Hello All,
I am trying to port the networking portion of linux to work with our
board. Our board uses a MPC850 and SCC2 as its ethernet interface ( almost
identical to the rpxlite ).
Port A, bit 13 - SCC2 Ethernet Rx
Port A, bit 12 - SCC2 Ethernet Tx
Port A, bit 6 (CLK2) - Ethernet Tx Clk
Port A, bit 4 (CLK4) - Ethernet Rx Clk
Port B, bit 18 (RTS2) - Ethernet Tx Enable
*Port B, bit 17 - Ethernet Loopback Enable
*Port B, bit 16 - Ethernet Full Duplex Disable
Port C, bit 8 (CD2) - Ethernet Rx Enable
Port C, bit 9 (CTS2) - SCC Ethernet Collision
* Differs from rpxlite
I am using ppcboot for our boot rom and compile the kernel targeting
a MBX board. I have modified the mbx.h file with a customized board info
structure to work with ppcboot. I have also stuck a #define ESTEEM192E so I
could stick my mods within the #ifdef CONFIG_MBX contained in the kernel as
needed. The only mods I think I need to make are in commproc.h and enet.c. I
have made them as follows...
from /arch/ppc/m8xx_io/commproc.h
Code was this ........
#ifdef CONFIG_MBX
/* Bits in parallel I/O port registers that have to be set/cleared
* to configure the pins for SCC1 use. The TCLK and RCLK seem unique
* to the MBX860 board. Any two of the four available clocks could be
* used, and the MPC860 cookbook manual has an example using different
* clock pins.
*/
#define PA_ENET_RXD ((ushort)0x0001)
#define PA_ENET_TXD ((ushort)0x0002)
#define PA_ENET_TCLK ((ushort)0x0200)
#define PA_ENET_RCLK ((ushort)0x0800)
#define PC_ENET_TENA ((ushort)0x0001)
#define PC_ENET_CLSN ((ushort)0x0010)
#define PC_ENET_RENA ((ushort)0x0020)
/* Control bits in the SICR to route TCLK (CLK2) and RCLK (CLK4) to
* SCC1. Also, make sure GR1 (bit 24) and SC1 (bit 25) are zero.
*/
#define SICR_ENET_MASK ((uint)0x000000ff)
#define SICR_ENET_CLKRT ((uint)0x0000003d)
#endif
Code is now this ........
#ifdef CONFIG_MBX
#ifndef ESTEEM192E
/* Bits in parallel I/O port registers that have to be set/cleared
* to configure the pins for SCC1 use. The TCLK and RCLK seem unique
* to the MBX860 board. Any two of the four available clocks could be
* used, and the MPC860 cookbook manual has an example using different
* clock pins.
*/
#define PA_ENET_RXD ((ushort)0x0001)
#define PA_ENET_TXD ((ushort)0x0002)
#define PA_ENET_TCLK ((ushort)0x0200)
#define PA_ENET_RCLK ((ushort)0x0800)
#define PC_ENET_TENA ((ushort)0x0001)
#define PC_ENET_CLSN ((ushort)0x0010)
#define PC_ENET_RENA ((ushort)0x0020)
/* Control bits in the SICR to route TCLK (CLK2) and RCLK (CLK4) to
* SCC1. Also, make sure GR1 (bit 24) and SC1 (bit 25) are zero.
*/
#define SICR_ENET_MASK ((uint)0x000000ff)
#define SICR_ENET_CLKRT ((uint)0x0000003d)
#else
/* ESTEEM192E
* This ENET stuff is for the MPC850 with ethernet on SCC2. This
* is very similar to the RPX-Lite configuration.
* Note TENA , LOOPBACK , FDPLEX_DIS on Port B.
*/
#define PA_ENET_RXD ((ushort)0x0004)
#define PA_ENET_TXD ((ushort)0x0008)
#define PA_ENET_TCLK ((ushort)0x0200)
#define PA_ENET_RCLK ((ushort)0x0800)
#define PB_ENET_TENA ((uint)0x00002000)
#define PC_ENET_CLSN ((ushort)0x0040)
#define PC_ENET_RENA ((ushort)0x0080)
#define SICR_ENET_MASK ((uint)0x0000ff00)
#define SICR_ENET_CLKRT ((uint)0x00003d00)
#define PB_ENET_LOOPBACK ((uint)0x00004000)
#define PB_ENET_FDPLEX_DIS ((uint)0x00008000)
#endif
from /arch/ppc/m8xx_io/enet.c in function cpm_enet_init
Code was this ........
#ifdef CONFIG_MBX
immap->im_ioport.iop_pcpar |= PC_ENET_TENA;
immap->im_ioport.iop_pcdir &= ~PC_ENET_TENA;
#endif
Code is now this ........
#ifdef CONFIG_MBX
#ifndef ESTEEM192E
/* MBX stuff */
immap->im_ioport.iop_pcpar |= PC_ENET_TENA;
immap->im_ioport.iop_pcdir &= ~PC_ENET_TENA;
#else
/* ESTEEM192E stuff */
cp->cp_pbpar |= PB_ENET_TENA; /* same as rpxlite */
cp->cp_pbdir |= PB_ENET_TENA; /* same as rpxlite */
/*** Make sure that the loopback and full duplex are disabled
just like the rpxlite code. ***/
cp->cp_pbpar &= ~(PB_ENET_LOOPBACK | PB_ENET_FDPLEX_DIS);
cp->cp_pbdir |= PB_ENET_LOOPBACK | PB_ENET_FDPLEX_DIS;
cp->cp_pbdat &= ~PB_ENET_LOOPBACK;
cp->cp_pbdat |= PB_ENET_FDPLEX_DIS;
#endif
#endif
This is what it looks like when booting.
Initializing...
CPU: XPC850xxZTB at 49 MHz: 2 kB I-Cache 1 kB D-Cache
Board: Esteem 192E
DRAM:
bank 0 size 8388608
bank 1 size 8388608
16 MB
Command "go": 0x4000dd28 => 0x00fedd28
Command "bootm": 0x4000e8dc => 0x00fee8dc
Command "net": 0x4000f3d4 => 0x00fef3d4
Command "loads": 0x4000ddd8 => 0x00feddd8
Command "md": 0x4000cf74 => 0x00fecf74
Command "mm": 0x4000d1f8 => 0x00fed1f8
Command "nm": 0x4000d228 => 0x00fed228
Command "mw": 0x4000d25c => 0x00fed25c
Command "cp": 0x4000d3e8 => 0x00fed3e8
Command "crc32": 0x4000dabc => 0x00fedabc
Command "base": 0x4000d610 => 0x00fed610
Command "printenv": 0x4000f56c => 0x00fef56c
Command "setenv": 0x4000f6e4 => 0x00fef6e4
Command "saveenv": 0x4000f998 => 0x00fef998
Command "bdinfo": 0x4000db70 => 0x00fedb70
Command "flinfo": 0x4000e0cc => 0x00fee0cc
Command "iminfo": 0x4000ee4c => 0x00feee4c
Command "erase": 0x4000e1a4 => 0x00fee1a4
Command "protect": 0x4000e4dc => 0x00fee4dc
Command "loop": 0x4000d674 => 0x00fed674
Command "mtest": 0x4000d820 => 0x00fed820
Command "icache": 0x4000cdb8 => 0x00fecdb8
Command "dcache": 0x4000ce5c => 0x00fece5c
Command "reset": 0x40013528 => 0x00ff3528
Command "version": 0x4000cc34 => 0x00fecc34
Command "help": 0x4000cc80 => 0x00fecc80
Command "?": 0x4000cc80 => 0x00fecc80
FLASH: 4 MB
Hit any key to stop autoboot: 0
## Booting Linux kernel at 40020000 ...
Image Name: 2.2.14 for EST192E
Image Type: PowerPC Linux Kernel Image (gzip compressed)
Data Size: 413162 Bytes = 403 kB = 0 MB
Load Address: 00000000
Entry Point: 00000000
Verifying Checksum ... OK
Uncompressing Kernel Image ... OK
Linux version 2.2.14 (root at pokey.esteem.com) (gcc version 2.95.2 19991030
(2.9
5.3 prerelease/franzo)) #14 Thu Oct 5 13:24:22 PDT 2000
Boot arguments: root=/dev/nfs rw nfsaddrs=100.100.100.21:100.100.100.14 nfsroo
t=100.100.100.14:/targ
time_init: decrementer frequency = 183750000/60
Calibrating delay loop... 48.74 BogoMIPS
Memory: 15120k available (748k kernel code, 476k data, 40k init) [c0000000,c10
00000]
Dentry hash table entries: 2048 (order 2, 16k)
Buffer cache hash table entries: 16384 (order 4, 64k)
Page cache hash table entries: 4096 (order 2, 16k)
POSIX conformance testing by UNIFIX
Linux NET4.0 for Linux 2.2
Based upon Swansea University Computer Society NET3.039
NET4: Unix domain sockets 1.0 for Linux NET4.0.
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP, IGMP
TCP: Hash tables configured (ehash 16384 bhash 16384)
Starting kswapd v 1.5
CPM UART driver version 0.03
ttyS00 at 0x0280 is a SMC
ttyS01 at 0x0380 is a SMC
pty: 256 Unix98 ptys configured
RAM disk driver initialized: 16 RAM disks of 4096K size
loop: registered device at major 7
NET4: Ethernet Bridge 007 for NET4.0
early initialization of device brg0 is deferred
eth0: CPM ENET Version 0.2, 00:10:ec:00:1b:ba
PPP: version 2.3.7 (demand dialling)
TCP compression code copyright 1989 Regents of the University of California
PPP line discipline registered.
brg0: network interface for Ethernet Bridge 007/NET4.0
brg0: generated MAC address FE:FD:00:30:24:76
brg0: attached to bridge instance 0
Sending BOOTP and RARP requests............. timed out!
IP-Config: Auto-configuration of network failed.
dev_close: bug eth0 still running
Looking up port of RPC 100003/2 on 100.100.100.14
RPC: sendmsg returned error 101
RPC: sendmsg returned error 101
RPC: sendmsg returned error 101
RPC: sendmsg returned error 101
portmap: server 100.100.100.14 not responding, timed out
Root-NFS: Unable to get nfsd port number from server, using default
Looking up port of RPC 100005/1 on 100.100.100.14
RPC: sendmsg returned error 101
RPC: sendmsg returned error 101
RPC: sendmsg returned error 101
RPC: sendmsg returned error 101
Blah..Blah..Blah..Blah..Blah..Blah..... reboot in 180 seconds
I know that my BOOTP and NFS server work because I have tested it
with another machine. I don't know if the BOOTP requests are making it to
the server okay.
At least some of the code works because I can see the TENA go high
and bits being sent (may not be the correct bits ;-) on the transmit line
going to the PHY(I can also see a signal going to the jack from the PHY).
When a packet (any ethernet activity from the net) comes in through the PHY
I can see the RENA line go high and bits coming in on the receive line.
I noticed that the rpxlite has a colision test disable line in its
BSCR. Is this needed to get the transmit to work right?
Note: I have not ruled out the PHY to Jack interface as the problem.
Any help appreciated. Many MAny thanks in advance.
Conn Clark
** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
