Danct12 changed the topic of #msm8937-mainline to: Boot Linux on your MSM8917/37/40 and QM215 mobile! | GitHub: https://github.com/msm89x7-mainline | Logs: https://oftc.irclog.whitequark.org/msm8937-mainline
<Danct12[m]> <barni2000[m]> "is ```qcom,init-seq``` in 28nm..." <- no but it can be supported by editing the code
<Danct12[m]> a little more improvement
<Danct12[m]> added CCI (Cache Control Interface) mailbox btw
<lanik123[m]> I'm alive... I don’t recomend anyone to get sick with new covid
<lanik123[m]> <Danct12[m]> "added CCI (Cache Control..." <- lol, you use gpll0 twice as clk on cci
<Danct12[m]> lanik123[m]: that's what downstream says
<lanik123[m]> Ye, I know
<lanik123[m]> This is a 140+ iq move
<Danct12[m]> <lanik123[m]> "I'm alive... I don’t recomend..." <- get better soon :\
<lanik123[m]> ty
<lanik123[m]> <Danct12[m]> "i already have a interconnect..." <- https://github.com/konradybcio/linux-interconnect-driver-generator
<Danct12[m]> oh wow that exists
<Danct12[m]> this is where i got my interconnect driver from
<Danct12[m]> it unfortunately doesn't seem to work yet as i get a blue screen as soon the driver initializes
<Danct12[m]> * unfortunately i haven't hooked it up to dts yet and i get a blue screen as soon the driver initializes
<lanik123[m]> <hacker420[m]> "limiting 8937 to 4 cores makes..." <- bruh
<lanik123[m]> 1 cluster is faster then two without cpr?
<lanik123[m]> s/1/one/
<Danct12[m]> my guess is that the bootloader set the clock speed on little cores because thats the core gets booted first
<lanik123[m]> You right but that doesn't explain why 4 cores faster than 8
<lanik123[m]> Hm
<lanik123[m]> Actually I have encountered this problem
<jojo_autoboy[m]> wasn't it because of internconnect
<jojo_autoboy[m]> posted a screenshot somewhere above
<lanik123[m]> jojo_autoboy[m]: Yes
<lanik123[m]> Y need add cpr driver
<lanik123[m]> s/Y/You/
<jojo_autoboy[m]> ah
<Danct12[m]> speaking of interconnect the newer platforms require them, i was working on fxtec pro1x (sm6115) yesterday and the touch screen i2c bus keeps timing out and never recovered until i added interconnect driver
<lanik123[m]> lanik123[m]: This worked on my old mainline
<lanik123[m]> <Danct12[m]> "it unfortunately doesn't seem to..." <- Have you already launched into PMOS?
<Danct12[m]> lanik123[m]: i use buildroot inside initramfs for now
<Danct12[m]> so not in pmos
<Danct12[m]> i know it sounds.. weird but that's what i do the most
<lanik123[m]> Hm, ok. I have a black screen on 8917 and 8937
<Danct12[m]> does it just show the framebuffer console then everything fades?
<Danct12[m]> or splash then it fades
<lanik123[m]> lanik123[m]: Kicuk says 8953 have the same problem after 6.5
<lanik123[m]> Danct12[m]: +
<lanik123[m]> lanik123[m]: I probably know why...
<Danct12[m]> lanik123[m]: mdp clocks are disabled
<Danct12[m]> * mdp clocks/pmdomain are
<lanik123[m]> Danct12[m]: Nope, it's working
<lanik123[m]> And clock have correct freq
<Danct12[m]> ah, this is the mdp driver
<lanik123[m]> Yep, but I think there may be problems with phy freq
<lanik123[m]> Early it can be fixed by CLK_GET_RATE_NOCACHE flag on pclk/byte clk in gcc