06:27 <davidlt> rwmjones, could you backport gdb changed to f40, and build with a new NVR?

06:28 <davidlt> my script attempted to build gdb a few times now and failed (because I built one from Rawhide, and that remove riscv64 tag from Release)

06:55 <davidlt> rwmjones, openmpi, valgrind: http://fedora.riscv.rocks:3000/rpms/openmpi/commit/db9334ea6c4055f21891e3023572836414e59820

06:58 <davidlt> rwmjones, updated commit for f40: http://fedora.riscv.rocks:3000/rpms/openmpi/commit/4a40bc80437b074db92604eeeb3e52aec2355666

06:59 <davidlt> rwmjones, petsc is another valgrind: http://fedora.riscv.rocks:3000/rpms/petsc/commit/570c35d3326370f098c2d85348420617d98fbf84

07:00 <davidlt> rwmjones, mpich, valgrind: http://fedora.riscv.rocks:3000/rpms/mpich/commit/d05ce2cc1f512b7c772f64301349660fcdda2779

07:19 <davidlt> Oh, GDB 14.2 was just released this morning

10:32 <davidlt> rwmjones, jpegxl, http://fedora.riscv.rocks/koji/taskinfo?taskID=1631326

10:32 <davidlt> libjxl-0.8.2/lib/jxl/enc_xyb.cc:32:1: note: ‘std::atomic’ is defined in header ‘<atomic>’; this is probably fixable by adding ‘#include <atomic>’

10:33 <davidlt> rwmjones, there is a fix upstream, we need a backport: https://github.com/libjxl/libjxl/commit/22d12d74e7bc56b09cfb1973aa89ec8d714fa3fc

10:34 <davidlt> the fix is part of v0.9.0 and beyond

10:34 <davidlt> f40 is at 0.8.2

10:36 <davidlt> rwmjones, libtasn1, valgrind: http://fedora.riscv.rocks:3000/rpms/libtasn1/commit/f48a6e2306ebd91664fc793ef0373fd089fe29b8

11:51 <davidlt> and Pioneer nuked itself again

11:52 <davidlt> I don't really want to run it with maxjobs=1

11:54 <rwmjones> davidlt: hi, I'll do the gdb thing right now, will do the other packages tomorrow

11:55 <rwmjones> gdb build - https://koji.fedoraproject.org/koji/taskinfo?taskID=114384767

12:03 <davidlt> rwmjones, thanks

12:19 <hezhengyu> rwmjones, Sorry for the disturbance... Do you mean I should paste the link to this spec file(https://github.com/fedora-riscv/hdf5/blob/4766aac01021f846cb7d34cbc802ff55df713b99/hdf5.spec) link under the issue page(https://github.com/HDFGroup/hdf5/issues/4056)?

12:34 <rwmjones> hezhengyu: hi, yes, just put a link to the upstream bug in the spec file

12:34 <rwmjones> as a comment

12:35 <rwmjones> and then "git push --force ..." the change so that it updates the pull request

12:35 <rwmjones> this means in future people looking at the spec file can check if the upstream bug has been fixed (and so enable the tests again)

12:37 <hezhengyu> rwmjones: Many thanks for the details! I'll do it later

12:38 <rwmjones> no problem, once the spec file looks good I can merge the change

18:41 <davidlt> and pioneer nuked itself again

21:50 <rwmjones> https://www.aliexpress.com/item/1005006448285623.html

21:50 <rwmjones> only 512 MB of RAM unfortunately :-(

23:01 <conchuod> 74 quid to get something with 1.0 isn't a bad deal, 512 MiB aside.

23:02 <conchuod> At least for what I will do with it, should be sufficient for tyre kicking.