14:09 <z3ntu1> bamse: seems because of commit "ARM: dts: qcom: strip prefix from PMIC files" the dtb for HTC One Mini 2, msm8926-htc-memul (which you applied some hours ago) doesn't build. Can you fix up the #include yourself or shall I send a patch?

14:35 <aka_[m]> konradybcio: regarding that last discussion bout schema&compats is it how it supposed to be?

14:35 <aka_[m]> https://github.com/Kiciuk/linux/commit/e4957d734568abff3bc8c30376896f521ba0d20f

20:01 <aka_[m]> lumag: konradybcio i guess per-process page tables for adreno aren't really supported right?

20:01 <aka_[m]> i mean legacy adrenos

20:01 <aka_[m]> i was wondering where gtcu/gtbu clocks come from on 8976/8917/8909 and it seems those are related

20:02 <lumag> aka_[m], no, it only works on a6xx and smmu500

20:02 <lumag> There were patches for a5xx, if I remember correctly

20:02 <aka_[m]> so it would not hurt to not add them

20:02 <aka_[m]> on 8953 no longer these sits under adreno node

20:03 <aka_[m]> lumag: any idea if "alwayson" on adreno is consumed somewhere?

20:03 <aka_[m]> i have it specified and if i take it away adreno fails to start

20:03 <aka_[m]> yet dt-checker doesn't like it being here

20:03 <lumag> aka_[m], huh?

20:03 <aka_[m]> <&gcc GCC_GFX3D_OXILI_AON_CLK>; on A510

20:04 <lumag> Ah

20:04 <aka_[m]> 8953 with A508 also have AON clock

20:04 <lumag> Feel free to add them to the dtb

20:04 <aka_[m]> if its removed adreno fails to start

20:04 <lumag> And to the schema

20:04 <aka_[m]> is it activated by clk core or something?

20:04 <aka_[m]> i cannot find it in adreno_X.c

20:04 <lumag> Doesn't adreno activate all clocks from DT?

20:05 <aka_[m]> no idea how is that even performed, i tried to search for clk_get

20:06 <lumag> aka_[m], there is a call to devm_clk_bulk_get_all() in msm_gpu.c:get_clocks

20:06 <aka_[m]> ah

20:06 <aka_[m]> so i need just add to schemas

20:06 <aka_[m]> seems like gtcu/gtbu still usefull, comment just got changed

20:07 <aka_[m]> on 8909 source they said its required on a405

20:08 <aka_[m]> ok so more schema fun

20:09 <aka_[m]> oh boy i don't understand much yet half of commits are already for schemas

20:09 <lumag> you might want to check if they are applicable to the GPU itself or to SMMU

20:15 <aka_[m]> they all sit under gpu node

20:18 <aka_[m]> uh GCC_GFX3D_TCU_CLK seems now be under two

20:18 <aka_[m]> i guess i should drop it from gpu

20:25 <aka_[m]> ok im kinda content for today, almost everything is in almost upstreamable-state

20:26 <aka_[m]> now push myself to send these

20:26 <aka_[m]> and i can jump into getting icc prepared and camera&venus

22:25 <lumag> aka_[m], sounds great!