On 17 April 2011 18:11, Richard Henderson <r...@twiddle.net> wrote: > On 04/17/2011 10:03 AM, Peter Maydell wrote: >> On 17 April 2011 17:14, Richard Henderson <r...@twiddle.net> wrote: >>> pc-bios/palcode-sx164 | Bin 0 -> 107781 bytes >> >> My personal opinion is that if we have the binary blob in qemu git >> we should also have its corresponding source code.
> How does this differ from e.g. SeaBios? I think that ought to follow the same rule: 'git clone' and making a release tarball from a git repo should both either (a) not ship a binary blob or (b) include the sources which made up that binary blob I know this isn't exactly where we are today; to some extent I'm trying to see if anybody else agrees with this position... > We don't really have any configury for cross-compilation of the > bios bits. Yes, this is a difficult problem (which also holds for test cases); I'm not objecting to having the blobs in git and in tarballs, I just want the sources too. -- PMM