<rjo>
sb0: i can't get the m_state memory for su-servo (dual port, both are writable) to be READ_FIRST. is that known to be possible with migen? afaik 7 series bram can do it.
<sb0>
rjo, READ_FIRST vs. other modes only changes the verilog per-port, so as far as migen is concerned it should not matter how many ports the memory has
<sb0>
rjo, I'm not exactly sure what you are trying to achieve, but multi-port memory with the two ports clocked with the same clock is tricky, e.g. in some cases vivado will not do what the verilog describes when there is an address collision
<sb0>
rjo, can you post the migen code you wrote, and explain where the generated verilog is not what you want?
<rjo>
sb0: the docs explicitly say that the use case is ok. dual port, read-write addr collision, read_first is well defined.
<rjo>
sb0: the verilog is fine afaict. just vivado says it doesn't recognize the pattern iirc.
<sb0>
whitequark, you'll have to reduce the overclocking on the buildserver. it just crashed.
<sb0>
it also refuses to boot again now, quite annoying
<rjo>
whitequark: i don't think trying to handle overtemp or stability issues caused by overclocking by adding monitoring is a worthwhile approach. if you really want to spend time on that, i'd recommend influxdb, telegraf, kapacitor, grafana.
<sb0>
okay, it rebooted now, you need to cycle the mains input...
<sb0>
_florent_, you can look into the serwb problem now, i'm not using sayma further
<whitequark>
sb0: yeah, I warned that this may happen
<whitequark>
although I wonder why it's so much less stable, maybe the lab is hotter?
<sb0>
because I'm running 3 instances of vivado at the same time, probably
<whitequark>
oh.
bb-m-labs has quit [Ping timeout: 276 seconds]
bb-m-labs has joined #m-labs
rohitksingh_wor1 has joined #m-labs
<_florent_>
sb0: i'll try to look at that today, but have limited time and i'm not working the next
rohitksingh_work has quit [Ping timeout: 256 seconds]
<_florent_>
sb0: 3 days (holiday in france)
<_florent_>
sb0: it's probably SI issues or timing issues...
<sb0>
_florent_, this is happening on 2 different boards
<sb0>
and it is blocking an important milestone, being able to use Sayma in a lab to generate RF...
<_florent_>
sb0: a week ago, it was working fine with sawg
<sb0>
_florent_, yep. you should be used to this kind of behavior with sayma by now.
<sb0>
_florent_, also I suggest checking your timing calibration algo carefully. the one you wrote for jesd204 sc1 has some problems. you can look at e.g. the siphaser or ddr3 calibration algos for examples of solutions to jitter and annoying corner cases.
<GitHub13>
[smoltcp] m-labs-homu closed pull request #196: Fix ICMPv6 checksums and add address resolution for IPv6 (master...ipv6) https://github.com/m-labs/smoltcp/pull/196
<GitHub98>
[smoltcp] m-labs-homu commented on issue #186: :umbrella: The latest upstream changes (presumably 6c363762845e69dd316617f7ae0a083b4e534d5b) made this pull request unmergeable. Please resolve the merge conflicts. https://github.com/m-labs/smoltcp/pull/186#issuecomment-387113406
<GitHub4>
[smoltcp] m-labs-homu commented on issue #190: :umbrella: The latest upstream changes (presumably 6c363762845e69dd316617f7ae0a083b4e534d5b) made this pull request unmergeable. Please resolve the merge conflicts. https://github.com/m-labs/smoltcp/pull/190#issuecomment-387113478
<travis-ci>
m-labs/smoltcp#892 (master - 6c36376 : Dan Robertson): The build passed.
<GitHub156>
[smoltcp] pothos opened pull request #200: Update remote_last_win for all ACKs (master...update_remote_last_win_for_all_acks) https://github.com/m-labs/smoltcp/pull/200
<GitHub44>
[smoltcp] pothos opened pull request #201: Mention that Probing Zero Windows is not implemented for TCP (master...mention_probing_zero_windows) https://github.com/m-labs/smoltcp/pull/201
<GitHub36>
[smoltcp] m-labs-homu commented on issue #201: :pushpin: Commit ab2ef05 has been approved by `whitequark`
<GitHub65>
[smoltcp] m-labs-homu closed pull request #201: Mention that Probing Zero Windows is not implemented for TCP (master...mention_probing_zero_windows) https://github.com/m-labs/smoltcp/pull/201
<travis-ci>
m-labs/smoltcp#898 (master - 880fb73 : Kai Lüke): The build passed.