[ADMtek 5120] 64M sdram on board but only 16M is deteted and usable

[ADMtek 5120] 64M sdram on board but only 16M is deteted and usable

[Zhuang Yuyao]

Hi,

I've just upgraded the SDRAM on my adm5120 board from 16M to 64M, but 
while the board is booting, it still reports that only 16M sdram is 
availible.
Since I do not have the source code of the bootloader, is there any way 
to let the board to boot with 64M sdram, or should I change to another 
bootloader?

Thanks very much.

Here is the boot message.
Linux version 2.6.12 (root@debian) (gcc version 3.4.2) #28 Wed Aug 31 
20:51:01 MDT 2005
CPU revision is: 0001800b
ADM5120 board setup
Determined physical RAM map:
 memory: 00b0b000 @ 004f5000 (usable)
Built 1 zonelists
Kernel command line:
Primary instruction cache 8kB, physically tagged, 2-way, linesize 16 bytes.
Primary data cache 8kB, 2-way, linesize 16 bytes.
Synthesized TLB refill handler (19 instructions).
Synthesized TLB load handler fastpath (31 instructions).
Synthesized TLB store handler fastpath (31 instructions).
Synthesized TLB modify handler fastpath (30 instructions).
PID hash table entries: 128 (order: 7, 2048 bytes)
CPU clock: 175MHz
Dentry cache hash table entries: 4096 (order: 2, 16384 bytes)
Inode-cache hash table entries: 2048 (order: 1, 8192 bytes)
Memory: 11136k/11308k available (2369k kernel code, 160k reserved, 375k 
data, 2004k init, 0k highmem)
Mount-cache hash table entries: 512
Checking for 'wait' instruction...  available.
NET: Registered protocol family 16
System has PCI BIOS
JFFS2 version 2.2. (C) 2001-2003 Red Hat, Inc.
Initializing Cryptographic API
ADM5120 LED & GPIO driver
ttyS0 at I/O 0x12600000 (irq = 1) is a ADM5120
ttyS1 at I/O 0x12800000 (irq = 2) is a ADM5120
io scheduler noop registered
RAMDISK driver initialized: 16 RAM disks of 7168K size 1024 blocksize
PPP generic driver version 2.4.2
NET: Registered protocol family 24
eth0: ADM5120 switch port0
eth1: ADM5120 switch port1
eth2: ADM5120 switch port2
eth3: ADM5120 switch port3
eth4: ADM5120 switch port4
eth5: ADM5120 switch port5
ADM5120 board flash (0x400000 at 0x1fc00000)
ADM5120: Found 1 x16 devices at 0x0 in 16-bit bank
 Amd/Fujitsu Extended Query Table at 0x0040
number of CFI chips: 1
cfi_cmdset_0002: Disabling erase-suspend-program due to code brokenness.
Creating 1 MTD partitions on "ADM5120":
0x003e0000-0x00400000 : "p1"
NET: Registered protocol family 2
IP: routing cache hash table of 512 buckets, 4Kbytes
TCP established hash table entries: 1024 (order: 1, 8192 bytes)
TCP bind hash table entries: 1024 (order: 0, 4096 bytes)
TCP: Hash tables configured (established 1024 bind 1024)
IPv4 over IPv4 tunneling driver
ip_conntrack version 2.1 (128 buckets, 1024 max) - 248 bytes per conntrack
ip_tables: (C) 2000-2002 Netfilter core team
ipt_recent v0.3.1: Stephen Frost <sfrost@snowman.net>.  
http://snowman.net/projects/ipt_recent/
ClusterIP Version 0.6 loaded successfully
arp_tables: (C) 2002 David S. Miller
Initializing IPsec netlink socket
NET: Registered protocol family 1
NET: Registered protocol family 17
NET: Registered protocol family 15
Bridge firewalling registered
Ebtables v2.0 registered
802.1Q VLAN Support v1.8 Ben Greear <greearb@candelatech.com>
All bugs added by David S. Miller <davem@redhat.com>
Freeing unused kernel memory: 2101563k freed
Algorithmics/MIPS FPU Emulator v1.5
Mounting filesystems...

Please press Enter to activate this console.


While booted with pre-compiled image from 
http://sharon.esrac.ele.tue.nl/users/pe1rxq/linux-adm/vmlinuz-20050329.
The boot message looks like this (Still can not find the higher memory):

Linux version 2.6.12-rc1 (pe1rxq@laptop) (gcc version 3.4.2) #120 Mon 
Mar 28 01:10:40 CEST 2005
CPU revision is: 0001800b
ADM5120 board setup
System has PCI BIOS
Determined physical RAM map:
 memory: 00d1c000 @ 002e4000 (usable)
Built 1 zonelists
Kernel command line:
Primary instruction cache 8kB, physically tagged, 2-way, linesize 16 bytes.
Primary data cache 8kB, 2-way, linesize 16 bytes.
Synthesized TLB refill handler (19 instructions).
Synthesized TLB load handler fastpath (31 instructions).
Synthesized TLB store handler fastpath (31 instructions).
Synthesized TLB modify handler fastpath (30 instructions).
PID hash table entries: 128 (order: 7, 2048 bytes)
CPU clock: 175MHz
Dentry cache hash table entries: 4096 (order: 2, 16384 bytes)
Inode-cache hash table entries: 2048 (order: 1, 8192 bytes)
Memory: 13244k/13424k available (1908k kernel code, 160k reserved, 315k 
data, 628k init, 0k highmem)
Mount-cache hash table entries: 512
Checking for 'wait' instruction...  available.
NET: Registered protocol family 16
usbcore: registered new driver usbfs
usbcore: registered new driver hub
Initializing Cryptographic API
ADM5120 LED & GPIO driver
ttyS0 at I/O 0x12600000 (irq = 1) is a ADM5120
ttyS1 at I/O 0x12800000 (irq = 2) is a ADM5120
io scheduler noop registered
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
eth0: ADM5120 switch port0
eth1: ADM5120 switch port1
eth2: ADM5120 switch port2
eth3: ADM5120 switch port3
eth4: ADM5120 switch port4
eth5: ADM5120 switch port5
ADM5120 board flash (0x200000 at 0x1fc00000)
ADM5120: Found 1 x16 devices at 0x0 in 16-bit bank
 Amd/Fujitsu Extended Query Table at 0x0040
number of CFI chips: 1
cfi_cmdset_0002: Disabling erase-suspend-program due to code brokenness.
adm5120-hcd adm5120-hcd: new USB bus registered, assigned bus number 1
usb usb1: Product: adm5120-hcd
usb usb1: Manufacturer: Linux 2.6.12-rc1 adm5120-hcd
usb usb1: SerialNumber: adm5120-hcd
hub 1-0:1.0: USB hub found
hub 1-0:1.0: 2 ports detected
NET: Registered protocol family 2
IP: routing cache hash table of 512 buckets, 4Kbytes
TCP established hash table entries: 1024 (order: 1, 8192 bytes)
TCP bind hash table entries: 1024 (order: 0, 4096 bytes)
TCP: Hash tables configured (established 1024 bind 1024)
NET: Registered protocol family 17
Freeing unused kernel memory: 2099720k freed
Algorithmics/MIPS FPU Emulator v1.5
Relaxen und vatch das blinkenlights!!!
Mounting filesystems...

*********************************
 Its alive!!!!!!!!!!!!!!!!!!!!!!
*********************************

Re: [ADMtek 5120] 64M sdram on board but only 16M is deteted and usable

[Yoichi Yuasa]

Hi,

2005/9/1, Zhuang Yuyao <ihollo@tom.com>:
> Hi,
> 
> I've just upgraded the SDRAM on my adm5120 board from 16M to 64M, but
> while the board is booting, it still reports that only 16M sdram is
> availible.
> Since I do not have the source code of the bootloader, is there any way
> to let the board to boot with 64M sdram, or should I change to another
> bootloader?

If you are using add_memory_region() with fixed memory size,
you should change the value of it.

Yoichi

Re: [ADMtek 5120] 64M sdram on board but only 16M is deteted and usable

[Zhuang Yuyao]

Hi,

No, I am not using fixed memory size.

The original source code (part) in arch/mips/adm5120/memory.c is:
......
#define ADM5120_MEMCTRL                 0x1200001c
#define ADM5120_MEMCTRL_SDRAM_MASK      0x7

static const unsigned long adm_sdramsize[] __initdata = {
        0x0,            /* Reserved */
        0x0400000,      /* 4Mb */
        0x0800000,      /* 8Mb */
        0x1000000,      /* 16Mb */
        0x4000000,      /* 64Mb */
        0x8000000,      /* 128Mb */
};

void __init prom_meminit(void)
{
        unsigned long base=CPHYSADDR(PFN_ALIGN(&_end));
        unsigned long size;

        u32 memctrl = *(u32*)KSEG1ADDR(ADM5120_MEMCTRL);
        size = adm_sdramsize[memctrl & ADM5120_MEMCTRL_SDRAM_MASK];
         add_memory_region(base, size-base, BOOT_MEM_RAM);
}
......

memctrl value returns 0x50403 even though the actual memory size is 64M.

Then I tried to manually set KSEG1ADDR(ADM5120_MEMCTRL) to 0x50404 by 
inserting the following lines into prom_meminit(void).

        unsigned long size;

+      u32* pm = (u32*)KSEG1ADDR(ADM5120_MEMCTRL);
+      *pm = (*pm & ~ADM5120_MEMCTRL_SDRAM_MASK) + 0x4;

        u32 memctrl = *(u32*)KSEG1ADDR(ADM5120_MEMCTRL);

memctrl value returns 0x50404, the boot is failed but many messages 
printed to the console:

**********************************************************************
                            Linux Boot Loader


Copyright 2002-2003  ADMtek, Inc.



CPU: ADM5120 Home Gateway Processor
Boot Loader Version: 2.00.0125
Creation Date: 2003.07.09

Linux version 2.6.12 (root@debian) (gcc version 3.4.2) #36 Wed Aug 31 
22:52:06 MDT 2005
memctrl: 50404
CPU revision is: 0001800b
ADM5120 board setup
Determined physical RAM map:
 memory: 03b0b000 @ 004f5000 (usable)
Built 1 zonelists
Kernel command line:
Primary instruction cache 8kB, physically tagged, 2-way, linesize 16 bytes.
Primary data cache 8kB, 2-way, linesize 16 bytes.
Synthesized TLB refill handler (19 instructions).
Synthesized TLB load handler fastpath (31 instructions).
Synthesized TLB store handler fastpath (31 instructions).
Synthesized TLB modify handler fastpath (30 instructions).
PID hash table entries: 512 (order: 9, 8192 bytes)
CPU clock: 175MHz
Dentry cache hash table entries: 16384 (order: 4, 65536 bytes)
Inode-cache hash table entries: 8192 (order: 3, 32768 bytes)
Memory: 59776k/60460k available (2369k kernel code, 624k reserved, 375k 
data, 2004k init, 0k highmem)
Mount-cache hash table entries: 512
Data bus error, epc == 80108e50, ra == 8005c21c
Oops in arch/mips/kernel/traps.c::do_be, line 354[#1]:
Cpu 0
$ 0   : 00000000 10008c00 804fb410 8c8506e4
$ 4   : 804fb410 b31e0620 00000040 ffffffff
$ 8   : 00000002 00000000 00000000 804b0000
$12   : 80288214 00100100 00200200 804fc8f4
$16   : 804fb400 804fc960 804fc96c 8109eac0
$20   : 804b0000 00000000 000000d0 11110017
$24   : 00000000 80283de2                 
$28   : 80282000 80283e28 80ee1da0 8005c21c
Hi    : 00000000
Lo    : 00000ea0
epc   : 80108e50 both_aligned+0x10/0x64     Not tainted
ra    : 8005c21c cache_alloc_refill+0x254/0x690
Status: 10008c02    KERNEL EXL
Cause : 3080001c
PrId  : 0001800b
Modules linked in:
Process swapper (pid: 0, threadinfo=80282000, task=80284000)
Stack : 0000000d 804b0000 10008c01 000041ed 804fb000 800c2130 10008c01 
804f7ed4
        00000000 80255da0 804b0000 00000000 11110016 11110017 80ee1da0 
8005bc60
        804fb000 8009fffc 804fc86c 00000000 00000000 8009e67c 804fb000 
000041ed
        00000000 800a0814 00000000 804f7ed4 80000000 800c2130 00000000 
8009e924
        00000000 800849d4 8028994c 8025e89c 804f7ed4 804fb000 800c21a8 
800c2190
        ...
Call Trace:
 [<800c2130>] ramfs_fill_super+0x0/0xac
 [<8005bc60>] kmem_cache_alloc+0x94/0x9c
 [<8009fffc>] alloc_inode+0x11c/0x144
 [<8009e67c>] d_alloc+0x34/0x228
 [<800a0814>] new_inode+0x14/0x8c
 [<800c2130>] ramfs_fill_super+0x0/0xac
 [<8009e924>] d_alloc_root+0x30/0x68
 [<800849d4>] sget+0x324/0x3ac
 [<800c21a8>] ramfs_fill_super+0x78/0xac
 [<800c2190>] ramfs_fill_super+0x60/0xac
 [<800c2130>] ramfs_fill_super+0x0/0xac
 [<800859b0>] get_sb_nodev+0x84/0xdc
 [<800a31a8>] alloc_vfsmnt+0xa8/0xe8
 [<8005ddcc>] kmem_cache_create+0x770/0x7c4
 [<80085b54>] do_kern_mount+0x68/0x1b0
 [<802c58f0>] mnt_init+0xe4/0x280
 [<802c5848>] mnt_init+0x3c/0x280
 [<802c575c>] inode_init+0x4c/0xfc
 [<802c5538>] vfs_caches_init+0x100/0x1d0
 [<802c5528>] vfs_caches_init+0xf0/0x1d0
 [<800f88c0>] key_link+0x40/0x5c
 [<802b17c8>] start_kernel+0x1e8/0x260
 [<802b17c0>] start_kernel+0x1e0/0x260
 [<802b112c>] unknown_bootoption+0x0/0x310


Code: 11000017  30d8001f  00000000 <8ca80000> 8ca90004  8caa0008  
8cab000c  24c6ffe0  8cac0010
Kernel panic - not syncing: Attempted to kill the idle task!

******************************************************************

Any suggestions will be appreciated!

Yoichi Yuasa wrote:

>Hi,
>
>2005/9/1, Zhuang Yuyao <ihollo@tom.com>:
>  
>
>>Hi,
>>
>>I've just upgraded the SDRAM on my adm5120 board from 16M to 64M, but
>>while the board is booting, it still reports that only 16M sdram is
>>availible.
>>Since I do not have the source code of the bootloader, is there any way
>>to let the board to boot with 64M sdram, or should I change to another
>>bootloader?
>>    
>>
>
>If you are using add_memory_region() with fixed memory size,
>you should change the value of it.
>
>Yoichi
>
>
>  
>

Re: [ADMtek 5120] 64M sdram on board but only 16M is deteted and usable

[Vladislav Moskovets]

Hi, we have successful experience with bootloader patching, we have
working 32M
If you want we can publish our work

> I've just upgraded the SDRAM on my adm5120 board from 16M to 64M, but
> while the board is booting, it still reports that only 16M sdram is
> availible.
> Since I do not have the source code of the bootloader, is there any
> way
> to let the board to boot with 64M sdram, or should I change to another
> bootloader?


-- 
Regards, Vladislav Moskovets