On Mon, 14 Jul 2014, Johannes Weiner wrote:
> Reorder the members by input and output, then turn the individual
> integers for may_writepage, may_unmap, may_swap, compaction_ready,
> hibernation_mode into flags that fit into a single integer.
>
> Stack delta: +72/-296 -224 old new delta
> kswapd 104 176 +72
> try_to_free_pages 80 56 -24
> try_to_free_mem_cgroup_pages 80 56 -24
> shrink_all_memory 88 64 -24
> reclaim_clean_pages_from_list 168 144 -24
> mem_cgroup_shrink_node_zone 104 80 -24
> __zone_reclaim 176 152 -24
> balance_pgdat 152 - -152
>
> text data bss dec hex filename
> 38151 5641 16 43808 ab20 mm/vmscan.o.old
> 38047 5641 16 43704 aab8 mm/vmscan.o
>
> Suggested-by: Mel Gorman <mgorman@xxxxxxx>
> Signed-off-by: Johannes Weiner <hannes@xxxxxxxxxxx>
> ---
> mm/vmscan.c | 158 ++++++++++++++++++++++++++++++------------------------------
> 1 file changed, 78 insertions(+), 80 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index c28b8981e56a..73d8e69ff3eb 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -58,36 +58,28 @@
> #define CREATE_TRACE_POINTS
> #include <trace/events/vmscan.h>
>
> -struct scan_control {
> - /* Incremented by the number of inactive pages that were scanned */
> - unsigned long nr_scanned;
> -
> - /* Number of pages freed so far during a call to shrink_zones() */
> - unsigned long nr_reclaimed;
> -
> - /* One of the zones is ready for compaction */
> - int compaction_ready;
> +/* Scan control flags */
> +#define MAY_WRITEPAGE 0x1
> +#define MAY_UNMAP 0x2
> +#define MAY_SWAP 0x4
> +#define MAY_SKIP_CONGESTION 0x8
> +#define COMPACTION_READY 0x10
>
> +struct scan_control {
> /* How many pages shrink_list() should reclaim */
> unsigned long nr_to_reclaim;
>
> - unsigned long hibernation_mode;
> -
> /* This context's GFP mask */
> gfp_t gfp_mask;
>
> - int may_writepage;
> -
> - /* Can mapped pages be reclaimed? */
> - int may_unmap;
> -
> - /* Can pages be swapped as part of reclaim? */
> - int may_swap;
> -
> + /* Allocation order */
> int order;
>
> - /* Scan (total_size >> priority) pages at once */
> - int priority;
> + /*
> + * Nodemask of nodes allowed by the caller. If NULL, all nodes
> + * are scanned.
> + */
> + nodemask_t *nodemask;
>
> /*
> * The memory cgroup that hit its limit and as a result is the
> @@ -95,11 +87,17 @@ struct scan_control {
> */
> struct mem_cgroup *target_mem_cgroup;
>
> - /*
> - * Nodemask of nodes allowed by the caller. If NULL, all nodes
> - * are scanned.
> - */
> - nodemask_t *nodemask;
> + /* Scan (total_size >> priority) pages at once */
> + int priority;
> +
> + /* Scan control flags; see above */
> + unsigned int flags;
This seems to result in a fair amount of unnecessary churn:
why not just put may_writepage etc into an unsigned int bitfield,
then you get the saving without changing all the rest of the code.
Hugh
> +
> + /* Incremented by the number of inactive pages that were scanned */
> + unsigned long nr_scanned;
> +
> + /* Number of pages freed so far during a call to shrink_zones() */
> + unsigned long nr_reclaimed;
> };
>
> #define lru_to_page(_head) (list_entry((_head)->prev, struct page, lru))
> @@ -840,7 +838,7 @@ static unsigned long shrink_page_list(struct list_head *page_list,
> if (unlikely(!page_evictable(page)))
> goto cull_mlocked;
>
> - if (!sc->may_unmap && page_mapped(page))
> + if (!(sc->flags & MAY_UNMAP) && page_mapped(page))
> goto keep_locked;
>
> /* Double the slab pressure for mapped and swapcache pages */
> @@ -1014,7 +1012,7 @@ static unsigned long shrink_page_list(struct list_head *page_list,
> goto keep_locked;
> if (!may_enter_fs)
> goto keep_locked;
> - if (!sc->may_writepage)
> + if (!(sc->flags & MAY_WRITEPAGE))
> goto keep_locked;
>
> /* Page is dirty, try to write it out here */
> @@ -1146,7 +1144,7 @@ unsigned long reclaim_clean_pages_from_list(struct zone *zone,
> struct scan_control sc = {
> .gfp_mask = GFP_KERNEL,
> .priority = DEF_PRIORITY,
> - .may_unmap = 1,
> + .flags = MAY_UNMAP,
> };
> unsigned long ret, dummy1, dummy2, dummy3, dummy4, dummy5;
> struct page *page, *next;
> @@ -1489,9 +1487,9 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
>
> lru_add_drain();
>
> - if (!sc->may_unmap)
> + if (!(sc->flags & MAY_UNMAP))
> isolate_mode |= ISOLATE_UNMAPPED;
> - if (!sc->may_writepage)
> + if (!(sc->flags & MAY_WRITEPAGE))
> isolate_mode |= ISOLATE_CLEAN;
>
> spin_lock_irq(&zone->lru_lock);
> @@ -1593,7 +1591,7 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
> * is congested. Allow kswapd to continue until it starts encountering
> * unqueued dirty pages or cycling through the LRU too quickly.
> */
> - if (!sc->hibernation_mode && !current_is_kswapd() &&
> + if (!(sc->flags & MAY_SKIP_CONGESTION) && !current_is_kswapd() &&
> current_may_throttle())
> wait_iff_congested(zone, BLK_RW_ASYNC, HZ/10);
>
> @@ -1683,9 +1681,9 @@ static void shrink_active_list(unsigned long nr_to_scan,
>
> lru_add_drain();
>
> - if (!sc->may_unmap)
> + if (!(sc->flags & MAY_UNMAP))
> isolate_mode |= ISOLATE_UNMAPPED;
> - if (!sc->may_writepage)
> + if (!(sc->flags & MAY_WRITEPAGE))
> isolate_mode |= ISOLATE_CLEAN;
>
> spin_lock_irq(&zone->lru_lock);
> @@ -1897,7 +1895,7 @@ static void get_scan_count(struct lruvec *lruvec, int swappiness,
> force_scan = true;
>
> /* If we have no swap space, do not bother scanning anon pages. */
> - if (!sc->may_swap || (get_nr_swap_pages() <= 0)) {
> + if (!(sc->flags & MAY_SWAP) || (get_nr_swap_pages() <= 0)) {
> scan_balance = SCAN_FILE;
> goto out;
> }
> @@ -2406,7 +2404,7 @@ static bool shrink_zones(struct zonelist *zonelist, struct scan_control *sc)
> sc->order > PAGE_ALLOC_COSTLY_ORDER &&
> zonelist_zone_idx(z) <= requested_highidx &&
> compaction_ready(zone, sc->order)) {
> - sc->compaction_ready = true;
> + sc->flags |= COMPACTION_READY;
> continue;
> }
>
> @@ -2496,7 +2494,7 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist,
> if (sc->nr_reclaimed >= sc->nr_to_reclaim)
> break;
>
> - if (sc->compaction_ready)
> + if (sc->flags & COMPACTION_READY)
> break;
>
> /*
> @@ -2504,7 +2502,7 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist,
> * writepage even in laptop mode.
> */
> if (sc->priority < DEF_PRIORITY - 2)
> - sc->may_writepage = 1;
> + sc->flags |= MAY_WRITEPAGE;
>
> /*
> * Try to write back as many pages as we just scanned. This
> @@ -2517,7 +2515,7 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist,
> if (total_scanned > writeback_threshold) {
> wakeup_flusher_threads(laptop_mode ? 0 : total_scanned,
> WB_REASON_TRY_TO_FREE_PAGES);
> - sc->may_writepage = 1;
> + sc->flags |= MAY_WRITEPAGE;
> }
> } while (--sc->priority >= 0);
>
> @@ -2527,7 +2525,7 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist,
> return sc->nr_reclaimed;
>
> /* Aborted reclaim to try compaction? don't OOM, then */
> - if (sc->compaction_ready)
> + if (sc->flags & COMPACTION_READY)
> return 1;
>
> /* Any of the zones still reclaimable? Don't OOM. */
> @@ -2668,17 +2666,17 @@ unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
> {
> unsigned long nr_reclaimed;
> struct scan_control sc = {
> - .gfp_mask = (gfp_mask = memalloc_noio_flags(gfp_mask)),
> - .may_writepage = !laptop_mode,
> .nr_to_reclaim = SWAP_CLUSTER_MAX,
> - .may_unmap = 1,
> - .may_swap = 1,
> + .gfp_mask = (gfp_mask = memalloc_noio_flags(gfp_mask)),
> .order = order,
> - .priority = DEF_PRIORITY,
> - .target_mem_cgroup = NULL,
> .nodemask = nodemask,
> + .priority = DEF_PRIORITY,
> + .flags = MAY_UNMAP | MAY_SWAP,
> };
>
> + if (!laptop_mode)
> + sc.flags |= MAY_WRITEPAGE;
> +
> /*
> * Do not enter reclaim if fatal signal was delivered while throttled.
> * 1 is returned so that the page allocator does not OOM kill at this
> @@ -2688,7 +2686,7 @@ unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
> return 1;
>
> trace_mm_vmscan_direct_reclaim_begin(order,
> - sc.may_writepage,
> + sc.flags & MAY_WRITEPAGE,
> gfp_mask);
>
> nr_reclaimed = do_try_to_free_pages(zonelist, &sc);
> @@ -2706,23 +2704,22 @@ unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *memcg,
> unsigned long *nr_scanned)
> {
> struct scan_control sc = {
> - .nr_scanned = 0,
> .nr_to_reclaim = SWAP_CLUSTER_MAX,
> - .may_writepage = !laptop_mode,
> - .may_unmap = 1,
> - .may_swap = !noswap,
> - .order = 0,
> - .priority = 0,
> + .gfp_mask = (gfp_mask & GFP_RECLAIM_MASK) |
> + (GFP_HIGHUSER_MOVABLE & ~GFP_RECLAIM_MASK),
> .target_mem_cgroup = memcg,
> + .flags = MAY_UNMAP,
> };
> struct lruvec *lruvec = mem_cgroup_zone_lruvec(zone, memcg);
> int swappiness = mem_cgroup_swappiness(memcg);
>
> - sc.gfp_mask = (gfp_mask & GFP_RECLAIM_MASK) |
> - (GFP_HIGHUSER_MOVABLE & ~GFP_RECLAIM_MASK);
> + if (!laptop_mode)
> + sc.flags |= MAY_WRITEPAGE;
> + if (!noswap)
> + sc.flags |= MAY_SWAP;
>
> trace_mm_vmscan_memcg_softlimit_reclaim_begin(sc.order,
> - sc.may_writepage,
> + sc.flags & MAY_WRITEPAGE,
> sc.gfp_mask);
>
> /*
> @@ -2748,18 +2745,19 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
> unsigned long nr_reclaimed;
> int nid;
> struct scan_control sc = {
> - .may_writepage = !laptop_mode,
> - .may_unmap = 1,
> - .may_swap = !noswap,
> .nr_to_reclaim = SWAP_CLUSTER_MAX,
> - .order = 0,
> - .priority = DEF_PRIORITY,
> - .target_mem_cgroup = memcg,
> - .nodemask = NULL, /* we don't care the placement */
> .gfp_mask = (gfp_mask & GFP_RECLAIM_MASK) |
> - (GFP_HIGHUSER_MOVABLE & ~GFP_RECLAIM_MASK),
> + (GFP_HIGHUSER_MOVABLE & ~GFP_RECLAIM_MASK),
> + .target_mem_cgroup = memcg,
> + .priority = DEF_PRIORITY,
> + .flags = MAY_UNMAP,
> };
>
> + if (!laptop_mode)
> + sc.flags |= MAY_WRITEPAGE;
> + if (!noswap)
> + sc.flags |= MAY_SWAP;
> +
> /*
> * Unlike direct reclaim via alloc_pages(), memcg's reclaim doesn't
> * take care of from where we get pages. So the node where we start the
> @@ -2770,7 +2768,7 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
> zonelist = NODE_DATA(nid)->node_zonelists;
>
> trace_mm_vmscan_memcg_reclaim_begin(0,
> - sc.may_writepage,
> + sc.flags & MAY_WRITEPAGE,
> sc.gfp_mask);
>
> nr_reclaimed = do_try_to_free_pages(zonelist, &sc);
> @@ -3015,15 +3013,15 @@ static unsigned long balance_pgdat(pg_data_t *pgdat, int order,
> unsigned long nr_soft_scanned;
> struct scan_control sc = {
> .gfp_mask = GFP_KERNEL,
> - .priority = DEF_PRIORITY,
> - .may_unmap = 1,
> - .may_swap = 1,
> - .may_writepage = !laptop_mode,
> .order = order,
> - .target_mem_cgroup = NULL,
> + .priority = DEF_PRIORITY,
> + .flags = MAY_UNMAP | MAY_SWAP,
> };
> count_vm_event(PAGEOUTRUN);
>
> + if (!laptop_mode)
> + sc.flags |= MAY_WRITEPAGE;
> +
> do {
> unsigned long lru_pages = 0;
> unsigned long nr_attempted = 0;
> @@ -3104,7 +3102,7 @@ static unsigned long balance_pgdat(pg_data_t *pgdat, int order,
> * even in laptop mode.
> */
> if (sc.priority < DEF_PRIORITY - 2)
> - sc.may_writepage = 1;
> + sc.flags |= MAY_WRITEPAGE;
>
> /*
> * Now scan the zone in the dma->highmem direction, stopping
> @@ -3401,14 +3399,11 @@ unsigned long shrink_all_memory(unsigned long nr_to_reclaim)
> {
> struct reclaim_state reclaim_state;
> struct scan_control sc = {
> - .gfp_mask = GFP_HIGHUSER_MOVABLE,
> - .may_swap = 1,
> - .may_unmap = 1,
> - .may_writepage = 1,
> .nr_to_reclaim = nr_to_reclaim,
> - .hibernation_mode = 1,
> - .order = 0,
> + .gfp_mask = GFP_HIGHUSER_MOVABLE,
> .priority = DEF_PRIORITY,
> + .flags = MAY_WRITEPAGE | MAY_UNMAP | MAY_SWAP |
> + MAY_SKIP_CONGESTION,
> };
> struct zonelist *zonelist = node_zonelist(numa_node_id(), sc.gfp_mask);
> struct task_struct *p = current;
> @@ -3588,19 +3583,22 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
> struct task_struct *p = current;
> struct reclaim_state reclaim_state;
> struct scan_control sc = {
> - .may_writepage = !!(zone_reclaim_mode & RECLAIM_WRITE),
> - .may_unmap = !!(zone_reclaim_mode & RECLAIM_SWAP),
> - .may_swap = 1,
> .nr_to_reclaim = max(nr_pages, SWAP_CLUSTER_MAX),
> .gfp_mask = (gfp_mask = memalloc_noio_flags(gfp_mask)),
> .order = order,
> .priority = ZONE_RECLAIM_PRIORITY,
> + .flags = MAY_SWAP,
> };
> struct shrink_control shrink = {
> .gfp_mask = sc.gfp_mask,
> };
> unsigned long nr_slab_pages0, nr_slab_pages1;
>
> + if (zone_reclaim_mode & RECLAIM_WRITE)
> + sc.flags |= MAY_WRITEPAGE;
> + if (zone_reclaim_mode & RECLAIM_SWAP)
> + sc.flags |= MAY_UNMAP;
> +
> cond_resched();
> /*
> * We need to be able to allocate from the reserves for RECLAIM_SWAP
> --
> 2.0.0
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>