linaro-release team mailing list archive

[Nicolas Pitre <859473@xxxxxxxxxxxxxxxxxx>]

On Mon, 17 Oct 2011, James Westby wrote:

> asac asked in a private mail thread about how to enable thumb2 on an
> otherwise plain defconfig.

Two things:

1) If you want your kernel to support Thumb2 user space binaries, you 
   must have CONFIG_ARM_THUMB=y in your kernel config.  This is probably
   set as our user space has been compiled to Thumb2 for quite a while.

2) If you want your kernel itself to be a Thumb2 binary, then you must 
   have CONFIG_THUMB2_KERNEL=y.  This is however available only for 
   ARMv7 and above.  For example, in the OMAP case, you can't support 
   OMAP2 with a Thumb2 kernel, therefore the CONFIG_THUMB2_KERNEL option
   will be visible only if OMAP2 support is configured out (OMAP3 + 
   OMAP4 is fine).

Also it is not necessary to match the kernel and user space "thumbness". 
However there was a bug in the kernel compiled for ARM mode when a Thumb 
mode user space was used with it when that kernel was configured for 
both ARMv6 and ARMv7 (e.g. omap2plus_defconfig).  The fix was available 
in linux-linaro-3.0 but this was missing from linux-linaro-3.1 until a 
few hours ago.

-- 
You received this bug notification because you are a member of Linaro
Release Team, which is subscribed to the bug report.
https://bugs.launchpad.net/bugs/859473

Title:
  CI kernels causing many "Illegal Instruction"s

Status in LAVA Validation Lab:
  Invalid
Status in LAVA Automated Testing Framework:
  Invalid
Status in Linaro CI:
  Confirmed

Bug description:
  Recently there have been lots of validation failures for kernels from
  CI.

  These failures manifest as "Illegal Instruction" errors all over the logs
  (not just during tests, but booting as well.)

  It seems that the kernel is causing a lot of SIGILLs, but we don't know
  why yet.

  To reproduce
  ------------

  Combining

  http://ci.linaro.org/kernel_hwpack/omap3/hwpack_linaro-omap3_20111007-0028_armel_supported.tar.gz
  http://snapshots.linaro.org/11.05-daily/linaro-nano/20111005/1/images/tar/nano-n-tar-20111005-1.tar.gz

  with linaro-media-create (--rootfs ext2) should give you an image that
  shows the problem.

  Information wanted
  ------------------

  There are currently two things that we would like more information on to try
  and narrow down the cause of the problem.

  1) A core file from a crashing application

  To get this

    Create the image as described above
    Boot in to it
    Run "ulimit -c 1024"
    Run "hwclock"
    Watch it crash with "Illegal Instruction (core dumped)"
    Get that core from the filesystem.

  2) Test a defconfig build

  If this succeeds then it suggests a problem with the way that the kernel is
  being built on ci.linaro.org.

  To do this

    Get the tip of linus' tree
    Build with omap2plus_defconfig
    Boot the resulting kernel

  If that shows no SIGILL problems, then try a cross-build from an x86
  host.

  If neither of those show problems then we can concentrate on
  ci.linaro.org.

To manage notifications about this bug go to:
https://bugs.launchpad.net/lava-lab/+bug/859473/+subscriptions