On Wed, Jan 14, 2004 at 03:24:54PM +1100, matthew green wrote:
   
   > i've been looking at fixing up the remaining issues with compat/netbsd32.
   > one of them is ktrace/kdump.  currently this doesn't work very well.. there
   > is little interpretation of arguments, the arguments are wrong in some
   > cases anyway, etc..
   > 
   > i have patches to kdump that allow 32 bit apps on 64 bit kernels to be
   > traced - but so far only with the 64 bit kdump.  there are multiple cases
   > that could be supported but i think something like this should work:
   
   Did that break?

did what break?  i added code to kdump(1) to handle 32 bit argument
types... it works fine, given a 64 bit kdump(1).  though i think i
may be able to change this to require fewer changes...
   
   I don't think trying to get a 32-bit ktrace/kdump working on 64-bit
   kernels is worth the effort.

after playing around with it a bit last night, i've come to that
conclusion as well... i think the best idea is to basically make
kdump(1) just work.  the hard part is detection of the format..

i've been playing with a version record, but it's still awfully
hard because of the variable size of "struct ktr_header" (even
though userland _doesn't use_ the variable part!  oh the pain.)


.mrg.