On Tue, Oct 18, 2022 at 10:38:10AM +0300, Oliver Upton wrote: > If we ever get to the point that we can relax this restriction i think a > flag on the BITMAP_WITH_TABLE cap that says "I don't actually set any > bits in the bitmap" would do. We shouldn't hide the cap entirely, as > that would be ABI breakage for VMMs that expect bitmap+ring. I'd rather drop the cap directly if it's just a boolean that tells us "whether we need bitmap to back rings". Btw when I said "dropping it" I meant "don't return 1 for the cap anymore" - we definitely need to make the cap macro stable as part of kvm API.. But I think I misunderstood the proposal previously, sorry. I assumed we were discussing an internal HAVE_DIRTY_RING_WITH_BITMAP only. I noticed this only after I had a closer look at Gavin's patch. Having a cap exposed sounds always good to me. I'll comment on Gavin's patch directly, thanks. -- Peter Xu
