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> MoeIcenowy: I looked at the diff between the two MMC dumps: I think I forgot to set the bus-width to 4 bits in the DT
<apritzel> MoeIcenowy: can you try to add "bus-width = <4>;" to the mmc section in sun50i-a64-pine64-common.dtsi?
Andy-D has joined #linux-sunxi
Da_Coynul has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
kaspter has joined #linux-sunxi
diego71_ has joined #linux-sunxi
ninolein has quit [Ping timeout: 258 seconds]
ninolein has joined #linux-sunxi
diego71 has quit [Ping timeout: 244 seconds]
gianMOD has joined #linux-sunxi
egbert has joined #linux-sunxi
egbert_ has quit [Ping timeout: 264 seconds]
cnxsoft has joined #linux-sunxi
gianMOD has quit [Ping timeout: 276 seconds]
dave0x6d has joined #linux-sunxi
Da_Coynul has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
vagrantc has joined #linux-sunxi
cnxsoft1 has joined #linux-sunxi
cnxsoft has quit [Read error: Connection reset by peer]
cnxsoft1 is now known as cnxsoft
vagrantc has quit [Ping timeout: 252 seconds]
dave0x6d has quit [Read error: Connection reset by peer]
dave0x6d has joined #linux-sunxi
vagrantc has joined #linux-sunxi
p1u3sch1 has quit [Ping timeout: 260 seconds]
gianMOD has joined #linux-sunxi
p1u3sch1 has joined #linux-sunxi
gianMOD has quit [Ping timeout: 260 seconds]
kaspter has quit [Read error: Connection reset by peer]
vagrantc has quit [Quit: leaving]
ikmaak has quit [Ping timeout: 276 seconds]
fire219 has quit [Read error: Connection reset by peer]
gianMOD has joined #linux-sunxi
gianMOD has quit [Ping timeout: 250 seconds]
kaspter has joined #linux-sunxi
IgorPec has joined #linux-sunxi
jernej has joined #linux-sunxi
IgorPec has quit [Read error: Connection reset by peer]
IgorPec has joined #linux-sunxi
reinforce has joined #linux-sunxi
p1u3sch1 has quit [Ping timeout: 240 seconds]
p1u3sch1 has joined #linux-sunxi
IgorPec has quit [Ping timeout: 276 seconds]
kaspter has quit [Read error: Connection reset by peer]
kaspter has joined #linux-sunxi
jernej has quit [Ping timeout: 276 seconds]
<MoeIcenowy> apritzel: 134217728 bytes (134 MB, 128 MiB) copied, 11.398 s, 11.8 MB/s
<MoeIcenowy> Congrats!
<MoeIcenowy> but I think the line should be added into sun50i-a64.dtsi
<MoeIcenowy> mmc0/1 came with 4 lines, and mmc2 came with 8
<mripard> no, it depends on how the board is wired
<MoeIcenowy> mripard: oh it's right...
<dave0x6d> MoeIcenowy pine64?
<MoeIcenowy> dave0x6d: yes
<dave0x6d> I gave up trying to get mine to run Linux.
<MoeIcenowy> developing based apritzel's a64-v5 branch
<dave0x6d> Said screw it and went to android.
<MoeIcenowy> dave0x6d: you can try longsleep's modificated BSP
<dave0x6d> I did.
<dave0x6d> was extremely sluggish.
<dave0x6d> took several minutes on each boot.
<MoeIcenowy> dave0x6d: ah-oh
<MoeIcenowy> what's your distro
<MoeIcenowy> on my board booting needs only 22s (mainline kernel
<dave0x6d> I tried a few ubuntu and debian builds.
<dave0x6d> probably just gonna stick with armbian devices in the future.
<dave0x6d> the Orange Pi One is perfect
<dave0x6d> wait not the one, the Orange Pi that's $10.
p1u3sch1 has quit [Ping timeout: 250 seconds]
p1u3sch1 has joined #linux-sunxi
gianMOD has joined #linux-sunxi
<NiteHawk> jonkerj: after blanking your "user talk" page, are you okay with deleting it?
<longsleep> MoeIcenowy: certainly not the Kernel to blame if it takes minutes to boot
<longsleep> dave0x6d: ^^^
gianMOD has quit [Ping timeout: 260 seconds]
<longsleep> dave0x6d: why does it take minutes to boot? Most likely its waiting for dhcp on eth0 (5 minute timeout in Ubuntu by default)
mossroy has joined #linux-sunxi
apritzel has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
lemonzest has joined #linux-sunxi
<agraf> MoeIcenowy: 12MB/s is the max speed at 25Mhz
<agraf> MoeIcenowy: for 4 lines (which again is the max you can get)
<agraf> MoeIcenowy: are you using an SDHC card?
<agraf> MoeIcenowy: if so, it should be able to go to 50Mhz which should give you an x2 performance increase
yann has joined #linux-sunxi
<montjoie> mripard: so you want that I rollback to standalone wens phy driver ?
<mripard> If it's doable, yes
<mripard> if not, no
<mripard> but the phy should still be represented as one in the DT
<mripard> then don't create a different driver
<jonkerj> NiteHawk: yes
<wens> mripard: so you want a separate DT node for the ephy?
<jonkerj> I'll do that right away
<mripard> yes, and rob seems to be saying the same thing
<wens> however that node would be different from the mdio phy
<NiteHawk> jonkerj: you can't. that's why I'm asking - I'd remove it for you
<jonkerj> cool, tnx!
<wens> mripard: i don't see rob suggesting anything about it
<wens> mripard: he only wants the power sequencing stuff under the phy node, not under the mac node, to which i agree looks better
<wens> and in line with the generic power sequencing library some people are pushing for
<mripard> ok, I misunderstood him then
<mripard> but still, why should we model an external PHY and an internal one differently
<mripard> ?
<wens> why do you think we model them differently?
<mripard> well, in one case every property for the phy are merged in the emac node
<wens> the only difference is the "allwinner,use-internal-phy" flag, which tells the glue driver to turn on and switch over to the internal one
<mripard> and in the other one, it's in a separate node
<mripard> I'm not talking about use-internal-phy
<wens> you still have a phy node for the internal one
<mripard> I'm talking about the clocks, resets, leds
<wens> that's the binding of the mdio bus
<wens> oh
<wens> so you think we should move them into the phy node, correct?
Worf has joined #linux-sunxi
ricardocrudo has joined #linux-sunxi
matthias_bgg has joined #linux-sunxi
<mripard> yes, for example
tsuggs has quit [Ping timeout: 258 seconds]
Worf has quit [Quit: Konversation terminated!]
Mr__Anderson has joined #linux-sunxi
tsuggs has joined #linux-sunxi
<rellla> what is a good starting point to get information about drm/kms drivers, kernel/userspace etc. and how this all works and fits together in general?
<buZz> i'd say /Documentation in kernel tree, has some info
<buZz> then sourcecode
jstein has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
<MoeIcenowy> agraf: I'm using a Class 10 card
<jelle> is anyone working on mainlining the nanopi boards ( of friendlymarm) ?
Worf has joined #linux-sunxi
<buZz> jelle: those arent allwinners, are they?
<jelle> it's one big mess it seems :P
<buZz> thats not friendlyarm
<jelle> awesome I'm 100% confused
apritzel has joined #linux-sunxi
<jelle> buZz: sigh
<jelle> but nanopi?
gianMOD has joined #linux-sunxi
<buZz> but?
<buZz> nanopi or nanopi
<buZz> :D
<jelle> haha the h3 nanopi
<jelle> their nanopi-m3 is not an allwinner woah
<jelle> buZz: see this confused me http://wiki.friendlyarm.com/wiki/index.php/NanoPi_M1
<jelle> > The NanoPi M1 is an Allwinner H3 based ARM board designed and released by FriendlyARM for hobbyists
<jelle> I give up on logic
<buZz> ;)
<buZz> i think i saw some H3 mainlining efforts earlier
<buZz> dont really see anything special about this board that it wouldnt work with the existing efforts
<jelle> yup
<jelle> it should work just fine
gianMOD has quit [Ping timeout: 264 seconds]
Worf has quit [Quit: Konversation terminated!]
premoboss has joined #linux-sunxi
kronicd has quit [Ping timeout: 264 seconds]
premoboss has quit [Quit: Sto andando via]
premoboss has joined #linux-sunxi
<montjoie> mripard: wens So if I have understand, in sun8i-h3 I need to add an "ephy" node (with syscon clock/reset)?
<wens> my idea is to put it in the phy node that is a child of the emac node
<MoeIcenowy> I'm trying to play GBA game on my A33 tablet with VBA-M
<MoeIcenowy> but... I got 10 fps...
<MoeIcenowy> Then I tried NDS, and the result is also 10fps...
<MoeIcenowy> is this the performance issue of simplefb?
gianMOD has joined #linux-sunxi
<wens> mripard: fyi i'm going to do the clk notifier stuff for a31
<mripard> wens: great, that would be a nice test
afaerber has joined #linux-sunxi
<wens> i already tried settings multipliers first, wait for lock, set dividers (for cases where multipliers increase or dividers decrease), didn't work
gianMOD has quit [Remote host closed the connection]
Netlynx has joined #linux-sunxi
gianMOD has joined #linux-sunxi
<zoobab_> someone has already used the new GPIO dev entries provided by 4.6?
<wens> i only remember seeing a patch for gpio labels
massi has joined #linux-sunxi
Da_Coynul has joined #linux-sunxi
<zoobab_> let me check
<zoobab_> "The GPIO stack has been changed to make the GPIO drivers real
<zoobab_> devices. For this reason, a new userspace ABI has been created: the
<zoobab_> GPIO character device. This release takes small steps, so first a pure
<zoobab_> *information* ABI has been added, along with the tool "lsgpio" that
<zoobab_> will list all GPIO devices on the system and all lines on these
<zoobab_> devices. GPIOs can now be properly discovered from userspace (but not
aalm has quit [Ping timeout: 260 seconds]
<zoobab_> be used from userspace). The old sysfs ABI is still available opt-in
<zoobab_> (and can be used in parallel) and will be maintained around for the
<zoobab_> foreseeable future, but it will not be extended"
<jonkerj> zoobab_: just tried it, lsgpio seems to work
gianMOD has quit [Remote host closed the connection]
<slapin> as I understand you still can't use GPIOs from character device though, only sysfs for now
IgorPec has joined #linux-sunxi
Da_Coynul has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
gianMOD has joined #linux-sunxi
<zoobab_> @slapin speed is the main difference
<slapin> zoobab_: yes, but as I understand it is not implemented yet
<mripard> zoobab_: not really
<mripard> if you know that open / write / close is slow, and don't use that in your driver
<mripard> s/driver/code/
<mripard> it won't really make any difference
cptG has joined #linux-sunxi
<slapin> if that would implement gpio grouping, that would be useful interface,,,
<zoobab_> I thought that speed was the main difference :-)
<MoeIcenowy> apritzel: how to replace ATF in your built image?
<MoeIcenowy> (I don't know what U-Boot to use now, so I cannot build my own...
<KotCzarny> TheLinuxBug: powervr?
<MoeIcenowy> (but I want an ATF to enable eldo1 regulator
cptG_ has quit [Ping timeout: 264 seconds]
<slapin> is ATF required for A64 or I can get off with u-boot only?
<slapin> and where can I gate 3 x pine64 cheap?
gianMOD has quit [Remote host closed the connection]
<slapin> *get I mean buy...
cnxsoft has quit [Read error: Connection reset by peer]
IgorPec has quit [Ping timeout: 244 seconds]
cnxsoft has joined #linux-sunxi
gianMOD has joined #linux-sunxi
<MoeIcenowy> is "ssh_dispatch_run_fatal: Connection to 10.42.0.139 port 22: message authentication code incorrect" a problem rasied by ARM erratum in Cortex A53?
apritzel1 has joined #linux-sunxi
iamfrankenstein1 has joined #linux-sunxi
<apritzel> MoeIcenowy: I see that blaming silicon errata for any random problem you encounter is tempting, but usually it's not the case ;-)
iamfrankenstein1 has quit [Read error: Connection reset by peer]
<apritzel> MoeIcenowy: for the ATF issue: I had quite some success with U-Boot 2016.07-rc1 and boot0, so I suggest you try this one
<MoeIcenowy> slapin: welcome to the gaint hole of A64!
<apritzel> slapin: according to TLLim Pine64 boards are on stock and will be delivered promptly these days
<apritzel> slapin: so shop.pine64.com would be the answer
<apritzel> slapin: and yes: ATF is required for A64
<apritzel> slapin: but actually it's very easy and quick to build
gianMOD has quit [Remote host closed the connection]
JohnDoe_71Rus has joined #linux-sunxi
massi has quit [Remote host closed the connection]
afaerber has quit [Quit: Ex-Chat]
premoboss has quit [Quit: Sto andando via]
<KotCzarny> O.o
<slapin> whet set of stuff one should buy to get proper development environment with pine64?
<slapin> MoeIcenowy, apritzel: ^
<slapin> to be able to do everything, even after corrupting bootloaders and needing to debug one?
<slapin> KotCzarny: the alum box is easier to cool
<KotCzarny> i wonder how well it works
<slapin> KotCzarny: well, it worked of i7-in-a-box, it should work for H3 too
<MoeIcenowy> slapin: what do you want?
<KotCzarny> if you know anyone with cnc machine you can make one for yourself
<MoeIcenowy> BSP or patched mainline?
<slapin> KotCzarny: probably I shoud make such a box myself from alum scrap...
<KotCzarny> box from pentium radiators? ;)
chrishell has joined #linux-sunxi
<slapin> MoeIcenowy: I want to be able to debug ATF and u-boot, JTAG should be nice, serial console is a must
<MoeIcenowy> I don't think you can have UART
<MoeIcenowy> Pine64 have only a sd slot at SDC0
<MoeIcenowy> s/UART/JTAG
<slapin> KotCzarny: that is too dramatic, on one of previous jobs we put fanless PC boards into boxe made of 5mm aluminium plates crewed together with bolts
<MoeIcenowy> UART is available on both Euler Bus and Expansion Bus
<apritzel> slapin: so far I was quite happy with UART debugging, you can easily print something from within ATF and U-Boot
<slapin> MoeIcenowy: what should I buy to have it all working (I have a several dozen of USB-UART converters, so no problem there), is there any addon boards I need for that?
<apritzel> slapin: for quick turnarounds I'd recommend an USB-A-to-A cable to use FEL booting
<apritzel> slapin: any UART adapter that does 3.3V should work
<slapin> apritzel: I'm fine with UART debugging too, JTUG just makes things more quick (like from months of debugging to days)
fire219 has joined #linux-sunxi
<jonkerj> I noticed that all my fake FTDI/USB did not work with my orangepi+ at 115200
<jonkerj> while all my ch341's seem to work
afaerber has joined #linux-sunxi
<jonkerj> (5x FTDI, different sellers, 3x CH341, different sellers)
<jonkerj> so I would recommend a random eBay CH341, as long as it has the 3.3v jumper
<slapin> jonkerj: all mine work though, buyght all that about 5 years ago, got a ton though
<apritzel> slapin: I guess that depends what kind of issues you debug, but for the Pine64 development UART was good enough so far
<jonkerj> err, ch340
<apritzel> I tend to copy some three assembly lines into code to dump single characters into the TX register to see where it hangs
<slapin> apritzel: anyway do I need to buy anuthing in addition to board itself to have UART console, Ethernet, HDMI output?
<jonkerj> slapin: mine seem to work at any speed below 115200, they seem out of sync with my pi's
<apritzel> slapin: no
<jonkerj> I can send/receive characters, but it is all garbled
<jonkerj> bitflips all over
<apritzel> decent USB power supply with microB connector and probably an micro SD card
<apritzel> basically when you run an OrangePi or RaspberryPi already, you are all set
<apritzel> in this regard the Pine64 isn't different at all
<slapin> jonkerj: well, if these are crystal-less, then the chip is just buggy, otherwise i fixed all my FTDI and prolific ones replacing crystal with proper one.
<jonkerj> that's actually a very good idea
<jonkerj> probably best not to source those replacements from shenzen :-)
afaerber has quit [Ping timeout: 250 seconds]
<slapin> jonkerj: well, I sourced mine from Hongkong :)
afaerber has joined #linux-sunxi
<slapin> jonkerj: they just need to be proper ones
<jonkerj> how does one distuingish a proper one from a dodgy one before ordering?
<slapin> jonkerj: well, metal-cased ones were all good for me
<slapin> also it is a good idea when fixing $1 USB hub from aliexpress
<jonkerj> I've got several of these ones: http://tomeko.net/other/ESP8266/FTDI.jpg
<jonkerj> unless the xo is in a ceramic package, I'm afraid it's operating on internal osc
<slapin> ah, these are crystalless, they just have really bad capacitors IIRC
<jonkerj> well, the $1 fix is to get a ch340g :-)
<jonkerj> the ftdi's seem to do fine for my other hobby (cortex-M / libopencm3)
<slapin> I fixed these by changing some of capacitors, check that yours have all capacitors soldered
<slapin> also one of capacitors have to be replaced by tantalum one
<slapin> but these are promising, as you can make them work onlike green ones which pretend to be prolific and have frequencies so much off, I used them all as cheapo GPIO adaptor as they have CTS/RTS :)
<slapin> don't buy green ones...
<slapin> :)
gianMOD has joined #linux-sunxi
<slapin> well, I found-out that FTDI chips are expensive thing, and when you see the board for $2 with FTDI chip, it is either fake or did not pass quality control and is sold just because they can
<jonkerj> yeah I know
<jonkerj> real FTDI is kind of expensive
<slapin> however CH341 and CP2X are really cheap and decent
<jonkerj> check out CH340/CH341, afaik the genuine ones are cheap
<jonkerj> yeah
<slapin> and I never met device so badly made it did not work with them
<jonkerj> you could use ch341 as usb-twi/usb-spi/usb-1w
<MoeIcenowy> my FTDI board seems to be $5...
<jonkerj> probably fake
<MoeIcenowy> (bought from Waveshare
<slapin> on the red ones, there is nothing wrong with chip itself, it is problem with board assembly
<slapin> as I know, fake FTDIs all work, unlike prolifics...
<slapin> I mean fake ones
fl_0 has quit [Ping timeout: 272 seconds]
<slapin> prolific chips are cheap, fake or not, but they require complicated board design, so people don't like them
<slapin> so it is easy to find out by photo that you buy crap with prolifics
apritzel1 has quit [Ping timeout: 244 seconds]
<slapin> I think when buying from ali it is better choise to just get ch34c or cp2x* and not stuff your head
jelly has quit [Quit: ""]
<jonkerj> agree
fl_0 has joined #linux-sunxi
<slapin> what PSU is best for pine64? if I'm going to stuff it with HDMI and SATA?
<slapin> also do it have IR like orangepi?
gianMOD has quit [Remote host closed the connection]
apritzel1 has joined #linux-sunxi
<KotCzarny> slapin: sata?
<KotCzarny> as in usb-uasp?
<slapin> btw, do I need special USB-SATA for UAS to work?
<slapin> doesn't pine64 have SATA port?
<KotCzarny> only a10/a20 chip has real sata
<KotCzarny> everything else is just usb-sata chip slapped beside soc
<KotCzarny> and most of them (especially gl830) just suck
<MoeIcenowy> so there's no different with a good external USB portable hdd
<MoeIcenowy> maybe even worse
<KotCzarny> for uas you have to buy usb-uas adapter
<MoeIcenowy> if you're using a really good usb hdd
<KotCzarny> then you can reach 40MB/s instead of 30-35
<KotCzarny> and probably lower usb overhead
JohnDoe_71Rus has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
JohnDoe_71Rus has joined #linux-sunxi
<TheLinuxBug> KotCzarny: was looking at those H8 boards have you ever used one? are they worth it for the processing power or?
cnxsoft has quit [Quit: cnxsoft]
apritzel1 has quit [Ping timeout: 244 seconds]
<KotCzarny> i only tried a20/h3
<TheLinuxBug> ahh well I was seeing it 'on sale' bu pcDuino for $39 USD
<KotCzarny> but h8 has pvr, which means even worse support
<TheLinuxBug> couldn't find much info on them
<TheLinuxBug> ahh
<TheLinuxBug> thats info I was looking for
<TheLinuxBug> so even less GPU support?
<TheLinuxBug> so basically use it with Android if you want GPU
<TheLinuxBug> ?
<KotCzarny> yup
<TheLinuxBug> doesn't seem to have many 'extras' on board
cnxsoft has joined #linux-sunxi
<TheLinuxBug> 1 USB, HDMI, gigabit nic, 1gb ram
<TheLinuxBug> i was kinda wondering how gigabit worked since all those CPU
<TheLinuxBug> and if it can actually hit gigabit
<TheLinuxBug> or is restricted to 40M/sec like others
<TheLinuxBug> but don't see many reviews on the board
<huawei> suck pvr
<TheLinuxBug> and wiki documentation is lacking
<KotCzarny> i wouldnt expect stellar gbit performance on allwinner
cnxsoft has quit [Client Quit]
<TheLinuxBug> well is performance any better out of PVR than Mali as far as Android?
<TheLinuxBug> supposedly is does h265
<TheLinuxBug> what is funny is in description it says gigabit nic then on board next to the port it has '100M'
<KotCzarny> you know, a20 can do gbit one way
<KotCzarny> but in duplex its more to 900/600
<KotCzarny> or something
<KotCzarny> i mean total ~1600
<TheLinuxBug> hmm
<TheLinuxBug> fully tweaked I have never seen over 50M/sec
<TheLinuxBug> upload from it
<TheLinuxBug> I have a PcDuino Nano Lite
<TheLinuxBug> that drives my NAS project
<TheLinuxBug> has raid5 attached on SYBA pm/raid card
<TheLinuxBug> but
<TheLinuxBug> that may more be limitations of SATA
<TheLinuxBug> like high IO + high nic IO it just slows down
kronicd has joined #linux-sunxi
<TheLinuxBug> ill have to setup an iperf and see
<TheLinuxBug> ya max is def 50M/sec that I can hit [ 3] 0.0-60.0 sec 2970 MBytes 49.5 MBytes/sec
<TheLinuxBug> let me try other way
<TheLinuxBug> slightly more performant
<TheLinuxBug> but looks like it will be about 60M/sec
<TheLinuxBug> [ 3] 0.0-60.0 sec 3367 MBytes 56.1 MBytes/sec
<TheLinuxBug> from what it reported 415Mbits out by 471Mbits in
<TheLinuxBug> you know any tricks to get more?
<TheLinuxBug> I wouldn't mind 900/600
<huawei> TheLinuxBug, in a20?
<TheLinuxBug> yes, A20
jelly-home has joined #linux-sunxi
<TheLinuxBug> and thats with http://paste.ee/p/iriN4 in place
<huawei> TheLinuxBug, irq affert peroformance seriously in arm
reinforce has quit [Quit: Leaving.]
<huawei> i don't know why x86 can handle irq with little performance impact
chrishell has left #linux-sunxi [#linux-sunxi]
<apritzel> I don't think that "IRQ performance" in general is worse on ARM
<apritzel> it's mostly the rather poor design of those devices that affect performance
<apritzel> or for that matter the much more clever design of common devices in the x86 world
jelly-home is now known as jelly
<apritzel> for instance to avoid too many interrupts in the first place
<apritzel> not to speak of things like the bus connection and the limited overall system bandwidth that the common cheap ARM SoCs provide
dave0x6d has quit [Read error: Connection reset by peer]
BenG83 has joined #linux-sunxi
gianMOD has joined #linux-sunxi
gianMOD has quit [Ping timeout: 264 seconds]
apritzel1 has joined #linux-sunxi
leio_ has joined #linux-sunxi
leio has quit [Ping timeout: 250 seconds]
apritzel1 has quit [Ping timeout: 244 seconds]
vagrantc has joined #linux-sunxi
kronicd has quit [Ping timeout: 240 seconds]
jernej has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
<KotCzarny> TheLinuxBug: did you try iperf alone without disk access?
gianMOD has joined #linux-sunxi
<TheLinuxBug> KotCzarny: I am not aware that it actually stores the data at all
<TheLinuxBug> I don't see any options to not store data
<TheLinuxBug> you can choose what is transmitted
<TheLinuxBug> but I don't see a way to select where it would write data
<TheLinuxBug> I would assume /dev/null
<TheLinuxBug> I am however using TCP and not UDP
<TheLinuxBug> UDP may have different results
<TheLinuxBug> ewww
kronicd has joined #linux-sunxi
<TheLinuxBug> abysmally slow actually
<TheLinuxBug> using UDP
<TheLinuxBug> 3] 0.0- 4.3 sec 0.53 MBytes 0.13 MBytes/sec 0.423 ms
<TheLinuxBug> interesting though, pushing both directions at once seems almost more performant
<TheLinuxBug> well at least one direction
apritzel1 has joined #linux-sunxi
<TheLinuxBug> [ 5] 0.0-60.0 sec 858 MBytes 14.3 MBytes/sec
<TheLinuxBug> [ 4] 0.0-60.0 sec 3838 MBytes 64.0 MBytes/sec
<TheLinuxBug> when run up and download at same time
<TheLinuxBug> hmm
<TheLinuxBug> more interesting is running it again
<TheLinuxBug> doesn't replicate
<TheLinuxBug> slower second time
<KotCzarny> set governor to performance?
<TheLinuxBug> should already be there
<TheLinuxBug> ya
<TheLinuxBug> just checked and I manually set min freq also to max
<TheLinuxBug> so it should be running full throttle
<TheLinuxBug> looks like when doing -d I can get up to about 70M/sec for short periods incoming, but outgoing is then very slow, but alone it doesn't go past 58-60M/sec max
<TheLinuxBug> ok I may eat my hat
<TheLinuxBug> not sure what changed but just ran it again
<TheLinuxBug> and for a while saw
<TheLinuxBug> [ 3] 1.0- 2.0 sec 81.0 MBytes 81.0 MBytes/sec
<TheLinuxBug> [ 3] 2.0- 3.0 sec 81.5 MBytes 81.5 MBytes/sec
<TheLinuxBug> [ 3] 3.0- 4.0 sec 81.8 MBytes 81.8 MBytes/sec
<TheLinuxBug> [ 3] 4.0- 5.0 sec 80.9 MBytes 80.9 MBytes/sec
<TheLinuxBug> [ 3] 5.0- 6.0 sec 80.6 MBytes 80.6 MBytes/sec
<KotCzarny> magic
<TheLinuxBug> but then it dropped back to only 50M/sec again
<KotCzarny> try differnt switch?
<TheLinuxBug> hmm
<KotCzarny> unplug other devices?
<TheLinuxBug> the end result (average)
<TheLinuxBug> [ 3] 0.0-60.0 sec 3676 MBytes 61.3 MBytes/sec
<KotCzarny> kill x, unplug hdmi?
<TheLinuxBug> these are only two really moving any traffic on a 5 port switch
<KotCzarny> disable console
<TheLinuxBug> x isn't running
<TheLinuxBug> don't thin hdmi is plugged
ricardocrudo has quit [Remote host closed the connection]
<TheLinuxBug> seems I got about 20 secons at 80M/sec
<KotCzarny> check if there are other interrupt sources
<TheLinuxBug> then rest of the 60 second was at 50M/sec
cosm has joined #linux-sunxi
gianMOD has quit [Remote host closed the connection]
formruga has quit [Quit: Konversation terminated!]
TheSeven has quit [Ping timeout: 258 seconds]
TheSeven has joined #linux-sunxi
<KotCzarny> crazy setup, hehe
<KotCzarny> i would route cable beautifully though
tlwoerner has quit [Ping timeout: 272 seconds]
tlwoerner has joined #linux-sunxi
<TheLinuxBug> interesting
reinforce has joined #linux-sunxi
tlwoerner has quit [Ping timeout: 244 seconds]
JohnDoe_71Rus has quit [Remote host closed the connection]
JohnDoe_71Rus has joined #linux-sunxi
ricardocrudo has joined #linux-sunxi
yann has quit [Ping timeout: 240 seconds]
tlwoerner has joined #linux-sunxi
tlwoerner has quit [Ping timeout: 258 seconds]
apritzel1 has quit [Ping timeout: 244 seconds]
matthias_bgg has quit [Quit: Leaving]
tlwoerner has joined #linux-sunxi
popolon has quit [Ping timeout: 252 seconds]
lemonzest has quit [Quit: Leaving]
apritzel has quit [Ping timeout: 244 seconds]
apritzel has joined #linux-sunxi
Hahaburger has joined #linux-sunxi
yann has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
Mr__Anderson has quit [Remote host closed the connection]
IgorPec has joined #linux-sunxi
IgorPec2 has joined #linux-sunxi
IgorPec has quit [Ping timeout: 240 seconds]
popolon has joined #linux-sunxi
IgorPec has joined #linux-sunxi
IgorPec2 has quit [Ping timeout: 276 seconds]
IgorPec2 has joined #linux-sunxi
IgorPec has quit [Ping timeout: 250 seconds]
popolon has quit [Quit: WeeChat 1.4]
popolon has joined #linux-sunxi
<montjoie> TheLinuxBug: you bench which soC ?
<KotCzarny> i think something a20 based
Hahaburger has quit [Ping timeout: 250 seconds]
afaerber has quit [Quit: Ex-Chat]
<TheLinuxBug> BananaPi M1 (A20)
<TheLinuxBug> or sorry
<TheLinuxBug> in this case its the pcDuino Nano Lite (A20)
<TheLinuxBug> forgot I switched them out cause I needed the composite out somewhere else
<TheLinuxBug> But I do have the BPi M1 sitting around here too now
<vagrantc> MoeIcenowy: apritzel mentioned you had some patches for USB support on the pine64?
arossdotme has quit [Quit: Ex-Chat]
<TheLinuxBug> montjoie: ^
<vagrantc> sounds promising!
<sW`> however only the bottom USB port works
<vagrantc> one is better than none
Mr__Anderson has joined #linux-sunxi
<fire219> can confirm that it works :)
<fire219> it's sad that the top port doesn't work, but the reason why is perfectly understandable
IgorPec2 has quit [Read error: Connection reset by peer]
IgorPec has joined #linux-sunxi
JohnDoe_71Rus has quit [Quit: KVIrc KVIrc Aria 4.9.2, revision: git-6627-gf708225, sources date: 20160102, built on: 2016-05-12 17:32:15 UTC git-6627-gf708225 http://www.kvirc.net/]
cosm has quit [Quit: Leaving]
IgorPec has quit [Read error: Connection reset by peer]
IgorPec has joined #linux-sunxi
Netlynx has quit [Quit: Leaving]
IgorPec2 has joined #linux-sunxi
IgorPec has quit [Ping timeout: 272 seconds]
IgorPec2 has quit [Read error: Connection reset by peer]
IgorPec has joined #linux-sunxi
iamfrankenstein has quit [Quit: iamfrankenstein]
Mr__Anderson has quit [Remote host closed the connection]
reinforce has quit [Quit: Leaving.]
Gerwin_J has quit [Quit: Gerwin_J]
_whitelogger has joined #linux-sunxi
iamfrankenstein has joined #linux-sunxi
<slapin> ah, looks like if I could find some good 3d-printable parts for laptop case, that'd omtivate me to finish my 3D printer....
IgorPec has quit [Quit: Nettalk6 - www.ntalk.de]
tlwoerner has quit [Quit: Leaving]
apritzel has joined #linux-sunxi
ricardocrudo has quit [Ping timeout: 272 seconds]
bobryan has quit [Ping timeout: 240 seconds]
bobryan has joined #linux-sunxi
OverCR has quit [Quit: Leaving.]
mossroy has quit [Quit: Quitte]
ricardocrudo has joined #linux-sunxi
mosterta|2 has quit [Ping timeout: 240 seconds]
Seppoz has joined #linux-sunxi
Seppoz has quit []
Da_Coynul has joined #linux-sunxi
BenG83 has quit [Quit: Leaving]
Da_Coynul has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
mosterta|2 has joined #linux-sunxi
vagrantc has quit [Quit: leaving]
ssvb has joined #linux-sunxi
Da_Coynul has joined #linux-sunxi
kimberly_ has quit [Quit: Page closed]
jstein has quit [Read error: Connection reset by peer]
paulk-collins has quit [Quit: Leaving]
apritzel has quit [Ping timeout: 244 seconds]