dgilmore changed the topic of #fedora-riscv to: Fedora on RISC-V https://fedoraproject.org/wiki/Architectures/RISC-V || Logs: https://libera.irclog.whitequark.org/fedora-riscv || Alt Arch discussions are welcome in #fedora-alt-arches
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 264 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Remote host closed the connection]
Guoguo has joined #fedora-riscv
Guoguo has quit [Remote host closed the connection]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 260 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 260 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Remote host closed the connection]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 264 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 256 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 256 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 260 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 272 seconds]
djdelorie has quit [Ping timeout: 240 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 272 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 260 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Read error: Connection reset by peer]
Guoguo has joined #fedora-riscv
djdelorie has joined #fedora-riscv
davidlt has joined #fedora-riscv
<davidlt> rwmjones_, and a proper mass rebuilding is starting today
<davidlt> rwmjones_, I am thinking, as I rebuild stuff I could just send you stuff as I encounter .riscv64 packages
Guoguo has quit [Remote host closed the connection]
Guoguo has joined #fedora-riscv
Guoguo has quit [Remote host closed the connection]
<rwmjones_> davidlt: yes please
<davidlt> also need NVR for gdb
<davidlt> because this doesn't help me: https://src.fedoraproject.org/rpms/gdb/commits/rawhide
<davidlt> I still have to deal with dist-git overlay, which means this become even more work
<davidlt> This will become annoying relatively quickly, and using Koji to figure out which packages are upstreamed will stop working
<rwmjones_> davidlt: one sec, will do gdb bump first
<davidlt> technically I need this in F40, but I can take it from Rawhide if it has minimal differences
<davidlt> my script will not pick it from Rawhide, and if I pick NVR from Rawhide (i.e. without .riscv64) the script will attempt to override it with whatever is in F40
<davidlt> typically I wait for upstream Koji build to finish, but I must rebuild GDB now (I landed a new Boost version)
<rwmjones_> doing gcc now ..
<davidlt> the next one most likely is elfutils, testing now.
<davidlt> I am doing a clean build + patched build as this will allow to write better PR messages for dist-git
<rwmjones_> davidlt: there's no change to elfutils (except adding a commit message and a blank line)
<davidlt> Not yet
<davidlt> wait, did he added a patch already
* davidlt checking..
<davidlt> I will chat with Mark
<conchuod> davidlt: do you have a pioneer?
<conchuod> If so, did you replace the puny nvme on it?
<davidlt> conchuod, I have it, and yes, I replaced the NVMe.
<conchuod> I've been trying to put a new kernel on that thing, building in situ. Horrible idea.
<davidlt> conchuod, I decided not to touch anything until upstream catches up a bit :)
<davidlt> There are some instructions how to build a new disk image and boot with SD card.
<davidlt> rwmjones_, Mark said 0.191 elfutils (March 1st). He will try to fix the issue before this release.
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 264 seconds]
<rwmjones_> fyi
<rwmjones_> Planned Outage - koji upgrade - 2024-02-21 21:00 UTC
<davidlt> saw the email yesterday
zsun has joined #fedora-riscv
Guoguo has joined #fedora-riscv
zsun has quit [Quit: Leaving.]
zsun has joined #fedora-riscv
zsun has quit [Quit: Leaving.]
<rwmjones_> davidlt: what BR'd mold? I've fixed mold now
<davidlt> rwmjones_, let's check
<davidlt> rwmjones_, ceph
<rwmjones_> the easy ones heh
<rwmjones_> so it looks like ceph will only use mold on x86_64 aarch64
<rwmjones_> it's unclear why mold is better (faster linking?)
<rwmjones_> there are some other differences in the spec so I'll make a PR for ceph anyway
<davidlt> it is faster linking IIRC, but also mold folks are involved in riscv psABI
<rwmjones_> looks as if ceph takes about 30 hours to build
<rwmjones_> I will try a build with mold enabled to see if it works, but we won't find out til tomorrow or Friday probably
* rwmjones_ uses a 'screen' session for this build
<rwmjones_> ok time to check on fuwei's list again
<davidlt> This doesn't build since F35, removed in F37, and later back
<davidlt> F38-F41, no point?
<rwmjones_> davidlt: it does seem to be giving off a deathly odour
<rwmjones_> checks bugzilla ...
zsun has joined #fedora-riscv
zsun has quit [Quit: Leaving.]
Guoguo has quit [Remote host closed the connection]
cyberpear has joined #fedora-riscv
amusil has joined #fedora-riscv
<rwmjones_> davidlt: amusil is trying to get the fedora.riscv.rocks images working on qemu
<amusil> Hi, I'm trying to get one of the latest images from http://fedora.riscv.rocks/koji/tasks?state=closed&view=flat&method=createAppliance&order=-id workign on qemu
<amusil> *working
<amusil> However it seems there isn't boot.scr or extlinux config
<rwmjones_> I can't recall if the latest images require uefi or not
<amusil> It end on this:
<amusil> Device 0: 1af4 VirtIO Block Device
<amusil>             Type: Hard Disk
<amusil>             Capacity: 15360.0 MB = 15.0 GB (31457280 x 512)
<amusil> ... is now current device
<amusil> Scanning virtio 0:1...
<amusil> Failed to load '/'
<amusil> Loading Boot0000 'virtio 0' failed
<amusil> EFI boot manager: Cannot load any image
<rwmjones_> what's the qemu command line and/or libvirt XML?
<rwmjones_> looks like virtio-blk
<rwmjones_> (I'm in a meeting at the moment)
<amusil> u-boot can see it just fine
<amusil> => virtio info
<amusil> Device 0: 1af4 VirtIO Block Device
<amusil>             Type: Hard Disk
<amusil>             Capacity: 15360.0 MB = 15.0 GB (31457280 x 512)
<rwmjones_> it's been over a year since I tried installing on qemu, and also I think we moved the kernel to require UEFI fairly recently; davidlt will know when he's around
Guoguo has joined #fedora-riscv
Guoguo has quit [Remote host closed the connection]
amusil has quit [Quit: Client closed]
<davidlt> We don't really use the images from Koji directly (those are not Pungi images)
<davidlt> The kernels in up to (incl.) F38 are extlinux, non-UEFI stuff.
<davidlt> Anything above is basically requires UEFI, but I never finished the transition.
<davidlt> I does direct-boot with EDKII firmware on QEMU, that I tested.
<davidlt> It was part of Pungi effort, which was not finished.
<davidlt> oh, amusil quit
<davidlt> failed test information is here: http://fedora.riscv.rocks/koji/taskinfo?taskID=625333
<davidlt> it's probably the only patched Perl package we have
<rwmjones_> davidlt: sure
<davidlt> otherwise all Perl packages were submitted
<davidlt> according to the script 9 packages are already in, but those seem to fail in upstream Fedora
<davidlt> huh, perl-Tk failed 3 times in F40 on upstream Koji
<rwmjones_> damn CPAN is down ...
<davidlt> oh yeah, almost forgot Koji is going offline in a few hours too
<rwmjones_> ok there's a github, not touched for 11 years
<davidlt> oh yeah, it's some ancient stuff IIRC
<rwmjones_> I asked amusil to rejoin
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 260 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 256 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 240 seconds]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 260 seconds]
davidlt has quit [Ping timeout: 268 seconds]
cyberpear has quit [Quit: Connection closed for inactivity]
Guoguo has joined #fedora-riscv
Guoguo has quit [Ping timeout: 255 seconds]