freedreno: fence_server_sync() fixes
[mesa.git] / src / gallium / drivers / freedreno / freedreno_state.c
index 05717da95995e5cec6a50a2194e440c45d8ca234..9e1c1a1776c431db6c3e65e3f45db89a509e4924 100644 (file)
@@ -1,5 +1,3 @@
-/* -*- mode: C; c-file-style: "k&r"; tab-width 4; indent-tabs-mode: t; -*- */
-
 /*
  * Copyright (C) 2012 Rob Clark <robclark@freedesktop.org>
  *
@@ -80,6 +78,14 @@ fd_set_sample_mask(struct pipe_context *pctx, unsigned sample_mask)
        ctx->dirty |= FD_DIRTY_SAMPLE_MASK;
 }
 
+static void
+fd_set_min_samples(struct pipe_context *pctx, unsigned min_samples)
+{
+       struct fd_context *ctx = fd_context(pctx);
+       ctx->min_samples = min_samples;
+       ctx->dirty |= FD_DIRTY_MIN_SAMPLES;
+}
+
 /* notes from calim on #dri-devel:
  * index==0 will be non-UBO (ie. glUniformXYZ()) all packed together padded
  * out to vec4's
@@ -98,70 +104,63 @@ fd_set_constant_buffer(struct pipe_context *pctx,
 
        util_copy_constant_buffer(&so->cb[index], cb);
 
-       /* Note that the state tracker can unbind constant buffers by
+       /* Note that gallium frontends can unbind constant buffers by
         * passing NULL here.
         */
        if (unlikely(!cb)) {
                so->enabled_mask &= ~(1 << index);
-               so->dirty_mask &= ~(1 << index);
                return;
        }
 
        so->enabled_mask |= 1 << index;
-       so->dirty_mask |= 1 << index;
        ctx->dirty_shader[shader] |= FD_DIRTY_SHADER_CONST;
        ctx->dirty |= FD_DIRTY_CONST;
+
+       fd_resource_set_usage(cb->buffer, FD_DIRTY_CONST);
 }
 
 static void
 fd_set_shader_buffers(struct pipe_context *pctx,
                enum pipe_shader_type shader,
                unsigned start, unsigned count,
-               const struct pipe_shader_buffer *buffers)
+               const struct pipe_shader_buffer *buffers,
+               unsigned writable_bitmask)
 {
        struct fd_context *ctx = fd_context(pctx);
        struct fd_shaderbuf_stateobj *so = &ctx->shaderbuf[shader];
-       unsigned mask = 0;
+       const unsigned modified_bits = u_bit_consecutive(start, count);
 
-       if (buffers) {
-               for (unsigned i = 0; i < count; i++) {
-                       unsigned n = i + start;
-                       struct pipe_shader_buffer *buf = &so->sb[n];
+       so->enabled_mask &= ~modified_bits;
+       so->writable_mask &= ~modified_bits;
+       so->writable_mask |= writable_bitmask << start;
 
+       for (unsigned i = 0; i < count; i++) {
+               unsigned n = i + start;
+               struct pipe_shader_buffer *buf = &so->sb[n];
+
+               if (buffers && buffers[i].buffer) {
                        if ((buf->buffer == buffers[i].buffer) &&
                                        (buf->buffer_offset == buffers[i].buffer_offset) &&
                                        (buf->buffer_size == buffers[i].buffer_size))
                                continue;
 
-                       mask |= BIT(n);
-
                        buf->buffer_offset = buffers[i].buffer_offset;
                        buf->buffer_size = buffers[i].buffer_size;
                        pipe_resource_reference(&buf->buffer, buffers[i].buffer);
 
-                       if (buf->buffer)
-                               so->enabled_mask |= BIT(n);
-                       else
-                               so->enabled_mask &= ~BIT(n);
-               }
-       } else {
-               mask = (BIT(count) - 1) << start;
-
-               for (unsigned i = 0; i < count; i++) {
-                       unsigned n = i + start;
-                       struct pipe_shader_buffer *buf = &so->sb[n];
+                       fd_resource_set_usage(buffers[i].buffer, FD_DIRTY_SSBO);
 
+                       so->enabled_mask |= BIT(n);
+               } else {
                        pipe_resource_reference(&buf->buffer, NULL);
                }
-
-               so->enabled_mask &= ~mask;
        }
 
-       so->dirty_mask |= mask;
        ctx->dirty_shader[shader] |= FD_DIRTY_SHADER_SSBO;
+       ctx->dirty |= FD_DIRTY_SSBO;
 }
 
-static void
+void
 fd_set_shader_images(struct pipe_context *pctx,
                enum pipe_shader_type shader,
                unsigned start, unsigned count,
@@ -186,10 +185,12 @@ fd_set_shader_images(struct pipe_context *pctx,
                        mask |= BIT(n);
                        util_copy_image_view(buf, &images[i]);
 
-                       if (buf->resource)
+                       if (buf->resource) {
+                               fd_resource_set_usage(buf->resource, FD_DIRTY_IMAGE);
                                so->enabled_mask |= BIT(n);
-                       else
+                       } else {
                                so->enabled_mask &= ~BIT(n);
+                       }
                }
        } else {
                mask = (BIT(count) - 1) << start;
@@ -204,8 +205,8 @@ fd_set_shader_images(struct pipe_context *pctx,
                so->enabled_mask &= ~mask;
        }
 
-       so->dirty_mask |= mask;
        ctx->dirty_shader[shader] |= FD_DIRTY_SHADER_IMAGE;
+       ctx->dirty |= FD_DIRTY_IMAGE;
 }
 
 static void
@@ -215,18 +216,30 @@ fd_set_framebuffer_state(struct pipe_context *pctx,
        struct fd_context *ctx = fd_context(pctx);
        struct pipe_framebuffer_state *cso;
 
+       DBG("%ux%u, %u layers, %u samples",
+               framebuffer->width, framebuffer->height,
+               framebuffer->layers, framebuffer->samples);
+
+       fd_context_switch_from(ctx);
+
+       cso = &ctx->framebuffer;
+
+       if (util_framebuffer_state_equal(cso, framebuffer))
+               return;
+
+       util_copy_framebuffer_state(cso, framebuffer);
+
+       cso->samples = util_framebuffer_get_num_samples(cso);
+
        if (ctx->screen->reorder) {
-               struct fd_batch *batch, *old_batch = NULL;
+               struct fd_batch *old_batch = NULL;
 
                fd_batch_reference(&old_batch, ctx->batch);
 
                if (likely(old_batch))
                        fd_batch_set_stage(old_batch, FD_STAGE_NULL);
 
-               batch = fd_batch_from_fb(&ctx->screen->batch_cache, ctx, framebuffer);
                fd_batch_reference(&ctx->batch, NULL);
-               fd_reset_wfi(batch);
-               ctx->batch = batch;
                fd_context_all_dirty(ctx);
 
                if (old_batch && old_batch->blit && !old_batch->back_blit) {
@@ -235,20 +248,16 @@ fd_set_framebuffer_state(struct pipe_context *pctx,
                         * multiple times to the same surface), so we might as
                         * well go ahead and flush this one:
                         */
-                       fd_batch_flush(old_batch, false, false);
+                       fd_batch_flush(old_batch);
                }
 
                fd_batch_reference(&old_batch, NULL);
-       } else {
+       } else if (ctx->batch) {
                DBG("%d: cbufs[0]=%p, zsbuf=%p", ctx->batch->needs_flush,
                                framebuffer->cbufs[0], framebuffer->zsbuf);
-               fd_batch_flush(ctx->batch, false, false);
+               fd_batch_flush(ctx->batch);
        }
 
-       cso = &ctx->batch->framebuffer;
-
-       util_copy_framebuffer_state(cso, framebuffer);
-
        ctx->dirty |= FD_DIRTY_FRAMEBUFFER;
 
        ctx->disabled_scissor.minx = 0;
@@ -287,7 +296,36 @@ fd_set_viewport_states(struct pipe_context *pctx,
                const struct pipe_viewport_state *viewport)
 {
        struct fd_context *ctx = fd_context(pctx);
+       struct pipe_scissor_state *scissor = &ctx->viewport_scissor;
+       float minx, miny, maxx, maxy;
+
        ctx->viewport = *viewport;
+
+       /* see si_get_scissor_from_viewport(): */
+
+       /* Convert (-1, -1) and (1, 1) from clip space into window space. */
+       minx = -viewport->scale[0] + viewport->translate[0];
+       miny = -viewport->scale[1] + viewport->translate[1];
+       maxx = viewport->scale[0] + viewport->translate[0];
+       maxy = viewport->scale[1] + viewport->translate[1];
+
+       /* Handle inverted viewports. */
+       if (minx > maxx) {
+               swap(minx, maxx);
+       }
+       if (miny > maxy) {
+               swap(miny, maxy);
+       }
+
+       debug_assert(miny >= 0);
+       debug_assert(maxy >= 0);
+
+       /* Convert to integer and round up the max bounds. */
+       scissor->minx = minx;
+       scissor->miny = miny;
+       scissor->maxx = ceilf(maxx);
+       scissor->maxy = ceilf(maxy);
+
        ctx->dirty |= FD_DIRTY_VIEWPORT;
 }
 
@@ -320,7 +358,15 @@ fd_set_vertex_buffers(struct pipe_context *pctx,
        util_set_vertex_buffers_mask(so->vb, &so->enabled_mask, vb, start_slot, count);
        so->count = util_last_bit(so->enabled_mask);
 
+       if (!vb)
+               return;
+
        ctx->dirty |= FD_DIRTY_VTXBUF;
+
+       for (unsigned i = 0; i < count; i++) {
+               assert(!vb[i].is_user_buffer);
+               fd_resource_set_usage(vb[i].buffer.resource, FD_DIRTY_VTXBUF);
+       }
 }
 
 static void
@@ -351,10 +397,17 @@ fd_rasterizer_state_bind(struct pipe_context *pctx, void *hwcso)
 {
        struct fd_context *ctx = fd_context(pctx);
        struct pipe_scissor_state *old_scissor = fd_context_get_scissor(ctx);
+       bool discard = ctx->rasterizer && ctx->rasterizer->rasterizer_discard;
 
        ctx->rasterizer = hwcso;
        ctx->dirty |= FD_DIRTY_RASTERIZER;
 
+       if (ctx->rasterizer && ctx->rasterizer->scissor) {
+               ctx->current_scissor = &ctx->scissor;
+       } else {
+               ctx->current_scissor = &ctx->disabled_scissor;
+       }
+
        /* if scissor enable bit changed we need to mark scissor
         * state as dirty as well:
         * NOTE: we can do a shallow compare, since we only care
@@ -362,6 +415,9 @@ fd_rasterizer_state_bind(struct pipe_context *pctx, void *hwcso)
         */
        if (old_scissor != fd_context_get_scissor(ctx))
                ctx->dirty |= FD_DIRTY_SCISSOR;
+
+       if (ctx->rasterizer && (discard != ctx->rasterizer->rasterizer_discard))
+               ctx->dirty |= FD_DIRTY_RASTERIZER_DISCARD;
 }
 
 static void
@@ -433,7 +489,7 @@ fd_create_stream_output_target(struct pipe_context *pctx,
        target->buffer_size = buffer_size;
 
        assert(rsc->base.target == PIPE_BUFFER);
-       util_range_add(&rsc->valid_buffer_range,
+       util_range_add(&rsc->base, &rsc->valid_buffer_range,
                buffer_offset, buffer_offset + buffer_size);
 
        return target;
@@ -460,12 +516,14 @@ fd_set_stream_output_targets(struct pipe_context *pctx,
 
        for (i = 0; i < num_targets; i++) {
                boolean changed = targets[i] != so->targets[i];
-               boolean append = (offsets[i] == (unsigned)-1);
+               boolean reset = (offsets[i] != (unsigned)-1);
+
+               so->reset |= (reset << i);
 
-               if (!changed && append)
+               if (!changed && !reset)
                        continue;
 
-               if (!append)
+               if (reset)
                        so->offsets[i] = offsets[i];
 
                pipe_so_target_reference(&so->targets[i], targets[i]);
@@ -532,10 +590,6 @@ fd_set_global_binding(struct pipe_context *pctx,
 
                for (unsigned i = 0; i < count; i++) {
                        unsigned n = i + first;
-                       if (so->buf[n]) {
-                               struct fd_resource *rsc = fd_resource(so->buf[n]);
-                               fd_bo_put_iova(rsc->bo);
-                       }
                        pipe_resource_reference(&so->buf[n], NULL);
                }
 
@@ -551,6 +605,7 @@ fd_state_init(struct pipe_context *pctx)
        pctx->set_stencil_ref = fd_set_stencil_ref;
        pctx->set_clip_state = fd_set_clip_state;
        pctx->set_sample_mask = fd_set_sample_mask;
+       pctx->set_min_samples = fd_set_min_samples;
        pctx->set_constant_buffer = fd_set_constant_buffer;
        pctx->set_shader_buffers = fd_set_shader_buffers;
        pctx->set_shader_images = fd_set_shader_images;
@@ -570,7 +625,8 @@ fd_state_init(struct pipe_context *pctx)
        pctx->bind_depth_stencil_alpha_state = fd_zsa_state_bind;
        pctx->delete_depth_stencil_alpha_state = fd_zsa_state_delete;
 
-       pctx->create_vertex_elements_state = fd_vertex_state_create;
+       if (!pctx->create_vertex_elements_state)
+               pctx->create_vertex_elements_state = fd_vertex_state_create;
        pctx->delete_vertex_elements_state = fd_vertex_state_delete;
        pctx->bind_vertex_elements_state = fd_vertex_state_bind;