[Qemu-devel] [Bug 586221] [NEW] Linux on ARM/Mainstone machine fails at bootstrap

[Qemu-devel] [Bug 586221] [NEW] Linux on ARM/Mainstone machine fails at bootstrap

[Daniel Rossier]

Public bug reported:

When QEMU (0.12.4) starts a ARM machine which boots a Linux, it
immediately fails because the initial PC is wrong (is equal to 0 instead
of 0xa0000000). First investigations indicate that the reset handlers
queue does not return the correct opaque structure in main_cpu_reset()
which is passed during the insertion within arm_load_kernel(). I checked
the value of r15 and the value is correct at the insertion, but at 0x0
in main_cpu_reset().

There may be still a bug in the new queue management functions (in qemu-
queue.h).

** Affects: qemu
     Importance: Undecided
         Status: New

-- 
Linux on ARM/Mainstone machine fails at bootstrap
https://bugs.launchpad.net/bugs/586221
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.

Status in QEMU: New

Bug description:
When QEMU (0.12.4) starts a ARM machine which boots a Linux, it immediately fails because the initial PC is wrong (is equal to 0 instead of 0xa0000000). First investigations indicate that the reset handlers queue does not return the correct opaque structure in main_cpu_reset() which is passed during the insertion within arm_load_kernel(). I checked the value of r15 and the value is correct at the insertion, but at 0x0 in main_cpu_reset().

There may be still a bug in the new queue management functions (in qemu-queue.h).

[Qemu-devel] [Bug 586221] Re: Linux on ARM/Mainstone machine fails at bootstrap

[Daniel Rossier]

Workaround: just comment the last line
"qemu_register_reset(main_cpu_reset, env);" of the function
"arm_load_kernel()" in hw/arm_boot.c

-- 
Linux on ARM/Mainstone machine fails at bootstrap
https://bugs.launchpad.net/bugs/586221
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.

Status in QEMU: New

Bug description:
When QEMU (0.12.4) starts a ARM machine which boots a Linux, it immediately fails because the initial PC is wrong (is equal to 0 instead of 0xa0000000). First investigations indicate that the reset handlers queue does not return the correct opaque structure in main_cpu_reset() which is passed during the insertion within arm_load_kernel(). I checked the value of r15 and the value is correct at the insertion, but at 0x0 in main_cpu_reset().

There may be still a bug in the new queue management functions (in qemu-queue.h).

[Qemu-devel] [Bug 586221] Re: Linux on ARM/Mainstone machine fails at bootstrap

[Lars Munch]

There are already patches pending to solve these issues:

http://article.gmane.org/gmane.comp.emulators.qemu/69598
and
http://article.gmane.org/gmane.comp.emulators.qemu/69597

Hopefully they will be reviewed/applied soon.

-- 
Linux on ARM/Mainstone machine fails at bootstrap
https://bugs.launchpad.net/bugs/586221
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.

Status in QEMU: New

Bug description:
When QEMU (0.12.4) starts a ARM machine which boots a Linux, it immediately fails because the initial PC is wrong (is equal to 0 instead of 0xa0000000). First investigations indicate that the reset handlers queue does not return the correct opaque structure in main_cpu_reset() which is passed during the insertion within arm_load_kernel(). I checked the value of r15 and the value is correct at the insertion, but at 0x0 in main_cpu_reset().

There may be still a bug in the new queue management functions (in qemu-queue.h).

[Qemu-devel] [Bug 586221] Re: Linux on ARM/Mainstone machine fails at bootstrap

[Lars Munch]

Fixed in e03c22a98cf5deffd0dec2f9ff88a600aa330bc4

** Changed in: qemu
       Status: New => Fix Committed

-- 
Linux on ARM/Mainstone machine fails at bootstrap
https://bugs.launchpad.net/bugs/586221
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.

Status in QEMU: Fix Committed

Bug description:
When QEMU (0.12.4) starts a ARM machine which boots a Linux, it immediately fails because the initial PC is wrong (is equal to 0 instead of 0xa0000000). First investigations indicate that the reset handlers queue does not return the correct opaque structure in main_cpu_reset() which is passed during the insertion within arm_load_kernel(). I checked the value of r15 and the value is correct at the insertion, but at 0x0 in main_cpu_reset().

There may be still a bug in the new queue management functions (in qemu-queue.h).

[Qemu-devel] [Bug 586221] Re: Linux on ARM/Mainstone machine fails at bootstrap

[Aurelien Jarno]

** Changed in: qemu
       Status: Fix Committed => Fix Released

-- 
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/586221

Title:
  Linux on ARM/Mainstone machine fails at bootstrap

Status in QEMU:
  Fix Released

Bug description:
  When QEMU (0.12.4) starts a ARM machine which boots a Linux, it
  immediately fails because the initial PC is wrong (is equal to 0
  instead of 0xa0000000). First investigations indicate that the reset
  handlers queue does not return the correct opaque structure in
  main_cpu_reset() which is passed during the insertion within
  arm_load_kernel(). I checked the value of r15 and the value is correct
  at the insertion, but at 0x0 in main_cpu_reset().

  There may be still a bug in the new queue management functions (in
  qemu-queue.h).