From patchwork Mon Jul 3 12:46:44 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Alexandre Ghiti X-Patchwork-Id: 13300040 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 9B011EB64DC for ; Mon, 3 Jul 2023 12:54:08 +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=KlbNc+mBOsqCXTxRxJ784Gu710zYvZ8TcD8QL1UAuEw=; b=dGlMMrx82Yh9er G0IuDky4tkXPANZCCNzpB7pVTF1Cn43lCFo1PNgWCm1ldLnP7u9zN0Dy2nCqCELZcVGii7qlRieIp XA3vbOVlcxb6JHUSEI2RJ9zrr8ErkJGN/yuyoYzwFfVF1grWvicvahdJtFzVudrvlqWYP7+pMQaBt fdd/ujhC3lFVOSHI4oEYB18YeubcOqIgUd2ZvcJbQgBRPQY7IR398dqajAZVD/1oMoLpUQ4HEAgCH uFAJOHgngfp/MruXuEgfyqKjBDWHiLkl22qnp5KIbP65SqKIOEzmFORdF9aFYQJU57AaiTmct+DYf UJV8MpPRWf1bPYD7VJig==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.96 #2 (Red Hat Linux)) id 1qGJ48-00AYDg-0o; Mon, 03 Jul 2023 12:54:04 +0000 Received: from mail-wm1-x330.google.com ([2a00:1450:4864:20::330]) by bombadil.infradead.org with esmtps (Exim 4.96 #2 (Red Hat Linux)) id 1qGJ45-00AYBY-2M for linux-riscv@lists.infradead.org; Mon, 03 Jul 2023 12:54:03 +0000 Received: by mail-wm1-x330.google.com with SMTP id 5b1f17b1804b1-3fb4146e8fcso37893605e9.0 for ; Mon, 03 Jul 2023 05:53:58 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rivosinc-com.20221208.gappssmtp.com; s=20221208; t=1688388837; x=1690980837; 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=EFx4ptfWql1VCkY+QAKSy8enzH3xoMQdKeo5FdCgpdA=; b=WL2SlqdJy2nf7qhtXjDFpSfFI3W6EJ9d7QlBrrUszJrt8kjAGu7mTPhwdcM0MrKvNG cTV114dR4bi/HNUnYvuEeyFcBHizLfOagaaW9OWwU/wNHnT1JOXtJudZleOC6aprfoP2 xDsK5DSiwJu8OOBxr+Pof+IziLN9QmZ51aYT+ZkIDCGRMH68dfG8CxBT0PGHZvN4pjZ8 mQddNQJPJd6ZZtrxm8yiynv5uM+AOw0+UW+rIpfwTF/1JYE5Z1RkfAJrxfZJVrAn1BKo 2+OBiZpC4BdLbBlFREP/wdgpmge64Dd+aOZVJZ0c7OH1gGhm9g477EzFYr7kF4TbgPa1 lhDg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1688388837; x=1690980837; 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=EFx4ptfWql1VCkY+QAKSy8enzH3xoMQdKeo5FdCgpdA=; b=D+8O2+SzLfEXC7gkRZpbhLfNORpL0XsOgS8vL5zOqmLyFhVG2V5Mvl4v0snznFZhTj cAx9xypdA88bK+asJ11+SeXaHpBTjArD4tuVt2mw84q05nCOv9VgmBetnJAo7p78d+Xa 0VZDiVe4TvpVuwS+ec02JlqjuiOrJ8C95etsQoX5NSXUut0tE7KE1l3xeafkF26ojHSe z6jTUAvxbbk9VEieJugiXkzTi1lfjDQrrre9b0m2wRax9EsLbhA5KSdwWNbFR1KN3cX/ T5W72rQ61ZA+283wh7kzLhLKlW+RCOs+rZ4zOCWKn0m8izl6Nlv3wk49sRqrth+1//a7 cXDQ== X-Gm-Message-State: AC+VfDyaIaFD4zFwcpiDnJd5X+wq+LDMikpsJVYPxsEqG9qpfP/33s3P hlvaAtGMUWDE2QmckQjELzSsgw== X-Google-Smtp-Source: ACHHUZ4N2tuGb+Vcz31VHfkql4J0f3Ibfvl2eWl4wUdJuir2vrWSCKs+ZAQjToM65s53t7xVGhOxMA== X-Received: by 2002:a05:600c:3d9b:b0:3fb:c257:9f48 with SMTP id bi27-20020a05600c3d9b00b003fbc2579f48mr9245072wmb.16.1688388837525; Mon, 03 Jul 2023 05:53:57 -0700 (PDT) Received: from alex-rivos.home (amontpellier-656-1-456-62.w92-145.abo.wanadoo.fr. [92.145.124.62]) by smtp.gmail.com with ESMTPSA id a10-20020a05600c224a00b003faef96ee78sm20711360wmm.33.2023.07.03.05.53.56 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 03 Jul 2023 05:53:57 -0700 (PDT) From: Alexandre Ghiti To: Jonathan Corbet , Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo , Mark Rutland , Alexander Shishkin , Jiri Olsa , Namhyung Kim , Ian Rogers , Paul Walmsley , Palmer Dabbelt , Albert Ou , Atish Patra , Anup Patel , Will Deacon , Rob Herring , Andrew Jones , linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org, linux-riscv@lists.infradead.org, linux-arm-kernel@lists.infradead.org Cc: Alexandre Ghiti Subject: [PATCH v4 07/10] drivers: perf: Implement perf event mmap support in the SBI backend Date: Mon, 3 Jul 2023 14:46:44 +0200 Message-Id: <20230703124647.215952-8-alexghiti@rivosinc.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20230703124647.215952-1-alexghiti@rivosinc.com> References: <20230703124647.215952-1-alexghiti@rivosinc.com> MIME-Version: 1.0 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20230703_055401_769837_9380046E X-CRM114-Status: GOOD ( 33.87 ) 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 We used to unconditionnally expose the cycle and instret csrs to userspace, which gives rise to security concerns. So now we only allow access to hw counters from userspace through the perf framework which will handle context switches, per-task events...etc. But as we cannot break userspace, we give the user the choice to go back to the previous behaviour by setting the sysctl perf_user_access. Signed-off-by: Alexandre Ghiti Reviewed-by: Andrew Jones --- drivers/perf/riscv_pmu.c | 10 +- drivers/perf/riscv_pmu_sbi.c | 192 +++++++++++++++++++++++++++++++++-- 2 files changed, 195 insertions(+), 7 deletions(-) diff --git a/drivers/perf/riscv_pmu.c b/drivers/perf/riscv_pmu.c index 432ad2e80ce3..80c052e93f9e 100644 --- a/drivers/perf/riscv_pmu.c +++ b/drivers/perf/riscv_pmu.c @@ -38,7 +38,15 @@ void arch_perf_update_userpage(struct perf_event *event, userpg->cap_user_time_short = 0; userpg->cap_user_rdpmc = riscv_perf_user_access(event); - userpg->pmc_width = 64; +#ifdef CONFIG_RISCV_PMU + /* + * The counters are 64-bit but the priv spec doesn't mandate all the + * bits to be implemented: that's why, counter width can vary based on + * the cpu vendor. + */ + if (userpg->cap_user_rdpmc) + userpg->pmc_width = to_riscv_pmu(event->pmu)->ctr_get_width(event->hw.idx) + 1; +#endif do { rd = sched_clock_read_begin(&seq); diff --git a/drivers/perf/riscv_pmu_sbi.c b/drivers/perf/riscv_pmu_sbi.c index 83c3f1c4d2f1..2236cc9aa4b8 100644 --- a/drivers/perf/riscv_pmu_sbi.c +++ b/drivers/perf/riscv_pmu_sbi.c @@ -24,6 +24,14 @@ #include #include +#define SYSCTL_NO_USER_ACCESS 0 +#define SYSCTL_USER_ACCESS 1 +#define SYSCTL_LEGACY 2 + +#define PERF_EVENT_FLAG_NO_USER_ACCESS BIT(SYSCTL_NO_USER_ACCESS) +#define PERF_EVENT_FLAG_USER_ACCESS BIT(SYSCTL_USER_ACCESS) +#define PERF_EVENT_FLAG_LEGACY BIT(SYSCTL_LEGACY) + PMU_FORMAT_ATTR(event, "config:0-47"); PMU_FORMAT_ATTR(firmware, "config:63"); @@ -43,6 +51,9 @@ static const struct attribute_group *riscv_pmu_attr_groups[] = { NULL, }; +/* Allow user mode access by default */ +static int sysctl_perf_user_access __read_mostly = SYSCTL_USER_ACCESS; + /* * RISC-V doesn't have heterogeneous harts yet. This need to be part of * per_cpu in case of harts with different pmu counters @@ -301,6 +312,11 @@ int riscv_pmu_get_hpm_info(u32 *hw_ctr_width, u32 *num_hw_ctr) } EXPORT_SYMBOL_GPL(riscv_pmu_get_hpm_info); +static uint8_t pmu_sbi_csr_index(struct perf_event *event) +{ + return pmu_ctr_list[event->hw.idx].csr - CSR_CYCLE; +} + static unsigned long pmu_sbi_get_filter_flags(struct perf_event *event) { unsigned long cflags = 0; @@ -329,18 +345,34 @@ static int pmu_sbi_ctr_get_idx(struct perf_event *event) struct cpu_hw_events *cpuc = this_cpu_ptr(rvpmu->hw_events); struct sbiret ret; int idx; - uint64_t cbase = 0; + uint64_t cbase = 0, cmask = rvpmu->cmask; unsigned long cflags = 0; cflags = pmu_sbi_get_filter_flags(event); + + /* + * In legacy mode, we have to force the fixed counters for those events + * but not in the user access mode as we want to use the other counters + * that support sampling/filtering. + */ + if (hwc->flags & PERF_EVENT_FLAG_LEGACY) { + if (event->attr.config == PERF_COUNT_HW_CPU_CYCLES) { + cflags |= SBI_PMU_CFG_FLAG_SKIP_MATCH; + cmask = 1; + } else if (event->attr.config == PERF_COUNT_HW_INSTRUCTIONS) { + cflags |= SBI_PMU_CFG_FLAG_SKIP_MATCH; + cmask = 1UL << (CSR_INSTRET - CSR_CYCLE); + } + } + /* retrieve the available counter index */ #if defined(CONFIG_32BIT) ret = sbi_ecall(SBI_EXT_PMU, SBI_EXT_PMU_COUNTER_CFG_MATCH, cbase, - rvpmu->cmask, cflags, hwc->event_base, hwc->config, + cmask, cflags, hwc->event_base, hwc->config, hwc->config >> 32); #else ret = sbi_ecall(SBI_EXT_PMU, SBI_EXT_PMU_COUNTER_CFG_MATCH, cbase, - rvpmu->cmask, cflags, hwc->event_base, hwc->config, 0); + cmask, cflags, hwc->event_base, hwc->config, 0); #endif if (ret.error) { pr_debug("Not able to find a counter for event %lx config %llx\n", @@ -474,6 +506,22 @@ static u64 pmu_sbi_ctr_read(struct perf_event *event) return val; } +static void pmu_sbi_set_scounteren(void *arg) +{ + struct perf_event *event = (struct perf_event *)arg; + + csr_write(CSR_SCOUNTEREN, + csr_read(CSR_SCOUNTEREN) | (1 << pmu_sbi_csr_index(event))); +} + +static void pmu_sbi_reset_scounteren(void *arg) +{ + struct perf_event *event = (struct perf_event *)arg; + + csr_write(CSR_SCOUNTEREN, + csr_read(CSR_SCOUNTEREN) & ~(1 << pmu_sbi_csr_index(event))); +} + static void pmu_sbi_ctr_start(struct perf_event *event, u64 ival) { struct sbiret ret; @@ -490,6 +538,10 @@ static void pmu_sbi_ctr_start(struct perf_event *event, u64 ival) if (ret.error && (ret.error != SBI_ERR_ALREADY_STARTED)) pr_err("Starting counter idx %d failed with error %d\n", hwc->idx, sbi_err_map_linux_errno(ret.error)); + + if ((hwc->flags & PERF_EVENT_FLAG_USER_ACCESS) && + (hwc->flags & PERF_EVENT_FLAG_USER_READ_CNT)) + pmu_sbi_set_scounteren((void *)event); } static void pmu_sbi_ctr_stop(struct perf_event *event, unsigned long flag) @@ -497,6 +549,10 @@ static void pmu_sbi_ctr_stop(struct perf_event *event, unsigned long flag) struct sbiret ret; struct hw_perf_event *hwc = &event->hw; + if ((hwc->flags & PERF_EVENT_FLAG_USER_ACCESS) && + (hwc->flags & PERF_EVENT_FLAG_USER_READ_CNT)) + pmu_sbi_reset_scounteren((void *)event); + ret = sbi_ecall(SBI_EXT_PMU, SBI_EXT_PMU_COUNTER_STOP, hwc->idx, 1, flag, 0, 0, 0); if (ret.error && (ret.error != SBI_ERR_ALREADY_STOPPED) && flag != SBI_PMU_STOP_FLAG_RESET) @@ -704,10 +760,13 @@ static int pmu_sbi_starting_cpu(unsigned int cpu, struct hlist_node *node) struct cpu_hw_events *cpu_hw_evt = this_cpu_ptr(pmu->hw_events); /* - * Enable the access for CYCLE, TIME, and INSTRET CSRs from userspace, - * as is necessary to maintain uABI compatibility. + * We keep enabling userspace access to CYCLE, TIME and INSRET via the + * legacy option but that will be removed in the future. */ - csr_write(CSR_SCOUNTEREN, 0x7); + if (sysctl_perf_user_access == SYSCTL_LEGACY) + csr_write(CSR_SCOUNTEREN, 0x7); + else + csr_write(CSR_SCOUNTEREN, 0x2); /* Stop all the counters so that they can be enabled from perf */ pmu_sbi_stop_all(pmu); @@ -851,6 +910,121 @@ static void riscv_pmu_destroy(struct riscv_pmu *pmu) cpuhp_state_remove_instance(CPUHP_AP_PERF_RISCV_STARTING, &pmu->node); } +static void pmu_sbi_event_init(struct perf_event *event) +{ + /* + * The permissions are set at event_init so that we do not depend + * on the sysctl value that can change. + */ + if (sysctl_perf_user_access == SYSCTL_NO_USER_ACCESS) + event->hw.flags |= PERF_EVENT_FLAG_NO_USER_ACCESS; + else if (sysctl_perf_user_access == SYSCTL_USER_ACCESS) + event->hw.flags |= PERF_EVENT_FLAG_USER_ACCESS; + else + event->hw.flags |= PERF_EVENT_FLAG_LEGACY; +} + +static void pmu_sbi_event_mapped(struct perf_event *event, struct mm_struct *mm) +{ + if (event->hw.flags & PERF_EVENT_FLAG_NO_USER_ACCESS) + return; + + if (event->hw.flags & PERF_EVENT_FLAG_LEGACY) { + if (event->attr.config != PERF_COUNT_HW_CPU_CYCLES && + event->attr.config != PERF_COUNT_HW_INSTRUCTIONS) { + return; + } + } + + /* + * The user mmapped the event to directly access it: this is where + * we determine based on sysctl_perf_user_access if we grant userspace + * the direct access to this event. That means that within the same + * task, some events may be directly accessible and some other may not, + * if the user changes the value of sysctl_perf_user_accesss in the + * meantime. + */ + + event->hw.flags |= PERF_EVENT_FLAG_USER_READ_CNT; + + /* + * We must enable userspace access *before* advertising in the user page + * that it is possible to do so to avoid any race. + * And we must notify all cpus here because threads that currently run + * on other cpus will try to directly access the counter too without + * calling pmu_sbi_ctr_start. + */ + if (event->hw.flags & PERF_EVENT_FLAG_USER_ACCESS) + on_each_cpu_mask(mm_cpumask(mm), + pmu_sbi_set_scounteren, (void *)event, 1); +} + +static void pmu_sbi_event_unmapped(struct perf_event *event, struct mm_struct *mm) +{ + if (event->hw.flags & PERF_EVENT_FLAG_NO_USER_ACCESS) + return; + + if (event->hw.flags & PERF_EVENT_FLAG_LEGACY) { + if (event->attr.config != PERF_COUNT_HW_CPU_CYCLES && + event->attr.config != PERF_COUNT_HW_INSTRUCTIONS) { + return; + } + } + + /* + * Here we can directly remove user access since the user does not have + * access to the user page anymore so we avoid the racy window where the + * user could have read cap_user_rdpmc to true right before we disable + * it. + */ + event->hw.flags &= ~PERF_EVENT_FLAG_USER_READ_CNT; + + if (event->hw.flags & PERF_EVENT_FLAG_USER_ACCESS) + on_each_cpu_mask(mm_cpumask(mm), + pmu_sbi_reset_scounteren, (void *)event, 1); +} + +static void riscv_pmu_update_counter_access(void *info) +{ + if (sysctl_perf_user_access == SYSCTL_LEGACY) + csr_write(CSR_SCOUNTEREN, 0x7); + else + csr_write(CSR_SCOUNTEREN, 0x2); +} + +static int riscv_pmu_proc_user_access_handler(struct ctl_table *table, + int write, void *buffer, + size_t *lenp, loff_t *ppos) +{ + int prev = sysctl_perf_user_access; + int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + + /* + * Test against the previous value since we clear SCOUNTEREN when + * sysctl_perf_user_access is set to SYSCTL_USER_ACCESS, but we should + * not do that if that was already the case. + */ + if (ret || !write || prev == sysctl_perf_user_access) + return ret; + + on_each_cpu(riscv_pmu_update_counter_access, NULL, 1); + + return 0; +} + +static struct ctl_table sbi_pmu_sysctl_table[] = { + { + .procname = "perf_user_access", + .data = &sysctl_perf_user_access, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = riscv_pmu_proc_user_access_handler, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_TWO, + }, + { } +}; + static int pmu_sbi_device_probe(struct platform_device *pdev) { struct riscv_pmu *pmu = NULL; @@ -888,6 +1062,10 @@ static int pmu_sbi_device_probe(struct platform_device *pdev) pmu->ctr_get_width = pmu_sbi_ctr_get_width; pmu->ctr_clear_idx = pmu_sbi_ctr_clear_idx; pmu->ctr_read = pmu_sbi_ctr_read; + pmu->event_init = pmu_sbi_event_init; + pmu->event_mapped = pmu_sbi_event_mapped; + pmu->event_unmapped = pmu_sbi_event_unmapped; + pmu->csr_index = pmu_sbi_csr_index; ret = cpuhp_state_add_instance(CPUHP_AP_PERF_RISCV_STARTING, &pmu->node); if (ret) @@ -901,6 +1079,8 @@ static int pmu_sbi_device_probe(struct platform_device *pdev) if (ret) goto out_unregister; + register_sysctl("kernel", sbi_pmu_sysctl_table); + return 0; out_unregister: