00:18:53 <sommerfeld> toasterson: they've definitely done some network scalability work.   
19:24:16 <gitomat> [illumos-gate] 18375 bhyve: viona feature_mask option cannot be parsed -- Andy Fiddaman <illumos⊙fn>
19:24:16 <gitomat> [illumos-gate] 18420 vioif relies on the device's initial receive filtering mode -- Andy Fiddaman <illumos⊙fn>
19:38:04 <gitomat> [illumos-gate] 18451 loader mb2 indexed color setup is buggy -- Toomas Soome <tsoome⊙mc>
21:02:57 <gitomat> [illumos-gate] 18010 loader: add panic_action() for i386 loader -- Toomas Soome <tsoome⊙mc>
22:14:57 <gitomat> [illumos-gate] 18448 amdzen reads wrong number of bits to determine node/fabric IDs for DFv4D2 -- Luqman Aden <luqman⊙oc>
22:47:19 <jbk> hrm shoot..we don't have a popcount function in the kernel, do we?
23:06:25 <rmustacc> jbk: We do.
23:07:51 <rmustacc> See stdc_count_ones(9F).
23:08:06 <rmustacc> All of the C23 stdbit.h stuff is available in kernel.
23:29:14 <davicito> https://src.illumos.org/source/ bad gateway
23:32:50 <alanc> hmm, I suppose using gcc's _builtin_popcount in the kernel is risky if you don't know if it's going to use an implementation that requires FP registers or something else silly and not allowed in kernel mode