I am trying to compare performance of walking without a tick as well as walking with ticks, but SE does not do the latter.
> On Dec 9, 2015, at 9:34 AM, Nizamudheen Ahmed <[email protected]> wrote: > > I thought you wanted the walk without a tick... Which is what the SE seems to > do, per your claim. I am sure I am missing something.. > From: Arman Shan <mailto:[email protected]> > Sent: 09-12-2015 21:01 > To: gem5 users mailing list <mailto:[email protected]> > Subject: Re: [gem5-users] TranslateFunctional > > It turns out that in SE mode, there is no latency for page table walk. The > page table is internally a hash table, and it returns immediately, without > simulating memory accesses. I find this ridiculous, don’t you? > > >> On Dec 9, 2015, at 9:27 AM, Nizamudheen Ahmed <[email protected] >> <mailto:[email protected]>> wrote: >> >> Translate atomic should help you do the translation without tick. Isn't it? >> From: Arman Shan <mailto:[email protected]> >> Sent: 08-12-2015 23:56 >> To: [email protected] <mailto:[email protected]> >> Subject: [gem5-users] TranslateFunctional >> >> Hi, >> >> I was wondering if someone has already implemented translate functional for >> the TLB. Otherwise, does anyone know how to do a page table walk without >> using any ticks? >> >> Thank you, >> Arman >> _______________________________________________ >> gem5-users mailing list >> [email protected] <mailto:[email protected]> >> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >> _______________________________________________ >> gem5-users mailing list >> [email protected] <mailto:[email protected]> >> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > > _______________________________________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
_______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
