Re: [OpenWrt-Devel] [Cerowrt-devel] [RFC PATCH] packages: Smart Queue Management for AQM Packet Scheduling and Qos from CeroWrt
Sebastian , Dave, Thanks for these instructions. I just finished a round of 3 months testing this setup on a couple of MyNet N750 ( AR71xx based ) in several locations. The feedback from users is very positive and encouraging - Most of them say that they feel the networks behave a lot snappier than before [ 7.5Mbps Up/ 1Mbps Down]. All I had to do was install the 3 packages over BB 14.07 : ip-full from the OpenWRT BB repository and luci-app-sqm + sam-scripts from the CeroWRT repository. Will the later two packages ever make it to the main OpenWRT package list? or there other packages that on the main OpenWRT distribution that can be used to provide this functionality? ( the 2 CeroWRT packages above make really simple to setup ). Regards, Luis Garcia On Thu, Oct 9, 2014 at 2:59 PM, Sebastian Moeller moell...@gmx.de wrote: Hi Luis, On Oct 9, 2014, at 20:05 , Luis E. Garcia l...@bitamins.net wrote: Please excuse my dumb question or just point me to the right documentation. But is there an easy way to run AQM/Codel on BB 14.07 ?? I'm running BB on a MyNETN750 that looks like it meets the HW requirements. There probably is, I assume you need to install ip-full from the BB repository, and then copy http://snapon.lab.bufferbloat.net/~cero2/cerowrt/wndr/3.10.50-1/packages/luci-app-sqm_3-1_all.ipk and http://snapon.lab.bufferbloat.net/~cero2/cerowrt/wndr/3.10.50-1/packages/sqm-scripts_6-1_all.ipk onto your router and try: opkg install sqm-scripts_6-1_all.ipk and: opkg install luci-app-sqm_3-1_all.ipk You might need to play with opkg options, but in theory it should work… There probably also is a nicer way like adding a stanza similar to: src/gz toronto http://snapon.lab.bufferbloat.net/~cero2/cerowrt/wndr/3.10.50-1/packages to /etc/opkg.conf but I do not know the exact incantation of that, maybe hnyman, or Dave know the correct answer to this.. Best Regards Sebastian Regards, Luis Garcia On Thu, Oct 9, 2014 at 12:57 PM, Sebastian Moeller moell...@gmx.de wrote: Hi Richard, On Oct 9, 2014, at 18:42 , Richard Smith smithb...@gmail.com wrote: On 10/06/2014 03:41 PM, Sebastian Moeller wrote: I have spare routers that I can run OpenWRT or CeroWRt on and I'm setup to test with netperf, netperf-wrapper on my local network ( desktop - router - laptop ) it's Gbit so I can easily saturate the router. That sounds great. I think the first test should be to run SQM under cerowrt, so you get a feel of how things should look. I typically run netperf-wrapper rrul tests (for ipv4 and if available for ipv6) through cerowrt with different settings for SQM. A second step then is to instal SQM-scripts under openwrt and check whether the same settings produce the same results ;) Ok. I'll get my spare WNDR3700v2 up running the latest cerowrt. Should be in the next few days. I'm a bit strapped for time right now. No hurry, I really appreciate your help and there is no time pressure (rather the other way around, I do this for a hobby, and real life leaves almost no time for that ;) ) So just let me know what you are willing/ready to test and we will take it from there okay? (I would already be a happy camper if you could just install the current SQM-scripts on openwrt and just send me the output of “logread” after installing and activating SQM, as well as the output from “tc -d qdisc” before and after enabling SQM, and finally the output of running “/etc/init.d/sqm stop ; /etc/init.d/sqm start” on the router’s console; that hopefully works or at least gives some indication what might be off. If you could throw in a quick netperf-wrapper RRUL test through the router I will be most delighted ;)) Ok. That sounds like a good starter step. First question: Where are the current SQM scripts located and what's the process for installing them? Are they packages or stuff that I scp over and manually install? So the repository lives on: https://github.com/dtaht/ceropackages-3.10 I just followed Dave’s advice and used: git clone g...@github.com:dtaht/ceropackages-3.3.git to get my own version to play around with. Oh, SQM basically is the work of Dave (who wrote the scripts doing the actual AQMQOS work) and Toke you packaged all this nicely and created the GUI for it (all I ever did was shuffling the GUI around a bit). I just started hacking around the files from sqm-scripts and luck-app-sqm on my cerowrt (so I could immediately do some testing), so I never got around building actual packages (I assume you need to be setup to build whole openwrt/cerowrt images and that is outside the scope of my recreational coding), But that said if you copy the files from ceropackages-3.10/net/sqm-scripts/files to the matching directory on your router you should have a working sqm-script. Or just use a cerowrt built as SQM is standard and potentially
Re: [OpenWrt-Devel] [Cerowrt-devel] [RFC PATCH] packages: Smart Queue Management for AQM Packet Scheduling and Qos from CeroWrt
That's great news. Thanks again Dave - I found both SQM packages on the Chaos Calmer files. I'll run some more test with users. Regards, Luis On Thursday, February 12, 2015, Dave Taht dave.t...@gmail.com wrote: sqm-scripts entered the openwrt package repos for chaos calmer quite some time back. On Thu, Feb 12, 2015 at 6:02 PM, Luis E. Garcia l...@bitamins.net javascript:; wrote: Sebastian , Dave, Thanks for these instructions. I just finished a round of 3 months testing this setup on a couple of MyNet N750 ( AR71xx based ) in several locations. The feedback from users is very positive and encouraging - Most of them say that they feel the networks behave a lot snappier than before [ 7.5Mbps Up/ 1Mbps Down]. All I had to do was install the 3 packages over BB 14.07 : ip-full from the OpenWRT BB repository and luci-app-sqm + sam-scripts from the CeroWRT repository. Will the later two packages ever make it to the main OpenWRT package list? or there other packages that on the main OpenWRT distribution that can be used to provide this functionality? ( the 2 CeroWRT packages above make really simple to setup ). Regards, Luis Garcia On Thu, Oct 9, 2014 at 2:59 PM, Sebastian Moeller moell...@gmx.de javascript:; wrote: Hi Luis, On Oct 9, 2014, at 20:05 , Luis E. Garcia l...@bitamins.net javascript:; wrote: Please excuse my dumb question or just point me to the right documentation. But is there an easy way to run AQM/Codel on BB 14.07 ?? I'm running BB on a MyNETN750 that looks like it meets the HW requirements. There probably is, I assume you need to install ip-full from the BB repository, and then copy http://snapon.lab.bufferbloat.net/~cero2/cerowrt/wndr/3.10.50-1/packages/luci-app-sqm_3-1_all.ipk and http://snapon.lab.bufferbloat.net/~cero2/cerowrt/wndr/3.10.50-1/packages/sqm-scripts_6-1_all.ipk onto your router and try: opkg install sqm-scripts_6-1_all.ipk and: opkg install luci-app-sqm_3-1_all.ipk You might need to play with opkg options, but in theory it should work… There probably also is a nicer way like adding a stanza similar to: src/gz toronto http://snapon.lab.bufferbloat.net/~cero2/cerowrt/wndr/3.10.50-1/packages to /etc/opkg.conf but I do not know the exact incantation of that, maybe hnyman, or Dave know the correct answer to this.. Best Regards Sebastian Regards, Luis Garcia On Thu, Oct 9, 2014 at 12:57 PM, Sebastian Moeller moell...@gmx.de javascript:; wrote: Hi Richard, On Oct 9, 2014, at 18:42 , Richard Smith smithb...@gmail.com javascript:; wrote: On 10/06/2014 03:41 PM, Sebastian Moeller wrote: I have spare routers that I can run OpenWRT or CeroWRt on and I'm setup to test with netperf, netperf-wrapper on my local network ( desktop - router - laptop ) it's Gbit so I can easily saturate the router. That sounds great. I think the first test should be to run SQM under cerowrt, so you get a feel of how things should look. I typically run netperf-wrapper rrul tests (for ipv4 and if available for ipv6) through cerowrt with different settings for SQM. A second step then is to instal SQM-scripts under openwrt and check whether the same settings produce the same results ;) Ok. I'll get my spare WNDR3700v2 up running the latest cerowrt. Should be in the next few days. I'm a bit strapped for time right now. No hurry, I really appreciate your help and there is no time pressure (rather the other way around, I do this for a hobby, and real life leaves almost no time for that ;) ) So just let me know what you are willing/ready to test and we will take it from there okay? (I would already be a happy camper if you could just install the current SQM-scripts on openwrt and just send me the output of “logread” after installing and activating SQM, as well as the output from “tc -d qdisc” before and after enabling SQM, and finally the output of running “/etc/init.d/sqm stop ; /etc/init.d/sqm start” on the router’s console; that hopefully works or at least gives some indication what might be off. If you could throw in a quick netperf-wrapper RRUL test through the router I will be most delighted ;)) Ok. That sounds like a good starter step. First question: Where are the current SQM scripts located and what's the process for installing them? Are they packages or stuff that I scp over and manually install? So the repository lives on: https://github.com/dtaht/ceropackages-3.10 I just followed Dave’s advice and used: git clone g...@github.com:dtaht/ceropackages-3.3.git to get my own version to play around with. Oh, SQM basically is the work of Dave (who wrote the scripts doing the actual AQMQOS work) and Toke you packaged all this nicely and created
Re: [OpenWrt-Devel] miniupnpd state loss
Steve, Thanks for the fix. I had also experienced this issue. Do you know if it is possible to back-port the fix BB14.07?? Regards, Luis Garcia On Thu, Oct 16, 2014 at 2:20 PM, Kao Kevin kevin@technicolor.com wrote: Hi Steve, The solution works great. I have tested it. Thanks Kevin -Original Message- From: Steven Barth [mailto:cy...@openwrt.org] Sent: Thursday, October 16, 2014 4:48 AM To: openwrt-devel@lists.openwrt.org Cc: Kao Kevin Subject: Re: [OpenWrt-Devel] miniupnpd state loss Your change completely disables reloading which breaks in case of wan switching to a different interface or if lan interfaces are coming up. I commited a change https://github.com/openwrt-routing/packages/commit/8bc38fccc73b40e9599b897e335f2e7a5a67e879 which should avoid restarts in case wan interface flakes but stays the same. Please test and let me know if this solves your issue. Still miniunpd needs to be restarted in certain cases as noted above, this will cause loss of state however this problem must be resolved upstream as the program doesn't suppport a change of external or addition of internal interfaces without a restart. Cheers, Steven ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] DIR-657 - D-Link HD Media Router 1000
Ubicom support was removed a while back. That specific router is using a Ubicom IP7K chip. Within DLinks opensource files for this router there is a OpenWRT code base but I don't how old or what version it is. I have a small pet project to port OpenWRT to Ubicom but the more recent IP8K chpset. Luis Garcia On Mon, Sep 15, 2014 at 1:36 PM, Luis Soltero lsolt...@globalmarinenet.com wrote: Hello All, Looking for a router recommendation and wondering is someone might be able to help. The D-Link DIR-657 looks perfect but I don't see any support for this in OpenWRT. Anyone working on a port to the D-Link Media Routers? Looking for the following 128+M RAM 16+M Flash Built-in SD card 1+ USB 4 LAN 1 WAN 2.4Ghz WiFi 5.2Ghz WiFi 12V The built-in SD card support is critical. I appreciate any leads. Thanks, --luis -- Luis Soltero, Ph.D., MCS Director of Software Development, CTO Global Marine Networks, LLC StarPilot, LLC Tel: +1.865.379.8723 Fax: +1.865.681.5017 E-Mail: lsolt...@globalmarinenet.net Web: http://www.globalmarinenet.net Web: http://www.redportglobal.com Web: http://www.starpilotllc.com ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] DIR-657 - D-Link HD Media Router 1000
You might be able to get by with the WD MyNet N750. It has most of the requirements except the Built-in SD - but instead you get 2 USB ports. You could use the extra USB with an SD Reader. Luis G. On Mon, Sep 15, 2014 at 2:42 PM, Luis Soltero lsolt...@globalmarinenet.com wrote: Hello All, It would be nice to get Ubicom support into OpenWRT. However.. our specific requirements is for any router that meets the specifications listed below. 128+M RAM 16+M Flash Built-in SD card 1+ USB 4 LAN 1 WAN 2.4Ghz WiFi 5.2Ghz WiFi 12V Any recommendations would greatly appreciated. If no router meeting this specs is available then we will need to build one (probably using an MT7620a) although would really rather not go down this road. Suggestions are greatly appreciated. --luis On 9/15/14, 3:11 PM, Luis E. Garcia wrote: Ubicom support was removed a while back. That specific router is using a Ubicom IP7K chip. Within DLinks opensource files for this router there is a OpenWRT code base but I don't how old or what version it is. I have a small pet project to port OpenWRT to Ubicom but the more recent IP8K chpset. Luis Garcia On Mon, Sep 15, 2014 at 1:36 PM, Luis Soltero lsolt...@globalmarinenet.com wrote: Hello All, Looking for a router recommendation and wondering is someone might be able to help. The D-Link DIR-657 looks perfect but I don't see any support for this in OpenWRT. Anyone working on a port to the D-Link Media Routers? Looking for the following 128+M RAM 16+M Flash Built-in SD card 1+ USB 4 LAN 1 WAN 2.4Ghz WiFi 5.2Ghz WiFi 12V The built-in SD card support is critical. I appreciate any leads. Thanks, --luis -- Luis Soltero, Ph.D., MCS Director of Software Development, CTO Global Marine Networks, LLC StarPilot, LLC Tel: +1.865.379.8723 Fax: +1.865.681.5017 E-Mail: lsolt...@globalmarinenet.net Web: http://www.globalmarinenet.net Web: http://www.redportglobal.com Web: http://www.starpilotllc.com ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel -- Luis Soltero, Ph.D., MCS Director of Software Development, CTO Global Marine Networks, LLC StarPilot, LLC Tel: +1.865.379.8723 Fax: +1.865.681.5017 E-Mail: lsolt...@globalmarinenet.net Web: http://www.globalmarinenet.net Web: http://www.redportglobal.com Web: http://www.starpilotllc.com ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] dnsmasq 2.71 dies silently (hangs) / how to debug
I've noticed this issue with dnsmasq, I think that miniupnp is also having a similar issue where it dies silently after a few days. Can I apply the same method to debug miniupnp ?? Regards, Luis Garcia On Tue, Aug 19, 2014 at 10:45 AM, Bastian Bittorf bitt...@bluebottle.com wrote: * Steven Barth cy...@openwrt.org [19.08.2014 12:49]: procd_set_param limits core=unlimited thanks, this works fine here when the 'root' part of dnsmasq gets a -SIGSEGV (or during a real crash) and produces coredumps. i will keep the list updated, when i catch a real hang. bye, bastian ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
[OpenWrt-Devel] Unable to disable option in MenuConfig Compile Kernel with Debug Filesystem enabled
Hello, When I type make menuconfig, then select Global Build settings, then in there the is one compile kernel with debug filesystem enabled is marked as enabled by dependencies. Its permanently checked, how do I get it to uncheck, no matter what I do it will NOT un-check? What are it's dependencies? Also when selecting Remove unnecessary symbols from Kernel there is no noticeable change in the resulting image. Maybe I'm doing something wrong. What is best way to optimize an image for size and performance? I did google for this and tried the solutions only without much results while generation images for the DLink DIR-600 B2. Regards, Luis Garcia ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
[OpenWrt-Devel] Regarding USB and the Dlink DIR-600 Bx
Good evening, I just tried the BB RC2 build for this router and it seems the image is much larger than the one for AA. Checking the installed kernel modules I found the kmod-usb-core and an other related module. Is it possible to eliminate these modules for this router ( and probably for the DIR-300 Bx ) since enabling the USB support on them is near to imposible ( unless you can hack ball grid joints from the logic board). Since they're build using the ralink generic target - would I need to create a separate router file for these devices in order to avoid ruining other working devices that do nees USB support? The resulting image without USB support is 0.7 MB smaller and allow other ( more useful or better suited) packages to be installed. Regards, Luis Garcia ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] Any plans to support WD My Net N900 (Ubicom IP8K 600MHz, 8x GigE, 256MB)?
Here are the environment variables for U-Boot Ubicom =printenv bootcmd=bootm bootdelay=1 baudrate=115200 ethaddr=xx:xx:xx:xx:xx:xx ipaddr=192.168.1.1 serverip=192.168.1.254 bootfile=vmlinux.ub loadram=0xC100 bootargs=mtdparts=ubicom32fc.0:512k(bootloader)ro,14080k(upgrade),1536k(rootfs_data),256k(fw_env) mtdparts=mtdparts=ubicom32fc.0:512k(bootloader)ro,14080k(upgrade),1536k(rootfs_data),256k(fw_env) stdin=serial stdout=serial stderr=serial Environment size: 386/65532 bytes On Mon, Apr 7, 2014 at 8:30 AM, Luis E. Garcia l...@bitamins.net wrote: As Jose stated - we should be able to load both the u-boot and the custom images without bricking the device. U-Boot gives you the options right after it boots - you would need to connect to the serial port to be able to access them ( as of now). Boot output from N900. Taken from the JP4 Serial port from the motherboard. -- function : 1e1 Serial: Ubicom32 serdes uart serial driver initialized ba038000 u-boot started DRAM: 256 MB flash_init flash_init: we got MX25L128 Flash: 16 MB ubi32-nand-spi-er: switch at gpio 162 ubi32-nand-spi-er: found device id 2018 No SPI-ER NAND detected *** Warning - bad CRC, using default environment ipaddr = 192.168.1.1 U-Boot 2009.06-rc3 (Apr 21 2012 - 03:03:33) SVN revision: 2823 Target board: WRGND-ND15 In:serial Out: serial Err: serial Choose a operation or press SPACE to abort autoboot in 1 seconds 2: Load runtime code via TFTP. 9: Load boot loader code via TFTP. image addr = b008 ## Booting kernel from Legacy Image at b0080064 ... Image Name: WRG-ND15 Image Type: UBICOM32 Linux Multi-File Image (lzma compressed) Data Size:12779862 Bytes = 12.2 MB Load Address: c010 Entry Point: c010 Contents: Image 0: 33891 Bytes = 33.1 kB Image 1: 10424320 Bytes = 9.9 MB Image 2: 2321634 Bytes = 2.2 MB Verifying Checksum ... OK ## Loading init Ramdisk from multi component Legacy Image at b0080064 ... Uncompressing (LZMA) Multi-File Image ... OK Starting kernel ... Linux version 2.6.36+ (kevin@BSD2_6_BUILD_SEV) (gcc version 4.4.1 (GCC) ) #1 SMP Fri May 11 21:33:47 CST 2012 processor dram c010-d000, expecting c010-c200 processor dram c010-d000, expecting c010-c200 processor ocm bffc5200-bffedb00, expecting bffc5200-bffedb00 IP8K Processor, Ubicom, Inc. www.ubicom.com Device Tree: bfcc: sendirq=255, recvirq=255, name=NA bf7c: sendirq=255, recvirq=255, name=board bcec: sendirq=255, recvirq=255, name=bootargs bf30: sendirq=255, recvirq=255, name=gpio bc04: sendirq=255, recvirq=093, name=pciej bbb4: sendirq=255, recvirq=095, name=pciek bb3c: sendirq=255, recvirq=255, name=processor bc54: sendirq=255, recvirq=255, name=storage bca4: sendirq=255, recvirq=024, name=traps PERCPU: Embedded 2 pages/cpu @c085 s4808 r8192 d19768 u65536 pcpu-alloc: s4808 r8192 d19768 u65536 alloc=4*16384 pcpu-alloc: [0] 0 [0] 1 [0] 2 [0] 3 [0] 4 cpu[0]: bootstrap CPU online - tid: 6 Built 1 zonelists in Zone order, mobility grouping off. Total pages: 16367 Kernel command line: console=ttyUM0 console=ttyUS0 serdes=0xba038000,61,25000 root=/dev/mtdblock0 noinitrd PID hash table entries: 1024 (order: -2, 4096 bytes) Dentry cache hash table entries: 32768 (order: 3, 131072 bytes) Inode-cache hash table entries: 16384 (order: 2, 65536 bytes) Memory available: 253280k/261376k RAM, (4998k kernel code, 1547k data) Hierarchical RCU implementation. RCU-based detection of stalled CPUs is disabled. Verbose stalled-CPUs detection is disabled. NR_IRQS:96 timer[33]: timer-primary - created timer[33]: shutdown timer[33]: periodic: 600 cycles diff: 75970028, timer has lost 13 ticks [rounded up] console [ttyUS0] enabled Linux version 2.6.36+ (kevin@BSD2_6_BUILD_SEV) (gcc version 4.4.1 (GCC) ) #1 SMP Fri May 11 21:33:47 CST 2012 processor dram c010-d000, expecting c010-c200 processor dram c010-d000, expecting c010-c200 processor ocm bffc5200-bffedb00, expecting bffc5200-bffedb00 IP8K Processor, Ubicom, Inc. www.ubicom.com Device Tree: bfcc: sendirq=255, recvirq=255, name=NA bf7c: sendirq=255, recvirq=255, name=board bcec: sendirq=255, recvirq=255, name=bootargs bf30: sendirq=255, recvirq=255, name=gpio bc04: sendirq=255, recvirq=093, name=pciej bbb4: sendirq=255, recvirq=095, name=pciek bb3c: sendirq=255, recvirq=255, name=processor bc54: sendirq=255, recvirq=255, name=storage bca4: sendirq=255, recvirq=024, name=traps PERCPU: Embedded 2 pages/cpu @c085 s4808 r8192 d19768 u65536 pcpu-alloc: s4808 r8192 d19768
Re: [OpenWrt-Devel] Any plans to support WD My Net N900 (Ubicom IP8K 600MHz, 8x GigE, 256MB)?
Constantine, The JP1 headers are present on the N900 just as they are on the N750. I'll put a copy of the boot output on the console next week. The original firmware for these devices seems to be linux based ( the GPL code from the manufacturer ). I've used the recovery procedure to install my custom version of the original firmware on my N900. The WD MyNet range are really developer friendly and very hard to brick. Regards, Luis E. Garcia On Sat, Apr 5, 2014 at 2:16 PM, Constantine A. Murenin muren...@gmail.comwrote: Hi Luis, I'm also interested in trying out the waters by porting OpenWrt over. I've noticed reports that WD My Net routers appear to be developer-friendly -- according to the reports like https://wikidevi.com/wiki/Western_Digital_My_Net_N750, the models sold to end-users still do have the serial header present (JP1 on N750); and also have the special recovery procedure from a failed firmware update (http://wiki.openwrt.org/toh/wd/n900#recovery.procedure and http://wdc.custhelp.com/app/answers/detail/a_id/9618/~/how-to-recover-a-my-net-router-from-a-failed-firmware-update ). Do you at all know if the recovery procedure could be used to test out custom firmwares without the possibility of bricking the device? Cheers, Constantine. On 3 April 2014 14:31, Luis E. Garcia l...@bitamins.net wrote: Constantine, I'm looking into porting OpenWRT to the WD MyNet N900. Since the IP8K has an mmu it should be able to compile without too much issues. I'm currently trying to get the toolchain from trunk to work. Regards, Luis E. Garcia On Thu, Apr 3, 2014 at 3:55 AM, José Vázquez ppvazquez...@gmail.com wrote: 2014-04-03 6:58 GMT+02:00, Constantine A. Murenin muren...@gmail.com: Hello, It has come to my attention that the recently discontinued WD My Net line of dual-band routers is just about the best bang for the buck -- they're currently selling N900 (w/ 8x GigE and 256MB of RAM) for 49,99 USD at Staples, or 39,99 USD if you order it in person from the store with the 20% in-store coupon, off from the original price of 149,99 USD. This firesale appears to come and go since late 2013, even before the whole line was discontinued in early 2014 (for example, Staples still appears to have some N750 left, but it's not on sale this week, unlike N900). http://wiki.openwrt.org/toh/wd/n900 http://www.staples.com/WD-My-Net-N900-HD-Dual-Band-Router/product_937135 However, N900 it's not supported by OpenWRT -- the whole Ubicom IP8K platform is not supported. Is there a reason for that? What would it take to add the support? WD My Net N900 looks like just about the best piece of hardware over at http://wiki.openwrt.org/toh/start list, this repeated firesale price of 39,99 USD is just insane and is easily 3x less than the competition -- might be a good motivator for adding support to OpenWRT. Cheers, Constantine. ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel AFAIK ubicom32 was deprecated because it was hard to maintain, need a different toolchain, ip6k and ip5k had not mmu and some other things. IP8k seems to have mmu and some interesting features, but still has its own architecture. Take a look at this link to check the status: https://dev.openwrt.org/browser/branches/attitude_adjustment/target/linux/ubicom32 The GCC version is 4.4.7. Regards: Pepe ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] Any plans to support WD My Net N900 (Ubicom IP8K 600MHz, 8x GigE, 256MB)?
Constantine, I'm looking into porting OpenWRT to the WD MyNet N900. Since the IP8K has an mmu it should be able to compile without too much issues. I'm currently trying to get the toolchain from trunk to work. Regards, Luis E. Garcia On Thu, Apr 3, 2014 at 3:55 AM, José Vázquez ppvazquez...@gmail.com wrote: 2014-04-03 6:58 GMT+02:00, Constantine A. Murenin muren...@gmail.com: Hello, It has come to my attention that the recently discontinued WD My Net line of dual-band routers is just about the best bang for the buck -- they're currently selling N900 (w/ 8x GigE and 256MB of RAM) for 49,99 USD at Staples, or 39,99 USD if you order it in person from the store with the 20% in-store coupon, off from the original price of 149,99 USD. This firesale appears to come and go since late 2013, even before the whole line was discontinued in early 2014 (for example, Staples still appears to have some N750 left, but it's not on sale this week, unlike N900). http://wiki.openwrt.org/toh/wd/n900 http://www.staples.com/WD-My-Net-N900-HD-Dual-Band-Router/product_937135 However, N900 it's not supported by OpenWRT -- the whole Ubicom IP8K platform is not supported. Is there a reason for that? What would it take to add the support? WD My Net N900 looks like just about the best piece of hardware over at http://wiki.openwrt.org/toh/start list, this repeated firesale price of 39,99 USD is just insane and is easily 3x less than the competition -- might be a good motivator for adding support to OpenWRT. Cheers, Constantine. ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel AFAIK ubicom32 was deprecated because it was hard to maintain, need a different toolchain, ip6k and ip5k had not mmu and some other things. IP8k seems to have mmu and some interesting features, but still has its own architecture. Take a look at this link to check the status: https://dev.openwrt.org/browser/branches/attitude_adjustment/target/linux/ubicom32 The GCC version is 4.4.7. Regards: Pepe ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
[OpenWrt-Devel] Porting OpenWRT to Ubicom32 for the IP8K
I've done some research about OpenWRT and saw that thew Ubicom32 targets were removed a while back. It was my understanding that IP5K and IP7K didn't have a Memory Management Unit (MMU) and that caused this architecture to be not very maintainable. The IP8K does have a MMU and in that sense it would behave much like other processors used for other OpenWRT supported devices - fork() shouldn't be an issue since now we could do context switching and the main quirkiness would be how to manage the threading. In this case - what would be a good way for me to start trying to Add support for this processor to the Cross-Compilation configurations. I'm willing to maintain this target, but I need some guidance to help me get started. From the GPL code for the MyNet N900 I can see that the latest version they have for the Kernel is 2.6.36 - Is this a good starting point? or should I try to get a newer version Kernel to compile. The GPL code with the BootStrap published by Western Digital compiles cleanly and generates a working image - I was able to modify it and plug a DNS Leak on their implementation. Regards, Luis Garcia ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] Found the issue with the N750
Felix, The GPIOs in trunk are working correctly - they behave like the Original Firmware. When the Radio 0 (2.4GHz) is enabled the GPIOs for the LNA0 and LNA1 are set to HIGH and when the driver for the Radio 0 is disabled the GPIO's are set to LOW. The command: cat /sys/kernel/debug/gpio was very useful. For Radio1 (5 GHz) what is the way to test if the LNAs for that radio are enabled/disabled?? Luis On Wed, Feb 12, 2014 at 10:13 PM, Luis E. Garcia l...@bitamins.net wrote: It would seem that I get a bit better signal results with the patch. The computer I used to test this is a Mac Mini that is stationary about 6 meter away from the N750 (There are few concrete walls in the way). For the patch I'm using the code that you submitted on r39213: err = gpio_request_one(gpio, GPIOF_OUT_INIT_HIGH | GPIOF_EXPORT_DIR_FIXED, label); but on the file: /trunk/target/linux/ar71xx/patches-3.10/613-MIPS-ath79-add-ath79_wmac_setup_ext_lna_gpio-helper.patch The GPIOs status for both firmwares is always High|Out so my guess is that the driver is setting this - similar to what is seen in the original firmware. Results for TRUNK: # cat /sys/kernel/debug/gpio GPIOs 0-22, ath79: gpio-11 (wd:blue:wireless) out lo gpio-12 (wd:blue:internet) out lo gpio-13 (wd:blue:wps ) out lo gpio-14 (wd:blue:power ) out hi gpio-15 (external LNA0 ) out hi gpio-17 (Reset button) in hi gpio-18 (external LNA1 ) out hi gpio-19 (WPS button ) in hi # iw dev wlan0 station dump Station XX:XX:XX:XX:XX:61 (on wlan0) inactive time: 60 ms rx bytes: 990712 rx packets: 5310 tx bytes: 1877767 tx packets: 4214 tx retries: 1361 tx failed: 0 signal: -77 [-84, -79] dBm signal avg: -74 [-78, -77] dBm tx bitrate: 57.8 MBit/s MCS 5 short GI rx bitrate: 1.0 MBit/s authorized: yes authenticated: yes preamble: short WMM/WME: yes MFP: no TDLS peer: no == Results for TRUNK + Patch: # cat /sys/kernel/debug/gpio GPIOs 0-22, ath79: gpio-11 (wd:blue:wireless) out lo gpio-12 (wd:blue:internet) out hi gpio-13 (wd:blue:wps ) out lo gpio-14 (wd:blue:power ) out hi gpio-15 (external LNA0 ) out hi gpio-17 (Reset button) in hi gpio-18 (external LNA1 ) out hi gpio-19 (WPS button ) in hi # iw dev wlan0 station dump Station XX:XX:XX:XX:XX:61 (on wlan0) inactive time: 20 ms rx bytes: 623951 rx packets: 3515 tx bytes: 1005503 tx packets: 2255 tx retries: 774 tx failed: 0 signal: -72 [-74, -77] dBm signal avg: -73 [-75, -77] dBm tx bitrate: 52.0 MBit/s MCS 11 rx bitrate: 52.0 MBit/s MCS 11 authorized: yes authenticated: yes preamble: short WMM/WME: yes MFP: no TDLS peer: no ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] Found the issue with the N750
Felix, If I've understood correctly this fix should only be tested with Clients on the 2.4 GHz bands right? Let me know and I'll run the tests. Luis On Wed, Feb 12, 2014 at 3:40 AM, Felix Kaechele fe...@fetzig.org wrote: I cannot reproduce the issue. To test this it would be great if you could send me the output of iw dev wlan0 station dump from the device for both versions (i.e. with fix applied, without fix applied). Check that there is at least one station connected to the 2.4 GHz wmac and also make sure that you don't move this station between the two tests so we can get a direct comparison of signal strength. For me the signal strength to a Jolla phone from about 4m distance is about -44 dBm with LNAs enabled and below -68 dBm with LNAs disabled. The former is what I currently get with running trunk without any further patches applied. By the way: The 5 GHz band should be totally unaffected by this as it doesn't control it's LNAs via GPIO lines. Regards, Felix ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] Found the issue with the N750
Felix, Do you know of a way to check the status of the GPIOs from the command line? I tried: echo 15 /sys/class/gpio/export This returned an error: write error: Device or resource busy Luis On Wed, Feb 12, 2014 at 6:56 AM, Luis E. Garcia l...@bitamins.net wrote: Felix, If I've understood correctly this fix should only be tested with Clients on the 2.4 GHz bands right? Let me know and I'll run the tests. Luis On Wed, Feb 12, 2014 at 3:40 AM, Felix Kaechele fe...@fetzig.org wrote: I cannot reproduce the issue. To test this it would be great if you could send me the output of iw dev wlan0 station dump from the device for both versions (i.e. with fix applied, without fix applied). Check that there is at least one station connected to the 2.4 GHz wmac and also make sure that you don't move this station between the two tests so we can get a direct comparison of signal strength. For me the signal strength to a Jolla phone from about 4m distance is about -44 dBm with LNAs enabled and below -68 dBm with LNAs disabled. The former is what I currently get with running trunk without any further patches applied. By the way: The 5 GHz band should be totally unaffected by this as it doesn't control it's LNAs via GPIO lines. Regards, Felix ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] Found the issue with the N750
It would seem that I get a bit better signal results with the patch. The computer I used to test this is a Mac Mini that is stationary about 6 meter away from the N750 (There are few concrete walls in the way). For the patch I'm using the code that you submitted on r39213: err = gpio_request_one(gpio, GPIOF_OUT_INIT_HIGH | GPIOF_EXPORT_DIR_FIXED, label); but on the file: /trunk/target/linux/ar71xx/patches-3.10/613-MIPS-ath79-add-ath79_wmac_setup_ext_lna_gpio-helper.patch The GPIOs status for both firmwares is always High|Out so my guess is that the driver is setting this - similar to what is seen in the original firmware. Results for TRUNK: # cat /sys/kernel/debug/gpio GPIOs 0-22, ath79: gpio-11 (wd:blue:wireless) out lo gpio-12 (wd:blue:internet) out lo gpio-13 (wd:blue:wps ) out lo gpio-14 (wd:blue:power ) out hi gpio-15 (external LNA0 ) out hi gpio-17 (Reset button) in hi gpio-18 (external LNA1 ) out hi gpio-19 (WPS button ) in hi # iw dev wlan0 station dump Station XX:XX:XX:XX:XX:61 (on wlan0) inactive time: 60 ms rx bytes: 990712 rx packets: 5310 tx bytes: 1877767 tx packets: 4214 tx retries: 1361 tx failed: 0 signal: -77 [-84, -79] dBm signal avg: -74 [-78, -77] dBm tx bitrate: 57.8 MBit/s MCS 5 short GI rx bitrate: 1.0 MBit/s authorized: yes authenticated: yes preamble: short WMM/WME: yes MFP: no TDLS peer: no == Results for TRUNK + Patch: # cat /sys/kernel/debug/gpio GPIOs 0-22, ath79: gpio-11 (wd:blue:wireless) out lo gpio-12 (wd:blue:internet) out hi gpio-13 (wd:blue:wps ) out lo gpio-14 (wd:blue:power ) out hi gpio-15 (external LNA0 ) out hi gpio-17 (Reset button) in hi gpio-18 (external LNA1 ) out hi gpio-19 (WPS button ) in hi # iw dev wlan0 station dump Station XX:XX:XX:XX:XX:61 (on wlan0) inactive time: 20 ms rx bytes: 623951 rx packets: 3515 tx bytes: 1005503 tx packets: 2255 tx retries: 774 tx failed: 0 signal: -72 [-74, -77] dBm signal avg: -73 [-75, -77] dBm tx bitrate: 52.0 MBit/s MCS 11 rx bitrate: 52.0 MBit/s MCS 11 authorized: yes authenticated: yes preamble: short WMM/WME: yes MFP: no TDLS peer: no ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
[OpenWrt-Devel] Found the issue with the N750
Steve, Chirag, I found where the issue is in the code. Chirag - you're right the code in the patch submited by Gabor in the helper-patch is where we should be putting our corrections. File: https://dev.openwrt.org/browser/trunk/target/linux/ar71xx/patches-3.10/613-MIPS-ath79-add-ath79_wmac_setup_ext_lna_gpio-helper.patch Line 325: err = gpio_request_one(gpio, GPIOF_DIR_OUT | GPIOF_INIT_LOW, label); Replace with: err = gpio_request_one(gpio, GPIOF_DIR_OUT | GPIOF_INIT_HIGH, label); ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] MyNET N750 - Actually turn the LNA on.
Steve, Chirag, I'll be testing the signal strength on both bands this week ( I don't want to rush any results, while we wait for Gabor's answer). For the test I'll use 3 firmware images: - original WD N750 firmware (latest version). - r39542 with patch. - r39542 without patch. I'll take note of the signal strength in dB (RSI) and also do a short speedtest on a file transfer. Luis ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
[OpenWrt-Devel] MyNET N750 - Actually turn the LNA on.
Images found here: http://openwrt.crc.id.au/ Steve The code you first sent was merged with revision 39213 for the N750. If I use your image with this patch ( the one marked ) r39211 - I can get connected with both radios. LuCI responds in a timely fashion. There is no lag when applying changes. They cleaned up the code in revision r39217, I couldn't find the definition for the function they're using ath79_wmac_set_ext_lna_gpio But all builds after that have issues with the radios, poor performance on the 5GHz radios - LuCI lags on the page that setups the WiFi radios. With the image you have as r39499 with the patch - the router behaves as with the r39213 with patch image. ( the first fix you sent ) I wanted to verify the state of the GPIOs on the kernel under the running linux system - that is why I asked if there was a way to verify the State of the GPIOs in your blog. Luis Garcia ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] MyNET N750 - Actually turn the LNA on.
Chirag, Can you test the reception on the radios with the patch at r39213. Also can you check the behavior of LuCI with revision on the r394xx range. I get better reception on the 5GHz radio and a resposive LuCI [Configuration of radios WLan0 WLan1] with the changes made on patch r39213. Builds with revisons r39217 or later introduce lag on LuCI [Configuration of radios WLan0 WLan1] and the radio reception on the 5GHz is much lower in my tests. If you can point me in the right direction to check the state of the GPIOs from the console on the router I can test the images and see the status, and post back the results. Luis On Thu, Feb 6, 2014 at 4:46 PM, Chirag Chhatriwala cchha...@gmail.comwrote: Very strange. I don't have the problem of the router being deaf anymore. However further investigation into the code leads me to believe that the value is never set to GPIOF_INIT_HIGH. Which is confusing. Or am I not seeing the obvious? ath79_wmac_set_ext_lna_gpio calls ar934x_set_ext_lna_gpio calls gpio_request_one and sets to GPIOF_DIR_OUT | GPIOF_INIT_LOW for the particular LNA GPIO. I reckon we need this to be GPIOF_INIT_HIGH in order for this to be in line with Felix's original patch. What bothers me is that if this patch isn't working, how is my device able to hear other devices just fine. I also agree with Felix, that value being passed in ath79_wmac_set_ext_lna_gpio as 0 or 1 is simply cosmetic, to produce an output of LNA0 or LNA1. It IS then later used to select AR934X_GPIO_OUT_EXT_LNA0 or AR934X_GPIO_OUT_EXT_LNA1 based on the provided value. I got lost in the ath79_gpio_output_select (maybe this is where the magic happens?) Best Regards, Chirag PS I own 2 N750s and they run in WDS AP and WDS AP+CLIENT mode and the devices see each other just fine. On Thu, Feb 6, 2014 at 5:37 PM, Luis E. Garcia l...@bitamins.net wrote: Images found here: http://openwrt.crc.id.au/ Steve The code you first sent was merged with revision 39213 for the N750. If I use your image with this patch ( the one marked ) r39211 - I can get connected with both radios. LuCI responds in a timely fashion. There is no lag when applying changes. They cleaned up the code in revision r39217, I couldn't find the definition for the function they're using ath79_wmac_set_ext_lna_gpio But all builds after that have issues with the radios, poor performance on the 5GHz radios - LuCI lags on the page that setups the WiFi radios. With the image you have as r39499 with the patch - the router behaves as with the r39213 with patch image. ( the first fix you sent ) I wanted to verify the state of the GPIOs on the kernel under the running linux system - that is why I asked if there was a way to verify the State of the GPIOs in your blog. Luis Garcia ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] MyNET N750 - Actually turn the LNA on.
I'm copying Gabor so that he might help us. Chirag, I'm trying to read the status of the GPIOs by using the /sys/class/gpio/export root@OpenWrt:/# echo 15 /sys/class/gpio/export ash: write error: Device or resource busy This doesn’t work, as the system complains that resource is busy. I do have confirmation from another tester that the patch has now enabled the 5GHz radio on his N750 ( Check comment 25 26 on the following page: https://www.crc.id.au/2014/01/10/wd-mynet-n600-n750-openwrt-update/ ) Luis On Thu, Feb 6, 2014 at 4:59 PM, Luis E. Garcia l...@bitamins.net wrote: Chirag, Can you test the reception on the radios with the patch at r39213. Also can you check the behavior of LuCI with revision on the r394xx range. I get better reception on the 5GHz radio and a resposive LuCI [Configuration of radios WLan0 WLan1] with the changes made on patch r39213. Builds with revisons r39217 or later introduce lag on LuCI [Configuration of radios WLan0 WLan1] and the radio reception on the 5GHz is much lower in my tests. If you can point me in the right direction to check the state of the GPIOs from the console on the router I can test the images and see the status, and post back the results. Luis On Thu, Feb 6, 2014 at 4:46 PM, Chirag Chhatriwala cchha...@gmail.comwrote: Very strange. I don't have the problem of the router being deaf anymore. However further investigation into the code leads me to believe that the value is never set to GPIOF_INIT_HIGH. Which is confusing. Or am I not seeing the obvious? ath79_wmac_set_ext_lna_gpio calls ar934x_set_ext_lna_gpio calls gpio_request_one and sets to GPIOF_DIR_OUT | GPIOF_INIT_LOW for the particular LNA GPIO. I reckon we need this to be GPIOF_INIT_HIGH in order for this to be in line with Felix's original patch. What bothers me is that if this patch isn't working, how is my device able to hear other devices just fine. I also agree with Felix, that value being passed in ath79_wmac_set_ext_lna_gpio as 0 or 1 is simply cosmetic, to produce an output of LNA0 or LNA1. It IS then later used to select AR934X_GPIO_OUT_EXT_LNA0 or AR934X_GPIO_OUT_EXT_LNA1 based on the provided value. I got lost in the ath79_gpio_output_select (maybe this is where the magic happens?) Best Regards, Chirag PS I own 2 N750s and they run in WDS AP and WDS AP+CLIENT mode and the devices see each other just fine. On Thu, Feb 6, 2014 at 5:37 PM, Luis E. Garcia l...@bitamins.net wrote: Images found here: http://openwrt.crc.id.au/ Steve The code you first sent was merged with revision 39213 for the N750. If I use your image with this patch ( the one marked ) r39211 - I can get connected with both radios. LuCI responds in a timely fashion. There is no lag when applying changes. They cleaned up the code in revision r39217, I couldn't find the definition for the function they're using ath79_wmac_set_ext_lna_gpio But all builds after that have issues with the radios, poor performance on the 5GHz radios - LuCI lags on the page that setups the WiFi radios. With the image you have as r39499 with the patch - the router behaves as with the r39213 with patch image. ( the first fix you sent ) I wanted to verify the state of the GPIOs on the kernel under the running linux system - that is why I asked if there was a way to verify the State of the GPIOs in your blog. Luis Garcia ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel