Messages in this thread | | | Date | Thu, 21 Mar 2019 11:22:22 -0700 | Subject | Re: [RFC][PATCH] tracing/x86: Save CR2 before tracing irqsoff on error_entry | From | hpa@zytor ... |
| |
On March 21, 2019 11:18:53 AM PDT, Linus Torvalds <torvalds@linux-foundation.org> wrote: >On Thu, Mar 21, 2019 at 11:05 AM Andy Lutomirski <luto@amacapital.net> >wrote: >> >> In the long run, I think the right solution is to rewrite even more >of >> this mess in C. We really ought to be able to put the IRQ flag >> tracing and the context tracking into C code. > >Tangentially about this long run thing - can we start discussing just >dropping PV mode entirely in the long run? Or even not-so-long run? > >What are the pressing advantages of paravirt these days? Because I can >point to a lot of pressing _disadvantages_, and not all of them are >about more complex code as shown by this patch... > > Linus
God only knows how nice that would be (for things that aren't naturally driverized.) Last I heard, which was quite a while ago, the big problem was that the Amazon cloud runs Xen in PV mode... -- Sent from my Android device with K-9 Mail. Please excuse my brevity.
| |