Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752628AbaG2IlO (ORCPT ); Tue, 29 Jul 2014 04:41:14 -0400 Received: from szxga03-in.huawei.com ([119.145.14.66]:64229 "EHLO szxga03-in.huawei.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751430AbaG2IlK (ORCPT ); Tue, 29 Jul 2014 04:41:10 -0400 Message-ID: <53D75E13.8000702@huawei.com> Date: Tue, 29 Jul 2014 16:40:51 +0800 From: Zhang Zhen User-Agent: Mozilla/5.0 (Windows NT 6.1; rv:16.0) Gecko/20121010 Thunderbird/16.0.1 MIME-Version: 1.0 To: David Rientjes CC: Dave Hansen , , Ingo Molnar , Andrew Morton , , Linux MM , Subject: Re: [PATCH v2] memory hotplug: update the variables after memory removed References: <1406619310-20555-1-git-send-email-zhenzhang.zhang@huawei.com> <53D74EE5.1070308@huawei.com> In-Reply-To: Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit X-Originating-IP: [10.111.69.77] X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020205.53D75E20.0022,ss=1,re=0.000,fgs=0, ip=0.0.0.0, so=2013-05-26 15:14:31, dmn=2011-05-27 18:58:46 X-Mirapoint-Loop-Id: 20d01a034e1369a02210e2d476771ae7 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 2014/7/29 15:53, David Rientjes wrote: > On Tue, 29 Jul 2014, Zhang Zhen wrote: > >> Commit ea0854170c95245a258b386c7a9314399c949fe0 added a fuction > > This would normally be written as > > Commit ea0854170c95 ("memory hotplug: fix a bug on /dev/mem for 64-bit > kernels") ... > ok. >> update_end_of_memory_vars() to update high_memory, max_pfn and >> max_low_pfn. >> >> I modified the function according to Dave Hansen and David Rientjes's >> suggestions. >> And call it in arch_remove_memory() to update these variables too. >> > > When the x86 maintainers merge this patch, they'll need to make a judgment > call on how urgent the fix is and that will guide them in whether they > want it backported to stable kernels as well. It would be useful to > provide the rationale for the change; in other words, why is the change > needed and what breaks if we don't have it? Got it, here we may access wrong memory via /dev/mem without this patch. I will provide the rationale for the change. > >> Change v1->v2: >> - according to Dave Hansen and David Rientjes's suggestions modified >> update_end_of_memory_vars(). >> Signed-off-by: Zhang Zhen > > Acked-by: David Rientjes > > You'll want to email all the x86 maintainers who would handle this patch, > check the output of scripts/get_maintainer.pl when run on this diff. Ok, Thanks! > >> --- >> arch/x86/mm/init_64.c | 23 ++++++++++++++--------- >> 1 file changed, 14 insertions(+), 9 deletions(-) >> >> diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c >> index df1a992..fd7bd6b 100644 >> --- a/arch/x86/mm/init_64.c >> +++ b/arch/x86/mm/init_64.c >> @@ -673,15 +673,11 @@ void __init paging_init(void) >> * After memory hotplug the variables max_pfn, max_low_pfn and high_memory need >> * updating. >> */ >> -static void update_end_of_memory_vars(u64 start, u64 size) >> +static void update_end_of_memory_vars(u64 end_pfn) > > Extra space that can be removed here at the same time as a cleanup. > Sorry, where is the extra space here? >> { >> - unsigned long end_pfn = PFN_UP(start + size); >> - >> - if (end_pfn > max_pfn) { >> - max_pfn = end_pfn; >> - max_low_pfn = end_pfn; >> - high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1; >> - } >> + max_pfn = end_pfn; >> + max_low_pfn = end_pfn; >> + high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1; >> } >> >> /* >> @@ -694,6 +690,7 @@ int arch_add_memory(int nid, u64 start, u64 size) >> struct zone *zone = pgdat->node_zones + ZONE_NORMAL; >> unsigned long start_pfn = start >> PAGE_SHIFT; >> unsigned long nr_pages = size >> PAGE_SHIFT; >> + unsigned long end_pfn; >> int ret; >> >> init_memory_mapping(start, start + size); >> @@ -702,7 +699,9 @@ int arch_add_memory(int nid, u64 start, u64 size) >> WARN_ON_ONCE(ret); >> >> /* update max_pfn, max_low_pfn and high_memory */ >> - update_end_of_memory_vars(start, size); >> + end_pfn = start_pfn + nr_pages; >> + if (end_pfn > max_pfn) >> + update_end_of_memory_vars(end_pfn); >> >> return ret; >> } >> @@ -1018,6 +1017,7 @@ int __ref arch_remove_memory(u64 start, u64 size) >> unsigned long start_pfn = start >> PAGE_SHIFT; >> unsigned long nr_pages = size >> PAGE_SHIFT; >> struct zone *zone; >> + unsigned long end_pfn; >> int ret; >> >> zone = page_zone(pfn_to_page(start_pfn)); >> @@ -1025,6 +1025,11 @@ int __ref arch_remove_memory(u64 start, u64 size) >> ret = __remove_pages(zone, start_pfn, nr_pages); >> WARN_ON_ONCE(ret); >> >> + /* update max_pfn, max_low_pfn and high_memory */ >> + end_pfn = start_pfn + nr_pages; >> + if ((max_pfn >= start_pfn) && (max_pfn < end_pfn)) >> + update_end_of_memory_vars(start_pfn); > > Not sure if we really need the new variable here; if you choose to > repropose this patch then you may want to consider just using > start_pfn + nr_pages in the conditional. yeah, i will remove it in v3. > >> + >> return ret; >> } >> #endif > > . > -- 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/