From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:44598) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dKldE-0002eA-Ix for qemu-devel@nongnu.org; Tue, 13 Jun 2017 09:13:20 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dKldA-00019A-Gt for qemu-devel@nongnu.org; Tue, 13 Jun 2017 09:13:16 -0400 Received: from mta01.ornl.gov ([128.219.177.137]:6402) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dKldA-00018l-61 for qemu-devel@nongnu.org; Tue, 13 Jun 2017 09:13:12 -0400 From: "Nutaro, James J." Date: Tue, 13 Jun 2017 13:13:06 +0000 Message-ID: <1497359587814.46408@ornl.gov> References: <1487097476-14259-1-git-send-email-nutarojj@ornl.gov>, <20170602144233.GA4868@flamenco> In-Reply-To: <20170602144233.GA4868@flamenco> Content-Language: en-US Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [Qemu-devel] [PATCH V8] qqq: module for synchronizing with a simulation List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Emilio G. Cota" Cc: "qemu-devel@nongnu.org" , "pbonzini@redhat.com" Thanks for the review Emilio. =0A= =0A= >> Isn't this a KVM-only feature though? I don't see calls to qqq_sync()=0A= >> anywhere but from kvm-all.c, but maybe I'm missing something.=0A= =0A= There is another call in the timer callback in qqq.c that gets invoked when= using =0A= the -icount mode.=0A= =0A= >> Also, a few more comments:=0A= =0A= >> - Why "qqq" as the name for this feature? I fail to see any=0A= >> connection between the name and the feature.=0A= =0A= There was a similar package called qbox, and I got stuck thinking of a name= . =0A= So I somewhat unimaginatively just stuttered on the q. I'm open to a better= =0A= name.=0A= =0A= >> If Windows users ever wanted this feature, would we have to=0A= >> change the user-facing API (i.e. the -qqq parameter)?=0A= =0A= It will (or should) ignore the -qqq option on Windows. But I'll double chec= k that.=0A= =0A= >> As is, isn't the code Unix-only rather than Linux-only?=0A= =0A= Yes, Unix-only. You are correct.=0A= =0A= Jim=0A= ________________________________________=0A= From: Emilio G. Cota =0A= Sent: Friday, June 02, 2017 10:42 AM=0A= To: Nutaro, James J.=0A= Cc: qemu-devel@nongnu.org; pbonzini@redhat.com=0A= Subject: Re: [Qemu-devel] [PATCH V8] qqq: module for synchronizing with a s= imulation=0A= =0A= On Tue, Feb 14, 2017 at 13:37:56 -0500, James J. Nutaro wrote:=0A= > This patch adds an interface for pacing the execution of QEMU to match an= external=0A= > simulation clock. Its aim is to permit QEMU to be used as a module within= a=0A= > larger simulation system.=0A= (snip)=0A= > +++ b/docs/simulation-sync.txt=0A= > @@ -0,0 +1,59 @@=0A= > +=3D Synchronizing the virtual clock with an external source =3D=0A= > +=0A= > +QEMU has a protocol for synchronizing its virtual clock=0A= > +with the clock of a simulator in which QEMU is embedded=0A= > +as a component. This options is enabled with the -qqq=0A= > +argument, and it should generally be accompanied by the=0A= > +following additional command line arguments:=0A= > +=0A= > +-icount 1,sleep=3Doff -rtc clock=3Dvm=0A= > + or=0A= > +-enable-kvm -rtc clock=3Dvm=0A= =0A= Isn't this a KVM-only feature though? I don't see calls to qqq_sync()=0A= anywhere but from kvm-all.c, but maybe I'm missing something.=0A= =0A= Also, a few more comments:=0A= =0A= - Why "qqq" as the name for this feature? I fail to see any=0A= connection between the name and the feature.=0A= =0A= - If Windows users ever wanted this feature, would we have to=0A= change the user-facing API (i.e. the -qqq parameter)?=0A= =0A= - As is, isn't the code Unix-only rather than Linux-only?=0A= =0A= Thanks,=0A= =0A= Emilio=0A= =0A=