From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id F2E4AC77B61 for ; Thu, 13 Apr 2023 11:52:56 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230312AbjDMLwz (ORCPT ); Thu, 13 Apr 2023 07:52:55 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:54400 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229742AbjDMLwx (ORCPT ); Thu, 13 Apr 2023 07:52:53 -0400 Received: from loongson.cn (mail.loongson.cn [114.242.206.163]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id CC8161B3; Thu, 13 Apr 2023 04:52:50 -0700 (PDT) Received: from loongson.cn (unknown [10.20.42.120]) by gateway (Coremail) with SMTP id _____8BxYU8R7Tdk+eIbAA--.43450S3; Thu, 13 Apr 2023 19:52:49 +0800 (CST) Received: from [10.20.42.120] (unknown [10.20.42.120]) by localhost.localdomain (Coremail) with SMTP id AQAAf8DxxrwQ7Tdkl1oiAA--.36338S3; Thu, 13 Apr 2023 19:52:48 +0800 (CST) Subject: Re: [PATCH v6 29/30] LoongArch: KVM: Enable kvm config and add the makefile To: Xi Ruoyao , kernel test robot , Paolo Bonzini References: <20230412083023.1693910-30-zhaotianrui@loongson.cn> <202304131526.iXfLaVZc-lkp@intel.com> Cc: oe-kbuild-all@lists.linux.dev, Huacai Chen , WANG Xuerui , Greg Kroah-Hartman , loongarch@lists.linux.dev, linux-kernel@vger.kernel.org, kvm@vger.kernel.org, Jens Axboe , Mark Brown , Alex Deucher , Oliver Upton , maobibo@loongson.cn From: Tianrui Zhao Message-ID: Date: Thu, 13 Apr 2023 19:52:58 +0800 User-Agent: Mozilla/5.0 (X11; Linux mips64; rv:45.0) Gecko/20100101 Thunderbird/45.4.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit X-CM-TRANSID: AQAAf8DxxrwQ7Tdkl1oiAA--.36338S3 X-CM-SenderInfo: p2kd03xldq233l6o00pqjv00gofq/ X-Coremail-Antispam: 1Uk129KBjvJXoW7KF4UKF1rWw45Cw4fGFW8Xrb_yoW8uFyxpa srAr4UKF4vqryjkw1jyF1xt3Z5tryfGay8JF1jvwn8ZF4Ut3s7Jr4UGryrKFWaqas8GF4Y gaySqws0kw18JF7anT9S1TB71UUUUjJqnTZGkaVYY2UrUUUUj1kv1TuYvTs0mT0YCTnIWj qI5I8CrVACY4xI64kE6c02F40Ex7xfYxn0WfASr-VFAUDa7-sFnT9fnUUIcSsGvfJTRUUU bq8YFVCjjxCrM7AC8VAFwI0_Jr0_Gr1l1xkIjI8I6I8E6xAIw20EY4v20xvaj40_Wr0E3s 1l1IIY67AEw4v_Jrv_JF1l8cAvFVAK0II2c7xJM28CjxkF64kEwVA0rcxSw2x7M28EF7xv wVC0I7IYx2IY67AKxVW8JVW5JwA2z4x0Y4vE2Ix0cI8IcVCY1x0267AKxVW8JVWxJwA2z4 x0Y4vEx4A2jsIE14v26F4UJVW0owA2z4x0Y4vEx4A2jsIEc7CjxVAFwI0_Cr1j6rxdM2kK e7AKxVWUAVWUtwAS0I0E0xvYzxvE52x082IY62kv0487Mc804VCY07AIYIkI8VC2zVCFFI 0UMc02F40EFcxC0VAKzVAqx4xG6I80ewAv7VC0I7IYx2IY67AKxVWUAVWUtwAv7VC2z280 aVAFwI0_Jr0_Gr1lOx8S6xCaFVCjc4AY6r1j6r4UM4x0Y48IcVAKI48JMxk0xIA0c2IEe2 xFo4CEbIxvr21lc7CjxVAaw2AFwI0_Jw0_GFyl42xK82IYc2Ij64vIr41l4I8I3I0E4IkC 6x0Yz7v_Jr0_Gr1l4IxYO2xFxVAFwI0_JF0_Jw1lx2IqxVAqx4xG67AKxVWUJVWUGwC20s 026x8GjcxK67AKxVWUGVWUWwC2zVAF1VAY17CE14v26r1q6r43MIIYrxkI7VAKI48JMIIF 0xvE2Ix0cI8IcVAFwI0_JFI_Gr1lIxAIcVC0I7IYx2IY6xkF7I0E14v26r4j6F4UMIIF0x vE42xK8VAvwI8IcIk0rVWUJVWUCwCI42IY6I8E87Iv67AKxVWUJVW8JwCI42IY6I8E87Iv 6xkF7I0E14v26r4j6r4UJbIYCTnIWIevJa73UjIFyTuYvjxU2-VyUUUUU Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 在 2023年04月13日 16:33, Xi Ruoyao 写道: > On Thu, 2023-04-13 at 15:32 +0800, kernel test robot wrote: > >> vim +/asm +244 arch/loongarch/include/asm/loongarch.h >> >> f2ac457a61389b Huacai Chen 2022-05-31 238 >> 2c87b678639d25 Tianrui Zhao 2023-04-12 239 /* GCSR */ >> 2c87b678639d25 Tianrui Zhao 2023-04-12 240 static inline u64 gcsr_read(u32 reg) > Try __always_inline instead of inline because this "function" will > definitely break up if it's not inlined. > > And I guess we should have something like: > > extern u64 __compiletime_error("gcsr_read parameter is not a constant") > __gcsr_read_non_const(void); > > static __always_inline u64 gcsr_read(u32 reg) > { > u64 val = 0; > > if (!__builtin_constant_p(reg)) > return __gcsr_read_non_const(); > > asm volatile ( > ... ... > > Or perhaps we should just rewrite gcsr_read to a macro. > >> 2c87b678639d25 Tianrui Zhao 2023-04-12 241 { >> 2c87b678639d25 Tianrui Zhao 2023-04-12 242 u64 val = 0; >> 2c87b678639d25 Tianrui Zhao 2023-04-12 243 /* Instructions will be available in binutils later */ >> 2c87b678639d25 Tianrui Zhao 2023-04-12 @244 asm volatile ( >> 2c87b678639d25 Tianrui Zhao 2023-04-12 245 "parse_r __reg, %[val]\n\t" >> 2c87b678639d25 Tianrui Zhao 2023-04-12 246 /* >> 2c87b678639d25 Tianrui Zhao 2023-04-12 247 * read val from guest csr register %[reg] >> 2c87b678639d25 Tianrui Zhao 2023-04-12 248 * gcsrrd %[val], %[reg] >> 2c87b678639d25 Tianrui Zhao 2023-04-12 249 */ >> 2c87b678639d25 Tianrui Zhao 2023-04-12 250 ".word 0x5 << 24 | %[reg] << 10 | 0 << 5 | __reg\n\t" >> 2c87b678639d25 Tianrui Zhao 2023-04-12 251 : [val] "+r" (val) >> 2c87b678639d25 Tianrui Zhao 2023-04-12 252 : [reg] "i" (reg) >> 2c87b678639d25 Tianrui Zhao 2023-04-12 253 : "memory"); >> 2c87b678639d25 Tianrui Zhao 2023-04-12 254 >> 2c87b678639d25 Tianrui Zhao 2023-04-12 255 return val; >> 2c87b678639d25 Tianrui Zhao 2023-04-12 256 } >> 2c87b678639d25 Tianrui Zhao 2023-04-12 257 >> Thanks, it really should use __always_inline there, and I will reconsider to fix this function mistakes about compiling. Thanks Tianrui Zhao