/boot/vmlinux: not a valid ELF image

Bug #329400 reported by Matteo Settenvini
4
Affects Status Importance Assigned to Milestone
linux-ports (Ubuntu)
Fix Released
Undecided
Luke Yelavich

Bug Description

Kernel 2.6.28-1.6 on PPC64 (iMac G5) fails immediately at boot with error "/boot/vmlinux: not a valid ELF image".
Last image known surely booting and working: linux-image-2.6.27-1-powerpc64-smp (2.6.27-1.1).

matteo@Dahlia:/boot$ ls -l
totale 46608
-rw-r--r-- 1 root root 481920 2008-11-07 04:40 abi-2.6.27-1-powerpc64-smp
-rw-r--r-- 1 root root 511053 2009-02-13 18:52 abi-2.6.28-1-powerpc64-smp
-rw-r--r-- 1 root root 79113 2008-11-07 04:40 config-2.6.27-1-powerpc64-smp
-rw-r--r-- 1 root root 81717 2009-02-13 18:52 config-2.6.28-1-powerpc64-smp
lrwxrwxrwx 1 root root 33 2009-02-14 11:39 initrd.img -> initrd.img-2.6.28-1-powerpc64-smp
-rw-r--r-- 1 root root 9726549 2009-02-13 15:25 initrd.img-2.6.27-1-powerpc64-smp
-rw-r--r-- 1 root root 10086168 2009-02-14 11:40 initrd.img-2.6.28-1-powerpc64-smp
lrwxrwxrwx 1 root root 33 2008-11-28 22:30 initrd.img.old -> initrd.img-2.6.27-1-powerpc64-smp
-rw-r--r-- 1 root root 1769032 2008-11-07 04:40 System.map-2.6.27-1-powerpc64-smp
-rw-r--r-- 1 root root 1955137 2009-02-13 18:52 System.map-2.6.28-1-powerpc64-smp
lrwxrwxrwx 1 root root 30 2009-02-14 11:39 vmlinux -> vmlinux-2.6.28-1-powerpc64-smp
-rw-r--r-- 1 root root 9460295 2008-11-07 04:40 vmlinux-2.6.27-1-powerpc64-smp
-rw-r--r-- 1 root root 13513102 2009-02-13 18:52 vmlinux-2.6.28-1-powerpc64-smp
lrwxrwxrwx 1 root root 30 2008-11-28 22:30 vmlinux.old -> vmlinux-2.6.27-1-powerpc64-smp

Related branches

Revision history for this message
Matteo Settenvini (tchernobog) wrote :

I'm attaching all the relevant files as per request. However, notice they refer to 2.6.27 since 2.6.28 doesn't boot.

Revision history for this message
Matteo Settenvini (tchernobog) wrote :
Revision history for this message
Matteo Settenvini (tchernobog) wrote :
Revision history for this message
Matteo Settenvini (tchernobog) wrote :
Revision history for this message
Luke Yelavich (themuso) wrote : Re: [Bug 329400] Re: /boot/vmlinux: not a valid ELF image

Confirmed on a PowerMac G5 dual 1.8 machine. WIll look into it.

affects ubuntu/linux-ports
 status confirmed
 assignee themuso

Changed in linux-ports:
assignee: nobody → themuso
status: New → Confirmed
Revision history for this message
Luke Yelavich (themuso) wrote : Re: [Bug 329400] [NEW] /boot/vmlinux: not a valid ELF image

 affects ubuntu/linux-ports
 status fixcommitted

Changed in linux-ports:
status: Confirmed → Fix Committed
Revision history for this message
Launchpad Janitor (janitor) wrote :
Download full text (9.1 KiB)

This bug was fixed in the package linux-ports - 2.6.28-2.7

---------------
linux-ports (2.6.28-2.7) jaunty; urgency=low

  [ Luke Yelavich ]

  * Disable RELOCATABLE on powerpc64-smp
    - LP: #329400
  * Turn ABI/module checking back on
  * Adjust ia64 to reduce kernel size, and increase functionality
  * add the ext4 module for Debian Installer
  * add some more IDE modules for Debian Installer
  * enable Ubuntu modules for powerpc
    - LP: #329704
  * add the dm-raid4-5 module for Debian Installer
  * fix up some differences in config for modules that should be buildable
    on all powerpc variants
  * enable maple platform support in the powerpc64-smp kernel
  * a few more ia64 configuration tweaks
  * make the Ubuntu added drbd module compile properly on powerpc
  * move rtc-ppc into the powerpc kernels
    - LP: #26658
  * re-add device-tree-compiler for powerpc again

  [ Michael Casadevall ]

  * [sparc] Prevent __LITTLE_ENDIAN from being defined in the kernel
    headers

  Rebase on top of Jaunty 2.6.28-8.24:

  [ Scott James Remnant ]

  * Change CPU_FREQ_DEFAULT_GOV_ONDEMAND to y
  * SAUCE: Link acpi-cpufreq.o first

  [ Tim Gardner ]

  * Build in CPU Frequency scaling drivers

  [ Andy Whitcroft ]

  * include the kernel configuration in the sub-flavour images
    - LP: #328859

  [ Tim Gardner ]

  * Revert "SAUCE: (drop after 2.6.28) [eCryptfs] Regression in unencrypted
    filename symlinks" in favor of upstream commit.
  * Fix compile issues with qc-usb
  * SAUCE: (remove after 2.6.28) V4L/DVB (10216): saa7127: fix broken
    S-Video with saa7129
    - LP: #329267

  [ Upstream Kernel Changes ]

  * Subject:SAUCE: LP#193970 iwlagn: fix hw-rfkill while the interface is
    down
    - LP: #193970
  * x86, vmi: put a missing paravirt_release_pmd in pgd_dtor
  * nbd: fix I/O hang on disconnected nbds
  * mac80211: restrict to AP in outgoing interface heuristic
  * w1: w1 temp calculation overflow fix
  * zd1211rw: adding 0ace:0xa211 as a ZD1211 device
  * zd1211rw: treat MAXIM_NEW_RF(0x08) as UW2453_RF(0x09) for TP-Link
    WN322/422G
  * parport: parport_serial, don't bind netmos ibm 0299
  * syscall define: fix uml compile bug
  * kernel-doc: fix syscall wrapper processing
  * Fix page writeback thinko, causing Berkeley DB slowdown
  * write-back: fix nr_to_write counter
  * writeback: fix break condition
  * mm: rearrange exit_mmap() to unlock before arch_exit_mmap
  * powerpc/fsl-booke: Fix mapping functions to use phys_addr_t
  * lockd: fix regression in lockd's handling of blocked locks
  * sctp: Fix crc32c calculations on big-endian arhes.
  * sctp: Correctly start rtx timer on new packet transmissions.
  * sctp: Properly timestamp outgoing data chunks for rtx purposes
  * net: Fix frag_list handling in skb_seq_read
  * net: Fix OOPS in skb_seq_read().
  * drivers/net/skfp: if !capable(CAP_NET_ADMIN): inverted logic
  * ipv4: fix infinite retry loop in IP-Config
  * net: Fix userland breakage wrt. linux/if_tunnel.h
  * net: packet socket packet_lookup_frame fix
  * packet: Avoid lock_sock in mmap handler
  * sungem: Soft lockup in sungem on Netra AC200 when switching interface
    up
  * udp: Fix UDP short packet false ...

Read more...

Changed in linux-ports:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.