From patchwork Wed Feb 1 16:34:19 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Florent Revest X-Patchwork-Id: 13124734 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 F066DC05027 for ; Wed, 1 Feb 2023 16:38:28 +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=WhWI2tI0aXXA4C6qJoy/NGMF8OLJplDOtXmCX6YpArM=; b=CYy5W2hvJSO+Ck DIP45yP5O/Q6m5F38rNLyYERe2CMpA9EigRfY4qQFsTWJL1VWZBVbjJ04yTZLDOb/pV1lfRqUiy8t bTKej39zmkGCp4f7r4tlEJakoRXbg1YmFoTU0r6xzgLbDJSLJFrBNiksuJ7761Hs5pI5Sh6G89WYy aLqcCVPmF7SgNRlRTJGtn6y2tNUDg091PET9wM8dHlps3m4TstVQeUc6zh8Y56yikYmAfomaoSAw0 avBWulqBpyVRV+ADuVRpWB9svdIoZK2t4zvFN/OKdcQyG1yoxKXpvY2fzgqs+jlV9+YrGNuW6RzYp L423H99kSXPs93Djy4mw==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.94.2 #2 (Red Hat Linux)) id 1pNG6r-00CpLW-B9; Wed, 01 Feb 2023 16:37:21 +0000 Received: from mail-wm1-x32e.google.com ([2a00:1450:4864:20::32e]) by bombadil.infradead.org with esmtps (Exim 4.94.2 #2 (Red Hat Linux)) id 1pNG5A-00CoUK-Rb for linux-arm-kernel@lists.infradead.org; Wed, 01 Feb 2023 16:35:38 +0000 Received: by mail-wm1-x32e.google.com with SMTP id bg26so7277318wmb.0 for ; Wed, 01 Feb 2023 08:35:34 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=chromium.org; s=google; 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=MbHY+Gmtwa2HHtiXzFaehEcw/QkGNqN5J6j1VumLZ5w=; b=B5Kk5qpWjq/NWChUoJdLAZ15WLm2v1L//Jj1sE7X8MZC3AVQVVYby77BqsgXYZk/qt b8j//orYYdKgUmjIVviXcRXUnAipPwUDBe7ZWqHGBPP1TTVK+p6qMOTArnZw+wO7ZigT 3gGskC/gE5tLjsWGLwCN3/IVdjIhSxpQLR9bo= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; 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=MbHY+Gmtwa2HHtiXzFaehEcw/QkGNqN5J6j1VumLZ5w=; b=H4KoTt+DUBvMUqvVR4j2gFlJWS/kcuoKIZc7jA70BDKUMT6DsWIixROKLOwvu034c3 88t0NDWv/zjq7zq3iTZf71MOJP+i3AMyymlUVviO/qWY2sBZ3mB21TK/J2Ng5z5UH2XH jnOLwXYGBrDnmrv4ul7WNgYH3ZHps+9HUTQoZafHNhj5RDSpP7lpVeDPQDmYwDHBcQKr LP1khpl9EePBug8/SxBcYcCM4ij9nRGUY0wWoRsqGCXZfBH8ivXS6X2nknYZmPG7D+nI sJagfYdpC5+G8f2Z41+k2VMIlRVTM3HP+UT6xc8OMHLMLg6t4004U+DCH0NBaRIubI0u 5axw== X-Gm-Message-State: AO0yUKVWtZVE8kiLSc6HRcxR96jorrYTHZGR/qd/y+8a5R1Sp/FYxjSO zecxYmnkB1Xd1emb0r6KvGo0z39VASCyIBd//1c= X-Google-Smtp-Source: AK7set9M34kub5/FQQZ2rEhTdT78awUB8bbp7/omx+W0n7FyqEelSYMdNpH/uDndWUi0UrY9espvyQ== X-Received: by 2002:a7b:c5ca:0:b0:3dd:e38b:a12f with SMTP id n10-20020a7bc5ca000000b003dde38ba12fmr5384458wmk.40.1675269332734; Wed, 01 Feb 2023 08:35:32 -0800 (PST) Received: from revest.zrh.corp.google.com ([2a00:79e0:9d:6:4399:89a1:4a86:9630]) by smtp.gmail.com with ESMTPSA id r38-20020a05600c322600b003dd7edcc960sm2058522wmp.45.2023.02.01.08.35.31 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 01 Feb 2023 08:35:32 -0800 (PST) From: Florent Revest To: linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, linux-trace-kernel@vger.kernel.org, bpf@vger.kernel.org Cc: catalin.marinas@arm.com, will@kernel.org, rostedt@goodmis.org, mhiramat@kernel.org, mark.rutland@arm.com, ast@kernel.org, daniel@iogearbox.net, andrii@kernel.org, kpsingh@kernel.org, jolsa@kernel.org, xukuohai@huaweicloud.com, Florent Revest Subject: [PATCH 7/8] arm64: ftrace: Add direct call support Date: Wed, 1 Feb 2023 17:34:19 +0100 Message-Id: <20230201163420.1579014-8-revest@chromium.org> X-Mailer: git-send-email 2.39.1.519.gcb327c4b5f-goog In-Reply-To: <20230201163420.1579014-1-revest@chromium.org> References: <20230201163420.1579014-1-revest@chromium.org> MIME-Version: 1.0 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20230201_083536_948728_A7D67BAB X-CRM114-Status: GOOD ( 26.53 ) 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 builds up on the CALL_OPS work which extends the ftrace patchsite on arm64 with an ops pointer usable by the ftrace trampoline. This ops pointer is valid at all time. Indeed, it is either pointing to ftrace_list_ops or to the single ops which should be called from that patchsite. There are a few cases to distinguish: - If a direct call ops is the only one tracing a function: - If the direct called trampoline is within the reach of a BL instruction -> the ftrace patchsite jumps to the trampoline - Else -> the ftrace patchsite jumps to the ftrace_caller trampoline which reads the ops pointer in the patchsite and jumps to the direct call address stored in the ops - Else -> the ftrace patchsite jumps to the ftrace_caller trampoline and its ops literal points to ftrace_list_ops so it iterates over all registered ftrace ops, including the direct call ops and calls its call_direct_funcs handler which stores the direct called trampoline's address in the ftrace_regs and the ftrace_caller trampoline will return to that address instead of returning to the traced function Signed-off-by: Florent Revest Suggested-by: Mark Rutland --- arch/arm64/Kconfig | 2 ++ arch/arm64/include/asm/ftrace.h | 17 +++++++++ arch/arm64/kernel/asm-offsets.c | 6 ++++ arch/arm64/kernel/entry-ftrace.S | 60 +++++++++++++++++++++++--------- arch/arm64/kernel/ftrace.c | 36 ++++++++++++++++--- 5 files changed, 100 insertions(+), 21 deletions(-) diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 6f6f37161cf6..7deafd653c42 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -188,6 +188,8 @@ config ARM64 select HAVE_DYNAMIC_FTRACE select HAVE_DYNAMIC_FTRACE_WITH_ARGS \ if $(cc-option,-fpatchable-function-entry=2) + select HAVE_DYNAMIC_FTRACE_WITH_DIRECT_CALLS \ + if DYNAMIC_FTRACE_WITH_ARGS && DYNAMIC_FTRACE_WITH_CALL_OPS select HAVE_DYNAMIC_FTRACE_WITH_CALL_OPS \ if (DYNAMIC_FTRACE_WITH_ARGS && !CFI_CLANG) select FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY \ diff --git a/arch/arm64/include/asm/ftrace.h b/arch/arm64/include/asm/ftrace.h index cf6d9c42ff36..9fb95966b6d5 100644 --- a/arch/arm64/include/asm/ftrace.h +++ b/arch/arm64/include/asm/ftrace.h @@ -80,6 +80,10 @@ struct ftrace_regs { unsigned long sp; unsigned long pc; + +#ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS + unsigned long custom_tramp; +#endif }; static __always_inline unsigned long @@ -136,6 +140,19 @@ int ftrace_init_nop(struct module *mod, struct dyn_ftrace *rec); void ftrace_graph_func(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct ftrace_regs *fregs); #define ftrace_graph_func ftrace_graph_func + +#ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS +static inline void arch_ftrace_set_direct_caller(struct ftrace_regs *fregs, + unsigned long addr) +{ + /* + * The ftrace trampoline will return to this address instead of the + * instrumented function. + */ + fregs->custom_tramp = addr; +} +#endif /* CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS */ + #endif #define ftrace_return_address(n) return_address(n) diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c index ae345b06e9f7..c67f0373b88c 100644 --- a/arch/arm64/kernel/asm-offsets.c +++ b/arch/arm64/kernel/asm-offsets.c @@ -93,6 +93,9 @@ int main(void) DEFINE(FREGS_LR, offsetof(struct ftrace_regs, lr)); DEFINE(FREGS_SP, offsetof(struct ftrace_regs, sp)); DEFINE(FREGS_PC, offsetof(struct ftrace_regs, pc)); +#ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS + DEFINE(FREGS_CUSTOM_TRAMP, offsetof(struct ftrace_regs, custom_tramp)); +#endif DEFINE(FREGS_SIZE, sizeof(struct ftrace_regs)); BLANK(); #endif @@ -197,6 +200,9 @@ int main(void) #endif #ifdef CONFIG_FUNCTION_TRACER DEFINE(FTRACE_OPS_FUNC, offsetof(struct ftrace_ops, func)); +#ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS + DEFINE(FTRACE_OPS_DIRECT_CALL, offsetof(struct ftrace_ops, direct_call)); +#endif #endif return 0; } diff --git a/arch/arm64/kernel/entry-ftrace.S b/arch/arm64/kernel/entry-ftrace.S index 9869debd22fb..0576f38e6362 100644 --- a/arch/arm64/kernel/entry-ftrace.S +++ b/arch/arm64/kernel/entry-ftrace.S @@ -36,6 +36,30 @@ SYM_CODE_START(ftrace_caller) bti c +#ifdef CONFIG_DYNAMIC_FTRACE_WITH_CALL_OPS + /* + * The literal pointer to the ops is at an 8-byte aligned boundary + * which is either 12 or 16 bytes before the BL instruction in the call + * site. See ftrace_call_adjust() for details. + * + * Therefore here the LR points at `literal + 16` or `literal + 20`, + * and we can find the address of the literal in either case by + * aligning to an 8-byte boundary and subtracting 16. We do the + * alignment first as this allows us to fold the subtraction into the + * LDR. + */ + bic x11, x30, 0x7 + ldr x11, [x11, #-(4 * AARCH64_INSN_SIZE)] // op + +#ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS + /* If the op has a direct call address, return to it */ + ldr x12, [x11, #FTRACE_OPS_DIRECT_CALL] // op->direct_call + cbz x12, 1f + ret x12 +1: +#endif +#endif + /* Save original SP */ mov x10, sp @@ -57,6 +81,11 @@ SYM_CODE_START(ftrace_caller) /* Save the PC after the ftrace callsite */ str x30, [sp, #FREGS_PC] +#ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS + /* Set custom_tramp to zero */ + str xzr, [sp, #FREGS_CUSTOM_TRAMP] +#endif + /* Create a frame record for the callsite above the ftrace regs */ stp x29, x9, [sp, #FREGS_SIZE + 16] add x29, sp, #FREGS_SIZE + 16 @@ -71,20 +100,7 @@ SYM_CODE_START(ftrace_caller) mov x3, sp // regs #ifdef CONFIG_DYNAMIC_FTRACE_WITH_CALL_OPS - /* - * The literal pointer to the ops is at an 8-byte aligned boundary - * which is either 12 or 16 bytes before the BL instruction in the call - * site. See ftrace_call_adjust() for details. - * - * Therefore here the LR points at `literal + 16` or `literal + 20`, - * and we can find the address of the literal in either case by - * aligning to an 8-byte boundary and subtracting 16. We do the - * alignment first as this allows us to fold the subtraction into the - * LDR. - */ - bic x2, x30, 0x7 - ldr x2, [x2, #-16] // op - + mov x2, x11 // op ldr x4, [x2, #FTRACE_OPS_FUNC] // op->func blr x4 // op->func(ip, parent_ip, op, regs) @@ -110,12 +126,24 @@ SYM_INNER_LABEL(ftrace_call, SYM_L_GLOBAL) /* Restore the callsite's FP, LR, PC */ ldr x29, [sp, #FREGS_FP] ldr x30, [sp, #FREGS_LR] - ldr x9, [sp, #FREGS_PC] + ldr x10, [sp, #FREGS_PC] + +#ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS + ldr x11, [sp, #FREGS_CUSTOM_TRAMP] + cbz x11, 1f + /* Set x9 to parent ip before jump to custom trampoline */ + mov x9, x30 + /* Set lr to self ip */ + ldr x30, [sp, #FREGS_PC] + /* Set x10 (used for return address) to custom trampoline */ + mov x10, x11 +1: +#endif /* Restore the callsite's SP */ add sp, sp, #FREGS_SIZE + 32 - ret x9 + ret x10 SYM_CODE_END(ftrace_caller) #if defined(CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS) && \ diff --git a/arch/arm64/kernel/ftrace.c b/arch/arm64/kernel/ftrace.c index 5545fe1a9012..758436727fba 100644 --- a/arch/arm64/kernel/ftrace.c +++ b/arch/arm64/kernel/ftrace.c @@ -206,6 +206,13 @@ static struct plt_entry *get_ftrace_plt(struct module *mod, unsigned long addr) return NULL; } +static bool reachable_by_bl(unsigned long addr, unsigned long pc) +{ + long offset = (long)addr - (long)pc; + + return offset >= -SZ_128M && offset < SZ_128M; +} + /* * Find the address the callsite must branch to in order to reach '*addr'. * @@ -220,14 +227,21 @@ static bool ftrace_find_callable_addr(struct dyn_ftrace *rec, unsigned long *addr) { unsigned long pc = rec->ip; - long offset = (long)*addr - (long)pc; struct plt_entry *plt; + /* + * If a custom trampoline is unreachable, rely on the ftrace_caller + * trampoline which knows how to indirectly reach that trampoline + * through ops->direct_call. + */ + if (*addr != FTRACE_ADDR && !reachable_by_bl(*addr, pc)) + *addr = FTRACE_ADDR; + /* * When the target is within range of the 'BL' instruction, use 'addr' * as-is and branch to that directly. */ - if (offset >= -SZ_128M && offset < SZ_128M) + if (reachable_by_bl(*addr, pc)) return true; /* @@ -330,12 +344,24 @@ int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr, unsigned long addr) { - if (WARN_ON_ONCE(old_addr != (unsigned long)ftrace_caller)) + unsigned long pc = rec->ip; + u32 old, new; + int ret; + + ret = ftrace_rec_set_ops(rec, arm64_rec_get_ops(rec)); + if (ret) + return ret; + + if (!ftrace_find_callable_addr(rec, NULL, &old_addr)) return -EINVAL; - if (WARN_ON_ONCE(addr != (unsigned long)ftrace_caller)) + if (!ftrace_find_callable_addr(rec, NULL, &addr)) return -EINVAL; - return ftrace_rec_update_ops(rec); + old = aarch64_insn_gen_branch_imm(pc, old_addr, + AARCH64_INSN_BRANCH_LINK); + new = aarch64_insn_gen_branch_imm(pc, addr, AARCH64_INSN_BRANCH_LINK); + + return ftrace_modify_code(pc, old, new, true); } #endif