The patch titled Subject: mm: free device private pages have zero refcount has been added to the -mm mm-unstable branch. Its filename is mm-free-device-private-pages-have-zero-refcount.patch This patch will shortly appear at https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-free-device-private-pages-have-zero-refcount.patch This patch will later appear in the mm-unstable branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/process/submit-checklist.rst when testing your code *** The -mm tree is included into linux-next via the mm-everything branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm and is updated there every 2-3 working days ------------------------------------------------------ From: Alistair Popple <apopple@xxxxxxxxxx> Subject: mm: free device private pages have zero refcount Date: Wed, 28 Sep 2022 22:01:16 +1000 Since 27674ef6c73f ("mm: remove the extra ZONE_DEVICE struct page refcount") device private pages have no longer had an extra reference count when the page is in use. However before handing them back to the owning device driver we add an extra reference count such that free pages have a reference count of one. This makes it difficult to tell if a page is free or not because both free and in use pages will have a non-zero refcount. Instead we should return pages to the drivers page allocator with a zero reference count. Kernel code can then safely use kernel functions such as get_page_unless_zero(). Link: https://lkml.kernel.org/r/cf70cf6f8c0bdb8aaebdbfb0d790aea4c683c3c6.1664366292.git-series.apopple@xxxxxxxxxx Signed-off-by: Alistair Popple <apopple@xxxxxxxxxx> Cc: Jason Gunthorpe <jgg@xxxxxxxxxx> Cc: Michael Ellerman <mpe@xxxxxxxxxxxxxx> Cc: Felix Kuehling <Felix.Kuehling@xxxxxxx> Cc: Alex Deucher <alexander.deucher@xxxxxxx> Cc: Christian König <christian.koenig@xxxxxxx> Cc: Ben Skeggs <bskeggs@xxxxxxxxxx> Cc: Lyude Paul <lyude@xxxxxxxxxx> Cc: Ralph Campbell <rcampbell@xxxxxxxxxx> Cc: Alex Sierra <alex.sierra@xxxxxxx> Cc: John Hubbard <jhubbard@xxxxxxxxxx> Cc: Dan Williams <dan.j.williams@xxxxxxxxx> Cc: David Hildenbrand <david@xxxxxxxxxx> Cc: "Huang, Ying" <ying.huang@xxxxxxxxx> Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx> Cc: Yang Shi <shy828301@xxxxxxxxx> Cc: Zi Yan <ziy@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- --- a/arch/powerpc/kvm/book3s_hv_uvmem.c~mm-free-device-private-pages-have-zero-refcount +++ a/arch/powerpc/kvm/book3s_hv_uvmem.c @@ -718,7 +718,7 @@ static struct page *kvmppc_uvmem_get_pag dpage = pfn_to_page(uvmem_pfn); dpage->zone_device_data = pvt; - lock_page(dpage); + zone_device_page_init(dpage); return dpage; out_clear: spin_lock(&kvmppc_uvmem_bitmap_lock); --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c~mm-free-device-private-pages-have-zero-refcount +++ a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c @@ -223,7 +223,7 @@ svm_migrate_get_vram_page(struct svm_ran page = pfn_to_page(pfn); svm_range_bo_ref(prange->svm_bo); page->zone_device_data = prange->svm_bo; - lock_page(page); + zone_device_page_init(page); } static void --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c~mm-free-device-private-pages-have-zero-refcount +++ a/drivers/gpu/drm/nouveau/nouveau_dmem.c @@ -326,7 +326,7 @@ nouveau_dmem_page_alloc_locked(struct no return NULL; } - lock_page(page); + zone_device_page_init(page); return page; } --- a/include/linux/memremap.h~mm-free-device-private-pages-have-zero-refcount +++ a/include/linux/memremap.h @@ -187,6 +187,7 @@ static inline bool folio_is_device_coher } #ifdef CONFIG_ZONE_DEVICE +void zone_device_page_init(struct page *page); void *memremap_pages(struct dev_pagemap *pgmap, int nid); void memunmap_pages(struct dev_pagemap *pgmap); void *devm_memremap_pages(struct device *dev, struct dev_pagemap *pgmap); --- a/lib/test_hmm.c~mm-free-device-private-pages-have-zero-refcount +++ a/lib/test_hmm.c @@ -627,8 +627,8 @@ static struct page *dmirror_devmem_alloc goto error; } + zone_device_page_init(dpage); dpage->zone_device_data = rpage; - lock_page(dpage); return dpage; error: --- a/mm/memremap.c~mm-free-device-private-pages-have-zero-refcount +++ a/mm/memremap.c @@ -505,8 +505,17 @@ void free_zone_device_page(struct page * /* * Reset the page count to 1 to prepare for handing out the page again. */ + if (page->pgmap->type != MEMORY_DEVICE_PRIVATE && + page->pgmap->type != MEMORY_DEVICE_COHERENT) + set_page_count(page, 1); +} + +void zone_device_page_init(struct page *page) +{ set_page_count(page, 1); + lock_page(page); } +EXPORT_SYMBOL_GPL(zone_device_page_init); #ifdef CONFIG_FS_DAX bool __put_devmap_managed_page_refs(struct page *page, int refs) --- a/mm/page_alloc.c~mm-free-device-private-pages-have-zero-refcount +++ a/mm/page_alloc.c @@ -6811,6 +6811,14 @@ static void __ref __init_zone_device_pag set_pageblock_migratetype(page, MIGRATE_MOVABLE); cond_resched(); } + + /* + * ZONE_DEVICE pages are released directly to the driver page allocator + * which will set the page count to 1 when allocating the page. + */ + if (pgmap->type == MEMORY_DEVICE_PRIVATE || + pgmap->type == MEMORY_DEVICE_COHERENT) + set_page_count(page, 0); } /* _ Patches currently in -mm which might be from apopple@xxxxxxxxxx are mm-memoryc-fix-race-when-faulting-a-device-private-page.patch mm-free-device-private-pages-have-zero-refcount.patch mm-memremapc-take-a-pgmap-reference-on-page-allocation.patch mm-migrate_devicec-refactor-migrate_vma-and-migrate_deivce_coherent_page.patch mm-migrate_devicec-add-migrate_device_range.patch nouveau-dmem-refactor-nouveau_dmem_fault_copy_one.patch nouveau-dmem-evict-device-private-memory-during-release.patch hmm-tests-add-test-for-migrate_device_range.patch