Rajkumar S

After adding one more app at commit b7b14cc the performance dropped drastically to around 600kpps. Most of the time snabb spent on interpreted code. I am wondering what could have...

Just wondering which aborted trace could be behind the high use of interpreted code? How can that be identified? It may be very well be possible for many trace aborts...

I am seeing quite a lot of trace aborts related to loop, not just in `intel_mp`, but in other modules also. The code in PR is just a small sample...

Merged lukego/jit-tracebarrier and ran the tests again. Dumpfile is attached. [dump-with-1242.txt.gz](https://github.com/snabbco/snabb/files/1445269/dump-with-1242.txt.gz)

Ah, ok thanks, this gives me some thing to start looking!

I added an assert and a check for length before calling new_from_mem to figure out what was happening. ``` while not link.empty(input) do local p = link.receive(input) local data, length...

Update: I have tried using `apps.intel.intel_app` but I am still getting this error. So this may not be a but in `intel_mp`. I have added a condition to ignore packets...

Some more updates: I printed the packet count from the NIC till my app to check if any packets are getting lost on the way. Printed the following stats from...

On Sat, Dec 2, 2017 at 7:27 PM, Luke Gorrie wrote: > I am nailing down the roadmap for RaptorJIT 1.0. Currently a > work-in-progress with a Milestone > and...