search for: walk_page_range

hmm_range_fault() calls find_vma() and walk_page_range() in a loop. This is unnecessary duplication since walk_page_range() calls find_vma() in a loop already. Simplify hmm_range_fault() by defining a walk_test() callback function to filter unhandled vmas. This also fixes a bug where hmm_range_fault() was not checking start >= vma->vm_start befor...

On Fri, Aug 23, 2019 at 03:17:52PM -0700, Ralph Campbell wrote: > Although hmm_range_fault() calls find_vma() to make sure that a vma exists > before calling walk_page_range(), hmm_vma_walk_hole() can still be called > with walk->vma == NULL if the start and end address are not contained > within the vma range. Should we convert to walk_vma_range instead? Or keep walk_page_range but drop searching the vma ourselves? Except for that the patch looks good to m...

hmm_range_fault() calls find_vma() and walk_page_range() in a loop. This is unnecessary duplication since walk_page_range() calls find_vma() in a loop already. Simplify hmm_range_fault() by defining a walk_test() callback function to filter unhandled vmas. Signed-off-by: Ralph Campbell <rcampbell at nvidia.com> Cc: "Jérôme Glisse" <...

..., 2019 at 11:02:12AM -0700, Ralph Campbell wrote: > > On 8/24/19 3:37 PM, Christoph Hellwig wrote: > > On Fri, Aug 23, 2019 at 03:17:52PM -0700, Ralph Campbell wrote: > > > Although hmm_range_fault() calls find_vma() to make sure that a vma exists > > > before calling walk_page_range(), hmm_vma_walk_hole() can still be called > > > with walk->vma == NULL if the start and end address are not contained > > > within the vma range. > > > > Should we convert to walk_vma_range instead? Or keep walk_page_range > > but drop searching the vma our...

> +static int hmm_pfns_fill(unsigned long addr, > + unsigned long end, > + struct hmm_range *range, > + enum hmm_pfn_value_e value) Nit: can we use the space a little more efficient, e.g.: static int hmm_pfns_fill(unsigned long addr, unsigned long end, struct hmm_range *range, enum hmm_pfn_value_e value) > +static int hmm_vma_walk_test(unsigned long start, > +

I have been working on converting Jerome's hmm_dummy driver and self tests into a stand-alone set of tests to be included in tools/testing/selftests/vm and came across these two bug fixes in the process. The tests aren't quite ready to be posted as a patch. I'm posting the fixes now since I thought they shouldn't wait. They should probably have a fixes line but with all the HMM

On 8/24/19 3:37 PM, Christoph Hellwig wrote: > On Fri, Aug 23, 2019 at 03:17:52PM -0700, Ralph Campbell wrote: >> Although hmm_range_fault() calls find_vma() to make sure that a vma exists >> before calling walk_page_range(), hmm_vma_walk_hole() can still be called >> with walk->vma == NULL if the start and end address are not contained >> within the vma range. > > Should we convert to walk_vma_range instead? Or keep walk_page_range > but drop searching the vma ourselves? > > Except f...

...oph Hellwig (2): mm/hmm: replace the block argument to hmm_range_fault with a flags value mm: merge hmm_range_snapshot into hmm_range_fault Ralph Campbell (5): mm/hmm: replace hmm_update with mmu_notifier_range mm/hmm: a few more C style and comment clean ups mm/hmm: make full use of walk_page_range() mm/hmm: remove hugetlbfs check in hmm_vma_walk_pmd mm/hmm: remove hmm_range vma Documentation/vm/hmm.rst | 17 +- drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c | 8 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 2 +- drivers/gpu/drm/nouveau/nouveau_svm.c | 13 +- include/lin...

Although hmm_range_fault() calls find_vma() to make sure that a vma exists before calling walk_page_range(), hmm_vma_walk_hole() can still be called with walk->vma == NULL if the start and end address are not contained within the vma range. hmm_range_fault() /* calls find_vma() but no range check */ walk_page_range() /* calls find_vma(), sets walk->vma = NULL */ __walk_page_range() wal...

walk_page_range() will only call hmm_vma_walk_hugetlb_entry() for hugetlbfs pages and doesn't call hmm_vma_walk_pmd() in this case. Therefore, it is safe to remove the check for vma->vm_flags & VM_HUGETLB in hmm_vma_walk_pmd(). Signed-off-by: Ralph Campbell <rcampbell at nvidia.com> Cc: "Jér...

...heck the vma->vm_flags first. hmm_range_fault() checks for VM_READ but doesn't check for VM_WRITE if the caller requests a page to be faulted in with write permission (via the hmm_range.pfns[] value). If the vma is write protected, this can result in an infinite loop: hmm_range_fault() walk_page_range() ... hmm_vma_walk_hole() hmm_vma_walk_hole_() hmm_vma_do_fault() handle_mm_fault(FAULT_FLAG_WRITE) /* returns VM_FAULT_WRITE */ /* returns -EBUSY */ /* returns -EBUSY */ /* returns -EBUSY */ /* loops on -EBUSY and ra...

...ge mappings that won't be migrated. + */ + mmu_notifier_range_init(&range, MMU_NOTIFY_MIGRATE, 0, migrate->vma, migrate->vma->vm_mm, migrate->start, migrate->end); + range.migrate_pgmap_owner = migrate->src_owner; mmu_notifier_invalidate_range_start(&range); walk_page_range(migrate->vma->vm_mm, migrate->start, migrate->end, -- 2.20.1

...+ * private page mappings that won't be migrated. + */ + mmu_notifier_range_init(&range, MMU_NOTIFY_MIGRATE, 0, migrate->vma, migrate->vma->vm_mm, migrate->start, migrate->end); + range.data = migrate->src_owner; mmu_notifier_invalidate_range_start(&range); walk_page_range(migrate->vma->vm_mm, migrate->start, migrate->end, -- 2.20.1

...mappings that won't be migrated. + */ + mmu_notifier_range_init(&range, MMU_NOTIFY_MIGRATE, 0, migrate->vma, migrate->vma->vm_mm, migrate->start, migrate->end); + range.migrate_pgmap_owner = migrate->pgmap_owner; mmu_notifier_invalidate_range_start(&range); walk_page_range(migrate->vma->vm_mm, migrate->start, migrate->end, -- 2.20.1

...mappings that won't be migrated. + */ + mmu_notifier_range_init(&range, MMU_NOTIFY_MIGRATE, 0, migrate->vma, migrate->vma->vm_mm, migrate->start, migrate->end); + range.migrate_pgmap_owner = migrate->pgmap_owner; mmu_notifier_invalidate_range_start(&range); walk_page_range(migrate->vma->vm_mm, migrate->start, migrate->end, -- 2.20.1

...pages, clearing the dcache for * them and setting the cache-inhibit bit. @@ -124,7 +124,7 @@ or1k_dma_free(struct device *dev, size_t size, void *vaddr, .mm = &init_mm }; - if (!dma_get_attr(DMA_ATTR_NON_CONSISTENT, attrs)) { + if ((attrs & DMA_ATTR_NON_CONSISTENT) == 0) { /* walk_page_range shouldn't be able to fail here */ WARN_ON(walk_page_range(va, va + size, &walk)); } diff --git a/arch/parisc/kernel/pci-dma.c b/arch/parisc/kernel/pci-dma.c index 845fdd52e4c5..02d9ed0f3949 100644 --- a/arch/parisc/kernel/pci-dma.c +++ b/arch/parisc/kernel/pci-dma.c @@ -581,7 +581,7 @@...

.../hmm-tests.c line 868 to > ASSERT_EQ(ret, -1); > Then the test will abort, core dump, and cause two problems, > 1) the migrated page will be faulted back to system memory in order to write > it to the core dump. This triggers lockdep_assert_held(&walk.mm->mmap_sem) > in walk_page_range(). Has the migration stuff become entangled with the xarray? > [ 137.980718] Code: 80 2f 1a 83 c6 05 e9 8d 7b 01 01 e8 3e b1 b1 fe e9 05 ff ff ff 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 41 56 41 55 41 54 55 <48> 89 fd 53 4c 8d 6d 10 e8 3c fc ff ff 49 89 c4 4c 89 e0 83 e0 03 > [ 13...

...* without mutating them (ie causing faults). */ -long hmm_range_fault(struct hmm_range *range) +int hmm_range_fault(struct hmm_range *range) { struct hmm_vma_walk hmm_vma_walk = { .range = range, @@ -590,10 +580,13 @@ long hmm_range_fault(struct hmm_range *range) return -EBUSY; ret = walk_page_range(mm, hmm_vma_walk.last, range->end, &hmm_walk_ops, &hmm_vma_walk); + /* + * When -EBUSY is returned the loop restarts with + * hmm_vma_walk.last set to an address that has not been stored + * in pfns. All entries < last in the pfn array are set to their + * output,...

...* without mutating them (ie causing faults). */ -long hmm_range_fault(struct hmm_range *range) +int hmm_range_fault(struct hmm_range *range) { struct hmm_vma_walk hmm_vma_walk = { .range = range, @@ -590,10 +580,13 @@ long hmm_range_fault(struct hmm_range *range) return -EBUSY; ret = walk_page_range(mm, hmm_vma_walk.last, range->end, &hmm_walk_ops, &hmm_vma_walk); + /* + * When -EBUSY is returned the loop restarts with + * hmm_vma_walk.last set to an address that has not been stored + * in pfns. All entries < last in the pfn array are set to their + * output,...

...ls/testing/selftests/vm/hmm-tests.c line 868 to ASSERT_EQ(ret, -1); Then the test will abort, core dump, and cause two problems, 1) the migrated page will be faulted back to system memory in order to write it to the core dump. This triggers lockdep_assert_held(&walk.mm->mmap_sem) in walk_page_range(). 2) Then after a delay, I get: [ 137.852986] rcu: INFO: rcu_sched self-detected stall on CPU [ 137.858594] rcu: 0-....: (26000 ticks this GP) idle=69e/1/0x4000000000000002 softirq=34555/34555 fqs=6497 [ 137.868439] (t=26007 jiffies g=14653 q=271) [ 137.872711] NMI backtrace for cpu 0 [ 137...