ChanServ changed the topic of #linux-msm to:
Daanct12 has joined #linux-msm
Daanct12 has quit [Quit: WeeChat 4.3.5]
Daanct12 has joined #linux-msm
marvin24_ has joined #linux-msm
marvin24 has quit [Ping timeout: 480 seconds]
mripard has joined #linux-msm
zstas has joined #linux-msm
pespin has joined #linux-msm
<krzk> Neither next nor next/pending-fixes boot on armv7 Exynos boards due to NULL ptr around memory management.
<krzk> Anyone also saw it on armv7 qcom boards?
<Rayyan> which commit breaks it?
<marc|gonzalez> konradybcio: your email bounced with "Konrad Dybcio no longer works for Linaro." hence my confusion.
<konradybcio> ah.. konradybcio@kernel.org
<konradybcio> krzk i can't even get earlycon lol
<marc|gonzalez> lumag: thank you for being an island of rationality
<marc|gonzalez> I was about to rage quit
<marc|gonzalez> konradybcio: then you didn't see my iommu RFC? :(
<marc|gonzalez> I'll resend it just to you
<marc|gonzalez> lumag: did you get the impression the TDP158 was sitting on an i2c bus but could not be reached over the bus? (confused)
<krzk> konradybcio: ok, so everyone is ducked up, not only Exynos. Good.
<lumag> marc|gonzalez, how is that possible?
<phh> krzk: it doesn't look like https://pastebin.com/EVfLpHEj does it?
<lumag> marc|gonzalez, re IOMMU. I haven't seen anything requiring comments. In other words, yes, it's known that some of -v2 require additional handling. Yes, there were patches from kholk. Yes, they were not accepted.
<konradybcio> krzk let me quickly spin up v6.10 to make sure it's not an issue with my setup..
<phh> okay nothing like what I had with Android Common Kernel. (on my armv7, Linux v6.6 boots, but Android Common Kernel v6.6 doesn't, so if you had the same crash as I do, that would mean a regression from a @google.com, but that's not the case)
<krzk> phh: sorry, we do not talk about forks here. I said explicitly NEXT.
<krzk> Whatever you have in Android common kernel is not here relevant - it's android's fork.
<phh> i thought my parenthesis was clear enough why i mentioned it, but nevermind, ok
<marc|gonzalez> lumag: re: iommu - I don't know about kholk's patches. I tested Angelo's. As I stated, he declares a list of context banks not available to Linux, a bit like "reserved-memory"
<lumag> Yeah, Angelo
<marc|gonzalez> kholk = Angelo ?
<lumag> His IRC nick was kholk if I'm not mistaken
<marc|gonzalez> Do you think his solution would be acceptable to robclark?
<lumag> marc|gonzalez, have you checked email threads wrt those patches?
<marc|gonzalez> I have not. Thanks for the suggestion. I assumed they had never been submitted
<marc|gonzalez> Because a few of the fixes I submitted that had also been proposed by these guys were never proposed
<marc|gonzalez> I suppose because mainlining is sometimes an exhausting ordeal
pevik has quit [Remote host closed the connection]
<Marijn[m]> Yep, IOMMU patches seem to never have made it through even though we have a lot of HW patches depending on it :)
pevik has joined #linux-msm
pevik has quit [Remote host closed the connection]
Daanct12 has quit [Quit: WeeChat 4.3.5]
<aka_[m]> marc|gonzalez: phh secured context banks ain't anything new
<aka_[m]> same thing on even older socs like 8976
<aka_[m]> and 8953, FW capped MMU500 impl
<aka_[m]> their smmu_v2 probably is even worse
<aka_[m]> oh thats fun, the lpass one is qsmmu LPASS_Q6_SMMU_QSMMUV2
<marc|gonzalez> aka_[m]: it does not give me comfort to know that others have stumbled upon that trap before :)
<aka_[m]> lumag: as you reviewed series before mind taking a look again at these?
<aka_[m]> i don't want do do some epic fail once again.
marc|gonzalez has quit [Quit: Leaving.]
f_ has quit [Ping timeout: 480 seconds]
f_ has joined #linux-msm
zstas has quit [Ping timeout: 480 seconds]
zstas has joined #linux-msm
zstas has quit [Ping timeout: 480 seconds]
<JIaxyga[m]> narmstrong: krzk :... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/ufBnDjMVawHRNDjAuzcpjNlJ>)
<JIaxyga[m]> I would send a patch myself, but idk where to fix it better
<aka_[m]> i would say fix bindings not driver
<aka_[m]> out of wonder
<aka_[m]> can two drivers utilize same compatible?
<aka_[m]> goodix,gt9916 is used in both i2c and spi driver
<JIaxyga[m]> aka_[m]: Yes, can
<bamse> aka_[m]: i2c busses only matches i2c drivers and spi busses only matches spi drivers
pespin has quit [Remote host closed the connection]
mripard is now known as Guest1699
mripard has joined #linux-msm
Guest1699 has quit [Ping timeout: 480 seconds]
<lumag> aka_[m], excuse me, I started looking at it, but then got sidetracked. LGTM, single nit: I'd prefer to see msm8916_qdsp6_add_ops() setting the q6afe_clk_ver to v1