| --- Log opened Tue Dec 24 00:00:22 2013 | ||
| maxpaln | stekern: looks like you're onto the right | 08:34 |
|---|---|---|
| maxpaln | track | 08:34 |
| maxpaln | It wasn't straight forward to disable bursts in the DDR controller so I disabled the cache and it now works | 08:34 |
| maxpaln | :-) | 08:34 |
| maxpaln | however, at the console output I am getting these messages which indicate something is wrong at the Ethernet Rx side | 08:35 |
| maxpaln | net eth0: RX: wrong CRC | 08:36 |
| maxpaln | net eth0: RX: frame too short | 08:36 |
| maxpaln | net eth0: RX: dribble nibble | 08:36 |
| maxpaln | hmmm, reading around on the web - not sure if this is a Linux thing or a HW thing. Any thoughs? | 08:44 |
| maxpaln | thoughts? | 08:45 |
| stekern | hmm, haven't seen that before | 09:30 |
| maxpaln | A ping yields no Rx packets | 09:30 |
| maxpaln | and it's difficult to say whether and are being Tx | 09:31 |
| maxpaln | so | 09:31 |
| maxpaln | I think this must be a GMII vs MII thing | 09:31 |
| stekern | but it *has* to be straight forward to disable bursts in the controller, since that's essentially what you did when you disabled bursts in the master | 09:31 |
| maxpaln | I would agree - but the controller hasn't been written in a way that I can find an easy method to disable the burst | 09:32 |
| maxpaln | it's a kinda crappy implementation if I am being honest | 09:32 |
| maxpaln | it assumes an incremental burst - | 09:32 |
| maxpaln | and waits for the controller to deassert STD to stop the burst | 09:32 |
| maxpaln | STB | 09:32 |
| stekern | hmm, I see | 09:32 |
| stekern | that's not really wb compliant behaviour | 09:33 |
| maxpaln | I think a rewrite of the controller (with the Xilinx DDR2 controller as a template) is probably the best method | 09:33 |
| maxpaln | yeah - I would agree with you | 09:33 |
| maxpaln | but, back on the ethernet | 09:33 |
| maxpaln | the board has a GMII phy | 09:33 |
| maxpaln | I think the problem must be to do with how the MAC is setting up the PHY | 09:34 |
| maxpaln | The router is showing the connection as a Gbps link | 09:34 |
| maxpaln | which is almost certainly the problem | 09:34 |
| maxpaln | the MII ethmac only uses four of the eight data bits | 09:35 |
| maxpaln | for Tx and Rx | 09:35 |
| maxpaln | I need to get the MAC to put the PHy into 10/100 mode | 09:35 |
| maxpaln | probably need to do a bit of research there | 09:35 |
| maxpaln | although, interestingly ifconfig shows a valid IP address | 09:42 |
| maxpaln | # ifconfig | 09:42 |
| maxpaln | eth0 Link encap:Ethernet HWaddr 16:D3:2E:BA:D6:BC | 09:42 |
| maxpaln | inet addr:192.168.1.100 Bcast:192.168.1.255 Mask:255.255.255.0 | 09:42 |
| maxpaln | UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 | 09:42 |
| maxpaln | RX packets:0 errors:0 dropped:345 overruns:0 frame:760 | 09:42 |
| maxpaln | TX packets:1758 errors:1 dropped:0 overruns:0 carrier:0 | 09:42 |
| maxpaln | collisions:0 txqueuelen:1000 | 09:42 |
| maxpaln | RX bytes:0 (0.0 B) TX bytes:73836 (72.1 KiB) | 09:42 |
| maxpaln | Interrupt:4 Memory:1200000-1220000 | 09:42 |
| maxpaln | lo Link encap:Local Loopback | 09:42 |
| maxpaln | inet addr:127.0.0.1 Mask:255.0.0.0 | 09:42 |
| maxpaln | UP LOOPBACK RUNNING MTU:65536 Metric:1 | 09:42 |
| maxpaln | although a LOT of dropped Rx kets | 09:43 |
| maxpaln | packets | 09:43 |
| stekern | isn't that a static ip though? | 09:49 |
| stekern | but, yes, if the phy is set to autosense and the router too, then it'll probably autosense to gigabit | 09:50 |
| stekern | what phy is it? | 09:50 |
| maxpaln | its a marvell PHY | 09:54 |
| maxpaln | 88E1119R | 09:54 |
| maxpaln | ah, you might be right about the static IP | 09:54 |
| maxpaln | it fits in well with my network | 09:54 |
| maxpaln | BUT | 09:54 |
| maxpaln | the Bcast address is wrontg | 09:54 |
| maxpaln | the Bcast address is wrong | 09:54 |
| maxpaln | oh, hang on | 09:55 |
| maxpaln | nope it's ok | 09:55 |
| stekern | ok, that's the same that's on the atlys board | 09:55 |
| maxpaln | ah, ok | 09:56 |
| stekern | but I only have 100Mbit switches | 09:56 |
| maxpaln | well, its definitely a static IP - I disconnected from the router and the IP adress remains | 09:57 |
| maxpaln | well, I've asked our ethernet team to let me know how the PHY should be configured to force it to 10/100 mode | 09:58 |
| maxpaln | I don't have the datasheet here | 09:59 |
| stekern | iirc, it's in the "standard" regs | 10:08 |
| stekern | but I might remember wrong | 10:08 |
| stekern | yes, it's in reg 0: basic control | 10:09 |
| maxpaln | where are you looking? | 10:11 |
| stekern | in any random phy datasheet | 10:12 |
| stekern | http://www.ti.com/lit/ds/symlink/dp83848c.pdf | 10:12 |
| maxpaln | ah, ok | 10:12 |
| maxpaln | :-) | 10:12 |
| stekern | for instance | 10:12 |
| maxpaln | SO I assume this needs to be implemented in the Linux build somewhere - perhaps in one of the drivers | 10:13 |
| stekern | but I'm not sure how that is handled on gbit phys, never poked around those in-depth | 10:13 |
| maxpaln | or perhaps even while Linux is running - should be possible to reconfigure the PHY I would have thought | 10:16 |
| stekern | there's ethtool and mii-tool to access mdio under linux | 10:16 |
| _franck_web_ | same thing on a 88E1512 gigabit phy | 10:16 |
| _franck_web_ | reg 0 | 10:16 |
| stekern | but you can probably hack something into the ethmac driver too | 10:17 |
| maxpaln | in the driver would be better long term | 10:23 |
| maxpaln | but it would be good to be able to modify in the running Linux to see what settings are needed | 10:23 |
| maxpaln | hmmm, mii-tool and ethtool are not i the vmlinux binary | 10:25 |
| maxpaln | I'd download it and install it but .... | 10:25 |
| maxpaln | ok, so I think my lack of Linux admin knowledge is exposing me here [blushes] | 10:27 |
| maxpaln | I'm running out of time today so I think I will have to return to this after the festive period | 10:28 |
| maxpaln | perhaps I'll recruit a Linux expert from our apps team :-) | 10:28 |
| maxpaln | thank you for all your help so far - | 10:28 |
| maxpaln | you have been very helpful and definitely allowing me to progress a lot quicker than I could have otherwise | 10:29 |
| maxpaln | I'm wondering if drivers for the Marvell phy have actually been included | 10:32 |
| maxpaln | I can't see one in the drivers/net/ethernet folder | 10:33 |
| stekern | well, you don't need any special phy drivers to change the regs | 10:34 |
| maxpaln | and if I run 'make menuconfig' to run the Linux configuration GUI thing (not really sure what to call it - it's kinda like a GUI) I can see that Marvell PHY drivers are not set for inclusion - just Micrel PHYs | 10:34 |
| maxpaln | right - time to dash | 10:37 |
| maxpaln | I'll pick this up again in a few days | 10:37 |
| stekern | see you | 10:38 |
| -!- Netsplit *.net <-> *.split quits: Amadiro, jeremybennett, FreezingCold | 14:06 | |
| -!- Netsplit over, joins: FreezingCold, Amadiro | 14:11 | |
| --- Log closed Wed Dec 25 00:00:24 2013 | ||
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!