April 2002 - ixp1200 - lists.cs.princeton.edu

IXP1200 standalone
by Charles Chiou 03 May '02

03 May '02

Hi, I have been trying the VERA package for IXP1200. I have a couple of questions and would really appreciate it if you could respond to them. I followed the instructions and got everything to build, changed jumper settings, update firmware, etc. The IPmm demo program works when IXP1200 is plugged into a PC PCI in passive backplane. Now, I am trying to use the libraries to compile a standalone application instead of using a linux host to download. I compile using a big-endian arm cross-compiler. I am getting some errors in Libcii (fmt.c), as well as ethernet port initialization, and systimer and a few other initialization routines. Has anybody successfully compiled IPmm or others to run as a standalone (with cygmon as boot loader) in big-endian mode? Are there any firmware-specific or big/little-endian portion of the libraries that needs to be ported? Thanks so much! Regards, Charles

2 3

unsubmit
by yh143＠mail1.sjtu.edu.cn 22 Apr '02

22 Apr '02

1 0

RE: [ixp1200] Booting Intel ixp boards from tftp
by Gregory Prier 22 Apr '02

22 Apr '02

Thank you all for the help. I didn't understand how all of this fit together and was taking a completely incorrect approach. I was able to get the board to boot just fine with your suggestions. The PCI network drivers seemed to install successfully. I am able to ping from the target from the host but I can't ping the host from the target. The network interfaces are set up as follows: Host: 6 > ifconfig -a eth0 Link encap:Ethernet HWaddr 00:06:5B:2F:24:44 inet addr:131.179.187.20 Bcast:131.179.187.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:5239573 errors:0 dropped:0 overruns:161 frame:0 TX packets:1347440 errors:0 dropped:0 overruns:0 carrier:837 collisions:2858 txqueuelen:100 Interrupt:11 Base address:0xe880 eth1 Link encap:Ethernet HWaddr AA:00:00:00:00:00 inet addr:192.168.0.4 Bcast:192.168.0.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:30 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 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:46895 errors:0 dropped:0 overruns:0 frame:0 TX packets:46895 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 -note: I removed the vmware interfaces- and for the target: [root@Linux /]$ifconfig -a eth0 Link encap:Ethernet HWaddr AA:00:00:00:00:01 inet addr:192.168.0.4 Bcast:192.168.0.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:30 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:3924 Metric:1 RX packets:24 errors:0 dropped:0 overruns:0 frame:0 TX packets:24 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 Routing is setup as follows: Host: 7 > route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 131.179.187.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 192.168.0.0 0.0.0.0 255.255.255.0 U 0 0 0 eth1 172.16.230.0 0.0.0.0 255.255.255.0 U 0 0 0 vmnet8 192.168.139.0 0.0.0.0 255.255.255.0 U 0 0 0 vmnet1 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo 0.0.0.0 131.179.187.1 0.0.0.0 UG 0 0 0 eth0 Target: [root@Linux /]$route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 192.168.0.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 0.0.0.0 0.0.0.0 0.0.0.0 U 0 0 0 eth0 Any suggestions what could the be problem? Also, I wanted to clarify exactly how this system works. The host now has an ethernet interface called eth1 that is really a link into the pci bus. This interface talks to the target's interface (eth0) through the pci bus. These packets never leave the network. Why did intel and Radisys decide to have this communication occur over ethernet interfaces? Is this to make communication between the host and target easier? I assumed that ifconfig on the target would also show interfaces for each of the four RJ-45 jacks that are on the card . Since they didn't show up I assume that I access these through the strong arm? Do these ports get interfaces in linux? Thanks for the help. GKP ---------------------------------------------------------------------- Greg Prier Previously: A Mudder 3564 Boelter Hall Currently: A Bruin working UCLA CA 90095 towards a PhD (310) 825-8899 After that: No idea "Hell, there are no rules here. We're trying to accomplish something." Thomas Edison On Wed, 17 Apr 2002, Brandewie, Dirk J wrote: > The ENP2505 does not have an on board 82559, the tftp option > requires an ethernet card to be installed in the PMC site. > > The firmware on the board you have is setup to boot the board > by downloading the kernel and ramdisk over the PCI interface. > There is a bug in the cygmon on the board that keeps the > ethernet interface from working. This is my bug :-( but you > are the first person of all the boards we have donated who > has wanted to use tftp for download. You should have gotten > a CD with the linux drivers needed to boot the board in a > Linux host. > > --Dirk > > > -----Original Message----- > > From: Gregory Prier [mailto:greg@lever.cs.ucla.edu] > > Sent: Tuesday, April 16, 2002 3:06 PM > > To: ixp1200(a)CS.Princeton.EDU > > Subject: [ixp1200] Booting Intel ixp boards from tftp > > > > > > Our lab has two intel IXP 1200 evaluation boards. These fit > > in a standard PC > > (they have pci slots) and have serial access like an > > evaluation board. The > > boards are made by RadiSys and have the part number ENP2505. > > I have been trying > > for a few weeks to get them to boot linux via tftp and I > > haven't had any luck. > > > > When the board starts it immediately boots into Cygmon. When > > HOST mode is > > enabled Cygmon then counts down and starts automatically > > booting linux. At this > > point I get this error message: > > > > Failed to initialize TFTP transfer, check network connection > > and IP addr. > > > > When I am in FTP mode I immediately get this message: > > > > Press space bar to abort host download mode.... > > Press space bar to abort host download mode.... > > > > If I don't push space then this will continue forever. > > > > The bootoptions are set as follows: > > > > cygmon> bo > > [BootOptions]: > > 1 BOOTP/Manual (manual) > > 2 Local IP (131.179.187.21) > > 3 Remote IP (131.179.187.20) > > 4 Linux kernel file name (zImage) > > 5 Linux ramdisk file name (ramdisk_img.gz) > > 6 Linux kernel load offset (C4008000) > > 7 Linux boot command line string () > > 8 Countdown to auto-boot linux (60) > > 9 Starting SDRAM offset for host PCI access (0 MB) > > 10 SDRAM window size for host PCI access (256 MB) > > 11 Linux download mode [1] 0=FTP 1=HOST > > Enter option number (1 to 11), "q" to quit, "s" to save: > > > > 131.179.187.20 is a machine on the same ethernet segment as > > the ixp card. It > > is running a tftp server that has been tested and seems to work fine. > > Furthermore, I can run tcpdump on 131.179.187.20 and the ixp > > card never tries > > to get anything from the tftp server. It doesn't send any > > traffic onto the > > network, not even arp. > > > > I tried using the System Monitor mode on the board but this > > just hung when the > > board tried to boot. I also can't get to the boot manager > > because Cygmon seems > > to immediately start up instead. I can't find a way to make > > sure that the > > ethernet interfaces are up and that I have configured > > everything correctly > > before declaring the ROM bad and reflashing it. > > > > Specifically, my questions are: > > > > - What exactly is the "on-board 82559 (or 82559ER) Ethernet Interface" > > mentioned in the manuals? Does this refer to all of the 4 > > RJ-45 jacks on the > > board or some jack that I don't have or don't see? Since the > > board supposedly > > uses this for the tftp transfer I want to make sure that I am > > plugging the > > cables into the right jacks on the card. Is there a specific > > jack that I am > > supposed to use? I have tried all four and haven't had any luck yet. > > > > - Any idea why I can't get to the boot manager or the system > > monitor when > > booting the board? Were these removes to make room for Cygmon? > > > > - Does any one know of a way to hard reboot the card from > > inside cygmon? It > > would be a big help for testing. > > > > - Any idea why the board won't boot from the tftp server? > > > > - Has anyone seen better documentation on setting up the > > boards and what to do > > when they don't work? > > > > Any suggestions would be greatly appreciated. > > > > Thank you, > > Greg Prier > > > > ---------------------------------------------------------------------- > > Greg Prier Previously: A Mudder > > 3564 Boelter Hall Currently: A > > Bruin working > > UCLA CA 90095 towards a PhD > > (310) 825-8899 After that: No idea > > > > "Hell, there are no rules here. We're trying to accomplish > > something." > > Thomas Edison > > > > > > >

2 2

RE: [ixp1200] Booting Intel ixp boards from tftp
by Brandewie, Dirk J 22 Apr '02

22 Apr '02

Both the target and the host have the same IP address. The target should be 192.168.0.2 and the host 192.168.0.4. The exact IP addresses are not important but they need to be different :-) Other comments inline > -----Original Message----- > From: Gregory Prier [mailto:greg@lever.cs.ucla.edu] > Sent: Monday, April 22, 2002 1:33 PM > To: ixp mailing list > Subject: RE: [ixp1200] Booting Intel ixp boards from tftp > > > Thank you all for the help. I didn't understand how all of > this fit together > and was taking a completely incorrect approach. > > I was able to get the board to boot just fine with your > suggestions. The PCI > network drivers seemed to install successfully. I am able to > ping from the > target from the host but I can't ping the host from the > target. The network > interfaces are set up as follows: > > Host: > > 6 > ifconfig -a > eth0 Link encap:Ethernet HWaddr 00:06:5B:2F:24:44 > inet addr:131.179.187.20 Bcast:131.179.187.255 > Mask:255.255.255.0 > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > RX packets:5239573 errors:0 dropped:0 overruns:161 frame:0 > TX packets:1347440 errors:0 dropped:0 overruns:0 carrier:837 > collisions:2858 txqueuelen:100 > Interrupt:11 Base address:0xe880 > > eth1 Link encap:Ethernet HWaddr AA:00:00:00:00:00 > inet addr:192.168.0.4 Bcast:192.168.0.255 > Mask:255.255.255.0 > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > RX packets:30 errors:0 dropped:0 overruns:0 frame:0 > TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:100 > > 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:46895 errors:0 dropped:0 overruns:0 frame:0 > TX packets:46895 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:0 > > -note: I removed the vmware interfaces- > > and for the target: > > [root@Linux /]$ifconfig -a > eth0 Link encap:Ethernet HWaddr AA:00:00:00:00:01 > inet addr:192.168.0.4 Bcast:192.168.0.255 > Mask:255.255.255.0 > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > RX packets:0 errors:0 dropped:0 overruns:0 frame:0 > TX packets:30 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:100 > > lo Link encap:Local Loopback > inet addr:127.0.0.1 Mask:255.0.0.0 > UP LOOPBACK RUNNING MTU:3924 Metric:1 > RX packets:24 errors:0 dropped:0 overruns:0 frame:0 > TX packets:24 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:0 > > Routing is setup as follows: > > Host: > > 7 > route -n > Kernel IP routing table > Destination Gateway Genmask Flags Metric > Ref Use Iface > 131.179.187.0 0.0.0.0 255.255.255.0 U 0 > 0 0 eth0 > 192.168.0.0 0.0.0.0 255.255.255.0 U 0 > 0 0 eth1 > 172.16.230.0 0.0.0.0 255.255.255.0 U 0 > 0 0 vmnet8 > 192.168.139.0 0.0.0.0 255.255.255.0 U 0 > 0 0 vmnet1 > 127.0.0.0 0.0.0.0 255.0.0.0 U 0 > 0 0 lo > 0.0.0.0 131.179.187.1 0.0.0.0 UG 0 > 0 0 eth0 > > Target: > > [root@Linux /]$route -n > Kernel IP routing table > Destination Gateway Genmask Flags Metric > Ref Use Iface > 192.168.0.0 0.0.0.0 255.255.255.0 U 0 > 0 0 eth0 > 0.0.0.0 0.0.0.0 0.0.0.0 U 0 > 0 0 eth0 > > Any suggestions what could the be problem? > > Also, I wanted to clarify exactly how this system works. The > host now has an > ethernet interface called eth1 that is really a link into the > pci bus. This > interface talks to the target's interface (eth0) through the > pci bus. These > packets never leave the network. > > Why did intel and Radisys decide to have this communication > occur over ethernet > interfaces? Is this to make communication between the host > and target easier? We built it this way to mimic the way the SDK system is used with development described in the standard SDK documentation. > > I assumed that ifconfig on the target would also show > interfaces for each of the > four RJ-45 jacks that are on the card . Since they didn't > show up I assume that > I access these through the strong arm? Do these ports get > interfaces in linux? > The IXP1200 card is not a NIC it is a self contained router on a card for lack of a better term. The eth1 interface is there to make it easy for the tools to commuicate with the card. > Thanks for the help. > > GKP > --Dirk

1 0

router performance/stability test tool?
by Mirko Benz 22 Apr '02

22 Apr '02

Hi, Is there any open source toolkit for evaluating router implementations? (E.g. for performance measurements, functionality and conformance checking and so on) Thanks in advance, Mirko

1 0

bridalveil+linux+peth+L3f
by Istvan Matyasovszki 22 Apr '02

22 Apr '02

Hi, I'm trying to run/debug the L3 forwarding reference design on an Bridalveil board (with driver ver. 1.4). The ue.o,peth.o,octmac.o,rs_udebug and rtmapp files are from IXA Kit ver 2.01 (as the ue.o supplied with the ramdisk conflicts with the latest rtmapp supplied with SDK 2.01). After starting the hardware debug from the ixp1200 developer workbench in about 5 seconds I either get a frozen bridalveil board (" Get Microengine Status failed due to remote service timeout. ... To proceed, stop debugging and reboot the chip.") and a frozen hub or an error from the board "Bad descriptor fffffb00, dropping outgoing packet pethFreeDescriptor: Bad descriptor fffffb00" and after a minute or so the board is frozen and the hub in which is plugged-in is frozen too. Could you please give me an advice ? Has anybody a working example that uses the L3f reference design ? Thank You -istvan

2 4

RE: [ixp1200] question on behavior of IXF440/IXF1002
by Johnson, Erik J 17 Apr '02

17 Apr '02

It isn't correct for the software to send two receive requests as you have described in this scenario. When the MAC asserts the receive ready bit for a port it means (unless configured differently) that 1 mpacket is available. Before issuing the second receive request the software must wait for the receive ready bit to both get updated (i.e., the ready bus sequencer must poll the MAC and update the receive ready CSR) and be set. You will get unpredictable results when doing this on the IXF440. Of course, this is only true of the IXF440, not the IXF1002, since the IXF1002 can be sent speculative receive requests. In your scenario, the IXF1002 will send a cancelled receive control message for the second request. erik > -----Original Message----- > From: Wen Xu [mailto:wxu@CS.Princeton.EDU] > Sent: Wednesday, April 17, 2002 7:57 AM > To: ixp1200(a)CS.Princeton.EDU > Subject: [ixp1200] question on behavior of IXF440/IXF1002 > > > Hi, > > My question is like this: > > In the middle of receiving a 1500 byte packet, if at some > point the MAC > has 100 bytes in its Receive FIFO (but the end of packet is not there > yet) and set the ready flag, but two data requests come to the port > instead of one. > > Will the MAC send 36 bytes of data for the second request? Or will it > simply assert error for the second request, keeping the 36 byte in the > Receive FIFO for next data request? > > Wen >

1 0

question on behavior of IXF440/IXF1002
by Wen Xu 17 Apr '02

17 Apr '02

Hi, My question is like this: In the middle of receiving a 1500 byte packet, if at some point the MAC has 100 bytes in its Receive FIFO (but the end of packet is not there yet) and set the ready flag, but two data requests come to the port instead of one. Will the MAC send 36 bytes of data for the second request? Or will it simply assert error for the second request, keeping the 36 byte in the Receive FIFO for next data request? Wen

1 0

Booting intel ixp boards from tftp
by Gregory Prier 16 Apr '02

16 Apr '02

Our lab has two intel IXP 1200 evaluation boards. These fit in a standard PC (they have pci slots) and have serial access like an evaluation board. The boards are made by RadiSys and have the part number ENP2505. I have been trying for a few weeks to get them to boot linux via tftp and I haven't had any luck. When the board starts it immediately boots into Cygmon. When HOST mode is enabled Cygmon then counts down and starts automatically booting linux. At this point I get this error message: Failed to initialize TFTP transfer, check network connection and IP addr. When I am in FTP mode I immediately get this message: Press space bar to abort host download mode.... Press space bar to abort host download mode.... If I don't push space then this will continue forever. The bootoptions are set as follows: cygmon> bo [BootOptions]: 1 BOOTP/Manual (manual) 2 Local IP (131.179.187.21) 3 Remote IP (131.179.187.20) 4 Linux kernel file name (zImage) 5 Linux ramdisk file name (ramdisk_img.gz) 6 Linux kernel load offset (C4008000) 7 Linux boot command line string () 8 Countdown to auto-boot linux (60) 9 Starting SDRAM offset for host PCI access (0 MB) 10 SDRAM window size for host PCI access (256 MB) 11 Linux download mode [1] 0=FTP 1=HOST Enter option number (1 to 11), "q" to quit, "s" to save: 131.179.187.20 is a machine on the same ethernet segment as the ixp card. It is running a tftp server that has been tested and seems to work fine. Furthermore, I can run tcpdump on 131.179.187.20 and the ixp card never tries to get anything from the tftp server. It doesn't send any traffic onto the network, not even arp. I tried using the System Monitor mode on the board but this just hung when the board tried to boot. I also can't get to the boot manager because Cygmon seems to immediately start up instead. I can't find a way to make sure that the ethernet interfaces are up and that I have configured everything correctly before declaring the ROM bad and reflashing it. Specifically, my questions are: - What exactly is the "on-board 82559 (or 82559ER) Ethernet Interface" mentioned in the manuals? Does this refer to all of the 4 RJ-45 jacks on the board or some jack that I don't have or don't see? Since the board supposedly uses this for the tftp transfer I want to make sure that I am plugging the cables into the right jacks on the card. Is there a specific jack that I am supposed to use? I have tried all four and haven't had any luck yet. - Any idea why I can't get to the boot manager or the system monitor when booting the board? Were these removes to make room for Cygmon? - Does any one know of a way to hard reboot the card from inside cygmon? It would be a big help for testing. - Any idea why the board won't boot from the tftp server? - Has anyone seen better documentation on setting up the boards and what to do when they don't work? Any suggestions would be greatly appreciated. Thank you, Greg Prier ---------------------------------------------------------------------- Greg Prier Previously: A Mudder 3564 Boelter Hall Currently: A Bruin working UCLA CA 90095 towards a PhD (310) 825-8899 After that: No idea "Hell, there are no rules here. We're trying to accomplish something." Thomas Edison

2 1

Re: [ixp1200] Large packets on the IXP
by Michael E. Kounavis 15 Apr '02

15 Apr '02

Hi Abhijeet, Thanks for your interest on Genesis/NetBind and feedback. Here are some quick answers to your questions: > Right Thanks. So figure 12 reports 900,000pps , each packet = 64 bytes, > across all 4 ports, which means a bandwidth of 115Mbps across EACH port. > > Is this interpretation correct? Not exactly. We receive packets from 4 ports and forward them to the same output port. So these numbers are from a four-to-one configuration. Also the receiving rate is not ~100Mbps but mucher larger. To measure the maximum possible throughput we assume that the rfifo always has a packet received. The size of this packet is 64 bytes. So our code repeatedly gets a minimum size packet from the RFIFO and forwards the packet to the same output port. Our measured maximum throughput is in accordance with the maximum throughput measured by our Princeton colleagues. However in their measurements they use sixteen contexts for fowarding as opposed to four in our case. > Also, how do you saturate the IXP's 100 Mbps links through external > sources? Do you use packet generating hardware or just plain PC's with One way to do it is to use IXPs as packet generators, if you have many boxes to play with. So the packet generating code runs in the microengines of an IXP1200 system. > some kernel/user level packet generators? I am finding it preety hard to > get high rate sources on PC's without missing packet send deadlines on the > senders. > > Hmm.., maybe I could use multiple sources per port !! > > Also, how did you measure the 225 cycles per packet processing time? In > the paper you mention you measure it through some program on the > StrongArm. So basically do you pump a large number of packets, so that > your IP code runs through a large number of iterations and then take > coarse grained measurements of packet count from the StrongArm and then > divide it by the total time taken to forward packets? correct. Hope this helps, best, Michael

1 0