01:42 <zignig> nice, hey all.

01:43 <zignig> Catherine , .. et al. Just managed to boot a rv32i control plane from flash on my tinybx.

01:43 <zignig> so cool.

01:43 <zignig> pure Amaranth; with source from others... ofcourse.

01:50 <zignig> ... anyway. @tpwrules I ported your icepanel spi interface and rewrote the driver from boneless asm to rust.

01:50 <zignig> work on my TODO list continuues.

08:47 <nates93[m]> https://amaranth-lang.org/play/#eyJhdiI6InYwLjQuNSIsInMiOiJmcm9tIGFtYXJhbnRoIGltcG9ydCAqXG5pbXBvcnQgYW1hcmFudGgubGliLmRhdGEgYXMgZGF0YVxuaW1wb3J0IGFtYXJhbnRoX3BsYXlncm91bmRcblxuY2xhc3MgSUVFRTc1NFNpbmdsZShkYXRhLlN0cnVjdCk6XG4gICAgZnJhY3Rpb246IDIzXG4gICAgZXhwb25lbnQ6ICA4ID0gMHg3ZlxuICAgIHNpZ246ICAgICAgMVxuXG5jbGFzcyBTb21lU3RydWN0KGRhdGEuU3RydWN0KTpcbiAgICBvcHRpb25zOiBBcnJheVtJRUVFNzU0U2luZ2xlLCAyXVxuXG5zb21lX3N0cnVjdCA9IFNpZ25hbChTb21

08:47 <nates93[m]> lU3RydWN0LCByZXNldD17XG4gICAgXCJvcHRpb25zXCI6IFtcbiAgICAgICAge1wiZnJhY3Rpb25cIjogMHgxRkZGRjcsIFwiZXhwb25lbnRcIjogMHg3RSwgXCJzaWduXCI6IDB9LFxuICAgICAgICB7XCJmcmFjdGlvblwiOiAweDFGRkZGNywgXCJleHBvbmVudFwiOiAweDdFLCBcInNpZ25cIjogMX1cbiAgICBdXG59KVxuXG4ifQ==

08:47 <whitequark[cis]> nates93: what's that on line 7?

08:48 <whitequark[cis]> I've actually just added a note to the docs about this exact mixup, in https://amaranth-lang.org/docs/amaranth/latest/guide.html#arrays

08:48 <whitequark[cis]> > Arrays, amaranth.hdl.Array, are distinct from and serve a different function than amaranth.lib.data.ArrayLayout.

08:48 <whitequark[cis]> you want data.ArrayLayout(IEEE754Single, 2)

09:04 <nates93[m]> Noted! To be honest, I glazed over that bit. It came from an LLM. I appreciate the note in the docs! Actually, while trying to create a new MVRE I figured out my issue. It seems when using a nested data.Union from within a data.Struct, I cannot just initialize it with 0, I must specify `{"member_that_is_a_union" : {"subfield_of_union" : 0}}`

09:04 <nates93[m]> * Noted! To be honest, I glazed over that bit. It came from an LLM. I appreciate the note in the docs! Actually, while trying to create a new MVRE I figured out my issue. It seems when using a nested `data.Union` from within a `data.Struct`, I cannot just initialize it with 0, I must set one of the members of the union: `{"member_that_is_a_union" : {"subfield_of_union" : 0}}`

09:08 <nates93[m]> https://amaranth-lang.org/play/#eyJhdiI6InYwLjQuNSIsInMiOiJmcm9tIGFtYXJhbnRoIGltcG9ydCAqXG5pbXBvcnQgYW1hcmFudGgubGliLmRhdGEgYXMgZGF0YVxuaW1wb3J0IGFtYXJhbnRoX3BsYXlncm91bmRcblxuXG5jbGFzcyBBKGRhdGEuVW5pb24pOlxuICAgIGFkZHIgOiAzMlxuXG5jbGFzcyBCKGRhdGEuU3RydWN0KTpcbiAgICBiICAgICAgICA6IEFcblxuY2xhc3MgQyhkYXRhLlN0cnVjdCk6XG4gICAgYyAgICAgIDogQlxuXG5cbmMgPSBTaWduYWwoQywgXG4gICAgcmVzZXQ9e1xuICAgICAgICBcImNcIjoge1xuICAgICAgICAgICAgXCJiXCI6IHt

09:08 <nates93[m]> cImFkZHJcIiA6IDB9LFxuICAgICAgICAgICAgI1wiYlwiOiAwLCAjIHRoaXMgd2lsbCBmYWlsIFxuICAgICAgICB9LFxuICAgIH0pXG5cbiJ9

09:30 <whitequark[cis]> oh, well, don't take advice from an LLM

09:30 <whitequark[cis]> or it'll continue to cause you issues like this

13:34 <whitequark[cis]> in principle we could extend the existing ArrayLayout to multiple dimensions; you could submit an RFC for that

13:37 <sohaiber[m]> alright