From patchwork Tue Aug 20 15:24:23 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jesse Taube X-Patchwork-Id: 13770422 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 bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 5A6EBC531DC for ; Tue, 20 Aug 2024 15:27:47 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender: Content-Transfer-Encoding:Content-Type:List-Subscribe:List-Help:List-Post: List-Archive:List-Unsubscribe:List-Id:MIME-Version:References:In-Reply-To: Message-ID:Date:Subject:Cc:To:From:Reply-To:Content-ID:Content-Description: Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID: List-Owner; bh=JbU82MKsBXkhLqnROapLkTz6ZclfjQZPnhyInF/ZEak=; b=kGJvYgvuG1ofdz mksmcdsLz3Nzv3QEsSGz3tvlYLTNDn7ekLaTK5iE/0501wvkq+aHNjFS5Mni9BHY6AEQpxjeWsyCY L7YfRJIu89cdiNyFI95k+2FXwvTPNkk7BcDw72lzwxcAyC+4lbPzMYf7O101NshEg9QtzFXFFkyYG 0R9t1zFFsxGWlZ/0apPcuuWAU0b0ygccCNV0TZkrE4jDKAuKC0aRJbGQpReUTM68LeF9j0SHOkVLk z7xhXbMRircFs/EP21XDkdpTDWGWff0SOBIOk7RTPDEpqBlwghZHuSAwmtWYiJXaeg35lcwOCOhAS BtpGMCNPUEF2hsnKekhA==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.97.1 #2 (Red Hat Linux)) id 1sgQlr-00000005orJ-3N1T; Tue, 20 Aug 2024 15:27:43 +0000 Received: from mail-pj1-x102c.google.com ([2607:f8b0:4864:20::102c]) by bombadil.infradead.org with esmtps (Exim 4.97.1 #2 (Red Hat Linux)) id 1sgQj6-00000005nMk-1sDd for linux-riscv@lists.infradead.org; Tue, 20 Aug 2024 15:24:53 +0000 Received: by mail-pj1-x102c.google.com with SMTP id 98e67ed59e1d1-2d3b36f5366so3950362a91.0 for ; Tue, 20 Aug 2024 08:24:52 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rivosinc-com.20230601.gappssmtp.com; s=20230601; t=1724167492; x=1724772292; darn=lists.infradead.org; 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=rJi2ebhbP9zlbLGPWODvcw9blh6x5jyJs+zGiuzhdIE=; b=1CFDcf/uEcP/sfc6sc2WhmApCOBlJZWPpn8TCudbUB+klJpzL7o88rNiE+Z9mOIESi 1MxEwDjmJ+BnfHkA0E9s2tmkurA8nMhrKBEFczm2jdozBgofgJWuCiw9znEwlL3CtQSz qbqGG0zmvbqDdhbahNMaicpy4CoBdLz818/QXWrFi/0sPhxGBRWpv0T9enNyRsMu1lb4 86Y5HCAbCrWP93BcxhC7QWmOYJMilbm3yraZ8T/VxAvtAirnQ8Sl/p4HdcXwqO57NuIi Am6zb8O0aFJ2emWZAIqyXoIWNVlxdET7jZnS0jG0d28Oz/VYs2J9aJriTiUDbBby3rfE 2WdA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1724167492; x=1724772292; 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=rJi2ebhbP9zlbLGPWODvcw9blh6x5jyJs+zGiuzhdIE=; b=RRnw3ME/DMcK+qo3DOmmxGe749FuoqVxzKb6h+ZJ0w4ynUVeMrNGYHQjx0idjzWtnP DUX8cGwTUMfphOMCHKUanOx6GIiDbOWSo6EiLwYSkcz4bL4LZLG/l0e/HH0TJHMcZjKu KI+NBChwHQ1OP6O27lyoG+Y12JEQ2lOkyg9rs7hyasWh2hoLMwl7hPvdc25zFa9/pRFR mOH+IFwJ7vgDJVbjxVb8Lw81CzFDyDtLl3lXHEm/npAhCe/V2LBOl30Dvhn8ceSEw+Qt Z1lwJZEzaEKtjjgLyW2dnmECERwMcjD4eu/faruT8YYa6iCwic/8BONfBv/DysQrVwXH dmcg== X-Gm-Message-State: AOJu0YwEXi692gAfFRc7h5h2R4MLH7qbz0/sDxsgSvNoeZ+y4UeNxmxj Bd47l9sGXCj+gmFI0yA3SqjzD2/Ndz+MZ/CCtGiZkBqtvcCNfhdC6D/0oARymK9rdQi1xJ8kB/X 3 X-Google-Smtp-Source: AGHT+IEclpseZglRZhbnRuBkQilDDbPp9TBiZLIEf2DItUIwluK/RElu63vQyqZY6QZ4MDhVhG0uhw== X-Received: by 2002:a17:90a:8c94:b0:2d1:ca16:5555 with SMTP id 98e67ed59e1d1-2d3e03e83e5mr13286356a91.37.1724167491452; Tue, 20 Aug 2024 08:24:51 -0700 (PDT) Received: from jesse-desktop.ba.rivosinc.com (pool-108-26-179-17.bstnma.fios.verizon.net. [108.26.179.17]) by smtp.gmail.com with ESMTPSA id 98e67ed59e1d1-2d40bea7cb3sm7258157a91.25.2024.08.20.08.24.47 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 20 Aug 2024 08:24:51 -0700 (PDT) From: Jesse Taube To: linux-riscv@lists.infradead.org Cc: Jonathan Corbet , Paul Walmsley , Palmer Dabbelt , Albert Ou , Conor Dooley , Rob Herring , Krzysztof Kozlowski , =?utf-8?b?Q2zDqW1lbnQgTMOpZ2Vy?= , Evan Green , Andrew Jones , Jesse Taube , Charlie Jenkins , Xiao Wang , Andy Chiu , Eric Biggers , Greentime Hu , =?utf-8?b?QmrDtnJuIFTDtnBlbA==?= , Heiko Stuebner , Costa Shulyupin , Andrew Morton , Baoquan He , Anup Patel , Zong Li , Sami Tolvanen , Ben Dooks , Alexandre Ghiti , "Gustavo A. R. Silva" , Erick Archer , Joel Granados , linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, devicetree@vger.kernel.org Subject: [PATCH v9 5/6] RISC-V: Report vector unaligned access speed hwprobe Date: Tue, 20 Aug 2024 11:24:23 -0400 Message-ID: <20240820152424.1973078-6-jesse@rivosinc.com> X-Mailer: git-send-email 2.45.2 In-Reply-To: <20240820152424.1973078-1-jesse@rivosinc.com> References: <20240820152424.1973078-1-jesse@rivosinc.com> MIME-Version: 1.0 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20240820_082452_517422_D718B77A X-CRM114-Status: GOOD ( 30.07 ) X-BeenThere: linux-riscv@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-riscv" Errors-To: linux-riscv-bounces+linux-riscv=archiver.kernel.org@lists.infradead.org Detect if vector misaligned accesses are faster or slower than equivalent vector byte accesses. This is useful for usermode to know whether vector byte accesses or vector misaligned accesses have a better bandwidth for operations like memcpy. Signed-off-by: Jesse Taube Reviewed-by: Charlie Jenkins --- V1 -> V2: - Add Kconfig options - Add WORD_EEW to vec-copy-unaligned.S V2 -> V3: - Remove unnecessary comment - Remove local_irq_enable V3 -> V4: - Add preempt_disable/enable - Alphabetize includes in vec-copy-unaligned.S and unaligned_access_speed.c - Add duplicate comments above mb() to please checkpatch - change all_cpus_vec_supported to all_cpus_vec_unsupported so speed is tested if any cpus support unaligned vector accesses - Spell out _VECTOR_ in macros V4 -> V5: - Change void *unused to void *unused __always_unused V5 -> V6: - Check for vector misaligned access support in hotplug V6 -> V7: - Change SLOW to UNKNOWN when used as a placeholder V7 -> V8: - Rebase onto fixes - s/RISCV_HWPROBE_VECTOR_MISALIGNED/RISCV_HWPROBE_MISALIGNED_VECTOR/g V8 -> V9: - No changes --- arch/riscv/Kconfig | 18 +++ arch/riscv/kernel/Makefile | 3 +- arch/riscv/kernel/copy-unaligned.h | 5 + arch/riscv/kernel/sys_hwprobe.c | 6 + arch/riscv/kernel/unaligned_access_speed.c | 141 ++++++++++++++++++++- arch/riscv/kernel/vec-copy-unaligned.S | 58 +++++++++ 6 files changed, 228 insertions(+), 3 deletions(-) create mode 100644 arch/riscv/kernel/vec-copy-unaligned.S diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig index 3bb7bf0e9ddc..db1393ba5258 100644 --- a/arch/riscv/Kconfig +++ b/arch/riscv/Kconfig @@ -855,6 +855,24 @@ config RISCV_PROBE_VECTOR_UNALIGNED_ACCESS will dynamically determine the speed of vector unaligned accesses on the underlying system if they are supported. +config RISCV_SLOW_VECTOR_UNALIGNED_ACCESS + bool "Assume the system supports slow vector unaligned memory accesses" + depends on NONPORTABLE + help + Assume that the system supports slow vector unaligned memory accesses. The + kernel and userspace programs may not be able to run at all on systems + that do not support unaligned memory accesses. + +config RISCV_EFFICIENT_VECTOR_UNALIGNED_ACCESS + bool "Assume the system supports fast vector unaligned memory accesses" + depends on NONPORTABLE + help + Assume that the system supports fast vector unaligned memory accesses. When + enabled, this option improves the performance of the kernel on such + systems. However, the kernel and userspace programs will run much more + slowly, or will not be able to run at all, on systems that do not + support efficient unaligned memory accesses. + endchoice source "arch/riscv/Kconfig.vendor" diff --git a/arch/riscv/kernel/Makefile b/arch/riscv/kernel/Makefile index 06d407f1b30b..c72ca98a1a1c 100644 --- a/arch/riscv/kernel/Makefile +++ b/arch/riscv/kernel/Makefile @@ -66,7 +66,8 @@ obj-$(CONFIG_MMU) += vdso.o vdso/ obj-$(CONFIG_RISCV_MISALIGNED) += traps_misaligned.o obj-$(CONFIG_RISCV_MISALIGNED) += unaligned_access_speed.o -obj-$(CONFIG_RISCV_PROBE_UNALIGNED_ACCESS) += copy-unaligned.o +obj-$(CONFIG_RISCV_PROBE_UNALIGNED_ACCESS) += copy-unaligned.o +obj-$(CONFIG_RISCV_PROBE_VECTOR_UNALIGNED_ACCESS) += vec-copy-unaligned.o obj-$(CONFIG_FPU) += fpu.o obj-$(CONFIG_FPU) += kernel_mode_fpu.o diff --git a/arch/riscv/kernel/copy-unaligned.h b/arch/riscv/kernel/copy-unaligned.h index e3d70d35b708..85d4d11450cb 100644 --- a/arch/riscv/kernel/copy-unaligned.h +++ b/arch/riscv/kernel/copy-unaligned.h @@ -10,4 +10,9 @@ void __riscv_copy_words_unaligned(void *dst, const void *src, size_t size); void __riscv_copy_bytes_unaligned(void *dst, const void *src, size_t size); +#ifdef CONFIG_RISCV_PROBE_VECTOR_UNALIGNED_ACCESS +void __riscv_copy_vec_words_unaligned(void *dst, const void *src, size_t size); +void __riscv_copy_vec_bytes_unaligned(void *dst, const void *src, size_t size); +#endif + #endif /* __RISCV_KERNEL_COPY_UNALIGNED_H */ diff --git a/arch/riscv/kernel/sys_hwprobe.c b/arch/riscv/kernel/sys_hwprobe.c index 6441baada36b..6673278e84d5 100644 --- a/arch/riscv/kernel/sys_hwprobe.c +++ b/arch/riscv/kernel/sys_hwprobe.c @@ -228,6 +228,12 @@ static u64 hwprobe_vec_misaligned(const struct cpumask *cpus) #else static u64 hwprobe_vec_misaligned(const struct cpumask *cpus) { + if (IS_ENABLED(CONFIG_RISCV_EFFICIENT_VECTOR_UNALIGNED_ACCESS)) + return RISCV_HWPROBE_MISALIGNED_VECTOR_FAST; + + if (IS_ENABLED(CONFIG_RISCV_SLOW_VECTOR_UNALIGNED_ACCESS)) + return RISCV_HWPROBE_MISALIGNED_VECTOR_SLOW; + return RISCV_HWPROBE_MISALIGNED_VECTOR_UNKNOWN; } #endif diff --git a/arch/riscv/kernel/unaligned_access_speed.c b/arch/riscv/kernel/unaligned_access_speed.c index 0b8b5e17453a..91f189cf1611 100644 --- a/arch/riscv/kernel/unaligned_access_speed.c +++ b/arch/riscv/kernel/unaligned_access_speed.c @@ -6,11 +6,13 @@ #include #include #include +#include #include #include #include #include #include +#include #include "copy-unaligned.h" @@ -268,12 +270,147 @@ static int check_unaligned_access_speed_all_cpus(void) } #endif +#ifdef CONFIG_RISCV_PROBE_VECTOR_UNALIGNED_ACCESS +static void check_vector_unaligned_access(struct work_struct *work __always_unused) +{ + int cpu = smp_processor_id(); + u64 start_cycles, end_cycles; + u64 word_cycles; + u64 byte_cycles; + int ratio; + unsigned long start_jiffies, now; + struct page *page; + void *dst; + void *src; + long speed = RISCV_HWPROBE_MISALIGNED_VECTOR_SLOW; + + if (per_cpu(vector_misaligned_access, cpu) != RISCV_HWPROBE_MISALIGNED_VECTOR_UNKNOWN) + return; + + page = alloc_pages(GFP_KERNEL, MISALIGNED_BUFFER_ORDER); + if (!page) { + pr_warn("Allocation failure, not measuring vector misaligned performance\n"); + return; + } + + /* Make an unaligned destination buffer. */ + dst = (void *)((unsigned long)page_address(page) | 0x1); + /* Unalign src as well, but differently (off by 1 + 2 = 3). */ + src = dst + (MISALIGNED_BUFFER_SIZE / 2); + src += 2; + word_cycles = -1ULL; + + /* Do a warmup. */ + kernel_vector_begin(); + __riscv_copy_vec_words_unaligned(dst, src, MISALIGNED_COPY_SIZE); + + start_jiffies = jiffies; + while ((now = jiffies) == start_jiffies) + cpu_relax(); + + /* + * For a fixed amount of time, repeatedly try the function, and take + * the best time in cycles as the measurement. + */ + while (time_before(jiffies, now + (1 << MISALIGNED_ACCESS_JIFFIES_LG2))) { + start_cycles = get_cycles64(); + /* Ensure the CSR read can't reorder WRT to the copy. */ + mb(); + __riscv_copy_vec_words_unaligned(dst, src, MISALIGNED_COPY_SIZE); + /* Ensure the copy ends before the end time is snapped. */ + mb(); + end_cycles = get_cycles64(); + if ((end_cycles - start_cycles) < word_cycles) + word_cycles = end_cycles - start_cycles; + } + + byte_cycles = -1ULL; + __riscv_copy_vec_bytes_unaligned(dst, src, MISALIGNED_COPY_SIZE); + start_jiffies = jiffies; + while ((now = jiffies) == start_jiffies) + cpu_relax(); + + while (time_before(jiffies, now + (1 << MISALIGNED_ACCESS_JIFFIES_LG2))) { + start_cycles = get_cycles64(); + /* Ensure the CSR read can't reorder WRT to the copy. */ + mb(); + __riscv_copy_vec_bytes_unaligned(dst, src, MISALIGNED_COPY_SIZE); + /* Ensure the copy ends before the end time is snapped. */ + mb(); + end_cycles = get_cycles64(); + if ((end_cycles - start_cycles) < byte_cycles) + byte_cycles = end_cycles - start_cycles; + } + + kernel_vector_end(); + + /* Don't divide by zero. */ + if (!word_cycles || !byte_cycles) { + pr_warn("cpu%d: rdtime lacks granularity needed to measure unaligned vector access speed\n", + cpu); + + return; + } + + if (word_cycles < byte_cycles) + speed = RISCV_HWPROBE_MISALIGNED_VECTOR_FAST; + + ratio = div_u64((byte_cycles * 100), word_cycles); + pr_info("cpu%d: Ratio of vector byte access time to vector unaligned word access is %d.%02d, unaligned accesses are %s\n", + cpu, + ratio / 100, + ratio % 100, + (speed == RISCV_HWPROBE_MISALIGNED_VECTOR_FAST) ? "fast" : "slow"); + + per_cpu(vector_misaligned_access, cpu) = speed; +} + +static int riscv_online_cpu_vec(unsigned int cpu) +{ + if (!has_vector()) + return 0; + + if (per_cpu(vector_misaligned_access, cpu) != RISCV_HWPROBE_MISALIGNED_VECTOR_UNSUPPORTED) + return 0; + + check_vector_unaligned_access_emulated(NULL); + check_vector_unaligned_access(NULL); + return 0; +} + +/* Measure unaligned access speed on all CPUs present at boot in parallel. */ +static int vec_check_unaligned_access_speed_all_cpus(void *unused __always_unused) +{ + schedule_on_each_cpu(check_vector_unaligned_access); + + /* + * Setup hotplug callbacks for any new CPUs that come online or go + * offline. + */ + cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN, "riscv:online", + riscv_online_cpu_vec, NULL); + + return 0; +} +#else /* CONFIG_RISCV_PROBE_VECTOR_UNALIGNED_ACCESS */ +static int vec_check_unaligned_access_speed_all_cpus(void *unused __always_unused) +{ + return 0; +} +#endif + static int check_unaligned_access_all_cpus(void) { - bool all_cpus_emulated; + bool all_cpus_emulated, all_cpus_vec_unsupported; all_cpus_emulated = check_unaligned_access_emulated_all_cpus(); - check_vector_unaligned_access_emulated_all_cpus(); + all_cpus_vec_unsupported = check_vector_unaligned_access_emulated_all_cpus(); + + if (!all_cpus_vec_unsupported && + IS_ENABLED(CONFIG_RISCV_PROBE_VECTOR_UNALIGNED_ACCESS)) { + kthread_run(vec_check_unaligned_access_speed_all_cpus, + NULL, "vec_check_unaligned_access_speed_all_cpus"); + } if (!all_cpus_emulated) return check_unaligned_access_speed_all_cpus(); diff --git a/arch/riscv/kernel/vec-copy-unaligned.S b/arch/riscv/kernel/vec-copy-unaligned.S new file mode 100644 index 000000000000..d16f19f1b3b6 --- /dev/null +++ b/arch/riscv/kernel/vec-copy-unaligned.S @@ -0,0 +1,58 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (C) 2024 Rivos Inc. */ + +#include +#include +#include + + .text + +#define WORD_EEW 32 + +#define WORD_SEW CONCATENATE(e, WORD_EEW) +#define VEC_L CONCATENATE(vle, WORD_EEW).v +#define VEC_S CONCATENATE(vle, WORD_EEW).v + +/* void __riscv_copy_vec_words_unaligned(void *, const void *, size_t) */ +/* Performs a memcpy without aligning buffers, using word loads and stores. */ +/* Note: The size is truncated to a multiple of WORD_EEW */ +SYM_FUNC_START(__riscv_copy_vec_words_unaligned) + andi a4, a2, ~(WORD_EEW-1) + beqz a4, 2f + add a3, a1, a4 + .option push + .option arch, +zve32x +1: + vsetivli t0, 8, WORD_SEW, m8, ta, ma + VEC_L v0, (a1) + VEC_S v0, (a0) + addi a0, a0, WORD_EEW + addi a1, a1, WORD_EEW + bltu a1, a3, 1b + +2: + .option pop + ret +SYM_FUNC_END(__riscv_copy_vec_words_unaligned) + +/* void __riscv_copy_vec_bytes_unaligned(void *, const void *, size_t) */ +/* Performs a memcpy without aligning buffers, using only byte accesses. */ +/* Note: The size is truncated to a multiple of 8 */ +SYM_FUNC_START(__riscv_copy_vec_bytes_unaligned) + andi a4, a2, ~(8-1) + beqz a4, 2f + add a3, a1, a4 + .option push + .option arch, +zve32x +1: + vsetivli t0, 8, e8, m8, ta, ma + vle8.v v0, (a1) + vse8.v v0, (a0) + addi a0, a0, 8 + addi a1, a1, 8 + bltu a1, a3, 1b + +2: + .option pop + ret +SYM_FUNC_END(__riscv_copy_vec_bytes_unaligned)