From nobody Sun Sep 14 11:08:52 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1CDE2C05027 for ; Mon, 23 Jan 2023 21:07:18 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232963AbjAWVHR (ORCPT ); Mon, 23 Jan 2023 16:07:17 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41258 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232599AbjAWVHO (ORCPT ); Mon, 23 Jan 2023 16:07:14 -0500 Received: from desiato.infradead.org (desiato.infradead.org [IPv6:2001:8b0:10b:1:d65d:64ff:fe57:4e05]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 060A938667 for ; Mon, 23 Jan 2023 13:07:12 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=infradead.org; s=desiato.20200630; h=Content-Type:MIME-Version:References: Subject:Cc:To:From:Date:Message-ID:Sender:Reply-To:Content-Transfer-Encoding: Content-ID:Content-Description:In-Reply-To; bh=KKXDnmc2Es+6ZmW+DV2eYtRfLFf2Gu02YvShw/3lB4Y=; b=PT4NpcyeoJrAlUFkM+JxIKvOzE wat/l+Lls2OmdG/RSotNTpPyBaTs30NSFT3/OKTFTFQwWiMnNzcxsP2g8kVvUaBDZZQFUfFcby1DU HDyhwUB2Huqfc6yTcFDqh4mpzpV21524uoQ4mVDXvKYdQWV7zBd5uyoOACAuO5EAjrg8OtgbUIebV 9vA8P000to1tKOp8gWyG8+fQswPZhYiJjr0CFD7hOWlXGxFJM6HNCCAFcJXEg7zHUzErZ7FWw/Kci 1ZVm89rNWAv/Od7R1wtLVfUR3Nj+5Of6XgWxgfOt+XO4/L/HMrI7DnNX9TwiNBm7ftLk/TFJsCDPn cYfx0lKg==; Received: from j130084.upc-j.chello.nl ([24.132.130.84] helo=noisy.programming.kicks-ass.net) by desiato.infradead.org with esmtpsa (Exim 4.96 #2 (Red Hat Linux)) id 1pK41C-001e3t-05; Mon, 23 Jan 2023 21:06:19 +0000 Received: from hirez.programming.kicks-ass.net (hirez.programming.kicks-ass.net [192.168.1.225]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits)) (Client did not present a certificate) by noisy.programming.kicks-ass.net (Postfix) with ESMTPS id 5D84B30008D; Mon, 23 Jan 2023 22:06:45 +0100 (CET) Received: by hirez.programming.kicks-ass.net (Postfix, from userid 0) id 300842038B0DB; Mon, 23 Jan 2023 22:06:45 +0100 (CET) Message-ID: <20230123210607.173715335@infradead.org> User-Agent: quilt/0.66 Date: Mon, 23 Jan 2023 21:59:18 +0100 From: Peter Zijlstra To: mingo@kernel.org Cc: tglx@linutronix.de, bp@alien8.de, dave.hansen@linux.intel.com, x86@kernel.org, hpa@zytor.com, peterz@infradead.org, jpoimboe@kernel.org, jbaron@akamai.com, rostedt@goodmis.org, ardb@kernel.org, linux-kernel@vger.kernel.org, erhard_f@mailbox.org, ndesaulniers@google.com, mhiramat@kernel.org, sandipan.das@amd.com Subject: [PATCH 3/3] x86/static_call: Add support for Jcc tail-calls References: <20230123205915.751729592@infradead.org> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Clang likes to create conditional tail calls like: 0000000000000350 : 350: 0f 1f 44 00 00 nopl 0x0(%rax,%rax,1) 351: R_X86_64_NO= NE __fentry__-0x4 355: 48 83 bf 20 01 00 00 00 cmpq $0x0,0x120(%rdi) 35d: 0f 85 00 00 00 00 jne 363 = 35f: R_X86_64_PLT32 __SCT__amd_pmu_branch_add-0x4 363: e9 00 00 00 00 jmp 368 = 364: R_X86_64_PLT32 __x86_return_thunk-0x4 Teach the in-line static call text patching about this. Notably, since there is no conditional-ret, in that caes patch the Jcc to point at an empty stub function that does the ret -- or the return thunk when needed. Reported-by: "Erhard F." Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Masami Hiramatsu (Google) --- arch/x86/kernel/static_call.c | 50 +++++++++++++++++++++++++++++++++++++= ++--- 1 file changed, 47 insertions(+), 3 deletions(-) --- a/arch/x86/kernel/static_call.c +++ b/arch/x86/kernel/static_call.c @@ -9,6 +9,7 @@ enum insn_type { NOP =3D 1, /* site cond-call */ JMP =3D 2, /* tramp / site tail-call */ RET =3D 3, /* tramp / site cond-tail-call */ + JCC =3D 4, }; =20 /* @@ -25,12 +26,40 @@ static const u8 xor5rax[] =3D { 0x2e, 0x2e =20 static const u8 retinsn[] =3D { RET_INSN_OPCODE, 0xcc, 0xcc, 0xcc, 0xcc }; =20 +static u8 __is_Jcc(u8 *insn) /* Jcc.d32 */ +{ + u8 ret =3D 0; + + if (insn[0] =3D=3D 0x0f) { + u8 tmp =3D insn[1]; + if ((tmp & 0xf0) =3D=3D 0x80) + ret =3D tmp; + } + + return ret; +} + +extern void __static_call_return(void); + +asm (".global __static_call_return\n\t" + ".type __static_call_return, @function\n\t" + ASM_FUNC_ALIGN "\n\t" + "__static_call_return:\n\t" + ANNOTATE_NOENDBR + ANNOTATE_RETPOLINE_SAFE + "ret; int3\n\t" + ".size __static_call_return, . - __static_call_return \n\t"); + static void __ref __static_call_transform(void *insn, enum insn_type type, void *func, bool modinit) { const void *emulate =3D NULL; int size =3D CALL_INSN_SIZE; const void *code; + u8 op, buf[6]; + + if ((type =3D=3D JMP || type =3D=3D RET) && (op =3D __is_Jcc(insn))) + type =3D JCC; =20 switch (type) { case CALL: @@ -57,6 +86,20 @@ static void __ref __static_call_transfor else code =3D &retinsn; break; + + case JCC: + if (!func) { + func =3D __static_call_return; + if (cpu_feature_enabled(X86_FEATURE_RETHUNK)) + func =3D x86_return_thunk; + } + + buf[0] =3D 0x0f; + __text_gen_insn(buf+1, op, insn+1, func, 5); + code =3D buf; + size =3D 6; + + break; } =20 if (memcmp(insn, code, size) =3D=3D 0) @@ -68,9 +111,9 @@ static void __ref __static_call_transfor text_poke_bp(insn, code, size, emulate); } =20 -static void __static_call_validate(void *insn, bool tail, bool tramp) +static void __static_call_validate(u8 *insn, bool tail, bool tramp) { - u8 opcode =3D *(u8 *)insn; + u8 opcode =3D insn[0]; =20 if (tramp && memcmp(insn+5, tramp_ud, 3)) { pr_err("trampoline signature fail"); @@ -79,7 +122,8 @@ static void __static_call_validate(void =20 if (tail) { if (opcode =3D=3D JMP32_INSN_OPCODE || - opcode =3D=3D RET_INSN_OPCODE) + opcode =3D=3D RET_INSN_OPCODE || + __is_Jcc(insn)) return; } else { if (opcode =3D=3D CALL_INSN_OPCODE ||