<DuClare>
Aha, looks like these are really new since grep didn't find such things on my end :)
<DuClare>
Still wondering about why mosi would show up as an input
<wens>
yeah, it's quite new
lurchi_ has quit [Ping timeout: 252 seconds]
<DuClare>
Luckily there shouldn't be anything to stop me from updating my kernel (it's pretty new anyway), but I think bitbanged spi should work even without pulls
lurchi_ has joined #linux-sunxi
lurchi_ is now known as lurchi__
<wens>
looks like the spi bit-bang system sets the mosi to input in rx mode
<wens>
was done in commit 4b859db2c6069 to support 3 wire mode
<wens>
probably harmless for normal SPI
mpmc has quit [Ping timeout: 268 seconds]
[TheBug] has quit [Ping timeout: 268 seconds]
reinforce has joined #linux-sunxi
tl_lim has quit [Quit: Leaving]
clemens3_ has joined #linux-sunxi
jstein_ has joined #linux-sunxi
jstein_ is now known as jstein
mpmc has joined #linux-sunxi
<DuClare>
Ok, I don't know why, but simply updating kernel made bitbanged spi work
<DuClare>
(From 5.1-rc1 to 5.1-rc7)
tnovotny has joined #linux-sunxi
yann has quit [Ping timeout: 246 seconds]
_whitelogger has joined #linux-sunxi
mpmc has quit [Ping timeout: 268 seconds]
mpmc has joined #linux-sunxi
yann has joined #linux-sunxi
<fALSO>
bom dia
tnovotny has quit [Quit: Leaving]
jstein has quit [Quit: quit]
tnovotny has joined #linux-sunxi
suprothunderbolt has quit [Ping timeout: 246 seconds]
Rafael1980 has joined #linux-sunxi
s_frit has quit [Remote host closed the connection]
s_frit has joined #linux-sunxi
agraf has quit [Read error: Connection reset by peer]
agraf has joined #linux-sunxi
tnovotny has quit [Ping timeout: 255 seconds]
Mangy_Dog has joined #linux-sunxi
victhor has joined #linux-sunxi
tnovotny has joined #linux-sunxi
megi has joined #linux-sunxi
mpmc has quit [Ping timeout: 246 seconds]
mpmc has joined #linux-sunxi
mpmc has quit [Ping timeout: 245 seconds]
voodoovood2 has joined #linux-sunxi
voodoovood has quit [Ping timeout: 250 seconds]
mpmc has joined #linux-sunxi
Nyuutwo has quit [Quit: No Ping reply in 180 seconds.]
Nyuutwo has joined #linux-sunxi
mpmc has quit [Ping timeout: 246 seconds]
mpmc has joined #linux-sunxi
victhor has quit [Ping timeout: 252 seconds]
<DuClare>
GPIO_PULL_UP doesn't actually work in 5.1rc7
<DuClare>
Nevermind, I'm stupid
shfil has joined #linux-sunxi
<willmore>
DuClare, specifically how so that the next person won't make that mistake.
<wens>
I suggest working on linux-next
dev1990 has joined #linux-sunxi
<DuClare>
willmore: You need parenthesis around the operands, <&pio 2 0 GPIO_ACTIVE_HIGH | GPIO_PULL_UP> won't work :)
JohnDoe_71Rus has quit [Quit: KVIrc KVIrc Aria 5.0.0, revision: 5.0.0+git-7416-9b35d44fd, build type: debug, sources date: 20160102, built on: 2019-01-23 17:02:15 UTC 5.0.0+git-7416-9b35d44fd http://www.kvirc.net/]
<willmore>
That's a good thing to make note of, thanks, DuClare.
mpmc has quit [Ping timeout: 244 seconds]
[TheBug] has joined #linux-sunxi
mpmc has joined #linux-sunxi
mpmc has quit [Ping timeout: 252 seconds]
pgreco has quit [Ping timeout: 246 seconds]
[TheBug] has quit [Read error: Connection timed out]
pgreco has joined #linux-sunxi
mpmc has joined #linux-sunxi
mhlavink has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
_whitelogger has joined #linux-sunxi
mpmc has quit [Ping timeout: 246 seconds]
mpmc has joined #linux-sunxi
gnufan_home has joined #linux-sunxi
nuuuciano_ has joined #linux-sunxi
gnufan_home1 has quit [Ping timeout: 245 seconds]
lurchi__ is now known as lurchi_
jaho has joined #linux-sunxi
mpmc has quit [Ping timeout: 245 seconds]
mpmc has joined #linux-sunxi
nuuuciano_ has quit [Ping timeout: 246 seconds]
reinforce has joined #linux-sunxi
lo has joined #linux-sunxi
lo is now known as Guest60566
[TheBug] has joined #linux-sunxi
Guest60566 has quit [Quit: Leaving]
mpmc has quit [Ping timeout: 246 seconds]
JohnDoe_71Rus has joined #linux-sunxi
lurchi_ is now known as lurchi__
mpmc has joined #linux-sunxi
<DuClare>
Yeah this kinda baffles me
mpmc has quit [Ping timeout: 245 seconds]
<DuClare>
I have a bunch of spi devices on one bus
<DuClare>
All but one of them work with allwinner,sun8i-h3-spi
<DuClare>
All of them appear to work with bitbanged spi
<DuClare>
The one that doesn't work with the built-in spi always reads high for every bit
mpmc has joined #linux-sunxi
Andy-D has joined #linux-sunxi
mpmc has quit [Ping timeout: 246 seconds]
afaerber has joined #linux-sunxi
mpmc has joined #linux-sunxi
mpmc has quit [Ping timeout: 258 seconds]
mpmc has joined #linux-sunxi
tllim has joined #linux-sunxi
lurchi__ is now known as lurchi_
Gerwin_J has joined #linux-sunxi
lurchi_ has quit [Ping timeout: 258 seconds]
selfbg has quit [Ping timeout: 245 seconds]
Gerwin_J has quit [Quit: Gerwin_J]
nuuuciano has joined #linux-sunxi
<ndufresne>
paulk-leonov, _cedrus_write_ref_list is doing a reference frame (capture buffer) lookup to pick the field and check if it has bottom field in it
<ndufresne>
paulk-leonov, that one is completely wrong
<paulk-leonov>
ndufresne, most likely indeed
<paulk-leonov>
interlaced wasn't tested on h264
<ndufresne>
in the ref list, you tell the decode to use top or bottom
<paulk-leonov>
it's very likely broken
<ndufresne>
with that code, as soon as bottom is decoded it will alway use bottom
<paulk-leonov>
keeping note of that
<paulk-leonov>
will fix when revising h264
<ndufresne>
what that uAPI meant I believe for using .field, was to read the .field value on the slice buffer (on OUTPUT queue)
<ndufresne>
or something, but I think it just does not work
<paulk-leonov>
ndufresne, I'm not sure our VPU does deinterlacing though
<paulk-leonov>
(cedrus)
<ndufresne>
it does not matter really
<ndufresne>
deinterlacing is external
<ndufresne>
got a go
<paulk-leonov>
alright then
<ndufresne>
paulk-leonov, along with that information, we also need to pass the view id (MVC), so if we extend we should make space for that one too
<paulk-leonov>
great, noted
<ndufresne>
we then need to decide if we limit to 2 views, which is what most decoder do
<paulk-leonov>
ideally we should allow more with more controls
<paulk-leonov>
but not sure how well that can fit in
cnxsoft1 has quit [Quit: cnxsoft1]
<paulk-leonov>
basically, anything that's an h264 extension should get its own separate control IMO
fl_0 has quit [Ping timeout: 246 seconds]
fl_0 has joined #linux-sunxi
lurchi_ has joined #linux-sunxi
lurchi_ is now known as lurchi__
mpmc has quit [Ping timeout: 255 seconds]
mpmc has joined #linux-sunxi
lurchi_ has joined #linux-sunxi
lurchi__ has quit [Ping timeout: 252 seconds]
netlynx has joined #linux-sunxi
mpmc has quit [Ping timeout: 245 seconds]
mpmc has joined #linux-sunxi
grw has quit [Quit: WeeChat 1.4]
gnufan_home1 has joined #linux-sunxi
gnufan_home has quit [Ping timeout: 246 seconds]
victhor has joined #linux-sunxi
wwilly has joined #linux-sunxi
lurchi_ has quit [Ping timeout: 245 seconds]
lurchi_ has joined #linux-sunxi
lurchi_ is now known as lurchi__
lurchi__ has quit [Quit: Konversation terminated!]
lurchi__ has joined #linux-sunxi
popolon has joined #linux-sunxi
clemens3_ has quit [Ping timeout: 244 seconds]
jstefanop has joined #linux-sunxi
yann has quit [Remote host closed the connection]
lurchi__ has quit [Ping timeout: 258 seconds]
<ndufresne>
paulk-leonov, so you'd make multiple array of ref-list in seperate control, it's a bit ...
<ndufresne>
but it would be extensible
lurchi__ has joined #linux-sunxi
<paulk-leonov>
ndufresne, I guess we would have the extra ref lists in the control as well
<ndufresne>
"the extra" ?
<ndufresne>
I was just saying that we have N ref, which right now are index in the dpb entry, but for each of the ref, we also need some more data, right now we need to know if it's a top or bottom reference, and if it's view 0 or 1
<ndufresne>
that being said, for h264, I doubt anything will be added in the future
<paulk-leonov>
ah sorry I mixed things up -- we should definitely support interleaved in the main controls
lurchi_ has joined #linux-sunxi
lurchi__ has quit [Ping timeout: 246 seconds]
tnovotny has quit [Quit: Leaving]
<ndufresne>
paulk-leonov, something that bugs me if the nuance between a reference list and the dpb, normally those two have similar info, but the order is different, I haven't checked the spec though yet
<ndufresne>
do you know if we really pass dpb order to the kernel atm ?
voodoovood has joined #linux-sunxi
voodoovood2 has quit [Ping timeout: 255 seconds]
lurchi_ has quit [Read error: Connection reset by peer]
lurchi_ has joined #linux-sunxi
vagrantc has joined #linux-sunxi
<paulk-leonov>
ndufresne, I'm pretty sure it's reworked to not evict entries that are not needed for the current frame but might be needed later, and the list is ordered to that each buffer keeps the same index in it
<ndufresne>
ok, I'll keep that in mind
zoums has quit [Ping timeout: 255 seconds]
zoums has joined #linux-sunxi
mpmc has quit [Ping timeout: 258 seconds]
mpmc has joined #linux-sunxi
AneoX_ has joined #linux-sunxi
AneoX has quit [Ping timeout: 250 seconds]
lurchi_ has quit [Ping timeout: 245 seconds]
lurchi_ has joined #linux-sunxi
clemens3_ has joined #linux-sunxi
HeavyMetal has quit [Remote host closed the connection]