Messages in this thread | | | From | Andy Lutomirski <> | Date | Fri, 22 May 2020 15:13:57 -0700 | Subject | Re: [RFC][PATCH 0/4] x86/entry: disallow #DB more |
| |
On Fri, May 22, 2020 at 1:49 PM Peter Zijlstra <peterz@infradead.org> wrote: > > Hai, this kills #DB during NMI/#MC and with that allows removing all the nasty > IST rewrite crud. >
This is great, except that the unconditional DR7 write is going to seriously hurt perf performance. Fortunately, no one cares about perf, right? :) Even just reading first won't help enough because DR7 reads are likely to be VM exits. Can we have a percpu dr7 shadow (with careful ordering) or even just a percpu count of dr7 users so we can skip this if there are no breakpoints? We have cpu_dr7, and some minor changes would make this work. Maybe replace all the direct cpu_dr7 access with helpers like dr7_set_bits() and dr7_clear_bits()?
Also, I like raving at DR7 :)
| |