diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index e8a083743e09276920a009dab4a6eeea79138cc9..49ce5777c51152804046b98d804922c311fc43b4 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -255,8 +255,30 @@ static void *i915_gem_object_map(struct drm_i915_gem_object *obj,
 		return NULL;
 
 	/* A single page can always be kmapped */
-	if (n_pte == 1 && type == I915_MAP_WB)
-		return kmap(sg_page(sgt->sgl));
+	if (n_pte == 1 && type == I915_MAP_WB) {
+		struct page *page = sg_page(sgt->sgl);
+
+		/*
+		 * On 32b, highmem using a finite set of indirect PTE (i.e.
+		 * vmap) to provide virtual mappings of the high pages.
+		 * As these are finite, map_new_virtual() must wait for some
+		 * other kmap() to finish when it runs out. If we map a large
+		 * number of objects, there is no method for it to tell us
+		 * to release the mappings, and we deadlock.
+		 *
+		 * However, if we make an explicit vmap of the page, that
+		 * uses a larger vmalloc arena, and also has the ability
+		 * to tell us to release unwanted mappings. Most importantly,
+		 * it will fail and propagate an error instead of waiting
+		 * forever.
+		 *
+		 * So if the page is beyond the 32b boundary, make an explicit
+		 * vmap. On 64b, this check will be optimised away as we can
+		 * directly kmap any page on the system.
+		 */
+		if (!PageHighMem(page))
+			return kmap(page);
+	}
 
 	mem = stack;
 	if (n_pte > ARRAY_SIZE(stack)) {