From patchwork Tue Apr 9 20:43:42 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Mathieu Desnoyers X-Patchwork-Id: 32231 Received: (qmail 36876 invoked by alias); 9 Apr 2019 20:43:47 -0000 Mailing-List: contact libc-alpha-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Unsubscribe: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: libc-alpha-owner@sourceware.org Delivered-To: mailing list libc-alpha@sourceware.org Received: (qmail 36863 invoked by uid 89); 9 Apr 2019 20:43:46 -0000 Authentication-Results: sourceware.org; auth=none X-Spam-SWARE-Status: No, score=-17.0 required=5.0 tests=AWL, BAYES_00, GIT_PATCH_0, GIT_PATCH_1, GIT_PATCH_2, GIT_PATCH_3, KAM_MANYTO, RCVD_IN_DNSWL_NONE, SPF_PASS autolearn=ham version=3.3.1 spammy= X-HELO: mail.efficios.com DKIM-Filter: OpenDKIM Filter v2.10.3 mail.efficios.com 166D97D6A8 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=efficios.com; s=default; t=1554842623; bh=rpljNT4W03Re1hwIRD5UA8RXQeJsMbdUOXrWL26XfWc=; h=Date:From:To:Message-ID:MIME-Version; b=GZIlWj+Fjejq5UsSqSSpfNqadXVBDPLnqEOHyxdlUpi2Ky8+2aUl5KhjPzAG15Xqr 3CRfhgFzRZ9gv/hYnCXF5s8CDJivGSMvY4hbMSMwV2OphyV+64QWjHJJMYkOM+9n6T mhtT54nU9uKAbAx9LJCeMsTZC/ceyjHrPC5rLwLMB+9A/NdjtJz9+j/YLX/0jtgAli NP5HeQf7CUVkTrVVur4lrLVmyVjAU4vXEpbaoxJ697TTqILKBtRu/uRVQAGX/6eSWQ 2q4Q77G4vC6O+3OvuTRGf7nsNiMLyC2pPFJKdXMdkVhsos2WSaY1cxRHiMnkcxe3iA U8zZShhOPQFeQ== Date: Tue, 9 Apr 2019 16:43:42 -0400 (EDT) From: Mathieu Desnoyers To: Thomas Gleixner , Andy Lutomirski , Peter Zijlstra , "H. Peter Anvin" , Andi Kleen , Ingo Molnar , Borislav Petkov Cc: libc-alpha , linux-kernel , Carlos O'Donell , x86 Message-ID: <913288111.2663.1554842622822.JavaMail.zimbra@efficios.com> In-Reply-To: <11513896.2624.1554838336494.JavaMail.zimbra@efficios.com> References: <11513896.2624.1554838336494.JavaMail.zimbra@efficios.com> Subject: Re: rseq/x86: choosing rseq code signature MIME-Version: 1.0 ----- On Apr 9, 2019, at 3:32 PM, Mathieu Desnoyers mathieu.desnoyers@efficios.com wrote: > Hi, > > We are about to include the code signature required prior to restartable > sequences abort handlers into glibc, which will make this ABI choice final. > We need architecture maintainer input on that signature value. > > That code signature is placed before each abort handler, so the kernel can > validate that it is indeed jumping to an abort handler (and not some > arbitrary attacker-chosen code). The signature is never executed. > > Currently, tools/testing/selftests/rseq/rseq-x86.h defines RSEQ_SIG > as 0x53053053, and uses it as an immediate operand to the following > instruction opcodes (as suggested by Andy Lutomirski): > > x86-32: > - .byte 0x0f, 0x1f, 0x05: nopl > > x86-64: > - .byte 0x0f, 0x1f, 0x05: nopl (%rip) > > The current discussion thread on the glibc mailing list leads us towards > using a trap with uncommon immediate operand, which simplifies integration > with disassemblers, emulators, makes it easier to debug if the control > flow gets redirected there by mistake, and is nicer for some architecture's > speculative execution. > > The main advantage of choosing a trap instruction over a no-op is to ensure the > program traps if the execution flow gets redirected to the signature by mistake > (makes it easier to debug). It's not a hard requirement, but it would be a > bonus. > > Are there trap instructions that take an uncommon 4-byte immediate > operand you would recommend on x86 32/64 ? Or is the current choice of > nopl confirmed to be right one ? > > Here is an example of rseq signature definition template: > > /* > * TODO: document trap instruction objdump output on each sub-architecture > * instruction sets, as well as instruction set extensions. > */ > #define RSEQ_SIG 0x######## Peter Zijlstra suggested to use "invlpg" in user-space, which should generate a trap. The only concern would be emulators, but ideally they would not try to decode an instruction that is never executed. This would lead to the following patch. Any objections/ack ? diff --git a/tools/testing/selftests/rseq/rseq-x86.h b/tools/testing/selftests/rseq/rseq-x86.h index 2d4887b5d3f0..e9c8a9879e18 100644 --- a/tools/testing/selftests/rseq/rseq-x86.h +++ b/tools/testing/selftests/rseq/rseq-x86.h @@ -7,6 +7,11 @@ #include +/* + * RSEQ_SIG is used with the following privileged instructions, which trap in user-space: + * x86-32: 0f 01 3d 53 30 05 53 invlpg 0x53053053 + * x86-64: 0f 01 3d 53 30 05 53 invlpg 0x53053053(%rip) + */ #define RSEQ_SIG 0x53053053 #ifdef __x86_64__ @@ -78,8 +83,8 @@ do { \ #define RSEQ_ASM_DEFINE_ABORT(label, teardown, abort_label) \ ".pushsection __rseq_failure, \"ax\"\n\t" \ - /* Disassembler-friendly signature: nopl (%rip). */\ - ".byte 0x0f, 0x1f, 0x05\n\t" \ + /* Disassembler-friendly signature: invlpg (%rip). */\ + ".byte 0x0f, 0x01, 0x3d\n\t" \ ".long " __rseq_str(RSEQ_SIG) "\n\t" \ __rseq_str(label) ":\n\t" \ teardown \ @@ -605,8 +610,8 @@ do { \ #define RSEQ_ASM_DEFINE_ABORT(label, teardown, abort_label) \ ".pushsection __rseq_failure, \"ax\"\n\t" \ - /* Disassembler-friendly signature: nopl . */ \ - ".byte 0x0f, 0x1f, 0x05\n\t" \ + /* Disassembler-friendly signature: invlpg . */ \ + ".byte 0x0f, 0x1f, 0x3d\n\t" \ ".long " __rseq_str(RSEQ_SIG) "\n\t" \ __rseq_str(label) ":\n\t" \ teardown \