Turl changed the topic of #linux-sunxi to: Allwinner/sunxi /development discussion - did you try looking at our wiki? https://linux-sunxi.org - Don't ask to ask. Just ask and wait! - https://github.com/linux-sunxi/ - Logs at http://irclog.whitequark.org/linux-sunxi
<apritzel> the H3 has the same new controller as the A64
<apritzel> (the A64 is mostly an H3 with the A7 cores replaced with A53s)
<lennyraposo> we get no indication what it is?
<apritzel> I couldn't find anything similar, at least by looking at Linux drivers
<lennyraposo> it says realtek
<apritzel> but I am basically clueless about networking and MAC IP in general ;-)
<apritzel> who say that?
<lennyraposo> pine wiki
<lennyraposo> datasheet
<apritzel> that's the WiFi
<lennyraposo> it says 1gbe
<apritzel> oh, that is the PHY
<apritzel> this is the part that connects the SoC to the Ethernet cable
matthias_bgg has quit [Ping timeout: 264 seconds]
<apritzel> the MAC is the actual network controller, it sits in the SoC
<lennyraposo> doh
souther has quit [Ping timeout: 264 seconds]
souther has joined #linux-sunxi
<apritzel> the PHY is a no-brainer, since it bascially needs no setup
<lennyraposo> it's the mac in the soc
<lennyraposo> I am reading it right now
<apritzel> well, some setup is needed, but this is all standard for that PHY
<lennyraposo> boot0 calls upon u-boot
<lennyraposo> u-boot points to kernel
<apritzel> not directly on the A64
<lennyraposo> u-boot initilizes hardware (voltage regulation ie 3.0 to 3.3 pmc
<apritzel> boot0 does a 64-bit reset into ARM trusted firmware, which drops back to 32-bit and into U-Boot
<lennyraposo> pmic)
<lennyraposo> ok
<lennyraposo> so my understanding is u-boot acts in place of a bios in otherwords
<apritzel> and U-Boot shouldn't do the PMIC stuff, I plan to do this from ATF (ARM trusted firmware)
<apritzel> well, this is a good topic
nemunaire has quit [Quit: quit]
<apritzel> traditionally U-Boot is often both firmware and bootloader
<lennyraposo> didn't longsleep backport and include ATF
<apritzel> for the 32-bit Allwinner chips it really is
<apritzel> but actually U-Boot should just be the boot loader and shouldn't be bothered with hardware initialization
<apritzel> and the A64 firmware setup has everything to do it that (the right) way
<lennyraposo> so oyu are telling me that allwinner is basically doing things half assed and doublin gup when there is no need to
<apritzel> kind of
<lennyraposo> so the boot source does away with that once we get it
<apritzel> they seem to have had some right ideas this time
<lennyraposo> on the hardware side
<lennyraposo> but not in the software side
<apritzel> but then really struggled with properly implementing this
<lennyraposo> so in an idealistic world
<lennyraposo> it shouldn't be this tennis match of back and forth with u-boot having to initialize and then boot
<apritzel> U-Boot shouldn't really initialise stuff expect say the MMC controller
<lennyraposo> the kernel should be initializing the hardware in other words?
<lennyraposo> or I should say the ATF
<apritzel> the ATF, really
<lennyraposo> I get it now
<lennyraposo> trusted firmware
<lennyraposo> get it going
<lennyraposo> u-boot turn on the mmc
<lennyraposo> point to the kernel
<apritzel> it should be more like x86: the BIOS cares about the board specific setup
<lennyraposo> and done
vishnup has quit [Remote host closed the connection]
<apritzel> than loads the boot loader, which could be U-Boot or something else
<apritzel> and if it is something else, the board setup shouldn't be replicated there
<lennyraposo> from what I have been reading it seems like a hand off from boot0 to u-boot then back to boot0
<lennyraposo> or am I reading oncorrectly
<lennyraposo> incorrectly
<apritzel> normally U-Boot just loads the kernel and jumps to it
<apritzel> boot0 is done once U-Boot starts
<apritzel> the only exception is that FEL booting, but that usually does not involve boot0
<lennyraposo> I can see how the source to boot0 kills alot of the upstart
hulu1522 has joined #linux-sunxi
<lennyraposo> leave ATF to do it's thing
<lennyraposo> on call and entering into EL3 properly
<lennyraposo> boot
<lennyraposo> done
<lennyraposo> that means we can further change things up properly with U-boot say pointing to a usb drive for booting
<lennyraposo> instead of just the MMC
<lennyraposo> I am assuming
<lennyraposo> u-boot can do that
<apritzel> yes, U-Boot supports USB as well
<lennyraposo> perhaps have a boot order
<lennyraposo> fallback to this if this is not present
<apritzel> think of U-Boot being similar to grub
<apritzel> just loads the kernel from {disk, MMC, USB, network}
<lennyraposo> so oyu can edit a settings file and alter to your content
<lennyraposo> your liking I mean
<lennyraposo> now that is useful
<apritzel> U-Boot is very versatile and flexible
<apritzel> you get a real shell
<lennyraposo> just like grub 2
hulu1522 has quit [Remote host closed the connection]
<apritzel> yes, similar
mosterta has quit [Ping timeout: 276 seconds]
<lennyraposo> just reading up on u-boot
<apritzel> but U-Boot is not really modular, you do all the configuration at compile time and load that single binary
<lennyraposo> sort of like initrd
<lennyraposo> exceptt with compiling
<lennyraposo> a package with everything needed for the sbc
<lennyraposo> the trusted firmware required aswell
<lennyraposo> brb gotta bring my daughter to bed
<lennyraposo> back in black
<apritzel> on a proper system (ARM server) you have ATF in flash, doing the board initialisation
<apritzel> it the loads EFI, which could be in flash as well (or on some disk)
<apritzel> this does further setup, probably launching grub
<apritzel> the point is that the Pine64 (and many other SBCs) have no flash on board
<apritzel> so everything comes from the SD card, for instance
<apritzel> including the firmware
<apritzel> so traditionally there is no need for a proper firmware stack with clear and separated responsibilities
<apritzel> instead U-Boot just does all the stuff ;-)
<lennyraposo> the first parition (fat32 or any supported fs)
<apritzel> Allwinner SoCs are hardwired to load 32KB from sector 16 of the SD card and execute this
<apritzel> (if there is no NAND flash, that is)
<lennyraposo> it's only personal fallback
<lennyraposo> own* not only
<apritzel> this is either boot0 or U-Boot's SPL
<lennyraposo> I see why you want to discard the bsp provided
<lennyraposo> unlocking this with mainline u-boot/kernel there would be no need for allwinner depndency
<apritzel> indeed
<lennyraposo> and if chip has done what it claims it has witht he vid driver
<lennyraposo> mali that is
<lennyraposo> wow
<apritzel> traditionally booting on ARM is very board specific, there are no real standards
<lennyraposo> dude I have learned a world of things today
<apritzel> ARM64 tries to fix this
<apritzel> so I'd rather go with that and do it right
<lennyraposo> so as I see there are 3 main areas of focus required
<lennyraposo> the boot0 source
<lennyraposo> the ethernet
<lennyraposo> and usb
<lennyraposo> we have the cedarx stuff in hand
<apritzel> and getting the firmware story right
<lennyraposo> the dtb
<lennyraposo> si that correct?
<lennyraposo> dts
<lennyraposo> I pretty much think of it as acpi like
<apritzel> kind of
<apritzel> .dts (the source) compiles into .dtb (the binary)
<apritzel> and describes the hardware that is not plug-and-play
<apritzel> which is basically everything on most ARM chips ;-)
<lennyraposo> I can see the settings to everything
<apritzel> also ARM has no standards, so the .dtb describes where the serial port is for instance
<apritzel> has no standards => has no standard hardware addresses
<apritzel> on x86 the first serial port is on 0x3f8 and the keyboard controller is port 0x60/0x61, VGA is at 0xA0000 and so on
<lennyraposo> so theoretically I can assign whatever
<lennyraposo> to any hex address?
<apritzel> on ARM this is different for every board, and this is described in the DT
<apritzel> the SoC vendor does this
<apritzel> this is (hopefully) documented in the memory map
<lennyraposo> so we have to match the soc vendors settings
<apritzel> exactly
<lennyraposo> strange how things play ball in android and the very same doesn't in linux
<apritzel> so basically you take the manual and create a DT out of it
<lennyraposo> good to have that on hand then
<apritzel> I think modern Android uses DT as well
<apritzel> you don't want to know what it did before DT ;-)
<lennyraposo> longsleep had to play with a bunch of settings
<lennyraposo> to fine tune it
<lennyraposo> I htink for the ethernet and the thermal trip points
afaerber has quit [Ping timeout: 250 seconds]
<apritzel> yes, DT is much more versatile than say ACPI, you can describe basically everything you want (with key=value pairs)
<apritzel> that's why it is important to agree on the key part at least
<apritzel> to avoid everyone defining their own
<lennyraposo> don't tell me about acpi
<lennyraposo> have had my fair share of battles with fixing cheap bios's that offer poor support outside the windows sphere
<apritzel> me too ;-)
<apritzel> go to arch/arm64/boot/dts and check out some examples
<lennyraposo> I foudn that out the hardway with my toshiba purchase
<lennyraposo> mainly becasue at the time for 300 bucks I got quad core apu
<lennyraposo> 5 years ago
<lennyraposo> which repo
<lennyraposo> main
<apritzel> yes, mainline Linux
<lennyraposo> I am browsing via browser atm
<apritzel> enjoy, time to bring myself to bed here ;-)
<lennyraposo> nice to see the supported vendors
<lennyraposo> thanks apritzel
<lennyraposo> between you and longsleep I should be caught when my boards are here
<lennyraposo> and you will eb one more stronger in the ranks
<apritzel> cu later!
<lennyraposo> ciao
<lennyraposo> haha
<lennyraposo> no allwinner listed
<lennyraposo> wish I knew back then what I know now
<lennyraposo> then again hindsight is always 20/20
apritzel has quit [Ping timeout: 244 seconds]
afaerber has joined #linux-sunxi
hipboi has joined #linux-sunxi
<lennyraposo> brb
lennyraposo has quit [Quit: Leaving.]
lennyraposo has joined #linux-sunxi
khuey is now known as khuey|away
hipboi has quit [Ping timeout: 260 seconds]
hipboi has joined #linux-sunxi
<lennyraposo> just wanted to ask anyone who may know
<lennyraposo> how many allwniner socs are near mainline ready (if I am using the term correctly)?
TheSeven has quit [Ping timeout: 250 seconds]
TheSeven has joined #linux-sunxi
keh has quit [Ping timeout: 244 seconds]
mosterta has joined #linux-sunxi
cnxsoft has joined #linux-sunxi
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
ninolein has quit [Ping timeout: 248 seconds]
ninolein has joined #linux-sunxi
KotCzarny has quit [Ping timeout: 260 seconds]
mosterta has quit [Ping timeout: 244 seconds]
Nyuutwo has quit [Remote host closed the connection]
Nyuutwo has joined #linux-sunxi
bwarff has quit [Read error: Connection reset by peer]
reev has joined #linux-sunxi
bwarff has joined #linux-sunxi
fredom is now known as poooooofs
KotCzarny has joined #linux-sunxi
reev has quit [Read error: Connection reset by peer]
p1u3sch1 has quit [Ping timeout: 246 seconds]
p1u3sch1 has joined #linux-sunxi
Shirasaka-Hazumi has quit [Ping timeout: 264 seconds]
hipboi_ has joined #linux-sunxi
hipboi has quit [Ping timeout: 250 seconds]
reev has joined #linux-sunxi
Shirasaka-Hazumi has joined #linux-sunxi
HeavyMetal has quit [Ping timeout: 250 seconds]
HeavyMetal has joined #linux-sunxi
HeavyMetal has joined #linux-sunxi
hipboi_ has quit [Ping timeout: 248 seconds]
hipboi has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
reinforce has joined #linux-sunxi
clonak has joined #linux-sunxi
<lennyraposo> answered my own quesiton
<lennyraposo> any recommendations for a git client under debian
<wens> git?
<lennyraposo> lol
<lennyraposo> got that
<lennyraposo> was look at some of the clients out there and was wondering if any clients come to mind under linux
<lennyraposo> gui clients
<lennyraposo> never used one before was just wondering
<wens> i only use cli
<lennyraposo> :)
<montjoie> it is official the EMAC H3 answer ping now
<montjoie> ssh comes soon
<lennyraposo> wow
<lennyraposo> apritzel and I were talking abotu that today
<wens> montjoie: yay!
<lennyraposo> if I am understanding correctly
<lennyraposo> so you got the ethernet working
<montjoie> yes
<lennyraposo> wonder if the a64 can use the same approach
<montjoie> just ssh wont work and I dont see why
<wens> lennyraposo: it's the same hardware
<montjoie> but ping work and ring buffer works
<wens> try tcpdump?
<lennyraposo> hence why apritzel mentioned it
<wens> put it this way, if we can clean it up so it works on a83t, it should also work on a64
<lennyraposo> figured that :)
<montjoie> wens for a83t I need ac200
<montjoie> and I2C
nemunaire has joined #linux-sunxi
<montjoie> I will publich the last patch today
<wens> montjoie: i have cubietruck plus, which uses rtl phy :)
<montjoie> ok
<lennyraposo> they use that rtl 8211e
<lennyraposo> interesting
IgorPec has joined #linux-sunxi
<lennyraposo> need to look into usb as well
<lennyraposo> maybe datasheet o rmanual somewhere has some insight
iamfrankenstein has joined #linux-sunxi
leviathancn_szu has joined #linux-sunxi
<wens> lennyraposo: usb for? a64?
<lennyraposo> yep
<lennyraposo> did not see any specific info
<lennyraposo> just generic banter
<lennyraposo> as I have no boards yet to tryout an dtest
<lennyraposo> msot of my research is through reading and locating
<wens> it mostly involes poking undocumented bits in the usb register space
<wens> looking through the bsp is the only option
<lennyraposo> so in other words
<lennyraposo> kernel from bsp
<lennyraposo> and poke it with a stick to see what happens
<lennyraposo> get data
<lennyraposo> adn identify
<wens> bsp should work, just look through its code
<lennyraposo> downloaded allwinners 24gb mess for the pine
<lennyraposo> time to start goign through it all
<lennyraposo> lol
Gerwin_J has joined #linux-sunxi
<lennyraposo> I am impressed with you guys work on and fully understadn this ecosphere
<lennyraposo> does anyoen here communicate with allwinner about anything?
<montjoie> 200 ping and after the void...
<lennyraposo> it just drops?
<lennyraposo> no response I mena
<lennyraposo> mean
<lennyraposo> man I need sleep
IgorPec has quit [Ping timeout: 268 seconds]
<montjoie> no idea
<montjoie> and I got a reply to the first ssh packet but my mainpc doesnt answer
<lennyraposo> does your mainpc register any activity?
<lennyraposo> from teh interface?
<lennyraposo> porbably a stupid question but I had to ask
domidumont has joined #linux-sunxi
<montjoie> tcpdump see nothing
domidumont has quit [Remote host closed the connection]
domidumont has joined #linux-sunxi
IgorPec has joined #linux-sunxi
jernej has quit [Ping timeout: 252 seconds]
paulk-collins has joined #linux-sunxi
<KotCzarny> montjoie: try pinging with different packet sizes?
cajg has joined #linux-sunxi
<mripard_> montjoie: you spent a lot of time on this already, is there really a point in trying to have a driver working from scratch when allwinner's just need a bit of cleanup? ?
IgorPec has quit [Quit: Nettalk6 - www.ntalk.de]
<montjoie> KotCzarny: will do that
<montjoie> mripard_: more than a bit of cleanup since no DT, etc.. and I wanted to be sure to understand all what I do
<mripard_> montjoie: no DT is easy
<mripard_> you've been struggling to get a few packets out of the thing for monthers
<mripard_> and you'll probably have a lot of other bugs down the road
akaizen has quit [Remote host closed the connection]
paulk-collins has quit [Remote host closed the connection]
<montjoie> the bugs that make me lose time was undocumented...
<montjoie> so now we know it
poooooofs has quit [Ping timeout: 276 seconds]
<mripard_> which is exactly my point.
<mripard_> all the undocumented bits are already there
<KotCzarny> in one he is right
<KotCzarny> once undocumented become documented
mzki has joined #linux-sunxi
clonak has quit [Ping timeout: 240 seconds]
FDCX has quit [Ping timeout: 244 seconds]
<mripard_> no one ever complained that the A10 EMAC had undocumented bits, however, everyone is very happy that it works
FDCX has joined #linux-sunxi
poffs has joined #linux-sunxi
clonak has joined #linux-sunxi
engideavr has joined #linux-sunxi
hipboi has quit [Ping timeout: 250 seconds]
<montjoie> mripard_: I understand well your point of view, but I wanted to do something that can I be proud of it
hipboi has joined #linux-sunxi
<maz> montjoie: doing something you are proud of is definitely commendable. doing something that is useful to others is even better.
enrico_ has joined #linux-sunxi
hipboi has quit [Ping timeout: 268 seconds]
cptG_ has joined #linux-sunxi
cptG has quit [Ping timeout: 248 seconds]
matthias_bgg has joined #linux-sunxi
matthias_bgg has quit [Remote host closed the connection]
matthias_bgg has joined #linux-sunxi
tomboy65 has quit [Ping timeout: 248 seconds]
tomboy65 has joined #linux-sunxi
diego_r has joined #linux-sunxi
hipboi has joined #linux-sunxi
tomboy64 has joined #linux-sunxi
tomboy65 has quit [Ping timeout: 248 seconds]
<KotCzarny> let him code for fun
<maz> KotCzarny: I'm not telling him what to do - that'd be stupid. I'm just stating that sometimes, it is better to gradually improve a code base (and get something off the ground quickly) than to start from scratch (and be stuck in the mud for too long).
<KotCzarny> maz, how about linux-sunxi statement to ignore 3.4 b0rkage and focus on mainline only?
<maz> KotCzarny: mainline is the only thing that matter. why do you think I'd think otherwise?
<buZz> no mainline no glory
<KotCzarny> maz, because that's the same story
<maz> buZz: it's not about glory. it is about being useful in the long run. if it was about glory, I'd have stopped 20 years ago! ;-)
<maz> KotCzarny: not quite. we're talking about lifting a random driver, improving it to make it useable with mainline.
<buZz> maz: no use = no glory
<maz> buZz: I don't get your drift, but nevermind.
<mripard_> KotCzarny: discarding everything that Allwinner did just because it's Allwinner that did it and not someone else is a bit stupid
<mripard_> they did some good stuff too
engideavr has quit [Quit: Konversation terminated!]
<KotCzarny> is gmac driver by allwinner any good?
engideavr has joined #linux-sunxi
<keesj> I have a few more "intersting???" questions.
<keesj> I enabled spi and spidev on mainline kernel (a10 olimex lime board) and the device is functional however I have the kernel not happy about how I modified the device tree
<keesj> http://paste.ubuntu.com/15390916/ buggy DT: spidev listed directly in DT
<keesj> any hints on fixing that warning (and possibly renaming /dev/spidev32766.1 to /dev/spidev2.1 or similar)
ricardocrudo has joined #linux-sunxi
<mripard_> to rename it, you have to set an alias
<mripard_> in the aliases node
adj_ has quit [Quit: Leaving]
hipboi has quit [Ping timeout: 248 seconds]
<keesj> so just define a new root node with alias : name{} and use that in the DT?
hipboi has joined #linux-sunxi
hipboi has quit [Remote host closed the connection]
PaddyF has left #linux-sunxi ["Leaving"]
<keesj> No, I don't understand :(
tomboy65 has joined #linux-sunxi
tomboy64 has quit [Ping timeout: 260 seconds]
tomboy64 has joined #linux-sunxi
tomboy65 has quit [Ping timeout: 250 seconds]
adj_ has joined #linux-sunxi
leviathancn_szu has quit [Remote host closed the connection]
iamfrankenstein has quit [Ping timeout: 246 seconds]
Amit_T has quit [Ping timeout: 260 seconds]
hansg has joined #linux-sunxi
Sieb has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
<hansg> Hi Sieb, shall I create a #gsl1680 ?
<Sieb> Yes that is alright
avph has joined #linux-sunxi
michelv has joined #linux-sunxi
michelv has quit [Client Quit]
tomboy65 has joined #linux-sunxi
tomboy64 has quit [Ping timeout: 244 seconds]
Gerwin_J has joined #linux-sunxi
adj_ has quit [Quit: Leaving]
iamfrankenstein has joined #linux-sunxi
montjoie has quit [Ping timeout: 260 seconds]
montjoie has joined #linux-sunxi
reev has quit [Ping timeout: 244 seconds]
reev has joined #linux-sunxi
cptG_ is now known as cptG
reev has quit [Ping timeout: 244 seconds]
<keesj> mripard_: yea I looked a that but it is the same kind of device (e.g. a bus driver) while I need to tell spidev that the second port of of the SPI driver
cnxsoft has quit [Read error: Connection reset by peer]
cnxsoft has joined #linux-sunxi
tomboy65 has quit [Ping timeout: 248 seconds]
yann|work has quit [Ping timeout: 260 seconds]
tomboy65 has joined #linux-sunxi
JohnDoe_71Rus has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
<mripard_> keesj: you said you wanted /dev/spidev32766.1 to be renamed as /dev/spidev2.1
<mripard_> this is exactly how you do that
<mripard_> you add / change a line there
<mripard_> to add a spi2 = <&whatever spi controller you want to use>
Netlynx has joined #linux-sunxi
Netlynx has joined #linux-sunxi
IgorPec7 has joined #linux-sunxi
phil42 has quit [Quit: Leaving]
IgorPec7 has quit [Ping timeout: 240 seconds]
bwarff has quit [Ping timeout: 276 seconds]
iamfrankenstein has quit [Ping timeout: 246 seconds]
iamfrankenstein has joined #linux-sunxi
cnxsoft has quit [Quit: cnxsoft]
adj_ has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
JohnDoe_71Rus has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
JohnDoe_71Rus has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
vishnup has joined #linux-sunxi
<lennyraposo> just got all caught up
<lennyraposo> the mac for the h3 still needs work I see
vishnup has quit [Ping timeout: 246 seconds]
afaerber has quit [Quit: Ex-Chat]
Gerwin_J has quit [Ping timeout: 244 seconds]
afaerber has joined #linux-sunxi
reinforce has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
apritzel has joined #linux-sunxi
JohnDoe7 has joined #linux-sunxi
<montjoie> arp/ping works, just a strange problem that my pc doesnt finalize TCP 3way handshake with it
<maz> montjoie: sounds like a checksumming issue.
tkaiser has joined #linux-sunxi
<tkaiser> montjoie: http://sunxi.montjoie.ovh/patchs_current/ is pretty outdated isn't it?
<montjoie> tonigth wireshark will confirm it
JohnDoe_71Rus has quit [Ping timeout: 260 seconds]
<montjoie> tkaiser: I will update it soon
<montjoie> I rebase my commits
mzki has quit [Quit: leaving]
Gerwin_J has quit [Ping timeout: 244 seconds]
Gerwin_J has joined #linux-sunxi
<cptG> ...my touch LCD is going crazy (reports input events even if I disconnect the touch panel completely) - is there a way to change sensitivity values that I might've accidentally changed?
<cptG> sorry: allwinner A20, module sun4i-ts...
Hahaburger has joined #linux-sunxi
<lennyraposo> reverse what was done done comes to mind
<lennyraposo> sorry still a newb at all this here
Hahaburger has quit [Client Quit]
Gerwin_J has quit [Read error: Connection reset by peer]
Hahaburger has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
domidumont has quit [Read error: Connection reset by peer]
<lennyraposo> looks liek I am getting a pine next week
<lennyraposo> so I have ot grow upa lot faster
<lennyraposo> ;)
Netlynx has quit [Quit: Leaving]
khuey|away is now known as khuey
<montjoie> ssh working:)
<montjoie> maz: it was CRC:)
<tkaiser> :)
<maz> montjoie: a classic.
<maz> montjoie: been there quite a few times.
<montjoie> but long output like dmesg doesnt work
<montjoie> I know what is it
hansg has quit [Quit: Leaving]
premoboss has joined #linux-sunxi
Sieb has quit [Quit: Page closed]
hulu1522 has joined #linux-sunxi
<apritzel> montjoie: can you post a fixed version somewhere?
zuikis has joined #linux-sunxi
<lennyraposo> was thinking the same thing apritzel
<lennyraposo> need usb next
tomboy64 has joined #linux-sunxi
poffs has quit [Ping timeout: 244 seconds]
tomboy65 has quit [Ping timeout: 248 seconds]
diego_r has quit [Ping timeout: 248 seconds]
tomboy64 has quit [Ping timeout: 244 seconds]
nove has joined #linux-sunxi
slapin has joined #linux-sunxi
tomboy64 has joined #linux-sunxi
hulu1522 has quit []
<slapin> hi, all!
<slapin> I see clock gates in device tree as <&ahb_gates 39> etc. how can I distinguish which is which
<slapin> and how can I match clock output-names to gate bits?
<slapin> also I see that clock gates are handled a bit differently on different hardware platforms, any ideas?
<slapin> mripard_: ^^^
tomboy65 has joined #linux-sunxi
dev1990 has joined #linux-sunxi
tomboy64 has quit [Ping timeout: 260 seconds]
<apritzel> slapin: in the ahb_gates nodes both clock-indices and clock-output-names are ordered in the same way
<apritzel> slapin: the indention is the same, that helps you to match them
apritzel has quit [Ping timeout: 244 seconds]
tomboy64 has joined #linux-sunxi
tomboy65 has quit [Ping timeout: 264 seconds]
enrico_ has quit [Quit: Bye]
paulk-collins has joined #linux-sunxi
<montjoie> apritzel http://sunxi.montjoie.ovh/ethernet/ (still needing to setup a public git repo)
<slapin> apritzel: I need to find clock gate and setting for CSI0 on Allwinner A13, per manual, it is bit 8 in AHB_GATING_REG1, offset 0x64, address 0x01C20064, and clock source / divider setting is at CSI_CFG_REG, 0ffset 0x134, addr 0x01C20134
vagrantc has joined #linux-sunxi
<slapin> apritzel: the problem is which clock setting should I use for the peripheral in my device tree
tomboy65 has joined #linux-sunxi
apritzel has joined #linux-sunxi
tomboy64 has quit [Ping timeout: 260 seconds]
<apritzel> slapin: bit 8 in REG1 would be 40 then
Hahaburger has quit [Quit: Page closed]
<apritzel> so that's already named as ahb_csi in the A13 .dtsi
<slapin> apritzel: how are reserved bits hadled for output-names ? I see reserved values are just skipped, but there is no information about which are reserved...
<apritzel> for the clock the usual approach is to find a similar clock and use that
<apritzel> slapin: well, we just enumerate those that are documented and useful
<apritzel> everything else is not needed anyway
<slapin> apritzel: output-names usage of values in ouput-names is prohibited, clock names should be defined properly to be used by names
<slapin> apritzel: there are reserved bits in the middle
<apritzel> are you talking about the gates or the CSI clock?
<slapin> apritzel: I need both
<montjoie> normal, I post the patch and found a missing part (txstats)
<slapin> apritzel: I need to enable CSI clock, so I need to both enable it and set to useful value
<apritzel> slapin: I was wondering about what reserved bits you were talking
<apritzel> so for the gate everything is fine, just use <ahb_gates 40>
<slapin> apritzel: gating register does have a set of bits reserved, these are 4, 3, 11, 12, 15-19, 23-27, I wonder if output-names handles these
<apritzel> slapin: why do you care those? they are just not enumerated and never used
<apritzel> *about* those
<slapin> apritzel: yeah, <ahb_gates 40> should work, but how about main clock? do we have PLL7 on A13? no such thing in DT...
<slapin> apritzel: now my main interest is to enable CSI clock
<slapin> apritzel: via DT
<apritzel> Unless you find a similar clock, I am afraid you have to define your own one here
<slapin> CSI_CFG_REG is quite similar to others...
<apritzel> really? the reset bit 30 in there looks unique to me
<mripard_> slapin: if you want pll7 you'll need my DRM series
<apritzel> yeah, the DE clock looks similar
<slapin> mripard_: where I can get these? any repository?
<mripard_> 2s
<mripard_> as usual, work in progress, anything might change in the future
<slapin> mripard_: thanks a lot!
<slapin> apritzel: where are clock handles defined? are they all in DT or some in code?
<mripard_> what do yo umean by clock handle?
apritzel has quit [Ping timeout: 244 seconds]
<slapin> mripard_: I mean clock bits and registers
<mripard_> the base address of clocks are in the DT
<mripard_> bits, it depends
<mripard_> if it's just an array of gates, in the DT
<mripard_> if it's a clock taht does more than that, in the driver
<slapin> it seems I need to upgrade from 4.1 to master to be on the same wave
lamer14580665872 has joined #linux-sunxi
domidumont has joined #linux-sunxi
tkaiser has quit [Ping timeout: 244 seconds]
<slapin> mripard_: how can I implement enabling of csi0 clock, do I need to add it to kernel or is it enough to change DT?
IgorPec7 has joined #linux-sunxi
IgorPec8 has joined #linux-sunxi
IgorPec7 has quit [Ping timeout: 240 seconds]
<mripard_> no idea, i never looked at it
afaerber has quit [Quit: Ex-Chat]
<mripard_> it is very similar to the display clocks
<mripard_> but a few things change
<mripard_> you have a wider divider
<mripard_> more parents
<mripard_> you'll probably need to code there :)
jernej has joined #linux-sunxi
premoboss has quit [Ping timeout: 260 seconds]
IgorPec8 has quit [Ping timeout: 276 seconds]
premoboss has joined #linux-sunxi
vovcia has joined #linux-sunxi
<vovcia> hello :)) im wondering - anyone now what are is FMIN pin on A20 SoC? i tried to google it but nothing comes
<vovcia> this one: PI21 (FMIN-R/PS2SDA0/UART7-RX/HSDA)
<maz> vovcia: FM Input? ;-)
<vovcia> maz: radio input? audio input? ;)
akaizen has joined #linux-sunxi
<maz> vovcia: your guess is as good as mine...
lamer14580665872 has quit [Quit: jIRCii - http://www.oldschoolirc.com]
tkaiser has joined #linux-sunxi
IgorPec8 has joined #linux-sunxi
jstein has joined #linux-sunxi
<tkaiser> montjoie: Should your patches work against a clean 4.5?
<montjoie> tkaiser: I think
<montjoie> I am based on 4.5
akaizen_ has joined #linux-sunxi
akaizen has quit [Ping timeout: 250 seconds]
IgorPec8 has quit [Ping timeout: 244 seconds]
Gerwin_J has quit [Quit: Gerwin_J]
apritzel has joined #linux-sunxi
arokux has joined #linux-sunxi
bwarff has joined #linux-sunxi
edolnx has joined #linux-sunxi
<tkaiser> montjoie: I had to add 0000-sun8i-h3.patch: http://sprunge.us/QFXd (I would believe you rely on http://moinejf.free.fr/opi2/sun8i-h3.dtsi)?
bwarff has quit [Ping timeout: 264 seconds]
tomboy65 has quit [Ping timeout: 276 seconds]
tomboy65 has joined #linux-sunxi
bgardner has quit [Ping timeout: 248 seconds]
bgardner has joined #linux-sunxi
gzamboni has quit [Read error: Connection reset by peer]
bwarff has joined #linux-sunxi
akaizen_ has quit [Ping timeout: 276 seconds]
IgorPec8 has joined #linux-sunxi
akaizen has joined #linux-sunxi
bwarff has quit [Ping timeout: 276 seconds]
<tkaiser> montjoie: Doesn't work for me: http://pastebin.com/nQKZdp7r
<tkaiser> What about creating a Github account right now (4.5 just released) and push your work up there?
zuikis has quit [Remote host closed the connection]
<tkaiser> That's on an Orange Pi PC with internal Ethernet PHY
<longsleep> tkaiser: i will increase the Pine64 critical trip point to 108C or would you suggest lower?
tomboy64 has joined #linux-sunxi
<lennyraposo> what's it at now?
<longsleep> 100C
<lennyraposo> hwo often does it reach that temp without heatsink
<montjoie> tkaiser: the phy seems ok (id 44), perhaps you hit the same bug then me with big frame
<montjoie> tkaiser: does arp works ?
<longsleep> lennyraposo: probably only with artifical load, so never as throttling will happen earlier
<longsleep> lennyraposo: but i want to increase all trip points by 5C eventually
<lennyraposo> why not 105
<lennyraposo> you still stay within near a 20% temp buffer to max
tomboy65 has quit [Ping timeout: 244 seconds]
<lennyraposo> just thinking from an engineers standpoint
<lennyraposo> leave a decent safety margin
<longsleep> lennyraposo: well the higher the better, and i am currently not using any hysteresis
<lennyraposo> besides I am sur emost will add their own heat sinks or cooling methods if they are concerned about temps
<longsleep> lennyraposo: and the board will just shut down at that level
<lennyraposo> I go with your judegment
<lennyraposo> as I am boardless fo rthe moment ;0
<lennyraposo> and still elarning
<lennyraposo> learning*
<longsleep> i am only concerned about people using the pine without thinking anything and getting whatever enclosure and putting it next to their tv where it is already hot
<lennyraposo> especially that biege looking thing
<lennyraposo> I am assuming it has a cover
<lennyraposo> so heat is just ogign to build up on those things
<longsleep> yes
<lennyraposo> along with that playbox case
tkaiser has quit [Ping timeout: 250 seconds]
<lennyraposo> a case package like the acryliic one with small fan plus heatsink package would be a wise design for these things
<lennyraposo> or at the very least some sort of heatsink to go with the playbox
<lennyraposo> a thermal pad to disappate thing s a bit better
tkaiser has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
apritzel has quit [Ping timeout: 244 seconds]
<lennyraposo> it's hard to tell what the case has ot offer space wise for addons
<longsleep> no fan please, those make noise
<longsleep> i hate noise
<lennyraposo> and add to more power consumption
<lennyraposo> heatsink
<longsleep> fans are for overclockers if you ask me
<lennyraposo> and properly ventillated case design
<longsleep> decent heatsink should be recommended though
<tkaiser> montjoie: More feedback tomorrow. I was lazy and let just built an Armbian image with your patches and no network came up
<lennyraposo> the acryllic seems ot be the best the pine has thus far
<tkaiser> longsleep: Why not 110¡C?
<lennyraposo> and if it is working correctly it should be throttling before hand am I right?
domidumont has quit [Remote host closed the connection]
<longsleep> tkaiser: sure, but the temperature sensor might be off in the wrong direction and i was choosing the middle between 105 and 110
<tkaiser> 107.5 then ;)
<lennyraposo> don't they say 1 - 3 degree variance
<longsleep> well i hate decimals as well
<lennyraposo> I liekmultiples of five
<tkaiser> lennyraposo: Did you read through the most misused github 'issue' ever already?
domidumont has joined #linux-sunxi
<lennyraposo> nope
<longsleep> hehe i was thinking about closing it :D
<longsleep> err locking it, its already closed
* slapin wonders why [PATCH] clk: Add regmap support is needed...
<lennyraposo> perhaps a little guidance tkaiser
<tkaiser> longsleep: It's a pull request, you can't close it ;)
<lennyraposo> I am assimilating alot here ;)
<mripard_> slapin: because we're using a clock backed by a regmap ?
bwarff has joined #linux-sunxi
<longsleep> its better there than the stupid whatever BB board they have
<lennyraposo> point me to materials and I will read and assimilate
<lennyraposo> then ask questions to be sure that I am on the right track
<longsleep> lennyraposo: well, you could read up about the thermal zone in mainline kernel and send apritzel merge requests for the mainline dts reflecting our changes :)
domidumont has quit [Ping timeout: 276 seconds]
<lennyraposo> reading
<lennyraposo> keep this in mind
IgorPec8 has quit [Ping timeout: 240 seconds]
<lennyraposo> I go back and forth here during the days (Eastern Standard Time) as my kids are home for march break
<lennyraposo> help the wife a lot
<lennyraposo> lol
<lennyraposo> in otherwords rl is hectic with my kids ;)
<lennyraposo> reading into the exxynnos
<lennyraposo> exynos*
akaizen has quit [Ping timeout: 252 seconds]
<longsleep> lennyraposo: sysfs-api is probaly the most helpful and you find references there to what you find in the device tree
iamfrankenstein has quit [Ping timeout: 246 seconds]
iamfrankenstein has joined #linux-sunxi
bwarff has quit [Ping timeout: 250 seconds]
<lennyraposo> I am also getting use to git too
<lennyraposo> cvs is all used before
<lennyraposo> although I do notice similar principles
<tkaiser> longsleep: BTW: I hate fans also and try to avoid where possible. For the Pine64 tests it was necessary to use one.
<longsleep> tkaiser: sure absolutely agreeed
<longsleep> -e
<longsleep> lennyraposo: then continue reading https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/plain/Documentation/devicetree/bindings/thermal/thermal.txt which gives you the idea about the device tree details regarding cooling
<tkaiser> And today I realised that only vibrations matter. As soon as the small fan was hanging around on its cables in the air since BluTack wasn't strong enough I couldn't barely hear anything.
<tkaiser> Logical consequence: I touched it with my finger later. That was fun :)
<lennyraposo> I hate geting nipped by fans
bwarff has joined #linux-sunxi
<longsleep> tkaiser: i liked the picture with the heatpipe most :)
<tkaiser> It worked pretty well. I learned quite a lot about heat dissipation the last days.
<tkaiser> And 'enclosure design' also
<lennyraposo> be back in a bit
<lennyraposo> liek the mathod of assigning trip points and the actions defined on what to do
<lennyraposo> method*
<lennyraposo> brb
<slapin> mripard_: what benefits regmap give? as the support for it is not mainlined?
<slapin> mripard_: also, do you plan to submit pll3 and pll5 support for next window?
<slapin> where do we get A64 boards with worldwide delivery?
<longsleep> tkaiser: commited 108C, with that the board should no longer shut down
apritzel has joined #linux-sunxi
avph has quit [Ping timeout: 246 seconds]
<tkaiser> longsleep: thx (still in hex? ;) )
<longsleep> tkaiser: yeah .. send PR :P
JohnDoe7 has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
willmore has joined #linux-sunxi
<longsleep> tkaiser: what about your HID problems? I checked and i think most of the stuff should already be enabled or do you need multi touch for magic mouse?
avph has joined #linux-sunxi
<mripard_> slapin: you can access the same register space without worrying about concurrency
mossroy has joined #linux-sunxi
<tkaiser> longsleep: No idea, but I'll fix it myself and only if it's a real issue I get back to you. Have to focus on other stuff the next time :)
<longsleep> tkaiser: ok thanks - let me know if i have missed some important module
ricardocrudo has quit [Remote host closed the connection]
bonbons has joined #linux-sunxi
bwarff has quit [Ping timeout: 276 seconds]
paulk-collins has quit [Quit: Quitte]
jstein has quit [Remote host closed the connection]
<mripard_> slapin: and yeah, I'd like to send a v3 sometime in the next weeks
tkaiser has quit [Quit: jIRCii - http://www.oldschoolirc.com]
xypron has joined #linux-sunxi
bonbons has quit [Quit: Leaving]
<lennyraposo> back in black
<lennyraposo> got a free laptop on my to the store
<lennyraposo> my way to the store*
<lennyraposo> old vaio dual core centrino
<lennyraposo> toy for my son
<lennyraposo> anyways
<lennyraposo> back to more reading
<slapin> mripard_: why do we need to access the same register space from several places? blobs?
<mripard_> slapin: no, we have a clock rigt in the middle of the TCON : the pixel clock
<slapin> mripard_: ah, I see
<mripard_> so if we want to share the tcon registers between the clock driver and the display driver, regmap is really helpful
<lennyraposo> be back in a bit
lennyraposo has quit [Quit: Leaving.]
yann|work has joined #linux-sunxi
mossroy has quit [Quit: Quitte]
premoboss has quit [Remote host closed the connection]
ricardocrudo has joined #linux-sunxi
nove has quit [Quit: nove]
arokux has quit [Ping timeout: 246 seconds]
ricardocrudo has quit [Ping timeout: 244 seconds]
dev1990 has quit [Quit: Konversation terminated!]
ricardocrudo has joined #linux-sunxi
bwarff has joined #linux-sunxi
iamfrankenstein has quit [Quit: iamfrankenstein]
ricardocrudo has quit [Remote host closed the connection]
apritzel has quit [Ping timeout: 244 seconds]