drm/gud: Add async_flush module parameter

Provide a way for userspace to choose synchronous flushing/pageflips.
This helps save CPU and power.

It is also useful for test scripts since userspace can know when a flush
has happended and wait before doing the next visual test.

Cc: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Noralf Trønnes <noralf@tronnes.org>
Reviewed-by: Peter Stuge <peter@stuge.se>
Link: https://patchwork.freedesktop.org/patch/msgid/20210703141321.35494-2-noralf@tronnes.org

diff --git a/drivers/gpu/drm/gud/gud_pipe.c b/drivers/gpu/drm/gud/gud_pipe.c
index e0fb6cc..8f56bf6 100644 (file)
--- a/drivers/gpu/drm/gud/gud_pipe.c
+++ b/drivers/gpu/drm/gud/gud_pipe.c
@@ -24,6 +24,19 @@
 #include "gud_internal.h"
 
 /*
+ * Some userspace rendering loops runs all displays in the same loop.
+ * This means that a fast display will have to wait for a slow one.
+ * For this reason gud does flushing asynchronous by default.
+ * The down side is that in e.g. a single display setup userspace thinks
+ * the display is insanely fast since the driver reports back immediately
+ * that the flush/pageflip is done. This wastes CPU and power.
+ * Such users might want to set this module parameter to false.
+ */
+static bool gud_async_flush = true;
+module_param_named(async_flush, gud_async_flush, bool, 0644);
+MODULE_PARM_DESC(async_flush, "Enable asynchronous flushing [default=true]");
+
+/*
  * FIXME: The driver is probably broken on Big Endian machines.
  * See discussion:
  * https://lore.kernel.org/dri-devel/CAKb7UvihLX0hgBOP3VBG7O+atwZcUVCPVuBdfmDMpg0NjXe-cQ@mail.gmail.com/
@@ -578,6 +591,8 @@ void gud_pipe_update(struct drm_simple_display_pipe *pipe,
                if (gdrm->flags & GUD_DISPLAY_FLAG_FULL_UPDATE)
                        drm_rect_init(&damage, 0, 0, fb->width, fb->height);
                gud_fb_queue_damage(gdrm, fb, &damage);
+               if (!gud_async_flush)
+                       flush_work(&gdrm->work);
        }
 
        if (!crtc->state->enable)