sb10q

sb10q merged pull request M-Labs/artiq-zynq#308

2024-07-22 16:53:07 +08:00

kernel/api: add nalgebra::linalg methods

sb10q commented on pull request M-Labs/artiq-zynq#308

2024-07-22 11:55:49 +08:00

kernel/api: add nalgebra::linalg methods

OK.

sb10q commented on pull request M-Labs/artiq-zynq#308

2024-07-22 11:45:06 +08:00

kernel/api: add nalgebra::linalg methods

And isn't this check expensive?

sb10q commented on pull request M-Labs/artiq-zynq#308

2024-07-22 11:44:48 +08:00

kernel/api: add nalgebra::linalg methods

I don't think crashing the firmware in response to incorrect user input is a sensible thing to do.

sb10q pushed to master at M-Labs/artiq-zynq

2024-07-22 11:13:48 +08:00

bab938c563 add nalgebra dependency

sb10q merged pull request M-Labs/artiq-zynq#307

2024-07-22 11:13:46 +08:00

kernel/api: add nalgebra::linalg try_invert_to api

sb10q commented on pull request M-Labs/nac3#477

2024-07-22 10:43:54 +08:00

Fix codegen issues for 32-bit targets

Sure, but with this PR introduces two cslices, one with uint32 and one with usize. Looks messy if not incorrect.

sb10q commented on pull request M-Labs/nac3#477

2024-07-22 08:34:46 +08:00

Fix codegen issues for 32-bit targets

Quite messy. And also this isn't what clang -m32 does - the latter selects 32-bit x86 ISA, so there is no need to name such flags in the same way. The previous CLI with -s was better.

sb10q commented on pull request M-Labs/nac3#477

2024-07-22 08:29:02 +08:00

Fix codegen issues for 32-bit targets

Why the x?

sb10q commented on pull request M-Labs/nac3#477

2024-07-22 08:27:51 +08:00

Fix codegen issues for 32-bit targets

Uh, why?

sb10q commented on pull request M-Labs/nac3#468

2024-07-21 19:21:41 +08:00

standalone: -s to set codegen size_t & ./check_demo.sh to test both 32-bits and 64-bits

No, the default should be host machine dependent. GCC doesn't force you to set one.

sb10q commented on pull request M-Labs/nac3#468

2024-07-21 19:20:55 +08:00

standalone: -s to set codegen size_t & ./check_demo.sh to test both 32-bits and 64-bits

Obviously

sb10q commented on pull request M-Labs/nac3#468

2024-07-21 19:20:34 +08:00

standalone: -s to set codegen size_t & ./check_demo.sh to test both 32-bits and 64-bits

reference output

sb10q commented on pull request M-Labs/nac3#468

2024-07-21 19:19:20 +08:00

standalone: -s to set codegen size_t & ./check_demo.sh to test both 32-bits and 64-bits

These two comments are obvious, remove.

sb10q commented on issue M-Labs/nac3#466

2024-07-21 13:32:41 +08:00

incorrect ndarray indexing crashes compiler

This actually happens because of the incorrect indexing in print_rpc.

sb10q opened issue M-Labs/nac3#467

2024-07-21 13:31:12 +08:00

wrong error message on matrix multiplication type error

sb10q commented on issue M-Labs/nac3#466

2024-07-21 13:29:57 +08:00

incorrect ndarray indexing crashes compiler

Same bug with y = 2.0*x.

sb10q opened issue M-Labs/nac3#466

2024-07-21 13:28:15 +08:00

2D ndarray addition crashes compiler

sb10q opened issue M-Labs/nac3#465

2024-07-21 13:09:50 +08:00

print_rpc of ndarray freezes

sb10q opened issue M-Labs/nac3#464

2024-07-21 13:07:44 +08:00

Cannot type-annotate ndarray RPC parameter in ARTIQ