procyberian/gcc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

[ARM] Remove remaining Neon DImode support

Browse source 
Remove the remaining Neon adddi3, subdi3 and negdi2 patterns.  As a result
adddi3, subdi3 and negdi2 can now always be expanded early irrespectively of
whether Neon is available.  Also expand the extenddi patterns at the same
time.  Several Neon arch attributes are no longer used and removed.

Code generation is improved in all cases, saving another 400-500 instructions
from the PR77308 testcase (total improvement is over 1700 instructions with
-mcpu=cortex-a57 -O2).

Bootstrap & regress OK on arm-none-linux-gnueabihf --with-cpu=cortex-a57

    gcc/
	* config/arm/arm.md (neon_for_64bits): Remove.
	(avoid_neon_for_64bits): Remove.
	(arm_adddi3): Always split early.
	(arm_subdi3): Always split early.
	(negdi2): Remove Neon expansion.
	(split zero_extend): Split before reload.
	(split sign_extend): Split before reload.

From-SVN: r274825
This commit is contained in:
Wilco Dijkstra 2019-08-22 15:19:58 +00:00 • committed by Wilco Dijkstra
parent 5c7c6c5fc1
commit ef27f40f48
3 changed files with 32 additions and 157 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
gcc/ChangeLog
View file

@ -1,3 +1,13 @@
2019-08-22 Wilco Dijkstra <wdijkstr@arm.com>
* config/arm/arm.md (neon_for_64bits): Remove.
(avoid_neon_for_64bits): Remove.
(arm_adddi3): Always split early.
(arm_subdi3): Always split early.
(negdi2): Remove Neon expansion.
(split zero_extend): Split before reload.
(split sign_extend): Split before reload.
2019-08-22 Wilco Dijkstra <wdijkstr@arm.com>
* config/arm/iterators.md (qhs_extenddi_cstr): Update.

90
gcc/config/arm/arm.md
View file

@ -125,7 +125,7 @@
; arm_arch6. "v6t2" for Thumb-2 with arm_arch6 and "v8mb" for ARMv8-M
; Baseline. This attribute is used to compute attribute "enabled",
; use type "any" to enable an alternative in all cases.
(define_attr "arch" "any,a,t,32,t1,t2,v6,nov6,v6t2,v8mb,neon_for_64bits,avoid_neon_for_64bits,iwmmxt,iwmmxt2,armv6_or_vfpv3,neon"
(define_attr "arch" "any,a,t,32,t1,t2,v6,nov6,v6t2,v8mb,iwmmxt,iwmmxt2,armv6_or_vfpv3,neon"
(const_string "any"))
(define_attr "arch_enabled" "no,yes"
@ -168,16 +168,6 @@
(match_test "TARGET_THUMB1 && arm_arch8"))
(const_string "yes")
(and (eq_attr "arch" "avoid_neon_for_64bits")
(match_test "TARGET_NEON")
(not (match_test "TARGET_PREFER_NEON_64BITS")))
(const_string "yes")
(and (eq_attr "arch" "neon_for_64bits")
(match_test "TARGET_NEON")
(match_test "TARGET_PREFER_NEON_64BITS"))
(const_string "yes")
(and (eq_attr "arch" "iwmmxt2")
(match_test "TARGET_REALLY_IWMMXT2"))
(const_string "yes")
@ -450,13 +440,8 @@
(clobber (reg:CC CC_REGNUM))])]
"TARGET_EITHER"
"
if (TARGET_THUMB1)
{
if (!REG_P (operands[1]))
operands[1] = force_reg (DImode, operands[1]);
if (!REG_P (operands[2]))
operands[2] = force_reg (DImode, operands[2]);
}
if (TARGET_THUMB1 && !REG_P (operands[2]))
operands[2] = force_reg (DImode, operands[2]);
"
)
@ -465,9 +450,9 @@
(plus:DI (match_operand:DI 1 "arm_general_register_operand" "%0, 0, r, 0, r")
(match_operand:DI 2 "arm_general_adddi_operand" "r, 0, r, Dd, Dd")))
(clobber (reg:CC CC_REGNUM))]
"TARGET_32BIT && !TARGET_NEON"
"TARGET_32BIT"
"#"
"TARGET_32BIT && ((!TARGET_NEON && !TARGET_IWMMXT) || reload_completed)"
"TARGET_32BIT"
[(parallel [(set (reg:CC_C CC_REGNUM)
(compare:CC_C (plus:SI (match_dup 1) (match_dup 2))
(match_dup 1)))
@ -1304,24 +1289,16 @@
(clobber (reg:CC CC_REGNUM))])]
"TARGET_EITHER"
"
if (TARGET_THUMB1)
{
if (!REG_P (operands[1]))
operands[1] = force_reg (DImode, operands[1]);
if (!REG_P (operands[2]))
operands[2] = force_reg (DImode, operands[2]);
}
"
)
")
(define_insn_and_split "*arm_subdi3"
[(set (match_operand:DI 0 "arm_general_register_operand" "=&r,&r,&r")
(minus:DI (match_operand:DI 1 "arm_general_register_operand" "0,r,0")
(match_operand:DI 2 "arm_general_register_operand" "r,0,0")))
(clobber (reg:CC CC_REGNUM))]
"TARGET_32BIT && !TARGET_NEON"
"TARGET_32BIT"
"#" ; "subs\\t%Q0, %Q1, %Q2\;sbc\\t%R0, %R1, %R2"
"&& (!TARGET_IWMMXT || reload_completed)"
"TARGET_32BIT"
[(parallel [(set (reg:CC CC_REGNUM)
(compare:CC (match_dup 1) (match_dup 2)))
(set (match_dup 0) (minus:SI (match_dup 1) (match_dup 2)))])
@ -4184,13 +4161,6 @@
(neg:DI (match_operand:DI 1 "s_register_operand")))
(clobber (reg:CC CC_REGNUM))])]
"TARGET_EITHER"
{
if (TARGET_NEON)
{
emit_insn (gen_negdi2_neon (operands[0], operands[1]));
DONE;
}
}
)
;; The constraints here are to prevent a *partial* overlap (where %Q0 == %R1).
@ -4202,7 +4172,7 @@
"TARGET_32BIT"
"#" ; rsbs %Q0, %Q1, #0; rsc %R0, %R1, #0 (ARM)
; negs %Q0, %Q1 ; sbc %R0, %R1, %R1, lsl #1 (Thumb-2)
"&& reload_completed"
"TARGET_32BIT"
[(parallel [(set (reg:CC CC_REGNUM)
(compare:CC (const_int 0) (match_dup 1)))
(set (match_dup 0) (minus:SI (const_int 0) (match_dup 1)))])
@ -4734,25 +4704,17 @@
(define_split
[(set (match_operand:DI 0 "s_register_operand" "")
(zero_extend:DI (match_operand 1 "nonimmediate_operand" "")))]
"TARGET_32BIT && reload_completed && !IS_VFP_REGNUM (REGNO (operands[0]))"
"TARGET_32BIT"
[(set (match_dup 0) (match_dup 1))]
{
rtx lo_part = gen_lowpart (SImode, operands[0]);
machine_mode src_mode = GET_MODE (operands[1]);
if (REG_P (operands[0])
&& !reg_overlap_mentioned_p (operands[0], operands[1]))
emit_clobber (operands[0]);
if (!REG_P (lo_part) || src_mode != SImode
|| !rtx_equal_p (lo_part, operands[1]))
{
if (src_mode == SImode)
emit_move_insn (lo_part, operands[1]);
else
emit_insn (gen_rtx_SET (lo_part,
gen_rtx_ZERO_EXTEND (SImode, operands[1])));
operands[1] = lo_part;
}
if (src_mode == SImode)
emit_move_insn (lo_part, operands[1]);
else
emit_insn (gen_rtx_SET (lo_part,
gen_rtx_ZERO_EXTEND (SImode, operands[1])));
operands[0] = gen_highpart (SImode, operands[0]);
operands[1] = const0_rtx;
})
@ -4760,26 +4722,18 @@
(define_split
[(set (match_operand:DI 0 "s_register_operand" "")
(sign_extend:DI (match_operand 1 "nonimmediate_operand" "")))]
"TARGET_32BIT && reload_completed && !IS_VFP_REGNUM (REGNO (operands[0]))"
"TARGET_32BIT"
[(set (match_dup 0) (ashiftrt:SI (match_dup 1) (const_int 31)))]
{
rtx lo_part = gen_lowpart (SImode, operands[0]);
machine_mode src_mode = GET_MODE (operands[1]);
if (REG_P (operands[0])
&& !reg_overlap_mentioned_p (operands[0], operands[1]))
emit_clobber (operands[0]);
if (!REG_P (lo_part) || src_mode != SImode
|| !rtx_equal_p (lo_part, operands[1]))
{
if (src_mode == SImode)
emit_move_insn (lo_part, operands[1]);
else
emit_insn (gen_rtx_SET (lo_part,
gen_rtx_SIGN_EXTEND (SImode, operands[1])));
operands[1] = lo_part;
}
if (src_mode == SImode)
emit_move_insn (lo_part, operands[1]);
else
emit_insn (gen_rtx_SET (lo_part,
gen_rtx_SIGN_EXTEND (SImode, operands[1])));
operands[1] = lo_part;
operands[0] = gen_highpart (SImode, operands[0]);
})

89
gcc/config/arm/neon.md
View file

@ -527,32 +527,6 @@
(const_string "neon_add<q>")))]
)
(define_insn "adddi3_neon"
[(set (match_operand:DI 0 "s_register_operand" "=w,?&r,?&r,?w,?&r,?&r,?&r")
(plus:DI (match_operand:DI 1 "s_register_operand" "%w,0,0,w,r,0,r")
(match_operand:DI 2 "arm_adddi_operand" "w,r,0,w,r,Dd,Dd")))
(clobber (reg:CC CC_REGNUM))]
"TARGET_NEON"
{
switch (which_alternative)
{
case 0: /* fall through */
case 3: return "vadd.i64\t%P0, %P1, %P2";
case 1: return "#";
case 2: return "#";
case 4: return "#";
case 5: return "#";
case 6: return "#";
default: gcc_unreachable ();
}
}
[(set_attr "type" "neon_add,multiple,multiple,neon_add,\
multiple,multiple,multiple")
(set_attr "conds" "*,clob,clob,*,clob,clob,clob")
(set_attr "length" "*,8,8,*,8,8,8")
(set_attr "arch" "neon_for_64bits,*,*,avoid_neon_for_64bits,*,*,*")]
)
(define_insn "*sub<mode>3_neon"
[(set (match_operand:VDQ 0 "s_register_operand" "=w")
(minus:VDQ (match_operand:VDQ 1 "s_register_operand" "w")
@ -587,29 +561,6 @@
[(set_attr "type" "neon_sub<q>")]
)
(define_insn "subdi3_neon"
[(set (match_operand:DI 0 "s_register_operand" "=w,?&r,?&r,?&r,?w")
(minus:DI (match_operand:DI 1 "s_register_operand" "w,0,r,0,w")
(match_operand:DI 2 "s_register_operand" "w,r,0,0,w")))
(clobber (reg:CC CC_REGNUM))]
"TARGET_NEON"
{
switch (which_alternative)
{
case 0: /* fall through */
case 4: return "vsub.i64\t%P0, %P1, %P2";
case 1: /* fall through */
case 2: /* fall through */
case 3: return "subs\\t%Q0, %Q1, %Q2\;sbc\\t%R0, %R1, %R2";
default: gcc_unreachable ();
}
}
[(set_attr "type" "neon_sub,multiple,multiple,multiple,neon_sub")
(set_attr "conds" "*,clob,clob,clob,*")
(set_attr "length" "*,8,8,8,*")
(set_attr "arch" "neon_for_64bits,*,*,*,avoid_neon_for_64bits")]
)
(define_insn "*mul<mode>3_neon"
[(set (match_operand:VDQW 0 "s_register_operand" "=w")
(mult:VDQW (match_operand:VDQW 1 "s_register_operand" "w")
@ -886,46 +837,6 @@
(const_string "neon_neg<q>")))]
)
(define_insn "negdi2_neon"
[(set (match_operand:DI 0 "s_register_operand" "=&w, w,r,&r")
(neg:DI (match_operand:DI 1 "s_register_operand" " w, w,0, r")))
(clobber (match_scratch:DI 2 "= X,&w,X, X"))
(clobber (reg:CC CC_REGNUM))]
"TARGET_NEON"
"#"
[(set_attr "length" "8")
(set_attr "type" "multiple")]
)
; Split negdi2_neon for vfp registers
(define_split
[(set (match_operand:DI 0 "s_register_operand" "")
(neg:DI (match_operand:DI 1 "s_register_operand" "")))
(clobber (match_scratch:DI 2 ""))
(clobber (reg:CC CC_REGNUM))]
"TARGET_NEON && reload_completed && IS_VFP_REGNUM (REGNO (operands[0]))"
[(set (match_dup 2) (const_int 0))
(parallel [(set (match_dup 0) (minus:DI (match_dup 2) (match_dup 1)))
(clobber (reg:CC CC_REGNUM))])]
{
if (!REG_P (operands[2]))
operands[2] = operands[0];
}
)
; Split negdi2_neon for core registers
(define_split
[(set (match_operand:DI 0 "s_register_operand" "")
(neg:DI (match_operand:DI 1 "s_register_operand" "")))
(clobber (match_scratch:DI 2 ""))
(clobber (reg:CC CC_REGNUM))]
"TARGET_32BIT && reload_completed
&& arm_general_register_operand (operands[0], DImode)"
[(parallel [(set (match_dup 0) (neg:DI (match_dup 1)))
(clobber (reg:CC CC_REGNUM))])]
""
)
(define_insn "<absneg_str><mode>2"
[(set (match_operand:VH 0 "s_register_operand" "=w")
(ABSNEG:VH (match_operand:VH 1 "s_register_operand" "w")))]