[casper] ERROR:Xflow - map: application received signal 9
hi all, I am trying to compile the correlator of tutorial 4 but i changed the adc to the acd083000x1. I have an error in the Starting program map. i went to gen_xps_files (line 636) and there is an error with the system(['xps -nw -scr run_xps.tcl system.xmp']) command. I do not know what this command do, any ideas? this is the matlab log #--# # Starting program map # map -timing -detail -ol high -xe n -register_duplication -o system_map.ncd -w -pr b system.ngd system.pcf #--# Release 14.2 - Map P.28xd (lin64) Copyright (c) 1995-2012 Xilinx, Inc. All rights reserved. PMSPEC -- Overriding Xilinx file /home/arturo/xilinx/14.2/ISE_DS/EDK/data/Xdh_PrimTypeLib.xda with local file /home/arturo/xilinx/14.2/ISE_DS/ISE/data/Xdh_PrimTypeLib.xda Using target part 5vsx95tff1136-1. Mapping design into LUTs... WARNING:MapLib:701 - Signal epb_addr_gp5 connected to top level port epb_addr_gp5 has been removed. WARNING:MapLib:701 - Signal epb_addr_gp4 connected to top level port epb_addr_gp4 has been removed. WARNING:MapLib:701 - Signal epb_addr_gp3 connected to top level port epb_addr_gp3 has been removed. WARNING:MapLib:701 - Signal aux0_clk_n connected to top level port aux0_clk_n has been removed. WARNING:MapLib:701 - Signal aux0_clk_p connected to top level port aux0_clk_p has been removed. WARNING:MapLib:701 - Signal aux1_clk_n connected to top level port aux1_clk_n has been removed. WARNING:MapLib:701 - Signal aux1_clk_p connected to top level port aux1_clk_p has been removed. WARNING:MapLib:41 - All members of TNM group infrastructure_inst_infrastructure_inst_sys_clk2x_int have been optimized out of the design. WARNING:MapLib:50 - The period specification TS_infrastructure_inst_infrastructure_inst_sys_clk2x_int has been discarded because the group infrastructure_inst_infrastructure_inst_sys_clk2x_int has been optimized away. *ERROR:Xflow - map: application received signal 9 ERROR:Xflow:42 - Aborting flow execution... * gmake: *** [__xps/system_routed] Error 1 ERROR:EDK - Error while running gmake -f system.make bits. Error using gen_xps_files (line 636) XPS failed. -- Atte. Arturo
Re: [casper] ERROR:Xflow - map: application received signal 9
Hi. Not sure of your error, but signal 9 is a sigkill by the system. Did someone kill your process or did you run completely out of memory? Running out of memory is a common cause of a spontaneous signal 9. John hi all, I am trying to compile the correlator of tutorial 4 but i changed the adc to the acd083000x1. I have an error in the Starting program map. i went to gen_xps_files (line 636) and there is an error with the system(['xps -nw -scr run_xps.tcl system.xmp']) command. I do not know what this command do, any ideas? this is the matlab log #--# # Starting program map # map -timing -detail -ol high -xe n -register_duplication -o system_map.ncd -w -pr b system.ngd system.pcf #--# Release 14.2 - Map P.28xd (lin64) Copyright (c) 1995-2012 Xilinx, Inc. All rights reserved. PMSPEC -- Overriding Xilinx file /home/arturo/xilinx/14.2/ISE_DS/EDK/data/Xdh_PrimTypeLib.xda with local file /home/arturo/xilinx/14.2/ISE_DS/ISE/data/Xdh_PrimTypeLib.xda Using target part 5vsx95tff1136-1. Mapping design into LUTs... WARNING:MapLib:701 - Signal epb_addr_gp5 connected to top level port epb_addr_gp5 has been removed. WARNING:MapLib:701 - Signal epb_addr_gp4 connected to top level port epb_addr_gp4 has been removed. WARNING:MapLib:701 - Signal epb_addr_gp3 connected to top level port epb_addr_gp3 has been removed. WARNING:MapLib:701 - Signal aux0_clk_n connected to top level port aux0_clk_n has been removed. WARNING:MapLib:701 - Signal aux0_clk_p connected to top level port aux0_clk_p has been removed. WARNING:MapLib:701 - Signal aux1_clk_n connected to top level port aux1_clk_n has been removed. WARNING:MapLib:701 - Signal aux1_clk_p connected to top level port aux1_clk_p has been removed. WARNING:MapLib:41 - All members of TNM group infrastructure_inst_infrastructure_inst_sys_clk2x_int have been optimized out of the design. WARNING:MapLib:50 - The period specification TS_infrastructure_inst_infrastructure_inst_sys_clk2x_int has been discarded because the group infrastructure_inst_infrastructure_inst_sys_clk2x_int has been optimized away. *ERROR:Xflow - map: application received signal 9 ERROR:Xflow:42 - Aborting flow execution... * gmake: *** [__xps/system_routed] Error 1 ERROR:EDK - Error while running gmake -f system.make bits. Error using gen_xps_files (line 636) XPS failed. -- Atte. Arturo
Re: [casper] ERROR:Xflow - map: application received signal 9
Hi, Now you mention is probably that i ran out of memory. I am going to check out. 2013/2/13 John Ford jf...@nrao.edu Hi. Not sure of your error, but signal 9 is a sigkill by the system. Did someone kill your process or did you run completely out of memory? Running out of memory is a common cause of a spontaneous signal 9. John hi all, I am trying to compile the correlator of tutorial 4 but i changed the adc to the acd083000x1. I have an error in the Starting program map. i went to gen_xps_files (line 636) and there is an error with the system(['xps -nw -scr run_xps.tcl system.xmp']) command. I do not know what this command do, any ideas? this is the matlab log #--# # Starting program map # map -timing -detail -ol high -xe n -register_duplication -o system_map.ncd -w -pr b system.ngd system.pcf #--# Release 14.2 - Map P.28xd (lin64) Copyright (c) 1995-2012 Xilinx, Inc. All rights reserved. PMSPEC -- Overriding Xilinx file /home/arturo/xilinx/14.2/ISE_DS/EDK/data/Xdh_PrimTypeLib.xda with local file /home/arturo/xilinx/14.2/ISE_DS/ISE/data/Xdh_PrimTypeLib.xda Using target part 5vsx95tff1136-1. Mapping design into LUTs... WARNING:MapLib:701 - Signal epb_addr_gp5 connected to top level port epb_addr_gp5 has been removed. WARNING:MapLib:701 - Signal epb_addr_gp4 connected to top level port epb_addr_gp4 has been removed. WARNING:MapLib:701 - Signal epb_addr_gp3 connected to top level port epb_addr_gp3 has been removed. WARNING:MapLib:701 - Signal aux0_clk_n connected to top level port aux0_clk_n has been removed. WARNING:MapLib:701 - Signal aux0_clk_p connected to top level port aux0_clk_p has been removed. WARNING:MapLib:701 - Signal aux1_clk_n connected to top level port aux1_clk_n has been removed. WARNING:MapLib:701 - Signal aux1_clk_p connected to top level port aux1_clk_p has been removed. WARNING:MapLib:41 - All members of TNM group infrastructure_inst_infrastructure_inst_sys_clk2x_int have been optimized out of the design. WARNING:MapLib:50 - The period specification TS_infrastructure_inst_infrastructure_inst_sys_clk2x_int has been discarded because the group infrastructure_inst_infrastructure_inst_sys_clk2x_int has been optimized away. *ERROR:Xflow - map: application received signal 9 ERROR:Xflow:42 - Aborting flow execution... * gmake: *** [__xps/system_routed] Error 1 ERROR:EDK - Error while running gmake -f system.make bits. Error using gen_xps_files (line 636) XPS failed. -- Atte. Arturo -- Atte. Arturo
[casper] Uploading bof files.
Hi all, there is a new rom file system (for soloboot) and tcpborphserver3 (copy to nfsroot/usr/local/sbin) for nfsboot in github ( https://github.com/ska-sa/roach2_nfs_uboot/) There is an email detailing how to upload a new romfs but it's fairly straight forward, set up a tftp server, use a terminal emulator like minicom and boot into uboot. Then type run tftproot. You can also use run newroot and then y-modem the file across, so no need for tftp but this takes a very long time! tcpborphserver3 now supports borph upload. There are 2 ways of using this command: 1. As Paul explained: you can use the upload_bof method in the FpgaClient class in corr to upload a bof file to your ROACH. Once it's uploaded, though, it will skip uploading it unless you force it to do so. Use progdev to program the device as usual. 2. Use the katcp command ?uploadbof port filename. The boffile can now be sent to that port using socat or netcat or similiar. Regards Alec On Tue, Feb 12, 2013 at 5:31 PM, Marc Welz m...@ska.ac.za wrote: Hello 1/ a- how can I find the version of my romfs and also how can I find the version of my Tcpborphserver3 if you telnet to port 7147, there should be a version string for tcpborphserver3. The romfs contains a timestamp in its filesystem label, but since it such a small system, you can pretty much use the tcpborphserver3 version string as a proxy for it b- which version for the uboot and the tcborphserver3 should I use, should I keep the current version above ? where is it located? The source is in git at github.com/ska-sa 2/ I heard that the configuration with the SD card is not yet working ? is it still the case ? Yes. A roach2 has sufficient onboard flash to keep a number of (optionally compressed) bof files there without having to use external storage, so we not allocated anybody to work on the MMC logic. Also note that MMC cards can be interesting, see http://www.bunniestudios.com/blog/?p=2297 So we recommend you soloboot, ie boot off the onboard flash. 3/I want to load my boffiles, can I use ssh ? or should Iuse the telnet to tranfer my files ? How should I do If I want to keep my boffiles alive? I used to store them on the SD card, even when shutting down my roach board. Do I need to re-load or to copy my boffiles after switching on my roach2 boards? The content of /usr on the onboard flash is preserved, so the bof files in /usr/bof are retained. I read somewhere that it could be possible to upload from the python ?(I've never done it) How long does it take to the boffile to be copied ? before launching the fpgaprogdev() ? I think Paul answered that (but note that you will need a pretty recent version of tcpborphserver3 to use the uploadbof request). If you are concerned about speed, you can transfer a bof file beforehand (say using wget or ftpget or an nfs mountpoint) - then the progdev is pretty quick, and you can then even arrange that the file is programmed at startup (by adding a progdev command to etc/tcpborphserver3.init) regards marc
Re: [casper] casper Digest, Vol 63, Issue 9
I recently got a ROACH2 from digicom and I'm trying to upload a bof file. My tcpborphserver does not support the KATCP command upload_bof. I do have the command upload which will upload and run a bof file but when I read a register something crashes. According to the previous posts, I believe I need to update to tcpborphserver3. Can someone give me an idea of how to do this? Do I have to find the source then compile and install? Does someone have a how-to? Thanks, Jason Castro NRAO On 2/12/2013 3:14 PM, casper-requ...@lists.berkeley.edu wrote: Send casper mailing list submissions to casper@lists.berkeley.edu To subscribe or unsubscribe via the World Wide Web, visit https://calmail.berkeley.edu/manage/list/listinfo/casper@lists.berkeley.edu or, via email, send a message with subject or body 'help' to casper-requ...@lists.berkeley.edu You can reach the person managing the list at casper-ow...@lists.berkeley.edu When replying, please edit your Subject line so it is more specific than Re: Contents of casper digest... Today's Topics: 1. Re: Roach2 configuration and setup (Paul Prozesky) 2. Re: Roach2 configuration and setup (Marc Welz) -- Message: 1 Date: Tue, 12 Feb 2013 15:26:11 +0200 From: Paul Prozesky paul.proze...@gmail.com Subject: Re: [casper] Roach2 configuration and setup To: LIST CASPER casper@lists.berkeley.edu Message-ID: CAASPtuoUXfPBh==XarFqvSkS5d=+oymrdce3e9ircmy67jt...@mail.gmail.com Content-Type: text/plain; charset=iso-8859-1 Afternoon Guy Regarding question 3... if you're running tcpborphserver3 you can use the upload_bof method in the FpgaClient class in corr to upload a bof file to your ROACH. I just checked and it took about 12s to copy a boffile using this. Once it's uploaded, though, it will skip uploading it unless you force it to do so. Use progdev to program the device as usual. Regards Paul On 12 February 2013 00:32, Guy kenfack guy.kenf...@gmail.com wrote: Hello, we received 2 roach2 boards from digicom (20th dec 2012).we managed to configure our minicom with centos6.2 final, and tty usb2. I've several questions about the advised version for the uboot and the tcborphserver3 u-boot versionof my boards: u-boot 2011.06-rc2-0-gd422dc0-dirty(Nov 8 2012 - :16:04:14) #SN : Roach2.2 batch = D#6#10 #kernel : Linux roach 02 06 0A 3.4.0 - rc3+ Tcpborphserver3 1/ a- how can I find the version of my romfs and also how can I find the version of my Tcpborphserver3 b- which version for the uboot and the tcborphserver3 should I use, should I keep the current version above ? where is it located? 2/ I heard that the configuration with the SD card is not yet working ? is it still the case ? 3/I want to load my boffiles, can I use ssh ? or should Iuse the telnet to tranfer my files ? How should I do If I want to keep my boffiles alive? I used to store them on the SD card, even when shutting down my roach board. Do I need to re-load or to copy my boffiles after switching on my roach2 boards? I read somewhere that it could be possible to upload from the python ?(I've never done it) How long does it take to the boffile to be copied ? before launching the fpgaprogdev() ? Thank you for your help, -- next part -- An HTML attachment scrubbed and removed. HTML attachments are only available in MIME digests. -- Message: 2 Date: Tue, 12 Feb 2013 15:31:30 + From: Marc Welz m...@ska.ac.za Subject: Re: [casper] Roach2 configuration and setup To: Guy kenfack guy.kenf...@gmail.com Cc: casper@lists.berkeley.edu Message-ID: cagrhwasfwdi5haxzo8h4jlsvq7hmf5ow_x7jkyafodzzilw...@mail.gmail.com Content-Type: text/plain; charset=ISO-8859-1 Hello 1/ a- how can I find the version of my romfs and also how can I find the version of my Tcpborphserver3 if you telnet to port 7147, there should be a version string for tcpborphserver3. The romfs contains a timestamp in its filesystem label, but since it such a small system, you can pretty much use the tcpborphserver3 version string as a proxy for it b- which version for the uboot and the tcborphserver3 should I use, should I keep the current version above ? where is it located? The source is in git at github.com/ska-sa 2/ I heard that the configuration with the SD card is not yet working ? is it still the case ? Yes. A roach2 has sufficient onboard flash to keep a number of (optionally compressed) bof files there without having to use external storage, so we not allocated anybody to work on the MMC logic. Also note that MMC cards can be interesting, see http://www.bunniestudios.com/blog/?p=2297 So we recommend you soloboot, ie boot off the onboard flash. 3/I want to load my boffiles, can I use ssh ? or should Iuse the telnet to tranfer my files ? How should I do If I want to keep my boffiles alive? I used to store them on the SD card, even
Re: [casper] casper Digest, Vol 63, Issue 9
Hi John. Grab the latest versions from https://github.com/ska-sa/roach2_nfs_uboot/. If you netboot: copy tcpborphserver3 to nfs_root_directory/usr/local/sbin. Tcpbporphserver3 will load automatically when the roach boots. If you soloboot (boot the root file system stored in flash), which is the easiest please follow these instructions: You need to update to the latest romfs which includes the new tcpborphserver3. To update: ROACH2 Test machine set-up instructions: https://docs.google.com/a/ska.ac.za/document/d/1tqw4C6uZ6EULl1OykTFL_vQTnK52UBr0aYqTg44E5wg/edit ) 1. Set up a tftp and dhcp server (see the tftp section in the test machine setup instructions). 2. Create a symlink or rename the romfs image downloaded from github to romfs e.g. ln -s roach2-root-readdebug-2013-02-12.romfs romfs in the tftp directory. 3. Connect to the roach via USB using a terminal emulator: In linux we use minicom, please follow the minicom setup instructions in the document. 4. Switch the roach on or reset and interrupt the boot process 5. At the uboot prompt type: run tftproot 6. This takes a long time! About 5 minutes. 7. type reset and let the board boot linux 8. Login using root (password is blank) 9. rm /usr/.keep 10. reboot -f If you do not which to run a dhcp and tftp server you can use run newroot. This macro starts a y-modem receiver, the file can now be sent using minicom via y-modem protocol, but this takes a long time and you really need a dhcp server unless you are going to set the roach ip address manually. You can set the IP address by editing /etc/network/interfaces To measure temperatures, voltages and currents: All sensors are available using kcp commands. You can telnet to the board to run kcp commands (telnet board_ip 7147) or if you have access to the console you can run the commands using the program kcpcmd (kcpcmd sensor-list note that there is no ?) ?sensor-list - lists all the sensors ?sensor-value - lists all the current values ?sensor-sampling sensor_name event - this will give updates when the sensor changes. If anything is unclear please don't hesitate to ask. Regards Alec On Wed, Feb 13, 2013 at 8:15 PM, Jason Castro jcas...@nrao.edu wrote: I recently got a ROACH2 from digicom and I'm trying to upload a bof file. My tcpborphserver does not support the KATCP command upload_bof. I do have the command upload which will upload and run a bof file but when I read a register something crashes. According to the previous posts, I believe I need to update to tcpborphserver3. Can someone give me an idea of how to do this? Do I have to find the source then compile and install? Does someone have a how-to? Thanks, Jason Castro NRAO On 2/12/2013 3:14 PM, casper-request@lists.berkeley.**educasper-requ...@lists.berkeley.eduwrote: Send casper mailing list submissions to casper@lists.berkeley.edu To subscribe or unsubscribe via the World Wide Web, visit https://calmail.berkeley.edu/**manage/list/listinfo/casper@** lists.berkeley.eduhttps://calmail.berkeley.edu/manage/list/listinfo/casper@lists.berkeley.edu or, via email, send a message with subject or body 'help' to casper-request@lists.berkeley.**educasper-requ...@lists.berkeley.edu You can reach the person managing the list at casper-owner@lists.berkeley.**educasper-ow...@lists.berkeley.edu When replying, please edit your Subject line so it is more specific than Re: Contents of casper digest... Today's Topics: 1. Re: Roach2 configuration and setup (Paul Prozesky) 2. Re: Roach2 configuration and setup (Marc Welz) --**--** -- Message: 1 Date: Tue, 12 Feb 2013 15:26:11 +0200 From: Paul Prozesky paul.proze...@gmail.com Subject: Re: [casper] Roach2 configuration and setup To: LIST CASPER casper@lists.berkeley.edu Message-ID: CAASPtuoUXfPBh==XarFqvSkS5d=+**OyMRDCE3E9iRcmy67jTn8w@mail.** gmail.com oymrdce3e9ircmy67jt...@mail.gmail.com Content-Type: text/plain; charset=iso-8859-1 Afternoon Guy Regarding question 3... if you're running tcpborphserver3 you can use the upload_bof method in the FpgaClient class in corr to upload a bof file to your ROACH. I just checked and it took about 12s to copy a boffile using this. Once it's uploaded, though, it will skip uploading it unless you force it to do so. Use progdev to program the device as usual. Regards Paul On 12 February 2013 00:32, Guy kenfack guy.kenf...@gmail.com wrote: Hello, we received 2 roach2 boards from digicom (20th dec 2012).we managed to configure our minicom with centos6.2 final, and tty usb2. I've several questions about the advised version for the uboot and the tcborphserver3 u-boot versionof my boards: u-boot 2011.06-rc2-0-gd422dc0-**dirty(Nov 8 2012 - :16:04:14) #SN : Roach2.2 batch = D#6#10 #kernel : Linux roach 02 06 0A 3.4.0 - rc3+ Tcpborphserver3 1/ a- how can I find the version of my romfs
Re: [casper] casper Digest, Vol 63, Issue 9
I just dumped this info on the wiki on the ROACH 2 page since it seems very useful and didn't appear to be there already. Glenn On Wed, Feb 13, 2013 at 2:21 PM, Alec Rust alec.r...@ska.ac.za wrote: Hi John. Grab the latest versions from https://github.com/ska-sa/roach2_nfs_uboot/. If you netboot: copy tcpborphserver3 to nfs_root_directory/usr/local/sbin. Tcpbporphserver3 will load automatically when the roach boots. If you soloboot (boot the root file system stored in flash), which is the easiest please follow these instructions: You need to update to the latest romfs which includes the new tcpborphserver3. To update: ROACH2 Test machine set-up instructions: https://docs.google.com/a/ska.ac.za/document/d/1tqw4C6uZ6EULl1OykTFL_vQTnK52UBr0aYqTg44E5wg/edit) 1. Set up a tftp and dhcp server (see the tftp section in the test machine setup instructions). 2. Create a symlink or rename the romfs image downloaded from github to romfs e.g. ln -s roach2-root-readdebug-2013-02-12.romfs romfs in the tftp directory. 3. Connect to the roach via USB using a terminal emulator: In linux we use minicom, please follow the minicom setup instructions in the document. 4. Switch the roach on or reset and interrupt the boot process 5. At the uboot prompt type: run tftproot 6. This takes a long time! About 5 minutes. 7. type reset and let the board boot linux 8. Login using root (password is blank) 9. rm /usr/.keep 10. reboot -f If you do not which to run a dhcp and tftp server you can use run newroot. This macro starts a y-modem receiver, the file can now be sent using minicom via y-modem protocol, but this takes a long time and you really need a dhcp server unless you are going to set the roach ip address manually. You can set the IP address by editing /etc/network/interfaces To measure temperatures, voltages and currents: All sensors are available using kcp commands. You can telnet to the board to run kcp commands (telnet board_ip 7147) or if you have access to the console you can run the commands using the program kcpcmd (kcpcmd sensor-list note that there is no ?) ?sensor-list - lists all the sensors ?sensor-value - lists all the current values ?sensor-sampling sensor_name event - this will give updates when the sensor changes. If anything is unclear please don't hesitate to ask. Regards Alec On Wed, Feb 13, 2013 at 8:15 PM, Jason Castro jcas...@nrao.edu wrote: I recently got a ROACH2 from digicom and I'm trying to upload a bof file. My tcpborphserver does not support the KATCP command upload_bof. I do have the command upload which will upload and run a bof file but when I read a register something crashes. According to the previous posts, I believe I need to update to tcpborphserver3. Can someone give me an idea of how to do this? Do I have to find the source then compile and install? Does someone have a how-to? Thanks, Jason Castro NRAO On 2/12/2013 3:14 PM, casper-requ...@lists.berkeley.edu wrote: Send casper mailing list submissions to casper@lists.berkeley.edu To subscribe or unsubscribe via the World Wide Web, visit https://calmail.berkeley.edu/manage/list/listinfo/casper@lists.berkeley.edu or, via email, send a message with subject or body 'help' to casper-requ...@lists.berkeley.edu You can reach the person managing the list at casper-ow...@lists.berkeley.edu When replying, please edit your Subject line so it is more specific than Re: Contents of casper digest... Today's Topics: 1. Re: Roach2 configuration and setup (Paul Prozesky) 2. Re: Roach2 configuration and setup (Marc Welz) -- Message: 1 Date: Tue, 12 Feb 2013 15:26:11 +0200 From: Paul Prozesky paul.proze...@gmail.com Subject: Re: [casper] Roach2 configuration and setup To: LIST CASPER casper@lists.berkeley.edu Message-ID: CAASPtuoUXfPBh==XarFqvSkS5d=+oymrdce3e9ircmy67jt...@mail.gmail.com Content-Type: text/plain; charset=iso-8859-1 Afternoon Guy Regarding question 3... if you're running tcpborphserver3 you can use the upload_bof method in the FpgaClient class in corr to upload a bof file to your ROACH. I just checked and it took about 12s to copy a boffile using this. Once it's uploaded, though, it will skip uploading it unless you force it to do so. Use progdev to program the device as usual. Regards Paul On 12 February 2013 00:32, Guy kenfack guy.kenf...@gmail.com wrote: Hello, we received 2 roach2 boards from digicom (20th dec 2012).we managed to configure our minicom with centos6.2 final, and tty usb2. I've several questions about the advised version for the uboot and the tcborphserver3 u-boot versionof my boards: u-boot 2011.06-rc2-0-gd422dc0-dirty(Nov 8 2012 - :16:04:14) #SN : Roach2.2 batch = D#6#10 #kernel : Linux roach 02 06 0A 3.4.0 - rc3+ Tcpborphserver3 1/ a- how can I find the version of my romfs and also how can
[casper] CASPER 10 GbE yellow block
Hi All, Using the latest 10 GbE yellow block, has anyone achieved a sustained data rate of greater than 8 Gb/s? Thanks, Rich
Re: [casper] casper Digest, Vol 63, Issue 9
Thanks Glenn... I am preparing a whole lot of documentation for the wiki. Hope to have it up soon! On Wed, Feb 13, 2013 at 9:38 PM, G Jones glenn.calt...@gmail.com wrote: I just dumped this info on the wiki on the ROACH 2 page since it seems very useful and didn't appear to be there already. Glenn On Wed, Feb 13, 2013 at 2:21 PM, Alec Rust alec.r...@ska.ac.za wrote: Hi John. Grab the latest versions from https://github.com/ska-sa/roach2_nfs_uboot/. If you netboot: copy tcpborphserver3 to nfs_root_directory/usr/local/sbin. Tcpbporphserver3 will load automatically when the roach boots. If you soloboot (boot the root file system stored in flash), which is the easiest please follow these instructions: You need to update to the latest romfs which includes the new tcpborphserver3. To update: ROACH2 Test machine set-up instructions: https://docs.google.com/a/ska.ac.za/document/d/1tqw4C6uZ6EULl1OykTFL_vQTnK52UBr0aYqTg44E5wg/edit ) 1. Set up a tftp and dhcp server (see the tftp section in the test machine setup instructions). 2. Create a symlink or rename the romfs image downloaded from github to romfs e.g. ln -s roach2-root-readdebug-2013-02-12.romfs romfs in the tftp directory. 3. Connect to the roach via USB using a terminal emulator: In linux we use minicom, please follow the minicom setup instructions in the document. 4. Switch the roach on or reset and interrupt the boot process 5. At the uboot prompt type: run tftproot 6. This takes a long time! About 5 minutes. 7. type reset and let the board boot linux 8. Login using root (password is blank) 9. rm /usr/.keep 10. reboot -f If you do not which to run a dhcp and tftp server you can use run newroot. This macro starts a y-modem receiver, the file can now be sent using minicom via y-modem protocol, but this takes a long time and you really need a dhcp server unless you are going to set the roach ip address manually. You can set the IP address by editing /etc/network/interfaces To measure temperatures, voltages and currents: All sensors are available using kcp commands. You can telnet to the board to run kcp commands (telnet board_ip 7147) or if you have access to the console you can run the commands using the program kcpcmd (kcpcmd sensor-list note that there is no ?) ?sensor-list - lists all the sensors ?sensor-value - lists all the current values ?sensor-sampling sensor_name event - this will give updates when the sensor changes. If anything is unclear please don't hesitate to ask. Regards Alec On Wed, Feb 13, 2013 at 8:15 PM, Jason Castro jcas...@nrao.edu wrote: I recently got a ROACH2 from digicom and I'm trying to upload a bof file. My tcpborphserver does not support the KATCP command upload_bof. I do have the command upload which will upload and run a bof file but when I read a register something crashes. According to the previous posts, I believe I need to update to tcpborphserver3. Can someone give me an idea of how to do this? Do I have to find the source then compile and install? Does someone have a how-to? Thanks, Jason Castro NRAO On 2/12/2013 3:14 PM, casper-requ...@lists.berkeley.edu wrote: Send casper mailing list submissions to casper@lists.berkeley.edu To subscribe or unsubscribe via the World Wide Web, visit https://calmail.berkeley.edu/manage/list/listinfo/casper@lists.berkeley.edu or, via email, send a message with subject or body 'help' to casper-requ...@lists.berkeley.edu You can reach the person managing the list at casper-ow...@lists.berkeley.edu When replying, please edit your Subject line so it is more specific than Re: Contents of casper digest... Today's Topics: 1. Re: Roach2 configuration and setup (Paul Prozesky) 2. Re: Roach2 configuration and setup (Marc Welz) -- Message: 1 Date: Tue, 12 Feb 2013 15:26:11 +0200 From: Paul Prozesky paul.proze...@gmail.com Subject: Re: [casper] Roach2 configuration and setup To: LIST CASPER casper@lists.berkeley.edu Message-ID: CAASPtuoUXfPBh==XarFqvSkS5d=+oymrdce3e9ircmy67jt...@mail.gmail.com Content-Type: text/plain; charset=iso-8859-1 Afternoon Guy Regarding question 3... if you're running tcpborphserver3 you can use the upload_bof method in the FpgaClient class in corr to upload a bof file to your ROACH. I just checked and it took about 12s to copy a boffile using this. Once it's uploaded, though, it will skip uploading it unless you force it to do so. Use progdev to program the device as usual. Regards Paul On 12 February 2013 00:32, Guy kenfack guy.kenf...@gmail.com wrote: Hello, we received 2 roach2 boards from digicom (20th dec 2012).we managed to configure our minicom with centos6.2 final, and tty usb2. I've several
Re: [casper] CASPER 10 GbE yellow block
Hi, Rich, We've sustained 9.9 Gpbs simultaneously from two 10 GbE ports (19.8 Gbps aggregate) on a ROACH into a PC with two 10 GbE NICs without dropping any packets. This was direct connect (i.e. no switch) and transmit only (i.e. no packets were sent back to the ROACH). Hope this helps, Dave On Feb 13, 2013, at 1:06 PM, Rich Lacasse wrote: Hi All, Using the latest 10 GbE yellow block, has anyone achieved a sustained data rate of greater than 8 Gb/s? Thanks, Rich
Re: [casper] CASPER 10 GbE yellow block
Hi, Rich, We've sustained 9.9 Gpbs simultaneously from two 10 GbE ports (19.8 Gbps aggregate) on a ROACH into a PC with two 10 GbE NICs without dropping any packets. This was direct connect (i.e. no switch) and transmit only (i.e. no packets were sent back to the ROACH). Rich, Dave, can you tell us the versions of the libraries you have? Thanks! John Hope this helps, Dave On Feb 13, 2013, at 1:06 PM, Rich Lacasse wrote: Hi All, Using the latest 10 GbE yellow block, has anyone achieved a sustained data rate of greater than 8 Gb/s? Thanks, Rich
Re: [casper] CASPER 10 GbE yellow block
Hi, John, The benchmarking tests were done here by Christopher Schollar who was visiting from the University of Cape Town. I'm not sure which library version he used, but I can find out. It was mostly just curiosity on my part, but are you using the same blocks for your work? John Dave On Feb 13, 2013, at 1:30 PM, John Ford wrote: Hi, Rich, We've sustained 9.9 Gpbs simultaneously from two 10 GbE ports (19.8 Gbps aggregate) on a ROACH into a PC with two 10 GbE NICs without dropping any packets. This was direct connect (i.e. no switch) and transmit only (i.e. no packets were sent back to the ROACH). Rich, Dave, can you tell us the versions of the libraries you have? Thanks! John Hope this helps, Dave On Feb 13, 2013, at 1:06 PM, Rich Lacasse wrote: Hi All, Using the latest 10 GbE yellow block, has anyone achieved a sustained data rate of greater than 8 Gb/s? Thanks, Rich
Re: [casper] CASPER 10 GbE yellow block
Hi, John, My current work is on ROACH2 and the 10_Gbe_v2 block is the only 10 GbE block supported on ROACH2 so that's the one I'm currently designing with (but not yet using myself). I'm not sure which block Christopher used in the packet generator. His design ran on a ROACH so it could have been using the ten_GbE or ten_Gbe_v2 block. Dave On Feb 13, 2013, at 2:01 PM, John Ford wrote: Hi, John, The benchmarking tests were done here by Christopher Schollar who was visiting from the University of Cape Town. I'm not sure which library version he used, but I can find out. It was mostly just curiosity on my part, but are you using the same blocks for your work? John Dave On Feb 13, 2013, at 1:30 PM, John Ford wrote: Hi, Rich, We've sustained 9.9 Gpbs simultaneously from two 10 GbE ports (19.8 Gbps aggregate) on a ROACH into a PC with two 10 GbE NICs without dropping any packets. This was direct connect (i.e. no switch) and transmit only (i.e. no packets were sent back to the ROACH). Rich, Dave, can you tell us the versions of the libraries you have? Thanks! John Hope this helps, Dave On Feb 13, 2013, at 1:06 PM, Rich Lacasse wrote: Hi All, Using the latest 10 GbE yellow block, has anyone achieved a sustained data rate of greater than 8 Gb/s? Thanks, Rich
Re: [casper] CASPER 10 GbE yellow block
Hi Dave and others, I was under the impression that ten_Gbe_v2 was suppose to support ROACH2 as well. At least, its mask supports parameters specific to ROACH2 (such as choice between CX4, SFP+ and ports etc.). Rather, these options are available in the mask only for ROACH2. Is 10_Gbe_v2 part of SKA-SA github repository now and is that supposed to be used instead of ten_Gbe_v2? I have not updated the repository of late and have been using ten_Gbe_v2 for my ROACH2 design. Thanks, Nimish On Wed, Feb 13, 2013 at 5:07 PM, David MacMahon dav...@astro.berkeley.eduwrote: Hi, John, My current work is on ROACH2 and the 10_Gbe_v2 block is the only 10 GbE block supported on ROACH2 so that's the one I'm currently designing with (but not yet using myself). I'm not sure which block Christopher used in the packet generator. His design ran on a ROACH so it could have been using the ten_GbE or ten_Gbe_v2 block. Dave On Feb 13, 2013, at 2:01 PM, John Ford wrote: Hi, John, The benchmarking tests were done here by Christopher Schollar who was visiting from the University of Cape Town. I'm not sure which library version he used, but I can find out. It was mostly just curiosity on my part, but are you using the same blocks for your work? John Dave On Feb 13, 2013, at 1:30 PM, John Ford wrote: Hi, Rich, We've sustained 9.9 Gpbs simultaneously from two 10 GbE ports (19.8 Gbps aggregate) on a ROACH into a PC with two 10 GbE NICs without dropping any packets. This was direct connect (i.e. no switch) and transmit only (i.e. no packets were sent back to the ROACH). Rich, Dave, can you tell us the versions of the libraries you have? Thanks! John Hope this helps, Dave On Feb 13, 2013, at 1:06 PM, Rich Lacasse wrote: Hi All, Using the latest 10 GbE yellow block, has anyone achieved a sustained data rate of greater than 8 Gb/s? Thanks, Rich
Re: [casper] CASPER 10 GbE yellow block
I think it's just a typo and ten_gbe_v2 == 10_gbe_v2 On Wed, Feb 13, 2013 at 5:47 PM, Nimish Sane nimishs...@gmail.com wrote: Hi Dave and others, I was under the impression that ten_Gbe_v2 was suppose to support ROACH2 as well. At least, its mask supports parameters specific to ROACH2 (such as choice between CX4, SFP+ and ports etc.). Rather, these options are available in the mask only for ROACH2. Is 10_Gbe_v2 part of SKA-SA github repository now and is that supposed to be used instead of ten_Gbe_v2? I have not updated the repository of late and have been using ten_Gbe_v2 for my ROACH2 design. Thanks, Nimish On Wed, Feb 13, 2013 at 5:07 PM, David MacMahon dav...@astro.berkeley.edu wrote: Hi, John, My current work is on ROACH2 and the 10_Gbe_v2 block is the only 10 GbE block supported on ROACH2 so that's the one I'm currently designing with (but not yet using myself). I'm not sure which block Christopher used in the packet generator. His design ran on a ROACH so it could have been using the ten_GbE or ten_Gbe_v2 block. Dave On Feb 13, 2013, at 2:01 PM, John Ford wrote: Hi, John, The benchmarking tests were done here by Christopher Schollar who was visiting from the University of Cape Town. I'm not sure which library version he used, but I can find out. It was mostly just curiosity on my part, but are you using the same blocks for your work? John Dave On Feb 13, 2013, at 1:30 PM, John Ford wrote: Hi, Rich, We've sustained 9.9 Gpbs simultaneously from two 10 GbE ports (19.8 Gbps aggregate) on a ROACH into a PC with two 10 GbE NICs without dropping any packets. This was direct connect (i.e. no switch) and transmit only (i.e. no packets were sent back to the ROACH). Rich, Dave, can you tell us the versions of the libraries you have? Thanks! John Hope this helps, Dave On Feb 13, 2013, at 1:06 PM, Rich Lacasse wrote: Hi All, Using the latest 10 GbE yellow block, has anyone achieved a sustained data rate of greater than 8 Gb/s? Thanks, Rich
Re: [casper] CASPER 10 GbE yellow block
Hi, Nimish, Sorry for my 10_Gbe_v2 typo. I meant ten_Gbe_v2. There is no 10_Gbe_v2 block. I am using the ten_Gbe_v2 block for ROACH2. Dave On Feb 13, 2013, at 2:47 PM, Nimish Sane wrote: Hi Dave and others, I was under the impression that ten_Gbe_v2 was suppose to support ROACH2 as well. At least, its mask supports parameters specific to ROACH2 (such as choice between CX4, SFP+ and ports etc.). Rather, these options are available in the mask only for ROACH2. Is 10_Gbe_v2 part of SKA-SA github repository now and is that supposed to be used instead of ten_Gbe_v2? I have not updated the repository of late and have been using ten_Gbe_v2 for my ROACH2 design. Thanks, Nimish On Wed, Feb 13, 2013 at 5:07 PM, David MacMahon dav...@astro.berkeley.edu wrote: Hi, John, My current work is on ROACH2 and the 10_Gbe_v2 block is the only 10 GbE block supported on ROACH2 so that's the one I'm currently designing with (but not yet using myself). I'm not sure which block Christopher used in the packet generator. His design ran on a ROACH so it could have been using the ten_GbE or ten_Gbe_v2 block. Dave On Feb 13, 2013, at 2:01 PM, John Ford wrote: Hi, John, The benchmarking tests were done here by Christopher Schollar who was visiting from the University of Cape Town. I'm not sure which library version he used, but I can find out. It was mostly just curiosity on my part, but are you using the same blocks for your work? John Dave On Feb 13, 2013, at 1:30 PM, John Ford wrote: Hi, Rich, We've sustained 9.9 Gpbs simultaneously from two 10 GbE ports (19.8 Gbps aggregate) on a ROACH into a PC with two 10 GbE NICs without dropping any packets. This was direct connect (i.e. no switch) and transmit only (i.e. no packets were sent back to the ROACH). Rich, Dave, can you tell us the versions of the libraries you have? Thanks! John Hope this helps, Dave On Feb 13, 2013, at 1:06 PM, Rich Lacasse wrote: Hi All, Using the latest 10 GbE yellow block, has anyone achieved a sustained data rate of greater than 8 Gb/s? Thanks, Rich
Re: [casper] CASPER 10 GbE yellow block
Ok! Thanks! Nimish On Feb 13, 2013, at 6:10 PM, David MacMahon dav...@astro.berkeley.edu wrote: Hi, Nimish, Sorry for my 10_Gbe_v2 typo. I meant ten_Gbe_v2. There is no 10_Gbe_v2 block. I am using the ten_Gbe_v2 block for ROACH2. Dave On Feb 13, 2013, at 2:47 PM, Nimish Sane wrote: Hi Dave and others, I was under the impression that ten_Gbe_v2 was suppose to support ROACH2 as well. At least, its mask supports parameters specific to ROACH2 (such as choice between CX4, SFP+ and ports etc.). Rather, these options are available in the mask only for ROACH2. Is 10_Gbe_v2 part of SKA-SA github repository now and is that supposed to be used instead of ten_Gbe_v2? I have not updated the repository of late and have been using ten_Gbe_v2 for my ROACH2 design. Thanks, Nimish On Wed, Feb 13, 2013 at 5:07 PM, David MacMahon dav...@astro.berkeley.edu wrote: Hi, John, My current work is on ROACH2 and the 10_Gbe_v2 block is the only 10 GbE block supported on ROACH2 so that's the one I'm currently designing with (but not yet using myself). I'm not sure which block Christopher used in the packet generator. His design ran on a ROACH so it could have been using the ten_GbE or ten_Gbe_v2 block. Dave On Feb 13, 2013, at 2:01 PM, John Ford wrote: Hi, John, The benchmarking tests were done here by Christopher Schollar who was visiting from the University of Cape Town. I'm not sure which library version he used, but I can find out. It was mostly just curiosity on my part, but are you using the same blocks for your work? John Dave On Feb 13, 2013, at 1:30 PM, John Ford wrote: Hi, Rich, We've sustained 9.9 Gpbs simultaneously from two 10 GbE ports (19.8 Gbps aggregate) on a ROACH into a PC with two 10 GbE NICs without dropping any packets. This was direct connect (i.e. no switch) and transmit only (i.e. no packets were sent back to the ROACH). Rich, Dave, can you tell us the versions of the libraries you have? Thanks! John Hope this helps, Dave On Feb 13, 2013, at 1:06 PM, Rich Lacasse wrote: Hi All, Using the latest 10 GbE yellow block, has anyone achieved a sustained data rate of greater than 8 Gb/s? Thanks, Rich
Re: [casper] CASPER 10 GbE yellow block
hi rich, we've transmitted 9.9 Gbit/sec on roach1, with the 10Gbe yellow block from the library about 4 months ago. i'm not sure if there have been revisions since then. best, dan On Wed, Feb 13, 2013 at 1:06 PM, Rich Lacasse rlaca...@nrao.edu wrote: Hi All, Using the latest 10 GbE yellow block, has anyone achieved a sustained data rate of greater than 8 Gb/s? Thanks, Rich
Re: [casper] parallel high bandwidtth FFT from synopsis/synphony
I hope we published it. Mel. On Wed, Feb 13, 2013 at 5:58 PM, Dan Werthimer d...@ssl.berkeley.edu wrote: these guys at synopsis re-invented and implemented casper's parameterized parallel FFT, (about a decade after we did): Using the Parallel FFT for Multigigahertz FPGA Signal Processing http://issuu.com/xcelljournal/docs/xcell_journal_issue_82/51?mode=windowviewMode=doublePage dan
Re: [casper] parallel high bandwidtth FFT from synopsis/synphony
We did: http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4840623 On Wed, Feb 13, 2013 at 7:45 PM, melvyn wright melvyn.wri...@gmail.comwrote: I hope we published it. Mel. On Wed, Feb 13, 2013 at 5:58 PM, Dan Werthimer d...@ssl.berkeley.edu wrote: these guys at synopsis re-invented and implemented casper's parameterized parallel FFT, (about a decade after we did): Using the Parallel FFT for Multigigahertz FPGA Signal Processing http://issuu.com/xcelljournal/docs/xcell_journal_issue_82/51?mode=windowviewMode=doublePage dan -- Aaron Parsons 510-306-4322 Hearst Field Annex B54, UCB
[casper] Virtex 6 BRAM timing
I've noticed that the Virtex 6 BRAM clock-to-output time is kind of long when NOT using the optional output register (Trcko_DO = 2.08 ns) as compared to using the optional output register (Trcko_DOA_REG = 0.75 ns). Using the optional output register adds an extra cycle of latency, but that extra 1.33 ns could be worth it especially since the optional output register uses (or at least sounds like it uses) dedicated flip-flops in the BRAM rather than CLB flip-flops. Setting a Xilinx Single Port RAM block's latency to 2 will enable use of the optional output register. Unfortunately, the CASPER Shared BRAM block does not have a latency setting (defaults to 1?) so the tools do not end up using the underlying BRAM's optional output register (even if there is a register on the Shared BRAM's output that could, in theory, be absorbed into the underlying BRAM). It would be good to add an optional Latency parameter to the Shared BRAM block and allow the user to select 1 (current value that does not use the BRAMs optional output register) or 2 (new value that does use the BRAM's optional output register). I think this would help ROACH2 designs meet timing more easily. I will look at this in more detail to see how involved it would be to add this feature. Once added, I think we'll want to set the default to 2. It would also be good to make sure the PPC side of the shared BRAM also uses the optional output registers. I think we should also be recommending that regular (i.e. non-yellow) BRAM blocks be set to a latency of 2 at a minimum. Maybe we are already? This also seems like an issue for ROACH as well (though the smaller chip seems faster to cross so maybe not so important as on ROACH2?). Dave
Re: [casper] Virtex 6 BRAM timing
Hi Dave It would be good to add an optional Latency parameter to the Shared BRAM block and allow the user to select 1 (current value that does not use the BRAMs optional output register) or 2 (new value that does use the BRAM's optional output register). I think this would help ROACH2 designs meet timing more easily. I will look at this in more detail to see how involved it would be to add this feature. Once added, I think we'll want to set the default to 2. It would also be good to make sure the PPC side of the shared BRAM also uses the optional output registers. This is on a (rather long) list of library upgrades we are keen on here in SA. Other BRAM related things are; * distributed RAM implementation option (for small, 32 bit BRAMs) * configurable initial values Not sure when we will get to them and would love it if others could help out. I think we should also be recommending that regular (i.e. non-yellow) BRAM blocks be set to a latency of 2 at a minimum. Maybe we are already? This also seems like an issue for ROACH as well (though the smaller chip seems faster to cross so maybe not so important as on ROACH2?). The majority of timing problems I have run across in ROACH are BRAM to multiplier paths. BRAM followed by logic often seems ok. Maybe a separate 'BRAM to multiplier latency' parameter with a default of 2? Would that be worth the effort? Probably not but worth looking at. Cheers Andrew