From nobody Sun Feb 8 05:23:07 2026 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 5D3EEC6FD1F for ; Tue, 14 Mar 2023 18:34:33 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230260AbjCNSec (ORCPT ); Tue, 14 Mar 2023 14:34:32 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:49840 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230242AbjCNSe2 (ORCPT ); Tue, 14 Mar 2023 14:34:28 -0400 Received: from mail-pj1-x1031.google.com (mail-pj1-x1031.google.com [IPv6:2607:f8b0:4864:20::1031]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7B5D51A660 for ; Tue, 14 Mar 2023 11:33:45 -0700 (PDT) Received: by mail-pj1-x1031.google.com with SMTP id p13-20020a17090a284d00b0023d2e945aebso1736958pjf.0 for ; Tue, 14 Mar 2023 11:33:45 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rivosinc-com.20210112.gappssmtp.com; s=20210112; t=1678818753; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=tga5shMHyjc0TB9sW2hC3q3qhXdH8jnVVmQsmewODbI=; b=fqdENcGiupBXLWJzcac75bZHKj70wOfrBAIlfGPvxgr7ssgFuE2Pf58C/qEZjdxlZw Dt4Fn4iy/JmCqs+dRf5tBs676FfzjpVWwNOBa/NduDczdGAGUHdLmd67xbvFLUOEyI35 pSz5+LCMRdzzljXC068womITkk7ozLxH5T3+0JohxZIunhBOITQEGM7okiH0VW5lHSU3 MPI4aaMN53SpTf6XUApffLj9DNn6W1WOxT6iPtwlA6vn8DlKpMIlY7PG7ACn5IYJl/Eb VF4hyc0UJFBJi+OLGqtpdE1C3TfFAYepbKoZXDLzzO7I/9BNf7dzlS7F7gGEhuESTwqn OWgg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1678818753; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=tga5shMHyjc0TB9sW2hC3q3qhXdH8jnVVmQsmewODbI=; b=hUgz5Ez7i0B5Ns0zawzepAfivOsH/A+RId7bICXJjUXwJECNAMjdYxCWbt+yxB0HUg 8ohO7+D+hkSWGY2F1HphoKRWLUASXfKK7PiEvo/gHDIKFOekbtIZTcRMgiMzOc7NCRx+ GFc3B4JQRgb7w0SRUH2YcuzAXM/B/krA7W0WFIGrcPZAjAsSh70X9gmjYOKI0SbXDokR dtbXnhmzCfO2Q4SyGGYmagsQkVEk9ILBHtNcwh88A8oY3sV7Mm6Ifu8O1iuQW166bVQO Hf56iwAjHXPSxPOdXS/2kjGR5O8lZLYmq6/xF4cjkme1OL6s7pILti8Fkk/a+U9ethrs NSBQ== X-Gm-Message-State: AO0yUKVv37q2l36oEwzeKL0j4ZHS4VHDxTVibDjoamWLxMP8m2ZpgtGt zK8bUN22cxLNOUZp48xEEGtpWQ== X-Google-Smtp-Source: AK7set9v9zIkmYb9ikpU9Da6c1URJ3+lBxePQISeL3aEwed0jfmqNAFJLGyZD2Ib8PdTkZhgH0knog== X-Received: by 2002:a17:902:facf:b0:19e:dc0e:1269 with SMTP id ld15-20020a170902facf00b0019edc0e1269mr55859plb.7.1678818752836; Tue, 14 Mar 2023 11:32:32 -0700 (PDT) Received: from evan.ba.rivosinc.com ([66.220.2.162]) by smtp.gmail.com with ESMTPSA id kn3-20020a170903078300b0019d397b0f18sm2072019plb.214.2023.03.14.11.32.31 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 14 Mar 2023 11:32:32 -0700 (PDT) From: Evan Green To: Palmer Dabbelt Cc: slewis@rivosinc.com, heiko@sntech.de, Conor Dooley , vineetg@rivosinc.com, Evan Green , Conor Dooley , Albert Ou , Andrew Jones , Anup Patel , Atish Patra , Dao Lu , Mayuresh Chitale , Palmer Dabbelt , Paul Walmsley , Randy Dunlap , Sunil V L , linux-kernel@vger.kernel.org, linux-riscv@lists.infradead.org Subject: [PATCH v4 1/6] RISC-V: Move struct riscv_cpuinfo to new header Date: Tue, 14 Mar 2023 11:32:15 -0700 Message-Id: <20230314183220.513101-2-evan@rivosinc.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20230314183220.513101-1-evan@rivosinc.com> References: <20230314183220.513101-1-evan@rivosinc.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" In preparation for tracking and exposing microarchitectural details to userspace (like whether or not unaligned accesses are fast), move the riscv_cpuinfo struct out to its own new cpufeatures.h header. It will need to be used by more than just cpu.c. Signed-off-by: Evan Green Reviewed-by: Conor Dooley Reviewed-by: Heiko Stuebner Tested-by: Heiko Stuebner --- (no changes since v3) Changes in v3: - Updated copyright date in cpufeature.h - Fixed typo in cpufeature.h comment (Conor) Changes in v2: - Factored the move of struct riscv_cpuinfo to its own header arch/riscv/include/asm/cpufeature.h | 21 +++++++++++++++++++++ arch/riscv/kernel/cpu.c | 8 ++------ 2 files changed, 23 insertions(+), 6 deletions(-) create mode 100644 arch/riscv/include/asm/cpufeature.h diff --git a/arch/riscv/include/asm/cpufeature.h b/arch/riscv/include/asm/c= pufeature.h new file mode 100644 index 000000000000..66ebaae449c8 --- /dev/null +++ b/arch/riscv/include/asm/cpufeature.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright 2022-2023 Rivos, Inc + */ + +#ifndef _ASM_CPUFEATURE_H +#define _ASM_CPUFEATURE_H + +/* + * These are probed via a device_initcall(), via either the SBI or directly + * from the corresponding CSRs. + */ +struct riscv_cpuinfo { + unsigned long mvendorid; + unsigned long marchid; + unsigned long mimpid; +}; + +DECLARE_PER_CPU(struct riscv_cpuinfo, riscv_cpuinfo); + +#endif diff --git a/arch/riscv/kernel/cpu.c b/arch/riscv/kernel/cpu.c index 1b9a5a66e55a..684e5419d37d 100644 --- a/arch/riscv/kernel/cpu.c +++ b/arch/riscv/kernel/cpu.c @@ -7,6 +7,7 @@ #include #include #include +#include #include #include #include @@ -70,12 +71,7 @@ int riscv_of_parent_hartid(struct device_node *node, uns= igned long *hartid) return -1; } =20 -struct riscv_cpuinfo { - unsigned long mvendorid; - unsigned long marchid; - unsigned long mimpid; -}; -static DEFINE_PER_CPU(struct riscv_cpuinfo, riscv_cpuinfo); +DEFINE_PER_CPU(struct riscv_cpuinfo, riscv_cpuinfo); =20 unsigned long riscv_cached_mvendorid(unsigned int cpu_id) { --=20 2.25.1 From nobody Sun Feb 8 05:23:07 2026 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 5D9C9C6FD1F for ; Tue, 14 Mar 2023 18:34:59 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230475AbjCNSen (ORCPT ); Tue, 14 Mar 2023 14:34:43 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:49850 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229590AbjCNSea (ORCPT ); Tue, 14 Mar 2023 14:34:30 -0400 Received: from mail-pl1-x62f.google.com (mail-pl1-x62f.google.com [IPv6:2607:f8b0:4864:20::62f]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 6908B769D for ; Tue, 14 Mar 2023 11:33:52 -0700 (PDT) Received: by mail-pl1-x62f.google.com with SMTP id p6so17613455plf.0 for ; Tue, 14 Mar 2023 11:33:52 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rivosinc-com.20210112.gappssmtp.com; s=20210112; t=1678818758; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=OcmFf0A4Qs8AV2mYWu4nAZTRoGwpoQN38ASi1el4j1g=; b=NG91jaugy2VVTwA/f47s723t29ASCtpUInf9ILIhHoYmCxz7yVWpoOK1RW8A+bhVIk syGM8TndHkbxeZ33UFqZ/eq8em6AVd8xKXZlJjIxi1drDb3I6YljoHiDLmzmoVkVBnek pyp6/taiZw474WsFb1i/vQHmLYjJ5kr0O2y1t9E1DAFrSN9oCIvw4x0czZktJ09fR+PD iKRHUQK+h8QwujdqvDEaV4Y66FkTVs9e4eLEahJRemJGR+owe7JVBXAA2pSOsA9rMPCr V3Gs9U/7KFbw5nNY5P8uGOYGsGnVgeI7NrRNbLNLZReiknUGvBim4QoculpP9Em99IXu ePNw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1678818758; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=OcmFf0A4Qs8AV2mYWu4nAZTRoGwpoQN38ASi1el4j1g=; b=i7OLDApf3qEOcrL9VFq9md2k1HKKLX22x84Pl+k31lEu2U4uJcMFybT8C1HXXjkWl1 BkFJgCGgVbK2sJ/etJbXAzSYmWnj9lKiwiO5V/bh/Qpt0W+LhQIsSqI0E6BG+0S5o32d 2aeb8BBWSCAZYnFPMruoUjvTW++p4MOLUZq/gkGESg5H1MT4Lp9WhHx8a8M076mHW7fO DYU2b0zx06Y6vPfqqt0WMyRBb9DkLoDYe4Nsuvu/Jppz9x5ONcXnUBP0RalMvRBZZwW6 fMgH5TqNwMzDh2sXikEY5JUxSxQgm6O7/qI644+4YSnogDPb6Vhgjwa0RlI6eYjU6Ju9 5dpw== X-Gm-Message-State: AO0yUKUugMIqITwBwD+/XHORs6usu9W+DNRTBDefda8h9BWtsCowFulh uto7qv9zSWAbPj9josUsKbJqgg== X-Google-Smtp-Source: AK7set8lOG/soxr3HbCDivUIzfs2ntoxWQ2RiARC8CqSbkFzXcbelJYmWHm/+gHYwFrVACKM+sJqqQ== X-Received: by 2002:a17:902:d486:b0:19c:a866:6a76 with SMTP id c6-20020a170902d48600b0019ca8666a76mr346plg.42.1678818757848; Tue, 14 Mar 2023 11:32:37 -0700 (PDT) Received: from evan.ba.rivosinc.com ([66.220.2.162]) by smtp.gmail.com with ESMTPSA id kn3-20020a170903078300b0019d397b0f18sm2072019plb.214.2023.03.14.11.32.36 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 14 Mar 2023 11:32:37 -0700 (PDT) From: Evan Green To: Palmer Dabbelt Cc: slewis@rivosinc.com, heiko@sntech.de, Conor Dooley , vineetg@rivosinc.com, Evan Green , Conor Dooley , Albert Ou , Andrew Bresticker , Arnd Bergmann , Atish Patra , Bagas Sanjaya , Celeste Liu , Guo Ren , Jonathan Corbet , Palmer Dabbelt , Paul Walmsley , Tobias Klauser , linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, linux-riscv@lists.infradead.org Subject: [PATCH v4 2/6] RISC-V: Add a syscall for HW probing Date: Tue, 14 Mar 2023 11:32:16 -0700 Message-Id: <20230314183220.513101-3-evan@rivosinc.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20230314183220.513101-1-evan@rivosinc.com> References: <20230314183220.513101-1-evan@rivosinc.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" We don't have enough space for these all in ELF_HWCAP{,2} and there's no system call that quite does this, so let's just provide an arch-specific one to probe for hardware capabilities. This currently just provides m{arch,imp,vendor}id, but with the key-value pairs we can pass more in the future. Co-developed-by: Palmer Dabbelt Signed-off-by: Palmer Dabbelt Signed-off-by: Evan Green Reviewed-by: Conor Dooley Reviewed-by: Heiko Stuebner Tested-by: Heiko Stuebner --- Changes in v4: - Used real types in syscall prototypes (Arnd) - Fixed static line break in do_riscv_hwprobe() (Conor) - Added newlines between documentation lists (Conor) - Crispen up size types to size_t, and cpu indices to int (Joe) - Fix copy_from_user() return logic bug (found via kselftests!) - Add __user to SYSCALL_DEFINE() to fix warning Changes in v3: - Refactored functions so that kernel mode can query too, in preparation for the vDSO data population. - Changed the vendor/arch/imp IDs to return a value of -1 on mismatch rather than failing the whole call. - Const cpumask pointer in hwprobe_mid() - Embellished documentation WRT cpu_set and the returned values. - Renamed hwprobe_mid() to hwprobe_arch_id() (Conor) - Fixed machine ID doc warnings, changed elements to c:macro:. - Completed dangling unistd.h comment (Conor) - Fixed line breaks and minor logic optimization (Conor). - Use riscv_cached_mxxxid() (Conor) Changes in v2: - Changed the interface to look more like poll(). Rather than supplying key_offset and getting back an array of values with numerically contiguous keys, have the user pre-fill the key members of the array, and the kernel will fill in the corresponding values. For any key it doesn't recognize, it will set the key of that element to -1. This allows usermode to quickly ask for exactly the elements it cares about, and not get bogged down in a back and forth about newer keys that older kernels might not recognize. In other words, the kernel can communicate that it doesn't recognize some of the keys while still providing the data for the keys it does know. - Added a shortcut to the cpuset parameters that if a size of 0 and NULL is provided for the CPU set, the kernel will use a cpu mask of all online CPUs. This is convenient because I suspect most callers will only want to act on a feature if it's supported on all CPUs, and it's a headache to dynamically allocate an array of all 1s, not to mention a waste to have the kernel loop over all of the offline bits. --- Documentation/riscv/hwprobe.rst | 41 ++++++++ Documentation/riscv/index.rst | 1 + arch/riscv/include/asm/hwprobe.h | 13 +++ arch/riscv/include/asm/syscall.h | 4 + arch/riscv/include/uapi/asm/hwprobe.h | 25 +++++ arch/riscv/include/uapi/asm/unistd.h | 9 ++ arch/riscv/kernel/sys_riscv.c | 136 +++++++++++++++++++++++++- 7 files changed, 228 insertions(+), 1 deletion(-) create mode 100644 Documentation/riscv/hwprobe.rst create mode 100644 arch/riscv/include/asm/hwprobe.h create mode 100644 arch/riscv/include/uapi/asm/hwprobe.h diff --git a/Documentation/riscv/hwprobe.rst b/Documentation/riscv/hwprobe.= rst new file mode 100644 index 000000000000..211828f706e3 --- /dev/null +++ b/Documentation/riscv/hwprobe.rst @@ -0,0 +1,41 @@ +.. SPDX-License-Identifier: GPL-2.0 + +RISC-V Hardware Probing Interface +--------------------------------- + +The RISC-V hardware probing interface is based around a single syscall, wh= ich +is defined in :: + + struct riscv_hwprobe { + __s64 key; + __u64 value; + }; + + long sys_riscv_hwprobe(struct riscv_hwprobe *pairs, size_t pair_count, + size_t cpu_count, cpu_set_t *cpus, + unsigned int flags); + +The arguments are split into three groups: an array of key-value pairs, a = CPU +set, and some flags. The key-value pairs are supplied with a count. Usersp= ace +must prepopulate the key field for each element, and the kernel will fill = in the +value if the key is recognized. If a key is unknown to the kernel, its key= field +will be cleared to -1, and its value set to 0. The CPU set is defined by +CPU_SET(3). For value-like keys (eg. vendor/arch/impl), the returned value= will +be only be valid if all CPUs in the given set have the same value. Otherwi= se -1 +will be returned. For boolean-like keys, the value returned will be a logi= cal +AND of the values for the specified CPUs. Usermode can supply NULL for cpu= s and +0 for cpu_count as a shortcut for all online CPUs. There are currently no = flags, +this value must be zero for future compatibility. + +On success 0 is returned, on failure a negative error code is returned. + +The following keys are defined: + +* :c:macro:`RISCV_HWPROBE_KEY_MVENDORID`: Contains the value of ``mvendori= d``, + as defined by the RISC-V privileged architecture specification. + +* :c:macro:`RISCV_HWPROBE_KEY_MARCHID`: Contains the value of ``marchid``,= as + defined by the RISC-V privileged architecture specification. + +* :c:macro:`RISCV_HWPROBE_KEY_MIMPLID`: Contains the value of ``mimplid``,= as + defined by the RISC-V privileged architecture specification. diff --git a/Documentation/riscv/index.rst b/Documentation/riscv/index.rst index 2e5b18fbb145..175a91db0200 100644 --- a/Documentation/riscv/index.rst +++ b/Documentation/riscv/index.rst @@ -7,6 +7,7 @@ RISC-V architecture =20 boot-image-header vm-layout + hwprobe patch-acceptance uabi =20 diff --git a/arch/riscv/include/asm/hwprobe.h b/arch/riscv/include/asm/hwpr= obe.h new file mode 100644 index 000000000000..08d1c3bdd78a --- /dev/null +++ b/arch/riscv/include/asm/hwprobe.h @@ -0,0 +1,13 @@ +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +/* + * Copyright 2022 Rivos, Inc + */ + +#ifndef _ASM_HWPROBE_H +#define _ASM_HWPROBE_H + +#include + +#define RISCV_HWPROBE_MAX_KEY 2 + +#endif diff --git a/arch/riscv/include/asm/syscall.h b/arch/riscv/include/asm/sysc= all.h index 384a63b86420..3b5a667928bf 100644 --- a/arch/riscv/include/asm/syscall.h +++ b/arch/riscv/include/asm/syscall.h @@ -10,6 +10,7 @@ #ifndef _ASM_RISCV_SYSCALL_H #define _ASM_RISCV_SYSCALL_H =20 +#include #include #include #include @@ -75,4 +76,7 @@ static inline int syscall_get_arch(struct task_struct *ta= sk) } =20 asmlinkage long sys_riscv_flush_icache(uintptr_t, uintptr_t, uintptr_t); + +asmlinkage long sys_riscv_hwprobe(struct riscv_hwprobe *, size_t, size_t, + unsigned long *, unsigned int); #endif /* _ASM_RISCV_SYSCALL_H */ diff --git a/arch/riscv/include/uapi/asm/hwprobe.h b/arch/riscv/include/uap= i/asm/hwprobe.h new file mode 100644 index 000000000000..591802047460 --- /dev/null +++ b/arch/riscv/include/uapi/asm/hwprobe.h @@ -0,0 +1,25 @@ +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +/* + * Copyright 2022 Rivos, Inc + */ + +#ifndef _UAPI_ASM_HWPROBE_H +#define _UAPI_ASM_HWPROBE_H + +#include + +/* + * Interface for probing hardware capabilities from userspace, see + * Documentation/riscv/hwprobe.rst for more information. + */ +struct riscv_hwprobe { + __s64 key; + __u64 value; +}; + +#define RISCV_HWPROBE_KEY_MVENDORID 0 +#define RISCV_HWPROBE_KEY_MARCHID 1 +#define RISCV_HWPROBE_KEY_MIMPID 2 +/* Increase RISCV_HWPROBE_MAX_KEY when adding items. */ + +#endif diff --git a/arch/riscv/include/uapi/asm/unistd.h b/arch/riscv/include/uapi= /asm/unistd.h index 73d7cdd2ec49..950ab3fd4409 100644 --- a/arch/riscv/include/uapi/asm/unistd.h +++ b/arch/riscv/include/uapi/asm/unistd.h @@ -43,3 +43,12 @@ #define __NR_riscv_flush_icache (__NR_arch_specific_syscall + 15) #endif __SYSCALL(__NR_riscv_flush_icache, sys_riscv_flush_icache) + +/* + * Allows userspace to query the kernel for CPU architecture and + * microarchitecture details across a given set of CPUs. + */ +#ifndef __NR_riscv_hwprobe +#define __NR_riscv_hwprobe (__NR_arch_specific_syscall + 14) +#endif +__SYSCALL(__NR_riscv_hwprobe, sys_riscv_hwprobe) diff --git a/arch/riscv/kernel/sys_riscv.c b/arch/riscv/kernel/sys_riscv.c index 5d3f2fbeb33c..981d23457f13 100644 --- a/arch/riscv/kernel/sys_riscv.c +++ b/arch/riscv/kernel/sys_riscv.c @@ -6,8 +6,11 @@ */ =20 #include -#include #include +#include +#include +#include +#include #include =20 static long riscv_sys_mmap(unsigned long addr, unsigned long len, @@ -69,3 +72,134 @@ SYSCALL_DEFINE3(riscv_flush_icache, uintptr_t, start, u= intptr_t, end, =20 return 0; } + +/* + * The hwprobe interface, for allowing userspace to probe to see which fea= tures + * are supported by the hardware. See Documentation/riscv/hwprobe.rst for= more + * details. + */ +static void hwprobe_arch_id(struct riscv_hwprobe *pair, + const struct cpumask *cpus) +{ + u64 id =3D -1ULL; + bool first =3D true; + int cpu; + + for_each_cpu(cpu, cpus) { + u64 cpu_id; + + switch (pair->key) { + case RISCV_HWPROBE_KEY_MVENDORID: + cpu_id =3D riscv_cached_mvendorid(cpu); + break; + case RISCV_HWPROBE_KEY_MIMPID: + cpu_id =3D riscv_cached_mimpid(cpu); + break; + case RISCV_HWPROBE_KEY_MARCHID: + cpu_id =3D riscv_cached_marchid(cpu); + break; + } + + if (first) + id =3D cpu_id; + + /* + * If there's a mismatch for the given set, return -1 in the + * value. + */ + if (id !=3D cpu_id) { + id =3D -1ULL; + break; + } + } + + pair->value =3D id; +} + +static void hwprobe_one_pair(struct riscv_hwprobe *pair, + const struct cpumask *cpus) +{ + switch (pair->key) { + case RISCV_HWPROBE_KEY_MVENDORID: + case RISCV_HWPROBE_KEY_MARCHID: + case RISCV_HWPROBE_KEY_MIMPID: + hwprobe_arch_id(pair, cpus); + break; + + /* + * For forward compatibility, unknown keys don't fail the whole + * call, but get their element key set to -1 and value set to 0 + * indicating they're unrecognized. + */ + default: + pair->key =3D -1; + pair->value =3D 0; + break; + } +} + +static int do_riscv_hwprobe(struct riscv_hwprobe __user *pairs, + size_t pair_count, size_t cpu_count, + unsigned long __user *cpus_user, + unsigned int flags) +{ + size_t out; + int ret; + cpumask_t cpus; + + /* Check the reserved flags. */ + if (flags !=3D 0) + return -EINVAL; + + /* + * The interface supports taking in a CPU mask, and returns values that + * are consistent across that mask. Allow userspace to specify NULL and + * 0 as a shortcut to all online CPUs. + */ + cpumask_clear(&cpus); + if (!cpu_count && !cpus_user) { + cpumask_copy(&cpus, cpu_online_mask); + } else { + if (cpu_count > cpumask_size()) + cpu_count =3D cpumask_size(); + + ret =3D copy_from_user(&cpus, cpus_user, cpu_count); + if (ret) + return -EFAULT; + + /* + * Userspace must provide at least one online CPU, without that + * there's no way to define what is supported. + */ + cpumask_and(&cpus, &cpus, cpu_online_mask); + if (cpumask_empty(&cpus)) + return -EINVAL; + } + + for (out =3D 0; out < pair_count; out++, pairs++) { + struct riscv_hwprobe pair; + + if (get_user(pair.key, &pairs->key)) + return -EFAULT; + + pair.value =3D 0; + hwprobe_one_pair(&pair, &cpus); + ret =3D put_user(pair.key, &pairs->key); + if (ret =3D=3D 0) + ret =3D put_user(pair.value, &pairs->value); + + if (ret) + return -EFAULT; + } + + return 0; + +} + +SYSCALL_DEFINE5(riscv_hwprobe, struct riscv_hwprobe __user *, pairs, + size_t, pair_count, size_t, cpu_count, unsigned long __user *, + cpus, unsigned int, flags) +{ + return do_riscv_hwprobe(pairs, pair_count, cpu_count, + cpus, flags); +} --=20 2.25.1 From nobody Sun Feb 8 05:23:07 2026 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 103CAC05027 for ; Tue, 14 Mar 2023 18:35:05 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230310AbjCNSfC (ORCPT ); Tue, 14 Mar 2023 14:35:02 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:49808 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230350AbjCNSek (ORCPT ); Tue, 14 Mar 2023 14:34:40 -0400 Received: from mail-pl1-x62f.google.com (mail-pl1-x62f.google.com [IPv6:2607:f8b0:4864:20::62f]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 92F7158B68 for ; Tue, 14 Mar 2023 11:34:01 -0700 (PDT) Received: by mail-pl1-x62f.google.com with SMTP id y11so17590614plg.1 for ; Tue, 14 Mar 2023 11:34:01 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rivosinc-com.20210112.gappssmtp.com; s=20210112; t=1678818763; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=4gg5yhIsRAGMuaeMyFJ06s8c0oFC+MQHoWhWknilX1Q=; b=sBtSlB+lZsAsGllgiJ9SAhM2pJxvABPKjHR+PXydOZfl00x7fYbhWEmek9Rl+3Cj/U zhtdv1At3d4IKBkkDGSH1nrM6LSJmK5yfjW+MikxAxwAhV1CbOtVLuQX6spjHh2S2c8c Dc+00eTZ/QiwzmtgQtwTBap+1T3oNzFhy9y84DMOFYAgtvaEnJCY+ApCe4QtYI1a3wwI shLW+59aonY8eqW7atE8CnmCA6TqaQwfnMis3ypnEloX4AD1WeecLhfC8L3VHDEVIToV 46zBzULSp7/ozEichcyM44IC3dsFCz3KKw3ZdS08bdtgVgq7N1TS3i+70lg1N93iXSv5 i7fA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1678818763; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=4gg5yhIsRAGMuaeMyFJ06s8c0oFC+MQHoWhWknilX1Q=; b=cGbgk+hk7ktTSEPZyQZGSn2fx9vnHqg2uy0wRvoeXaFmq2ZvNbN1/I2v7XD/JeSYLU O7lcdATGQL4OdlSi8Tklq2KXVtR58svZAUGc2xSnZFVqKST0F0T2j2Py2HjNl2xGMboI DyzpcF2PbeNejiBymRKqTAI57l0ItMXnmhMK/2tpG8Jguk4t4G9bpusT0awI+/tsvhcT 30YOKFaUA8797OSa/XJbx2bhmrGnEtPKtNMb7B8GwLJ9T2XSFuJyxcX8CG/iVsTHERTI UaEIbgEQplhy/0jj59Aijsmtfh4PB5Qo5Zw5HkoBb6+hA9pvTwHc+1HOYxGpC07BvxKt BC3A== X-Gm-Message-State: AO0yUKWc5h5iMi0129Bpk4m3OqRjYwUDmNMbWg7wOkh8XOa0/Yt0kkjc dXwI6a0vvUKpXSGy9NOkqOuGMw== X-Google-Smtp-Source: AK7set9+xqikL2wUga0QqTbqna0f4KaIShGQoOHrfjpuM1uSkcKKKzH0akHZ3/0/gSHEVEyio7wiTg== X-Received: by 2002:a17:903:1105:b0:19e:6c02:801c with SMTP id n5-20020a170903110500b0019e6c02801cmr48114plh.14.1678818763557; Tue, 14 Mar 2023 11:32:43 -0700 (PDT) Received: from evan.ba.rivosinc.com ([66.220.2.162]) by smtp.gmail.com with ESMTPSA id kn3-20020a170903078300b0019d397b0f18sm2072019plb.214.2023.03.14.11.32.42 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 14 Mar 2023 11:32:43 -0700 (PDT) From: Evan Green To: Palmer Dabbelt Cc: slewis@rivosinc.com, heiko@sntech.de, Conor Dooley , vineetg@rivosinc.com, Evan Green , Conor Dooley , Albert Ou , Andrew Bresticker , Celeste Liu , Guo Ren , Jonathan Corbet , Palmer Dabbelt , Paul Walmsley , linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, linux-riscv@lists.infradead.org Subject: [PATCH v4 3/6] RISC-V: hwprobe: Add support for RISCV_HWPROBE_BASE_BEHAVIOR_IMA Date: Tue, 14 Mar 2023 11:32:17 -0700 Message-Id: <20230314183220.513101-4-evan@rivosinc.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20230314183220.513101-1-evan@rivosinc.com> References: <20230314183220.513101-1-evan@rivosinc.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" We have an implicit set of base behaviors that userspace depends on, which are mostly defined in various ISA specifications. Co-developed-by: Palmer Dabbelt Signed-off-by: Palmer Dabbelt Signed-off-by: Evan Green Reviewed-by: Conor Dooley Reviewed-by: Heiko Stuebner Tested-by: Heiko Stuebner --- Changes in v4: - More newlines in BASE_BEHAVIOR_IMA documentation (Conor) Changes in v3: - Refactored base ISA behavior probe to allow kernel probing as well, in prep for vDSO data initialization. - Fixed doc warnings in IMA text list, use :c:macro:. Documentation/riscv/hwprobe.rst | 24 ++++++++++++++++++++++++ arch/riscv/include/asm/hwprobe.h | 2 +- arch/riscv/include/uapi/asm/hwprobe.h | 5 +++++ arch/riscv/kernel/sys_riscv.c | 20 ++++++++++++++++++++ 4 files changed, 50 insertions(+), 1 deletion(-) diff --git a/Documentation/riscv/hwprobe.rst b/Documentation/riscv/hwprobe.= rst index 211828f706e3..945d44683c40 100644 --- a/Documentation/riscv/hwprobe.rst +++ b/Documentation/riscv/hwprobe.rst @@ -39,3 +39,27 @@ The following keys are defined: =20 * :c:macro:`RISCV_HWPROBE_KEY_MIMPLID`: Contains the value of ``mimplid``,= as defined by the RISC-V privileged architecture specification. + +* :c:macro:`RISCV_HWPROBE_KEY_BASE_BEHAVIOR`: A bitmask containing the base + user-visible behavior that this kernel supports. The following base use= r ABIs + are defined: + + * :c:macro:`RISCV_HWPROBE_BASE_BEHAVIOR_IMA`: Support for rv32ima or + rv64ima, as defined by version 2.2 of the user ISA and version 1.10 of= the + privileged ISA, with the following known exceptions (more exceptions m= ay be + added, but only if it can be demonstrated that the user ABI is not bro= ken): + + * The :fence.i: instruction cannot be directly executed by userspace + programs (it may still be executed in userspace via a + kernel-controlled mechanism such as the vDSO). + +* :c:macro:`RISCV_HWPROBE_KEY_IMA_EXT_0`: A bitmask containing the extensi= ons + that are compatible with the :c:macro:`RISCV_HWPROBE_BASE_BEHAVIOR_IMA`: + base system behavior. + + * :c:macro:`RISCV_HWPROBE_IMA_FD`: The F and D extensions are supported,= as + defined by commit cd20cee ("FMIN/FMAX now implement + minimumNumber/maximumNumber, not minNum/maxNum") of the RISC-V ISA man= ual. + + * :c:macro:`RISCV_HWPROBE_IMA_C`: The C extension is supported, as defin= ed + by version 2.2 of the RISC-V ISA manual. diff --git a/arch/riscv/include/asm/hwprobe.h b/arch/riscv/include/asm/hwpr= obe.h index 08d1c3bdd78a..7e52f1e1fe10 100644 --- a/arch/riscv/include/asm/hwprobe.h +++ b/arch/riscv/include/asm/hwprobe.h @@ -8,6 +8,6 @@ =20 #include =20 -#define RISCV_HWPROBE_MAX_KEY 2 +#define RISCV_HWPROBE_MAX_KEY 4 =20 #endif diff --git a/arch/riscv/include/uapi/asm/hwprobe.h b/arch/riscv/include/uap= i/asm/hwprobe.h index 591802047460..fc5665411782 100644 --- a/arch/riscv/include/uapi/asm/hwprobe.h +++ b/arch/riscv/include/uapi/asm/hwprobe.h @@ -20,6 +20,11 @@ struct riscv_hwprobe { #define RISCV_HWPROBE_KEY_MVENDORID 0 #define RISCV_HWPROBE_KEY_MARCHID 1 #define RISCV_HWPROBE_KEY_MIMPID 2 +#define RISCV_HWPROBE_KEY_BASE_BEHAVIOR 3 +#define RISCV_HWPROBE_BASE_BEHAVIOR_IMA (1 << 0) +#define RISCV_HWPROBE_KEY_IMA_EXT_0 4 +#define RISCV_HWPROBE_IMA_FD (1 << 0) +#define RISCV_HWPROBE_IMA_C (1 << 1) /* Increase RISCV_HWPROBE_MAX_KEY when adding items. */ =20 #endif diff --git a/arch/riscv/kernel/sys_riscv.c b/arch/riscv/kernel/sys_riscv.c index 981d23457f13..1c118438b1b3 100644 --- a/arch/riscv/kernel/sys_riscv.c +++ b/arch/riscv/kernel/sys_riscv.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -125,6 +126,25 @@ static void hwprobe_one_pair(struct riscv_hwprobe *pai= r, case RISCV_HWPROBE_KEY_MIMPID: hwprobe_arch_id(pair, cpus); break; + /* + * The kernel already assumes that the base single-letter ISA + * extensions are supported on all harts, and only supports the + * IMA base, so just cheat a bit here and tell that to + * userspace. + */ + case RISCV_HWPROBE_KEY_BASE_BEHAVIOR: + pair->value =3D RISCV_HWPROBE_BASE_BEHAVIOR_IMA; + break; + + case RISCV_HWPROBE_KEY_IMA_EXT_0: + pair->value =3D 0; + if (has_fpu()) + pair->value |=3D RISCV_HWPROBE_IMA_FD; + + if (elf_hwcap & RISCV_ISA_EXT_c) + pair->value |=3D RISCV_HWPROBE_IMA_C; + + break; =20 /* * For forward compatibility, unknown keys don't fail the whole --=20 2.25.1 From nobody Sun Feb 8 05:23:07 2026 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 607FCC05027 for ; Tue, 14 Mar 2023 18:35:09 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230408AbjCNSfH (ORCPT ); Tue, 14 Mar 2023 14:35:07 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:50272 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230395AbjCNSel (ORCPT ); Tue, 14 Mar 2023 14:34:41 -0400 Received: from mail-pj1-x102a.google.com (mail-pj1-x102a.google.com [IPv6:2607:f8b0:4864:20::102a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 8109B9F231 for ; Tue, 14 Mar 2023 11:34:06 -0700 (PDT) Received: by mail-pj1-x102a.google.com with SMTP id u3-20020a17090a450300b00239db6d7d47so16070989pjg.4 for ; Tue, 14 Mar 2023 11:34:06 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rivosinc-com.20210112.gappssmtp.com; s=20210112; t=1678818768; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=QKTjl3uMnu6bKpkbPPZuKBQ5mmKPjtDK6wHW5Zya6zw=; b=39YY02HqSJn+6mUv2/JWrTbXr/QuCXFqwwq4KpWilvQ17c6oM5SFIHSqw7qNcQikul Cq6qqNqCXlKGtDijf3gQvF48xXDO4BcTn0ldNXqVRuDQYaJ2dNqpT94Z0WskSsc9wHbw WMcJB/FbEvCFg+RTEx6j80CllH/jY9++BqJIuqxIPUC6ep1GzZKflbBa7udPGjSLeZAm PRemcyNRZM5qi6Gdx3ehAwQx5Ii5tpP1vbBTNwHrZgNJ6zHSee0J+i1SZAlo18E7/kYB uv+AJ1SGlh/S4pAUEXFTWtWQFL+OLsM9cwy9zZpfd91TYG5FespyxNJgL5Rj9v1tCp1d gkTw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1678818768; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=QKTjl3uMnu6bKpkbPPZuKBQ5mmKPjtDK6wHW5Zya6zw=; b=S/Yicr1cKfcW/H99Z39yPI//KcGk58qs4dZh6jVOEfmUsgu359VRgEYFn6vSfDNfss kj2Ih/A1lXRQYO7yQXPo/ue+oLmBca1okzsHRjQibitLsOmbW0hU0cuZ8KAqSQ5gosjY zAc+ECkQVNtCvBM5jqmBLgj1vjfnqtyoph2ac6M6O97ONdKHXaY8T90uei79W9Vq/+pO A6qypHCGWsvaL9bK8ty2bfKSd3boYOD+qktXPfyktL8Xi/gJro0l3bXIBHaVMwOnwxQc Om1QEoqPJbiR59W2nYJ+6wfI+soEUGuCdYUouWPrQJ4JH7kticjAibaCRVvsn6iWM80m 8JuQ== X-Gm-Message-State: AO0yUKXsz23mLzLy/ynGDSyk0WMHMZv//W9TSQCOrQZzrPXnGDNyBGPW CtdMC0wHOF72ZaVH+e46AaD/AQ== X-Google-Smtp-Source: AK7set+ry6vHcj3C81Rix70L4IbdTgGhEeKDGXvyF/M7BBOHf56MAKJDt0fcqjyPN/zr2uWing1KqA== X-Received: by 2002:a17:902:d4cf:b0:19e:500b:517a with SMTP id o15-20020a170902d4cf00b0019e500b517amr45847871plg.69.1678818768035; Tue, 14 Mar 2023 11:32:48 -0700 (PDT) Received: from evan.ba.rivosinc.com ([66.220.2.162]) by smtp.gmail.com with ESMTPSA id kn3-20020a170903078300b0019d397b0f18sm2072019plb.214.2023.03.14.11.32.46 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 14 Mar 2023 11:32:47 -0700 (PDT) From: Evan Green To: Palmer Dabbelt Cc: slewis@rivosinc.com, heiko@sntech.de, Conor Dooley , vineetg@rivosinc.com, Evan Green , Albert Ou , Andrew Bresticker , Andrew Jones , Anup Patel , Atish Patra , Celeste Liu , Conor Dooley , Guo Ren , Jisheng Zhang , Jonathan Corbet , Palmer Dabbelt , Paul Walmsley , Philipp Tomsich , Sudeep Holla , Sunil V L , Tsukasa OI , Wei Fu , linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, linux-riscv@lists.infradead.org Subject: [PATCH v4 4/6] RISC-V: hwprobe: Support probing of misaligned access performance Date: Tue, 14 Mar 2023 11:32:18 -0700 Message-Id: <20230314183220.513101-5-evan@rivosinc.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20230314183220.513101-1-evan@rivosinc.com> References: <20230314183220.513101-1-evan@rivosinc.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" This allows userspace to select various routines to use based on the performance of misaligned access on the target hardware. Co-developed-by: Palmer Dabbelt Signed-off-by: Palmer Dabbelt Signed-off-by: Evan Green Reviewed-by: Conor Dooley Reviewed-by: Heiko Stuebner Tested-by: Heiko Stuebner --- Changes in v4: - Add newlines to CPUPERF_0 documentation (Conor) - Add UNSUPPORTED value (Conor) - Switched from DT to alternatives-based probing (Rob) - Crispen up cpu index type to always be int (Conor) Changes in v3: - Have hwprobe_misaligned return int instead of long. - Constify cpumask pointer in hwprobe_misaligned() - Fix warnings in _PERF_O list documentation, use :c:macro:. - Move include cpufeature.h to misaligned patch. - Fix documentation mismatch for RISCV_HWPROBE_KEY_CPUPERF_0 (Conor) - Use for_each_possible_cpu() instead of NR_CPUS (Conor) - Break early in misaligned access iteration (Conor) - Increase MISALIGNED_MASK from 2 bits to 3 for possible UNSUPPORTED future value (Conor) Changes in v2: - Fixed logic error in if(of_property_read_string...) that caused crash - Include cpufeature.h in cpufeature.h to avoid undeclared variable warning. - Added a _MASK define - Fix random checkpatch complaints Documentation/riscv/hwprobe.rst | 21 ++++++++++++++++++++ arch/riscv/errata/thead/errata.c | 9 +++++++++ arch/riscv/include/asm/alternative.h | 5 +++++ arch/riscv/include/asm/cpufeature.h | 2 ++ arch/riscv/include/asm/hwprobe.h | 2 +- arch/riscv/include/uapi/asm/hwprobe.h | 7 +++++++ arch/riscv/kernel/alternative.c | 19 ++++++++++++++++++ arch/riscv/kernel/cpufeature.c | 3 +++ arch/riscv/kernel/smpboot.c | 1 + arch/riscv/kernel/sys_riscv.c | 28 +++++++++++++++++++++++++++ 10 files changed, 96 insertions(+), 1 deletion(-) diff --git a/Documentation/riscv/hwprobe.rst b/Documentation/riscv/hwprobe.= rst index 945d44683c40..9f0dd62dcb5d 100644 --- a/Documentation/riscv/hwprobe.rst +++ b/Documentation/riscv/hwprobe.rst @@ -63,3 +63,24 @@ The following keys are defined: =20 * :c:macro:`RISCV_HWPROBE_IMA_C`: The C extension is supported, as defin= ed by version 2.2 of the RISC-V ISA manual. + +* :c:macro:`RISCV_HWPROBE_KEY_CPUPERF_0`: A bitmask that contains performa= nce + information about the selected set of processors. + + * :c:macro:`RISCV_HWPROBE_MISALIGNED_UNKNOWN`: The performance of misali= gned + accesses is unknown. + + * :c:macro:`RISCV_HWPROBE_MISALIGNED_EMULATED`: Misaligned accesses are + emulated via software, either in or below the kernel. These accesses = are + always extremely slow. + + * :c:macro:`RISCV_HWPROBE_MISALIGNED_SLOW`: Misaligned accesses are supp= orted + in hardware, but are slower than the cooresponding aligned accesses + sequences. + + * :c:macro:`RISCV_HWPROBE_MISALIGNED_FAST`: Misaligned accesses are supp= orted + in hardware and are faster than the cooresponding aligned accesses + sequences. + + * :c:macro:`RISCV_HWPROBE_MISALIGNED_UNSUPPORTED`: Misaligned accesses a= re + not supported at all and will generate a misaligned address fault. diff --git a/arch/riscv/errata/thead/errata.c b/arch/riscv/errata/thead/err= ata.c index fac5742d1c1e..f41a45af5607 100644 --- a/arch/riscv/errata/thead/errata.c +++ b/arch/riscv/errata/thead/errata.c @@ -10,7 +10,9 @@ #include #include #include +#include #include +#include #include #include =20 @@ -108,3 +110,10 @@ void __init_or_module thead_errata_patch_func(struct a= lt_entry *begin, struct al if (stage =3D=3D RISCV_ALTERNATIVES_EARLY_BOOT) local_flush_icache_all(); } + +void thead_feature_probe_func(unsigned int cpu, unsigned long archid, + unsigned long impid) +{ + if ((archid =3D=3D 0) && (impid =3D=3D 0)) + per_cpu(misaligned_access_speed, cpu) =3D RISCV_HWPROBE_MISALIGNED_FAST; +} diff --git a/arch/riscv/include/asm/alternative.h b/arch/riscv/include/asm/= alternative.h index 6511dd73e812..7be6d4c6a27d 100644 --- a/arch/riscv/include/asm/alternative.h +++ b/arch/riscv/include/asm/alternative.h @@ -23,6 +23,7 @@ #define RISCV_ALTERNATIVES_MODULE 1 /* alternatives applied during module-= init */ #define RISCV_ALTERNATIVES_EARLY_BOOT 2 /* alternatives applied before mmu= start */ =20 +void probe_vendor_features(unsigned int cpu); void __init apply_boot_alternatives(void); void __init apply_early_boot_alternatives(void); void apply_module_alternatives(void *start, size_t length); @@ -47,11 +48,15 @@ void thead_errata_patch_func(struct alt_entry *begin, s= truct alt_entry *end, unsigned long archid, unsigned long impid, unsigned int stage); =20 +void thead_feature_probe_func(unsigned int cpu, unsigned long archid, + unsigned long impid); + void riscv_cpufeature_patch_func(struct alt_entry *begin, struct alt_entry= *end, unsigned int stage); =20 #else /* CONFIG_RISCV_ALTERNATIVE */ =20 +static inline void probe_vendor_features(unsigned int cpu) { } static inline void apply_boot_alternatives(void) { } static inline void apply_early_boot_alternatives(void) { } static inline void apply_module_alternatives(void *start, size_t length) {= } diff --git a/arch/riscv/include/asm/cpufeature.h b/arch/riscv/include/asm/c= pufeature.h index 66ebaae449c8..808d5403f2ac 100644 --- a/arch/riscv/include/asm/cpufeature.h +++ b/arch/riscv/include/asm/cpufeature.h @@ -18,4 +18,6 @@ struct riscv_cpuinfo { =20 DECLARE_PER_CPU(struct riscv_cpuinfo, riscv_cpuinfo); =20 +DECLARE_PER_CPU(long, misaligned_access_speed); + #endif diff --git a/arch/riscv/include/asm/hwprobe.h b/arch/riscv/include/asm/hwpr= obe.h index 7e52f1e1fe10..4e45e33015bc 100644 --- a/arch/riscv/include/asm/hwprobe.h +++ b/arch/riscv/include/asm/hwprobe.h @@ -8,6 +8,6 @@ =20 #include =20 -#define RISCV_HWPROBE_MAX_KEY 4 +#define RISCV_HWPROBE_MAX_KEY 5 =20 #endif diff --git a/arch/riscv/include/uapi/asm/hwprobe.h b/arch/riscv/include/uap= i/asm/hwprobe.h index fc5665411782..2968bb0984b5 100644 --- a/arch/riscv/include/uapi/asm/hwprobe.h +++ b/arch/riscv/include/uapi/asm/hwprobe.h @@ -25,6 +25,13 @@ struct riscv_hwprobe { #define RISCV_HWPROBE_KEY_IMA_EXT_0 4 #define RISCV_HWPROBE_IMA_FD (1 << 0) #define RISCV_HWPROBE_IMA_C (1 << 1) +#define RISCV_HWPROBE_KEY_CPUPERF_0 5 +#define RISCV_HWPROBE_MISALIGNED_UNKNOWN (0 << 0) +#define RISCV_HWPROBE_MISALIGNED_EMULATED (1 << 0) +#define RISCV_HWPROBE_MISALIGNED_SLOW (2 << 0) +#define RISCV_HWPROBE_MISALIGNED_FAST (3 << 0) +#define RISCV_HWPROBE_MISALIGNED_UNSUPPORTED (4 << 0) +#define RISCV_HWPROBE_MISALIGNED_MASK (7 << 0) /* Increase RISCV_HWPROBE_MAX_KEY when adding items. */ =20 #endif diff --git a/arch/riscv/kernel/alternative.c b/arch/riscv/kernel/alternativ= e.c index a7d26a00beea..522d3d11e0c3 100644 --- a/arch/riscv/kernel/alternative.c +++ b/arch/riscv/kernel/alternative.c @@ -23,6 +23,8 @@ struct cpu_manufacturer_info_t { void (*patch_func)(struct alt_entry *begin, struct alt_entry *end, unsigned long archid, unsigned long impid, unsigned int stage); + void (*feature_probe_func)(unsigned int cpu, unsigned long archid, + unsigned long impid); }; =20 static void __init_or_module riscv_fill_cpu_mfr_info(struct cpu_manufactur= er_info_t *cpu_mfr_info) @@ -37,6 +39,7 @@ static void __init_or_module riscv_fill_cpu_mfr_info(stru= ct cpu_manufacturer_inf cpu_mfr_info->imp_id =3D sbi_get_mimpid(); #endif =20 + cpu_mfr_info->feature_probe_func =3D NULL; switch (cpu_mfr_info->vendor_id) { #ifdef CONFIG_ERRATA_SIFIVE case SIFIVE_VENDOR_ID: @@ -46,6 +49,7 @@ static void __init_or_module riscv_fill_cpu_mfr_info(stru= ct cpu_manufacturer_inf #ifdef CONFIG_ERRATA_THEAD case THEAD_VENDOR_ID: cpu_mfr_info->patch_func =3D thead_errata_patch_func; + cpu_mfr_info->feature_probe_func =3D thead_feature_probe_func; break; #endif default: @@ -53,6 +57,20 @@ static void __init_or_module riscv_fill_cpu_mfr_info(str= uct cpu_manufacturer_inf } } =20 +/* Called on each CPU as it starts */ +void probe_vendor_features(unsigned int cpu) +{ + struct cpu_manufacturer_info_t cpu_mfr_info; + + riscv_fill_cpu_mfr_info(&cpu_mfr_info); + if (!cpu_mfr_info.feature_probe_func) + return; + + cpu_mfr_info.feature_probe_func(cpu, + cpu_mfr_info.arch_id, + cpu_mfr_info.imp_id); +} + /* * This is called very early in the boot process (directly after we run * a feature detect on the boot CPU). No need to worry about other CPUs @@ -82,6 +100,7 @@ void __init apply_boot_alternatives(void) /* If called on non-boot cpu things could go wrong */ WARN_ON(smp_processor_id() !=3D 0); =20 + probe_vendor_features(0); _apply_alternatives((struct alt_entry *)__alt_start, (struct alt_entry *)__alt_end, RISCV_ALTERNATIVES_BOOT); diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c index 93e45560af30..8ccf260e8b02 100644 --- a/arch/riscv/kernel/cpufeature.c +++ b/arch/riscv/kernel/cpufeature.c @@ -32,6 +32,9 @@ static DECLARE_BITMAP(riscv_isa, RISCV_ISA_EXT_MAX) __rea= d_mostly; DEFINE_STATIC_KEY_ARRAY_FALSE(riscv_isa_ext_keys, RISCV_ISA_EXT_KEY_MAX); EXPORT_SYMBOL(riscv_isa_ext_keys); =20 +/* Performance information */ +DEFINE_PER_CPU(long, misaligned_access_speed); + /** * riscv_isa_extension_base() - Get base extension word * diff --git a/arch/riscv/kernel/smpboot.c b/arch/riscv/kernel/smpboot.c index 3373df413c88..1291ab5ba4c3 100644 --- a/arch/riscv/kernel/smpboot.c +++ b/arch/riscv/kernel/smpboot.c @@ -167,6 +167,7 @@ asmlinkage __visible void smp_callin(void) notify_cpu_starting(curr_cpuid); numa_add_cpu(curr_cpuid); set_cpu_online(curr_cpuid, 1); + probe_vendor_features(curr_cpuid); =20 /* * Remote TLB flushes are ignored while the CPU is offline, so emit diff --git a/arch/riscv/kernel/sys_riscv.c b/arch/riscv/kernel/sys_riscv.c index 1c118438b1b3..76d5b468914c 100644 --- a/arch/riscv/kernel/sys_riscv.c +++ b/arch/riscv/kernel/sys_riscv.c @@ -7,6 +7,7 @@ =20 #include #include +#include #include #include #include @@ -117,6 +118,29 @@ static void hwprobe_arch_id(struct riscv_hwprobe *pair, pair->value =3D id; } =20 +static u64 hwprobe_misaligned(const struct cpumask *cpus) +{ + int cpu; + u64 perf =3D -1ULL; + + for_each_cpu(cpu, cpus) { + int this_perf =3D per_cpu(misaligned_access_speed, cpu); + + if (perf =3D=3D -1ULL) + perf =3D this_perf; + + if (perf !=3D this_perf) { + perf =3D RISCV_HWPROBE_MISALIGNED_UNKNOWN; + break; + } + } + + if (perf =3D=3D -1ULL) + return RISCV_HWPROBE_MISALIGNED_UNKNOWN; + + return perf; +} + static void hwprobe_one_pair(struct riscv_hwprobe *pair, const struct cpumask *cpus) { @@ -146,6 +170,10 @@ static void hwprobe_one_pair(struct riscv_hwprobe *pai= r, =20 break; =20 + case RISCV_HWPROBE_KEY_CPUPERF_0: + pair->value =3D hwprobe_misaligned(cpus); + break; + /* * For forward compatibility, unknown keys don't fail the whole * call, but get their element key set to -1 and value set to 0 --=20 2.25.1 From nobody Sun Feb 8 05:23:07 2026 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 C0129C05027 for ; Tue, 14 Mar 2023 18:35:13 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230472AbjCNSfM (ORCPT ); Tue, 14 Mar 2023 14:35:12 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:50118 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230435AbjCNSel (ORCPT ); Tue, 14 Mar 2023 14:34:41 -0400 Received: from mail-pl1-x629.google.com (mail-pl1-x629.google.com [IPv6:2607:f8b0:4864:20::629]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id EF3865617E for ; Tue, 14 Mar 2023 11:34:09 -0700 (PDT) Received: by mail-pl1-x629.google.com with SMTP id i5so17575139pla.2 for ; Tue, 14 Mar 2023 11:34:09 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rivosinc-com.20210112.gappssmtp.com; s=20210112; t=1678818771; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=x9pHjuYHTV1jwIa+yTaMhteDvvpQj1ERo7dHlvgskt4=; b=KiUg5097LeF/j7i5KvuYGtASQQhq3N8JtMhG38XK8+76BA/FxIZjIKIlXkdOFsuRnj Li1odfe9+ia+MHLeIJVd3J2xwZfzk2KxtMQx3Ul92KkdQmqNxmYS0fozuFA4rU3E+rGT kSg92KYLKSNAJRxXLADE7PtPq2vEMoNukbjJyTPAQaB8VhuLrBLxLiERelw2iBZ4FGbs OVvBngBK8n7BhcLMYHsbOsg8CLfHSdfOEvbM+Cb4KIckp8A4tw1r5doc+yKdSpMEx9vU +CcPivC49xx4wgAQvK8F5xq53vDHkXcVVfAKMXhVKs+SGAj3QP5KBLDAIrVbGJ3Tq/Q7 gbnw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1678818771; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=x9pHjuYHTV1jwIa+yTaMhteDvvpQj1ERo7dHlvgskt4=; b=lRX8DhncoKstCBX5zeoK1FJllekmOkxjkxLgzs4rrL5/50UyO8rYPqUA7DIfFiM/qs 3UFeVfUiTZru1hABBMutn7B3n14zqRTMRlPWpaYg/WxxxC5S4vPHFD2xF6b1En45SIYQ hKT++4tpqgf/fEzqpnjFU+3XXF5H52Ns4cegORgTzJHKo4YKwdMz/QDwEkNjgLNysDEY XS9SaT/OblABx1dSPOcWE8+hmuNBzvjclPjcXzaKz+SvlHW1SDucT4eS9cs+/7IIkDa9 ++7YnJxYTlALnIKp/eC68DWuc6qKFcxfd2HOKGfrk60CugpOnRw2vfz6PckPE2Btj8m1 wwUA== X-Gm-Message-State: AO0yUKWwmZYDd3PpHHhMEH3GeJgR0oVA6JldGC0GOVJy/tMThBGI8du3 kJ7MbZG/WbMpKi0L0rPHNxUseg== X-Google-Smtp-Source: AK7set/wQMmEaZj2KMkqT56JMMbAGHCfckB9OTdTe5KEZjFBskKFirBxhIFoD3oaI8Z10C6al/d4Qw== X-Received: by 2002:a17:903:234a:b0:196:595b:2580 with SMTP id c10-20020a170903234a00b00196595b2580mr141744plh.0.1678818771704; Tue, 14 Mar 2023 11:32:51 -0700 (PDT) Received: from evan.ba.rivosinc.com ([66.220.2.162]) by smtp.gmail.com with ESMTPSA id kn3-20020a170903078300b0019d397b0f18sm2072019plb.214.2023.03.14.11.32.50 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 14 Mar 2023 11:32:51 -0700 (PDT) From: Evan Green To: Palmer Dabbelt Cc: slewis@rivosinc.com, heiko@sntech.de, Conor Dooley , vineetg@rivosinc.com, Evan Green , Albert Ou , Catalin Marinas , Mark Brown , Palmer Dabbelt , Paul Walmsley , Shuah Khan , linux-kernel@vger.kernel.org, linux-kselftest@vger.kernel.org, linux-riscv@lists.infradead.org Subject: [PATCH v4 5/6] selftests: Test the new RISC-V hwprobe interface Date: Tue, 14 Mar 2023 11:32:19 -0700 Message-Id: <20230314183220.513101-6-evan@rivosinc.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20230314183220.513101-1-evan@rivosinc.com> References: <20230314183220.513101-1-evan@rivosinc.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" This adds a test for the recently added RISC-V interface for probing hardware capabilities. It happens to be the first selftest we have for RISC-V, so I've added some infrastructure for those as well. Co-developed-by: Palmer Dabbelt Signed-off-by: Palmer Dabbelt Signed-off-by: Evan Green Tested-by: Heiko Stuebner --- Changes in v4: - Fixed selftests commit description, no more tiny libc (Mark Brown) - Fixed selftest syscall prototype types to match v4. Changes in v2: - Updated the selftests to the new API and added some more. - Fixed indentation, comments in .S, and general checkpatch complaints. --- tools/testing/selftests/Makefile | 1 + tools/testing/selftests/riscv/Makefile | 58 ++++++++++++ .../testing/selftests/riscv/hwprobe/Makefile | 10 +++ .../testing/selftests/riscv/hwprobe/hwprobe.c | 90 +++++++++++++++++++ .../selftests/riscv/hwprobe/sys_hwprobe.S | 12 +++ 5 files changed, 171 insertions(+) create mode 100644 tools/testing/selftests/riscv/Makefile create mode 100644 tools/testing/selftests/riscv/hwprobe/Makefile create mode 100644 tools/testing/selftests/riscv/hwprobe/hwprobe.c create mode 100644 tools/testing/selftests/riscv/hwprobe/sys_hwprobe.S diff --git a/tools/testing/selftests/Makefile b/tools/testing/selftests/Mak= efile index 41b649452560..a599ef726310 100644 --- a/tools/testing/selftests/Makefile +++ b/tools/testing/selftests/Makefile @@ -62,6 +62,7 @@ TARGETS +=3D pstore TARGETS +=3D ptrace TARGETS +=3D openat2 TARGETS +=3D resctrl +TARGETS +=3D riscv TARGETS +=3D rlimits TARGETS +=3D rseq TARGETS +=3D rtc diff --git a/tools/testing/selftests/riscv/Makefile b/tools/testing/selftes= ts/riscv/Makefile new file mode 100644 index 000000000000..32a72902d045 --- /dev/null +++ b/tools/testing/selftests/riscv/Makefile @@ -0,0 +1,58 @@ +# SPDX-License-Identifier: GPL-2.0 +# Originally tools/testing/arm64/Makefile + +# When ARCH not overridden for crosscompiling, lookup machine +ARCH ?=3D $(shell uname -m 2>/dev/null || echo not) + +ifneq (,$(filter $(ARCH),riscv)) +RISCV_SUBTARGETS ?=3D hwprobe +else +RISCV_SUBTARGETS :=3D +endif + +CFLAGS :=3D -Wall -O2 -g + +# A proper top_srcdir is needed by KSFT(lib.mk) +top_srcdir =3D $(realpath ../../../../) + +# Additional include paths needed by kselftest.h and local headers +CFLAGS +=3D -I$(top_srcdir)/tools/testing/selftests/ + +CFLAGS +=3D $(KHDR_INCLUDES) + +export CFLAGS +export top_srcdir + +all: + @for DIR in $(RISCV_SUBTARGETS); do \ + BUILD_TARGET=3D$(OUTPUT)/$$DIR; \ + mkdir -p $$BUILD_TARGET; \ + $(MAKE) OUTPUT=3D$$BUILD_TARGET -C $$DIR $@; \ + done + +install: all + @for DIR in $(RISCV_SUBTARGETS); do \ + BUILD_TARGET=3D$(OUTPUT)/$$DIR; \ + $(MAKE) OUTPUT=3D$$BUILD_TARGET -C $$DIR $@; \ + done + +run_tests: all + @for DIR in $(RISCV_SUBTARGETS); do \ + BUILD_TARGET=3D$(OUTPUT)/$$DIR; \ + $(MAKE) OUTPUT=3D$$BUILD_TARGET -C $$DIR $@; \ + done + +# Avoid any output on non riscv on emit_tests +emit_tests: all + @for DIR in $(RISCV_SUBTARGETS); do \ + BUILD_TARGET=3D$(OUTPUT)/$$DIR; \ + $(MAKE) OUTPUT=3D$$BUILD_TARGET -C $$DIR $@; \ + done + +clean: + @for DIR in $(RISCV_SUBTARGETS); do \ + BUILD_TARGET=3D$(OUTPUT)/$$DIR; \ + $(MAKE) OUTPUT=3D$$BUILD_TARGET -C $$DIR $@; \ + done + +.PHONY: all clean install run_tests emit_tests diff --git a/tools/testing/selftests/riscv/hwprobe/Makefile b/tools/testing= /selftests/riscv/hwprobe/Makefile new file mode 100644 index 000000000000..ebdbb3c22e54 --- /dev/null +++ b/tools/testing/selftests/riscv/hwprobe/Makefile @@ -0,0 +1,10 @@ +# SPDX-License-Identifier: GPL-2.0 +# Copyright (C) 2021 ARM Limited +# Originally tools/testing/arm64/abi/Makefile + +TEST_GEN_PROGS :=3D hwprobe + +include ../../lib.mk + +$(OUTPUT)/hwprobe: hwprobe.c sys_hwprobe.S + $(CC) -o$@ $(CFLAGS) $(LDFLAGS) $^ diff --git a/tools/testing/selftests/riscv/hwprobe/hwprobe.c b/tools/testin= g/selftests/riscv/hwprobe/hwprobe.c new file mode 100644 index 000000000000..09f290a67420 --- /dev/null +++ b/tools/testing/selftests/riscv/hwprobe/hwprobe.c @@ -0,0 +1,90 @@ +// SPDX-License-Identifier: GPL-2.0-only +#include +#include + +/* + * Rather than relying on having a new enough libc to define this, just do= it + * ourselves. This way we don't need to be coupled to a new-enough libc to + * contain the call. + */ +long riscv_hwprobe(struct riscv_hwprobe *pairs, size_t pair_count, + size_t cpu_count, unsigned long *cpus, unsigned int flags); + +int main(int argc, char **argv) +{ + struct riscv_hwprobe pairs[8]; + unsigned long cpus; + long out; + + /* Fake the CPU_SET ops. */ + cpus =3D -1; + + /* + * Just run a basic test: pass enough pairs to get up to the base + * behavior, and then check to make sure it's sane. + */ + for (long i =3D 0; i < 8; i++) + pairs[i].key =3D i; + out =3D riscv_hwprobe(pairs, 8, 1, &cpus, 0); + if (out !=3D 0) + return -1; + for (long i =3D 0; i < 4; ++i) { + /* Fail if the kernel claims not to recognize a base key. */ + if ((i < 4) && (pairs[i].key !=3D i)) + return -2; + + if (pairs[i].key !=3D RISCV_HWPROBE_KEY_BASE_BEHAVIOR) + continue; + + if (pairs[i].value & RISCV_HWPROBE_BASE_BEHAVIOR_IMA) + continue; + + return -3; + } + + /* + * This should also work with a NULL CPU set, but should not work + * with an improperly supplied CPU set. + */ + out =3D riscv_hwprobe(pairs, 8, 0, 0, 0); + if (out !=3D 0) + return -4; + + out =3D riscv_hwprobe(pairs, 8, 0, &cpus, 0); + if (out =3D=3D 0) + return -5; + + out =3D riscv_hwprobe(pairs, 8, 1, 0, 0); + if (out =3D=3D 0) + return -6; + + /* + * Check that keys work by providing one that we know exists, and + * checking to make sure the resultig pair is what we asked for. + */ + pairs[0].key =3D RISCV_HWPROBE_KEY_BASE_BEHAVIOR; + out =3D riscv_hwprobe(pairs, 1, 1, &cpus, 0); + if (out !=3D 0) + return -7; + if (pairs[0].key !=3D RISCV_HWPROBE_KEY_BASE_BEHAVIOR) + return -8; + + /* + * Check that an unknown key gets overwritten with -1, + * but doesn't block elements after it. + */ + pairs[0].key =3D 0x5555; + pairs[1].key =3D 1; + pairs[1].value =3D 0xAAAA; + out =3D riscv_hwprobe(pairs, 2, 0, 0, 0); + if (out !=3D 0) + return -9; + + if (pairs[0].key !=3D -1) + return -10; + + if ((pairs[1].key !=3D 1) || (pairs[1].value =3D=3D 0xAAAA)) + return -11; + + return 0; +} diff --git a/tools/testing/selftests/riscv/hwprobe/sys_hwprobe.S b/tools/te= sting/selftests/riscv/hwprobe/sys_hwprobe.S new file mode 100644 index 000000000000..ed8d28863b27 --- /dev/null +++ b/tools/testing/selftests/riscv/hwprobe/sys_hwprobe.S @@ -0,0 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (C) 2022 Rivos, Inc */ + +.text +.global riscv_hwprobe +riscv_hwprobe: + # Put __NR_riscv_hwprobe in the syscall number register, then just shim + # back the kernel's return. This doesn't do any sort of errno + # handling, the caller can deal with it. + li a7, 258 + ecall + ret --=20 2.25.1 From nobody Sun Feb 8 05:23:07 2026 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 9D774C6FD1F for ; Tue, 14 Mar 2023 18:35:16 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231161AbjCNSfP (ORCPT ); Tue, 14 Mar 2023 14:35:15 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:50296 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231173AbjCNSeq (ORCPT ); Tue, 14 Mar 2023 14:34:46 -0400 Received: from mail-pl1-x62d.google.com (mail-pl1-x62d.google.com [IPv6:2607:f8b0:4864:20::62d]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 5579BA8816 for ; Tue, 14 Mar 2023 11:34:19 -0700 (PDT) Received: by mail-pl1-x62d.google.com with SMTP id i5so17575293pla.2 for ; Tue, 14 Mar 2023 11:34:19 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rivosinc-com.20210112.gappssmtp.com; s=20210112; t=1678818776; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=L2I6pzb9mfcEjWOn9wUkhHEqFGs76E697ygZBc6Zpec=; b=SJ9H+8cq/o5ePemMP/iGfo9qWAMMQSokbVTQyJh1hPK9R7xoxRxMQQSac1ZYoj07/8 pvDG8QHI83PX5z7VVQ1YwGApZcYNJqC8X6ESSRAJvN1fBkQUaFXbHvPn6RQMJ/K2AKdc aP+b5nA5kS8n+CfDsdhKgILlZ0kRDkOArahw1NOe08u+PJ5ne7RG6MthaMWDm/cSJkL0 6FgrsW5i1yBDZJ0A19M+ouwzIzVFcwU17NYY9MVNWUmBSHO5Gajt6TCqDvcKeU6PG9Vo DS3DXTgiXRiPvMu//X4+L/eL1YXX8s0GeOMQwpegAA4nfh+vw5t9vo3kAnNf3UfEPWSm vXoA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1678818776; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=L2I6pzb9mfcEjWOn9wUkhHEqFGs76E697ygZBc6Zpec=; b=e8R0+sq2y+VQ9OGgPsJJzZEtXhNtpXm8IUtIyMBafLU4MorPWlVTn9jmzBj9vyyMCD S14OpZnvjcIq3QOjpgxuK7uLzJDkl4JCbitsfoGsDpjUrt6MmyAYtp1ORoP8LsYuTFWo WMjEt5YhDF8h2zkFWGzqRGobjDddC3d+sVpux2SYUKGykfFy+olRLDUHFDzJwRwKyNRX XruZcq6Z+EQHAtvF3UDgA7aomyDKG0qFN01vIjer03YxKw+7KFUttvsIFGzWJcARmXy+ imsuEsJ5TMY+IVTuQHY+jSwqDs3DfVZ/g/VH7ZmR0vgDX0JwQE5qxfenH70f5D2AQYC2 XEXA== X-Gm-Message-State: AO0yUKWavYMPK5lYKXqA4zksHsWgA9GzLiJIGLA6xPXjz448CzL7r8vy SSHqa7l7QubeoUCCoqmfM+wziA== X-Google-Smtp-Source: AK7set8Fsg5s7kt3oEpyj8SZxQYfKabHPUu4a6pAtMNhvgpOMLq8DQjgipHpvPFZ3gjwql6/4NPk5g== X-Received: by 2002:a17:902:e847:b0:19a:81c1:e743 with SMTP id t7-20020a170902e84700b0019a81c1e743mr92302plg.2.1678818775814; Tue, 14 Mar 2023 11:32:55 -0700 (PDT) Received: from evan.ba.rivosinc.com ([66.220.2.162]) by smtp.gmail.com with ESMTPSA id kn3-20020a170903078300b0019d397b0f18sm2072019plb.214.2023.03.14.11.32.54 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 14 Mar 2023 11:32:55 -0700 (PDT) From: Evan Green To: Palmer Dabbelt Cc: slewis@rivosinc.com, heiko@sntech.de, Conor Dooley , vineetg@rivosinc.com, Evan Green , Albert Ou , Andrew Bresticker , Atish Patra , Celeste Liu , Conor Dooley , Davidlohr Bueso , Guo Ren , Jann Horn , Jisheng Zhang , "Liam R. Howlett" , Nathan Chancellor , Palmer Dabbelt , Paul Walmsley , linux-kernel@vger.kernel.org, linux-riscv@lists.infradead.org Subject: [PATCH v4 6/6] RISC-V: Add hwprobe vDSO function and data Date: Tue, 14 Mar 2023 11:32:20 -0700 Message-Id: <20230314183220.513101-7-evan@rivosinc.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20230314183220.513101-1-evan@rivosinc.com> References: <20230314183220.513101-1-evan@rivosinc.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Add a vDSO function __vdso_riscv_hwprobe, which can sit in front of the riscv_hwprobe syscall and answer common queries. We stash a copy of static answers for the "all CPUs" case in the vDSO data page. This data is private to the vDSO, so we can decide later to change what's stored there or under what conditions we defer to the syscall. Currently all data can be discovered at boot, so the vDSO function answers all queries when the cpumask is set to the "all CPUs" hint. There's also a boolean in the data that lets the vDSO function know that all CPUs are the same. In that case, the vDSO will also answer queries for arbitrary CPU masks in addition to the "all CPUs" hint. Signed-off-by: Evan Green Tested-by: Heiko Stuebner --- Changes in v4: - Added a prototype to fix -Wmissing-prototype warning (lkp@intel.com) - Fixed rv32 build failure (lkp@intel.com) - Make vdso prototype match syscall types update Changes in v3: - Introduced vDSO function One aspect of this that's less than perfect is that there are two copies of the arch_vdso_data in the data page. This stems from the fact that vdso_data is arrayed by CS_BASES. Since we're very far away from filling up the page, and this is not ABI, I left it for now. If things get full we can move this data out to its own page, or refactor vdso_data tree-wide to allow for non-arrayed data. --- arch/riscv/Kconfig | 1 + arch/riscv/include/asm/vdso/data.h | 17 +++++++ arch/riscv/include/asm/vdso/gettimeofday.h | 8 ++++ arch/riscv/kernel/sys_riscv.c | 41 +++++++++++++++++ arch/riscv/kernel/vdso.c | 6 --- arch/riscv/kernel/vdso/Makefile | 2 + arch/riscv/kernel/vdso/hwprobe.c | 52 ++++++++++++++++++++++ arch/riscv/kernel/vdso/sys_hwprobe.S | 15 +++++++ arch/riscv/kernel/vdso/vdso.lds.S | 1 + 9 files changed, 137 insertions(+), 6 deletions(-) create mode 100644 arch/riscv/include/asm/vdso/data.h create mode 100644 arch/riscv/kernel/vdso/hwprobe.c create mode 100644 arch/riscv/kernel/vdso/sys_hwprobe.S diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig index e2b656043abf..218e75975a15 100644 --- a/arch/riscv/Kconfig +++ b/arch/riscv/Kconfig @@ -33,6 +33,7 @@ config RISCV select ARCH_HAS_STRICT_MODULE_RWX if MMU && !XIP_KERNEL select ARCH_HAS_TICK_BROADCAST if GENERIC_CLOCKEVENTS_BROADCAST select ARCH_HAS_UBSAN_SANITIZE_ALL + select ARCH_HAS_VDSO_DATA select ARCH_OPTIONAL_KERNEL_RWX if ARCH_HAS_STRICT_KERNEL_RWX select ARCH_OPTIONAL_KERNEL_RWX_DEFAULT select ARCH_STACKWALK diff --git a/arch/riscv/include/asm/vdso/data.h b/arch/riscv/include/asm/vd= so/data.h new file mode 100644 index 000000000000..dc2f76f58b76 --- /dev/null +++ b/arch/riscv/include/asm/vdso/data.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __RISCV_ASM_VDSO_DATA_H +#define __RISCV_ASM_VDSO_DATA_H + +#include +#include +#include + +struct arch_vdso_data { + /* Stash static answers to the hwprobe queries when all CPUs are selected= . */ + __u64 all_cpu_hwprobe_values[RISCV_HWPROBE_MAX_KEY + 1]; + + /* Boolean indicating all CPUs have the same static hwprobe values. */ + __u8 homogeneous_cpus; +}; + +#endif /* __RISCV_ASM_VDSO_DATA_H */ diff --git a/arch/riscv/include/asm/vdso/gettimeofday.h b/arch/riscv/includ= e/asm/vdso/gettimeofday.h index 77d9c2f721c4..ba3283cf7acc 100644 --- a/arch/riscv/include/asm/vdso/gettimeofday.h +++ b/arch/riscv/include/asm/vdso/gettimeofday.h @@ -9,6 +9,12 @@ #include #include =20 +/* + * 32-bit land is lacking generic time vsyscalls as well as the legacy 32-= bit + * time syscalls like gettimeofday. Skip these definitions since on 32-bit. + */ +#ifdef CONFIG_GENERIC_TIME_VSYSCALL + #define VDSO_HAS_CLOCK_GETRES 1 =20 static __always_inline @@ -60,6 +66,8 @@ int clock_getres_fallback(clockid_t _clkid, struct __kern= el_timespec *_ts) return ret; } =20 +#endif /* CONFIG_GENERIC_TIME_VSYSCALL */ + static __always_inline u64 __arch_get_hw_counter(s32 clock_mode, const struct vdso_data *vd) { diff --git a/arch/riscv/kernel/sys_riscv.c b/arch/riscv/kernel/sys_riscv.c index 76d5b468914c..0711dc4afe73 100644 --- a/arch/riscv/kernel/sys_riscv.c +++ b/arch/riscv/kernel/sys_riscv.c @@ -14,6 +14,7 @@ #include #include #include +#include =20 static long riscv_sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, @@ -244,6 +245,46 @@ static int do_riscv_hwprobe(struct riscv_hwprobe __use= r *pairs, =20 } =20 +static int __init init_hwprobe_vdso_data(void) +{ + struct vdso_data *vd =3D __arch_get_k_vdso_data(); + struct arch_vdso_data *avd =3D &vd->arch_data; + u64 id_bitsmash =3D 0; + struct riscv_hwprobe pair; + int key; + + /* + * Initialize vDSO data with the answers for the "all CPUs" case, to + * save a syscall in the common case. + */ + for (key =3D 0; key <=3D RISCV_HWPROBE_MAX_KEY; key++) { + pair.key =3D key; + hwprobe_one_pair(&pair, cpu_online_mask); + + WARN_ON_ONCE(pair.key < 0); + + avd->all_cpu_hwprobe_values[key] =3D pair.value; + /* + * Smash together the vendor, arch, and impl IDs to see if + * they're all 0 or any negative. + */ + if (key <=3D RISCV_HWPROBE_KEY_MIMPID) + id_bitsmash |=3D pair.value; + } + + /* + * If the arch, vendor, and implementation ID are all the same across + * all harts, then assume all CPUs are the same, and allow the vDSO to + * answer queries for arbitrary masks. However if all values are 0 (not + * populated) or any value returns -1 (varies across CPUs), then the + * vDSO should defer to the kernel for exotic cpu masks. + */ + avd->homogeneous_cpus =3D (id_bitsmash > 0); + return 0; +} + +arch_initcall_sync(init_hwprobe_vdso_data); + SYSCALL_DEFINE5(riscv_hwprobe, struct riscv_hwprobe __user *, pairs, size_t, pair_count, size_t, cpu_count, unsigned long __user *, cpus, unsigned int, flags) diff --git a/arch/riscv/kernel/vdso.c b/arch/riscv/kernel/vdso.c index e410275918ac..52a847eacd79 100644 --- a/arch/riscv/kernel/vdso.c +++ b/arch/riscv/kernel/vdso.c @@ -14,13 +14,7 @@ #include #include #include - -#ifdef CONFIG_GENERIC_TIME_VSYSCALL #include -#else -struct vdso_data { -}; -#endif =20 extern char vdso_start[], vdso_end[]; #ifdef CONFIG_COMPAT diff --git a/arch/riscv/kernel/vdso/Makefile b/arch/riscv/kernel/vdso/Makef= ile index 06e6b27f3bcc..107fd2162cdf 100644 --- a/arch/riscv/kernel/vdso/Makefile +++ b/arch/riscv/kernel/vdso/Makefile @@ -12,6 +12,8 @@ vdso-syms +=3D vgettimeofday endif vdso-syms +=3D getcpu vdso-syms +=3D flush_icache +vdso-syms +=3D hwprobe +vdso-syms +=3D sys_hwprobe =20 # Files to link into the vdso obj-vdso =3D $(patsubst %, %.o, $(vdso-syms)) note.o diff --git a/arch/riscv/kernel/vdso/hwprobe.c b/arch/riscv/kernel/vdso/hwpr= obe.c new file mode 100644 index 000000000000..d40bec6ac078 --- /dev/null +++ b/arch/riscv/kernel/vdso/hwprobe.c @@ -0,0 +1,52 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright 2023 Rivos, Inc + */ + +#include +#include +#include + +extern int riscv_hwprobe(struct riscv_hwprobe *pairs, size_t pair_count, + size_t cpu_count, unsigned long *cpus, + unsigned int flags); + +/* Add a prototype to avoid -Wmissing-prototypes warning. */ +int __vdso_riscv_hwprobe(struct riscv_hwprobe *pairs, size_t pair_count, + size_t cpu_count, unsigned long *cpus, + unsigned int flags); + +int __vdso_riscv_hwprobe(struct riscv_hwprobe *pairs, size_t pair_count, + size_t cpu_count, unsigned long *cpus, + unsigned int flags) +{ + const struct vdso_data *vd =3D __arch_get_vdso_data(); + const struct arch_vdso_data *avd =3D &vd->arch_data; + bool all_cpus =3D !cpu_count && !cpus; + struct riscv_hwprobe *p =3D pairs; + struct riscv_hwprobe *end =3D pairs + pair_count; + + /* + * Defer to the syscall for exotic requests. The vdso has answers + * stashed away only for the "all cpus" case. If all CPUs are + * homogeneous, then this function can handle requests for arbitrary + * masks. + */ + if ((flags !=3D 0) || (!all_cpus && !avd->homogeneous_cpus)) + return riscv_hwprobe(pairs, pair_count, cpu_count, cpus, flags); + + /* This is something we can handle, fill out the pairs. */ + while (p < end) { + if (p->key <=3D RISCV_HWPROBE_MAX_KEY) { + p->value =3D avd->all_cpu_hwprobe_values[p->key]; + + } else { + p->key =3D -1; + p->value =3D 0; + } + + p++; + } + + return 0; +} diff --git a/arch/riscv/kernel/vdso/sys_hwprobe.S b/arch/riscv/kernel/vdso/= sys_hwprobe.S new file mode 100644 index 000000000000..2511c8ba8f86 --- /dev/null +++ b/arch/riscv/kernel/vdso/sys_hwprobe.S @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (C) 2022 Rivos, Inc */ + +#include +#include + +.text +ENTRY(riscv_hwprobe) + .cfi_startproc + li a7, __NR_riscv_hwprobe + ecall + ret + + .cfi_endproc +ENDPROC(riscv_hwprobe) diff --git a/arch/riscv/kernel/vdso/vdso.lds.S b/arch/riscv/kernel/vdso/vds= o.lds.S index 150b1a572e61..dfe63fb71041 100644 --- a/arch/riscv/kernel/vdso/vdso.lds.S +++ b/arch/riscv/kernel/vdso/vdso.lds.S @@ -75,6 +75,7 @@ VERSION #endif __vdso_getcpu; __vdso_flush_icache; + __vdso_riscv_hwprobe; local: *; }; } --=20 2.25.1