Commit Graph

6270 Commits

Author SHA1 Message Date
David Nadlinger
f79b9d9e1e ttl: Expand input gate/count API docstrings 2018-11-03 20:33:19 +08:00
David Nadlinger
f02ceee626 language: Clarify now_mu() docstring [nfc] 2018-11-03 20:33:19 +08:00
David Nadlinger
e7d3f36b91 doc: Fix ancient trigger code in slides
These examples were already broken before my recent changes to the
artiq.coredevice.ttl API.
2018-11-03 20:33:19 +08:00
David Nadlinger
5d2e3f975f coredevice: Add get_rtio_counter_mu() docstring [nfc] 2018-11-03 20:33:19 +08:00
David Nadlinger
d6fcc0529f coredevice: Imperative mood in docstrings [nfc]
This follows Python conventions (PEP257) and unifies the style with
other comments.
2018-11-03 20:33:19 +08:00
David Nadlinger
cbdef0225c ttl: Add target RTIO time argument to timestamp/count functions
Software-based tracking of timestamps is problematic (e.g. when
using DMA, see GitHub #1113).
2018-11-03 20:33:19 +08:00
David Nadlinger
2a0e1dabfb ttl: Remove unused attribute [nfc] 2018-11-03 20:33:19 +08:00
David Nadlinger
17a5fb2dce ttl: Remove error-prone sync() calls
These methods are problematic, as with DMA in the picture, the
timestamp member variables did not necessarily reflect the last
submitted event timestamp (see GitHub #1113).

sync() is only very rarely used in typical experimental code, so
the methods are removed without a transition period.
Core.wait_until() can be used to busy-wait for a specified RTIO
timestamp value on the core device CPU instead.
2018-11-03 20:33:19 +08:00
David Nadlinger
11e8c9d5f7 coredevice: Add Core.wait_until_mu()
(This supersedes TTLOut.sync(), see see GitHub #1113.)
2018-11-03 20:33:19 +08:00
David Nadlinger
cbfbe24d7a ttl: Remove broken TTLClockGen.sync
The code currently doesn't compile because of a typo in the timestamp
field name. However, tracking event timestamps in software is
problematic anyway (e.g. with DMA, see GitHub #1113), so just remove
`sync()` altogether.
2018-11-03 20:33:19 +08:00
7f11411127 RELEASE_NOTES: 3.7 2018-11-03 20:17:28 +08:00
bc4a8157c0 kasli: add tsinghua2 2018-11-01 18:26:37 +08:00
David Nadlinger
91f0d640e2
.github: Match suggested commit message format to established convention
This changes the suggested format for commit messages to match the
convention actually used here (no square brackets).
2018-10-25 22:01:14 +01:00
89a961fb00 urukul, ad9912, ad9910: expose CFG RF switch better
* conincident setting of multiple switches
* per channel setting
2018-10-24 13:04:46 +02:00
Drew
4e04b6b4dd [template]: fix 3_question.md template use of ":"
Question.md didn't display on ARTIQ new issue page.
Solution: Cannot use ":" in Markdown header

Signed-off-by: Drew Risinger <drewrisinger@users.noreply.github.com>
2018-10-23 23:01:58 +08:00
Drew
761d30d998 [template]: Reorder pull request template
PR template wasn't in a logical order, hard for maintainers to read. Reorganized.
Fix directory where docs are built from.

Signed-off-by: Drew Risinger <drewrisinger@users.noreply.github.com>
2018-10-23 23:01:22 +08:00
0b43ec4719 conda: bump migen 2018-10-21 14:33:14 +08:00
48a142ed63 use FutureWarning instead of DeprecationWarning
DeprecationWarning is disabled by default and too easy to ignore.
2018-10-21 12:14:51 +08:00
9793632282 enviromnment: rename 'save' in set_dataset to 'archive'. Closes #1171 2018-10-21 12:08:34 +08:00
Marius Weber
029f9d983a Tpz fixes (#1178)
*   flake8
*   fix TPZ constructor after move to asyncio
*   Tcube fix docummentation in set_channel_enable_state
2018-10-20 20:49:15 +08:00
Drew
a84c6c6d74 Add issue & pull request templates (#1163) 2018-10-20 20:43:19 +08:00
c52a6ca8e9 typo (#1179) 2018-10-20 20:40:07 +08:00
Florent Kermarrec
3318925f4e firmware/liboard_misoc/sdram: use similar loops on read_level_scan and read_level for consistent results 2018-10-16 09:40:39 +02:00
6357a50d33 kasli: update nudt variant 2018-10-15 18:04:57 +08:00
David Nadlinger
e3cfbfed06 master: Add minimal docstring to worker_impl [nfc] 2018-10-14 10:41:32 +08:00
David Nadlinger
64b9a377da master: Factor RIDCounter out into own module, explain worker_db module [nfc]
The docstrings are quite minimal still, but should already
help with navigating the different layers when getting
accustomed with the code base.

RIDCounter was moved to its own module, as it isn't really
related to the other classes (used from master only).
2018-10-14 10:41:32 +08:00
David Nadlinger
4641ddf002 master: Remove unused import [nfc] 2018-10-14 10:41:32 +08:00
661dd00c4c ad9912: phase offset is 14 bit LSB aligned
c.f. sinara-hw/Urukul#15
2018-10-11 15:16:08 +02:00
hartytp
08074d5275 Urukul: add support for hardware v1.3 clocking options 2018-10-10 09:26:35 +02:00
469a66db61 drtio: monitor RTIOClockMultiplier PLL (#1155)
Debugging by Tom Harty
2018-10-08 14:50:02 +02:00
9cf88329b2 CONTRIBUTING: correct default licensing 2018-10-08 11:20:37 +02:00
hartytp
1a1b454ed9 Urukul: flake8 (NFC) 2018-10-08 10:09:49 +01:00
5de319d76a bit2bin: don't print string terminator 2018-10-08 10:13:03 +02:00
hartytp
9a509e5070 Zotino: increase delay after register read in init method to avoid underflows 2018-10-06 21:45:17 +08:00
86fe6b0594 kasli: add NUDT variant 2018-10-04 23:20:09 +08:00
a89bd6b684 kasli: swap Urukul EEMs for Tester
Updated to Urukul 1.3.
2018-10-04 23:19:31 +08:00
d0ee2c2955 opticlock: external 100 MHz 2018-09-28 19:05:18 +02:00
Drew Risinger
0fef2afccb make.bat: fix typo 2018-09-27 23:22:19 +08:00
Drew Risinger
fad7128a52 Create Docs Makefile for Windows. #1159
Allows building Sphinx documentation on Windows,
(not just Linux as previously).
Closes #1159.
2018-09-27 23:22:19 +08:00
3b3fddb5a4 kasli: add mitll2 2018-09-27 23:21:52 +08:00
998a468983 examples: add grabber to device databases 2018-09-27 16:09:25 +08:00
c71e442929 documentation improvements
Based on PR #1101 by @drewrisinger
2018-09-26 12:12:37 +08:00
53c7a5f2c6 dashboard: fix TTL moninj level display 2018-09-26 10:32:08 +08:00
20cddb6a25 tester: handle no available ttl outputs 2018-09-24 09:19:28 +00:00
69d060b639 drtio: fix satellite i_status handling 2018-09-19 20:57:21 +08:00
David Nadlinger
b482f5feae firmware: Use larger ARP cache
This works around a problematic interaction between ARP cache
expiry in smoltcp (with its 3 seconds timeout before a discovery
request is sent) and our TCP keepalive settings, where the timeout
is reached before the keepalive had a chance to be sent.

GitHub: Closes #1150.
2018-09-18 14:46:52 +00:00
7565d816e4 frontend: remove artiq_pcap. Closes #1152 2018-09-15 12:09:37 +08:00
eaeab0c9bd frontend: add artiq_rtiomon 2018-09-15 12:09:32 +08:00
312256a18d grabber: fix frame size off-by-1 2018-09-07 16:55:43 +02:00
ec62eb9373 drtio: minor cleanup 2018-09-07 17:51:38 +08:00