nouveau: Fix nv20-40 swizzled miptree RTs

I just coded a patch that does this and seems to work fine. It must be
fixed since it breaks OpenGL (or the state tracker can be changed, but
it seems better to do it in the driver).

The patch also fixes NV20 and NV30 in the same way. They compile but
are untested.

I would guess that using the 3D engine is faster for the larger
levels, but the 2D engine is faster for the smaller ones (and lacks
this issue).

2 files changed, 23 insertions, 1 deletions

diff --git a/src/gallium/drivers/nv40/nv40_miptree.c b/src/gallium/drivers/nv40/nv40_miptree.c
index b974e68a07..89bd155ff4 100644
--- a/src/gallium/drivers/nv40/nv40_miptree.c
+++ b/src/gallium/drivers/nv40/nv40_miptree.c
@@ -5,6 +5,7 @@
 #include "util/u_math.h"
 
 #include "nv40_context.h"
+#include "../nv04/nv04_surface_2d.h"
 
 
 
@@ -105,6 +106,12 @@ nv40_miptree_create(struct pipe_screen *pscreen, const struct pipe_texture *pt)
 	if (pt->tex_usage & PIPE_TEXTURE_USAGE_DYNAMIC)
 		buf_usage |= PIPE_BUFFER_USAGE_CPU_READ_WRITE;
 
+	/* apparently we can't render to swizzled surfaces smaller than 64 bytes, so make them linear.
+	 * If the user did not ask for a render target, they can still render to it, but it will cost them an extra copy.
+	 * This also happens for small mipmaps of large textures. */
+	if (pt->tex_usage & PIPE_TEXTURE_USAGE_RENDER_TARGET && util_format_get_stride(pt->format, pt->width0) < 64)
+		mt->base.tex_usage |= NOUVEAU_TEXTURE_USAGE_LINEAR;
+
 	nv40_miptree_layout(mt);
 
 	mt->buffer = pscreen->buffer_create(pscreen, 256, buf_usage, mt->total_size);
@@ -191,12 +198,27 @@ nv40_miptree_surface_new(struct pipe_screen *pscreen, struct pipe_texture *pt,
 		ns->base.offset = mt->level[level].image_offset[0];
 	}
 
+	/* create a linear temporary that we can render into if necessary.
+	 * Note that ns->pitch is always a multiple of 64 for linear surfaces and swizzled surfaces are POT, so
+	 * ns->pitch & 63 is equivalent to (ns->pitch < 64 && swizzled)*/
+	if((ns->pitch & 63) && (ns->base.usage & (PIPE_BUFFER_USAGE_GPU_WRITE | NOUVEAU_BUFFER_USAGE_NO_RENDER)) == PIPE_BUFFER_USAGE_GPU_WRITE)
+		return &nv04_surface_wrap_for_render(pscreen, ((struct nv40_screen*)pscreen)->eng2d, ns)->base;
+
 	return &ns->base;
 }
 
 static void
 nv40_miptree_surface_del(struct pipe_surface *ps)
 {
+	struct nv04_surface* ns = (struct nv04_surface*)ps;
+	if(ns->backing)
+	{
+		struct nv40_screen* screen = (struct nv40_screen*)ps->texture->screen;
+		if(ns->backing->base.usage & PIPE_BUFFER_USAGE_GPU_WRITE)
+			screen->eng2d->copy(screen->eng2d, &ns->backing->base, 0, 0, ps, 0, 0, ns->base.width, ns->base.height);
+		nv40_miptree_surface_del(&ns->backing->base);
+	}
+
 	pipe_texture_reference(&ps->texture, NULL);
 	FREE(ps);
 }
diff --git a/src/gallium/drivers/nv40/nv40_transfer.c b/src/gallium/drivers/nv40/nv40_transfer.c
index b084a38b48..adfd035621 100644
--- a/src/gallium/drivers/nv40/nv40_transfer.c
+++ b/src/gallium/drivers/nv40/nv40_transfer.c
@@ -126,7 +126,7 @@ nv40_transfer_del(struct pipe_transfer *ptx)
 
 		dst = pscreen->get_tex_surface(pscreen, ptx->texture,
 	                                       ptx->face, ptx->level, ptx->zslice,
-	                                       PIPE_BUFFER_USAGE_GPU_WRITE);
+	                                       PIPE_BUFFER_USAGE_GPU_WRITE | NOUVEAU_BUFFER_USAGE_NO_RENDER);
 
 		/* TODO: Check if SIFM can deal with x,y,w,h when swizzling */
 		nvscreen->eng2d->copy(nvscreen->eng2d,