Hi All,
About 13 years ago the reg-to-reg patterns were split up, before that time
output_move_double could actually handle this case.
After the split was done most patterns were updated except for *neon_mov
which incorrectly retained reg,reg as a valid alternative.
However output_move_double can not handle this and simply returns ""
and asserts.
This pattern is essentially dead and I'm removing it for clarity.
Regtested on armeb-none-eabi and no regressions.
Bootstrapped on arm-none-linux-gnueabihf and no issues.
Ok for trunk?
Thanks,
Tamar
gcc/
2018-07-23 Tamar Christina <tamar.christ...@arm.com>
* config/arm/neon.md (*neon_mov<mode>): Remove reg-to-reg alternative.
--
diff --git a/gcc/config/arm/neon.md b/gcc/config/arm/neon.md
index 57df17fe18dfeaa82e890fd339e2104ad27ee13b..55fc9a02ba526538cbc51e07e7d4add1a9c38002 100644
--- a/gcc/config/arm/neon.md
+++ b/gcc/config/arm/neon.md
@@ -25,9 +25,9 @@
(define_insn "*neon_mov<mode>"
[(set (match_operand:VDX 0 "nonimmediate_operand"
- "=w,Un,w, w, ?r,?w,?r,?r, ?Us")
+ "=w,Un,w, w, ?r,?w,?r, ?Us")
(match_operand:VDX 1 "general_operand"
- " w,w, Dn,Uni, w, r, r, Usi,r"))]
+ " w,w, Dn,Uni, w, r, Usi,r"))]
"TARGET_NEON
&& (register_operand (operands[0], <MODE>mode)
|| register_operand (operands[1], <MODE>mode))"
@@ -61,12 +61,12 @@
}
}
[(set_attr "type" "neon_move<q>,neon_store1_1reg,neon_move<q>,\
- neon_load1_1reg, neon_to_gp<q>,neon_from_gp<q>,mov_reg,\
+ neon_load1_1reg, neon_to_gp<q>,neon_from_gp<q>,\
neon_load1_2reg, neon_store1_2reg")
- (set_attr "length" "4,4,4,4,4,4,8,8,8")
- (set_attr "arm_pool_range" "*,*,*,1020,*,*,*,1020,*")
- (set_attr "thumb2_pool_range" "*,*,*,1018,*,*,*,1018,*")
- (set_attr "neg_pool_range" "*,*,*,1004,*,*,*,1004,*")])
+ (set_attr "length" "4,4,4,4,4,4,8,8")
+ (set_attr "arm_pool_range" "*,*,*,1020,*,*,1020,*")
+ (set_attr "thumb2_pool_range" "*,*,*,1018,*,*,1018,*")
+ (set_attr "neg_pool_range" "*,*,*,1004,*,*,1004,*")])
(define_insn "*neon_mov<mode>"
[(set (match_operand:VQXMOV 0 "nonimmediate_operand"
