Martin Husemann writes: | On Wed, Aug 22, 2018 at 02:35:15PM +0000, Christos Zoulas wrote: |> But it was already fixed, why did you break it again?
It was more or less put back to yesterday - what was in asan.h for the !KASAN case is more or less irrelevant, as all the calls are #ifdef'd and never get seen. | Don't think it really was fixed (but maybe I missed some fixes The issue with the types remains. KASAN is only turned on for amd64 (and only for maxv's private builds right now) so it doesn't matter right now, but some decision will be needed. | my last cvs update and local build failure). evbarm 32bit builds did still | fail for me. i386 is working from b5 The buildbots are building now for sources from between maxv's last commit and Christos' later one - if those show any issues, and if Max doesn't fix them, I will take a look. But I don't get to see any of the logs until all 67 builds are done, so it will be many hours yet. kre
