From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-it0-f72.google.com (mail-it0-f72.google.com [209.85.214.72]) by kanga.kvack.org (Postfix) with ESMTP id 3B6EE6B0253 for ; Wed, 29 Jun 2016 03:57:42 -0400 (EDT) Received: by mail-it0-f72.google.com with SMTP id g8so92829672itb.2 for ; Wed, 29 Jun 2016 00:57:42 -0700 (PDT) Received: from lgeamrelo12.lge.com (LGEAMRELO12.lge.com. [156.147.23.52]) by mx.google.com with ESMTP id n64si258916itb.52.2016.06.29.00.57.40 for ; Wed, 29 Jun 2016 00:57:41 -0700 (PDT) Date: Wed, 29 Jun 2016 17:00:38 +0900 From: Joonsoo Kim Subject: Re: [PATCH v3 3/6] mm/cma: populate ZONE_CMA Message-ID: <20160629080038.GB25102@js1304-P5Q-DELUXE> References: <1464243748-16367-1-git-send-email-iamjoonsoo.kim@lge.com> <1464243748-16367-4-git-send-email-iamjoonsoo.kim@lge.com> <576A58FA.8040101@hisilicon.com> <20160623025238.GB30438@js1304-P5Q-DELUXE> <57725E2B.3000201@hisilicon.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <57725E2B.3000201@hisilicon.com> Sender: owner-linux-mm@kvack.org List-ID: To: Chen Feng Cc: Andrew Morton , Rik van Riel , Johannes Weiner , mgorman@techsingularity.net, Laura Abbott , Minchan Kim , Marek Szyprowski , Michal Nazarewicz , "Aneesh Kumar K.V" , Vlastimil Babka , Rui Teng , linux-mm@kvack.org, linux-kernel@vger.kernel.org, "fujun (F)" , Zhuangluan Su , Yiping Xu , Dan Zhao On Tue, Jun 28, 2016 at 07:23:23PM +0800, Chen Feng wrote: > Hello, > > On 2016/6/23 10:52, Joonsoo Kim wrote: > > On Wed, Jun 22, 2016 at 05:23:06PM +0800, Chen Feng wrote: > >> Hello, > >> > >> On 2016/5/26 14:22, js1304@gmail.com wrote: > >>> From: Joonsoo Kim > >>> > >>> Until now, reserved pages for CMA are managed in the ordinary zones > >>> where page's pfn are belong to. This approach has numorous problems > >>> and fixing them isn't easy. (It is mentioned on previous patch.) > >>> To fix this situation, ZONE_CMA is introduced in previous patch, but, > >>> not yet populated. This patch implement population of ZONE_CMA > >>> by stealing reserved pages from the ordinary zones. > >>> > >>> Unlike previous implementation that kernel allocation request with > >>> __GFP_MOVABLE could be serviced from CMA region, allocation request only > >>> with GFP_HIGHUSER_MOVABLE can be serviced from CMA region in the new > >>> approach. This is an inevitable design decision to use the zone > >>> implementation because ZONE_CMA could contain highmem. Due to this > >>> decision, ZONE_CMA will work like as ZONE_HIGHMEM or ZONE_MOVABLE. > >>> > >>> I don't think it would be a problem because most of file cache pages > >>> and anonymous pages are requested with GFP_HIGHUSER_MOVABLE. It could > >>> be proved by the fact that there are many systems with ZONE_HIGHMEM and > >>> they work fine. Notable disadvantage is that we cannot use these pages > >>> for blockdev file cache page, because it usually has __GFP_MOVABLE but > >>> not __GFP_HIGHMEM and __GFP_USER. But, in this case, there is pros and > >>> cons. In my experience, blockdev file cache pages are one of the top > >>> reason that causes cma_alloc() to fail temporarily. So, we can get more > >>> guarantee of cma_alloc() success by discarding that case. > >>> > >>> Implementation itself is very easy to understand. Steal when cma area is > >>> initialized and recalculate various per zone stat/threshold. > >>> > >>> Signed-off-by: Joonsoo Kim > >>> --- > >>> include/linux/memory_hotplug.h | 3 --- > >>> mm/cma.c | 41 +++++++++++++++++++++++++++++++++++++++++ > >>> mm/internal.h | 3 +++ > >>> mm/page_alloc.c | 26 ++++++++++++++++++++++++-- > >>> 4 files changed, 68 insertions(+), 5 deletions(-) > >>> > >>> diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h > >>> index a864d79..6fde69b 100644 > >>> --- a/include/linux/memory_hotplug.h > >>> +++ b/include/linux/memory_hotplug.h > >>> @@ -198,9 +198,6 @@ void put_online_mems(void); > >>> void mem_hotplug_begin(void); > >>> void mem_hotplug_done(void); > >>> > >>> -extern void set_zone_contiguous(struct zone *zone); > >>> -extern void clear_zone_contiguous(struct zone *zone); > >>> - > >>> #else /* ! CONFIG_MEMORY_HOTPLUG */ > >>> /* > >>> * Stub functions for when hotplug is off > >>> diff --git a/mm/cma.c b/mm/cma.c > >>> index ea506eb..8684f50 100644 > >>> --- a/mm/cma.c > >>> +++ b/mm/cma.c > >>> @@ -38,6 +38,7 @@ > >>> #include > >>> > >>> #include "cma.h" > >>> +#include "internal.h" > >>> > >>> struct cma cma_areas[MAX_CMA_AREAS]; > >>> unsigned cma_area_count; > >>> @@ -145,6 +146,11 @@ err: > >>> static int __init cma_init_reserved_areas(void) > >>> { > >>> int i; > >>> + struct zone *zone; > >>> + unsigned long start_pfn = UINT_MAX, end_pfn = 0; > >>> + > >>> + if (!cma_area_count) > >>> + return 0; > >>> > >>> for (i = 0; i < cma_area_count; i++) { > >>> int ret = cma_activate_area(&cma_areas[i]); > >>> @@ -153,6 +159,41 @@ static int __init cma_init_reserved_areas(void) > >>> return ret; > >>> } > >>> > >>> + for (i = 0; i < cma_area_count; i++) { > >>> + if (start_pfn > cma_areas[i].base_pfn) > >>> + start_pfn = cma_areas[i].base_pfn; > >>> + if (end_pfn < cma_areas[i].base_pfn + cma_areas[i].count) > >>> + end_pfn = cma_areas[i].base_pfn + cma_areas[i].count; > >>> + } > >>> + > >>> + for_each_populated_zone(zone) { > >>> + if (!is_zone_cma(zone)) > >>> + continue; > >>> + > >>> + /* ZONE_CMA doesn't need to exceed CMA region */ > >>> + zone->zone_start_pfn = max(zone->zone_start_pfn, start_pfn); > >>> + zone->spanned_pages = min(zone_end_pfn(zone), end_pfn) - > >>> + zone->zone_start_pfn; > >>> + } > >>> + > >>> + /* > >>> + * Reserved pages for ZONE_CMA are now activated and this would change > >>> + * ZONE_CMA's managed page counter and other zone's present counter. > >>> + * We need to re-calculate various zone information that depends on > >>> + * this initialization. > >>> + */ > >>> + build_all_zonelists(NULL, NULL); > >>> + for_each_populated_zone(zone) { > >>> + zone_pcp_update(zone); > >>> + set_zone_contiguous(zone); > >>> + } > >>> + > >>> + /* > >>> + * We need to re-init per zone wmark by calling > >>> + * init_per_zone_wmark_min() but doesn't call here because it is > >>> + * registered on module_init and it will be called later than us. > >>> + */ > >>> + > >>> return 0; > >>> } > >>> core_initcall(cma_init_reserved_areas); > >>> diff --git a/mm/internal.h b/mm/internal.h > >>> index b6ead95..4c37234 100644 > >>> --- a/mm/internal.h > >>> +++ b/mm/internal.h > >>> @@ -155,6 +155,9 @@ extern void __free_pages_bootmem(struct page *page, unsigned long pfn, > >>> extern void prep_compound_page(struct page *page, unsigned int order); > >>> extern int user_min_free_kbytes; > >>> > >>> +extern void set_zone_contiguous(struct zone *zone); > >>> +extern void clear_zone_contiguous(struct zone *zone); > >>> + > >>> #if defined CONFIG_COMPACTION || defined CONFIG_CMA > >>> > >>> /* > >>> diff --git a/mm/page_alloc.c b/mm/page_alloc.c > >>> index 0197d5d..796b271 100644 > >>> --- a/mm/page_alloc.c > >>> +++ b/mm/page_alloc.c > >>> @@ -1572,16 +1572,38 @@ void __init page_alloc_init_late(void) > >>> } > >>> > >>> #ifdef CONFIG_CMA > >>> +static void __init adjust_present_page_count(struct page *page, long count) > >>> +{ > >>> + struct zone *zone = page_zone(page); > >>> + > >>> + /* We don't need to hold a lock since it is boot-up process */ > >>> + zone->present_pages += count; > >>> +} > >>> + > >>> /* Free whole pageblock and set its migration type to MIGRATE_CMA. */ > >>> void __init init_cma_reserved_pageblock(struct page *page) > >>> { > >>> unsigned i = pageblock_nr_pages; > >>> + unsigned long pfn = page_to_pfn(page); > >>> struct page *p = page; > >>> + int nid = page_to_nid(page); > >>> + > >>> + /* > >>> + * ZONE_CMA will steal present pages from other zones by changing > >>> + * page links so page_zone() is changed. Before that, > >>> + * we need to adjust previous zone's page count first. > >>> + */ > >>> + adjust_present_page_count(page, -pageblock_nr_pages); > >>> > >>> do { > >>> __ClearPageReserved(p); > >>> set_page_count(p, 0); > >>> - } while (++p, --i); > >>> + > >>> + /* Steal pages from other zones */ > >>> + set_page_links(p, ZONE_CMA, nid, pfn); > >>> + } while (++p, ++pfn, --i); > >>> + > >>> + adjust_present_page_count(page, pageblock_nr_pages); > >>> > >>> set_pageblock_migratetype(page, MIGRATE_CMA); > >> > >> The ZONE_CMA should depends on sparse_mem. > >> > >> Because the zone size is not fixed when init the buddy core. > >> The pageblock_flags will be NULL when setup_usemap. > > > > Before setup_usemap(), range of ZONE_CMA is set conservatively, from > > min_start_pfn of the node to max_end_pfn of the node. So, > > pageblock_flags will be allocated and assigned properly. > > > > Unfortunately, I found a bug for FLATMEM system. If you'd like to > > test ZONE_CMA on FLATMEM system, please apply below one. > > > > The filesystem, inode map is also GFP_HIGHUSER_MOVABLE. > > SyS_write filemap_fault will also use cma memory. > > This may also make cma migrate failed. What's your idea on this type? Purpose of this patchset is not improving success rate. It can be solved separately if it is a real problem. Could you share how above usage makes CMA migration failed and how long does it continue to fail? Thanks. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org