Hi
Kernel is 2.6.30-03984-g45e3e19
It is Ultra10 (UP) with 1Gb RAM
PROMLIB: Sun IEEE Boot Prom 'OBP 3.25.3 2000/06/29 14:12'
PROMLIB: Root node compatible:
Linux version 2.6.30-03984-g45e3e19 (alexb@sparky) (gcc version 4.3.3
(Gentoo 4.3.3 p1.0) ) #27 PREEMPT Mon Jun 15 13:42:41 MSD 2009
console [earlyprom0] enabled
ARCH: SUN4U
Ethernet address: 08:00:20:ff:e6:ff
Kernel: Using 4 locked TLB entries for main kernel image.
Remapping the kernel... done.
OF stdout device is: /pci@1f,0/pci@1,1/SUNW,m64B@2
PROM: Built device tree with 41922 bytes of memory.
Top of RAM: 0x3ff44000, Total RAM: 0x3ff34000
Memory hole size: 0MB
[0000000200000000-fffff80001400000] page_structs=131072 node=0 entry=0/0
[0000000200000000-fffff80001800000] page_structs=131072 node=0 entry=1/0
Allocated 1056768 bytes for kernel page tables.
Zone PFN ranges:
Normal 0x00000000 -> 0x0001ffa2
Movable zone start PFN for each node
early_node_map[3] active PFN ranges
0: 0x00000000 -> 0x0001ff7f
0: 0x0001ff80 -> 0x0001ff98
0: 0x0001ff9f -> 0x0001ffa2
On node 0 totalpages: 130970
Normal zone: 1024 pages used for memmap
Normal zone: 0 pages reserved
Normal zone: 129946 pages, LIFO batch:15
Booting Linux...
Built 1 zonelists in Zone order, mobility grouping on. Total pages: 129946
Kernel command line: root=/dev/sda2
PID hash table entries: 4096 (order: 12, 32768 bytes)
Dentry cache hash table entries: 131072 (order: 7, 1048576 bytes)
Inode-cache hash table entries: 65536 (order: 6, 524288 bytes)
------------[ cut here ]------------
WARNING: at kernel/lockdep.c:2282 lockdep_trace_alloc+0xd0/0xf8()
Modules linked in:
Call Trace:
[0000000000450cf8] warn_slowpath_common+0x50/0x84
[0000000000450d48] warn_slowpath_null+0x1c/0x2c
[0000000000474148] lockdep_trace_alloc+0xd0/0xf8
[000000000049149c] __alloc_pages_internal+0x30/0x434
[00000000007dad60] mem_init+0x234/0x304
[00000000007d6638] start_kernel+0x16c/0x30c
[000000000066b14c] tlb_fixup_done+0x88/0x90
[0000000000000000] (null)
---[ end trace 139ce121c98e96c9 ]---
Memory: 1023456k available (2536k kernel code, 1328k data, 144k init)
[fffff80000000000,000000003ff44000]
SLUB: Genslabs=14, HWalign=32, Order=0-3, MinObjects=0, CPUs=1, Nodes=1
Preemptible RCU implementation.
NR_IRQS:255
------------[ cut here ]------------
WARNING: at mm/bootmem.c:535 alloc_arch_preferred_bootmem+0x34/0x64()
Modules linked in:
Call Trace:
[0000000000450cf8] warn_slowpath_common+0x50/0x84
[0000000000450d48] warn_slowpath_null+0x1c/0x2c
[00000000007ddf9c] alloc_arch_preferred_bootmem+0x34/0x64
[00000000007de75c] ___alloc_bootmem_nopanic+0x20/0xc0
[00000000007de8d8] ___alloc_bootmem+0x10/0x44
[00000000007dea8c] __alloc_bootmem+0x10/0x20
[00000000007d7388] init_IRQ+0xcc/0x1d4
[00000000007d6690] start_kernel+0x1c4/0x30c
[000000000066b14c] tlb_fixup_done+0x88/0x90
[0000000000000000] (null)
---[ end trace 139ce121c98e96ca ]---
clocksource: mult[245d1] shift[16]
clockevent: mult[70a3d70a] shift[32]
I know about them, I saw these messages too.
kmalloc() initialization got moved earlier so large swaths of sparc
early initialization need to move from bootmem to kmalloc or similar.
It's just a harmless warning as the bootmem code calls kmalloc()
if that's available already.