Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932935Ab2K2AyK (ORCPT ); Wed, 28 Nov 2012 19:54:10 -0500 Received: from e9.ny.us.ibm.com ([32.97.182.139]:48854 "EHLO e9.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932414Ab2K2AyJ (ORCPT ); Wed, 28 Nov 2012 19:54:09 -0500 Subject: [RFC][PATCH 1/2] create slow_virt_to_phys() To: linux-kernel@vger.kernel.org Cc: linux-mm@kvack.org, Gleb Natapov , Avi Kivity , Dave Hansen From: Dave Hansen Date: Thu, 29 Nov 2012 00:53:55 +0000 Message-Id: <20121129005355.966BD487@kernel.stglabs.ibm.com> X-Content-Scanned: Fidelis XPS MAILER x-cbid: 12112900-7182-0000-0000-00000365D89D Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3767 Lines: 102 This is necessary because __pa() does not work on some kinds of memory, like vmalloc() or the alloc_remap() areas on 32-bit NUMA systems. We have some functions to do conversions _like_ this in the vmalloc() code (like vmalloc_to_page()), but they do not work on sizes other than 4k pages. We would potentially need to be able to handle all the page sizes that we use for the kernel linear mapping (4k, 2M, 1G). In practice, on 32-bit NUMA systems, the percpu areas get stuck in the alloc_remap() area. Any __pa() call on them will break and basically return garbage. This patch introduces a new function slow_virt_to_phys(), which walks the kernel page tables on x86 and should do precisely the same logical thing as __pa(), but actually work on a wider range of memory. It should work on the normal linear mapping, vmalloc(), kmap(), etc... Signed-off-by: Dave Hansen --- linux-2.6.git-dave/arch/x86/include/asm/pgtable_types.h | 1 linux-2.6.git-dave/arch/x86/mm/pageattr.c | 47 ++++++++++++++++ 2 files changed, 48 insertions(+) diff -puN arch/x86/include/asm/pgtable_types.h~create-slow_virt_to_phys arch/x86/include/asm/pgtable_types.h --- linux-2.6.git/arch/x86/include/asm/pgtable_types.h~create-slow_virt_to_phys 2012-11-29 00:26:58.583830776 +0000 +++ linux-2.6.git-dave/arch/x86/include/asm/pgtable_types.h 2012-11-29 00:26:58.595830876 +0000 @@ -332,6 +332,7 @@ static inline void update_page_count(int * as a pte too. */ extern pte_t *lookup_address(unsigned long address, unsigned int *level); +extern phys_addr_t slow_virt_to_phys(void *__address); #endif /* !__ASSEMBLY__ */ diff -puN arch/x86/mm/pageattr.c~create-slow_virt_to_phys arch/x86/mm/pageattr.c --- linux-2.6.git/arch/x86/mm/pageattr.c~create-slow_virt_to_phys 2012-11-29 00:26:58.591830844 +0000 +++ linux-2.6.git-dave/arch/x86/mm/pageattr.c 2012-11-29 00:26:58.595830876 +0000 @@ -364,6 +364,53 @@ pte_t *lookup_address(unsigned long addr EXPORT_SYMBOL_GPL(lookup_address); /* + * This is necessary because __pa() does not work on some + * kinds of memory, like vmalloc() or the alloc_remap() + * areas on 32-bit NUMA systems. The percpu areas can + * end up in this kind of memory, for instance. + * + * This could be optimized, but it is only intended to be + * used at inititalization time, and keeping it + * unoptimized should increase the testing coverage for + * the more obscure platforms. + */ +phys_addr_t slow_virt_to_phys(void *__virt_addr) +{ + unsigned long virt_addr = (unsigned long)__virt_addr; + phys_addr_t phys_addr; + unsigned long offset; + unsigned int level = -1; + unsigned long psize = 0; + unsigned long pmask = 0; + pte_t *pte; + + pte = lookup_address(virt_addr, &level); + BUG_ON(!pte); + switch (level) { + case PG_LEVEL_4K: + psize = PAGE_SIZE; + pmask = PAGE_MASK; + break; + case PG_LEVEL_2M: + psize = PMD_PAGE_SIZE; + pmask = PMD_PAGE_MASK; + break; +#ifdef CONFIG_X86_64 + case PG_LEVEL_1G: + psize = PUD_PAGE_SIZE; + pmask = PUD_PAGE_MASK; + break; +#endif + default: + BUG(); + } + offset = virt_addr & ~pmask; + phys_addr = pte_pfn(*pte) << PAGE_SHIFT; + return (phys_addr | offset); +} +EXPORT_SYMBOL_GPL(slow_virt_to_phys); + +/* * Set the new pmd in all the pgds we know about: */ static void __set_pmd_pte(pte_t *kpte, unsigned long address, pte_t pte) _ -- 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/