2 * Copyright © 2011 Intel Corporation
4 * Permission is hereby granted, free of charge, to any person obtaining a
5 * copy of this software and associated documentation files (the "Software"),
6 * to deal in the Software without restriction, including without limitation
7 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
8 * and/or sell copies of the Software, and to permit persons to whom the
9 * Software is furnished to do so, subject to the following conditions:
11 * The above copyright notice and this permission notice (including the next
12 * paragraph) shall be included in all copies or substantial portions of the
15 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
18 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
20 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
25 #include "brw_context.h"
26 #include "brw_state.h"
27 #include "brw_defines.h"
30 #include "program/prog_parameter.h"
31 #include "program/prog_statevars.h"
32 #include "intel_batchbuffer.h"
35 upload_wm_state(struct brw_context
*brw
)
37 struct intel_context
*intel
= &brw
->intel
;
38 struct gl_context
*ctx
= &intel
->ctx
;
39 const struct brw_fragment_program
*fp
=
40 brw_fragment_program_const(brw
->fragment_program
);
41 bool writes_depth
= false;
45 dw1
|= GEN7_WM_STATISTICS_ENABLE
;
46 dw1
|= GEN7_WM_LINE_AA_WIDTH_1_0
;
47 dw1
|= GEN7_WM_LINE_END_CAP_AA_WIDTH_0_5
;
50 if (ctx
->Line
.StippleFlag
)
51 dw1
|= GEN7_WM_LINE_STIPPLE_ENABLE
;
54 if (ctx
->Polygon
.StippleFlag
)
55 dw1
|= GEN7_WM_POLYGON_STIPPLE_ENABLE
;
57 /* BRW_NEW_FRAGMENT_PROGRAM */
58 if (fp
->program
.Base
.InputsRead
& FRAG_BIT_WPOS
)
59 dw1
|= GEN7_WM_USES_SOURCE_DEPTH
| GEN7_WM_USES_SOURCE_W
;
60 if (fp
->program
.Base
.OutputsWritten
& BITFIELD64_BIT(FRAG_RESULT_DEPTH
)) {
62 dw1
|= GEN7_WM_PSCDEPTH_ON
;
64 /* CACHE_NEW_WM_PROG */
65 dw1
|= brw
->wm
.prog_data
->barycentric_interp_modes
<<
66 GEN7_WM_BARYCENTRIC_INTERPOLATION_MODE_SHIFT
;
69 if (fp
->program
.UsesKill
|| ctx
->Color
.AlphaEnabled
)
70 dw1
|= GEN7_WM_KILL_ENABLE
;
73 if (brw_color_buffer_write_enabled(brw
) || writes_depth
||
74 dw1
& GEN7_WM_KILL_ENABLE
) {
75 dw1
|= GEN7_WM_DISPATCH_ENABLE
;
79 OUT_BATCH(_3DSTATE_WM
<< 16 | (3 - 2));
85 const struct brw_tracked_state gen7_wm_state
= {
87 .mesa
= (_NEW_LINE
| _NEW_POLYGON
|
88 _NEW_COLOR
| _NEW_BUFFERS
),
89 .brw
= (BRW_NEW_FRAGMENT_PROGRAM
|
91 .cache
= CACHE_NEW_WM_PROG
,
93 .emit
= upload_wm_state
,
97 upload_ps_state(struct brw_context
*brw
)
99 struct intel_context
*intel
= &brw
->intel
;
100 uint32_t dw2
, dw4
, dw5
;
102 /* BRW_NEW_PS_BINDING_TABLE */
104 OUT_BATCH(_3DSTATE_BINDING_TABLE_POINTERS_PS
<< 16 | (2 - 2));
105 OUT_BATCH(brw
->bind
.bo_offset
);
108 /* CACHE_NEW_SAMPLER */
110 OUT_BATCH(_3DSTATE_SAMPLER_STATE_POINTERS_PS
<< 16 | (2 - 2));
111 OUT_BATCH(brw
->sampler
.offset
);
114 /* CACHE_NEW_WM_PROG */
115 if (brw
->wm
.prog_data
->nr_params
== 0) {
116 /* Disable the push constant buffers. */
118 OUT_BATCH(_3DSTATE_CONSTANT_PS
<< 16 | (7 - 2));
128 OUT_BATCH(_3DSTATE_CONSTANT_PS
<< 16 | (7 - 2));
130 OUT_BATCH(ALIGN(brw
->wm
.prog_data
->nr_params
,
131 brw
->wm
.prog_data
->dispatch_width
) / 8);
133 /* Pointer to the WM constant buffer. Covered by the set of
134 * state flags from gen6_upload_wm_push_constants.
136 OUT_BATCH(brw
->wm
.push_const_offset
);
145 /* CACHE_NEW_SAMPLER */
146 dw2
|= (ALIGN(brw
->sampler
.count
, 4) / 4) << GEN7_PS_SAMPLER_COUNT_SHIFT
;
148 /* Use ALT floating point mode for ARB fragment programs, because they
149 * require 0^0 == 1. Even though _CurrentFragmentProgram is used for
150 * rendering, CurrentFragmentProgram is used for this check to
151 * differentiate between the GLSL and non-GLSL cases.
153 if (intel
->ctx
.Shader
.CurrentFragmentProgram
== NULL
)
154 dw2
|= GEN7_PS_FLOATING_POINT_MODE_ALT
;
156 dw4
|= (brw
->max_wm_threads
- 1) << GEN7_PS_MAX_THREADS_SHIFT
;
158 /* CACHE_NEW_WM_PROG */
159 if (brw
->wm
.prog_data
->nr_params
> 0)
160 dw4
|= GEN7_PS_PUSH_CONSTANT_ENABLE
;
162 /* BRW_NEW_FRAGMENT_PROGRAM */
163 if (brw
->fragment_program
->Base
.InputsRead
!= 0)
164 dw4
|= GEN7_PS_ATTRIBUTE_ENABLE
;
166 if (brw
->wm
.prog_data
->dispatch_width
== 8) {
167 dw4
|= GEN7_PS_8_DISPATCH_ENABLE
;
168 if (brw
->wm
.prog_data
->prog_offset_16
)
169 dw4
|= GEN7_PS_16_DISPATCH_ENABLE
;
171 dw4
|= GEN7_PS_16_DISPATCH_ENABLE
;
174 dw5
|= (brw
->wm
.prog_data
->first_curbe_grf
<<
175 GEN7_PS_DISPATCH_START_GRF_SHIFT_0
);
176 dw5
|= (brw
->wm
.prog_data
->first_curbe_grf_16
<<
177 GEN7_PS_DISPATCH_START_GRF_SHIFT_2
);
180 OUT_BATCH(_3DSTATE_PS
<< 16 | (8 - 2));
181 OUT_BATCH(brw
->wm
.prog_offset
);
183 if (brw
->wm
.prog_data
->total_scratch
) {
184 OUT_RELOC(brw
->wm
.scratch_bo
,
185 I915_GEM_DOMAIN_RENDER
, I915_GEM_DOMAIN_RENDER
,
186 ffs(brw
->wm
.prog_data
->total_scratch
) - 11);
192 OUT_BATCH(0); /* kernel 1 pointer */
193 OUT_BATCH(brw
->wm
.prog_offset
+ brw
->wm
.prog_data
->prog_offset_16
);
197 const struct brw_tracked_state gen7_ps_state
= {
199 .mesa
= _NEW_PROGRAM_CONSTANTS
,
200 .brw
= (BRW_NEW_FRAGMENT_PROGRAM
|
201 BRW_NEW_PS_BINDING_TABLE
|
203 .cache
= (CACHE_NEW_SAMPLER
|
206 .emit
= upload_ps_state
,