<GitHub183>
[smoltcp] whitequark commented on issue #246: No, this is not wrong. There is a special element in the metadata ring that means "there is a packet after this element, but this element itself does not signify anything, it only consumes space in the tail end of the buffer because this space was not sufficient to insert the next packet". Aka "padding". https://github.com/m-labs/smoltcp/issues/246#issuecomment-39897
<GitHub-m-labs>
[artiq] whitequark commented on issue #1075: The reason this isn't done is because Rust's [`format!`](https://doc.rust-lang.org/nightly/std/fmt/index.html) facility does not support providing a format string at runtime. I could do something like `.replace("{0}", args[0].to_string()`, but that will miss valid cases like `{0:#p}` and result in an even more obscure message. https://github.com/m-labs/artiq/issu
<GitHub-m-labs>
[artiq] whitequark commented on issue #1074: Fixed. But this makes builds take 8 seconds regardless of whether they're fresh or not because of some issue with dependency tracking I never figured out. https://github.com/m-labs/artiq/issues/1074#issuecomment-398979257
<GitHub-m-labs>
[artiq] hartytp commented on commit 05e908a: Nit: am I right in thinking that this is not just due to the HMC7043 but also due to the HMC830 when the output dividers are enabled (DAC clock < 1.5GHz), since there is no way of deterministically resetting those dividers. i.e. if we increased the HMC830 output divider to give 600MHz output and ran the HMC7043 in fundamental mode for the DAC channels then we wou
<GitHub-m-labs>
[artiq] sbourdeauducq commented on commit 05e908a: If we configured the HMC830 to output 600MHz, we would not need this alignment here. But then we have a problem because the HMC7043 analog delay range is less than one half-period of 600MHz, and the combination of analog+coarse digital delay gets a "hole". https://github.com/m-labs/artiq/commit/05e908a0fd66709bc143587bb843730a1133535d#commitcomment-2944
<GitHub-m-labs>
[artiq] mfe5003 commented on issue #1078: Thanks for that link. The updated manual is helpful. It still seems to be behind since artiq_coreconfig appears to not be a command anymore.... https://github.com/m-labs/artiq/issues/1078#issuecomment-399239193
<travis-ci>
batonius/smoltcp#135 (layers - 27cccc0 : Egor Karavaev): The build is still failing.
<GitHub-m-labs>
[artiq] marmeladapk commented on issue #1078: @mfe5003 Can you run idle kernel you mentioned in your first post? (artiq_run name_of_experiment.py in the same directory as device_db.py). https://github.com/m-labs/artiq/issues/1078#issuecomment-399242078
<GitHub-m-labs>
[artiq] mfe5003 commented on issue #1078: @marmeladapk I haven't defined and experiment file yet, but I suspect because I haven't been able to write the idle_kernel to the device yet that it would not work. I will try to make an experiment though and let you know. https://github.com/m-labs/artiq/issues/1078#issuecomment-399243126
<GitHub-m-labs>
[artiq] marmeladapk commented on issue #1078: @mfe5003 Ok, if this doesn't work (and no LED is flashing on front panel nor near power supply LEDs) we need to find out if you're using right variant of gateware with right variant of software. If you're flashed different variant of gateware to the board it will have different numbering of channels. ... https://github.com/m-labs/artiq/issues/1078#issuecomment-3
<GitHub-m-labs>
[artiq] marmeladapk commented on issue #1078: @mfe5003 Ok, if this doesn't work (and no LED is flashing on front panel nor near power supply LEDs) we need to find out if you're using right variant of gateware with right variant of software. If you're flashed different variant of gateware to the board it will have different numbering of channels. ... https://github.com/m-labs/artiq/issues/1078#issuecomment-3
<GitHub-m-labs>
[artiq] marmeladapk commented on issue #1078: Could you check what variants are present in C:\\Users\\ebert\\Anaconda2\\envs\\artiq-dev\\lib\\site-packages\\artiq\\binaries? Perhaps we could try with opticlock. https://github.com/m-labs/artiq/issues/1078#issuecomment-399264736
<GitHub-m-labs>
[artiq] mfe5003 commented on issue #1078: There is only kasli-master in the directory, however the actual path is `envs/artiq-dev/**Lib**/site-packages` which might be part of the problem.... https://github.com/m-labs/artiq/issues/1078#issuecomment-399266354