nvfx: support an unlimited number of occlusion queries

Currently on nv30/nv40 an assert will be triggered once 32 queries are
outstanding.

This violates the OpenGL/Gallium interface, which requires support for
an unlimited number of fences.

This patch fixes the problem by putting queries in a linked list and
waiting on the oldest one if allocation fails.

nVidia seems to use a similar strategy, but with 1024 instead of 32 fences.
The next patch will improve this.

diff --git a/src/gallium/drivers/nvfx/nvfx_query.c b/src/gallium/drivers/nvfx/nvfx_query.c
index acbaf75a2368a6c0d15a250379294e81870d9231..1b20b5245d7ff51c8e4f7cb2f8554791045d2153 100644 (file)
--- a/src/gallium/drivers/nvfx/nvfx_query.c
+++ b/src/gallium/drivers/nvfx/nvfx_query.c
@@ -3,6 +3,7 @@
 #include "nvfx_context.h"
 
 struct nvfx_query {
+       struct list_head list;
        struct nouveau_resource *object;
        unsigned type;
        boolean ready;
@@ -23,6 +24,8 @@ nvfx_query_create(struct pipe_context *pipe, unsigned query_type)
        q = CALLOC(1, sizeof(struct nvfx_query));
        q->type = query_type;
 
+       assert(q->type == PIPE_QUERY_OCCLUSION_COUNTER);
+
        return (struct pipe_query *)q;
 }
 
@@ -32,7 +35,10 @@ nvfx_query_destroy(struct pipe_context *pipe, struct pipe_query *pq)
        struct nvfx_query *q = nvfx_query(pq);
 
        if (q->object)
+       {
                nouveau_resource_free(&q->object);
+               LIST_DEL(&q->list);
+       }
        FREE(q);
 }
 
@@ -44,20 +50,25 @@ nvfx_query_begin(struct pipe_context *pipe, struct pipe_query *pq)
        struct nvfx_screen *screen = nvfx->screen;
        struct nouveau_channel *chan = screen->base.channel;
        struct nouveau_grobj *eng3d = screen->eng3d;
-
-       assert(q->type == PIPE_QUERY_OCCLUSION_COUNTER);
+       uint64_t tmp;
 
        /* Happens when end_query() is called, then another begin_query()
         * without querying the result in-between.  For now we'll wait for
         * the existing query to notify completion, but it could be better.
         */
-       if (q->object) {
-               uint64_t tmp;
+       if (q->object)
                pipe->get_query_result(pipe, pq, 1, &tmp);
+
+       while (nouveau_resource_alloc(nvfx->screen->query_heap, 1, NULL, &q->object))
+       {
+               struct nvfx_query* oldestq;
+               assert(!LIST_IS_EMPTY(&nvfx->screen->query_list));
+               oldestq = LIST_ENTRY(struct nvfx_query, nvfx->screen->query_list.next, list);
+               pipe->get_query_result(pipe, (struct pipe_query*)oldestq, 1, &tmp);
        }
 
-       if (nouveau_resource_alloc(nvfx->screen->query_heap, 1, NULL, &q->object))
-               assert(0);
+       LIST_ADDTAIL(&q->list, &nvfx->screen->query_list);
+
        nouveau_notifier_reset(nvfx->screen->query, q->object->start);
 
        BEGIN_RING(chan, eng3d, NV34TCL_QUERY_RESET, 1);
@@ -90,8 +101,6 @@ nvfx_query_result(struct pipe_context *pipe, struct pipe_query *pq,
        struct nvfx_context *nvfx = nvfx_context(pipe);
        struct nvfx_query *q = nvfx_query(pq);
 
-       assert(q->object && q->type == PIPE_QUERY_OCCLUSION_COUNTER);
-
        if (!q->ready) {
                unsigned status;
 
@@ -110,6 +119,7 @@ nvfx_query_result(struct pipe_context *pipe, struct pipe_query *pq,
                                                        q->object->start);
                q->ready = TRUE;
                nouveau_resource_free(&q->object);
+               LIST_DEL(&q->list);
        }
 
        *result = q->result;

diff --git a/src/gallium/drivers/nvfx/nvfx_screen.c b/src/gallium/drivers/nvfx/nvfx_screen.c
index 9236c8d256a0a6670af25f25ec4bc573b23a6ada..651e6ee64b6cdaf58dd0f47a08564769dc1e9866 100644 (file)
--- a/src/gallium/drivers/nvfx/nvfx_screen.c
+++ b/src/gallium/drivers/nvfx/nvfx_screen.c
@@ -411,6 +411,8 @@ nvfx_screen_create(struct pipe_winsys *ws, struct nouveau_device *dev)
                return NULL;
        }
 
+       LIST_INITHEAD(&screen->query_list);
+
        /* Vtxprog resources */
        if (nouveau_resource_init(&screen->vp_exec_heap, 0, screen->is_nv4x ? 512 : 256) ||
            nouveau_resource_init(&screen->vp_data_heap, 0, 256)) {

diff --git a/src/gallium/drivers/nvfx/nvfx_screen.h b/src/gallium/drivers/nvfx/nvfx_screen.h
index ba9a6993ece14c42675e63d41bb6f128a40f7005..678ee406016a9434eb82d43b31ede7afb261ca60 100644 (file)
--- a/src/gallium/drivers/nvfx/nvfx_screen.h
+++ b/src/gallium/drivers/nvfx/nvfx_screen.h
@@ -1,6 +1,7 @@
 #ifndef __NVFX_SCREEN_H__
 #define __NVFX_SCREEN_H__
 
+#include <util/u_double_list.h>
 #include "nouveau/nouveau_screen.h"
 #include "nv04_surface_2d.h"
 #include "nvfx_context.h"
@@ -24,6 +25,7 @@ struct nvfx_screen {
        /* Query object resources */
        struct nouveau_notifier *query;
        struct nouveau_resource *query_heap;
+       struct list_head query_list;
 
        /* Vtxprog resources */
        struct nouveau_resource *vp_exec_heap;