Linaro-mm-sig November 2012

linaro-mm-sig@lists.linaro.org

14 participants
9 discussions

[PATCH] mm: dmapool: use provided gfp flags for all dma_alloc_coherent() calls

by Marek Szyprowski

dmapool always calls dma_alloc_coherent() with GFP_ATOMIC flag, regardless the flags provided by the caller. This causes excessive pruning of emergency memory pools without any good reason. This patch changes the code to correctly use gfp flags provided by the dmapool caller. This should solve the dmapool usage on ARM architecture, where GFP_ATOMIC DMA allocations can be served only from the special, very limited memory pool. Reported-by: Soren Moch <smoch(a)web.de> Reported-by: Thomas Petazzoni <thomas.petazzoni(a)free-electrons.com> Signed-off-by: Marek Szyprowski <m.szyprowski(a)samsung.com> --- mm/dmapool.c | 27 +++++++-------------------- 1 file changed, 7 insertions(+), 20 deletions(-) diff --git a/mm/dmapool.c b/mm/dmapool.c index c5ab33b..86de9b2 100644 --- a/mm/dmapool.c +++ b/mm/dmapool.c @@ -62,8 +62,6 @@ struct dma_page { /* cacheable header for 'allocation' bytes */ unsigned int offset; }; -#define POOL_TIMEOUT_JIFFIES ((100 /* msec */ * HZ) / 1000) - static DEFINE_MUTEX(pools_lock); static ssize_t @@ -227,7 +225,6 @@ static struct dma_page *pool_alloc_page(struct dma_pool *pool, gfp_t mem_flags) memset(page->vaddr, POOL_POISON_FREED, pool->allocation); #endif pool_initialise_page(pool, page); - list_add(&page->page_list, &pool->page_list); page->in_use = 0; page->offset = 0; } else { @@ -315,30 +312,21 @@ void *dma_pool_alloc(struct dma_pool *pool, gfp_t mem_flags, might_sleep_if(mem_flags & __GFP_WAIT); spin_lock_irqsave(&pool->lock, flags); - restart: list_for_each_entry(page, &pool->page_list, page_list) { if (page->offset < pool->allocation) goto ready; } - page = pool_alloc_page(pool, GFP_ATOMIC); - if (!page) { - if (mem_flags & __GFP_WAIT) { - DECLARE_WAITQUEUE(wait, current); - __set_current_state(TASK_UNINTERRUPTIBLE); - __add_wait_queue(&pool->waitq, &wait); - spin_unlock_irqrestore(&pool->lock, flags); + /* pool_alloc_page() might sleep, so temporarily drop &pool->lock */ + spin_unlock_irqrestore(&pool->lock, flags); - schedule_timeout(POOL_TIMEOUT_JIFFIES); + page = pool_alloc_page(pool, mem_flags); + if (!page) + return NULL; - spin_lock_irqsave(&pool->lock, flags); - __remove_wait_queue(&pool->waitq, &wait); - goto restart; - } - retval = NULL; - goto done; - } + spin_lock_irqsave(&pool->lock, flags); + list_add(&page->page_list, &pool->page_list); ready: page->in_use++; offset = page->offset; @@ -348,7 +336,6 @@ void *dma_pool_alloc(struct dma_pool *pool, gfp_t mem_flags, #ifdef DMAPOOL_DEBUG memset(retval, POOL_POISON_ALLOCATED, pool->size); #endif - done: spin_unlock_irqrestore(&pool->lock, flags); return retval; } -- 1.7.9.5

10 years, 11 months

[RFC 0/5] ARM: dma-mapping: New dma_map_ops to control IOVA more precisely

by Hiroshi Doyu

Hi, The following APIs are needed for us to support the legacy Tegra memory manager for devices("NvMap") with *DMA mapping API*. New API: ->iova_alloc(): To allocate IOVA area. ->iova_alloc_at(): To allocate IOVA area at specific address. ->iova_free(): To free IOVA area. ->map_page_at(): To map page at specific IOVA. misc: ->iova_get_free_total(): To return how much IOVA is available totally. ->iova_get_free_max(): To return the size of biggest IOVA area. Although NvMap itself will be replaced soon, there are cases for the above API where we need to specify IOVA explicitly. (1) HWAs may require the address for special purpose, like reset vector. (2) IOVA linear mapping: ex: [RFC 5/5] ARM: dma-mapping: Introduce dma_map_linear_attrs() for IOVA linear map (3) To support different heaps. To have allocation and mapping independently. Some of them could be supported with creating different mappings, but currently a device can have a single contiguous mapping, and we cannot specifiy any address inside of a map since all IOVA alloction is done implicitly now. This is the revised version of: http://lists.linaro.org/pipermail/linaro-mm-sig/2012-May/001947.html http://lists.linaro.org/pipermail/linaro-mm-sig/2012-May/001948.html http://lists.linaro.org/pipermail/linaro-mm-sig/2012-May/001949.html Any comment would be really appreciated. Hiroshi Doyu (5): ARM: dma-mapping: New dma_map_ops->iova_get_free_{total,max} functions ARM: dma-mapping: New dma_map_ops->iova_{alloc,free}() functions ARM: dma-mapping: New dma_map_ops->iova_alloc*_at* function ARM: dma-mapping: New dma_map_ops->map_page*_at* function ARM: dma-mapping: Introduce dma_map_linear_attrs() for IOVA linear map arch/arm/include/asm/dma-mapping.h | 55 +++++++++++++ arch/arm/mm/dma-mapping.c | 124 ++++++++++++++++++++++++++++++ include/asm-generic/dma-mapping-common.h | 20 +++++ include/linux/dma-mapping.h | 14 ++++ 4 files changed, 213 insertions(+), 0 deletions(-) -- 1.7.5.4

11 years, 4 months

[PATCH] dma-mapping: fix dma_common_get_sgtable() conditional compilation

by Marek Szyprowski

dma_common_get_sgtable() function doesn't depend on ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY, so it must not be compiled conditionally. Reported-by: Stephen Rothwell <sfr(a)canb.auug.org.au> Signed-off-by: Marek Szyprowski <m.szyprowski(a)samsung.com> --- drivers/base/dma-mapping.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/base/dma-mapping.c b/drivers/base/dma-mapping.c index 3fbedc7..0ce39a3 100644 --- a/drivers/base/dma-mapping.c +++ b/drivers/base/dma-mapping.c @@ -218,6 +218,8 @@ void dmam_release_declared_memory(struct device *dev) } EXPORT_SYMBOL(dmam_release_declared_memory); +#endif + /* * Create scatter-list for the already allocated DMA buffer. */ @@ -236,8 +238,6 @@ int dma_common_get_sgtable(struct device *dev, struct sg_table *sgt, } EXPORT_SYMBOL(dma_common_get_sgtable); -#endif - /* * Create userspace mapping for the DMA-coherent memory. */ -- 1.7.9.5

11 years, 4 months

DMABUF V4L2 patches got merged

by Mauro Carvalho Chehab

Hi all, Today, I finally merged the DMABUF V4L2 patches from Tomasz. The DMABUF allows replacing the old V4L2 Overlay method by something more robust and safer. It was a long road to get them ready for their upstream inclusion, and to be able to test on both embedded and personal computers. Along this weekend, I was able to test it using 4 different test scenarios: - vivi + s5p-tv; - uvcvideo + fimc (m2m) + s5p-tv; - s5k4ecgx + fimc (m2m) + s5p-tv; - uvcvideo + i915. The first 3 tests ran on a Samsung Origen Rev. A board; the 4th one on a notebook, with a Sandy Bridge i5core processor with GPU, and an embedded UVC camera. While testing the s5k4ecgx sensor driver, I also added support for multiplane at libv4l, via its plugin interface: http://git.linuxtv.org/v4l-utils.git/commit/ced1be346fe4f61c864cba9d81f6608… Such tests wouldn't be possible without the help of Linaro and Samsung, with donated me some hardware for the tests, and Ideas on Board for making uvcvideo + i915 driver to work especially for this test. Thank you all for your support! In particular, Sylwester helped me a lot to fix several non-related issues with the Origen board, that was not running with an upstream Kernel. There are a number of patches required to make the Origen board to work with an Upstream Kernel. Also, its sensor driver (s5k4ecgx) was not submitted upstream yet. In order to help others that may need to do similar tests, I added the needed patches on my experimental tree, at branch origen+dmabuf: http://git.linuxtv.org/mchehab/experimental.git/shortlog/refs/heads/origen%… Still missing there are the wireless/bluetooth support. It seems that there are some patches for it already, but they aren't submitted upstream, nor I didn't test they. I expect that Linaro and Samsung will be able to submit real soon the pending patches needed by Origen in time for its addition on 3.8. Thank you all! Mauro

11 years, 5 months

[PATCH] mm: cma: allocate pages from CMA if NR_FREE_PAGES approaches low water mark

by Marek Szyprowski

It has been observed that system tends to keep a lot of CMA free pages even in very high memory pressure use cases. The CMA fallback for movable pages is used very rarely, only when system is completely pruned from MOVABLE pages, what usually means that the out-of-memory even will be triggered very soon. To avoid such situation and make better use of CMA pages, a heuristics is introduced which turns on CMA fallback for movable pages when the real number of free pages (excluding CMA free pages) approaches low water mark. Signed-off-by: Marek Szyprowski <m.szyprowski(a)samsung.com> Reviewed-by: Kyungmin Park <kyungmin.park(a)samsung.com> CC: Michal Nazarewicz <mina86(a)mina86.com> --- mm/page_alloc.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index fcb9719..90b51f3 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1076,6 +1076,15 @@ static struct page *__rmqueue(struct zone *zone, unsigned int order, { struct page *page; +#ifdef CONFIG_CMA + unsigned long nr_free = zone_page_state(zone, NR_FREE_PAGES); + unsigned long nr_cma_free = zone_page_state(zone, NR_FREE_CMA_PAGES); + + if (migratetype == MIGRATE_MOVABLE && nr_cma_free && + nr_free - nr_cma_free < 2 * low_wmark_pages(zone)) + migratetype = MIGRATE_CMA; +#endif /* CONFIG_CMA */ + retry_reserve: page = __rmqueue_smallest(zone, order, migratetype); -- 1.7.9.5

11 years, 5 months

[PATCH] mm: skip watermarks check for already isolated blocks in split_free_page()

by Marek Szyprowski

Since commit 2139cbe627b8 ("cma: fix counting of isolated pages") free pages in isolated pageblocks are not accounted to NR_FREE_PAGES counters, so watermarks check is not required if one operates on a free page in isolated pageblock. Signed-off-by: Marek Szyprowski <m.szyprowski(a)samsung.com> --- mm/page_alloc.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index fd154fe..43ab09f 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1394,10 +1394,12 @@ int capture_free_page(struct page *page, int alloc_order, int migratetype) zone = page_zone(page); order = page_order(page); - /* Obey watermarks as if the page was being allocated */ - watermark = low_wmark_pages(zone) + (1 << order); - if (!zone_watermark_ok(zone, 0, watermark, 0, 0)) - return 0; + if (get_pageblock_migratetype(page) != MIGRATE_ISOLATE) { + /* Obey watermarks as if the page was being allocated */ + watermark = low_wmark_pages(zone) + (1 << order); + if (!zone_watermark_ok(zone, 0, watermark, 0, 0)) + return 0; + } /* Remove page from free list */ list_del(&page->lru); -- 1.7.9.5

11 years, 5 months

[PATCH] mm: cma: WARN if freed memory is still in use

by Marek Szyprowski

Memory return to free_contig_range() must have no other references. Let kernel to complain loudly if page reference count is not equal to 1. Signed-off-by: Marek Szyprowski <m.szyprowski(a)samsung.com> Reviewed-by: Kyungmin Park <kyungmin.park(a)samsung.com> CC: Michal Nazarewicz <mina86(a)mina86.com> --- mm/page_alloc.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 022e4ed..290c2eb 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -5888,8 +5888,13 @@ done: void free_contig_range(unsigned long pfn, unsigned nr_pages) { - for (; nr_pages--; ++pfn) - __free_page(pfn_to_page(pfn)); + struct page *page = pfn_to_page(pfn); + int refcount = nr_pages; + for (; nr_pages--; page++) { + refcount -= page_count(page) == 1; + __free_page(page); + } + WARN(refcount != 0, "some pages are still in use!\n"); } #endif -- 1.7.9.5

11 years, 5 months

[PATCH] mm: use migrate_prep() instead of migrate_prep_local()

by Marek Szyprowski

__alloc_contig_migrate_range() should use all possible ways to get all the pages migrated from the given memory range, so pruning per-cpu lru lists for all CPUs is required, regadless the cost of such operation. Otherwise some pages which got stuck at per-cpu lru list might get missed by migration procedure causing the contiguous allocation to fail. Reported-by: SeongHwan Yoon <sunghwan.yun(a)samsung.com> Signed-off-by: Marek Szyprowski <m.szyprowski(a)samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park(a)samsung.com> --- mm/page_alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 1bfe2b0..fcb9719 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -5677,7 +5677,7 @@ static int __alloc_contig_migrate_range(struct compact_control *cc, unsigned int tries = 0; int ret = 0; - migrate_prep_local(); + migrate_prep(); while (pfn < end || !list_empty(&cc->migratepages)) { if (fatal_signal_pending(current)) { -- 1.7.9.5

11 years, 5 months

[PATCH] mm: remove watermark hacks for CMA

by Marek Szyprowski

Commits 2139cbe627b89 ("cma: fix counting of isolated pages") and d95ea5d18e69951 ("cma: fix watermark checking") introduced a reliable method of free page accounting when memory is being allocated from CMA regions, so the workaround introduced earlier by commit 49f223a9cd96c72 ("mm: trigger page reclaim in alloc_contig_range() to stabilise watermarks") can be finally removed. Signed-off-by: Marek Szyprowski <m.szyprowski(a)samsung.com> --- include/linux/mmzone.h | 9 -------- mm/page_alloc.c | 57 ------------------------------------------------ 2 files changed, 66 deletions(-) diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index c9fcd8f..f010b23 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -63,10 +63,8 @@ enum { #ifdef CONFIG_CMA # define is_migrate_cma(migratetype) unlikely((migratetype) == MIGRATE_CMA) -# define cma_wmark_pages(zone) zone->min_cma_pages #else # define is_migrate_cma(migratetype) false -# define cma_wmark_pages(zone) 0 #endif #define for_each_migratetype_order(order, type) \ @@ -372,13 +370,6 @@ struct zone { /* see spanned/present_pages for more description */ seqlock_t span_seqlock; #endif -#ifdef CONFIG_CMA - /* - * CMA needs to increase watermark levels during the allocation - * process to make sure that the system is not starved. - */ - unsigned long min_cma_pages; -#endif struct free_area free_area[MAX_ORDER]; #ifndef CONFIG_SPARSEMEM diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 43ab09f..5028a18 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -5217,10 +5217,6 @@ static void __setup_per_zone_wmarks(void) zone->watermark[WMARK_LOW] = min_wmark_pages(zone) + (tmp >> 2); zone->watermark[WMARK_HIGH] = min_wmark_pages(zone) + (tmp >> 1); - zone->watermark[WMARK_MIN] += cma_wmark_pages(zone); - zone->watermark[WMARK_LOW] += cma_wmark_pages(zone); - zone->watermark[WMARK_HIGH] += cma_wmark_pages(zone); - setup_zone_migrate_reserve(zone); spin_unlock_irqrestore(&zone->lock, flags); } @@ -5765,54 +5761,6 @@ static int __alloc_contig_migrate_range(struct compact_control *cc, return ret > 0 ? 0 : ret; } -/* - * Update zone's cma pages counter used for watermark level calculation. - */ -static inline void __update_cma_watermarks(struct zone *zone, int count) -{ - unsigned long flags; - spin_lock_irqsave(&zone->lock, flags); - zone->min_cma_pages += count; - spin_unlock_irqrestore(&zone->lock, flags); - setup_per_zone_wmarks(); -} - -/* - * Trigger memory pressure bump to reclaim some pages in order to be able to - * allocate 'count' pages in single page units. Does similar work as - *__alloc_pages_slowpath() function. - */ -static int __reclaim_pages(struct zone *zone, gfp_t gfp_mask, int count) -{ - enum zone_type high_zoneidx = gfp_zone(gfp_mask); - struct zonelist *zonelist = node_zonelist(0, gfp_mask); - int did_some_progress = 0; - int order = 1; - - /* - * Increase level of watermarks to force kswapd do his job - * to stabilise at new watermark level. - */ - __update_cma_watermarks(zone, count); - - /* Obey watermarks as if the page was being allocated */ - while (!zone_watermark_ok(zone, 0, low_wmark_pages(zone), 0, 0)) { - wake_all_kswapd(order, zonelist, high_zoneidx, zone_idx(zone)); - - did_some_progress = __perform_reclaim(gfp_mask, order, zonelist, - NULL); - if (!did_some_progress) { - /* Exhausted what can be done so it's blamo time */ - out_of_memory(zonelist, gfp_mask, order, NULL, false); - } - } - - /* Restore original watermark levels. */ - __update_cma_watermarks(zone, -count); - - return count; -} - /** * alloc_contig_range() -- tries to allocate given range of pages * @start: start PFN to allocate @@ -5921,11 +5869,6 @@ int alloc_contig_range(unsigned long start, unsigned long end, goto done; } - /* - * Reclaim enough pages to make sure that contiguous allocation - * will not starve the system. - */ - __reclaim_pages(zone, GFP_HIGHUSER_MOVABLE, end-start); /* Grab isolated pages from freelists. */ outer_end = isolate_freepages_range(&cc, outer_start, end); -- 1.7.9.5

11 years, 5 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

Linaro-mm-sig November 2012