JRepin has quit [Remote host closed the connection]
JRepin has joined #riscv
naoki has joined #riscv
Stat_headcrabbed has joined #riscv
Andre_Z has joined #riscv
peepsalot has quit [Quit: Connection reset by peep]
prabhakalad has quit [Ping timeout: 252 seconds]
prabhakalad has joined #riscv
Armand|5800x has joined #riscv
cleger has joined #riscv
Stat_headcrabbed has quit [Remote host closed the connection]
alifib has joined #riscv
naoki has quit [Ping timeout: 260 seconds]
alexandre has joined #riscv
<alexandre>
@palmer: The first PR I sent was based on 6.15-rc3, in the meantime, we added fixes in 6.15-rc6 and the second PR depends on it. So to have "clean" merges, I think you should simply move your for-next head to 6.15-rc6, let me know if I can help.
seasharp_ has joined #riscv
seasharp has quit [Ping timeout: 248 seconds]
naoki has joined #riscv
naoki has quit [Client Quit]
armand_ has joined #riscv
Armand|5800x has quit [Read error: Connection reset by peer]
JRepin has quit [Quit: Konversation terminated!]
JRepin has joined #riscv
cousteau has joined #riscv
jacklsw has joined #riscv
armand_ is now known as Armand|5800x
Andre_Z has quit [Quit: Leaving.]
tgamblin_ is now known as tgamblin
inochi has quit [Quit: inochi]
<palmer>
OK, I'll try -- I'm not getting any conflicts, though, just a build error
<palmer>
alexandre: do you have a merge that works on your end? I think it's mcount related, so it could just be some tool chain weirdness
cleger has quit [Remote host closed the connection]
<alexandre>
I answered your mail on the ML: that's because you picked up the VDSO getrandom patch from the ML, but it requires fixes. I fixed it and sent it in the second PR, so just remove it.
Andre_Z has joined #riscv
netmonk has joined #riscv
guerby has quit [Remote host closed the connection]
guerby has joined #riscv
jacklsw has quit [Quit: Back to the real world]
Armand|5800x has quit [Quit: Leaving]
BootLayer has joined #riscv
peepsalot has joined #riscv
alperak has joined #riscv
JRepin has joined #riscv
JRepin has quit [Quit: Konversation terminated!]
alifib has quit [Quit: .]
JRepin has joined #riscv
JRepin has quit [Quit: Konversation terminated!]
JRepin has joined #riscv
JRepin has quit [Quit: Konversation terminated!]
JRepin has joined #riscv
<palmer>
alexandre: OK, thanks
<palmer>
I’ll look in a bit, I’ve got some work stuff to try and deal with...
matrixbrain has joined #riscv
<palmer>
alexandre: actually that’s pretty straight-forward, I’m going to revert the VDSO one I picked up because it’s been on -next for a bit and then merge the PR (it’s all testing, I’ll reply to the thread in a bit)
<drewfustini>
palmer: conchuod: there are more TH1520 peripheral drivers that I'd like to enable by default. Is it best to create patches for riscv defconfig or add selects to ARCH_THEAD in Kconfig.socs?
raym has joined #riscv
JRepin has quit [Quit: Konversation terminated!]
JRepin has joined #riscv
<drewfustini>
Hmm, I suppose select in ARCH_THEAD has the problem of making it hard to turn off.
<drewfustini>
What is the best way to get riscv defconfig patches merged? Should I create a thead defconfig tree and send PRs?
<alexandre>
What's the point of defconfig? To me it's just a minimal set of configs needed to boot a very simple kernel
<drewfustini>
I'd like there to be a default where the peripheral drivers are enabled for the SoC. There is already TH1520 pinctrl in defconfig. We now have mailbox, pm domain, and reset drivers.
<drewfustini>
Are you thinking that SoC drivers shouldn't be in defconfig?
<drewfustini>
I normally make defconfig to test so I prefer to have all those drivers enabled
<drewfustini>
Is ARCH_THEAD in Kconfig.socs better for that puprose?
<drewfustini>
Esmil: do you use defconfig as a starting pointing for what drivers to enable for a given SoC in Ubuntu? Or does it not matter because you create your own configs?
<palmer>
alexandre: I don’t think there’s any real concrete answer to “what’s the point of defconfig”, everyone says different things. I think having what’s required to boot common distros on common silicon is sane, though
<palmer>
drewfustini: distros should be picking their own drivers, but having something in defconfig can be a good signal to flip it on
<conchuod>
Definitely don't select drivers
BootLayer has quit [Quit: Leaving]
<conchuod>
Put stuff the platform reasonably needs in defconfig, as a module if possible
<drewfustini>
Thanks for the tips. I I guess I should just submit patches for riscv defconfig