On 4/19/20 2:14 PM, Romain Dolbeau wrote:
Le ven. 17 avr. 2020 à 02:30, Dennis Clarke <dclarke@blastwave.org> a écrit :I have no idea how that works.Normally it's the default, quoting the doc: "The default is to enable multi-threading where both the back-end and front-ends support it and no incompatible TCG features have been enabled (e.g. icount/replay)."I see terrible horrific performanceYou need some pretty fast host cores to get an interactively usable emulated host. My home test system is a dual E5-2690v4...
I have a dedicated Intel i7 based machine with 16G DDR4 and Samsung SSD's to run this instance. Exclusively. I did carefully walk through systemd to remove services that are mostly useless anyways.
ppc64@styx:~$ ./ppc64le.shIs your host PPC64 as the username implies? I assume it's not PPC64LE otherwise you'd run KVM rather than TCG, so PPC64BE ... I'm not sure you can get MTTCG doing PPC64LE-on-PPC64BE... and most PPC64BE systems are old and not particularly fast, emulating PPC64LE on them isn't going to be fast :-(
Ha .. no way. No. That is just the username that runs the instance. Also I did build qemu 5.0.0-rc3 with debug enabled and that will cause a 2x to 5x slow down I am told. However it did allow us to see an opcode problem which has since been fixed and I re-built with the patch. Seems to be running ( briskly walking ) fine now : https://lists.nongnu.org/archive/html/qemu-discuss/2020-04/msg00078.html If I see anything curious I will be sure to report. Also I will rebuild QEMU without the debug and check the speed that I get. If any. -- Dennis Clarke RISC-V/SPARC/PPC/ARM/CISC UNIX and Linux spoken GreyBeard and suspenders optional