On Tue, Apr 18, 2023, Metin Kaya wrote: > Implement in-KVM support for Xen's HVMOP_flush_tlbs hypercall, which > allows the guest to flush all vCPU's TLBs. KVM doesn't provide an > ioctl() to precisely flush guest TLBs, and punting to userspace would > likely negate the performance benefits of avoiding a TLB shootdown in > the guest. > > Signed-off-by: Metin Kaya <metikaya@xxxxxxxxxxxx> > > --- > v3: > - Addressed comments for v2. > - Verified with XTF/invlpg test case. > > v2: > - Removed an irrelevant URL from commit message. > --- I had applied this and even generated the "thank you", but then I actually ran the KUT access test[*]. It may very well be a pre-existing bug, but that doesn't change the fact that this patch breaks a test. I apologize for being snippy, especially if it turns out this is unique to HSW (or worse, my system), and your systems don't exhibit issues. Getting the test to run took way too long and I'm bit grumpy. Anyways, whatever is going wrong needs to be sorted out before this can be applied. [*] https://lore.kernel.org/all/ZMxDRg6gWsVLeYFL@xxxxxxxxxx
