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: [patch] Add DFA-based pipeline descriptions for MIPS 24K core.


On Wed, 2005-05-04 at 09:18 -0700, Eric Christopher wrote:
> >; The 24kx is a 24k configuered with 1:1 cpu and fpu, so use
> >; the unscaled timings
> 
> I'm not seeing this comment reflected in the scheduling code.

If I understood your question correctly, you are trying to find where
the difference is in the code? 
Then I guess this is what you are looking for?

for the 24k

(define_insn_reservation "r24k_fdiv_df" 64 
  (and (eq_attr "cpu" "24k")
       (and (eq_attr "type" "fdiv,fsqrt")
	    (eq_attr "mode" "DF")))
  "r24k_fpu_iss,(r24k_fpu_arith*56)")

and for the 24kx,

(define_insn_reservation "r24kx_fdiv_df" 32 
  (and (eq_attr "cpu" "24kx")
       (and (eq_attr "type" "fdiv,fsqrt")
	    (eq_attr "mode" "DF")))
  "r24kx_fpu_iss,(r24k_fpu_arith*28)")

The numbers are simply halve.

David.


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