My previous change to the Cortex-A53 scheduler resulted in a 13% regression on a...

My previous change to the Cortex-A53 scheduler resulted in a 13% regression on a
proprietary benchmark.  This turned out to be due to non-optimal scheduling of int
to float conversions.  This patch separates int to FP transfers from int to float
conversions based on experiments to determine the best schedule.  As a result of
these tweaks the performance of the benchmark improves by 20%.

    gcc/
	* config/arm/cortex-a53.md: Add bypasses for
	cortex_a53_r2f_cvt.
	(cortex_a53_r2f): Only use for transfers.
	(cortex_a53_f2r): Likewise.
	(cortex_a53_r2f_cvt): Add reservation for conversions.
	(cortex_a53_f2r_cvt): Likewise.

From-SVN: r244322

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index bdf19ee..d0aae19 100644 (file)
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,12 @@
+2017-01-11  Wilco Dijkstra  <wdijkstr@arm.com>
+
+       * config/arm/cortex-a53.md: Add bypasses for
+       cortex_a53_r2f_cvt.
+       (cortex_a53_r2f): Only use for transfers.
+       (cortex_a53_f2r): Likewise.
+       (cortex_a53_r2f_cvt): Add reservation for conversions.
+       (cortex_a53_f2r_cvt): Likewise.
+
 2017-01-11  Tamar Christina  <tamar.christina@arm.com>
 
        * config/arm/arm_neon.h: Add __artificial__ and gnu_inline

diff --git a/gcc/config/arm/cortex-a53.md b/gcc/config/arm/cortex-a53.md
index fbec5df..7cf5fc5 100644 (file)
--- a/gcc/config/arm/cortex-a53.md
+++ b/gcc/config/arm/cortex-a53.md
@@ -260,9 +260,18 @@
                 "cortex_a53_r2f")
 
 (define_bypass 1 "cortex_a53_mul,
-                 cortex_a53_load*"
+                 cortex_a53_load1,
+                 cortex_a53_load2"
                 "cortex_a53_r2f")
 
+(define_bypass 2 "cortex_a53_alu*"
+                "cortex_a53_r2f_cvt")
+
+(define_bypass 3 "cortex_a53_mul,
+                 cortex_a53_load1,
+                 cortex_a53_load2"
+                "cortex_a53_r2f_cvt")
+
 ;; Model flag forwarding to branches.
 
 (define_bypass 0 "cortex_a53_alu*,cortex_a53_shift*"
@@ -522,19 +531,25 @@
 ;; Floating-point to/from core transfers.
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 
-(define_insn_reservation "cortex_a53_r2f" 6
+(define_insn_reservation "cortex_a53_r2f" 2
+  (and (eq_attr "tune" "cortexa53")
+       (eq_attr "type" "f_mcr,f_mcrr"))
+  "cortex_a53_slot_any,cortex_a53_fp_alu")
+
+(define_insn_reservation "cortex_a53_f2r" 4
+  (and (eq_attr "tune" "cortexa53")
+       (eq_attr "type" "f_mrc,f_mrrc"))
+  "cortex_a53_slot_any,cortex_a53_fp_alu")
+
+(define_insn_reservation "cortex_a53_r2f_cvt" 4
   (and (eq_attr "tune" "cortexa53")
-       (eq_attr "type" "f_mcr,f_mcrr,f_cvti2f,
-                       neon_from_gp, neon_from_gp_q"))
-  "cortex_a53_slot_any,cortex_a53_store,
-   nothing,cortex_a53_fp_alu")
+       (eq_attr "type" "f_cvti2f, neon_from_gp, neon_from_gp_q"))
+  "cortex_a53_slot_any,cortex_a53_fp_alu")
 
-(define_insn_reservation "cortex_a53_f2r" 6
+(define_insn_reservation "cortex_a53_f2r_cvt" 5
   (and (eq_attr "tune" "cortexa53")
-       (eq_attr "type" "f_mrc,f_mrrc,f_cvtf2i,
-                       neon_to_gp, neon_to_gp_q"))
-  "cortex_a53_slot_any,cortex_a53_fp_alu,
-   nothing,cortex_a53_store")
+       (eq_attr "type" "f_cvtf2i, neon_to_gp, neon_to_gp_q"))
+  "cortex_a53_slot_any,cortex_a53_fp_alu")
 
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;; Floating-point flag transfer.