From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758602Ab3BSKHv (ORCPT ); Tue, 19 Feb 2013 05:07:51 -0500 Received: from mail-gh0-f173.google.com ([209.85.160.173]:51512 "EHLO mail-gh0-f173.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755812Ab3BSKHt (ORCPT ); Tue, 19 Feb 2013 05:07:49 -0500 Message-ID: <51234EEC.3010700@gmail.com> Date: Tue, 19 Feb 2013 18:07:40 +0800 From: Simon Jeons User-Agent: Mozilla/5.0 (X11; Linux i686; rv:17.0) Gecko/20130106 Thunderbird/17.0.2 MIME-Version: 1.0 To: Minchan Kim CC: Greg Kroah-Hartman , Matt Sealey , linux-mm@kvack.org, linux-kernel@vger.kernel.org, stable@vger.kernel.org, Dan Magenheimer , Russell King , Konrad Rzeszutek Wilk , Nitin Gupta , Seth Jennings Subject: Re: [PATCH] zsmalloc: Fix TLB coherency and build problem References: <1359334808-19794-1-git-send-email-minchan@kernel.org> In-Reply-To: <1359334808-19794-1-git-send-email-minchan@kernel.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 01/28/2013 09:00 AM, Minchan Kim wrote: > Recently, Matt Sealey reported he fail to build zsmalloc caused by > using of local_flush_tlb_kernel_range which are architecture dependent > function so !CONFIG_SMP in ARM couldn't implement it so it ends up > build error following as. Confuse me! 1) Why I see flush_tlb_kernel_range is different in different architecture? 2) Does local here means local cpu? If the answer is yes, why ARM doesn't support it? > > MODPOST 216 modules > LZMA arch/arm/boot/compressed/piggy.lzma > AS arch/arm/boot/compressed/lib1funcs.o > ERROR: "v7wbi_flush_kern_tlb_range" > [drivers/staging/zsmalloc/zsmalloc.ko] undefined! > make[1]: *** [__modpost] Error 1 > make: *** [modules] Error 2 > make: *** Waiting for unfinished jobs.... > > The reason we used that function is copy method by [1] > was really slow in ARM but at that time. > > More severe problem is ARM can prefetch speculatively on other CPUs > so under us, other TLBs can have an entry only if we do flush local > CPU. Russell King pointed that. Thanks! > We don't have many choices except using flush_tlb_kernel_range. > > My experiment in ARMv7 processor 4 core didn't make any difference with > zsmapbench[2] between local_flush_tlb_kernel_range and flush_tlb_kernel_range > but still page-table based is much better than copy-based. > > * bigger is better. > > 1. local_flush_tlb_kernel_range: 3918795 mappings > 2. flush_tlb_kernel_range : 3989538 mappings > 3. copy-based: 635158 mappings > > This patch replace local_flush_tlb_kernel_range with > flush_tlb_kernel_range which are avaialbe in all architectures > because we already have used it in vmalloc allocator which are > generic one so build problem should go away and performane loss > shoud be void. > > [1] f553646, zsmalloc: add page table mapping method > [2] https://github.com/spartacus06/zsmapbench > > Cc: stable@vger.kernel.org > Cc: Dan Magenheimer > Cc: Russell King > Cc: Konrad Rzeszutek Wilk > Cc: Nitin Gupta > Cc: Seth Jennings > Reported-by: Matt Sealey > Signed-off-by: Minchan Kim > --- > > Matt, Could you test this patch? > > drivers/staging/zsmalloc/zsmalloc-main.c | 10 ++++------ > 1 file changed, 4 insertions(+), 6 deletions(-) > > diff --git a/drivers/staging/zsmalloc/zsmalloc-main.c b/drivers/staging/zsmalloc/zsmalloc-main.c > index eb00772..82e627c 100644 > --- a/drivers/staging/zsmalloc/zsmalloc-main.c > +++ b/drivers/staging/zsmalloc/zsmalloc-main.c > @@ -222,11 +222,9 @@ struct zs_pool { > /* > * By default, zsmalloc uses a copy-based object mapping method to access > * allocations that span two pages. However, if a particular architecture > - * 1) Implements local_flush_tlb_kernel_range() and 2) Performs VM mapping > - * faster than copying, then it should be added here so that > - * USE_PGTABLE_MAPPING is defined. This causes zsmalloc to use page table > - * mapping rather than copying > - * for object mapping. > + * performs VM mapping faster than copying, then it should be added here > + * so that USE_PGTABLE_MAPPING is defined. This causes zsmalloc to use > + * page table mapping rather than copying for object mapping. > */ > #if defined(CONFIG_ARM) > #define USE_PGTABLE_MAPPING > @@ -663,7 +661,7 @@ static inline void __zs_unmap_object(struct mapping_area *area, > > flush_cache_vunmap(addr, end); > unmap_kernel_range_noflush(addr, PAGE_SIZE * 2); > - local_flush_tlb_kernel_range(addr, end); > + flush_tlb_kernel_range(addr, end); > } > > #else /* USE_PGTABLE_MAPPING */