search for: __gfp_highmem

From: Michal Hocko <mhocko at suse.com> b92b1b89a33c ("virtio: force vring descriptors to be allocated from lowmem") tried to exclude highmem pages for descriptors so it cleared __GFP_HIGHMEM from a given gfp mask. The patch also cleared __GFP_HIGH which doesn't make much sense for this fix because __GFP_HIGH only controls access to memory reserves and it doesn't have any influence on the zone selection. Some of the call paths use GFP_ATOMIC and dropping __GFP_HIGH will reduce t...

From: Michal Hocko <mhocko at suse.com> b92b1b89a33c ("virtio: force vring descriptors to be allocated from lowmem") tried to exclude highmem pages for descriptors so it cleared __GFP_HIGHMEM from a given gfp mask. The patch also cleared __GFP_HIGH which doesn't make much sense for this fix because __GFP_HIGH only controls access to memory reserves and it doesn't have any influence on the zone selection. Some of the call paths use GFP_ATOMIC and dropping __GFP_HIGH will reduce t...

On Tue, Dec 01, 2015 at 03:32:49PM +0100, Michal Hocko wrote: > From: Michal Hocko <mhocko at suse.com> > > b92b1b89a33c ("virtio: force vring descriptors to be allocated from > lowmem") tried to exclude highmem pages for descriptors so it cleared > __GFP_HIGHMEM from a given gfp mask. The patch also cleared __GFP_HIGH > which doesn't make much sense for this fix because __GFP_HIGH only > controls access to memory reserves and it doesn't have any influence > on the zone selection. Some of the call paths use GFP_ATOMIC and > dropping __GF...

.../zram_drv.c index 370c2f76016d..10f6ff1cf6a0 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -514,7 +514,8 @@ static struct zram_meta *zram_meta_alloc(char *pool_name, u64 disksize) goto out_error; } - meta->mem_pool = zs_create_pool(pool_name, GFP_NOIO | __GFP_HIGHMEM); + meta->mem_pool = zs_create_pool(pool_name, GFP_NOIO|__GFP_HIGHMEM + |__GFP_MOVABLE); if (!meta->mem_pool) { pr_err("Error creating memory pool\n"); goto out_error; diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index b9ff698115a1..a1f47a7b3a99 100644 --- a/mm/zsmalloc.c...

...gfp) > > +{ > > + struct vring_packed_desc *desc; > > + > > + /* > > + * We require lowmem mappings for the descriptors because > > + * otherwise virt_to_phys will give us bogus addresses in the > > + * virtqueue. > > + */ > > + gfp &= ~__GFP_HIGHMEM; > > + > > + desc = kmalloc(total_sg * sizeof(struct vring_packed_desc), gfp); > > Can we simply check vq->packed here to avoid duplicating helpers? Then it would be something like this: static void *alloc_indirect(struct virtqueue *_vq, unsigned int total_sg, gfp_t g...

...gfp) > > +{ > > + struct vring_packed_desc *desc; > > + > > + /* > > + * We require lowmem mappings for the descriptors because > > + * otherwise virt_to_phys will give us bogus addresses in the > > + * virtqueue. > > + */ > > + gfp &= ~__GFP_HIGHMEM; > > + > > + desc = kmalloc(total_sg * sizeof(struct vring_packed_desc), gfp); > > Can we simply check vq->packed here to avoid duplicating helpers? Then it would be something like this: static void *alloc_indirect(struct virtqueue *_vq, unsigned int total_sg, gfp_t g...

...((unsigned long)desc & ~PAGE_MASK); > vq->vring.desc[head].len = i * sizeof(struct vring_desc); Gah, virt_to_phys_harder()? What's the performance effect? If it's negligible, why doesn't virt_to_phys() just do this for us? We do have an alternate solution: masking out __GFP_HIGHMEM from the kmalloc of desc. If it fails, we will fall back to laying out the virtio request directly inside the ring; if it doesn't fit, we'll wait for the device to consume more buffers. > @@ -325,7 +326,7 @@ static int p9_get_mapped_pages(struct virtio_chan *chan, > int count = n...

...((unsigned long)desc & ~PAGE_MASK); > vq->vring.desc[head].len = i * sizeof(struct vring_desc); Gah, virt_to_phys_harder()? What's the performance effect? If it's negligible, why doesn't virt_to_phys() just do this for us? We do have an alternate solution: masking out __GFP_HIGHMEM from the kmalloc of desc. If it fails, we will fall back to laying out the virtio request directly inside the ring; if it doesn't fit, we'll wait for the device to consume more buffers. > @@ -325,7 +326,7 @@ static int p9_get_mapped_pages(struct virtio_chan *chan, > int count = n...

...gt;> + struct vring_packed_desc *desc; >>> + >>> + /* >>> + * We require lowmem mappings for the descriptors because >>> + * otherwise virt_to_phys will give us bogus addresses in the >>> + * virtqueue. >>> + */ >>> + gfp &= ~__GFP_HIGHMEM; >>> + >>> + desc = kmalloc(total_sg * sizeof(struct vring_packed_desc), gfp); >> Can we simply check vq->packed here to avoid duplicating helpers? > Then it would be something like this: > > static void *alloc_indirect(struct virtqueue *_vq, unsigned int total_s...

...============================================== --- linux-4.7-rc7.orig/drivers/virtio/virtio_ring.c 2016-05-09 17:45:04.000000000 +0200 +++ linux-4.7-rc7/drivers/virtio/virtio_ring.c 2016-07-11 17:30:20.000000000 +0200 @@ -237,6 +237,12 @@ static struct vring_desc *alloc_indirect */ gfp &= ~__GFP_HIGHMEM; + /* + * Don't warn if the allocation fails, because the driver + * handles allocation failures gracefully. + */ + gfp |= __GFP_NOWARN; + desc = kmalloc(total_sg * sizeof(struct vring_desc), gfp); if (!desc) return NULL;

...============================================== --- linux-4.7-rc7.orig/drivers/virtio/virtio_ring.c 2016-05-09 17:45:04.000000000 +0200 +++ linux-4.7-rc7/drivers/virtio/virtio_ring.c 2016-07-11 17:30:20.000000000 +0200 @@ -237,6 +237,12 @@ static struct vring_desc *alloc_indirect */ gfp &= ~__GFP_HIGHMEM; + /* + * Don't warn if the allocation fails, because the driver + * handles allocation failures gracefully. + */ + gfp |= __GFP_NOWARN; + desc = kmalloc(total_sg * sizeof(struct vring_desc), gfp); if (!desc) return NULL;

VMware balloon enhancements: adding support for memory compaction, memory shrinker (to prevent OOM) and splitting of refused pages to prevent recurring inflations. Patches 1-2: Support for compaction Patch 3: Support for memory shrinker - disabled by default Patch 4: Split refused pages to improve performance v3->v4: * "get around to" comment [Michael] * Put list_add under page lock

VMware balloon enhancements: adding support for memory compaction, memory shrinker (to prevent OOM) and splitting of refused pages to prevent recurring inflations. Patches 1-2: Support for compaction Patch 3: Support for memory shrinker - disabled by default Patch 4: Split refused pages to improve performance v3->v4: * "get around to" comment [Michael] * Put list_add under page lock

...--- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -961,6 +961,13 @@ static int btree_releasepage(struct page *page, gfp_t gfp_flags) tree = &BTRFS_I(page->mapping->host)->io_tree; map = &BTRFS_I(page->mapping->host)->extent_tree; + /* + * We need to mask out eg. __GFP_HIGHMEM and __GFP_DMA32 as we''re doing + * slab allocation from alloc_extent_state down the callchain where + * it''d hit a BUG_ON as those flags are not allowed. + */ + gfp_flags &= ~GFP_SLAB_BUG_MASK; + ret = try_release_extent_state(map, tree, page, gfp_flags); if (!ret) r...

VMware balloon enhancements: adding support for memory compaction, memory shrinker (to prevent OOM) and splitting of refused pages to prevent recurring inflations. Patches 1-2: Support for compaction Patch 3: Support for memory shrinker - disabled by default Patch 4: Split refused pages to improve performance v2->v3: * Fixing wrong argument type (int->size_t) [Michael] * Fixing a comment

Some virtio device drivers (9p) need to translate high virtual addresses to physical addresses, which are inserted into the virtqueue for processing by userspace. This patch exports the kmap_to_page symbol, so that the affected drivers can be compiled as modules. Signed-off-by: Will Deacon <will.deacon at arm.com> --- mm/highmem.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-)

Some virtio device drivers (9p) need to translate high virtual addresses to physical addresses, which are inserted into the virtqueue for processing by userspace. This patch exports the kmap_to_page symbol, so that the affected drivers can be compiled as modules. Signed-off-by: Will Deacon <will.deacon at arm.com> --- mm/highmem.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-)

...t; > > + > > > > + /* > > > > + * We require lowmem mappings for the descriptors because > > > > + * otherwise virt_to_phys will give us bogus addresses in the > > > > + * virtqueue. > > > > + */ > > > > + gfp &= ~__GFP_HIGHMEM; > > > > + > > > > + desc = kmalloc(total_sg * sizeof(struct vring_packed_desc), gfp); > > > Can we simply check vq->packed here to avoid duplicating helpers? > > Then it would be something like this: > > > > static void *alloc_indirect(struct...

...t; > > + > > > > + /* > > > > + * We require lowmem mappings for the descriptors because > > > > + * otherwise virt_to_phys will give us bogus addresses in the > > > > + * virtqueue. > > > > + */ > > > > + gfp &= ~__GFP_HIGHMEM; > > > > + > > > > + desc = kmalloc(total_sg * sizeof(struct vring_packed_desc), gfp); > > > Can we simply check vq->packed here to avoid duplicating helpers? > > Then it would be something like this: > > > > static void *alloc_indirect(struct...

...f --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 409aeaa49246..34b6b694298c 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -247,8 +247,7 @@ static struct vring_desc *alloc_indirect(struct virtqueue *_vq, * virtqueue. */ gfp &= ~__GFP_HIGHMEM; - - desc = kmalloc(total_sg * sizeof(struct vring_desc), gfp); + desc = kmalloc_array(total_sg, sizeof(*desc), gfp); if (!desc) return NULL; -- 2.11.0