<sorear>
heat: it's part of the ratified sbi spec so it has to still be there, yes. conceptually it's just providing a s-mode csr name for the same time comparator used by the sbi calls, you don't get two independent comparators
<jrtc27>
sorear: I hadn't, and I don't know if I have the energy today to reply
<jrtc27>
but it sure would be nice if someone could say "do we really want to irreversibly bake a linuxism into *the* ifunc ABI"
<jrtc27>
non-linux will never have the same interface, and having two would be very sad
<sorear>
no "rev", no complex permuations, no ternaries
<davidlt>
cousteau, from what I understand finally we would know what misa.B means, which would be Zba + Zbb + Zbs.
Kyuvi has quit [Ping timeout: 250 seconds]
<davidlt>
and A would be Zaamo + Zalrsc
Andre_Z has quit [Quit: Leaving.]
<dh`>
muurkha: that's what I thought, but a couple apparently exist now
<sorear>
is Zbs actually useful?
<sorear>
how would i know which 32 bit softcores with mmus dh`wants to use
Stat_headcrabed has joined #riscv
Kyuvi has joined #riscv
Andre_Z has joined #riscv
Andre_Z has quit [Client Quit]
Stat_headcrabed has quit [Quit: Stat_headcrabed]
<sorear>
is it just me or do the memory model operational semantics forbid fetching instructions after an unconditional jump, thus effectively forbidding a line-based instruction cache or instruction prefetching?
Stat_headcrabed has joined #riscv
Kyuvi has quit [Quit: Client closed]
Andre_Z has joined #riscv
EchelonX has joined #riscv
prabhakar has joined #riscv
prabhakarlad has joined #riscv
ntwk has quit [Read error: Connection reset by peer]
meritamen has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
BootLayer has joined #riscv
Stat_headcrabed has quit [Quit: Stat_headcrabed]
davidlt has quit [Remote host closed the connection]
davidlt has joined #riscv
hightower2 has joined #riscv
alexghiti has quit [Ping timeout: 256 seconds]
epony has joined #riscv
psydroid has joined #riscv
SpaceCoaster has quit [Ping timeout: 264 seconds]
JanC has quit [Remote host closed the connection]
jmdaemon has joined #riscv
JanC has joined #riscv
jmdaemon has quit [Ping timeout: 268 seconds]
SpaceCoaster has joined #riscv
<dh`>
sorear: "use" in this case (as often) implies a certain degree of maturity/stability, iow, you can use the widget without having to spend time debugging it or coping with flakinesses
<dh`>
re the semantics, happy to look at it, remind me this weekend if you want
<dh`>
(but can't really before then)
Kyuvi has joined #riscv
jmdaemon has joined #riscv
frkazoid333 has quit [Ping timeout: 260 seconds]
Andre_Z has quit [Quit: Leaving.]
frkazoid333 has joined #riscv
frkazoid333 has quit [Remote host closed the connection]
n_crm_ has joined #riscv
<drewfustini>
conchuod:
<drewfustini>
oops
<drewfustini>
conchuod: i have a clk driver for th1520 from jisheng
<drewfustini>
i'll post it today
n_crm has quit [Ping timeout: 252 seconds]
<drewfustini>
there were some issues I was trying to work out with the emmc clock but it's gone on for too long
<drewfustini>
so i will post what I currently have
<conchuod>
drewfustini: 👍
<drewfustini>
I just replied
<drewfustini>
basically, I found it is not working for the emmc. The documentation is a bit confusing on that point and why the T-Head SDK just uses a fixed clock even though there is a clock driver in the SDK. I have not gotten any responses to my initial inquiries into t-head but I need to followup
<drewfustini>
Anyways, no reason to not share what I currently have on the list as RFC
<drewfustini>
it is timely for other people's work like Emil
<drewfustini>
I'm hoping clk and ethernet might be able to make it into 6.9 so starting a thread with current clk driver is probably a good idea
Kyuvi has quit [Quit: Client closed]
BootLayer has quit [Quit: Leaving]
sevan has quit [Changing host]
sevan has joined #riscv
ezulian has quit [Ping timeout: 276 seconds]
Narrat has joined #riscv
cousteau has quit [Read error: Connection reset by peer]
cousteau has joined #riscv
cousteau_ has joined #riscv
cousteau has quit [Ping timeout: 268 seconds]
Kyuvi has joined #riscv
vagrantc has joined #riscv
ntwk has joined #riscv
JanC_ has joined #riscv
JanC has quit [Killed (zirconium.libera.chat (Nickname regained by services))]
JanC_ is now known as JanC
Kyuvi has quit [Ping timeout: 250 seconds]
<drewfustini>
Esmil: do you know if the Ubuntu kernel supports PCIe for the VF2?
alexghiti has joined #riscv
davidlt has quit [Ping timeout: 246 seconds]
jmdaemon has quit [Ping timeout: 264 seconds]
jmdaemon has joined #riscv
jmdaemon has quit [Ping timeout: 246 seconds]
Kyuvi has joined #riscv
<Tenkawa>
drewfustini: give me a few and I can tell you... I need to rebuild my box and I can use that Ubuntu image as the base. I'm going to be updating the kernel later but for now I can use it to test
<Tenkawa>
Let me throw the Ubuntu server one on a sd and get booted up.
cousteau_ has quit [Remote host closed the connection]
cousteau has joined #riscv
jfsimon1981_c has quit [Remote host closed the connection]
<Tenkawa>
drewfustini: just found it in the Ubuntu release notes:
<Tenkawa>
USB and PCIe are not supported by the 6.2 kernel. and The 1.2A and 1.2B board revisions do not work with the 6.5 kernel.
___nick___ has joined #riscv
<Tenkawa>
so that will be the two most relevant areas for you (checking what version the current image is now)
<Tenkawa>
drwxr-xr-x 5 root root 4096 Oct 10 00:13 6.5.0-9-generic
<Tenkawa>
I'm thinking that looks hopeful.. will know shortly
cousteau has quit [Read error: Connection reset by peer]
cousteau has joined #riscv
cousteau has quit [Ping timeout: 256 seconds]
<xypron>
drewfustini: Yes, PCIe is enable on the VisionFive 2 since the Ubuntu 23.10 release. NVMe is working fine, extenal GPU not yet.
<Esmil>
drewfustini: yes, i cherry picked the PCI patches a bit early for our Ubuntu risc-v kernel
Kyuvi has quit [Ping timeout: 250 seconds]
<Esmil>
drewfustini: my jh7110 branch is the patches that I plan on merging into the ubuntu risc-v kernel
<drewfustini>
thanks, so it may have more functionality than the current ubuntu risc-v kernel?
Narrat has quit [Quit: They say a little knowledge is a dangerous thing, but it's not one half so bad as a lot of ignorance.]
<Esmil>
yes, the pci patches are updated to the latest version and so are the pwm patches. i'm still debating if Keith's display system patches make sense to add since they don't really give a working console on HDMI for me
<Esmil>
..but they're in the jh7110 branch
<Esmil>
i'm also hoping Jisheng will post an updated version of the MilkV Mars device tree