whitequark
3cf5cef168
artiq_pcap: still grab the file if the command fails.
2017-12-08 07:36:56 +00:00
whitequark
9e8bb1d51c
runtime: update smoltcp.
2017-12-08 07:36:56 +00:00
d6f62c5697
DEVELOPER_NOTES: add old release branch deprecation
2017-12-07 12:32:58 +01:00
63f65b6e53
doc: clean up artiq-dev installation instructions
...
Add a heading to the openocd setup instruction so that it is
clearly distinguishable from the openocd installation. Otherwise people
"re-install" openocd the wrong way.
2017-12-07 12:31:30 +01:00
69b2d1abfc
conda/artiq-dev: fix channel list
...
Now, with conda 4.1 packages are sorted by channel, version, build
number in decreasing priority. The highest matching package is
taken. https://conda.io/docs/user-guide/tasks/manage-channels.html
For the artiq-dev environment, the m-labs/label/dev channel should be
first, then the main channel, then defaults, and then conda-forge
(community supported packages).
closes #864
2017-12-07 12:31:30 +01:00
825f1ff1ce
move RELEASING into DEVELOPER_NOTES
2017-12-07 13:24:06 +08:00
e5d8314aa7
RELEASE_NOTES: 3.1
2017-12-07 13:23:11 +08:00
a44f8282dc
test_performance: relax network speed to 2 MB/s
...
At QUARTIQ I am getting 2.4/2.3 MB/s and with single switch at M-Labs we
apparently regularly met 2.2/2.2 MB/s. But with the current multiple
switches and one of them being a problematic switch that triggered #837
it looks like it is a tad slower.
http://buildbot.m-labs.hk/builders/artiq/builds/1818/steps/python_coverage_1/logs/stdio
2017-12-06 17:21:41 +01:00
c47006b154
conda: bump migen
...
Relevant changes:
* Kasli
2017-12-06 16:55:27 +01:00
fcbf6fde8a
conda: bump misoc again
...
(wrong version in previous commit)
shot at #837
2017-12-06 16:50:44 +01:00
41e30a5ddf
conda: bump misoc dependency
...
* fixed LiteETH preamble/SFD checker (this is a shot at #837 )
2017-12-06 16:48:21 +01:00
whitequark
1c25f7ef52
gateware: make software builds spew less junk on the console.
...
[ci skip]
2017-12-04 14:19:35 +00:00
whitequark
277b00eb9d
artiq_devtool: --config is no longer supported by artiq_flash.
...
[ci skip]
2017-12-04 14:19:31 +00:00
66e089b376
libboard/serwb: more explicit retry log message
2017-12-02 10:51:13 +08:00
Florent Kermarrec
0eb9f93611
libboard/serwb: retry serwb initialization on error
2017-12-01 22:45:05 +01:00
2e5b20d53d
libboard/hmc830_7043: enable HMC830
2017-11-30 12:06:54 +08:00
ad2a32d381
libboard/ad9154: small cleanup
2017-11-30 11:33:58 +08:00
Florent Kermarrec
bd75954192
libboard/ad9154: update for sayma (spi, jesd parameters, linerate), breaks kc705/ad9154 fmc support
2017-11-29 18:09:39 +01:00
Florent Kermarrec
8b8da39a8f
libboard/hmc830_7043.rs: fix HMC7043 comments
2017-11-29 17:59:42 +01:00
19498e59f9
libboard: fix JESD reset release
2017-11-29 18:18:41 +08:00
bb3d6ef84a
sayma: remove ad9154 from mem_map
...
Address is autogenerated by CSR system.
2017-11-29 18:17:25 +08:00
1426ecad64
dashboard: tidy dashboard naming logic
2017-11-28 08:56:04 +08:00
fbd4d8eb6f
dashboard: log server address even if friendly name given
2017-11-28 08:56:04 +08:00
9a1adfe65c
RELEASE_NOTES: master friendly name, dashboard config one-per-master
2017-11-28 08:56:04 +08:00
b6ad510007
master: style (NFC)
2017-11-28 08:56:04 +08:00
00efb0ccd2
master: expand --name help
2017-11-28 08:56:04 +08:00
10ee3f19a8
dashboard: close config RPC Client cleanly at exit
2017-11-28 08:56:04 +08:00
0999f2dd86
dashboard: explain default configuration file path
2017-11-28 08:56:04 +08:00
84b5e68349
dashboard: use master's friendly name in dashboard title
2017-11-28 08:56:04 +08:00
fafabacf97
master: add friendly name
2017-11-28 08:56:04 +08:00
2852e793ab
dashboard: store separate configuration data for each master, keyed by server and port
2017-11-28 08:56:04 +08:00
cfb41e71a8
libboard: print serwb settings at the debug log level
2017-11-28 01:23:33 +08:00
whitequark
cf5d50de0c
conda: do not compile .py files to .pyc.
...
We are producing a noarch package, and conda will strip all .pyc
files, so this is a significant waste of time.
2017-11-26 23:59:53 +00:00
whitequark
30f47994db
artiq-dev: fix typo.
2017-11-26 23:48:41 +00:00
whitequark
8e8f037971
runtime: print ethmac preamble error count.
...
Also, print total error count, not the difference (but still compute
the difference to determine *when* to print).
2017-11-26 22:35:26 +00:00
whitequark
e5995794be
artiq_devtool: actually hold the lock until script termination.
...
The open channel that ran flock was garbage collected, which caused
the flock command to terminate.
2017-11-26 20:54:24 +00:00
whitequark
b5d098c6db
Add lockfile support to artiq_devtool.
2017-11-26 15:17:35 +00:00
b89929565c
manual: fix formatting problem
2017-11-25 14:46:31 +08:00
14e4e1cf22
Hack-patch Sphinx so that ARTIQ-Python types are correctly printed.
...
Modification proposed to Sphinx but my issue is getting ignored.
Closes #741
2017-11-25 14:45:31 +08:00
7f15f50878
coredevice/ad9912: add [untested, wip]
2017-11-24 18:47:46 +01:00
280392708d
sawg: fix typo
2017-11-22 20:06:02 +08:00
29181b1586
artiq_influxdb: use aiohttp.ClientSession. Closes #829
2017-11-22 17:31:09 +08:00
whitequark
26fdd42f8f
runtime: update smoltcp.
2017-11-22 08:09:06 +00:00
ecfe2e40ee
sayma_amc_standalone: rtio channels for both sawg groups
2017-11-19 18:32:42 +01:00
d1a7c1c3a1
sayma_amc_standalone: connect sawg to jesd again
2017-11-19 14:36:20 +01:00
Florent Kermarrec
dfdd2dd9e6
gateware/targets/sayma_amc_standalone: revert self.add_wb_slave on serwb
2017-11-19 09:01:20 +01:00
Florent Kermarrec
cd83b71d92
gateware/targets/sayma_amc_standalone: serwb working, need fixing on AD9154 data mapping
2017-11-18 18:10:28 +01:00
Florent Kermarrec
a3383c340c
firmware/satman/lib.rs: has_serwb_phy to has_serwb_phy_amc
2017-11-18 18:09:35 +01:00
Florent Kermarrec
39a8fc682d
artiq/firmware/libboard/hmc830_7043.rs: some cleanup, don't use hmc830 for now
2017-11-18 18:08:31 +01:00
Florent Kermarrec
f003566e52
serwb: fix rx_delay_inc on ultrascale, this was the issue serwb issue...
...
rx_delay_inc and rx_delay_ce were set for only one cycle, on ultrascale, these signals are translated to serwb_serdes_5x clock domain and we now set rx_delay_inc always to 1 (MultiReg), rx_delay_ce for one cycle (PulseSynchronizer)
2017-11-18 18:01:46 +01:00
