PPC440GX IBM EMAC2/3 RGMII mode selection

PPC440GX IBM EMAC2/3 RGMII mode selection

[David Grab]

Hello,

i have some problems to configure emac2, emac3 support of my ibm ppc440gx
for linux kernel 2.6.11.6. I have two RGMII Phy´s connected to the ppc an no
other PHY´s are connected. PPC Strapping is selected for Group4 (SMII, SMII,
RGMII/RTBI, RGMII/RTBI). I only want to configure the RGMII support of emac2
and emac3. But it seems like the PHY´s are used for eth0 (emac1) and eth1
(emac2) which are configured as SMII. Could someone tell me, what i have to
configure to select only emac2 and emac3 in RGMII mode?

Thx for help,

David

console output:

=> bootd
## Booting image at ff000000 ...
   Image Name:   Linux-2.6.11.6
   Image Type:   PowerPC Linux Kernel Image (gzip compressed)
   Data Size:    890881 Bytes = 870 kB
   Load Address: 00000000
   Entry Point:  00000000
   Verifying Checksum ... OK
   Uncompressing Kernel Image ... OK
## Current stack ends at 0x1FF9D898 => set upper limit to 0x00800000
## cmdline at 0x007FFF00 ... 0x007FFF25
bd address  = 0x1FF9DF54
memstart    = 0x00000000
memsize     = 0x20000000
flashstart  = 0xFF000000
flashsize   = 0x01000000
flashoffset = 0x00000000
sramstart   = 0x00000000
sramsize    = 0x00000000
bootflags   = 0xC01D2D00
intfreq     =    500 MHz
busfreq     = 166.666 MHz
ethaddr     = 00:04:AC:E3:28:8A
eth1addr    = 00:04:AC:E3:28:8B
eth2addr    = 00:04:AC:E3:28:8C
eth3addr    = 00:04:AC:E3:28:8D
IP addr     = 192.168.0.116
baudrate    = 115200 bps
## Loading RAMDisk Image at ff800000 ...
   Image Name:   Test Ramdisk
   Image Type:   PowerPC Linux RAMDisk Image (gzip compressed)
   Data Size:    1629365 Bytes =  1.6 MB
   Load Address: 00000000
   Entry Point:  00000000
   Verifying Checksum ... OK
## initrd at 0xFF800040 ... 0xFF98DCF4 (len=1629365=0x18DCB5)
   Loading Ramdisk to 1fe0f000, end 1ff9ccb5 ... OK
Linux version 2.6.11.6 (david@FedoraDG) (gcc-Version 3.3.3 (DENX ELDK 3.1
3.3.3-8)) #2 Thu Jul 7 01:33:11 CEST 2005
IBM
Built 1 zonelists
Kernel command line: console=ttyS1,115200 root=/dev/ram rw
PID hash table entries: 4096 (order: 12, 65536 bytes)
Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
Memory: 515712k available (1372k kernel code, 464k data, 108k init, 0k
highmem)
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
checking if image is initramfs...it isn't (no cpio magic); looks like an
initrd
Freeing initrd memory: 1591k freed
NET: Registered protocol family 16
PCI: Probing PCI hardware
PCI: bridge rsrc 0..ffff (100), parent c0195018
PCI: bridge rsrc 80000000..ffffefff (200), parent c0195034
SCSI subsystem initialized
Serial: 8250/16550 driver $Revision: 1.90 $ 6 ports, IRQ sharing enabled
ttyS0 at MMIO 0x0 (irq = 0) is a 16550A
ttyS1 at MMIO 0x0 (irq = 1) is a 16550A
io scheduler noop registered
io scheduler anticipatory registered
io scheduler deadline registered
io scheduler cfq registered
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
loop: loaded (max 8 devices)
mal0: Initialized, 4 tx channels, 4 rx channels
emac: IBM EMAC Ethernet driver, version 2.0
Maintained by Benjamin Herrenschmidt <benh@kernel.crashing.org>
zmii0: input 0 in SMII mode
eth0: IBM emac, MAC 00:04:ac:e3:28:8a
eth0: Found Generic MII PHY (0x05)
zmii0: input 1 in SMII mode
eth1: IBM emac, MAC 00:04:ac:e3:28:8b
eth1: Found Generic MII PHY (0x06)
zmii0: input 2 in SMII mode
rgmii0: input 0 in RGMII mode
emac2: Can't find PHY.
zmii0: input 3 in SMII mode
rgmii0: input 1 in RGMII mode
emac3: Can't find PHY.
NET: Registered protocol family 2
IP: routing cache hash table of 4096 buckets, 32Kbytes
TCP established hash table entries: 131072 (order: 8, 1048576 bytes)
TCP bind hash table entries: 65536 (order: 6, 262144 bytes)
TCP: Hash tables configured (established 131072 bind 65536)
NET: Registered protocol family 1
NET: Registered protocol family 17
RAMDISK: Compressed image found at block 0
RAMDISK: incomplete write (-28 != 1024) 4194304
VFS: Mounted root (ext2 filesystem).
Freeing unused kernel memory: 108k init
eth0: Link is Up
eth0: Cannot reset EMAC
### Application running ...

BusyBox v0.60.5 (2004.11.10-15:02+0000) Built-in shell (msh)
Enter 'help' for a list of built-in commands.

Re: PPC440GX IBM EMAC2/3 RGMII mode selection

[Eugene Surovegin]

On Thu, Jul 07, 2005 at 04:28:57PM +0200, David Grab wrote:
> i have some problems to configure emac2, emac3 support of my ibm ppc440gx
> for linux kernel 2.6.11.6. I have two RGMII Phy?s connected to the ppc an no
> other PHY?s are connected. PPC Strapping is selected for Group4 (SMII, SMII,
> RGMII/RTBI, RGMII/RTBI). I only want to configure the RGMII support of emac2
> and emac3. But it seems like the PHY?s are used for eth0 (emac1) and eth1
> (emac2) which are configured as SMII. Could someone tell me, what i have to
> configure to select only emac2 and emac3 in RGMII mode?

Quick hack would be commenting out EMAC0 and EMAC1 ocp_def in 
arch/ppc/platforms/4xx/ibm440gx.c.

Let me know if it works, and I'll think about more clean way to do 
this.

-- 
Eugene

Re: PPC440GX IBM EMAC2/3 RGMII mode selection

[Eugene Surovegin]

On Thu, Jul 07, 2005 at 11:36:55AM -0400, Scott Coulter wrote:
> What about:
> 
> ocp_remove_one_device(OCP_VENDOR_IBM, OCP_FUNC_EMAC, 0);
> ocp_remove_one_device(OCP_VENDOR_IBM, OCP_FUNC_EMAC, 1);
> 
> dropped in xxx_setup_arch()?
> 

Of course, too early in the morning for me, I guess. I forgot I asked 
Matt to add this function :)

-- 
Eugene

AW: PPC440GX IBM EMAC2/3 RGMII mode selection

[David Grab]

I solved the PHY assertion with my board specific function. Thanks to Travis
B. Sawyer for this hint.

		if (i < 2) {
			emacdata->phy_map = 0xffffffff;
			emacdata->phy_mode = PHY_MODE_NA;
		}
		else {
			emacdata->phy_map = 0x00000001;
			emacdata->phy_mode = PHY_MODE_RGMII;
		}

But now i have a new problem. As you can see my PHY is detected and working
properly. It respond on link down and link up by (un)connecting ethernet
cable. Also i get RX Stuff from ethernet. But if i ping after setting
network, then it results in No response and even i can´t measure an TX
Enable or TX Clock signal from PPC440 to my PHY. I debugged a litte bit and
i think it´s becaus my ethernet is set up with zmii0: input 0 in SMII mode
and rgmii0: input 0 in RGMII mode. Is it possible that eth0 uses SMII mode
instead of RGMII so that no TX CLK / Enable is generated from PPC?

bootd
## Booting image at ff000000 ...
   Image Name:   Linux-2.6.11.6
   Image Type:   PowerPC Linux Kernel Image (gzip compressed)
   Data Size:    891020 Bytes = 870.1 kB
   Load Address: 00000000
   Entry Point:  00000000
   Verifying Checksum ... OK
   Uncompressing Kernel Image ... OK
## Current stack ends at 0x1FF9D898 => set upper limit to 0x00800000
## cmdline at 0x007FFF00 ... 0x007FFF25
bd address  = 0x1FF9DF54
memstart    = 0x00000000
memsize     = 0x20000000
flashstart  = 0xFF000000
flashsize   = 0x01000000
flashoffset = 0x00000000
sramstart   = 0x00000000
sramsize    = 0x00000000
bootflags   = 0xC01D2260
intfreq     =    500 MHz
busfreq     = 166.666 MHz
ethaddr     = 00:04:AC:E3:28:8A
eth1addr    = 00:04:AC:E3:28:8B
eth2addr    = 00:04:AC:E3:28:8C
eth3addr    = 00:04:AC:E3:28:8D
IP addr     = 192.168.0.116
baudrate    = 115200 bps
## Loading RAMDisk Image at ff800000 ...
   Image Name:   Test Ramdisk
   Image Type:   PowerPC Linux RAMDisk Image (gzip compressed)
   Data Size:    1629365 Bytes =  1.6 MB
   Load Address: 00000000
   Entry Point:  00000000
   Verifying Checksum ... OK
## initrd at 0xFF800040 ... 0xFF98DCF4 (len=1629365=0x18DCB5)
   Loading Ramdisk to 1fe0f000, end 1ff9ccb5 ... OK
## Transferring control to Linux (at address 00000000) ...
## Transferring control to Linux (at address 00000000) ...
Adresse: 7ffe50
Linux version 2.6.11.6 (david@FedoraDG) (gcc-Version 3.3.3 (DENX ELDK 3.1
3.3.3-8)) #12 Fri Jul 8 00:54:21 CEST 2005

IBM
Built 1 zonelists
Kernel command line: console=ttyS1,115200 root=/dev/ram rw
PID hash table entries: 4096 (order: 12, 65536 bytes)
Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
Memory: 515712k available (1372k kernel code, 464k data, 108k init, 0k
highmem)
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
checking if image is initramfs...it isn't (no cpio magic); looks like an
initrd
Freeing initrd memory: 1591k freed
NET: Registered protocol family 16
PCI: Probing PCI hardware
PCI: bridge rsrc 0..ffff (100), parent c0195018
PCI: bridge rsrc 80000000..ffffefff (200), parent c0195034
SCSI subsystem initialized
Serial: 8250/16550 driver $Revision: 1.90 $ 6 ports, IRQ sharing enabled
ttyS0 at MMIO 0x0 (irq = 0) is a 16550A
ttyS1 at MMIO 0x0 (irq = 1) is a 16550A
io scheduler noop registered
io scheduler anticipatory registered
io scheduler deadline registered
io scheduler cfq registered
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
loop: loaded (max 8 devices)
mal0: Initialized, 4 tx channels, 4 rx channels
emac: IBM EMAC Ethernet driver, version 2.0
Maintained by Benjamin Herrenschmidt <benh@kernel.crashing.org>
zmii0: input 0 in SMII mode
emac0: Can't find PHY.
zmii0: input 1 in SMII mode
emac1: Can't find PHY.
zmii0: input 2 in SMII mode
rgmii0: input 0 in RGMII mode
eth0: IBM emac, MAC 00:04:ac:e3:28:8c
eth0: Found Generic MII PHY (0x05)
zmii0: input 3 in SMII mode
rgmii0: input 1 in RGMII mode
eth1: IBM emac, MAC 00:04:ac:e3:28:8d
eth1: Found Generic MII PHY (0x06)
NET: Registered protocol family 2
IP: routing cache hash table of 4096 buckets, 32Kbytes
TCP established hash table entries: 131072 (order: 8, 1048576 bytes)
TCP bind hash table entries: 65536 (order: 6, 262144 bytes)
TCP: Hash tables configured (established 131072 bind 65536)
NET: Registered protocol family 1
NET: Registered protocol family 17
RAMDISK: Compressed image found at block 0
RAMDISK: incomplete write (-28 != 1024) 4194304
VFS: Mounted root (ext2 filesystem).
Freeing unused kernel memory: 108k init
eth0: Link is Up
eth0: Speed: 100, Full duplex.
### Application running ...


BusyBox v0.60.5 (2004.11.10-15:02+0000) Built-in shell (msh)
Enter 'help' for a list of built-in commands.

# ifconfig eth0 192.168.0.116 netmask 255.255.252.0
# eth0: Link is Up
eth0: Speed: 100, Full duplex.

# ifconfig
eth0      Link encap:Ethernet  HWaddr 00:04:AC:E3:28:8C
          inet addr:192.168.0.116  Bcast:192.168.0.255  Mask:255.255.252.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:134 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:9225 (9.0 kiB)  TX bytes:0 (0.0 iB)
          Interrupt:64

lo        Link encap:Local Loopback
          inet addr:127.0.0.1  Mask:255.0.0.0
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 iB)  TX bytes:0 (0.0 iB)

# ping 192.168.121
emac_start_xmit()
exitn
emac_start_xmit()
exitn
emac_start_xmit()
exitn
No response from 192.168.0.121
#

Re: PPC440GX IBM EMAC2/3 RGMII mode selection

[Eugene Surovegin]

On Fri, Jul 08, 2005 at 01:14:54PM +0200, David Grab wrote:
> I solved the PHY assertion with my board specific function. Thanks to Travis
> B. Sawyer for this hint.
> 
> 		if (i < 2) {
> 			emacdata->phy_map = 0xffffffff;
> 			emacdata->phy_mode = PHY_MODE_NA;
> 		}
> 		else {
> 			emacdata->phy_map = 0x00000001;
> 			emacdata->phy_mode = PHY_MODE_RGMII;
> 		}
> 
> But now i have a new problem. As you can see my PHY is detected and working
> properly. It respond on link down and link up by (un)connecting ethernet
> cable. Also i get RX Stuff from ethernet. But if i ping after setting
> network, then it results in No response and even i can?t measure an TX
> Enable or TX Clock signal from PPC440 to my PHY. I debugged a litte bit and
> i think it?s becaus my ethernet is set up with zmii0: input 0 in SMII mode
> and rgmii0: input 0 in RGMII mode. Is it possible that eth0 uses SMII mode
> instead of RGMII so that no TX CLK / Enable is generated from PPC?

[snip]

> mal0: Initialized, 4 tx channels, 4 rx channels
> emac: IBM EMAC Ethernet driver, version 2.0
> Maintained by Benjamin Herrenschmidt <benh@kernel.crashing.org>
> zmii0: input 0 in SMII mode
> emac0: Can't find PHY.
> zmii0: input 1 in SMII mode
> emac1: Can't find PHY.
> zmii0: input 2 in SMII mode
> rgmii0: input 0 in RGMII mode
> eth0: IBM emac, MAC 00:04:ac:e3:28:8c
> eth0: Found Generic MII PHY (0x05)
> zmii0: input 3 in SMII mode
> rgmii0: input 1 in RGMII mode
> eth1: IBM emac, MAC 00:04:ac:e3:28:8d
> eth1: Found Generic MII PHY (0x06)

Are you using gigabit connection or ordinary fast ethernet (100 Mbps)?
If gige, then you have to add your gig PHY definition. Generic MII PHY 
in the current driver doesn't support GigE (my version supports it).

Another question, is ethernet on your board working in U-Boot?

There is alternative EMAC driver I wrote, which fixed numerous 
problems with current driver version (http://kernel.ebshome.net). You 
can try it and I can help you getting it working on your board, 
unfortunately I cannot help you much with the current driver, because 
it's known to be buggy and IMHO fighting it is a waste of time.

-- 
Eugene