From patchwork Wed Jul 21 19:59:29 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Richard Henderson X-Patchwork-Id: 483317 Delivered-To: patch@linaro.org Received: by 2002:a02:c94a:0:0:0:0:0 with SMTP id u10csp6511818jao; Wed, 21 Jul 2021 13:00:58 -0700 (PDT) X-Google-Smtp-Source: ABdhPJxnouwq5EygEa4BVnzpnSSFdbDBl3rs7wqPg8POlsfxIw7E5p7LacXu+4R13FRcUmcvLw1o X-Received: by 2002:a05:6e02:d93:: with SMTP id i19mr13049847ilj.72.1626897658893; Wed, 21 Jul 2021 13:00:58 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1626897658; cv=none; d=google.com; s=arc-20160816; b=FEZTduwdZBnenAS0pCgP2iu0DFAd2NmQNcmbe608vy/c3fiXYuh6MFbvs1nxgZsjEr sYBxfFuTRD0f31MFRcwaqul+aW713sc/G/fLDLcr8pv+WPJzvO2q2l7tMf1qnXMxa7HC Sttm6H+M8Ted0LsRjO8aRmbkqnoNLTvEogTVIqTLQDGyLrQA3ziRjMiCsU1bykLQQOMO YyqhRve6DrnVoATIRqpLLFrW+NnYyx3nSye5aiPSM50eG1v9oNF02hXX9OTfm9sWa5sE elOuAt8m2j2FBOjyA00F7qffBxeEOjgqkYaNz0EQ1dTFHBJd08A/4Ybz7Dz+96WfhFQI ey5Q== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:cc:list-subscribe:list-help:list-post:list-archive :list-unsubscribe:list-id:precedence:content-transfer-encoding :mime-version:references:in-reply-to:message-id:date:subject:to:from :dkim-signature; bh=Qsy1TulgZgPi35LzO24guGsv4onLBv9K5/zzVnSrTIU=; b=hA4l60Jr907liaInOZXtuTsENCf2mOHwiQowHpdHRzNyKnev5aNPt1vWJfr8qAjNYx C/52i79gUpN1uUWhyfdlaIUDnmWkGPisQ1rsx7G0csFGEcNVX9erBH+bjXNtzxOCoJNV RKY8W2KxWM8CqK1sSfkFu+j6oNVwhMl2gCdPBpQh3mZdGpnyPOh/B9QesOJg4dAaLexl QpdVTjCnNzmDR5RRz5mHeJXZiqcC/gk+meUTtVmQL1PzUWYB7wt1KYsrtbYzZDy2WfDa vn4aufJLkk4DiXlQskeYtCjBeY2YbU3xdS+22uP2TqpsYz6jM8nHIaZtYw0uBvhsBTUx SZrA== ARC-Authentication-Results: i=1; mx.google.com; dkim=fail header.i=@linaro.org header.s=google header.b=QRByAOkk; spf=pass (google.com: domain of qemu-devel-bounces+patch=linaro.org@nongnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom="qemu-devel-bounces+patch=linaro.org@nongnu.org"; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=linaro.org Return-Path: Received: from lists.gnu.org (lists.gnu.org. [209.51.188.17]) by mx.google.com with ESMTPS id x14si26752005ilm.100.2021.07.21.13.00.58 for (version=TLS1_2 cipher=ECDHE-ECDSA-CHACHA20-POLY1305 bits=256/256); Wed, 21 Jul 2021 13:00:58 -0700 (PDT) Received-SPF: pass (google.com: domain of qemu-devel-bounces+patch=linaro.org@nongnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; Authentication-Results: mx.google.com; dkim=fail header.i=@linaro.org header.s=google header.b=QRByAOkk; spf=pass (google.com: domain of qemu-devel-bounces+patch=linaro.org@nongnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom="qemu-devel-bounces+patch=linaro.org@nongnu.org"; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=linaro.org Received: from localhost ([::1]:49276 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1m6IOo-0002Pu-2G for patch@linaro.org; Wed, 21 Jul 2021 16:00:58 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:36488) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1m6INw-0002OX-4o for qemu-devel@nongnu.org; Wed, 21 Jul 2021 16:00:04 -0400 Received: from mail-pl1-x630.google.com ([2607:f8b0:4864:20::630]:36816) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1m6INt-0005av-G9 for qemu-devel@nongnu.org; Wed, 21 Jul 2021 16:00:03 -0400 Received: by mail-pl1-x630.google.com with SMTP id x16so1650052plg.3 for ; Wed, 21 Jul 2021 13:00:01 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=Qsy1TulgZgPi35LzO24guGsv4onLBv9K5/zzVnSrTIU=; b=QRByAOkkN+6TUvyldPDTtYpInxWSRayy9aBteaLzDRUBNO3TIH56SDNlXkr8MA6rFj T8T84jGKQYaGvn+dx3y9fToPR/5kyIkhgH3ZD0jXdtDgTK7kLXTh0aSnHGpeBFFkjIfO zVLB3eIHCVX7Sd4LyP5UOVZe9BQOX6HR+IDT9l0qLNhU/gatARy7KtiIVWzBm+tcgSXe bUNR/a0BqyAtUXXtKdNt/C+Ow/YhuTVNOmVe5txQ+EzVsV2ydpUuXETfIYgdSAJ059Kw dJdvTrDCinj407WGDBeJ0s/5AkWZQ0WhSskgWlGJ7xY17Xgq+4lJrhHohat48wsSZuKI kOvw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=Qsy1TulgZgPi35LzO24guGsv4onLBv9K5/zzVnSrTIU=; b=LPv/yHkPqx0HWtGDUTEktVbsFcu0gQHtp7iD8nryxt57GZ4fPX//AbedWCaMdbx6EX 7p8k9tkkngYEOyNNvppA1FTBKzYOBIuhAJG4oJ3UxfS1wO6K8uzAyxlg+japMetHASNk 07i87sJdPjpT6H6oObSprFUgPndnILH2ROS+Bl5rXyxRnIqM8VzHuMVNPaMWcEMfSRK9 luZGVBQCjJaj3L+MlHWoHtUAJicvadNtAiQJ15PSCwVEIlpeMgRRrT8CxkeLla3Y6PSw 1o0mpdhRK5ROnZq2uIgJHmxMcLk+cc6UcIhvLdo+RH3m69STnmIhCMpgJsctW8ilsO2Z 1CoQ== X-Gm-Message-State: AOAM530xt4noCfsZ/OqbawztZ7Zqtuy8c4RiJ2StUPak+4nOMMz6wLFi 193ec4X4nB3otNcpHgKZSWSJAofEr8Yu8w== X-Received: by 2002:a05:6a00:189e:b029:32b:9f66:dcbb with SMTP id x30-20020a056a00189eb029032b9f66dcbbmr38263852pfh.72.1626897600003; Wed, 21 Jul 2021 13:00:00 -0700 (PDT) Received: from localhost.localdomain (204-210-126-223.res.spectrum.com. [204.210.126.223]) by smtp.gmail.com with ESMTPSA id w3sm706028pjq.12.2021.07.21.12.59.58 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 21 Jul 2021 12:59:59 -0700 (PDT) From: Richard Henderson To: qemu-devel@nongnu.org Subject: [PULL 02/27] qemu/atomic: Remove pre-C11 atomic fallbacks Date: Wed, 21 Jul 2021 09:59:29 -1000 Message-Id: <20210721195954.879535-3-richard.henderson@linaro.org> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20210721195954.879535-1-richard.henderson@linaro.org> References: <20210721195954.879535-1-richard.henderson@linaro.org> MIME-Version: 1.0 Received-SPF: pass client-ip=2607:f8b0:4864:20::630; envelope-from=richard.henderson@linaro.org; helo=mail-pl1-x630.google.com X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: peter.maydell@linaro.org, =?utf-8?q?Alex_Benn=C3=A9e?= , Cole Robinson Errors-To: qemu-devel-bounces+patch=linaro.org@nongnu.org Sender: "Qemu-devel" We now require c11, so the fallbacks are now dead code Tested-by: Cole Robinson Reviewed-by: Alex Bennée Reviewed-by: Peter Maydell Signed-off-by: Richard Henderson --- configure | 7 -- include/qemu/atomic.h | 204 +++--------------------------------------- 2 files changed, 10 insertions(+), 201 deletions(-) -- 2.25.1 diff --git a/configure b/configure index 232c54dcc1..b5965b159f 100755 --- a/configure +++ b/configure @@ -3991,18 +3991,11 @@ cat > $TMPC << EOF int main(void) { uint64_t x = 0, y = 0; -#ifdef __ATOMIC_RELAXED y = __atomic_load_n(&x, __ATOMIC_RELAXED); __atomic_store_n(&x, y, __ATOMIC_RELAXED); __atomic_compare_exchange_n(&x, &y, x, 0, __ATOMIC_RELAXED, __ATOMIC_RELAXED); __atomic_exchange_n(&x, y, __ATOMIC_RELAXED); __atomic_fetch_add(&x, y, __ATOMIC_RELAXED); -#else - typedef char is_host64[sizeof(void *) >= sizeof(uint64_t) ? 1 : -1]; - __sync_lock_test_and_set(&x, y); - __sync_val_compare_and_swap(&x, y, 0); - __sync_fetch_and_add(&x, y); -#endif return 0; } EOF diff --git a/include/qemu/atomic.h b/include/qemu/atomic.h index a7654d2a33..fe5467d193 100644 --- a/include/qemu/atomic.h +++ b/include/qemu/atomic.h @@ -60,8 +60,9 @@ (unsigned short)1, \ (expr)+0)))))) -#ifdef __ATOMIC_RELAXED -/* For C11 atomic ops */ +#ifndef __ATOMIC_RELAXED +#error "Expecting C11 atomic ops" +#endif /* Manual memory barriers * @@ -239,193 +240,8 @@ #define qatomic_xor(ptr, n) \ ((void) __atomic_fetch_xor(ptr, n, __ATOMIC_SEQ_CST)) -#else /* __ATOMIC_RELAXED */ - -#ifdef __alpha__ -#define smp_read_barrier_depends() asm volatile("mb":::"memory") -#endif - -#if defined(__i386__) || defined(__x86_64__) || defined(__s390x__) - -/* - * Because of the strongly ordered storage model, wmb() and rmb() are nops - * here (a compiler barrier only). QEMU doesn't do accesses to write-combining - * qemu memory or non-temporal load/stores from C code. - */ -#define smp_mb_release() barrier() -#define smp_mb_acquire() barrier() - -/* - * __sync_lock_test_and_set() is documented to be an acquire barrier only, - * but it is a full barrier at the hardware level. Add a compiler barrier - * to make it a full barrier also at the compiler level. - */ -#define qatomic_xchg(ptr, i) (barrier(), __sync_lock_test_and_set(ptr, i)) - -#elif defined(_ARCH_PPC) - -/* - * We use an eieio() for wmb() on powerpc. This assumes we don't - * need to order cacheable and non-cacheable stores with respect to - * each other. - * - * smp_mb has the same problem as on x86 for not-very-new GCC - * (http://patchwork.ozlabs.org/patch/126184/, Nov 2011). - */ -#define smp_wmb() ({ asm volatile("eieio" ::: "memory"); (void)0; }) -#if defined(__powerpc64__) -#define smp_mb_release() ({ asm volatile("lwsync" ::: "memory"); (void)0; }) -#define smp_mb_acquire() ({ asm volatile("lwsync" ::: "memory"); (void)0; }) -#else -#define smp_mb_release() ({ asm volatile("sync" ::: "memory"); (void)0; }) -#define smp_mb_acquire() ({ asm volatile("sync" ::: "memory"); (void)0; }) -#endif -#define smp_mb() ({ asm volatile("sync" ::: "memory"); (void)0; }) - -#endif /* _ARCH_PPC */ - -/* - * For (host) platforms we don't have explicit barrier definitions - * for, we use the gcc __sync_synchronize() primitive to generate a - * full barrier. This should be safe on all platforms, though it may - * be overkill for smp_mb_acquire() and smp_mb_release(). - */ -#ifndef smp_mb -#define smp_mb() __sync_synchronize() -#endif - -#ifndef smp_mb_acquire -#define smp_mb_acquire() __sync_synchronize() -#endif - -#ifndef smp_mb_release -#define smp_mb_release() __sync_synchronize() -#endif - -#ifndef smp_read_barrier_depends -#define smp_read_barrier_depends() barrier() -#endif - -#ifndef signal_barrier -#define signal_barrier() barrier() -#endif - -/* These will only be atomic if the processor does the fetch or store - * in a single issue memory operation - */ -#define qatomic_read__nocheck(p) (*(__typeof__(*(p)) volatile*) (p)) -#define qatomic_set__nocheck(p, i) ((*(__typeof__(*(p)) volatile*) (p)) = (i)) - -#define qatomic_read(ptr) qatomic_read__nocheck(ptr) -#define qatomic_set(ptr, i) qatomic_set__nocheck(ptr,i) - -/** - * qatomic_rcu_read - reads a RCU-protected pointer to a local variable - * into a RCU read-side critical section. The pointer can later be safely - * dereferenced within the critical section. - * - * This ensures that the pointer copy is invariant thorough the whole critical - * section. - * - * Inserts memory barriers on architectures that require them (currently only - * Alpha) and documents which pointers are protected by RCU. - * - * qatomic_rcu_read also includes a compiler barrier to ensure that - * value-speculative optimizations (e.g. VSS: Value Speculation - * Scheduling) does not perform the data read before the pointer read - * by speculating the value of the pointer. - * - * Should match qatomic_rcu_set(), qatomic_xchg(), qatomic_cmpxchg(). - */ -#define qatomic_rcu_read(ptr) ({ \ - typeof(*ptr) _val = qatomic_read(ptr); \ - smp_read_barrier_depends(); \ - _val; \ -}) - -/** - * qatomic_rcu_set - assigns (publicizes) a pointer to a new data structure - * meant to be read by RCU read-side critical sections. - * - * Documents which pointers will be dereferenced by RCU read-side critical - * sections and adds the required memory barriers on architectures requiring - * them. It also makes sure the compiler does not reorder code initializing the - * data structure before its publication. - * - * Should match qatomic_rcu_read(). - */ -#define qatomic_rcu_set(ptr, i) do { \ - smp_wmb(); \ - qatomic_set(ptr, i); \ -} while (0) - -#define qatomic_load_acquire(ptr) ({ \ - typeof(*ptr) _val = qatomic_read(ptr); \ - smp_mb_acquire(); \ - _val; \ -}) - -#define qatomic_store_release(ptr, i) do { \ - smp_mb_release(); \ - qatomic_set(ptr, i); \ -} while (0) - -#ifndef qatomic_xchg -#if defined(__clang__) -#define qatomic_xchg(ptr, i) __sync_swap(ptr, i) -#else -/* __sync_lock_test_and_set() is documented to be an acquire barrier only. */ -#define qatomic_xchg(ptr, i) (smp_mb(), __sync_lock_test_and_set(ptr, i)) -#endif -#endif -#define qatomic_xchg__nocheck qatomic_xchg - -/* Provide shorter names for GCC atomic builtins. */ -#define qatomic_fetch_inc(ptr) __sync_fetch_and_add(ptr, 1) -#define qatomic_fetch_dec(ptr) __sync_fetch_and_add(ptr, -1) - -#define qatomic_fetch_add(ptr, n) __sync_fetch_and_add(ptr, n) -#define qatomic_fetch_sub(ptr, n) __sync_fetch_and_sub(ptr, n) -#define qatomic_fetch_and(ptr, n) __sync_fetch_and_and(ptr, n) -#define qatomic_fetch_or(ptr, n) __sync_fetch_and_or(ptr, n) -#define qatomic_fetch_xor(ptr, n) __sync_fetch_and_xor(ptr, n) - -#define qatomic_inc_fetch(ptr) __sync_add_and_fetch(ptr, 1) -#define qatomic_dec_fetch(ptr) __sync_add_and_fetch(ptr, -1) -#define qatomic_add_fetch(ptr, n) __sync_add_and_fetch(ptr, n) -#define qatomic_sub_fetch(ptr, n) __sync_sub_and_fetch(ptr, n) -#define qatomic_and_fetch(ptr, n) __sync_and_and_fetch(ptr, n) -#define qatomic_or_fetch(ptr, n) __sync_or_and_fetch(ptr, n) -#define qatomic_xor_fetch(ptr, n) __sync_xor_and_fetch(ptr, n) - -#define qatomic_cmpxchg(ptr, old, new) \ - __sync_val_compare_and_swap(ptr, old, new) -#define qatomic_cmpxchg__nocheck(ptr, old, new) qatomic_cmpxchg(ptr, old, new) - -/* And even shorter names that return void. */ -#define qatomic_inc(ptr) ((void) __sync_fetch_and_add(ptr, 1)) -#define qatomic_dec(ptr) ((void) __sync_fetch_and_add(ptr, -1)) -#define qatomic_add(ptr, n) ((void) __sync_fetch_and_add(ptr, n)) -#define qatomic_sub(ptr, n) ((void) __sync_fetch_and_sub(ptr, n)) -#define qatomic_and(ptr, n) ((void) __sync_fetch_and_and(ptr, n)) -#define qatomic_or(ptr, n) ((void) __sync_fetch_and_or(ptr, n)) -#define qatomic_xor(ptr, n) ((void) __sync_fetch_and_xor(ptr, n)) - -#endif /* __ATOMIC_RELAXED */ - -#ifndef smp_wmb #define smp_wmb() smp_mb_release() -#endif -#ifndef smp_rmb #define smp_rmb() smp_mb_acquire() -#endif - -/* This is more efficient than a store plus a fence. */ -#if !defined(__SANITIZE_THREAD__) -#if defined(__i386__) || defined(__x86_64__) || defined(__s390x__) -#define qatomic_mb_set(ptr, i) ((void)qatomic_xchg(ptr, i)) -#endif -#endif /* qatomic_mb_read/set semantics map Java volatile variables. They are * less expensive on some platforms (notably POWER) than fully @@ -435,16 +251,16 @@ * use. See docs/devel/atomics.rst for more discussion. */ -#ifndef qatomic_mb_read #define qatomic_mb_read(ptr) \ qatomic_load_acquire(ptr) -#endif -#ifndef qatomic_mb_set -#define qatomic_mb_set(ptr, i) do { \ - qatomic_store_release(ptr, i); \ - smp_mb(); \ -} while(0) +#if !defined(__SANITIZE_THREAD__) && \ + (defined(__i386__) || defined(__x86_64__) || defined(__s390x__)) +/* This is more efficient than a store plus a fence. */ +# define qatomic_mb_set(ptr, i) ((void)qatomic_xchg(ptr, i)) +#else +# define qatomic_mb_set(ptr, i) \ + ({ qatomic_store_release(ptr, i); smp_mb(); }) #endif #define qatomic_fetch_inc_nonzero(ptr) ({ \