Please let's go with hardfloat pps support, it's really a good feature to implement. Even if in a first step it could lead to inaccuracy results, later it could solved with other patches.
I think it's important for qemu to as global as possible and don't target only recent hardware. Regards, Dino Papararo Da: Qemu-ppc <qemu-ppc-bounces+skizzato73=msn....@nongnu.org> Per conto di luigi burdo Inviato: mercoledì 26 febbraio 2020 14:01 A: BALATON Zoltan <bala...@eik.bme.hu>; Programmingkid <programmingk...@gmail.com> Cc: David Gibson <da...@gibson.dropbear.id.au>; qemu-...@nongnu.org; qemu-devel qemu-devel <qemu-devel@nongnu.org>; Howard Spoelstra <hsp.c...@gmail.com> Oggetto: R: [RFC PATCH v2] target/ppc: Enable hardfloat for PPC Hi Zoltan, i can say MacOs Leopard use multiple cores on PowerMac G5 Quad the most of the apps did for Panter/Tiger/leopard use for sure 2 Core in smtp only apps did for Tiger/leopard use more than 2 Cores. Ciao and thenks Luigi ________________________________ Da: Qemu-ppc <qemu-ppc-bounces+intermediadc=hotmail....@nongnu.org<mailto:qemu-ppc-bounces+intermediadc=hotmail....@nongnu.org>> per conto di BALATON Zoltan <bala...@eik.bme.hu<mailto:bala...@eik.bme.hu>> Inviato: mercoledì 26 febbraio 2020 12:28 A: Programmingkid <programmingk...@gmail.com<mailto:programmingk...@gmail.com>> Cc: Howard Spoelstra <hsp.c...@gmail.com<mailto:hsp.c...@gmail.com>>; qemu-...@nongnu.org<mailto:qemu-...@nongnu.org> <qemu-...@nongnu.org<mailto:qemu-...@nongnu.org>>; qemu-devel qemu-devel <qemu-devel@nongnu.org<mailto:qemu-devel@nongnu.org>>; David Gibson <da...@gibson.dropbear.id.au<mailto:da...@gibson.dropbear.id.au>> Oggetto: Re: [RFC PATCH v2] target/ppc: Enable hardfloat for PPC On Wed, 26 Feb 2020, Programmingkid wrote: > I think a timeout takes place and that is why audio stops playing. It is > probably an USB OHCI issue. The other USB controller seems to work > better. Which other USB controller? Maybe you could try enabling some usb_ohci* traces and see if they reveal anything. >> The Amiga like OSes I'm interested in don't use multiple cores so I'm >> mainly interested in improving single core performance. Also I'm not >> sure if (part of) your problem is slow FPU preventing fast enough audio >> decoding then having multiple CPUs with slow FPU would help as this may >> use a single thread anyway. > > Good point. MTTCG might be the option that really helps with speed > improvements. Only if you have multithreaded workload in the guest because AFAIK MTTCG only runs different vcpus in parallel, it won't make single emulated CPU faster in any way. OSX probably can benefit from having multiple cores emulated but I don't think MacOS would use it apart from some apps maybe. Regards, BALATON Zoltan
