From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id C77CC537FD; Mon, 8 Jan 2024 15:42:43 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="jMsrLN2X" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 2E135C433C9; Mon, 8 Jan 2024 15:42:42 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1704728563; bh=H/R/nhiya/oFUBkyRvt/rXR8LicDkrGN0ai8oDSTnE8=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=jMsrLN2X0CTGev0CpXebBcmEWYB+g9AW216/0Q++s7zIvng/98cHtbt0BMvYKvF0V khz9k/XCrO8pQL89g6tnUz9fHeV+zusy6hoFFl2lSRcOs/l84jhvEIVkLc7oB4R7Ua LkOngYlGlDtAnw95Mt55lcqpTiniW1RE3bmnaFYA= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Thomas Richter , Sumanth Korikkar , Heiko Carstens , Sasha Levin Subject: [PATCH 6.1 089/150] s390/cpumf: support user space events for counting Date: Mon, 8 Jan 2024 16:35:40 +0100 Message-ID: <20240108153515.337386634@linuxfoundation.org> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240108153511.214254205@linuxfoundation.org> References: <20240108153511.214254205@linuxfoundation.org> User-Agent: quilt/0.67 X-stable: review X-Patchwork-Hint: ignore Precedence: bulk X-Mailing-List: patches@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit 6.1-stable review patch. If anyone has any objections, please let me know. ------------------ From: Thomas Richter [ Upstream commit 91d5364dc673fa9cf3a5b7b30cf33c70803eb3a4 ] CPU Measurement counting facility events PROBLEM_STATE_CPU_CYCLES(32) and PROBLEM_STATE_INSTRUCTIONS(33) are valid events. However the device driver returns error -EOPNOTSUPP when these event are to be installed. Fix this and allow installation of events PROBLEM_STATE_CPU_CYCLES, PROBLEM_STATE_CPU_CYCLES:u, PROBLEM_STATE_INSTRUCTIONS and PROBLEM_STATE_INSTRUCTIONS:u. Kernel space counting only is still not supported by s390. Signed-off-by: Thomas Richter Acked-by: Sumanth Korikkar Signed-off-by: Heiko Carstens Stable-dep-of: 09cda0a40051 ("s390/mm: add missing arch_set_page_dat() call to vmem_crst_alloc()") Signed-off-by: Sasha Levin --- arch/s390/kernel/perf_cpum_cf.c | 35 ++++++++++++++++++++++----------- 1 file changed, 24 insertions(+), 11 deletions(-) diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c index f043a7ff220b7..28fa80fd69fa0 100644 --- a/arch/s390/kernel/perf_cpum_cf.c +++ b/arch/s390/kernel/perf_cpum_cf.c @@ -2,7 +2,7 @@ /* * Performance event support for s390x - CPU-measurement Counter Facility * - * Copyright IBM Corp. 2012, 2021 + * Copyright IBM Corp. 2012, 2022 * Author(s): Hendrik Brueckner * Thomas Richter */ @@ -434,6 +434,12 @@ static void cpumf_hw_inuse(void) mutex_unlock(&pmc_reserve_mutex); } +static int is_userspace_event(u64 ev) +{ + return cpumf_generic_events_user[PERF_COUNT_HW_CPU_CYCLES] == ev || + cpumf_generic_events_user[PERF_COUNT_HW_INSTRUCTIONS] == ev; +} + static int __hw_perf_event_init(struct perf_event *event, unsigned int type) { struct perf_event_attr *attr = &event->attr; @@ -456,19 +462,26 @@ static int __hw_perf_event_init(struct perf_event *event, unsigned int type) if (is_sampling_event(event)) /* No sampling support */ return -ENOENT; ev = attr->config; - /* Count user space (problem-state) only */ if (!attr->exclude_user && attr->exclude_kernel) { - if (ev >= ARRAY_SIZE(cpumf_generic_events_user)) - return -EOPNOTSUPP; - ev = cpumf_generic_events_user[ev]; - - /* No support for kernel space counters only */ + /* + * Count user space (problem-state) only + * Handle events 32 and 33 as 0:u and 1:u + */ + if (!is_userspace_event(ev)) { + if (ev >= ARRAY_SIZE(cpumf_generic_events_user)) + return -EOPNOTSUPP; + ev = cpumf_generic_events_user[ev]; + } } else if (!attr->exclude_kernel && attr->exclude_user) { + /* No support for kernel space counters only */ return -EOPNOTSUPP; - } else { /* Count user and kernel space */ - if (ev >= ARRAY_SIZE(cpumf_generic_events_basic)) - return -EOPNOTSUPP; - ev = cpumf_generic_events_basic[ev]; + } else { + /* Count user and kernel space, incl. events 32 + 33 */ + if (!is_userspace_event(ev)) { + if (ev >= ARRAY_SIZE(cpumf_generic_events_basic)) + return -EOPNOTSUPP; + ev = cpumf_generic_events_basic[ev]; + } } break; -- 2.43.0