vc4: Add a userspace BO cache.

Since our kernel BOs require CMA allocation, and the use of them requires
new mmaps, it's pretty expensive and we should avoid it if possible.
Copying my original design for Intel, make a userspace cache that reuses
BOs that haven't been shared to other processes but frees BOs that have
sat in the cache for over a second.

Improves glxgears framerate on RPi by around 30%.

1 files changed, 2 insertions, 0 deletions

diff --git a/src/gallium/drivers/vc4/vc4_screen.c b/src/gallium/drivers/vc4/vc4_screen.c
index b532cc6782f..8d216338bf7 100644
--- a/src/gallium/drivers/vc4/vc4_screen.c
+++ b/src/gallium/drivers/vc4/vc4_screen.c
@@ -76,6 +76,7 @@ vc4_screen_get_vendor(struct pipe_screen *pscreen)
 static void
 vc4_screen_destroy(struct pipe_screen *pscreen)
 {
+        vc4_bufmgr_destroy(pscreen);
         ralloc_free(pscreen);
 }
 
@@ -449,6 +450,7 @@ vc4_screen_create(int fd)
         pscreen->is_format_supported = vc4_screen_is_format_supported;
 
         screen->fd = fd;
+        make_empty_list(&screen->bo_cache.time_list);
 
         vc4_fence_init(screen);