On 12/06/2019 13:14, David Woodhouse wrote: > On Wed, 2019-06-12 at 06:00 -0600, Jan Beulich wrote: >>>>> Reported-by: David Woodhouse <d...@amazon.co.uk> >>>> Does this mean there was an actual problem resulting from this code >>>> being there, or simply the observation that this code is all dead? >>>> Clarifying one way or the other by half a sentence would be nice. >>> It was more an observation that when you kexec Xen, it blindly writes >>> into the BDA even when that wasn't set up properly by the tools. >>> >>> Arguably that may have been kexec setup problem more than a Xen problem, >>> but after looking at this code, it is obviously that what Xen was doing >>> definitely wasn't right to do unconditionally. It just so happens that >>> it safe to unconditionally drop the logic, rather than to make it >>> dependant on other system properties. >>> >>> I'm not sure how best to phrase this. >> Maybe "In practice issues with this were observed only with kexec"? > Not sure that's true either, is it? I found *lots* of issues when doing > kexec, and I should resend that series of boot code cleanups — but this > wasn't one of the ones I remember spotting :)
You definitely complained about the BDA on IRC, which is why I started looking, but I'm happy to leave you out of the patch if you'd prefer. ~Andrew _______________________________________________ Xen-devel mailing list Xen-devel@lists.xenproject.org https://lists.xenproject.org/mailman/listinfo/xen-devel
