From patchwork Thu Oct 3 18:43:01 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Leo Yan X-Patchwork-Id: 13821416 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 E1AE0CF34BE for ; Thu, 3 Oct 2024 18:49:49 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender:List-Subscribe:List-Help :List-Post:List-Archive:List-Unsubscribe:List-Id:Content-Transfer-Encoding: MIME-Version:References:In-Reply-To:Message-Id:Date:Subject:Cc:To:From: Reply-To:Content-Type:Content-ID:Content-Description:Resent-Date:Resent-From: Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Owner; bh=Lus3xhLtWPkANrgrAdCVUTx+taYiST+DRJMUr7VgpO4=; b=rcAun8asi4lG/4kjbOfPRsCUfw zVRPXlcwiHCypIp6h2Dv2eDRy3qPkdIfsZA1pQeHazpaangXECt5yYqMnSUgVC1kE9Tu6aaN/NJ5P ZhLYW9wxSjy9COoBAJIIke2CyvH2JX6LRJXVj1In4e0GpJmnPgf1U+2w9QXRi/GxADBv7rToRL6JX 8PCaXZSEywDHdW0z2056j7M4aB/pCYeI40MF8/qVpTYVyWSHXu9IyhwftF4ahwkwmVnOsFG2RX1P1 hGxQJSULwRtQ3XOeraTWFDD7hFNONJbKLX5zmIvIgIxaCEFBGb2fN0HN1fnSmLAgH4XOlCu1ha0VF 90fQzZUQ==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.98 #2 (Red Hat Linux)) id 1swQtQ-0000000A7kI-48tW; Thu, 03 Oct 2024 18:49:40 +0000 Received: from foss.arm.com ([217.140.110.172]) by bombadil.infradead.org with esmtp (Exim 4.98 #2 (Red Hat Linux)) id 1swQnO-0000000A6g4-3fyY for linux-arm-kernel@lists.infradead.org; Thu, 03 Oct 2024 18:43:28 +0000 Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 77002497; Thu, 3 Oct 2024 11:43:54 -0700 (PDT) Received: from e132581.cambridge.arm.com (e132581.arm.com [10.2.76.71]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id E377F3F58B; Thu, 3 Oct 2024 11:43:22 -0700 (PDT) From: Leo Yan To: Arnaldo Carvalho de Melo , Namhyung Kim , Mark Rutland , Alexander Shishkin , Jiri Olsa , Ian Rogers , Adrian Hunter , "Liang, Kan" , Besar Wicaksono , linux-arm-kernel@lists.infradead.org, linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, Will Deacon , John Garry , James Clark , Mike Leach Cc: Leo Yan Subject: [PATCH v4 4/5] perf arm-spe: Support metadata version 2 Date: Thu, 3 Oct 2024 19:43:01 +0100 Message-Id: <20241003184302.190806-5-leo.yan@arm.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20241003184302.190806-1-leo.yan@arm.com> References: <20241003184302.190806-1-leo.yan@arm.com> MIME-Version: 1.0 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20241003_114327_192761_88DD7FC6 X-CRM114-Status: GOOD ( 20.16 ) X-BeenThere: linux-arm-kernel@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-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=archiver.kernel.org@lists.infradead.org This commit is to support metadata version 2 and at the meantime it is backward compatible for version 1's format. The metadata version 1 doesn't include the ARM_SPE_HEADER_VERSION field. As version 1 is fixed with two u64 fields, by checking the metadata size, it distinguishes the metadata is version 1 or version 2 (and any new versions if later will have). For version 2, it reads out CPU number and retrieves the metadata info for every CPU. Signed-off-by: Leo Yan Reviewed-by: James Clark --- tools/perf/util/arm-spe.c | 99 +++++++++++++++++++++++++++++++++++++-- 1 file changed, 95 insertions(+), 4 deletions(-) diff --git a/tools/perf/util/arm-spe.c b/tools/perf/util/arm-spe.c index 70989b1bae47..8b2af1c7dc9b 100644 --- a/tools/perf/util/arm-spe.c +++ b/tools/perf/util/arm-spe.c @@ -78,6 +78,10 @@ struct arm_spe { unsigned long num_events; u8 use_ctx_pkt_for_pid; + + u64 **metadata; + u64 metadata_ver; + u64 metadata_nr_cpu; }; struct arm_spe_queue { @@ -1016,6 +1020,73 @@ static int arm_spe_flush(struct perf_session *session __maybe_unused, return 0; } +static u64 *arm_spe__alloc_per_cpu_metadata(u64 *buf, int per_cpu_size) +{ + u64 *metadata; + + metadata = zalloc(per_cpu_size); + if (!metadata) + return NULL; + + memcpy(metadata, buf, per_cpu_size); + return metadata; +} + +static void arm_spe__free_metadata(u64 **metadata, int nr_cpu) +{ + int i; + + for (i = 0; i < nr_cpu; i++) + zfree(&metadata[i]); + free(metadata); +} + +static u64 **arm_spe__alloc_metadata(struct perf_record_auxtrace_info *info, + u64 *ver, int *nr_cpu) +{ + u64 *ptr = (u64 *)info->priv; + u64 metadata_size; + u64 **metadata = NULL; + int hdr_sz, per_cpu_sz, i; + + metadata_size = info->header.size - + sizeof(struct perf_record_auxtrace_info); + + /* Metadata version 1 */ + if (metadata_size == ARM_SPE_AUXTRACE_V1_PRIV_SIZE) { + *ver = 1; + *nr_cpu = 0; + /* No per CPU metadata */ + return NULL; + } + + *ver = ptr[ARM_SPE_HEADER_VERSION]; + hdr_sz = ptr[ARM_SPE_HEADER_SIZE]; + *nr_cpu = ptr[ARM_SPE_CPUS_NUM]; + + metadata = calloc(*nr_cpu, sizeof(*metadata)); + if (!metadata) + return NULL; + + /* Locate the start address of per CPU metadata */ + ptr += hdr_sz; + per_cpu_sz = (metadata_size - (hdr_sz * sizeof(u64))) / (*nr_cpu); + + for (i = 0; i < *nr_cpu; i++) { + metadata[i] = arm_spe__alloc_per_cpu_metadata(ptr, per_cpu_sz); + if (!metadata[i]) + goto err_per_cpu_metadata; + + ptr += per_cpu_sz / sizeof(u64); + } + + return metadata; + +err_per_cpu_metadata: + arm_spe__free_metadata(metadata, *nr_cpu); + return NULL; +} + static void arm_spe_free_queue(void *priv) { struct arm_spe_queue *speq = priv; @@ -1050,6 +1121,7 @@ static void arm_spe_free(struct perf_session *session) auxtrace_heap__free(&spe->heap); arm_spe_free_events(session); session->auxtrace = NULL; + arm_spe__free_metadata(spe->metadata, spe->metadata_nr_cpu); free(spe); } @@ -1267,15 +1339,26 @@ int arm_spe_process_auxtrace_info(union perf_event *event, const char *cpuid = perf_env__cpuid(session->evlist->env); u64 midr = strtol(cpuid, NULL, 16); struct arm_spe *spe; - int err; + u64 **metadata = NULL; + u64 metadata_ver; + int nr_cpu, err; if (auxtrace_info->header.size < sizeof(struct perf_record_auxtrace_info) + min_sz) return -EINVAL; + metadata = arm_spe__alloc_metadata(auxtrace_info, &metadata_ver, + &nr_cpu); + if (!metadata && metadata_ver != 1) { + pr_err("Failed to parse Arm SPE metadata.\n"); + return -EINVAL; + } + spe = zalloc(sizeof(struct arm_spe)); - if (!spe) - return -ENOMEM; + if (!spe) { + err = -ENOMEM; + goto err_free_metadata; + } err = auxtrace_queues__init(&spe->queues); if (err) @@ -1284,8 +1367,14 @@ int arm_spe_process_auxtrace_info(union perf_event *event, spe->session = session; spe->machine = &session->machines.host; /* No kvm support */ spe->auxtrace_type = auxtrace_info->type; - spe->pmu_type = auxtrace_info->priv[ARM_SPE_PMU_TYPE]; + if (metadata_ver == 1) + spe->pmu_type = auxtrace_info->priv[ARM_SPE_PMU_TYPE]; + else + spe->pmu_type = auxtrace_info->priv[ARM_SPE_PMU_TYPE_V2]; spe->midr = midr; + spe->metadata = metadata; + spe->metadata_ver = metadata_ver; + spe->metadata_nr_cpu = nr_cpu; spe->timeless_decoding = arm_spe__is_timeless_decoding(spe); @@ -1346,5 +1435,7 @@ int arm_spe_process_auxtrace_info(union perf_event *event, session->auxtrace = NULL; err_free: free(spe); +err_free_metadata: + arm_spe__free_metadata(metadata, nr_cpu); return err; }