15:20 <mtrbot-ml> [mattermost] <sb10q> @astro it seems that the only thing that prevents building the zc706 firmware with regular rustc are "--release-channel=stable" passed to the rustc configure, and bypassing rust-lld...

15:22 <mtrbot-ml> [mattermost] <sb10q> with release-channel=stable, xargo just prints "The sysroot can't be built for the Stable channel. Switch to nightly."

15:22 <mtrbot-ml> [mattermost] <sb10q> this stable/nightly stuff is one of the most frustrating parts of rust imo

15:24 <mtrbot-ml> [mattermost] <sb10q> in a number of cases you can just STFU it by abusing the RUSTC_BOOTSTRAP environment variable, but not here

15:27 <mtrbot-ml> [mattermost] <sb10q> you actually have to recompile rustc (but you can still use the release 1.41, just tell it --release-channel=nightly...)

17:27 <mtrbot-ml> [mattermost] <hartytp> @rjo is the Fastino "wide" interface complete now?

17:27 <mtrbot-ml> [mattermost] <hartytp> if so ace!

17:27 <mtrbot-ml> [mattermost] <hartytp> would be interested to have a play with it... (cc @pathfinder49 see recent ARTIQ commits)

19:49 <rjo> yes. play with it.

19:53 <rjo> It did compile and work fine for a single Fastino on Kasli and 32-channel bus. I didn't try any DMA things with it. It seemed resource hungry. If you play with it, look at resource utilization. Check how many Fastino's you can get onto Kasli, compensating with SED depth once the defaults don't work.

21:34 <mtrbot-ml> [mattermost] <hartytp> @rjo ace, thanks!

21:35 <mtrbot-ml> [mattermost] <hartytp> you mean decreasing the SED depth to compensate for the extra resources needed for the wider RTIO bus?

21:38 <mtrbot-ml> [mattermost] <hartytp> and idea what dominates the increased resource utilization? Is it just the wider RTIO fifos? If so, how does that scale with the number of Fastinos/other EEMs (presumably it doesn't change much since the FIFOs are shared?)

21:39 <mtrbot-ml> [mattermost] <hartytp> @pathfinder49 you should now be able to use the fast interface to finish your testing