On 16.01.2018 19:45, Emil Velikov wrote:
Hi Tapani,
On 15 January 2018 at 12:31, Tapani Pälli <tapani.pa...@intel.com> wrote:
+static void
+update_blob_cache_functions(struct dri2_egl_display *dri2_dpy,
+ struct dri2_egl_context *dri2_ctx)
+{
+ if (!dri2_dpy || !dri2_ctx)
+ return;
AFAICT dri2_dpy can never be NULL.
Yeah, that check can be removed.
It was because of my 'debug framework' (see eariler RFC). It was hit
when running PIGLIT_PLATFORM=gbm and during Piglit teardown we end up
here from eglMakeCurrent that called drv->API.SetBlobCacheFuncsANDROID
(and which is called during __run_exit_handlers) and display is then
0x0. But it won't ever happen without my 'debug framework' addition.
+
+ /* No blob support. */
+ if (!dri2_dpy->blob)
+ return;
+
+ /* No functions to set. */
+ if (!dri2_dpy->blob_cache_set)
+ return;
+
+ dri2_dpy->blob->set_cache_funcs(dri2_ctx->dri_context,
+ dri2_dpy->blob_cache_set,
+ dri2_dpy->blob_cache_get);
+}
+
I'm wondering why you opted to make set_cache_funcs dri_context
specific as opposed to dri_screen.
The latter seems to align better to EGLDisplay.
Plus doing so will simplify the existing code - no hunk in
dri2_make_current, no dri2_dpy->blob/blob_cache_set checks, etc.
It's an extension so we would still need dri2_dpy->blob check but maybe
not so much checks overall .. I'll see how that would look. Thanks!
@@ -230,6 +231,9 @@ struct dri2_egl_display
bool is_render_node;
bool is_different_gpu;
+
+ EGLSetBlobFuncANDROID blob_cache_set;
+ EGLGetBlobFuncANDROID blob_cache_get;
These two are part of the EGL API, so they are better suited in struct
_egl_display.
OK, will move them there.
// Tapani
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
