On Fri, Jun 19, 2015 at 01:39:53PM +0100, Michael Brown wrote: > On 19/06/15 13:33, Matt Fleming wrote: > >>However, there doesn't seem to be to be much value in breaking the > >>ability of EDK2-based builds to load such kernels. In particular, this > >>is likely to mean that future versions of OVMF will be unable to boot > >>some current Linux distributions. > > > >Yeah, OK, that's a very good point. > > > >My concern here is that we have enough divergence between different UEFI > >implementations as it is, so if we get a chance to reduce that (by > >making things fail consistently across platforms) I'm all for it. > > > >Perhaps what we should do is hide these checks inside a DEBUG/DIAGNOSTIC > >guard? That way I can enable it on my machines when verifying new EFI > >kernel patches, and at the same time we won't go breaking people's > >working setups. > > Sounds good to me!
Sounds good to me as well. I did not actually have a problem on a system refusing to load the kernel, rather I noticed it while using another application that was checking PE/COFF headers. ------------------------------------------------------------------------------ _______________________________________________ edk2-devel mailing list edk2-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/edk2-devel
