X-Git-Url: https://git.libre-soc.org/?a=blobdiff_plain;f=src%2Fmesa%2Fdrivers%2Fdri%2Fi965%2Fgen6_clip_state.c;h=004ecebc69eba5eb55d572d1d755641d63fb1839;hb=88d28aa4d9edec33ef7bcf1f56b77fbb756a24f8;hp=acc4b7f1019130814826b166c3ee5ac50d15822b;hpb=35489ef285f1fde234b2b9bbb91fdc41fddefc02;p=mesa.git diff --git a/src/mesa/drivers/dri/i965/gen6_clip_state.c b/src/mesa/drivers/dri/i965/gen6_clip_state.c index acc4b7f1019..004ecebc69e 100644 --- a/src/mesa/drivers/dri/i965/gen6_clip_state.c +++ b/src/mesa/drivers/dri/i965/gen6_clip_state.c @@ -28,48 +28,195 @@ #include "brw_context.h" #include "brw_state.h" #include "brw_defines.h" +#include "brw_util.h" #include "intel_batchbuffer.h" +#include "main/fbobject.h" +#include "main/framebuffer.h" static void upload_clip_state(struct brw_context *brw) { - struct intel_context *intel = &brw->intel; - GLcontext *ctx = &intel->ctx; - uint32_t depth_clamp = 0; - uint32_t provoking; + struct gl_context *ctx = &brw->ctx; + /* BRW_NEW_META_IN_PROGRESS */ + uint32_t dw1 = brw->meta_in_progress ? 0 : GEN6_CLIP_STATISTICS_ENABLE; + uint32_t dw2 = 0; - if (!ctx->Transform.DepthClamp) - depth_clamp = GEN6_CLIP_Z_TEST; + /* _NEW_BUFFERS */ + struct gl_framebuffer *fb = ctx->DrawBuffer; + /* BRW_NEW_FS_PROG_DATA */ + if (brw->wm.prog_data->barycentric_interp_modes & + BRW_WM_NONPERSPECTIVE_BARYCENTRIC_BITS) { + dw2 |= GEN6_CLIP_NON_PERSPECTIVE_BARYCENTRIC_ENABLE; + } + + if (brw->gen >= 7) + dw1 |= GEN7_CLIP_EARLY_CULL; + + if (brw->gen == 7) { + /* _NEW_POLYGON */ + if (ctx->Polygon._FrontBit == _mesa_is_user_fbo(fb)) + dw1 |= GEN7_CLIP_WINDING_CCW; + + if (ctx->Polygon.CullFlag) { + switch (ctx->Polygon.CullFaceMode) { + case GL_FRONT: + dw1 |= GEN7_CLIP_CULLMODE_FRONT; + break; + case GL_BACK: + dw1 |= GEN7_CLIP_CULLMODE_BACK; + break; + case GL_FRONT_AND_BACK: + dw1 |= GEN7_CLIP_CULLMODE_BOTH; + break; + default: + unreachable("Should not get here: invalid CullFlag"); + } + } else { + dw1 |= GEN7_CLIP_CULLMODE_NONE; + } + } + + if (brw->gen < 8 && !ctx->Transform.DepthClamp) + dw2 |= GEN6_CLIP_Z_TEST; + + /* _NEW_LIGHT */ if (ctx->Light.ProvokingVertex == GL_FIRST_VERTEX_CONVENTION) { - provoking = 0; + dw2 |= + (0 << GEN6_CLIP_TRI_PROVOKE_SHIFT) | + (1 << GEN6_CLIP_TRIFAN_PROVOKE_SHIFT) | + (0 << GEN6_CLIP_LINE_PROVOKE_SHIFT); } else { - provoking = + dw2 |= (2 << GEN6_CLIP_TRI_PROVOKE_SHIFT) | (2 << GEN6_CLIP_TRIFAN_PROVOKE_SHIFT) | (1 << GEN6_CLIP_LINE_PROVOKE_SHIFT); } + /* _NEW_TRANSFORM */ + dw2 |= (ctx->Transform.ClipPlanesEnabled << + GEN6_USER_CLIP_CLIP_DISTANCES_SHIFT); + if (ctx->Transform.ClipDepthMode == GL_ZERO_TO_ONE) + dw2 |= GEN6_CLIP_API_D3D; + else + dw2 |= GEN6_CLIP_API_OGL; + + dw2 |= GEN6_CLIP_GB_TEST; + + /* We need to disable guardband clipping if the guardband (which we always + * program to the maximum screen-space bounding box of 8K x 8K) will be + * smaller than the viewport. + * + * Closely examining the clip determination formulas in the documentation + * reveals that objects will be discarded entirely if they're outside the + * (small) guardband, even if they're within the (large) viewport: + * + * TR = TR_GB || TR_VPXY || TR_VPZ || TR_UC || TR_NEGW + * TA = !TR && TA_GB && TA_VPZ && TA_NEGW + * MC = !(TA || TR) + * + * (TA is "Trivial Accept", TR is "Trivial Reject", MC is "Must Clip".) + * + * Disabling guardband clipping removes the TR_GB condition, which means + * they'll be considered MC ("Must Clip") unless they're rejected for + * some other reason. + * + * Note that there is no TA_VPXY condition. If there were, objects entirely + * inside a 16384x16384 viewport would be trivially accepted, breaking the + * "objects must have a screenspace bounding box not exceeding 8K in the X + * or Y direction" restriction. Instead, they're clipped. + */ + for (unsigned i = 0; i < ctx->Const.MaxViewports; i++) { + if (ctx->ViewportArray[i].Width > 8192 || + ctx->ViewportArray[i].Height > 8192) { + dw2 &= ~GEN6_CLIP_GB_TEST; + break; + } + } + + /* If the viewport dimensions are smaller than the drawable dimensions, + * we have to disable guardband clipping prior to Gen8. We always program + * the guardband to a fixed size, which is almost always larger than the + * viewport. Any geometry which intersects the viewport but lies within + * the guardband would bypass the 3D clipping stage, so it wouldn't be + * clipped to the viewport. Rendering would happen beyond the viewport, + * but still inside the drawable. + * + * Gen8+ introduces a viewport extents test which restricts rendering to + * the viewport, so we can ignore this restriction. + */ + if (brw->gen < 8) { + const float fb_width = (float)_mesa_geometric_width(fb); + const float fb_height = (float)_mesa_geometric_height(fb); + + for (unsigned i = 0; i < ctx->Const.MaxViewports; i++) { + if (ctx->ViewportArray[i].X != 0 || + ctx->ViewportArray[i].Y != 0 || + ctx->ViewportArray[i].Width != fb_width || + ctx->ViewportArray[i].Height != fb_height) { + dw2 &= ~GEN6_CLIP_GB_TEST; + break; + } + } + } + + /* BRW_NEW_RASTERIZER_DISCARD */ + if (ctx->RasterDiscard) { + dw2 |= GEN6_CLIP_MODE_REJECT_ALL; + perf_debug("Rasterizer discard is currently implemented via the clipper; " + "%s be faster.\n", brw->gen >= 7 ? "using the SOL unit may" : + "having the GS not write primitives would likely"); + } + + uint32_t enable; + if (brw->primitive == _3DPRIM_RECTLIST) + enable = 0; + else + enable = GEN6_CLIP_ENABLE; + + if (!is_drawing_points(brw) && !is_drawing_lines(brw)) + dw2 |= GEN6_CLIP_XY_TEST; + BEGIN_BATCH(4); - OUT_BATCH(CMD_3D_CLIP_STATE << 16 | (4 - 2)); - OUT_BATCH(GEN6_CLIP_STATISTICS_ENABLE); - OUT_BATCH(GEN6_CLIP_ENABLE | - GEN6_CLIP_API_OGL | + OUT_BATCH(_3DSTATE_CLIP << 16 | (4 - 2)); + OUT_BATCH(dw1); + OUT_BATCH(enable | GEN6_CLIP_MODE_NORMAL | - GEN6_CLIP_XY_TEST | - depth_clamp | - provoking); - OUT_BATCH(0); + dw2); + OUT_BATCH(U_FIXED(0.125, 3) << GEN6_CLIP_MIN_POINT_WIDTH_SHIFT | + U_FIXED(255.875, 3) << GEN6_CLIP_MAX_POINT_WIDTH_SHIFT | + (_mesa_geometric_layers(fb) > 0 ? 0 : GEN6_CLIP_FORCE_ZERO_RTAINDEX) | + ((ctx->Const.MaxViewports - 1) & GEN6_CLIP_MAX_VP_INDEX_MASK)); ADVANCE_BATCH(); - - intel_batchbuffer_emit_mi_flush(intel->batch); } const struct brw_tracked_state gen6_clip_state = { .dirty = { - .mesa = _NEW_TRANSFORM, - .brw = BRW_NEW_CONTEXT, - .cache = 0 + .mesa = _NEW_BUFFERS | + _NEW_LIGHT | + _NEW_TRANSFORM, + .brw = BRW_NEW_CONTEXT | + BRW_NEW_FS_PROG_DATA | + BRW_NEW_GEOMETRY_PROGRAM | + BRW_NEW_META_IN_PROGRESS | + BRW_NEW_PRIMITIVE | + BRW_NEW_RASTERIZER_DISCARD, + }, + .emit = upload_clip_state, +}; + +const struct brw_tracked_state gen7_clip_state = { + .dirty = { + .mesa = _NEW_BUFFERS | + _NEW_LIGHT | + _NEW_POLYGON | + _NEW_TRANSFORM, + .brw = BRW_NEW_CONTEXT | + BRW_NEW_FS_PROG_DATA | + BRW_NEW_GEOMETRY_PROGRAM | + BRW_NEW_META_IN_PROGRESS | + BRW_NEW_PRIMITIVE | + BRW_NEW_RASTERIZER_DISCARD, }, .emit = upload_clip_state, };