From patchwork Wed Aug 23 15:00:27 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Sergio Gonzalez Monroy X-Patchwork-Id: 27817 Return-Path: X-Original-To: patchwork@dpdk.org Delivered-To: patchwork@dpdk.org Received: from [92.243.14.124] (localhost [IPv6:::1]) by dpdk.org (Postfix) with ESMTP id A2BF37D5A; Wed, 23 Aug 2017 17:00:35 +0200 (CEST) Received: from mga04.intel.com (mga04.intel.com [192.55.52.120]) by dpdk.org (Postfix) with ESMTP id F2AE57D3A for ; Wed, 23 Aug 2017 17:00:31 +0200 (CEST) Received: from fmsmga001.fm.intel.com ([10.253.24.23]) by fmsmga104.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 23 Aug 2017 08:00:31 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos; i="5.41,417,1498546800"; d="scan'208"; a="1187349799" Received: from silpixa00397517.ir.intel.com (HELO silpixa00397517.ger.corp.intel.com) ([10.237.222.54]) by fmsmga001.fm.intel.com with ESMTP; 23 Aug 2017 08:00:30 -0700 From: Sergio Gonzalez Monroy To: dev@dpdk.org Cc: konstantin.ananyev@intel.com, bruce.richardson@intel.com Date: Wed, 23 Aug 2017 16:00:27 +0100 Message-Id: <20170823150027.70565-2-sergio.gonzalez.monroy@intel.com> X-Mailer: git-send-email 2.9.5 In-Reply-To: <20170823150027.70565-1-sergio.gonzalez.monroy@intel.com> References: <20170823150027.70565-1-sergio.gonzalez.monroy@intel.com> Subject: [dpdk-dev] [PATCH] eal/x86: use cpuid builtin X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" GCC does have the __get_cpuid_count builtin which checks for maximum supported leaf, but implementations differ between CLANG and GCC. This change provides an implementation compatible with both GCC and CLANG 3.4+. Signed-off-by: Sergio Gonzalez Monroy Reviewed-by: Ferruh Yigit --- lib/librte_eal/common/arch/x86/rte_cpuflags.c | 40 ++++++--------------------- 1 file changed, 8 insertions(+), 32 deletions(-) diff --git a/lib/librte_eal/common/arch/x86/rte_cpuflags.c b/lib/librte_eal/common/arch/x86/rte_cpuflags.c index 0138257..7d4a0fe 100644 --- a/lib/librte_eal/common/arch/x86/rte_cpuflags.c +++ b/lib/librte_eal/common/arch/x86/rte_cpuflags.c @@ -36,6 +36,7 @@ #include #include #include +#include enum cpu_register_t { RTE_REG_EAX = 0, @@ -156,38 +157,12 @@ const struct feature_entry rte_cpu_feature_table[] = { FEAT_DEF(INVTSC, 0x80000007, 0, RTE_REG_EDX, 8) }; -/* - * Execute CPUID instruction and get contents of a specific register - * - * This function, when compiled with GCC, will generate architecture-neutral - * code, as per GCC manual. - */ -static void -rte_cpu_get_features(uint32_t leaf, uint32_t subleaf, cpuid_registers_t out) -{ -#if defined(__i386__) && defined(__PIC__) - /* %ebx is a forbidden register if we compile with -fPIC or -fPIE */ - asm volatile("movl %%ebx,%0 ; cpuid ; xchgl %%ebx,%0" - : "=r" (out[RTE_REG_EBX]), - "=a" (out[RTE_REG_EAX]), - "=c" (out[RTE_REG_ECX]), - "=d" (out[RTE_REG_EDX]) - : "a" (leaf), "c" (subleaf)); -#else - asm volatile("cpuid" - : "=a" (out[RTE_REG_EAX]), - "=b" (out[RTE_REG_EBX]), - "=c" (out[RTE_REG_ECX]), - "=d" (out[RTE_REG_EDX]) - : "a" (leaf), "c" (subleaf)); -#endif -} - int rte_cpu_get_flag_enabled(enum rte_cpu_flag_t feature) { const struct feature_entry *feat; cpuid_registers_t regs; + unsigned int maxleaf; if (feature >= RTE_CPUFLAG_NUMFLAGS) /* Flag does not match anything in the feature tables */ @@ -199,13 +174,14 @@ rte_cpu_get_flag_enabled(enum rte_cpu_flag_t feature) /* This entry in the table wasn't filled out! */ return -EFAULT; - rte_cpu_get_features(feat->leaf & 0xffff0000, 0, regs); - if (((regs[RTE_REG_EAX] ^ feat->leaf) & 0xffff0000) || - regs[RTE_REG_EAX] < feat->leaf) + maxleaf = __get_cpuid_max(feat->leaf & 0x80000000, NULL); + + if (maxleaf < feat->leaf) return 0; - /* get the cpuid leaf containing the desired feature */ - rte_cpu_get_features(feat->leaf, feat->subleaf, regs); + __cpuid_count(feat->leaf, feat->subleaf, + regs[RTE_REG_EAX], regs[RTE_REG_EBX], + regs[RTE_REG_ECX], regs[RTE_REG_EDX]); /* check if the feature is enabled */ return (regs[feat->reg] >> feat->bit) & 1;