stash

[fips] / TODO

diff --git a/TODO b/TODO
index 36b970916150f7747971b03e06ce4a0c20d61f03..e7f4276772d3fe8f1d3700721d51b157825162e8 100644 (file)
--- a/TODO
+++ b/TODO
@@ -1,20 +1,3 @@
-Known bugs
-==========
-undefined symbol: glGenQueries
-
-       The problem here is code in libfips directly calling OpenGL
-       functions like glGenQueries but not linking against any OpenGL
-       library.
-
-       We don't want to link against any OpenGL library so that the
-       application itself can choose which OpenGL to use (and how to
-       link/load it).
-
-       The trick is to instead make these calls indirectly by first
-       calling glXGetProcAddressARB/eglGetProcAddress. There's some
-       proof-of-concept code for this in the stash-egl-lookup-fixups
-       branch, (which needs some cleaning up).
-
 Feature requests (small-scale, near-term changes)
 =================================================
 
@@ -36,10 +19,6 @@ Add Eric's tiny hash table for collecting per-shader statistics
 
        people.freedesktop.org:~anholt/hash_table
 
-Sort list of shaders in output
-
-Use better units for shader activity (eg. absolute time, relative percentage)
-
 Capture GPU performance counters.
 
 Allow dumping of shader source for investigation
@@ -84,3 +63,17 @@ Explore using perf/LTTng probing instead of LD_PRELOAD wrapper
 
        This has the advantage of allowing full-system,
        multi-process data collection.
+
+Things to change while replaying to isolate impact of different pieces
+of the pipeline (list by Eero)
+==============================
+    The interesting "what is a bottleneck" experiments are
+    (from back of 3D pipeline to front):
+    - Disable blend (= disable related dst reads)
+    - Binding 1x1 / NULL texture instead of real ones [1]
+    - Using simple pass-through pixel shader instead of real ones [2]
+    - Use kill / discard pixel shader instead of real ones
+    - Use 0x0 / 1x1 scissor rect
+    - Front+backface cull
+    - Disable draws
+