search for: find_get_mmu_notifi

...mas for the address range given in the ioctl() data, and migrates the pages to the GPU memory. mmu_notifier_get() would try to lock mmap_sem for writing so that would deadlock. But it is similar in that the GPU specific process context (nouveau_svmm) needs to be found for the given ioctl caller. If find_get_mmu_notifier() was exported, I think that could work. Now that I look at this again, there is an easier way to find the svmm and I see some other bugs that need fixing. I'll post a v3 as soon as I get those written and tested. Thanks for the review.

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: