mmind00 changed the topic of #linux-rockchip to: Rockchip development discussion | public log at https://libera.irclog.whitequark.org/linux-rockchip
Danct12 has quit [Quit: ZNC 1.9.1 - https://znc.in]
Danct12 has joined #linux-rockchip
asriel has quit [Quit: Don't drink the water. They put something in it to make you forget.]
Danct12 has quit [Remote host closed the connection]
Danct12 has joined #linux-rockchip
asriel has joined #linux-rockchip
<dsimic> mmind00: I was really surprised to see QNAP using an RK3568, and I was a bit disappointed to see only 4 GB of DRAM for the price they charge for it :/
<dsimic> if it had 8 GB of RAM, it would be perfect :)
<dsimic> it would still be a bit expensive, but you've got to pay for its proprietary software as well
<dsimic> regarding the MaskROM header, perhaps QNAP followed the reference Rockchip design :)
<dsimic> of course, nice job on making the TS-433 supported better in mainline kernel :)
<dlg> yes, nice work
Livio has joined #linux-rockchip
Livio has quit [Ping timeout: 260 seconds]
warpme has joined #linux-rockchip
eballetbo has joined #linux-rockchip
Livio has joined #linux-rockchip
franoosh has joined #linux-rockchip
erg_ has joined #linux-rockchip
ldevulder has joined #linux-rockchip
<mmind00> dsimic: that "proprietary software" is definitly overpayed though ... looking at their firmware made my eyes bleed
<dsimic> frankly, I'm not surprised to hear that
<dsimic> what was actually wrong with its firmware? what do they actually use?
stikonas has joined #linux-rockchip
<mmind00> dsimic: I just don't like vendor firmwares in general ... the reason I bought the device was the rk3568 and thus the ability to hopefully replace everything without going mad
<mmind00> dsimic: but from what I read qnap software has a tendency to phone home a lot
<mmind00> so I actually never ran its intended firmware and just did go to the replace-it step directly ;-)
raster has joined #linux-rockchip
<mmind00> their rescue-system uses library versions that Debian doesn't even have in oldoldstable anymore ... base-kernel is of course a Rockchip 5.10 and some "hal" binary crap on top that abstracts talking to an embedded controller that does things like fan control
<mmind00> (over a serial port :-) )
stikonas has quit [Remote host closed the connection]
<dsimic> oh, I don't like vendor software in general :)
<dsimic> phoning home is especially bad :/
<dsimic> yes, I read about the serial-port-connected MCU in your patches... do you know what MCU is that?
<dsimic> have you taken the NAS apart to see it?
<mps> dsimic: I wonder who likes it (vendor software)
<dsimic> vendors, probably :)
<mps> haha
<dsimic> :D
<mps> you made me to smile after bad night, thanks
<dsimic> hearing that makes me smile, too :)
<mps> nice to hear
kevery has quit [Ping timeout: 260 seconds]
kevery has joined #linux-rockchip
<mmind00> dsimic: nope, haven't taken the device apart ... qnap engineers had the foresight to provide all the necessary headers easily accessible through the drive bay opening ;-)
<mmind00> dsimic: https://www.neowin.net/reviews/review-of-the-qnap-ts-233-a-two-bay-nas-for-modest-needs/ ... similar model (2 bay, 2GB, quite a bit cheaper) ... on the 2nd image you can even see the maskrom header labeled ... now that is service :-D
warpme has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<dsimic> I see, headers are nicely labeled :)
<dsimic> it seemingly uses a Foresee eMMC chip
<dlg> is it an rk356x?
<diederik> yes, an rk3568 to be exact
<dlg> nice
<diederik> mmind00: ran `rngtest` yet ? ;-)
Livio has quit [Ping timeout: 260 seconds]
<mmind00> diederik: haha nope ... only marginally followed the rng series so far
<diederik> thanks to it I discovered that my new AMD CPU/APU has a bug wrt HWRNG ... it's not working (while it does work on my older AMD CPU)
<qschulz> diederik: which AMD CPU :) I'm "about to" build my new NAS based on one
<naoki> about ROCK Pi E, I can see "irq29: nobody cared" oops. how can I check what is IRQ29?
<diederik> qschulz: AMD Ryzen 5 5500GT
<diederik> ... which I bought for my (to be build) NAS system ;-)
<naoki> GIC_SPI 29?
<naoki> (I guess no)
<CounterPillow> Probably, when reading the TRM make sure you don't take the interrupt numbers at face value but for SPI interrupt factor in the offset those have
<diederik> qschulz: I'm currently triaging it, but it could be that the latest microcode update fixes it ... but currently working to figure out which one I'm actually running
<naoki> ... oh I found a bug in rk3328-rock-pi-e.dts...
<qschulz> diederik: nice :) i'll be going for a PRO 8300GE i think for myself, want an APU in there (replace my RPi4 with LibreElec) and ECC and use AM5 socket for future-proof-ability :)
<diederik> yeah, I had 64G of RAM laying about, so figured I'd better find a use for that ;)
<diederik> (and concluded, possibly incorrectly, that the GE variants weren't for sale to 'normal' consumers)
<qschulz> diederik: they aren't, but you can buy them from "OEMs"
<diederik> yeah, as part of a complete system (IIUC) ... which is not what I was looking for
<qschulz> nah, I also don't want a complete system
<qschulz> you are not having AMD warranty though through that reseller channel, so...
<qschulz> basically the reseller buys loads of those SoCs on trays, then they repackage them into invidiual ones
<qschulz> AIFAU
<qschulz> AFAIU
<diederik> ah ok, the shop where I usually buy stuff, doesn't have those GE variants. I wouldn't care about no AMD warranty as you'd rarely if ever need it. But in this case I'm glad I bought it at my local shop
<qschulz> diederik: it was actually a quite painful experience to come up with the components for the NAS I want to build. And I still haven't bought any parts :)
<qschulz> I was waiting for the Jonsbo N4 to be available in Europe, but it wasn't for two months so was even considering taking the risk to buy it from AliExpress
<qschulz> but it is now available in Europe, and even cheaper than on AliExpress, make it make sense
<diederik> I'm just re-using old parts, I just replaced MB+CPU+RAM
<qschulz> :) I have a microserver gen8 right now, not really upgradable :)
<diederik> To make it more/back on-topic, I'd been advocating for sth like the Rock 5 ITX for years, so that people could re-use their old PC case/PSU/etc to build their own energy efficient NAS
<diederik> Hence why I recently remarked that I thought that board was the first product I thought was actually usuable
<naoki> (I'm noooooob...)
warpme has joined #linux-rockchip
naoki has quit [Ping timeout: 255 seconds]
naoki has joined #linux-rockchip
mripard has joined #linux-rockchip
System_Error has quit [Remote host closed the connection]
System_Error has joined #linux-rockchip
warpme has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
warpme has joined #linux-rockchip
<wens> re-did my quartz64 sd card, and now ethernet doesn't work ...
<wens> qschulz: why the APU version? the standard versions after ryzen 7000 also have iGPUs?
<qschulz> wens: Yes, standard versions to have the APU. But IIRC you have non-APU versions now (which arent' the default anymore)
<qschulz> s/to/now do/
urja has quit [Read error: Connection reset by peer]
Stat_headcrabed has joined #linux-rockchip
Stat_headcrabed has quit [Quit: Stat_headcrabed]
naoki has quit [Quit: naoki]
franoosh has quit [Quit: Leaving]
urja has joined #linux-rockchip
warpme has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<wens> ah, right. I'm missing the PHY driver
<wens> diederik: thanks!
<wens> qschulz: right, but only the lowest Ryzen 7000 lacks the iGPU. I guess the iGPU in the APU is a bit better?
<qschulz> wens: For me iGPU = APU, did I miss something?
<qschulz> wens: I'm not sure to understand what you're asking/discussing :)
<phh> APU refers to the whole SoC, with both iGPU and CPU
<wens> qschulz: ah, for me APU is the ones with the G suffix, which AMD specifically markets as APU.
<qschulz> phh: thx
<qschulz> wens: marketing decided to change the naming recently IIRC
<qschulz> wens: so 7-series doesn't have the G, but has the iGPU
Stat_headcrabed has joined #linux-rockchip
<wens> qschulz: yeah. I don't think it's marketing. the GPU included is bare minimum.
<wens> probably just to be on par with Intel offerings
<wens> qschulz: so I was wondering why you would go for non retail Ryzen 8000 PRO APU instead of a standard Ryzen 7000 with internal GPU
<qschulz> wens: and the 8-series doesn't have a non-<insert letter> variant, G(e) or F it seems?
<qschulz> wens: more recent, more better :D ?
<wens> qschulz: 8-series is all APUs, so G or F (which lacks GPU)
<qschulz> wens: wqould that make the F variant not an APU?
<wens> qschulz: it's still the same architecture as 7-series. AMD's APU offerings are one step behind
<wens> qschulz: that's where it's confusing :( IIRC the older APU lines didn't have the -F variant
<qschulz> wens: I need to check the "TDP"
<qschulz> the GE is capped to 35W
<qschulz> and since it's going to run 24/7, it's nice to not consume too much
<qschulz> though TDP != idle, but you know...
<wens> lol
<wens> you can tweak the actual power cap in the BIOS though
<qschulz> IIRC, you cannot buy PRO variant anyway, so G or GE won't change much for my case
<wens> I'm waiting for the 9-series to come out at the end of the month
<qschulz> i don't remember if the 7-series had official ECC support though
<qschulz> wens: I can wait too, considering the 8-series APU I want to buy will have me wait for a month at least :)
<wens> All the Ryzen CPUs, minus the non-PRO APUs, have ECC support
<qschulz> wens: no
<qschulz> not the newest series
<wens> 9 or 7?
<qschulz> 8 for sure
<qschulz> it was months ago I looked into it :/
<wens> That's the APU series, for which only the PRO variants do ECC
<wens> APUs are the even numbered series
<qschulz> i'm getting really confused now, if the 7-series is not an APU but has an iGPU in it, what's what?
<wens> 8-series = 7-series + beefed up GPU
<wens> the APU distinction kind of fails after the 7-series
warpme has joined #linux-rockchip
<wens> think of APUs as AMD's latest CPU + GPU (shrunk down mobile variant) offerings in one budget package?
<qschulz> wens: ok, are there any public prices already?
<wens> for 9-series?
<qschulz> yup
<wens> I don't know how realistic these are though
<qschulz> yeah, this is like almost twice the price of my 8300GE :D
<wens> probably twice the power as well, which you probably don't need for a NAS
<diederik> wens: does that mean there's a(n easy) fix?
<qschulz> wens: yes, i'm highly expecting the 8300GE to be way overkill but the thrill of building your first PC is calling me :)
<wens> diederik: easy fix for ethernet on quartz64? IIRC it was having the PHY driver builtin?
<wens> will try it later
<wens> I don't think it needed any changes, otherwise it would've been in my local tree and wouldn't have broken :p
<wens> mmind00: FYI VPD seems to be "Vital Product Data" or data on a specific componet, like stuff you get when you run dmidecode on a PC or server
<mmind00> wens: thanks, that make sense ... from looking what normally lives on those addresses, it's probably some eeprom(s) I guess
<diederik> wens: with me it was enabling networking in u-boot that broke networking in Linux
dsimic has quit [Ping timeout: 252 seconds]
dsimic has joined #linux-rockchip
<diederik> oh, did you mean "having the PHY driver builtin" *into the kernel*? Because that's something that Debian (and I) normally don't do ...
<diederik> with "builtin into the kernel" I thought it needed to be =y, while I/Debian uses =m
<diederik> I'll go through the mails again, maybe I notice sth now which I didn't before
psydroid2 has joined #linux-rockchip
warpme has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
erg_ has quit [Ping timeout: 252 seconds]
warpme has joined #linux-rockchip
warpme has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
mripard has quit [Quit: mripard]
Stat_headcrabed has quit [Quit: Stat_headcrabed]
Stat_headcrabed has joined #linux-rockchip
Stat_headcrabed has quit [Client Quit]
Livio has joined #linux-rockchip
raster has quit [Quit: Gettin' stinky!]
kevery has quit [Ping timeout: 265 seconds]
<wens> diederik: OK, that didn't fix things for me
Livio has quit [Ping timeout: 264 seconds]
kevery has joined #linux-rockchip
Livio has joined #linux-rockchip
kevery1 has joined #linux-rockchip
stikonas has joined #linux-rockchip
kevery has quit [Read error: Connection reset by peer]
kevery1 is now known as kevery
kevery1 has joined #linux-rockchip
kevery has quit [Ping timeout: 245 seconds]
kevery1 is now known as kevery
psydroid2 has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/]
naoki has joined #linux-rockchip
naoki has quit [Quit: naoki]
kevery1 has joined #linux-rockchip
kevery has quit [Ping timeout: 260 seconds]
kevery1 is now known as kevery
kevery has quit [Remote host closed the connection]
kevery has joined #linux-rockchip
eballetbo has quit [Quit: Connection closed for inactivity]
kevery1 has joined #linux-rockchip
kevery has quit [Ping timeout: 252 seconds]
kevery1 is now known as kevery
kevery1 has joined #linux-rockchip
kevery has quit [Ping timeout: 276 seconds]
kevery1 is now known as kevery
krei-se has quit [Ping timeout: 244 seconds]
krei-se has joined #linux-rockchip
ldevulder has quit [Quit: Leaving]
naoki has joined #linux-rockchip
Livio has quit [Ping timeout: 276 seconds]
<naoki> mmind00: rk356x mipi csi is usable state? or we need rkisp1(?) driver?
kevery1 has joined #linux-rockchip
kevery has quit [Ping timeout: 276 seconds]
kevery1 is now known as kevery