• Joined on 2019-03-28
sb10q commented on pull request M-Labs/nac3#477 2024-07-25 21:11:21 +08:00
Fix codegen issues for 32-bit targets

This isn't only about size_t.

sb10q commented on pull request M-Labs/nac3#477 2024-07-25 21:10:15 +08:00
Fix codegen issues for 32-bit targets

Is this correct?

sb10q commented on pull request M-Labs/nac3#477 2024-07-25 21:09:42 +08:00
Fix codegen issues for 32-bit targets

Won't this break the automatic tests? They don't have the 686 packages unless the flake is updated.

sb10q commented on pull request M-Labs/nac3#477 2024-07-25 21:07:51 +08:00
Fix codegen issues for 32-bit targets

Are all those three needed? My code was just an unrefined proof of concept.

sb10q commented on pull request M-Labs/nac3#478 2024-07-25 21:04:33 +08:00
add nalgebra::linalg methods

Removing LLI should (1) be done more thoroughly, we don't want to drag dead code along (2) be a separate PR.

sb10q commented on pull request M-Labs/nac3#478 2024-07-25 21:03:36 +08:00
add nalgebra::linalg methods

Why the globbing?

sb10q commented on pull request M-Labs/nac3#478 2024-07-25 21:02:32 +08:00
add nalgebra::linalg methods

Does not belong in this PR

sb10q commented on pull request M-Labs/nac3#478 2024-07-25 21:01:23 +08:00
add nalgebra::linalg methods

We don't support exceptions in the host demo and we don't plan to. Just call abort(), assert() and similar. Please keep everything as simple as possible.

sb10q commented on pull request M-Labs/nac3#478 2024-07-25 21:00:21 +08:00
add nalgebra::linalg methods

code formatting

sb10q commented on pull request M-Labs/nac3#478 2024-07-25 20:58:27 +08:00
add nalgebra::linalg methods

What crate is used is a firmware implementation detail. Remove this comment and remove any code that goes along that line.

sb10q commented on pull request M-Labs/nac3#478 2024-07-25 20:57:30 +08:00
add nalgebra::linalg methods

Will this just crash the compiler if the user tries to do it on integer matrices?

sb10q commented on pull request M-Labs/nac3#478 2024-07-25 20:56:23 +08:00
add nalgebra::linalg methods

And can it be moved to demo? Does it need to be at the top level cargo workspace?

sb10q commented on pull request M-Labs/nac3#478 2024-07-25 20:55:54 +08:00
add nalgebra::linalg methods

Just call it linalg or similar, this name is too long.

sb10q commented on pull request M-Labs/nac3#478 2024-07-25 20:54:27 +08:00
add nalgebra::linalg methods

The resulting matrices are slightly different

What does that means exactly? And why?

sb10q commented on pull request M-Labs/nac3#478 2024-07-25 20:53:49 +08:00
add nalgebra::linalg methods

the current implementation is limited to 1D case

Why?

sb10q commented on pull request M-Labs/artiq-extrapkg#19 2024-07-25 20:52:46 +08:00
Remove scipy dependency for MSYS2 ARTIQ

Needs backport and testing on release-8.

sb10q commented on pull request M-Labs/artiq-extrapkg#19 2024-07-25 08:09:46 +08:00
Remove scipy dependency for MSYS2 ARTIQ

What tests were done?

sb10q opened issue sinara-hw/kirdy#49 2024-07-23 12:06:08 +08:00
generate panel outputs from flake
sb10q opened issue sinara-hw/kirdy#48 2024-07-23 12:05:53 +08:00
merge flake files
sb10q commented on pull request M-Labs/nac3#478 2024-07-23 09:05:21 +08:00
add nalgebra::linalg methods

Again, move this out of nac3core. This is only used by nac3standalone/demo and this is where it should go. I also doubt that clang xxy.so works to link an executable against a DSO. Did you…