On 28/07/2020 16:31, Chris Wilson wrote:
kmalloc uses power-of-two slab buckets for small allocations (up to a few pages). Since i915_page_directory is a page of pointers, plus a couple more, this is rounded up to 8K, and we waste nearly 50% of that allocation. Long terms this leads to poor memory utilisation, bloating the kernel footpoint, but the problem is excerbated by our conservative
footprint exacerbated
preallocation scheme for binding VMA. As we require to allocate all levels for each vma just in case we need to insert them upon binding, this leads to a large multiplication factor for single page vma. By halving the allocation we need for the page directory structure, we halve the multipliation factor, bringing workloads that once fitted into
multiplication
memory, hopefully back to fitting into memory. We maintain the split between i915_page_directory and i915_page_table as we only need half the allocation for the lowest, most populous, level. Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> Cc: Mika Kuoppala <mika.kuoppala@xxxxxxxxxxxxxxx> Cc: Matthew Auld <matthew.auld@xxxxxxxxx>
Nice find, Reviewed-by: Matthew Auld <matthew.auld@xxxxxxxxx> _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
