<bjdooks>
^ I can't believe they pulled riscv support because they insist on using RDCYCLE directly instead of a perfectly good vdso inteface... idiots
agentcasey has quit [Quit: Lost terminal]
<bjdooks>
that wasted an hour of yesterday morning looking into a fdsdk failure on a milkv
Trifton has joined #riscv
Trifton has quit [Ping timeout: 276 seconds]
Trifton has joined #riscv
kilic has joined #riscv
craigo has joined #riscv
Trifton_ has joined #riscv
Trifton has quit [Ping timeout: 260 seconds]
Trifton2 has joined #riscv
Trifton_ has quit [Ping timeout: 255 seconds]
ldevulder has joined #riscv
danilogondolfo has joined #riscv
Armand|X230 has joined #riscv
jacklsw has quit [Quit: Back to the real world]
bbbju has quit [Quit: Client closed]
Andre_Z has joined #riscv
jedesa has joined #riscv
<gurki>
well. it _does_ sound like youre blaming the wrong guy who just had to deal with what was handed down to him ...
<bjdooks>
I'm not sure, it's difficult to track stuff across three issue trackers
<bjdooks>
I'm wondering about adding a test for RDCYCLE aborting and just emulate it and return
<bjdooks>
we can't seem to get the sbi_pmu to enable it, so wondering if the SBI on this board has disabled it higher up or something
damian101 has joined #riscv
damian101_ has quit [Ping timeout: 264 seconds]
jacklsw has joined #riscv
fuwei has joined #riscv
paddymahoney has quit [Ping timeout: 260 seconds]
jacklsw has quit [Quit: Back to the real life]
fuwei has quit [Quit: Konversation terminated!]
ldevulder has quit [Quit: Leaving]
aethlas has quit [Ping timeout: 248 seconds]
aethlas has joined #riscv
fuwei has joined #riscv
Guest9beatty has joined #riscv
heat has joined #riscv
heat has quit [Read error: Connection reset by peer]
<conchuod>
I probably should have tagged it for stable, but I think Sasha picked it up earlier today.
<mps>
conchuod: you are Conor Dooley
<mps>
?
<conchuod>
Yup, that I am.
<mps>
nice, thanks again
<conchuod>
No worries, nice to have it in before 6.11 was finalised.
<mps>
yes, now I don't have to maintain separate kernel for alpine
<Tenkawa>
this Spacemit(R) X60 is quite a mystery
<mps>
Tenkawa: agree fully
<Tenkawa>
mps: found some "unique" behaviour in it today
<Tenkawa>
this is what I posted to edolnx over on Discord earlier about its high cpu load problem I knew about already
<mps>
I found that 6.6 crashes very often :)
<Tenkawa>
You may/not have seen the cpu high load avg I told you about.. After some reading/disussion with others we have been able to stabilize it to 0.. but found an interesting side effect... maxcpus has to be set to 4... which is very "intriguing"
<Tenkawa>
Once you allow the 5th core to go online the RCU immediately becomes unstable
<Tenkawa>
It acts like it "is" a split core system even though it says its not
<mps>
ah, interesting
<Tenkawa>
And I can't find good docs to qualify/rule in either direction
<Tenkawa>
its completely stable currently on 4 cpus with 6.6.36 now though
<Tenkawa>
but if I enable the 5th... boom
<mps>
hm, thanks for info, I will test and see
<Tenkawa>
look at this post too
<mps>
but with their 6.1.15 it is stable except this high load