On 07/09/15 10:27, Ramana Radhakrishnan wrote: > > On 24/07/15 11:55, Kyrill Tkachov wrote: >> commit d562629e36ba013b8f77956a74139330d191bc30 >> Author: Kyrylo Tkachov >> Date: Fri Jul 17 16:30:01 2015 +0100 >> >> [ARM][3/3] Expand mod by power of 2 >> >> diff --git a/gcc/config/arm/arm.c b/gcc/config/arm/arm.c >> index e1bc727..6ade07c 100644 >> --- a/gcc/config/arm/arm.c >> +++ b/gcc/config/arm/arm.c >> @@ -9556,6 +9556,22 @@ arm_new_rtx_costs (rtx x, enum rtx_code code, enum rtx_code outer_code, >> >> case MOD: >> case UMOD: >> + /* MOD by a power of 2 can be expanded as: >> + rsbs r1, r0, #0 >> + and r0, r0, #(n - 1) >> + and r1, r1, #(n - 1) >> + rsbpl r0, r1, #0. */ >> + if (code == MOD >> + && CONST_INT_P (XEXP (x, 1)) >> + && exact_log2 (INTVAL (XEXP (x, 1))) > 0 >> + && mode == SImode) >> + { >> + *cost += COSTS_N_INSNS (3) >> + + 2 * extra_cost->alu.logical >> + + extra_cost->alu.arith; >> + return true; >> + } >> + >> *cost = LIBCALL_COST (2); >> return false; /* All arguments must be in registers. */ >> >> diff --git a/gcc/config/arm/arm.md b/gcc/config/arm/arm.md >> index f341109..8301648 100644 >> --- a/gcc/config/arm/arm.md >> +++ b/gcc/config/arm/arm.md >> @@ -1229,7 +1229,7 @@ (define_peephole2 >> "" >> ) >> >> -(define_insn "*subsi3_compare0" >> +(define_insn "subsi3_compare0" >> [(set (reg:CC_NOOV CC_REGNUM) >> (compare:CC_NOOV >> (minus:SI (match_operand:SI 1 "arm_rhs_operand" "r,r,I") >> @@ -2158,7 +2158,7 @@ (define_expand "andsi3" >> ) >> >> ; ??? Check split length for Thumb-2 >> -(define_insn_and_split "*arm_andsi3_insn" >> +(define_insn_and_split "arm_andsi3_insn" >> [(set (match_operand:SI 0 "s_register_operand" "=r,l,r,r,r") >> (and:SI (match_operand:SI 1 "s_register_operand" "%r,0,r,r,r") >> (match_operand:SI 2 "reg_or_int_operand" "I,l,K,r,?n")))] >> @@ -11105,6 +11105,78 @@ (define_expand "thumb_legacy_rev" >> "" >> ) > This shouldn't be necessary - you are just adding another interface to produce an and insn. I see what you mean. Ok, I'll drop this. > >> >> +;; ARM-specific expansion of signed mod by power of 2 >> +;; using conditional negate. >> +;; For r0 % n where n is a power of 2 produce: >> +;; rsbs r1, r0, #0 >> +;; and r0, r0, #(n - 1) >> +;; and r1, r1, #(n - 1) >> +;; rsbpl r0, r1, #0 >> + >> +(define_expand "modsi3" >> + [(match_operand:SI 0 "register_operand" "") >> + (match_operand:SI 1 "register_operand" "") >> + (match_operand:SI 2 "const_int_operand" "")] >> + "TARGET_32BIT" >> + { >> + HOST_WIDE_INT val = INTVAL (operands[2]); >> + >> + if (val <= 0 >> + || exact_log2 (INTVAL (operands[2])) <= 0 >> + || !const_ok_for_arm (INTVAL (operands[2]) - 1)) >> + FAIL; >> + >> + rtx mask = GEN_INT (val - 1); >> + >> + /* In the special case of x0 % 2 we can do the even shorter: >> + cmp r0, #0 >> + and r0, r0, #1 >> + rsblt r0, r0, #0. */ >> + >> + if (val == 2) >> + { >> + rtx cc_reg = gen_rtx_REG (CCmode, CC_REGNUM); >> + rtx cond = gen_rtx_LT (SImode, cc_reg, const0_rtx); >> + >> + emit_insn (gen_rtx_SET (cc_reg, >> + gen_rtx_COMPARE (CCmode, operands[1], const0_rtx))); >> + >> + rtx masked = gen_reg_rtx (SImode); >> + emit_insn (gen_arm_andsi3_insn (masked, operands[1], mask)); > Use emit_insn (gen_andsi3 (masked, operands[1], mask) instead and likewise below. Ok, done that. A side effect of this is that since the andsi3 expander handles any reg_or_int we can catch more masks this way. Also, due to the way the andsi3 expander is written, for the mask 255 it will generate a zero_extend instead of an and. This may or may not be optimal on some cores but perhaps we should look at the andsi3 expander to make it more robust as a separate task. > > >> + emit_move_insn (operands[0], >> + gen_rtx_IF_THEN_ELSE (SImode, cond, >> + gen_rtx_NEG (SImode, >> + masked), >> + masked)); >> + DONE; >> + } >> + >> + rtx neg_op = gen_reg_rtx (SImode); >> + rtx_insn *insn = emit_insn (gen_subsi3_compare0 (neg_op, const0_rtx, >> + operands[1])); >> + >> + /* Extract the condition register and mode. */ >> + rtx cmp = XVECEXP (PATTERN (insn), 0, 0); >> + rtx cc_reg = SET_DEST (cmp); >> + rtx cond = gen_rtx_GE (SImode, cc_reg, const0_rtx); >> + >> + emit_insn (gen_arm_andsi3_insn (operands[0], operands[1], mask)); >> + >> + rtx masked_neg = gen_reg_rtx (SImode); >> + emit_insn (gen_arm_andsi3_insn (masked_neg, neg_op, mask)); >> + >> + /* We want a conditional negate here, but emitting COND_EXEC rtxes >> + during expand does not always work. Do an IF_THEN_ELSE instead. */ >> + emit_move_insn (operands[0], >> + gen_rtx_IF_THEN_ELSE (SImode, cond, >> + gen_rtx_NEG (SImode, masked_neg), >> + operands[0])); >> + >> + >> + DONE; >> + } >> +) >> + >> (define_expand "bswapsi2" >> [(set (match_operand:SI 0 "s_register_operand" "=r") >> (bswap:SI (match_operand:SI 1 "s_register_operand" "r")))] >> diff --git a/gcc/testsuite/gcc.target/aarch64/mod_2.c b/gcc/testsuite/gcc.target/aarch64/mod_2.c >> new file mode 100644 >> index 0000000..2645c18 >> --- /dev/null >> +++ b/gcc/testsuite/gcc.target/aarch64/mod_2.c >> @@ -0,0 +1,7 @@ >> +/* { dg-do compile } */ >> +/* { dg-options "-O2 -mcpu=cortex-a57 -save-temps" } */ >> + >> +#include "mod_2.x" >> + >> +/* { dg-final { scan-assembler "csneg\t\[wx\]\[0-9\]*" } } */ >> +/* { dg-final { scan-assembler-times "and\t\[wx\]\[0-9\]*" 1 } } */ >> diff --git a/gcc/testsuite/gcc.target/aarch64/mod_2.x b/gcc/testsuite/gcc.target/aarch64/mod_2.x >> new file mode 100644 >> index 0000000..2b079a4 >> --- /dev/null >> +++ b/gcc/testsuite/gcc.target/aarch64/mod_2.x >> @@ -0,0 +1,5 @@ >> +int >> +f (int x) >> +{ >> + return x % 2; >> +} >> diff --git a/gcc/testsuite/gcc.target/aarch64/mod_256.c b/gcc/testsuite/gcc.target/aarch64/mod_256.c >> new file mode 100644 >> index 0000000..567332c >> --- /dev/null >> +++ b/gcc/testsuite/gcc.target/aarch64/mod_256.c >> @@ -0,0 +1,6 @@ >> +/* { dg-do compile } */ >> +/* { dg-options "-O2 -mcpu=cortex-a57 -save-temps" } */ >> + >> +#include "mod_256.x" >> + >> +/* { dg-final { scan-assembler "csneg\t\[wx\]\[0-9\]*" } } */ >> diff --git a/gcc/testsuite/gcc.target/aarch64/mod_256.x b/gcc/testsuite/gcc.target/aarch64/mod_256.x >> new file mode 100644 >> index 0000000..c1de42c >> --- /dev/null >> +++ b/gcc/testsuite/gcc.target/aarch64/mod_256.x >> @@ -0,0 +1,5 @@ >> +int >> +f (int x) >> +{ >> + return x % 256; >> +} >> diff --git a/gcc/testsuite/gcc.target/arm/mod_2.c b/gcc/testsuite/gcc.target/arm/mod_2.c >> new file mode 100644 >> index 0000000..93017a1 >> --- /dev/null >> +++ b/gcc/testsuite/gcc.target/arm/mod_2.c >> @@ -0,0 +1,8 @@ >> +/* { dg-do compile } */ >> +/* { dg-require-effective-target arm32 } */ >> +/* { dg-options "-O2 -mcpu=cortex-a57 -save-temps" } */ >> + >> +#include "../aarch64/mod_2.x" >> + >> +/* { dg-final { scan-assembler "rsblt\tr\[0-9\]*" } } */ >> +/* { dg-final { scan-assembler-times "and\tr\[0-9\].*1" 1 } } */ >> diff --git a/gcc/testsuite/gcc.target/arm/mod_256.c b/gcc/testsuite/gcc.target/arm/mod_256.c >> new file mode 100644 >> index 0000000..92ab05a >> --- /dev/null >> +++ b/gcc/testsuite/gcc.target/arm/mod_256.c >> @@ -0,0 +1,8 @@ >> +/* { dg-do compile } */ >> +/* { dg-require-effective-target arm32 } */ >> +/* { dg-options "-O2 -mcpu=cortex-a57 -save-temps" } */ >> + >> +#include "../aarch64/mod_256.x" >> + >> +/* { dg-final { scan-assembler "rsbpl\tr\[0-9\]*" } } */ >> +/* { dg-final { scan-assembler "and\tr\[0-9\].*255" } } */ > > OK with those changes if no regressions. Thanks, I'm attaching the updated patch. I'll commit it when the aarch64 one is ok'd as the tests are partly shared with this patch. Kyrill 2015-09-08 Kyrylo Tkachov * config/arm/arm.md (*subsi3_compare0): Rename to... (subsi3_compare0): ... This. (modsi3): New define_expand. * config/arm/arm.c (arm_new_rtx_costs, MOD case): Handle case when operand is power of 2. 2015-09-08 Kyrylo Tkachov * gcc.target/aarch64/mod_2.x: New file. * gcc.target/aarch64/mod_256.x: Likewise. * gcc.target/arm/mod_2.c: New test. * gcc.target/arm/mod_256.c: Likewise. * gcc.target/aarch64/mod_2.c: Likewise. * gcc.target/aarch64/mod_256.c: Likewise. > Ramana >