physis has quit [Read error: Connection reset by peer]
physis has joined #linux-sunxi
hansg has joined #linux-sunxi
ricardocrudo has joined #linux-sunxi
<wens>
hansg: do you have plans to add pinctrl/gpio support for axp?
<wens>
i know i volunteered to do it some time ago, but other interesting things came up
<hansg>
wens, not atm, I agree we will need it eventually, so if you can do it that would be great
tomboy64 has quit [Ping timeout: 256 seconds]
<hansg>
wens, I would like you to get rsb support to work in the kernel first though, as that together with the axp-vbus-det + otg patches I posted yesterday will enable us to do proper otg on a23 / a33 tablets
<hansg>
What is the status of your rsb work ?
<wens>
hansg: agreed
<wens>
hansg: haven't started with the next version, which is likely a custom bus
<hansg>
Ah
<wens>
i2c maintainer made it clear he wouldn't merge it
tomboy64 has joined #linux-sunxi
<hansg>
Right I saw those mails, but I was under the assumption that you had already started on the custom bus work
kaspter has quit [Remote host closed the connection]
<wens>
not yet, but i'll take spmi as an example to work with
kaspter has joined #linux-sunxi
<NiteHawk>
hansg: would it be useful to introduce CONFIG_SYS_64BIT_LBA as a u-boot default for some (or all) allwinner socs? SUNXI_AHCI seems to handle it nicely, and it's currently not the default. Kasreyn ran into trouble with that yesterday with a large (>2TB) disk on banana pi (sun7i/A20)
<rellla>
NiteHawk: i don't know how to solve that but i think, the "FAQ" is a mediawiki special page.
<rellla>
but i don't have the rights to maybe change that. just revert it :p
gianMOD has quit [Remote host closed the connection]
<NiteHawk>
rellla: i would, but i'm lacking rights to remove a page - i could transform that page into a "manual" redirect / link, but that's a bit pointless :)
<hansg>
NiteHawk, sounds like a good idea to enable that on all boards which enable SUNXI_AHCi, should be a simple matter of having a "selects SYS_64BIT_LBA" in the Kconfig, assuming that SYS_64BIT_LBA has been Kconfigified, can you submit a patch ?
<NiteHawk>
hansg: i'm not too familiar with it, that why i wanted your opinion as custodian on where it would have to go (sun7i/sunxi-common). i suppose sun7i alone isn't sufficient, as e.g. A10 has SATA/AHCI too?
<rellla>
NiteHawk: i deleted it.
<NiteHawk>
hansg: would it harm to enable it globally in sunxi-common.h? i'd expect those devices that don't offer AHCI to be unaffected
<NiteHawk>
rellla: thanks!
simosx has joined #linux-sunxi
<NiteHawk>
hansg: ah, okay - now i see your point. you mean: SUNXI_AHCI selected would auto-activate SYS_64BIT_LBA via Kconfig. that sounds neat / right
<NiteHawk>
hansg: a search for SYS_64BIT_LBA in all Kconfig came up empty, so it's probably not in the Kconfig system :/
sehraf has joined #linux-sunxi
<NiteHawk>
hansg: other platforms just set it globally via #ifdef, e.g. include/configs/x86-common.h
<hansg>
NiteHawk, ok, so then we need a #ifdef SUNXI_AHCI #define CONFIG_SYS_64BIT_LBA 1 #endif in include/configs/sunxi-common.h Thinking about this I think that SUNXI_AHCI itself also has not yet been converted to Kconfig ...
<NiteHawk>
yup. i'm just having a look at include/configs/sunxi-common.h - there's a #ifdef CONFIG_AHCI block that looks like it would be a good place to add CONFIG_SYS_64BIT_LBA
<NiteHawk>
(it also sets CONFIG_SCSI_AHCI and CONFIG_SCSI_AHCI_PLAT)
vpeter has quit [Quit: Konversation terminated!]
<hansg>
Ack, as for why not simply enable it always, I expect the CONFIG_SYS_64BIT_LBA setting to influence the u-boot block layer in general, making the code larger, and we do not need this on systems which only have an sdcard slot
gianMOD has joined #linux-sunxi
<hansg>
So if you can send me a patch for this, I can combine it with another pending bugfix, and send a pull-req so that this will go into the upcoming v2015.07 release
<NiteHawk>
hansg: as far as i understood it, basically it just redefines lbaint_t to be 64-bit
paulk-collins has joined #linux-sunxi
<hansg>
Right, but that probably gets used in a lot of places
<NiteHawk>
but i agree to the general idea: keep its use to a minimum
kaspter has quit [Remote host closed the connection]
<hansg>
NiteHawk, you don't need to be subscribed to send to it, please do a git send-email to me with the list in the Cc, then others will get a chance to comment on the patch, more eyes is always good
<NiteHawk>
hansg: ok, i'm new to that - so i'll have to setup things first and hope i get it straight...
<hansg>
NiteHawk, ok, learning howto use git send-email is always a good idea, as that is how most projects want to receive their patches.
<hansg>
Let me know if you've any questions
<hansg>
also please add a Signed-off-by line to the commit message
<hansg>
(just a random commit I had open in my browser)
<NiteHawk>
problem is i'm working with sunxi on my *nix machines which don't have mail accounts set up, and otherwise am on windows where i don't have git readily available :D :P
<bbrezillon>
hansg: ouhouh, I've been randomly chosen :-D
<NiteHawk>
but i'll try to jump through all the hoops... :)
<hansg>
NiteHawk, if you can at least do a "git commit" write a decent commit message including S-o-b, and then "git format-patch HEAD~" that would already help
<hansg>
Then send a mail with your regular mail client with the 0001-foo.patch generated by git-format-patch both inline and *attached* and I can pick it up that way. The attached part is necessary because mail-clients tend to mangle patches (which is why git-send-email is preferred_
<NiteHawk>
hansg: i'll try to straighten up my stuff accordingly and get things in place properly, i'll eventually get there
awe00 has joined #linux-sunxi
<plaes>
NiteHawk: msmtp
<libv>
NiteHawk: i have my git send-emails talk to my isp smtp directly
kaspter has quit [Remote host closed the connection]
<NiteHawk>
thanks for the tips :) as i said, it's all new to me, i'm sure i'll get it sorted out sooner or later
diego_r has joined #linux-sunxi
<simosx>
NiteHawk, if you take the Eudyptula Challenge (Linux kernel programming), you deal with emailing patches on Excercise 5.
<NiteHawk>
hansg: btw, any chance you would know about the current state of netconsole support? 2015.07-rc1 failed to compile for me with CONFIG_NETCONSOLE set (http://pastebin.com/wKbDsJba)
<NiteHawk>
hansg: i think you might want to fix that patch before committing it upstream. :/ it doesn't need "#define CONFIG_SYS_64BIT_LBA 1", "#define CONFIG_SYS_64BIT_LBA" alone should be sufficient
<hansg>
NiteHawk, I do not know about CONFIG_NETCONSOLE, I will fix the patch
<NiteHawk>
hansg: no problem. it seems this is work in progress (e.g. http://git.denx.de/?p=u-boot/u-boot-dm.git;a=commit;h=d8abb46b37fadff0349adb376df6d3ecd09ee7d1), so i expect it to get fixed sooner or later
flok420 has quit [Ping timeout: 265 seconds]
mrnuke has quit [Remote host closed the connection]
mrnuke has joined #linux-sunxi
gianMOD has quit [Remote host closed the connection]
leviathancn has joined #linux-sunxi
gianMOD has joined #linux-sunxi
flok420 has joined #linux-sunxi
naobsd has quit [Quit: naobsd]
arete74_ has joined #linux-sunxi
arete74 has quit [Read error: Connection reset by peer]
<RSpliet>
bbrezillon: no feedback on patch 2? ;-) I'll fix #1 later today... playing around with DMA right now
<MY123>
Is Lima usable now?
<MY123>
libv, seeing your blog post, I see that you may no longer work on Lima...
naobsd has joined #linux-sunxi
naobsd has quit [Client Quit]
MY123 has quit [Quit: Leaving]
cubeast has quit [Quit: Leaving]
gianMOD has quit [Remote host closed the connection]
kaspter has joined #linux-sunxi
simosx has quit [Quit: Leaving]
kaspter has quit [Remote host closed the connection]
<bbrezillon>
RSpliet: done
MY123 has joined #linux-sunxi
<RSpliet>
bbrezillon: thanks!
<philippe_fouquet>
RSpliet, bbrezillon thanks for your nand path it work fine
<bbrezillon>
philippe_fouquet: you mean the patch RSpliet just sent ?
bgola has quit [Quit: leaving]
bgola has joined #linux-sunxi
<philippe_fouquet>
No 1 week ago
leviathancn has quit [Ping timeout: 265 seconds]
<plaes>
hrm.. the banana pi guy is back
gianMOD has joined #linux-sunxi
<libv>
plaes: want me to block his account?
<libv>
nah
<libv>
i will just block
<libv>
he earned as much
<MY123>
libv, Is there Mali-450 cheap boards?
<libv>
MY123: odroid c1
<MY123>
libv, tolerable kernel versions?
<libv>
i am not sure whether i already did order that, or whether i dreamt ordering that
<libv>
no idea
<libv>
at best, it lies around my pile of hardware-todos
<MY123>
libv, as 3.4 in 2015 for a new board is not good :-)
<MY123>
libv, and the C1 doesn't have hardware VT
<MY123>
(unlike sunXi and RPi2 boards)
<MY123>
*sun7i
<MY123>
sun8i
<NiteHawk>
maybe those folks are upset because we label it "LeMaker" Banana Pi? should we move to more neutral "Banana Pi" / "Banana Pro" pages and just mention the various manufacturers / distributors (LeMaker, Lenovator, Sinovoip / banana-pi.com)?
domidumont has quit [Read error: Connection reset by peer]
reinforce has joined #linux-sunxi
<NiteHawk>
the branding and model mixup on those bpi variants is a real PITA - but then, they brought exactly that upon themselves :P
<libv>
NiteHawk: i do not care
<libv>
indeed
gianMOD has quit [Ping timeout: 258 seconds]
<libv>
they are a pest, and they are taking customers away from better/nicer players
<MY123>
is there a Mali-400 simulator(ARM propretary is not a problem)
<MY123>
?
<libv>
and this pseudo spamming is all that they contribute
<NiteHawk>
full ack - that's just awful
<MY123>
NiteHawk, they just want bucks :-)
<libv>
BPI-forum has been blocked.
<MY123>
libv, really?
<MY123>
if the BPI has no forums, it means EOL
<MY123>
$marsboard_a20
<libv>
that's a wiki username
<NiteHawk>
no, that wiki-spamming user was blocked (with a 1 year expiry)
<libv>
and this message at the top should be spread to those other pages
<NiteHawk>
no it wasn't. that redirect was created quite some time ago by me. the BPI-forum dude just "forked" the page, which was inacceptable
<libv>
ah
<NiteHawk>
libv, it might be preferable to have it 'the other way around' - i.e. rename the page to http://linux-sunxi.org/Banana_Pi and have "LeMaker Banana Pi" redirect there
<libv>
NiteHawk: perhaps in future
<libv>
let these idiots fight it out first
<NiteHawk>
same goes for "Banana Pro" / "LeMaker Banana Pro" - thought that's actually a name they claim (I think sinovoip has that one as "BPi-M1+")
domidumont has joined #linux-sunxi
<plaes>
yup
<MY123>
libv, Are all Fish/Banana/Apple/Orange/... Pi the same?
<NiteHawk>
yup, i'm all for that - keep what we have until they show up with some 'sane' approach
<plaes>
MY123: nope
<NiteHawk>
no, Orange Pi is yet another company and a different SoC (H3)
<plaes>
though, it might be that Sinovoip Banana Pi (or BPi-M1+) is same as Lemaker's Banana Pi
<rellla>
omg. poor BPI-forum. thanks libv for the hammer :) i can't understand why those guys don't get it. my personal entry page in the wiki is the recent changes page...
<MY123>
NiteHawk, there is as many $(cat /dev/random) Pi attempts as possible
<NiteHawk>
plaes: that's part of the whole charade that they can't get themselves together and straighten out that mess they've created with their licensing/lawsuits bickering and model mashup
<libv>
if this continues, i will create a libv_shit_pi device page
<libv>
and start off with a big rant about companies ripping off the rpi and linux-sunxi community
<NiteHawk>
Insert-random-ripoff Pi
Net147 has joined #linux-sunxi
<MY123>
libv, there will be more disk space taking all the Pi "clones" attempts than all sunxi boards
* rellla
decides to trash his bananapi and his new orangepi \o/
<rellla>
libv: kind of a "New *Pi Device Howto"
<MY123>
libv, for the RPi case, it's separate as the BCM2836 is sold only for the RPF
<MY123>
(it isn't even on broadcom.com)
<MY123>
rellla, use what you bought. Just don't make free advertising for them
<libv>
rellla: that would be an attempt to be constructive
<MY123>
libv, there is too many *Pi to document all of them
<Wizzup_>
in fact, Pi is infinite.
<libv>
rellla: and wouldn't be read by any of those idiots anyway :)
<rellla>
libv: that is the main problem. even hints are ignored.
<libv>
rellla: at least a shit pi page would give those of us who try to keep the wiki in shape a nice outlet :)
<rellla>
imho the biggest mess is, that EVERY device (cubieboard included) creates it's own "community", forum, github repos ...
<rellla>
i get angry, when i read some threads there ... urgh
<libv>
rellla: true, but that actually plays to our advantage
<MY123>
libv, for SBCs, the RPF, Olimex, Cubietech and Hardkernel are enough
<libv>
rellla: it keeps the worst of the worst away from us
<libv>
the clueful people will eventually graduate to linux-sunxi
<libv>
MY123: cubietech is no longer our friend now that tom cubie and benn huang left