From c24ad77d962c31af92f2b731dad2104cbf3fbb03 Mon Sep 17 00:00:00 2001 From: Lucas Stach Date: Thu, 14 Dec 2017 15:32:55 -0800 Subject: mm/page_alloc.c: avoid excessive IRQ disabled times in free_unref_page_list() Since commit 9cca35d42eb6 ("mm, page_alloc: enable/disable IRQs once when freeing a list of pages") we see excessive IRQ disabled times of up to 25ms on an embedded ARM system (tracing overhead included). This is due to graphics buffers being freed back to the system via release_pages(). Graphics buffers can be huge, so it's not hard to hit cases where the list of pages to free has 2048 entries. Disabling IRQs while freeing all those pages is clearly not a good idea. Introduce a batch limit, which allows IRQ servicing once every few pages. The batch count is the same as used in other parts of the MM subsystem when dealing with IRQ disabled regions. Link: http://lkml.kernel.org/r/20171207170314.4419-1-l.stach@pengutronix.de Fixes: 9cca35d42eb6 ("mm, page_alloc: enable/disable IRQs once when freeing a list of pages") Signed-off-by: Lucas Stach Acked-by: Mel Gorman Cc: Michal Hocko Cc: Vlastimil Babka Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 73f5d4556b3d..7e5e775e97f4 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2684,6 +2684,7 @@ void free_unref_page_list(struct list_head *list) { struct page *page, *next; unsigned long flags, pfn; + int batch_count = 0; /* Prepare pages for freeing */ list_for_each_entry_safe(page, next, list, lru) { @@ -2700,6 +2701,16 @@ void free_unref_page_list(struct list_head *list) set_page_private(page, 0); trace_mm_page_free_batched(page); free_unref_page_commit(page, pfn); + + /* + * Guard against excessive IRQ disabled times when we get + * a large list of pages to free. + */ + if (++batch_count == SWAP_CLUSTER_MAX) { + local_irq_restore(flags); + batch_count = 0; + local_irq_save(flags); + } } local_irq_restore(flags); } -- cgit v1.2.3-59-g8ed1b From a99583e780c751003ac9c0105eec9a3b23ec3bc4 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 29 Dec 2017 08:53:57 +0100 Subject: mm: pass the vmem_altmap to memmap_init_zone Pass the vmem_altmap two levels down instead of needing a lookup. Signed-off-by: Christoph Hellwig Signed-off-by: Dan Williams --- arch/ia64/mm/init.c | 9 +++++---- include/linux/memory_hotplug.h | 2 +- include/linux/mm.h | 4 ++-- kernel/memremap.c | 2 +- mm/hmm.c | 2 +- mm/memory_hotplug.c | 9 +++++---- mm/page_alloc.c | 6 +++--- 7 files changed, 18 insertions(+), 16 deletions(-) (limited to 'mm/page_alloc.c') diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index 6a8ce9e1536e..18278b448530 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -501,7 +501,7 @@ virtual_memmap_init(u64 start, u64 end, void *arg) if (map_start < map_end) memmap_init_zone((unsigned long)(map_end - map_start), args->nid, args->zone, page_to_pfn(map_start), - MEMMAP_EARLY); + MEMMAP_EARLY, NULL); return 0; } @@ -509,9 +509,10 @@ void __meminit memmap_init (unsigned long size, int nid, unsigned long zone, unsigned long start_pfn) { - if (!vmem_map) - memmap_init_zone(size, nid, zone, start_pfn, MEMMAP_EARLY); - else { + if (!vmem_map) { + memmap_init_zone(size, nid, zone, start_pfn, MEMMAP_EARLY, + NULL); + } else { struct page *start; struct memmap_init_callback_data args; diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index 20dd98ad44a0..aba5f86eb038 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -324,7 +324,7 @@ extern int add_memory_resource(int nid, struct resource *resource, bool online); extern int arch_add_memory(int nid, u64 start, u64 size, struct vmem_altmap *altmap, bool want_memblock); extern void move_pfn_range_to_zone(struct zone *zone, unsigned long start_pfn, - unsigned long nr_pages); + unsigned long nr_pages, struct vmem_altmap *altmap); extern int offline_pages(unsigned long start_pfn, unsigned long nr_pages); extern bool is_memblock_offlined(struct memory_block *mem); extern void remove_memory(int nid, u64 start, u64 size); diff --git a/include/linux/mm.h b/include/linux/mm.h index 9d4cd4c1dc6d..fd01135324b6 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -2069,8 +2069,8 @@ static inline void zero_resv_unavail(void) {} #endif extern void set_dma_reserve(unsigned long new_dma_reserve); -extern void memmap_init_zone(unsigned long, int, unsigned long, - unsigned long, enum memmap_context); +extern void memmap_init_zone(unsigned long, int, unsigned long, unsigned long, + enum memmap_context, struct vmem_altmap *); extern void setup_per_zone_wmarks(void); extern int __meminit init_per_zone_wmark_min(void); extern void mem_init(void); diff --git a/kernel/memremap.c b/kernel/memremap.c index 380fca1c4a02..64b12c806cc5 100644 --- a/kernel/memremap.c +++ b/kernel/memremap.c @@ -432,7 +432,7 @@ void *devm_memremap_pages(struct device *dev, struct resource *res, if (!error) move_pfn_range_to_zone(&NODE_DATA(nid)->node_zones[ZONE_DEVICE], align_start >> PAGE_SHIFT, - align_size >> PAGE_SHIFT); + align_size >> PAGE_SHIFT, altmap); mem_hotplug_done(); if (error) goto err_add_memory; diff --git a/mm/hmm.c b/mm/hmm.c index 5d17ba89062f..2f2e13c61040 100644 --- a/mm/hmm.c +++ b/mm/hmm.c @@ -942,7 +942,7 @@ static int hmm_devmem_pages_create(struct hmm_devmem *devmem) } move_pfn_range_to_zone(&NODE_DATA(nid)->node_zones[ZONE_DEVICE], align_start >> PAGE_SHIFT, - align_size >> PAGE_SHIFT); + align_size >> PAGE_SHIFT, NULL); mem_hotplug_done(); for (pfn = devmem->pfn_first; pfn < devmem->pfn_last; pfn++) { diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index a8dde9734120..12df8a5fadcc 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -798,8 +798,8 @@ static void __meminit resize_pgdat_range(struct pglist_data *pgdat, unsigned lon pgdat->node_spanned_pages = max(start_pfn + nr_pages, old_end_pfn) - pgdat->node_start_pfn; } -void __ref move_pfn_range_to_zone(struct zone *zone, - unsigned long start_pfn, unsigned long nr_pages) +void __ref move_pfn_range_to_zone(struct zone *zone, unsigned long start_pfn, + unsigned long nr_pages, struct vmem_altmap *altmap) { struct pglist_data *pgdat = zone->zone_pgdat; int nid = pgdat->node_id; @@ -824,7 +824,8 @@ void __ref move_pfn_range_to_zone(struct zone *zone, * expects the zone spans the pfn range. All the pages in the range * are reserved so nobody should be touching them so we should be safe */ - memmap_init_zone(nr_pages, nid, zone_idx(zone), start_pfn, MEMMAP_HOTPLUG); + memmap_init_zone(nr_pages, nid, zone_idx(zone), start_pfn, + MEMMAP_HOTPLUG, altmap); set_zone_contiguous(zone); } @@ -896,7 +897,7 @@ static struct zone * __meminit move_pfn_range(int online_type, int nid, struct zone *zone; zone = zone_for_pfn_range(online_type, nid, start_pfn, nr_pages); - move_pfn_range_to_zone(zone, start_pfn, nr_pages); + move_pfn_range_to_zone(zone, start_pfn, nr_pages, NULL); return zone; } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 7e5e775e97f4..1748dd4a4b1b 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -5314,9 +5314,9 @@ void __ref build_all_zonelists(pg_data_t *pgdat) * done. Non-atomic initialization, single-pass. */ void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone, - unsigned long start_pfn, enum memmap_context context) + unsigned long start_pfn, enum memmap_context context, + struct vmem_altmap *altmap) { - struct vmem_altmap *altmap = to_vmem_altmap(__pfn_to_phys(start_pfn)); unsigned long end_pfn = start_pfn + size; pg_data_t *pgdat = NODE_DATA(nid); unsigned long pfn; @@ -5417,7 +5417,7 @@ static void __meminit zone_init_free_lists(struct zone *zone) #ifndef __HAVE_ARCH_MEMMAP_INIT #define memmap_init(size, nid, zone, start_pfn) \ - memmap_init_zone((size), (nid), (zone), (start_pfn), MEMMAP_EARLY) + memmap_init_zone((size), (nid), (zone), (start_pfn), MEMMAP_EARLY, NULL) #endif static int zone_batchsize(struct zone *zone) -- cgit v1.2.3-59-g8ed1b