Turl changed the topic of #linux-sunxi to: Allwinner/sunxi development discussion - Don't ask to ask. Just ask and wait! - See http://linux-sunxi.org | https://github.com/linux-sunxi/ | Logs at http://irclog.whitequark.org/linux-sunxi
arokux2 has quit [Remote host closed the connection]
<Turl> libv: a23 without adb? seriously? :\
<Turl> libv: how are app developers supposed to use such a device? does it ship with play store et al?
<bfree> hahaha, $15.99 for the card, $43.35 for the shipping, $52.14 for "duties and tax" for a grand total of $111.48 :-p $59.34 "pay on delivery" and take care of duty/tax myself (p.s. that was to Ireland, not brazil or russia or anywhere "special")
Gerwin_J has quit [Quit: Gerwin_J]
pepee has quit [Quit: goto Quit;]
<bsdfox> libv, can you get any output on the uart using terminal emulator?
tomboy64 has joined #linux-sunxi
egbert has quit [Disconnected by services]
egbert_ has joined #linux-sunxi
cajg has quit [Ping timeout: 260 seconds]
pwhalen has quit [Ping timeout: 245 seconds]
prasannapete has joined #linux-sunxi
cajg has joined #linux-sunxi
pwhalen has joined #linux-sunxi
pwhalen has joined #linux-sunxi
pwhalen has quit [Changing host]
parabyte has quit [Remote host closed the connection]
drachensun_ has quit [Ping timeout: 248 seconds]
windyan has joined #linux-sunxi
hipboi has joined #linux-sunxi
HdkR has quit [Read error: Connection timed out]
drachensun_ has joined #linux-sunxi
HdkR has joined #linux-sunxi
tomboy64 has quit [Ping timeout: 248 seconds]
TheSeven has quit [Disconnected by services]
[7] has joined #linux-sunxi
Md has quit [Quit: si hoc legere scis nimium eruditionis habes]
tomboy64 has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
pwhalen has quit [Ping timeout: 240 seconds]
prasannapete has quit [Quit: prasannapete]
pwhalen has joined #linux-sunxi
Quarx has joined #linux-sunxi
LoveMHz has quit [Ping timeout: 246 seconds]
biniyas has joined #linux-sunxi
<biniyas> Can any body tell me how can I activate hibernation for sunxi board ?
rellla has joined #linux-sunxi
wingrime has joined #linux-sunxi
rellla2 has joined #linux-sunxi
rellla has quit [Ping timeout: 245 seconds]
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has quit [Changing host]
<oliv3r> biniyas: don't think that's implemented yet
wolfy has joined #linux-sunxi
BluesBoy has joined #linux-sunxi
biniyas has quit [Quit: Page closed]
<oliv3r> libv: isn't it automatically generated when you use the tag on something that doesn't exist yet?
<oliv3r> category{sun8i} or whatever it is
<oliv3r> libv: yeah it does create the (empty) category, you can simply edit it from there
<oliv3r> hno: could a13/a23's do 1 gb of memory if there would be chips with those densities? Or won't it ever work due to the address line thing?
gzamboni has quit [Ping timeout: 265 seconds]
n01 has joined #linux-sunxi
gzamboni has joined #linux-sunxi
Quarx has quit []
_massi_ has joined #linux-sunxi
<oliv3r> wens: if your around, let me know so I can test gmac on cb2; i can't get it to work, though even the led seems to blink;
atiti has quit [Ping timeout: 265 seconds]
<ganbold_> oliv3r: I think there is no gmac on cb2
<wens> oliv3r: hi
<oliv3r> ganbold_: cb2 has GMAC, but a ethernet PHY
<ganbold_> yeah, right
<oliv3r> you can use the new GMAC driver over MII on the CB2 :)
<wens> oliv3r: can you try manually setting ethaddr, ipaddr, netmask, then try ping?
<oliv3r> i did, but no ping reply
<oliv3r> i checked for an unused IP address on our work network
<oliv3r> (in the same subnet as my test device)
<oliv3r> i have the following setup company network port -> gbit switch -> port 1 'uplink', port 2 is my test pc, port 3 is CB2
<oliv3r> so my test pc and the cb2 are directly connected so to say
<wens> sounds good
<oliv3r> my test pc has as IP inet addr: Bcast: Mask:
<oliv3r> i gave my cb2 as IP addr; same netmask
<oliv3r> and as mac 12:23:34:45:56:67
<wens> and using emac works, right?
<oliv3r> i know it's not the best mac in the world; but it should be quite unique to this switch :p
<wens> should work
<wens> don't get it :(
<oliv3r> it might be my network
<oliv3r> can you get me a binary that I can put on an SD card?
<oliv3r> just so i can test that ti's not my build
<wens> sure
<wens> got the wifi chip to respond
<wens> device id says broadcom 43362
<oliv3r> oh very cool
<oliv3r> wens: that works
<oliv3r> ping says its alive
<oliv3r> dhcp now works too
<wens> :)
<oliv3r> so there i go trying dhcp all day yesterday
<oliv3r> right
<oliv3r> then
<oliv3r> i did a git pull from sunxi master; then rebased your v3 ontop of that
<oliv3r> i'll checkout your v3 directly and compile that
<oliv3r> hmm
<oliv3r> yeah i'll do that
<oliv3r> restart the chain
<wens> my v3 is over linux-sunxi or hans' sunxi branch
<wolfy> oliv3r: try using arping from the test pc to the cb2
<wolfy> and monitor the traffic using tcpdump
<wolfy> if you get no arp replies ...
<wens> it's most likely something in the hardware wasn't setup correctly, likely the pins
<wolfy> if they are connected to the same switch, at least arp level should be functional
<wolfy> assuming all drivers work
<oliv3r> wolfy: arp table was empty, but remember, this is a new driver i'm trying to test ;) so all might fail
<oliv3r> wens: i doubt it was the wrong u-boot bin, as something old should have the emac driver still
<oliv3r> wens: or, i used your v3, rebased it, but never compiled it! though i doubt i did that
<oliv3r> hmm, i have to build it twice
<wens> always make distclean with u-boot :p
<wens> no response from bluetooth :(
<oliv3r> wens: i build out of tree, so i do rm -r build/u-boot :)
<oliv3r> even cleaner then distclean ;)
<oliv3r> ok, step 1 complete, i can build your tree; time to cherry pick your patches
<oliv3r> or rebase probably
<oliv3r> still fails on first build; got to run build 2x
<oliv3r> maybe it's just more verbose and i miss things
<oliv3r> moving on
<oliv3r> ok now your v3 tree works rebased ontop of sunxi; good
<oliv3r> wens: the bigger question si now, the designware specific patches; we really should be pushing those via upstream, don't you think?
atiti has joined #linux-sunxi
<oliv3r> testing cb3 now but i expect that to just work
JohnDoe_71Rus has quit [Ping timeout: 240 seconds]
FR^2 has quit [Quit: Connection reset by peer]
JohnDoe_71Rus has joined #linux-sunxi
<oliv3r> wens: do you think it's technically possible to have 1 u-boot with both net drivers in it?
<oliv3r> ENET Speed is 1000 Mbps - FULL duplex connection
<oliv3r> DHCP client bound to address
<oliv3r> works :)
<oliv3r> gonna setup tftp server to test that too
<wens> oliv3r: yeah, the designware stuff should be upstreamed
<oliv3r> wens: did anybody else review the patches?
<wens> oliv3r: as for both net drivers, I'm sure it would work. it's just the board setup varies
<oliv3r> well i just noticed that I accidentally used the cb2 u-boot on cb3 and it worked (besides net of course) hence the brainfart
<wens> oliv3r: Turl started to look at it. hans maybe, not sure.
<oliv3r> but then, who decides what PHY's get included etc
<oliv3r> ok i'll try to go over them and understand them, but I assume the gross of the patches are disgnware stuff; so the indepth review should come from upstreaming that
<wens> oliv3r: the pins/lines are seperate, only one mac gets the correct MDIO bus
<oliv3r> probably not easily auto detectable
<oliv3r> you'd need either som rom somewhere that stores that
<wens> the phy stuff is auto-probed
<wens> which mac gets the phy, depends on which pins the phy is connected to, and for PA*, whether gmac or emac is active.
<wens> not sure how u-boot handles a mac without a phy, though
notmart has joined #linux-sunxi
<wens> oh, and the u-boot patches are sunxi specific, all but one.
<oliv3r> wens: i'll look for them on the ML and try to review them; then push it upstream
<wens> cold today.. 12 degrees C, but no heater
<oliv3r> wens: i was just thinking if we can have a config struct embedded in the binary, that tells u-boot what PHY and how it is connected
<oliv3r> 9C, kinda warm :p
<wens> oliv3r: upstream all of it? I never sent it to the ML as a patch series
AreaScout has joined #linux-sunxi
FR^2 has joined #linux-sunxi
<oliv3r> wens: oh, so how will we review it!
<oliv3r> and i'll push it to sunxi repo
<oliv3r> hmm, that's pretty silly, on arm, the default tftp filename is boot\x86\pxeboot.com
<oliv3r> probably make that arm, and maybe reverse the slashes :D
jinzo has joined #linux-sunxi
jinzo has joined #linux-sunxi
<wens> oliv3r: i'll push it to ML today
lioka has quit [Ping timeout: 252 seconds]
lioka has joined #linux-sunxi
<oliv3r> can't find the string :S
<oliv3r> wens: very cool
<hno> oliv3r, a13 is limited to 512MB from what I remember.
<oliv3r> hno: but where is that limitation from?
<oliv3r> hno: i first thought it was because of the maximum allowed number of chips
<wens> oliv3r: sent
<oliv3r> hno: but with a higher chip density, it could work? not that any body would look at a13 for those kinds of experiments now; but a23 seems to have the same limitation
<oliv3r> wens: lets pray ;)
<oliv3r> my patch got ignored, twice
<oliv3r> so still hopeing for hno to weight in on it whne he has more time
<atsampson> oliv3r: argh, sorry, those merges shouldn't have been in that pull request -- I'll redo them -- apologies
<oliv3r> atsampson: ;)
<atsampson> it should just have been the N90 stuff
<oliv3r> also double check the machine name ontop of the fiel
<atsampson> well, the machine name is obviously wrong, but that's how it was already and that's what it was set to on mine as well ;)
<atsampson> it's not really clear what the machine is actually called -- is it an Ampe A90 or an Anpei A90 or a Sanei N90 or an M9701 or...
<oliv3r> heheh
<oliv3r> true
<oliv3r> but it's deff. not an 'allwinner evaulation board' :p
<atsampson> the kernel name had "crane-anpei" in it so I suspect Ampe/Anpei would actually be the best bet
<oliv3r> yeah it doesn't have to be perfect
<oliv3r> if it is in relation to the iwki page, that's even better
Montjoie has quit [Remote host closed the connection]
juanfont has left #linux-sunxi ["Saliendo"]
juanfont has joined #linux-sunxi
Montjoie has joined #linux-sunxi
<atsampson> oliv3r: have another look now (https://github.com/linux-sunxi/sunxi-boards/pull/14) -- I've added one more cleanup (csi1) that I had sitting around
<oliv3r> merged
xeros has quit [Ping timeout: 245 seconds]
arokux1 has joined #linux-sunxi
arokux has quit [Read error: Connection reset by peer]
xeros has joined #linux-sunxi
<oliv3r> god atftpd is annoying, it refuses to start, but does not tell you why whatsoever
hipboi has quit [Ping timeout: 252 seconds]
<n01> strace
arokux has joined #linux-sunxi
arokux1 has quit [Read error: Connection reset by peer]
<atsampson> oliv3r: cheers :)
<oliv3r> yeah installing as we speak;)
<oliv3r> but this is a very very slow 250 Mhz Geode :p
<oliv3r> i don't see it :(
<oliv3r> anybody noticeing it?
naobsd has joined #linux-sunxi
<n01> oliv3r: it looks like you are passing a wrong argv (it prints "usage: ..")
<oliv3r> yeah but it's copy pasted :p
<oliv3r> --no-fork that is
<oliv3r> lets try without any arguments
<libv> oliv3r: thanks for creating the category :)
<oliv3r> libv: :p
<atsampson> oliv3r: looking at the source, you need to pass --daemon as well as --no-fork (else it exit(1)s if stdin is a tty)
<atsampson> also, it handily reports later errors to syslog...
<oliv3r> atsampson: heh; thanks for checking that :D
<libv> oliv3r: so you just typed: Category:...?
<libv> or is there some info somewhere
<oliv3r> atsampson: that was the key :)
<oliv3r> libv: no, you added the category tag, and that created the (empty) category, I just clicked on the red a23-tablet category, which brings up the edit page for that category
<libv> aah
<oliv3r> there's quite a few empty ones
<libv> right :)
<oliv3r> wens: tftp works on cb3; for completeness i'll test cb2 with tftp too
<oliv3r> a nother u-boot oddity; when issueing the dhcp command, it requests a dhcp address, but doesn't set that add in the ipaddr var; so doing tftp 0x123 uImage fails due to missing ipaddr :S
<oliv3r> and bootfile doesn't override boot\x86\pxeboot.com ...
naobsd has quit [Ping timeout: 272 seconds]
<wens> oliv3r: :)
<oliv3r> wens: you happen to know where the default bootfile env. is set? i can't find it with grep strangely
<oliv3r> (in our u-boot source tree)
<oliv3r> wens: cb2 works too; pushing your patch now
<oliv3r> couldn't sunxi: fix gpio drive mask have been done with a define? for prettier code? :)
<oliv3r> wens: could you possibly do turls suggestion and do a sunxi_gmac_initialize() for arch/arm/cpu/armv7/sunxi/board.c
<wens> oliv3r: default env is in include/configs/sunxi-common.h
Black_Horseman has quit [Quit: Zwi se logou mou!!!]
<oliv3r> wens: yeah but i odn't see bootfile defined there, so it uses some built in default, but i don't find it; anyway, i'll add a default Filename :)
kz1 has joined #linux-sunxi
rz2k has joined #linux-sunxi
\\Mr_C\\ has joined #linux-sunxi
<oliv3r> wens: i added +#define CONFIG_BOOTFILE "uImage" /* Boot file name */
<oliv3r> but gets completly ignored
<oliv3r> it should be CONFIG_BOOTP_
<oliv3r> i copied from the one board that does it worng
<wens> best read u-boot's README
<wens> tons of options
<oliv3r> yeah i have
<oliv3r> but it isn't documented :)
<oliv3r> and the env var is ignored it seems
<oliv3r> i do env print bootfile; it prints that boot\x86\pxeboot.com; i change that with env set, that works fine, but dhcp still uses the old bootfile
kivutar has quit [Ping timeout: 264 seconds]
<oliv3r> bootp_bootpath doesn't do it either; very weird; but a minor thing
<kz1> does the sunxi uboot have suuport for adb reboot bootloader command?
<oliv3r> of course not
<oliv3r> there's no android or adb stuff in u-boot
<oliv3r> android however, may accept that command and boot to u-boot
<libv> ok, so cydia impactor is also a dead end as the q8h doesn't do adb properly
<oliv3r> libv: crap :(
<kz1> ok, but will the sunxi version of uboot boot to the bootloader?
<oliv3r> kz1: of course not
<oliv3r> kz1: u-boot IS the bootloader
<kz1> the problem I have is that if I call adb reboot bootloader it boots to the system not to fastboot
<kz1> the method I have been told to use is with UART at boot
<oliv3r> kz1: your bootloader probably skips really fast passed that and continues to boot
<kz1> but that is not an option for my devices
<oliv3r> kz1: quite common default behavior, also sunxi bootloader doesn't do anything graphical nor offer you some sort of upload menu
pwhalen has quit [Read error: Connection reset by peer]
<oliv3r> kz1: what are you planning to do
<kz1> I need to be able to get to fastboot from the adb command.
<oliv3r> kz1: to do what
<kz1> to run fastboot commands :-) update the firmware etc...
<oliv3r> kz1: you probably better start reading our wiki at linux-sunxi.org
<oliv3r> kz1: as you don't need fastboot commands for any of that
<oliv3r> kz1: allwinner updates by default their images via livesuit
<oliv3r> kz1: to update only the android portion you'd run CWM or whatever comes preinstalled with your device
jinzo has quit [Read error: Operation timed out]
<oliv3r> kz1: the allwinner based devices don't do 'fastboot' the only thing that even comes close to that is 'FEL' (which is not what you want)
<kz1> we are using the lichee kernel and official 4.2.2 sdk from allwinner
<kz1> However we need to be able to batch update a lot of devices at the same time
<oliv3r> kz1: then you are asking the wrong place :)
<kz1> we would like to use fastboot
<oliv3r> kz1: this is about the sunxi kernel which is at 3.4
<oliv3r> kz1: there is no fastboot, it does not exist for allwinner based devices
<oliv3r> kz1: you really have to talk to allwinner about allt his
<oliv3r> the allwinner way is to either use livesuit from a PC
<kz1> The problem is that we cannot get the devices to fastboot mode and Allwinner are not resonding so I am llooking at fixing uboot myself using the linux -sunxi version
<kz1> according to allwinner fastboot is accessible via uart
<oliv3r> kz1: we don't have fastboot
<oliv3r> you can access u-boot via the uart
<kz1> ok, so the linux-sunxi tools will not be useful for batch firmware updates ?
<oliv3r> far from
<oliv3r> so no
<oliv3r> which tool specifically where you looking at?
<oliv3r> sunxi-tools are for working with FEL mode mostly (or some of them are)
<kz1> basically I would like to replace phoenixsuit and run a batch update on a linux system
<oliv3r> and yeah, in the future we'd want to use those for that
<kz1> I should be able to get some funds if you think it willbe possible
<oliv3r> kz1: you can boot linux via fel mode and run whatever from there (so automated scripted stuff)
<oliv3r> you could boot a workiung linux kernel with initramfs, that overwrites nand
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
<kz1> The other option we have is OTA updates but I would prefer to enhance the sunxi tools if you think it is viable
<oliv3r> if you have the skills and the time, sure that's viable
<oliv3r> what we currently can do, is create an SD card, that always forces a device into FEL mode
<oliv3r> so for forcing lots of devices into fel mode, that is one easy way
<oliv3r> (fel-sdboot); should work on all sunxi devices
<oliv3r> then, you can use fel mode to upload stuff directly into the memory into the device
<oliv3r> e.g. a known working kernel + an initramfs
<oliv3r> if you make your initramfs so that it formats, downloads and writes a partition
<oliv3r> you'd be pretty much done
<oliv3r> you can push the kernel etc via USB; so that makes it easy-ish
<oliv3r> but i guess making an SD card with said kernel + initramfs might be easier?
<kz1> the issue is we cannot access the internals of the devices so no hot swappable sdcards
BluesBoy- has joined #linux-sunxi
<oliv3r> ok then FEL mode is what you want
<oliv3r> as even tftp may not be useable?
<oliv3r> (no wired ethernet?)
BluesBoy has quit [Ping timeout: 264 seconds]
BluesBoy- is now known as BluesBoy
<kz1> unfortunately no ;-)
<kz1> only usb
<oliv3r> ok then you need fel mode
<oliv3r> if possible, kernel + wifi then pull data via wifi
<kz1> yeah we have looked at the otA option but our techs prefer to use a linux push method reather than the OTA method
<oliv3r> ok, yeah you can push via FEL quite easily
<oliv3r> http://linux-sunxi.org/FEL/USBBoot that should tell you everything you want
<oliv3r> step 1 being the trickiest one
<oliv3r> you can not directly write to nand yet though
<oliv3r> well you could, but we don't have that code in place yet
<kz1> what would it take (approx) to get boot to FEL directly from USB working?
<oliv3r> you can in theory, write a fel-program, that gets data pushed from the USB side, and writes that onto the nand; but we dno't have that written yet
<oliv3r> A10 only boots into fel mode if one of the following criteria are met:
<oliv3r> you push the FEL button during boot
<oliv3r> you push a key combination during boot
<oliv3r> volume up or some such
<kz1> yep, we dont have any buttons on our devices
<kz1> there is a reed switch for on/off
<kz1> :-)
<kz1> just to make things easy
<oliv3r> i don't think you can force the A10 into fel mode via USB
<oliv3r> if sd, nand, mmc and SPI-NOR all fail to boot, it will always go into FEL mode
<kz1> essentially we need to replace phoenixsuit with a linux version that can do batch updates to multiple devices
<oliv3r> you could make that happen, if you can force your devices into fel mode
<kz1> how does phoenixsuit do it?
<wolfy> a large cricket bat !
<oliv3r> if you have root on your active device, you could force the CPU to execute from 0xfffff0000 iirc (that's fel mode if i'm not mistaken)
<kz1> yes we do have root
<oliv3r> kz1: it doesn't, it forces you to either put the device into fel mode (often via sd)
<oliv3r> kz1: since a corrupt flash (e.g. the first sector broken) forces fel mode also, you can always recover
<kz1> btw, these are a20 devices
<oliv3r> that should be okay
<oliv3r> kz1: a10, a10s, a13 and a20 all work identically with regards to FEL mode
<oliv3r> kz1: so yeah, run 'update.sh' or something, that forces the CPU to execute instructions at 0xffff0000; that should be run FEL mode in theory anyway as FEL mode is actually inside the CPU (BROM, see wiki)
<oliv3r> once in fel mode, you can upload anything to USB, but if you want to write directly to the nand, you'd have to write that bit yourself, quicker is possibly boot a linux kernel +initramfs, that then takes the data either from USB or via wifi
<oliv3r> if you don't have wifi either; then probably only via usb ;)
<oliv3r> dunno how scriptable it is
<oliv3r> since you'd have to wait for the USB to boot, wait for a device to show up that you can mount etc
<kz1> That's really helpful. So it sounds like that is very similar to the phoenixsuit method
<oliv3r> yeah kinda obviously ;)
<oliv3r> I suppose you could implement fastboot in u-boot, I don't know if u-boot does fast boot at all
<oliv3r> googling for it suggest it does
<oliv3r> but since we don't have a complete mtd/flash driver yet, that's a little low on the priority list
<kz1> sure.
<kz1> what would it take to write the code to write to nand ?
<kz1> approx?
<oliv3r> it depends, it'll require some effort
<oliv3r> we HAVE mtd driver both for u-boot and the kernel
<oliv3r> they require some cleanup but we do have that
<oliv3r> there is however a problem/bug with OOB that isn't implemented yet, so you can only use a flash filesystem that doesn't use OOB's, UBFS is good, yaffs2 not as it requires oob
<oliv3r> if you'd fix that, you can then integrate the mtd driver into the kernel and u-boot and then I suppose it's just a matter of enabeling fastboot in the u-boot compile
<kz1> hmm, so that would exclude the allwinner images
<oliv3r> as those are all based on the internal allwinner nand driver, yeah
<oliv3r> you could take the lichee allwinner nand driver from u-boot that we have somewhere, and try to enable fastboot there; but it's an old messy code base
<oliv3r> the other option would be to take the allwinner lichee driver and see if you can make something of that for FEL mode
<oliv3r> so the code does exist I suppose
<oliv3r> a skilled dev will probably need a few days to get it working possibly
popolon has joined #linux-sunxi
popolon has joined #linux-sunxi
<kz1> would you have some time for that if the funds were available?
<oliv3r> kz1: unfortunatly no
<oliv3r> kz1: i don't have the skills/knowledge and desire to open that can of worms :p
<kz1> :-) fair enough
<kz1> what about writing to nand over usbfel ?
<oliv3r> you need the nand driver that i mentioned above
<oliv3r> so you need to incoperate that
<oliv3r> once you have that, anything is game
<kz1> thats from the lchee driver or uboot ?
<oliv3r> lichee-kernel and lichee-u-boot :p
<oliv3r> so you would want lichee-u-boot
* kz1 vomits
* kz1 punches self in face
hansg has joined #linux-sunxi
<oliv3r> hey hans
<oliv3r> wens: just done testing wens's patches for gmac; works on cb2 and 3 so want to push them to sunxi u-boot; just needs a little cleanup
<oliv3r> puneet B is at it again :S
<oliv3r> that guy is so frustrating and lazy to read
ganbold__ has joined #linux-sunxi
popolon has quit [Quit: Quitte]
kivutar has joined #linux-sunxi
<hansg> oliv3r, hi. I've been looking at wens u-boot patches too, but I don't have any test setup for testing *mac under u-boot (ie tftp) and I'm too lazy to set it up, so I'm glad you're doing it. Have you discussed pushing this with hno ?
<hansg> oliv3r, I'm fine with you pushing it once you're happy. Having this in the main u-boot-sunxi repo would be great.
<oliv3r> i replied to wens's patches
<oliv3r> wens sent the designware specifics upstream
<oliv3r> hansg: i agree; i specifically brought a gbit switch and an old thinclient (as linux box) to work to set it all up and test
<oliv3r> but yeah, i can dhcp, tftp download data, so all looks well functional
<oliv3r> hansg: once my comments are addressed or debunked; i'll happily push it all
<oliv3r> hansg: i saw you pushed emilio's fast_mboot patches, was gonna push it all
<hansg> oliv3r, yeah, with hno seemingly a bit inactive atm and no negative feedback + my ack, I thought it would be good to just push emilip
<hansg> * emilio's * patches
<oliv3r> hansg: yeah i was a little slow; I did test them, and wanted to push them together with wens's stuff, but i couldn't get it to work so was delayed; but alls good :)
<oliv3r> progress == win
nove has joined #linux-sunxi
<wens> inbox just got blasted with 49 patches from TI
<oliv3r> ouch loo
<oliv3r> i unsubbed from u-boot ML as those where really just patch bombs
<wens> nope. LAKML
<oliv3r> ah
<oliv3r> same :p
<oliv3r> too noise a list
<wens> oliv3r: designware stuff hasn't been sent upstream yet. I will push that seperately
<wens> oliv3r: didn't think it too good to blast u-boot ML with all the sunxi stuff
<oliv3r> wens: no it's not :p
<oliv3r> Turl: the ahb gates (0x60) for A20 are they accurate? and are those the same ones on A10?
<oliv3r> i notices they don't match the wiki; which is plausible, just oduble checking
<wens> oliv3r: about picking out the gmac init code, and the defs, since it's jemk's patch, I would prefer to add another patch over it, instead of squashing the fixes
<wens> oliv3r: what do you think?
<oliv3r> wens: personally, i prefer the squash; dunno what hno prefers :)
<oliv3r> but a cleanup patch is good to add to that
<hno> hansg, oliv3r there is no need to ask me if you think a patch is fine.
<wens> oliv3r: i prefer to keep other people's stuff intact, unless the author says otherwise
<oliv3r> wens: that is true; best would have been for jens to fixup his patch i guess ;)
<oliv3r> wens: i can do the define patch ontop of it all too if you want
FR^2 has quit [Quit: Connection reset by peer]
<wens> ok, i'll add a patch over it.
pwhalen has joined #linux-sunxi
pwhalen has joined #linux-sunxi
pwhalen has quit [Changing host]
<nove> wingrime, ping
Gerwin_J has joined #linux-sunxi
<oliv3r> wens: you raise a valid point with gate 17; since none of the other reg1 gates are listed there
<oliv3r> wens: so having it as a dupe is good; but having it in order is still better maybe?
<wens> oliv3r: maybe adding a header comment to it is better?
<oliv3r> well there's really no logical reason why we want to seperate them
<oliv3r> they are seperated in the chip, because there's only 32bit registers, if there where 64bit regs, then it wouldn't have mattered :)
<oliv3r> but a comment at the least makes it less confusing to go over it
<wens> btw, pulling the gmac init code into a seperate function that's still in the same file, is this what you want?
<oliv3r> but yeah, adding a comment header 'block 0; block 1; is usefull
<wens> the emac code puts the pin mux stuff in the driver. not really good :(
<oliv3r> wens: hmm, goog question; sunxi_emac_init is in sunxi_emac.c; so you can't put it there obviously :p
<oliv3r> maybe create a small sunxi_gmac.c that acts as a shim for the gmac related stuff
<oliv3r> wens: yeah all of u-boot-sunxi needs quite some cleanup
wolfy has quit [Quit: Statisticile sunt precum costumele de baie:iti arata ce stie toata lumea si iti ascund exact ceea ce ai fi vrut sa afli. (Georges Vedel)]
<wens> oliv3r: I think it belongs with the SoC code though.
<wens> oliv3r: about ahb reg1, none of the devices controlled by it, except gmac, would be used by u-boot
<wens> oliv3r: so none of them are defined
<oliv3r> hdmi will be in the (far) future, i think wingrime was looking in adding some basic framebuff output support to u-boot at some point
<oliv3r> but there comes the question, define all we know, or only what we use
<oliv3r> i personally prefer complete defines tbh; but i'm in no place to answer that
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
FDCX has quit [Read error: Operation timed out]
bsdfox has quit [Ping timeout: 264 seconds]
bsdfox has joined #linux-sunxi
bsdfox has quit [Changing host]
bsdfox has joined #linux-sunxi
FDCX has joined #linux-sunxi
<libv> hah
<libv> i have adb.
<oliv3r> YAY!
hansg has quit [Quit: Leaving]
<oliv3r> hometime
<wens> oliv3r: done with the modifications, pushed to my tree, will resend tomorrow
<drachensun_> I'm trying to setup the sunxi-bsp with linaro alip 13.04 looks like its hanging on an X config issue
<drachensun_> anyone run into that?
rellla2 has quit [Quit: Nettalk6 - www.ntalk.de]
<libv> haha!
<libv> root!
<wens> libv: congrats!
AreaScout has quit [Ping timeout: 250 seconds]
dwilkins has quit [Read error: Connection reset by peer]
aep has quit [Read error: Connection reset by peer]
kivutar has quit [Ping timeout: 246 seconds]
drachensun_ has quit [Ping timeout: 260 seconds]
dwilkins has joined #linux-sunxi
rz2k has quit [Remote host closed the connection]
<wens> what do you know, found datasheet for BCM43362
HeHoPMaJIeH has quit [Remote host closed the connection]
drachensun_ has joined #linux-sunxi
atiti has quit [Read error: Operation timed out]
<libv> ooh, 552MHz dram
<libv> mali should clock up to 600MHz, crickey
<libv> 1.5GHz cpu
n01 has quit [Ping timeout: 248 seconds]
pwhalen has quit [Ping timeout: 246 seconds]
n01 has joined #linux-sunxi
atiti has joined #linux-sunxi
woprr_ has joined #linux-sunxi
pwhalen has joined #linux-sunxi
wingrime has quit [*.net *.split]
[7] has quit [*.net *.split]
rwmjones has quit [*.net *.split]
woprr has quit [*.net *.split]
tomboy64 has quit [Ping timeout: 245 seconds]
<bfree> libv (or anyone): it may be some other pebkac (kernel command line arguments maybe?), but I've finally got myself new sunxi-3.4 kernel packages to test but I'm just getting a blank screen (not no signal though). I suspect I might be loading the modules in the wrong order? I'm loading them like http://linux-sunxi.org/Initial_Ramdisk which worked back when I wrote that, any ideas?
[7] has joined #linux-sunxi
wingrime has joined #linux-sunxi
BluesBoy has quit [Quit: HydraIRC -> http://www.hydrairc.com <- Nine out of ten l33t h4x0rz prefer it]
wingrime has quit [*.net *.split]
[7] has quit [*.net *.split]
aep has joined #linux-sunxi
AreaScout has joined #linux-sunxi
[7] has joined #linux-sunxi
wingrime has joined #linux-sunxi
[7] has quit [Disconnected by services]
TheSeven has joined #linux-sunxi
wingrime has quit [*.net *.split]
n01 has quit [Read error: Operation timed out]
wingrime has joined #linux-sunxi
jinzo has joined #linux-sunxi
jinzo has joined #linux-sunxi
jinzo has quit [Changing host]
notmart has quit [Quit: notmart terminated!]
jinzo has quit [Ping timeout: 250 seconds]
hramrach has quit [Ping timeout: 240 seconds]
jinzo has joined #linux-sunxi
jinzo has quit [Changing host]
jinzo has joined #linux-sunxi
<oliv3r> wens: i'll pull your tree then :)
<oliv3r> drachensun_: what make command do you run? i admit i don't do more then make; or make u-boot or make dtb or make linux
<oliv3r> libv: success!
<oliv3r> libv: with 552 MHz dram; that could mean it outperforms all < sun8i's
<oliv3r> to bad it lacks hdmi
<oliv3r> libv: the CPu truely at 1.5GHz?!
<drachensun_> oliv3r: scripts/sunxi-media-create.sh
<oliv3r> ah, isn't that one of the make things?
<oliv3r> drachensun_: invoked from the make file
<wens> wifi tree is slightly cleaned up
<wens> still can't get bt to respond
<wens> wifi needs support from broadcom guys
<wens> bedtime
theskilledworker has joined #linux-sunxi
netlynx has joined #linux-sunxi
netlynx has quit [Changing host]
netlynx has joined #linux-sunxi
wingrime has quit [Read error: Operation timed out]
<drachensun_> oliv3r: Im not sure, I wasn't calling it that way
<drachensun_> I'm working on two machines in parallel, on the ubuntu machine it doesn't happen
rz2k has joined #linux-sunxi
<drachensun_> but it does on my mint desktop
<drachensun_> however I have a bigger issue, X is crashing for one machines cards and not the others, I thought they were identical :/
<drachensun_> pretty neat tool though, do the livesuit images work? have you tried them?
<Turl> mripard: give "Can someone Ack and queue a patch for RTC subsytem?" a read when you get a chance
<Turl> (on lakml)
<Turl> bfree: I thought they were all =y these days?
<bfree> Turl: not in my kernel package, which is a very very modular kitchen sink config ;)
<Turl> oliv3r: they're probably not; check the user manuals :p
<Turl> bfree: do you have serial to debug?
deasy has joined #linux-sunxi
<bfree> Turl: yes, have uart hooked up (hence know it's alive just not displaying on hdmi)
<Turl> is there any error or messages regarding hdmi on dmesg?
<bfree> there are related messages, some look like warnings not errors ... one sec, installing openssh-client so I can easily post the dmesg from the cubie2
<Turl> microcom :)
<nove> does a libjpeg-sunxi makes sense? is that the hardware appears to only be able to encode/decode mjpeg type jpegs
<oliv3r> libv: those performance improvments make it a very interesting tablet SoC; to bad there's no 1 GiB tablets with that yet
<bfree> or just remember to add openssh-client to my rootfs builder ;) or something to replace my sata psu which seems to have died so it's not crawling along on the sd :-p http://dl.linux-sunxi.org/users/niall/dmesg.sun7i
<Turl> nove: would it be faster than libjpeg-turbo?
_massi_ has quit [Quit: Leaving]
<oliv3r> Turl: if it's very similar, it'll be good; as it unloads the CPU
<bfree> Turl: easier read: http://paste.debian.net/71818/
<Turl> bfree: I don't see anything horribly wrong tbh
<Turl> bfree: did you start x or load the framebuffer console module to actually show something?
<bfree> Turl: no X ... load the framebuffer console module?
<nove> Turl, in a A13 i got 18fps with 2048x2048
<Turl> fbcon.ko or something like that
<Turl> also you may want an extra console=tty0 on your cmdline
* bfree <3 Turl
<bfree> fbcon.ko gets me a display
pwhalen has quit [Ping timeout: 240 seconds]
<oliv3r> drachensun_: the bsp with the livesuit images? i'd expect it to not work
<Turl> bfree: :) great
\\Mr_C\\ has quit [Quit: .]
pwhalen has joined #linux-sunxi
pwhalen has joined #linux-sunxi
pwhalen has quit [Changing host]
<bfree> my sun4i (initramfs) isn't bringing up /dev/mmcblk0p1, not sure if I've chosen the wrong mmc in the config or if I've to tweak what I'm putting into /etc/initramfs-tools/modules or ??? but heading out too soon to dive in now ... http://dl.linux-sunxi.org/users/niall/debian.preview/incoming/ updated (my sunxi-boot->sunxi-update fixed a little for when changing boards and above kernel added)
<bfree> p.s. ftr above packages are built on/for debian unstable
nove has quit [Quit: nove]
atiti has quit [Ping timeout: 245 seconds]
tomboy64 has joined #linux-sunxi
<jelly-home> can sunxi-mmc be built and work as module at all?
jinzo has quit [Ping timeout: 252 seconds]
n01 has joined #linux-sunxi
FR^2 has joined #linux-sunxi
jinzo has joined #linux-sunxi
fredy has quit [Excess Flood]
bluse has quit [Read error: Connection reset by peer]
bluse has joined #linux-sunxi
fredy has joined #linux-sunxi
netlynx has quit [Quit: Leaving]
<oliv3r> jelly-home: possibly; but not sure
n01 has quit [Ping timeout: 240 seconds]
n01 has joined #linux-sunxi
HdkR has quit [Read error: Connection timed out]
theskilledworker has left #linux-sunxi ["Leaving"]
n01 has quit [Ping timeout: 264 seconds]
n01 has joined #linux-sunxi
rz2k has quit [Read error: Connection reset by peer]
tomboy64 has quit [Ping timeout: 240 seconds]
tomboy64 has joined #linux-sunxi
nedko has joined #linux-sunxi
HdkR has joined #linux-sunxi
maz_ has joined #linux-sunxi
rings_VII has joined #linux-sunxi
rings_IIV has quit [Ping timeout: 264 seconds]
popolon has joined #linux-sunxi
popolon has joined #linux-sunxi
n01 has quit [Remote host closed the connection]
maz_ has quit [Ping timeout: 250 seconds]
FR^2 has quit [Quit: und weg...]
pwhalen has quit [Ping timeout: 246 seconds]
woprr_ has quit [Quit: Ex-Chat]
pwhalen has joined #linux-sunxi
pwhalen has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
ganbold__ has quit [Remote host closed the connection]
woprr has joined #linux-sunxi
jinzo has quit [Remote host closed the connection]
ZetaNeta has joined #linux-sunxi
AreaScout has quit []