Bill Allombert on Wed, 01 Feb 2006 00:18:24 +0100

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: possible solution for the asm kernel breakage

On Tue, Jan 31, 2006 at 10:49:15AM -0500, Igor Schein wrote:
> On Sun, Jan 29, 2006 at 06:04:40PM +0100, Bill Allombert wrote:
> Bill,
> can you please define for the potential patch testers what the
> symptoms of the problem are and the scope of testing?  ./Configure
> --kernel=none works fine on amd64 with CVS sources, so how would the
> tester know whether the patch fixes anything.  Or is the goal here to
> make sure the patch doesn't break anything?  

Well, without the patch, PARI will not build on ppc with the asm
kernel, which is already a bug.

Whether the patch work is a matter of running nm -D and
check the resulting symbol list against another build (in particular
nm -D | grep divll should return something).

However Karim is apparently considering another solution: a
single kernel.c which will specifically include the level0 and level1
kernel rather than the LEVEL0/LEVEL1 symbol tricks.

However there is the issue of asm level0 non-inline kernels.