Nils Goroll
Nils Goroll
FTR: today's bugwash seemed to be generally in favor, Dridi wants to have a closer look still.
bugwash: implement vcc_trace differently and move to new PR: * use debug flag * add vcl control over it
force-push: Update to remove vcl tracing change I have polished the initial commit of this PR so I am now happy with it and thus removed the Draft status. I...
IRC notes - in panic, dump the VCL code line instead of "token", if possible - @bsdphk questions if the VCL line info for panic can be achieved more simply...
I was asked to provide some numbers on the relevance of the performance improvement enabled by this PR. This PR itself does not aim at performance improvements, but it is...
@Dridi you have a point, sorry.
_if_ we want to tackle this, should we also address a unit for (m)onths?
FTR: clarified with @bsdphk via IRC: I should have written a better description: He proposed that the preamble should be in the backend, but putting it in the tcp pool...
> maybe we could move in this direction first and review that change in isolation? @Dridi OK, but can we _please_ tick off the rather trivial #2680 and #2845 first?...