Greetings! "Mike Thomas" <[EMAIL PROTECTED]> writes:
> Hi Camm. > > I just got around to trying your new recompilation facility. > Great! Please be advised it is currently a moving target. > Could you please comment on the recompilation spawned during the build > of GCL as logged in the attached file. > > The recompilation occurs during initialisation and required me to > rearrange the init*_gcl.lsp.in files so that compiler::*cc* was set up > in time for the initialisation stage. > > This does not seem right. > You log looks good up to the assembler warnings. The idea, as I'm sure you know, is that GCL can compile faster code if it knows the arguments and types on function entry and return. Different images can compile different functions in different files during different sessions, then to be loaded into the same image -- hence the only time conflicts can be resolved is at load time, including here in GCL initialization. At some point we will predump the signature hash table to avoid the time consuming recompiles, but right now it serves as a critical cross check. Please let me know if any of this is causing problems. YOu should be able to restore the previous behavior with: (setq compiler::*compiler-auto-proclaim* nil si::*disable-recompile* t) Take care, > Cheers > > Mike Thomas. > -- Camm Maguire [EMAIL PROTECTED] ========================================================================== "The earth is but one country, and mankind its citizens." -- Baha'u'llah _______________________________________________ Gcl-devel mailing list Gcl-devel@gnu.org http://lists.gnu.org/mailman/listinfo/gcl-devel
