f_ has quit [Read error: Connection reset by peer]
f_ has joined #linux-msm
marvin24 has joined #linux-msm
marvin24_ has quit [Ping timeout: 480 seconds]
deathmist has quit [Remote host closed the connection]
jhovold has joined #linux-msm
deathmist has joined #linux-msm
pespin has joined #linux-msm
rawoul has quit [Remote host closed the connection]
<aka_[m]>
lumag: at this point im not sure if i should cry or laugh,
<aka_[m]>
Got some more schemas done with select it does not select any other anymore but on adreno-506.yaml does not validate a510, however second yaml with a530 validates 8996 dtb just fine
<aka_[m]>
a530 has single enum(should be const i guess)
<lumag>
that's a matter of preference
<lumag>
aka_[m], what's the problem with a506 schema?
<aka_[m]>
no idea its almost exactly same schema, same spaces i even compared in meld
<aka_[m]>
haven't tried to validate against 8953
<aka_[m]>
now rebuilding kernel just to be sure it didn't had cached dts
<aka_[m]>
if i get it sorted i would probably send it as rfc
<aka_[m]>
on 8996 it warns about clock orders well
<aka_[m]>
only a530 seems to validate
<aka_[m]>
a306 also fails
<aka_[m]>
lumag:
<aka_[m]>
im going to do something
<aka_[m]>
imagine DT-schema does not validate nodes with status="disabled"
<lumag>
Ah :D
<aka_[m]>
not sure but it seems so
<aka_[m]>
checked 8916 dts i wanted to validate and it had no gpu enabled
<aka_[m]>
which is weird because i expected it to have it for sure
<krzk>
aka_[m]: disabled nodes are incomplete, so they should not be validated, of course
<krzk>
you would have warnings all over the place otherwise
<aka_[m]>
krzk: i hope there would be one place with tips&tricks about yaml docs with various examples for avg dumb duck
<krzk>
or I misunderstood and you actually experience disabled node being validated, thus warnings? This was fixed loyear ago or so, so maybe you use old dtschema?
<aka_[m]>
nah
<aka_[m]>
disabled was not validated
<aka_[m]>
and i wondered why it didn't pick up even after removing props
<krzk>
so what is the surprise? How do you imagine disabled node being validated? how could it possibly work?
<aka_[m]>
its still compiled in dtb right
<aka_[m]>
why not for each node validate
<krzk>
it is, and it is incomplete
<krzk>
Because you would have millions of unrelevant errors
<aka_[m]>
so if i do status= "disabled" i can send incomplete dts and mark it "unfinished"?
<krzk>
It would make dtschema useless - printing all the errors which are not relevant, because board did not enable something
<aka_[m]>
well, now i know, better late than ever
<krzk>
That's not dtschema problem, but writing DTS. Please look at DTS coding style
<aka_[m]>
*never
<aka_[m]>
ok seems i should slowly get my shit together and prepare for job
<krzk>
The entire point of DTSI and DTS is to represent (in some way) the hardware, thus you have SoC=DTSI and board=DTS.
<krzk>
The DTSI has pieces from the SoC. The easiest example is: supplies are usually board specific, so they are in DTS, but they are required by schema. So imagine a board which did not enable display, because there is no display, so display nodes are disabled and no supplies are provided.
<krzk>
What do you expect dtschema to do with it? supplies are requires, so should it complain? No. dtschema here follows the convention how DTS (and DTSI) should be written
Daanct12 has joined #linux-msm
svarbanov has quit [Ping timeout: 480 seconds]
Daaanct12 has joined #linux-msm
Daanct12 has quit [Read error: Connection reset by peer]
<robher>
krzk: Disabled nodes are validated except for 'required'. IOW, if a property is there, it still has to be correct.
Daaanct12 has quit [Quit: WeeChat 4.2.1]
<krzk>
robher: ah, right, that's the change I forgot
svarbanov has joined #linux-msm
svarbanov has quit [Remote host closed the connection]
svarbanov has joined #linux-msm
pespin has quit [Remote host closed the connection]