WARNING: vmlinux.o (.PPC.EMB.apuinfo): unexpected non-allocatable section.

WARNING: vmlinux.o (.PPC.EMB.apuinfo): unexpected non-allocatable section.

[Timur Tabi]

When trying to build a 64-bit kernel, I'm getting this warning message:

WARNING: vmlinux.o (.PPC.EMB.apuinfo): unexpected non-allocatable section.
Did you forget to use "ax"/"aw" in a .S file?
Note that for example <linux/init.h> contains
section definitions for use in .S files.

This kernel continues to compile and link, but when I try to boot it, it stops
here (this is under the hypervisor):

Linux version 3.0.4-31-00029-g06d86c8-dirty (b04825@efes) (gcc version 4.5.1
(Sourcery G++ Lite 2010.09-55) ) #1 SMP Mon Sep 19 14:49:48 CDT 2011
[boot]0012 Setup Arch
Found FSL PCI host bridge at 0x0000000ffe200000. Firmware bus number: 0->1
PCI host bridge /devices/pci0  ranges:
 MEM 0x0000000c00000000..0x0000000c1fffffff -> 0x00000000e0000000
  IO 0x0000000ff8000000..0x0000000ff800ffff -> 0x0000000000000000
/devices/pci0: PCICSRBAR @ 0xdf000000
Found FSL PCI host bridge at 0x0000000ffe202000. Firmware bus number: 0->1
PCI host bridge /devices/pci2  ranges:
 MEM 0x0000000c40000000..0x0000000c5fffffff -> 0x00000000e0000000
  IO 0x0000000ff8020000..0x0000000ff802ffff -> 0x0000000000000000
/devices/pci2: PCICSRBAR @ 0xdf000000
P5020 DS board from Freescale Semiconductor
Zone PFN ranges:
  DMA      0x00000000 -> 0x00020000
  Normal   empty
Movable zone start PFN for each node
early_node_map[1] active PFN ranges
    0: 0x00000000 -> 0x00020000
MMU: Allocated 2112 bytes of context maps for 255 contexts
[boot]0015 Setup Done
PERCPU: Embedded 13 pages/cpu @c000000002d00000 s23872 r0 d29376 u524288
Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 129280
Kernel command line:
PID hash table entries: 2048 (order: 2, 16384 bytes)
Dentry cache hash table entries: 65536 (order: 7, 524288 bytes)
Inode-cache hash table entries: 32768 (order: 6, 262144 bytes)
Memory: 425104k/524288k available (8500k kernel code, 99184k reserved, 968k
data, 702k bss, 300k init)
Hierarchical RCU implementation.
	RCU debugfs-based tracing is enabled.
	CONFIG_RCU_FANOUT set to non-default value of 32
NR_IRQS:512 nr_irqs:512 16
clocksource: timebase mult[a000000] shift[22] registered
Console: colour dummy device 80x25
console [tty0] enabled
console [ttyEHV0] enabled
ehv-bc: registered console driver for byte channel 300
pid_max: default: 32768 minimum: 301
Mount-cache hash table entries: 256
e500 family performance monitor hardware support registered

If I disable performance counter support, it just stops at the "Mount-cache hash
table entries: 256" line.


-- 
Timur Tabi
Linux kernel developer at Freescale

Re: WARNING: vmlinux.o (.PPC.EMB.apuinfo): unexpected non-allocatable section.

[Scott Wood]

On 09/19/2011 03:18 PM, Timur Tabi wrote:
> This kernel continues to compile and link, but when I try to boot it, it stops
> here (this is under the hypervisor):

"The hypervisor" is ambiguous.  We have Topaz, and KVM.

> Linux version 3.0.4-31-00029-g06d86c8-dirty (b04825@efes) (gcc version 4.5.1
> (Sourcery G++ Lite 2010.09-55) ) #1 SMP Mon Sep 19 14:49:48 CDT 2011
> [boot]0012 Setup Arch
> Found FSL PCI host bridge at 0x0000000ffe200000. Firmware bus number: 0->1
> PCI host bridge /devices/pci0  ranges:
>  MEM 0x0000000c00000000..0x0000000c1fffffff -> 0x00000000e0000000
>   IO 0x0000000ff8000000..0x0000000ff800ffff -> 0x0000000000000000
> /devices/pci0: PCICSRBAR @ 0xdf000000
> Found FSL PCI host bridge at 0x0000000ffe202000. Firmware bus number: 0->1
> PCI host bridge /devices/pci2  ranges:
>  MEM 0x0000000c40000000..0x0000000c5fffffff -> 0x00000000e0000000
>   IO 0x0000000ff8020000..0x0000000ff802ffff -> 0x0000000000000000
> /devices/pci2: PCICSRBAR @ 0xdf000000
> P5020 DS board from Freescale Semiconductor
> Zone PFN ranges:
>   DMA      0x00000000 -> 0x00020000
>   Normal   empty
> Movable zone start PFN for each node
> early_node_map[1] active PFN ranges
>     0: 0x00000000 -> 0x00020000
> MMU: Allocated 2112 bytes of context maps for 255 contexts
> [boot]0015 Setup Done
> PERCPU: Embedded 13 pages/cpu @c000000002d00000 s23872 r0 d29376 u524288
> Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 129280
> Kernel command line:
> PID hash table entries: 2048 (order: 2, 16384 bytes)
> Dentry cache hash table entries: 65536 (order: 7, 524288 bytes)
> Inode-cache hash table entries: 32768 (order: 6, 262144 bytes)
> Memory: 425104k/524288k available (8500k kernel code, 99184k reserved, 968k
> data, 702k bss, 300k init)
> Hierarchical RCU implementation.
> 	RCU debugfs-based tracing is enabled.
> 	CONFIG_RCU_FANOUT set to non-default value of 32
> NR_IRQS:512 nr_irqs:512 16
> clocksource: timebase mult[a000000] shift[22] registered
> Console: colour dummy device 80x25
> console [tty0] enabled
> console [ttyEHV0] enabled
> ehv-bc: registered console driver for byte channel 300
> pid_max: default: 32768 minimum: 301
> Mount-cache hash table entries: 256
> e500 family performance monitor hardware support registered
> 
> If I disable performance counter support, it just stops at the "Mount-cache hash
> table entries: 256" line.

These messages generally tell you what the kernel just did, not what
it's about to do (and thus what hung).

Have you tried using CCS (or Topaz's debug functionality) to see where
it's hung?

-Scott