similar to: turn hmm migrate_vma upside down v2

Hi Jérôme, Ben and Jason, below is a series against the hmm tree which starts revamping the migrate_vma functionality. The prime idea is to export three slightly lower level functions and thus avoid the need for migrate_vma_ops callbacks. Diffstat: 4 files changed, 285 insertions(+), 602 deletions(-) A git tree is also available at: git://git.infradead.org/users/hch/misc.git

Hi Jérôme, Ben and Jason, below is a series against the hmm tree which starts revamping the migrate_vma functionality. The prime idea is to export three slightly lower level functions and thus avoid the need for migrate_vma_ops callbacks. Diffstat: 7 files changed, 282 insertions(+), 614 deletions(-) A git tree is also available at: git://git.infradead.org/users/hch/misc.git

This series adds support for transparent huge page migration to migrate_vma_*() and adds nouveau SVM and HMM selftests as consumers. Earlier versions were posted previously [1] and [2]. The patches apply cleanly to the linux-mm 5.10.0-rc2 tree. There are a lot of other THP patches being posted. I don't think there are any semantic conflicts but there may be some merge conflicts depending on

This series adds support for transparent huge page migration to migrate_vma_*() and adds nouveau SVM and HMM selftests as consumers. An earlier version was posted previously [1]. This version now supports splitting a THP midway in the migration process which led to a number of changes. The patches apply cleanly to the current linux-mm tree. Since there are a couple of patches in linux-mm from Dan

These patches apply to linux-5.8.0-rc1. Patches 1-3 should probably go into 5.8, the others can be queued for 5.9. Patches 4-6 improve the HMM self tests. Patch 7-8 prepare nouveau for the meat of this series which adds support and testing for compound page mapping of system memory (patches 9-11) and compound page migration to device private memory (patches 12-16). Since these changes are split

The goal for this series is to avoid device private memory TLB invalidations when migrating a range of addresses from system memory to device private memory and some of those pages have already been migrated. The approach taken is to introduce a new mmu notifier invalidation event type and use that in the device driver to skip invalidation callbacks from migrate_vma_setup(). The device driver is

The goal for this series is to avoid device private memory TLB invalidations when migrating a range of addresses from system memory to device private memory and some of those pages have already been migrated. The approach taken is to introduce a new mmu notifier invalidation event type and use that in the device driver to skip invalidation callbacks from migrate_vma_setup(). The device driver is

The goal for this series is to avoid device private memory TLB invalidations when migrating a range of addresses from system memory to device private memory and some of those pages have already been migrated. The approach taken is to introduce a new mmu notifier invalidation event type and use that in the device driver to skip invalidation callbacks from migrate_vma_setup(). The device driver is

The goal for this series is to avoid device private memory TLB invalidations when migrating a range of addresses from system memory to device private memory and some of those pages have already been migrated. The approach taken is to introduce a new mmu notifier invalidation event type and use that in the device driver to skip invalidation callbacks from migrate_vma_setup(). The device driver is

These changes are based on Jason's latest hmm branch. Patch 1 was previously posted here [1] but was dropped from the orginal series. Hopefully, the tests will reduce concerns about edge conditions. I'm sure more tests could be usefully added but I thought this was a good starting point. [1] https://lore.kernel.org/linux-mm/20190726005650.2566-6-rcampbell at nvidia.com/ Ralph Campbell

When memory is migrated to the GPU it is likely to be accessed by GPU code soon afterwards. Instead of waiting for a GPU fault, map the migrated memory into the GPU page tables with the same access permissions as the source CPU page table entries. This preserves copy on write semantics. Signed-off-by: Ralph Campbell <rcampbell at nvidia.com> Cc: Christoph Hellwig <hch at lst.de> Cc:

Originally patch 4 was targeted for Jason's rdma tree since other HMM related changes were queued there. Now that those have been merged, these patches just contain changes to nouveau so they could go through any tree. I guess Ben Skeggs' tree would be appropriate. Changes since v2: Added patches 1-3 to fix some minor issues. Eliminated nouveau_find_svmm() since it is easily found.

On 2020-06-22 16:38, Ralph Campbell wrote: > The OpenCL function clEnqueueSVMMigrateMem(), without any flags, will > migrate memory in the given address range to device private memory. The > source pages might already have been migrated to device private memory. > In that case, the source struct page is not checked to see if it is > a device private page and incorrectly computes the

These are based on 5.8.0-rc2 and intended for Ben Skeggs' nouveau tree. I believe the changes can be queued for 5.8-rcX after being reviewed. These were part of a larger series but I'm resending them separately as suggested by Jason Gunthorpe. https://lore.kernel.org/linux-mm/20200619215649.32297-1-rcampbell at nvidia.com/ Note that in order to exercise/test patch 2 here, you will need a

Factor the main copy page to ram routine out into a helper that acts on a single page and which doesn't require the nouveau_dmem_fault structure for argument passing. Also remove the loop over multiple pages as we only handle one at the moment, although the structure of the main worker function makes it relatively easy to add multi page support back if needed in the future. But at least for

On 8/8/19 8:33 AM, Christoph Hellwig wrote: > Factor the main copy page to ram routine out into a helper that acts on > a single page and which doesn't require the nouveau_dmem_fault > structure for argument passing. Also remove the loop over multiple > pages as we only handle one at the moment, although the structure of > the main worker function makes it relatively easy to

Factor the main copy page to vram routine out into a helper that acts on a single page and which doesn't require the nouveau_dmem_migrate structure for argument passing. As an added benefit the new version only allocates the dma address array once and reuses it for each subsequent chunk of work. Signed-off-by: Christoph Hellwig <hch at lst.de> ---

Factor the main copy page to ram routine out into a helper that acts on a single page and which doesn't require the nouveau_dmem_fault structure for argument passing. Also remove the loop over multiple pages as we only handle one at the moment, although the structure of the main worker function makes it relatively easy to add multi page support back if needed in the future. But at least for

The kcalloc() in nouveau_dmem_evict_chunk() will return null if the physical memory has run out. As a result, if we dereference src_pfns, dst_pfns or dma_addrs, the null pointer dereference bugs will happen. Moreover, the GPU is going away. If the kcalloc() fails, we could not evict all pages mapping a chunk. So this patch adds a __GFP_NOFAIL flag in kcalloc(). Finally, as there is no need to

On Mon, Jul 29, 2019 at 05:28:39PM +0300, Christoph Hellwig wrote: > Factor the main copy page to ram routine out into a helper that acts on > a single page and which doesn't require the nouveau_dmem_fault > structure for argument passing. Also remove the loop over multiple > pages as we only handle one at the moment, although the structure of > the main worker function makes