Commit Graph

7237 Commits

Author SHA1 Message Date
3645a6424e sayma: fix Master build 2018-07-17 18:56:33 +08:00
9b016dcd6d eem: support specifying I/O standard
Xilinx FPGAs require different LVDS I/O standard names depending on I/O bank voltage.
2018-07-17 18:55:17 +08:00
3168b193e6 kc705: remove Zotino and Urukul
* use Kasli instead for using EEMs
* code required outdated VHDCI adapter 1.0
2018-07-17 17:48:57 +08:00
13984385a8 firmware: version → ident 2018-07-15 17:40:17 +08:00
b2695d03ed sayma: remove with_sawg from Master variant 2018-07-15 17:38:29 +08:00
123e7bc054 pyon: sort string dicts by key when pretty-printing. Closes #1010 2018-07-15 17:38:09 +08:00
b27fa8964b add variant in identifier string
Also add without-sawg suffixes on Sayma.

Closes #1060
Closes #1059
2018-07-15 17:21:17 +08:00
b6c70b3cb0 eem: add Zotino monitoring. Closes #1095 2018-07-15 15:35:04 +08:00
8bcba82b65 grabber: reset *_good signals on end of frame
This reduces the amount of time the ROI engine produces invalid output after
being reconfigured.
2018-07-15 15:34:00 +08:00
ea7f925852 Revert "worker_db: Only warn on repeated archive read if dataset changed"
Breaks numpy arrays.

This reverts commit 141fcaaa8a.
2018-07-13 10:41:06 +08:00
46fb5adac3 grabber: fix frequency counter formula 2018-07-12 20:14:38 +08:00
82def6b535 grabber: add frequency counter
Cameras are a bit obscure about what they output, this can help with troubleshooting.
2018-07-12 17:05:18 +08:00
29c35ee553 hmc7043: fix dumb mistake in previous commit 2018-07-12 13:01:41 +08:00
8802b930de hmc7043: add delay after init
Delay required at step 9 of the "Typical Programming Sequence" (page 24 of the datasheet)
2018-07-12 12:37:12 +08:00
c66f9483f8 hmc7043: wait after changing delays
Allows for the SPI transaction to finish, and for the delay to stabilize.
2018-07-12 12:33:53 +08:00
1c191a62bf sayma: tune SYSREF phases 2018-07-12 12:33:35 +08:00
773240bef4 hmc7043: test GPO before using
Based on code by David.
2018-07-12 11:30:24 +08:00
David Nadligner
141fcaaa8a worker_db: Only warn on repeated archive read if dataset changed
In larger experiments, it is quite natural for the same dataset
to be read from multiple unrelated components. The only situation
where multiple reads from an archived dataset are problematic is
when the valeu actually changes between reads. Hence, this commit
restricts the warning to the latter situation.
2018-07-12 10:15:42 +08:00
4843832329 hmc7043: check phase status on init. Closes #1055
Troubleshooting by David.
2018-07-11 19:45:24 +08:00
9397fa7f5a hmc7043: unstick SYSREF FSM (#1055)
Troubleshooting by David.

Additionally, register 7D is broken.
Checking phase init state has to be done through another means.
2018-07-11 19:11:01 +08:00
88fb9ce4d6 sayma_rtm: add hmc7043_gpo monitoring 2018-07-11 19:04:29 +08:00
29e5c95afa sayma_rtm: minor cleanup 2018-07-11 19:02:59 +08:00
7f05e0c121 sayma_rtm: remove UART loopback
RTM power supply issues are fixed now, plus this will get in the way of satman support.
2018-07-11 19:00:18 +08:00
f8ceea20d0 grabber: add new ROI engine (untested) 2018-07-10 17:06:17 +08:00
d82beee540 grabber: make parser EOP a pulse 2018-07-10 17:04:07 +08:00
701c93d46c grabber: add false path constraints 2018-07-10 14:28:23 +08:00
6a77032fa5 grabber: use BUFR/BUFIO
Less jitter and frees up BUFGs.
2018-07-10 13:30:38 +08:00
208dc7c218 grabber: prevent glitches in last_x/last_y cdc 2018-07-10 12:56:37 +08:00
c4e3c66265 grabber: add clock constraint 2018-07-10 12:37:32 +08:00
d51294649e conda: bump migen/misoc 2018-07-10 12:26:24 +08:00
David Nadlinger
768b970deb Fixup 4359a437 (tuples of lists), add regression tests 2018-07-10 01:18:51 +01:00
David Nadlinger
edc314524c test_embedding: Remove unused reference to led device 2018-07-10 01:11:47 +01:00
4f56710e4b grabber: add parser, report detected frame size in core device log 2018-07-10 02:06:37 +08:00
37e303dafc gitmodules: remove empty file 2018-07-09 19:32:29 +02:00
2d4af509c2 readthedocs.yml: drop (out of date, not working) 2018-07-09 19:31:27 +02:00
David Nadlinger
4359a43732 compiler: Indirection status of TTuple depends on elements
For instance, TTuple(TList(TInt32())) has indirections, while
TTuple(TInt32()) does not.

This fixes memory corruption with RPCs that return tuples of lists.

Signed-off-by: David Nadlinger <code@klickverbot.at>
2018-07-09 18:49:50 +08:00
d2c8e62cb7 test_rtio: relax ClockGeneratorLoopback performance requirements 2018-07-09 18:07:25 +08:00
423929a125 test: relax min transfer rates from 2MB/s to 1.9MB/s 2018-07-09 18:00:24 +08:00
9153c4d8a3 use tokenize.open() to open Python source files
Fixes encoding issues especially with device databases modified in obscure editors.
2018-07-07 17:04:56 +08:00
4420046502 kasli_tester: support mixed AD9910/AD9912 systems 2018-07-06 15:43:38 +08:00
ac3f360c26 kasli_tester: fix AD9912 support 2018-07-06 15:43:25 +08:00
509562ddbf kasli: add WIPM target 2018-07-06 15:41:28 +08:00
4eb26c0050 hmc7043: enable group 5 2018-07-03 14:16:31 +02:00
16b917be5d doc: add reminder of what positive slack means. Closes #1084 2018-07-02 16:23:12 +08:00
540bdae99c grabber: enable DIFF_TERM on inputs 2018-07-01 09:28:51 +08:00
0483b8d14c sayma_drtio: ditto 2018-06-28 17:03:32 +08:00
04d6ff45c8 kasli_sawgmaster: reset SAWGs
Most importantly this resets the phase accumulators.
2018-06-28 17:01:48 +08:00
729ce58f98 sayma: use GTP_CLK1 to clock DRTIO satellite transceiver
This is required to get constant skew between the DRTIO transceiver clock
(which then generates the RTIO clock) and the siphaser reference clock.

Both the Si5324 and the RTM clock tree have non-deterministic in-to-out skew
at 150MHz due to dividers.
2018-06-28 11:23:40 +08:00
a65721d649 sayma: put RTM clock tree into the siphaser loop
* Fixes one bug where siphaser was one Si5324 output and the rest of the
system was clocked by the other. With the Si5324 settings we have, skew
between the outputs is not controlled.
* Puts the coaxial cable between AMC and RTM into the siphaser loop.
2018-06-27 21:46:55 +08:00
d49716dfac satman: tune Sayma SYSREF phases 2018-06-27 18:09:35 +08:00