search for: memory_device_priv

...allow to pin such memory so that it can always be evicted. - * * MEMORY_DEVICE_FS_DAX: * Host memory that has similar access semantics as System RAM i.e. DMA * coherent and supports page pinning. In support of coordinating page @@ -58,7 +51,6 @@ struct vmem_altmap { */ enum memory_type { MEMORY_DEVICE_PRIVATE = 1, - MEMORY_DEVICE_PUBLIC, MEMORY_DEVICE_FS_DAX, MEMORY_DEVICE_PCI_P2PDMA, }; diff --git a/include/linux/mm.h b/include/linux/mm.h index dd0b5f4e1e45..6e4b9be08b13 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -944,7 +944,6 @@ static inline bool put_devmap_managed_page(struc...

...evmap_managed_key); > - > -static inline bool page_is_devmap_managed(struct page *page) > -{ > - if (!static_branch_unlikely(&devmap_managed_key)) > - return false; > - if (!is_zone_device_page(page)) > - return false; > - switch (page->pgmap->type) { > - case MEMORY_DEVICE_PRIVATE: > - case MEMORY_DEVICE_FS_DAX: > - return true; > - default: > - break; > - } > - return false; > -} > - > -void put_devmap_managed_page(struct page *page); > - > -#else /* CONFIG_DEV_PAGEMAP_OPS */ > -static inline bool page_is_devmap_managed(struct page *...

...o free_devmap_managed_page can move to mm/internal.h. > +#ifdef CONFIG_DEV_PAGEMAP_OPS > +static void __put_devmap_managed_page(struct page *page) > +{ > + if (!static_branch_unlikely(&devmap_managed_key)) > + return; > + > + switch (page->pgmap->type) { > + case MEMORY_DEVICE_PRIVATE: > + case MEMORY_DEVICE_FS_DAX: > + free_devmap_managed_page(page); > + break; > + default: > + break; > + } > +} > +#else > +static inline void __put_devmap_managed_page(struct page *page) > +{ > +} > +#endif I think this should be moved to mm/memremap.c...

....h. Good suggestion. >> +#ifdef CONFIG_DEV_PAGEMAP_OPS >> +static void __put_devmap_managed_page(struct page *page) >> +{ >> + if (!static_branch_unlikely(&devmap_managed_key)) >> + return; >> + >> + switch (page->pgmap->type) { >> + case MEMORY_DEVICE_PRIVATE: >> + case MEMORY_DEVICE_FS_DAX: >> + free_devmap_managed_page(page); >> + break; >> + default: >> + break; >> + } >> +} >> +#else >> +static inline void __put_devmap_managed_page(struct page *page) >> +{ >> +} >> +#endif...

...ruct page *page); DECLARE_STATIC_KEY_FALSE(devmap_managed_key); - -static inline bool page_is_devmap_managed(struct page *page) -{ - if (!static_branch_unlikely(&devmap_managed_key)) - return false; - if (!is_zone_device_page(page)) - return false; - switch (page->pgmap->type) { - case MEMORY_DEVICE_PRIVATE: - case MEMORY_DEVICE_FS_DAX: - return true; - default: - break; - } - return false; -} - -void put_devmap_managed_page(struct page *page); - -#else /* CONFIG_DEV_PAGEMAP_OPS */ -static inline bool page_is_devmap_managed(struct page *page) -{ - return false; -} - -static inline void put_devmap...

...o + * MEMORY_DEVICE_FS_DAX, this memory is access via a device-dax + * character device. + * * MEMORY_DEVICE_PCI_P2PDMA: * Device memory residing in a PCI BAR intended for use with Peer-to-Peer * transactions. */ enum memory_type { + /* 0 is reserved to catch uninitialized type fields */ MEMORY_DEVICE_PRIVATE = 1, MEMORY_DEVICE_FS_DAX, + MEMORY_DEVICE_DEVDAX, MEMORY_DEVICE_PCI_P2PDMA, }; diff --git a/kernel/memremap.c b/kernel/memremap.c index 6e1970719dc2..abda62d1e5a3 100644 --- a/kernel/memremap.c +++ b/kernel/memremap.c @@ -157,6 +157,28 @@ void *devm_memremap_pages(struct device *dev, str...

...1 insertions(+) diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c index 79b1202b1c62..67fefb03b9b7 100644 --- a/arch/powerpc/kvm/book3s_hv_uvmem.c +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c @@ -779,6 +779,8 @@ int kvmppc_uvmem_init(void) kvmppc_uvmem_pgmap.type = MEMORY_DEVICE_PRIVATE; kvmppc_uvmem_pgmap.res = *res; kvmppc_uvmem_pgmap.ops = &kvmppc_uvmem_ops; + /* just one global instance: */ + kvmppc_uvmem_pgmap.owner = &kvmppc_uvmem_pgmap; addr = memremap_pages(&kvmppc_uvmem_pgmap, NUMA_NO_NODE); if (IS_ERR(addr)) { ret = PTR_ERR(addr); diff --git a/d...

...de/linux/mm.h index ef360fe70aaf..7a7013c57a4a 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1110,14 +1110,7 @@ static inline bool page_is_devmap_managed(struct page *page) return false; if (!is_zone_device_page(page)) return false; - switch (page->pgmap->type) { - case MEMORY_DEVICE_PRIVATE: - case MEMORY_DEVICE_FS_DAX: - return true; - default: - break; - } - return false; + return page->pgmap->type == MEMORY_DEVICE_FS_DAX; } void put_devmap_managed_page(struct page *page); diff --git a/lib/test_hmm.c b/lib/test_hmm.c index e151a7f10519..bf92a261fa6f 100644 --- a/lib/t...

...ruct page *page); -DECLARE_STATIC_KEY_FALSE(devmap_managed_key); - -static inline bool page_is_devmap_managed(struct page *page) -{ - if (!static_branch_unlikely(&devmap_managed_key)) - return false; - if (!is_zone_device_page(page)) - return false; - switch (page->pgmap->type) { - case MEMORY_DEVICE_PRIVATE: - case MEMORY_DEVICE_FS_DAX: - return true; - default: - break; - } - return false; -} - -void put_devmap_managed_page(struct page *page); - -#else /* CONFIG_DEV_PAGEMAP_OPS */ -static inline bool page_is_devmap_managed(struct page *page) -{ - return false; -} - -static inline void put_devmap...

Matthew Wilcox, Ira Weiny, and others have complained that ZONE_DEVICE struct page reference counting is ugly because they are "free" when the reference count is one instead of zero. This leads to explicit checks for ZONE_DEVICE pages in places like put_page(), GUP, THP splitting, and page migration which have to adjust the expected reference count when determining if the page is

....end = end; mig.src = &src_pfn; mig.dst = &dst_pfn; + mig.dev_private_owner = &kvmppc_uvmem_pgmap; mutex_lock(&kvm->arch.uvmem_lock); /* The requested page is already paged-out, nothing to do */ @@ -779,6 +781,8 @@ int kvmppc_uvmem_init(void) kvmppc_uvmem_pgmap.type = MEMORY_DEVICE_PRIVATE; kvmppc_uvmem_pgmap.res = *res; kvmppc_uvmem_pgmap.ops = &kvmppc_uvmem_ops; + /* just one global instance: */ + kvmppc_uvmem_pgmap.owner = &kvmppc_uvmem_pgmap; addr = memremap_pages(&kvmppc_uvmem_pgmap, NUMA_NO_NODE); if (IS_ERR(addr)) { ret = PTR_ERR(addr); diff --git a/d...

...de/linux/mm.h index d6b8e30dce2e..ac848eeb2a1d 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1110,14 +1110,7 @@ static inline bool page_is_devmap_managed(struct page *page) return false; if (!is_zone_device_page(page)) return false; - switch (page->pgmap->type) { - case MEMORY_DEVICE_PRIVATE: - case MEMORY_DEVICE_FS_DAX: - return true; - default: - break; - } - return false; + return page->pgmap->type == MEMORY_DEVICE_FS_DAX; } void put_devmap_managed_page(struct page *page); diff --git a/lib/test_hmm.c b/lib/test_hmm.c index e151a7f10519..bf92a261fa6f 100644 --- a/lib/t...

...ake progress on the THP migration code in the mean time. > > I think P2P is pretty trivial, given that ZONE_DEVICE pages are used like > a normal memory allocator. DAX is the interesting case, any specific > help that you need with that? There are 4 types of ZONE_DEVICE struct pages: MEMORY_DEVICE_PRIVATE, MEMORY_DEVICE_FS_DAX, MEMORY_DEVICE_GENERIC, and MEMORY_DEVICE_PCI_P2PDMA. Currently, memremap_pages() allocates struct pages for a physical address range with a page_ref_count(page) of one and increments the pgmap->ref per CPU reference count by the number of pages created since each ZONE_...

...ruct page *page); -DECLARE_STATIC_KEY_FALSE(devmap_managed_key); - -static inline bool page_is_devmap_managed(struct page *page) -{ - if (!static_branch_unlikely(&devmap_managed_key)) - return false; - if (!is_zone_device_page(page)) - return false; - switch (page->pgmap->type) { - case MEMORY_DEVICE_PRIVATE: - case MEMORY_DEVICE_FS_DAX: - return true; - default: - break; - } - return false; -} - -void put_devmap_managed_page(struct page *page); - -#else /* CONFIG_DEV_PAGEMAP_OPS */ -static inline bool page_is_devmap_managed(struct page *page) -{ - return false; -} - -static inline void put_devmap...

Hi Dan, Jérôme and Jason, below is a series that cleans up the dev_pagemap interface so that it is more easily usable, which removes the need to wrap it in hmm and thus allowing to kill a lot of code Note: this series is on top of Linux 5.2-rc5 and has some minor conflicts with the hmm tree that are easy to resolve. Diffstat summary: 32 files changed, 361 insertions(+), 1012 deletions(-) Git

...gmap); if (IS_ERR(addr)) { diff --git a/include/linux/memremap.h b/include/linux/memremap.h index f0628660d541..5f7f40875b35 100644 --- a/include/linux/memremap.h +++ b/include/linux/memremap.h @@ -63,39 +63,40 @@ enum memory_type { MEMORY_DEVICE_PCI_P2PDMA, }; -/* - * Additional notes about MEMORY_DEVICE_PRIVATE may be found in - * include/linux/hmm.h and Documentation/vm/hmm.rst. There is also a brief - * explanation in include/linux/memory_hotplug.h. - * - * The page_free() callback is called once the page refcount reaches 1 - * (ZONE_DEVICE pages never reach 0 refcount unless there is a refcount bug....

...gmap); if (IS_ERR(addr)) { diff --git a/include/linux/memremap.h b/include/linux/memremap.h index 1732dea030b2..1cdcfd595770 100644 --- a/include/linux/memremap.h +++ b/include/linux/memremap.h @@ -63,41 +63,45 @@ enum memory_type { MEMORY_DEVICE_PCI_P2PDMA, }; -/* - * Additional notes about MEMORY_DEVICE_PRIVATE may be found in - * include/linux/hmm.h and Documentation/vm/hmm.rst. There is also a brief - * explanation in include/linux/memory_hotplug.h. - * - * The page_free() callback is called once the page refcount reaches 1 - * (ZONE_DEVICE pages never reach 0 refcount unless there is a refcount bug....

On Fri, Nov 06, 2020 at 01:26:50PM -0800, Ralph Campbell wrote: > > On 11/6/20 12:03 AM, Christoph Hellwig wrote: >> I hate the extra pin count magic here. IMHO we really need to finish >> off the series to get rid of the extra references on the ZONE_DEVICE >> pages first. > > First, thanks for the review comments. > > I don't like the extra refcount

...c b/kernel/memremap.c index 6e1970719dc2..6a2dd31a6250 100644 --- a/kernel/memremap.c +++ b/kernel/memremap.c @@ -157,6 +157,33 @@ void *devm_memremap_pages(struct device *dev, struct dev_pagemap *pgmap) pgprot_t pgprot = PAGE_KERNEL; int error, nid, is_ram; + switch (pgmap->type) { + case MEMORY_DEVICE_PRIVATE: + if (!IS_ENABLED(CONFIG_DEVICE_PRIVATE)) { + WARN(1, "Device private memory not supported\n"); + return ERR_PTR(-EINVAL); + } + break; + case MEMORY_DEVICE_PUBLIC: + if (!IS_ENABLED(CONFIG_DEVICE_PUBLIC)) { + WARN(1, "Device public memory not supported\n"); + re...

...> --- a/kernel/memremap.c > +++ b/kernel/memremap.c > @@ -157,6 +157,33 @@ void *devm_memremap_pages(struct device *dev, struct dev_pagemap *pgmap) > pgprot_t pgprot = PAGE_KERNEL; > int error, nid, is_ram; > > + switch (pgmap->type) { > + case MEMORY_DEVICE_PRIVATE: > + if (!IS_ENABLED(CONFIG_DEVICE_PRIVATE)) { > + WARN(1, "Device private memory not supported\n"); > + return ERR_PTR(-EINVAL); > + } > + break; > + case MEMORY_DEVICE_PUBLIC...