Luke Gorrie

Fixed in lukego/raptorjit@473cb199e7741e88f210b904fecccf0129b4da59. Certain ctype IDs are not printable but will be attempted by the polling loop anyway. These would cause an assertion failure in the VM or else an...

> so the idea would be simply that blacklisting could be tagges case-by-case, like what went wrong. The challenge here is non-determinism. Each time you try to record a trace...

I'd say your descriptions are quite well aligned with the way the JIT actually works today: > those rules that should be applied to an input can be collected, like...

(Just quietly fixed a small bug that would prevent installation from working.)

@raj2569 Well #101 is part of the story anyway. I don't think the heuristic changes there are complete enough for your issue. The other major bit we need - which...

@nnikolaev-virtualopensystems The normal operation of `packetblaster` is to create transmit descriptors and then reuse them in a loop. This works on Intel NICs. Hopefully works on Mellanox NICs too. I...

Added a link to Mellanox [firmware release notes](http://www.mellanox.com/pdf/firmware/ConnectX4-FW-12_12_1240-release_notes.pdf). Interesting reading. Gives some insight into the line between hardware/firmware on the cards i.e. which bugs are fixed and features added by...

I have pushed a major update of the ConnectX-4 driver in commit 7659eb61fcf01cc9db342dfdc040644c23b45186. I have been able to initialize the card and transmit and receive packets now. I am reformulating...

I pushed a big update to the Mellanox driver with commit 21d0dc36d1f3dd1c022d19c3777339f52c3609b2. There is still work to do but most things are in place now. The driver is now designed...

Hi @tsuraan. Yes, I am actually planning to loop back this month and try to get the driver branch ready for upstream. There is a quite complete driver here: [connectx_4.lua](https://github.com/lukego/snabb/blob/debug100g/src/apps/mellanox/connectx4.lua)....