Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754145AbcLIPcj (ORCPT ); Fri, 9 Dec 2016 10:32:39 -0500 Received: from mail-wj0-f195.google.com ([209.85.210.195]:36042 "EHLO mail-wj0-f195.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753889AbcLIPch (ORCPT ); Fri, 9 Dec 2016 10:32:37 -0500 Date: Fri, 9 Dec 2016 18:32:33 +0300 From: "Kirill A. Shutemov" To: Borislav Petkov Cc: "Kirill A. Shutemov" , Linus Torvalds , Andrew Morton , x86@kernel.org, Thomas Gleixner , Ingo Molnar , Arnd Bergmann , "H. Peter Anvin" , Andi Kleen , Dave Hansen , Andy Lutomirski , linux-arch@vger.kernel.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: Re: [RFC, PATCHv1 15/28] x86: detect 5-level paging support Message-ID: <20161209153233.GA8932@node.shutemov.name> References: <20161208162150.148763-1-kirill.shutemov@linux.intel.com> <20161208162150.148763-17-kirill.shutemov@linux.intel.com> <20161208200505.c6xiy56oufg6d24m@pd.tnic> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20161208200505.c6xiy56oufg6d24m@pd.tnic> User-Agent: Mutt/1.5.23.1 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2398 Lines: 69 On Thu, Dec 08, 2016 at 09:05:05PM +0100, Borislav Petkov wrote: > On Thu, Dec 08, 2016 at 07:21:37PM +0300, Kirill A. Shutemov wrote: > > 5-level paging support is required from hardware when compiled with > > CONFIG_X86_5LEVEL=y. We may implement runtime switch support later. > > > > Signed-off-by: Kirill A. Shutemov > > ... > > > diff --git a/arch/x86/boot/cpuflags.c b/arch/x86/boot/cpuflags.c > > index 6687ab953257..26e9a287805f 100644 > > --- a/arch/x86/boot/cpuflags.c > > +++ b/arch/x86/boot/cpuflags.c > > @@ -80,6 +80,17 @@ static inline void cpuid(u32 id, u32 *a, u32 *b, u32 *c, u32 *d) > > ); > > } > > > > +static inline void cpuid_count(u32 id, u32 count, > > + u32 *a, u32 *b, u32 *c, u32 *d) > > +{ > > + asm volatile(".ifnc %%ebx,%3 ; movl %%ebx,%3 ; .endif \n\t" > > + "cpuid \n\t" > > + ".ifnc %%ebx,%3 ; xchgl %%ebx,%3 ; .endif \n\t" > > + : "=a" (*a), "=c" (*c), "=d" (*d), EBX_REG (*b) > > + : "a" (id), "c" (count) > > + ); > > +} > > Pls make those like cpuid() and cpuid_count() in > arch/x86/include/asm/processor.h, which explicitly assign ecx and then > call the underlying helper. > > The cpuid() in cpuflags.c doesn't zero ecx which, if we have to be > pedantic, it should do. It calls CPUID now with the ptr value of its 4th > on 64-bit and 3rd arg on 32-bit, respectively, IINM. Something like this? diff --git a/arch/x86/boot/cpuflags.c b/arch/x86/boot/cpuflags.c index 6687ab953257..366aad972025 100644 --- a/arch/x86/boot/cpuflags.c +++ b/arch/x86/boot/cpuflags.c @@ -70,16 +70,22 @@ int has_eflag(unsigned long mask) # define EBX_REG "=b" #endif -static inline void cpuid(u32 id, u32 *a, u32 *b, u32 *c, u32 *d) +static inline void cpuid_count(u32 id, u32 count, + u32 *a, u32 *b, u32 *c, u32 *d) { + *a = id; + *c = count; + asm volatile(".ifnc %%ebx,%3 ; movl %%ebx,%3 ; .endif \n\t" "cpuid \n\t" ".ifnc %%ebx,%3 ; xchgl %%ebx,%3 ; .endif \n\t" : "=a" (*a), "=c" (*c), "=d" (*d), EBX_REG (*b) - : "a" (id) + : "a" (id), "c" (count) ); } +#define cpuid(id, a, b, c, d) cpuid_count(id, 0, a, b, c, d) + void get_cpuflags(void) { u32 max_intel_level, max_amd_level; -- Kirill A. Shutemov