Mesa (master): freedreno/a6xx: Fix indirect+patches draws

[GitLab Mirror]

Module: Mesa
Branch: master
Commit: c14e059d9fc5531a81bb42306dcd564f98251c19
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=c14e059d9fc5531a81bb42306dcd564f98251c19

Author: Rob Clark <robdclark at chromium.org>
Date:   Fri Mar 26 13:49:49 2021 -0700

freedreno/a6xx: Fix indirect+patches draws

Without TC, we'd get draw->count==0.. which is obviously not correct.
With TC we get random garbage for draw->count.  Which turns into
exciting things like trying to allocate multi-gigabyte buffers for
tess param/factor buffers.

But we can just tell the CP to split up large tess draws, and put an
upper bound on the tess param/factor buffer sizes.

Signed-off-by: Rob Clark <robdclark at chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9864>

---

 src/gallium/drivers/freedreno/a6xx/fd6_draw.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

diff --git a/src/gallium/drivers/freedreno/a6xx/fd6_draw.c b/src/gallium/drivers/freedreno/a6xx/fd6_draw.c
index 6904e1b6363..aea565459d1 100644
--- a/src/gallium/drivers/freedreno/a6xx/fd6_draw.c
+++ b/src/gallium/drivers/freedreno/a6xx/fd6_draw.c
@@ -276,11 +276,28 @@ fd6_draw_vbo(struct fd_context *ctx, const struct pipe_draw_info *info,
 		draw0.prim_type = DI_PT_PATCHES0 + info->vertices_per_patch;
 		draw0.tess_enable = true;
 
+		const unsigned max_count = 2048;
+		unsigned count;
+
+		/**
+		 * We can cap tessparam/tessfactor buffer sizes at the sub-draw
+		 * limit.  But in the indirect-draw case we must assume the worst.
+		 */
+		if (indirect && indirect->buffer) {
+			count = ALIGN_NPOT(max_count, info->vertices_per_patch);
+		} else {
+			count = MIN2(max_count, draw->count);
+			count = ALIGN_NPOT(count, info->vertices_per_patch);
+		}
+
+		OUT_PKT7(ring, CP_SET_SUBDRAW_SIZE, 1);
+		OUT_RING(ring, count);
+
 		ctx->batch->tessellation = true;
 		ctx->batch->tessparam_size = MAX2(ctx->batch->tessparam_size,
-				emit.hs->output_size * 4 * draw->count);
+				emit.hs->output_size * 4 * count);
 		ctx->batch->tessfactor_size = MAX2(ctx->batch->tessfactor_size,
-				factor_stride * draw->count);
+				factor_stride * count);
 
 		if (!ctx->batch->tess_addrs_constobj) {
 			/* Reserve space for the bo address - we'll write them later in