[v5] Enable THP support in drm_pagemap

[PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Francois Dugast 3 weeks, 4 days ago

From: Matthew Brost <matthew.brost@intel.com>

Reinitialize metadata for large zone device private folios in
zone_device_page_init prior to creating a higher-order zone device
private folio. This step is necessary when the folio’s order changes
dynamically between zone_device_page_init calls to avoid building a
corrupt folio. As part of the metadata reinitialization, the dev_pagemap
must be passed in from the caller because the pgmap stored in the folio
page may have been overwritten with a compound head.

Cc: Zi Yan <ziy@nvidia.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: adhavan Srinivasan <maddy@linux.ibm.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona@ffwll.ch>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Lyude Paul <lyude@redhat.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: David Hildenbrand <david@kernel.org>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Leon Romanovsky <leon@kernel.org>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Balbir Singh <balbirs@nvidia.com>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: kvm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: amd-gfx@lists.freedesktop.org
Cc: dri-devel@lists.freedesktop.org
Cc: nouveau@lists.freedesktop.org
Cc: linux-mm@kvack.org
Cc: linux-cxl@vger.kernel.org
Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
---
 arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
 drivers/gpu/drm/drm_pagemap.c            |  2 +-
 drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
 include/linux/memremap.h                 |  9 ++++++---
 lib/test_hmm.c                           |  4 +++-
 mm/memremap.c                            | 20 +++++++++++++++++++-
 7 files changed, 32 insertions(+), 9 deletions(-)

diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
index e5000bef90f2..7cf9310de0ec 100644
--- a/arch/powerpc/kvm/book3s_hv_uvmem.c
+++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
@@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
 
 	dpage = pfn_to_page(uvmem_pfn);
 	dpage->zone_device_data = pvt;
-	zone_device_page_init(dpage, 0);
+	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
 	return dpage;
 out_clear:
 	spin_lock(&kvmppc_uvmem_bitmap_lock);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index af53e796ea1b..6ada7b4af7c6 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
 	page = pfn_to_page(pfn);
 	svm_range_bo_ref(prange->svm_bo);
 	page->zone_device_data = prange->svm_bo;
-	zone_device_page_init(page, 0);
+	zone_device_page_init(page, page_pgmap(page), 0);
 }
 
 static void
diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
index 03ee39a761a4..c497726b0147 100644
--- a/drivers/gpu/drm/drm_pagemap.c
+++ b/drivers/gpu/drm/drm_pagemap.c
@@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
 					struct drm_pagemap_zdd *zdd)
 {
 	page->zone_device_data = drm_pagemap_zdd_get(zdd);
-	zone_device_page_init(page, 0);
+	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
 }
 
 /**
diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
index 58071652679d..3d8031296eed 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
@@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
 			order = ilog2(DMEM_CHUNK_NPAGES);
 	}
 
-	zone_device_folio_init(folio, order);
+	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
 	return page;
 }
 
diff --git a/include/linux/memremap.h b/include/linux/memremap.h
index 713ec0435b48..e3c2ccf872a8 100644
--- a/include/linux/memremap.h
+++ b/include/linux/memremap.h
@@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
 }
 
 #ifdef CONFIG_ZONE_DEVICE
-void zone_device_page_init(struct page *page, unsigned int order);
+void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
+			   unsigned int order);
 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);
@@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
 
 unsigned long memremap_compat_align(void);
 
-static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
+static inline void zone_device_folio_init(struct folio *folio,
+					  struct dev_pagemap *pgmap,
+					  unsigned int order)
 {
-	zone_device_page_init(&folio->page, order);
+	zone_device_page_init(&folio->page, pgmap, order);
 	if (order)
 		folio_set_large_rmappable(folio);
 }
diff --git a/lib/test_hmm.c b/lib/test_hmm.c
index 8af169d3873a..455a6862ae50 100644
--- a/lib/test_hmm.c
+++ b/lib/test_hmm.c
@@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
 			goto error;
 	}
 
-	zone_device_folio_init(page_folio(dpage), order);
+	zone_device_folio_init(page_folio(dpage),
+			       page_pgmap(folio_page(page_folio(dpage), 0)),
+			       order);
 	dpage->zone_device_data = rpage;
 	return dpage;
 
diff --git a/mm/memremap.c b/mm/memremap.c
index 63c6ab4fdf08..6f46ab14662b 100644
--- a/mm/memremap.c
+++ b/mm/memremap.c
@@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
 	}
 }
 
-void zone_device_page_init(struct page *page, unsigned int order)
+void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
+			   unsigned int order)
 {
+	struct page *new_page = page;
+	unsigned int i;
+
 	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
 
+	for (i = 0; i < (1UL << order); ++i, ++new_page) {
+		struct folio *new_folio = (struct folio *)new_page;
+
+		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
+#ifdef NR_PAGES_IN_LARGE_FOLIO
+		((struct folio *)(new_page - 1))->_nr_pages = 0;
+#endif
+		new_folio->mapping = NULL;
+		new_folio->pgmap = pgmap;	/* Also clear compound head */
+		new_folio->share = 0;   /* fsdax only, unused for device private */
+		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
+		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
+	}
+
 	/*
 	 * Drivers shouldn't be allocating pages after calling
 	 * memunmap_pages().
-- 
2.43.0

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Rodrigo Vivi 3 weeks, 3 days ago

On Wed, Jan 14, 2026 at 08:19:52PM +0100, Francois Dugast wrote:
> From: Matthew Brost <matthew.brost@intel.com>
> 
> Reinitialize metadata for large zone device private folios in
> zone_device_page_init prior to creating a higher-order zone device
> private folio. This step is necessary when the folio’s order changes
> dynamically between zone_device_page_init calls to avoid building a
> corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> must be passed in from the caller because the pgmap stored in the folio
> page may have been overwritten with a compound head.
> 
> Cc: Zi Yan <ziy@nvidia.com>
> Cc: Alistair Popple <apopple@nvidia.com>
> Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> Cc: Nicholas Piggin <npiggin@gmail.com>
> Cc: Michael Ellerman <mpe@ellerman.id.au>
> Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> Cc: Alex Deucher <alexander.deucher@amd.com>
> Cc: "Christian König" <christian.koenig@amd.com>
> Cc: David Airlie <airlied@gmail.com>
> Cc: Simona Vetter <simona@ffwll.ch>
> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> Cc: Maxime Ripard <mripard@kernel.org>
> Cc: Thomas Zimmermann <tzimmermann@suse.de>
> Cc: Lyude Paul <lyude@redhat.com>
> Cc: Danilo Krummrich <dakr@kernel.org>
> Cc: David Hildenbrand <david@kernel.org>
> Cc: Oscar Salvador <osalvador@suse.de>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Jason Gunthorpe <jgg@ziepe.ca>
> Cc: Leon Romanovsky <leon@kernel.org>
> Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> Cc: Vlastimil Babka <vbabka@suse.cz>
> Cc: Mike Rapoport <rppt@kernel.org>
> Cc: Suren Baghdasaryan <surenb@google.com>
> Cc: Michal Hocko <mhocko@suse.com>
> Cc: Balbir Singh <balbirs@nvidia.com>
> Cc: linuxppc-dev@lists.ozlabs.org
> Cc: kvm@vger.kernel.org
> Cc: linux-kernel@vger.kernel.org
> Cc: amd-gfx@lists.freedesktop.org
> Cc: dri-devel@lists.freedesktop.org
> Cc: nouveau@lists.freedesktop.org
> Cc: linux-mm@kvack.org
> Cc: linux-cxl@vger.kernel.org
> Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> ---
>  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
>  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-

AMD folks, ack to get this patch through drm-misc-next or even perhaps some mm
related tree?

>  drivers/gpu/drm/drm_pagemap.c            |  2 +-
>  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-

Nouveau folks, ack as well?

>  include/linux/memremap.h                 |  9 ++++++---
>  lib/test_hmm.c                           |  4 +++-
>  mm/memremap.c                            | 20 +++++++++++++++++++-
>  7 files changed, 32 insertions(+), 9 deletions(-)
> 
> diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> index e5000bef90f2..7cf9310de0ec 100644
> --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
>  
>  	dpage = pfn_to_page(uvmem_pfn);
>  	dpage->zone_device_data = pvt;
> -	zone_device_page_init(dpage, 0);
> +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
>  	return dpage;
>  out_clear:
>  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> index af53e796ea1b..6ada7b4af7c6 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
>  	page = pfn_to_page(pfn);
>  	svm_range_bo_ref(prange->svm_bo);
>  	page->zone_device_data = prange->svm_bo;
> -	zone_device_page_init(page, 0);
> +	zone_device_page_init(page, page_pgmap(page), 0);
>  }
>  
>  static void
> diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> index 03ee39a761a4..c497726b0147 100644
> --- a/drivers/gpu/drm/drm_pagemap.c
> +++ b/drivers/gpu/drm/drm_pagemap.c
> @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
>  					struct drm_pagemap_zdd *zdd)
>  {
>  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> -	zone_device_page_init(page, 0);
> +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
>  }
>  
>  /**
> diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> index 58071652679d..3d8031296eed 100644
> --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
>  			order = ilog2(DMEM_CHUNK_NPAGES);
>  	}
>  
> -	zone_device_folio_init(folio, order);
> +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
>  	return page;
>  }
>  
> diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> index 713ec0435b48..e3c2ccf872a8 100644
> --- a/include/linux/memremap.h
> +++ b/include/linux/memremap.h
> @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
>  }
>  
>  #ifdef CONFIG_ZONE_DEVICE
> -void zone_device_page_init(struct page *page, unsigned int order);
> +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> +			   unsigned int order);
>  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);
> @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
>  
>  unsigned long memremap_compat_align(void);
>  
> -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> +static inline void zone_device_folio_init(struct folio *folio,
> +					  struct dev_pagemap *pgmap,
> +					  unsigned int order)
>  {
> -	zone_device_page_init(&folio->page, order);
> +	zone_device_page_init(&folio->page, pgmap, order);
>  	if (order)
>  		folio_set_large_rmappable(folio);
>  }
> diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> index 8af169d3873a..455a6862ae50 100644
> --- a/lib/test_hmm.c
> +++ b/lib/test_hmm.c
> @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
>  			goto error;
>  	}
>  
> -	zone_device_folio_init(page_folio(dpage), order);
> +	zone_device_folio_init(page_folio(dpage),
> +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> +			       order);
>  	dpage->zone_device_data = rpage;
>  	return dpage;
>  
> diff --git a/mm/memremap.c b/mm/memremap.c
> index 63c6ab4fdf08..6f46ab14662b 100644
> --- a/mm/memremap.c
> +++ b/mm/memremap.c
> @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
>  	}
>  }
>  
> -void zone_device_page_init(struct page *page, unsigned int order)
> +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> +			   unsigned int order)
>  {
> +	struct page *new_page = page;
> +	unsigned int i;
> +
>  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
>  
> +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> +		struct folio *new_folio = (struct folio *)new_page;
> +
> +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> +#ifdef NR_PAGES_IN_LARGE_FOLIO
> +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> +#endif
> +		new_folio->mapping = NULL;
> +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> +		new_folio->share = 0;   /* fsdax only, unused for device private */
> +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> +	}
> +
>  	/*
>  	 * Drivers shouldn't be allocating pages after calling
>  	 * memunmap_pages().
> -- 
> 2.43.0
>

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Kuehling, Felix 3 weeks, 3 days ago

On 2026-01-16 11:43, Rodrigo Vivi wrote:
> On Wed, Jan 14, 2026 at 08:19:52PM +0100, Francois Dugast wrote:
>> From: Matthew Brost <matthew.brost@intel.com>
>>
>> Reinitialize metadata for large zone device private folios in
>> zone_device_page_init prior to creating a higher-order zone device
>> private folio. This step is necessary when the folio’s order changes
>> dynamically between zone_device_page_init calls to avoid building a
>> corrupt folio. As part of the metadata reinitialization, the dev_pagemap
>> must be passed in from the caller because the pgmap stored in the folio
>> page may have been overwritten with a compound head.
>>
>> Cc: Zi Yan <ziy@nvidia.com>
>> Cc: Alistair Popple <apopple@nvidia.com>
>> Cc: adhavan Srinivasan <maddy@linux.ibm.com>
>> Cc: Nicholas Piggin <npiggin@gmail.com>
>> Cc: Michael Ellerman <mpe@ellerman.id.au>
>> Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
>> Cc: Felix Kuehling <Felix.Kuehling@amd.com>
>> Cc: Alex Deucher <alexander.deucher@amd.com>
>> Cc: "Christian König" <christian.koenig@amd.com>
>> Cc: David Airlie <airlied@gmail.com>
>> Cc: Simona Vetter <simona@ffwll.ch>
>> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
>> Cc: Maxime Ripard <mripard@kernel.org>
>> Cc: Thomas Zimmermann <tzimmermann@suse.de>
>> Cc: Lyude Paul <lyude@redhat.com>
>> Cc: Danilo Krummrich <dakr@kernel.org>
>> Cc: David Hildenbrand <david@kernel.org>
>> Cc: Oscar Salvador <osalvador@suse.de>
>> Cc: Andrew Morton <akpm@linux-foundation.org>
>> Cc: Jason Gunthorpe <jgg@ziepe.ca>
>> Cc: Leon Romanovsky <leon@kernel.org>
>> Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
>> Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
>> Cc: Vlastimil Babka <vbabka@suse.cz>
>> Cc: Mike Rapoport <rppt@kernel.org>
>> Cc: Suren Baghdasaryan <surenb@google.com>
>> Cc: Michal Hocko <mhocko@suse.com>
>> Cc: Balbir Singh <balbirs@nvidia.com>
>> Cc: linuxppc-dev@lists.ozlabs.org
>> Cc: kvm@vger.kernel.org
>> Cc: linux-kernel@vger.kernel.org
>> Cc: amd-gfx@lists.freedesktop.org
>> Cc: dri-devel@lists.freedesktop.org
>> Cc: nouveau@lists.freedesktop.org
>> Cc: linux-mm@kvack.org
>> Cc: linux-cxl@vger.kernel.org
>> Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
>> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
>> Signed-off-by: Francois Dugast <francois.dugast@intel.com>
>> ---
>>   arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
>>   drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
> AMD folks, ack to get this patch through drm-misc-next or even perhaps some mm
> related tree?

The kfd_migrate change looks straight-forward enough for me. We don't 
use large device pages yet, so I guess the changes in 
zone_device_page_init should be safe for us. Feel free to add  my

Acked-by: Felix Kuehling <felix.kuehling@amd.com>

I have no objections to merging this through drm-misc-next. @Alex, how 
will this flow back into our amd-staging-drm-next branch? I guess we'll 
get it on the next branch rebase. There should be no rush as I don't 
think we're affected by the bug being fixed here.

Thanks,
   Felix


>
>>   drivers/gpu/drm/drm_pagemap.c            |  2 +-
>>   drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
> Nouveau folks, ack as well?
>
>>   include/linux/memremap.h                 |  9 ++++++---
>>   lib/test_hmm.c                           |  4 +++-
>>   mm/memremap.c                            | 20 +++++++++++++++++++-
>>   7 files changed, 32 insertions(+), 9 deletions(-)
>>
>> diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
>> index e5000bef90f2..7cf9310de0ec 100644
>> --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
>> +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
>> @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
>>   
>>   	dpage = pfn_to_page(uvmem_pfn);
>>   	dpage->zone_device_data = pvt;
>> -	zone_device_page_init(dpage, 0);
>> +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
>>   	return dpage;
>>   out_clear:
>>   	spin_lock(&kvmppc_uvmem_bitmap_lock);
>> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
>> index af53e796ea1b..6ada7b4af7c6 100644
>> --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
>> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
>> @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
>>   	page = pfn_to_page(pfn);
>>   	svm_range_bo_ref(prange->svm_bo);
>>   	page->zone_device_data = prange->svm_bo;
>> -	zone_device_page_init(page, 0);
>> +	zone_device_page_init(page, page_pgmap(page), 0);
>>   }
>>   
>>   static void
>> diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
>> index 03ee39a761a4..c497726b0147 100644
>> --- a/drivers/gpu/drm/drm_pagemap.c
>> +++ b/drivers/gpu/drm/drm_pagemap.c
>> @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
>>   					struct drm_pagemap_zdd *zdd)
>>   {
>>   	page->zone_device_data = drm_pagemap_zdd_get(zdd);
>> -	zone_device_page_init(page, 0);
>> +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
>>   }
>>   
>>   /**
>> diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
>> index 58071652679d..3d8031296eed 100644
>> --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
>> +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
>> @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
>>   			order = ilog2(DMEM_CHUNK_NPAGES);
>>   	}
>>   
>> -	zone_device_folio_init(folio, order);
>> +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
>>   	return page;
>>   }
>>   
>> diff --git a/include/linux/memremap.h b/include/linux/memremap.h
>> index 713ec0435b48..e3c2ccf872a8 100644
>> --- a/include/linux/memremap.h
>> +++ b/include/linux/memremap.h
>> @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
>>   }
>>   
>>   #ifdef CONFIG_ZONE_DEVICE
>> -void zone_device_page_init(struct page *page, unsigned int order);
>> +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
>> +			   unsigned int order);
>>   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);
>> @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
>>   
>>   unsigned long memremap_compat_align(void);
>>   
>> -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
>> +static inline void zone_device_folio_init(struct folio *folio,
>> +					  struct dev_pagemap *pgmap,
>> +					  unsigned int order)
>>   {
>> -	zone_device_page_init(&folio->page, order);
>> +	zone_device_page_init(&folio->page, pgmap, order);
>>   	if (order)
>>   		folio_set_large_rmappable(folio);
>>   }
>> diff --git a/lib/test_hmm.c b/lib/test_hmm.c
>> index 8af169d3873a..455a6862ae50 100644
>> --- a/lib/test_hmm.c
>> +++ b/lib/test_hmm.c
>> @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
>>   			goto error;
>>   	}
>>   
>> -	zone_device_folio_init(page_folio(dpage), order);
>> +	zone_device_folio_init(page_folio(dpage),
>> +			       page_pgmap(folio_page(page_folio(dpage), 0)),
>> +			       order);
>>   	dpage->zone_device_data = rpage;
>>   	return dpage;
>>   
>> diff --git a/mm/memremap.c b/mm/memremap.c
>> index 63c6ab4fdf08..6f46ab14662b 100644
>> --- a/mm/memremap.c
>> +++ b/mm/memremap.c
>> @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
>>   	}
>>   }
>>   
>> -void zone_device_page_init(struct page *page, unsigned int order)
>> +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
>> +			   unsigned int order)
>>   {
>> +	struct page *new_page = page;
>> +	unsigned int i;
>> +
>>   	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
>>   
>> +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
>> +		struct folio *new_folio = (struct folio *)new_page;
>> +
>> +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
>> +#ifdef NR_PAGES_IN_LARGE_FOLIO
>> +		((struct folio *)(new_page - 1))->_nr_pages = 0;
>> +#endif
>> +		new_folio->mapping = NULL;
>> +		new_folio->pgmap = pgmap;	/* Also clear compound head */
>> +		new_folio->share = 0;   /* fsdax only, unused for device private */
>> +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
>> +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
>> +	}
>> +
>>   	/*
>>   	 * Drivers shouldn't be allocating pages after calling
>>   	 * memunmap_pages().
>> -- 
>> 2.43.0
>>

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Alistair Popple 3 weeks, 4 days ago

On 2026-01-15 at 06:19 +1100, Francois Dugast <francois.dugast@intel.com> wrote...
> From: Matthew Brost <matthew.brost@intel.com>
> 
> Reinitialize metadata for large zone device private folios in
> zone_device_page_init prior to creating a higher-order zone device
> private folio. This step is necessary when the folio’s order changes
> dynamically between zone_device_page_init calls to avoid building a
> corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> must be passed in from the caller because the pgmap stored in the folio
> page may have been overwritten with a compound head.

Thanks for fixing, a couple of minor comments below.

> Cc: Zi Yan <ziy@nvidia.com>
> Cc: Alistair Popple <apopple@nvidia.com>
> Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> Cc: Nicholas Piggin <npiggin@gmail.com>
> Cc: Michael Ellerman <mpe@ellerman.id.au>
> Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> Cc: Alex Deucher <alexander.deucher@amd.com>
> Cc: "Christian König" <christian.koenig@amd.com>
> Cc: David Airlie <airlied@gmail.com>
> Cc: Simona Vetter <simona@ffwll.ch>
> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> Cc: Maxime Ripard <mripard@kernel.org>
> Cc: Thomas Zimmermann <tzimmermann@suse.de>
> Cc: Lyude Paul <lyude@redhat.com>
> Cc: Danilo Krummrich <dakr@kernel.org>
> Cc: David Hildenbrand <david@kernel.org>
> Cc: Oscar Salvador <osalvador@suse.de>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Jason Gunthorpe <jgg@ziepe.ca>
> Cc: Leon Romanovsky <leon@kernel.org>
> Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> Cc: Vlastimil Babka <vbabka@suse.cz>
> Cc: Mike Rapoport <rppt@kernel.org>
> Cc: Suren Baghdasaryan <surenb@google.com>
> Cc: Michal Hocko <mhocko@suse.com>
> Cc: Balbir Singh <balbirs@nvidia.com>
> Cc: linuxppc-dev@lists.ozlabs.org
> Cc: kvm@vger.kernel.org
> Cc: linux-kernel@vger.kernel.org
> Cc: amd-gfx@lists.freedesktop.org
> Cc: dri-devel@lists.freedesktop.org
> Cc: nouveau@lists.freedesktop.org
> Cc: linux-mm@kvack.org
> Cc: linux-cxl@vger.kernel.org
> Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> ---
>  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
>  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
>  drivers/gpu/drm/drm_pagemap.c            |  2 +-
>  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
>  include/linux/memremap.h                 |  9 ++++++---
>  lib/test_hmm.c                           |  4 +++-
>  mm/memremap.c                            | 20 +++++++++++++++++++-
>  7 files changed, 32 insertions(+), 9 deletions(-)
> 
> diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> index e5000bef90f2..7cf9310de0ec 100644
> --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
>  
>  	dpage = pfn_to_page(uvmem_pfn);
>  	dpage->zone_device_data = pvt;
> -	zone_device_page_init(dpage, 0);
> +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
>  	return dpage;
>  out_clear:
>  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> index af53e796ea1b..6ada7b4af7c6 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
>  	page = pfn_to_page(pfn);
>  	svm_range_bo_ref(prange->svm_bo);
>  	page->zone_device_data = prange->svm_bo;
> -	zone_device_page_init(page, 0);
> +	zone_device_page_init(page, page_pgmap(page), 0);
>  }
>  
>  static void
> diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> index 03ee39a761a4..c497726b0147 100644
> --- a/drivers/gpu/drm/drm_pagemap.c
> +++ b/drivers/gpu/drm/drm_pagemap.c
> @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
>  					struct drm_pagemap_zdd *zdd)
>  {
>  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> -	zone_device_page_init(page, 0);
> +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
>  }
>  
>  /**
> diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> index 58071652679d..3d8031296eed 100644
> --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
>  			order = ilog2(DMEM_CHUNK_NPAGES);
>  	}
>  
> -	zone_device_folio_init(folio, order);
> +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
>  	return page;
>  }
>  
> diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> index 713ec0435b48..e3c2ccf872a8 100644
> --- a/include/linux/memremap.h
> +++ b/include/linux/memremap.h
> @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
>  }
>  
>  #ifdef CONFIG_ZONE_DEVICE
> -void zone_device_page_init(struct page *page, unsigned int order);
> +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> +			   unsigned int order);
>  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);
> @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
>  
>  unsigned long memremap_compat_align(void);
>  
> -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> +static inline void zone_device_folio_init(struct folio *folio,
> +					  struct dev_pagemap *pgmap,
> +					  unsigned int order)
>  {
> -	zone_device_page_init(&folio->page, order);
> +	zone_device_page_init(&folio->page, pgmap, order);
>  	if (order)
>  		folio_set_large_rmappable(folio);
>  }
> diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> index 8af169d3873a..455a6862ae50 100644
> --- a/lib/test_hmm.c
> +++ b/lib/test_hmm.c
> @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
>  			goto error;
>  	}
>  
> -	zone_device_folio_init(page_folio(dpage), order);
> +	zone_device_folio_init(page_folio(dpage),
> +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> +			       order);
>  	dpage->zone_device_data = rpage;
>  	return dpage;
>  
> diff --git a/mm/memremap.c b/mm/memremap.c
> index 63c6ab4fdf08..6f46ab14662b 100644
> --- a/mm/memremap.c
> +++ b/mm/memremap.c
> @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
>  	}
>  }
>  
> -void zone_device_page_init(struct page *page, unsigned int order)
> +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> +			   unsigned int order)
>  {
> +	struct page *new_page = page;
> +	unsigned int i;
> +
>  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
>  
> +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> +		struct folio *new_folio = (struct folio *)new_page;
> +
> +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */

This seems odd to me, mainly due to the "magic" number. Why not just clear
the flags entirely? Or at least explicitly just clear the flags you care about
which would remove the need for the comment and  should let you use the proper
PageFlag functions.

> +#ifdef NR_PAGES_IN_LARGE_FOLIO
> +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> +#endif
> +		new_folio->mapping = NULL;
> +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> +		new_folio->share = 0;   /* fsdax only, unused for device private */

It would be nice if the FS DAX code actually used this as well. Is there a
reason that change was dropped from the series?

> +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> +	}
> +
>  	/*
>  	 * Drivers shouldn't be allocating pages after calling
>  	 * memunmap_pages().
> -- 
> 2.43.0
>

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Matthew Brost 3 weeks, 4 days ago

On Thu, Jan 15, 2026 at 04:27:26PM +1100, Alistair Popple wrote:
> On 2026-01-15 at 06:19 +1100, Francois Dugast <francois.dugast@intel.com> wrote...
> > From: Matthew Brost <matthew.brost@intel.com>
> > 
> > Reinitialize metadata for large zone device private folios in
> > zone_device_page_init prior to creating a higher-order zone device
> > private folio. This step is necessary when the folio’s order changes
> > dynamically between zone_device_page_init calls to avoid building a
> > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > must be passed in from the caller because the pgmap stored in the folio
> > page may have been overwritten with a compound head.
> 
> Thanks for fixing, a couple of minor comments below.
> 
> > Cc: Zi Yan <ziy@nvidia.com>
> > Cc: Alistair Popple <apopple@nvidia.com>
> > Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> > Cc: Nicholas Piggin <npiggin@gmail.com>
> > Cc: Michael Ellerman <mpe@ellerman.id.au>
> > Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> > Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> > Cc: Alex Deucher <alexander.deucher@amd.com>
> > Cc: "Christian König" <christian.koenig@amd.com>
> > Cc: David Airlie <airlied@gmail.com>
> > Cc: Simona Vetter <simona@ffwll.ch>
> > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > Cc: Maxime Ripard <mripard@kernel.org>
> > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > Cc: Lyude Paul <lyude@redhat.com>
> > Cc: Danilo Krummrich <dakr@kernel.org>
> > Cc: David Hildenbrand <david@kernel.org>
> > Cc: Oscar Salvador <osalvador@suse.de>
> > Cc: Andrew Morton <akpm@linux-foundation.org>
> > Cc: Jason Gunthorpe <jgg@ziepe.ca>
> > Cc: Leon Romanovsky <leon@kernel.org>
> > Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> > Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> > Cc: Vlastimil Babka <vbabka@suse.cz>
> > Cc: Mike Rapoport <rppt@kernel.org>
> > Cc: Suren Baghdasaryan <surenb@google.com>
> > Cc: Michal Hocko <mhocko@suse.com>
> > Cc: Balbir Singh <balbirs@nvidia.com>
> > Cc: linuxppc-dev@lists.ozlabs.org
> > Cc: kvm@vger.kernel.org
> > Cc: linux-kernel@vger.kernel.org
> > Cc: amd-gfx@lists.freedesktop.org
> > Cc: dri-devel@lists.freedesktop.org
> > Cc: nouveau@lists.freedesktop.org
> > Cc: linux-mm@kvack.org
> > Cc: linux-cxl@vger.kernel.org
> > Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> > ---
> >  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
> >  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
> >  drivers/gpu/drm/drm_pagemap.c            |  2 +-
> >  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
> >  include/linux/memremap.h                 |  9 ++++++---
> >  lib/test_hmm.c                           |  4 +++-
> >  mm/memremap.c                            | 20 +++++++++++++++++++-
> >  7 files changed, 32 insertions(+), 9 deletions(-)
> > 
> > diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > index e5000bef90f2..7cf9310de0ec 100644
> > --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> > +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
> >  
> >  	dpage = pfn_to_page(uvmem_pfn);
> >  	dpage->zone_device_data = pvt;
> > -	zone_device_page_init(dpage, 0);
> > +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
> >  	return dpage;
> >  out_clear:
> >  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > index af53e796ea1b..6ada7b4af7c6 100644
> > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
> >  	page = pfn_to_page(pfn);
> >  	svm_range_bo_ref(prange->svm_bo);
> >  	page->zone_device_data = prange->svm_bo;
> > -	zone_device_page_init(page, 0);
> > +	zone_device_page_init(page, page_pgmap(page), 0);
> >  }
> >  
> >  static void
> > diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> > index 03ee39a761a4..c497726b0147 100644
> > --- a/drivers/gpu/drm/drm_pagemap.c
> > +++ b/drivers/gpu/drm/drm_pagemap.c
> > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> >  					struct drm_pagemap_zdd *zdd)
> >  {
> >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > -	zone_device_page_init(page, 0);
> > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> >  }
> >  
> >  /**
> > diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > index 58071652679d..3d8031296eed 100644
> > --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
> >  			order = ilog2(DMEM_CHUNK_NPAGES);
> >  	}
> >  
> > -	zone_device_folio_init(folio, order);
> > +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
> >  	return page;
> >  }
> >  
> > diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> > index 713ec0435b48..e3c2ccf872a8 100644
> > --- a/include/linux/memremap.h
> > +++ b/include/linux/memremap.h
> > @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
> >  }
> >  
> >  #ifdef CONFIG_ZONE_DEVICE
> > -void zone_device_page_init(struct page *page, unsigned int order);
> > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > +			   unsigned int order);
> >  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);
> > @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
> >  
> >  unsigned long memremap_compat_align(void);
> >  
> > -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> > +static inline void zone_device_folio_init(struct folio *folio,
> > +					  struct dev_pagemap *pgmap,
> > +					  unsigned int order)
> >  {
> > -	zone_device_page_init(&folio->page, order);
> > +	zone_device_page_init(&folio->page, pgmap, order);
> >  	if (order)
> >  		folio_set_large_rmappable(folio);
> >  }
> > diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> > index 8af169d3873a..455a6862ae50 100644
> > --- a/lib/test_hmm.c
> > +++ b/lib/test_hmm.c
> > @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
> >  			goto error;
> >  	}
> >  
> > -	zone_device_folio_init(page_folio(dpage), order);
> > +	zone_device_folio_init(page_folio(dpage),
> > +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> > +			       order);
> >  	dpage->zone_device_data = rpage;
> >  	return dpage;
> >  
> > diff --git a/mm/memremap.c b/mm/memremap.c
> > index 63c6ab4fdf08..6f46ab14662b 100644
> > --- a/mm/memremap.c
> > +++ b/mm/memremap.c
> > @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
> >  	}
> >  }
> >  
> > -void zone_device_page_init(struct page *page, unsigned int order)
> > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > +			   unsigned int order)
> >  {
> > +	struct page *new_page = page;
> > +	unsigned int i;
> > +
> >  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> >  
> > +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > +		struct folio *new_folio = (struct folio *)new_page;
> > +
> > +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> 
> This seems odd to me, mainly due to the "magic" number. Why not just clear
> the flags entirely? Or at least explicitly just clear the flags you care about
> which would remove the need for the comment and  should let you use the proper
> PageFlag functions.
> 

I'm copying this from folio_reset_order [1]. My paranoia about touching
anything related to struct page is high, so I did the same thing
folio_reset_order does here.

[1] https://elixir.bootlin.com/linux/v6.18.5/source/include/linux/mm.h#L1075

> > +#ifdef NR_PAGES_IN_LARGE_FOLIO
> > +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> > +#endif
> > +		new_folio->mapping = NULL;
> > +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> > +		new_folio->share = 0;   /* fsdax only, unused for device private */
> 
> It would be nice if the FS DAX code actually used this as well. Is there a
> reason that change was dropped from the series?
> 

I don't have a test platform for FS DAX. In prior revisions, I was just
moving existing FS DAX code to a helper, which I felt confident about.

This revision is slightly different, and I don't feel comfortable
modifying FS DAX code without a test platform. I agree we should update
FS DAX, but that should be done in a follow-up with coordinated testing.

Matt 

> > +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> > +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> > +	}
> > +
> >  	/*
> >  	 * Drivers shouldn't be allocating pages after calling
> >  	 * memunmap_pages().
> > -- 
> > 2.43.0
> >

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Matthew Brost 3 weeks, 4 days ago

On Wed, Jan 14, 2026 at 09:57:31PM -0800, Matthew Brost wrote:
> On Thu, Jan 15, 2026 at 04:27:26PM +1100, Alistair Popple wrote:
> > On 2026-01-15 at 06:19 +1100, Francois Dugast <francois.dugast@intel.com> wrote...
> > > From: Matthew Brost <matthew.brost@intel.com>
> > > 
> > > Reinitialize metadata for large zone device private folios in
> > > zone_device_page_init prior to creating a higher-order zone device
> > > private folio. This step is necessary when the folio’s order changes
> > > dynamically between zone_device_page_init calls to avoid building a
> > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > must be passed in from the caller because the pgmap stored in the folio
> > > page may have been overwritten with a compound head.
> > 
> > Thanks for fixing, a couple of minor comments below.
> > 
> > > Cc: Zi Yan <ziy@nvidia.com>
> > > Cc: Alistair Popple <apopple@nvidia.com>
> > > Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> > > Cc: Nicholas Piggin <npiggin@gmail.com>
> > > Cc: Michael Ellerman <mpe@ellerman.id.au>
> > > Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> > > Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> > > Cc: Alex Deucher <alexander.deucher@amd.com>
> > > Cc: "Christian König" <christian.koenig@amd.com>
> > > Cc: David Airlie <airlied@gmail.com>
> > > Cc: Simona Vetter <simona@ffwll.ch>
> > > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > > Cc: Maxime Ripard <mripard@kernel.org>
> > > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > > Cc: Lyude Paul <lyude@redhat.com>
> > > Cc: Danilo Krummrich <dakr@kernel.org>
> > > Cc: David Hildenbrand <david@kernel.org>
> > > Cc: Oscar Salvador <osalvador@suse.de>
> > > Cc: Andrew Morton <akpm@linux-foundation.org>
> > > Cc: Jason Gunthorpe <jgg@ziepe.ca>
> > > Cc: Leon Romanovsky <leon@kernel.org>
> > > Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> > > Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> > > Cc: Vlastimil Babka <vbabka@suse.cz>
> > > Cc: Mike Rapoport <rppt@kernel.org>
> > > Cc: Suren Baghdasaryan <surenb@google.com>
> > > Cc: Michal Hocko <mhocko@suse.com>
> > > Cc: Balbir Singh <balbirs@nvidia.com>
> > > Cc: linuxppc-dev@lists.ozlabs.org
> > > Cc: kvm@vger.kernel.org
> > > Cc: linux-kernel@vger.kernel.org
> > > Cc: amd-gfx@lists.freedesktop.org
> > > Cc: dri-devel@lists.freedesktop.org
> > > Cc: nouveau@lists.freedesktop.org
> > > Cc: linux-mm@kvack.org
> > > Cc: linux-cxl@vger.kernel.org
> > > Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> > > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > > Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> > > ---
> > >  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
> > >  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
> > >  drivers/gpu/drm/drm_pagemap.c            |  2 +-
> > >  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
> > >  include/linux/memremap.h                 |  9 ++++++---
> > >  lib/test_hmm.c                           |  4 +++-
> > >  mm/memremap.c                            | 20 +++++++++++++++++++-
> > >  7 files changed, 32 insertions(+), 9 deletions(-)
> > > 
> > > diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > index e5000bef90f2..7cf9310de0ec 100644
> > > --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
> > >  
> > >  	dpage = pfn_to_page(uvmem_pfn);
> > >  	dpage->zone_device_data = pvt;
> > > -	zone_device_page_init(dpage, 0);
> > > +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
> > >  	return dpage;
> > >  out_clear:
> > >  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > index af53e796ea1b..6ada7b4af7c6 100644
> > > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
> > >  	page = pfn_to_page(pfn);
> > >  	svm_range_bo_ref(prange->svm_bo);
> > >  	page->zone_device_data = prange->svm_bo;
> > > -	zone_device_page_init(page, 0);
> > > +	zone_device_page_init(page, page_pgmap(page), 0);
> > >  }
> > >  
> > >  static void
> > > diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> > > index 03ee39a761a4..c497726b0147 100644
> > > --- a/drivers/gpu/drm/drm_pagemap.c
> > > +++ b/drivers/gpu/drm/drm_pagemap.c
> > > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> > >  					struct drm_pagemap_zdd *zdd)
> > >  {
> > >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > > -	zone_device_page_init(page, 0);
> > > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> > >  }
> > >  
> > >  /**
> > > diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > index 58071652679d..3d8031296eed 100644
> > > --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
> > >  			order = ilog2(DMEM_CHUNK_NPAGES);
> > >  	}
> > >  
> > > -	zone_device_folio_init(folio, order);
> > > +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
> > >  	return page;
> > >  }
> > >  
> > > diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> > > index 713ec0435b48..e3c2ccf872a8 100644
> > > --- a/include/linux/memremap.h
> > > +++ b/include/linux/memremap.h
> > > @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
> > >  }
> > >  
> > >  #ifdef CONFIG_ZONE_DEVICE
> > > -void zone_device_page_init(struct page *page, unsigned int order);
> > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > +			   unsigned int order);
> > >  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);
> > > @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
> > >  
> > >  unsigned long memremap_compat_align(void);
> > >  
> > > -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> > > +static inline void zone_device_folio_init(struct folio *folio,
> > > +					  struct dev_pagemap *pgmap,
> > > +					  unsigned int order)
> > >  {
> > > -	zone_device_page_init(&folio->page, order);
> > > +	zone_device_page_init(&folio->page, pgmap, order);
> > >  	if (order)
> > >  		folio_set_large_rmappable(folio);
> > >  }
> > > diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> > > index 8af169d3873a..455a6862ae50 100644
> > > --- a/lib/test_hmm.c
> > > +++ b/lib/test_hmm.c
> > > @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
> > >  			goto error;
> > >  	}
> > >  
> > > -	zone_device_folio_init(page_folio(dpage), order);
> > > +	zone_device_folio_init(page_folio(dpage),
> > > +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> > > +			       order);
> > >  	dpage->zone_device_data = rpage;
> > >  	return dpage;
> > >  
> > > diff --git a/mm/memremap.c b/mm/memremap.c
> > > index 63c6ab4fdf08..6f46ab14662b 100644
> > > --- a/mm/memremap.c
> > > +++ b/mm/memremap.c
> > > @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
> > >  	}
> > >  }
> > >  
> > > -void zone_device_page_init(struct page *page, unsigned int order)
> > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > +			   unsigned int order)
> > >  {
> > > +	struct page *new_page = page;
> > > +	unsigned int i;
> > > +
> > >  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> > >  
> > > +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > > +		struct folio *new_folio = (struct folio *)new_page;
> > > +
> > > +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> > 
> > This seems odd to me, mainly due to the "magic" number. Why not just clear
> > the flags entirely? Or at least explicitly just clear the flags you care about
> > which would remove the need for the comment and  should let you use the proper
> > PageFlag functions.
> > 
> 
> I'm copying this from folio_reset_order [1]. My paranoia about touching
> anything related to struct page is high, so I did the same thing
> folio_reset_order does here.
> 
> [1] https://elixir.bootlin.com/linux/v6.18.5/source/include/linux/mm.h#L1075
> 

This immediately hangs my first SVM test...

diff --git a/mm/memremap.c b/mm/memremap.c
index 6f46ab14662b..ef8c56876cf5 100644
--- a/mm/memremap.c
+++ b/mm/memremap.c
@@ -488,7 +488,7 @@ void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
        for (i = 0; i < (1UL << order); ++i, ++new_page) {
                struct folio *new_folio = (struct folio *)new_page;

-               new_page->flags.f &= ~0xffUL;   /* Clear possible order, page head */
+               new_page->flags.f = 0;
 #ifdef NR_PAGES_IN_LARGE_FOLIO
                ((struct folio *)(new_page - 1))->_nr_pages = 0;
 #endif

I can walk through exactly which flags need to be cleared, but my
feeling is that likely any flag that the order field overloads and can
possibly encode should be cleared—so bits 0–7 based on the existing
code.

How about in a follow-up we normalize setting / clearing the order flag
field with a #define and an inline helper?

Matt

> > > +#ifdef NR_PAGES_IN_LARGE_FOLIO
> > > +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> > > +#endif
> > > +		new_folio->mapping = NULL;
> > > +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> > > +		new_folio->share = 0;   /* fsdax only, unused for device private */
> > 
> > It would be nice if the FS DAX code actually used this as well. Is there a
> > reason that change was dropped from the series?
> > 
> 
> I don't have a test platform for FS DAX. In prior revisions, I was just
> moving existing FS DAX code to a helper, which I felt confident about.
> 
> This revision is slightly different, and I don't feel comfortable
> modifying FS DAX code without a test platform. I agree we should update
> FS DAX, but that should be done in a follow-up with coordinated testing.
> 
> Matt 
> 
> > > +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> > > +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> > > +	}
> > > +
> > >  	/*
> > >  	 * Drivers shouldn't be allocating pages after calling
> > >  	 * memunmap_pages().
> > > -- 
> > > 2.43.0
> > >

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Vlastimil Babka 3 weeks, 3 days ago

On 1/15/26 07:18, Matthew Brost wrote:
> On Wed, Jan 14, 2026 at 09:57:31PM -0800, Matthew Brost wrote:
>> On Thu, Jan 15, 2026 at 04:27:26PM +1100, Alistair Popple wrote:
>> > On 2026-01-15 at 06:19 +1100, Francois Dugast <francois.dugast@intel.com> wrote...
>> > > From: Matthew Brost <matthew.brost@intel.com>
>> > > 
>> > > Reinitialize metadata for large zone device private folios in
>> > > zone_device_page_init prior to creating a higher-order zone device
>> > > private folio. This step is necessary when the folio’s order changes
>> > > dynamically between zone_device_page_init calls to avoid building a
>> > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
>> > > must be passed in from the caller because the pgmap stored in the folio
>> > > page may have been overwritten with a compound head.
>> > 
>> > Thanks for fixing, a couple of minor comments below.
>> > 
>> > > Cc: Zi Yan <ziy@nvidia.com>
>> > > Cc: Alistair Popple <apopple@nvidia.com>
>> > > Cc: adhavan Srinivasan <maddy@linux.ibm.com>
>> > > Cc: Nicholas Piggin <npiggin@gmail.com>
>> > > Cc: Michael Ellerman <mpe@ellerman.id.au>
>> > > Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
>> > > Cc: Felix Kuehling <Felix.Kuehling@amd.com>
>> > > Cc: Alex Deucher <alexander.deucher@amd.com>
>> > > Cc: "Christian König" <christian.koenig@amd.com>
>> > > Cc: David Airlie <airlied@gmail.com>
>> > > Cc: Simona Vetter <simona@ffwll.ch>
>> > > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
>> > > Cc: Maxime Ripard <mripard@kernel.org>
>> > > Cc: Thomas Zimmermann <tzimmermann@suse.de>
>> > > Cc: Lyude Paul <lyude@redhat.com>
>> > > Cc: Danilo Krummrich <dakr@kernel.org>
>> > > Cc: David Hildenbrand <david@kernel.org>
>> > > Cc: Oscar Salvador <osalvador@suse.de>
>> > > Cc: Andrew Morton <akpm@linux-foundation.org>
>> > > Cc: Jason Gunthorpe <jgg@ziepe.ca>
>> > > Cc: Leon Romanovsky <leon@kernel.org>
>> > > Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
>> > > Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
>> > > Cc: Vlastimil Babka <vbabka@suse.cz>
>> > > Cc: Mike Rapoport <rppt@kernel.org>
>> > > Cc: Suren Baghdasaryan <surenb@google.com>
>> > > Cc: Michal Hocko <mhocko@suse.com>
>> > > Cc: Balbir Singh <balbirs@nvidia.com>
>> > > Cc: linuxppc-dev@lists.ozlabs.org
>> > > Cc: kvm@vger.kernel.org
>> > > Cc: linux-kernel@vger.kernel.org
>> > > Cc: amd-gfx@lists.freedesktop.org
>> > > Cc: dri-devel@lists.freedesktop.org
>> > > Cc: nouveau@lists.freedesktop.org
>> > > Cc: linux-mm@kvack.org
>> > > Cc: linux-cxl@vger.kernel.org
>> > > Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
>> > > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
>> > > Signed-off-by: Francois Dugast <francois.dugast@intel.com>
>> > > ---
>> > >  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
>> > >  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
>> > >  drivers/gpu/drm/drm_pagemap.c            |  2 +-
>> > >  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
>> > >  include/linux/memremap.h                 |  9 ++++++---
>> > >  lib/test_hmm.c                           |  4 +++-
>> > >  mm/memremap.c                            | 20 +++++++++++++++++++-
>> > >  7 files changed, 32 insertions(+), 9 deletions(-)
>> > > 
>> > > diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
>> > > index e5000bef90f2..7cf9310de0ec 100644
>> > > --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
>> > > +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
>> > > @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
>> > >  
>> > >  	dpage = pfn_to_page(uvmem_pfn);
>> > >  	dpage->zone_device_data = pvt;
>> > > -	zone_device_page_init(dpage, 0);
>> > > +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
>> > >  	return dpage;
>> > >  out_clear:
>> > >  	spin_lock(&kvmppc_uvmem_bitmap_lock);
>> > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
>> > > index af53e796ea1b..6ada7b4af7c6 100644
>> > > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
>> > > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
>> > > @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
>> > >  	page = pfn_to_page(pfn);
>> > >  	svm_range_bo_ref(prange->svm_bo);
>> > >  	page->zone_device_data = prange->svm_bo;
>> > > -	zone_device_page_init(page, 0);
>> > > +	zone_device_page_init(page, page_pgmap(page), 0);
>> > >  }
>> > >  
>> > >  static void
>> > > diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
>> > > index 03ee39a761a4..c497726b0147 100644
>> > > --- a/drivers/gpu/drm/drm_pagemap.c
>> > > +++ b/drivers/gpu/drm/drm_pagemap.c
>> > > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
>> > >  					struct drm_pagemap_zdd *zdd)
>> > >  {
>> > >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
>> > > -	zone_device_page_init(page, 0);
>> > > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
>> > >  }
>> > >  
>> > >  /**
>> > > diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
>> > > index 58071652679d..3d8031296eed 100644
>> > > --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
>> > > +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
>> > > @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
>> > >  			order = ilog2(DMEM_CHUNK_NPAGES);
>> > >  	}
>> > >  
>> > > -	zone_device_folio_init(folio, order);
>> > > +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
>> > >  	return page;
>> > >  }
>> > >  
>> > > diff --git a/include/linux/memremap.h b/include/linux/memremap.h
>> > > index 713ec0435b48..e3c2ccf872a8 100644
>> > > --- a/include/linux/memremap.h
>> > > +++ b/include/linux/memremap.h
>> > > @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
>> > >  }
>> > >  
>> > >  #ifdef CONFIG_ZONE_DEVICE
>> > > -void zone_device_page_init(struct page *page, unsigned int order);
>> > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
>> > > +			   unsigned int order);
>> > >  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);
>> > > @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
>> > >  
>> > >  unsigned long memremap_compat_align(void);
>> > >  
>> > > -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
>> > > +static inline void zone_device_folio_init(struct folio *folio,
>> > > +					  struct dev_pagemap *pgmap,
>> > > +					  unsigned int order)
>> > >  {
>> > > -	zone_device_page_init(&folio->page, order);
>> > > +	zone_device_page_init(&folio->page, pgmap, order);
>> > >  	if (order)
>> > >  		folio_set_large_rmappable(folio);
>> > >  }
>> > > diff --git a/lib/test_hmm.c b/lib/test_hmm.c
>> > > index 8af169d3873a..455a6862ae50 100644
>> > > --- a/lib/test_hmm.c
>> > > +++ b/lib/test_hmm.c
>> > > @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
>> > >  			goto error;
>> > >  	}
>> > >  
>> > > -	zone_device_folio_init(page_folio(dpage), order);
>> > > +	zone_device_folio_init(page_folio(dpage),
>> > > +			       page_pgmap(folio_page(page_folio(dpage), 0)),
>> > > +			       order);
>> > >  	dpage->zone_device_data = rpage;
>> > >  	return dpage;
>> > >  
>> > > diff --git a/mm/memremap.c b/mm/memremap.c
>> > > index 63c6ab4fdf08..6f46ab14662b 100644
>> > > --- a/mm/memremap.c
>> > > +++ b/mm/memremap.c
>> > > @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
>> > >  	}
>> > >  }
>> > >  
>> > > -void zone_device_page_init(struct page *page, unsigned int order)
>> > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
>> > > +			   unsigned int order)
>> > >  {
>> > > +	struct page *new_page = page;
>> > > +	unsigned int i;
>> > > +
>> > >  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
>> > >  
>> > > +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
>> > > +		struct folio *new_folio = (struct folio *)new_page;
>> > > +
>> > > +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
>> > 
>> > This seems odd to me, mainly due to the "magic" number. Why not just clear
>> > the flags entirely? Or at least explicitly just clear the flags you care about
>> > which would remove the need for the comment and  should let you use the proper
>> > PageFlag functions.
>> > 
>> 
>> I'm copying this from folio_reset_order [1]. My paranoia about touching
>> anything related to struct page is high, so I did the same thing
>> folio_reset_order does here.
>> 
>> [1] https://elixir.bootlin.com/linux/v6.18.5/source/include/linux/mm.h#L1075
>> 
> 
> This immediately hangs my first SVM test...
> 
> diff --git a/mm/memremap.c b/mm/memremap.c
> index 6f46ab14662b..ef8c56876cf5 100644
> --- a/mm/memremap.c
> +++ b/mm/memremap.c
> @@ -488,7 +488,7 @@ void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
>         for (i = 0; i < (1UL << order); ++i, ++new_page) {
>                 struct folio *new_folio = (struct folio *)new_page;
> 
> -               new_page->flags.f &= ~0xffUL;   /* Clear possible order, page head */
> +               new_page->flags.f = 0;
>  #ifdef NR_PAGES_IN_LARGE_FOLIO
>                 ((struct folio *)(new_page - 1))->_nr_pages = 0;
>  #endif

the flags field includes also zone and node id, so clearing that will break
the pages's node+zone association, that's probably why...

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Alistair Popple 3 weeks, 4 days ago

On 2026-01-15 at 17:18 +1100, Matthew Brost <matthew.brost@intel.com> wrote...
> On Wed, Jan 14, 2026 at 09:57:31PM -0800, Matthew Brost wrote:
> > On Thu, Jan 15, 2026 at 04:27:26PM +1100, Alistair Popple wrote:
> > > On 2026-01-15 at 06:19 +1100, Francois Dugast <francois.dugast@intel.com> wrote...
> > > > From: Matthew Brost <matthew.brost@intel.com>
> > > > 
> > > > Reinitialize metadata for large zone device private folios in
> > > > zone_device_page_init prior to creating a higher-order zone device
> > > > private folio. This step is necessary when the folio’s order changes
> > > > dynamically between zone_device_page_init calls to avoid building a
> > > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > > must be passed in from the caller because the pgmap stored in the folio
> > > > page may have been overwritten with a compound head.
> > > 
> > > Thanks for fixing, a couple of minor comments below.
> > > 
> > > > Cc: Zi Yan <ziy@nvidia.com>
> > > > Cc: Alistair Popple <apopple@nvidia.com>
> > > > Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> > > > Cc: Nicholas Piggin <npiggin@gmail.com>
> > > > Cc: Michael Ellerman <mpe@ellerman.id.au>
> > > > Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> > > > Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> > > > Cc: Alex Deucher <alexander.deucher@amd.com>
> > > > Cc: "Christian König" <christian.koenig@amd.com>
> > > > Cc: David Airlie <airlied@gmail.com>
> > > > Cc: Simona Vetter <simona@ffwll.ch>
> > > > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > > > Cc: Maxime Ripard <mripard@kernel.org>
> > > > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > > > Cc: Lyude Paul <lyude@redhat.com>
> > > > Cc: Danilo Krummrich <dakr@kernel.org>
> > > > Cc: David Hildenbrand <david@kernel.org>
> > > > Cc: Oscar Salvador <osalvador@suse.de>
> > > > Cc: Andrew Morton <akpm@linux-foundation.org>
> > > > Cc: Jason Gunthorpe <jgg@ziepe.ca>
> > > > Cc: Leon Romanovsky <leon@kernel.org>
> > > > Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> > > > Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> > > > Cc: Vlastimil Babka <vbabka@suse.cz>
> > > > Cc: Mike Rapoport <rppt@kernel.org>
> > > > Cc: Suren Baghdasaryan <surenb@google.com>
> > > > Cc: Michal Hocko <mhocko@suse.com>
> > > > Cc: Balbir Singh <balbirs@nvidia.com>
> > > > Cc: linuxppc-dev@lists.ozlabs.org
> > > > Cc: kvm@vger.kernel.org
> > > > Cc: linux-kernel@vger.kernel.org
> > > > Cc: amd-gfx@lists.freedesktop.org
> > > > Cc: dri-devel@lists.freedesktop.org
> > > > Cc: nouveau@lists.freedesktop.org
> > > > Cc: linux-mm@kvack.org
> > > > Cc: linux-cxl@vger.kernel.org
> > > > Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> > > > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > > > Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> > > > ---
> > > >  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
> > > >  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
> > > >  drivers/gpu/drm/drm_pagemap.c            |  2 +-
> > > >  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
> > > >  include/linux/memremap.h                 |  9 ++++++---
> > > >  lib/test_hmm.c                           |  4 +++-
> > > >  mm/memremap.c                            | 20 +++++++++++++++++++-
> > > >  7 files changed, 32 insertions(+), 9 deletions(-)
> > > > 
> > > > diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > index e5000bef90f2..7cf9310de0ec 100644
> > > > --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
> > > >  
> > > >  	dpage = pfn_to_page(uvmem_pfn);
> > > >  	dpage->zone_device_data = pvt;
> > > > -	zone_device_page_init(dpage, 0);
> > > > +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
> > > >  	return dpage;
> > > >  out_clear:
> > > >  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> > > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > index af53e796ea1b..6ada7b4af7c6 100644
> > > > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
> > > >  	page = pfn_to_page(pfn);
> > > >  	svm_range_bo_ref(prange->svm_bo);
> > > >  	page->zone_device_data = prange->svm_bo;
> > > > -	zone_device_page_init(page, 0);
> > > > +	zone_device_page_init(page, page_pgmap(page), 0);
> > > >  }
> > > >  
> > > >  static void
> > > > diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> > > > index 03ee39a761a4..c497726b0147 100644
> > > > --- a/drivers/gpu/drm/drm_pagemap.c
> > > > +++ b/drivers/gpu/drm/drm_pagemap.c
> > > > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> > > >  					struct drm_pagemap_zdd *zdd)
> > > >  {
> > > >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > > > -	zone_device_page_init(page, 0);
> > > > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> > > >  }
> > > >  
> > > >  /**
> > > > diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > index 58071652679d..3d8031296eed 100644
> > > > --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
> > > >  			order = ilog2(DMEM_CHUNK_NPAGES);
> > > >  	}
> > > >  
> > > > -	zone_device_folio_init(folio, order);
> > > > +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
> > > >  	return page;
> > > >  }
> > > >  
> > > > diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> > > > index 713ec0435b48..e3c2ccf872a8 100644
> > > > --- a/include/linux/memremap.h
> > > > +++ b/include/linux/memremap.h
> > > > @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
> > > >  }
> > > >  
> > > >  #ifdef CONFIG_ZONE_DEVICE
> > > > -void zone_device_page_init(struct page *page, unsigned int order);
> > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > +			   unsigned int order);
> > > >  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);
> > > > @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
> > > >  
> > > >  unsigned long memremap_compat_align(void);
> > > >  
> > > > -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> > > > +static inline void zone_device_folio_init(struct folio *folio,
> > > > +					  struct dev_pagemap *pgmap,
> > > > +					  unsigned int order)
> > > >  {
> > > > -	zone_device_page_init(&folio->page, order);
> > > > +	zone_device_page_init(&folio->page, pgmap, order);
> > > >  	if (order)
> > > >  		folio_set_large_rmappable(folio);
> > > >  }
> > > > diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> > > > index 8af169d3873a..455a6862ae50 100644
> > > > --- a/lib/test_hmm.c
> > > > +++ b/lib/test_hmm.c
> > > > @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
> > > >  			goto error;
> > > >  	}
> > > >  
> > > > -	zone_device_folio_init(page_folio(dpage), order);
> > > > +	zone_device_folio_init(page_folio(dpage),
> > > > +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> > > > +			       order);
> > > >  	dpage->zone_device_data = rpage;
> > > >  	return dpage;
> > > >  
> > > > diff --git a/mm/memremap.c b/mm/memremap.c
> > > > index 63c6ab4fdf08..6f46ab14662b 100644
> > > > --- a/mm/memremap.c
> > > > +++ b/mm/memremap.c
> > > > @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
> > > >  	}
> > > >  }
> > > >  
> > > > -void zone_device_page_init(struct page *page, unsigned int order)
> > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > +			   unsigned int order)
> > > >  {
> > > > +	struct page *new_page = page;
> > > > +	unsigned int i;
> > > > +
> > > >  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> > > >  
> > > > +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > > > +		struct folio *new_folio = (struct folio *)new_page;
> > > > +
> > > > +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> > > 
> > > This seems odd to me, mainly due to the "magic" number. Why not just clear
> > > the flags entirely? Or at least explicitly just clear the flags you care about
> > > which would remove the need for the comment and  should let you use the proper
> > > PageFlag functions.
> > > 
> > 
> > I'm copying this from folio_reset_order [1]. My paranoia about touching
> > anything related to struct page is high, so I did the same thing
> > folio_reset_order does here.

So why not just use folio_reset_order() below?

> > 
> > [1] https://elixir.bootlin.com/linux/v6.18.5/source/include/linux/mm.h#L1075
> > 
> 
> This immediately hangs my first SVM test...
> 
> diff --git a/mm/memremap.c b/mm/memremap.c
> index 6f46ab14662b..ef8c56876cf5 100644
> --- a/mm/memremap.c
> +++ b/mm/memremap.c
> @@ -488,7 +488,7 @@ void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
>         for (i = 0; i < (1UL << order); ++i, ++new_page) {
>                 struct folio *new_folio = (struct folio *)new_page;
> 
> -               new_page->flags.f &= ~0xffUL;   /* Clear possible order, page head */
> +               new_page->flags.f = 0;
>  #ifdef NR_PAGES_IN_LARGE_FOLIO
>                 ((struct folio *)(new_page - 1))->_nr_pages = 0;

This seems wrong to me - I saw your reply to Balbir but for an order-0 page
isn't this going to access a completely different, possibly already allocated,
page?

>  #endif
> 
> I can walk through exactly which flags need to be cleared, but my
> feeling is that likely any flag that the order field overloads and can
> possibly encode should be cleared—so bits 0–7 based on the existing
> code.
> 
> How about in a follow-up we normalize setting / clearing the order flag
> field with a #define and an inline helper?

Ie: Would something like the following work:

		ClearPageHead(new_page);
		clear_compound_head(new_page);
		folio_reset_order(new_folio);

Which would also deal with setting _nr_pages.

> Matt
> 
> > > > +#ifdef NR_PAGES_IN_LARGE_FOLIO
> > > > +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> > > > +#endif
> > > > +		new_folio->mapping = NULL;
> > > > +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> > > > +		new_folio->share = 0;   /* fsdax only, unused for device private */
> > > 
> > > It would be nice if the FS DAX code actually used this as well. Is there a
> > > reason that change was dropped from the series?
> > > 
> > 
> > I don't have a test platform for FS DAX. In prior revisions, I was just
> > moving existing FS DAX code to a helper, which I felt confident about.
> > 
> > This revision is slightly different, and I don't feel comfortable
> > modifying FS DAX code without a test platform. I agree we should update
> > FS DAX, but that should be done in a follow-up with coordinated testing.

Fair enough, I figured something like this might be your answer :-) You
could update it and ask people with access to such a system to test it though
(unfortunately my setup has bit-rotted beyond repair).

But I'm ok leaving to for a future change.

> > 
> > Matt 
> > 
> > > > +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> > > > +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> > > > +	}
> > > > +
> > > >  	/*
> > > >  	 * Drivers shouldn't be allocating pages after calling
> > > >  	 * memunmap_pages().
> > > > -- 
> > > > 2.43.0
> > > >

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Matthew Brost 3 weeks, 4 days ago

On Thu, Jan 15, 2026 at 06:07:08PM +1100, Alistair Popple wrote:
> On 2026-01-15 at 17:18 +1100, Matthew Brost <matthew.brost@intel.com> wrote...
> > On Wed, Jan 14, 2026 at 09:57:31PM -0800, Matthew Brost wrote:
> > > On Thu, Jan 15, 2026 at 04:27:26PM +1100, Alistair Popple wrote:
> > > > On 2026-01-15 at 06:19 +1100, Francois Dugast <francois.dugast@intel.com> wrote...
> > > > > From: Matthew Brost <matthew.brost@intel.com>
> > > > > 
> > > > > Reinitialize metadata for large zone device private folios in
> > > > > zone_device_page_init prior to creating a higher-order zone device
> > > > > private folio. This step is necessary when the folio’s order changes
> > > > > dynamically between zone_device_page_init calls to avoid building a
> > > > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > > > must be passed in from the caller because the pgmap stored in the folio
> > > > > page may have been overwritten with a compound head.
> > > > 
> > > > Thanks for fixing, a couple of minor comments below.
> > > > 
> > > > > Cc: Zi Yan <ziy@nvidia.com>
> > > > > Cc: Alistair Popple <apopple@nvidia.com>
> > > > > Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> > > > > Cc: Nicholas Piggin <npiggin@gmail.com>
> > > > > Cc: Michael Ellerman <mpe@ellerman.id.au>
> > > > > Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> > > > > Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> > > > > Cc: Alex Deucher <alexander.deucher@amd.com>
> > > > > Cc: "Christian König" <christian.koenig@amd.com>
> > > > > Cc: David Airlie <airlied@gmail.com>
> > > > > Cc: Simona Vetter <simona@ffwll.ch>
> > > > > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > > > > Cc: Maxime Ripard <mripard@kernel.org>
> > > > > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > > > > Cc: Lyude Paul <lyude@redhat.com>
> > > > > Cc: Danilo Krummrich <dakr@kernel.org>
> > > > > Cc: David Hildenbrand <david@kernel.org>
> > > > > Cc: Oscar Salvador <osalvador@suse.de>
> > > > > Cc: Andrew Morton <akpm@linux-foundation.org>
> > > > > Cc: Jason Gunthorpe <jgg@ziepe.ca>
> > > > > Cc: Leon Romanovsky <leon@kernel.org>
> > > > > Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> > > > > Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> > > > > Cc: Vlastimil Babka <vbabka@suse.cz>
> > > > > Cc: Mike Rapoport <rppt@kernel.org>
> > > > > Cc: Suren Baghdasaryan <surenb@google.com>
> > > > > Cc: Michal Hocko <mhocko@suse.com>
> > > > > Cc: Balbir Singh <balbirs@nvidia.com>
> > > > > Cc: linuxppc-dev@lists.ozlabs.org
> > > > > Cc: kvm@vger.kernel.org
> > > > > Cc: linux-kernel@vger.kernel.org
> > > > > Cc: amd-gfx@lists.freedesktop.org
> > > > > Cc: dri-devel@lists.freedesktop.org
> > > > > Cc: nouveau@lists.freedesktop.org
> > > > > Cc: linux-mm@kvack.org
> > > > > Cc: linux-cxl@vger.kernel.org
> > > > > Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> > > > > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > > > > Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> > > > > ---
> > > > >  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
> > > > >  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
> > > > >  drivers/gpu/drm/drm_pagemap.c            |  2 +-
> > > > >  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
> > > > >  include/linux/memremap.h                 |  9 ++++++---
> > > > >  lib/test_hmm.c                           |  4 +++-
> > > > >  mm/memremap.c                            | 20 +++++++++++++++++++-
> > > > >  7 files changed, 32 insertions(+), 9 deletions(-)
> > > > > 
> > > > > diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > index e5000bef90f2..7cf9310de0ec 100644
> > > > > --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
> > > > >  
> > > > >  	dpage = pfn_to_page(uvmem_pfn);
> > > > >  	dpage->zone_device_data = pvt;
> > > > > -	zone_device_page_init(dpage, 0);
> > > > > +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
> > > > >  	return dpage;
> > > > >  out_clear:
> > > > >  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> > > > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > index af53e796ea1b..6ada7b4af7c6 100644
> > > > > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
> > > > >  	page = pfn_to_page(pfn);
> > > > >  	svm_range_bo_ref(prange->svm_bo);
> > > > >  	page->zone_device_data = prange->svm_bo;
> > > > > -	zone_device_page_init(page, 0);
> > > > > +	zone_device_page_init(page, page_pgmap(page), 0);
> > > > >  }
> > > > >  
> > > > >  static void
> > > > > diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> > > > > index 03ee39a761a4..c497726b0147 100644
> > > > > --- a/drivers/gpu/drm/drm_pagemap.c
> > > > > +++ b/drivers/gpu/drm/drm_pagemap.c
> > > > > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> > > > >  					struct drm_pagemap_zdd *zdd)
> > > > >  {
> > > > >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > > > > -	zone_device_page_init(page, 0);
> > > > > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> > > > >  }
> > > > >  
> > > > >  /**
> > > > > diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > index 58071652679d..3d8031296eed 100644
> > > > > --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
> > > > >  			order = ilog2(DMEM_CHUNK_NPAGES);
> > > > >  	}
> > > > >  
> > > > > -	zone_device_folio_init(folio, order);
> > > > > +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
> > > > >  	return page;
> > > > >  }
> > > > >  
> > > > > diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> > > > > index 713ec0435b48..e3c2ccf872a8 100644
> > > > > --- a/include/linux/memremap.h
> > > > > +++ b/include/linux/memremap.h
> > > > > @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
> > > > >  }
> > > > >  
> > > > >  #ifdef CONFIG_ZONE_DEVICE
> > > > > -void zone_device_page_init(struct page *page, unsigned int order);
> > > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > > +			   unsigned int order);
> > > > >  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);
> > > > > @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
> > > > >  
> > > > >  unsigned long memremap_compat_align(void);
> > > > >  
> > > > > -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> > > > > +static inline void zone_device_folio_init(struct folio *folio,
> > > > > +					  struct dev_pagemap *pgmap,
> > > > > +					  unsigned int order)
> > > > >  {
> > > > > -	zone_device_page_init(&folio->page, order);
> > > > > +	zone_device_page_init(&folio->page, pgmap, order);
> > > > >  	if (order)
> > > > >  		folio_set_large_rmappable(folio);
> > > > >  }
> > > > > diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> > > > > index 8af169d3873a..455a6862ae50 100644
> > > > > --- a/lib/test_hmm.c
> > > > > +++ b/lib/test_hmm.c
> > > > > @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
> > > > >  			goto error;
> > > > >  	}
> > > > >  
> > > > > -	zone_device_folio_init(page_folio(dpage), order);
> > > > > +	zone_device_folio_init(page_folio(dpage),
> > > > > +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> > > > > +			       order);
> > > > >  	dpage->zone_device_data = rpage;
> > > > >  	return dpage;
> > > > >  
> > > > > diff --git a/mm/memremap.c b/mm/memremap.c
> > > > > index 63c6ab4fdf08..6f46ab14662b 100644
> > > > > --- a/mm/memremap.c
> > > > > +++ b/mm/memremap.c
> > > > > @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
> > > > >  	}
> > > > >  }
> > > > >  
> > > > > -void zone_device_page_init(struct page *page, unsigned int order)
> > > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > > +			   unsigned int order)
> > > > >  {
> > > > > +	struct page *new_page = page;
> > > > > +	unsigned int i;
> > > > > +
> > > > >  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> > > > >  
> > > > > +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > > > > +		struct folio *new_folio = (struct folio *)new_page;
> > > > > +
> > > > > +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> > > > 
> > > > This seems odd to me, mainly due to the "magic" number. Why not just clear
> > > > the flags entirely? Or at least explicitly just clear the flags you care about
> > > > which would remove the need for the comment and  should let you use the proper
> > > > PageFlag functions.
> > > > 
> > > 
> > > I'm copying this from folio_reset_order [1]. My paranoia about touching
> > > anything related to struct page is high, so I did the same thing
> > > folio_reset_order does here.
> 
> So why not just use folio_reset_order() below?
> 
> > > 
> > > [1] https://elixir.bootlin.com/linux/v6.18.5/source/include/linux/mm.h#L1075
> > > 
> > 
> > This immediately hangs my first SVM test...
> > 
> > diff --git a/mm/memremap.c b/mm/memremap.c
> > index 6f46ab14662b..ef8c56876cf5 100644
> > --- a/mm/memremap.c
> > +++ b/mm/memremap.c
> > @@ -488,7 +488,7 @@ void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> >         for (i = 0; i < (1UL << order); ++i, ++new_page) {
> >                 struct folio *new_folio = (struct folio *)new_page;
> > 
> > -               new_page->flags.f &= ~0xffUL;   /* Clear possible order, page head */
> > +               new_page->flags.f = 0;
> >  #ifdef NR_PAGES_IN_LARGE_FOLIO
> >                 ((struct folio *)(new_page - 1))->_nr_pages = 0;
> 
> This seems wrong to me - I saw your reply to Balbir but for an order-0 page
> isn't this going to access a completely different, possibly already allocated,
> page?
> 

No — it accesses itself (new_page). It just uses some odd memory tricks
for this, which I agree isn’t the best thing I’ve ever written, but it
was the least-worst idea I had. I didn’t design the folio/page field
aliasing; I understand why it exists, but it still makes my head hurt.

folio->_nr_pages is page + 1 for reference (new_page after this math).
Again, if I touched this memory directly in new_page, it’s most likely
memcg_data, but that is hidden behind a Kconfig.

This just blindly implementing part of folio_reset_order which clears
_nr_pages.

> >  #endif
> > 
> > I can walk through exactly which flags need to be cleared, but my
> > feeling is that likely any flag that the order field overloads and can
> > possibly encode should be cleared—so bits 0–7 based on the existing
> > code.
> > 
> > How about in a follow-up we normalize setting / clearing the order flag
> > field with a #define and an inline helper?
> 
> Ie: Would something like the following work:
> 
> 		ClearPageHead(new_page);

Any of these bit could possibly be set the order field in a folio, which
modifies page + 1 flags field.

	PG_locked,		/* Page is locked. Don't touch. */
	PG_writeback,		/* Page is under writeback */
	PG_referenced,
	PG_uptodate,
	PG_dirty,
	PG_lru,
	PG_head,		/* Must be in bit 6 */
	PG_waiters,		/* Page has waiters, check its waitqueue. Must be bit #7 and in the same byte as "PG_locked" */

So a common order-9 (2MB) folio would have PG_locked | PG_uptodate set.
Now we get stuck on the next page lock because PG_locked is set.
Offhand, I don’t know if different orders—which set different bits—cause
any nasty issues either. So I figured the safest thing was clear any
bits which folio order can set within subsequent page's memory flags
like folio_reset_order does.

> 		clear_compound_head(new_page);
> 		folio_reset_order(new_folio);
> 
> Which would also deal with setting _nr_pages.
>

folio_reset_order(new_folio) would set _nr_pages in the memory that is
new_page + 1. So let's say that page has a ref count + memcg_data, now
that memory is corrupted and will crash the kernel.

All of the above is why is took me multiple hours to write 6 lines of
code :).
 
> > Matt
> > 
> > > > > +#ifdef NR_PAGES_IN_LARGE_FOLIO
> > > > > +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> > > > > +#endif
> > > > > +		new_folio->mapping = NULL;
> > > > > +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> > > > > +		new_folio->share = 0;   /* fsdax only, unused for device private */
> > > > 
> > > > It would be nice if the FS DAX code actually used this as well. Is there a
> > > > reason that change was dropped from the series?
> > > > 
> > > 
> > > I don't have a test platform for FS DAX. In prior revisions, I was just
> > > moving existing FS DAX code to a helper, which I felt confident about.
> > > 
> > > This revision is slightly different, and I don't feel comfortable
> > > modifying FS DAX code without a test platform. I agree we should update
> > > FS DAX, but that should be done in a follow-up with coordinated testing.
> 
> Fair enough, I figured something like this might be your answer :-) You
> could update it and ask people with access to such a system to test it though
> (unfortunately my setup has bit-rotted beyond repair).
> 
> But I'm ok leaving to for a future change.
>

I did a quick grep in fs/dax.c and don’t see zone_device_page_init
called there. It probably could be used if it’s creating compound pages
and drop the open-coded reinit when shared == 0, but yeah, that’s not
something I can blindly code without testing.

I can try to put something together for people to test soonish.

Matt

> > > 
> > > Matt 
> > > 
> > > > > +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> > > > > +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> > > > > +	}
> > > > > +
> > > > >  	/*
> > > > >  	 * Drivers shouldn't be allocating pages after calling
> > > > >  	 * memunmap_pages().
> > > > > -- 
> > > > > 2.43.0
> > > > >

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Alistair Popple 3 weeks, 4 days ago

On 2026-01-15 at 18:43 +1100, Matthew Brost <matthew.brost@intel.com> wrote...
> On Thu, Jan 15, 2026 at 06:07:08PM +1100, Alistair Popple wrote:
> > On 2026-01-15 at 17:18 +1100, Matthew Brost <matthew.brost@intel.com> wrote...
> > > On Wed, Jan 14, 2026 at 09:57:31PM -0800, Matthew Brost wrote:
> > > > On Thu, Jan 15, 2026 at 04:27:26PM +1100, Alistair Popple wrote:
> > > > > On 2026-01-15 at 06:19 +1100, Francois Dugast <francois.dugast@intel.com> wrote...
> > > > > > From: Matthew Brost <matthew.brost@intel.com>
> > > > > > 
> > > > > > Reinitialize metadata for large zone device private folios in
> > > > > > zone_device_page_init prior to creating a higher-order zone device
> > > > > > private folio. This step is necessary when the folio’s order changes
> > > > > > dynamically between zone_device_page_init calls to avoid building a
> > > > > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > > > > must be passed in from the caller because the pgmap stored in the folio
> > > > > > page may have been overwritten with a compound head.
> > > > > 
> > > > > Thanks for fixing, a couple of minor comments below.
> > > > > 
> > > > > > Cc: Zi Yan <ziy@nvidia.com>
> > > > > > Cc: Alistair Popple <apopple@nvidia.com>
> > > > > > Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> > > > > > Cc: Nicholas Piggin <npiggin@gmail.com>
> > > > > > Cc: Michael Ellerman <mpe@ellerman.id.au>
> > > > > > Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> > > > > > Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> > > > > > Cc: Alex Deucher <alexander.deucher@amd.com>
> > > > > > Cc: "Christian König" <christian.koenig@amd.com>
> > > > > > Cc: David Airlie <airlied@gmail.com>
> > > > > > Cc: Simona Vetter <simona@ffwll.ch>
> > > > > > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > > > > > Cc: Maxime Ripard <mripard@kernel.org>
> > > > > > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > > > > > Cc: Lyude Paul <lyude@redhat.com>
> > > > > > Cc: Danilo Krummrich <dakr@kernel.org>
> > > > > > Cc: David Hildenbrand <david@kernel.org>
> > > > > > Cc: Oscar Salvador <osalvador@suse.de>
> > > > > > Cc: Andrew Morton <akpm@linux-foundation.org>
> > > > > > Cc: Jason Gunthorpe <jgg@ziepe.ca>
> > > > > > Cc: Leon Romanovsky <leon@kernel.org>
> > > > > > Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> > > > > > Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> > > > > > Cc: Vlastimil Babka <vbabka@suse.cz>
> > > > > > Cc: Mike Rapoport <rppt@kernel.org>
> > > > > > Cc: Suren Baghdasaryan <surenb@google.com>
> > > > > > Cc: Michal Hocko <mhocko@suse.com>
> > > > > > Cc: Balbir Singh <balbirs@nvidia.com>
> > > > > > Cc: linuxppc-dev@lists.ozlabs.org
> > > > > > Cc: kvm@vger.kernel.org
> > > > > > Cc: linux-kernel@vger.kernel.org
> > > > > > Cc: amd-gfx@lists.freedesktop.org
> > > > > > Cc: dri-devel@lists.freedesktop.org
> > > > > > Cc: nouveau@lists.freedesktop.org
> > > > > > Cc: linux-mm@kvack.org
> > > > > > Cc: linux-cxl@vger.kernel.org
> > > > > > Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> > > > > > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > > > > > Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> > > > > > ---
> > > > > >  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
> > > > > >  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
> > > > > >  drivers/gpu/drm/drm_pagemap.c            |  2 +-
> > > > > >  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
> > > > > >  include/linux/memremap.h                 |  9 ++++++---
> > > > > >  lib/test_hmm.c                           |  4 +++-
> > > > > >  mm/memremap.c                            | 20 +++++++++++++++++++-
> > > > > >  7 files changed, 32 insertions(+), 9 deletions(-)
> > > > > > 
> > > > > > diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > > index e5000bef90f2..7cf9310de0ec 100644
> > > > > > --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > > +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > > @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
> > > > > >  
> > > > > >  	dpage = pfn_to_page(uvmem_pfn);
> > > > > >  	dpage->zone_device_data = pvt;
> > > > > > -	zone_device_page_init(dpage, 0);
> > > > > > +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
> > > > > >  	return dpage;
> > > > > >  out_clear:
> > > > > >  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> > > > > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > > index af53e796ea1b..6ada7b4af7c6 100644
> > > > > > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > > @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
> > > > > >  	page = pfn_to_page(pfn);
> > > > > >  	svm_range_bo_ref(prange->svm_bo);
> > > > > >  	page->zone_device_data = prange->svm_bo;
> > > > > > -	zone_device_page_init(page, 0);
> > > > > > +	zone_device_page_init(page, page_pgmap(page), 0);
> > > > > >  }
> > > > > >  
> > > > > >  static void
> > > > > > diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> > > > > > index 03ee39a761a4..c497726b0147 100644
> > > > > > --- a/drivers/gpu/drm/drm_pagemap.c
> > > > > > +++ b/drivers/gpu/drm/drm_pagemap.c
> > > > > > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> > > > > >  					struct drm_pagemap_zdd *zdd)
> > > > > >  {
> > > > > >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > > > > > -	zone_device_page_init(page, 0);
> > > > > > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> > > > > >  }
> > > > > >  
> > > > > >  /**
> > > > > > diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > > index 58071652679d..3d8031296eed 100644
> > > > > > --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > > +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > > @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
> > > > > >  			order = ilog2(DMEM_CHUNK_NPAGES);
> > > > > >  	}
> > > > > >  
> > > > > > -	zone_device_folio_init(folio, order);
> > > > > > +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
> > > > > >  	return page;
> > > > > >  }
> > > > > >  
> > > > > > diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> > > > > > index 713ec0435b48..e3c2ccf872a8 100644
> > > > > > --- a/include/linux/memremap.h
> > > > > > +++ b/include/linux/memremap.h
> > > > > > @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
> > > > > >  }
> > > > > >  
> > > > > >  #ifdef CONFIG_ZONE_DEVICE
> > > > > > -void zone_device_page_init(struct page *page, unsigned int order);
> > > > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > > > +			   unsigned int order);
> > > > > >  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);
> > > > > > @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
> > > > > >  
> > > > > >  unsigned long memremap_compat_align(void);
> > > > > >  
> > > > > > -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> > > > > > +static inline void zone_device_folio_init(struct folio *folio,
> > > > > > +					  struct dev_pagemap *pgmap,
> > > > > > +					  unsigned int order)
> > > > > >  {
> > > > > > -	zone_device_page_init(&folio->page, order);
> > > > > > +	zone_device_page_init(&folio->page, pgmap, order);
> > > > > >  	if (order)
> > > > > >  		folio_set_large_rmappable(folio);
> > > > > >  }
> > > > > > diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> > > > > > index 8af169d3873a..455a6862ae50 100644
> > > > > > --- a/lib/test_hmm.c
> > > > > > +++ b/lib/test_hmm.c
> > > > > > @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
> > > > > >  			goto error;
> > > > > >  	}
> > > > > >  
> > > > > > -	zone_device_folio_init(page_folio(dpage), order);
> > > > > > +	zone_device_folio_init(page_folio(dpage),
> > > > > > +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> > > > > > +			       order);
> > > > > >  	dpage->zone_device_data = rpage;
> > > > > >  	return dpage;
> > > > > >  
> > > > > > diff --git a/mm/memremap.c b/mm/memremap.c
> > > > > > index 63c6ab4fdf08..6f46ab14662b 100644
> > > > > > --- a/mm/memremap.c
> > > > > > +++ b/mm/memremap.c
> > > > > > @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
> > > > > >  	}
> > > > > >  }
> > > > > >  
> > > > > > -void zone_device_page_init(struct page *page, unsigned int order)
> > > > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > > > +			   unsigned int order)
> > > > > >  {
> > > > > > +	struct page *new_page = page;
> > > > > > +	unsigned int i;
> > > > > > +
> > > > > >  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> > > > > >  
> > > > > > +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > > > > > +		struct folio *new_folio = (struct folio *)new_page;
> > > > > > +
> > > > > > +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> > > > > 
> > > > > This seems odd to me, mainly due to the "magic" number. Why not just clear
> > > > > the flags entirely? Or at least explicitly just clear the flags you care about
> > > > > which would remove the need for the comment and  should let you use the proper
> > > > > PageFlag functions.
> > > > > 
> > > > 
> > > > I'm copying this from folio_reset_order [1]. My paranoia about touching
> > > > anything related to struct page is high, so I did the same thing
> > > > folio_reset_order does here.
> > 
> > So why not just use folio_reset_order() below?
> > 
> > > > 
> > > > [1] https://elixir.bootlin.com/linux/v6.18.5/source/include/linux/mm.h#L1075
> > > > 
> > > 
> > > This immediately hangs my first SVM test...
> > > 
> > > diff --git a/mm/memremap.c b/mm/memremap.c
> > > index 6f46ab14662b..ef8c56876cf5 100644
> > > --- a/mm/memremap.c
> > > +++ b/mm/memremap.c
> > > @@ -488,7 +488,7 @@ void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > >         for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > >                 struct folio *new_folio = (struct folio *)new_page;
> > > 
> > > -               new_page->flags.f &= ~0xffUL;   /* Clear possible order, page head */
> > > +               new_page->flags.f = 0;
> > >  #ifdef NR_PAGES_IN_LARGE_FOLIO
> > >                 ((struct folio *)(new_page - 1))->_nr_pages = 0;
> > 
> > This seems wrong to me - I saw your reply to Balbir but for an order-0 page
> > isn't this going to access a completely different, possibly already allocated,
> > page?
> > 
> 
> No — it accesses itself (new_page). It just uses some odd memory tricks
> for this, which I agree isn’t the best thing I’ve ever written, but it
> was the least-worst idea I had. I didn’t design the folio/page field
> aliasing; I understand why it exists, but it still makes my head hurt.

And obviously mine, because I (was) still not getting it and had typed up a
whole response and code walk through to show what was wrong, in the hope it
would help settle the misunderstanding. Which it did, because I discovered
where I was getting things wrong. But I've left the analysis below because it's
probably useful for others following along:

Walking through the code we have:

void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
			   unsigned int order)
{

The first argument, page, is the first in a set of 1 << order contiguous
struct page. In the simplest case order == 0, meaning this function should only
initialise (ie. touch) a single struct page pointer which is passed as the first
argument to the function.

	struct page *new_page = page;

So now *new_page points to the single struct page we should touch.
	
	unsigned int i;

	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);

	for (i = 0; i < (1UL << order); ++i, ++new_page) {

order == 0, so this loop will only execute once.

		struct folio *new_folio = (struct folio *)new_page;

new_page still points to the single page we're initialising, and new_folio
points to the same page. Ie: &new_folio->page == new_page. There is a hazard
here because new_folio->__page_1, __page_2, etc. all point to pages we shouldn't
touch.

		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */

Clears the flags, makes sense.

#ifdef NR_PAGES_IN_LARGE_FOLIO
		((struct folio *)(new_page - 1))->_nr_pages = 0;

If we break this down we have:

struct page *tmp_new_page = new_page - 1;

Which is the page before the one we're initialising and shouldn't be touched.
Then we cast to a folio:

struct folio *tmp_new_folio = (struct folio *) tmp_new_page;

And reset _nr_pages:

tmp_new_folio->_nr_pages = 0

And now I can see where I was confused - &tmp_new_folio->_nr_pages == &tmp_new_folio->__page_1->memcg_data == &new_page->memcg_data

So after both Balbir, probably yourself, and definitely myself scratching our
heads for way too long over this change I think we can conclude that the code as
is is way too confusing to merge without a lot more comments :-)

However why go through all this magic in the first place? Why not just treat
everything here as a page and just do

	new_page->memcg_data = 0

directly? That seems like the more straight forward approach. In fact given
all the confusion I wonder if it wouldn't be better to just do
memset(new_page, 0, sizeof(*new_page)) and reinitialise everything from
scratch.

> folio->_nr_pages is page + 1 for reference (new_page after this math).
> Again, if I touched this memory directly in new_page, it’s most likely
> memcg_data, but that is hidden behind a Kconfig.
> 
> This just blindly implementing part of folio_reset_order which clears
> _nr_pages.

Yeah, I get it now. But I think just clearing memcg_data would be the easiest to
understand approach, especially if it had a comment explaining that it may have
previously been used for _nr_pages.

> > >  #endif
> > > 
> > > I can walk through exactly which flags need to be cleared, but my
> > > feeling is that likely any flag that the order field overloads and can
> > > possibly encode should be cleared—so bits 0–7 based on the existing
> > > code.
> > > 
> > > How about in a follow-up we normalize setting / clearing the order flag
> > > field with a #define and an inline helper?
> > 
> > Ie: Would something like the following work:
> > 
> > 		ClearPageHead(new_page);
> 
> Any of these bit could possibly be set the order field in a folio, which
> modifies page + 1 flags field.
> 
> 	PG_locked,		/* Page is locked. Don't touch. */
> 	PG_writeback,		/* Page is under writeback */
> 	PG_referenced,
> 	PG_uptodate,
> 	PG_dirty,
> 	PG_lru,
> 	PG_head,		/* Must be in bit 6 */
> 	PG_waiters,		/* Page has waiters, check its waitqueue. Must be bit #7 and in the same byte as "PG_locked" */
> 
> So a common order-9 (2MB) folio would have PG_locked | PG_uptodate set.
> Now we get stuck on the next page lock because PG_locked is set.
> Offhand, I don’t know if different orders—which set different bits—cause
> any nasty issues either. So I figured the safest thing was clear any
> bits which folio order can set within subsequent page's memory flags
> like folio_reset_order does.

Oh, I get the above. I was thinking folio_reset_order() below would clear the
flags, but I see the folly there - that resets the flags for the next page.

> 
> > 		clear_compound_head(new_page);
> > 		folio_reset_order(new_folio);
> > 
> > Which would also deal with setting _nr_pages.
> >
> 
> folio_reset_order(new_folio) would set _nr_pages in the memory that is
> new_page + 1. So let's say that page has a ref count + memcg_data, now
> that memory is corrupted and will crash the kernel.

Yep, I just noticed that. Thanks for pointing that out.

> All of the above is why is took me multiple hours to write 6 lines of
> code :).

And to review :) Good thing we don't get paid per SLOC of code right?

 - Alistair

> > > Matt
> > > 
> > > > > > +#ifdef NR_PAGES_IN_LARGE_FOLIO
> > > > > > +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> > > > > > +#endif
> > > > > > +		new_folio->mapping = NULL;
> > > > > > +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> > > > > > +		new_folio->share = 0;   /* fsdax only, unused for device private */
> > > > > 
> > > > > It would be nice if the FS DAX code actually used this as well. Is there a
> > > > > reason that change was dropped from the series?
> > > > > 
> > > > 
> > > > I don't have a test platform for FS DAX. In prior revisions, I was just
> > > > moving existing FS DAX code to a helper, which I felt confident about.
> > > > 
> > > > This revision is slightly different, and I don't feel comfortable
> > > > modifying FS DAX code without a test platform. I agree we should update
> > > > FS DAX, but that should be done in a follow-up with coordinated testing.
> > 
> > Fair enough, I figured something like this might be your answer :-) You
> > could update it and ask people with access to such a system to test it though
> > (unfortunately my setup has bit-rotted beyond repair).
> > 
> > But I'm ok leaving to for a future change.
> >
> 
> I did a quick grep in fs/dax.c and don’t see zone_device_page_init
> called there. It probably could be used if it’s creating compound pages
> and drop the open-coded reinit when shared == 0, but yeah, that’s not
> something I can blindly code without testing.
> 
> I can try to put something together for people to test soonish.
> 
> Matt
> 
> > > > 
> > > > Matt 
> > > > 
> > > > > > +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> > > > > > +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> > > > > > +	}
> > > > > > +
> > > > > >  	/*
> > > > > >  	 * Drivers shouldn't be allocating pages after calling
> > > > > >  	 * memunmap_pages().
> > > > > > -- 
> > > > > > 2.43.0
> > > > > >

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Matthew Brost 3 weeks, 3 days ago

On Thu, Jan 15, 2026 at 10:05:00PM +1100, Alistair Popple wrote:
> On 2026-01-15 at 18:43 +1100, Matthew Brost <matthew.brost@intel.com> wrote...
> > On Thu, Jan 15, 2026 at 06:07:08PM +1100, Alistair Popple wrote:
> > > On 2026-01-15 at 17:18 +1100, Matthew Brost <matthew.brost@intel.com> wrote...
> > > > On Wed, Jan 14, 2026 at 09:57:31PM -0800, Matthew Brost wrote:
> > > > > On Thu, Jan 15, 2026 at 04:27:26PM +1100, Alistair Popple wrote:
> > > > > > On 2026-01-15 at 06:19 +1100, Francois Dugast <francois.dugast@intel.com> wrote...
> > > > > > > From: Matthew Brost <matthew.brost@intel.com>
> > > > > > > 
> > > > > > > Reinitialize metadata for large zone device private folios in
> > > > > > > zone_device_page_init prior to creating a higher-order zone device
> > > > > > > private folio. This step is necessary when the folio’s order changes
> > > > > > > dynamically between zone_device_page_init calls to avoid building a
> > > > > > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > > > > > must be passed in from the caller because the pgmap stored in the folio
> > > > > > > page may have been overwritten with a compound head.
> > > > > > 
> > > > > > Thanks for fixing, a couple of minor comments below.
> > > > > > 
> > > > > > > Cc: Zi Yan <ziy@nvidia.com>
> > > > > > > Cc: Alistair Popple <apopple@nvidia.com>
> > > > > > > Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> > > > > > > Cc: Nicholas Piggin <npiggin@gmail.com>
> > > > > > > Cc: Michael Ellerman <mpe@ellerman.id.au>
> > > > > > > Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> > > > > > > Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> > > > > > > Cc: Alex Deucher <alexander.deucher@amd.com>
> > > > > > > Cc: "Christian König" <christian.koenig@amd.com>
> > > > > > > Cc: David Airlie <airlied@gmail.com>
> > > > > > > Cc: Simona Vetter <simona@ffwll.ch>
> > > > > > > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > > > > > > Cc: Maxime Ripard <mripard@kernel.org>
> > > > > > > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > > > > > > Cc: Lyude Paul <lyude@redhat.com>
> > > > > > > Cc: Danilo Krummrich <dakr@kernel.org>
> > > > > > > Cc: David Hildenbrand <david@kernel.org>
> > > > > > > Cc: Oscar Salvador <osalvador@suse.de>
> > > > > > > Cc: Andrew Morton <akpm@linux-foundation.org>
> > > > > > > Cc: Jason Gunthorpe <jgg@ziepe.ca>
> > > > > > > Cc: Leon Romanovsky <leon@kernel.org>
> > > > > > > Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> > > > > > > Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> > > > > > > Cc: Vlastimil Babka <vbabka@suse.cz>
> > > > > > > Cc: Mike Rapoport <rppt@kernel.org>
> > > > > > > Cc: Suren Baghdasaryan <surenb@google.com>
> > > > > > > Cc: Michal Hocko <mhocko@suse.com>
> > > > > > > Cc: Balbir Singh <balbirs@nvidia.com>
> > > > > > > Cc: linuxppc-dev@lists.ozlabs.org
> > > > > > > Cc: kvm@vger.kernel.org
> > > > > > > Cc: linux-kernel@vger.kernel.org
> > > > > > > Cc: amd-gfx@lists.freedesktop.org
> > > > > > > Cc: dri-devel@lists.freedesktop.org
> > > > > > > Cc: nouveau@lists.freedesktop.org
> > > > > > > Cc: linux-mm@kvack.org
> > > > > > > Cc: linux-cxl@vger.kernel.org
> > > > > > > Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> > > > > > > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > > > > > > Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> > > > > > > ---
> > > > > > >  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
> > > > > > >  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
> > > > > > >  drivers/gpu/drm/drm_pagemap.c            |  2 +-
> > > > > > >  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
> > > > > > >  include/linux/memremap.h                 |  9 ++++++---
> > > > > > >  lib/test_hmm.c                           |  4 +++-
> > > > > > >  mm/memremap.c                            | 20 +++++++++++++++++++-
> > > > > > >  7 files changed, 32 insertions(+), 9 deletions(-)
> > > > > > > 
> > > > > > > diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > > > index e5000bef90f2..7cf9310de0ec 100644
> > > > > > > --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > > > +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > > > @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
> > > > > > >  
> > > > > > >  	dpage = pfn_to_page(uvmem_pfn);
> > > > > > >  	dpage->zone_device_data = pvt;
> > > > > > > -	zone_device_page_init(dpage, 0);
> > > > > > > +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
> > > > > > >  	return dpage;
> > > > > > >  out_clear:
> > > > > > >  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> > > > > > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > > > index af53e796ea1b..6ada7b4af7c6 100644
> > > > > > > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > > > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > > > @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
> > > > > > >  	page = pfn_to_page(pfn);
> > > > > > >  	svm_range_bo_ref(prange->svm_bo);
> > > > > > >  	page->zone_device_data = prange->svm_bo;
> > > > > > > -	zone_device_page_init(page, 0);
> > > > > > > +	zone_device_page_init(page, page_pgmap(page), 0);
> > > > > > >  }
> > > > > > >  
> > > > > > >  static void
> > > > > > > diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> > > > > > > index 03ee39a761a4..c497726b0147 100644
> > > > > > > --- a/drivers/gpu/drm/drm_pagemap.c
> > > > > > > +++ b/drivers/gpu/drm/drm_pagemap.c
> > > > > > > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> > > > > > >  					struct drm_pagemap_zdd *zdd)
> > > > > > >  {
> > > > > > >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > > > > > > -	zone_device_page_init(page, 0);
> > > > > > > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> > > > > > >  }
> > > > > > >  
> > > > > > >  /**
> > > > > > > diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > > > index 58071652679d..3d8031296eed 100644
> > > > > > > --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > > > +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > > > @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
> > > > > > >  			order = ilog2(DMEM_CHUNK_NPAGES);
> > > > > > >  	}
> > > > > > >  
> > > > > > > -	zone_device_folio_init(folio, order);
> > > > > > > +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
> > > > > > >  	return page;
> > > > > > >  }
> > > > > > >  
> > > > > > > diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> > > > > > > index 713ec0435b48..e3c2ccf872a8 100644
> > > > > > > --- a/include/linux/memremap.h
> > > > > > > +++ b/include/linux/memremap.h
> > > > > > > @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
> > > > > > >  }
> > > > > > >  
> > > > > > >  #ifdef CONFIG_ZONE_DEVICE
> > > > > > > -void zone_device_page_init(struct page *page, unsigned int order);
> > > > > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > > > > +			   unsigned int order);
> > > > > > >  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);
> > > > > > > @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
> > > > > > >  
> > > > > > >  unsigned long memremap_compat_align(void);
> > > > > > >  
> > > > > > > -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> > > > > > > +static inline void zone_device_folio_init(struct folio *folio,
> > > > > > > +					  struct dev_pagemap *pgmap,
> > > > > > > +					  unsigned int order)
> > > > > > >  {
> > > > > > > -	zone_device_page_init(&folio->page, order);
> > > > > > > +	zone_device_page_init(&folio->page, pgmap, order);
> > > > > > >  	if (order)
> > > > > > >  		folio_set_large_rmappable(folio);
> > > > > > >  }
> > > > > > > diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> > > > > > > index 8af169d3873a..455a6862ae50 100644
> > > > > > > --- a/lib/test_hmm.c
> > > > > > > +++ b/lib/test_hmm.c
> > > > > > > @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
> > > > > > >  			goto error;
> > > > > > >  	}
> > > > > > >  
> > > > > > > -	zone_device_folio_init(page_folio(dpage), order);
> > > > > > > +	zone_device_folio_init(page_folio(dpage),
> > > > > > > +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> > > > > > > +			       order);
> > > > > > >  	dpage->zone_device_data = rpage;
> > > > > > >  	return dpage;
> > > > > > >  
> > > > > > > diff --git a/mm/memremap.c b/mm/memremap.c
> > > > > > > index 63c6ab4fdf08..6f46ab14662b 100644
> > > > > > > --- a/mm/memremap.c
> > > > > > > +++ b/mm/memremap.c
> > > > > > > @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
> > > > > > >  	}
> > > > > > >  }
> > > > > > >  
> > > > > > > -void zone_device_page_init(struct page *page, unsigned int order)
> > > > > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > > > > +			   unsigned int order)
> > > > > > >  {
> > > > > > > +	struct page *new_page = page;
> > > > > > > +	unsigned int i;
> > > > > > > +
> > > > > > >  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> > > > > > >  
> > > > > > > +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > > > > > > +		struct folio *new_folio = (struct folio *)new_page;
> > > > > > > +
> > > > > > > +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> > > > > > 
> > > > > > This seems odd to me, mainly due to the "magic" number. Why not just clear
> > > > > > the flags entirely? Or at least explicitly just clear the flags you care about
> > > > > > which would remove the need for the comment and  should let you use the proper
> > > > > > PageFlag functions.
> > > > > > 
> > > > > 
> > > > > I'm copying this from folio_reset_order [1]. My paranoia about touching
> > > > > anything related to struct page is high, so I did the same thing
> > > > > folio_reset_order does here.
> > > 
> > > So why not just use folio_reset_order() below?
> > > 
> > > > > 
> > > > > [1] https://elixir.bootlin.com/linux/v6.18.5/source/include/linux/mm.h#L1075
> > > > > 
> > > > 
> > > > This immediately hangs my first SVM test...
> > > > 
> > > > diff --git a/mm/memremap.c b/mm/memremap.c
> > > > index 6f46ab14662b..ef8c56876cf5 100644
> > > > --- a/mm/memremap.c
> > > > +++ b/mm/memremap.c
> > > > @@ -488,7 +488,7 @@ void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > >         for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > > >                 struct folio *new_folio = (struct folio *)new_page;
> > > > 
> > > > -               new_page->flags.f &= ~0xffUL;   /* Clear possible order, page head */
> > > > +               new_page->flags.f = 0;
> > > >  #ifdef NR_PAGES_IN_LARGE_FOLIO
> > > >                 ((struct folio *)(new_page - 1))->_nr_pages = 0;
> > > 
> > > This seems wrong to me - I saw your reply to Balbir but for an order-0 page
> > > isn't this going to access a completely different, possibly already allocated,
> > > page?
> > > 
> > 
> > No — it accesses itself (new_page). It just uses some odd memory tricks
> > for this, which I agree isn’t the best thing I’ve ever written, but it
> > was the least-worst idea I had. I didn’t design the folio/page field
> > aliasing; I understand why it exists, but it still makes my head hurt.
> 
> And obviously mine, because I (was) still not getting it and had typed up a
> whole response and code walk through to show what was wrong, in the hope it
> would help settle the misunderstanding. Which it did, because I discovered
> where I was getting things wrong. But I've left the analysis below because it's
> probably useful for others following along:
> 
> Walking through the code we have:
> 
> void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> 			   unsigned int order)
> {
> 
> The first argument, page, is the first in a set of 1 << order contiguous
> struct page. In the simplest case order == 0, meaning this function should only
> initialise (ie. touch) a single struct page pointer which is passed as the first
> argument to the function.

Yes.

> 
> 	struct page *new_page = page;
> 
> So now *new_page points to the single struct page we should touch.
> 	
> 	unsigned int i;
> 
> 	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> 
> 	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> 
> order == 0, so this loop will only execute once.
> 

Yes.

> 		struct folio *new_folio = (struct folio *)new_page;
> 
> new_page still points to the single page we're initialising, and new_folio
> points to the same page. Ie: &new_folio->page == new_page. There is a hazard
> here because new_folio->__page_1, __page_2, etc. all point to pages we shouldn't
> touch.
> 

Yes.

> 		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> 
> Clears the flags, makes sense.
> 

+1

> #ifdef NR_PAGES_IN_LARGE_FOLIO
> 		((struct folio *)(new_page - 1))->_nr_pages = 0;
> 
> If we break this down we have:
> 
> struct page *tmp_new_page = new_page - 1;
> 
> Which is the page before the one we're initialising and shouldn't be touched.
> Then we cast to a folio:
> 
> struct folio *tmp_new_folio = (struct folio *) tmp_new_page;
> 
> And reset _nr_pages:
> 
> tmp_new_folio->_nr_pages = 0
> 
> And now I can see where I was confused - &tmp_new_folio->_nr_pages == &tmp_new_folio->__page_1->memcg_data == &new_page->memcg_data
> 

Not 100% right, as _nr_pages is 4 bytes and memcg_data is 8, but the
pointer base address is the same.

> So after both Balbir, probably yourself, and definitely myself scratching our
> heads for way too long over this change I think we can conclude that the code as
> is is way too confusing to merge without a lot more comments :-)
> 

I think more comments is the way to go. More below.

> However why go through all this magic in the first place? Why not just treat
> everything here as a page and just do
> 
> 	new_page->memcg_data = 0
> 

Well, memcg_data is 8 bytes and _nr_pages is 4. They also have different
#ifdef conditions around each field, etc.

I’ve also seen failures in our testing, and so has François, with the
memcg_data change. I wish I had a stack trace to share or explain, but
the times I hit the error I didn’t capture the dmesg, and I’ve been
having issues with my dev machine today. If I catch the error again,
I’ll reply with a stack trace and analysis.

> directly? That seems like the more straight forward approach. In fact given
> all the confusion I wonder if it wouldn't be better to just do
> memset(new_page, 0, sizeof(*new_page)) and reinitialise everything from
> scratch.

I had considered this option too, but I’d be a little concerned about
the performance. Reinitializing a zone page/folio is a hot path, as this
is typically done in a GPU fault handler. I think adding verbose
comments explaining why this works, plus some follow-up helpers, might
be the better option.

> 
> > folio->_nr_pages is page + 1 for reference (new_page after this math).
> > Again, if I touched this memory directly in new_page, it’s most likely
> > memcg_data, but that is hidden behind a Kconfig.
> > 
> > This just blindly implementing part of folio_reset_order which clears
> > _nr_pages.
> 
> Yeah, I get it now. But I think just clearing memcg_data would be the easiest to
> understand approach, especially if it had a comment explaining that it may have
> previously been used for _nr_pages.
> 

See above — the different sizes, the failure I’m seeing, and the
conflicting #ifdefs are why this is not my preferred option.

> > > >  #endif
> > > > 
> > > > I can walk through exactly which flags need to be cleared, but my
> > > > feeling is that likely any flag that the order field overloads and can
> > > > possibly encode should be cleared—so bits 0–7 based on the existing
> > > > code.
> > > > 
> > > > How about in a follow-up we normalize setting / clearing the order flag
> > > > field with a #define and an inline helper?
> > > 
> > > Ie: Would something like the following work:
> > > 
> > > 		ClearPageHead(new_page);
> > 
> > Any of these bit could possibly be set the order field in a folio, which
> > modifies page + 1 flags field.
> > 
> > 	PG_locked,		/* Page is locked. Don't touch. */
> > 	PG_writeback,		/* Page is under writeback */
> > 	PG_referenced,
> > 	PG_uptodate,
> > 	PG_dirty,
> > 	PG_lru,
> > 	PG_head,		/* Must be in bit 6 */
> > 	PG_waiters,		/* Page has waiters, check its waitqueue. Must be bit #7 and in the same byte as "PG_locked" */
> > 
> > So a common order-9 (2MB) folio would have PG_locked | PG_uptodate set.
> > Now we get stuck on the next page lock because PG_locked is set.
> > Offhand, I don’t know if different orders—which set different bits—cause
> > any nasty issues either. So I figured the safest thing was clear any
> > bits which folio order can set within subsequent page's memory flags
> > like folio_reset_order does.
> 
> Oh, I get the above. I was thinking folio_reset_order() below would clear the
> flags, but I see the folly there - that resets the flags for the next page.
> 

Correct.

> > 
> > > 		clear_compound_head(new_page);
> > > 		folio_reset_order(new_folio);
> > > 
> > > Which would also deal with setting _nr_pages.
> > >
> > 
> > folio_reset_order(new_folio) would set _nr_pages in the memory that is
> > new_page + 1. So let's say that page has a ref count + memcg_data, now
> > that memory is corrupted and will crash the kernel.
> 
> Yep, I just noticed that. Thanks for pointing that out.
> 
> > All of the above is why is took me multiple hours to write 6 lines of
> > code :).
> 
> And to review :) Good thing we don't get paid per SLOC of code right?
>

I don’t think anyone would touch core MM if pay were based on SLOC — it
would be a terrible career choice. :)

All joking aside, I think the next revision should use this version,
plus more comments and helpers/defines in a follow-up—which I’ll commit
to—along with fixing the branch mismatch Andrew pointed out between
drm-tip (which this series is based on) and 6.19 (where this patch needs
to apply).

Matt

>  - Alistair
> 
> > > > Matt
> > > > 
> > > > > > > +#ifdef NR_PAGES_IN_LARGE_FOLIO
> > > > > > > +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> > > > > > > +#endif
> > > > > > > +		new_folio->mapping = NULL;
> > > > > > > +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> > > > > > > +		new_folio->share = 0;   /* fsdax only, unused for device private */
> > > > > > 
> > > > > > It would be nice if the FS DAX code actually used this as well. Is there a
> > > > > > reason that change was dropped from the series?
> > > > > > 
> > > > > 
> > > > > I don't have a test platform for FS DAX. In prior revisions, I was just
> > > > > moving existing FS DAX code to a helper, which I felt confident about.
> > > > > 
> > > > > This revision is slightly different, and I don't feel comfortable
> > > > > modifying FS DAX code without a test platform. I agree we should update
> > > > > FS DAX, but that should be done in a follow-up with coordinated testing.
> > > 
> > > Fair enough, I figured something like this might be your answer :-) You
> > > could update it and ask people with access to such a system to test it though
> > > (unfortunately my setup has bit-rotted beyond repair).
> > > 
> > > But I'm ok leaving to for a future change.
> > >
> > 
> > I did a quick grep in fs/dax.c and don’t see zone_device_page_init
> > called there. It probably could be used if it’s creating compound pages
> > and drop the open-coded reinit when shared == 0, but yeah, that’s not
> > something I can blindly code without testing.
> > 
> > I can try to put something together for people to test soonish.
> > 
> > Matt
> > 
> > > > > 
> > > > > Matt 
> > > > > 
> > > > > > > +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> > > > > > > +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> > > > > > > +	}
> > > > > > > +
> > > > > > >  	/*
> > > > > > >  	 * Drivers shouldn't be allocating pages after calling
> > > > > > >  	 * memunmap_pages().
> > > > > > > -- 
> > > > > > > 2.43.0
> > > > > > >

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Rodrigo Vivi 3 weeks, 3 days ago

On Thu, Jan 15, 2026 at 10:35:56PM -0800, Matthew Brost wrote:
> On Thu, Jan 15, 2026 at 10:05:00PM +1100, Alistair Popple wrote:
> > On 2026-01-15 at 18:43 +1100, Matthew Brost <matthew.brost@intel.com> wrote...
> > > On Thu, Jan 15, 2026 at 06:07:08PM +1100, Alistair Popple wrote:
> > > > On 2026-01-15 at 17:18 +1100, Matthew Brost <matthew.brost@intel.com> wrote...
> > > > > On Wed, Jan 14, 2026 at 09:57:31PM -0800, Matthew Brost wrote:
> > > > > > On Thu, Jan 15, 2026 at 04:27:26PM +1100, Alistair Popple wrote:
> > > > > > > On 2026-01-15 at 06:19 +1100, Francois Dugast <francois.dugast@intel.com> wrote...
> > > > > > > > From: Matthew Brost <matthew.brost@intel.com>
> > > > > > > > 
> > > > > > > > Reinitialize metadata for large zone device private folios in
> > > > > > > > zone_device_page_init prior to creating a higher-order zone device
> > > > > > > > private folio. This step is necessary when the folio’s order changes
> > > > > > > > dynamically between zone_device_page_init calls to avoid building a
> > > > > > > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > > > > > > must be passed in from the caller because the pgmap stored in the folio
> > > > > > > > page may have been overwritten with a compound head.
> > > > > > > 
> > > > > > > Thanks for fixing, a couple of minor comments below.
> > > > > > > 
> > > > > > > > Cc: Zi Yan <ziy@nvidia.com>
> > > > > > > > Cc: Alistair Popple <apopple@nvidia.com>
> > > > > > > > Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> > > > > > > > Cc: Nicholas Piggin <npiggin@gmail.com>
> > > > > > > > Cc: Michael Ellerman <mpe@ellerman.id.au>
> > > > > > > > Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> > > > > > > > Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> > > > > > > > Cc: Alex Deucher <alexander.deucher@amd.com>
> > > > > > > > Cc: "Christian König" <christian.koenig@amd.com>
> > > > > > > > Cc: David Airlie <airlied@gmail.com>
> > > > > > > > Cc: Simona Vetter <simona@ffwll.ch>
> > > > > > > > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > > > > > > > Cc: Maxime Ripard <mripard@kernel.org>
> > > > > > > > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > > > > > > > Cc: Lyude Paul <lyude@redhat.com>
> > > > > > > > Cc: Danilo Krummrich <dakr@kernel.org>
> > > > > > > > Cc: David Hildenbrand <david@kernel.org>
> > > > > > > > Cc: Oscar Salvador <osalvador@suse.de>
> > > > > > > > Cc: Andrew Morton <akpm@linux-foundation.org>
> > > > > > > > Cc: Jason Gunthorpe <jgg@ziepe.ca>
> > > > > > > > Cc: Leon Romanovsky <leon@kernel.org>
> > > > > > > > Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> > > > > > > > Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> > > > > > > > Cc: Vlastimil Babka <vbabka@suse.cz>
> > > > > > > > Cc: Mike Rapoport <rppt@kernel.org>
> > > > > > > > Cc: Suren Baghdasaryan <surenb@google.com>
> > > > > > > > Cc: Michal Hocko <mhocko@suse.com>
> > > > > > > > Cc: Balbir Singh <balbirs@nvidia.com>
> > > > > > > > Cc: linuxppc-dev@lists.ozlabs.org
> > > > > > > > Cc: kvm@vger.kernel.org
> > > > > > > > Cc: linux-kernel@vger.kernel.org
> > > > > > > > Cc: amd-gfx@lists.freedesktop.org
> > > > > > > > Cc: dri-devel@lists.freedesktop.org
> > > > > > > > Cc: nouveau@lists.freedesktop.org
> > > > > > > > Cc: linux-mm@kvack.org
> > > > > > > > Cc: linux-cxl@vger.kernel.org
> > > > > > > > Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> > > > > > > > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > > > > > > > Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> > > > > > > > ---
> > > > > > > >  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
> > > > > > > >  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
> > > > > > > >  drivers/gpu/drm/drm_pagemap.c            |  2 +-
> > > > > > > >  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
> > > > > > > >  include/linux/memremap.h                 |  9 ++++++---
> > > > > > > >  lib/test_hmm.c                           |  4 +++-
> > > > > > > >  mm/memremap.c                            | 20 +++++++++++++++++++-
> > > > > > > >  7 files changed, 32 insertions(+), 9 deletions(-)
> > > > > > > > 
> > > > > > > > diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > > > > index e5000bef90f2..7cf9310de0ec 100644
> > > > > > > > --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > > > > +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > > > > > @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
> > > > > > > >  
> > > > > > > >  	dpage = pfn_to_page(uvmem_pfn);
> > > > > > > >  	dpage->zone_device_data = pvt;
> > > > > > > > -	zone_device_page_init(dpage, 0);
> > > > > > > > +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
> > > > > > > >  	return dpage;
> > > > > > > >  out_clear:
> > > > > > > >  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> > > > > > > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > > > > index af53e796ea1b..6ada7b4af7c6 100644
> > > > > > > > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > > > > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > > > > > @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
> > > > > > > >  	page = pfn_to_page(pfn);
> > > > > > > >  	svm_range_bo_ref(prange->svm_bo);
> > > > > > > >  	page->zone_device_data = prange->svm_bo;
> > > > > > > > -	zone_device_page_init(page, 0);
> > > > > > > > +	zone_device_page_init(page, page_pgmap(page), 0);
> > > > > > > >  }
> > > > > > > >  
> > > > > > > >  static void
> > > > > > > > diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> > > > > > > > index 03ee39a761a4..c497726b0147 100644
> > > > > > > > --- a/drivers/gpu/drm/drm_pagemap.c
> > > > > > > > +++ b/drivers/gpu/drm/drm_pagemap.c
> > > > > > > > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> > > > > > > >  					struct drm_pagemap_zdd *zdd)
> > > > > > > >  {
> > > > > > > >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > > > > > > > -	zone_device_page_init(page, 0);
> > > > > > > > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> > > > > > > >  }
> > > > > > > >  
> > > > > > > >  /**
> > > > > > > > diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > > > > index 58071652679d..3d8031296eed 100644
> > > > > > > > --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > > > > +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > > > > > @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
> > > > > > > >  			order = ilog2(DMEM_CHUNK_NPAGES);
> > > > > > > >  	}
> > > > > > > >  
> > > > > > > > -	zone_device_folio_init(folio, order);
> > > > > > > > +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
> > > > > > > >  	return page;
> > > > > > > >  }
> > > > > > > >  
> > > > > > > > diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> > > > > > > > index 713ec0435b48..e3c2ccf872a8 100644
> > > > > > > > --- a/include/linux/memremap.h
> > > > > > > > +++ b/include/linux/memremap.h
> > > > > > > > @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
> > > > > > > >  }
> > > > > > > >  
> > > > > > > >  #ifdef CONFIG_ZONE_DEVICE
> > > > > > > > -void zone_device_page_init(struct page *page, unsigned int order);
> > > > > > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > > > > > +			   unsigned int order);
> > > > > > > >  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);
> > > > > > > > @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
> > > > > > > >  
> > > > > > > >  unsigned long memremap_compat_align(void);
> > > > > > > >  
> > > > > > > > -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> > > > > > > > +static inline void zone_device_folio_init(struct folio *folio,
> > > > > > > > +					  struct dev_pagemap *pgmap,
> > > > > > > > +					  unsigned int order)
> > > > > > > >  {
> > > > > > > > -	zone_device_page_init(&folio->page, order);
> > > > > > > > +	zone_device_page_init(&folio->page, pgmap, order);
> > > > > > > >  	if (order)
> > > > > > > >  		folio_set_large_rmappable(folio);
> > > > > > > >  }
> > > > > > > > diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> > > > > > > > index 8af169d3873a..455a6862ae50 100644
> > > > > > > > --- a/lib/test_hmm.c
> > > > > > > > +++ b/lib/test_hmm.c
> > > > > > > > @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
> > > > > > > >  			goto error;
> > > > > > > >  	}
> > > > > > > >  
> > > > > > > > -	zone_device_folio_init(page_folio(dpage), order);
> > > > > > > > +	zone_device_folio_init(page_folio(dpage),
> > > > > > > > +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> > > > > > > > +			       order);
> > > > > > > >  	dpage->zone_device_data = rpage;
> > > > > > > >  	return dpage;
> > > > > > > >  
> > > > > > > > diff --git a/mm/memremap.c b/mm/memremap.c
> > > > > > > > index 63c6ab4fdf08..6f46ab14662b 100644
> > > > > > > > --- a/mm/memremap.c
> > > > > > > > +++ b/mm/memremap.c
> > > > > > > > @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
> > > > > > > >  	}
> > > > > > > >  }
> > > > > > > >  
> > > > > > > > -void zone_device_page_init(struct page *page, unsigned int order)
> > > > > > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > > > > > +			   unsigned int order)
> > > > > > > >  {
> > > > > > > > +	struct page *new_page = page;
> > > > > > > > +	unsigned int i;
> > > > > > > > +
> > > > > > > >  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> > > > > > > >  
> > > > > > > > +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > > > > > > > +		struct folio *new_folio = (struct folio *)new_page;
> > > > > > > > +
> > > > > > > > +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> > > > > > > 
> > > > > > > This seems odd to me, mainly due to the "magic" number. Why not just clear
> > > > > > > the flags entirely? Or at least explicitly just clear the flags you care about
> > > > > > > which would remove the need for the comment and  should let you use the proper
> > > > > > > PageFlag functions.
> > > > > > > 
> > > > > > 
> > > > > > I'm copying this from folio_reset_order [1]. My paranoia about touching
> > > > > > anything related to struct page is high, so I did the same thing
> > > > > > folio_reset_order does here.
> > > > 
> > > > So why not just use folio_reset_order() below?
> > > > 
> > > > > > 
> > > > > > [1] https://elixir.bootlin.com/linux/v6.18.5/source/include/linux/mm.h#L1075
> > > > > > 
> > > > > 
> > > > > This immediately hangs my first SVM test...
> > > > > 
> > > > > diff --git a/mm/memremap.c b/mm/memremap.c
> > > > > index 6f46ab14662b..ef8c56876cf5 100644
> > > > > --- a/mm/memremap.c
> > > > > +++ b/mm/memremap.c
> > > > > @@ -488,7 +488,7 @@ void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > >         for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > > > >                 struct folio *new_folio = (struct folio *)new_page;
> > > > > 
> > > > > -               new_page->flags.f &= ~0xffUL;   /* Clear possible order, page head */
> > > > > +               new_page->flags.f = 0;
> > > > >  #ifdef NR_PAGES_IN_LARGE_FOLIO
> > > > >                 ((struct folio *)(new_page - 1))->_nr_pages = 0;
> > > > 
> > > > This seems wrong to me - I saw your reply to Balbir but for an order-0 page
> > > > isn't this going to access a completely different, possibly already allocated,
> > > > page?
> > > > 
> > > 
> > > No — it accesses itself (new_page). It just uses some odd memory tricks
> > > for this, which I agree isn’t the best thing I’ve ever written, but it
> > > was the least-worst idea I had. I didn’t design the folio/page field
> > > aliasing; I understand why it exists, but it still makes my head hurt.
> > 
> > And obviously mine, because I (was) still not getting it and had typed up a
> > whole response and code walk through to show what was wrong, in the hope it
> > would help settle the misunderstanding. Which it did, because I discovered
> > where I was getting things wrong. But I've left the analysis below because it's
> > probably useful for others following along:
> > 
> > Walking through the code we have:
> > 
> > void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > 			   unsigned int order)
> > {
> > 
> > The first argument, page, is the first in a set of 1 << order contiguous
> > struct page. In the simplest case order == 0, meaning this function should only
> > initialise (ie. touch) a single struct page pointer which is passed as the first
> > argument to the function.
> 
> Yes.
> 
> > 
> > 	struct page *new_page = page;
> > 
> > So now *new_page points to the single struct page we should touch.
> > 	
> > 	unsigned int i;
> > 
> > 	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> > 
> > 	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > 
> > order == 0, so this loop will only execute once.
> > 
> 
> Yes.
> 
> > 		struct folio *new_folio = (struct folio *)new_page;
> > 
> > new_page still points to the single page we're initialising, and new_folio
> > points to the same page. Ie: &new_folio->page == new_page. There is a hazard
> > here because new_folio->__page_1, __page_2, etc. all point to pages we shouldn't
> > touch.
> > 
> 
> Yes.
> 
> > 		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> > 
> > Clears the flags, makes sense.
> > 
> 
> +1
> 
> > #ifdef NR_PAGES_IN_LARGE_FOLIO
> > 		((struct folio *)(new_page - 1))->_nr_pages = 0;
> > 
> > If we break this down we have:
> > 
> > struct page *tmp_new_page = new_page - 1;
> > 
> > Which is the page before the one we're initialising and shouldn't be touched.
> > Then we cast to a folio:
> > 
> > struct folio *tmp_new_folio = (struct folio *) tmp_new_page;
> > 
> > And reset _nr_pages:
> > 
> > tmp_new_folio->_nr_pages = 0
> > 
> > And now I can see where I was confused - &tmp_new_folio->_nr_pages == &tmp_new_folio->__page_1->memcg_data == &new_page->memcg_data
> > 
> 
> Not 100% right, as _nr_pages is 4 bytes and memcg_data is 8, but the
> pointer base address is the same.
> 
> > So after both Balbir, probably yourself, and definitely myself scratching our
> > heads for way too long over this change I think we can conclude that the code as
> > is is way too confusing to merge without a lot more comments :-)
> > 
> 
> I think more comments is the way to go. More below.
> 
> > However why go through all this magic in the first place? Why not just treat
> > everything here as a page and just do
> > 
> > 	new_page->memcg_data = 0
> > 
> 
> Well, memcg_data is 8 bytes and _nr_pages is 4. They also have different
> #ifdef conditions around each field, etc.
> 
> I’ve also seen failures in our testing, and so has François, with the
> memcg_data change. I wish I had a stack trace to share or explain, but
> the times I hit the error I didn’t capture the dmesg, and I’ve been
> having issues with my dev machine today. If I catch the error again,
> I’ll reply with a stack trace and analysis.
> 
> > directly? That seems like the more straight forward approach. In fact given
> > all the confusion I wonder if it wouldn't be better to just do
> > memset(new_page, 0, sizeof(*new_page)) and reinitialise everything from
> > scratch.
> 
> I had considered this option too, but I’d be a little concerned about
> the performance. Reinitializing a zone page/folio is a hot path, as this
> is typically done in a GPU fault handler. I think adding verbose
> comments explaining why this works, plus some follow-up helpers, might
> be the better option.
> 
> > 
> > > folio->_nr_pages is page + 1 for reference (new_page after this math).
> > > Again, if I touched this memory directly in new_page, it’s most likely
> > > memcg_data, but that is hidden behind a Kconfig.
> > > 
> > > This just blindly implementing part of folio_reset_order which clears
> > > _nr_pages.
> > 
> > Yeah, I get it now. But I think just clearing memcg_data would be the easiest to
> > understand approach, especially if it had a comment explaining that it may have
> > previously been used for _nr_pages.
> > 
> 
> See above — the different sizes, the failure I’m seeing, and the
> conflicting #ifdefs are why this is not my preferred option.
> 
> > > > >  #endif
> > > > > 
> > > > > I can walk through exactly which flags need to be cleared, but my
> > > > > feeling is that likely any flag that the order field overloads and can
> > > > > possibly encode should be cleared—so bits 0–7 based on the existing
> > > > > code.
> > > > > 
> > > > > How about in a follow-up we normalize setting / clearing the order flag
> > > > > field with a #define and an inline helper?
> > > > 
> > > > Ie: Would something like the following work:
> > > > 
> > > > 		ClearPageHead(new_page);
> > > 
> > > Any of these bit could possibly be set the order field in a folio, which
> > > modifies page + 1 flags field.
> > > 
> > > 	PG_locked,		/* Page is locked. Don't touch. */
> > > 	PG_writeback,		/* Page is under writeback */
> > > 	PG_referenced,
> > > 	PG_uptodate,
> > > 	PG_dirty,
> > > 	PG_lru,
> > > 	PG_head,		/* Must be in bit 6 */
> > > 	PG_waiters,		/* Page has waiters, check its waitqueue. Must be bit #7 and in the same byte as "PG_locked" */
> > > 
> > > So a common order-9 (2MB) folio would have PG_locked | PG_uptodate set.
> > > Now we get stuck on the next page lock because PG_locked is set.
> > > Offhand, I don’t know if different orders—which set different bits—cause
> > > any nasty issues either. So I figured the safest thing was clear any
> > > bits which folio order can set within subsequent page's memory flags
> > > like folio_reset_order does.
> > 
> > Oh, I get the above. I was thinking folio_reset_order() below would clear the
> > flags, but I see the folly there - that resets the flags for the next page.
> > 
> 
> Correct.
> 
> > > 
> > > > 		clear_compound_head(new_page);
> > > > 		folio_reset_order(new_folio);
> > > > 
> > > > Which would also deal with setting _nr_pages.
> > > >
> > > 
> > > folio_reset_order(new_folio) would set _nr_pages in the memory that is
> > > new_page + 1. So let's say that page has a ref count + memcg_data, now
> > > that memory is corrupted and will crash the kernel.
> > 
> > Yep, I just noticed that. Thanks for pointing that out.
> > 
> > > All of the above is why is took me multiple hours to write 6 lines of
> > > code :).
> > 
> > And to review :) Good thing we don't get paid per SLOC of code right?
> >
> 
> I don’t think anyone would touch core MM if pay were based on SLOC — it
> would be a terrible career choice. :)
> 
> All joking aside, I think the next revision should use this version,
> plus more comments and helpers/defines in a follow-up—which I’ll commit
> to—along with fixing the branch mismatch Andrew pointed out between
> drm-tip (which this series is based on) and 6.19 (where this patch needs
> to apply).

I believe the best branch for this series would be drm-misc-next indeed.

But this patch in particular needs multiple acks to get through drm trees.
At least one from each block:

## arch/powerpc/kvm/book3s_hv_uvmem.c
Madhavan Srinivasan <maddy@linux.ibm.com> (maintainer:KERNEL VIRTUAL MACHINE FOR POWERPC (KVM/powerpc))
Nicholas Piggin <npiggin@gmail.com> (reviewer:KERNEL VIRTUAL MACHINE FOR POWERPC (KVM/powerpc))
Michael Ellerman <mpe@ellerman.id.au> (maintainer:LINUX FOR POWERPC (32-BIT AND 64-BIT))
"Christophe Leroy (CS GROUP)" <chleroy@kernel.org> (reviewer:LINUX FOR POWERPC (32-BIT AND 64-BIT))

## include/linux/memremap.h
David Hildenbrand <david@kernel.org> (maintainer:MEMORY HOT(UN)PLUG)
Oscar Salvador <osalvador@suse.de> (maintainer:MEMORY HOT(UN)PLUG)

## lib/test_hmm.c
Andrew Morton <akpm@linux-foundation.org> (maintainer:LIBRARY CODE)
Jason Gunthorpe <jgg@ziepe.ca> (maintainer:HMM - Heterogeneous Memory Management)
Leon Romanovsky <leon@kernel.org> (maintainer:HMM - Heterogeneous Memory Management)

## mm/memremap.c
David Hildenbrand <david@kernel.org> (maintainer:MEMORY HOT(UN)PLUG)
Oscar Salvador <osalvador@suse.de> (maintainer:MEMORY HOT(UN)PLUG)
Andrew Morton <akpm@linux-foundation.org> (maintainer:MEMORY MANAGEMENT)


On the other hand we would also need Max to do one extra last pull-request
towards 7.0 after we get this merged. Because our window in drm closed
earlier.

Or this patch goes to any regular mm tree, and we hold the drm ones
after we backmerge 7.0-rc1

> 
> Matt
> 
> >  - Alistair
> > 
> > > > > Matt
> > > > > 
> > > > > > > > +#ifdef NR_PAGES_IN_LARGE_FOLIO
> > > > > > > > +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> > > > > > > > +#endif
> > > > > > > > +		new_folio->mapping = NULL;
> > > > > > > > +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> > > > > > > > +		new_folio->share = 0;   /* fsdax only, unused for device private */
> > > > > > > 
> > > > > > > It would be nice if the FS DAX code actually used this as well. Is there a
> > > > > > > reason that change was dropped from the series?
> > > > > > > 
> > > > > > 
> > > > > > I don't have a test platform for FS DAX. In prior revisions, I was just
> > > > > > moving existing FS DAX code to a helper, which I felt confident about.
> > > > > > 
> > > > > > This revision is slightly different, and I don't feel comfortable
> > > > > > modifying FS DAX code without a test platform. I agree we should update
> > > > > > FS DAX, but that should be done in a follow-up with coordinated testing.
> > > > 
> > > > Fair enough, I figured something like this might be your answer :-) You
> > > > could update it and ask people with access to such a system to test it though
> > > > (unfortunately my setup has bit-rotted beyond repair).
> > > > 
> > > > But I'm ok leaving to for a future change.
> > > >
> > > 
> > > I did a quick grep in fs/dax.c and don’t see zone_device_page_init
> > > called there. It probably could be used if it’s creating compound pages
> > > and drop the open-coded reinit when shared == 0, but yeah, that’s not
> > > something I can blindly code without testing.
> > > 
> > > I can try to put something together for people to test soonish.
> > > 
> > > Matt
> > > 
> > > > > > 
> > > > > > Matt 
> > > > > > 
> > > > > > > > +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> > > > > > > > +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> > > > > > > > +	}
> > > > > > > > +
> > > > > > > >  	/*
> > > > > > > >  	 * Drivers shouldn't be allocating pages after calling
> > > > > > > >  	 * memunmap_pages().
> > > > > > > > -- 
> > > > > > > > 2.43.0
> > > > > > > >

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Balbir Singh 3 weeks, 4 days ago

On 1/15/26 18:07, Alistair Popple wrote:
> On 2026-01-15 at 17:18 +1100, Matthew Brost <matthew.brost@intel.com> wrote...
>> On Wed, Jan 14, 2026 at 09:57:31PM -0800, Matthew Brost wrote:
>>> On Thu, Jan 15, 2026 at 04:27:26PM +1100, Alistair Popple wrote:
>>>> On 2026-01-15 at 06:19 +1100, Francois Dugast <francois.dugast@intel.com> wrote...
>>>>> From: Matthew Brost <matthew.brost@intel.com>
>>>>>
>>>>> Reinitialize metadata for large zone device private folios in
>>>>> zone_device_page_init prior to creating a higher-order zone device
>>>>> private folio. This step is necessary when the folio’s order changes
>>>>> dynamically between zone_device_page_init calls to avoid building a
>>>>> corrupt folio. As part of the metadata reinitialization, the dev_pagemap
>>>>> must be passed in from the caller because the pgmap stored in the folio
>>>>> page may have been overwritten with a compound head.
>>>>
>>>> Thanks for fixing, a couple of minor comments below.
>>>>
>>>>> Cc: Zi Yan <ziy@nvidia.com>
>>>>> Cc: Alistair Popple <apopple@nvidia.com>
>>>>> Cc: adhavan Srinivasan <maddy@linux.ibm.com>
>>>>> Cc: Nicholas Piggin <npiggin@gmail.com>
>>>>> Cc: Michael Ellerman <mpe@ellerman.id.au>
>>>>> Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
>>>>> Cc: Felix Kuehling <Felix.Kuehling@amd.com>
>>>>> Cc: Alex Deucher <alexander.deucher@amd.com>
>>>>> Cc: "Christian König" <christian.koenig@amd.com>
>>>>> Cc: David Airlie <airlied@gmail.com>
>>>>> Cc: Simona Vetter <simona@ffwll.ch>
>>>>> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
>>>>> Cc: Maxime Ripard <mripard@kernel.org>
>>>>> Cc: Thomas Zimmermann <tzimmermann@suse.de>
>>>>> Cc: Lyude Paul <lyude@redhat.com>
>>>>> Cc: Danilo Krummrich <dakr@kernel.org>
>>>>> Cc: David Hildenbrand <david@kernel.org>
>>>>> Cc: Oscar Salvador <osalvador@suse.de>
>>>>> Cc: Andrew Morton <akpm@linux-foundation.org>
>>>>> Cc: Jason Gunthorpe <jgg@ziepe.ca>
>>>>> Cc: Leon Romanovsky <leon@kernel.org>
>>>>> Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
>>>>> Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
>>>>> Cc: Vlastimil Babka <vbabka@suse.cz>
>>>>> Cc: Mike Rapoport <rppt@kernel.org>
>>>>> Cc: Suren Baghdasaryan <surenb@google.com>
>>>>> Cc: Michal Hocko <mhocko@suse.com>
>>>>> Cc: Balbir Singh <balbirs@nvidia.com>
>>>>> Cc: linuxppc-dev@lists.ozlabs.org
>>>>> Cc: kvm@vger.kernel.org
>>>>> Cc: linux-kernel@vger.kernel.org
>>>>> Cc: amd-gfx@lists.freedesktop.org
>>>>> Cc: dri-devel@lists.freedesktop.org
>>>>> Cc: nouveau@lists.freedesktop.org
>>>>> Cc: linux-mm@kvack.org
>>>>> Cc: linux-cxl@vger.kernel.org
>>>>> Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
>>>>> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
>>>>> Signed-off-by: Francois Dugast <francois.dugast@intel.com>
>>>>> ---
>>>>>  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
>>>>>  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
>>>>>  drivers/gpu/drm/drm_pagemap.c            |  2 +-
>>>>>  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
>>>>>  include/linux/memremap.h                 |  9 ++++++---
>>>>>  lib/test_hmm.c                           |  4 +++-
>>>>>  mm/memremap.c                            | 20 +++++++++++++++++++-
>>>>>  7 files changed, 32 insertions(+), 9 deletions(-)
>>>>>
>>>>> diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
>>>>> index e5000bef90f2..7cf9310de0ec 100644
>>>>> --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
>>>>> +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
>>>>> @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
>>>>>  
>>>>>  	dpage = pfn_to_page(uvmem_pfn);
>>>>>  	dpage->zone_device_data = pvt;
>>>>> -	zone_device_page_init(dpage, 0);
>>>>> +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
>>>>>  	return dpage;
>>>>>  out_clear:
>>>>>  	spin_lock(&kvmppc_uvmem_bitmap_lock);
>>>>> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
>>>>> index af53e796ea1b..6ada7b4af7c6 100644
>>>>> --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
>>>>> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
>>>>> @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
>>>>>  	page = pfn_to_page(pfn);
>>>>>  	svm_range_bo_ref(prange->svm_bo);
>>>>>  	page->zone_device_data = prange->svm_bo;
>>>>> -	zone_device_page_init(page, 0);
>>>>> +	zone_device_page_init(page, page_pgmap(page), 0);
>>>>>  }
>>>>>  
>>>>>  static void
>>>>> diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
>>>>> index 03ee39a761a4..c497726b0147 100644
>>>>> --- a/drivers/gpu/drm/drm_pagemap.c
>>>>> +++ b/drivers/gpu/drm/drm_pagemap.c
>>>>> @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
>>>>>  					struct drm_pagemap_zdd *zdd)
>>>>>  {
>>>>>  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
>>>>> -	zone_device_page_init(page, 0);
>>>>> +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
>>>>>  }
>>>>>  
>>>>>  /**
>>>>> diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
>>>>> index 58071652679d..3d8031296eed 100644
>>>>> --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
>>>>> +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
>>>>> @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
>>>>>  			order = ilog2(DMEM_CHUNK_NPAGES);
>>>>>  	}
>>>>>  
>>>>> -	zone_device_folio_init(folio, order);
>>>>> +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
>>>>>  	return page;
>>>>>  }
>>>>>  
>>>>> diff --git a/include/linux/memremap.h b/include/linux/memremap.h
>>>>> index 713ec0435b48..e3c2ccf872a8 100644
>>>>> --- a/include/linux/memremap.h
>>>>> +++ b/include/linux/memremap.h
>>>>> @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
>>>>>  }
>>>>>  
>>>>>  #ifdef CONFIG_ZONE_DEVICE
>>>>> -void zone_device_page_init(struct page *page, unsigned int order);
>>>>> +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
>>>>> +			   unsigned int order);
>>>>>  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);
>>>>> @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
>>>>>  
>>>>>  unsigned long memremap_compat_align(void);
>>>>>  
>>>>> -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
>>>>> +static inline void zone_device_folio_init(struct folio *folio,
>>>>> +					  struct dev_pagemap *pgmap,
>>>>> +					  unsigned int order)
>>>>>  {
>>>>> -	zone_device_page_init(&folio->page, order);
>>>>> +	zone_device_page_init(&folio->page, pgmap, order);
>>>>>  	if (order)
>>>>>  		folio_set_large_rmappable(folio);
>>>>>  }
>>>>> diff --git a/lib/test_hmm.c b/lib/test_hmm.c
>>>>> index 8af169d3873a..455a6862ae50 100644
>>>>> --- a/lib/test_hmm.c
>>>>> +++ b/lib/test_hmm.c
>>>>> @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
>>>>>  			goto error;
>>>>>  	}
>>>>>  
>>>>> -	zone_device_folio_init(page_folio(dpage), order);
>>>>> +	zone_device_folio_init(page_folio(dpage),
>>>>> +			       page_pgmap(folio_page(page_folio(dpage), 0)),
>>>>> +			       order);
>>>>>  	dpage->zone_device_data = rpage;
>>>>>  	return dpage;
>>>>>  
>>>>> diff --git a/mm/memremap.c b/mm/memremap.c
>>>>> index 63c6ab4fdf08..6f46ab14662b 100644
>>>>> --- a/mm/memremap.c
>>>>> +++ b/mm/memremap.c
>>>>> @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
>>>>>  	}
>>>>>  }
>>>>>  
>>>>> -void zone_device_page_init(struct page *page, unsigned int order)
>>>>> +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
>>>>> +			   unsigned int order)
>>>>>  {
>>>>> +	struct page *new_page = page;
>>>>> +	unsigned int i;
>>>>> +
>>>>>  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
>>>>>  
>>>>> +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
>>>>> +		struct folio *new_folio = (struct folio *)new_page;
>>>>> +
>>>>> +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
>>>>
>>>> This seems odd to me, mainly due to the "magic" number. Why not just clear
>>>> the flags entirely? Or at least explicitly just clear the flags you care about
>>>> which would remove the need for the comment and  should let you use the proper
>>>> PageFlag functions.
>>>>
>>>
>>> I'm copying this from folio_reset_order [1]. My paranoia about touching
>>> anything related to struct page is high, so I did the same thing
>>> folio_reset_order does here.
> 
> So why not just use folio_reset_order() below?
> 
>>>
>>> [1] https://elixir.bootlin.com/linux/v6.18.5/source/include/linux/mm.h#L1075
>>>
>>
>> This immediately hangs my first SVM test...
>>
>> diff --git a/mm/memremap.c b/mm/memremap.c
>> index 6f46ab14662b..ef8c56876cf5 100644
>> --- a/mm/memremap.c
>> +++ b/mm/memremap.c
>> @@ -488,7 +488,7 @@ void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
>>         for (i = 0; i < (1UL << order); ++i, ++new_page) {
>>                 struct folio *new_folio = (struct folio *)new_page;
>>
>> -               new_page->flags.f &= ~0xffUL;   /* Clear possible order, page head */
>> +               new_page->flags.f = 0;
>>  #ifdef NR_PAGES_IN_LARGE_FOLIO
>>                 ((struct folio *)(new_page - 1))->_nr_pages = 0;
> 
> This seems wrong to me - I saw your reply to Balbir but for an order-0 page
> isn't this going to access a completely different, possibly already allocated,
> page?
> 
>>  #endif
>>
>> I can walk through exactly which flags need to be cleared, but my
>> feeling is that likely any flag that the order field overloads and can
>> possibly encode should be cleared—so bits 0–7 based on the existing
>> code.
>>
>> How about in a follow-up we normalize setting / clearing the order flag
>> field with a #define and an inline helper?
> 
> Ie: Would something like the following work:
> 
> 		ClearPageHead(new_page);
> 		clear_compound_head(new_page);
> 		folio_reset_order(new_folio);
> 
> Which would also deal with setting _nr_pages.
> 

I thought about this, but folio_reset_order works only for larger folios otherwise
there is a VM_WARN_ON.

>> Matt
>>
>>>>> +#ifdef NR_PAGES_IN_LARGE_FOLIO
>>>>> +		((struct folio *)(new_page - 1))->_nr_pages = 0;
>>>>> +#endif
>>>>> +		new_folio->mapping = NULL;
>>>>> +		new_folio->pgmap = pgmap;	/* Also clear compound head */
>>>>> +		new_folio->share = 0;   /* fsdax only, unused for device private */
>>>>
>>>> It would be nice if the FS DAX code actually used this as well. Is there a
>>>> reason that change was dropped from the series?
>>>>
>>>
>>> I don't have a test platform for FS DAX. In prior revisions, I was just
>>> moving existing FS DAX code to a helper, which I felt confident about.
>>>
>>> This revision is slightly different, and I don't feel comfortable
>>> modifying FS DAX code without a test platform. I agree we should update
>>> FS DAX, but that should be done in a follow-up with coordinated testing.
> 
> Fair enough, I figured something like this might be your answer :-) You
> could update it and ask people with access to such a system to test it though
> (unfortunately my setup has bit-rotted beyond repair).
> 
> But I'm ok leaving to for a future change.
> 
>>>
>>> Matt 
>>>
>>>>> +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
>>>>> +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
>>>>> +	}
>>>>> +
>>>>>  	/*
>>>>>  	 * Drivers shouldn't be allocating pages after calling
>>>>>  	 * memunmap_pages().
>>>>> -- 
>>>>> 2.43.0
>>>>>

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Andrew Morton 3 weeks, 4 days ago

On Wed, 14 Jan 2026 20:19:52 +0100 Francois Dugast <francois.dugast@intel.com> wrote:

> From: Matthew Brost <matthew.brost@intel.com>
> 
> Reinitialize metadata for large zone device private folios in
> zone_device_page_init prior to creating a higher-order zone device
> private folio. This step is necessary when the folio’s order changes
> dynamically between zone_device_page_init calls to avoid building a
> corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> must be passed in from the caller because the pgmap stored in the folio
> page may have been overwritten with a compound head.
> 
> --- a/drivers/gpu/drm/drm_pagemap.c
> +++ b/drivers/gpu/drm/drm_pagemap.c
> @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
>  					struct drm_pagemap_zdd *zdd)
>  {
>  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> -	zone_device_page_init(page, 0);
> +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
>  }

drivers/gpu/drm/drm_pagemap.c:200:40: error: 'struct drm_pagemap_zdd' has no member named 'dpagemap'

I guess this was accidentally fixed in a later patch?

Please let's decide whether to fast-track the [1/N] fix into mainline
and if so, prepare something which compiles!

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Matthew Brost 3 weeks, 4 days ago

On Wed, Jan 14, 2026 at 07:01:54PM -0800, Andrew Morton wrote:
> On Wed, 14 Jan 2026 20:19:52 +0100 Francois Dugast <francois.dugast@intel.com> wrote:
> 
> > From: Matthew Brost <matthew.brost@intel.com>
> > 
> > Reinitialize metadata for large zone device private folios in
> > zone_device_page_init prior to creating a higher-order zone device
> > private folio. This step is necessary when the folio’s order changes
> > dynamically between zone_device_page_init calls to avoid building a
> > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > must be passed in from the caller because the pgmap stored in the folio
> > page may have been overwritten with a compound head.
> > 
> > --- a/drivers/gpu/drm/drm_pagemap.c
> > +++ b/drivers/gpu/drm/drm_pagemap.c
> > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> >  					struct drm_pagemap_zdd *zdd)
> >  {
> >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > -	zone_device_page_init(page, 0);
> > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> >  }
> 
> drivers/gpu/drm/drm_pagemap.c:200:40: error: 'struct drm_pagemap_zdd' has no member named 'dpagemap'
> 
> I guess this was accidentally fixed in a later patch?
> 

Ah, no. This is because we merged some in drm-tip which is not 6.19,
this is based on the drm-tip branch.

> Please let's decide whether to fast-track the [1/N] fix into mainline
> and if so, prepare something which compiles!

Maybe we just take this through the MM repo then? I suppose I should
send out patch which applies to the MM repo? I just cloned that repo.

Matt

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Matthew Brost 3 weeks, 4 days ago

On Wed, Jan 14, 2026 at 07:07:37PM -0800, Matthew Brost wrote:
> On Wed, Jan 14, 2026 at 07:01:54PM -0800, Andrew Morton wrote:
> > On Wed, 14 Jan 2026 20:19:52 +0100 Francois Dugast <francois.dugast@intel.com> wrote:
> > 
> > > From: Matthew Brost <matthew.brost@intel.com>
> > > 
> > > Reinitialize metadata for large zone device private folios in
> > > zone_device_page_init prior to creating a higher-order zone device
> > > private folio. This step is necessary when the folio’s order changes
> > > dynamically between zone_device_page_init calls to avoid building a
> > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > must be passed in from the caller because the pgmap stored in the folio
> > > page may have been overwritten with a compound head.
> > > 
> > > --- a/drivers/gpu/drm/drm_pagemap.c
> > > +++ b/drivers/gpu/drm/drm_pagemap.c
> > > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> > >  					struct drm_pagemap_zdd *zdd)
> > >  {
> > >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > > -	zone_device_page_init(page, 0);
> > > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> > >  }
> > 
> > drivers/gpu/drm/drm_pagemap.c:200:40: error: 'struct drm_pagemap_zdd' has no member named 'dpagemap'
> > 
> > I guess this was accidentally fixed in a later patch?
> > 
> 
> Ah, no. This is because we merged some in drm-tip which is not 6.19,
> this is based on the drm-tip branch.
> 
> > Please let's decide whether to fast-track the [1/N] fix into mainline
> > and if so, prepare something which compiles!
> 
> Maybe we just take this through the MM repo then? I suppose I should
> send out patch which applies to the MM repo? I just cloned that repo.
> 

Sorry, typing to fast. I believe have a patch structure that applies to
6.19, MM branches, and drm-tip. Just need to run CI on 3 branches :).

Matt

> Matt

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Balbir Singh 3 weeks, 4 days ago

On 1/15/26 06:19, Francois Dugast wrote:
> From: Matthew Brost <matthew.brost@intel.com>
> 
> Reinitialize metadata for large zone device private folios in
> zone_device_page_init prior to creating a higher-order zone device
> private folio. This step is necessary when the folio’s order changes
> dynamically between zone_device_page_init calls to avoid building a
> corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> must be passed in from the caller because the pgmap stored in the folio
> page may have been overwritten with a compound head.
> 
> Cc: Zi Yan <ziy@nvidia.com>
> Cc: Alistair Popple <apopple@nvidia.com>
> Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> Cc: Nicholas Piggin <npiggin@gmail.com>
> Cc: Michael Ellerman <mpe@ellerman.id.au>
> Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> Cc: Alex Deucher <alexander.deucher@amd.com>
> Cc: "Christian König" <christian.koenig@amd.com>
> Cc: David Airlie <airlied@gmail.com>
> Cc: Simona Vetter <simona@ffwll.ch>
> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> Cc: Maxime Ripard <mripard@kernel.org>
> Cc: Thomas Zimmermann <tzimmermann@suse.de>
> Cc: Lyude Paul <lyude@redhat.com>
> Cc: Danilo Krummrich <dakr@kernel.org>
> Cc: David Hildenbrand <david@kernel.org>
> Cc: Oscar Salvador <osalvador@suse.de>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Jason Gunthorpe <jgg@ziepe.ca>
> Cc: Leon Romanovsky <leon@kernel.org>
> Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> Cc: Vlastimil Babka <vbabka@suse.cz>
> Cc: Mike Rapoport <rppt@kernel.org>
> Cc: Suren Baghdasaryan <surenb@google.com>
> Cc: Michal Hocko <mhocko@suse.com>
> Cc: Balbir Singh <balbirs@nvidia.com>
> Cc: linuxppc-dev@lists.ozlabs.org
> Cc: kvm@vger.kernel.org
> Cc: linux-kernel@vger.kernel.org
> Cc: amd-gfx@lists.freedesktop.org
> Cc: dri-devel@lists.freedesktop.org
> Cc: nouveau@lists.freedesktop.org
> Cc: linux-mm@kvack.org
> Cc: linux-cxl@vger.kernel.org
> Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> ---
>  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
>  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
>  drivers/gpu/drm/drm_pagemap.c            |  2 +-
>  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
>  include/linux/memremap.h                 |  9 ++++++---
>  lib/test_hmm.c                           |  4 +++-
>  mm/memremap.c                            | 20 +++++++++++++++++++-
>  7 files changed, 32 insertions(+), 9 deletions(-)
> 
> diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> index e5000bef90f2..7cf9310de0ec 100644
> --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
>  
>  	dpage = pfn_to_page(uvmem_pfn);
>  	dpage->zone_device_data = pvt;
> -	zone_device_page_init(dpage, 0);
> +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
>  	return dpage;
>  out_clear:
>  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> index af53e796ea1b..6ada7b4af7c6 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
>  	page = pfn_to_page(pfn);
>  	svm_range_bo_ref(prange->svm_bo);
>  	page->zone_device_data = prange->svm_bo;
> -	zone_device_page_init(page, 0);
> +	zone_device_page_init(page, page_pgmap(page), 0);
>  }
>  
>  static void
> diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> index 03ee39a761a4..c497726b0147 100644
> --- a/drivers/gpu/drm/drm_pagemap.c
> +++ b/drivers/gpu/drm/drm_pagemap.c
> @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
>  					struct drm_pagemap_zdd *zdd)
>  {
>  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> -	zone_device_page_init(page, 0);
> +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
>  }
>  
>  /**
> diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> index 58071652679d..3d8031296eed 100644
> --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
>  			order = ilog2(DMEM_CHUNK_NPAGES);
>  	}
>  
> -	zone_device_folio_init(folio, order);
> +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
>  	return page;
>  }
>  
> diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> index 713ec0435b48..e3c2ccf872a8 100644
> --- a/include/linux/memremap.h
> +++ b/include/linux/memremap.h
> @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
>  }
>  
>  #ifdef CONFIG_ZONE_DEVICE
> -void zone_device_page_init(struct page *page, unsigned int order);
> +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> +			   unsigned int order);
>  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);
> @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
>  
>  unsigned long memremap_compat_align(void);
>  
> -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> +static inline void zone_device_folio_init(struct folio *folio,
> +					  struct dev_pagemap *pgmap,
> +					  unsigned int order)
>  {
> -	zone_device_page_init(&folio->page, order);
> +	zone_device_page_init(&folio->page, pgmap, order);
>  	if (order)
>  		folio_set_large_rmappable(folio);
>  }
> diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> index 8af169d3873a..455a6862ae50 100644
> --- a/lib/test_hmm.c
> +++ b/lib/test_hmm.c
> @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
>  			goto error;
>  	}
>  
> -	zone_device_folio_init(page_folio(dpage), order);
> +	zone_device_folio_init(page_folio(dpage),
> +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> +			       order);
>  	dpage->zone_device_data = rpage;
>  	return dpage;
>  
> diff --git a/mm/memremap.c b/mm/memremap.c
> index 63c6ab4fdf08..6f46ab14662b 100644
> --- a/mm/memremap.c
> +++ b/mm/memremap.c
> @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
>  	}
>  }
>  
> -void zone_device_page_init(struct page *page, unsigned int order)
> +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> +			   unsigned int order)
>  {
> +	struct page *new_page = page;
> +	unsigned int i;
> +
>  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
>  
> +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> +		struct folio *new_folio = (struct folio *)new_page;
> +
> +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> +#ifdef NR_PAGES_IN_LARGE_FOLIO
> +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> +#endif

Not sure I follow the new_page - 1? What happens when order is 0?

> +		new_folio->mapping = NULL;
> +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> +		new_folio->share = 0;   /* fsdax only, unused for device private */
> +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> +	}
> +
>  	/*
>  	 * Drivers shouldn't be allocating pages after calling
>  	 * memunmap_pages().

I wish we did not have to pass in the pgmap, but I can see why
we can't rely on the existing pgmap

Balbir

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Matthew Brost 3 weeks, 4 days ago

On Thu, Jan 15, 2026 at 01:36:11PM +1100, Balbir Singh wrote:
> On 1/15/26 06:19, Francois Dugast wrote:
> > From: Matthew Brost <matthew.brost@intel.com>
> > 
> > Reinitialize metadata for large zone device private folios in
> > zone_device_page_init prior to creating a higher-order zone device
> > private folio. This step is necessary when the folio’s order changes
> > dynamically between zone_device_page_init calls to avoid building a
> > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > must be passed in from the caller because the pgmap stored in the folio
> > page may have been overwritten with a compound head.
> > 
> > Cc: Zi Yan <ziy@nvidia.com>
> > Cc: Alistair Popple <apopple@nvidia.com>
> > Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> > Cc: Nicholas Piggin <npiggin@gmail.com>
> > Cc: Michael Ellerman <mpe@ellerman.id.au>
> > Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> > Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> > Cc: Alex Deucher <alexander.deucher@amd.com>
> > Cc: "Christian König" <christian.koenig@amd.com>
> > Cc: David Airlie <airlied@gmail.com>
> > Cc: Simona Vetter <simona@ffwll.ch>
> > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > Cc: Maxime Ripard <mripard@kernel.org>
> > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > Cc: Lyude Paul <lyude@redhat.com>
> > Cc: Danilo Krummrich <dakr@kernel.org>
> > Cc: David Hildenbrand <david@kernel.org>
> > Cc: Oscar Salvador <osalvador@suse.de>
> > Cc: Andrew Morton <akpm@linux-foundation.org>
> > Cc: Jason Gunthorpe <jgg@ziepe.ca>
> > Cc: Leon Romanovsky <leon@kernel.org>
> > Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> > Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> > Cc: Vlastimil Babka <vbabka@suse.cz>
> > Cc: Mike Rapoport <rppt@kernel.org>
> > Cc: Suren Baghdasaryan <surenb@google.com>
> > Cc: Michal Hocko <mhocko@suse.com>
> > Cc: Balbir Singh <balbirs@nvidia.com>
> > Cc: linuxppc-dev@lists.ozlabs.org
> > Cc: kvm@vger.kernel.org
> > Cc: linux-kernel@vger.kernel.org
> > Cc: amd-gfx@lists.freedesktop.org
> > Cc: dri-devel@lists.freedesktop.org
> > Cc: nouveau@lists.freedesktop.org
> > Cc: linux-mm@kvack.org
> > Cc: linux-cxl@vger.kernel.org
> > Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> > ---
> >  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
> >  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
> >  drivers/gpu/drm/drm_pagemap.c            |  2 +-
> >  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
> >  include/linux/memremap.h                 |  9 ++++++---
> >  lib/test_hmm.c                           |  4 +++-
> >  mm/memremap.c                            | 20 +++++++++++++++++++-
> >  7 files changed, 32 insertions(+), 9 deletions(-)
> > 
> > diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > index e5000bef90f2..7cf9310de0ec 100644
> > --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> > +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
> >  
> >  	dpage = pfn_to_page(uvmem_pfn);
> >  	dpage->zone_device_data = pvt;
> > -	zone_device_page_init(dpage, 0);
> > +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
> >  	return dpage;
> >  out_clear:
> >  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > index af53e796ea1b..6ada7b4af7c6 100644
> > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
> >  	page = pfn_to_page(pfn);
> >  	svm_range_bo_ref(prange->svm_bo);
> >  	page->zone_device_data = prange->svm_bo;
> > -	zone_device_page_init(page, 0);
> > +	zone_device_page_init(page, page_pgmap(page), 0);
> >  }
> >  
> >  static void
> > diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> > index 03ee39a761a4..c497726b0147 100644
> > --- a/drivers/gpu/drm/drm_pagemap.c
> > +++ b/drivers/gpu/drm/drm_pagemap.c
> > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> >  					struct drm_pagemap_zdd *zdd)
> >  {
> >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > -	zone_device_page_init(page, 0);
> > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> >  }
> >  
> >  /**
> > diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > index 58071652679d..3d8031296eed 100644
> > --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
> >  			order = ilog2(DMEM_CHUNK_NPAGES);
> >  	}
> >  
> > -	zone_device_folio_init(folio, order);
> > +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
> >  	return page;
> >  }
> >  
> > diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> > index 713ec0435b48..e3c2ccf872a8 100644
> > --- a/include/linux/memremap.h
> > +++ b/include/linux/memremap.h
> > @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
> >  }
> >  
> >  #ifdef CONFIG_ZONE_DEVICE
> > -void zone_device_page_init(struct page *page, unsigned int order);
> > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > +			   unsigned int order);
> >  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);
> > @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
> >  
> >  unsigned long memremap_compat_align(void);
> >  
> > -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> > +static inline void zone_device_folio_init(struct folio *folio,
> > +					  struct dev_pagemap *pgmap,
> > +					  unsigned int order)
> >  {
> > -	zone_device_page_init(&folio->page, order);
> > +	zone_device_page_init(&folio->page, pgmap, order);
> >  	if (order)
> >  		folio_set_large_rmappable(folio);
> >  }
> > diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> > index 8af169d3873a..455a6862ae50 100644
> > --- a/lib/test_hmm.c
> > +++ b/lib/test_hmm.c
> > @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
> >  			goto error;
> >  	}
> >  
> > -	zone_device_folio_init(page_folio(dpage), order);
> > +	zone_device_folio_init(page_folio(dpage),
> > +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> > +			       order);
> >  	dpage->zone_device_data = rpage;
> >  	return dpage;
> >  
> > diff --git a/mm/memremap.c b/mm/memremap.c
> > index 63c6ab4fdf08..6f46ab14662b 100644
> > --- a/mm/memremap.c
> > +++ b/mm/memremap.c
> > @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
> >  	}
> >  }
> >  
> > -void zone_device_page_init(struct page *page, unsigned int order)
> > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > +			   unsigned int order)
> >  {
> > +	struct page *new_page = page;
> > +	unsigned int i;
> > +
> >  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> >  
> > +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > +		struct folio *new_folio = (struct folio *)new_page;
> > +
> > +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> > +#ifdef NR_PAGES_IN_LARGE_FOLIO
> > +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> > +#endif
> 
> Not sure I follow the new_page - 1? What happens when order is 0?
> 

This is just to get _nr_pages in the new_page as folio->_nr_pages is in
the folio's second page. So it just modifying itself. I agree this is a
bit goofy but couldn't think of a better way to do this. In the page
structure this is the memcg_data field on most builds.

Matt

> > +		new_folio->mapping = NULL;
> > +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> > +		new_folio->share = 0;   /* fsdax only, unused for device private */
> > +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> > +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> > +	}
> > +
> >  	/*
> >  	 * Drivers shouldn't be allocating pages after calling
> >  	 * memunmap_pages().
> 
> I wish we did not have to pass in the pgmap, but I can see why
> we can't rely on the existing pgmap
> 
> Balbir
>

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Alistair Popple 3 weeks, 4 days ago

On 2026-01-15 at 13:41 +1100, Matthew Brost <matthew.brost@intel.com> wrote...
> On Thu, Jan 15, 2026 at 01:36:11PM +1100, Balbir Singh wrote:
> > On 1/15/26 06:19, Francois Dugast wrote:
> > > From: Matthew Brost <matthew.brost@intel.com>
> > > 
> > > Reinitialize metadata for large zone device private folios in
> > > zone_device_page_init prior to creating a higher-order zone device
> > > private folio. This step is necessary when the folio’s order changes
> > > dynamically between zone_device_page_init calls to avoid building a
> > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > must be passed in from the caller because the pgmap stored in the folio
> > > page may have been overwritten with a compound head.
> > > 
> > > Cc: Zi Yan <ziy@nvidia.com>
> > > Cc: Alistair Popple <apopple@nvidia.com>
> > > Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> > > Cc: Nicholas Piggin <npiggin@gmail.com>
> > > Cc: Michael Ellerman <mpe@ellerman.id.au>
> > > Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> > > Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> > > Cc: Alex Deucher <alexander.deucher@amd.com>
> > > Cc: "Christian König" <christian.koenig@amd.com>
> > > Cc: David Airlie <airlied@gmail.com>
> > > Cc: Simona Vetter <simona@ffwll.ch>
> > > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > > Cc: Maxime Ripard <mripard@kernel.org>
> > > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > > Cc: Lyude Paul <lyude@redhat.com>
> > > Cc: Danilo Krummrich <dakr@kernel.org>
> > > Cc: David Hildenbrand <david@kernel.org>
> > > Cc: Oscar Salvador <osalvador@suse.de>
> > > Cc: Andrew Morton <akpm@linux-foundation.org>
> > > Cc: Jason Gunthorpe <jgg@ziepe.ca>
> > > Cc: Leon Romanovsky <leon@kernel.org>
> > > Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> > > Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> > > Cc: Vlastimil Babka <vbabka@suse.cz>
> > > Cc: Mike Rapoport <rppt@kernel.org>
> > > Cc: Suren Baghdasaryan <surenb@google.com>
> > > Cc: Michal Hocko <mhocko@suse.com>
> > > Cc: Balbir Singh <balbirs@nvidia.com>
> > > Cc: linuxppc-dev@lists.ozlabs.org
> > > Cc: kvm@vger.kernel.org
> > > Cc: linux-kernel@vger.kernel.org
> > > Cc: amd-gfx@lists.freedesktop.org
> > > Cc: dri-devel@lists.freedesktop.org
> > > Cc: nouveau@lists.freedesktop.org
> > > Cc: linux-mm@kvack.org
> > > Cc: linux-cxl@vger.kernel.org
> > > Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> > > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > > Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> > > ---
> > >  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
> > >  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
> > >  drivers/gpu/drm/drm_pagemap.c            |  2 +-
> > >  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
> > >  include/linux/memremap.h                 |  9 ++++++---
> > >  lib/test_hmm.c                           |  4 +++-
> > >  mm/memremap.c                            | 20 +++++++++++++++++++-
> > >  7 files changed, 32 insertions(+), 9 deletions(-)
> > > 
> > > diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > index e5000bef90f2..7cf9310de0ec 100644
> > > --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
> > >  
> > >  	dpage = pfn_to_page(uvmem_pfn);
> > >  	dpage->zone_device_data = pvt;
> > > -	zone_device_page_init(dpage, 0);
> > > +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
> > >  	return dpage;
> > >  out_clear:
> > >  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > index af53e796ea1b..6ada7b4af7c6 100644
> > > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
> > >  	page = pfn_to_page(pfn);
> > >  	svm_range_bo_ref(prange->svm_bo);
> > >  	page->zone_device_data = prange->svm_bo;
> > > -	zone_device_page_init(page, 0);
> > > +	zone_device_page_init(page, page_pgmap(page), 0);
> > >  }
> > >  
> > >  static void
> > > diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> > > index 03ee39a761a4..c497726b0147 100644
> > > --- a/drivers/gpu/drm/drm_pagemap.c
> > > +++ b/drivers/gpu/drm/drm_pagemap.c
> > > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> > >  					struct drm_pagemap_zdd *zdd)
> > >  {
> > >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > > -	zone_device_page_init(page, 0);
> > > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> > >  }
> > >  
> > >  /**
> > > diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > index 58071652679d..3d8031296eed 100644
> > > --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
> > >  			order = ilog2(DMEM_CHUNK_NPAGES);
> > >  	}
> > >  
> > > -	zone_device_folio_init(folio, order);
> > > +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
> > >  	return page;
> > >  }
> > >  
> > > diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> > > index 713ec0435b48..e3c2ccf872a8 100644
> > > --- a/include/linux/memremap.h
> > > +++ b/include/linux/memremap.h
> > > @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
> > >  }
> > >  
> > >  #ifdef CONFIG_ZONE_DEVICE
> > > -void zone_device_page_init(struct page *page, unsigned int order);
> > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > +			   unsigned int order);
> > >  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);
> > > @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
> > >  
> > >  unsigned long memremap_compat_align(void);
> > >  
> > > -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> > > +static inline void zone_device_folio_init(struct folio *folio,
> > > +					  struct dev_pagemap *pgmap,
> > > +					  unsigned int order)
> > >  {
> > > -	zone_device_page_init(&folio->page, order);
> > > +	zone_device_page_init(&folio->page, pgmap, order);
> > >  	if (order)
> > >  		folio_set_large_rmappable(folio);
> > >  }
> > > diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> > > index 8af169d3873a..455a6862ae50 100644
> > > --- a/lib/test_hmm.c
> > > +++ b/lib/test_hmm.c
> > > @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
> > >  			goto error;
> > >  	}
> > >  
> > > -	zone_device_folio_init(page_folio(dpage), order);
> > > +	zone_device_folio_init(page_folio(dpage),
> > > +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> > > +			       order);
> > >  	dpage->zone_device_data = rpage;
> > >  	return dpage;
> > >  
> > > diff --git a/mm/memremap.c b/mm/memremap.c
> > > index 63c6ab4fdf08..6f46ab14662b 100644
> > > --- a/mm/memremap.c
> > > +++ b/mm/memremap.c
> > > @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
> > >  	}
> > >  }
> > >  
> > > -void zone_device_page_init(struct page *page, unsigned int order)
> > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > +			   unsigned int order)
> > >  {
> > > +	struct page *new_page = page;
> > > +	unsigned int i;
> > > +
> > >  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> > >  
> > > +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > > +		struct folio *new_folio = (struct folio *)new_page;
> > > +
> > > +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> > > +#ifdef NR_PAGES_IN_LARGE_FOLIO
> > > +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> > > +#endif
> > 
> > Not sure I follow the new_page - 1? What happens when order is 0?
> > 
> 
> This is just to get _nr_pages in the new_page as folio->_nr_pages is in
> the folio's second page. So it just modifying itself. I agree this is a
> bit goofy but couldn't think of a better way to do this. In the page
> structure this is the memcg_data field on most builds.

I still don't follow - page == new_page == new_folio so isn't
&new_page->_nr_pages the same as &new_folio->_nr_pages? I don't understand why
we would care about the a second page here.

 - Alistair

> 
> Matt
> 
> > > +		new_folio->mapping = NULL;
> > > +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> > > +		new_folio->share = 0;   /* fsdax only, unused for device private */
> > > +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> > > +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> > > +	}
> > > +
> > >  	/*
> > >  	 * Drivers shouldn't be allocating pages after calling
> > >  	 * memunmap_pages().
> > 
> > I wish we did not have to pass in the pgmap, but I can see why
> > we can't rely on the existing pgmap
> > 
> > Balbir
> >

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Matthew Brost 3 weeks, 4 days ago

On Thu, Jan 15, 2026 at 06:13:15PM +1100, Alistair Popple wrote:
> On 2026-01-15 at 13:41 +1100, Matthew Brost <matthew.brost@intel.com> wrote...
> > On Thu, Jan 15, 2026 at 01:36:11PM +1100, Balbir Singh wrote:
> > > On 1/15/26 06:19, Francois Dugast wrote:
> > > > From: Matthew Brost <matthew.brost@intel.com>
> > > > 
> > > > Reinitialize metadata for large zone device private folios in
> > > > zone_device_page_init prior to creating a higher-order zone device
> > > > private folio. This step is necessary when the folio’s order changes
> > > > dynamically between zone_device_page_init calls to avoid building a
> > > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > > must be passed in from the caller because the pgmap stored in the folio
> > > > page may have been overwritten with a compound head.
> > > > 
> > > > Cc: Zi Yan <ziy@nvidia.com>
> > > > Cc: Alistair Popple <apopple@nvidia.com>
> > > > Cc: adhavan Srinivasan <maddy@linux.ibm.com>
> > > > Cc: Nicholas Piggin <npiggin@gmail.com>
> > > > Cc: Michael Ellerman <mpe@ellerman.id.au>
> > > > Cc: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
> > > > Cc: Felix Kuehling <Felix.Kuehling@amd.com>
> > > > Cc: Alex Deucher <alexander.deucher@amd.com>
> > > > Cc: "Christian König" <christian.koenig@amd.com>
> > > > Cc: David Airlie <airlied@gmail.com>
> > > > Cc: Simona Vetter <simona@ffwll.ch>
> > > > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > > > Cc: Maxime Ripard <mripard@kernel.org>
> > > > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > > > Cc: Lyude Paul <lyude@redhat.com>
> > > > Cc: Danilo Krummrich <dakr@kernel.org>
> > > > Cc: David Hildenbrand <david@kernel.org>
> > > > Cc: Oscar Salvador <osalvador@suse.de>
> > > > Cc: Andrew Morton <akpm@linux-foundation.org>
> > > > Cc: Jason Gunthorpe <jgg@ziepe.ca>
> > > > Cc: Leon Romanovsky <leon@kernel.org>
> > > > Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> > > > Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
> > > > Cc: Vlastimil Babka <vbabka@suse.cz>
> > > > Cc: Mike Rapoport <rppt@kernel.org>
> > > > Cc: Suren Baghdasaryan <surenb@google.com>
> > > > Cc: Michal Hocko <mhocko@suse.com>
> > > > Cc: Balbir Singh <balbirs@nvidia.com>
> > > > Cc: linuxppc-dev@lists.ozlabs.org
> > > > Cc: kvm@vger.kernel.org
> > > > Cc: linux-kernel@vger.kernel.org
> > > > Cc: amd-gfx@lists.freedesktop.org
> > > > Cc: dri-devel@lists.freedesktop.org
> > > > Cc: nouveau@lists.freedesktop.org
> > > > Cc: linux-mm@kvack.org
> > > > Cc: linux-cxl@vger.kernel.org
> > > > Fixes: d245f9b4ab80 ("mm/zone_device: support large zone device private folios")
> > > > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > > > Signed-off-by: Francois Dugast <francois.dugast@intel.com>
> > > > ---
> > > >  arch/powerpc/kvm/book3s_hv_uvmem.c       |  2 +-
> > > >  drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
> > > >  drivers/gpu/drm/drm_pagemap.c            |  2 +-
> > > >  drivers/gpu/drm/nouveau/nouveau_dmem.c   |  2 +-
> > > >  include/linux/memremap.h                 |  9 ++++++---
> > > >  lib/test_hmm.c                           |  4 +++-
> > > >  mm/memremap.c                            | 20 +++++++++++++++++++-
> > > >  7 files changed, 32 insertions(+), 9 deletions(-)
> > > > 
> > > > diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > index e5000bef90f2..7cf9310de0ec 100644
> > > > --- a/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > +++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
> > > > @@ -723,7 +723,7 @@ static struct page *kvmppc_uvmem_get_page(unsigned long gpa, struct kvm *kvm)
> > > >  
> > > >  	dpage = pfn_to_page(uvmem_pfn);
> > > >  	dpage->zone_device_data = pvt;
> > > > -	zone_device_page_init(dpage, 0);
> > > > +	zone_device_page_init(dpage, &kvmppc_uvmem_pgmap, 0);
> > > >  	return dpage;
> > > >  out_clear:
> > > >  	spin_lock(&kvmppc_uvmem_bitmap_lock);
> > > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > index af53e796ea1b..6ada7b4af7c6 100644
> > > > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
> > > > @@ -217,7 +217,7 @@ svm_migrate_get_vram_page(struct svm_range *prange, unsigned long pfn)
> > > >  	page = pfn_to_page(pfn);
> > > >  	svm_range_bo_ref(prange->svm_bo);
> > > >  	page->zone_device_data = prange->svm_bo;
> > > > -	zone_device_page_init(page, 0);
> > > > +	zone_device_page_init(page, page_pgmap(page), 0);
> > > >  }
> > > >  
> > > >  static void
> > > > diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
> > > > index 03ee39a761a4..c497726b0147 100644
> > > > --- a/drivers/gpu/drm/drm_pagemap.c
> > > > +++ b/drivers/gpu/drm/drm_pagemap.c
> > > > @@ -201,7 +201,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
> > > >  					struct drm_pagemap_zdd *zdd)
> > > >  {
> > > >  	page->zone_device_data = drm_pagemap_zdd_get(zdd);
> > > > -	zone_device_page_init(page, 0);
> > > > +	zone_device_page_init(page, zdd->dpagemap->pagemap, 0);
> > > >  }
> > > >  
> > > >  /**
> > > > diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > index 58071652679d..3d8031296eed 100644
> > > > --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
> > > > @@ -425,7 +425,7 @@ nouveau_dmem_page_alloc_locked(struct nouveau_drm *drm, bool is_large)
> > > >  			order = ilog2(DMEM_CHUNK_NPAGES);
> > > >  	}
> > > >  
> > > > -	zone_device_folio_init(folio, order);
> > > > +	zone_device_folio_init(folio, page_pgmap(folio_page(folio, 0)), order);
> > > >  	return page;
> > > >  }
> > > >  
> > > > diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> > > > index 713ec0435b48..e3c2ccf872a8 100644
> > > > --- a/include/linux/memremap.h
> > > > +++ b/include/linux/memremap.h
> > > > @@ -224,7 +224,8 @@ static inline bool is_fsdax_page(const struct page *page)
> > > >  }
> > > >  
> > > >  #ifdef CONFIG_ZONE_DEVICE
> > > > -void zone_device_page_init(struct page *page, unsigned int order);
> > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > +			   unsigned int order);
> > > >  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);
> > > > @@ -234,9 +235,11 @@ bool pgmap_pfn_valid(struct dev_pagemap *pgmap, unsigned long pfn);
> > > >  
> > > >  unsigned long memremap_compat_align(void);
> > > >  
> > > > -static inline void zone_device_folio_init(struct folio *folio, unsigned int order)
> > > > +static inline void zone_device_folio_init(struct folio *folio,
> > > > +					  struct dev_pagemap *pgmap,
> > > > +					  unsigned int order)
> > > >  {
> > > > -	zone_device_page_init(&folio->page, order);
> > > > +	zone_device_page_init(&folio->page, pgmap, order);
> > > >  	if (order)
> > > >  		folio_set_large_rmappable(folio);
> > > >  }
> > > > diff --git a/lib/test_hmm.c b/lib/test_hmm.c
> > > > index 8af169d3873a..455a6862ae50 100644
> > > > --- a/lib/test_hmm.c
> > > > +++ b/lib/test_hmm.c
> > > > @@ -662,7 +662,9 @@ static struct page *dmirror_devmem_alloc_page(struct dmirror *dmirror,
> > > >  			goto error;
> > > >  	}
> > > >  
> > > > -	zone_device_folio_init(page_folio(dpage), order);
> > > > +	zone_device_folio_init(page_folio(dpage),
> > > > +			       page_pgmap(folio_page(page_folio(dpage), 0)),
> > > > +			       order);
> > > >  	dpage->zone_device_data = rpage;
> > > >  	return dpage;
> > > >  
> > > > diff --git a/mm/memremap.c b/mm/memremap.c
> > > > index 63c6ab4fdf08..6f46ab14662b 100644
> > > > --- a/mm/memremap.c
> > > > +++ b/mm/memremap.c
> > > > @@ -477,10 +477,28 @@ void free_zone_device_folio(struct folio *folio)
> > > >  	}
> > > >  }
> > > >  
> > > > -void zone_device_page_init(struct page *page, unsigned int order)
> > > > +void zone_device_page_init(struct page *page, struct dev_pagemap *pgmap,
> > > > +			   unsigned int order)
> > > >  {
> > > > +	struct page *new_page = page;
> > > > +	unsigned int i;
> > > > +
> > > >  	VM_WARN_ON_ONCE(order > MAX_ORDER_NR_PAGES);
> > > >  
> > > > +	for (i = 0; i < (1UL << order); ++i, ++new_page) {
> > > > +		struct folio *new_folio = (struct folio *)new_page;
> > > > +
> > > > +		new_page->flags.f &= ~0xffUL;	/* Clear possible order, page head */
> > > > +#ifdef NR_PAGES_IN_LARGE_FOLIO
> > > > +		((struct folio *)(new_page - 1))->_nr_pages = 0;
> > > > +#endif
> > > 
> > > Not sure I follow the new_page - 1? What happens when order is 0?
> > > 
> > 
> > This is just to get _nr_pages in the new_page as folio->_nr_pages is in
> > the folio's second page. So it just modifying itself. I agree this is a
> > bit goofy but couldn't think of a better way to do this. In the page
> > structure this is the memcg_data field on most builds.
> 
> I still don't follow - page == new_page == new_folio so isn't
> &new_page->_nr_pages the same as &new_folio->_nr_pages? I don't understand why
> we would care about the a second page here.
> 

I just replied to another email—this is quite confusing, but let me try
here...

Memory layout of a folio:

page0
page1 <-- this is where _nr_pages is
...

So ((struct folio *)(new_page - 1))->_nr_pages is pointing to memory at
new_page but using casting to determine the _nr_pages location.

At this point, we have no idea if _nr_pages in new_page was set by a
prior larger folio, so we just blindly clear it, which is safe. This is
no different than what folio_reset_order() does; we just do it for every
single page’s memory within the orderi passed in.

Matt

>  - Alistair
> 
> > 
> > Matt
> > 
> > > > +		new_folio->mapping = NULL;
> > > > +		new_folio->pgmap = pgmap;	/* Also clear compound head */
> > > > +		new_folio->share = 0;   /* fsdax only, unused for device private */
> > > > +		VM_WARN_ON_FOLIO(folio_ref_count(new_folio), new_folio);
> > > > +		VM_WARN_ON_FOLIO(!folio_is_zone_device(new_folio), new_folio);
> > > > +	}
> > > > +
> > > >  	/*
> > > >  	 * Drivers shouldn't be allocating pages after calling
> > > >  	 * memunmap_pages().
> > > 
> > > I wish we did not have to pass in the pgmap, but I can see why
> > > we can't rely on the existing pgmap
> > > 
> > > Balbir
> > >

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Andrew Morton 3 weeks, 4 days ago

On Wed, 14 Jan 2026 20:19:52 +0100 Francois Dugast <francois.dugast@intel.com> wrote:

> Reinitialize metadata for large zone device private folios in
> zone_device_page_init prior to creating a higher-order zone device
> private folio. This step is necessary when the folio’s order changes
> dynamically between zone_device_page_init calls to avoid building a
> corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> must be passed in from the caller because the pgmap stored in the folio
> page may have been overwritten with a compound head.

Thanks.  What are the worst-case userspace-visible effects of the bug?

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Matthew Brost 3 weeks, 4 days ago

On Wed, Jan 14, 2026 at 01:48:25PM -0800, Andrew Morton wrote:
> On Wed, 14 Jan 2026 20:19:52 +0100 Francois Dugast <francois.dugast@intel.com> wrote:
> 
> > Reinitialize metadata for large zone device private folios in
> > zone_device_page_init prior to creating a higher-order zone device
> > private folio. This step is necessary when the folio’s order changes
> > dynamically between zone_device_page_init calls to avoid building a
> > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > must be passed in from the caller because the pgmap stored in the folio
> > page may have been overwritten with a compound head.
> 
> Thanks.  What are the worst-case userspace-visible effects of the bug?

If you reallocate a subset of pages from what was originally a large
device folio, the pgmap mapping becomes invalid because it was
overwritten by the compound head, and this can crash the kernel.

Alternatively, consider the case where the original folio had an order
of 9 and _nr_pages was set. If you then reallocate the folio plus one as
an individual page, the flags would still have PG_locked set, causing a
hang the next time you try to lock the page.

This is pretty bad if drivers implement a buddy allocator for device
pages (Xe does; Nouveau doesn’t, which is why they haven’t hit this
issue). Only Nouveau enables large device pages in 6.19 but probably
best to have kernel flying around with known issues.

Matt

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Matthew Brost 3 weeks, 4 days ago

On Wed, Jan 14, 2026 at 03:34:21PM -0800, Matthew Brost wrote:
> On Wed, Jan 14, 2026 at 01:48:25PM -0800, Andrew Morton wrote:
> > On Wed, 14 Jan 2026 20:19:52 +0100 Francois Dugast <francois.dugast@intel.com> wrote:
> > 
> > > Reinitialize metadata for large zone device private folios in
> > > zone_device_page_init prior to creating a higher-order zone device
> > > private folio. This step is necessary when the folio’s order changes
> > > dynamically between zone_device_page_init calls to avoid building a
> > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > must be passed in from the caller because the pgmap stored in the folio
> > > page may have been overwritten with a compound head.
> > 
> > Thanks.  What are the worst-case userspace-visible effects of the bug?
> 
> If you reallocate a subset of pages from what was originally a large
> device folio, the pgmap mapping becomes invalid because it was
> overwritten by the compound head, and this can crash the kernel.
> 
> Alternatively, consider the case where the original folio had an order
> of 9 and _nr_pages was set. If you then reallocate the folio plus one as

s/_nr_pages/the order was encoded the page flags.

Not clearing _nr_pages is probably bad too, not sure what the side
affect of that is, but it can't be good. 

> an individual page, the flags would still have PG_locked set, causing a
> hang the next time you try to lock the page.
> 
> This is pretty bad if drivers implement a buddy allocator for device
> pages (Xe does; Nouveau doesn’t, which is why they haven’t hit this
> issue). Only Nouveau enables large device pages in 6.19 but probably
> best to have kernel flying around with known issues.

s/best to have kernel/best to not have kernels

Matt

> 
> Matt

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Andrew Morton 3 weeks, 4 days ago

On Wed, 14 Jan 2026 15:51:16 -0800 Matthew Brost <matthew.brost@intel.com> wrote:

> On Wed, Jan 14, 2026 at 03:34:21PM -0800, Matthew Brost wrote:
> > On Wed, Jan 14, 2026 at 01:48:25PM -0800, Andrew Morton wrote:
> > > On Wed, 14 Jan 2026 20:19:52 +0100 Francois Dugast <francois.dugast@intel.com> wrote:
> > > 
> > > > Reinitialize metadata for large zone device private folios in
> > > > zone_device_page_init prior to creating a higher-order zone device
> > > > private folio. This step is necessary when the folio’s order changes
> > > > dynamically between zone_device_page_init calls to avoid building a
> > > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > > must be passed in from the caller because the pgmap stored in the folio
> > > > page may have been overwritten with a compound head.
> > > 
> > > Thanks.  What are the worst-case userspace-visible effects of the bug?
> > 
> > If you reallocate a subset of pages from what was originally a large
> > device folio, the pgmap mapping becomes invalid because it was
> > overwritten by the compound head, and this can crash the kernel.
> > 
> > Alternatively, consider the case where the original folio had an order
> > of 9 and _nr_pages was set. If you then reallocate the folio plus one as
> 
> s/_nr_pages/the order was encoded the page flags.
> 
> ...
>
> s/best to have kernel/best to not have kernels
> 

Great, thanks.  I pasted all the above into the changelog to help
explain our reasons.  I'll retain the patch in mm-hotfixes, targeting
6.19-rcX.  The remainder of the series is DRM stuff, NotMyProblem.  I
assume that getting this into 6.19-rcX is helpful to DRM - if not, and
if taking this via the DRM tree is preferable then let's discuss.

Can reviewers please take a look at this reasonably promptly?


btw, this patch uses

+		struct folio *new_folio = (struct folio *)new_page;

Was page_folio() unsuitable?

Re: [PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios

Posted by Matthew Brost 3 weeks, 4 days ago

On Wed, Jan 14, 2026 at 06:40:42PM -0800, Andrew Morton wrote:
> On Wed, 14 Jan 2026 15:51:16 -0800 Matthew Brost <matthew.brost@intel.com> wrote:
> 
> > On Wed, Jan 14, 2026 at 03:34:21PM -0800, Matthew Brost wrote:
> > > On Wed, Jan 14, 2026 at 01:48:25PM -0800, Andrew Morton wrote:
> > > > On Wed, 14 Jan 2026 20:19:52 +0100 Francois Dugast <francois.dugast@intel.com> wrote:
> > > > 
> > > > > Reinitialize metadata for large zone device private folios in
> > > > > zone_device_page_init prior to creating a higher-order zone device
> > > > > private folio. This step is necessary when the folio’s order changes
> > > > > dynamically between zone_device_page_init calls to avoid building a
> > > > > corrupt folio. As part of the metadata reinitialization, the dev_pagemap
> > > > > must be passed in from the caller because the pgmap stored in the folio
> > > > > page may have been overwritten with a compound head.
> > > > 
> > > > Thanks.  What are the worst-case userspace-visible effects of the bug?
> > > 
> > > If you reallocate a subset of pages from what was originally a large
> > > device folio, the pgmap mapping becomes invalid because it was
> > > overwritten by the compound head, and this can crash the kernel.
> > > 
> > > Alternatively, consider the case where the original folio had an order
> > > of 9 and _nr_pages was set. If you then reallocate the folio plus one as
> > 
> > s/_nr_pages/the order was encoded the page flags.
> > 
> > ...
> >
> > s/best to have kernel/best to not have kernels
> > 
> 
> Great, thanks.  I pasted all the above into the changelog to help
> explain our reasons.  I'll retain the patch in mm-hotfixes, targeting
> 6.19-rcX.  The remainder of the series is DRM stuff, NotMyProblem.  I
> assume that getting this into 6.19-rcX is helpful to DRM - if not, and
> if taking this via the DRM tree is preferable then let's discuss.
> 

I would prefer to take this through DRM since our window for 7.0 closes
earlier than the rest of Linux (typically this Friday), which makes it
easier for me to merge the other four patches and include them in the
next PR. If we can't take it through DRM, I'm sure we can figure
something out - new as a maintainer here, so still figuring out all DRM
flows.

> Can reviewers please take a look at this reasonably promptly?
> 
> 
> btw, this patch uses
> 
> +		struct folio *new_folio = (struct folio *)new_page;
> 
> Was page_folio() unsuitable?
>

The compound head might be pointing somewhere else here, and we are
trying to clear the metadata from new_page up to order << 1. So we
explictly do not want to use page_folio here.

Matt

[PATCH v5 1/5] mm/zone_device: Reinitialize large zone device private folios