This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Shrink struct basic_block, get rid of NOTE_INSN_DISABLE_SCHED_OF_BLOCK


On Tue, Aug 24, 2004 at 05:54:13PM -0700, Mark Mitchell wrote:
> What would you think about forcing two blocks to be in the same 
> partition of there are no registers free across the edge between them?  

Does not solve the problem.  You have to expose all symbolic
references before the target exposes GP references.  Which
often happens in the first post-reload split, or in peephole2.

If you try to do add symbolic references too late, you may 
not have a live GP register, and if you don't have that, you
can't load an address even if you've got free registers.


r~


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]