<MoeIcenowy>
consider the suniv family, which is a newly created SoC with ARM926EJ-S
<libv>
KotCzarny: mimicking the "community" section on the wiki?
<MoeIcenowy>
and there are still very cheap Allwinner-based OTT boxes
<MoeIcenowy>
(considering Amlogic is well developing recently, Allwinner has now only the lowest end market
<KotCzarny>
libv: for example. but really any info
<libv>
but would you really waste your time with them if you can buy a fanless atom based intel machine for 100?
<MoeIcenowy>
100 what?
<MoeIcenowy>
CNY?
<KotCzarny>
usd
<libv>
usd/eur/gbp, whatever
<MoeIcenowy>
but I can buy an Allwinner for 100 CNY.
<libv>
not worth the hassle tbh
<libv>
anyway, KotCzarny, good point
<libv>
mnemoc is spending his spare time reworking the server now
<libv>
and it makes sense to go there
<MoeIcenowy>
To be honest I don't think Allwinner have gain a lot of market from linux-sunxi
<MoeIcenowy>
many of their mass customers are still part of the strange Chinese electronics
<MoeIcenowy>
which just accepts their weird BSP
<libv>
but for the rest of the world, what market is left?
<KotCzarny>
tinkerers
<libv>
development boards only
<libv>
and people get disappointed quick with what bpi/opi provides itself
<libv>
and then land with armbian, or with linux-sunxi directly
<libv>
in as much as bpi/opi does not already mostly depend on the sunxi community
<KotCzarny>
xunlong is a nice bsp provider
<MoeIcenowy>
well Allwinner is although weak, not so weak that they will care individual customers
<KotCzarny>
but they are mostly hw company
<libv>
KotCzarny: to that point, just like with ati, arm, etc, why not sell more hw and give some support to the software people
<MoeIcenowy>
they still have OEMs that make products
<MoeIcenowy>
because Allwinner always try to give out a full solution (although always dirty too)
<libv>
a full solution never is throwing something bad over the fence
<MoeIcenowy>
their solution-based business model is also why they name the same SoC under different series letters
<libv>
a full solution for development boards is providing full docs, and some development resources, but the first is needed to ensure actual long term support
<MoeIcenowy>
because each letter will be distributed with different SDK
<MoeIcenowy>
in addition, to prevent SDK being used on the same chip with other letter, they closed libdram
<MoeIcenowy>
(because they added identification code there)
<MoeIcenowy>
libv: they never care development boards market
<libv>
but the development boards, and the products being made from them, is soon all they that will have left
<gediz0x539>
interesting complaint. hardware vendor complaining about a fully volunteered community effort? eh...
<libv>
gediz0x539: not fully
<KotCzarny>
gediz: i bet on hw vendor not knowing who owns the site
<gediz0x539>
libv: sorry, some obfuscated SDKs?
<gediz0x539>
KotCzarny: highly likely
<libv>
bootlin got paid by c.h.i.p. a allwinner spin-off kickstarter as that was a .us company doing a kickstarter with hw with a well known and then very prominent GPL violation shit throwing contest
<libv>
which they only got wise to a week or so into the kickstarter
<MoeIcenowy>
some SBC vendors do pay
<MoeIcenowy>
but nothing is got from Allwinner itself
<MoeIcenowy>
well, someone with @allwinnertech.com did the mainline of A100, but he's just doing this in part time
<libv>
MoeIcenowy: which is a smart thing to do, this employee might get a better paid job elsewhere because of that
<gediz0x539>
libv: oh i did not know chip had a relation with AW
<libv>
gediz0x539: some former employees were at the now long defunct Next Thing Co, and no-one else was getting such a low price for those chips
<libv>
but really, if we had not stepped up the gpl violation noise a few months earlier, next thing co would've continued on its original path
<libv>
(before anyone complains, we had 3 years trying to convince allwinner of anything, and communication always ended up stalling even when playing nice)
<MoeIcenowy>
By the way it's said that Allwinner will deliver a RISC-V chip the next year
<libv>
anyway, nice to know that allwinner complains
<MoeIcenowy>
let's wonder how they can do on that chip
<libv>
but then, nothing will come of that either, it's allwinner, we've been there before countless times
<libv>
MoeIcenowy: allwinner is far from alone
<libv>
so no edge to be had there
<MoeIcenowy>
yes, they're far from alone
_whitelogger has joined #linux-sunxi
s_frit_ has quit [Ping timeout: 260 seconds]
s_frit has joined #linux-sunxi
camus1 has joined #linux-sunxi
kaspter has quit [Ping timeout: 272 seconds]
camus1 is now known as kaspter
jstein has quit [Ping timeout: 272 seconds]
shailangsa has quit [Ping timeout: 240 seconds]
_whitelogger has joined #linux-sunxi
niceplace has quit [Ping timeout: 260 seconds]
niceplace has joined #linux-sunxi
s_frit_ has joined #linux-sunxi
s_frit has quit [Ping timeout: 240 seconds]
jstein has joined #linux-sunxi
jstein has quit [Ping timeout: 260 seconds]
Mangy_Dog has joined #linux-sunxi
cnxsoft has joined #linux-sunxi
mps has quit [Ping timeout: 272 seconds]
cnxsoft1 has quit [Ping timeout: 265 seconds]
bauen1 has quit [Read error: Connection reset by peer]
netlynx has joined #linux-sunxi
bauen1 has joined #linux-sunxi
<pgreco>
mripard: any idea if all the Ethernet fixes will make their way to 5.9.x and/or 5.10.x?
<pgreco>
ack, thanks. I'll keep them in my trees for now
<mripard>
eventually I guess they will, but the contentious part seems to be in the driver
<mripard>
and the DT changes are being held until it's figured out
jonkerj has quit [Quit: brb]
<pgreco>
are there any other clear "victims" of the patch?
jonkerj has joined #linux-sunxi
<MoeIcenowy>
I think everyone is the victim
<MoeIcenowy>
except part of original Pine A64+ boards can benefit
<MoeIcenowy>
mripard: I think the driver patch at least broke DT backward compatibility
<MoeIcenowy>
and as I know Fedora now depends on DT backward compatibility (they do not switch DT when update kernel, but include a new kernel DT in U-Boot when distributing system)
<pgreco>
MoeIcenowy: you mean that if they were to revert the patch, dt fixes be reverted too?
<MoeIcenowy>
I don't know what will happen
<MoeIcenowy>
at least this patch shouldn't break so many boards
<pgreco>
I think Fedora updates dtb symlink to the "latest" installed
hanni76 has quit [Remote host closed the connection]
lurchi_ is now known as lurchi__
lurchi__ is now known as lurchi_
<jernej>
mripard: I'm not in one of those threads, but at least for pine64-plus board we have to be able to override HW delay settings - that was actually the fix provided by Realtek
lkcl has quit [Ping timeout: 265 seconds]
<jernej>
can you give that info to net maintainers?