Skip to forum content
Discussions relating to Kosagi's products and services
You are not logged in. Please login or register.
Search options (Page 1 of 3)
Posts found: 1 to 25 of 55
It was discharged yesterday, and charged up fine. Left it plugged in overnight. Now software says not charging. I asked senoko for stats and it says:
Temperature: 21.9 C
Voltage: 12373 mV
Current: 0 mA
Average current: 1 mA
Target voltage: 12600 mV
Target current: 0 mA
Number of cells: 3 cells
Cell 1 voltage: 4129 mV
Cell 2 voltage: 4118 mV
Cell 3 voltage: 4126 mV
Cell 4 voltage: 0 mV
Charge status: 0x1000
Charging allowed? yes
Can suspend? no
Can precharge? no
Can maintenance? yes
Temperature limit? no
Temperature limit? no
Can fastcharge? no
Pulse charging? no
Pulse disable CHG? no
Cell balancing? no
Precharge timeout? no
Fastcharge timeout? no
Overcharge OV? no
Overcharge OC? no
Battery empty? no
TERMINATE CHARGE ALARM
No errors detected
No safety alerts
No safety status messages
So I guess that alarm is the issue? If I use chg pause and then chg set I can force it to charge, but once I chg resume it stops charging again (tried this only once, but that is what happened).
@dileepvr a week or two ago? Basically the day I got the email from CrowdSupply.
Follow on question: how few cells can senoko handle? Could it handle a single-cell battery? That's much less voltage.
Right, I did figure I'd need a step-down board for the power rail.
I got an extra senoko from crowdsupply, and I want to try using it to power other motherboards (like, say, a raspberry pi). What is the minimum connections I will need to get something functional? Power rail, of course, and I assume a UART so that I can change the settings to match the battery I get. Anything else that is absolutely necessary for a first pass?
Barely related question: where did you get the pixelQi display from?
I tried to enable acceleration/OMTC in iceweasel and got:
(process:1845): GLib-CRITICAL **: g_slice_set_config: assertion 'sys_page_size == 0' failed
libGL error: MESA-LOADER: malformed or no PCI ID
libGL error: unable to load driver: etnaviv_dri.so
libGL error: driver pointer missing
libGL error: failed to load driver: etnaviv
I'm using the etnaviv/armada stuff from xobs' NAS
Certainly one of the more sluggish things I run. A bit faster now that I have 2D acceleration on, but still terrible when there are lots of images on a page, not sure why as my RAM doesn't fill up.
@stim my novena always did that. 2D acceleration fixed it. Or just turn off showing window contents while dragging.
Hmm, mplayer just crashes now, @ballanux -- what video player are you using?
I can drag windows around with their contents showing, so something is definitely accelerated!
@ballanux installing just the kernel image and two xserver packages from that link worked for me, thanks!
I now get the same error as you. etnadrm_dri is definitely a thing in that source tree, but it doesn't seem to compile to its own .so?
I'm not sure how these binaries are different from the kernel/xserver packages I was building from source. I definitely needed both, though, since the xserver packages without the kernel clearly tried to use GPU, but then would fail and X would not start, until I installed the kernel from that link.
I grabbed the source package for the X11 driver, and narrowed the problem down to this line in etnadrm.c :
conn->fd = drmOpen("etnaviv", NULL);
I'm reading about drmOpen now, but I'm not really sure what's likely to be failing.
My 60-novena.conf looks like this:
Option "UseGPU" "true" # change to "false" to disable 2D acceleration
Option "XvPreferOverlay" "FALSE" # FALSE enables xv acceleration
# The next two options switch it from galcore to etnadrm mode
Option "Screen" "0"
Option "AccelModule" "etnadrm_gpu"
Full Xorg.0.log after removing it and rebooting: http://pastie.org/10302549
I switched etnaviv to be built as a loadable module so that I can clearly see it in my lsmod. I still get the same in my X logs:
[ 7.821] (EE) armada(0): etnaviv: unable to open: No such file or directory
[ 7.822] (WW) armada(0): [drm] Vivante initialization failed, running unaccelerated
I am now running on branch v3.19-novena-etnadrm-r3 from xobs' novena-linux git, which has etnaviv enabled as a built-in module, but I get the same error from X...
Hmm, it seems the packages are built using a script from the wiki here: http://www.kosagi.com/w/index.php?title … nux-kernel
A real source package would be nice, but this should do
apt-get source linux-image-novena
E: Unable to find a source package for linux-upstream
I know the kernel tree is on xobs' github, but what about the packaging? I'd really like the source package, or the tree with the debian packaging at least, so that I can easily try changes. Am I missing something? I don't see a kernel source package on the novena.io repo
Hmm, seems that I cannot build that module for the existing kernel because it needs CONFIG IOMMU API on ... maybe I'll try doing a full kernel build later, if I get some time.
Seems likely that the issue is I lack https://github.com/austriancoder/linux/ … ng/etnaviv in my kernel. Going to try playing with that.
Seems to be working for these people on related hardware: http://forum.solid-run.com/linux-on-cub … -s240.html -- not sure what I'm missing yet.
Using that config I get:
[ 11967.876] (EE) armada(0): etnaviv: unable to open: No such file or directory
[ 11967.876] (WW) armada(0): [drm] Vivante initialization failed, running unaccelerated
What's the status of 2D acceleration? How hard is it to enable experimentally on my own machine?
I had no problem running it as non-root, but the senoko can take quite some time in a reboot loop before it stabilises after a flash. At least, it did for me
Posts found: 1 to 25 of 55