drm/i915: Only initialize partially filled pagetables

If we know that we will completely fill a pagetable (i.e. we are
inserting a complete set of 512 pages), we can skip prefilling that PT
with scratch entries. If we have to abort the insertion prior to writing
the real entries, we will teardown the pagetable and remove it from the
page directory (so that we will restart the allocation next time).

We could do similar tricks for the PD and PDP, but the likelihood of a
single insertion covering the entire 512 entries diminishes, as do the
cycle savings. The saving are even greater (relatively) when we are
preallocating page tables for huge pages, as then we never need to fill
the page table.

Signed-off-by: Chris Wilson <[email protected]>
Cc: Matthew Auld <[email protected]>
Cc: Joonas Lahtinen <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Reviewed-by: Matthew Auld <[email protected]>

1 files changed, 5 insertions, 2 deletions

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index c33c2f97c814..33181d6eac3f 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1168,19 +1168,22 @@ static int gen8_ppgtt_alloc_pd(struct i915_address_space *vm,
 	unsigned int pde;
 
 	gen8_for_each_pde(pt, pd, start, length, pde) {
+		int count = gen8_pte_count(start, length);
+
 		if (pt == vm->scratch_pt) {
 			pt = alloc_pt(vm);
 			if (IS_ERR(pt))
 				goto unwind;
 
-			gen8_initialize_pt(vm, pt);
+			if (count < GEN8_PTES)
+				gen8_initialize_pt(vm, pt);
 
 			gen8_ppgtt_set_pde(vm, pd, pt, pde);
 			pd->used_pdes++;
 			GEM_BUG_ON(pd->used_pdes > I915_PDES);
 		}
 
-		pt->used_ptes += gen8_pte_count(start, length);
+		pt->used_ptes += count;
 	}
 	return 0;