> -----Original Message----- > From: Jan Beulich [mailto:jbeul...@suse.com] > Sent: 28 November 2017 11:26 > To: Paul Durrant <paul.durr...@citrix.com> > Cc: Julien Grall <julien.gr...@arm.com>; Andrew Cooper > <andrew.coop...@citrix.com>; xen-devel <xen- > de...@lists.xenproject.org> > Subject: RE: [PATCH] x86/HVM: fix interaction between internal and extern > emulation > > >>> On 28.11.17 at 12:06, <paul.durr...@citrix.com> wrote: > > Yes, it appears that mmio_retry is only set when the underlying emulation > > returned X86EMUL_OKAY but not all reps were completed. If the > underlying > > emulation did not return X86EMUL_RETRY then I can't figure out why > > vio->io_completion should need to be set to anything other than > > HVMIO_no_completion since any other return value indicates there should > be > > nothing pending. > > So am I getting it right that you're suggesting to remove the > mmio_retry part of the condition in hvm_emulate_one_insn()? > That looks like it might work (I was previously only considering > to get rid of mmio_retry altogether, and that didn't look like a > viable route).
Yes, that's what I'm suggesting. I really can't see why it is needed. It could have been a mistake in my original patches or a semantic change in a subsequent patch, but it certainly looks wrong in current context. Andrew has just sent me his xtf repro so I'll give the change a go with that. Paul Paul > > Jan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xenproject.org https://lists.xenproject.org/mailman/listinfo/xen-devel
