From: Michal Hocko <mhocko@kernel.org> To: Mel Gorman <mgorman@suse.de> Cc: linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>, Johannes Weiner <hannes@cmpxchg.org>, Vlastimil Babka <vbabka@suse.cz>, LKML <linux-kernel@vger.kernel.org> Subject: Re: [PATCH 6/9] mm, page_alloc: simplify zonelist initialization Date: Fri, 14 Jul 2017 15:02:42 +0200 Message-ID: <20170714130242.GQ2618@dhcp22.suse.cz> (raw) In-Reply-To: <20170714124645.i3duhuie6cczlybr@suse.de> On Fri 14-07-17 13:46:46, Mel Gorman wrote: > On Fri, Jul 14, 2017 at 10:00:03AM +0200, Michal Hocko wrote: > > From: Michal Hocko <mhocko@suse.com> > > > > build_zonelists gradually builds zonelists from the nearest to the most > > distant node. As we do not know how many populated zones we will have in > > each node we rely on the _zoneref to terminate initialized part of the > > zonelist by a NULL zone. While this is functionally correct it is quite > > suboptimal because we cannot allow updaters to race with zonelists > > users because they could see an empty zonelist and fail the allocation > > or hit the OOM killer in the worst case. > > > > We can do much better, though. We can store the node ordering into an > > already existing node_order array and then give this array to > > build_zonelists_in_node_order and do the whole initialization at once. > > zonelists consumers still might see halfway initialized state but that > > should be much more tolerateable because the list will not be empty and > > they would either see some zone twice or skip over some zone(s) in the > > worst case which shouldn't lead to immediate failures. > > > > This patch alone doesn't introduce any functional change yet, though, it > > is merely a preparatory work for later changes. > > > > Signed-off-by: Michal Hocko <mhocko@suse.com> > > --- > > mm/page_alloc.c | 42 ++++++++++++++++++------------------------ > > 1 file changed, 18 insertions(+), 24 deletions(-) > > > > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > > index 00e117922b3f..78bd62418380 100644 > > --- a/mm/page_alloc.c > > +++ b/mm/page_alloc.c > > @@ -4913,17 +4913,20 @@ static int find_next_best_node(int node, nodemask_t *used_node_mask) > > * This results in maximum locality--normal zone overflows into local > > * DMA zone, if any--but risks exhausting DMA zone. > > */ > > -static void build_zonelists_in_node_order(pg_data_t *pgdat, int node) > > +static void build_zonelists_in_node_order(pg_data_t *pgdat, int *node_order) > > { > > - int j; > > struct zonelist *zonelist; > > + int i, zoneref_idx = 0; > > > > zonelist = &pgdat->node_zonelists[ZONELIST_FALLBACK]; > > - for (j = 0; zonelist->_zonerefs[j].zone != NULL; j++) > > - ; > > - j = build_zonelists_node(NODE_DATA(node), zonelist, j); > > - zonelist->_zonerefs[j].zone = NULL; > > - zonelist->_zonerefs[j].zone_idx = 0; > > + > > + for (i = 0; i < MAX_NUMNODES; i++) { > > + pg_data_t *node = NODE_DATA(node_order[i]); > > + > > + zoneref_idx = build_zonelists_node(node, zonelist, zoneref_idx); > > + } > > The naming here is weird to say the least and makes this a lot more > confusing than it needs to be. Primarily, it's because the zoneref_idx > parameter gets renamed to nr_zones in build_zonelists_node where it's > nothing to do with the number of zones at all. you are right. I just wanted to get rid of `j' and didn't realize nr_zones would fit much better. > It also iterates for longer than it needs to. MAX_NUMNODES can be a > large value of mostly empty nodes but it happily goes through them > anyway. Pass zoneref_idx in as a pointer that is updated by the function > and use the return value to break the loop when an empty node is > encountered? > > > + zonelist->_zonerefs[zoneref_idx].zone = NULL; > > + zonelist->_zonerefs[zoneref_idx].zone_idx = 0; > > } > > > > It *might* be safer given the next patch to zero out the remainder of > the _zonerefs to that there is no combination of node add/remove that has > an iterator working with a semi-valid _zoneref which is beyond the last > correct value. It *should* be safe as the very last entry will always > be null but if you don't zero it out, it is possible for iterators to be > working beyond the "end" of the zonelist for a short window. yes that is true but there will always be terminating NULL zone and I found that acceptable. It is basically the same thing as accessing an empty zone or a zone twice. Or do you think this is absolutely necessary to handle? > Otherwise think it's ok including my stupid comment about node_order > stack usage. What do you think about this on top? --- diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 49bade7ff049..3b98524c04ec 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -4913,20 +4913,21 @@ static int find_next_best_node(int node, nodemask_t *used_node_mask) * This results in maximum locality--normal zone overflows into local * DMA zone, if any--but risks exhausting DMA zone. */ -static void build_zonelists_in_node_order(pg_data_t *pgdat, int *node_order) +static void build_zonelists_in_node_order(pg_data_t *pgdat, int *node_order, + unsigned nr_nodes) { struct zonelist *zonelist; - int i, zoneref_idx = 0; + int i, nr_zones = 0; zonelist = &pgdat->node_zonelists[ZONELIST_FALLBACK]; - for (i = 0; i < MAX_NUMNODES; i++) { + for (i = 0; i < nr_nodes; i++) { pg_data_t *node = NODE_DATA(node_order[i]); - zoneref_idx = build_zonelists_node(node, zonelist, zoneref_idx); + nr_zones = build_zonelists_node(node, zonelist, nr_zones); } - zonelist->_zonerefs[zoneref_idx].zone = NULL; - zonelist->_zonerefs[zoneref_idx].zone_idx = 0; + zonelist->_zonerefs[nr_zones].zone = NULL; + zonelist->_zonerefs[nr_zones].zone_idx = 0; } /* @@ -4935,12 +4936,12 @@ static void build_zonelists_in_node_order(pg_data_t *pgdat, int *node_order) static void build_thisnode_zonelists(pg_data_t *pgdat) { struct zonelist *zonelist; - int zoneref_idx = 0; + int nr_zones = 0; zonelist = &pgdat->node_zonelists[ZONELIST_NOFALLBACK]; - zoneref_idx = build_zonelists_node(pgdat, zonelist, zoneref_idx); - zonelist->_zonerefs[zoneref_idx].zone = NULL; - zonelist->_zonerefs[zoneref_idx].zone_idx = 0; + nr_zones = build_zonelists_node(pgdat, zonelist, nr_zones); + zonelist->_zonerefs[nr_zones].zone = NULL; + zonelist->_zonerefs[nr_zones].zone_idx = 0; } /* @@ -4979,7 +4980,7 @@ static void build_zonelists(pg_data_t *pgdat) load--; } - build_zonelists_in_node_order(pgdat, node_order); + build_zonelists_in_node_order(pgdat, node_order, i); build_thisnode_zonelists(pgdat); } -- Michal Hocko SUSE Labs -- 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: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply index Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top 2017-07-14 7:59 [PATCH 0/9] cleanup zonelists initialization Michal Hocko 2017-07-14 7:59 ` [PATCH 1/9] mm, page_alloc: rip out ZONELIST_ORDER_ZONE Michal Hocko 2017-07-14 9:36 ` Mel Gorman 2017-07-14 10:47 ` Michal Hocko 2017-07-14 11:16 ` Mel Gorman 2017-07-14 11:38 ` Michal Hocko 2017-07-14 12:56 ` Mel Gorman 2017-07-14 13:01 ` Mel Gorman 2017-07-14 13:08 ` Michal Hocko 2017-07-19 9:33 ` Vlastimil Babka 2017-07-19 13:44 ` Michal Hocko 2017-07-14 7:59 ` [PATCH 2/9] mm, page_alloc: remove boot pageset initialization from memory hotplug Michal Hocko 2017-07-14 9:39 ` Mel Gorman 2017-07-19 13:15 ` Vlastimil Babka 2017-07-14 8:00 ` [PATCH 3/9] mm, page_alloc: do not set_cpu_numa_mem on empty nodes initialization Michal Hocko 2017-07-14 9:48 ` Mel Gorman 2017-07-14 10:50 ` Michal Hocko 2017-07-14 12:32 ` Mel Gorman 2017-07-14 12:39 ` Michal Hocko 2017-07-14 12:56 ` Mel Gorman 2017-07-19 13:19 ` Vlastimil Babka 2017-07-14 8:00 ` [PATCH 4/9] mm, memory_hotplug: drop zone from build_all_zonelists Michal Hocko 2017-07-19 13:33 ` Vlastimil Babka 2017-07-20 8:15 ` Michal Hocko 2017-07-14 8:00 ` [PATCH 5/9] mm, memory_hotplug: remove explicit build_all_zonelists from try_online_node Michal Hocko 2017-07-14 12:14 ` Michal Hocko 2017-07-20 6:13 ` Vlastimil Babka 2017-07-14 8:00 ` [PATCH 6/9] mm, page_alloc: simplify zonelist initialization Michal Hocko 2017-07-14 9:55 ` Mel Gorman 2017-07-14 10:51 ` Michal Hocko 2017-07-14 12:46 ` Mel Gorman 2017-07-14 13:02 ` Michal Hocko [this message] 2017-07-14 14:18 ` Mel Gorman 2017-07-17 6:06 ` Michal Hocko 2017-07-17 8:07 ` Mel Gorman 2017-07-17 8:19 ` Michal Hocko 2017-07-17 8:58 ` Mel Gorman 2017-07-17 9:15 ` Michal Hocko 2017-07-20 6:55 ` Vlastimil Babka 2017-07-20 7:19 ` Michal Hocko 2017-07-14 8:00 ` [PATCH 7/9] mm, page_alloc: remove stop_machine from build_all_zonelists Michal Hocko 2017-07-14 9:59 ` Mel Gorman 2017-07-14 11:00 ` Michal Hocko 2017-07-14 12:47 ` Mel Gorman 2017-07-14 11:29 ` Vlastimil Babka 2017-07-14 11:43 ` Michal Hocko 2017-07-14 11:45 ` Michal Hocko 2017-07-20 6:16 ` Vlastimil Babka 2017-07-20 7:24 ` Vlastimil Babka 2017-07-20 9:21 ` Michal Hocko 2017-07-14 8:00 ` [PATCH 8/9] mm, memory_hotplug: get rid of zonelists_mutex Michal Hocko 2017-07-14 8:00 ` [PATCH 9/9] mm, sparse, page_ext: drop ugly N_HIGH_MEMORY branches for allocations Michal Hocko 2017-07-20 8:04 ` Vlastimil Babka 2017-07-21 14:39 [PATCH -v1 0/9] cleanup zonelists initialization Michal Hocko 2017-07-21 14:39 ` [PATCH 6/9] mm, page_alloc: simplify zonelist initialization Michal Hocko 2017-07-24 9:25 ` Vlastimil Babka
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20170714130242.GQ2618@dhcp22.suse.cz \ --to=mhocko@kernel.org \ --cc=akpm@linux-foundation.org \ --cc=hannes@cmpxchg.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=mgorman@suse.de \ --cc=vbabka@suse.cz \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: link
Linux-mm Archive on lore.kernel.org Archives are clonable: git clone --mirror https://lore.kernel.org/linux-mm/0 linux-mm/git/0.git # If you have public-inbox 1.1+ installed, you may # initialize and index your mirror using the following commands: public-inbox-init -V2 linux-mm linux-mm/ https://lore.kernel.org/linux-mm \ linux-mm@kvack.org public-inbox-index linux-mm Example config snippet for mirrors Newsgroup available over NNTP: nntp://nntp.lore.kernel.org/org.kvack.linux-mm AGPL code for this site: git clone https://public-inbox.org/public-inbox.git