diff --git a/src/mesa/main/glthread.h b/src/mesa/main/glthread.h
index d752ca9ef7a..05d8ebd6b53 100644
--- a/src/mesa/main/glthread.h
+++ b/src/mesa/main/glthread.h
@@ -102,6 +102,7 @@ struct glthread_vao {
    GLbitfield BufferEnabled; /**< "Enabled" converted to buffer bindings. */
    GLbitfield BufferInterleaved; /**< Bitmask of buffers used by multiple attribs. */
    GLbitfield UserPointerMask; /**< Bitmask of buffer bindings. */
+   GLbitfield NonNullPointerMask; /**< Bitmask of buffer bindings with non-NULL user pointers. */
    GLbitfield NonZeroDivisorMask; /**< Bitmask of buffer bindings. */
 
    struct glthread_attrib Attrib[VERT_ATTRIB_MAX];
diff --git a/src/mesa/main/glthread_draw.c b/src/mesa/main/glthread_draw.c
index a8b8b1b7498..bc7c33bb4c0 100644
--- a/src/mesa/main/glthread_draw.c
+++ b/src/mesa/main/glthread_draw.c
@@ -427,15 +427,30 @@ draw_arrays_async_user(struct gl_context *ctx, GLenum mode, GLint first,
       memcpy(cmd + 1, buffers, buffers_size);
 }
 
+static inline unsigned
+get_user_buffer_mask(struct gl_context *ctx)
+{
+   struct glthread_vao *vao = ctx->GLThread.CurrentVAO;
+
+   /* BufferEnabled means which attribs are enabled in terms of buffer
+    * binding slots (not attrib slots).
+    *
+    * UserPointerMask means which buffer bindings don't have a buffer bound.
+    *
+    * NonNullPointerMask means which buffer bindings have a NULL pointer.
+    * Those are not uploaded. This can happen when an attrib is enabled, but
+    * the shader doesn't use it, so it's ignored by mesa/state_tracker.
+    */
+   return vao->BufferEnabled & vao->UserPointerMask & vao->NonNullPointerMask;
+}
+
 static ALWAYS_INLINE void
 draw_arrays(GLenum mode, GLint first, GLsizei count, GLsizei instance_count,
             GLuint baseinstance, bool compiled_into_dlist)
 {
    GET_CURRENT_CONTEXT(ctx);
-
-   struct glthread_vao *vao = ctx->GLThread.CurrentVAO;
    unsigned user_buffer_mask =
-      ctx->API == API_OPENGL_CORE ? 0 : vao->UserPointerMask & vao->BufferEnabled;
+      ctx->API == API_OPENGL_CORE ? 0 : get_user_buffer_mask(ctx);
 
    if (compiled_into_dlist && ctx->GLThread.ListMode) {
       _mesa_glthread_finish_before(ctx, "DrawArrays");
@@ -552,11 +567,9 @@ _mesa_marshal_MultiDrawArrays(GLenum mode, const GLint *first,
                               const GLsizei *count, GLsizei draw_count)
 {
    GET_CURRENT_CONTEXT(ctx);
-
-   struct glthread_vao *vao = ctx->GLThread.CurrentVAO;
    unsigned user_buffer_mask =
       ctx->API == API_OPENGL_CORE || draw_count <= 0 ?
-            0 : vao->UserPointerMask & vao->BufferEnabled;
+            0 : get_user_buffer_mask(ctx);
 
    if (ctx->GLThread.ListMode)
       goto sync;
@@ -898,10 +911,9 @@ draw_elements(GLenum mode, GLsizei count, GLenum type, const GLvoid *indices,
               bool compiled_into_dlist)
 {
    GET_CURRENT_CONTEXT(ctx);
-
    struct glthread_vao *vao = ctx->GLThread.CurrentVAO;
    unsigned user_buffer_mask =
-      ctx->API == API_OPENGL_CORE ? 0 : vao->UserPointerMask & vao->BufferEnabled;
+      ctx->API == API_OPENGL_CORE ? 0 : get_user_buffer_mask(ctx);
    bool has_user_indices = vao->CurrentElementBufferName == 0 && indices;
 
    if (compiled_into_dlist && ctx->GLThread.ListMode)
@@ -1151,10 +1163,9 @@ _mesa_marshal_MultiDrawElementsBaseVertex(GLenum mode, const GLsizei *count,
                                           const GLsizei *basevertex)
 {
    GET_CURRENT_CONTEXT(ctx);
-
    struct glthread_vao *vao = ctx->GLThread.CurrentVAO;
    unsigned user_buffer_mask =
-      ctx->API == API_OPENGL_CORE ? 0 : vao->UserPointerMask & vao->BufferEnabled;
+      ctx->API == API_OPENGL_CORE ? 0 : get_user_buffer_mask(ctx);
    bool has_user_indices = vao->CurrentElementBufferName == 0;
 
    if (ctx->GLThread.ListMode)
diff --git a/src/mesa/main/glthread_varray.c b/src/mesa/main/glthread_varray.c
index f4f2bde2158..a309ffe21aa 100644
--- a/src/mesa/main/glthread_varray.c
+++ b/src/mesa/main/glthread_varray.c
@@ -51,6 +51,7 @@ _mesa_glthread_reset_vao(struct glthread_vao *vao)
    vao->Enabled = 0;
    vao->BufferEnabled = 0;
    vao->UserPointerMask = 0;
+   vao->NonNullPointerMask = 0;
    vao->NonZeroDivisorMask = 0;
 
    for (unsigned i = 0; i < ARRAY_SIZE(vao->Attrib); i++) {
@@ -363,6 +364,11 @@ attrib_pointer(struct glthread_state *glthread, struct glthread_vao *vao,
       vao->UserPointerMask &= ~(1u << attrib);
    else
       vao->UserPointerMask |= 1u << attrib;
+
+   if (pointer)
+      vao->NonNullPointerMask |= 1u << attrib;
+   else
+      vao->NonNullPointerMask &= ~(1u << attrib);
 }
 
 void
@@ -447,6 +453,11 @@ bind_vertex_buffer(struct glthread_state *glthread, struct glthread_vao *vao,
       vao->UserPointerMask &= ~(1u << i);
    else
       vao->UserPointerMask |= 1u << i;
+
+   if (offset)
+      vao->NonNullPointerMask |= 1u << i;
+   else
+      vao->NonNullPointerMask &= ~(1u << i);
 }
 
 void