cnxsoft has quit [Read error: Connection reset by peer]
cnxsoft1 has joined #linux-sunxi
cnxsoft1 is now known as cnxsoft
victhor has quit [Remote host closed the connection]
tl_lim has quit [Quit: Leaving]
sunshavi has quit [Ping timeout: 260 seconds]
<wens>
forkbomb: sounds like a waste of hardware compositing resources :p
TheSeven has quit [Ping timeout: 240 seconds]
TheSeven has joined #linux-sunxi
lurchi_ has joined #linux-sunxi
lurchi__ has quit [Ping timeout: 260 seconds]
IgorPec has joined #linux-sunxi
TestLinux has joined #linux-sunxi
montjoie has joined #linux-sunxi
cnxsoft has quit [Remote host closed the connection]
cnxsoft has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
foxx_ has joined #linux-sunxi
fredl has joined #linux-sunxi
xes has joined #linux-sunxi
<icenowy[m]>
icenowy[m]: see my github h6-init branch
<icenowy[m]>
mmc, usb2/3
<icenowy[m]>
and eth
reinforce has joined #linux-sunxi
<mripard>
forkbomb: I'm not sure it does, or it changed compared to when it was introduced then
jernej has quit [Ping timeout: 252 seconds]
doppo has quit [Ping timeout: 252 seconds]
<forkbomb>
mripard: hm, ok. it's been a while since i tried to use it, so i could be wrong
<forkbomb>
last time i tried it also needed GLES 3, so mali 400 couldn't use it out-of-the-box anyway
<wens>
icenowy[m]: nice, that is a lot of stuff
<mripard>
forkbomb: maybe I'm confusing it with something else though
jstein has joined #linux-sunxi
<wens>
icenowy[m]: I have some comments about the AXP805 stuff (basically, it was already submitted before for an A80 board, which is why we have the "master-mode" property for the AXP806)
naggety has joined #linux-sunxi
fl_0 has quit [Quit: STRG + Q]
foxx_ has quit [Ping timeout: 264 seconds]
fl_0 has joined #linux-sunxi
foxx_ has joined #linux-sunxi
<icenowy[m]>
wens: just comment on github ;-)
<wens>
hmm, I remembered the wrong model number :/
<mripard>
icenowy[m]: can you put the actual list of items you're working on for the H6 ?
<mripard>
claiming support for a whole SoC doesn't really help, since we can't realistically expect you to work on everything at the same time
<mripard>
but no one can work on what you're not working on either
Gerwin_J has joined #linux-sunxi
<wens>
I put that entry there, as I was being lazy
<wens>
but to her credit, she did already get through a number of the basics to get a nice headless system
<icenowy[m]>
mripard: maybe I claimed nothing now
TestLinux has quit [Quit: Page closed]
<mripard>
wens: oh, so you did it, sorry icenowy[m] :)
<wens>
icenowy[m]: sorry for the confusion
<wens>
mripard: I updated it with specifics
<wens>
and the 4.17 entry should be mostly correct (I didn't go through all the driver commits)
<icenowy[m]>
wens: are you still WIP on the LInux_mainlining_effort page?
<icenowy[m]>
I think the matrix should be updated
<icenowy[m]>
for H6
<wens>
icenowy[m]: I'm done with it for now
<wens>
I might have to go through the matrix later to update the stuff for 4.17
<icenowy[m]>
and for H6 PCIe I will make it black now
<naggety>
Hi guys. Yesterday I asked for some help, and I didn't get any response. If you don't mind, I copy & paste my question from yesterdaty and let's see if someone can tell me something helpful.
<naggety>
Hi. I hope someone can help me with this. I'm making a sunxi-linux system for an A20 SoC using Buildroot. I want to have 2 monitors: one LCD and one analog TV (CVBS, PAL). At the same time I want to use TV decoder to input image from 4 cameras.
<naggety>
My problem is that when I modprobe that driver for TV decoder, the second monitor (the TV one) stop working
<naggety>
However, LCD + TV decoder work fine. LCD + TV output work fine. TV output + TV decoder work fine. All three at the same time doesn't work
<naggety>
That's it. Any help or guidance about where to look would be appreciated.
<wens>
you're probably not going to get much help with the 3.4 kernel :|
<KotCzarny>
a20 has tv-in?
<KotCzarny>
or its some custom hw?
Baschto has joined #linux-sunxi
<wens>
it does, 4 channel cvbs
<KotCzarny>
never seen it exposed
<naggety>
I know, but unfortunately I need to use that version. I don't expect any kind of support, but maybe an advice that let me look to the right place.
<icenowy[m]>
KotCzarny: some vehicle-based solutions with T2(A20)/T3(R40) may use them
<icenowy[m]>
ok created the first black block in the mainline status matrix :-(
<naggety>
The board exposes all TV-in channels, and I have it working yet. The problem is to use the 3 things at the same time: LCD output + TV output + TV input
<mripard>
icenowy[m]: can we get the amount of space a device would require when it's detected on the PCIe bus ?
<naggety>
I was wondering if it could be not enough CMA
<naggety>
But I couldn't increase it because it collides with G2D memory address and LCD memory address, and I don't know how to move them
<naggety>
(I neither know if it makes any sense that CMA can be the problem)
<KotCzarny>
increase via boot param and test?
<icenowy[m]>
mripard: no, as everything are muxed
<mripard>
icenowy[m]: ok :/ too bad
<wens>
naggety: you might be hitting clock issues, as there are only 2 PLLs for all things video-related
<icenowy[m]>
wens: will it work if a of compatible is attached to different drivers for different bus?
<naggety>
KotCzarny: I tried that, adding cma=256M to bootargs. But it says that cma could not be allocated, it collides with G2D and LCD memory regions.
<KotCzarny>
try something smaller then
<wens>
naggety: so if you have 3 incompatible dot-clocks, it's not going to work out
<gnarface>
90M is a sane value for CMA
<icenowy[m]>
mripard: I'm thinking of using hypervisor to wrap it, but I'm not familiar with ARM hypervisor
jtf has quit [Ping timeout: 240 seconds]
clemens3 has joined #linux-sunxi
<icenowy[m]>
yes I also think it's dot-clock issue
<gnarface>
64M <i>should</i> be a sane value for CMA, but i ran into something that it wasn't enough for (i forget what)
<icenowy[m]>
usual Allwinner SoCs have only two video PLLs
<wens>
icenowy[m]: yes it will work, the bus is in charge of registering new devices and match it against some list of drivers, typically its own bus
Baschto has quit [Ping timeout: 264 seconds]
<wens>
icenowy[m]: so the list for i2c and rsb is separate, and the i2c core will only register devices that match against i2c drivers
<naggety>
wens: thanks, I'm going to read about clocks and PLL in the SoC manual. If I stuck, I will ask again
<pmpp>
gnarface: for 2K hevc need 80MiB
<gnarface>
pmpp: good to know but whatever it was definitely wasn't running at 2k
<icenowy[m]>
wens: ok
fkluknav has joined #linux-sunxi
<icenowy[m]>
P.S. for RSB chips if they're started from I2C
<icenowy[m]>
how did them switch to RSB?
xerpi has joined #linux-sunxi
<wens>
the RSB driver commands the RSB controller to send the init sequence
<wens>
which is basically an I2C write to the interface mode register of all chips
Baschto has joined #linux-sunxi
<icenowy[m]>
interesting
<icenowy[m]>
wens: so for "all chips" you mean send it to every I2C address?
<naggety>
Sorry, I have no experience with Linux. devmem2 lets me to read registers?
<KotCzarny>
hmm, why is there no copy on sunxi page too?
foxx_ has quit [Remote host closed the connection]
<KotCzarny>
naggety: yes
<naggety>
Thanks, KotCzarny, I already had the manual
<naggety>
I'm going to try with devmem. If anybody knows an easier way (maybe where to look to the default values) please let me know.
<KotCzarny>
no one has written subxi-info tool, so its the only way i guess
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
BenG83 has joined #linux-sunxi
leviathan has joined #linux-sunxi
Gerwin_J has quit [Remote host closed the connection]
yann has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
mavkhimenia has joined #linux-sunxi
Gerwin_J has quit [Client Quit]
Gerwin_J has joined #linux-sunxi
Gerwin_J has quit [Client Quit]
Baschto has quit [Ping timeout: 276 seconds]
Gerwin_J has joined #linux-sunxi
Gerwin_J has quit [Client Quit]
mavkhimenia has quit [Ping timeout: 264 seconds]
<Pe3ucTop>
TheSeven: about your question SPI speed , working with ~24..25 Mhz without problems.
megi has joined #linux-sunxi
<Pe3ucTop>
Hey, could anybody check SPI Dual mode patch on Allwinner sun6i/sun8i other then V3s CPU ?
<BenG83>
does anyone have a Yuntab B102? with a A64...
<Pe3ucTop>
TheSeven: My research about SPI MBC / MWTC / STC registers -> MBC : Total number of bursts affect include all together Tx, Dummy, Rx ; MWTC -> Effect count of Tx data will be taken from TX FiFo, Tx part length. Put attention, by default sun6i/8i SPI TX is in Dual Mode (transmit on both pins MISO & MOSI), if needed Single mode -> STC need be defined , if it is equal to MWTC then all in Single mode, otherwice Single mode transmit go first (untill empty
<Pe3ucTop>
After transmit, goes Dummy burst (do not know behaviour dual/single mode, did not experimented with it).
<naggety>
OK, I have it working. I found that LCD was using video_pll1 at 297MHZ, and TV-out video_pll0 at 270MHz. TV decoder driver was using pll0 at 297MHz, hard-coded, which messed TV-out.
<naggety>
Changing TV decoder driver to use pll1, which already have the same freq it needs, everything seems to work
<Pe3ucTop>
and after SPI dummy burst goes RX, by default RX is in Single mode, to switch to SPI Dual Mode DRM bit need be set.
<naggety>
Thanks wens, hanetzer and KotCzarny
yann has quit [Ping timeout: 264 seconds]
BenG83 has quit [Remote host closed the connection]
BenG83 has joined #linux-sunxi
yann has joined #linux-sunxi
dddddd has joined #linux-sunxi
naggety has quit [Ping timeout: 260 seconds]
naggety has joined #linux-sunxi
<naggety>
btw, if it can be useful for someone: I found what pll were used and with what freq in function disp_clk_cfg of file drivers/video/sunxi/disp/disp_clk.c. I have had to add some messages printing.
yann has quit [Ping timeout: 268 seconds]
jtf has joined #linux-sunxi
BenG83_ has joined #linux-sunxi
fkluknav has quit [Ping timeout: 256 seconds]
BenG83 has quit [Ping timeout: 245 seconds]
Gerwin_J has joined #linux-sunxi
BenG83_ has quit [Remote host closed the connection]
bonbons has joined #linux-sunxi
yann has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
BenG83_ has joined #linux-sunxi
aalm has quit [Ping timeout: 264 seconds]
mkid has joined #linux-sunxi
BenG83_ has quit [Remote host closed the connection]
lucascastro has quit [Read error: Connection reset by peer]
bonbons has quit [Quit: Leaving]
yann has joined #linux-sunxi
afaerber has quit [Quit: Leaving]
fkluknav has joined #linux-sunxi
naggety has quit [Quit: Page closed]
mavkhimenia has joined #linux-sunxi
matthias_bgg has quit [Ping timeout: 256 seconds]
afaerber has joined #linux-sunxi
lucascastro has joined #linux-sunxi
return0e has joined #linux-sunxi
Baschto has joined #linux-sunxi
elros_ has quit [Remote host closed the connection]
BenG83 has quit [Quit: Leaving]
Xalius_Ph has joined #linux-sunxi
jstein has quit [Read error: Connection reset by peer]
lucascastro has quit [Ping timeout: 260 seconds]
vagrantc has joined #linux-sunxi
lurchi_ is now known as lurchi__
cnxsoft has quit [Quit: cnxsoft]
xerpi has joined #linux-sunxi
jernej has joined #linux-sunxi
matthias_bgg has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
lucascastro has joined #linux-sunxi
tllim has joined #linux-sunxi
doppo has joined #linux-sunxi
yann has quit [Ping timeout: 252 seconds]
yann has joined #linux-sunxi
mavkhime_ has joined #linux-sunxi
return0e has quit [Ping timeout: 256 seconds]
mavkhimenia has quit [Ping timeout: 248 seconds]
kevans91 has quit [Remote host closed the connection]
JohnDoe7 has joined #linux-sunxi
mcan has quit [Ping timeout: 240 seconds]
JohnDoe_71Rus has quit [Ping timeout: 260 seconds]
mcan has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
jernej has quit [Ping timeout: 256 seconds]
yann has quit [Ping timeout: 264 seconds]
kevans91 has joined #linux-sunxi
doppo has quit [Ping timeout: 256 seconds]
return0e has joined #linux-sunxi
kevans91 has quit [Changing host]
kevans91 has joined #linux-sunxi
fkluknav has quit [Ping timeout: 240 seconds]
yann has joined #linux-sunxi
jernej has joined #linux-sunxi
lucas_ has joined #linux-sunxi
lucas_ has quit [Remote host closed the connection]
lucas_ has joined #linux-sunxi
lucascastro has quit [Ping timeout: 260 seconds]
xerpi has quit [Quit: Leaving]
clemens3 has quit [Remote host closed the connection]
fkluknav has joined #linux-sunxi
yann has quit [Ping timeout: 240 seconds]
lucas_ has quit [Ping timeout: 264 seconds]
victhor has joined #linux-sunxi
tl_lim has joined #linux-sunxi
tllim has quit [Ping timeout: 245 seconds]
Xalius_Ph has quit [Quit: Bye]
Xalius_Ph has joined #linux-sunxi
Xalius_Ph has quit [Client Quit]
lucascastro has joined #linux-sunxi
lucascastro has quit [Ping timeout: 240 seconds]
msimpson has quit [Quit: Leaving]
nuuuciano__ has quit [Ping timeout: 263 seconds]
nuuuciano_ has joined #linux-sunxi
f0xx has joined #linux-sunxi
nuuuciano_ has quit [Remote host closed the connection]
doppo has joined #linux-sunxi
Baschto has quit [Remote host closed the connection]
doppo has quit [Ping timeout: 240 seconds]
jernej has quit [Ping timeout: 256 seconds]
doppo has joined #linux-sunxi
kevans91 has quit [Remote host closed the connection]
kevans91 has joined #linux-sunxi
kevans91 has quit [Changing host]
kevans91 has joined #linux-sunxi
lucascastro has joined #linux-sunxi
nuuuciano has joined #linux-sunxi
yann has joined #linux-sunxi
fkluknav has quit [Ping timeout: 264 seconds]
jernej has joined #linux-sunxi
lucascastro has quit [Ping timeout: 248 seconds]
fkluknav has joined #linux-sunxi
ak_hepcat has joined #linux-sunxi
reinforce has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
Gerwin_J has joined #linux-sunxi
anarsoul|2 has joined #linux-sunxi
victhor has quit [Remote host closed the connection]
yann has quit [Ping timeout: 256 seconds]
Gerwin_J has quit [Read error: No route to host]
Gerwin_J_ has joined #linux-sunxi
yann has joined #linux-sunxi
lucascastro has joined #linux-sunxi
leio has quit [Remote host closed the connection]
leio has joined #linux-sunxi
fkluknav has quit [Ping timeout: 240 seconds]
<ak_hepcat>
Good morning from Alaska: My pine64+(2g) on any mainline kernel (armbian stable/testing, or latest 4.16.0-rc7) doesn't see USB connections on either port. Any ideas for further troubleshooting, or specific kernel driver tweaking?
jernej has quit [Ping timeout: 276 seconds]
yann has quit [Ping timeout: 264 seconds]
mavkhimenia has joined #linux-sunxi
mavkhime_ has quit [Ping timeout: 264 seconds]
lucascastro has quit [Remote host closed the connection]
mavkhimenia has quit [Ping timeout: 240 seconds]
f0xx has quit [Ping timeout: 264 seconds]
<KotCzarny>
hmm, sun8i_ths driver seems to be doing its job, but why it's not exposing temperature anywhere in /sys ?
<pmpp>
cooling by obfuscating :)
<KotCzarny>
pah
a|3x has quit [Ping timeout: 264 seconds]
a|3x has joined #linux-sunxi
<KotCzarny>
opipc idles ~33C with 4.16 (with heatsink and in closed case
<KotCzarny>
not bad
<KotCzarny>
~23C ambient
<KotCzarny>
friggin' allwinner, why did they abandon cooling by lowering voltage/freq
<webuser5224>
I have an Orange Pi Zero, but the wifi seems a bit unreliable
<webuser5224>
I also read that this is a common problem
<KotCzarny>
yes, get another wifi card
<webuser5224>
Is there any board with similar specifications that has a better wifi chip?
<webuser5224>
Or does that not make much sense and should I rather add a wifi dongle?
<webuser5224>
I need all the ports :(
<KotCzarny>
you are better with usb dongle
<KotCzarny>
use usb hub?
<KotCzarny>
for slow interfaces
<webuser5224>
I want to keep a small form factor
<webuser5224>
Are RTL8189 or AP6212 any better?
<KotCzarny>
a bit
<KotCzarny>
but not much
<KotCzarny>
i like ralink 5370 or 5572
<DonkeyHotei>
ralink 11n drivers are bad
<KotCzarny>
not as bad as realteks cheapest of the cheap hw
IgorPec has quit [Ping timeout: 256 seconds]
<webuser5224>
Maybe I shouldn't use the default antenna on the OPi Zero, but the wifi reception is literally very bad
<webuser5224>
I had issues getting a few resources via apt etc.
<KotCzarny>
no, xr819 is bad
<DonkeyHotei>
the xradio driver was never finished
<KotCzarny>
and driver even worse
<webuser5224>
oh ok
<DonkeyHotei>
ap6212 is broadcom, no?
<webuser5224>
I read it is
<webuser5224>
A combination of wifi and bt from broadcom, it said
<DonkeyHotei>
if i had no choices other than ap6212, xr819, or rtl8189, i'd probably pick rtl8189, but ugh
<webuser5224>
Well, those are the ones I found so far
<webuser5224>
But I'd like some solution that is not too large or doesn't draw too much power
<DonkeyHotei>
there might be other choices on usb
<KotCzarny>
i've never had any stability problems with rt5370
<KotCzarny>
and will soon be testing it as a ap too
<DonkeyHotei>
ralink 11n drivers need a ground-up rewrite
<KotCzarny>
wfm
<KotCzarny>
which i cant say about realteks random disconnects
florianH has quit [Quit: Connection closed for inactivity]
<DonkeyHotei>
completely unusable as an ap unless you use the ancient ralink vendor driver
<KotCzarny>
DonkeyHotei: preliminary tests show that it works just fine (tested briefly with nokia n900 as a client)
<DonkeyHotei>
leave it up for a couple of weeks and you'll see
<KotCzarny>
anyway, bedtime
<DonkeyHotei>
it simply stops passing traffic after a while
<DonkeyHotei>
only a reboot fixes that
<webuser5224>
Are those mini antennas on the OPi Zero models also a problem? Or does it depend on the wifi chip mostly?
<DonkeyHotei>
antenna may affect range, but it's one of several factors
<DonkeyHotei>
look for adapters with the ar9271 chipset
<DonkeyHotei>
that will give you the least headache in ap mode on 2.4GHz
<webuser5224>
I actually have one, yeah
<webuser5224>
But it's huge in comparison
jernej has joined #linux-sunxi
<webuser5224>
I'd also need more than one USB port on my board then
fl_0 has quit [Quit: STRG + Q]
reinforce has quit [Quit: Leaving.]
<hanni76>
webuser5224, you need OrangePi PC+
<DonkeyHotei>
hanni76: which wifi chip does that have?
<hanni76>
RTL
<webuser5224>
I'm a bit limited because I want to run it from a battery
<TheSeven>
hmm I'm having some issues with DMA losing bursts of data
<TheSeven>
someone ever had that before?
<hanni76>
rtl8189fs, to be exact
<DonkeyHotei>
there are opi-like devices that come with a battery, too
<DonkeyHotei>
travel routers etc
<TheSeven>
it seems like occasionally one RAM=>SPI 32bit transfer gets lost, and around the same time one SPI=>RAM 512bit burst gets lost
<TheSeven>
seems to happen every couple thousand DMA transfers
<hanni76>
webuser5224, power consumption is something that can be tuned
Gerwin_J_ has quit [Ping timeout: 265 seconds]
Gerwin_J has joined #linux-sunxi
<BenG83>
TheSeven, maybe some error handling is missing there?
<BenG83>
are there flags for DMA errors or sth?
<TheSeven>
I can't see how an error could even occur there
fl_0 has joined #linux-sunxi
<TheSeven>
I mean this would have to be an AHB bus level error
adj_ has quit [Remote host closed the connection]
<TheSeven>
and why should the SPI slave ever return that?
adj_ has joined #linux-sunxi
<TheSeven>
it looks like the DMA fails to increment an internal counter or is doing it twice in some race condition
<TheSeven>
or maybe an issue with the SPI FIFOs
<BenG83>
mmmh
<TheSeven>
neither DMA nor SPI really have any error conditions
<TheSeven>
and this seems to happen between every 100k to 10M transfers, randomly
<TheSeven>
and I don't mean bytes but actual DMA operations
<TheSeven>
I'm seeing it every couple of minutes while pushing 50Mbit/s through SPI
<TheSeven>
but sometimes it lasts for half an hour
<BenG83>
does it happen at much lower speeds too?
f0xx has quit [Ping timeout: 246 seconds]
<BenG83>
like 5Mbit/s ?
<TheSeven>
it's always the same resulting state though: TX DMA is finished, SPI TX buffer is empty, SPI transfer size says 4 bytes left (where did those go missing?), SPI RX buffer contains 16 bytes (and would only trigger DMA at 32), RX DMA still wants to receive 52 bytes
<TheSeven>
haven't tried yet - would take ages to reproduce at slower speeds
<TheSeven>
I may try that overnight
Gerwin_J has quit [Quit: Gerwin_J]
<TheSeven>
hmm I'm wondering if this may be TX DMA clashing with RX DMA somehow while competing for the SPI peripheral on the bus
<BenG83>
we don't have any errata sheets for AW SoC sadly
<TheSeven>
I doubt that they would have even discovered this
<TheSeven>
these peripherals probably don't see a lot of high-load testing/applications
<BenG83>
true
<TheSeven>
let me play with the DMA trigger thresholds...
<BenG83>
I have the problem on RK3328 that SPI DMA is completely broken
<TheSeven>
maybe I can offset the TX vs. RX DMA bursts by adding half a burst to the RX watermark
<TheSeven>
heh and on MT7621 SPI is completely broken ;)
Putti has quit [Ping timeout: 252 seconds]
<DonkeyHotei>
openwrt seems to read/write nor flash just fine on mt7621 for me
<TheSeven>
as in the hardware is a complete misconception, specifically designed to access flash memory and nothing else, and not supporting accesses of more than 32 bytes with one request (which is pretty pointless especially for flash), and on top of that it has a really weird bus-side glitch where it mixes up bits in some cases
<TheSeven>
that only happens with full-duplex transfers though, so it's indeed fine (but slow as hell) for flash access
<TheSeven>
getting that to talk to NRF24L01 or CC1101 or similar sub-GHz RF modules required a major rework
<DonkeyHotei>
at least it's not broadcom, right?
<TheSeven>
heh never had to mess with broadcom yet, but even from a higher level view the first-gen RPi was just urgh...
<DonkeyHotei>
right
<TheSeven>
this H3 that I'm working with right now consumes less power running full tilt than the RPi was drawing idle
<TheSeven>
not to even talk about performance...
* TheSeven
hopes nanopi neo core boards will be back in stock soon
<TheSeven>
hmm so far it's going strong with TX and RX dma watermarks shifted apart by 16 (half a burst)
<TheSeven>
but it will take hours until I can be somewhat sure that it doesn't happen anymore
<TheSeven>
I've had it run just fine for half an hour before it suddenly crashed
fdcx has joined #linux-sunxi
hanetzer has joined #linux-sunxi
fdcx has quit [Ping timeout: 248 seconds]
* willmore
is following TheSeven's work.
<TheSeven>
willmore: too bad that most of what I'm doing won't be terribly useful to others - it's highly application specific
<TheSeven>
sure, it will likely produce a bit of insight about the hardware
<TheSeven>
but what I'm doing right now is wrapping the SPI and DMA peripherals into a userspace I/O device (that also exposes a contiguous coherent buffer), then using that to pump data to an FPGA from userspace
<TheSeven>
the regular kernel SPI API just doesn't provide the performance required by my application - keeping CPU overhead minimal is crucial here
<BenG83>
does anyone know if spidev can support extra chipselects via GPIOs?
<TheSeven>
yes, it should support that already
fdcx has joined #linux-sunxi
<TheSeven>
it doesn't use HW chip select at all
hanetzer has quit [Ping timeout: 240 seconds]
<TheSeven>
(i.e. drives the HW chip select pin like a GPIO)
<TheSeven>
I haven't tried, but I'd assume that you can just plug arbitrary GPIOs into the device tree there and it should just work
<TheSeven>
half an hour (~2 million DMA ops) and it still hasn't crashed, this is starting to look promising
lurchi__ is now known as lurchi_
<BenG83>
TheSeven, if you have time maybe add those findings to the wiki...
<TheSeven>
hmm there aren't really that many findings, other than that DMA has at least 2 bugs when talking to SPI (doesn't like non-incrementing address on RAM side, doesn't like TX and RX bursts at the same time?), and SPI seems to generally fail above 50MHz (and I don't have a fast enough scope yet to check what's going on there)
<TheSeven>
but let's wait for a night and see if this is actually stable now, or if I'm just chasing the next red herring here...
<TheSeven>
this is like the 3rd time now that I'm thinking that I've finally caught the root cause, but it seems to just be happening more rarely
Mr__Anderson has quit [Remote host closed the connection]
victhor has joined #linux-sunxi
tl_lim has quit [Ping timeout: 276 seconds]
leviathan has quit [Remote host closed the connection]
tl_lim has joined #linux-sunxi
scream has quit [Remote host closed the connection]
tl_lim has quit [Ping timeout: 240 seconds]
tl_lim has joined #linux-sunxi
clemens3 has quit [Ping timeout: 240 seconds]
lerc has quit [Read error: Connection reset by peer]
lerc has joined #linux-sunxi
fdcx has quit [Ping timeout: 264 seconds]
hanetzer has joined #linux-sunxi
tl_lim has quit [Ping timeout: 276 seconds]
tl_lim has joined #linux-sunxi
xerpi has quit [Quit: Leaving]
hanni76 has quit [Quit: Leaving]
fredl has quit [Ping timeout: 240 seconds]
hanni76 has joined #linux-sunxi
fredl has joined #linux-sunxi
tl_lim has quit [Quit: Leaving]
Kwiboo has quit [Quit: .]
tllim has joined #linux-sunxi
Kwiboo has joined #linux-sunxi
Kwiboo has quit [Client Quit]
Kwiboo has joined #linux-sunxi
Kwiboo has quit [Client Quit]
Kwiboo has joined #linux-sunxi
<hanetzer>
lvrp16: heyo, any news regarind hisi?
<lvrp16>
what about hisi?
<lvrp16>
960 970?
hanni76 has quit [Quit: Leaving]
mavkhimenia has joined #linux-sunxi
<lvrp16>
hanetzer:
<hanetzer>
lvrp16: you had mentioned that there was some outsourcing done to get the hi3559a upstreamed a while back