15:02 <RobTaylor[m]> does anyone know how to stop UnusedElaboratable messages when a test fails? # amaranth: UnusedElaboratable=no at the top of the test file doesn't seem to do the job

15:19 <whitequark[cis]> probably a bug

15:21 <RobTaylor[m]> i should get my head around how that mechanism works. its still a bit of a mystery to me =)

15:22 <whitequark[cis]> grabs a backtrace, extracts the file, and checks the first few lines

15:23 <RobTaylor[m]> Catherine: btw for my own purposes, i updated amaranth-soc#58 to be current against main . Should i submit a pull request to amaranth-soc?

15:24 <whitequark[cis]> <RobTaylor[m]> does anyone know how to stop UnusedElaboratable messages when a test fails? # amaranth: UnusedElaboratable=no at the top of the test file doesn't seem to do the job

15:24 <whitequark[cis]> you might need to add src_loc_at where applicable, i don't recall if wiring.Component has it as a parameter but if it doesn't it probably should

15:25 <whitequark[cis]> RobTaylor[m]: sure

15:25 <RobTaylor[m]> whitequark[cis]: thanks, will do

15:26 <RobTaylor[m]> whitequark[cis]: i'll dig in and see whats actually going on. might be that it needs to step back though the trace

18:03 <_whitenotifier-8> [amaranth] robtaylor opened pull request #1621: Modify MustUse to iterate the stack frame and collate options from all frames. - https://github.com/amaranth-lang/amaranth/pull/1621

18:04 <RobTaylor[m]> Catherine: for discussion ^ :)

18:06 <_whitenotifier-8> [amaranth] codecov[bot] commented on pull request #1621: Modify MustUse to iterate the stack frame and collate options from all frames. - https://github.com/amaranth-lang/amaranth/pull/1621#issuecomment-3137351980

18:19 <_whitenotifier-8> [amaranth-soc] robtaylor opened pull request #104: Add annotations for memory maps, wishbone and CSR primitives [updated] - https://github.com/amaranth-lang/amaranth-soc/pull/104

18:23 <_whitenotifier-8> [amaranth-soc] robtaylor reviewed pull request #58 commit - https://github.com/amaranth-lang/amaranth-soc/pull/58#discussion_r2243544453

18:50 <zyp[m]> <RobTaylor[m]> Catherine: btw for my own purposes, i updated amaranth-soc#58 to be current against main . Should i submit a pull request to amaranth-soc?

18:50 <zyp[m]> I have some proposed changes to that, but I'll need to go through what I did and compare

18:50 <RobTaylor[m]> zyp[m]: nice! Happy to have a look at your branch and integrate

19:01 <zyp[m]> it's not a branch, I built against latest main and monkeypatched the changes instead: https://github.com/zyp/katsuo-bridge/tree/main/katsuo/bridge/_monkeypatch

19:02 <vup> is it intentional, that `data.Const` returns a boolean when comparing with another `data.Const`, while `_ast.Const` returns a `Operator("==", ...)`?

19:03 <vup> Furthermore, is there any reason, `data.Const` is not hashable?

19:07 <zyp[m]> the main change I want is the ability to attach additional annotations to a memory map, and my implementation is still half baked because we need to sort out how to do that before a memory map is frozen, ref: https://github.com/zyp/katsuo-bridge/blob/main/katsuo/bridge/_monkeypatch/amaranth_soc_memory.py#L20

20:30 <_whitenotifier-8> [amaranth] rroohhh opened issue #1622: Spaces in names generates invalid `RTLIL` - https://github.com/amaranth-lang/amaranth/issues/1622

21:51 <RobTaylor[m]> <zyp[m]> it's not a branch, I built against latest main and monkeypatched the changes instead: https://github.com/zyp/katsuo-bridge/tree/main/katsuo/bridge/_monkeypatch

21:51 <RobTaylor[m]> i'll have a look asap :)

21:52 <RobTaylor[m]> <zyp[m]> the main change I want is the ability to attach additional annotations to a memory map, and my implementation is still half baked because we need to sort out how to do that before a memory map is frozen, ref: https://github.com/zyp/katsuo-bridge/blob/main/katsuo/bridge/_monkeypatch/...

21:52 <RobTaylor[m]> yeah. I'm unsure. It feels to me like there are valid reasons to attach annotations to the memory map after it has been frozen.

21:53 <RobTaylor[m]> e.g. configuration for drivers, or things like that

22:02 <zyp[m]> I don't think partially frozen makes sense, I figure we either want to freeze it later, add a way to add annotations earlier or maybe bypass the whole issue by wrapping the frozen object in another object that adds the additional annotations

22:05 <RobTaylor[m]> yep, i'm thinking more along the lines of a separate object for user annotations. Could inject its data so that as_json picks it up

22:06 <RobTaylor[m]> i haven't put much thought in yet tho 😅

22:08 <RobTaylor[m]> RobTaylor[m]: or indeed, could just be completely seperate *shrug*

22:10 <zyp[m]> while we're at it, I published https://github.com/zyp/katsuo-ila as well

22:10 <zyp[m]> it's pretty rudimentary and probably not really usable yet, but it's an example of stuff built on katsuo-bridge

22:11 <RobTaylor[m]> zyp: on a seperate line, here's what i'm doing for my annotations using pydantic: https://github.com/ChipFlow/chipflow-lib/blob/gen-sim-main/chipflow_lib/platforms/_annotate.py

22:11 <RobTaylor[m]> makes annotating much lighter - e.g. https://github.com/ChipFlow/chipflow-lib/blob/gen-sim-main/chipflow_lib/platforms/_signatures.py#L112

22:14 <zyp[m]> ah, that's nice, I'll have to look into that

22:14 <RobTaylor[m]> zyp[m]: i'll have to have a good look, looks interesting!

22:15 <RobTaylor[m]> zyp[m]: thought you might like it :) Let me know how you get on