From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 7856) id 6762D385843A; Wed, 6 Jul 2022 16:32:28 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 6762D385843A Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit From: Xi Ruoyao To: gcc-cvs@gcc.gnu.org Subject: [gcc(refs/users/xry111/heads/xry111/loongarch-div-opt)] loongarch: avoid unnecessary sign-extend after 32-bit division X-Act-Checkin: gcc X-Git-Author: Xi Ruoyao X-Git-Refname: refs/users/xry111/heads/xry111/loongarch-div-opt X-Git-Oldrev: 06bf853ae271eb138743d80b8a6085e9bdbe4d25 X-Git-Newrev: f98ca8fbb9188f01532effbed4342f94167a5582 Message-Id: <20220706163228.6762D385843A@sourceware.org> Date: Wed, 6 Jul 2022 16:32:28 +0000 (GMT) X-BeenThere: gcc-cvs@gcc.gnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Gcc-cvs mailing list List-Unsubscribe: , List-Archive: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 06 Jul 2022 16:32:28 -0000 https://gcc.gnu.org/g:f98ca8fbb9188f01532effbed4342f94167a5582 commit f98ca8fbb9188f01532effbed4342f94167a5582 Author: Xi Ruoyao Date: Wed Jul 6 23:22:29 2022 +0800 loongarch: avoid unnecessary sign-extend after 32-bit division Like add.w/sub.w/mul.w, div.w/mod.w/div.wu/mod.wu also sign-extend the output on LA64. But, LoongArch v1.00 mandates that the inputs of 32-bit division to be sign-extended so we have to expand 32-bit division into RTL sequences. We defined div.w/mod.w/div.wu/mod.wu as a (DI, DI) -> SI instruction. This definition does not indicate the fact that these instructions will store the result as sign-extended value in a 64-bit GR. Then the compiler would emit unnecessary sign-extend operations. For example: int div(int a, int b) { return a / b; } was compiled to: div.w $r4, $r4, $r5 slli.w $r4, $r4, 0 # this is unnecessary jr $r1 To remove this unnecessary operation, we change the division instructions to (DI, DI) -> DI and describe the sign-extend behavior explicitly in the RTL template. In the expander for 32-bit division we then use simplify_gen_subreg to extract the lower 32 bits. gcc/ChangeLog: * config/loongarch/loongarch.md (di3_fake): Describe the sign-extend of result in the RTL template. (3): Adjust for di3_fake change. gcc/testsuite/ChangeLog: * gcc.target/loongarch/div-4.c: New test. Diff: --- gcc/config/loongarch/loongarch.md | 12 ++++++++---- gcc/testsuite/gcc.target/loongarch/div-4.c | 9 +++++++++ 2 files changed, 17 insertions(+), 4 deletions(-) diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md index b002eb2ac22..0202f73efae 100644 --- a/gcc/config/loongarch/loongarch.md +++ b/gcc/config/loongarch/loongarch.md @@ -752,6 +752,7 @@ { rtx reg1 = gen_reg_rtx (DImode); rtx reg2 = gen_reg_rtx (DImode); + rtx rd = gen_reg_rtx (DImode); operands[1] = gen_rtx_SIGN_EXTEND (word_mode, operands[1]); operands[2] = gen_rtx_SIGN_EXTEND (word_mode, operands[2]); @@ -759,7 +760,9 @@ emit_insn (gen_rtx_SET (reg1, operands[1])); emit_insn (gen_rtx_SET (reg2, operands[2])); - emit_insn (gen_di3_fake (operands[0], reg1, reg2)); + emit_insn (gen_di3_fake (rd, reg1, reg2)); + emit_insn (gen_rtx_SET (operands[0], + simplify_gen_subreg (SImode, rd, DImode, 0))); DONE; } }) @@ -781,9 +784,10 @@ (const_string "no")))]) (define_insn "di3_fake" - [(set (match_operand:SI 0 "register_operand" "=r,&r,&r") - (any_div:SI (match_operand:DI 1 "register_operand" "r,r,0") - (match_operand:DI 2 "register_operand" "r,r,r")))] + [(set (match_operand:DI 0 "register_operand" "=r,&r,&r") + (sign_extend:DI + (any_div:SI (match_operand:DI 1 "register_operand" "r,r,0") + (match_operand:DI 2 "register_operand" "r,r,r"))))] "" { return loongarch_output_division (".w\t%0,%1,%2", operands); diff --git a/gcc/testsuite/gcc.target/loongarch/div-4.c b/gcc/testsuite/gcc.target/loongarch/div-4.c new file mode 100644 index 00000000000..a52f87d6caf --- /dev/null +++ b/gcc/testsuite/gcc.target/loongarch/div-4.c @@ -0,0 +1,9 @@ +/* { dg-do compile } */ +/* { dg-options "-O2" } */ +/* { dg-final { scan-assembler-not "slli" } } */ + +int +div(int a, int b) +{ + return a / b; +}