn.debian.org/wsvn/kernel/dists/sid/linux-2.6/?op=log
diff -u linux-2.6-2.6.38/debian/templates/control.main.in 
linux-2.6-2.6.38/debian/templates/control.main.in
--- linux-2.6-2.6.38/debian/templates/control.main.in
+++ linux-2.6-2.6.38/debian/templates/control.main.in
@@ -1,5 +1,5 @@
 Package: linux-tools-@version@
-Architecture: alpha amd64 armel hppa i386 powerpc ppc64 s390 sh4 sparc sparc64
+Architecture: alpha amd64 armel armhf hppa i386 powerpc ppc64 s390 sh4 sparc 
sparc64
 Depends: ${shlibs:Depends}, ${misc:Depends}, ${perl:Depends}, ${python:Depends}
 Recommends: linux-base (= 2.6.32-16)
 Suggests: linux-doc-@version@
diff -u linux-2.6-2.6.38/debian/config/defines 
linux-2.6-2.6.38/debian/config/defines
--- linux-2.6-2.6.38/debian/config/defines
+++ linux-2.6-2.6.38/debian/config/defines
@@ -6,6 +6,7 @@
  alpha
  amd64
  armel
+ armhf
  hppa
  i386
  ia64
only in patch2:
unchanged:
--- linux-2.6-2.6.38.orig/debian/config/armhf/config.mx5
+++ linux-2.6-2.6.38/debian/config/armhf/config.mx5
@@ -0,0 +1,253 @@
+##
+## file: arch/arm/Kconfig
+##
+## choice: ARM system type
+CONFIG_ARCH_MXC=y
+
+CONFIG_IMX_HAVE_PLATFORM_FEC=y
+CONFIG_IMX_HAVE_PLATFORM_GPIO_KEYS=y
+CONFIG_IMX_HAVE_PLATFORM_IMX2_WDT=y
+CONFIG_IMX_HAVE_PLATFORM_IMX_I2C=y
+CONFIG_IMX_HAVE_PLATFORM_IMX_SSI=y
+CONFIG_IMX_HAVE_PLATFORM_IMX_UART=y
+CONFIG_IMX_HAVE_PLATFORM_MXC_PWM=y
+CONFIG_IMX_HAVE_PLATFORM_SDHCI_ESDHC_IMX=y
+CONFIG_IMX_HAVE_PLATFORM_SPI_IMX=y
+# CONFIG_IMX_HAVE_PLATFORM_IMX_IPUV3 is not set
+
+##
+## file: arch/arm/plat-mxc/Kconfig
+##
+CONFIG_ARCH_MX5=y
+##
+## file: arch/arm/mach-mx5/Kconfig
+##
+CONFIG_ARCH_MX51=y
+CONFIG_SOC_IMX51=y
+CONFIG_MACH_MX51_EFIKAMX=y
+# CONFIG_MACH_MX51_EFIKASB is not set
+##
+## file: arch/arm/mm/Kconfig
+##
+CONFIG_CPU_32v6K=y
+CONFIG_CPU_32v7=y
+CONFIG_CPU_ABRT_EV7=y
+CONFIG_CPU_PABRT_V7=y
+CONFIG_CPU_CACHE_V7=y
+CONFIG_CPU_CACHE_VIPT=y
+CONFIG_CPU_COPY_V6=y
+CONFIG_CPU_TLB_V7=y
+CONFIG_CPU_HAS_ASID=y
+CONFIG_CPU_CP15=y
+CONFIG_CPU_CP15_MMU=y
+# CONFIG_ARM_THUMBEE is not set
+# CONFIG_SWP_EMULATE is not set
+# CONFIG_CPU_ICACHE_DISABLE is not set
+# CONFIG_CPU_DCACHE_DISABLE is not set
+# CONFIG_CPU_BPREDICT_DISABLE is not set
+CONFIG_ARM_L1_CACHE_SHIFT=6
+CONFIG_ARM_DMA_MEM_BUFFERABLE=y
+# CONFIG_ARM_ERRATA_430973 is not set
+# CONFIG_ARM_ERRATA_458693 is not set
+# CONFIG_ARM_ERRATA_460075 is not set
+# CONFIG_ARM_ERRATA_743622 is not set
+
+# 

Bug#621032: linux-2.6: please add armhf support

[Vagrant Cascadian]

On Fri, Apr 15, 2011 at 09:36:40AM +0200, Bastian Blank wrote:
 On Fri, Apr 15, 2011 at 12:51:29AM +0200, Hector Oron wrote:
I propose this patch which enables basic functionality of Efika MX nettop 
which it is currently in mainline 2.6.38.
 
 The configs looks strange.

agreed.


 - Why is network support system specific?

it isn't, and should be removed. i've been working on trimming the config
considerably to get it closer to debian defaults...

i was kind of confused to see hamradio stuff defined in 
debian/config/armel/config; hector's patch just copied that, and 
then added debian/config/armhf/config.mx5 (which needs some considerable 
cleanup).


 - What is the difference between armel and armhf kernel-wise?

main differences would be:

  ## file: arch/arm/mm/Kconfig
  CONFIG_CPU_V7=y
  CONFIG_ARM_THUMB=y
  
  ## file: arch/arm/Kconfig
  CONFIG_VFP=y
  CONFIG_VFPv3=y

probably a few more configs related to those, as well.


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110415224504.GT10524@talon.fglan

Bug#621032: linux-2.6: please add armhf support

[Vagrant Cascadian]

On Fri, Apr 15, 2011 at 03:45:04PM -0700, Vagrant Cascadian wrote:
 On Fri, Apr 15, 2011 at 09:36:40AM +0200, Bastian Blank wrote:
  On Fri, Apr 15, 2011 at 12:51:29AM +0200, Hector Oron wrote:
 I propose this patch which enables basic functionality of Efika MX 
   nettop 
 which it is currently in mainline 2.6.38.
  
  The configs looks strange.
 
 agreed.
 
 
  - Why is network support system specific?
 
 it isn't, and should be removed. i've been working on trimming the config
 considerably to get it closer to debian defaults...

tested the following changes, with many config options trimmed out. it boots to 
mmc ok. it probably could be even further simplified.

patch for trimmed down debian/config/armhf/*:

 debian/config/armhf/config  |   59 +++
 debian/config/armhf/config.mx5  |  343 
 debian/config/armhf/defines |   17
 linux-2.6-2.6.38/debian/changelog   |   11
 linux-2.6-2.6.38/debian/config/defines  |1
 linux-2.6-2.6.38/debian/rules.real  |3
 linux-2.6-2.6.38/debian/templates/control.main.in   |2
 linux-2.6-2.6.38/debian/templates/control.source.in |2
 8 files changed, 435 insertions(+), 3 deletions(-)

diff -u linux-2.6-2.6.38/debian/changelog linux-2.6-2.6.38/debian/changelog
--- linux-2.6-2.6.38/debian/changelog
+++ linux-2.6-2.6.38/debian/changelog
@@ -1,3 +1,14 @@
+linux-2.6 (2.6.38-3+armhf.2) UNRELEASED; urgency=low
+
+  [ Hector Oron ]
+  * Add armhf support: enables efikamx nettop basic support. 
+
+  [ Vagrant Cascadian ]
+  * Simplify debian/config/armhf/config* 
+  * Drop check-size from debian/config/armhf/defines
+
+ -- Vagrant Cascadian vagr...@debian.org  Wed, 13 Apr 2011 17:51:28 -0700
+
 linux-2.6 (2.6.38-3) unstable; urgency=low
 
   [ Ben Hutchings ]
diff -u linux-2.6-2.6.38/debian/rules.real linux-2.6-2.6.38/debian/rules.real
--- linux-2.6-2.6.38/debian/rules.real
+++ linux-2.6-2.6.38/debian/rules.real
@@ -39,7 +39,7 @@
 #
 binary-arch-arch: install-headers_$(ARCH)
 binary-arch-arch: install-libc-dev_$(ARCH)
-ifneq ($(filter alpha amd64 armel hppa i386 powerpc ppc64 s390 sh4 sparc 
sparc64,$(ARCH)),)
+ifneq ($(filter alpha amd64 armel armhf hppa i386 powerpc ppc64 s390 sh4 sparc 
sparc64,$(ARCH)),)
   binary-arch-arch: install-tools_$(ARCH)
 endif
 binary-arch-featureset: install-headers_$(ARCH)_$(FEATURESET)
@@ -364,6 +364,7 @@
 
 install-image_arm_$(FEATURESET)_$(FLAVOUR)_plain_image \
 install-image_armel_$(FEATURESET)_$(FLAVOUR)_plain_image \
+install-image_armhf_$(FEATURESET)_$(FLAVOUR)_plain_image \
 install-image_sparc_$(FEATURESET)_$(FLAVOUR)_plain_image \
 install-image_sparc64_$(FEATURESET)_$(FLAVOUR)_plain_image \
 install-image_sh4_$(FEATURESET)_$(FLAVOUR)_plain_image:
diff -u linux-2.6-2.6.38/debian/templates/control.source.in 
linux-2.6-2.6.38/debian/templates/control.source.in
--- linux-2.6-2.6.38/debian/templates/control.source.in
+++ linux-2.6-2.6.38/debian/templates/control.source.in
@@ -3,7 +3,7 @@
 Maintainer: Debian Kernel Team debian-kernel@lists.debian.org
 Uploaders: Bastian Blank wa...@debian.org, Frederik Schüler 
f...@debian.org, maximilian attems m...@debian.org, Ben Hutchings 
b...@decadent.org.uk
 Standards-Version: 3.9.1
-Build-Depends: debhelper ( 7), cpio, module-init-tools, python, lzma 
[armel], libelf-dev, libdw-dev, binutils-dev, asciidoc, xmlto, libperl-dev, 
python-dev, libnewt-dev, python-support
+Build-Depends: debhelper ( 7), cpio, module-init-tools, python, lzma [armel 
armhf], libelf-dev, libdw-dev, binutils-dev, asciidoc, xmlto, libperl-dev, 
python-dev, libnewt-dev, python-support
 Build-Depends-Indep: bzip2
 Vcs-Svn: svn://svn.debian.org/svn/kernel/dists/sid/linux-2.6/
 Vcs-Browser: http://svn.debian.org/wsvn/kernel/dists/sid/linux-2.6/?op=log
diff -u linux-2.6-2.6.38/debian/templates/control.main.in 
linux-2.6-2.6.38/debian/templates/control.main.in
--- linux-2.6-2.6.38/debian/templates/control.main.in
+++ linux-2.6-2.6.38/debian/templates/control.main.in
@@ -1,5 +1,5 @@
 Package: linux-tools-@version@
-Architecture: alpha amd64 armel hppa i386 powerpc ppc64 s390 sh4 sparc sparc64
+Architecture: alpha amd64 armel armhf hppa i386 powerpc ppc64 s390 sh4 sparc 
sparc64
 Depends: ${shlibs:Depends}, ${misc:Depends}, ${perl:Depends}, ${python:Depends}
 Recommends: linux-base (= 2.6.32-16)
 Suggests: linux-doc-@version@
diff -u linux-2.6-2.6.38/debian/config/defines 
linux-2.6-2.6.38/debian/config/defines
--- linux-2.6-2.6.38/debian/config/defines
+++ linux-2.6-2.6.38/debian/config/defines
@@ -6,6 +6,7 @@
  alpha
  amd64
  armel
+ armhf
  hppa
  i386
  ia64
only in patch2:
unchanged:
--- linux-2.6-2.6.38.orig/debian/config/armhf/config.mx5
+++ linux-2.6-2.6.38/debian/config/armhf/config.mx5
@@ -0,0 +1,343 @@
+##
+## file: arch/arm/Kconfig
+##
+## choice: ARM system type
+# CONFIG_ARCH_AAEC2000 is not set
+# CONFIG_ARCH_INTEGRATOR is not set
+# CONFIG_ARCH_REALVIEW is not set

Bug#609926: linux-image-2.6.37-trunk-686: resume from disk reboots sometimes

[Vagrant Cascadian]

Package: linux-2.6
Version: 2.6.37-1~experimental.1
Severity: important

i've started to run the experimental kernel recently and version 2.6.37-rc7-686
seemed to be working fine (as well as a previous version), although since
updating to linux-image-2.6.37-trunk-686 it *sometimes* (1 out of 3?) reboots
immediately after resuming from disk. this results in filesystems being
uncleanly unmounted and such. i think it happens too fast for anything to hit
log files.

maybe this is related, maybe not, but i started running the experimental
kernels due to #595187, though kernel updates have been too frequent to really
know weather that's fixed (with kernel in squeeze, i could go two weeks without
the bug getting triggered), or this may be an exadgerrated version of the same
thing?


live well,
  vagrant


-- Package-specific info:
** Version:
Linux version 2.6.37-trunk-686 (Debian 2.6.37-1~experimental.1) 
(b...@decadent.org.uk) (gcc version 4.4.5 (Debian 4.4.5-10) ) #1 SMP Thu Jan 6 
14:39:08 UTC 2011

** Command line:
BOOT_IMAGE=/vmlinuz-2.6.37-trunk-686 root=/dev/mapper/mneme-rwt ro quiet 
memory_corruption_check=1

** Not tainted

** Kernel log:
[   20.589142] ACPI: AC Adapter [AC] (on-line)
[   20.766545] pci_hotplug: PCI Hot Plug PCI Core version: 0.5
[   20.777220] dcdbas dcdbas: Dell Systems Management Base Driver (version 
5.6.0-3.2)
[   21.106747] intel_rng: FWH not detected
[   21.135744] input: Video Bus as 
/devices/LNXSYSTM:00/device:00/PNP0A03:00/LNXVIDEO:00/input/input4
[   21.135891] ACPI: Video Device [VID] (multi-head: yes  rom: no  post: no)
[   21.135925] [Firmware Bug]: Duplicate ACPI video bus devices for the same 
VGA controller, please try module parameter video.allow_duplicates=1if the 
current driver doesn't work.
[   21.141563] yenta_cardbus :02:01.0: CardBus bridge found [1028:01d6]
[   21.269279] yenta_cardbus :02:01.0: ISA IRQ mask 0x0cb8, PCI irq 19
[   21.269287] yenta_cardbus :02:01.0: Socket status: 3006
[   21.269294] pci_bus :02: Raising subordinate bus# of parent bus (#02) 
from #03 to #06
[   21.269308] yenta_cardbus :02:01.0: pcmcia: parent PCI bridge window: 
[io  0x5000-0x5fff]
[   21.269314] pcmcia_socket pcmcia_socket0: cs: IO port probe 0x5000-0x5fff: 
excluding 0x5000-0x50ff 0x5400-0x54ff
[   21.276596] yenta_cardbus :02:01.0: pcmcia: parent PCI bridge window: 
[mem 0xdfb0-0xdfbf]
[   21.276603] pcmcia_socket pcmcia_socket0: cs: memory probe 
0xdfb0-0xdfbf: excluding 0xdfb0-0xdfb0 0xdfbf-0xdfbf
[   21.276628] yenta_cardbus :02:01.0: pcmcia: parent PCI bridge window: 
[mem 0x4000-0x43ff pref]
[   21.276634] pcmcia_socket pcmcia_socket0: cs: memory probe 
0x4000-0x43ff: excluding 0x4000-0x43ff
[   21.303060] shpchp: Standard Hot Plug PCI Controller Driver version: 0.4
[   21.337352] i801_smbus :00:1f.3: PCI INT B - GSI 17 (level, low) - IRQ 
17
[   21.696309] leds_ss4200: no LED devices found
[   22.012383] input: DualPoint Stick as 
/devices/platform/i8042/serio1/input/input5
[   22.032781] input: AlpsPS/2 ALPS DualPoint TouchPad as 
/devices/platform/i8042/serio1/input/input6
[   22.113067] cfg80211: Calling CRDA to update world regulatory domain
[   22.259558] [drm] Initialized drm 1.1.0 20060810
[   22.878474] i915 :00:02.0: PCI INT A - GSI 16 (level, low) - IRQ 16
[   22.878484] i915 :00:02.0: setting latency timer to 64
[   23.088566] vgaarb: device changed decodes: 
PCI::00:02.0,olddecodes=io+mem,decodes=io+mem:owns=io+mem
[   23.089313] [drm] initialized overlay support
[   23.168725] pcmcia_socket pcmcia_socket0: cs: IO port probe 0x100-0x3af: 
excluding 0x170-0x177 0x1f0-0x1f7 0x370-0x377
[   23.170934] pcmcia_socket pcmcia_socket0: cs: IO port probe 0x3e0-0x4ff: 
excluding 0x3f0-0x3f7 0x4d0-0x4d7
[   23.171892] pcmcia_socket pcmcia_socket0: cs: IO port probe 0x820-0x8ff: 
clean.
[   23.172726] pcmcia_socket pcmcia_socket0: cs: IO port probe 0xc00-0xcf7: 
excluding 0xc80-0xcbf
[   23.173405] pcmcia_socket pcmcia_socket0: cs: memory probe 
0x0c-0x0f: excluding 0xc-0xc7fff 0xcc000-0xc 0xf-0xf
[   23.173504] pcmcia_socket pcmcia_socket0: cs: memory probe 
0xa000-0xa0ff: clean.
[   23.173602] pcmcia_socket pcmcia_socket0: cs: memory probe 
0x6000-0x60ff: clean.
[   23.173705] pcmcia_socket pcmcia_socket0: cs: IO port probe 0xa00-0xaff: 
clean.
[   23.516938] Console: switching to colour frame buffer device 160x50
[   23.523560] b43-phy0: Broadcom 4311 WLAN found (core revision 10)
[   23.524106] fb0: inteldrmfb frame buffer device
[   23.524109] drm: registered panic notifier
[   23.524315] [drm] Initialized i915 1.6.0 20080730 for :00:02.0 on minor 0
[   23.851901] ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
[   23.852923] Registered led device: b43-phy0::tx
[   23.852969] Registered led device: b43-phy0::rx
[   23.853023] Registered led device: b43-phy0::radio
[   23.853049] Broadcom 43xx driver 

Bug#595187: linux-2.6: lots of OOPses on resume (NULL pointer dereference in file_ra_state_init)

[Vagrant Cascadian]

On Mon, Sep 27, 2010 at 09:29:28AM -0700, Vagrant Cascadian wrote:
 On Tue, Sep 21, 2010 at 01:37:42PM -0700, Vagrant Cascadian wrote:
  On Mon, Sep 20, 2010 at 02:04:23AM +0100, Ben Hutchings wrote:
   Please add 'memory_corruption_check=1' to the kernel parameters and
   report whether that fixes the problem and whether it results in any new
   log messages.
  
  added this now, will reboot and see if that fixes it... presuming it just
  doesn't hide itself for another several weeks.
 
 ok, didn't have to wait for weeks, apparently.

still getting crashes every week or two, but not sure if they're related to the
initial problem...

so i've been running with memory_corruption_check=1:

cat /proc/cmdline
BOOT_IMAGE=/vmlinuz-2.6.32-5-686 root=/dev/mapper/mneme-rwt ro quiet 
memory_corruption_check=1


Oct  5 19:14:30 mneme kernel: [ 8525.133061] BUG: unable to handle kernel NULL 
pointer dereference at (null)
Oct  5 19:14:30 mneme kernel: [ 8525.133072] IP: [c113a802] strlen+0x8/0x11
Oct  5 19:14:30 mneme kernel: [ 8525.133083] *pde = 
Oct  5 19:14:30 mneme kernel: [ 8525.133088] Oops:  [#1] SMP
Oct  5 19:14:30 mneme kernel: [ 8525.133094] last sysfs file: 
/sys/devices/pci:00/:00:02.1/resource
Oct  5 19:14:30 mneme kernel: [ 8525.133099] Modules linked in: usbhid hid tun 
ip6table_filter ip6_tables iptable_filter ip_tables x_tables sco bridge stp bnep
rfcomm l2cap crc16 bluetooth acpi_cpufreq parport_pc ppdev lp cpufreq_stats 
cpufreq_powersave parport cpufreq_conservative cpufreq_userspace kvm_intel kvm 
uinpu
t fuse dm_snapshot firewire_sbp2 loop snd_hda_codec_idt snd_hda_intel 
snd_hda_codec snd_hwdep snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_midi 
snd_rawmidi snd_seq
_midi_event i915 snd_seq arc4 drm_kms_helper ecb snd_timer b43 snd_seq_device 
drm snd yenta_socket rsrc_nonstatic i2c_i801 i2c_algo_bit soundcore mac80211 
rng_c
ore snd_page_alloc dell_laptop joydev i2c_core cfg80211 rfkill video battery 
processor button output dcdbas psmouse ac evdev serio_raw ext3 jbd mbcache 
sha256_g
eneric aes_i586 aes_generic cbc dm_crypt dm_mod sd_mod crc_t10dif ata_generic 
ata_piix sdhci_pci sdhci thermal ssb firewire_ohci pcmcia libata mmc_core tg3 
uhci
_hcd firewire_core crc_itu_t pcmcia_core scsi_mod led_class ehci_h
Oct  5 19:14:30 mneme kernel: cd thermal_sys libphy usbcore nls_base [last 
unloaded: scsi_wait_scan]
Oct  5 19:14:30 mneme kernel: [ 8525.133242]
Oct  5 19:14:30 mneme kernel: [ 8525.133248] Pid: 10491, comm: Xorg Not tainted 
(2.6.32-5-686 #1) Latitude D420
Oct  5 19:14:30 mneme kernel: [ 8525.133254] EIP: 0060:[c113a802] EFLAGS: 
00213246 CPU: 0
Oct  5 19:14:30 mneme kernel: [ 8525.133259] EIP is at strlen+0x8/0x11
Oct  5 19:14:30 mneme kernel: [ 8525.133263] EAX:  EBX: 0fd4 ECX: 
 EDX: 0005
Oct  5 19:14:30 mneme kernel: [ 8525.133268] ESI:  EDI:  EBP: 
f6f533f0 ESP: d5871f4c
Oct  5 19:14:30 mneme kernel: [ 8525.133272]  DS: 007b ES: 007b FS: 00d8 GS: 
00e0 SS: 0068
Oct  5 19:14:30 mneme kernel: [ 8525.133278] Process Xorg (pid: 10491, 
ti=d587 task=f335d540 task.ti=d587)
Oct  5 19:14:30 mneme kernel: [ 8525.133282] Stack:
Oct  5 19:14:30 mneme kernel: [ 8525.133285]  f0790e00 c10f2aee c10bd768 
d5871f90  c128167c f0790e00 e0537d88
Oct  5 19:14:30 mneme kernel: [ 8525.133295] 0 e0537e00 c10bd9fb d5871f90 
c10bd768 fff7 f0790e00  8000
Oct  5 19:14:30 mneme kernel: [ 8525.133306] 0 c10bda8b 0a0882a8 0a088288 
7d28 ffea 000c 0a087fd0 b76afff4
Oct  5 19:14:30 mneme kernel: [ 8525.133318] Call Trace:
Oct  5 19:14:30 mneme kernel: [ 8525.133328]  [c10f2aee] ? 
sysfs_readdir+0xe0/0x13a
Oct  5 19:14:30 mneme kernel: [ 8525.16]  [c10bd768] ? filldir64+0x0/0xc5
Oct  5 19:14:30 mneme kernel: [ 8525.133342]  [c10bd9fb] ? 
vfs_readdir+0x62/0x8c
Oct  5 19:14:30 mneme kernel: [ 8525.133349]  [c10bd768] ? filldir64+0x0/0xc5
Oct  5 19:14:30 mneme kernel: [ 8525.133355]  [c10bda8b] ? 
sys_getdents64+0x66/0xa5
Oct  5 19:14:30 mneme kernel: [ 8525.133363]  [c10030fb] ? 
sysenter_do_call+0x12/0x28
Oct  5 19:14:30 mneme kernel: [ 8525.133367] Code: eb 04 19 c0 0c 01 5e 5f c3 
56 89 c6 89 d0 88 c4 ac 38 e0 74 09 84 c0 75 f7 be 01 00 00 00 89 f0 48 5e c3 
57 8
3 c9 ff 89 c7 31 c0 f2 ae f7 d1 49 89 c8 5f c3 57 31 ff 85 c9 74 0e 89 c7 89 
d0 f2
Oct  5 19:14:30 mneme kernel: [ 8525.133427] EIP: [c113a802] strlen+0x8/0x11 
SS:ESP 0068:d5871f4c
Oct  5 19:14:30 mneme kernel: [ 8525.133434] CR2: 
Oct  5 19:14:30 mneme kernel: [ 8525.133439] ---[ end trace e4f8adeee260d138 
]---

and again today:

Oct  6 08:20:00 mneme kernel: [ 5643.886857] BUG: unable to handle kernel NULL 
pointer dereference at 0010
Oct  6 08:20:00 mneme kernel: [ 5643.887055] IP: [c108cf4b] 
file_ra_state_init+0x3/0x18
Oct  6 08:20:00 mneme kernel: [ 5643.890573] *pde =  
Oct  6 08:20:00 mneme kernel: [ 5643.890573] Oops:  [#1] SMP 
Oct  6 08:20:00state_init+0x3/0x18
Oct  6 08:20:00 mneme kernel: [ 5643.890573

Bug#595187: linux-2.6: lots of OOPses on resume (NULL pointer dereference in file_ra_state_init)

[Vagrant Cascadian]

On Tue, Sep 21, 2010 at 01:37:42PM -0700, Vagrant Cascadian wrote:
 On Mon, Sep 20, 2010 at 02:04:23AM +0100, Ben Hutchings wrote:
 
  Please add 'memory_corruption_check=1' to the kernel parameters and
  report whether that fixes the problem and whether it results in any new
  log messages.
 
 added this now, will reboot and see if that fixes it... presuming it just
 doesn't hide itself for another several weeks.

ok, didn't have to wait for weeks, apparently.

the error message was different this time...  also different is that it failed
on starting X.org, rather than starting my window manager (tritium)
sucessfully, but failing to start an x-terminal-emulator.

is it a new issue, or the same issue triggered differently due to
memory_corruption_check=1 ?

this froze the system with periodic disk activity shortly after resuming from
disk on with linux-image-2.6.32-5-686 2.6.32-23:


Sep 26 19:51:52 mneme kernel: [92514.768255] [ cut here 
]
Sep 26 19:51:52 mneme kernel: [92514.768268] WARNING: at 
/build/buildd-linux-2.6_2.6.32-23-i386-x1D1UQ/linux-2.6-2.6.32/debian/build/source_i386_none/fs/sysfs/file.c:355
 sysfs_open_file+0x91/0x259()
Sep 26 19:51:52 mneme kernel: [92514.768275] Hardware name: Latitude D420   

Sep 26 19:51:52 mneme kernel: [92514.768278] missing sysfs attribute operations 
for kobject: NULL
Sep 26 19:51:52 mneme kernel: [92514.768283] Modules linked in: mct_u232 
usbserial ext2 hfs hfsplus vfat fat isofs nls_utf8 udf usb_storage usbhid hid 
tun ip6table_filter ip6_tables iptable_filter ip_tables x_tables sco bridge stp 
bnep parport_pc ppdev lp parport l2cap crc16 bluetooth acpi_cpufreq 
cpufreq_stats cpufreq_powersave cpufreq_conservative cpufreq_userspace 
kvm_intel kvm uinput fuse dm_snapshot firewire_sbp2 loop snd_hda_codec_idt 
snd_hda_intel snd_hda_codec snd_hwdep snd_pcm_oss arc4 snd_mixer_oss ecb 
snd_pcm snd_seq_midi snd_rawmidi snd_seq_midi_event b43 i915 snd_seq 
drm_kms_helper snd_timer joydev snd_seq_device drm mac80211 yenta_socket 
rsrc_nonstatic i2c_i801 snd i2c_algo_bit soundcore snd_page_alloc psmouse 
i2c_core cfg80211 video battery processor button ac dell_laptop serio_raw evdev 
rng_core dcdbas output rfkill ext3 jbd mbcache sha256_generic aes_i586 
aes_generic cbc dm_crypt dm_mod sd_mod crc_t10dif ata_generic ssb sdhci_pci 
firewire_ohci pcmcia thermal tg3 ata_piix uhci_hcd sdhci firewire
Sep 26 19:51:52 mneme kernel: _core crc_itu_t mmc_core led_class thermal_sys 
libphy libata scsi_mod ehci_hcd pcmcia_core usbcore nls_base [last unloaded: 
scsi_wait_scan]
Sep 26 19:51:52 mneme kernel: [92514.768448] Pid: 12333, comm: Xorg Not tainted 
2.6.32-5-686 #1
Sep 26 19:51:52 mneme kernel: [92514.768452] Call Trace:
Sep 26 19:51:52 mneme kernel: [92514.768461]  [c103014d] ? 
warn_slowpath_common+0x5e/0x8a
Sep 26 19:51:52 mneme kernel: [92514.768468]  [c10301ab] ? 
warn_slowpath_fmt+0x26/0x2a
Sep 26 19:51:52 mneme kernel: [92514.768475]  [c10f2050] ? 
sysfs_open_file+0x91/0x259
Sep 26 19:51:52 mneme kernel: [92514.768482]  [c10b1347] ? 
__dentry_open+0x156/0x246
Sep 26 19:51:52 mneme kernel: [92514.768489]  [c10b14c8] ? 
nameidata_to_filp+0x29/0x3c
Sep 26 19:51:52 mneme kernel: [92514.768495]  [c10f1fbf] ? 
sysfs_open_file+0x0/0x259
Sep 26 19:51:52 mneme kernel: [92514.768502]  [c10bb470] ? 
do_filp_open+0x43f/0x802
Sep 26 19:51:52 mneme kernel: [92514.768509]  [c113ae03] ? 
copy_to_user+0x29/0xf8
Sep 26 19:51:52 mneme kernel: [92514.768515]  [c10b9edc] ? 
vfs_readlink+0x2f/0x40
Sep 26 19:51:52 mneme kernel: [92514.768521]  [c10b9f7a] ? 
generic_readlink+0x48/0x6f
Sep 26 19:51:52 mneme kernel: [92514.768528]  [c10c3181] ? alloc_fd+0x52/0xb7
Sep 26 19:51:52 mneme kernel: [92514.768534]  [c10b10ff] ? 
do_sys_open+0x4c/0xdf
Sep 26 19:51:52 mneme kernel: [92514.768540]  [c10b11d6] ? sys_open+0x1e/0x23
Sep 26 19:51:52 mneme kernel: [92514.768547]  [c10030fb] ? 
sysenter_do_call+0x12/0x28
Sep 26 19:51:52 mneme kernel: [92514.768552] ---[ end trace 9cabd45097cf0d9d 
]---
Sep 26 19:51:52 mneme kernel: [92514.768704] [ cut here 
]
Sep 26 19:51:52 mneme kernel: [92514.768712] WARNING: at 
/build/buildd-linux-2.6_2.6.32-23-i386-x1D1UQ/linux-2.6-2.6.32/debian/build/source_i386_none/fs/sysfs/file.c:355
 sysfs_open_file+0x91/0x259()
Sep 26 19:51:52 mneme kernel: [92514.768718] Hardware name: Latitude D420   

Sep 26 19:51:52 mneme kernel: [92514.768722] missing sysfs attribute operations 
for kobject: NULL
Sep 26 19:51:52 mneme kernel: [92514.768725] Modules linked in: mct_u232 
usbserial ext2 hfs hfsplus vfat fat isofs nls_utf8 udf usb_storage usbhid hid 
tun ip6table_filter ip6_tables iptable_filter ip_tables x_tables sco bridge stp 
bnep parport_pc ppdev lp parport l2cap crc16 bluetooth acpi_cpufreq 
cpufreq_stats cpufreq_powersave cpufreq_conservative cpufreq_userspace 
kvm_intel kvm uinput fuse dm_snapshot firewire_sbp2 loop snd_hda_codec_idt 
snd_hda_intel snd_hda_codec snd_hwdep snd_pcm_oss arc4

Bug#595187: linux-2.6: lots of OOPses on resume (NULL pointer dereference in file_ra_state_init)

[Vagrant Cascadian]

On Mon, Sep 20, 2010 at 02:04:23AM +0100, Ben Hutchings wrote:
 On Mon, 2010-09-06 at 17:05 -0700, Vagrant Cascadian wrote:
  On Tue, Sep 07, 2010 at 12:10:09AM +0100, Ben Hutchings wrote:
   On Mon, 2010-09-06 at 12:27 -0700, Vagrant Cascadian wrote:
...snip...
  i also didn't mention, largely because i'm pretty unsure if it's really the
  case, but the issue *seems* to occur more reliably when hibernating (to 
  disk)
  while on battery, but i'm really unsure of that.

that didn't appear to really make any difference... i tried all combinations of
hibernate with AC, with battery, and resume with AC, with battery, and was
unable to reproduce it reliably. in fact, i hadn't seen it since my last
comment on the bug report nearly two weeks ago... until today. gah.

so it's not exactly easy to reproduce. :(


  i'll see if i can more reliably trigger the same problem on a clean squeeze
  install on the same hardware in another partition... where i'm not as 
  worried
  about crashing.

i was also unable to reproduce it on a clean squeeze install either, so it's
probably something particular with my existing configuration or useage
patterns.

the test install was just on another partition, and my typical environment is
using lvm on an encrypted volume, if that seems likely to make any difference.
my typical environment is also upgraded from a lenny install.

 
 Please add 'memory_corruption_check=1' to the kernel parameters and
 report whether that fixes the problem and whether it results in any new
 log messages.

added this now, will reboot and see if that fixes it... presuming it just
doesn't hide itself for another several weeks.


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100921203742.gb3...@talon.fglan

Bug#595187: linux-2.6: lots of OOPses on resume (NULL pointer dereference in file_ra_state_init)

[Vagrant Cascadian]

Package: linux-2.6
Version: 2.6.32-21

i'm experiencing very similar issues, although also with earlier versions of
linux-image-2.6.32-5-686.

i first started having this problem after upgrading to squeeze from lenny,
although i was running the exact same linux-image-2.6.32-* directly on lenny
for quite some time without problems. 

i had been using uswsusp, but after experiencing this problem several times, i
purged uswsusp and switched to using the in-kernel resume with pm-hibernate.

an example OOPS below, if it's at all useful:

Sep  5 19:00:35 mneme kernel: [ 7750.522234] BUG: unable to handle kernel NULL 
pointer dereference at 0010
Sep  5 19:00:35 mneme kernel: [ 7750.522243] IP: [c108cdc7] 
file_ra_state_init+0x3/0x18
Sep  5 19:00:35 mneme kernel: [ 7750.522256] *pde =  
Sep  5 19:00:35 mneme kernel: [ 7750.522261] Oops:  [#3] SMP 
Sep  5 19:00:35 mneme kernel: [ 7750.522266] last sysfs file: 
/sys/devices/system/cpu/cpu1/cache/index2/shared_cpu_map
Sep  5 19:00:35 mneme kernel: [ 7750.522272] Modules linked in: mct_u232 
usbserial tun ip6table_filter ip6_tables iptable_filter ip_tables x_tables 
parport_pc ppdev lp parport sco bridge stp bnep rfcomm l2cap crc16 bluetooth 
acpi_cpufreq cpufreq_stats cpufreq_powersave cpufreq_conservative 
cpufreq_userspace kvm_intel kvm uinput fuse firewire_sbp2 loop 
snd_hda_codec_idt snd_hda_intel snd_hda_codec snd_hwdep snd_pcm_oss 
snd_mixer_oss snd_pcm snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq i915 
snd_timer drm_kms_helper joydev snd_seq_device arc4 ecb drm b43 snd 
yenta_socket soundcore i2c_i801 i2c_algo_bit psmouse snd_page_alloc 
rsrc_nonstatic mac80211 dell_laptop i2c_core rng_core serio_raw evdev dcdbas 
video cfg80211 output button processor battery ac rfkill ext3 jbd mbcache 
sha256_generic aes_i586 aes_generic cbc dm_crypt dm_mod sd_mod crc_t10dif 
ata_generic ssb sdhci_pci sdhci ata_piix mmc_core firewire_ohci tg3 libata 
pcmcia thermal uhci_hcd led_class firewire_core crc_itu_t libphy scsi_mod 
ehci_hcd therma
Sep  5 19:00:35 mneme kernel: l_sys pcmcia_core usbcore nls_base [last 
unloaded: scsi_wait_scan]
Sep  5 19:00:35 mneme kernel: [ 7750.522412] 
Sep  5 19:00:35 mneme kernel: [ 7750.522418] Pid: 10665, comm: x-terminal-emul 
Tainted: G  D(2.6.32-5-686 #1) Latitude D420   
Sep  5 19:00:35 mneme kernel: [ 7750.522424] EIP: 0060:[c108cdc7] EFLAGS: 
00210202 CPU: 1
Sep  5 19:00:35 mneme kernel: [ 7750.522429] EIP is at 
file_ra_state_init+0x3/0x18
Sep  5 19:00:35 mneme kernel: [ 7750.522434] EAX: ef7f45c8 EBX:  ECX: 
d5699800 EDX: 
Sep  5 19:00:35 mneme kernel: [ 7750.522438] ESI: ef7f4580 EDI:  EBP: 
f6f73b18 ESP: c49a3ea0
Sep  5 19:00:35 mneme kernel: [ 7750.522443]  DS: 007b ES: 007b FS: 00d8 GS: 
00e0 SS: 0068
Sep  5 19:00:35 mneme kernel: [ 7750.522448] Process x-terminal-emul (pid: 
10665, ti=c49a2000 task=efaa1100 task.ti=c49a2000)
Sep  5 19:00:35 mneme kernel: [ 7750.522452] Stack:
Sep  5 19:00:35 mneme kernel: [ 7750.522455]  c10b114b f6f28080 f68bb220 
c49a3f00 ef7f4580 c49a3f00 c49a3f00 0003
Sep  5 19:00:35 mneme kernel: [ 7750.522465] 0 c10b12ac ef7f4580 c10b4c40 
f6cdfe80  c49a3f00 c10bb254 
Sep  5 19:00:35 mneme kernel: [ 7750.522476] 0 0002 efa7a000 ff9c 
c1c7cbc0 b7112000  fffa0844 ef992898
Sep  5 19:00:35 mneme kernel: [ 7750.522488] Call Trace:
Sep  5 19:00:35 mneme kernel: [ 7750.522495]  [c10b114b] ? 
__dentry_open+0x176/0x246
Sep  5 19:00:35 mneme kernel: [ 7750.522502]  [c10b12ac] ? 
nameidata_to_filp+0x29/0x3c
Sep  5 19:00:35 mneme kernel: [ 7750.522509]  [c10b4c40] ? 
chrdev_open+0x0/0x116
Sep  5 19:00:35 mneme kernel: [ 7750.522516]  [c10bb254] ? 
do_filp_open+0x43f/0x802
Sep  5 19:00:35 mneme kernel: [ 7750.522524]  [c10c2f65] ? alloc_fd+0x52/0xb7
Sep  5 19:00:35 mneme kernel: [ 7750.522530]  [c10b0ee3] ? 
do_sys_open+0x4c/0xdf
Sep  5 19:00:35 mneme kernel: [ 7750.522536]  [c10b0fba] ? sys_open+0x1e/0x23
Sep  5 19:00:35 mneme kernel: [ 7750.522543]  [c10030fb] ? 
sysenter_do_call+0x12/0x28
Sep  5 19:00:35 mneme kernel: [ 7750.522547] Code: c3 53 89 d3 ff 74 24 0c ff 
74 24 0c e8 0e 93 fa ff 5a 59 85 c0 75 0e 85 db 74 0a c7 05 34 fc 4a c1 00 00 
00 00 5b c3 90 8b 52 40 8b 52 10 c7 40 14 ff ff ff ff c7 40 18 ff ff ff ff 89 
50 0c c3 
Sep  5 19:00:35 mneme kernel: [ 7750.522607] EIP: [c108cdc7] 
file_ra_state_init+0x3/0x18 SS:ESP 0068:c49a3ea0
Sep  5 19:00:35 mneme kernel: [ 7750.522616] CR2: 0010
Sep  5 19:00:35 mneme kernel: [ 7750.522620] ---[ end trace f383376f90ced1d2 
]---

live well,
  vagrant

-- Package-specific info:
** Version:
Linux version 2.6.32-5-686 (Debian 2.6.32-21) (b...@decadent.org.uk) (gcc 
version 4.3.5 (Debian 4.3.5-2) ) #1 SMP Wed Aug 25 14:28:12 UTC 2010

** Command line:
BOOT_IMAGE=/vmlinuz-2.6.32-5-686 root=/dev/mapper/mneme-rwt ro quiet

** Not tainted

** Kernel log:
[ 5400.080272] CPU: Physical Processor ID: 0
[ 5400.080272] CPU: Processor Core ID: 1
[ 5400.080272] CPU1: 

Bug#595187: linux-2.6: lots of OOPses on resume (NULL pointer dereference in file_ra_state_init)

[Vagrant Cascadian]

On Tue, Sep 07, 2010 at 12:10:09AM +0100, Ben Hutchings wrote:
 On Mon, 2010-09-06 at 12:27 -0700, Vagrant Cascadian wrote:
  Package: linux-2.6
  Version: 2.6.32-21
  
  i'm experiencing very similar issues, although also with earlier versions of
  linux-image-2.6.32-5-686.
  
  i first started having this problem after upgrading to squeeze from lenny,
  although i was running the exact same linux-image-2.6.32-* directly on lenny
  for quite some time without problems. 
  
  i had been using uswsusp, but after experiencing this problem several 
  times, i
  purged uswsusp and switched to using the in-kernel resume with pm-hibernate.
 [...]
 
 And did that make any difference?

heh. sorry for being unclear :)

unfortunately, no, it's still Oops'ing after resume.

i also didn't mention, largely because i'm pretty unsure if it's really the
case, but the issue *seems* to occur more reliably when hibernating (to disk)
while on battery, but i'm really unsure of that.

i'll see if i can more reliably trigger the same problem on a clean squeeze
install on the same hardware in another partition... where i'm not as worried
about crashing.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100907000545.gx28...@claws.fglan

Bug#584583: initramfs-tools: configure_networking function: repeatedly makes DHCP requests

[Vagrant Cascadian]

thanks to Petter Reinholdtsen, i think i've figured out the initramfs-tools
portion of this problem (there may still be outstanding issues with ipconfig).

tested the attached patch, which seems to address the issue for me at least.

live well,
  vagrant
commit 71e197686aa9048633dac7f853a74ecbf6ac6d4a
Author: Vagrant Cascadian vagr...@freegeek.org
Date:   Fri Aug 20 12:35:43 2010 -0700

configure_networking function: Look for presence of /tmp/net-*.conf files when
$DEVICE is not set (which is now default), otherwise ipconfig may receive a
valid DHCP response, but fails to break out of the loop and is called
repeatedly.

Closes: #584583

diff --git a/scripts/functions b/scripts/functions
index 1e2aeee..c497ffa 100644
--- a/scripts/functions
+++ b/scripts/functions
@@ -372,9 +372,11 @@ configure_networking()
 		# The NIC is to be configured if this file does not exist.
 		# Ip-Config tries to create this file and when it succeds
 		# creating the file, ipconfig is not run again.
-		if [ -e /tmp/net-${DEVICE}.conf ]; then
-			break;
-		fi
+		for x in /tmp/net-${DEVICE}.conf /tmp/net-*.conf ; do
+			if [ -e $x ]; then
+break 2
+			fi
+		done
 
 		case ${IP} in
 		none|off)

Bug#574510: NMU to fix: statd running as root after remove/install

[Vagrant Cascadian]

tags 574510 pending
thanks

Uploaded an NMU to DELAYED 3, with the patch previously submitted applied (with
some whitespace sanitizing) to fix this RC bug:

diff -Nru nfs-utils-1.2.2/debian/changelog nfs-utils-1.2.2/debian/changelog
--- nfs-utils-1.2.2/debian/changelog2010-04-06 02:15:29.0 -0700
+++ nfs-utils-1.2.2/debian/changelog2010-07-12 13:53:04.0 -0700
@@ -1,3 +1,11 @@
+nfs-utils (1:1.2.2-1.1) unstable; urgency=low
+
+  * Non-maintainer upload.
+  * Ensure files under /var/lib/nfs/ are owned by statd user. 
+Closes: #574510 
+
+ -- Vagrant Cascadian vagr...@debian.org  Mon, 12 Jul 2010 20:52:13 +
+
 nfs-utils (1:1.2.2-1) unstable; urgency=low
 
   [ Anibal Monsalve Salazar ]
diff -Nru nfs-utils-1.2.2/debian/nfs-common.postinst 
nfs-utils-1.2.2/debian/nfs-common.postinst
--- nfs-utils-1.2.2/debian/nfs-common.postinst  2010-01-12 19:18:00.0 
-0800
+++ nfs-utils-1.2.2/debian/nfs-common.postinst  2010-07-12 13:51:59.0 
-0700
@@ -23,15 +23,13 @@
 rmdir --ignore-fail-on-non-empty /home/statd
 fi
fi
-   if [ $2 =  ] || dpkg --compare-versions $2 lt 1:1.0.7-16; then
-   chown statd /var/lib/nfs/sm \
+   chown statd /var/lib/nfs/sm \
/var/lib/nfs/sm.bak \
/var/lib/nfs/rpc_pipefs \
/var/lib/nfs
-if [ -f /var/lib/nfs/state ]; then
+if [ -f /var/lib/nfs/state ]; then
chown statd /var/lib/nfs/state
-fi
-   fi
+fi

if [ $2 !=  ] || dpkg --compare-versions $2 lt 1:1.1.0-14; then
if dpkg-statoverride --list /sbin/mount.nfs /dev/null 21; then


live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100712215308.ge9...@claws.fglan

Bug#574510: #574510: statd running as root after remove/install

[Vagrant Cascadian]

Package: nfs-common
Followup-For: Bug #574510

this patch partially addresses the issue by ensuring that the files are owned
by statd no matter what version is installed, even if a previous version was
installed.

--- nfs-common.postinst.orig2010-06-20 18:29:11.0 -0700
+++ nfs-common.postinst 2010-06-20 18:28:44.0 -0700
@@ -23,7 +23,6 @@
 rmdir --ignore-fail-on-non-empty /home/statd
 fi
fi
-   if [ $2 =  ] || dpkg --compare-versions $2 lt 1:1.0.7-16; then
chown statd /var/lib/nfs/sm \
/var/lib/nfs/sm.bak \
/var/lib/nfs/rpc_pipefs \
@@ -31,7 +30,6 @@
 if [ -f /var/lib/nfs/state ]; then
chown statd /var/lib/nfs/state
 fi
-   fi

if [ $2 !=  ] || dpkg --compare-versions $2 lt 1:1.1.0-14; then
if dpkg-statoverride --list /sbin/mount.nfs /dev/null 21; then

it looks like the comparison with an empty $2 argument works for an initial
install, but not when a package was removed and later re-installed, as postinst
will be called with the previously installed version as an argument. other
parts of the postinst may be affected by similar uses of tests against $2.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100621020413.go9...@claws.fglan

Bug#509077: initramfs-tools: support alternate DHCP port and DHCP vendor-class-identifier

[Vagrant Cascadian]

On Sat, Jun 19, 2010 at 06:43:16PM +0200, maximilian attems wrote:
 On Tue, 08 Jun 2010, Michael Prokop wrote:
  * Vagrant Cascadian vagrant+debianb...@freegeek.org [Mit Dez 17, 2008 at 
  07:01:51 -0800]:
  
   please consider the attached patch, which adds boot prompt parameters for 
   two
   of ipconfig's commandline options in the configure_networking function:
  
 - using an alternate DHCP port (dhcpport=NNN, ipconfig -p NNN)
 - specifying the vendor-class-identifier (dhcpvci=XXX, ipconfig -i XXX)
  
   supporting these options gives additional flexibility in distinguishing 
   thin
   clients from conventional workstations in a mixed network.
  
  Sorry that no one took care of your bugreport for so long, Vagrant.
  Patch looks fine to me.
  
  maks, any objections against inclusion of the patches?
 
 I had been dragging that patch as I do not like to add to many
 bootparams that are outside of documented upstream linux-2.6 handling
 and thus initramfs-tools specific. waited to see if a second voice
 would raise..

understandable.

perhaps it would be better to rewrite in such a way that ipconfig could be
passed arbitrary arguments? of course, if it ever switched away from ipconfig
to some other dhcp client, it'd be nice to not have to change syntax or
configuration names...
 
 I do not see so much the point of setting ipconfig vendor class id.
 why would that be needed for booting?

in order to distinguish a machine that's doing network boot for
debian-installer, LTSP or PXE. for example, in each case, you might want to
hand it a different dhcp filename option. pxe should get pxelinux.0, d-i gets a
preseeding URL, and ltsp could get a configuration file...

 the alternate DHCP port looks indeed more interesting, but no idea
 how common that is in the wild? 

i don't know that it's incredibly common, but is not infrequently asked for in
LTSP environments, though there are other alternatives using dnsmasq's dhcp
proxy support now, so it's maybe less important.

 and if there is not a place in the ip= monster bootparam?

don't believe so.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100621021141.gp9...@claws.fglan

Bug#584583: initramfs-tools: configure_networking function: repeatedly makes DHCP requests

[Vagrant Cascadian]

On Tue, Jun 08, 2010 at 03:45:09PM +0200, Michael Prokop wrote:
 * Vagrant Cascadian vagrant+debianb...@freegeek.org [Fre Jun 04, 2010 at 
 01:18:04 -0700]:
 
  an alternate possibility is that i've been testing with qemu and qemu-kvm,
  which may have strange timing issues. i will attempt to test with real 
  hardware
  and follow up with that...
 
 Can you please provide output of:
 
   dpkg --list klibc-utils module-init-tools udev busybox\*

  ltsp-chroot dpkg --list klibc-utils module-init-tools udev 'busybox*'
  Desired=Unknown/Install/Remove/Purge/Hold
  | Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
  |/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
  ||/ Name   VersionDescription
  +++-==-==-
  ii  busybox1:1.14.2-2 Tiny utilities for small and embedded system
  un  busybox-initra none (no description available)
  un  busybox-static none (no description available)
  ii  klibc-utils1.5.18-1   small utilities built with klibc for early b
  ii  module-init-to 3.12~pre2-3tools for managing Linux kernel modules
  ii  udev   154-1  /dev/ and hotplug management daemon
   
 Which qemu/kvm cmdline are you using? Is the e1000 module involved?

  qemu -enable-kvm -name ltsp-client -net nic,vlan=1,model=pcnet -net 
tap,vlan=1,ifname=ltsp1,script=no -boot n -soundhw es1370 -usb

same behavior with model=virtio, model=ne2k_pci. e1000 doesn't even boot far 
enough to test...

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100608224012.gd4...@claws.fglan

Bug#584583: initramfs-tools: configure_networking function: repeatedly makes DHCP requests

[Vagrant Cascadian]

Package: initramfs-tools
Version: 0.95.1
Severity: normal

i've been testing the newest initramfs-tools network booting support, though it
seems to make 10 DHCP requests in rapid sucession. fortunately, the last DHCP
request does work. switching back to hard-coded DEVICE=eth0 in
/etc/initramfs-tools/initramfs.conf doesn't trigger this issue, oddly enough.

it think something in the following code is the issue:

for ROUNDTTT in 2 3 4 6 9 16 25 36 64 100; do

# The NIC is to be configured if this file does not exist.
# Ip-Config tries to create this file and when it succeds
# creating the file, ipconfig is not run again.
if [ -e /tmp/net-${DEVICE}.conf ]; then
break;
fi

case ${IP} in
none|off)
# Do nothing
;;
|on|any)
# Bring up device
ipconfig -t ${ROUNDTTT} ${DEVICE}
;;
dhcp|bootp|rarp|both)
ipconfig -t ${ROUNDTTT} -c ${IP} -d ${DEVICE}
;;
*)
ipconfig -t ${ROUNDTTT} -d $IP

it loops through all of the ROUNDTTT's in 1 second, which doesn't seem like the
intended behavior.

so either ipconfig -t is broken, or initramfs-tools is expecting the wrong
behavior from ipconfig.

an alternate possibility is that i've been testing with qemu and qemu-kvm,
which may have strange timing issues. i will attempt to test with real hardware
and follow up with that...

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100604201804.gn12...@claws.fglan

Bug#566295: initramfs-tools: Deviation from Documentation/filesystems/nfs/nfsroot.txt WRT multiple net devs

[Vagrant Cascadian]

On Sat, Apr 10, 2010 at 07:08:47AM +0200, maximilian attems wrote:
 On Fri, 09 Apr 2010, Vagrant Cascadian wrote:
 
  On Wed, Feb 24, 2010 at 06:22:47PM +0100, maximilian attems wrote:
   could you please review belows patch.
   if you see no objection, I'd just merge it.
  
  i finally got around to testing the proposed patch (after adjusting slightly
  for newer version). seems to work fine with 0.94.4.
  
  i'd probably leave out the support for DEVICE=all (and just rely on 
  DEVICE=
  working properly), but that's just personal preference.
  
  live well,
vagrant
 
 thanks for your review, as I currently lack the capability of testing
 would that look at this patch, please correct me if I missed something,
 or post me your variant, that worked, thanks:

looks good to me. :)

live well,
  vagrant

 From 4a0b7bc34031bd669d94ce2e8766e4fb5d85f2c2 Mon Sep 17 00:00:00 2001
 From: maximilian attems m...@debian.org
 Date: Sat, 10 Apr 2010 07:04:05 +0200
 Subject: [PATCH] configure_networking(): work with empty DEVICE string
 
 initramfs-tools currently requires a device to be hard-coded, but this
 is not much use if the network device is not known ahead of time.  If
 the device specified in either /etc/initramfs-tools/initramfs.conf or
 on the ip=xxx kernel command line.
 
 usefull for multiple net devices.
 
 Closes: #566295, #575766
 Signed-off-by: maximilian attems m...@debian.org
 ---
  conf/initramfs.conf |5 +++--
  scripts/functions   |7 ---
  2 files changed, 7 insertions(+), 5 deletions(-)
 
 diff --git a/conf/initramfs.conf b/conf/initramfs.conf
 index 2aa380f..0a108a8 100644
 --- a/conf/initramfs.conf
 +++ b/conf/initramfs.conf
 @@ -58,10 +58,11 @@ BOOT=local
  #
  # DEVICE: ...
  #
 -# Specify the network interface, like eth0
 +# Specify a specific network interface, like eth0
 +# Overriden by optional ip= bootarg
  #
  
 -DEVICE=eth0
 +DEVICE=
  
  #
  # NFSROOT: [ auto | HOST:MOUNT ]
 diff --git a/scripts/functions b/scripts/functions
 index 88f1bbd..1cc9be6 100644
 --- a/scripts/functions
 +++ b/scripts/functions
 @@ -363,10 +363,10 @@ configure_networking()
   ;;
   |on|any)
   # Bring up device
 - ipconfig -t ${ROUNDTTT} ${DEVICE}
 + ipconfig -t ${ROUNDTTT} ${DEVICE}
   ;;
   dhcp|bootp|rarp|both)
 - ipconfig -t ${ROUNDTTT} -c ${IP} -d ${DEVICE}
 + ipconfig -t ${ROUNDTTT} -c ${IP} -d ${DEVICE}
   ;;
   *)
   ipconfig -t ${ROUNDTTT} -d $IP
 @@ -391,7 +391,8 @@ configure_networking()
   # source specific bootdevice
   . /tmp/net-${DEVICE}.conf
   else
 - # source any interface as not exaclty specified
 + # source any interface...
 + # ipconfig should have quit after first response
   . /tmp/net-*.conf
   fi
  }
 -- 
 1.7.0.4
 



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100410071352.gm27...@claws.fglan

Bug#577149: configure_networking: pxelinux BOOTIF fixes

[Vagrant Cascadian]

Package: initramfs-tools
Version: 0.94.4
Severity: normal
Tags: patch

looks like i missed a few things on implementing BOOTIF support properly.

somehow i managed to get the order of the mac address backwards. i know i
tested it many times, but somehow the patch i submitted got it backwards:

diff --git a/scripts/functions b/scripts/functions
index 7b68255..f854495 100644
--- a/scripts/functions
+++ b/scripts/functions
@@ -323,7 +323,7 @@ configure_networking()
if [ -z $bootif_mac ]; then
bootif_mac=$x
else
-   bootif_mac=$x:$bootif_mac
+   bootif_mac=$bootif_mac:$x
fi
done
unset IFS


as mentioned by Christoph Bussenius in #535008, it does appear that setting
BOOTIF from the value in /proc/cmdline is needed. at the time i wrote the
patch, i'm not sure it was, but it definitely is now:

diff --git a/init b/init
index 7d0db07..142eb14 100755
--- a/init
+++ b/init
@@ -175,6 +175,9 @@ for x in $(cat /proc/cmdline); do
netconsole=*)
netconsole=${x#netconsole=}
;;
+   BOOTIF=*)
+   BOOTIF=${x#BOOTIF=}
+   ;;
esac
 done
 

with these patchese applied, it seems to work. at least for the moment.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100410033223.gj27...@claws.fglan

Bug#566295: initramfs-tools: Deviation from Documentation/filesystems/nfs/nfsroot.txt WRT multiple net devs

[Vagrant Cascadian]

On Wed, Feb 24, 2010 at 06:22:47PM +0100, maximilian attems wrote:
 could you please review belows patch.
 if you see no objection, I'd just merge it.

i finally got around to testing the proposed patch (after adjusting slightly
for newer version). seems to work fine with 0.94.4.

i'd probably leave out the support for DEVICE=all (and just rely on DEVICE=
working properly), but that's just personal preference.

live well,
  vagrant

  --- initramfs-tools-0.93.4/conf/initramfs.conf.old  2010-01-22 
  17:52:02.0 +
  +++ initramfs-tools-0.93.4/conf/initramfs.conf  2010-01-22 
  17:54:55.0 +
  @@ -52,10 +52,14 @@
   #
   # DEVICE: ...
   #
  -# Specify the network interface, like eth0
  +# Hard code a device for autoconfiguration.  If the empty
  +# string, or the word all then autoconfiguration will occur
  +# on all devices, and the first reply will be used.  May be
  +# overridden using the ip=... argument as per the Linux kernel
  +# Documentation/filesystems/nfs/nfsroot.txt
   #
   
  -DEVICE=eth0
  +DEVICE=all
   
   #
   # NFSROOT: [ auto | HOST:MOUNT ]
  --- lenny/initramfs-tools/scripts/functions.orig2009-01-07 
  14:14:37.0 +
  +++ lenny/initramfs-tools/scripts/functions 2010-01-22 16:37:26.0 
  +
  @@ -269,10 +269,10 @@
  ;;
  |on|any)
  # Bring up device
  -   ipconfig -t 180 ${DEVICE}
  +   ipconfig -t 180 ${DEVICE}
  ;;
  dhcp|bootp|rarp|both)
  -   ipconfig -t 180 -c ${IPOPTS} -d ${DEVICE}
  +   ipconfig -t 180 -c ${IPOPTS} -d ${DEVICE}
  ;;
  *)
  ipconfig -t 180 -d $IPOPTS
  @@ -292,11 +292,12 @@
  esac
   
  # source ipconfig output
  -   if [ -n ${DEVICE} ]; then
  +   if [ -n ${DEVICE} ]  [ ${DEVICE} != all ]; then
  # source specific bootdevice
  . /tmp/net-${DEVICE}.conf
  else
  -   # source any interface as not exaclty specified
  +   # source any interface...
  +   # ipconfig should have quit after first response
  . /tmp/net-*.conf
  fi
   }
  --- initramfs-tools-0.93.4/initramfs.conf.5.orig2010-01-22 
  17:45:51.0 +
  +++ initramfs-tools-0.93.4/initramfs.conf.5 2010-01-22 18:11:43.0 
  +
  @@ -57,7 +57,20 @@
   
   .TP
   \fB DEVICE
  -Specifies the network interface, like eth0.
  +Specifies the network interface which should be used during IP 
  configuration.
  +This can be overridden by arguments supplied on the kernel command line 
  using
  +the 'ip=...' argument as described in
  +\fIDocumentation/filesystems/nfs/nfsroot.txt\fP from the Linux kernel
  +documentation.
  +
  +If this is set to the empty string, or the value all i.e.
  +\fIDEVICE\fP=\fIall\fP then autoconfiguration will be carried out on all
  +network interfaces, and only the first reply will be used.
  +
  +Note that the network devices will have been renamed as specified in
  +\fI/etc/udev/rules.d/70-persistent-net.rules\fP prior to autoconfiguration,
  +and that a copy of this file (as it was at the time that
  +update-initramfs/mkinitramfs was executed) will be included in the 
  initramfs.
   
   .TP
   \fB ROOT



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100410043449.gl27...@claws.fglan

Bug#575766: initramfs-tools: configure_networking function fails if no DEVICE is specified

[Vagrant Cascadian]

On Sun, Apr 04, 2010 at 02:32:43AM +0200, maximilian attems wrote:
 On Sun, 28 Mar 2010, Vagrant Cascadian wrote:
 
  the configure_networking function relies on DEVICE being set, and
  conf/initramfs.conf hard-codes this to eth0. though some scripts may call 
  this
  from /scripts/local-top, and DEVICE may not be set. even when it is set, it
  hard-coding it to eth0 may be the wrong network interface.
  
  the following patch should partially address this by checking for DEVICE 
  when
  not defined, and using the first interface with an active link.
 
 this looks very similar to #566295
 
 did you have a look at the proposed patch of
 http://bugs.debian.org/566295

yes, that addresses the issue much more simply. i'd recommend going with that
instead of my patch.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100410043322.gk27...@claws.fglan

Bug#573912: linux-2.6: please enable ramzswap/compcache

[Vagrant Cascadian]

On Sun, Apr 04, 2010 at 08:06:23PM +0100, Ben Hutchings wrote:
 On Sun, 2010-04-04 at 11:21 -0700, Vagrant Cascadian wrote:
  On Sun, Apr 04, 2010 at 07:16:36AM +0100, Ben Hutchings wrote:
   On Thu, 2010-03-18 at 14:36 +1100, Trent W. Buck wrote:
Vagrant Cascadian wrote:
 Please consider setting CONFIG_RAMZSWAP (a.k.a. compcache), which
 enables a ram based compressed swap device. i think it can be
 enabled as a module, but i'm not sure.
...snip...
   Would you like me to add this to 2.6.32 for squeeze, then?
  
  if it's not too difficult, sure! i had thought it only made it upstream as
  recently as 2.6.33.
 
 Indeed, but it seems trivial to backport.  Please test it when 2.6.32-11
 is released.

thanks!

it works, although needs rzscontrol from compcache-tools (in the NEW queue
since december) in order to initialize the ramzswap device.

i had the impression from the documentation that the module could initialize
the device itself with the defaults, but this apparently isn't the case, at
least not with my initial testing, or the version that was backported.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100407041811.gd27...@claws.fglan

Bug#573912: linux-2.6: please enable ramzswap/compcache

[Vagrant Cascadian]

On Sun, Apr 04, 2010 at 07:16:36AM +0100, Ben Hutchings wrote:
 On Thu, 2010-03-18 at 14:36 +1100, Trent W. Buck wrote:
  Vagrant Cascadian wrote:
   Please consider setting CONFIG_RAMZSWAP (a.k.a. compcache), which
   enables a ram based compressed swap device. i think it can be
   enabled as a module, but i'm not sure.
  
   This feature is very useful to thin clients, embedded devices, or
   other environments with limited ram and no conventional swap device
   available.
  
  This is also useful for Debian Live images generated using the
  live-helper and live-initramfs packages.  Using Sid's live-helper to
  build an Ubuntu 10.04 live image already results in a 25% ramzswap, so
  I guess Ubuntu have patched compcache into their default .32 kernel.
 
 Would you like me to add this to 2.6.32 for squeeze, then?

if it's not too difficult, sure! i had thought it only made it upstream as
recently as 2.6.33.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100404182112.gv27...@claws.fglan

Bug#575766: initramfs-tools: configure_networking function fails if no DEVICE is specified

[Vagrant Cascadian]

Package: initramfs-tools
Severity: normal
Tags: patch

the configure_networking function relies on DEVICE being set, and
conf/initramfs.conf hard-codes this to eth0. though some scripts may call this
from /scripts/local-top, and DEVICE may not be set. even when it is set, it
hard-coding it to eth0 may be the wrong network interface.

the following patch should partially address this by checking for DEVICE when
not defined, and using the first interface with an active link.

diff --git a/conf/initramfs.conf b/conf/initramfs.conf
index 23dd249..471dfea 100644
--- a/conf/initramfs.conf
+++ b/conf/initramfs.conf
@@ -55,7 +55,7 @@ BOOT=local
 # Specify the network interface, like eth0
 #
 
-DEVICE=eth0
+#DEVICE=eth0
 
 #
 # NFSROOT: [ auto | HOST:MOUNT ]
diff --git a/scripts/functions b/scripts/functions
index 0ecb1d6..5444c11 100644
--- a/scripts/functions
+++ b/scripts/functions
@@ -294,6 +294,24 @@ configure_networking()
done
fi
 
+   if [ -z ${DEVICE} ]; then
+   # attempt to detect the first device with an established link.
+   for net_device_path in /sys/class/net/* ; do
+   net_device=${net_device_path##*/}
+   case $net_device in
+   lo) ;;
+   *)
+   if read carrier  
$net_device_path/carrier ; then
+   if [ $carrier = 1 ]; then
+   DEVICE=$net_device
+   break
+   fi
+   fi
+   ;;
+   esac
+   done
+   fi
+
# networking already configured thus bail out
[ -n ${DEVICE} ]  [ -e /tmp/net-${DEVICE}.conf ]  return 0
 

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100329042325.gf27...@claws.fglan

Bug#573912: linux-2.6: please enable ramzswap/compcache

[Vagrant Cascadian]

Package: linux-2.6
Version: consider enabling CONFIG_RAMZSWAP in 2.6.33+
Severity: wishlist

please consider setting CONFIG_RAMZSWAP (a.k.a. compcache), which enables a ram
based compressed swap device. i think it can be enabled as a module, but i'm
not sure.

this feature is very useful to thin clients, embedded devices, or other
environments with limited ram and no conventional swap device available.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100314222759.ga2...@claws.fglan

Bug#535008: configure_networking: support BOOTIF variable set by pxelinux

[Vagrant Cascadian]

On Wed, Feb 24, 2010 at 06:16:45PM +0100, maximilian attems wrote:
  it may need special handling to make sure BOOTIF is exported, such as if run
  from the scripts/*-top/*.
 
 hmmm ?? don't get that part, do you want an 
 export BOOTIF= in init?

i *think* that would be required, but i'm not sure. it would need the BOOTIF
variable available in some scripts such as the NBD support, which calls
configure_networking from scripts/local-top/nbd-client.

thanks!

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100226184239.gj8...@claws.fglan

Bug#535008: configure_networking: support BOOTIF variable set by pxelinux

[Vagrant Cascadian]

updated patch against current master, using only shell, and with a cleaner
method to convert BOOTF to a typical mac address.

diff --git a/scripts/functions b/scripts/functions
index 77de8f3..cb6cf08 100644
--- a/scripts/functions
+++ b/scripts/functions
@@ -258,6 +258,42 @@ parse_numeric() {
 
 configure_networking()
 {
+   if [ -n ${BOOTIF} ]; then
+   # pxelinux sets BOOTIF to a value based on the mac address of 
the
+   # network card used to PXE boot, so use this value for DEVICE 
rather
+   # than a hard-coded device name from initramfs.conf. this 
facilitates
+   # network booting when machines may have multiple network cards.
+   # pxelinux sets BOOTIF to 01-$mac_address
+
+   # strip off the leading 01-, which isn't part of the mac
+   # address
+   temp_mac=${BOOTIF#*-}
+
+   # convert to typical mac address format by replacing - with 
:
+   bootif_mac=
+   IFS='-'
+   for x in $temp_mac ; do
+   if [ -z $bootif_mac ]; then
+   bootif_mac=$x
+   else
+   bootif_mac=$x:$bootif_mac
+   fi
+   done
+   unset IFS
+
+   # look for devices with matching mac address, and set DEVICE to
+   # appropriate value if match is found.
+   for device in /sys/class/net/* ; do
+   if [ -f $device/address ]; then
+   current_mac=$(cat $device/address)
+   if [ $bootif_mac = $current_mac ]; then
+   DEVICE=${device##*/}
+   break
+   fi
+   fi
+   done
+   fi
+
# networking already configured thus bail out
[ -n ${DEVICE} ]  [ -e /tmp/net-${DEVICE}.conf ]  return 0
 

it may need special handling to make sure BOOTIF is exported, such as if run
from the scripts/*-top/*.

since it's a whole block of code, maybe it could go into it's own function.

please consider applying this patch for squeeze. :)

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#534880: linux-image-xen still hanging

[Vagrant Cascadian]

On Tue, Nov 17, 2009 at 01:49:56PM -0800, Vagrant Cascadian wrote:
 i'm happy to report that we've been running linux-image-2.6.26-2-xen-686
 2.6.26-20 from lenny-proposed-updates for a week on one of our dom0 and two of
 our domU servers, and haven't experienced the mysterious hanging problem
 anymore.

i spoke too soon.

that domU machine is currently unresponsive to ssh, ping, xm console, etc.
after 20 some days of uptime.

*sigh*

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#534880: linux-image-xen from stable-proposed-updates, 1 week without freeze

[Vagrant Cascadian]

On Thu, Nov 19, 2009 at 02:17:32PM +0100, Axel Beckert wrote:
  i'm happy to report that we've been running
  linux-image-2.6.26-2-xen-686 2.6.26-20 from lenny-proposed-updates for
  a week on one of our dom0 and two of our domU servers, and haven't
  experienced the mysterious hanging problem anymore.
 
 Have you rebooted only the DomU or also the Dom0 after installing
 2.6.26-20?

both dom0 and domU running 2.6.26-20.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#534880: linux-image-xen from stable-proposed-updates, 1 week without freeze

[Vagrant Cascadian]

i'm happy to report that we've been running linux-image-2.6.26-2-xen-686
2.6.26-20 from lenny-proposed-updates for a week on one of our dom0 and two of
our domU servers, and haven't experienced the mysterious hanging problem
anymore.

will start trying it out on more servers to see if we still experience the
problem...

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#534880: Debian Bug report logs - #534880

[Vagrant Cascadian]

On Tue, Jul 28, 2009 at 09:46:55AM +0200, Daniel Krause wrote:
 I'm experiencing the same problem. Did your solution of restriting the number 
 of CPUs solve the problem fully?

yes, haven't seen a similar crash since commenting out vcpus in all
/etc/xen/*.cfg.

it does take a performance hit, as now we can only use single-CPU VMs, but that
beats crashing seemingly randomly.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#534880: linux-image-2.6.26-2-xen-686: domU hang and are unresponsive

[Vagrant Cascadian]

 unfortunately, in the last couple weeks, i've been experiencing some issues
 with some domU machines crashing on lenny xen dom0.
 
 at some point, the domU just hangs, and xm console is unresponsive. still
 able to ping the domU server, but not log in via ssh, and existing ssh
 connections stop working. xm reboot and xm shutdown are useless.
 eventually, we xm destroy it, and it continues to work for another 1-3 days.

the machines that were failing all had vcpus = 2 in their respective
/etc/xen/*.cfg files. after commenting it out and restarting these domU, i've
seen uptimes of 7-12 days, with no hangs thus far, although they are now
effectively single-cpu systems, so are a bit slower...

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#535008: configure_networking: support BOOTIF variable set by pxelinux

[Vagrant Cascadian]

updated patch that only using only shell parameter expansion and cat:

diff --git a/scripts/functions b/scripts/functions
index 77de8f3..e9e94b3 100644
--- a/scripts/functions
+++ b/scripts/functions
@@ -258,6 +258,42 @@ parse_numeric() {
 
 configure_networking()
 {
+
+   # pxelinux sets BOOTIF to a value based on the mac address of the
+   # network card used to PXE boot, so use this value for DEVICE rather
+   # than a hard-coded device name from initramfs.conf. this facilitates
+   # network booting when machines may have multiple network cards.
+   # pxelinux sets BOOTIF to 01-$mac_address
+   if [ -n ${BOOTIF} ]; then
+   # strip off the leading 01-, which isn't part of the mac
+   # address
+   temp_mac=${BOOTIF#*-}
+   bootif_mac=
+   # given value is the right number of characters, so
+   # loop through replacing - with : 6 times...
+   for x in 1 2 3 4 5 6 ; do
+   # strip off everything before and including the last -
+   x=${temp_mac##*-}
+   # strip off everything before and including the
+   # first occurrance of $x
+   temp_mac=${temp_mac%%-$x}
+   if [ -z $bootif_mac ]; then
+   bootif_mac=$x
+   else
+   bootif_mac=$x:$bootif_mac
+   fi
+   done
+   for device in /sys/class/net/* ; do
+   if [ -f $device/address ]; then
+   current_mac=$(cat $device/address)
+   if [ $bootif_mac = $current_mac ]; then
+   DEVICE=${device##*/}
+   break
+   fi
+   fi
+   done
+   fi
+
# networking already configured thus bail out
[ -n ${DEVICE} ]  [ -e /tmp/net-${DEVICE}.conf ]  return 0
 

sure is starting to look ugly.

tested that it basically works, though it may have lingering bugs if unexpected
values end up in $BOOTIF.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#535008: configure_networking: support BOOTIF variable set by pxelinux

[Vagrant Cascadian]

Package: initramfs-tools
Version: 0.93.3
Severity: wishlist
Tags: patch

when booting thin clients with multiple network cards, initramfs-tools/ipconfig
have a hard time figuring out which interface to use. attached is a patch which
sets DEVICE to the interface matching the mac address that pxelinux passes when
using the IPAPPEND 2 or 3 options.

diff --git a/scripts/functions b/scripts/functions
index 77de8f3..ec0ffd2 100644
--- a/scripts/functions
+++ b/scripts/functions
@@ -258,6 +258,19 @@ parse_numeric() {
 
 configure_networking()
 {
+
+   # pxelinux sets BOOTIF to a value based on the mac address of the 
network
+   # card used to PXE boot, so use this value for DEVICE rather than a
+   # hard-coded device name from initramfs.conf. this facilitates network
+   # booting when machines may have multiple network cards.
+   if [ -n ${BOOTIF} ]; then
+   normal_mac=$(echo $BOOTIF | sed -e 's,^01-,,g' -e 's,-,:,g' | 
tr [a-z] [A-Z])
+   new_device=$(ifconfig -a | awk /$normal_mac/'{print $1}')
+   if [ -n $new_device ]; then
+   DEVICE=$new_device
+   fi
+   fi
+
# networking already configured thus bail out
[ -n ${DEVICE} ]  [ -e /tmp/net-${DEVICE}.conf ]  return 0

 
live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#535008: configure_networking: support BOOTIF variable set by pxelinux

[Vagrant Cascadian]

On Sun, Jun 28, 2009 at 11:51:32PM +0100, Ben Hutchings wrote:
 On Sun, 2009-06-28 at 14:50 -0700, Vagrant Cascadian wrote:
  Package: initramfs-tools
  Version: 0.93.3
  Severity: wishlist
  Tags: patch
  
  when booting thin clients with multiple network cards, 
  initramfs-tools/ipconfig
  have a hard time figuring out which interface to use. attached is a patch 
  which
  sets DEVICE to the interface matching the mac address that pxelinux passes 
  when
  using the IPAPPEND 2 or 3 options.
 
 This sounds reasonable, but your patch makes unnecessary use of several
 commands.
 
  diff --git a/scripts/functions b/scripts/functions
  index 77de8f3..ec0ffd2 100644
  --- a/scripts/functions
  +++ b/scripts/functions
  @@ -258,6 +258,19 @@ parse_numeric() {
   
   configure_networking()
   {
  +
  +   # pxelinux sets BOOTIF to a value based on the mac address of the 
  network
  +   # card used to PXE boot, so use this value for DEVICE rather than a
  +   # hard-coded device name from initramfs.conf. this facilitates network
  +   # booting when machines may have multiple network cards.
  +   if [ -n ${BOOTIF} ]; then
  +   normal_mac=$(echo $BOOTIF | sed -e 's,^01-,,g' -e 's,-,:,g' | 
  tr [a-z] [A-Z])
 
 You can use sed's y/// command instead of piping to tr.
 
  +   new_device=$(ifconfig -a | awk /$normal_mac/'{print $1}')
 [...]
 
 You can use the files /sys/class/net/*/address instead of ifconfig.

thanks for the suggestions, updated patch:

diff --git a/scripts/functions b/scripts/functions
index 77de8f3..f99e779 100644
--- a/scripts/functions
+++ b/scripts/functions
@@ -258,6 +258,25 @@ parse_numeric() {
 
 configure_networking()
 {
+
+   # pxelinux sets BOOTIF to a value based on the mac address of the
+   # network card used to PXE boot, so use this value for DEVICE rather
+   # than a hard-coded device name from initramfs.conf. this facilitates
+   # network booting when machines may have multiple network cards.
+   # pxelinux sets BOOTIF to 01-$mac_address
+   if [ -n ${BOOTIF} ]; then
+   bootif_mac=$(echo ${BOOTIF#*-} | sed -e 'y,-,:,')
+   for device in /sys/class/net/* ; do
+   if [ -f $device/address ]; then
+   current_mac=$(cat $device/address)
+   if [ $bootif_mac = $current_mac ]; then
+   DEVICE=${device##*/}
+   break
+   fi
+   fi
+   done
+   fi
+
# networking already configured thus bail out
[ -n ${DEVICE} ]  [ -e /tmp/net-${DEVICE}.conf ]  return 0
 

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#534880: linux-image-2.6.26-2-xen-686: domU hang and are unresponsive

[Vagrant Cascadian]

Package: linux-image-2.6.26-2-xen-686
Version: 2.6.26-15lenny3
Severity: important

we've been using xen on debian extensively for the last couple years, and it's
overall been great for freegeek!

unfortunately, in the last couple weeks, i've been experiencing some issues
with some domU machines crashing on lenny xen dom0.

at some point, the domU just hangs, and xm console is unresponsive. still
able to ping the domU server, but not log in via ssh, and existing ssh
connections stop working. xm reboot and xm shutdown are useless.
eventually, we xm destroy it, and it continues to work for another 1-3 days.

load on the dom0 doesn't seem high in any way. load levels on the domU seem
reasonable (usually less than 1-2 with a 2 processor system).

nothing obvious in the dom1 or domU syslog entries, in /var/log/xend/*.log. 

we've seen this happen on 3 separate dom0 servers, with 3 separate domU xen
servers. other domU servers running on the same dom0 continue to work fine.

it's happening on two lenny xen servers, one of which was cleanly installed as
a lenny server post-release, one of which was upgraded from etch, and one of
which is just etch running the lenny xen kernel.

i noticed that we upgraded the kernel to the current version around june 2nd
(though the problems didn't really start till mid-june), which seems to have a
xen-specific changelog entry:

  [ Ian Campbell ]
  * xen: Fix missing check of interrupted code's code selector
(CVE-2009-1758)

no idea if that could possibly be relevent. i might try running an older kernel
and see if that stabalizes it at all.

i know this isn't much to go on for a bug report, but it's been kind of
crippling and i don't really know what to do. any suggestions on how to get
more useful troubleshooting information would be greatly appreciated.

live well,
  vagrant

-- System Information:
Debian Release: 5.0.1
  APT prefers stable
  APT policy: (500, 'stable')
Architecture: i386 (i686)

Kernel: Linux 2.6.26-2-xen-686 (SMP w/2 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) (ignored: LC_ALL 
set to en_US.utf8)
Shell: /bin/sh linked to /bin/bash

Versions of packages linux-image-2.6.26-2-xen-686 depends on:
ii  initramfs-tools  0.92o   tools for generating an initramfs
ii  linux-modules-2.6.26-2-x 2.6.26-15lenny3 Linux 2.6.26 modules on i686

Versions of packages linux-image-2.6.26-2-xen-686 recommends:
ii  libc6-xen 2.7-18 GNU C Library: Shared libraries [X

Versions of packages linux-image-2.6.26-2-xen-686 suggests:
ii  grub   0.97-47lenny2 GRand Unified Bootloader (Legacy v
pn  linux-doc-2.6.26   none(no description available)

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#405135: no longer have hardware to test: 409822, 405135

[Vagrant Cascadian]

i no longer have the hardware necessary to test the following bugs:

405135: linux-image-2.6.18-3-xen-vserver-686: kernel freezes on pcmcia card 
insertion

409822: linux-image-2.6.18-4-vserver-686: pcmcia wireless card doesn't work 
until inserting usb ethernet

please feel free to close them or mark as unreproducible.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#352891: linux-image-2.6.18-2-vserver-686: does not power off on shutdown

[Vagrant Cascadian]

On Fri, Dec 26, 2008 at 11:30:51AM +0100, Moritz Muehlenhoff wrote:
 On Wed, Nov 22, 2006 at 12:42:26PM -0800, Vagrant Cascadian wrote:
  Package: linux-image-2.6.18-2-vserver-686
  Version: 2.6.18-5
  Followup-For: Bug #352891
  
  this kernel also seems to fail to power off.
  
  linux-image-2.6.18-2-486 (version 2.6.18-5) works fine, though.
  
  perhaps this bug is related to: 
  
  #378323: linux-image-2.6.17-1-686: CONFIG_SMP=y keeps APM poweroff from 
  working
  
  as the -486 kernel does not have CONFIG_SMP set.
 
 Does this error still occur with more recent kernel versions?

i might be able to find the hardware that experienced this problem, but i
haven't been using it recently.  haven't experienced it with other hardware
with more recent kernels. as far as i'm concerned, feel free to close the bug.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#426843: hald: general protection fault

[Vagrant Cascadian]

On Fri, Dec 26, 2008 at 12:05:56AM +0100, Moritz Muehlenhoff wrote:
 On Thu, May 31, 2007 at 02:06:01AM -0700, Vagrant Cascadian wrote:
  so, ever since installing hal, periodically i get these messages
  displayed on every terminal(every xterm, every screen session, every
  console session, etc):
  
  May 31 10:22:06 tyst kernel: general protection fault: ea30 [#1]
  May 31 10:22:06 tyst kernel: Modules linked in: ipt_LOG iptable_mangle
  iptable_filter pegasus mii orinoco_cs orinoco hermes rfcomm
...snip...
  May 31 10:22:06 tyst kernel:0216530a 5f57 00b8 530a 
  0067 0060cc9b 530a 0033
  May 31 10:22:06 tyst kernel:007b 007bc015 5a40 c02c 
  2040 0282c118  
  May 31 10:22:06 tyst kernel: Call Trace:
  May 31 10:22:06 tyst kernel: Code:  Bad EIP value.
  May 31 10:22:06 tyst kernel: EIP: [86f1] 0x86f1 SS:ESP 0068:c4d75e8a
 
 Does this error still occur with more recent kernel versions?

i might be able to find the hardware that experienced this problem, but i
haven't been using it recently.  haven't experienced it with other hardware
with more recent kernels. as far as i'm concerned, feel free to close the bug.

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#509077: initramfs-tools: support alternate DHCP port and DHCP vendor-class-identifier

[Vagrant Cascadian]

patches also available in a git branch:

  http://dev.freegeek.org/~vagrant/git/initramfs-tools/

live well,
  vagrant



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org