Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753467AbbH3Mua (ORCPT ); Sun, 30 Aug 2015 08:50:30 -0400 Received: from e28smtp08.in.ibm.com ([122.248.162.8]:56068 "EHLO e28smtp08.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753358AbbH3Mu2 (ORCPT ); Sun, 30 Aug 2015 08:50:28 -0400 X-Helo: d28dlp02.in.ibm.com X-MailFrom: aneesh.kumar@linux.vnet.ibm.com X-RcptTo: linux-kernel@vger.kernel.org From: "Aneesh Kumar K.V" To: Andrey Ryabinin Cc: Benjamin Herrenschmidt , paulus@samba.org, mpe@ellerman.id.au, linuxppc-dev@lists.ozlabs.org, LKML Subject: Re: [PATCH V2 07/10] kasan: Make INLINE KASan support arch selectable In-Reply-To: References: <1440577578-15813-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com> <1440577578-15813-8-git-send-email-aneesh.kumar@linux.vnet.ibm.com> User-Agent: Notmuch/0.17 (http://notmuchmail.org) Emacs/24.5.1 (x86_64-pc-linux-gnu) Date: Sun, 30 Aug 2015 18:20:21 +0530 Message-ID: <87613xq6sy.fsf@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain X-TM-AS-MML: disable X-Content-Scanned: Fidelis XPS MAILER x-cbid: 15083012-0029-0000-0000-000007C602BB Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5279 Lines: 156 Andrey Ryabinin writes: > 2015-08-26 11:26 GMT+03:00 Aneesh Kumar K.V : >> Some of the archs, may find it difficult to support inline KASan >> mode. Add HAVE_ARCH_KASAN_INLINE so that we can disable inline >> support at config time. >> >> Signed-off-by: Aneesh Kumar K.V >> --- >> arch/x86/Kconfig | 1 + >> lib/Kconfig.kasan | 2 ++ >> scripts/Makefile.kasan | 28 ++++++++++++++-------------- >> 3 files changed, 17 insertions(+), 14 deletions(-) >> >> diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig >> index b3a1a5d77d92..4416f80580fb 100644 >> --- a/arch/x86/Kconfig >> +++ b/arch/x86/Kconfig >> @@ -78,6 +78,7 @@ config X86 >> select HAVE_ARCH_HUGE_VMAP if X86_64 || X86_PAE >> select HAVE_ARCH_JUMP_LABEL >> select HAVE_ARCH_KASAN if X86_64 && SPARSEMEM_VMEMMAP >> + select HAVE_ARCH_KASAN_INLINE if X86_64 && SPARSEMEM_VMEMMAP > > This will not work because config HAVE_ARCH_KASAN_INLINE is not defined. > Instead of you can just add following in this file: > > config HAVE_ARCH_KASAN_INLINE > def_bool y > depends on KASAN > Missed cherry-pick diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan index e9d1bb1175b8..5dba03bc3f01 100644 --- a/lib/Kconfig.kasan +++ b/lib/Kconfig.kasan @@ -1,6 +1,9 @@ config HAVE_ARCH_KASAN bool +config HAVE_ARCH_KASAN_INLINE + bool + if HAVE_ARCH_KASAN config KASAN > >> select HAVE_ARCH_KGDB >> select HAVE_ARCH_KMEMCHECK >> select HAVE_ARCH_SECCOMP_FILTER >> diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan >> index 39f24d6721e5..e9d1bb1175b8 100644 >> --- a/lib/Kconfig.kasan >> +++ b/lib/Kconfig.kasan >> @@ -32,6 +32,7 @@ config KASAN_OUTLINE >> however it doesn't bloat size of kernel's .text section so >> much as inline does. >> >> +if HAVE_ARCH_KASAN_INLINE >> config KASAN_INLINE >> bool "Inline instrumentation" > > depends on HAVE_ARCH_KASAN_INLINE > >> help >> @@ -40,6 +41,7 @@ config KASAN_INLINE >> it gives about x2 boost over outline instrumentation), but >> make kernel's .text size much bigger. >> This requires a gcc version of 5.0 or later. >> +endif >> >> endchoice >> >> diff --git a/scripts/Makefile.kasan b/scripts/Makefile.kasan >> index 3f874d24234f..c1c06e9e107a 100644 >> --- a/scripts/Makefile.kasan >> +++ b/scripts/Makefile.kasan >> @@ -1,29 +1,29 @@ >> ifdef CONFIG_KASAN >> -ifdef CONFIG_KASAN_INLINE >> - call_threshold := 10000 >> -else >> - call_threshold := 0 >> -endif >> - >> -CFLAGS_KASAN_MINIMAL := -fsanitize=kernel-address >> >> CFLAGS_KASAN := $(call cc-option, -fsanitize=kernel-address \ >> - -fasan-shadow-offset=$(CONFIG_KASAN_SHADOW_OFFSET) \ >> - --param asan-stack=1 --param asan-globals=1 \ >> - --param asan-instrumentation-with-call-threshold=$(call_threshold)) >> - >> -ifeq ($(call cc-option, $(CFLAGS_KASAN_MINIMAL) -Werror),) >> + --param asan-instrumentation-with-call-threshold=0) >> +ifeq ($(CFLAGS_KASAN),) >> ifneq ($(CONFIG_COMPILE_TEST),y) >> $(warning Cannot use CONFIG_KASAN: \ >> -fsanitize=kernel-address is not supported by compiler) >> endif >> else >> - ifeq ($(CFLAGS_KASAN),) >> + >> + ifdef CONFIG_KASAN_INLINE >> + CFLAGS_KASAN_INLINE := $(call cc-option, -fsanitize=kernel-address \ >> + -fasan-shadow-offset=$(CONFIG_KASAN_SHADOW_OFFSET) \ >> + --param asan-stack=1 --param asan-globals=1 \ >> + --param asan-instrumentation-with-call-threshold=10000) >> + >> + ifeq ($(CFLAGS_KASAN_INLINE),) >> ifneq ($(CONFIG_COMPILE_TEST),y) >> $(warning CONFIG_KASAN: compiler does not support all options.\ >> Trying minimal configuration) >> endif >> - CFLAGS_KASAN := $(CFLAGS_KASAN_MINIMAL) >> + else >> + CFLAGS_KASAN := $(CFLAGS_KASAN_INLINE) >> endif >> + endif >> + > > This removes stack and globals for CONFIG_KASAN_OUTLINE=y. Why? > Those are completely separate features. So this patch shouldn't touch > this Makefile at all. > Depends on HAVE_ARCH_KASAN_INLINE in CONFIG_KASAN_INLINE should be enough. Correct. I have changes that will make kasan_global and kasan_stack as config items. I didn't add it as part of this series because I still haven't got outline global and stack work on ppc64. Right now the split config only applies to ppc64. Hence didn't want to make the patch more complicated. > > But you need to disable 'asan-stack' and 'asan-globals' for pcc64. > I'd suggest to introduce CFLAGS_ARCH_KASAN. > Define it in ppc64 Makefile: > > CFLAGS_ARCH_KASAN := --param asan-globals=0 --param asan-stack=0 > > and add these flags to CFLAGS_KASAN_MINIMAL and CFLAGS_KASAN in Makefile.kasan. > > >> endif >> endif >> -aneesh -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/