From patchwork Thu Nov 17 06:38:52 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jinyang He X-Patchwork-Id: 60738 Return-Path: X-Original-To: patchwork@sourceware.org Delivered-To: patchwork@sourceware.org Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 935FE38983A5 for ; Thu, 17 Nov 2022 06:40:47 +0000 (GMT) X-Original-To: gcc-patches@gcc.gnu.org Delivered-To: gcc-patches@gcc.gnu.org Received: from loongson.cn (mail.loongson.cn [114.242.206.163]) by sourceware.org (Postfix) with ESMTP id 1F21D388B68F for ; Thu, 17 Nov 2022 06:39:39 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.1 sourceware.org 1F21D388B68F Authentication-Results: sourceware.org; dmarc=none (p=none dis=none) header.from=loongson.cn Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=loongson.cn Received: from loongson.cn (unknown [111.9.175.10]) by gateway (Coremail) with SMTP id _____8AxTbcp13VjvTEIAA--.11773S3; Thu, 17 Nov 2022 14:39:37 +0800 (CST) Received: from localhost.localdomain (unknown [111.9.175.10]) by localhost.localdomain (Coremail) with SMTP id AQAAf8Dx_1cm13VjdcMVAA--.57467S2; Thu, 17 Nov 2022 14:39:36 +0800 (CST) From: Jinyang He To: Chenghua Xu , Lulu Cheng , Xi Ruoyao Cc: Weining Lu , Xing Li , yala , Peng Fan , Huang Pei , gcc-patches@gcc.gnu.org Subject: [PATCH v2] LoongArch: Fix atomic_exchange expanding [PR107713] Date: Thu, 17 Nov 2022 14:38:52 +0800 Message-Id: <20221117063852.29869-1-hejinyang@loongson.cn> X-Mailer: git-send-email 2.20.1 MIME-Version: 1.0 X-CM-TRANSID: AQAAf8Dx_1cm13VjdcMVAA--.57467S2 X-CM-SenderInfo: pkhmx0p1dqwqxorr0wxvrqhubq/ X-Coremail-Antispam: 1Uk129KBjvJXoWxCFW7Gr47tr1DGr1fKry7Awb_yoWruFW5pr ZrAr15tF48JrZ7G397Gay3JrsIyrn7CF4xua9rK34I9w13WryUZ3W8KF9I9FyUGw4aqr1f XF4S934jgF42qaUanT9S1TB71UUUUUUqnTZGkaVYY2UrUUUUj1kv1TuYvTs0mT0YCTnIWj qI5I8CrVACY4xI64kE6c02F40Ex7xfYxn0WfASr-VFAUDa7-sFnT9fnUUIcSsGvfJTRUUU b7kYFVCjjxCrM7AC8VAFwI0_Jr0_Gr1l1xkIjI8I6I8E6xAIw20EY4v20xvaj40_Wr0E3s 1l1IIY67AEw4v_Jrv_JF1l8cAvFVAK0II2c7xJM28CjxkF64kEwVA0rcxSw2x7M28EF7xv wVC0I7IYx2IY67AKxVW8JVW5JwA2z4x0Y4vE2Ix0cI8IcVCY1x0267AKxVW8JVWxJwA2z4 x0Y4vEx4A2jsIE14v26r4UJVWxJr1l84ACjcxK6I8E87Iv6xkF7I0E14v26F4UJVW0owAS 0I0E0xvYzxvE52x082IY62kv0487Mc804VCY07AIYIkI8VC2zVCFFI0UMc02F40EFcxC0V AKzVAqx4xG6I80ewAv7VC0I7IYx2IY67AKxVWUJVWUGwAv7VC2z280aVAFwI0_Gr0_Cr1l Ox8S6xCaFVCjc4AY6r1j6r4UM4x0Y48IcxkI7VAKI48JMxAIw28IcxkI7VAKI48JMxC20s 026xCaFVCjc4AY6r1j6r4UMI8I3I0E5I8CrVAFwI0_Jr0_Jr4lx2IqxVCjr7xvwVAFwI0_ JrI_JrWlx4CE17CEb7AF67AKxVWUtVW8ZwCIc40Y0x0EwIxGrwCI42IY6xIIjxv20xvE14 v26r1j6r1xMIIF0xvE2Ix0cI8IcVCY1x0267AKxVWUJVW8JwCI42IY6xAIw20EY4v20xva j40_Jr0_JF4lIxAIcVC2z280aVAFwI0_Jr0_Gr1lIxAIcVC2z280aVCY1x0267AKxVWUJV W8JbIYCTnIWIevJa73UjIFyTuYvjxUclApUUUUU X-Spam-Status: No, score=-11.5 required=5.0 tests=BAYES_00, GIT_PATCH_0, KAM_DMARC_STATUS, KAM_SHORT, SPF_HELO_PASS, SPF_PASS, TXREP autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org X-BeenThere: gcc-patches@gcc.gnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Gcc-patches mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: gcc-patches-bounces+patchwork=sourceware.org@gcc.gnu.org Sender: "Gcc-patches" We used to expand atomic_exchange_n(ptr, new, mem_order) for subword types into something like: { __typeof__(*ptr) t = atomic_load_n(ptr, mem_order); atomic_compare_exchange_n(ptr, &t, new, true, mem_order, mem_order); return t; } It's incorrect because another thread may store a different value into *ptr after atomic_load_n. Then atomic_compare_exchange_n will not store into *ptr, but atomic_exchange_n should always perform the store. gcc/ChangeLog: PR target/107713 * config/loongarch/sync.md (atomic_cas_value_exchange_7_): New define_insn. (atomic_exchange): Use atomic_cas_value_exchange_7_si instead of atomic_cas_value_cmp_and_7_si. gcc/testsuite/ChangeLog: PR target/107713 * gcc.target/loongarch/pr107713-1.c: New test. * gcc.target/loongarch/pr107713-2.c: New test. --- Thanks Ruoyao who helped me with community works, improved the submit message and gave a more straightforward test. Thanks Pei who gave me a further understanding of ll-sc offline. gcc/config/loongarch/sync.md | 27 +++++++++- .../gcc.target/loongarch/pr107713-1.c | 50 +++++++++++++++++++ .../gcc.target/loongarch/pr107713-2.c | 9 ++++ 3 files changed, 84 insertions(+), 2 deletions(-) create mode 100644 gcc/testsuite/gcc.target/loongarch/pr107713-1.c create mode 100644 gcc/testsuite/gcc.target/loongarch/pr107713-2.c diff --git a/gcc/config/loongarch/sync.md b/gcc/config/loongarch/sync.md index 0c4f1983e88..45be1442439 100644 --- a/gcc/config/loongarch/sync.md +++ b/gcc/config/loongarch/sync.md @@ -448,6 +448,29 @@ } [(set (attr "length") (const_int 32))]) +(define_insn "atomic_cas_value_exchange_7_" + [(set (match_operand:GPR 0 "register_operand" "=&r") + (match_operand:GPR 1 "memory_operand" "+ZC")) + (set (match_dup 1) + (unspec_volatile:GPR [(match_operand:GPR 2 "reg_or_0_operand" "rJ") + (match_operand:GPR 3 "reg_or_0_operand" "rJ") + (match_operand:GPR 4 "reg_or_0_operand" "rJ") + (match_operand:GPR 5 "reg_or_0_operand" "rJ") + (match_operand:SI 6 "const_int_operand")] ;; model + UNSPEC_SYNC_EXCHANGE)) + (clobber (match_scratch:GPR 7 "=&r"))] + "" +{ + return "%G6\\n\\t" + "1:\\n\\t" + "ll.\\t%0,%1\\n\\t" + "and\\t%7,%0,%z3\\n\\t" + "or%i5\\t%7,%7,%5\\n\\t" + "sc.\\t%7,%1\\n\\t" + "beqz\\t%7,1b\\n\\t"; +} + [(set (attr "length") (const_int 20))]) + (define_expand "atomic_exchange" [(set (match_operand:SHORT 0 "register_operand") (unspec_volatile:SHORT @@ -459,9 +482,9 @@ "" { union loongarch_gen_fn_ptrs generator; - generator.fn_7 = gen_atomic_cas_value_cmp_and_7_si; + generator.fn_7 = gen_atomic_cas_value_exchange_7_si; loongarch_expand_atomic_qihi (generator, operands[0], operands[1], - operands[1], operands[2], operands[3]); + const0_rtx, operands[2], operands[3]); DONE; }) diff --git a/gcc/testsuite/gcc.target/loongarch/pr107713-1.c b/gcc/testsuite/gcc.target/loongarch/pr107713-1.c new file mode 100644 index 00000000000..d1536c95b27 --- /dev/null +++ b/gcc/testsuite/gcc.target/loongarch/pr107713-1.c @@ -0,0 +1,50 @@ +/* { dg-do run } */ +/* { dg-require-effective-target pthread } */ +/* { dg-options "-pthread" } */ + +#include + +char x, x1, x2; + +void * +work1 (void *) +{ + for (int i = 0; i < 100; i++) + x1 = __atomic_exchange_n (&x, x1, __ATOMIC_SEQ_CST); + return NULL; +} + +void * +work2 (void *) +{ + for (int i = 0; i < 100; i++) + x2 = __atomic_exchange_n (&x, x2, __ATOMIC_SEQ_CST); + return NULL; +} + +void +test (void) +{ + x = 0; + x1 = 1; + x2 = 2; + pthread_t w1, w2; + if (pthread_create (&w1, NULL, work1, NULL) != 0) + __builtin_abort (); + if (pthread_create (&w2, NULL, work2, NULL) != 0) + __builtin_abort (); + if (pthread_join (w1, NULL) != 0) + __builtin_abort (); + if (pthread_join (w2, NULL) != 0) + __builtin_abort (); + if ((x ^ x1 ^ x2) != 3) + __builtin_abort (); +} + +int +main () +{ + int i; + for (i = 0; i < 10000; i++) + test (); +} diff --git a/gcc/testsuite/gcc.target/loongarch/pr107713-2.c b/gcc/testsuite/gcc.target/loongarch/pr107713-2.c new file mode 100644 index 00000000000..82d44db3d51 --- /dev/null +++ b/gcc/testsuite/gcc.target/loongarch/pr107713-2.c @@ -0,0 +1,9 @@ +/* { dg-do compile } */ +/* { dg-options "-O2" } */ +/* { dg-final { scan-assembler-times "beq|bne" 1 } } */ + +char +t (char *p, char x) +{ + return __atomic_exchange_n (p, x, __ATOMIC_RELAXED); +}