Am 26.02.2014 01:17, schrieb Antti Kantee: > On 25/02/14 19:28, Martin Unzner wrote: >> This is not a pressing issue, and you are quite right, I could easily >> use asynchronous communication. >> >> I actually do, in my own code, but I'm fiddling with netperf ATM, and >> I'm already glad that I got it working at all, thus nothing's going to >> change there. Thank you for confirming that host signals are not >> forwarded, though, because that eliminates the possibility of bugs in >> the important parts of the setup. > I don't think anyone truly understands what is going on with signals ;) > > But speaking of netperf, the standard network benchmarking tools are > something that is asked about a decent amount. Would you care to write > a howto on how you use netperf, perhaps on the wiki? I am not sure if my experience helps -- the average Rump user would probably want to use rumprun and thus port the whole of netperf, while I more or less only replaced the AF_INET-socket related syscalls in the netperf netserver source code, added an init function and ran it on Linux. If you think that is a valid use case, please tell me, then I could try and assemble a patchset (I do not want to get involved in any kind of legal issues with HP, thus no complete source tree) and document it somewhere, also in the wiki if you like. This could take a while, though (worst case: end of May), because I still have to finish my Diploma thesis.
HTH Martin ------------------------------------------------------------------------------ Flow-based real-time traffic analytics software. Cisco certified tool. Monitor traffic, SLAs, QoS, Medianet, WAAS etc. with NetFlow Analyzer Customize your own dashboards, set traffic alerts and generate reports. Network behavioral analysis & security monitoring. All-in-one tool. http://pubads.g.doubleclick.net/gampad/clk?id=126839071&iu=/4140/ostg.clktrk _______________________________________________ rumpkernel-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/rumpkernel-users
