lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
Open Source and information security mailing list archives
| ||
|
Date: Fri, 18 Nov 2016 16:46:02 +0000 From: Mark Rutland <mark.rutland@....com> To: Laura Abbott <labbott@...hat.com>, lorenzo.pieralisi@....com Cc: Ard Biesheuvel <ard.biesheuvel@...aro.org>, Catalin Marinas <catalin.marinas@....com>, x86@...nel.org, Will Deacon <will.deacon@....com>, linux-kernel@...r.kernel.org, linux-mm@...ck.org, Ingo Molnar <mingo@...hat.com>, "H. Peter Anvin" <hpa@...or.com>, Joonsoo Kim <iamjoonsoo.kim@....com>, Thomas Gleixner <tglx@...utronix.de>, Andrew Morton <akpm@...ux-foundation.org>, linux-arm-kernel@...ts.infradead.org, Marek Szyprowski <m.szyprowski@...sung.com> Subject: Re: [PATCHv3 5/6] arm64: Use __pa_symbol for kernel symbols On Fri, Nov 18, 2016 at 02:35:44PM +0000, Mark Rutland wrote: > Hi Laura, > > On Thu, Nov 17, 2016 at 05:16:55PM -0800, Laura Abbott wrote: > > > > __pa_symbol is technically the marco that should be used for kernel > > symbols. Switch to this as a pre-requisite for DEBUG_VIRTUAL which > > will do bounds checking. > > > > Signed-off-by: Laura Abbott <labbott@...hat.com> > > --- > > v3: Conversion of more sites besides just _end. Addition of __lm_sym_addr > > macro to take care of the _va(__pa_symbol(..)) idiom. > > > > Note that a copy of __pa_symbol was added to avoid a mess of headers > > since the #ifndef __pa_symbol case is defined in linux/mm.h > > I think we also need to fix up virt_to_phys(__cpu_soft_restart) in > arch/arm64/kernel/cpu-reset.h. Otherwise, this looks complete for uses > falling under arch/arm64/. I think I spoke too soon. :( In the kasan code, use of tmp_pg_dir, kasan_zero_{page,pte,pmd,pud} all need to be vetted, as those are in the image, but get passed directly to functions which will end up doing a virt_to_phys behind the scenes (e.g. cpu_replace_ttbr1(), pmd_populate_kernel()). There's also some virt_to_pfn(<symbol>) usage that needs to be fixed up in arch/arm64/kernel/hibernate.c. ... there's also more of that in common kernel code. :( Thanks, Mark.
Powered by blists - more mailing lists