• Joined on 2020-07-18
dpn created pull request M-Labs/artiq#1750 2026-02-02 02:23:29 +08:00
artiq_flash: Fix incorrect re.sub() call
dpn created branch flash-regex-fix in dpn/artiq 2026-02-02 02:23:16 +08:00
dpn pushed to flash-regex-fix at dpn/artiq 2026-02-02 02:23:16 +08:00
ec8ae26d51 artiq_flash: Fix incorrect re.sub() call
dpn deleted branch schema-disallow-unknown from dpn/artiq 2026-02-01 19:52:01 +08:00
dpn created pull request M-Labs/artiq#1748 2026-02-01 06:48:43 +08:00
jsondesc: Disallow unknown properties
dpn created branch schema-disallow-unknown in dpn/artiq 2026-02-01 06:47:01 +08:00
dpn pushed to schema-disallow-unknown at dpn/artiq 2026-02-01 06:47:01 +08:00
8a2d7dab94 jsondesc: Disallow unknown properties
dpn created pull request M-Labs/artiq#1746 2026-01-30 23:51:48 +08:00
firmware/ksupport: Add missing __floatundidf export
dpn created branch floatundidf in dpn/artiq 2026-01-30 23:49:04 +08:00
dpn pushed to floatundidf at dpn/artiq 2026-01-30 23:49:04 +08:00
fc0bb62f32 firmware/ksupport: Add missing __floatundidf export
dpn created repository dpn/artiq 2026-01-30 23:28:44 +08:00
dpn commented on issue M-Labs/artiq#1725 2026-01-18 19:06:29 +08:00
llvmlite 0.46 breaks optimization

https://github.com/numba/llvmlite/pull/1321#issuecomment-3384040409

As upstream thinks that "adding backward compatibility for this might not be ideal" (?!), we'll have to work around this…

dpn commented on issue M-Labs/artiq#1725 2026-01-18 19:06:29 +08:00
llvmlite 0.46 breaks optimization

This is not related to the new pass manager, but https://github.com/numba/llvmlite/issues/1306.

dpn commented on issue M-Labs/artiq#1725 2026-01-18 19:06:29 +08:00
llvmlite 0.46 breaks optimization

Yep – I understand that the Numba team may be primarily interested in maintaining llvmlite for Numba's sake, and they do not appear to use size_level. Still, changing the API in a mutually…

dpn commented on issue M-Labs/artiq#1724 2026-01-18 19:06:28 +08:00
List temporaries/literals in loops can cause stack overflows without compiler diagnostics

This is indeed quite a brittle area in the current implementation, in that the compiler barely helps in ensuring the correct lifetimes. That being said, the aspects at play can be understood from…

dpn commented on issue M-Labs/artiq#1707 2026-01-18 19:06:19 +08:00
Repository scans sometimes skip non-problematic experiments

We also have quite long repository scan times here (>20 s definitely), but never ran into this issue (unless it was introduced in the last couple of months). Maybe this is some sort of Windows-spec…

dpn commented on issue M-Labs/artiq#1702 2026-01-18 19:06:15 +08:00
now_mu() affects Urukul AD9910's output

Why do you say that "AD9910 output is clearly affected by call to now_mu()"? From the short description, what seems to be going on is that shifting the IO_UPDATE edge by 1 ns has an effect (the…

dpn commented on issue M-Labs/artiq#1702 2026-01-18 19:06:15 +08:00
now_mu() affects Urukul AD9910's output

tune_io_update_delay() is I think Robert Jördens's innovation when he integrated the multi-chip sync/phase calculations properly into upstream ARTIQ. We originally just mapped the four possible…

dpn opened issue M-Labs/artiq#1696 2026-01-18 19:06:13 +08:00
device_db entry keys ought to be sanitised