Bug#1011542: installation-reports: Armhf Installer Missing phy-gmii-sel.ko Module Required on BeagleBone Black

2022-05-24 Thread David Kaiser
Package: installation-reports
Severity: important
Tags: d-i
X-Debbugs-Cc: dwkai...@hotmail.com, vagr...@debian.org

(Please provide enough information to help the Debian
maintainers evaluate the report efficiently - e.g., by filling
in the sections below.)

Boot method: SD-Card Image
Image version: 
http://ftp.debian.org/debian/dists/bullseye/main/installer-armhf/20210731+deb11u3/images/netboot/SD-Card-images/
2022-03-22  [firmware.BeagleBoneBlack.img + partition.img]
Date: May 22, 2022

Machine: Circuitco BeagleBone Black, Rev C
Partitions: 


Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it

Initial boot:   [O]
Detect network card:[E]
Configure network:  [ ]
Detect media:   [ ]
Load installer modules: [ ]
Clock/timezone setup:   [ ]
User/password setup:[ ]
Detect hard drives: [ ]
Partition hard drives:  [ ]
Install base system:[ ]
Install tasks:  [ ]
Install boot loader:[ ]
Overall install:[ ]

Comments/Problems:

The d-i program fails to identify the network card and, therefore, cannot 
complete.  I can resolve this issue by manually editting the initrd.gz image, 
provided by the default installer, and appending the 
drivers/phy/ti/phy-gmii-sel.ko module from linux-image-5.10.0-13-armmp.  This 
same issue has been validated in each Bullseye release prior to Update 3, and 
in the previous major release.

Please make sure that any installation logs that you think would
be useful are attached to this report. Please compress large
files using gzip.


-- Package-specific info:

==
Installer lsb-release:
==
DISTRIB_ID=Debian
DISTRIB_DESCRIPTION="Debian GNU/Linux installer"
DISTRIB_RELEASE="11 (bullseye) - installer build 20210731+deb11u3"
X_INSTALLATION_MEDIUM=netboot

==
Installer hardware-summary:
==
uname -a: Linux BBBlack 5.10.0-13-armmp #1 SMP Debian 5.10.106-1 (2022-03-17) 
armv7l GNU/Linux
lsmod: Module  Size  Used by
lsmod: fuse  122880  0
lsmod: dm_mod114688  0
lsmod: md_mod147456  0
lsmod: jfs   180224  0
lsmod: btrfs1339392  0
lsmod: libcrc32c  16384  1 btrfs
lsmod: xor16384  1 btrfs
lsmod: xor_neon   16384  1 xor
lsmod: raid6_pq  106496  1 btrfs
lsmod: vfat   24576  0
lsmod: fat69632  1 vfat
lsmod: ext4  729088  2
lsmod: crc16  16384  1 ext4
lsmod: mbcache16384  1 ext4
lsmod: jbd2  114688  1 ext4
lsmod: crc32c_generic 16384  4
lsmod: usb_storage57344  0
lsmod: scsi_mod  196608  1 usb_storage
lsmod: sdhci_omap 24576  0
lsmod: sdhci_pltfm16384  1 sdhci_omap
lsmod: sdhci  61440  2 sdhci_pltfm,sdhci_omap
lsmod: smsc   16384  1
lsmod: phy_gmii_sel   20480  1
lsmod: davinci_mdio   16384  0
lsmod: ti_cpsw73728  0
lsmod: cpts   24576  1 ti_cpsw
lsmod: cpsw_common16384  1 ti_cpsw
lsmod: ptp24576  1 cpts
lsmod: pps_core   24576  1 ptp
lsmod: omap_rng   16384  0
lsmod: rng_core   20480  1 omap_rng
lsmod: musb_dsps  24576  0
lsmod: musb_hdrc 131072  1 musb_dsps
lsmod: udc_core   45056  1 musb_hdrc
lsmod: usbcore   225280  2 usb_storage,musb_hdrc
lsmod: phy_am335x 16384  9
lsmod: phy_generic20480  1 phy_am335x
lsmod: phy_am335x_control 16384  1 phy_am335x
lsmod: leds_gpio  16384  0
df: Filesystem   1K-blocks  Used Available Use% Mounted on
df: tmpfs4995260 49892   0% /run
df: devtmpfs229472 0229472   0% /dev
df: /dev/mmcblk0p229527944   2900368  25102280  10% /target
df: /dev/mmcblk0p1  480618 61220394464  13% /target/boot
df: /dev/mmcblk0p229527944   2900368  25102280  10% /dev/.static/dev
df: devtmpfs229472 0229472   0% /target/dev
free:   totalusedfree  shared  buff/cache   
available
free: Mem: 499500   68996  194436   65580  236068  
374060
free: Swap:487420   62208  425212
/proc/cmdline:  console=ttyO0,115200n8
/proc/cpuinfo: processor: 0
/proc/cpuinfo: model name   : ARMv7 Processor rev 2 (v7l)
/proc/cpuinfo: BogoMIPS : 995.32
/proc/cpuinfo: Features : half thumb fastmult vfp edsp thumbee neon vfpv3 tls 
vfpd32 
/proc/cpuinfo: CPU implementer  : 0x41
/proc/cpuinfo: CPU architecture: 7
/proc/cpuinfo: CPU variant  : 0x3
/proc/cpuinfo: CPU part : 0xc08
/proc/cpuinfo: 

Bug#923021: marked as done (Kernel oops using Buster Alpha 5 armhf installer onto ClearFog Base)

2020-02-04 Thread Debian Bug Tracking System
Your message dated Tue, 04 Feb 2020 22:54:56 -0700
with message-id 
and subject line Resolved in buster release images
has caused the Debian Bug report #923021,
regarding Kernel oops using Buster Alpha 5 armhf installer onto ClearFog Base
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact ow...@bugs.debian.org
immediately.)


-- 
923021: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=923021
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---

Package: installation-reports

Boot method: microSD card
Image version: 
http://ftp.debian.org/debian/dists/testing/main/installer-armhf/current/images/hd-media/hd-media.tar.gz

Date: 2019-02-22T20:57:53-07:00

Machine: ClearFog Base
Processor: Marvell ARMADA A388 ARM A9
Memory: 1GB
Partitions: single ext4 partition, primary part 1

Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it

Initial boot:   [O]
Detect network card:[E]
Configure network:  [ ]
Detect CD:  [O]
Load installer modules: [O]
Detect hard drives: [O]
Partition hard drives:  [ ]
Install base system:[ ]
Clock/timezone setup:   [ ]
User/password setup:[ ]
Install tasks:  [ ]
Install boot loader:[ ]
Overall install:[ ]

Comments/Problems:

I encounter a consistent kernel oops when trying to install Debian onto 
a SolidRun ClearFog Base device. I previously attempted installation 
using the Buster Alpha 4 version of the installer, and have just 
retested using Alpha 5 version and get the same result. This particular 
device is using a self-compiled U-Boot version based on upstream sources 
at v2018.11.


To prepare the installation media, I downloaded the armhf hd-media 
archive [1], and then unpacked the contents into a single ext4 partition 
on a micro SD card. I then copied the armhf netinst ISO image to the 
root of the filesystem.


The installer boots successfully, goes through the language selection 
and location prompts, then proceeds to mount the mmcblk0 device, find 
the ISO image, and cycles through "Loading additional components". When 
it completes component loading, it proceed to detecting network devices, 
at which point I get a kernel oops dump like the one below. The 
installation happens via serial (USB FTDI) connection, so I apologize 
for the corruption in the capture of the message.


[   68.717677] Unable to handle kernel paging request at virtual address 
2680

[   68.725981] pgd = (ptrval)
[   68.730086] [2680] *pgd=
[   68.735082] Internal error: Oops: 5 [#1] SMP ARM
[   68.741086] Modules linked in: marvell mvmdio(+) mvneta(+) mvneta_bm 
phylink sfp mdio_i2c nls_utf8 dm_mod loop isofs vfat fat ext4 crc16 
mbcache jbd2 crc32c_generic fscrypto usb_storage sd_mod gpio_pca953x 
ehci_orion ahci_mve bu ehci_hcd libahci_platform libahci sdhci_pxav3 
xhci_plat_hcd xhci_hcd sdhci_pltfm libata sdhci scsi_mod usbcore 
i2c_mv64xxx
[   68.772053] CPU: 0 PID: 13 Comm: kworker/0:1 Not tainted 
4.19.0-1-armmp #1 Debian 4.19.12-1

[   68.780422] Hardware name: Marvell Armada 380/385 (Device Tree)
[   68.786370] Workqueue: events_power_efficient phylink_resolve 
[phylink]

[   68.793033] PC is at mvneta_port_down+0x20/0x1a0 [mvneta]
[   68.798454] LR is at mvneta_mac_link_down+0x24/0x90 [mvneta]
[   68.804125] pc : []lr : []psr: 600b0013
[   68.810406] sp : ee96be60  ip : ee96be88  fp : ee96be84
[   68.815641] r10:   r9 : eef1c99c  r8 : eef1c900
[   68.820876] r7 : eef1c960  r6 : ede82000  r5 : 0002  r4 : 
ede82580
[   68.827417] r3 : 2680  r2 : 0004  r1 : 0002  r0 : 
ede82580
[   68.833959] Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  
Segment none

[   68.841109] Control: 10c5387d  Table: 0fe9804a  DAC: 0051
[   68.846867] Process kworker/0:1 (pid: 13, stack limit = 0x(ptrval))
[   68.853147] Stack: (0xee96be60 to 0xee96c000)
[   68.857514] be60: ede82000 0002 ede82000 eef1c960 eef1c900 
eef1c99c ee96bea4 ee96be88
[   68.865711] be80: bf51b490 bf51b2d8 eef1c998 c1105dcc ede82000 
eef1c960 ee96befc ee96bea8
[   68.873908] bea0: bf507000 bf51b478 c0aeb1e8 c037bc10 0001 
 ee96bee4 ee96bec8
[   68.882104] bec0: c04f637c c050ebdc  2e6b8000 c1018c40 
4bd6f77b ee8d6180 eef1c998
[   68.890301] bee0: ee8d6180 ef6d0840 ef6d3f00  ee96bf34 
ee96bf00 c036c090 bf506eac
[   68.898497] bf00: 0008 ef6d0858 c1104d00 ee8d6180 ee8d6194 
ef6d0840 0008 ef6d0858
[   68.906695] bf20: c1104d00 ef6d0840 ee96bf74 ee96bf38 c036d0a4 
c036bed4 ee8f6e40 c0c7436c
[   68.914891] bf40: c11f1428 e000 c0372638 ee8f6600 ee8f6e40

Bug#923021: Kernel oops using Buster Alpha 5 armhf installer onto ClearFog Base

2019-02-22 Thread Joel Johnson

Package: installation-reports

Boot method: microSD card
Image version: 
http://ftp.debian.org/debian/dists/testing/main/installer-armhf/current/images/hd-media/hd-media.tar.gz

Date: 2019-02-22T20:57:53-07:00

Machine: ClearFog Base
Processor: Marvell ARMADA A388 ARM A9
Memory: 1GB
Partitions: single ext4 partition, primary part 1

Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it

Initial boot:   [O]
Detect network card:[E]
Configure network:  [ ]
Detect CD:  [O]
Load installer modules: [O]
Detect hard drives: [O]
Partition hard drives:  [ ]
Install base system:[ ]
Clock/timezone setup:   [ ]
User/password setup:[ ]
Install tasks:  [ ]
Install boot loader:[ ]
Overall install:[ ]

Comments/Problems:

I encounter a consistent kernel oops when trying to install Debian onto 
a SolidRun ClearFog Base device. I previously attempted installation 
using the Buster Alpha 4 version of the installer, and have just 
retested using Alpha 5 version and get the same result. This particular 
device is using a self-compiled U-Boot version based on upstream sources 
at v2018.11.


To prepare the installation media, I downloaded the armhf hd-media 
archive [1], and then unpacked the contents into a single ext4 partition 
on a micro SD card. I then copied the armhf netinst ISO image to the 
root of the filesystem.


The installer boots successfully, goes through the language selection 
and location prompts, then proceeds to mount the mmcblk0 device, find 
the ISO image, and cycles through "Loading additional components". When 
it completes component loading, it proceed to detecting network devices, 
at which point I get a kernel oops dump like the one below. The 
installation happens via serial (USB FTDI) connection, so I apologize 
for the corruption in the capture of the message.


[   68.717677] Unable to handle kernel paging request at virtual address 
2680

[   68.725981] pgd = (ptrval)
[   68.730086] [2680] *pgd=
[   68.735082] Internal error: Oops: 5 [#1] SMP ARM
[   68.741086] Modules linked in: marvell mvmdio(+) mvneta(+) mvneta_bm 
phylink sfp mdio_i2c nls_utf8 dm_mod loop isofs vfat fat ext4 crc16 
mbcache jbd2 crc32c_generic fscrypto usb_storage sd_mod gpio_pca953x 
ehci_orion ahci_mve bu ehci_hcd libahci_platform libahci sdhci_pxav3 
xhci_plat_hcd xhci_hcd sdhci_pltfm libata sdhci scsi_mod usbcore 
i2c_mv64xxx
[   68.772053] CPU: 0 PID: 13 Comm: kworker/0:1 Not tainted 
4.19.0-1-armmp #1 Debian 4.19.12-1

[   68.780422] Hardware name: Marvell Armada 380/385 (Device Tree)
[   68.786370] Workqueue: events_power_efficient phylink_resolve 
[phylink]

[   68.793033] PC is at mvneta_port_down+0x20/0x1a0 [mvneta]
[   68.798454] LR is at mvneta_mac_link_down+0x24/0x90 [mvneta]
[   68.804125] pc : []lr : []psr: 600b0013
[   68.810406] sp : ee96be60  ip : ee96be88  fp : ee96be84
[   68.815641] r10:   r9 : eef1c99c  r8 : eef1c900
[   68.820876] r7 : eef1c960  r6 : ede82000  r5 : 0002  r4 : 
ede82580
[   68.827417] r3 : 2680  r2 : 0004  r1 : 0002  r0 : 
ede82580
[   68.833959] Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  
Segment none

[   68.841109] Control: 10c5387d  Table: 0fe9804a  DAC: 0051
[   68.846867] Process kworker/0:1 (pid: 13, stack limit = 0x(ptrval))
[   68.853147] Stack: (0xee96be60 to 0xee96c000)
[   68.857514] be60: ede82000 0002 ede82000 eef1c960 eef1c900 
eef1c99c ee96bea4 ee96be88
[   68.865711] be80: bf51b490 bf51b2d8 eef1c998 c1105dcc ede82000 
eef1c960 ee96befc ee96bea8
[   68.873908] bea0: bf507000 bf51b478 c0aeb1e8 c037bc10 0001 
 ee96bee4 ee96bec8
[   68.882104] bec0: c04f637c c050ebdc  2e6b8000 c1018c40 
4bd6f77b ee8d6180 eef1c998
[   68.890301] bee0: ee8d6180 ef6d0840 ef6d3f00  ee96bf34 
ee96bf00 c036c090 bf506eac
[   68.898497] bf00: 0008 ef6d0858 c1104d00 ee8d6180 ee8d6194 
ef6d0840 0008 ef6d0858
[   68.906695] bf20: c1104d00 ef6d0840 ee96bf74 ee96bf38 c036d0a4 
c036bed4 ee8f6e40 c0c7436c
[   68.914891] bf40: c11f1428 e000 c0372638 ee8f6600 ee8f6e40 
 ee96a000 ee8d6180
[   68.923088] bf60: c036d044 ee957e74 ee96bfac ee96bf78 c0372b58 
c036d050 ee8f661c ee8f661c
[   68.931285] bf80: ee96bfac ee8f6e40 c03729ec   
  
[   68.939482] bfa0:  ee96bfb0 c03010e8 c03729f8  
  
[   68.947678] bfc0:      
  
[   68.955875] bfe0:     0013 
  
[   68.964102] [] (mvneta_port_down [mvneta]) from 
[] (mvneta_mac_link_down+0x24/0x90 [mvneta]
) ┌─┤ Detecting network hardware 
├──┐
[ │  
   │ (phylink_resolve+0x160/0x2dc [phylin
k]│0%  

Re: Consolidating armhf installer flavours

2013-09-24 Thread Ian Campbell
On Sat, 2013-09-21 at 17:08 +0100, Ian Campbell wrote:
 With the current trunk/experimental 3.11 kernel the separate omap5, mx5
 and vexpress flavours have been removed in favour of the armmp flavour
 which supports all of those.
 
 So I think we need to switch the installer over too. Since the 3.10
 kernel already has the armmp flavour I don't think we even need wait for
 3.11 to be uploaded.

Perhaps it would be sensible to split this patch up and add the new
flavour now, so folks can test it, and remove the other flavours a bit
later on.

My only concern would be the short term impact on the build time.
Looking at [1] it seems that a successful build takes about 30 mins. In
the short term we would be adding 3 variants
(armmp-{netboot,network-console,netboot-gtk}) to the existing 4
(mx5-{netboot,network-console,netboot-gtk} + vexpress-netboot), which I
estimate would push the build time up to somewhere under an hour.
Eventually we would drop back down to just the three armmp variants.

Unless there are any objections I'll look at doing this shortly.

Ian.

[1]
https://buildd.debian.org/status/logs.php?pkg=debian-installerarch=armhf
 
 I merged the existing mx5 and vexpress mediums so armmp creates all of
 netboot network-console netboot-gtk. I also retained the efikamx uImage
 stuff (but using the new kernel) since I guess it is useful? 
 
 Should we have more of those uImage things for other armmp supported
 platforms?
 
 Ian.
 
 commit 7de47649d4f47225f740d262de8d51c83be73688
 Author: Ian Campbell i...@hellion.org.uk
 Date:   Mon Aug 26 11:25:56 2013 +0100
 
 Switch armhf to armmp kernel.
 
 The other flavours have been removed in the 3.11 kernel, even in 3.10 the 
 armmp
 flavour supports mx5 and vexpress.
 
 Merge all of the subtypes over.
 
 diff --git a/build/config/armhf.cfg b/build/config/armhf.cfg
 index 73b8a80..0287ef2 100644
 --- a/build/config/armhf.cfg
 +++ b/build/config/armhf.cfg
 @@ -1,4 +1,4 @@
 -SUBARCH_SUPPORTED = mx5 vexpress
 +SUBARCH_SUPPORTED = armmp
  
  MKLIBS = mklibs --ldlib=/lib/ld-linux-armhf.so.3
  
 diff --git a/build/config/armhf/mx5.cfg b/build/config/armhf/armmp.cfg
 similarity index 69%
 rename from build/config/armhf/mx5.cfg
 rename to build/config/armhf/armmp.cfg
 index a30c0a2..dd24206 100644
 --- a/build/config/armhf/mx5.cfg
 +++ b/build/config/armhf/armmp.cfg
 @@ -1,4 +1,4 @@
  MEDIUM_SUPPORTED = netboot network-console netboot-gtk
  
  # The version of the kernel to use.
 -KERNELVERSION := $(KERNELVERSION)-mx5
 +KERNELVERSION := $(KERNELVERSION)-armmp
 diff --git a/build/config/armhf/mx5/netboot-gtk.cfg 
 b/build/config/armhf/armmp/netboot-gtk.cfg
 similarity index 81%
 rename from build/config/armhf/mx5/netboot-gtk.cfg
 rename to build/config/armhf/armmp/netboot-gtk.cfg
 index cb0b074..3817048 100644
 --- a/build/config/armhf/mx5/netboot-gtk.cfg
 +++ b/build/config/armhf/armmp/netboot-gtk.cfg
 @@ -1,6 +1,6 @@
  MEDIA_TYPE = netboot image
  
 -TARGET = $(TEMP_INITRD) $(TEMP_KERNEL) efikamx
 +TARGET = $(INITRD) $(KERNEL) efikamx
  
  TYPE = netboot/gtk
  EXTRANAME = netboot/gtk/
 @@ -16,8 +16,8 @@ EXTRAUDEBS += cdebconf-gtk-entropy
  efikamx:
   mkdir -p $(SOME_DEST)/$(EXTRANAME)/efikamx
   install -m 744 boot/arm/bootscr.mx5 
 $(SOME_DEST)/$(EXTRANAME)/efikamx/bootscript
 - mkimage -A arm -O linux -T kernel -C none -a 0x90008000 -e 0x90008000 
 -n Debian kernel -d $(TEMP_KERNEL) $(SOME_DEST)/$(EXTRANAME)/efikamx/uImage
 - mkimage -A arm -O linux -T ramdisk -C gzip -a 0x0 -e 0x0 -n 
 debian-installer ramdisk -d $(TEMP_INITRD) 
 $(SOME_DEST)/$(EXTRANAME)/efikamx/uInitrd
 + mkimage -A arm -O linux -T kernel -C none -a 0x90008000 -e 0x90008000 
 -n Debian kernel -d $(KERNEL) $(SOME_DEST)/$(EXTRANAME)/efikamx/uImage
 + mkimage -A arm -O linux -T ramdisk -C gzip -a 0x0 -e 0x0 -n 
 debian-installer ramdisk -d $(INITRD) 
 $(SOME_DEST)/$(EXTRANAME)/efikamx/uInitrd
   mkimage -A arm -O linux -a 0 -e 0 -T script -C none -n Efika MX Boot 
 Script -d $(SOME_DEST)/$(EXTRANAME)/efikamx/bootscript 
 $(SOME_DEST)/$(EXTRANAME)/efikamx/boot.scr
   update-manifest $(SOME_DEST)/$(EXTRANAME)/efikamx/uImage Linux kernel 
 for Efika MX
   update-manifest $(SOME_DEST)/$(EXTRANAME)/efikamx/uInitrd initrd for 
 Efika MX
 diff --git a/build/config/armhf/mx5/netboot.cfg 
 b/build/config/armhf/armmp/netboot.cfg
 similarity index 77%
 rename from build/config/armhf/mx5/netboot.cfg
 rename to build/config/armhf/armmp/netboot.cfg
 index 2456c93..e6886ba 100644
 --- a/build/config/armhf/mx5/netboot.cfg
 +++ b/build/config/armhf/armmp/netboot.cfg
 @@ -1,7 +1,7 @@
  MEDIA_TYPE = netboot image
  
 -TARGET = $(TEMP_INITRD) $(TEMP_KERNEL) efikamx
 -EXTRANAME = $(MEDIUM)
 +TARGET = $(INITRD) $(KERNEL) efikamx
 +EXTRANAME = $(MEDIUM)/
  
  MANIFEST-INITRD = netboot initrd
  MANIFEST-KERNEL = kernel image to netboot
 @@ -10,8 +10,8 @@ MANIFEST-KERNEL = kernel image to netboot
  efikamx:
   mkdir -p $(SOME_DEST)/$(EXTRANAME)/efikamx
   

Re: Consolidating armhf installer flavours

2013-09-24 Thread Cyril Brulebois
Hi Ian,

[ disclaimer: I don't know anything about arm*… ]

Ian Campbell i...@hellion.org.uk (2013-09-24):
 Perhaps it would be sensible to split this patch up and add the new
 flavour now, so folks can test it, and remove the other flavours a bit
 later on.

looks OK to me.

 My only concern would be the short term impact on the build time.
 Looking at [1] it seems that a successful build takes about 30 mins. In
 the short term we would be adding 3 variants
 (armmp-{netboot,network-console,netboot-gtk}) to the existing 4
 (mx5-{netboot,network-console,netboot-gtk} + vexpress-netboot), which I
 estimate would push the build time up to somewhere under an hour.

I don't think that's a problem at all, is it? Trading CPU/build time for
more hardware support looks very OK to me. :-)

If you're concerned about random joe developers, they can still build
one (or a handful) of target(s) under build/, they don't need to build
the whole source package, so that shouldn't be a problem on this side
either.

 Eventually we would drop back down to just the three armmp variants.

That's the kind of decisions I'll leave up to the arm folks. (See
disclaimer above.)

Mraw,
KiBi.


signature.asc
Description: Digital signature


Re: Consolidating armhf installer flavours

2013-09-24 Thread Nigel Sollars
HI all,

Perhaps a bit off topic,  Ive gotten Wheezy up and running on Panda ES -
4460,

Ive seen in /lib/modules you have a 3.11 rc6 candidate.  my questions are:

1.  hows the status of this one?

2. id like to know the version of u-boot used to build it

3.  the LOADADDR value used for the uImage / any other info

4. the boot.scr values used for bootm etc.

Ive had many issues with kernel beyond 3.7 / 3.8 myself

Regards
Nigel


On Tue, Sep 24, 2013 at 6:40 AM, Cyril Brulebois k...@debian.org wrote:

 Hi Ian,

 [ disclaimer: I don't know anything about arm*… ]

 Ian Campbell i...@hellion.org.uk (2013-09-24):
  Perhaps it would be sensible to split this patch up and add the new
  flavour now, so folks can test it, and remove the other flavours a bit
  later on.

 looks OK to me.

  My only concern would be the short term impact on the build time.
  Looking at [1] it seems that a successful build takes about 30 mins. In
  the short term we would be adding 3 variants
  (armmp-{netboot,network-console,netboot-gtk}) to the existing 4
  (mx5-{netboot,network-console,netboot-gtk} + vexpress-netboot), which I
  estimate would push the build time up to somewhere under an hour.

 I don't think that's a problem at all, is it? Trading CPU/build time for
 more hardware support looks very OK to me. :-)

 If you're concerned about random joe developers, they can still build
 one (or a handful) of target(s) under build/, they don't need to build
 the whole source package, so that shouldn't be a problem on this side
 either.

  Eventually we would drop back down to just the three armmp variants.

 That's the kind of decisions I'll leave up to the arm folks. (See
 disclaimer above.)

 Mraw,
 KiBi.




-- 
“Science is a differential equation. Religion is a boundary condition.”

  Alan Turing


Re: Consolidating armhf installer flavours

2013-09-24 Thread Ian Campbell
On Tue, 2013-09-24 at 12:40 +0200, Cyril Brulebois wrote:
 Hi Ian,
 
 [ disclaimer: I don't know anything about arm*… ]
 
 Ian Campbell i...@hellion.org.uk (2013-09-24):
  Perhaps it would be sensible to split this patch up and add the new
  flavour now, so folks can test it, and remove the other flavours a bit
  later on.
 
 looks OK to me.
 
  My only concern would be the short term impact on the build time.
  Looking at [1] it seems that a successful build takes about 30 mins. In
  the short term we would be adding 3 variants
  (armmp-{netboot,network-console,netboot-gtk}) to the existing 4
  (mx5-{netboot,network-console,netboot-gtk} + vexpress-netboot), which I
  estimate would push the build time up to somewhere under an hour.
 
 I don't think that's a problem at all, is it? Trading CPU/build time for
 more hardware support looks very OK to me. :-)

It's more of a concern with the kernel where a new flavour means the
build takes 4 days intsead of 3 ;-)

 If you're concerned about random joe developers, they can still build
 one (or a handful) of target(s) under build/, they don't need to build
 the whole source package, so that shouldn't be a problem on this side
 either.

Ack.

  Eventually we would drop back down to just the three armmp variants.
 
 That's the kind of decisions I'll leave up to the arm folks. (See
 disclaimer above.)

It's already happened on the kernel side (in trunk) so I think the
installer will have to change when 3.11 hits sid.

Ian.


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1380038514.23688.60.ca...@kazak.uk.xensource.com



Consolidating armhf installer flavours

2013-09-21 Thread Ian Campbell
With the current trunk/experimental 3.11 kernel the separate omap5, mx5
and vexpress flavours have been removed in favour of the armmp flavour
which supports all of those.

So I think we need to switch the installer over too. Since the 3.10
kernel already has the armmp flavour I don't think we even need wait for
3.11 to be uploaded.

I merged the existing mx5 and vexpress mediums so armmp creates all of
netboot network-console netboot-gtk. I also retained the efikamx uImage
stuff (but using the new kernel) since I guess it is useful? 

Should we have more of those uImage things for other armmp supported
platforms?

Ian.

commit 7de47649d4f47225f740d262de8d51c83be73688
Author: Ian Campbell i...@hellion.org.uk
Date:   Mon Aug 26 11:25:56 2013 +0100

Switch armhf to armmp kernel.

The other flavours have been removed in the 3.11 kernel, even in 3.10 the 
armmp
flavour supports mx5 and vexpress.

Merge all of the subtypes over.

diff --git a/build/config/armhf.cfg b/build/config/armhf.cfg
index 73b8a80..0287ef2 100644
--- a/build/config/armhf.cfg
+++ b/build/config/armhf.cfg
@@ -1,4 +1,4 @@
-SUBARCH_SUPPORTED = mx5 vexpress
+SUBARCH_SUPPORTED = armmp
 
 MKLIBS = mklibs --ldlib=/lib/ld-linux-armhf.so.3
 
diff --git a/build/config/armhf/mx5.cfg b/build/config/armhf/armmp.cfg
similarity index 69%
rename from build/config/armhf/mx5.cfg
rename to build/config/armhf/armmp.cfg
index a30c0a2..dd24206 100644
--- a/build/config/armhf/mx5.cfg
+++ b/build/config/armhf/armmp.cfg
@@ -1,4 +1,4 @@
 MEDIUM_SUPPORTED = netboot network-console netboot-gtk
 
 # The version of the kernel to use.
-KERNELVERSION := $(KERNELVERSION)-mx5
+KERNELVERSION := $(KERNELVERSION)-armmp
diff --git a/build/config/armhf/mx5/netboot-gtk.cfg 
b/build/config/armhf/armmp/netboot-gtk.cfg
similarity index 81%
rename from build/config/armhf/mx5/netboot-gtk.cfg
rename to build/config/armhf/armmp/netboot-gtk.cfg
index cb0b074..3817048 100644
--- a/build/config/armhf/mx5/netboot-gtk.cfg
+++ b/build/config/armhf/armmp/netboot-gtk.cfg
@@ -1,6 +1,6 @@
 MEDIA_TYPE = netboot image
 
-TARGET = $(TEMP_INITRD) $(TEMP_KERNEL) efikamx
+TARGET = $(INITRD) $(KERNEL) efikamx
 
 TYPE = netboot/gtk
 EXTRANAME = netboot/gtk/
@@ -16,8 +16,8 @@ EXTRAUDEBS += cdebconf-gtk-entropy
 efikamx:
mkdir -p $(SOME_DEST)/$(EXTRANAME)/efikamx
install -m 744 boot/arm/bootscr.mx5 
$(SOME_DEST)/$(EXTRANAME)/efikamx/bootscript
-   mkimage -A arm -O linux -T kernel -C none -a 0x90008000 -e 0x90008000 
-n Debian kernel -d $(TEMP_KERNEL) $(SOME_DEST)/$(EXTRANAME)/efikamx/uImage
-   mkimage -A arm -O linux -T ramdisk -C gzip -a 0x0 -e 0x0 -n 
debian-installer ramdisk -d $(TEMP_INITRD) 
$(SOME_DEST)/$(EXTRANAME)/efikamx/uInitrd
+   mkimage -A arm -O linux -T kernel -C none -a 0x90008000 -e 0x90008000 
-n Debian kernel -d $(KERNEL) $(SOME_DEST)/$(EXTRANAME)/efikamx/uImage
+   mkimage -A arm -O linux -T ramdisk -C gzip -a 0x0 -e 0x0 -n 
debian-installer ramdisk -d $(INITRD) 
$(SOME_DEST)/$(EXTRANAME)/efikamx/uInitrd
mkimage -A arm -O linux -a 0 -e 0 -T script -C none -n Efika MX Boot 
Script -d $(SOME_DEST)/$(EXTRANAME)/efikamx/bootscript 
$(SOME_DEST)/$(EXTRANAME)/efikamx/boot.scr
update-manifest $(SOME_DEST)/$(EXTRANAME)/efikamx/uImage Linux kernel 
for Efika MX
update-manifest $(SOME_DEST)/$(EXTRANAME)/efikamx/uInitrd initrd for 
Efika MX
diff --git a/build/config/armhf/mx5/netboot.cfg 
b/build/config/armhf/armmp/netboot.cfg
similarity index 77%
rename from build/config/armhf/mx5/netboot.cfg
rename to build/config/armhf/armmp/netboot.cfg
index 2456c93..e6886ba 100644
--- a/build/config/armhf/mx5/netboot.cfg
+++ b/build/config/armhf/armmp/netboot.cfg
@@ -1,7 +1,7 @@
 MEDIA_TYPE = netboot image
 
-TARGET = $(TEMP_INITRD) $(TEMP_KERNEL) efikamx
-EXTRANAME = $(MEDIUM)
+TARGET = $(INITRD) $(KERNEL) efikamx
+EXTRANAME = $(MEDIUM)/
 
 MANIFEST-INITRD = netboot initrd
 MANIFEST-KERNEL = kernel image to netboot
@@ -10,8 +10,8 @@ MANIFEST-KERNEL = kernel image to netboot
 efikamx:
mkdir -p $(SOME_DEST)/$(EXTRANAME)/efikamx
install -m 744 boot/arm/bootscr.mx5 
$(SOME_DEST)/$(EXTRANAME)/efikamx/bootscript
-   mkimage -A arm -O linux -T kernel -C none -a 0x90008000 -e 0x90008000 
-n Debian kernel -d $(TEMP_KERNEL) $(SOME_DEST)/$(EXTRANAME)/efikamx/uImage
-   mkimage -A arm -O linux -T ramdisk -C gzip -a 0x0 -e 0x0 -n 
debian-installer ramdisk -d $(TEMP_INITRD) 
$(SOME_DEST)/$(EXTRANAME)/efikamx/uInitrd
+   mkimage -A arm -O linux -T kernel -C none -a 0x90008000 -e 0x90008000 
-n Debian kernel -d $(KERNEL) $(SOME_DEST)/$(EXTRANAME)/efikamx/uImage
+   mkimage -A arm -O linux -T ramdisk -C gzip -a 0x0 -e 0x0 -n 
debian-installer ramdisk -d $(INITRD) 
$(SOME_DEST)/$(EXTRANAME)/efikamx/uInitrd
mkimage -A arm -O linux -a 0 -e 0 -T script -C none -n Efika MX Boot 
Script -d $(SOME_DEST)/$(EXTRANAME)/efikamx/bootscript 
$(SOME_DEST)/$(EXTRANAME)/efikamx/boot.scr
update-manifest 

Re: armhf installer?

2012-02-10 Thread Hector Oron
Hello Peter,

2012/2/9 peter green plugw...@p10link.net:
 E: Unable to locate package kernel-image-3.1.0-1-mx5-di
 E: Couldn't find any package by regex 'kernel-image-3.1.0-1-mx5-di'

 Are the errors i'm seeing just caused by ongoing kernel transitions in sid
 or are they indicative of a bigger problem?

That's more likely the problem.

 What needs to be done to get armhf d-i into the official daily builds system
 (and eventually into releases)?

Most support for Efika MX and SB is there, just
flash-kernel/experimental yet does not work. Adding i.MX53 QSB LOCO
board is pending too.

 What needs to be done to support the beagleboard series in armhf d-i?

You need to add omap support to the debian-installer infrastructure.
It is already half supported, it is probably just missing machine
specific bits.

-- 
 Héctor Orón  -.. . -... .. .- -.   -.. . ...- . .-.. --- .--. . .-.


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CAODfWeE=biafjrh+celf0ffarawzlcynkgwx9e2o_uzzjq_...@mail.gmail.com



armhf installer?

2012-02-08 Thread peter green
It seems there are some unofficial builds at 
http://people.debian.org/~zumbi/di-armhf/ but only for one particular 
peice of hardware (and it's not the one I have)


I tried building d-i mx5_netboot following the instructioions on the 
wiki. However it failed with


Reading package lists... Done
Building dependency tree... Done
Need to download: anna archdetect bogl-bterm-udeb busybox-udeb 
cdebconf-newt-terminal cdebconf-newt-udeb cdebconf-priority 
cdebconf-text-udeb cdebconf-udeb choose-mirror choose-mirror-bin 
console-keymaps-at debian-archive-keyring-udeb di-utils di-utils-reboot 
di-utils-shell di-utils-terminfo download-installer env-preseed 
ethdetect file-preseed gpgv-udeb hw-detect initrd-preseed 
installation-locale kbd-chooser kernel-image-3.1.0-1-mx5-di 
libblkid1-udeb libcrypto1.0.0-udeb libdebconfclient0-udeb 
libdebian-installer4-udeb libfribidi0-udeb libiw30-udeb libnl-3-200-udeb 
libnl-genl-3-200-udeb libnss-dns-udeb libtextwrap1-udeb libuuid1-udeb 
localechooser lowmemcheck main-menu mmc-modules-3.1.0-1-mx5-di 
module-init-tools-udeb nano-udeb net-retriever netcfg network-preseed 
nic-modules-3.1.0-1-mx5-di pciutils-udeb preseed-common rescue-check 
rootskel save-logs udev-udeb udpkg util-linux-udeb wpasupplicant-udeb 
zlib1g-udeb

Reading package lists... Done
Building dependency tree... Done
E: Unable to locate package kernel-image-3.1.0-1-mx5-di
E: Couldn't find any package by regex 'kernel-image-3.1.0-1-mx5-di'
E: Unable to locate package mmc-modules-3.1.0-1-mx5-di
E: Couldn't find any package by regex 'mmc-modules-3.1.0-1-mx5-di'
E: Unable to locate package nic-modules-3.1.0-1-mx5-di
E: Couldn't find any package by regex 'nic-modules-3.1.0-1-mx5-di'
make[2]: *** [stamps/get_udebs-mx5_netboot-stamp] Error 100
make[1]: *** [_build] Error 2
make: *** [build_mx5_netboot] Error 2
root@debian:/debian-installer/installer/build#

Further even if it were to build I couldn't test it as it doesn't match 
my hardware (i'm running a beagleboard xm).


Are the errors i'm seeing just caused by ongoing kernel transitions in 
sid or are they indicative of a bigger problem?
What needs to be done to get armhf d-i into the official daily builds 
system (and eventually into releases)?

What needs to be done to support the beagleboard series in armhf d-i?


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4f333dca.6090...@p10link.net



Re: armhf installer?

2012-02-08 Thread harish badrinath
Hello,

You can
(a)build your own udebs and build you d-i image.
(b)Download the required udeb and place it in localudebs directory.
Please see 
http://d-i.alioth.debian.org/doc/talks/debconf6/paper/index.html#id2536380.

If both are not possible, I do not know how to proceed.


Regards,
Harish


On Thu, Feb 9, 2012 at 9:00 AM, peter green plugw...@p10link.net wrote:
 It seems there are some unofficial builds at
 http://people.debian.org/~zumbi/di-armhf/ but only for one particular peice
 of hardware (and it's not the one I have)

 I tried building d-i mx5_netboot following the instructioions on the wiki.
 However it failed with

 Reading package lists... Done
 Building dependency tree... Done
 Need to download: anna archdetect bogl-bterm-udeb busybox-udeb
 cdebconf-newt-terminal cdebconf-newt-udeb cdebconf-priority
 cdebconf-text-udeb cdebconf-udeb choose-mirror choose-mirror-bin
 console-keymaps-at debian-archive-keyring-udeb di-utils di-utils-reboot
 di-utils-shell di-utils-terminfo download-installer env-preseed ethdetect
 file-preseed gpgv-udeb hw-detect initrd-preseed installation-locale
 kbd-chooser kernel-image-3.1.0-1-mx5-di libblkid1-udeb libcrypto1.0.0-udeb
 libdebconfclient0-udeb libdebian-installer4-udeb libfribidi0-udeb
 libiw30-udeb libnl-3-200-udeb libnl-genl-3-200-udeb libnss-dns-udeb
 libtextwrap1-udeb libuuid1-udeb localechooser lowmemcheck main-menu
 mmc-modules-3.1.0-1-mx5-di module-init-tools-udeb nano-udeb net-retriever
 netcfg network-preseed nic-modules-3.1.0-1-mx5-di pciutils-udeb
 preseed-common rescue-check rootskel save-logs udev-udeb udpkg
 util-linux-udeb wpasupplicant-udeb zlib1g-udeb
 Reading package lists... Done
 Building dependency tree... Done
 E: Unable to locate package kernel-image-3.1.0-1-mx5-di
 E: Couldn't find any package by regex 'kernel-image-3.1.0-1-mx5-di'
 E: Unable to locate package mmc-modules-3.1.0-1-mx5-di
 E: Couldn't find any package by regex 'mmc-modules-3.1.0-1-mx5-di'
 E: Unable to locate package nic-modules-3.1.0-1-mx5-di
 E: Couldn't find any package by regex 'nic-modules-3.1.0-1-mx5-di'
 make[2]: *** [stamps/get_udebs-mx5_netboot-stamp] Error 100
 make[1]: *** [_build] Error 2
 make: *** [build_mx5_netboot] Error 2
 root@debian:/debian-installer/installer/build#

 Further even if it were to build I couldn't test it as it doesn't match my
 hardware (i'm running a beagleboard xm).

 Are the errors i'm seeing just caused by ongoing kernel transitions in sid
 or are they indicative of a bigger problem?
 What needs to be done to get armhf d-i into the official daily builds system
 (and eventually into releases)?
 What needs to be done to support the beagleboard series in armhf d-i?


 --
 To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
 with a subject of unsubscribe. Trouble? Contact
 listmas...@lists.debian.org
 Archive: http://lists.debian.org/4f333dca.6090...@p10link.net



-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CAFUoqbce5gAouthHC1Y7asUV4PKScj=qrz8rh9jn6zzhgrr...@mail.gmail.com