| 00:05 | Scotty2Shoes | Does anyone know how I can track my order? |
| 00:07 | Scotty2Shoes | or if this means it will be shipped by today or expected to arrive today "Your order is now on track and is expected to ship to you by January 15th 2014. |
| 00:07 | Scotty2Shoes | " |
| 00:14 | _rmk_ | hmm, hdmi audio is... bubbly :( |
| 00:15 | _rmk_ | ... but spdif is fine |
| 00:15 | _rmk_ | 84: 0 0 0 0 GIC 84 spdif |
| 00:15 | _rmk_ | no spdif buffer interrupts :) |
| 00:31 | jhonny | hey everyone, i think i managed to root the cubox-i2ultra |
| 00:31 | jhonny | i can open up a terminal session and "su root" without issues |
| 00:32 | _rmk_ | maybe it doesn't have a password set? |
| 00:32 | jhonny | however when i run the SuperSU tool it starts, finds su, but then says it needs to update. After clicking ok it says it failed to update |
| 00:32 | jhonny | any good app that requires root to run that i can test ? |
| 00:33 | _rmk_ | what exactly do you want to test? |
| 00:34 | jhonny | if SuperSU is working correctly ? normally when you run an app that requires root SuperSU comes up and asks if you want to grant it permission, doesn't it ? |
| 00:34 | _rmk_ | sorry, I can't help with supersu, I've never used it |
| 01:06 | jhonny | guys now that i have root access, or so i think i do, how can i remount the filesystem to be read-write ? |
| 01:06 | jhonny | i tried the mount command but it says i don't have permission to do so |
| 02:42 | i4pro | i just got my cubox-i 4 pro today, threw the geexbox image on it, and noticed it was all 32 bit, i was wondering if there is an img out there already for the i4pro that is 64bit |
| 02:45 | Scotty2Shoes | I can't answer that, but I'd love to know. I should be getting mine soon |
| 02:47 | i4pro | was hoping to run xbmc on it, but for me, its not quite there yet.. since playback is not very smooth on bluray isos |
| 02:51 | jmontleon | The specific Cortex-A9 processor used in the Freescale i.MX6 is a 32 bit, configurable device with enough power to run sophisticated applications while keeping power consumption to a mobile device level. |
| 02:51 | jmontleon | http://cubox-i.com/cubox-i-core-architecture-freescale-i-mx6-arm-cortex-a9-introduction/ |
| 02:52 | Scotty2Shoes | Thankfully I don't have that many blurays. I mainly want XMBC and a torrent interface |
| 02:52 | i4pro | http://cubox-i.com/ |
| 02:52 | jmontleon | http://fedoraproject.org/wiki/Architectures/ARM/AArch64 |
| 02:52 | i4pro | so why does it say there that the i2, i2ultra, and i4pro are 64 bit? |
| 02:52 | i4pro | what am i missing? |
| 02:55 | Scotty2Shoes | the lack of customer support is starting to annoy me. I don't see how as a business the solid run people don't see that its important |
| 02:59 | jmontleon | don't know for sure i4pro. If I had to guess, something to do with the explanation in the 3rd paragraph here and therefore affecting performance: http://en.wikipedia.org/wiki/DDR_SDRAM. |
| 03:00 | jmontleon | if you google 32 vs. 64 bit ddr3 you can see some explanations |
| 03:04 | i4pro | http://cache.freescale.com/files/32bit/doc/data_sheet/IMX6DQCEC.pdf?fpsp=1&WT_TYPE=Data%20Sheets&WT_VENDOR=FREESCALE&WT_FILE_FORMAT=pdf&WT_ASSET=Documentation&Parent_nodeId=1337637154535695831062&Parent_pageType=product |
| 03:06 | Scotty2Shoes | So is just 64bit ram but its rate limited at the processor? |
| 03:06 | i4pro | weird i totally missed that... I have no idea how that would be handled then, seems odd to xfer 64bit from memory to 32bit cou |
| 03:06 | i4pro | apparently so |
| 03:06 | Scotty2Shoes | Like the 1000mbps ethernet gets rate limited to 470 |
| 03:07 | i4pro | yeah not sure i fully understand that one either, but 470 is fast enough |
| 03:08 | Scotty2Shoes | yeah I have fiber, but 470 is still fast enough |
| 03:09 | Scotty2Shoes | I just can't wait to get mine. BTW what power supply is needed |
| 03:09 | i4pro | im using the one they sent (i ordered with it) |
| 03:10 | i4pro | i think it depends on if your using usb devices and such |
| 03:10 | Scotty2Shoes | nvm I see its 110v. Yeah I skimped on that because I was being cheap. I shoul dhave one lying around |
| 03:10 | Scotty2Shoes | just should of paid the convience cost |
| 03:10 | jmontleon | i4pro, I think you're confusing 64 bit memory addresses with a unit for data transfer. And since there are only 2GB of RAM in the cubox-i at most, I don't see what 64-bit addressing would get you. You can't have the more than 4gb of ram that would be needed for it to become an issue. |
| 03:11 | i4pro | yeah. i misread that initially, im betting a lot of other people did to.. i thought the i4pro was a 64bit cpu, that could handle 64bit instructions |
| 03:12 | Scotty2Shoes | its definitely the assumption most people will make |
| 03:13 | jmontleon | and 480Mbps sounds like USB 2.0 limit. It's probably on the USB bus. something along those lines. again, don't know for sure, though that PDF you link might show it in the diagram and make it clear |
| 03:15 | i4pro | im not too worried about the network speeds... they should be fine for anything, i have 100M boxes that can stream stuff just find |
| 03:15 | i4pro | might have to look at turning on jumbo frames though |
| 03:15 | iio7 | Is it possible to have cubox power down/spin down an idle usb hd? |
| 03:20 | jmontleon | iio7, don't see why not; google how to spin it down with sdparm... |
| 03:21 | jmontleon | bbl |
| 03:45 | bigjohnx | Anyone know that how to change mubox (hummingboard) that running on cubox-i2u? |
| 03:45 | bigjohnx | I tried changed the uboot and finally cannot boot to rootfs |
| 04:31 | Scotty2Shoes | So can anyone tell me how many amps the stock power supply is? |
| 04:31 | Scotty2Shoes | I know its 110v |
| 04:36 | shakesoda | I think input was 110/0.6a and output was 5v/3A |
| 04:36 | shakesoda | let me check |
| 04:36 | shakesoda | yep |
| 04:51 | iio7 | I changed my bootcmd and issued a "save", but when I disconnected the cubox and reattached to power the bootcmd was set back to default. |
| 04:51 | iio7 | How can I flash it so it doesn't change on power off? |
| 04:59 | iio7 | savenev is supposed to "save environment variables to persistent storage", but once plugged of power and re-plugged the env variables are back to defaults. |
| 05:28 | durandle | Hmm thats odd. Streaming 1080p video over wifi (Wireless N at 300mbps). the wideo stutters a fair bit, yet XBMC (Running under GeexBox) reports that the buffer is always full (so its not network lag). I thought the cubox was supposed to have good 1080p video decoding? |
| 08:31 | jnettlet | jmontleo, the ethernet is not on the USB 2.0 bus. The speed limitation is due to other limitations to how it is connected. In reality I have perf tested it up to 540Mbs receiving and 380Mbps transmitting. |
| 12:34 | _rmk | 12:34 * _rmk_ sighs... yet more lockdep warnings from 3.13 from the fsl specific code |
| 13:02 | _rmk_ | getting the bluetooth/wifi working is going to be a challenge too |
| 13:06 | hste_ | _rmk_: don't u like a challenge :) |
| 13:07 | _rmk_ | hste: the problem is going to be, how to deal with those GPIOs to power up and reset the device before the MMC stuff gets probed. |
| 13:07 | _rmk_ | to me that sounds like having to hack up the device tree code to prevent the MMC host device being created... which ain't going to be suitable for mainline |
| 13:08 | _rmk_ | or hacking something into the sdhc host controller driver to fiddle with those gpios - again, not suitable for mainline |
| 13:12 | hste_ | why does they need to powered up and reset before MMC? |
| 13:15 | _rmk_ | because it's a sdio device, and so we need it to be visible for the interface to probe it |
| 13:16 | _rmk_ | hmm, I think I can use a regulator for the power gpios, but that still presents a problem for the reset gpios |
| 13:19 | _rmk_ | now, I wonder what voltage gets supplied to the wifi/bt chip |
| 13:19 | _rmk_ | dt is annoying in that it needs this kind of detail :( |
| 13:20 | _rmk_ | I guess I need rabeeh for that |
| 14:34 | iio7 | I used "saveenv" to store a new bootcmd and some bootargs, but cubox doesn't retain the setup after a power down. Is it possible to have it retain the setup? |
| 14:50 | psergy | Hi All, |
| 14:50 | psergy | Can I ask about shipping ETA here? |
| 14:51 | psergy | How long it will take if I order CuBox-i4-Pro today? |
| 14:55 | _rmk_ | You need to talk to solidrun about that, probably the fastest answer will be by mailing their customer service people. |
| 14:57 | psergy | Is there email address or I can use contact form http://cubox-i.com/contact-us/ ? |
| 14:58 | psergy | OK. Found email address [email protected] |
| 15:05 | iio7 | Is it not possible to retain a bootcmd after power off? |
| 15:19 | iio7 | Nobody who knows this and who can give a pointer as to have to do it? |
| 15:21 | iio7 | have - how |
| 15:21 | _rmk_ | iio7: please be patient; I'm sure someone knows the answer, but they're probably busy doing other stuff. Many people here are volunteers who give their time for free. |
| 15:22 | iio7 | _rmk_, thanks, I'll try again later :) |
| 15:23 | _rmk_ | it's best if you can stay on irc and wait for a reply rather than disconnecting and returning. |
| 15:23 | jnettlet | _rmk_, did you have a problem doing cd detection via gpio's which is why you switched to the controller DAT3 method? |
| 15:23 | _rmk_ | jnettlet: the dat3 method is force-enabled in the driver as some kind of bug work-around |
| 15:24 | jnettlet | _rmk_, yeah but all the comments from freescale on their community site say that the GPIO method is preferred |
| 15:25 | jnettlet | which I understand because I am still seeing the DAT3 method spewing interrupts in anything but 1-bit mode. |
| 15:26 | _rmk_ | I wonder if it spews interrupts because it's seeing DAT3 toggle... which would mean that the workaround is buggy |
| 15:26 | _rmk_ | and I still don't understand why I don't see this problem |
| 15:26 | iio7 | Found the solution finally, the box has to be reflashed with a modified u-boot. |
| 15:27 | _rmk_ | on the cubox-i4, I get: |
| 15:27 | _rmk_ | 75: 49 0 0 0 GIC 75 ci_hdrc_imx |
| 15:27 | _rmk_ | wrong one |
| 15:27 | _rmk_ | 55: 1874 0 0 0 GIC 55 mmc1 |
| 15:27 | jnettlet | and I get. |
| 15:27 | jnettlet | 55: 165719 0 0 0 GIC mmc0 |
| 15:27 | _rmk_ | and that's after cat'ing all the files on the card |
| 15:28 | jnettlet | and that is just after me booting |
| 15:29 | _rmk_ | the other thing could be there's a difference between the v3.10 stuff you're using and v3.12/13 kernels here |
| 15:30 | jnettlet | nah, I have tried the 3.12 kernel and saw the same thing |
| 15:36 | _rmk_ | does the debug in /sys/kernel/debug/mmc0/ help? |
| 15:36 | _rmk_ | I see: |
| 15:36 | _rmk_ | clock: 50000000 Hz |
| 15:36 | _rmk_ | actual clock: 49500000 Hz |
| 15:36 | _rmk_ | vdd: 21 (3.3 ~ 3.4 V) |
| 15:36 | _rmk_ | bus mode: 2 (push-pull) |
| 15:36 | _rmk_ | chip select: 0 (don't care) |
| 15:37 | _rmk_ | power mode: 2 (on) |
| 15:37 | _rmk_ | bus width: 2 (4 bits) |
| 15:37 | _rmk_ | timing spec: 2 (sd high-speed) |
| 15:37 | _rmk_ | signal voltage: 0 (3.30 V) |
| 15:41 | jnettlet | clock: 50000000 Hz |
| 15:41 | jnettlet | actual clock: 49500000 Hz |
| 15:41 | jnettlet | vdd: 17 (2.9 ~ 3.0 V) |
| 15:41 | jnettlet | bus mode: 2 (push-pull) |
| 15:41 | jnettlet | chip select: 0 (don't care) |
| 15:41 | jnettlet | power mode: 2 (on) |
| 15:41 | jnettlet | bus width: 2 (4 bits) |
| 15:41 | jnettlet | timing spec: 2 (sd high-speed) |
| 15:41 | jnettlet | signal voltage: 0 (3.30 V) |
| 15:41 | jnettlet | vdd is different. |
| 15:42 | jmontleo | jnettlet, thanks for the clarification. I looked later in the evening, but I didn't see anything obvious to me. It was just a guess based on the speed they listed |
| 15:46 | jnettlet | _rmk_, so it looks like the 3.0.35 sources have a patch that explicitely disable the D3CD detection if it is an imx6 chip |
| 15:49 | jnettlet | what we really need to do is say if we have a gpio-cd setup then don't enable D3CD |
| 16:25 | rabeeh | _rmk_: is your SPL u-boot working fine all the way? |
| 16:25 | rabeeh | you have used hardfp toolchain - right? (the oldie 3.4.x) |
| 16:26 | _rmk_ | yes, my uboot is fine on the cubox-i and hummingboard |
| 16:31 | _rmk_ | just setting the sata and i2c rtc sorted |
| 16:32 | _rmk_ | is the pcf8523 fitted on the cubox-i, or is that just for the hummingboard? |
| 16:33 | rabeeh | pcf8523 is assembly option; it's valid in cubox-i2u and i4pro |
| 16:34 | _rmk_ | but at the moment it's unpopulated? |
| 16:34 | _rmk_ | if I declare that it's present in DT, I get: |
| 16:34 | _rmk_ | rtc-pcf8523: probe of 0-0068 failed with error -5 |
| 16:34 | _rmk_ | on the i4pro |
| 16:34 | rabeeh | https://github.com/SolidRun/linux-imx6/blob/imx_3.0.35_4.1.0/arch/arm/mach-mx6/board-mx6q_cubox-i.c#L111 |
| 16:34 | rabeeh | notice that it's on i2c3 on cubox-i |
| 16:35 | rabeeh | if you do i2cdetect you should see the device on 0x68 |
| 16:36 | rabeeh | and if you dump the content of the device few times you should see the seconds parameter rolling |
| 16:37 | i4pro | rabeeh: with geexbox, do you know what causes this in the xbmc logs? ERROR: GetPicture - player is ahead of time |
| 16:38 | rabeeh | i4pro; nop |
| 16:38 | rabeeh | what kind of content is this? what is the bitrate? |
| 16:39 | rabeeh | can you click on 'o' in playback and send the parameters out? |
| 16:39 | _rmk_ | hmm, I guess it found it that time... |
| 16:39 | _rmk_ | rtc-pcf8523 2-0068: rtc core: registered rtc-pcf8523 as rtc0 |
| 16:39 | _rmk_ | rtc-pcf8523 2-0068: setting system clock to 2000-01-02 22:45:53 UTC (946853153) |
| 16:39 | rabeeh | yes :) |
| 16:40 | rabeeh | you need to initialize it :) |
| 16:40 | _rmk_ | that's easy... |
| 16:40 | _rmk_ | root@cubox-i4:~# hwclock -uw |
| 16:40 | rabeeh | how many rtc clocks you have? |
| 16:40 | _rmk_ | two |
| 16:40 | _rmk_ | snvs_rtc 20cc034.snvs-rtc-lp: rtc core: registered 20cc034.snvs-rtc-lp as rtc1 |
| 16:41 | rabeeh | ok; so rtc0 is the pcf then |
| 16:41 | _rmk_ | and the pcf8523 came up as rtc0, so that's the default |
| 16:41 | rabeeh | in LK 3.0.35 it's the way around |
| 16:41 | rabeeh | and yes causes lots of headache |
| 16:42 | _rmk_ | I'm also getting for sata: |
| 16:42 | _rmk_ | ahci ahci: SSS flag set, parallel bus scan disabled |
| 16:42 | _rmk_ | ahci ahci: AHCI 0001.0300 32 slots 1 ports 3 Gbps 0x1 impl platform mode |
| 16:42 | _rmk_ | ahci ahci: flags: ncq sntf stag pm led clo only pmp pio slum part ccc apst |
| 16:42 | _rmk_ | scsi0 : ahci_platform |
| 16:42 | _rmk_ | ata1: SATA max UDMA/133 mmio [mem 0x02200000-0x02203fff] port 0x100 irq 71 |
| 16:42 | _rmk_ | which looks good - I need to connect my esata drive to it to check that it actually works |
| 16:42 | rabeeh | please try it; lets see if the default analog parameters for the upstream kernel is good enough |
| 16:43 | rabeeh | LK 3.0.35 removes the SATA interface if it has booted with nothing connected |
| 16:43 | _rmk_ | if it's marginal, my esata enclosure should show that quite well... because they didn't put an esata receiver chip on the interface |
| 16:43 | rabeeh | which is a typical configuration for mobile (to spare the few tens of mW waiting for connection) |
| 16:44 | rabeeh | is it a single drive? |
| 16:44 | rabeeh | multi drive? |
| 16:44 | rabeeh | notice that the i.mx6 supports port multiplier too (command based switching) |
| 16:45 | _rmk_ | yes, single drive - the esata tracks go from a usb-to-esata device, to the esata socket, and then to the drive. |
| 16:46 | _rmk_ | I've been looking for another esata enclosure, but all I can find is nasty cheap stuff which does similar things |
| 16:47 | _rmk_ | fyi - http://www.amazon.co.uk/gp/product/B003832EIC/sr=8-6/qid=1389634214/ref=olp_product_details?ie=UTF8&me=&qid=1389634214&seller=&sr=8-6 |
| 16:48 | _rmk_ | that's what I have on the dove cubox with a WD800 drive, which seems to work fine provided you don't want to use the USB (which is flakey as hell) |
| 16:48 | i4pro | rabeeh: its a bluray iso |
| 16:49 | rabeeh | _rmk_: wow; it has the feature 'HDD capacity unlimited' |
| 16:49 | rabeeh | i4pro - full rip? |
| 16:49 | rabeeh | how big the file? |
| 16:49 | _rmk_ | and if you wrap the cubox end of the esata cable it comes with in tin foil, if you have a logitec keyboard/mouse receiver plugged into the USB |
| 16:49 | rabeeh | streaming or through a usb stick / drive? |
| 16:49 | i4pro | full iso' |
| 16:50 | rabeeh | i4pro: and i'm assuming that you really have i4pro :) |
| 16:50 | _rmk_ | rabeeh: yes, the phrase "HDD capacity unlimited" makes you wonder how you can put twenty 3.5" drives into such a small box :) |
| 16:50 | i4pro | yes |
| 16:50 | i4pro | O |
| 16:50 | rabeeh | i4pro: now what do you see? i mean how long does it play? does it hang? stutter? |
| 16:51 | i4pro | rabeeh: with the O option what did you want? |
| 16:51 | i4pro | stutter |
| 16:51 | i4pro | and pixalation on some sceens |
| 16:51 | rabeeh | i4pro: in playback you can click 'o' letter and get statistics |
| 16:51 | rabeeh | which image was that? |
| 16:52 | i4pro | D( Audio: ac3 (AC -3 /0x332d4341),48000 Hz, 5.1 (side), fltp, 640 kb/s) |
| 16:53 | i4pro | p(aq:99% Kb/s:624.61, att:-270.6 dB) |
| 16:54 | i4pro | D(Video:vc1 (Advanced) VC-1 /0x312D4356),youv420p, 1920x1080 [SAR 1:! DAR 16:9] |
| 16:54 | rabeeh | oh; it's a vc1 codec !!!! |
| 16:54 | rabeeh | please continue |
| 16:55 | i4pro | P(fr:23.976, vq:88%, dc:iMX-vc1, MB/s:5.37, drop:648, skip:2 pc:none) (MB/s and drop are variable and changed depending on sceen) |
| 16:57 | i4pro | C(ad:0.000, a/v:0.0??, edi:-,dcpu:0% acpu 0% vcpu:0% cache0 B 96%) ?? is varies quiclky but looks like .012 - .083 |
| 16:58 | rabeeh | so a/v is quite in sync; which is great |
| 16:58 | i4pro | W(fps:19.46 CPU0 37.5 CPU1 12.0% CPU2 7.7% cpu3 15%) CPUs all vary |
| 16:59 | rabeeh | vc-1 codec was a bit of a surprise; i haven't seen bluray vc-1 quite a while |
| 16:59 | rabeeh | cpus as expected are resting |
| 17:00 | i4pro | yeah not to bad, just really jerky watching it |
| 17:01 | i4pro | and like when things blow up, i guess fast motion,things go pixelated, like an poor signal for satelite tv or something |
| 17:01 | i4pro | I can try some other BR.isos |
| 17:01 | i4pro | i have over 200 |
| 17:02 | rabeeh | try the ones with h.264 |
| 17:02 | rabeeh | those were probably most tested |
| 17:02 | rabeeh | and do you stream? |
| 17:04 | i4pro | tried one with h264, the audio was jacked |
| 17:04 | i4pro | yes from my NFS server |
| 17:04 | i4pro | but i can play all of these just fine on an older atom box |
| 17:07 | i4pro | so have one playing now... the audio messup was kinda wierd.. have the hdmi going through the tv, passthrough to a sound bar, the tv audio was all scratchy, on both the dts-hd ma, and dts 5.1 (handt seen that on other xbmc boxes) but gessing that is a setting somewhere |
| 17:08 | i4pro | this one is a bit off also, video h264 (high) (HDMV / 0x564D4448), yuv 420p |
| 17:29 | _rmk_ | rabeeh: does the partitioning of the microsom pinmux configuration from the hummingboard/cubox-i in your github repository reflect the actual hardware? |
| 17:29 | i4pro | rabeeh: h264 is also a jittery |
| 17:33 | xraxor_ | rabeeh: with the latest geexbox snapshot plaback seemed better, maybe worth a try for i4pro to test it |
| 17:39 | xraxor_ | some builds playback was really bad for me |
| 17:42 | i4pro | rabeeh: so when watching a BR ripped to MKV, playback is really pretty good... and i see no errors in the xbmc log file, then when i watch a BR ISO, i have jitters, some pixelation, and just kinda umm, dropped video frames, and the log starts filling with ERRORs http://pastebin.com/p33ZhDzT |
| 17:43 | rabeeh | _rmk_: yes. it reflects the hardware |
| 17:44 | jnettlet | i4pro, as there a sample video I can download and look at? |
| 17:44 | _rmk_ | rabeeh: ok, I'm updating the DT files to follow that partitioning as well - the only one which I question is the CEC line - what if it gets used as a gpio in some future change? |
| 17:45 | _rmk_ | same actually goes for i2c2 as well |
| 17:46 | rabeeh | _rmk_: cec true |
| 17:46 | rabeeh | i2c i'm not sure since they are pulled to 5V (HDMI standard) |
| 17:46 | _rmk_ | ah, those pullups are on the microsom? |
| 17:46 | rabeeh | well.. but on the other hand they are general purpose I2C that is just pulled up to 5V |
| 17:46 | rabeeh | yes; on the microsom |
| 17:47 | i4pro | jnettlet: the issue seems to show up when watching BR ISOs, Im not sure how to make a sample of that, and a BR iso, is like 26 - 40 Gig |
| 17:48 | jnettlet | sintel has a blueray iso image. I can test with that. |
| 17:48 | i4pro | If you have a BR you could make an ISO that it will play using anydvdhd |
| 17:48 | _rmk_ | rabeeh: btw, on the hummingboard, the 26 pin connector, does that carry lots of gpio (maybe even some i2c capable?) |
| 17:49 | rabeeh | _rmk_: yes |
| 17:49 | i4pro | you can get anydvdhd demo for like 14 days... |
| 17:49 | rabeeh | lots of gpios that can be confirued to lots of things; beyond gpio |
| 17:49 | i4pro | ive tried with some that are VC-1 and 264 |
| 17:50 | rabeeh | hd-dvd is a MS driven spec that is mainly VC-1 |
| 17:50 | i4pro | jnettlet: do you know what iso he has? I can see if i have the same just to verify it does it with that one |
| 17:50 | _rmk_ | if it's capable of two (or more) hardware-driven i2c buses via that 26 pin connector, I'll be interested in a non-dev HB. |
| 17:51 | rabee | 17:51 * rabeeh gets curious |
| 17:51 | jnettlet | i4pro, :-D sintel is an OSS movie made by the Blender team. Here is the link http://download.blender.org/durian/movies/ |
| 17:52 | _rmk_ | my boot test system, which generates nice graphs of power etc - that's all driven by a Versatile (ARM926) board bit-banging two GPIO buses to talk to 5 I2C devices per bus - four are INA226 devices, and one is a gpio expander which switches the power regulators on/off. |
| 17:52 | i4pro | ok, downloading now... |
| 17:53 | _rmk_ | bit-banging i2c buses isn't exactly cheap, especially on versatile hardware |
| 17:53 | nikitis | I finally got a tracking number for my order, but where do I enter it in? They did not say. |
| 17:53 | _rmk_ | plus, the versatile is a tad large to fix in the 19" box I have |
| 17:54 | i4pro | nikitis: you can use the usps page |
| 17:55 | nikitis | ok |
| 17:55 | _rmk_ | rabeeh: it's the silver 19" rack box just above the black terminal server in this post: https://plus.google.com/u/1/103895730806848715870/posts/cCq2CxDogW9 |
| 17:56 | jnettlet | looks like you could use a rack for your rack boxes |
| 17:58 | _rmk_ | jnettlet: that's... something I need to look into |
| 18:00 | _rmk_ | inside that silver fronted case are two of these boards (but cleaned up!) - http://www.home.arm.linux.org.uk/~rmk/IMG_1307.JPG |
| 18:01 | _rmk_ | the ina226's are U2/U12/U22/U32 which do all the voltage/current/power monitoring |
| 18:01 | _rmk_ | U51 is the i2c gpio expander which controls the regulators |
| 18:02 | _rmk_ | each one can produce up to about 4A at a (preset) 5V up to 12V with a 24-36V supply in. |
| 18:03 | _rmk_ | I wired the INA226s so they can operate as current trips as well, so I can program an over-current cut-off. |
| 18:04 | rabeeh | _rmk_: this board looks seriously reworked |
| 18:04 | _rmk_ | I decided in the end to toggle them between over-current and over-voltage detection |
| 18:04 | rabeeh | c15 broke? |
| 18:04 | _rmk_ | rabeeh: it was hand assembled :) |
| 18:07 | _rmk_ | rabeeh: I think that's a combination of the stupidly powerful flash and flux... that's the boost cap for the mosfet inside the chip |
| 18:08 | rabeeh | _rmk_: which capacitors feeds the LTCYM dc-dc? |
| 18:08 | rabeeh | which one should connect to pin 4? |
| 18:08 | rabeeh | and please don't say this is C13 in the middle dc-dc |
| 18:09 | rabeeh | or C2 or C24 |
| 18:09 | rabeeh | or C32 |
| 18:10 | _rmk_ | C13 is the feedback compensation |
| 18:10 | _rmk_ | pin 4 (supply) is C11/C12 (10u ceramic) C42 (47u) |
| 18:11 | _rmk_ | LTCYM is LT3680 btw |
| 18:14 | rabeeh | _rmk_: C12 looks to me like C13 |
| 18:14 | rabeeh | anyhow; those capacitors are way off |
| 18:15 | rabeeh | it's something like 10mm away from the Vin |
| 18:15 | _rmk_ | C12 is just below D11 |
| 18:16 | rabeeh | yeah; i noticed that |
| 18:16 | rabeeh | C12 and C11 which is probably the invisible capacitor in the picture are far away from the dc-dc |
| 18:17 | i4pro | jnettlet: ok the sintel movie is doing it to, you can see the error messages in the log file.. since its animated it a lot harder to tell though by watching the movie |
| 18:17 | rabeeh | when the transistor in the dc-dc opens it can draw very high current that can generate high voltages |
| 18:18 | _rmk_ | rabeeh: they're more like 5mm away |
| 18:19 | rabeeh | oh; it's a 3.5mm package |
| 18:19 | rabeeh | now i get the dimensions |
| 18:19 | i4pro | jnettlet: around 2:38 in sintel i start getting a lot of the errors in the xbmc log |
| 18:19 | _rmk_ | the critical path is the C11/C12, U2, L11, C16/C17 loop |
| 18:19 | _rmk_ | erm, U12 |
| 18:20 | _rmk_ | which is where the bulk of the current flow is when the switch is on |
| 18:21 | _rmk_ | moving C11/C12 closer to U12 would just move them further away from C16/C17 |
| 18:21 | _rmk_ | I played about with the positioning of them quite a bit before settling with that layout |
| 18:23 | _rmk_ | the package is a MSOP10... package body is 3x3mm |
| 18:23 | rabeeh | _rmk_: why do you care about c16/c17? they are second order filters |
| 18:24 | rabeeh | you need to take care of the high di/dt |
| 18:24 | rabeeh | so you need minimal inductance for the next power supplier (c11/c22) |
| 18:25 | rabeeh | i have never seen an application note or any simultion that refers to the second order capacitors |
| 18:25 | _rmk_ | yes, c16/c17 are the output caps - yes, they're not as stressed as the input caps but they're still part of the loop which has to be minimised |
| 18:33 | _rmk_ | however, the other loop which needs to be considered is the c11/c12/u12/d11 loop - and my loop is smaller than that in the data sheet for this device quite simply because I have the diode closer to u12. |
| 18:35 | _rmk_ | in the data sheet, the equivalent of d11 is about 10mm away from pin 3. |
| 18:37 | _rmk_ | and before you start asking about switching frequency, it's 200kHz which is at the low end. |
| 18:38 | _rmk_ | it's been running for two years now, and if there was a problem there, it would've failed by now. |
| 18:39 | rabeeh | _rmk_: not failure; but i was worried if you put your hand in that are you might get high voltage into your hand |
| 18:42 | _rmk_ | one of the first things I did when I built the first channel was to power it up and see how it behaved on the scope, particularly the amount of noise on pin 4 |
| 18:45 | _rmk_ | that was a while back, so I don't remember what the results were other than I was happy with it. |
| 18:50 | davorin | evnin'... |
| 18:50 | davorin | just a quick question (o; are there any c1 users here as well? or just cubox users. |
| 18:51 | _rmk_ | both :) |
| 18:51 | davorin | good ;-) |
| 18:51 | davorin | any1 knows which bottom pads is for msata and mini-pcie? |
| 18:54 | _rmk_ | you have an imx6 solo on there... |
| 18:56 | davorin | so..pcie only... |
| 18:56 | davorin | so i need to get a dual or quad module on it then... |
| 18:57 | davorin | or test with a mini-pcie ssd |
| 18:59 | _rmk_ | I think so... as for the connections... that's a question for rabeeh |
| 19:00 | davorin | well...haven't heard from rabeeh anything quite a while now (o; |
| 19:00 | davorin | at least my i4pros are on the way ;-) |
| 19:01 | jnettlet | _rmk_, you don't have any uhs-1 cards to test with do you? |
| 19:02 | _rmk_ | I don't think I do. |
| 19:04 | jnettlet | bummer okay |
| 19:05 | _rmk_ | does it appear only with uhs-1 cards? |
| 19:07 | jnettlet | another issue. the problem with the CD interrupt happens on all cards. If you are working on more device-tree stuff you need to add the support for the 100mhz and 200mhz pinctrl |
| 19:08 | jnettlet | and test it. I can't get my board to handle 50mhz so 100 and 200mhz is out of the question |
| 19:09 | _rmk_ | I'm just sorting out some patches for mainline - I need to move the flexcan stuff out of the microsom into the hummingboard file, preferably before it hits mainline this weekend |
| 19:10 | jnettlet | is the flexcan usable for anything? |
| 19:11 | _rmk_ | not on my hummingboard since I rewired the CEC line |
| 19:15 | _rmk_ | heh. "Both boards ship with the same relatively high-end, HD-ready, Cortex-A9 Freescale i.MX6 SoC , which we can safely call the unofficial embedded Linux ARM SoC of the year." |
| 19:16 | rabeeh | davorin: hi |
| 19:17 | rabeeh | you are looking for hooking up to the pci-e or the msata? |
| 19:17 | rabeeh | on C1 both are there but not assembled |
| 19:23 | davorin | ah hi rabeeh ;-) |
| 19:23 | davorin | you got my emails? |
| 19:24 | rabeeh | hmm... let me check |
| 19:24 | rabeeh | i'v got a ton of unread emails :( |
| 19:24 | davorin | can imagine ;-) |
| 19:24 | rabeeh | which one would that be? |
| 19:24 | rabeeh | subject? |
| 19:24 | davorin | hmm..several (o; |
| 19:24 | davorin | lemme check.. |
| 19:25 | davorin | ah...hummingboard pinouts...is one |
| 19:25 | davorin | dsi/lvds pinout goes into same... |
| 19:25 | _rmk_ | okay, that's the basic cubox-i dts sent |
| 19:26 | rabeeh | oh |
| 19:26 | rabeeh | damn it; it's marked as important :) |
| 19:26 | rabeeh | 11:31AM |
| 19:27 | davorin | yes..10:31am it was here (o; |
| 19:27 | rabeeh | maybe i can give some insights about the new HummingBoard before jumping in to C1 |
| 19:27 | davorin | wanna do some testing with my lads, dsi display..as well as with msata and mini-pcie ssds |
| 19:28 | davorin | that would be nice...as i want to start developing add-on boards as soon as possible.. |
| 19:28 | rabeeh | after checking the market with the availability of external displays; we found that the most two major types are parallel displays and LVDS |
| 19:28 | davorin | john from broadcom told me they are way behind bringing dsi up (o; |
| 19:28 | davorin | so you do then rob t/l after all? |
| 19:29 | davorin | and pleaaaasse: add 2 pin headers for usb ;-) |
| 19:29 | rabeeh | the DSI are almost non existent; so we have removed the connector but the pads are there if you want to hijack those |
| 19:29 | davorin | s/rob/rgb/ |
| 19:29 | rabeeh | LVDS stays with HDMI |
| 19:29 | davorin | most ddi displays you get are form tables and phones... |
| 19:29 | rabeeh | for the USB we will have two versions of HB; the higher end one with have a 1-4 USB hub |
| 19:30 | rabeeh | 2 of those ports will go to header (mainly for additional onboard wifi modules |
| 19:30 | rabeeh | 1 will go to the mini pci-e (mainly used to activate pci-e GSM modems and bluetooth chips on 11ac wifi modules) |
| 19:31 | davorin | that would be nice...as i would need them for add-on board touchscreen controller |
| 19:31 | rabeeh | the LVDS is exactly like the freescale design - 30 pin that exposes LVDS with i2c for touchscreen controllers |
| 19:31 | rabeeh | please try using the 5V rail for the back light |
| 19:32 | rabeeh | the two usb ports takes over the 8 pin header that in the c1 |
| 19:32 | rabeeh | and there is an additional reset button added - so no need to reset HB via jumpers on the 8 pin header |
| 19:33 | rabeeh | and for _rmk_ we think that all the mechanical issues should be resolved. |
| 19:33 | rabeeh | so you won't need your Dremel anymore |
| 19:34 | rabeeh | oh; and the pci-e is on top of the board now; make it much more accessible |
| 19:34 | davorin | nice (o; |
| 19:35 | davorin | but the cpu module is still replaceable? |
| 19:35 | rabeeh | yes :) |
| 19:35 | davorin | excellent :-) |
| 19:35 | davorin | still the most visited product on my site your products (o; |
| 19:36 | davorin | ah right..any estimates when the som modules can be delivered? |
| 19:36 | davorin | or when the docs are ready for own boards? |
| 19:36 | rabeeh | davorin: i'm glad about this. |
| 19:36 | rabee | 19:36 * rabeeh hides |
| 19:36 | davorin | (o; |
| 19:37 | rabeeh | the plan is getting the schematics first and then getting the docs second |
| 19:37 | davorin | you know...i'm planning on i.mx line cards which can be plugged into a rack case... |
| 19:37 | rabeeh | the carrier boards are really very simple to design |
| 19:37 | davorin | schematic and pads layout is fine... |
| 19:38 | davori | 19:38 * davorin has to switch to chat 2morrow...this web client is a pita... |
| 19:39 | davorin | dunno if any1 knows wide vine... |
| 19:39 | rabeeh | davorin: i use xchat that puts lines in bold red once people mention my name on the channel |
| 19:39 | rabeeh | so it's much more easier to read |
| 19:39 | davorin | but a large iso here uses it to distribute iptv |
| 19:40 | davorin | just got the google license and free scales i.mx6 ddm libraries to test with... |
| 19:40 | davorin | s/ddm/drm/ |
| 19:41 | davorin | so my plan is to use cubox-i to play back iptv at home in any room without requiring a pc or a tv... |
| 19:42 | rabeeh | nice :) |
| 19:42 | rabeeh | notice i.mx6 also has secure boot and other security measures |
| 19:43 | rabeeh | it can't do nds or nagra drm's but the internet drms are doable |
| 19:43 | davorin | i asked john form broadcom, and he mentioned it would be most probably impossible to get hands on their ddm libraries (o; |
| 19:44 | davorin | with google and freescale it was a matter of a week to have it sorted out... |
| 19:44 | davorin | just some tiny thing is missing...documentation (o; |
| 19:44 | davorin | otoh...as programmers say: the source code IS the documentation (o; |
| 19:47 | jnettlet | isn't the cubox-i basically a PC? |
| 19:51 | rabeeh | jnettlet: depends to whom |
| 19:51 | rabeeh | for most people a PC is something that can run windows |
| 19:51 | davorin | put your name on it then it is personal (o; |
| 19:52 | davori | 19:52 * davorin doesn't like: windows and doors...their music just puts me off |
| 19:53 | davorin | so..gotta go for 2day....coming back 2morrow with chat ;-) |
| 19:53 | davorin | cya l8r |
| 19:53 | slawlor | Is there anyone from sales on right now? I regretably need to cancel my order, but the automatic reply to my email said it could be a few days before the email is replied to. |
| 19:54 | slawlor | Just had an unexpected expense and I can't justify spending money on the cubox right now. |
| 19:56 | slawlor | I know its about 9pm for the solid run teams timezone, still I'll be logged in for a while if anyone sees this. Just tab me |
| 19:57 | iio7 | Anyone with experience in getting a usb harddrive to spin down? I have tried using sdparm, but somehow the drive keeps starting up after a few seconds. It's a seagate drive and AFAIK it already has some internal spin down power thing, but the drive keeps running when attached to the cubox. |
| 20:01 | slawlor | I had a similar problem with my 2tb WD my book and my xkeyed xbox |
| 20:01 | slawlor | I never found a tech solution, but for what I needed I just unplugged it when I was done playing. Doesn't help your problem though sorry |
| 20:24 | i4pro | jnettlet: did you try out the sintel-bluray.iso? |
| 20:40 | i4pro | iio7: what did you do? |
| 20:40 | jmontleo | rabeeh, is there a driver for X11 that would work with kernels besides 3.0.35? |
| 20:41 | jmontle | 20:41 * jmontleo has done some searches, but hasn't found much encouraging information |
| 20:42 | iio7 | i4pro, currently I'm trying to disable/remove anything that could effect the disk. I had samba running and have set that to use ramdisk. |
| 20:43 | i4pro | do you have hdparm on what ever os your are using? |
| 20:44 | iio7 | i4pro, I'm using Debian testing, hdparm doesn't work on USB drives AFAIK so I'm using sdparm instead. |
| 20:45 | i4pro | right, not on usb, but i think it will still work to check that it is actually spun down |
| 20:46 | iio7 | i4pro, yes, and it can give more data too. In this case however I can hear if the drive spins down. |
| 20:46 | i4pro | iio7: you can download and try http://hd-idle.sourceforge.net/ |
| 20:46 | i4pro | it works on usb |
| 20:47 | iio7 | i4pro, thanks! will test it! |
| 20:53 | _rmk_ | okay... |
| 20:53 | _rmk_ | http://www.home.arm.linux.org.uk/~rmk/cubox/hummingboard-cubox-i-v3.13-rc7-20140116.diff |
| 20:53 | _rmk_ | http://www.home.arm.linux.org.uk/~rmk/cubox/hummingboard-cubox-i-v3.13-rc7/ |
| 20:54 | i4pro | let me know if that works |
| 20:54 | _rmk_ | is the current patch queue against v3.13-rc7 |
| 20:54 | _rmk_ | I may not be around much this evening |
| 20:55 | iio7 | i4pro, I have just compiled hd-idle and installed it, however it does the same as with sdparm, it shuts down the drive, then 1 second after the drive spins up again. |
| 20:55 | iio7 | I am using ext4 on the drive if thats important. |
| 20:55 | i4pro | is that drive being written to? |
| 20:56 | iio7 | i4pro, I can't see any disk activity except for "jbd2" using iotop. |
| 20:56 | iio7 | every 5 seconds or something |
| 20:57 | iio7 | just for this test I have disabled every daemon that isn't needed too |
| 20:57 | i4pro | so thats a fs journaling thing, is it ext4? or something? |
| 20:57 | iio7 | yes, its ext4 |
| 20:57 | i4pro | you might have to add some stuff to your fstab, like noatime, and some other stuff |
| 20:58 | iio7 | I don't have anything in fstab, the disk boots from u-boot |
| 20:58 | i4pro | ok so maybe you need to add some params, and set it up in fstab |
| 20:58 | i4pro | give this a read, its for arch, but i think it will get you where you want to go eventually https://bbs.archlinux.org/viewtopic.php?id=113516 |
| 21:00 | i4pro | or you could go with a FS that is not journaled |
| 21:00 | i4pro | but dont know what your needs are |
| 21:01 | iio7 | i4pro, yeah, maybe that is the solution, I need to look into what FS cubox pro supports |
| 21:01 | iio7 | it definitely is jbd2 that is causing the problem! |
| 21:02 | iio7 | http://paste.debian.net/76654/ |
| 21:08 | jnettlet | i4pro, sorry got distracted will test now |
| 21:09 | iio7 | Does cubox support xfs and booting from it? |
| 21:09 | jnettlet | jmontleo, the basic driver should work with anything that uses the fbdriver. For KMS support you will need to wait just a bit longer |
| 21:12 | slawlor | Any solid run employees around right now? |
| 21:16 | rabeeh | slawlor: yes |
| 21:16 | slawlor | Hey rabeeh so I regretfully need to cancel my order for my pro model |
| 21:16 | slawlor | I sent an email detailing it, but the automatic reply mentioned you guys are overloaded and it could take a few days. |
| 21:17 | rabeeh | slawlor: please send email to [email protected] |
| 21:17 | rabeeh | to which email? |
| 21:17 | slawlor | from the account page the returns/exchange button links to the sales email |
| 21:17 | rabeeh | ok; sales is also fine |
| 21:17 | slawlor | I can CC it to there also. I was just worried it might ship before it was cancelled. I hit some economic hardship and cna't justify spending the money |
| 21:18 | slawlor | Last email I had from you said it would be shipping yesterday, not sure if thats still the case or if it will be from the november batch. |
| 21:18 | iio7 | I can't seem to locate information on supported FS. I know the CuBox installer recently added support for ext4. |
| 21:19 | rabeeh | slawlor: i'll PM |
| 21:37 | xraxor | slawlor: what you looking for? |
| 21:55 | iio7 | So the cubox installer has been updated to support ext4. Now, ext4 isn't any good if you want to spin down a harddrive because jdb2 keep writing to the disk every 10 seconds. So I changed the filesystem and copied the files and made a new boot.scr with ext3, but no.. the updated version of u-boot only uses ext4load and doesn't make any kind of fallback to ext2load. |
| 21:55 | rabeeh | _rmk_ what would be the best floating point build target for the imx6? |
| 21:55 | rabeeh | neon? vfpv3-d16? |
| 22:00 | iio7 | Gonna try a manual installation of debian.. no cubox installer. Which is better hard-float or soft-float? |
| 22:05 | jnettlet | hard-float should be fine |
| 22:08 | iio7 | Some dead links on solid-run pointing towards debian stuff. |
| 22:33 | iio7 | Wanting to run with ext3 seem to need a downgrade of SPI. |
| 22:39 | jnettlet | iio7, why not ext4? |
| 22:40 | iio7 | because of the journal problem. jdb2 keeps writing to disk every 10 seconds and I need the disk to power down after some time with idle. |
| 22:41 | iio7 | sorry, jbd2 |
| 22:42 | jmontleo | hmmm, system thinks it's playing sound, but nothing is coming out of the tv |
| 22:46 | jnettlet | jmontleo, volume up? |
| 22:47 | jmontleo | ya, very high both on the tv and system, if I don't remember I'm going to get a very loud surprise when I plug the roku back in later |
| 22:47 | jmontleo | but pavucontrol is showing the volume level go up and down when i try to play a video from youtube |
| 22:47 | jmontleo | it very much thinks it's playing and to the hdmi rather than spdif even |
| 22:51 | jnettlet | does your tv info give you info about the audio stream it thinks it is getting? |
| 22:54 | jmontleo | no, unfortunately not |
| 22:54 | jmontleo | I'll try the one in the basement later. i've seen it tell me things like unsupported codec, etc. in the past |
| 22:55 | jmontleo | maybe that will give me a hint |
| 22:55 | iio7 | Anyone who knows which version is before ext4 support? http://download.solid-run.com/pub/solidrun/cubox/u-boot/2009.08_5.4.4_SR1/ |
| 23:03 | niki1 | Anyone know about how long from the time you received "Item has been shipped Notice" till you can see it in usps.com? |
| 23:03 | niki1 | Just curious |
| 23:06 | iio7 | I just did a "dd if=u-boot-cubox_hynix_cubox_2GB_uart.bin of=/dev/mtdblock0", now the cubox wont turn on. When pluggin in the power cord no lights come on and the device is "dead" |
| 23:08 | iio7 | brick |
| 23:11 | iio7 | Anyone who knows if there is one of the cubox installers that use ext3 and not ext4 of these http://download.solid-run.com/pub/solidrun/cubox/installer/ |
| 23:11 | jnettlet | iio7, I doubt it. Really ext3 is a horrible filesystem to use on any sort of flash storage. |
| 23:12 | iio7 | I'm using it on a USB disk |
| 23:12 | iio7 | I need to have the disk power down which isn't possible using ext4 |
| 23:12 | jnettlet | why is that? |
| 23:12 | jnettlet | all my disks use ext4 and powerdown just fine |
| 23:13 | iio7 | I have tested several disks and jbd2 keeps the disks up because it writes to disk every 10 seconds |
| 23:13 | iio7 | The issues is discussed here https://bbs.archlinux.org/viewtopic.php?id=113516 |
| 23:20 | jnettlet | iio7, have you enabled laptopmode? |
| 23:21 | jmontleo | hmm, analog output; that seems not right |
| 23:21 | iio7 | jnettlet, I think I tried that without success too. Right now I need to unbrick the cubox. |