At Tue, 11 Apr 2000 11:24:02 -0700,
Rob Aagaard <rob@xxxxxxxxxxxxxxxxxxx> wrote:
>
>
> First off, is it possible for anyone outside of SGI to get write access
> to the CVS tree? (I'm not suggesting that I be given write access to the
> CVS tree, I know just enough about the kernel to really break stuff
> fast.)
For the moment source code control is being done via SGI internal
tools. (p tools) The cvs tree is a copy of the internal rcs files from
the p_tools tree. So to answer your question: nobody has write
access to the tree, not even internally.
>
> Secondly, how much i386 specific code is in XFS, I'm very interested in
> getting it ported to Alpha/PPC (the two alternative archs I have at
> home) and am wondering when would be an appropriate time to begin
> porting.
Most of your problems are going to arise from differences in types.
specifically unsigned long and __kernel_addr_t
there are bound to be others.
Dave Miller already managed to get the kernel side compiled on
an Ultra sparc, most of those changes should map to an alpha.
>
>
> But the main reason I sent this message is to thank all of you down at
> SGI, I am looking forward to having XFS on Linux, and I have it
> currently in my system, and a 7G partiton dedicated to it. It made it
> all the way through a kernel compile, but locked up my rm process when I
> was nuking the kernel source.i
Yup... we are looking into that one right now.
I am however, pleasantly supprised that
> this thing compiles consistantly, and was amazed when it actually worked
> (mostly). Thank you again for all of your efforts, and to a job well
> done.
>
> - Rob Aagaard
|