From patchwork Mon Dec 18 13:15:35 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Masami Hiramatsu (Google)" X-Patchwork-Id: 180395 Return-Path: Delivered-To: ouuuleilei@gmail.com Received: by 2002:a05:7300:24d3:b0:fb:cd0c:d3e with SMTP id r19csp1234685dyi; Mon, 18 Dec 2023 05:22:22 -0800 (PST) X-Google-Smtp-Source: AGHT+IFL4FXbUeFXdMa60X9d922MSaFTl9DOKgOW1c2jPH5nTIBvV3bpsA3aNr77ZOG/Usc47jlY X-Received: by 2002:a05:620a:298c:b0:77f:b26c:7901 with SMTP id r12-20020a05620a298c00b0077fb26c7901mr6686543qkp.117.1702905742209; Mon, 18 Dec 2023 05:22:22 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1702905742; cv=none; d=google.com; s=arc-20160816; b=o7J8dc7dfLl2wo60WIlT8qlGu2EJgPbF1i5LGLOltSOcie8c7af4fgZpg+4wFiaQ28 Uegt7BXmpmxZLz9C08pecmBrlCLnJw+FMtAk5FY5SECqHermdU16Ph5cVMYISLX3Xtxf xDwBFRSsCINcA2TVTsCFlNFxuvLcKRCdNfIS37uoTCzGJy7BtXo/97Enf0WN+ZnUVKDj XA9baYQ4QNTVgI3gc1wl2h1c3IU2mjxA6s5vlGzyz7vSFg9gnnci65pK7mK/8WQDRdid rE5i6tY3EFZ9YkS2Ah2UOq4tO+GSSB/BzflupsPTyS3HN5gaUoab8P3yFFsOjk2Ra2aN AzFg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=content-transfer-encoding:mime-version:list-unsubscribe :list-subscribe:list-id:precedence:user-agent:references:in-reply-to :message-id:date:subject:cc:to:from:dkim-signature; bh=isVw+NG/j29F9cKQWy0NiSxONjei2mHFXUwHldPqheA=; fh=SIgps5XdV0XNwjZfT2uAI7g3mrspDldK9Qs8qQAfoa4=; b=q1ttT4OhO2SAziejy3WkKbt8Ptp/Y+odDRJYokkVMIlpb0a3tA/VuFjnRApD2cXiaM FxQyE5Sa69iIbw0Z5VMCjv/7dwaHCDEY3QXVepXeS0im6i0D1R4YoAHegercQF7Y/53C 726uqKIoQg0nkusOszpy0K87Sw2/6+827JtfNFNmTza3OMNAsHLfKVF6S3Ji1OQGQrMT ZMuRQIKY4sLthzDdLW2tD5+t6z+FDcWAE9VdC4zXI0JgOTkSeCwbU7LGd13f3SwAa3Lo TA3+zeNnbiDGrFrZGssu8F4KWpqybNnH9BR5FvVQ8/jeoEcG/9MPkOGT/C/pYRCCW63X bd0w== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@kernel.org header.s=k20201202 header.b=s6TmIOk7; spf=pass (google.com: domain of linux-kernel+bounces-3726-ouuuleilei=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) smtp.mailfrom="linux-kernel+bounces-3726-ouuuleilei=gmail.com@vger.kernel.org"; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Received: from ny.mirrors.kernel.org (ny.mirrors.kernel.org. [2604:1380:45d1:ec00::1]) by mx.google.com with ESMTPS id l6-20020a05620a28c600b0077f64af994csi20427424qkp.555.2023.12.18.05.22.22 for (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 18 Dec 2023 05:22:22 -0800 (PST) Received-SPF: pass (google.com: domain of linux-kernel+bounces-3726-ouuuleilei=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1; Authentication-Results: mx.google.com; dkim=pass header.i=@kernel.org header.s=k20201202 header.b=s6TmIOk7; spf=pass (google.com: domain of linux-kernel+bounces-3726-ouuuleilei=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) smtp.mailfrom="linux-kernel+bounces-3726-ouuuleilei=gmail.com@vger.kernel.org"; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Received: from smtp.subspace.kernel.org (wormhole.subspace.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ny.mirrors.kernel.org (Postfix) with ESMTPS id EAA941C24764 for ; Mon, 18 Dec 2023 13:22:21 +0000 (UTC) Received: from localhost.localdomain (localhost.localdomain [127.0.0.1]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 4E58174092; Mon, 18 Dec 2023 13:15:45 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="s6TmIOk7" X-Original-To: linux-kernel@vger.kernel.org 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 750FD3A1AF; Mon, 18 Dec 2023 13:15:42 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 23E8DC433C8; Mon, 18 Dec 2023 13:15:37 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1702905342; bh=LOpm3UNcQ+E/4+YV3fM2gRBGhvVbnPGvdL7zuuXUPv8=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=s6TmIOk7bJ12Z3yXL7sAA0q3qFtQERDf4EvCYotSIBJbRxR5w37exvk5ArzS5Fqzo +cW5aEZSlH3BTd/a5b2ZrgOT4eFbaePxaQ/xhcAKk51W7a13yiyolxtVnf7bas9cBR eNYWyFEramLI/hhx+2dJAoq5pylCoPmzHhm49AS5qQ4B9lFKT/PPOLYEzOb54Y8QGX bdbs2z36Nb8OHQIgJIB6J4RQQd2haNKSejDsAL7lDYvSrqy1vN5B++Y6M9n9LYbZAe yjocvSWReh1mZqjtv1cuLeMjGdQ8fnG9/n3r2DoHnCEGRgZ3ITpivyZIS9tRnrEFan qoTJqX2Ayu23w== From: "Masami Hiramatsu (Google)" To: Alexei Starovoitov , Steven Rostedt , Florent Revest Cc: linux-trace-kernel@vger.kernel.org, LKML , Martin KaFai Lau , bpf , Sven Schnelle , Alexei Starovoitov , Jiri Olsa , Arnaldo Carvalho de Melo , Daniel Borkmann , Alan Maguire , Mark Rutland , Peter Zijlstra , Thomas Gleixner , Guo Ren Subject: [PATCH v5 20/34] function_graph: Add a new exit handler with parent_ip and ftrace_regs Date: Mon, 18 Dec 2023 22:15:35 +0900 Message-Id: <170290533538.220107.1881096843495121692.stgit@devnote2> X-Mailer: git-send-email 2.34.1 In-Reply-To: <170290509018.220107.1347127510564358608.stgit@devnote2> References: <170290509018.220107.1347127510564358608.stgit@devnote2> User-Agent: StGit/0.19 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-getmail-retrieved-from-mailbox: INBOX X-GMAIL-THRID: 1785626091306151762 X-GMAIL-MSGID: 1785626091306151762 From: Masami Hiramatsu (Google) Add a new return handler to fgraph_ops as 'retregfunc' which takes parent_ip and ftrace_regs instead of ftrace_graph_ret. This handler is available only if the arch support CONFIG_HAVE_FUNCTION_GRAPH_FREGS. Note that the 'retfunc' and 'reregfunc' are mutual exclusive. You can set only one of them. Signed-off-by: Masami Hiramatsu (Google) --- Changes in v3: - Update for new multiple fgraph. - Save the return address to instruction pointer in ftrace_regs. --- arch/x86/include/asm/ftrace.h | 2 + include/linux/ftrace.h | 10 +++++- kernel/trace/Kconfig | 5 ++- kernel/trace/fgraph.c | 70 ++++++++++++++++++++++++++++------------- 4 files changed, 63 insertions(+), 24 deletions(-) diff --git a/arch/x86/include/asm/ftrace.h b/arch/x86/include/asm/ftrace.h index 415cf7a2ec2c..0b306c82855d 100644 --- a/arch/x86/include/asm/ftrace.h +++ b/arch/x86/include/asm/ftrace.h @@ -72,6 +72,8 @@ arch_ftrace_get_regs(struct ftrace_regs *fregs) override_function_with_return(&(fregs)->regs) #define ftrace_regs_query_register_offset(name) \ regs_query_register_offset(name) +#define ftrace_regs_get_frame_pointer(fregs) \ + frame_pointer(&(fregs)->regs) struct ftrace_ops; #define ftrace_graph_func ftrace_graph_func diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index 6da6cc9aaeaf..79875a00c02b 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -43,7 +43,9 @@ struct dyn_ftrace; char *arch_ftrace_match_adjust(char *str, const char *search); -#ifdef CONFIG_HAVE_FUNCTION_GRAPH_RETVAL +#ifdef CONFIG_HAVE_FUNCTION_GRAPH_FREGS +unsigned long ftrace_return_to_handler(struct ftrace_regs *fregs); +#elif defined(CONFIG_HAVE_FUNCTION_GRAPH_RETVAL) struct fgraph_ret_regs; unsigned long ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs); #else @@ -157,6 +159,7 @@ struct ftrace_regs { #define ftrace_regs_set_instruction_pointer(fregs, ip) do { } while (0) #endif /* CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS */ + static __always_inline struct pt_regs *ftrace_get_regs(struct ftrace_regs *fregs) { if (!fregs) @@ -1067,6 +1070,10 @@ typedef int (*trace_func_graph_regs_ent_t)(unsigned long func, unsigned long parent_ip, struct ftrace_regs *fregs, struct fgraph_ops *); /* entry w/ regs */ +typedef void (*trace_func_graph_regs_ret_t)(unsigned long func, + unsigned long parent_ip, + struct ftrace_regs *, + struct fgraph_ops *); /* return w/ regs */ extern int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace, struct fgraph_ops *gops); @@ -1076,6 +1083,7 @@ struct fgraph_ops { trace_func_graph_ent_t entryfunc; trace_func_graph_ret_t retfunc; trace_func_graph_regs_ent_t entryregfunc; + trace_func_graph_regs_ret_t retregfunc; struct ftrace_ops ops; /* for the hash lists */ void *private; int idx; diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 61c541c36596..308b3bec01b1 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -34,6 +34,9 @@ config HAVE_FUNCTION_GRAPH_TRACER config HAVE_FUNCTION_GRAPH_RETVAL bool +config HAVE_FUNCTION_GRAPH_FREGS + bool + config HAVE_DYNAMIC_FTRACE bool help @@ -232,7 +235,7 @@ config FUNCTION_GRAPH_TRACER config FUNCTION_GRAPH_RETVAL bool "Kernel Function Graph Return Value" - depends on HAVE_FUNCTION_GRAPH_RETVAL + depends on HAVE_FUNCTION_GRAPH_RETVAL || HAVE_FUNCTION_GRAPH_FREGS depends on FUNCTION_GRAPH_TRACER default n help diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index 52fb63f8251c..088432b695a6 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -685,8 +685,8 @@ int function_graph_enter_ops(unsigned long ret, unsigned long func, /* Retrieve a function return address to the trace stack on thread info.*/ static struct ftrace_ret_stack * -ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret, - unsigned long frame_pointer, int *index) +ftrace_pop_return_trace(unsigned long *ret, unsigned long frame_pointer, + int *index) { struct ftrace_ret_stack *ret_stack; @@ -731,10 +731,6 @@ ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret, *index += FGRAPH_RET_INDEX; *ret = ret_stack->ret; - trace->func = ret_stack->func; - trace->calltime = ret_stack->calltime; - trace->overrun = atomic_read(¤t->trace_overrun); - trace->depth = current->curr_ret_depth; /* * We still want to trace interrupts coming in if * max_depth is set to 1. Make sure the decrement is @@ -773,21 +769,42 @@ static struct notifier_block ftrace_suspend_notifier = { /* fgraph_ret_regs is not defined without CONFIG_FUNCTION_GRAPH_RETVAL */ struct fgraph_ret_regs; +static void fgraph_call_retfunc(struct ftrace_regs *fregs, + struct fgraph_ret_regs *ret_regs, + struct ftrace_ret_stack *ret_stack, + struct fgraph_ops *gops) +{ + struct ftrace_graph_ret trace; + + trace.func = ret_stack->func; + trace.calltime = ret_stack->calltime; + trace.overrun = atomic_read(¤t->trace_overrun); + trace.depth = current->curr_ret_depth; + trace.rettime = trace_clock_local(); +#ifdef CONFIG_FUNCTION_GRAPH_RETVAL + if (fregs) + trace.retval = ftrace_regs_return_value(fregs); + else + trace.retval = fgraph_ret_regs_return_value(ret_regs); +#endif + gops->retfunc(&trace, gops); +} + /* * Send the trace to the ring-buffer. * @return the original return address. */ -static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs, +static unsigned long __ftrace_return_to_handler(struct ftrace_regs *fregs, + struct fgraph_ret_regs *ret_regs, unsigned long frame_pointer) { struct ftrace_ret_stack *ret_stack; - struct ftrace_graph_ret trace; unsigned long bitmap; unsigned long ret; int index; int i; - ret_stack = ftrace_pop_return_trace(&trace, &ret, frame_pointer, &index); + ret_stack = ftrace_pop_return_trace(&ret, frame_pointer, &index); if (unlikely(!ret_stack)) { ftrace_graph_stop(); @@ -796,10 +813,8 @@ static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs return (unsigned long)panic; } - trace.rettime = trace_clock_local(); -#ifdef CONFIG_FUNCTION_GRAPH_RETVAL - trace.retval = fgraph_ret_regs_return_value(ret_regs); -#endif + if (fregs) + ftrace_regs_set_instruction_pointer(fregs, ret); bitmap = get_fgraph_index_bitmap(current, index); for (i = 0; i < FGRAPH_ARRAY_SIZE; i++) { @@ -810,7 +825,10 @@ static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs if (gops == &fgraph_stub) continue; - gops->retfunc(&trace, gops); + if (gops->retregfunc) + gops->retregfunc(ret_stack->func, ret, fregs, gops); + else + fgraph_call_retfunc(fregs, ret_regs, ret_stack, gops); } /* @@ -825,20 +843,22 @@ static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs return ret; } -/* - * After all architecures have selected HAVE_FUNCTION_GRAPH_RETVAL, we can - * leave only ftrace_return_to_handler(ret_regs). - */ -#ifdef CONFIG_HAVE_FUNCTION_GRAPH_RETVAL +#ifdef CONFIG_HAVE_FUNCTION_GRAPH_FREGS +unsigned long ftrace_return_to_handler(struct ftrace_regs *fregs) +{ + return __ftrace_return_to_handler(fregs, NULL, + ftrace_regs_get_frame_pointer(fregs)); +} +#elif defined(CONFIG_HAVE_FUNCTION_GRAPH_RETVAL) unsigned long ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs) { - return __ftrace_return_to_handler(ret_regs, + return __ftrace_return_to_handler(NULL, ret_regs, fgraph_ret_regs_frame_pointer(ret_regs)); } #else unsigned long ftrace_return_to_handler(unsigned long frame_pointer) { - return __ftrace_return_to_handler(NULL, frame_pointer); + return __ftrace_return_to_handler(NULL, NULL, frame_pointer); } #endif @@ -1191,9 +1211,15 @@ int register_ftrace_graph(struct fgraph_ops *gops) int ret = 0; int i; - if (gops->entryfunc && gops->entryregfunc) + if ((gops->entryfunc && gops->entryregfunc) || + (gops->retfunc && gops->retregfunc)) return -EINVAL; +#ifndef CONFIG_HAVE_FUNCTION_GRAPH_FREGS + if (gops->retregfunc) + return -EOPNOTSUPP; +#endif + mutex_lock(&ftrace_lock); if (!gops->ops.func) {