2 * Copyright © 2017 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 shall be included
12 * in all copies or substantial portions of the Software.
14 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
15 * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
16 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
17 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
18 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
19 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
20 * DEALINGS IN THE SOFTWARE.
26 * ============================= GENXML CODE =============================
27 * [This file is compiled once per generation.]
28 * =======================================================================
30 * This is the main state upload code.
32 * Gallium uses Constant State Objects, or CSOs, for most state. Large,
33 * complex, or highly reusable state can be created once, and bound and
34 * rebound multiple times. This is modeled with the pipe->create_*_state()
35 * and pipe->bind_*_state() hooks. Highly dynamic or inexpensive state is
36 * streamed out on the fly, via pipe->set_*_state() hooks.
38 * OpenGL involves frequently mutating context state, which is mirrored in
39 * core Mesa by highly mutable data structures. However, most applications
40 * typically draw the same things over and over - from frame to frame, most
41 * of the same objects are still visible and need to be redrawn. So, rather
42 * than inventing new state all the time, applications usually mutate to swap
43 * between known states that we've seen before.
45 * Gallium isolates us from this mutation by tracking API state, and
46 * distilling it into a set of Constant State Objects, or CSOs. Large,
47 * complex, or typically reusable state can be created once, then reused
48 * multiple times. Drivers can create and store their own associated data.
49 * This create/bind model corresponds to the pipe->create_*_state() and
50 * pipe->bind_*_state() driver hooks.
52 * Some state is cheap to create, or expected to be highly dynamic. Rather
53 * than creating and caching piles of CSOs for these, Gallium simply streams
54 * them out, via the pipe->set_*_state() driver hooks.
56 * To reduce draw time overhead, we try to compute as much state at create
57 * time as possible. Wherever possible, we translate the Gallium pipe state
58 * to 3DSTATE commands, and store those commands in the CSO. At draw time,
59 * we can simply memcpy them into a batch buffer.
61 * No hardware matches the abstraction perfectly, so some commands require
62 * information from multiple CSOs. In this case, we can store two copies
63 * of the packet (one in each CSO), and simply | together their DWords at
64 * draw time. Sometimes the second set is trivial (one or two fields), so
65 * we simply pack it at draw time.
67 * There are two main components in the file below. First, the CSO hooks
68 * create/bind/track state. The second are the draw-time upload functions,
69 * iris_upload_render_state() and iris_upload_compute_state(), which read
70 * the context state and emit the commands into the actual batch.
81 #define __gen_validate_value(x) VALGRIND_CHECK_MEM_IS_DEFINED(&(x), sizeof(x))
87 #include "pipe/p_defines.h"
88 #include "pipe/p_state.h"
89 #include "pipe/p_context.h"
90 #include "pipe/p_screen.h"
91 #include "util/u_dual_blend.h"
92 #include "util/u_inlines.h"
93 #include "util/u_format.h"
94 #include "util/u_framebuffer.h"
95 #include "util/u_transfer.h"
96 #include "util/u_upload_mgr.h"
97 #include "util/u_viewport.h"
98 #include "drm-uapi/i915_drm.h"
100 #include "intel/compiler/brw_compiler.h"
101 #include "intel/common/gen_aux_map.h"
102 #include "intel/common/gen_l3_config.h"
103 #include "intel/common/gen_sample_positions.h"
104 #include "iris_batch.h"
105 #include "iris_context.h"
106 #include "iris_defines.h"
107 #include "iris_pipe.h"
108 #include "iris_resource.h"
110 #include "iris_genx_macros.h"
111 #include "intel/common/gen_guardband.h"
114 /* TODO: Set PTE to MOCS 61 when the kernel is ready */
115 #define MOCS_PTE (3 << 1)
116 #define MOCS_WB (2 << 1)
118 #define MOCS_PTE (1 << 1)
119 #define MOCS_WB (2 << 1)
121 #define MOCS_PTE 0x18
126 mocs(const struct iris_bo
*bo
)
128 return bo
&& bo
->external
? MOCS_PTE
: MOCS_WB
;
132 * Statically assert that PIPE_* enums match the hardware packets.
133 * (As long as they match, we don't need to translate them.)
135 UNUSED
static void pipe_asserts()
137 #define PIPE_ASSERT(x) STATIC_ASSERT((int)x)
139 /* pipe_logicop happens to match the hardware. */
140 PIPE_ASSERT(PIPE_LOGICOP_CLEAR
== LOGICOP_CLEAR
);
141 PIPE_ASSERT(PIPE_LOGICOP_NOR
== LOGICOP_NOR
);
142 PIPE_ASSERT(PIPE_LOGICOP_AND_INVERTED
== LOGICOP_AND_INVERTED
);
143 PIPE_ASSERT(PIPE_LOGICOP_COPY_INVERTED
== LOGICOP_COPY_INVERTED
);
144 PIPE_ASSERT(PIPE_LOGICOP_AND_REVERSE
== LOGICOP_AND_REVERSE
);
145 PIPE_ASSERT(PIPE_LOGICOP_INVERT
== LOGICOP_INVERT
);
146 PIPE_ASSERT(PIPE_LOGICOP_XOR
== LOGICOP_XOR
);
147 PIPE_ASSERT(PIPE_LOGICOP_NAND
== LOGICOP_NAND
);
148 PIPE_ASSERT(PIPE_LOGICOP_AND
== LOGICOP_AND
);
149 PIPE_ASSERT(PIPE_LOGICOP_EQUIV
== LOGICOP_EQUIV
);
150 PIPE_ASSERT(PIPE_LOGICOP_NOOP
== LOGICOP_NOOP
);
151 PIPE_ASSERT(PIPE_LOGICOP_OR_INVERTED
== LOGICOP_OR_INVERTED
);
152 PIPE_ASSERT(PIPE_LOGICOP_COPY
== LOGICOP_COPY
);
153 PIPE_ASSERT(PIPE_LOGICOP_OR_REVERSE
== LOGICOP_OR_REVERSE
);
154 PIPE_ASSERT(PIPE_LOGICOP_OR
== LOGICOP_OR
);
155 PIPE_ASSERT(PIPE_LOGICOP_SET
== LOGICOP_SET
);
157 /* pipe_blend_func happens to match the hardware. */
158 PIPE_ASSERT(PIPE_BLENDFACTOR_ONE
== BLENDFACTOR_ONE
);
159 PIPE_ASSERT(PIPE_BLENDFACTOR_SRC_COLOR
== BLENDFACTOR_SRC_COLOR
);
160 PIPE_ASSERT(PIPE_BLENDFACTOR_SRC_ALPHA
== BLENDFACTOR_SRC_ALPHA
);
161 PIPE_ASSERT(PIPE_BLENDFACTOR_DST_ALPHA
== BLENDFACTOR_DST_ALPHA
);
162 PIPE_ASSERT(PIPE_BLENDFACTOR_DST_COLOR
== BLENDFACTOR_DST_COLOR
);
163 PIPE_ASSERT(PIPE_BLENDFACTOR_SRC_ALPHA_SATURATE
== BLENDFACTOR_SRC_ALPHA_SATURATE
);
164 PIPE_ASSERT(PIPE_BLENDFACTOR_CONST_COLOR
== BLENDFACTOR_CONST_COLOR
);
165 PIPE_ASSERT(PIPE_BLENDFACTOR_CONST_ALPHA
== BLENDFACTOR_CONST_ALPHA
);
166 PIPE_ASSERT(PIPE_BLENDFACTOR_SRC1_COLOR
== BLENDFACTOR_SRC1_COLOR
);
167 PIPE_ASSERT(PIPE_BLENDFACTOR_SRC1_ALPHA
== BLENDFACTOR_SRC1_ALPHA
);
168 PIPE_ASSERT(PIPE_BLENDFACTOR_ZERO
== BLENDFACTOR_ZERO
);
169 PIPE_ASSERT(PIPE_BLENDFACTOR_INV_SRC_COLOR
== BLENDFACTOR_INV_SRC_COLOR
);
170 PIPE_ASSERT(PIPE_BLENDFACTOR_INV_SRC_ALPHA
== BLENDFACTOR_INV_SRC_ALPHA
);
171 PIPE_ASSERT(PIPE_BLENDFACTOR_INV_DST_ALPHA
== BLENDFACTOR_INV_DST_ALPHA
);
172 PIPE_ASSERT(PIPE_BLENDFACTOR_INV_DST_COLOR
== BLENDFACTOR_INV_DST_COLOR
);
173 PIPE_ASSERT(PIPE_BLENDFACTOR_INV_CONST_COLOR
== BLENDFACTOR_INV_CONST_COLOR
);
174 PIPE_ASSERT(PIPE_BLENDFACTOR_INV_CONST_ALPHA
== BLENDFACTOR_INV_CONST_ALPHA
);
175 PIPE_ASSERT(PIPE_BLENDFACTOR_INV_SRC1_COLOR
== BLENDFACTOR_INV_SRC1_COLOR
);
176 PIPE_ASSERT(PIPE_BLENDFACTOR_INV_SRC1_ALPHA
== BLENDFACTOR_INV_SRC1_ALPHA
);
178 /* pipe_blend_func happens to match the hardware. */
179 PIPE_ASSERT(PIPE_BLEND_ADD
== BLENDFUNCTION_ADD
);
180 PIPE_ASSERT(PIPE_BLEND_SUBTRACT
== BLENDFUNCTION_SUBTRACT
);
181 PIPE_ASSERT(PIPE_BLEND_REVERSE_SUBTRACT
== BLENDFUNCTION_REVERSE_SUBTRACT
);
182 PIPE_ASSERT(PIPE_BLEND_MIN
== BLENDFUNCTION_MIN
);
183 PIPE_ASSERT(PIPE_BLEND_MAX
== BLENDFUNCTION_MAX
);
185 /* pipe_stencil_op happens to match the hardware. */
186 PIPE_ASSERT(PIPE_STENCIL_OP_KEEP
== STENCILOP_KEEP
);
187 PIPE_ASSERT(PIPE_STENCIL_OP_ZERO
== STENCILOP_ZERO
);
188 PIPE_ASSERT(PIPE_STENCIL_OP_REPLACE
== STENCILOP_REPLACE
);
189 PIPE_ASSERT(PIPE_STENCIL_OP_INCR
== STENCILOP_INCRSAT
);
190 PIPE_ASSERT(PIPE_STENCIL_OP_DECR
== STENCILOP_DECRSAT
);
191 PIPE_ASSERT(PIPE_STENCIL_OP_INCR_WRAP
== STENCILOP_INCR
);
192 PIPE_ASSERT(PIPE_STENCIL_OP_DECR_WRAP
== STENCILOP_DECR
);
193 PIPE_ASSERT(PIPE_STENCIL_OP_INVERT
== STENCILOP_INVERT
);
195 /* pipe_sprite_coord_mode happens to match 3DSTATE_SBE */
196 PIPE_ASSERT(PIPE_SPRITE_COORD_UPPER_LEFT
== UPPERLEFT
);
197 PIPE_ASSERT(PIPE_SPRITE_COORD_LOWER_LEFT
== LOWERLEFT
);
202 translate_prim_type(enum pipe_prim_type prim
, uint8_t verts_per_patch
)
204 static const unsigned map
[] = {
205 [PIPE_PRIM_POINTS
] = _3DPRIM_POINTLIST
,
206 [PIPE_PRIM_LINES
] = _3DPRIM_LINELIST
,
207 [PIPE_PRIM_LINE_LOOP
] = _3DPRIM_LINELOOP
,
208 [PIPE_PRIM_LINE_STRIP
] = _3DPRIM_LINESTRIP
,
209 [PIPE_PRIM_TRIANGLES
] = _3DPRIM_TRILIST
,
210 [PIPE_PRIM_TRIANGLE_STRIP
] = _3DPRIM_TRISTRIP
,
211 [PIPE_PRIM_TRIANGLE_FAN
] = _3DPRIM_TRIFAN
,
212 [PIPE_PRIM_QUADS
] = _3DPRIM_QUADLIST
,
213 [PIPE_PRIM_QUAD_STRIP
] = _3DPRIM_QUADSTRIP
,
214 [PIPE_PRIM_POLYGON
] = _3DPRIM_POLYGON
,
215 [PIPE_PRIM_LINES_ADJACENCY
] = _3DPRIM_LINELIST_ADJ
,
216 [PIPE_PRIM_LINE_STRIP_ADJACENCY
] = _3DPRIM_LINESTRIP_ADJ
,
217 [PIPE_PRIM_TRIANGLES_ADJACENCY
] = _3DPRIM_TRILIST_ADJ
,
218 [PIPE_PRIM_TRIANGLE_STRIP_ADJACENCY
] = _3DPRIM_TRISTRIP_ADJ
,
219 [PIPE_PRIM_PATCHES
] = _3DPRIM_PATCHLIST_1
- 1,
222 return map
[prim
] + (prim
== PIPE_PRIM_PATCHES
? verts_per_patch
: 0);
226 translate_compare_func(enum pipe_compare_func pipe_func
)
228 static const unsigned map
[] = {
229 [PIPE_FUNC_NEVER
] = COMPAREFUNCTION_NEVER
,
230 [PIPE_FUNC_LESS
] = COMPAREFUNCTION_LESS
,
231 [PIPE_FUNC_EQUAL
] = COMPAREFUNCTION_EQUAL
,
232 [PIPE_FUNC_LEQUAL
] = COMPAREFUNCTION_LEQUAL
,
233 [PIPE_FUNC_GREATER
] = COMPAREFUNCTION_GREATER
,
234 [PIPE_FUNC_NOTEQUAL
] = COMPAREFUNCTION_NOTEQUAL
,
235 [PIPE_FUNC_GEQUAL
] = COMPAREFUNCTION_GEQUAL
,
236 [PIPE_FUNC_ALWAYS
] = COMPAREFUNCTION_ALWAYS
,
238 return map
[pipe_func
];
242 translate_shadow_func(enum pipe_compare_func pipe_func
)
244 /* Gallium specifies the result of shadow comparisons as:
246 * 1 if ref <op> texel,
251 * 0 if texel <op> ref,
254 * So we need to flip the operator and also negate.
256 static const unsigned map
[] = {
257 [PIPE_FUNC_NEVER
] = PREFILTEROPALWAYS
,
258 [PIPE_FUNC_LESS
] = PREFILTEROPLEQUAL
,
259 [PIPE_FUNC_EQUAL
] = PREFILTEROPNOTEQUAL
,
260 [PIPE_FUNC_LEQUAL
] = PREFILTEROPLESS
,
261 [PIPE_FUNC_GREATER
] = PREFILTEROPGEQUAL
,
262 [PIPE_FUNC_NOTEQUAL
] = PREFILTEROPEQUAL
,
263 [PIPE_FUNC_GEQUAL
] = PREFILTEROPGREATER
,
264 [PIPE_FUNC_ALWAYS
] = PREFILTEROPNEVER
,
266 return map
[pipe_func
];
270 translate_cull_mode(unsigned pipe_face
)
272 static const unsigned map
[4] = {
273 [PIPE_FACE_NONE
] = CULLMODE_NONE
,
274 [PIPE_FACE_FRONT
] = CULLMODE_FRONT
,
275 [PIPE_FACE_BACK
] = CULLMODE_BACK
,
276 [PIPE_FACE_FRONT_AND_BACK
] = CULLMODE_BOTH
,
278 return map
[pipe_face
];
282 translate_fill_mode(unsigned pipe_polymode
)
284 static const unsigned map
[4] = {
285 [PIPE_POLYGON_MODE_FILL
] = FILL_MODE_SOLID
,
286 [PIPE_POLYGON_MODE_LINE
] = FILL_MODE_WIREFRAME
,
287 [PIPE_POLYGON_MODE_POINT
] = FILL_MODE_POINT
,
288 [PIPE_POLYGON_MODE_FILL_RECTANGLE
] = FILL_MODE_SOLID
,
290 return map
[pipe_polymode
];
294 translate_mip_filter(enum pipe_tex_mipfilter pipe_mip
)
296 static const unsigned map
[] = {
297 [PIPE_TEX_MIPFILTER_NEAREST
] = MIPFILTER_NEAREST
,
298 [PIPE_TEX_MIPFILTER_LINEAR
] = MIPFILTER_LINEAR
,
299 [PIPE_TEX_MIPFILTER_NONE
] = MIPFILTER_NONE
,
301 return map
[pipe_mip
];
305 translate_wrap(unsigned pipe_wrap
)
307 static const unsigned map
[] = {
308 [PIPE_TEX_WRAP_REPEAT
] = TCM_WRAP
,
309 [PIPE_TEX_WRAP_CLAMP
] = TCM_HALF_BORDER
,
310 [PIPE_TEX_WRAP_CLAMP_TO_EDGE
] = TCM_CLAMP
,
311 [PIPE_TEX_WRAP_CLAMP_TO_BORDER
] = TCM_CLAMP_BORDER
,
312 [PIPE_TEX_WRAP_MIRROR_REPEAT
] = TCM_MIRROR
,
313 [PIPE_TEX_WRAP_MIRROR_CLAMP_TO_EDGE
] = TCM_MIRROR_ONCE
,
315 /* These are unsupported. */
316 [PIPE_TEX_WRAP_MIRROR_CLAMP
] = -1,
317 [PIPE_TEX_WRAP_MIRROR_CLAMP_TO_BORDER
] = -1,
319 return map
[pipe_wrap
];
323 * Allocate space for some indirect state.
325 * Return a pointer to the map (to fill it out) and a state ref (for
326 * referring to the state in GPU commands).
329 upload_state(struct u_upload_mgr
*uploader
,
330 struct iris_state_ref
*ref
,
335 u_upload_alloc(uploader
, 0, size
, alignment
, &ref
->offset
, &ref
->res
, &p
);
340 * Stream out temporary/short-lived state.
342 * This allocates space, pins the BO, and includes the BO address in the
343 * returned offset (which works because all state lives in 32-bit memory
347 stream_state(struct iris_batch
*batch
,
348 struct u_upload_mgr
*uploader
,
349 struct pipe_resource
**out_res
,
352 uint32_t *out_offset
)
356 u_upload_alloc(uploader
, 0, size
, alignment
, out_offset
, out_res
, &ptr
);
358 struct iris_bo
*bo
= iris_resource_bo(*out_res
);
359 iris_use_pinned_bo(batch
, bo
, false);
361 *out_offset
+= iris_bo_offset_from_base_address(bo
);
363 iris_record_state_size(batch
->state_sizes
, *out_offset
, size
);
369 * stream_state() + memcpy.
372 emit_state(struct iris_batch
*batch
,
373 struct u_upload_mgr
*uploader
,
374 struct pipe_resource
**out_res
,
381 stream_state(batch
, uploader
, out_res
, size
, alignment
, &offset
);
384 memcpy(map
, data
, size
);
390 * Did field 'x' change between 'old_cso' and 'new_cso'?
392 * (If so, we may want to set some dirty flags.)
394 #define cso_changed(x) (!old_cso || (old_cso->x != new_cso->x))
395 #define cso_changed_memcmp(x) \
396 (!old_cso || memcmp(old_cso->x, new_cso->x, sizeof(old_cso->x)) != 0)
399 flush_before_state_base_change(struct iris_batch
*batch
)
401 /* Flush before emitting STATE_BASE_ADDRESS.
403 * This isn't documented anywhere in the PRM. However, it seems to be
404 * necessary prior to changing the surface state base adress. We've
405 * seen issues in Vulkan where we get GPU hangs when using multi-level
406 * command buffers which clear depth, reset state base address, and then
409 * Normally, in GL, we would trust the kernel to do sufficient stalls
410 * and flushes prior to executing our batch. However, it doesn't seem
411 * as if the kernel's flushing is always sufficient and we don't want to
414 * We make this an end-of-pipe sync instead of a normal flush because we
415 * do not know the current status of the GPU. On Haswell at least,
416 * having a fast-clear operation in flight at the same time as a normal
417 * rendering operation can cause hangs. Since the kernel's flushing is
418 * insufficient, we need to ensure that any rendering operations from
419 * other processes are definitely complete before we try to do our own
420 * rendering. It's a bit of a big hammer but it appears to work.
422 iris_emit_end_of_pipe_sync(batch
,
423 "change STATE_BASE_ADDRESS (flushes)",
424 PIPE_CONTROL_RENDER_TARGET_FLUSH
|
425 PIPE_CONTROL_DEPTH_CACHE_FLUSH
|
426 PIPE_CONTROL_DATA_CACHE_FLUSH
);
430 flush_after_state_base_change(struct iris_batch
*batch
)
432 /* After re-setting the surface state base address, we have to do some
433 * cache flusing so that the sampler engine will pick up the new
434 * SURFACE_STATE objects and binding tables. From the Broadwell PRM,
435 * Shared Function > 3D Sampler > State > State Caching (page 96):
437 * Coherency with system memory in the state cache, like the texture
438 * cache is handled partially by software. It is expected that the
439 * command stream or shader will issue Cache Flush operation or
440 * Cache_Flush sampler message to ensure that the L1 cache remains
441 * coherent with system memory.
445 * Whenever the value of the Dynamic_State_Base_Addr,
446 * Surface_State_Base_Addr are altered, the L1 state cache must be
447 * invalidated to ensure the new surface or sampler state is fetched
448 * from system memory.
450 * The PIPE_CONTROL command has a "State Cache Invalidation Enable" bit
451 * which, according the PIPE_CONTROL instruction documentation in the
454 * Setting this bit is independent of any other bit in this packet.
455 * This bit controls the invalidation of the L1 and L2 state caches
456 * at the top of the pipe i.e. at the parsing time.
458 * Unfortunately, experimentation seems to indicate that state cache
459 * invalidation through a PIPE_CONTROL does nothing whatsoever in
460 * regards to surface state and binding tables. In stead, it seems that
461 * invalidating the texture cache is what is actually needed.
463 * XXX: As far as we have been able to determine through
464 * experimentation, shows that flush the texture cache appears to be
465 * sufficient. The theory here is that all of the sampling/rendering
466 * units cache the binding table in the texture cache. However, we have
467 * yet to be able to actually confirm this.
469 iris_emit_end_of_pipe_sync(batch
,
470 "change STATE_BASE_ADDRESS (invalidates)",
471 PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE
|
472 PIPE_CONTROL_CONST_CACHE_INVALIDATE
|
473 PIPE_CONTROL_STATE_CACHE_INVALIDATE
);
477 _iris_emit_lri(struct iris_batch
*batch
, uint32_t reg
, uint32_t val
)
479 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_IMM
), lri
) {
480 lri
.RegisterOffset
= reg
;
484 #define iris_emit_lri(b, r, v) _iris_emit_lri(b, GENX(r##_num), v)
487 _iris_emit_lrr(struct iris_batch
*batch
, uint32_t dst
, uint32_t src
)
489 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_REG
), lrr
) {
490 lrr
.SourceRegisterAddress
= src
;
491 lrr
.DestinationRegisterAddress
= dst
;
496 iris_load_register_reg32(struct iris_batch
*batch
, uint32_t dst
,
499 _iris_emit_lrr(batch
, dst
, src
);
503 iris_load_register_reg64(struct iris_batch
*batch
, uint32_t dst
,
506 _iris_emit_lrr(batch
, dst
, src
);
507 _iris_emit_lrr(batch
, dst
+ 4, src
+ 4);
511 iris_load_register_imm32(struct iris_batch
*batch
, uint32_t reg
,
514 _iris_emit_lri(batch
, reg
, val
);
518 iris_load_register_imm64(struct iris_batch
*batch
, uint32_t reg
,
521 _iris_emit_lri(batch
, reg
+ 0, val
& 0xffffffff);
522 _iris_emit_lri(batch
, reg
+ 4, val
>> 32);
526 * Emit MI_LOAD_REGISTER_MEM to load a 32-bit MMIO register from a buffer.
529 iris_load_register_mem32(struct iris_batch
*batch
, uint32_t reg
,
530 struct iris_bo
*bo
, uint32_t offset
)
532 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_MEM
), lrm
) {
533 lrm
.RegisterAddress
= reg
;
534 lrm
.MemoryAddress
= ro_bo(bo
, offset
);
539 * Load a 64-bit value from a buffer into a MMIO register via
540 * two MI_LOAD_REGISTER_MEM commands.
543 iris_load_register_mem64(struct iris_batch
*batch
, uint32_t reg
,
544 struct iris_bo
*bo
, uint32_t offset
)
546 iris_load_register_mem32(batch
, reg
+ 0, bo
, offset
+ 0);
547 iris_load_register_mem32(batch
, reg
+ 4, bo
, offset
+ 4);
551 iris_store_register_mem32(struct iris_batch
*batch
, uint32_t reg
,
552 struct iris_bo
*bo
, uint32_t offset
,
555 iris_emit_cmd(batch
, GENX(MI_STORE_REGISTER_MEM
), srm
) {
556 srm
.RegisterAddress
= reg
;
557 srm
.MemoryAddress
= rw_bo(bo
, offset
);
558 srm
.PredicateEnable
= predicated
;
563 iris_store_register_mem64(struct iris_batch
*batch
, uint32_t reg
,
564 struct iris_bo
*bo
, uint32_t offset
,
567 iris_store_register_mem32(batch
, reg
+ 0, bo
, offset
+ 0, predicated
);
568 iris_store_register_mem32(batch
, reg
+ 4, bo
, offset
+ 4, predicated
);
572 iris_store_data_imm32(struct iris_batch
*batch
,
573 struct iris_bo
*bo
, uint32_t offset
,
576 iris_emit_cmd(batch
, GENX(MI_STORE_DATA_IMM
), sdi
) {
577 sdi
.Address
= rw_bo(bo
, offset
);
578 sdi
.ImmediateData
= imm
;
583 iris_store_data_imm64(struct iris_batch
*batch
,
584 struct iris_bo
*bo
, uint32_t offset
,
587 /* Can't use iris_emit_cmd because MI_STORE_DATA_IMM has a length of
588 * 2 in genxml but it's actually variable length and we need 5 DWords.
590 void *map
= iris_get_command_space(batch
, 4 * 5);
591 _iris_pack_command(batch
, GENX(MI_STORE_DATA_IMM
), map
, sdi
) {
592 sdi
.DWordLength
= 5 - 2;
593 sdi
.Address
= rw_bo(bo
, offset
);
594 sdi
.ImmediateData
= imm
;
599 iris_copy_mem_mem(struct iris_batch
*batch
,
600 struct iris_bo
*dst_bo
, uint32_t dst_offset
,
601 struct iris_bo
*src_bo
, uint32_t src_offset
,
604 /* MI_COPY_MEM_MEM operates on DWords. */
605 assert(bytes
% 4 == 0);
606 assert(dst_offset
% 4 == 0);
607 assert(src_offset
% 4 == 0);
609 for (unsigned i
= 0; i
< bytes
; i
+= 4) {
610 iris_emit_cmd(batch
, GENX(MI_COPY_MEM_MEM
), cp
) {
611 cp
.DestinationMemoryAddress
= rw_bo(dst_bo
, dst_offset
+ i
);
612 cp
.SourceMemoryAddress
= ro_bo(src_bo
, src_offset
+ i
);
618 emit_pipeline_select(struct iris_batch
*batch
, uint32_t pipeline
)
620 #if GEN_GEN >= 8 && GEN_GEN < 10
621 /* From the Broadwell PRM, Volume 2a: Instructions, PIPELINE_SELECT:
623 * Software must clear the COLOR_CALC_STATE Valid field in
624 * 3DSTATE_CC_STATE_POINTERS command prior to send a PIPELINE_SELECT
625 * with Pipeline Select set to GPGPU.
627 * The internal hardware docs recommend the same workaround for Gen9
630 if (pipeline
== GPGPU
)
631 iris_emit_cmd(batch
, GENX(3DSTATE_CC_STATE_POINTERS
), t
);
635 /* From "BXML » GT » MI » vol1a GPU Overview » [Instruction]
636 * PIPELINE_SELECT [DevBWR+]":
640 * Software must ensure all the write caches are flushed through a
641 * stalling PIPE_CONTROL command followed by another PIPE_CONTROL
642 * command to invalidate read only caches prior to programming
643 * MI_PIPELINE_SELECT command to change the Pipeline Select Mode."
645 iris_emit_pipe_control_flush(batch
,
646 "workaround: PIPELINE_SELECT flushes (1/2)",
647 PIPE_CONTROL_RENDER_TARGET_FLUSH
|
648 PIPE_CONTROL_DEPTH_CACHE_FLUSH
|
649 PIPE_CONTROL_DATA_CACHE_FLUSH
|
650 PIPE_CONTROL_CS_STALL
);
652 iris_emit_pipe_control_flush(batch
,
653 "workaround: PIPELINE_SELECT flushes (2/2)",
654 PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE
|
655 PIPE_CONTROL_CONST_CACHE_INVALIDATE
|
656 PIPE_CONTROL_STATE_CACHE_INVALIDATE
|
657 PIPE_CONTROL_INSTRUCTION_INVALIDATE
);
659 iris_emit_cmd(batch
, GENX(PIPELINE_SELECT
), sel
) {
663 sel
.PipelineSelection
= pipeline
;
668 init_glk_barrier_mode(struct iris_batch
*batch
, uint32_t value
)
673 * "This chicken bit works around a hardware issue with barrier
674 * logic encountered when switching between GPGPU and 3D pipelines.
675 * To workaround the issue, this mode bit should be set after a
676 * pipeline is selected."
679 iris_pack_state(GENX(SLICE_COMMON_ECO_CHICKEN1
), ®_val
, reg
) {
680 reg
.GLKBarrierMode
= value
;
681 reg
.GLKBarrierModeMask
= 1;
683 iris_emit_lri(batch
, SLICE_COMMON_ECO_CHICKEN1
, reg_val
);
688 init_state_base_address(struct iris_batch
*batch
)
690 flush_before_state_base_change(batch
);
692 /* We program most base addresses once at context initialization time.
693 * Each base address points at a 4GB memory zone, and never needs to
694 * change. See iris_bufmgr.h for a description of the memory zones.
696 * The one exception is Surface State Base Address, which needs to be
697 * updated occasionally. See iris_binder.c for the details there.
699 iris_emit_cmd(batch
, GENX(STATE_BASE_ADDRESS
), sba
) {
700 sba
.GeneralStateMOCS
= MOCS_WB
;
701 sba
.StatelessDataPortAccessMOCS
= MOCS_WB
;
702 sba
.DynamicStateMOCS
= MOCS_WB
;
703 sba
.IndirectObjectMOCS
= MOCS_WB
;
704 sba
.InstructionMOCS
= MOCS_WB
;
705 sba
.SurfaceStateMOCS
= MOCS_WB
;
707 sba
.GeneralStateBaseAddressModifyEnable
= true;
708 sba
.DynamicStateBaseAddressModifyEnable
= true;
709 sba
.IndirectObjectBaseAddressModifyEnable
= true;
710 sba
.InstructionBaseAddressModifyEnable
= true;
711 sba
.GeneralStateBufferSizeModifyEnable
= true;
712 sba
.DynamicStateBufferSizeModifyEnable
= true;
714 sba
.BindlessSurfaceStateBaseAddressModifyEnable
= true;
715 sba
.BindlessSurfaceStateMOCS
= MOCS_WB
;
717 sba
.IndirectObjectBufferSizeModifyEnable
= true;
718 sba
.InstructionBuffersizeModifyEnable
= true;
720 sba
.InstructionBaseAddress
= ro_bo(NULL
, IRIS_MEMZONE_SHADER_START
);
721 sba
.DynamicStateBaseAddress
= ro_bo(NULL
, IRIS_MEMZONE_DYNAMIC_START
);
723 sba
.GeneralStateBufferSize
= 0xfffff;
724 sba
.IndirectObjectBufferSize
= 0xfffff;
725 sba
.InstructionBufferSize
= 0xfffff;
726 sba
.DynamicStateBufferSize
= 0xfffff;
729 flush_after_state_base_change(batch
);
733 iris_emit_l3_config(struct iris_batch
*batch
, const struct gen_l3_config
*cfg
,
734 bool has_slm
, bool wants_dc_cache
)
739 #define L3_ALLOCATION_REG GENX(L3ALLOC)
740 #define L3_ALLOCATION_REG_num GENX(L3ALLOC_num)
742 #define L3_ALLOCATION_REG GENX(L3CNTLREG)
743 #define L3_ALLOCATION_REG_num GENX(L3CNTLREG_num)
746 iris_pack_state(L3_ALLOCATION_REG
, ®_val
, reg
) {
748 reg
.SLMEnable
= has_slm
;
751 /* WA_1406697149: Bit 9 "Error Detection Behavior Control" must be set
752 * in L3CNTLREG register. The default setting of the bit is not the
753 * desirable behavior.
755 reg
.ErrorDetectionBehaviorControl
= true;
756 reg
.UseFullWays
= true;
758 reg
.URBAllocation
= cfg
->n
[GEN_L3P_URB
];
759 reg
.ROAllocation
= cfg
->n
[GEN_L3P_RO
];
760 reg
.DCAllocation
= cfg
->n
[GEN_L3P_DC
];
761 reg
.AllAllocation
= cfg
->n
[GEN_L3P_ALL
];
763 _iris_emit_lri(batch
, L3_ALLOCATION_REG_num
, reg_val
);
767 iris_emit_default_l3_config(struct iris_batch
*batch
,
768 const struct gen_device_info
*devinfo
,
771 bool wants_dc_cache
= true;
772 bool has_slm
= compute
;
773 const struct gen_l3_weights w
=
774 gen_get_default_l3_weights(devinfo
, wants_dc_cache
, has_slm
);
775 const struct gen_l3_config
*cfg
= gen_get_l3_config(devinfo
, w
);
776 iris_emit_l3_config(batch
, cfg
, has_slm
, wants_dc_cache
);
779 #if GEN_GEN == 9 || GEN_GEN == 10
781 iris_enable_obj_preemption(struct iris_batch
*batch
, bool enable
)
785 /* A fixed function pipe flush is required before modifying this field */
786 iris_emit_end_of_pipe_sync(batch
, enable
? "enable preemption"
787 : "disable preemption",
788 PIPE_CONTROL_RENDER_TARGET_FLUSH
);
790 /* enable object level preemption */
791 iris_pack_state(GENX(CS_CHICKEN1
), ®_val
, reg
) {
792 reg
.ReplayMode
= enable
;
793 reg
.ReplayModeMask
= true;
795 iris_emit_lri(batch
, CS_CHICKEN1
, reg_val
);
801 iris_upload_slice_hashing_state(struct iris_batch
*batch
)
803 const struct gen_device_info
*devinfo
= &batch
->screen
->devinfo
;
804 int subslices_delta
=
805 devinfo
->ppipe_subslices
[0] - devinfo
->ppipe_subslices
[1];
806 if (subslices_delta
== 0)
809 struct iris_context
*ice
= NULL
;
810 ice
= container_of(batch
, ice
, batches
[IRIS_BATCH_RENDER
]);
811 assert(&ice
->batches
[IRIS_BATCH_RENDER
] == batch
);
813 unsigned size
= GENX(SLICE_HASH_TABLE_length
) * 4;
814 uint32_t hash_address
;
815 struct pipe_resource
*tmp
= NULL
;
817 stream_state(batch
, ice
->state
.dynamic_uploader
, &tmp
,
818 size
, 64, &hash_address
);
819 pipe_resource_reference(&tmp
, NULL
);
821 struct GENX(SLICE_HASH_TABLE
) table0
= {
823 { 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1 },
824 { 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1 },
825 { 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0 },
826 { 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1 },
827 { 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1 },
828 { 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0 },
829 { 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1 },
830 { 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1 },
831 { 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0 },
832 { 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1 },
833 { 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1 },
834 { 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0 },
835 { 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1 },
836 { 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1 },
837 { 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0 },
838 { 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1 }
842 struct GENX(SLICE_HASH_TABLE
) table1
= {
844 { 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0 },
845 { 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0 },
846 { 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1 },
847 { 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0 },
848 { 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0 },
849 { 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1 },
850 { 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0 },
851 { 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0 },
852 { 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1 },
853 { 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0 },
854 { 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0 },
855 { 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1 },
856 { 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0 },
857 { 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0 },
858 { 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1 },
859 { 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0 }
863 const struct GENX(SLICE_HASH_TABLE
) *table
=
864 subslices_delta
< 0 ? &table0
: &table1
;
865 GENX(SLICE_HASH_TABLE_pack
)(NULL
, map
, table
);
867 iris_emit_cmd(batch
, GENX(3DSTATE_SLICE_TABLE_STATE_POINTERS
), ptr
) {
868 ptr
.SliceHashStatePointerValid
= true;
869 ptr
.SliceHashTableStatePointer
= hash_address
;
872 iris_emit_cmd(batch
, GENX(3DSTATE_3D_MODE
), mode
) {
873 mode
.SliceHashingTableEnable
= true;
879 iris_alloc_push_constants(struct iris_batch
*batch
)
881 /* For now, we set a static partitioning of the push constant area,
882 * assuming that all stages could be in use.
884 * TODO: Try lazily allocating the HS/DS/GS sections as needed, and
885 * see if that improves performance by offering more space to
886 * the VS/FS when those aren't in use. Also, try dynamically
887 * enabling/disabling it like i965 does. This would be more
888 * stalls and may not actually help; we don't know yet.
890 for (int i
= 0; i
<= MESA_SHADER_FRAGMENT
; i
++) {
891 iris_emit_cmd(batch
, GENX(3DSTATE_PUSH_CONSTANT_ALLOC_VS
), alloc
) {
892 alloc
._3DCommandSubOpcode
= 18 + i
;
893 alloc
.ConstantBufferOffset
= 6 * i
;
894 alloc
.ConstantBufferSize
= i
== MESA_SHADER_FRAGMENT
? 8 : 6;
900 * Upload the initial GPU state for a render context.
902 * This sets some invariant state that needs to be programmed a particular
903 * way, but we never actually change.
906 iris_init_render_context(struct iris_batch
*batch
)
908 UNUSED
const struct gen_device_info
*devinfo
= &batch
->screen
->devinfo
;
911 emit_pipeline_select(batch
, _3D
);
913 iris_emit_default_l3_config(batch
, devinfo
, false);
915 init_state_base_address(batch
);
918 iris_pack_state(GENX(CS_DEBUG_MODE2
), ®_val
, reg
) {
919 reg
.CONSTANT_BUFFERAddressOffsetDisable
= true;
920 reg
.CONSTANT_BUFFERAddressOffsetDisableMask
= true;
922 iris_emit_lri(batch
, CS_DEBUG_MODE2
, reg_val
);
924 iris_pack_state(GENX(INSTPM
), ®_val
, reg
) {
925 reg
.CONSTANT_BUFFERAddressOffsetDisable
= true;
926 reg
.CONSTANT_BUFFERAddressOffsetDisableMask
= true;
928 iris_emit_lri(batch
, INSTPM
, reg_val
);
932 iris_pack_state(GENX(CACHE_MODE_1
), ®_val
, reg
) {
933 reg
.FloatBlendOptimizationEnable
= true;
934 reg
.FloatBlendOptimizationEnableMask
= true;
935 reg
.PartialResolveDisableInVC
= true;
936 reg
.PartialResolveDisableInVCMask
= true;
938 iris_emit_lri(batch
, CACHE_MODE_1
, reg_val
);
940 if (devinfo
->is_geminilake
)
941 init_glk_barrier_mode(batch
, GLK_BARRIER_MODE_3D_HULL
);
945 iris_pack_state(GENX(SAMPLER_MODE
), ®_val
, reg
) {
946 reg
.HeaderlessMessageforPreemptableContexts
= 1;
947 reg
.HeaderlessMessageforPreemptableContextsMask
= 1;
949 iris_emit_lri(batch
, SAMPLER_MODE
, reg_val
);
951 /* Bit 1 must be set in HALF_SLICE_CHICKEN7. */
952 iris_pack_state(GENX(HALF_SLICE_CHICKEN7
), ®_val
, reg
) {
953 reg
.EnabledTexelOffsetPrecisionFix
= 1;
954 reg
.EnabledTexelOffsetPrecisionFixMask
= 1;
956 iris_emit_lri(batch
, HALF_SLICE_CHICKEN7
, reg_val
);
958 /* Hardware specification recommends disabling repacking for the
959 * compatibility with decompression mechanism in display controller.
961 if (devinfo
->disable_ccs_repack
) {
962 iris_pack_state(GENX(CACHE_MODE_0
), ®_val
, reg
) {
963 reg
.DisableRepackingforCompression
= true;
964 reg
.DisableRepackingforCompressionMask
= true;
966 iris_emit_lri(batch
, CACHE_MODE_0
, reg_val
);
969 iris_upload_slice_hashing_state(batch
);
972 /* 3DSTATE_DRAWING_RECTANGLE is non-pipelined, so we want to avoid
973 * changing it dynamically. We set it to the maximum size here, and
974 * instead include the render target dimensions in the viewport, so
975 * viewport extents clipping takes care of pruning stray geometry.
977 iris_emit_cmd(batch
, GENX(3DSTATE_DRAWING_RECTANGLE
), rect
) {
978 rect
.ClippedDrawingRectangleXMax
= UINT16_MAX
;
979 rect
.ClippedDrawingRectangleYMax
= UINT16_MAX
;
982 /* Set the initial MSAA sample positions. */
983 iris_emit_cmd(batch
, GENX(3DSTATE_SAMPLE_PATTERN
), pat
) {
984 GEN_SAMPLE_POS_1X(pat
._1xSample
);
985 GEN_SAMPLE_POS_2X(pat
._2xSample
);
986 GEN_SAMPLE_POS_4X(pat
._4xSample
);
987 GEN_SAMPLE_POS_8X(pat
._8xSample
);
989 GEN_SAMPLE_POS_16X(pat
._16xSample
);
993 /* Use the legacy AA line coverage computation. */
994 iris_emit_cmd(batch
, GENX(3DSTATE_AA_LINE_PARAMETERS
), foo
);
996 /* Disable chromakeying (it's for media) */
997 iris_emit_cmd(batch
, GENX(3DSTATE_WM_CHROMAKEY
), foo
);
999 /* We want regular rendering, not special HiZ operations. */
1000 iris_emit_cmd(batch
, GENX(3DSTATE_WM_HZ_OP
), foo
);
1002 /* No polygon stippling offsets are necessary. */
1003 /* TODO: may need to set an offset for origin-UL framebuffers */
1004 iris_emit_cmd(batch
, GENX(3DSTATE_POLY_STIPPLE_OFFSET
), foo
);
1006 iris_alloc_push_constants(batch
);
1009 /* Gen11+ is enabled for us by the kernel. */
1010 iris_enable_obj_preemption(batch
, true);
1015 iris_init_compute_context(struct iris_batch
*batch
)
1017 UNUSED
const struct gen_device_info
*devinfo
= &batch
->screen
->devinfo
;
1019 emit_pipeline_select(batch
, GPGPU
);
1021 iris_emit_default_l3_config(batch
, devinfo
, true);
1023 init_state_base_address(batch
);
1026 if (devinfo
->is_geminilake
)
1027 init_glk_barrier_mode(batch
, GLK_BARRIER_MODE_GPGPU
);
1031 struct iris_vertex_buffer_state
{
1032 /** The VERTEX_BUFFER_STATE hardware structure. */
1033 uint32_t state
[GENX(VERTEX_BUFFER_STATE_length
)];
1035 /** The resource to source vertex data from. */
1036 struct pipe_resource
*resource
;
1041 struct iris_depth_buffer_state
{
1042 /* Depth/HiZ/Stencil related hardware packets. */
1043 uint32_t packets
[GENX(3DSTATE_DEPTH_BUFFER_length
) +
1044 GENX(3DSTATE_STENCIL_BUFFER_length
) +
1045 GENX(3DSTATE_HIER_DEPTH_BUFFER_length
) +
1046 GENX(3DSTATE_CLEAR_PARAMS_length
)];
1050 * Generation-specific context state (ice->state.genx->...).
1052 * Most state can go in iris_context directly, but these encode hardware
1053 * packets which vary by generation.
1055 struct iris_genx_state
{
1056 struct iris_vertex_buffer_state vertex_buffers
[33];
1057 uint32_t last_index_buffer
[GENX(3DSTATE_INDEX_BUFFER_length
)];
1059 struct iris_depth_buffer_state depth_buffer
;
1061 uint32_t so_buffers
[4 * GENX(3DSTATE_SO_BUFFER_length
)];
1064 bool pma_fix_enabled
;
1068 /* Is object level preemption enabled? */
1069 bool object_preemption
;
1074 struct brw_image_param image_param
[PIPE_MAX_SHADER_IMAGES
];
1076 } shaders
[MESA_SHADER_STAGES
];
1080 * The pipe->set_blend_color() driver hook.
1082 * This corresponds to our COLOR_CALC_STATE.
1085 iris_set_blend_color(struct pipe_context
*ctx
,
1086 const struct pipe_blend_color
*state
)
1088 struct iris_context
*ice
= (struct iris_context
*) ctx
;
1090 /* Our COLOR_CALC_STATE is exactly pipe_blend_color, so just memcpy */
1091 memcpy(&ice
->state
.blend_color
, state
, sizeof(struct pipe_blend_color
));
1092 ice
->state
.dirty
|= IRIS_DIRTY_COLOR_CALC_STATE
;
1096 * Gallium CSO for blend state (see pipe_blend_state).
1098 struct iris_blend_state
{
1099 /** Partial 3DSTATE_PS_BLEND */
1100 uint32_t ps_blend
[GENX(3DSTATE_PS_BLEND_length
)];
1102 /** Partial BLEND_STATE */
1103 uint32_t blend_state
[GENX(BLEND_STATE_length
) +
1104 BRW_MAX_DRAW_BUFFERS
* GENX(BLEND_STATE_ENTRY_length
)];
1106 bool alpha_to_coverage
; /* for shader key */
1108 /** Bitfield of whether blending is enabled for RT[i] - for aux resolves */
1109 uint8_t blend_enables
;
1111 /** Bitfield of whether color writes are enabled for RT[i] */
1112 uint8_t color_write_enables
;
1114 /** Does RT[0] use dual color blending? */
1115 bool dual_color_blending
;
1118 static enum pipe_blendfactor
1119 fix_blendfactor(enum pipe_blendfactor f
, bool alpha_to_one
)
1122 if (f
== PIPE_BLENDFACTOR_SRC1_ALPHA
)
1123 return PIPE_BLENDFACTOR_ONE
;
1125 if (f
== PIPE_BLENDFACTOR_INV_SRC1_ALPHA
)
1126 return PIPE_BLENDFACTOR_ZERO
;
1133 * The pipe->create_blend_state() driver hook.
1135 * Translates a pipe_blend_state into iris_blend_state.
1138 iris_create_blend_state(struct pipe_context
*ctx
,
1139 const struct pipe_blend_state
*state
)
1141 struct iris_blend_state
*cso
= malloc(sizeof(struct iris_blend_state
));
1142 uint32_t *blend_entry
= cso
->blend_state
+ GENX(BLEND_STATE_length
);
1144 cso
->blend_enables
= 0;
1145 cso
->color_write_enables
= 0;
1146 STATIC_ASSERT(BRW_MAX_DRAW_BUFFERS
<= 8);
1148 cso
->alpha_to_coverage
= state
->alpha_to_coverage
;
1150 bool indep_alpha_blend
= false;
1152 for (int i
= 0; i
< BRW_MAX_DRAW_BUFFERS
; i
++) {
1153 const struct pipe_rt_blend_state
*rt
=
1154 &state
->rt
[state
->independent_blend_enable
? i
: 0];
1156 enum pipe_blendfactor src_rgb
=
1157 fix_blendfactor(rt
->rgb_src_factor
, state
->alpha_to_one
);
1158 enum pipe_blendfactor src_alpha
=
1159 fix_blendfactor(rt
->alpha_src_factor
, state
->alpha_to_one
);
1160 enum pipe_blendfactor dst_rgb
=
1161 fix_blendfactor(rt
->rgb_dst_factor
, state
->alpha_to_one
);
1162 enum pipe_blendfactor dst_alpha
=
1163 fix_blendfactor(rt
->alpha_dst_factor
, state
->alpha_to_one
);
1165 if (rt
->rgb_func
!= rt
->alpha_func
||
1166 src_rgb
!= src_alpha
|| dst_rgb
!= dst_alpha
)
1167 indep_alpha_blend
= true;
1169 if (rt
->blend_enable
)
1170 cso
->blend_enables
|= 1u << i
;
1173 cso
->color_write_enables
|= 1u << i
;
1175 iris_pack_state(GENX(BLEND_STATE_ENTRY
), blend_entry
, be
) {
1176 be
.LogicOpEnable
= state
->logicop_enable
;
1177 be
.LogicOpFunction
= state
->logicop_func
;
1179 be
.PreBlendSourceOnlyClampEnable
= false;
1180 be
.ColorClampRange
= COLORCLAMP_RTFORMAT
;
1181 be
.PreBlendColorClampEnable
= true;
1182 be
.PostBlendColorClampEnable
= true;
1184 be
.ColorBufferBlendEnable
= rt
->blend_enable
;
1186 be
.ColorBlendFunction
= rt
->rgb_func
;
1187 be
.AlphaBlendFunction
= rt
->alpha_func
;
1188 be
.SourceBlendFactor
= src_rgb
;
1189 be
.SourceAlphaBlendFactor
= src_alpha
;
1190 be
.DestinationBlendFactor
= dst_rgb
;
1191 be
.DestinationAlphaBlendFactor
= dst_alpha
;
1193 be
.WriteDisableRed
= !(rt
->colormask
& PIPE_MASK_R
);
1194 be
.WriteDisableGreen
= !(rt
->colormask
& PIPE_MASK_G
);
1195 be
.WriteDisableBlue
= !(rt
->colormask
& PIPE_MASK_B
);
1196 be
.WriteDisableAlpha
= !(rt
->colormask
& PIPE_MASK_A
);
1198 blend_entry
+= GENX(BLEND_STATE_ENTRY_length
);
1201 iris_pack_command(GENX(3DSTATE_PS_BLEND
), cso
->ps_blend
, pb
) {
1202 /* pb.HasWriteableRT is filled in at draw time.
1203 * pb.AlphaTestEnable is filled in at draw time.
1205 * pb.ColorBufferBlendEnable is filled in at draw time so we can avoid
1206 * setting it when dual color blending without an appropriate shader.
1209 pb
.AlphaToCoverageEnable
= state
->alpha_to_coverage
;
1210 pb
.IndependentAlphaBlendEnable
= indep_alpha_blend
;
1212 pb
.SourceBlendFactor
=
1213 fix_blendfactor(state
->rt
[0].rgb_src_factor
, state
->alpha_to_one
);
1214 pb
.SourceAlphaBlendFactor
=
1215 fix_blendfactor(state
->rt
[0].alpha_src_factor
, state
->alpha_to_one
);
1216 pb
.DestinationBlendFactor
=
1217 fix_blendfactor(state
->rt
[0].rgb_dst_factor
, state
->alpha_to_one
);
1218 pb
.DestinationAlphaBlendFactor
=
1219 fix_blendfactor(state
->rt
[0].alpha_dst_factor
, state
->alpha_to_one
);
1222 iris_pack_state(GENX(BLEND_STATE
), cso
->blend_state
, bs
) {
1223 bs
.AlphaToCoverageEnable
= state
->alpha_to_coverage
;
1224 bs
.IndependentAlphaBlendEnable
= indep_alpha_blend
;
1225 bs
.AlphaToOneEnable
= state
->alpha_to_one
;
1226 bs
.AlphaToCoverageDitherEnable
= state
->alpha_to_coverage
;
1227 bs
.ColorDitherEnable
= state
->dither
;
1228 /* bl.AlphaTestEnable and bs.AlphaTestFunction are filled in later. */
1231 cso
->dual_color_blending
= util_blend_state_is_dual(state
, 0);
1237 * The pipe->bind_blend_state() driver hook.
1239 * Bind a blending CSO and flag related dirty bits.
1242 iris_bind_blend_state(struct pipe_context
*ctx
, void *state
)
1244 struct iris_context
*ice
= (struct iris_context
*) ctx
;
1245 struct iris_blend_state
*cso
= state
;
1247 ice
->state
.cso_blend
= cso
;
1248 ice
->state
.blend_enables
= cso
? cso
->blend_enables
: 0;
1250 ice
->state
.dirty
|= IRIS_DIRTY_PS_BLEND
;
1251 ice
->state
.dirty
|= IRIS_DIRTY_BLEND_STATE
;
1252 ice
->state
.dirty
|= IRIS_DIRTY_RENDER_RESOLVES_AND_FLUSHES
;
1253 ice
->state
.dirty
|= ice
->state
.dirty_for_nos
[IRIS_NOS_BLEND
];
1256 ice
->state
.dirty
|= IRIS_DIRTY_PMA_FIX
;
1260 * Return true if the FS writes to any color outputs which are not disabled
1261 * via color masking.
1264 has_writeable_rt(const struct iris_blend_state
*cso_blend
,
1265 const struct shader_info
*fs_info
)
1270 unsigned rt_outputs
= fs_info
->outputs_written
>> FRAG_RESULT_DATA0
;
1272 if (fs_info
->outputs_written
& BITFIELD64_BIT(FRAG_RESULT_COLOR
))
1273 rt_outputs
= (1 << BRW_MAX_DRAW_BUFFERS
) - 1;
1275 return cso_blend
->color_write_enables
& rt_outputs
;
1279 * Gallium CSO for depth, stencil, and alpha testing state.
1281 struct iris_depth_stencil_alpha_state
{
1282 /** Partial 3DSTATE_WM_DEPTH_STENCIL. */
1283 uint32_t wmds
[GENX(3DSTATE_WM_DEPTH_STENCIL_length
)];
1286 uint32_t depth_bounds
[GENX(3DSTATE_DEPTH_BOUNDS_length
)];
1289 /** Outbound to BLEND_STATE, 3DSTATE_PS_BLEND, COLOR_CALC_STATE. */
1290 struct pipe_alpha_state alpha
;
1292 /** Outbound to resolve and cache set tracking. */
1293 bool depth_writes_enabled
;
1294 bool stencil_writes_enabled
;
1296 /** Outbound to Gen8-9 PMA stall equations */
1297 bool depth_test_enabled
;
1301 * The pipe->create_depth_stencil_alpha_state() driver hook.
1303 * We encode most of 3DSTATE_WM_DEPTH_STENCIL, and just save off the alpha
1304 * testing state since we need pieces of it in a variety of places.
1307 iris_create_zsa_state(struct pipe_context
*ctx
,
1308 const struct pipe_depth_stencil_alpha_state
*state
)
1310 struct iris_depth_stencil_alpha_state
*cso
=
1311 malloc(sizeof(struct iris_depth_stencil_alpha_state
));
1313 bool two_sided_stencil
= state
->stencil
[1].enabled
;
1315 cso
->alpha
= state
->alpha
;
1316 cso
->depth_writes_enabled
= state
->depth
.writemask
;
1317 cso
->depth_test_enabled
= state
->depth
.enabled
;
1318 cso
->stencil_writes_enabled
=
1319 state
->stencil
[0].writemask
!= 0 ||
1320 (two_sided_stencil
&& state
->stencil
[1].writemask
!= 0);
1322 /* The state tracker needs to optimize away EQUAL writes for us. */
1323 assert(!(state
->depth
.func
== PIPE_FUNC_EQUAL
&& state
->depth
.writemask
));
1325 iris_pack_command(GENX(3DSTATE_WM_DEPTH_STENCIL
), cso
->wmds
, wmds
) {
1326 wmds
.StencilFailOp
= state
->stencil
[0].fail_op
;
1327 wmds
.StencilPassDepthFailOp
= state
->stencil
[0].zfail_op
;
1328 wmds
.StencilPassDepthPassOp
= state
->stencil
[0].zpass_op
;
1329 wmds
.StencilTestFunction
=
1330 translate_compare_func(state
->stencil
[0].func
);
1331 wmds
.BackfaceStencilFailOp
= state
->stencil
[1].fail_op
;
1332 wmds
.BackfaceStencilPassDepthFailOp
= state
->stencil
[1].zfail_op
;
1333 wmds
.BackfaceStencilPassDepthPassOp
= state
->stencil
[1].zpass_op
;
1334 wmds
.BackfaceStencilTestFunction
=
1335 translate_compare_func(state
->stencil
[1].func
);
1336 wmds
.DepthTestFunction
= translate_compare_func(state
->depth
.func
);
1337 wmds
.DoubleSidedStencilEnable
= two_sided_stencil
;
1338 wmds
.StencilTestEnable
= state
->stencil
[0].enabled
;
1339 wmds
.StencilBufferWriteEnable
=
1340 state
->stencil
[0].writemask
!= 0 ||
1341 (two_sided_stencil
&& state
->stencil
[1].writemask
!= 0);
1342 wmds
.DepthTestEnable
= state
->depth
.enabled
;
1343 wmds
.DepthBufferWriteEnable
= state
->depth
.writemask
;
1344 wmds
.StencilTestMask
= state
->stencil
[0].valuemask
;
1345 wmds
.StencilWriteMask
= state
->stencil
[0].writemask
;
1346 wmds
.BackfaceStencilTestMask
= state
->stencil
[1].valuemask
;
1347 wmds
.BackfaceStencilWriteMask
= state
->stencil
[1].writemask
;
1348 /* wmds.[Backface]StencilReferenceValue are merged later */
1352 iris_pack_command(GENX(3DSTATE_DEPTH_BOUNDS
), cso
->depth_bounds
, depth_bounds
) {
1353 depth_bounds
.DepthBoundsTestValueModifyDisable
= false;
1354 depth_bounds
.DepthBoundsTestEnableModifyDisable
= false;
1355 depth_bounds
.DepthBoundsTestEnable
= state
->depth
.bounds_test
;
1356 depth_bounds
.DepthBoundsTestMinValue
= state
->depth
.bounds_min
;
1357 depth_bounds
.DepthBoundsTestMaxValue
= state
->depth
.bounds_max
;
1365 * The pipe->bind_depth_stencil_alpha_state() driver hook.
1367 * Bind a depth/stencil/alpha CSO and flag related dirty bits.
1370 iris_bind_zsa_state(struct pipe_context
*ctx
, void *state
)
1372 struct iris_context
*ice
= (struct iris_context
*) ctx
;
1373 struct iris_depth_stencil_alpha_state
*old_cso
= ice
->state
.cso_zsa
;
1374 struct iris_depth_stencil_alpha_state
*new_cso
= state
;
1377 if (cso_changed(alpha
.ref_value
))
1378 ice
->state
.dirty
|= IRIS_DIRTY_COLOR_CALC_STATE
;
1380 if (cso_changed(alpha
.enabled
))
1381 ice
->state
.dirty
|= IRIS_DIRTY_PS_BLEND
| IRIS_DIRTY_BLEND_STATE
;
1383 if (cso_changed(alpha
.func
))
1384 ice
->state
.dirty
|= IRIS_DIRTY_BLEND_STATE
;
1386 if (cso_changed(depth_writes_enabled
))
1387 ice
->state
.dirty
|= IRIS_DIRTY_RENDER_RESOLVES_AND_FLUSHES
;
1389 ice
->state
.depth_writes_enabled
= new_cso
->depth_writes_enabled
;
1390 ice
->state
.stencil_writes_enabled
= new_cso
->stencil_writes_enabled
;
1393 if (cso_changed(depth_bounds
))
1394 ice
->state
.dirty
|= IRIS_DIRTY_DEPTH_BOUNDS
;
1398 ice
->state
.cso_zsa
= new_cso
;
1399 ice
->state
.dirty
|= IRIS_DIRTY_CC_VIEWPORT
;
1400 ice
->state
.dirty
|= IRIS_DIRTY_WM_DEPTH_STENCIL
;
1401 ice
->state
.dirty
|= ice
->state
.dirty_for_nos
[IRIS_NOS_DEPTH_STENCIL_ALPHA
];
1404 ice
->state
.dirty
|= IRIS_DIRTY_PMA_FIX
;
1409 want_pma_fix(struct iris_context
*ice
)
1411 UNUSED
struct iris_screen
*screen
= (void *) ice
->ctx
.screen
;
1412 UNUSED
const struct gen_device_info
*devinfo
= &screen
->devinfo
;
1413 const struct brw_wm_prog_data
*wm_prog_data
= (void *)
1414 ice
->shaders
.prog
[MESA_SHADER_FRAGMENT
]->prog_data
;
1415 const struct pipe_framebuffer_state
*cso_fb
= &ice
->state
.framebuffer
;
1416 const struct iris_depth_stencil_alpha_state
*cso_zsa
= ice
->state
.cso_zsa
;
1417 const struct iris_blend_state
*cso_blend
= ice
->state
.cso_blend
;
1419 /* In very specific combinations of state, we can instruct Gen8-9 hardware
1420 * to avoid stalling at the pixel mask array. The state equations are
1421 * documented in these places:
1423 * - Gen8 Depth PMA Fix: CACHE_MODE_1::NP_PMA_FIX_ENABLE
1424 * - Gen9 Stencil PMA Fix: CACHE_MODE_0::STC PMA Optimization Enable
1426 * Both equations share some common elements:
1429 * !(3DSTATE_WM_HZ_OP::DepthBufferClear ||
1430 * 3DSTATE_WM_HZ_OP::DepthBufferResolve ||
1431 * 3DSTATE_WM_HZ_OP::Hierarchical Depth Buffer Resolve Enable ||
1432 * 3DSTATE_WM_HZ_OP::StencilBufferClear) &&
1435 * 3DSTATE_WM::ForceKillPix != ForceOff &&
1436 * (3DSTATE_PS_EXTRA::PixelShaderKillsPixels ||
1437 * 3DSTATE_PS_EXTRA::oMask Present to RenderTarget ||
1438 * 3DSTATE_PS_BLEND::AlphaToCoverageEnable ||
1439 * 3DSTATE_PS_BLEND::AlphaTestEnable ||
1440 * 3DSTATE_WM_CHROMAKEY::ChromaKeyKillEnable)
1442 * (Technically the stencil PMA treats ForceKillPix differently,
1443 * but I think this is a documentation oversight, and we don't
1444 * ever use it in this way, so it doesn't matter).
1447 * 3DSTATE_WM::ForceThreadDispatch != 1 &&
1448 * 3DSTATE_RASTER::ForceSampleCount == NUMRASTSAMPLES_0 &&
1449 * 3DSTATE_DEPTH_BUFFER::SURFACE_TYPE != NULL &&
1450 * 3DSTATE_DEPTH_BUFFER::HIZ Enable &&
1451 * 3DSTATE_WM::EDSC_Mode != EDSC_PREPS &&
1452 * 3DSTATE_PS_EXTRA::PixelShaderValid &&
1455 * These are always true:
1457 * 3DSTATE_RASTER::ForceSampleCount == NUMRASTSAMPLES_0
1458 * 3DSTATE_PS_EXTRA::PixelShaderValid
1460 * Also, we never use the normal drawing path for HiZ ops; these are true:
1462 * !(3DSTATE_WM_HZ_OP::DepthBufferClear ||
1463 * 3DSTATE_WM_HZ_OP::DepthBufferResolve ||
1464 * 3DSTATE_WM_HZ_OP::Hierarchical Depth Buffer Resolve Enable ||
1465 * 3DSTATE_WM_HZ_OP::StencilBufferClear)
1467 * This happens sometimes:
1469 * 3DSTATE_WM::ForceThreadDispatch != 1
1471 * However, we choose to ignore it as it either agrees with the signal
1472 * (dispatch was already enabled, so nothing out of the ordinary), or
1473 * there are no framebuffer attachments (so no depth or HiZ anyway,
1474 * meaning the PMA signal will already be disabled).
1480 struct iris_resource
*zres
, *sres
;
1481 iris_get_depth_stencil_resources(cso_fb
->zsbuf
->texture
, &zres
, &sres
);
1483 /* 3DSTATE_DEPTH_BUFFER::SURFACE_TYPE != NULL &&
1484 * 3DSTATE_DEPTH_BUFFER::HIZ Enable &&
1486 if (!zres
|| !iris_resource_level_has_hiz(zres
, cso_fb
->zsbuf
->u
.tex
.level
))
1489 /* 3DSTATE_WM::EDSC_Mode != EDSC_PREPS */
1490 if (wm_prog_data
->early_fragment_tests
)
1493 /* 3DSTATE_WM::ForceKillPix != ForceOff &&
1494 * (3DSTATE_PS_EXTRA::PixelShaderKillsPixels ||
1495 * 3DSTATE_PS_EXTRA::oMask Present to RenderTarget ||
1496 * 3DSTATE_PS_BLEND::AlphaToCoverageEnable ||
1497 * 3DSTATE_PS_BLEND::AlphaTestEnable ||
1498 * 3DSTATE_WM_CHROMAKEY::ChromaKeyKillEnable)
1500 bool killpixels
= wm_prog_data
->uses_kill
|| wm_prog_data
->uses_omask
||
1501 cso_blend
->alpha_to_coverage
|| cso_zsa
->alpha
.enabled
;
1503 /* The Gen8 depth PMA equation becomes:
1506 * 3DSTATE_WM_DEPTH_STENCIL::DepthWriteEnable &&
1507 * 3DSTATE_DEPTH_BUFFER::DEPTH_WRITE_ENABLE
1510 * 3DSTATE_WM_DEPTH_STENCIL::Stencil Buffer Write Enable &&
1511 * 3DSTATE_DEPTH_BUFFER::STENCIL_WRITE_ENABLE &&
1512 * 3DSTATE_STENCIL_BUFFER::STENCIL_BUFFER_ENABLE
1516 * 3DSTATE_WM_DEPTH_STENCIL::DepthTestEnable &&
1517 * ((killpixels && (depth_writes || stencil_writes)) ||
1518 * 3DSTATE_PS_EXTRA::PixelShaderComputedDepthMode != PSCDEPTH_OFF)
1521 if (!cso_zsa
->depth_test_enabled
)
1524 return wm_prog_data
->computed_depth_mode
!= PSCDEPTH_OFF
||
1525 (killpixels
&& (cso_zsa
->depth_writes_enabled
||
1526 (sres
&& cso_zsa
->stencil_writes_enabled
)));
1531 genX(update_pma_fix
)(struct iris_context
*ice
,
1532 struct iris_batch
*batch
,
1536 struct iris_genx_state
*genx
= ice
->state
.genx
;
1538 if (genx
->pma_fix_enabled
== enable
)
1541 genx
->pma_fix_enabled
= enable
;
1543 /* According to the Broadwell PIPE_CONTROL documentation, software should
1544 * emit a PIPE_CONTROL with the CS Stall and Depth Cache Flush bits set
1545 * prior to the LRI. If stencil buffer writes are enabled, then a Render * Cache Flush is also necessary.
1547 * The Gen9 docs say to use a depth stall rather than a command streamer
1548 * stall. However, the hardware seems to violently disagree. A full
1549 * command streamer stall seems to be needed in both cases.
1551 iris_emit_pipe_control_flush(batch
, "PMA fix change (1/2)",
1552 PIPE_CONTROL_CS_STALL
|
1553 PIPE_CONTROL_DEPTH_CACHE_FLUSH
|
1554 PIPE_CONTROL_RENDER_TARGET_FLUSH
);
1557 iris_pack_state(GENX(CACHE_MODE_1
), ®_val
, reg
) {
1558 reg
.NPPMAFixEnable
= enable
;
1559 reg
.NPEarlyZFailsDisable
= enable
;
1560 reg
.NPPMAFixEnableMask
= true;
1561 reg
.NPEarlyZFailsDisableMask
= true;
1563 iris_emit_lri(batch
, CACHE_MODE_1
, reg_val
);
1565 /* After the LRI, a PIPE_CONTROL with both the Depth Stall and Depth Cache
1566 * Flush bits is often necessary. We do it regardless because it's easier.
1567 * The render cache flush is also necessary if stencil writes are enabled.
1569 * Again, the Gen9 docs give a different set of flushes but the Broadwell
1570 * flushes seem to work just as well.
1572 iris_emit_pipe_control_flush(batch
, "PMA fix change (1/2)",
1573 PIPE_CONTROL_DEPTH_STALL
|
1574 PIPE_CONTROL_DEPTH_CACHE_FLUSH
|
1575 PIPE_CONTROL_RENDER_TARGET_FLUSH
);
1580 * Gallium CSO for rasterizer state.
1582 struct iris_rasterizer_state
{
1583 uint32_t sf
[GENX(3DSTATE_SF_length
)];
1584 uint32_t clip
[GENX(3DSTATE_CLIP_length
)];
1585 uint32_t raster
[GENX(3DSTATE_RASTER_length
)];
1586 uint32_t wm
[GENX(3DSTATE_WM_length
)];
1587 uint32_t line_stipple
[GENX(3DSTATE_LINE_STIPPLE_length
)];
1589 uint8_t num_clip_plane_consts
;
1590 bool clip_halfz
; /* for CC_VIEWPORT */
1591 bool depth_clip_near
; /* for CC_VIEWPORT */
1592 bool depth_clip_far
; /* for CC_VIEWPORT */
1593 bool flatshade
; /* for shader state */
1594 bool flatshade_first
; /* for stream output */
1595 bool clamp_fragment_color
; /* for shader state */
1596 bool light_twoside
; /* for shader state */
1597 bool rasterizer_discard
; /* for 3DSTATE_STREAMOUT and 3DSTATE_CLIP */
1598 bool half_pixel_center
; /* for 3DSTATE_MULTISAMPLE */
1599 bool line_stipple_enable
;
1600 bool poly_stipple_enable
;
1602 bool force_persample_interp
;
1603 bool conservative_rasterization
;
1604 bool fill_mode_point_or_line
;
1605 enum pipe_sprite_coord_mode sprite_coord_mode
; /* PIPE_SPRITE_* */
1606 uint16_t sprite_coord_enable
;
1610 get_line_width(const struct pipe_rasterizer_state
*state
)
1612 float line_width
= state
->line_width
;
1614 /* From the OpenGL 4.4 spec:
1616 * "The actual width of non-antialiased lines is determined by rounding
1617 * the supplied width to the nearest integer, then clamping it to the
1618 * implementation-dependent maximum non-antialiased line width."
1620 if (!state
->multisample
&& !state
->line_smooth
)
1621 line_width
= roundf(state
->line_width
);
1623 if (!state
->multisample
&& state
->line_smooth
&& line_width
< 1.5f
) {
1624 /* For 1 pixel line thickness or less, the general anti-aliasing
1625 * algorithm gives up, and a garbage line is generated. Setting a
1626 * Line Width of 0.0 specifies the rasterization of the "thinnest"
1627 * (one-pixel-wide), non-antialiased lines.
1629 * Lines rendered with zero Line Width are rasterized using the
1630 * "Grid Intersection Quantization" rules as specified by the
1631 * "Zero-Width (Cosmetic) Line Rasterization" section of the docs.
1640 * The pipe->create_rasterizer_state() driver hook.
1643 iris_create_rasterizer_state(struct pipe_context
*ctx
,
1644 const struct pipe_rasterizer_state
*state
)
1646 struct iris_rasterizer_state
*cso
=
1647 malloc(sizeof(struct iris_rasterizer_state
));
1649 cso
->multisample
= state
->multisample
;
1650 cso
->force_persample_interp
= state
->force_persample_interp
;
1651 cso
->clip_halfz
= state
->clip_halfz
;
1652 cso
->depth_clip_near
= state
->depth_clip_near
;
1653 cso
->depth_clip_far
= state
->depth_clip_far
;
1654 cso
->flatshade
= state
->flatshade
;
1655 cso
->flatshade_first
= state
->flatshade_first
;
1656 cso
->clamp_fragment_color
= state
->clamp_fragment_color
;
1657 cso
->light_twoside
= state
->light_twoside
;
1658 cso
->rasterizer_discard
= state
->rasterizer_discard
;
1659 cso
->half_pixel_center
= state
->half_pixel_center
;
1660 cso
->sprite_coord_mode
= state
->sprite_coord_mode
;
1661 cso
->sprite_coord_enable
= state
->sprite_coord_enable
;
1662 cso
->line_stipple_enable
= state
->line_stipple_enable
;
1663 cso
->poly_stipple_enable
= state
->poly_stipple_enable
;
1664 cso
->conservative_rasterization
=
1665 state
->conservative_raster_mode
== PIPE_CONSERVATIVE_RASTER_POST_SNAP
;
1667 cso
->fill_mode_point_or_line
=
1668 state
->fill_front
== PIPE_POLYGON_MODE_LINE
||
1669 state
->fill_front
== PIPE_POLYGON_MODE_POINT
||
1670 state
->fill_back
== PIPE_POLYGON_MODE_LINE
||
1671 state
->fill_back
== PIPE_POLYGON_MODE_POINT
;
1673 if (state
->clip_plane_enable
!= 0)
1674 cso
->num_clip_plane_consts
= util_logbase2(state
->clip_plane_enable
) + 1;
1676 cso
->num_clip_plane_consts
= 0;
1678 float line_width
= get_line_width(state
);
1680 iris_pack_command(GENX(3DSTATE_SF
), cso
->sf
, sf
) {
1681 sf
.StatisticsEnable
= true;
1682 sf
.AALineDistanceMode
= AALINEDISTANCE_TRUE
;
1683 sf
.LineEndCapAntialiasingRegionWidth
=
1684 state
->line_smooth
? _10pixels
: _05pixels
;
1685 sf
.LastPixelEnable
= state
->line_last_pixel
;
1686 sf
.LineWidth
= line_width
;
1687 sf
.SmoothPointEnable
= (state
->point_smooth
|| state
->multisample
) &&
1688 !state
->point_quad_rasterization
;
1689 sf
.PointWidthSource
= state
->point_size_per_vertex
? Vertex
: State
;
1690 sf
.PointWidth
= state
->point_size
;
1692 if (state
->flatshade_first
) {
1693 sf
.TriangleFanProvokingVertexSelect
= 1;
1695 sf
.TriangleStripListProvokingVertexSelect
= 2;
1696 sf
.TriangleFanProvokingVertexSelect
= 2;
1697 sf
.LineStripListProvokingVertexSelect
= 1;
1701 iris_pack_command(GENX(3DSTATE_RASTER
), cso
->raster
, rr
) {
1702 rr
.FrontWinding
= state
->front_ccw
? CounterClockwise
: Clockwise
;
1703 rr
.CullMode
= translate_cull_mode(state
->cull_face
);
1704 rr
.FrontFaceFillMode
= translate_fill_mode(state
->fill_front
);
1705 rr
.BackFaceFillMode
= translate_fill_mode(state
->fill_back
);
1706 rr
.DXMultisampleRasterizationEnable
= state
->multisample
;
1707 rr
.GlobalDepthOffsetEnableSolid
= state
->offset_tri
;
1708 rr
.GlobalDepthOffsetEnableWireframe
= state
->offset_line
;
1709 rr
.GlobalDepthOffsetEnablePoint
= state
->offset_point
;
1710 rr
.GlobalDepthOffsetConstant
= state
->offset_units
* 2;
1711 rr
.GlobalDepthOffsetScale
= state
->offset_scale
;
1712 rr
.GlobalDepthOffsetClamp
= state
->offset_clamp
;
1713 rr
.SmoothPointEnable
= state
->point_smooth
;
1714 rr
.AntialiasingEnable
= state
->line_smooth
;
1715 rr
.ScissorRectangleEnable
= state
->scissor
;
1717 rr
.ViewportZNearClipTestEnable
= state
->depth_clip_near
;
1718 rr
.ViewportZFarClipTestEnable
= state
->depth_clip_far
;
1719 rr
.ConservativeRasterizationEnable
=
1720 cso
->conservative_rasterization
;
1722 rr
.ViewportZClipTestEnable
= (state
->depth_clip_near
|| state
->depth_clip_far
);
1726 iris_pack_command(GENX(3DSTATE_CLIP
), cso
->clip
, cl
) {
1727 /* cl.NonPerspectiveBarycentricEnable is filled in at draw time from
1728 * the FS program; cl.ForceZeroRTAIndexEnable is filled in from the FB.
1730 cl
.EarlyCullEnable
= true;
1731 cl
.UserClipDistanceClipTestEnableBitmask
= state
->clip_plane_enable
;
1732 cl
.ForceUserClipDistanceClipTestEnableBitmask
= true;
1733 cl
.APIMode
= state
->clip_halfz
? APIMODE_D3D
: APIMODE_OGL
;
1734 cl
.GuardbandClipTestEnable
= true;
1735 cl
.ClipEnable
= true;
1736 cl
.MinimumPointWidth
= 0.125;
1737 cl
.MaximumPointWidth
= 255.875;
1739 if (state
->flatshade_first
) {
1740 cl
.TriangleFanProvokingVertexSelect
= 1;
1742 cl
.TriangleStripListProvokingVertexSelect
= 2;
1743 cl
.TriangleFanProvokingVertexSelect
= 2;
1744 cl
.LineStripListProvokingVertexSelect
= 1;
1748 iris_pack_command(GENX(3DSTATE_WM
), cso
->wm
, wm
) {
1749 /* wm.BarycentricInterpolationMode and wm.EarlyDepthStencilControl are
1750 * filled in at draw time from the FS program.
1752 wm
.LineAntialiasingRegionWidth
= _10pixels
;
1753 wm
.LineEndCapAntialiasingRegionWidth
= _05pixels
;
1754 wm
.PointRasterizationRule
= RASTRULE_UPPER_RIGHT
;
1755 wm
.LineStippleEnable
= state
->line_stipple_enable
;
1756 wm
.PolygonStippleEnable
= state
->poly_stipple_enable
;
1759 /* Remap from 0..255 back to 1..256 */
1760 const unsigned line_stipple_factor
= state
->line_stipple_factor
+ 1;
1762 iris_pack_command(GENX(3DSTATE_LINE_STIPPLE
), cso
->line_stipple
, line
) {
1763 if (state
->line_stipple_enable
) {
1764 line
.LineStipplePattern
= state
->line_stipple_pattern
;
1765 line
.LineStippleInverseRepeatCount
= 1.0f
/ line_stipple_factor
;
1766 line
.LineStippleRepeatCount
= line_stipple_factor
;
1774 * The pipe->bind_rasterizer_state() driver hook.
1776 * Bind a rasterizer CSO and flag related dirty bits.
1779 iris_bind_rasterizer_state(struct pipe_context
*ctx
, void *state
)
1781 struct iris_context
*ice
= (struct iris_context
*) ctx
;
1782 struct iris_rasterizer_state
*old_cso
= ice
->state
.cso_rast
;
1783 struct iris_rasterizer_state
*new_cso
= state
;
1786 /* Try to avoid re-emitting 3DSTATE_LINE_STIPPLE, it's non-pipelined */
1787 if (cso_changed_memcmp(line_stipple
))
1788 ice
->state
.dirty
|= IRIS_DIRTY_LINE_STIPPLE
;
1790 if (cso_changed(half_pixel_center
))
1791 ice
->state
.dirty
|= IRIS_DIRTY_MULTISAMPLE
;
1793 if (cso_changed(line_stipple_enable
) || cso_changed(poly_stipple_enable
))
1794 ice
->state
.dirty
|= IRIS_DIRTY_WM
;
1796 if (cso_changed(rasterizer_discard
))
1797 ice
->state
.dirty
|= IRIS_DIRTY_STREAMOUT
| IRIS_DIRTY_CLIP
;
1799 if (cso_changed(flatshade_first
))
1800 ice
->state
.dirty
|= IRIS_DIRTY_STREAMOUT
;
1802 if (cso_changed(depth_clip_near
) || cso_changed(depth_clip_far
) ||
1803 cso_changed(clip_halfz
))
1804 ice
->state
.dirty
|= IRIS_DIRTY_CC_VIEWPORT
;
1806 if (cso_changed(sprite_coord_enable
) ||
1807 cso_changed(sprite_coord_mode
) ||
1808 cso_changed(light_twoside
))
1809 ice
->state
.dirty
|= IRIS_DIRTY_SBE
;
1811 if (cso_changed(conservative_rasterization
))
1812 ice
->state
.dirty
|= IRIS_DIRTY_FS
;
1815 ice
->state
.cso_rast
= new_cso
;
1816 ice
->state
.dirty
|= IRIS_DIRTY_RASTER
;
1817 ice
->state
.dirty
|= IRIS_DIRTY_CLIP
;
1818 ice
->state
.dirty
|= ice
->state
.dirty_for_nos
[IRIS_NOS_RASTERIZER
];
1822 * Return true if the given wrap mode requires the border color to exist.
1824 * (We can skip uploading it if the sampler isn't going to use it.)
1827 wrap_mode_needs_border_color(unsigned wrap_mode
)
1829 return wrap_mode
== TCM_CLAMP_BORDER
|| wrap_mode
== TCM_HALF_BORDER
;
1833 * Gallium CSO for sampler state.
1835 struct iris_sampler_state
{
1836 union pipe_color_union border_color
;
1837 bool needs_border_color
;
1839 uint32_t sampler_state
[GENX(SAMPLER_STATE_length
)];
1843 * The pipe->create_sampler_state() driver hook.
1845 * We fill out SAMPLER_STATE (except for the border color pointer), and
1846 * store that on the CPU. It doesn't make sense to upload it to a GPU
1847 * buffer object yet, because 3DSTATE_SAMPLER_STATE_POINTERS requires
1848 * all bound sampler states to be in contiguous memor.
1851 iris_create_sampler_state(struct pipe_context
*ctx
,
1852 const struct pipe_sampler_state
*state
)
1854 struct iris_sampler_state
*cso
= CALLOC_STRUCT(iris_sampler_state
);
1859 STATIC_ASSERT(PIPE_TEX_FILTER_NEAREST
== MAPFILTER_NEAREST
);
1860 STATIC_ASSERT(PIPE_TEX_FILTER_LINEAR
== MAPFILTER_LINEAR
);
1862 unsigned wrap_s
= translate_wrap(state
->wrap_s
);
1863 unsigned wrap_t
= translate_wrap(state
->wrap_t
);
1864 unsigned wrap_r
= translate_wrap(state
->wrap_r
);
1866 memcpy(&cso
->border_color
, &state
->border_color
, sizeof(cso
->border_color
));
1868 cso
->needs_border_color
= wrap_mode_needs_border_color(wrap_s
) ||
1869 wrap_mode_needs_border_color(wrap_t
) ||
1870 wrap_mode_needs_border_color(wrap_r
);
1872 float min_lod
= state
->min_lod
;
1873 unsigned mag_img_filter
= state
->mag_img_filter
;
1875 // XXX: explain this code ported from ilo...I don't get it at all...
1876 if (state
->min_mip_filter
== PIPE_TEX_MIPFILTER_NONE
&&
1877 state
->min_lod
> 0.0f
) {
1879 mag_img_filter
= state
->min_img_filter
;
1882 iris_pack_state(GENX(SAMPLER_STATE
), cso
->sampler_state
, samp
) {
1883 samp
.TCXAddressControlMode
= wrap_s
;
1884 samp
.TCYAddressControlMode
= wrap_t
;
1885 samp
.TCZAddressControlMode
= wrap_r
;
1886 samp
.CubeSurfaceControlMode
= state
->seamless_cube_map
;
1887 samp
.NonnormalizedCoordinateEnable
= !state
->normalized_coords
;
1888 samp
.MinModeFilter
= state
->min_img_filter
;
1889 samp
.MagModeFilter
= mag_img_filter
;
1890 samp
.MipModeFilter
= translate_mip_filter(state
->min_mip_filter
);
1891 samp
.MaximumAnisotropy
= RATIO21
;
1893 if (state
->max_anisotropy
>= 2) {
1894 if (state
->min_img_filter
== PIPE_TEX_FILTER_LINEAR
) {
1895 samp
.MinModeFilter
= MAPFILTER_ANISOTROPIC
;
1896 samp
.AnisotropicAlgorithm
= EWAApproximation
;
1899 if (state
->mag_img_filter
== PIPE_TEX_FILTER_LINEAR
)
1900 samp
.MagModeFilter
= MAPFILTER_ANISOTROPIC
;
1902 samp
.MaximumAnisotropy
=
1903 MIN2((state
->max_anisotropy
- 2) / 2, RATIO161
);
1906 /* Set address rounding bits if not using nearest filtering. */
1907 if (state
->min_img_filter
!= PIPE_TEX_FILTER_NEAREST
) {
1908 samp
.UAddressMinFilterRoundingEnable
= true;
1909 samp
.VAddressMinFilterRoundingEnable
= true;
1910 samp
.RAddressMinFilterRoundingEnable
= true;
1913 if (state
->mag_img_filter
!= PIPE_TEX_FILTER_NEAREST
) {
1914 samp
.UAddressMagFilterRoundingEnable
= true;
1915 samp
.VAddressMagFilterRoundingEnable
= true;
1916 samp
.RAddressMagFilterRoundingEnable
= true;
1919 if (state
->compare_mode
== PIPE_TEX_COMPARE_R_TO_TEXTURE
)
1920 samp
.ShadowFunction
= translate_shadow_func(state
->compare_func
);
1922 const float hw_max_lod
= GEN_GEN
>= 7 ? 14 : 13;
1924 samp
.LODPreClampMode
= CLAMP_MODE_OGL
;
1925 samp
.MinLOD
= CLAMP(min_lod
, 0, hw_max_lod
);
1926 samp
.MaxLOD
= CLAMP(state
->max_lod
, 0, hw_max_lod
);
1927 samp
.TextureLODBias
= CLAMP(state
->lod_bias
, -16, 15);
1929 /* .BorderColorPointer is filled in by iris_bind_sampler_states. */
1936 * The pipe->bind_sampler_states() driver hook.
1939 iris_bind_sampler_states(struct pipe_context
*ctx
,
1940 enum pipe_shader_type p_stage
,
1941 unsigned start
, unsigned count
,
1944 struct iris_context
*ice
= (struct iris_context
*) ctx
;
1945 gl_shader_stage stage
= stage_from_pipe(p_stage
);
1946 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
1948 assert(start
+ count
<= IRIS_MAX_TEXTURE_SAMPLERS
);
1952 for (int i
= 0; i
< count
; i
++) {
1953 if (shs
->samplers
[start
+ i
] != states
[i
]) {
1954 shs
->samplers
[start
+ i
] = states
[i
];
1960 ice
->state
.dirty
|= IRIS_DIRTY_SAMPLER_STATES_VS
<< stage
;
1964 * Upload the sampler states into a contiguous area of GPU memory, for
1965 * for 3DSTATE_SAMPLER_STATE_POINTERS_*.
1967 * Also fill out the border color state pointers.
1970 iris_upload_sampler_states(struct iris_context
*ice
, gl_shader_stage stage
)
1972 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
1973 const struct shader_info
*info
= iris_get_shader_info(ice
, stage
);
1975 /* We assume the state tracker will call pipe->bind_sampler_states()
1976 * if the program's number of textures changes.
1978 unsigned count
= info
? util_last_bit(info
->textures_used
) : 0;
1983 /* Assemble the SAMPLER_STATEs into a contiguous table that lives
1984 * in the dynamic state memory zone, so we can point to it via the
1985 * 3DSTATE_SAMPLER_STATE_POINTERS_* commands.
1987 unsigned size
= count
* 4 * GENX(SAMPLER_STATE_length
);
1989 upload_state(ice
->state
.dynamic_uploader
, &shs
->sampler_table
, size
, 32);
1993 struct pipe_resource
*res
= shs
->sampler_table
.res
;
1994 shs
->sampler_table
.offset
+=
1995 iris_bo_offset_from_base_address(iris_resource_bo(res
));
1997 iris_record_state_size(ice
->state
.sizes
, shs
->sampler_table
.offset
, size
);
1999 /* Make sure all land in the same BO */
2000 iris_border_color_pool_reserve(ice
, IRIS_MAX_TEXTURE_SAMPLERS
);
2002 ice
->state
.need_border_colors
&= ~(1 << stage
);
2004 for (int i
= 0; i
< count
; i
++) {
2005 struct iris_sampler_state
*state
= shs
->samplers
[i
];
2006 struct iris_sampler_view
*tex
= shs
->textures
[i
];
2009 memset(map
, 0, 4 * GENX(SAMPLER_STATE_length
));
2010 } else if (!state
->needs_border_color
) {
2011 memcpy(map
, state
->sampler_state
, 4 * GENX(SAMPLER_STATE_length
));
2013 ice
->state
.need_border_colors
|= 1 << stage
;
2015 /* We may need to swizzle the border color for format faking.
2016 * A/LA formats are faked as R/RG with 000R or R00G swizzles.
2017 * This means we need to move the border color's A channel into
2018 * the R or G channels so that those read swizzles will move it
2021 union pipe_color_union
*color
= &state
->border_color
;
2022 union pipe_color_union tmp
;
2024 enum pipe_format internal_format
= tex
->res
->internal_format
;
2026 if (util_format_is_alpha(internal_format
)) {
2027 unsigned char swz
[4] = {
2028 PIPE_SWIZZLE_W
, PIPE_SWIZZLE_0
,
2029 PIPE_SWIZZLE_0
, PIPE_SWIZZLE_0
2031 util_format_apply_color_swizzle(&tmp
, color
, swz
, true);
2033 } else if (util_format_is_luminance_alpha(internal_format
) &&
2034 internal_format
!= PIPE_FORMAT_L8A8_SRGB
) {
2035 unsigned char swz
[4] = {
2036 PIPE_SWIZZLE_X
, PIPE_SWIZZLE_W
,
2037 PIPE_SWIZZLE_0
, PIPE_SWIZZLE_0
2039 util_format_apply_color_swizzle(&tmp
, color
, swz
, true);
2044 /* Stream out the border color and merge the pointer. */
2045 uint32_t offset
= iris_upload_border_color(ice
, color
);
2047 uint32_t dynamic
[GENX(SAMPLER_STATE_length
)];
2048 iris_pack_state(GENX(SAMPLER_STATE
), dynamic
, dyns
) {
2049 dyns
.BorderColorPointer
= offset
;
2052 for (uint32_t j
= 0; j
< GENX(SAMPLER_STATE_length
); j
++)
2053 map
[j
] = state
->sampler_state
[j
] | dynamic
[j
];
2056 map
+= GENX(SAMPLER_STATE_length
);
2060 static enum isl_channel_select
2061 fmt_swizzle(const struct iris_format_info
*fmt
, enum pipe_swizzle swz
)
2064 case PIPE_SWIZZLE_X
: return fmt
->swizzle
.r
;
2065 case PIPE_SWIZZLE_Y
: return fmt
->swizzle
.g
;
2066 case PIPE_SWIZZLE_Z
: return fmt
->swizzle
.b
;
2067 case PIPE_SWIZZLE_W
: return fmt
->swizzle
.a
;
2068 case PIPE_SWIZZLE_1
: return SCS_ONE
;
2069 case PIPE_SWIZZLE_0
: return SCS_ZERO
;
2070 default: unreachable("invalid swizzle");
2075 fill_buffer_surface_state(struct isl_device
*isl_dev
,
2076 struct iris_resource
*res
,
2078 enum isl_format format
,
2079 struct isl_swizzle swizzle
,
2083 const struct isl_format_layout
*fmtl
= isl_format_get_layout(format
);
2084 const unsigned cpp
= format
== ISL_FORMAT_RAW
? 1 : fmtl
->bpb
/ 8;
2086 /* The ARB_texture_buffer_specification says:
2088 * "The number of texels in the buffer texture's texel array is given by
2090 * floor(<buffer_size> / (<components> * sizeof(<base_type>)),
2092 * where <buffer_size> is the size of the buffer object, in basic
2093 * machine units and <components> and <base_type> are the element count
2094 * and base data type for elements, as specified in Table X.1. The
2095 * number of texels in the texel array is then clamped to the
2096 * implementation-dependent limit MAX_TEXTURE_BUFFER_SIZE_ARB."
2098 * We need to clamp the size in bytes to MAX_TEXTURE_BUFFER_SIZE * stride,
2099 * so that when ISL divides by stride to obtain the number of texels, that
2100 * texel count is clamped to MAX_TEXTURE_BUFFER_SIZE.
2102 unsigned final_size
=
2103 MIN3(size
, res
->bo
->size
- res
->offset
- offset
,
2104 IRIS_MAX_TEXTURE_BUFFER_SIZE
* cpp
);
2106 isl_buffer_fill_state(isl_dev
, map
,
2107 .address
= res
->bo
->gtt_offset
+ res
->offset
+ offset
,
2108 .size_B
= final_size
,
2112 .mocs
= mocs(res
->bo
));
2115 #define SURFACE_STATE_ALIGNMENT 64
2118 * Allocate several contiguous SURFACE_STATE structures, one for each
2119 * supported auxiliary surface mode.
2122 alloc_surface_states(struct u_upload_mgr
*mgr
,
2123 struct iris_state_ref
*ref
,
2124 unsigned aux_usages
)
2126 const unsigned surf_size
= 4 * GENX(RENDER_SURFACE_STATE_length
);
2128 /* If this changes, update this to explicitly align pointers */
2129 STATIC_ASSERT(surf_size
== SURFACE_STATE_ALIGNMENT
);
2131 assert(aux_usages
!= 0);
2134 upload_state(mgr
, ref
, util_bitcount(aux_usages
) * surf_size
,
2135 SURFACE_STATE_ALIGNMENT
);
2137 ref
->offset
+= iris_bo_offset_from_base_address(iris_resource_bo(ref
->res
));
2144 * Return an ISL surface for use with non-coherent render target reads.
2146 * In a few complex cases, we can't use the SURFACE_STATE for normal render
2147 * target writes. We need to make a separate one for sampling which refers
2148 * to the single slice of the texture being read.
2151 get_rt_read_isl_surf(const struct gen_device_info
*devinfo
,
2152 struct iris_resource
*res
,
2153 enum pipe_texture_target target
,
2154 struct isl_view
*view
,
2155 uint32_t *tile_x_sa
,
2156 uint32_t *tile_y_sa
,
2157 struct isl_surf
*surf
)
2162 const enum isl_dim_layout dim_layout
=
2163 iris_get_isl_dim_layout(devinfo
, res
->surf
.tiling
, target
);
2165 surf
->dim
= target_to_isl_surf_dim(target
);
2167 if (surf
->dim_layout
== dim_layout
)
2170 /* The layout of the specified texture target is not compatible with the
2171 * actual layout of the miptree structure in memory -- You're entering
2172 * dangerous territory, this can only possibly work if you only intended
2173 * to access a single level and slice of the texture, and the hardware
2174 * supports the tile offset feature in order to allow non-tile-aligned
2175 * base offsets, since we'll have to point the hardware to the first
2176 * texel of the level instead of relying on the usual base level/layer
2179 assert(view
->levels
== 1 && view
->array_len
== 1);
2180 assert(*tile_x_sa
== 0 && *tile_y_sa
== 0);
2182 res
->offset
+= iris_resource_get_tile_offsets(res
, view
->base_level
,
2183 view
->base_array_layer
,
2184 tile_x_sa
, tile_y_sa
);
2185 const unsigned l
= view
->base_level
;
2187 surf
->logical_level0_px
.width
= minify(surf
->logical_level0_px
.width
, l
);
2188 surf
->logical_level0_px
.height
= surf
->dim
<= ISL_SURF_DIM_1D
? 1 :
2189 minify(surf
->logical_level0_px
.height
, l
);
2190 surf
->logical_level0_px
.depth
= surf
->dim
<= ISL_SURF_DIM_2D
? 1 :
2191 minify(surf
->logical_level0_px
.depth
, l
);
2193 surf
->logical_level0_px
.array_len
= 1;
2195 surf
->dim_layout
= dim_layout
;
2197 view
->base_level
= 0;
2198 view
->base_array_layer
= 0;
2203 fill_surface_state(struct isl_device
*isl_dev
,
2205 struct iris_resource
*res
,
2206 struct isl_surf
*surf
,
2207 struct isl_view
*view
,
2212 struct isl_surf_fill_state_info f
= {
2215 .mocs
= mocs(res
->bo
),
2216 .address
= res
->bo
->gtt_offset
+ res
->offset
,
2217 .x_offset_sa
= tile_x_sa
,
2218 .y_offset_sa
= tile_y_sa
,
2221 assert(!iris_resource_unfinished_aux_import(res
));
2223 if (aux_usage
!= ISL_AUX_USAGE_NONE
) {
2224 f
.aux_surf
= &res
->aux
.surf
;
2225 f
.aux_usage
= aux_usage
;
2226 f
.aux_address
= res
->aux
.bo
->gtt_offset
+ res
->aux
.offset
;
2228 struct iris_bo
*clear_bo
= NULL
;
2229 uint64_t clear_offset
= 0;
2231 iris_resource_get_clear_color(res
, &clear_bo
, &clear_offset
);
2233 f
.clear_address
= clear_bo
->gtt_offset
+ clear_offset
;
2234 f
.use_clear_address
= isl_dev
->info
->gen
> 9;
2238 isl_surf_fill_state_s(isl_dev
, map
, &f
);
2242 * The pipe->create_sampler_view() driver hook.
2244 static struct pipe_sampler_view
*
2245 iris_create_sampler_view(struct pipe_context
*ctx
,
2246 struct pipe_resource
*tex
,
2247 const struct pipe_sampler_view
*tmpl
)
2249 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2250 struct iris_screen
*screen
= (struct iris_screen
*)ctx
->screen
;
2251 const struct gen_device_info
*devinfo
= &screen
->devinfo
;
2252 struct iris_sampler_view
*isv
= calloc(1, sizeof(struct iris_sampler_view
));
2257 /* initialize base object */
2259 isv
->base
.context
= ctx
;
2260 isv
->base
.texture
= NULL
;
2261 pipe_reference_init(&isv
->base
.reference
, 1);
2262 pipe_resource_reference(&isv
->base
.texture
, tex
);
2264 if (util_format_is_depth_or_stencil(tmpl
->format
)) {
2265 struct iris_resource
*zres
, *sres
;
2266 const struct util_format_description
*desc
=
2267 util_format_description(tmpl
->format
);
2269 iris_get_depth_stencil_resources(tex
, &zres
, &sres
);
2271 tex
= util_format_has_depth(desc
) ? &zres
->base
: &sres
->base
;
2274 isv
->res
= (struct iris_resource
*) tex
;
2276 void *map
= alloc_surface_states(ice
->state
.surface_uploader
,
2277 &isv
->surface_state
,
2278 isv
->res
->aux
.sampler_usages
);
2282 isl_surf_usage_flags_t usage
= ISL_SURF_USAGE_TEXTURE_BIT
;
2284 if (isv
->base
.target
== PIPE_TEXTURE_CUBE
||
2285 isv
->base
.target
== PIPE_TEXTURE_CUBE_ARRAY
)
2286 usage
|= ISL_SURF_USAGE_CUBE_BIT
;
2288 const struct iris_format_info fmt
=
2289 iris_format_for_usage(devinfo
, tmpl
->format
, usage
);
2291 isv
->clear_color
= isv
->res
->aux
.clear_color
;
2293 isv
->view
= (struct isl_view
) {
2295 .swizzle
= (struct isl_swizzle
) {
2296 .r
= fmt_swizzle(&fmt
, tmpl
->swizzle_r
),
2297 .g
= fmt_swizzle(&fmt
, tmpl
->swizzle_g
),
2298 .b
= fmt_swizzle(&fmt
, tmpl
->swizzle_b
),
2299 .a
= fmt_swizzle(&fmt
, tmpl
->swizzle_a
),
2304 /* Fill out SURFACE_STATE for this view. */
2305 if (tmpl
->target
!= PIPE_BUFFER
) {
2306 isv
->view
.base_level
= tmpl
->u
.tex
.first_level
;
2307 isv
->view
.levels
= tmpl
->u
.tex
.last_level
- tmpl
->u
.tex
.first_level
+ 1;
2308 // XXX: do I need to port f9fd0cf4790cb2a530e75d1a2206dbb9d8af7cb2?
2309 isv
->view
.base_array_layer
= tmpl
->u
.tex
.first_layer
;
2310 isv
->view
.array_len
=
2311 tmpl
->u
.tex
.last_layer
- tmpl
->u
.tex
.first_layer
+ 1;
2313 if (iris_resource_unfinished_aux_import(isv
->res
))
2314 iris_resource_finish_aux_import(&screen
->base
, isv
->res
);
2316 unsigned aux_modes
= isv
->res
->aux
.sampler_usages
;
2318 enum isl_aux_usage aux_usage
= u_bit_scan(&aux_modes
);
2320 /* If we have a multisampled depth buffer, do not create a sampler
2321 * surface state with HiZ.
2323 fill_surface_state(&screen
->isl_dev
, map
, isv
->res
, &isv
->res
->surf
,
2324 &isv
->view
, aux_usage
, 0, 0);
2326 map
+= SURFACE_STATE_ALIGNMENT
;
2329 fill_buffer_surface_state(&screen
->isl_dev
, isv
->res
, map
,
2330 isv
->view
.format
, isv
->view
.swizzle
,
2331 tmpl
->u
.buf
.offset
, tmpl
->u
.buf
.size
);
2338 iris_sampler_view_destroy(struct pipe_context
*ctx
,
2339 struct pipe_sampler_view
*state
)
2341 struct iris_sampler_view
*isv
= (void *) state
;
2342 pipe_resource_reference(&state
->texture
, NULL
);
2343 pipe_resource_reference(&isv
->surface_state
.res
, NULL
);
2348 * The pipe->create_surface() driver hook.
2350 * In Gallium nomenclature, "surfaces" are a view of a resource that
2351 * can be bound as a render target or depth/stencil buffer.
2353 static struct pipe_surface
*
2354 iris_create_surface(struct pipe_context
*ctx
,
2355 struct pipe_resource
*tex
,
2356 const struct pipe_surface
*tmpl
)
2358 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2359 struct iris_screen
*screen
= (struct iris_screen
*)ctx
->screen
;
2360 const struct gen_device_info
*devinfo
= &screen
->devinfo
;
2362 isl_surf_usage_flags_t usage
= 0;
2364 usage
= ISL_SURF_USAGE_STORAGE_BIT
;
2365 else if (util_format_is_depth_or_stencil(tmpl
->format
))
2366 usage
= ISL_SURF_USAGE_DEPTH_BIT
;
2368 usage
= ISL_SURF_USAGE_RENDER_TARGET_BIT
;
2370 const struct iris_format_info fmt
=
2371 iris_format_for_usage(devinfo
, tmpl
->format
, usage
);
2373 if ((usage
& ISL_SURF_USAGE_RENDER_TARGET_BIT
) &&
2374 !isl_format_supports_rendering(devinfo
, fmt
.fmt
)) {
2375 /* Framebuffer validation will reject this invalid case, but it
2376 * hasn't had the opportunity yet. In the meantime, we need to
2377 * avoid hitting ISL asserts about unsupported formats below.
2382 struct iris_surface
*surf
= calloc(1, sizeof(struct iris_surface
));
2383 struct pipe_surface
*psurf
= &surf
->base
;
2384 struct iris_resource
*res
= (struct iris_resource
*) tex
;
2389 pipe_reference_init(&psurf
->reference
, 1);
2390 pipe_resource_reference(&psurf
->texture
, tex
);
2391 psurf
->context
= ctx
;
2392 psurf
->format
= tmpl
->format
;
2393 psurf
->width
= tex
->width0
;
2394 psurf
->height
= tex
->height0
;
2395 psurf
->texture
= tex
;
2396 psurf
->u
.tex
.first_layer
= tmpl
->u
.tex
.first_layer
;
2397 psurf
->u
.tex
.last_layer
= tmpl
->u
.tex
.last_layer
;
2398 psurf
->u
.tex
.level
= tmpl
->u
.tex
.level
;
2400 uint32_t array_len
= tmpl
->u
.tex
.last_layer
- tmpl
->u
.tex
.first_layer
+ 1;
2402 struct isl_view
*view
= &surf
->view
;
2403 *view
= (struct isl_view
) {
2405 .base_level
= tmpl
->u
.tex
.level
,
2407 .base_array_layer
= tmpl
->u
.tex
.first_layer
,
2408 .array_len
= array_len
,
2409 .swizzle
= ISL_SWIZZLE_IDENTITY
,
2414 enum pipe_texture_target target
= (tex
->target
== PIPE_TEXTURE_3D
&&
2415 array_len
== 1) ? PIPE_TEXTURE_2D
:
2416 tex
->target
== PIPE_TEXTURE_1D_ARRAY
?
2417 PIPE_TEXTURE_2D_ARRAY
: tex
->target
;
2419 struct isl_view
*read_view
= &surf
->read_view
;
2420 *read_view
= (struct isl_view
) {
2422 .base_level
= tmpl
->u
.tex
.level
,
2424 .base_array_layer
= tmpl
->u
.tex
.first_layer
,
2425 .array_len
= array_len
,
2426 .swizzle
= ISL_SWIZZLE_IDENTITY
,
2427 .usage
= ISL_SURF_USAGE_TEXTURE_BIT
,
2431 surf
->clear_color
= res
->aux
.clear_color
;
2433 /* Bail early for depth/stencil - we don't want SURFACE_STATE for them. */
2434 if (res
->surf
.usage
& (ISL_SURF_USAGE_DEPTH_BIT
|
2435 ISL_SURF_USAGE_STENCIL_BIT
))
2439 void *map
= alloc_surface_states(ice
->state
.surface_uploader
,
2440 &surf
->surface_state
,
2441 res
->aux
.possible_usages
);
2442 if (!unlikely(map
)) {
2443 pipe_resource_reference(&surf
->surface_state
.res
, NULL
);
2448 void *map_read
= alloc_surface_states(ice
->state
.surface_uploader
,
2449 &surf
->surface_state_read
,
2450 res
->aux
.possible_usages
);
2451 if (!unlikely(map_read
)) {
2452 pipe_resource_reference(&surf
->surface_state_read
.res
, NULL
);
2457 if (!isl_format_is_compressed(res
->surf
.format
)) {
2458 if (iris_resource_unfinished_aux_import(res
))
2459 iris_resource_finish_aux_import(&screen
->base
, res
);
2461 /* This is a normal surface. Fill out a SURFACE_STATE for each possible
2462 * auxiliary surface mode and return the pipe_surface.
2464 unsigned aux_modes
= res
->aux
.possible_usages
;
2467 uint32_t offset
= res
->offset
;
2469 enum isl_aux_usage aux_usage
= u_bit_scan(&aux_modes
);
2470 fill_surface_state(&screen
->isl_dev
, map
, res
, &res
->surf
,
2471 view
, aux_usage
, 0, 0);
2472 map
+= SURFACE_STATE_ALIGNMENT
;
2475 struct isl_surf surf
;
2476 uint32_t tile_x_sa
= 0, tile_y_sa
= 0;
2477 get_rt_read_isl_surf(devinfo
, res
, target
, read_view
,
2478 &tile_x_sa
, &tile_y_sa
, &surf
);
2479 fill_surface_state(&screen
->isl_dev
, map_read
, res
, &surf
, read_view
,
2480 aux_usage
, tile_x_sa
, tile_y_sa
);
2481 /* Restore offset because we change offset in case of handling
2482 * non_coherent fb fetch
2484 res
->offset
= offset
;
2485 map_read
+= SURFACE_STATE_ALIGNMENT
;
2492 /* The resource has a compressed format, which is not renderable, but we
2493 * have a renderable view format. We must be attempting to upload blocks
2494 * of compressed data via an uncompressed view.
2496 * In this case, we can assume there are no auxiliary buffers, a single
2497 * miplevel, and that the resource is single-sampled. Gallium may try
2498 * and create an uncompressed view with multiple layers, however.
2500 assert(!isl_format_is_compressed(fmt
.fmt
));
2501 assert(res
->aux
.possible_usages
== 1 << ISL_AUX_USAGE_NONE
);
2502 assert(res
->surf
.samples
== 1);
2503 assert(view
->levels
== 1);
2505 struct isl_surf isl_surf
;
2506 uint32_t offset_B
= 0, tile_x_sa
= 0, tile_y_sa
= 0;
2508 if (view
->base_level
> 0) {
2509 /* We can't rely on the hardware's miplevel selection with such
2510 * a substantial lie about the format, so we select a single image
2511 * using the Tile X/Y Offset fields. In this case, we can't handle
2512 * multiple array slices.
2514 * On Broadwell, HALIGN and VALIGN are specified in pixels and are
2515 * hard-coded to align to exactly the block size of the compressed
2516 * texture. This means that, when reinterpreted as a non-compressed
2517 * texture, the tile offsets may be anything and we can't rely on
2520 * Return NULL to force the state tracker to take fallback paths.
2522 if (view
->array_len
> 1 || GEN_GEN
== 8)
2525 const bool is_3d
= res
->surf
.dim
== ISL_SURF_DIM_3D
;
2526 isl_surf_get_image_surf(&screen
->isl_dev
, &res
->surf
,
2528 is_3d
? 0 : view
->base_array_layer
,
2529 is_3d
? view
->base_array_layer
: 0,
2531 &offset_B
, &tile_x_sa
, &tile_y_sa
);
2533 /* We use address and tile offsets to access a single level/layer
2534 * as a subimage, so reset level/layer so it doesn't offset again.
2536 view
->base_array_layer
= 0;
2537 view
->base_level
= 0;
2539 /* Level 0 doesn't require tile offsets, and the hardware can find
2540 * array slices using QPitch even with the format override, so we
2541 * can allow layers in this case. Copy the original ISL surface.
2543 memcpy(&isl_surf
, &res
->surf
, sizeof(isl_surf
));
2546 /* Scale down the image dimensions by the block size. */
2547 const struct isl_format_layout
*fmtl
=
2548 isl_format_get_layout(res
->surf
.format
);
2549 isl_surf
.format
= fmt
.fmt
;
2550 isl_surf
.logical_level0_px
= isl_surf_get_logical_level0_el(&isl_surf
);
2551 isl_surf
.phys_level0_sa
= isl_surf_get_phys_level0_el(&isl_surf
);
2552 tile_x_sa
/= fmtl
->bw
;
2553 tile_y_sa
/= fmtl
->bh
;
2555 psurf
->width
= isl_surf
.logical_level0_px
.width
;
2556 psurf
->height
= isl_surf
.logical_level0_px
.height
;
2558 struct isl_surf_fill_state_info f
= {
2561 .mocs
= mocs(res
->bo
),
2562 .address
= res
->bo
->gtt_offset
+ offset_B
,
2563 .x_offset_sa
= tile_x_sa
,
2564 .y_offset_sa
= tile_y_sa
,
2567 isl_surf_fill_state_s(&screen
->isl_dev
, map
, &f
);
2573 fill_default_image_param(struct brw_image_param
*param
)
2575 memset(param
, 0, sizeof(*param
));
2576 /* Set the swizzling shifts to all-ones to effectively disable swizzling --
2577 * See emit_address_calculation() in brw_fs_surface_builder.cpp for a more
2578 * detailed explanation of these parameters.
2580 param
->swizzling
[0] = 0xff;
2581 param
->swizzling
[1] = 0xff;
2585 fill_buffer_image_param(struct brw_image_param
*param
,
2586 enum pipe_format pfmt
,
2589 const unsigned cpp
= util_format_get_blocksize(pfmt
);
2591 fill_default_image_param(param
);
2592 param
->size
[0] = size
/ cpp
;
2593 param
->stride
[0] = cpp
;
2596 #define isl_surf_fill_image_param(x, ...)
2597 #define fill_default_image_param(x, ...)
2598 #define fill_buffer_image_param(x, ...)
2602 * The pipe->set_shader_images() driver hook.
2605 iris_set_shader_images(struct pipe_context
*ctx
,
2606 enum pipe_shader_type p_stage
,
2607 unsigned start_slot
, unsigned count
,
2608 const struct pipe_image_view
*p_images
)
2610 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2611 struct iris_screen
*screen
= (struct iris_screen
*)ctx
->screen
;
2612 const struct gen_device_info
*devinfo
= &screen
->devinfo
;
2613 gl_shader_stage stage
= stage_from_pipe(p_stage
);
2614 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
2616 struct iris_genx_state
*genx
= ice
->state
.genx
;
2617 struct brw_image_param
*image_params
= genx
->shaders
[stage
].image_param
;
2620 shs
->bound_image_views
&= ~u_bit_consecutive(start_slot
, count
);
2622 for (unsigned i
= 0; i
< count
; i
++) {
2623 struct iris_image_view
*iv
= &shs
->image
[start_slot
+ i
];
2625 if (p_images
&& p_images
[i
].resource
) {
2626 const struct pipe_image_view
*img
= &p_images
[i
];
2627 struct iris_resource
*res
= (void *) img
->resource
;
2630 alloc_surface_states(ice
->state
.surface_uploader
,
2631 &iv
->surface_state
, 1 << ISL_AUX_USAGE_NONE
);
2635 util_copy_image_view(&iv
->base
, img
);
2637 shs
->bound_image_views
|= 1 << (start_slot
+ i
);
2639 res
->bind_history
|= PIPE_BIND_SHADER_IMAGE
;
2640 res
->bind_stages
|= 1 << stage
;
2642 isl_surf_usage_flags_t usage
= ISL_SURF_USAGE_STORAGE_BIT
;
2643 enum isl_format isl_fmt
=
2644 iris_format_for_usage(devinfo
, img
->format
, usage
).fmt
;
2646 bool untyped_fallback
= false;
2648 if (img
->shader_access
& PIPE_IMAGE_ACCESS_READ
) {
2649 /* On Gen8, try to use typed surfaces reads (which support a
2650 * limited number of formats), and if not possible, fall back
2653 untyped_fallback
= GEN_GEN
== 8 &&
2654 !isl_has_matching_typed_storage_image_format(devinfo
, isl_fmt
);
2656 if (untyped_fallback
)
2657 isl_fmt
= ISL_FORMAT_RAW
;
2659 isl_fmt
= isl_lower_storage_image_format(devinfo
, isl_fmt
);
2662 if (res
->base
.target
!= PIPE_BUFFER
) {
2663 struct isl_view view
= {
2665 .base_level
= img
->u
.tex
.level
,
2667 .base_array_layer
= img
->u
.tex
.first_layer
,
2668 .array_len
= img
->u
.tex
.last_layer
- img
->u
.tex
.first_layer
+ 1,
2669 .swizzle
= ISL_SWIZZLE_IDENTITY
,
2673 if (untyped_fallback
) {
2674 fill_buffer_surface_state(&screen
->isl_dev
, res
, map
,
2675 isl_fmt
, ISL_SWIZZLE_IDENTITY
,
2678 /* Images don't support compression */
2679 unsigned aux_modes
= 1 << ISL_AUX_USAGE_NONE
;
2681 enum isl_aux_usage usage
= u_bit_scan(&aux_modes
);
2683 fill_surface_state(&screen
->isl_dev
, map
, res
, &res
->surf
,
2684 &view
, usage
, 0, 0);
2686 map
+= SURFACE_STATE_ALIGNMENT
;
2690 isl_surf_fill_image_param(&screen
->isl_dev
,
2691 &image_params
[start_slot
+ i
],
2694 util_range_add(&res
->base
, &res
->valid_buffer_range
, img
->u
.buf
.offset
,
2695 img
->u
.buf
.offset
+ img
->u
.buf
.size
);
2697 fill_buffer_surface_state(&screen
->isl_dev
, res
, map
,
2698 isl_fmt
, ISL_SWIZZLE_IDENTITY
,
2699 img
->u
.buf
.offset
, img
->u
.buf
.size
);
2700 fill_buffer_image_param(&image_params
[start_slot
+ i
],
2701 img
->format
, img
->u
.buf
.size
);
2704 pipe_resource_reference(&iv
->base
.resource
, NULL
);
2705 pipe_resource_reference(&iv
->surface_state
.res
, NULL
);
2706 fill_default_image_param(&image_params
[start_slot
+ i
]);
2710 ice
->state
.dirty
|= IRIS_DIRTY_BINDINGS_VS
<< stage
;
2712 stage
== MESA_SHADER_COMPUTE
? IRIS_DIRTY_COMPUTE_RESOLVES_AND_FLUSHES
2713 : IRIS_DIRTY_RENDER_RESOLVES_AND_FLUSHES
;
2715 /* Broadwell also needs brw_image_params re-uploaded */
2717 ice
->state
.dirty
|= IRIS_DIRTY_CONSTANTS_VS
<< stage
;
2718 shs
->sysvals_need_upload
= true;
2724 * The pipe->set_sampler_views() driver hook.
2727 iris_set_sampler_views(struct pipe_context
*ctx
,
2728 enum pipe_shader_type p_stage
,
2729 unsigned start
, unsigned count
,
2730 struct pipe_sampler_view
**views
)
2732 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2733 gl_shader_stage stage
= stage_from_pipe(p_stage
);
2734 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
2736 shs
->bound_sampler_views
&= ~u_bit_consecutive(start
, count
);
2738 for (unsigned i
= 0; i
< count
; i
++) {
2739 struct pipe_sampler_view
*pview
= views
? views
[i
] : NULL
;
2740 pipe_sampler_view_reference((struct pipe_sampler_view
**)
2741 &shs
->textures
[start
+ i
], pview
);
2742 struct iris_sampler_view
*view
= (void *) pview
;
2744 view
->res
->bind_history
|= PIPE_BIND_SAMPLER_VIEW
;
2745 view
->res
->bind_stages
|= 1 << stage
;
2747 shs
->bound_sampler_views
|= 1 << (start
+ i
);
2751 ice
->state
.dirty
|= (IRIS_DIRTY_BINDINGS_VS
<< stage
);
2753 stage
== MESA_SHADER_COMPUTE
? IRIS_DIRTY_COMPUTE_RESOLVES_AND_FLUSHES
2754 : IRIS_DIRTY_RENDER_RESOLVES_AND_FLUSHES
;
2758 * The pipe->set_tess_state() driver hook.
2761 iris_set_tess_state(struct pipe_context
*ctx
,
2762 const float default_outer_level
[4],
2763 const float default_inner_level
[2])
2765 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2766 struct iris_shader_state
*shs
= &ice
->state
.shaders
[MESA_SHADER_TESS_CTRL
];
2768 memcpy(&ice
->state
.default_outer_level
[0], &default_outer_level
[0], 4 * sizeof(float));
2769 memcpy(&ice
->state
.default_inner_level
[0], &default_inner_level
[0], 2 * sizeof(float));
2771 ice
->state
.dirty
|= IRIS_DIRTY_CONSTANTS_TCS
;
2772 shs
->sysvals_need_upload
= true;
2776 iris_surface_destroy(struct pipe_context
*ctx
, struct pipe_surface
*p_surf
)
2778 struct iris_surface
*surf
= (void *) p_surf
;
2779 pipe_resource_reference(&p_surf
->texture
, NULL
);
2780 pipe_resource_reference(&surf
->surface_state
.res
, NULL
);
2781 pipe_resource_reference(&surf
->surface_state_read
.res
, NULL
);
2786 iris_set_clip_state(struct pipe_context
*ctx
,
2787 const struct pipe_clip_state
*state
)
2789 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2790 struct iris_shader_state
*shs
= &ice
->state
.shaders
[MESA_SHADER_VERTEX
];
2791 struct iris_shader_state
*gshs
= &ice
->state
.shaders
[MESA_SHADER_GEOMETRY
];
2792 struct iris_shader_state
*tshs
= &ice
->state
.shaders
[MESA_SHADER_TESS_EVAL
];
2794 memcpy(&ice
->state
.clip_planes
, state
, sizeof(*state
));
2796 ice
->state
.dirty
|= IRIS_DIRTY_CONSTANTS_VS
| IRIS_DIRTY_CONSTANTS_GS
|
2797 IRIS_DIRTY_CONSTANTS_TES
;
2798 shs
->sysvals_need_upload
= true;
2799 gshs
->sysvals_need_upload
= true;
2800 tshs
->sysvals_need_upload
= true;
2804 * The pipe->set_polygon_stipple() driver hook.
2807 iris_set_polygon_stipple(struct pipe_context
*ctx
,
2808 const struct pipe_poly_stipple
*state
)
2810 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2811 memcpy(&ice
->state
.poly_stipple
, state
, sizeof(*state
));
2812 ice
->state
.dirty
|= IRIS_DIRTY_POLYGON_STIPPLE
;
2816 * The pipe->set_sample_mask() driver hook.
2819 iris_set_sample_mask(struct pipe_context
*ctx
, unsigned sample_mask
)
2821 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2823 /* We only support 16x MSAA, so we have 16 bits of sample maks.
2824 * st/mesa may pass us 0xffffffff though, meaning "enable all samples".
2826 ice
->state
.sample_mask
= sample_mask
& 0xffff;
2827 ice
->state
.dirty
|= IRIS_DIRTY_SAMPLE_MASK
;
2831 * The pipe->set_scissor_states() driver hook.
2833 * This corresponds to our SCISSOR_RECT state structures. It's an
2834 * exact match, so we just store them, and memcpy them out later.
2837 iris_set_scissor_states(struct pipe_context
*ctx
,
2838 unsigned start_slot
,
2839 unsigned num_scissors
,
2840 const struct pipe_scissor_state
*rects
)
2842 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2844 for (unsigned i
= 0; i
< num_scissors
; i
++) {
2845 if (rects
[i
].minx
== rects
[i
].maxx
|| rects
[i
].miny
== rects
[i
].maxy
) {
2846 /* If the scissor was out of bounds and got clamped to 0 width/height
2847 * at the bounds, the subtraction of 1 from maximums could produce a
2848 * negative number and thus not clip anything. Instead, just provide
2849 * a min > max scissor inside the bounds, which produces the expected
2852 ice
->state
.scissors
[start_slot
+ i
] = (struct pipe_scissor_state
) {
2853 .minx
= 1, .maxx
= 0, .miny
= 1, .maxy
= 0,
2856 ice
->state
.scissors
[start_slot
+ i
] = (struct pipe_scissor_state
) {
2857 .minx
= rects
[i
].minx
, .miny
= rects
[i
].miny
,
2858 .maxx
= rects
[i
].maxx
- 1, .maxy
= rects
[i
].maxy
- 1,
2863 ice
->state
.dirty
|= IRIS_DIRTY_SCISSOR_RECT
;
2867 * The pipe->set_stencil_ref() driver hook.
2869 * This is added to 3DSTATE_WM_DEPTH_STENCIL dynamically at draw time.
2872 iris_set_stencil_ref(struct pipe_context
*ctx
,
2873 const struct pipe_stencil_ref
*state
)
2875 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2876 memcpy(&ice
->state
.stencil_ref
, state
, sizeof(*state
));
2878 ice
->state
.dirty
|= IRIS_DIRTY_COLOR_CALC_STATE
;
2880 ice
->state
.dirty
|= IRIS_DIRTY_WM_DEPTH_STENCIL
;
2884 viewport_extent(const struct pipe_viewport_state
*state
, int axis
, float sign
)
2886 return copysignf(state
->scale
[axis
], sign
) + state
->translate
[axis
];
2890 * The pipe->set_viewport_states() driver hook.
2892 * This corresponds to our SF_CLIP_VIEWPORT states. We can't calculate
2893 * the guardband yet, as we need the framebuffer dimensions, but we can
2894 * at least fill out the rest.
2897 iris_set_viewport_states(struct pipe_context
*ctx
,
2898 unsigned start_slot
,
2900 const struct pipe_viewport_state
*states
)
2902 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2904 memcpy(&ice
->state
.viewports
[start_slot
], states
, sizeof(*states
) * count
);
2906 ice
->state
.dirty
|= IRIS_DIRTY_SF_CL_VIEWPORT
;
2908 if (ice
->state
.cso_rast
&& (!ice
->state
.cso_rast
->depth_clip_near
||
2909 !ice
->state
.cso_rast
->depth_clip_far
))
2910 ice
->state
.dirty
|= IRIS_DIRTY_CC_VIEWPORT
;
2914 * The pipe->set_framebuffer_state() driver hook.
2916 * Sets the current draw FBO, including color render targets, depth,
2917 * and stencil buffers.
2920 iris_set_framebuffer_state(struct pipe_context
*ctx
,
2921 const struct pipe_framebuffer_state
*state
)
2923 struct iris_context
*ice
= (struct iris_context
*) ctx
;
2924 struct iris_screen
*screen
= (struct iris_screen
*)ctx
->screen
;
2925 struct isl_device
*isl_dev
= &screen
->isl_dev
;
2926 struct pipe_framebuffer_state
*cso
= &ice
->state
.framebuffer
;
2927 struct iris_resource
*zres
;
2928 struct iris_resource
*stencil_res
;
2930 unsigned samples
= util_framebuffer_get_num_samples(state
);
2931 unsigned layers
= util_framebuffer_get_num_layers(state
);
2933 if (cso
->samples
!= samples
) {
2934 ice
->state
.dirty
|= IRIS_DIRTY_MULTISAMPLE
;
2936 /* We need to toggle 3DSTATE_PS::32 Pixel Dispatch Enable */
2937 if (GEN_GEN
>= 9 && (cso
->samples
== 16 || samples
== 16))
2938 ice
->state
.dirty
|= IRIS_DIRTY_FS
;
2941 if (cso
->nr_cbufs
!= state
->nr_cbufs
) {
2942 ice
->state
.dirty
|= IRIS_DIRTY_BLEND_STATE
;
2945 if ((cso
->layers
== 0) != (layers
== 0)) {
2946 ice
->state
.dirty
|= IRIS_DIRTY_CLIP
;
2949 if (cso
->width
!= state
->width
|| cso
->height
!= state
->height
) {
2950 ice
->state
.dirty
|= IRIS_DIRTY_SF_CL_VIEWPORT
;
2953 if (cso
->zsbuf
|| state
->zsbuf
) {
2954 ice
->state
.dirty
|= IRIS_DIRTY_DEPTH_BUFFER
;
2957 util_copy_framebuffer_state(cso
, state
);
2958 cso
->samples
= samples
;
2959 cso
->layers
= layers
;
2961 struct iris_depth_buffer_state
*cso_z
= &ice
->state
.genx
->depth_buffer
;
2963 struct isl_view view
= {
2966 .base_array_layer
= 0,
2968 .swizzle
= ISL_SWIZZLE_IDENTITY
,
2971 struct isl_depth_stencil_hiz_emit_info info
= { .view
= &view
};
2974 iris_get_depth_stencil_resources(cso
->zsbuf
->texture
, &zres
,
2977 view
.base_level
= cso
->zsbuf
->u
.tex
.level
;
2978 view
.base_array_layer
= cso
->zsbuf
->u
.tex
.first_layer
;
2980 cso
->zsbuf
->u
.tex
.last_layer
- cso
->zsbuf
->u
.tex
.first_layer
+ 1;
2983 view
.usage
|= ISL_SURF_USAGE_DEPTH_BIT
;
2985 info
.depth_surf
= &zres
->surf
;
2986 info
.depth_address
= zres
->bo
->gtt_offset
+ zres
->offset
;
2987 info
.mocs
= mocs(zres
->bo
);
2989 view
.format
= zres
->surf
.format
;
2991 if (iris_resource_level_has_hiz(zres
, view
.base_level
)) {
2992 info
.hiz_usage
= zres
->aux
.usage
;
2993 info
.hiz_surf
= &zres
->aux
.surf
;
2994 info
.hiz_address
= zres
->aux
.bo
->gtt_offset
+ zres
->aux
.offset
;
2999 view
.usage
|= ISL_SURF_USAGE_STENCIL_BIT
;
3000 info
.stencil_aux_usage
= stencil_res
->aux
.usage
;
3001 info
.stencil_surf
= &stencil_res
->surf
;
3002 info
.stencil_address
= stencil_res
->bo
->gtt_offset
+ stencil_res
->offset
;
3004 view
.format
= stencil_res
->surf
.format
;
3005 info
.mocs
= mocs(stencil_res
->bo
);
3010 isl_emit_depth_stencil_hiz_s(isl_dev
, cso_z
->packets
, &info
);
3012 /* Make a null surface for unbound buffers */
3013 void *null_surf_map
=
3014 upload_state(ice
->state
.surface_uploader
, &ice
->state
.null_fb
,
3015 4 * GENX(RENDER_SURFACE_STATE_length
), 64);
3016 isl_null_fill_state(&screen
->isl_dev
, null_surf_map
,
3017 isl_extent3d(MAX2(cso
->width
, 1),
3018 MAX2(cso
->height
, 1),
3019 cso
->layers
? cso
->layers
: 1));
3020 ice
->state
.null_fb
.offset
+=
3021 iris_bo_offset_from_base_address(iris_resource_bo(ice
->state
.null_fb
.res
));
3023 /* Render target change */
3024 ice
->state
.dirty
|= IRIS_DIRTY_BINDINGS_FS
;
3026 ice
->state
.dirty
|= IRIS_DIRTY_RENDER_BUFFER
;
3028 ice
->state
.dirty
|= IRIS_DIRTY_RENDER_RESOLVES_AND_FLUSHES
;
3030 ice
->state
.dirty
|= ice
->state
.dirty_for_nos
[IRIS_NOS_FRAMEBUFFER
];
3033 ice
->state
.dirty
|= IRIS_DIRTY_PMA_FIX
;
3037 * The pipe->set_constant_buffer() driver hook.
3039 * This uploads any constant data in user buffers, and references
3040 * any UBO resources containing constant data.
3043 iris_set_constant_buffer(struct pipe_context
*ctx
,
3044 enum pipe_shader_type p_stage
, unsigned index
,
3045 const struct pipe_constant_buffer
*input
)
3047 struct iris_context
*ice
= (struct iris_context
*) ctx
;
3048 gl_shader_stage stage
= stage_from_pipe(p_stage
);
3049 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
3050 struct pipe_shader_buffer
*cbuf
= &shs
->constbuf
[index
];
3052 /* TODO: Only do this if the buffer changes? */
3053 pipe_resource_reference(&shs
->constbuf_surf_state
[index
].res
, NULL
);
3055 if (input
&& input
->buffer_size
&& (input
->buffer
|| input
->user_buffer
)) {
3056 shs
->bound_cbufs
|= 1u << index
;
3058 if (input
->user_buffer
) {
3060 pipe_resource_reference(&cbuf
->buffer
, NULL
);
3061 u_upload_alloc(ice
->ctx
.const_uploader
, 0, input
->buffer_size
, 64,
3062 &cbuf
->buffer_offset
, &cbuf
->buffer
, (void **) &map
);
3064 if (!cbuf
->buffer
) {
3065 /* Allocation was unsuccessful - just unbind */
3066 iris_set_constant_buffer(ctx
, p_stage
, index
, NULL
);
3071 memcpy(map
, input
->user_buffer
, input
->buffer_size
);
3072 } else if (input
->buffer
) {
3073 pipe_resource_reference(&cbuf
->buffer
, input
->buffer
);
3075 cbuf
->buffer_offset
= input
->buffer_offset
;
3079 MIN2(input
->buffer_size
,
3080 iris_resource_bo(cbuf
->buffer
)->size
- cbuf
->buffer_offset
);
3082 struct iris_resource
*res
= (void *) cbuf
->buffer
;
3083 res
->bind_history
|= PIPE_BIND_CONSTANT_BUFFER
;
3084 res
->bind_stages
|= 1 << stage
;
3086 shs
->bound_cbufs
&= ~(1u << index
);
3087 pipe_resource_reference(&cbuf
->buffer
, NULL
);
3090 ice
->state
.dirty
|= IRIS_DIRTY_CONSTANTS_VS
<< stage
;
3094 upload_sysvals(struct iris_context
*ice
,
3095 gl_shader_stage stage
)
3097 UNUSED
struct iris_genx_state
*genx
= ice
->state
.genx
;
3098 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
3100 struct iris_compiled_shader
*shader
= ice
->shaders
.prog
[stage
];
3101 if (!shader
|| shader
->num_system_values
== 0)
3104 assert(shader
->num_cbufs
> 0);
3106 unsigned sysval_cbuf_index
= shader
->num_cbufs
- 1;
3107 struct pipe_shader_buffer
*cbuf
= &shs
->constbuf
[sysval_cbuf_index
];
3108 unsigned upload_size
= shader
->num_system_values
* sizeof(uint32_t);
3109 uint32_t *map
= NULL
;
3111 assert(sysval_cbuf_index
< PIPE_MAX_CONSTANT_BUFFERS
);
3112 u_upload_alloc(ice
->ctx
.const_uploader
, 0, upload_size
, 64,
3113 &cbuf
->buffer_offset
, &cbuf
->buffer
, (void **) &map
);
3115 for (int i
= 0; i
< shader
->num_system_values
; i
++) {
3116 uint32_t sysval
= shader
->system_values
[i
];
3119 if (BRW_PARAM_DOMAIN(sysval
) == BRW_PARAM_DOMAIN_IMAGE
) {
3121 unsigned img
= BRW_PARAM_IMAGE_IDX(sysval
);
3122 unsigned offset
= BRW_PARAM_IMAGE_OFFSET(sysval
);
3123 struct brw_image_param
*param
=
3124 &genx
->shaders
[stage
].image_param
[img
];
3126 assert(offset
< sizeof(struct brw_image_param
));
3127 value
= ((uint32_t *) param
)[offset
];
3129 } else if (sysval
== BRW_PARAM_BUILTIN_ZERO
) {
3131 } else if (BRW_PARAM_BUILTIN_IS_CLIP_PLANE(sysval
)) {
3132 int plane
= BRW_PARAM_BUILTIN_CLIP_PLANE_IDX(sysval
);
3133 int comp
= BRW_PARAM_BUILTIN_CLIP_PLANE_COMP(sysval
);
3134 value
= fui(ice
->state
.clip_planes
.ucp
[plane
][comp
]);
3135 } else if (sysval
== BRW_PARAM_BUILTIN_PATCH_VERTICES_IN
) {
3136 if (stage
== MESA_SHADER_TESS_CTRL
) {
3137 value
= ice
->state
.vertices_per_patch
;
3139 assert(stage
== MESA_SHADER_TESS_EVAL
);
3140 const struct shader_info
*tcs_info
=
3141 iris_get_shader_info(ice
, MESA_SHADER_TESS_CTRL
);
3143 value
= tcs_info
->tess
.tcs_vertices_out
;
3145 value
= ice
->state
.vertices_per_patch
;
3147 } else if (sysval
>= BRW_PARAM_BUILTIN_TESS_LEVEL_OUTER_X
&&
3148 sysval
<= BRW_PARAM_BUILTIN_TESS_LEVEL_OUTER_W
) {
3149 unsigned i
= sysval
- BRW_PARAM_BUILTIN_TESS_LEVEL_OUTER_X
;
3150 value
= fui(ice
->state
.default_outer_level
[i
]);
3151 } else if (sysval
== BRW_PARAM_BUILTIN_TESS_LEVEL_INNER_X
) {
3152 value
= fui(ice
->state
.default_inner_level
[0]);
3153 } else if (sysval
== BRW_PARAM_BUILTIN_TESS_LEVEL_INNER_Y
) {
3154 value
= fui(ice
->state
.default_inner_level
[1]);
3156 assert(!"unhandled system value");
3162 cbuf
->buffer_size
= upload_size
;
3163 iris_upload_ubo_ssbo_surf_state(ice
, cbuf
,
3164 &shs
->constbuf_surf_state
[sysval_cbuf_index
], false);
3166 shs
->sysvals_need_upload
= false;
3170 * The pipe->set_shader_buffers() driver hook.
3172 * This binds SSBOs and ABOs. Unfortunately, we need to stream out
3173 * SURFACE_STATE here, as the buffer offset may change each time.
3176 iris_set_shader_buffers(struct pipe_context
*ctx
,
3177 enum pipe_shader_type p_stage
,
3178 unsigned start_slot
, unsigned count
,
3179 const struct pipe_shader_buffer
*buffers
,
3180 unsigned writable_bitmask
)
3182 struct iris_context
*ice
= (struct iris_context
*) ctx
;
3183 gl_shader_stage stage
= stage_from_pipe(p_stage
);
3184 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
3186 unsigned modified_bits
= u_bit_consecutive(start_slot
, count
);
3188 shs
->bound_ssbos
&= ~modified_bits
;
3189 shs
->writable_ssbos
&= ~modified_bits
;
3190 shs
->writable_ssbos
|= writable_bitmask
<< start_slot
;
3192 for (unsigned i
= 0; i
< count
; i
++) {
3193 if (buffers
&& buffers
[i
].buffer
) {
3194 struct iris_resource
*res
= (void *) buffers
[i
].buffer
;
3195 struct pipe_shader_buffer
*ssbo
= &shs
->ssbo
[start_slot
+ i
];
3196 struct iris_state_ref
*surf_state
=
3197 &shs
->ssbo_surf_state
[start_slot
+ i
];
3198 pipe_resource_reference(&ssbo
->buffer
, &res
->base
);
3199 ssbo
->buffer_offset
= buffers
[i
].buffer_offset
;
3201 MIN2(buffers
[i
].buffer_size
, res
->bo
->size
- ssbo
->buffer_offset
);
3203 shs
->bound_ssbos
|= 1 << (start_slot
+ i
);
3205 iris_upload_ubo_ssbo_surf_state(ice
, ssbo
, surf_state
, true);
3207 res
->bind_history
|= PIPE_BIND_SHADER_BUFFER
;
3208 res
->bind_stages
|= 1 << stage
;
3210 util_range_add(&res
->base
, &res
->valid_buffer_range
, ssbo
->buffer_offset
,
3211 ssbo
->buffer_offset
+ ssbo
->buffer_size
);
3213 pipe_resource_reference(&shs
->ssbo
[start_slot
+ i
].buffer
, NULL
);
3214 pipe_resource_reference(&shs
->ssbo_surf_state
[start_slot
+ i
].res
,
3219 ice
->state
.dirty
|= IRIS_DIRTY_BINDINGS_VS
<< stage
;
3223 iris_delete_state(struct pipe_context
*ctx
, void *state
)
3229 * The pipe->set_vertex_buffers() driver hook.
3231 * This translates pipe_vertex_buffer to our 3DSTATE_VERTEX_BUFFERS packet.
3234 iris_set_vertex_buffers(struct pipe_context
*ctx
,
3235 unsigned start_slot
, unsigned count
,
3236 const struct pipe_vertex_buffer
*buffers
)
3238 struct iris_context
*ice
= (struct iris_context
*) ctx
;
3239 struct iris_genx_state
*genx
= ice
->state
.genx
;
3241 ice
->state
.bound_vertex_buffers
&= ~u_bit_consecutive64(start_slot
, count
);
3243 for (unsigned i
= 0; i
< count
; i
++) {
3244 const struct pipe_vertex_buffer
*buffer
= buffers
? &buffers
[i
] : NULL
;
3245 struct iris_vertex_buffer_state
*state
=
3246 &genx
->vertex_buffers
[start_slot
+ i
];
3249 pipe_resource_reference(&state
->resource
, NULL
);
3253 /* We may see user buffers that are NULL bindings. */
3254 assert(!(buffer
->is_user_buffer
&& buffer
->buffer
.user
!= NULL
));
3256 pipe_resource_reference(&state
->resource
, buffer
->buffer
.resource
);
3257 struct iris_resource
*res
= (void *) state
->resource
;
3259 state
->offset
= (int) buffer
->buffer_offset
;
3262 ice
->state
.bound_vertex_buffers
|= 1ull << (start_slot
+ i
);
3263 res
->bind_history
|= PIPE_BIND_VERTEX_BUFFER
;
3266 iris_pack_state(GENX(VERTEX_BUFFER_STATE
), state
->state
, vb
) {
3267 vb
.VertexBufferIndex
= start_slot
+ i
;
3268 vb
.AddressModifyEnable
= true;
3269 vb
.BufferPitch
= buffer
->stride
;
3271 vb
.BufferSize
= res
->bo
->size
- (int) buffer
->buffer_offset
;
3272 vb
.BufferStartingAddress
=
3273 ro_bo(NULL
, res
->bo
->gtt_offset
+ (int) buffer
->buffer_offset
);
3274 vb
.MOCS
= mocs(res
->bo
);
3276 vb
.NullVertexBuffer
= true;
3281 ice
->state
.dirty
|= IRIS_DIRTY_VERTEX_BUFFERS
;
3285 * Gallium CSO for vertex elements.
3287 struct iris_vertex_element_state
{
3288 uint32_t vertex_elements
[1 + 33 * GENX(VERTEX_ELEMENT_STATE_length
)];
3289 uint32_t vf_instancing
[33 * GENX(3DSTATE_VF_INSTANCING_length
)];
3290 uint32_t edgeflag_ve
[GENX(VERTEX_ELEMENT_STATE_length
)];
3291 uint32_t edgeflag_vfi
[GENX(3DSTATE_VF_INSTANCING_length
)];
3296 * The pipe->create_vertex_elements() driver hook.
3298 * This translates pipe_vertex_element to our 3DSTATE_VERTEX_ELEMENTS
3299 * and 3DSTATE_VF_INSTANCING commands. The vertex_elements and vf_instancing
3300 * arrays are ready to be emitted at draw time if no EdgeFlag or SGVs are
3301 * needed. In these cases we will need information available at draw time.
3302 * We setup edgeflag_ve and edgeflag_vfi as alternatives last
3303 * 3DSTATE_VERTEX_ELEMENT and 3DSTATE_VF_INSTANCING that can be used at
3304 * draw time if we detect that EdgeFlag is needed by the Vertex Shader.
3307 iris_create_vertex_elements(struct pipe_context
*ctx
,
3309 const struct pipe_vertex_element
*state
)
3311 struct iris_screen
*screen
= (struct iris_screen
*)ctx
->screen
;
3312 const struct gen_device_info
*devinfo
= &screen
->devinfo
;
3313 struct iris_vertex_element_state
*cso
=
3314 malloc(sizeof(struct iris_vertex_element_state
));
3318 iris_pack_command(GENX(3DSTATE_VERTEX_ELEMENTS
), cso
->vertex_elements
, ve
) {
3320 1 + GENX(VERTEX_ELEMENT_STATE_length
) * MAX2(count
, 1) - 2;
3323 uint32_t *ve_pack_dest
= &cso
->vertex_elements
[1];
3324 uint32_t *vfi_pack_dest
= cso
->vf_instancing
;
3327 iris_pack_state(GENX(VERTEX_ELEMENT_STATE
), ve_pack_dest
, ve
) {
3329 ve
.SourceElementFormat
= ISL_FORMAT_R32G32B32A32_FLOAT
;
3330 ve
.Component0Control
= VFCOMP_STORE_0
;
3331 ve
.Component1Control
= VFCOMP_STORE_0
;
3332 ve
.Component2Control
= VFCOMP_STORE_0
;
3333 ve
.Component3Control
= VFCOMP_STORE_1_FP
;
3336 iris_pack_command(GENX(3DSTATE_VF_INSTANCING
), vfi_pack_dest
, vi
) {
3340 for (int i
= 0; i
< count
; i
++) {
3341 const struct iris_format_info fmt
=
3342 iris_format_for_usage(devinfo
, state
[i
].src_format
, 0);
3343 unsigned comp
[4] = { VFCOMP_STORE_SRC
, VFCOMP_STORE_SRC
,
3344 VFCOMP_STORE_SRC
, VFCOMP_STORE_SRC
};
3346 switch (isl_format_get_num_channels(fmt
.fmt
)) {
3347 case 0: comp
[0] = VFCOMP_STORE_0
; /* fallthrough */
3348 case 1: comp
[1] = VFCOMP_STORE_0
; /* fallthrough */
3349 case 2: comp
[2] = VFCOMP_STORE_0
; /* fallthrough */
3351 comp
[3] = isl_format_has_int_channel(fmt
.fmt
) ? VFCOMP_STORE_1_INT
3352 : VFCOMP_STORE_1_FP
;
3355 iris_pack_state(GENX(VERTEX_ELEMENT_STATE
), ve_pack_dest
, ve
) {
3356 ve
.EdgeFlagEnable
= false;
3357 ve
.VertexBufferIndex
= state
[i
].vertex_buffer_index
;
3359 ve
.SourceElementOffset
= state
[i
].src_offset
;
3360 ve
.SourceElementFormat
= fmt
.fmt
;
3361 ve
.Component0Control
= comp
[0];
3362 ve
.Component1Control
= comp
[1];
3363 ve
.Component2Control
= comp
[2];
3364 ve
.Component3Control
= comp
[3];
3367 iris_pack_command(GENX(3DSTATE_VF_INSTANCING
), vfi_pack_dest
, vi
) {
3368 vi
.VertexElementIndex
= i
;
3369 vi
.InstancingEnable
= state
[i
].instance_divisor
> 0;
3370 vi
.InstanceDataStepRate
= state
[i
].instance_divisor
;
3373 ve_pack_dest
+= GENX(VERTEX_ELEMENT_STATE_length
);
3374 vfi_pack_dest
+= GENX(3DSTATE_VF_INSTANCING_length
);
3377 /* An alternative version of the last VE and VFI is stored so it
3378 * can be used at draw time in case Vertex Shader uses EdgeFlag
3381 const unsigned edgeflag_index
= count
- 1;
3382 const struct iris_format_info fmt
=
3383 iris_format_for_usage(devinfo
, state
[edgeflag_index
].src_format
, 0);
3384 iris_pack_state(GENX(VERTEX_ELEMENT_STATE
), cso
->edgeflag_ve
, ve
) {
3385 ve
.EdgeFlagEnable
= true ;
3386 ve
.VertexBufferIndex
= state
[edgeflag_index
].vertex_buffer_index
;
3388 ve
.SourceElementOffset
= state
[edgeflag_index
].src_offset
;
3389 ve
.SourceElementFormat
= fmt
.fmt
;
3390 ve
.Component0Control
= VFCOMP_STORE_SRC
;
3391 ve
.Component1Control
= VFCOMP_STORE_0
;
3392 ve
.Component2Control
= VFCOMP_STORE_0
;
3393 ve
.Component3Control
= VFCOMP_STORE_0
;
3395 iris_pack_command(GENX(3DSTATE_VF_INSTANCING
), cso
->edgeflag_vfi
, vi
) {
3396 /* The vi.VertexElementIndex of the EdgeFlag Vertex Element is filled
3397 * at draw time, as it should change if SGVs are emitted.
3399 vi
.InstancingEnable
= state
[edgeflag_index
].instance_divisor
> 0;
3400 vi
.InstanceDataStepRate
= state
[edgeflag_index
].instance_divisor
;
3408 * The pipe->bind_vertex_elements_state() driver hook.
3411 iris_bind_vertex_elements_state(struct pipe_context
*ctx
, void *state
)
3413 struct iris_context
*ice
= (struct iris_context
*) ctx
;
3414 struct iris_vertex_element_state
*old_cso
= ice
->state
.cso_vertex_elements
;
3415 struct iris_vertex_element_state
*new_cso
= state
;
3417 /* 3DSTATE_VF_SGVs overrides the last VE, so if the count is changing,
3418 * we need to re-emit it to ensure we're overriding the right one.
3420 if (new_cso
&& cso_changed(count
))
3421 ice
->state
.dirty
|= IRIS_DIRTY_VF_SGVS
;
3423 ice
->state
.cso_vertex_elements
= state
;
3424 ice
->state
.dirty
|= IRIS_DIRTY_VERTEX_ELEMENTS
;
3428 * The pipe->create_stream_output_target() driver hook.
3430 * "Target" here refers to a destination buffer. We translate this into
3431 * a 3DSTATE_SO_BUFFER packet. We can handle most fields, but don't yet
3432 * know which buffer this represents, or whether we ought to zero the
3433 * write-offsets, or append. Those are handled in the set() hook.
3435 static struct pipe_stream_output_target
*
3436 iris_create_stream_output_target(struct pipe_context
*ctx
,
3437 struct pipe_resource
*p_res
,
3438 unsigned buffer_offset
,
3439 unsigned buffer_size
)
3441 struct iris_resource
*res
= (void *) p_res
;
3442 struct iris_stream_output_target
*cso
= calloc(1, sizeof(*cso
));
3446 res
->bind_history
|= PIPE_BIND_STREAM_OUTPUT
;
3448 pipe_reference_init(&cso
->base
.reference
, 1);
3449 pipe_resource_reference(&cso
->base
.buffer
, p_res
);
3450 cso
->base
.buffer_offset
= buffer_offset
;
3451 cso
->base
.buffer_size
= buffer_size
;
3452 cso
->base
.context
= ctx
;
3454 util_range_add(&res
->base
, &res
->valid_buffer_range
, buffer_offset
,
3455 buffer_offset
+ buffer_size
);
3457 upload_state(ctx
->stream_uploader
, &cso
->offset
, sizeof(uint32_t), 4);
3463 iris_stream_output_target_destroy(struct pipe_context
*ctx
,
3464 struct pipe_stream_output_target
*state
)
3466 struct iris_stream_output_target
*cso
= (void *) state
;
3468 pipe_resource_reference(&cso
->base
.buffer
, NULL
);
3469 pipe_resource_reference(&cso
->offset
.res
, NULL
);
3475 * The pipe->set_stream_output_targets() driver hook.
3477 * At this point, we know which targets are bound to a particular index,
3478 * and also whether we want to append or start over. We can finish the
3479 * 3DSTATE_SO_BUFFER packets we started earlier.
3482 iris_set_stream_output_targets(struct pipe_context
*ctx
,
3483 unsigned num_targets
,
3484 struct pipe_stream_output_target
**targets
,
3485 const unsigned *offsets
)
3487 struct iris_context
*ice
= (struct iris_context
*) ctx
;
3488 struct iris_genx_state
*genx
= ice
->state
.genx
;
3489 uint32_t *so_buffers
= genx
->so_buffers
;
3491 const bool active
= num_targets
> 0;
3492 if (ice
->state
.streamout_active
!= active
) {
3493 ice
->state
.streamout_active
= active
;
3494 ice
->state
.dirty
|= IRIS_DIRTY_STREAMOUT
;
3496 /* We only emit 3DSTATE_SO_DECL_LIST when streamout is active, because
3497 * it's a non-pipelined command. If we're switching streamout on, we
3498 * may have missed emitting it earlier, so do so now. (We're already
3499 * taking a stall to update 3DSTATE_SO_BUFFERS anyway...)
3502 ice
->state
.dirty
|= IRIS_DIRTY_SO_DECL_LIST
;
3505 for (int i
= 0; i
< PIPE_MAX_SO_BUFFERS
; i
++) {
3506 struct iris_stream_output_target
*tgt
=
3507 (void *) ice
->state
.so_target
[i
];
3509 struct iris_resource
*res
= (void *) tgt
->base
.buffer
;
3511 flush
|= iris_flush_bits_for_history(res
);
3512 iris_dirty_for_history(ice
, res
);
3515 iris_emit_pipe_control_flush(&ice
->batches
[IRIS_BATCH_RENDER
],
3516 "make streamout results visible", flush
);
3520 for (int i
= 0; i
< 4; i
++) {
3521 pipe_so_target_reference(&ice
->state
.so_target
[i
],
3522 i
< num_targets
? targets
[i
] : NULL
);
3525 /* No need to update 3DSTATE_SO_BUFFER unless SOL is active. */
3529 for (unsigned i
= 0; i
< 4; i
++,
3530 so_buffers
+= GENX(3DSTATE_SO_BUFFER_length
)) {
3532 struct iris_stream_output_target
*tgt
= (void *) ice
->state
.so_target
[i
];
3533 unsigned offset
= offsets
[i
];
3536 iris_pack_command(GENX(3DSTATE_SO_BUFFER
), so_buffers
, sob
) {
3538 sob
.SOBufferIndex
= i
;
3540 sob
._3DCommandOpcode
= 0;
3541 sob
._3DCommandSubOpcode
= SO_BUFFER_INDEX_0_CMD
+ i
;
3547 struct iris_resource
*res
= (void *) tgt
->base
.buffer
;
3549 /* Note that offsets[i] will either be 0, causing us to zero
3550 * the value in the buffer, or 0xFFFFFFFF, which happens to mean
3551 * "continue appending at the existing offset."
3553 assert(offset
== 0 || offset
== 0xFFFFFFFF);
3555 /* We might be called by Begin (offset = 0), Pause, then Resume
3556 * (offset = 0xFFFFFFFF) before ever drawing (where these commands
3557 * will actually be sent to the GPU). In this case, we don't want
3558 * to append - we still want to do our initial zeroing.
3563 iris_pack_command(GENX(3DSTATE_SO_BUFFER
), so_buffers
, sob
) {
3565 sob
.SOBufferIndex
= i
;
3567 sob
._3DCommandOpcode
= 0;
3568 sob
._3DCommandSubOpcode
= SO_BUFFER_INDEX_0_CMD
+ i
;
3570 sob
.SurfaceBaseAddress
=
3571 rw_bo(NULL
, res
->bo
->gtt_offset
+ tgt
->base
.buffer_offset
);
3572 sob
.SOBufferEnable
= true;
3573 sob
.StreamOffsetWriteEnable
= true;
3574 sob
.StreamOutputBufferOffsetAddressEnable
= true;
3575 sob
.MOCS
= mocs(res
->bo
);
3577 sob
.SurfaceSize
= MAX2(tgt
->base
.buffer_size
/ 4, 1) - 1;
3578 sob
.StreamOffset
= offset
;
3579 sob
.StreamOutputBufferOffsetAddress
=
3580 rw_bo(NULL
, iris_resource_bo(tgt
->offset
.res
)->gtt_offset
+
3581 tgt
->offset
.offset
);
3585 ice
->state
.dirty
|= IRIS_DIRTY_SO_BUFFERS
;
3589 * An iris-vtable helper for encoding the 3DSTATE_SO_DECL_LIST and
3590 * 3DSTATE_STREAMOUT packets.
3592 * 3DSTATE_SO_DECL_LIST is a list of shader outputs we want the streamout
3593 * hardware to record. We can create it entirely based on the shader, with
3594 * no dynamic state dependencies.
3596 * 3DSTATE_STREAMOUT is an annoying mix of shader-based information and
3597 * state-based settings. We capture the shader-related ones here, and merge
3598 * the rest in at draw time.
3601 iris_create_so_decl_list(const struct pipe_stream_output_info
*info
,
3602 const struct brw_vue_map
*vue_map
)
3604 struct GENX(SO_DECL
) so_decl
[MAX_VERTEX_STREAMS
][128];
3605 int buffer_mask
[MAX_VERTEX_STREAMS
] = {0, 0, 0, 0};
3606 int next_offset
[MAX_VERTEX_STREAMS
] = {0, 0, 0, 0};
3607 int decls
[MAX_VERTEX_STREAMS
] = {0, 0, 0, 0};
3609 STATIC_ASSERT(ARRAY_SIZE(so_decl
[0]) >= MAX_PROGRAM_OUTPUTS
);
3611 memset(so_decl
, 0, sizeof(so_decl
));
3613 /* Construct the list of SO_DECLs to be emitted. The formatting of the
3614 * command feels strange -- each dword pair contains a SO_DECL per stream.
3616 for (unsigned i
= 0; i
< info
->num_outputs
; i
++) {
3617 const struct pipe_stream_output
*output
= &info
->output
[i
];
3618 const int buffer
= output
->output_buffer
;
3619 const int varying
= output
->register_index
;
3620 const unsigned stream_id
= output
->stream
;
3621 assert(stream_id
< MAX_VERTEX_STREAMS
);
3623 buffer_mask
[stream_id
] |= 1 << buffer
;
3625 assert(vue_map
->varying_to_slot
[varying
] >= 0);
3627 /* Mesa doesn't store entries for gl_SkipComponents in the Outputs[]
3628 * array. Instead, it simply increments DstOffset for the following
3629 * input by the number of components that should be skipped.
3631 * Our hardware is unusual in that it requires us to program SO_DECLs
3632 * for fake "hole" components, rather than simply taking the offset
3633 * for each real varying. Each hole can have size 1, 2, 3, or 4; we
3634 * program as many size = 4 holes as we can, then a final hole to
3635 * accommodate the final 1, 2, or 3 remaining.
3637 int skip_components
= output
->dst_offset
- next_offset
[buffer
];
3639 while (skip_components
> 0) {
3640 so_decl
[stream_id
][decls
[stream_id
]++] = (struct GENX(SO_DECL
)) {
3642 .OutputBufferSlot
= output
->output_buffer
,
3643 .ComponentMask
= (1 << MIN2(skip_components
, 4)) - 1,
3645 skip_components
-= 4;
3648 next_offset
[buffer
] = output
->dst_offset
+ output
->num_components
;
3650 so_decl
[stream_id
][decls
[stream_id
]++] = (struct GENX(SO_DECL
)) {
3651 .OutputBufferSlot
= output
->output_buffer
,
3652 .RegisterIndex
= vue_map
->varying_to_slot
[varying
],
3654 ((1 << output
->num_components
) - 1) << output
->start_component
,
3657 if (decls
[stream_id
] > max_decls
)
3658 max_decls
= decls
[stream_id
];
3661 unsigned dwords
= GENX(3DSTATE_STREAMOUT_length
) + (3 + 2 * max_decls
);
3662 uint32_t *map
= ralloc_size(NULL
, sizeof(uint32_t) * dwords
);
3663 uint32_t *so_decl_map
= map
+ GENX(3DSTATE_STREAMOUT_length
);
3665 iris_pack_command(GENX(3DSTATE_STREAMOUT
), map
, sol
) {
3666 int urb_entry_read_offset
= 0;
3667 int urb_entry_read_length
= (vue_map
->num_slots
+ 1) / 2 -
3668 urb_entry_read_offset
;
3670 /* We always read the whole vertex. This could be reduced at some
3671 * point by reading less and offsetting the register index in the
3674 sol
.Stream0VertexReadOffset
= urb_entry_read_offset
;
3675 sol
.Stream0VertexReadLength
= urb_entry_read_length
- 1;
3676 sol
.Stream1VertexReadOffset
= urb_entry_read_offset
;
3677 sol
.Stream1VertexReadLength
= urb_entry_read_length
- 1;
3678 sol
.Stream2VertexReadOffset
= urb_entry_read_offset
;
3679 sol
.Stream2VertexReadLength
= urb_entry_read_length
- 1;
3680 sol
.Stream3VertexReadOffset
= urb_entry_read_offset
;
3681 sol
.Stream3VertexReadLength
= urb_entry_read_length
- 1;
3683 /* Set buffer pitches; 0 means unbound. */
3684 sol
.Buffer0SurfacePitch
= 4 * info
->stride
[0];
3685 sol
.Buffer1SurfacePitch
= 4 * info
->stride
[1];
3686 sol
.Buffer2SurfacePitch
= 4 * info
->stride
[2];
3687 sol
.Buffer3SurfacePitch
= 4 * info
->stride
[3];
3690 iris_pack_command(GENX(3DSTATE_SO_DECL_LIST
), so_decl_map
, list
) {
3691 list
.DWordLength
= 3 + 2 * max_decls
- 2;
3692 list
.StreamtoBufferSelects0
= buffer_mask
[0];
3693 list
.StreamtoBufferSelects1
= buffer_mask
[1];
3694 list
.StreamtoBufferSelects2
= buffer_mask
[2];
3695 list
.StreamtoBufferSelects3
= buffer_mask
[3];
3696 list
.NumEntries0
= decls
[0];
3697 list
.NumEntries1
= decls
[1];
3698 list
.NumEntries2
= decls
[2];
3699 list
.NumEntries3
= decls
[3];
3702 for (int i
= 0; i
< max_decls
; i
++) {
3703 iris_pack_state(GENX(SO_DECL_ENTRY
), so_decl_map
+ 3 + i
* 2, entry
) {
3704 entry
.Stream0Decl
= so_decl
[0][i
];
3705 entry
.Stream1Decl
= so_decl
[1][i
];
3706 entry
.Stream2Decl
= so_decl
[2][i
];
3707 entry
.Stream3Decl
= so_decl
[3][i
];
3715 iris_compute_sbe_urb_read_interval(uint64_t fs_input_slots
,
3716 const struct brw_vue_map
*last_vue_map
,
3717 bool two_sided_color
,
3718 unsigned *out_offset
,
3719 unsigned *out_length
)
3721 /* The compiler computes the first URB slot without considering COL/BFC
3722 * swizzling (because it doesn't know whether it's enabled), so we need
3723 * to do that here too. This may result in a smaller offset, which
3726 const unsigned first_slot
=
3727 brw_compute_first_urb_slot_required(fs_input_slots
, last_vue_map
);
3729 /* This becomes the URB read offset (counted in pairs of slots). */
3730 assert(first_slot
% 2 == 0);
3731 *out_offset
= first_slot
/ 2;
3733 /* We need to adjust the inputs read to account for front/back color
3734 * swizzling, as it can make the URB length longer.
3736 for (int c
= 0; c
<= 1; c
++) {
3737 if (fs_input_slots
& (VARYING_BIT_COL0
<< c
)) {
3738 /* If two sided color is enabled, the fragment shader's gl_Color
3739 * (COL0) input comes from either the gl_FrontColor (COL0) or
3740 * gl_BackColor (BFC0) input varyings. Mark BFC as used, too.
3742 if (two_sided_color
)
3743 fs_input_slots
|= (VARYING_BIT_BFC0
<< c
);
3745 /* If front color isn't written, we opt to give them back color
3746 * instead of an undefined value. Switch from COL to BFC.
3748 if (last_vue_map
->varying_to_slot
[VARYING_SLOT_COL0
+ c
] == -1) {
3749 fs_input_slots
&= ~(VARYING_BIT_COL0
<< c
);
3750 fs_input_slots
|= (VARYING_BIT_BFC0
<< c
);
3755 /* Compute the minimum URB Read Length necessary for the FS inputs.
3757 * From the Sandy Bridge PRM, Volume 2, Part 1, documentation for
3758 * 3DSTATE_SF DWord 1 bits 15:11, "Vertex URB Entry Read Length":
3760 * "This field should be set to the minimum length required to read the
3761 * maximum source attribute. The maximum source attribute is indicated
3762 * by the maximum value of the enabled Attribute # Source Attribute if
3763 * Attribute Swizzle Enable is set, Number of Output Attributes-1 if
3764 * enable is not set.
3765 * read_length = ceiling((max_source_attr + 1) / 2)
3767 * [errata] Corruption/Hang possible if length programmed larger than
3770 * Similar text exists for Ivy Bridge.
3772 * We find the last URB slot that's actually read by the FS.
3774 unsigned last_read_slot
= last_vue_map
->num_slots
- 1;
3775 while (last_read_slot
> first_slot
&& !(fs_input_slots
&
3776 (1ull << last_vue_map
->slot_to_varying
[last_read_slot
])))
3779 /* The URB read length is the difference of the two, counted in pairs. */
3780 *out_length
= DIV_ROUND_UP(last_read_slot
- first_slot
+ 1, 2);
3784 iris_emit_sbe_swiz(struct iris_batch
*batch
,
3785 const struct iris_context
*ice
,
3786 unsigned urb_read_offset
,
3787 unsigned sprite_coord_enables
)
3789 struct GENX(SF_OUTPUT_ATTRIBUTE_DETAIL
) attr_overrides
[16] = {};
3790 const struct brw_wm_prog_data
*wm_prog_data
= (void *)
3791 ice
->shaders
.prog
[MESA_SHADER_FRAGMENT
]->prog_data
;
3792 const struct brw_vue_map
*vue_map
= ice
->shaders
.last_vue_map
;
3793 const struct iris_rasterizer_state
*cso_rast
= ice
->state
.cso_rast
;
3795 /* XXX: this should be generated when putting programs in place */
3797 for (int fs_attr
= 0; fs_attr
< VARYING_SLOT_MAX
; fs_attr
++) {
3798 const int input_index
= wm_prog_data
->urb_setup
[fs_attr
];
3799 if (input_index
< 0 || input_index
>= 16)
3802 struct GENX(SF_OUTPUT_ATTRIBUTE_DETAIL
) *attr
=
3803 &attr_overrides
[input_index
];
3804 int slot
= vue_map
->varying_to_slot
[fs_attr
];
3806 /* Viewport and Layer are stored in the VUE header. We need to override
3807 * them to zero if earlier stages didn't write them, as GL requires that
3808 * they read back as zero when not explicitly set.
3811 case VARYING_SLOT_VIEWPORT
:
3812 case VARYING_SLOT_LAYER
:
3813 attr
->ComponentOverrideX
= true;
3814 attr
->ComponentOverrideW
= true;
3815 attr
->ConstantSource
= CONST_0000
;
3817 if (!(vue_map
->slots_valid
& VARYING_BIT_LAYER
))
3818 attr
->ComponentOverrideY
= true;
3819 if (!(vue_map
->slots_valid
& VARYING_BIT_VIEWPORT
))
3820 attr
->ComponentOverrideZ
= true;
3823 case VARYING_SLOT_PRIMITIVE_ID
:
3824 /* Override if the previous shader stage didn't write gl_PrimitiveID. */
3826 attr
->ComponentOverrideX
= true;
3827 attr
->ComponentOverrideY
= true;
3828 attr
->ComponentOverrideZ
= true;
3829 attr
->ComponentOverrideW
= true;
3830 attr
->ConstantSource
= PRIM_ID
;
3838 if (sprite_coord_enables
& (1 << input_index
))
3841 /* If there was only a back color written but not front, use back
3842 * as the color instead of undefined.
3844 if (slot
== -1 && fs_attr
== VARYING_SLOT_COL0
)
3845 slot
= vue_map
->varying_to_slot
[VARYING_SLOT_BFC0
];
3846 if (slot
== -1 && fs_attr
== VARYING_SLOT_COL1
)
3847 slot
= vue_map
->varying_to_slot
[VARYING_SLOT_BFC1
];
3849 /* Not written by the previous stage - undefined. */
3851 attr
->ComponentOverrideX
= true;
3852 attr
->ComponentOverrideY
= true;
3853 attr
->ComponentOverrideZ
= true;
3854 attr
->ComponentOverrideW
= true;
3855 attr
->ConstantSource
= CONST_0001_FLOAT
;
3859 /* Compute the location of the attribute relative to the read offset,
3860 * which is counted in 256-bit increments (two 128-bit VUE slots).
3862 const int source_attr
= slot
- 2 * urb_read_offset
;
3863 assert(source_attr
>= 0 && source_attr
<= 32);
3864 attr
->SourceAttribute
= source_attr
;
3866 /* If we are doing two-sided color, and the VUE slot following this one
3867 * represents a back-facing color, then we need to instruct the SF unit
3868 * to do back-facing swizzling.
3870 if (cso_rast
->light_twoside
&&
3871 ((vue_map
->slot_to_varying
[slot
] == VARYING_SLOT_COL0
&&
3872 vue_map
->slot_to_varying
[slot
+1] == VARYING_SLOT_BFC0
) ||
3873 (vue_map
->slot_to_varying
[slot
] == VARYING_SLOT_COL1
&&
3874 vue_map
->slot_to_varying
[slot
+1] == VARYING_SLOT_BFC1
)))
3875 attr
->SwizzleSelect
= INPUTATTR_FACING
;
3878 iris_emit_cmd(batch
, GENX(3DSTATE_SBE_SWIZ
), sbes
) {
3879 for (int i
= 0; i
< 16; i
++)
3880 sbes
.Attribute
[i
] = attr_overrides
[i
];
3885 iris_calculate_point_sprite_overrides(const struct brw_wm_prog_data
*prog_data
,
3886 const struct iris_rasterizer_state
*cso
)
3888 unsigned overrides
= 0;
3890 if (prog_data
->urb_setup
[VARYING_SLOT_PNTC
] != -1)
3891 overrides
|= 1 << prog_data
->urb_setup
[VARYING_SLOT_PNTC
];
3893 for (int i
= 0; i
< 8; i
++) {
3894 if ((cso
->sprite_coord_enable
& (1 << i
)) &&
3895 prog_data
->urb_setup
[VARYING_SLOT_TEX0
+ i
] != -1)
3896 overrides
|= 1 << prog_data
->urb_setup
[VARYING_SLOT_TEX0
+ i
];
3903 iris_emit_sbe(struct iris_batch
*batch
, const struct iris_context
*ice
)
3905 const struct iris_rasterizer_state
*cso_rast
= ice
->state
.cso_rast
;
3906 const struct brw_wm_prog_data
*wm_prog_data
= (void *)
3907 ice
->shaders
.prog
[MESA_SHADER_FRAGMENT
]->prog_data
;
3908 const struct shader_info
*fs_info
=
3909 iris_get_shader_info(ice
, MESA_SHADER_FRAGMENT
);
3911 unsigned urb_read_offset
, urb_read_length
;
3912 iris_compute_sbe_urb_read_interval(fs_info
->inputs_read
,
3913 ice
->shaders
.last_vue_map
,
3914 cso_rast
->light_twoside
,
3915 &urb_read_offset
, &urb_read_length
);
3917 unsigned sprite_coord_overrides
=
3918 iris_calculate_point_sprite_overrides(wm_prog_data
, cso_rast
);
3920 iris_emit_cmd(batch
, GENX(3DSTATE_SBE
), sbe
) {
3921 sbe
.AttributeSwizzleEnable
= true;
3922 sbe
.NumberofSFOutputAttributes
= wm_prog_data
->num_varying_inputs
;
3923 sbe
.PointSpriteTextureCoordinateOrigin
= cso_rast
->sprite_coord_mode
;
3924 sbe
.VertexURBEntryReadOffset
= urb_read_offset
;
3925 sbe
.VertexURBEntryReadLength
= urb_read_length
;
3926 sbe
.ForceVertexURBEntryReadOffset
= true;
3927 sbe
.ForceVertexURBEntryReadLength
= true;
3928 sbe
.ConstantInterpolationEnable
= wm_prog_data
->flat_inputs
;
3929 sbe
.PointSpriteTextureCoordinateEnable
= sprite_coord_overrides
;
3931 for (int i
= 0; i
< 32; i
++) {
3932 sbe
.AttributeActiveComponentFormat
[i
] = ACTIVE_COMPONENT_XYZW
;
3937 iris_emit_sbe_swiz(batch
, ice
, urb_read_offset
, sprite_coord_overrides
);
3940 /* ------------------------------------------------------------------- */
3943 * Populate VS program key fields based on the current state.
3946 iris_populate_vs_key(const struct iris_context
*ice
,
3947 const struct shader_info
*info
,
3948 gl_shader_stage last_stage
,
3949 struct brw_vs_prog_key
*key
)
3951 const struct iris_rasterizer_state
*cso_rast
= ice
->state
.cso_rast
;
3953 if (info
->clip_distance_array_size
== 0 &&
3954 (info
->outputs_written
& (VARYING_BIT_POS
| VARYING_BIT_CLIP_VERTEX
)) &&
3955 last_stage
== MESA_SHADER_VERTEX
)
3956 key
->nr_userclip_plane_consts
= cso_rast
->num_clip_plane_consts
;
3960 * Populate TCS program key fields based on the current state.
3963 iris_populate_tcs_key(const struct iris_context
*ice
,
3964 struct brw_tcs_prog_key
*key
)
3969 * Populate TES program key fields based on the current state.
3972 iris_populate_tes_key(const struct iris_context
*ice
,
3973 const struct shader_info
*info
,
3974 gl_shader_stage last_stage
,
3975 struct brw_tes_prog_key
*key
)
3977 const struct iris_rasterizer_state
*cso_rast
= ice
->state
.cso_rast
;
3979 if (info
->clip_distance_array_size
== 0 &&
3980 (info
->outputs_written
& (VARYING_BIT_POS
| VARYING_BIT_CLIP_VERTEX
)) &&
3981 last_stage
== MESA_SHADER_TESS_EVAL
)
3982 key
->nr_userclip_plane_consts
= cso_rast
->num_clip_plane_consts
;
3986 * Populate GS program key fields based on the current state.
3989 iris_populate_gs_key(const struct iris_context
*ice
,
3990 const struct shader_info
*info
,
3991 gl_shader_stage last_stage
,
3992 struct brw_gs_prog_key
*key
)
3994 const struct iris_rasterizer_state
*cso_rast
= ice
->state
.cso_rast
;
3996 if (info
->clip_distance_array_size
== 0 &&
3997 (info
->outputs_written
& (VARYING_BIT_POS
| VARYING_BIT_CLIP_VERTEX
)) &&
3998 last_stage
== MESA_SHADER_GEOMETRY
)
3999 key
->nr_userclip_plane_consts
= cso_rast
->num_clip_plane_consts
;
4003 * Populate FS program key fields based on the current state.
4006 iris_populate_fs_key(const struct iris_context
*ice
,
4007 const struct shader_info
*info
,
4008 struct brw_wm_prog_key
*key
)
4010 struct iris_screen
*screen
= (void *) ice
->ctx
.screen
;
4011 const struct pipe_framebuffer_state
*fb
= &ice
->state
.framebuffer
;
4012 const struct iris_depth_stencil_alpha_state
*zsa
= ice
->state
.cso_zsa
;
4013 const struct iris_rasterizer_state
*rast
= ice
->state
.cso_rast
;
4014 const struct iris_blend_state
*blend
= ice
->state
.cso_blend
;
4016 key
->nr_color_regions
= fb
->nr_cbufs
;
4018 key
->clamp_fragment_color
= rast
->clamp_fragment_color
;
4020 key
->alpha_to_coverage
= blend
->alpha_to_coverage
;
4022 key
->alpha_test_replicate_alpha
= fb
->nr_cbufs
> 1 && zsa
->alpha
.enabled
;
4024 key
->flat_shade
= rast
->flatshade
&&
4025 (info
->inputs_read
& (VARYING_BIT_COL0
| VARYING_BIT_COL1
));
4027 key
->persample_interp
= rast
->force_persample_interp
;
4028 key
->multisample_fbo
= rast
->multisample
&& fb
->samples
> 1;
4030 key
->coherent_fb_fetch
= GEN_GEN
>= 9;
4032 key
->force_dual_color_blend
=
4033 screen
->driconf
.dual_color_blend_by_location
&&
4034 (blend
->blend_enables
& 1) && blend
->dual_color_blending
;
4036 /* TODO: Respect glHint for key->high_quality_derivatives */
4040 iris_populate_cs_key(const struct iris_context
*ice
,
4041 struct brw_cs_prog_key
*key
)
4046 KSP(const struct iris_compiled_shader
*shader
)
4048 struct iris_resource
*res
= (void *) shader
->assembly
.res
;
4049 return iris_bo_offset_from_base_address(res
->bo
) + shader
->assembly
.offset
;
4052 /* Gen11 workaround table #2056 WABTPPrefetchDisable suggests to disable
4053 * prefetching of binding tables in A0 and B0 steppings. XXX: Revisit
4054 * this WA on C0 stepping.
4056 * TODO: Fill out SamplerCount for prefetching?
4059 #define INIT_THREAD_DISPATCH_FIELDS(pkt, prefix, stage) \
4060 pkt.KernelStartPointer = KSP(shader); \
4061 pkt.BindingTableEntryCount = GEN_GEN == 11 ? 0 : \
4062 shader->bt.size_bytes / 4; \
4063 pkt.FloatingPointMode = prog_data->use_alt_mode; \
4065 pkt.DispatchGRFStartRegisterForURBData = \
4066 prog_data->dispatch_grf_start_reg; \
4067 pkt.prefix##URBEntryReadLength = vue_prog_data->urb_read_length; \
4068 pkt.prefix##URBEntryReadOffset = 0; \
4070 pkt.StatisticsEnable = true; \
4071 pkt.Enable = true; \
4073 if (prog_data->total_scratch) { \
4074 struct iris_bo *bo = \
4075 iris_get_scratch_space(ice, prog_data->total_scratch, stage); \
4076 uint32_t scratch_addr = bo->gtt_offset; \
4077 pkt.PerThreadScratchSpace = ffs(prog_data->total_scratch) - 11; \
4078 pkt.ScratchSpaceBasePointer = rw_bo(NULL, scratch_addr); \
4082 * Encode most of 3DSTATE_VS based on the compiled shader.
4085 iris_store_vs_state(struct iris_context
*ice
,
4086 const struct gen_device_info
*devinfo
,
4087 struct iris_compiled_shader
*shader
)
4089 struct brw_stage_prog_data
*prog_data
= shader
->prog_data
;
4090 struct brw_vue_prog_data
*vue_prog_data
= (void *) prog_data
;
4092 iris_pack_command(GENX(3DSTATE_VS
), shader
->derived_data
, vs
) {
4093 INIT_THREAD_DISPATCH_FIELDS(vs
, Vertex
, MESA_SHADER_VERTEX
);
4094 vs
.MaximumNumberofThreads
= devinfo
->max_vs_threads
- 1;
4095 vs
.SIMD8DispatchEnable
= true;
4096 vs
.UserClipDistanceCullTestEnableBitmask
=
4097 vue_prog_data
->cull_distance_mask
;
4102 * Encode most of 3DSTATE_HS based on the compiled shader.
4105 iris_store_tcs_state(struct iris_context
*ice
,
4106 const struct gen_device_info
*devinfo
,
4107 struct iris_compiled_shader
*shader
)
4109 struct brw_stage_prog_data
*prog_data
= shader
->prog_data
;
4110 struct brw_vue_prog_data
*vue_prog_data
= (void *) prog_data
;
4111 struct brw_tcs_prog_data
*tcs_prog_data
= (void *) prog_data
;
4113 iris_pack_command(GENX(3DSTATE_HS
), shader
->derived_data
, hs
) {
4114 INIT_THREAD_DISPATCH_FIELDS(hs
, Vertex
, MESA_SHADER_TESS_CTRL
);
4116 hs
.InstanceCount
= tcs_prog_data
->instances
- 1;
4117 hs
.MaximumNumberofThreads
= devinfo
->max_tcs_threads
- 1;
4118 hs
.IncludeVertexHandles
= true;
4121 hs
.DispatchMode
= vue_prog_data
->dispatch_mode
;
4122 hs
.IncludePrimitiveID
= tcs_prog_data
->include_primitive_id
;
4128 * Encode 3DSTATE_TE and most of 3DSTATE_DS based on the compiled shader.
4131 iris_store_tes_state(struct iris_context
*ice
,
4132 const struct gen_device_info
*devinfo
,
4133 struct iris_compiled_shader
*shader
)
4135 struct brw_stage_prog_data
*prog_data
= shader
->prog_data
;
4136 struct brw_vue_prog_data
*vue_prog_data
= (void *) prog_data
;
4137 struct brw_tes_prog_data
*tes_prog_data
= (void *) prog_data
;
4139 uint32_t *te_state
= (void *) shader
->derived_data
;
4140 uint32_t *ds_state
= te_state
+ GENX(3DSTATE_TE_length
);
4142 iris_pack_command(GENX(3DSTATE_TE
), te_state
, te
) {
4143 te
.Partitioning
= tes_prog_data
->partitioning
;
4144 te
.OutputTopology
= tes_prog_data
->output_topology
;
4145 te
.TEDomain
= tes_prog_data
->domain
;
4147 te
.MaximumTessellationFactorOdd
= 63.0;
4148 te
.MaximumTessellationFactorNotOdd
= 64.0;
4151 iris_pack_command(GENX(3DSTATE_DS
), ds_state
, ds
) {
4152 INIT_THREAD_DISPATCH_FIELDS(ds
, Patch
, MESA_SHADER_TESS_EVAL
);
4154 ds
.DispatchMode
= DISPATCH_MODE_SIMD8_SINGLE_PATCH
;
4155 ds
.MaximumNumberofThreads
= devinfo
->max_tes_threads
- 1;
4156 ds
.ComputeWCoordinateEnable
=
4157 tes_prog_data
->domain
== BRW_TESS_DOMAIN_TRI
;
4159 ds
.UserClipDistanceCullTestEnableBitmask
=
4160 vue_prog_data
->cull_distance_mask
;
4166 * Encode most of 3DSTATE_GS based on the compiled shader.
4169 iris_store_gs_state(struct iris_context
*ice
,
4170 const struct gen_device_info
*devinfo
,
4171 struct iris_compiled_shader
*shader
)
4173 struct brw_stage_prog_data
*prog_data
= shader
->prog_data
;
4174 struct brw_vue_prog_data
*vue_prog_data
= (void *) prog_data
;
4175 struct brw_gs_prog_data
*gs_prog_data
= (void *) prog_data
;
4177 iris_pack_command(GENX(3DSTATE_GS
), shader
->derived_data
, gs
) {
4178 INIT_THREAD_DISPATCH_FIELDS(gs
, Vertex
, MESA_SHADER_GEOMETRY
);
4180 gs
.OutputVertexSize
= gs_prog_data
->output_vertex_size_hwords
* 2 - 1;
4181 gs
.OutputTopology
= gs_prog_data
->output_topology
;
4182 gs
.ControlDataHeaderSize
=
4183 gs_prog_data
->control_data_header_size_hwords
;
4184 gs
.InstanceControl
= gs_prog_data
->invocations
- 1;
4185 gs
.DispatchMode
= DISPATCH_MODE_SIMD8
;
4186 gs
.IncludePrimitiveID
= gs_prog_data
->include_primitive_id
;
4187 gs
.ControlDataFormat
= gs_prog_data
->control_data_format
;
4188 gs
.ReorderMode
= TRAILING
;
4189 gs
.ExpectedVertexCount
= gs_prog_data
->vertices_in
;
4190 gs
.MaximumNumberofThreads
=
4191 GEN_GEN
== 8 ? (devinfo
->max_gs_threads
/ 2 - 1)
4192 : (devinfo
->max_gs_threads
- 1);
4194 if (gs_prog_data
->static_vertex_count
!= -1) {
4195 gs
.StaticOutput
= true;
4196 gs
.StaticOutputVertexCount
= gs_prog_data
->static_vertex_count
;
4198 gs
.IncludeVertexHandles
= vue_prog_data
->include_vue_handles
;
4200 gs
.UserClipDistanceCullTestEnableBitmask
=
4201 vue_prog_data
->cull_distance_mask
;
4203 const int urb_entry_write_offset
= 1;
4204 const uint32_t urb_entry_output_length
=
4205 DIV_ROUND_UP(vue_prog_data
->vue_map
.num_slots
, 2) -
4206 urb_entry_write_offset
;
4208 gs
.VertexURBEntryOutputReadOffset
= urb_entry_write_offset
;
4209 gs
.VertexURBEntryOutputLength
= MAX2(urb_entry_output_length
, 1);
4214 * Encode most of 3DSTATE_PS and 3DSTATE_PS_EXTRA based on the shader.
4217 iris_store_fs_state(struct iris_context
*ice
,
4218 const struct gen_device_info
*devinfo
,
4219 struct iris_compiled_shader
*shader
)
4221 struct brw_stage_prog_data
*prog_data
= shader
->prog_data
;
4222 struct brw_wm_prog_data
*wm_prog_data
= (void *) shader
->prog_data
;
4224 uint32_t *ps_state
= (void *) shader
->derived_data
;
4225 uint32_t *psx_state
= ps_state
+ GENX(3DSTATE_PS_length
);
4227 iris_pack_command(GENX(3DSTATE_PS
), ps_state
, ps
) {
4228 ps
.VectorMaskEnable
= true;
4229 // XXX: WABTPPrefetchDisable, see above, drop at C0
4230 ps
.BindingTableEntryCount
= GEN_GEN
== 11 ? 0 :
4231 shader
->bt
.size_bytes
/ 4;
4232 ps
.FloatingPointMode
= prog_data
->use_alt_mode
;
4233 ps
.MaximumNumberofThreadsPerPSD
= 64 - (GEN_GEN
== 8 ? 2 : 1);
4235 ps
.PushConstantEnable
= prog_data
->ubo_ranges
[0].length
> 0;
4237 /* From the documentation for this packet:
4238 * "If the PS kernel does not need the Position XY Offsets to
4239 * compute a Position Value, then this field should be programmed
4240 * to POSOFFSET_NONE."
4242 * "SW Recommendation: If the PS kernel needs the Position Offsets
4243 * to compute a Position XY value, this field should match Position
4244 * ZW Interpolation Mode to ensure a consistent position.xyzw
4247 * We only require XY sample offsets. So, this recommendation doesn't
4248 * look useful at the moment. We might need this in future.
4250 ps
.PositionXYOffsetSelect
=
4251 wm_prog_data
->uses_pos_offset
? POSOFFSET_SAMPLE
: POSOFFSET_NONE
;
4253 if (prog_data
->total_scratch
) {
4254 struct iris_bo
*bo
=
4255 iris_get_scratch_space(ice
, prog_data
->total_scratch
,
4256 MESA_SHADER_FRAGMENT
);
4257 uint32_t scratch_addr
= bo
->gtt_offset
;
4258 ps
.PerThreadScratchSpace
= ffs(prog_data
->total_scratch
) - 11;
4259 ps
.ScratchSpaceBasePointer
= rw_bo(NULL
, scratch_addr
);
4263 iris_pack_command(GENX(3DSTATE_PS_EXTRA
), psx_state
, psx
) {
4264 psx
.PixelShaderValid
= true;
4265 psx
.PixelShaderComputedDepthMode
= wm_prog_data
->computed_depth_mode
;
4266 psx
.PixelShaderKillsPixel
= wm_prog_data
->uses_kill
;
4267 psx
.AttributeEnable
= wm_prog_data
->num_varying_inputs
!= 0;
4268 psx
.PixelShaderUsesSourceDepth
= wm_prog_data
->uses_src_depth
;
4269 psx
.PixelShaderUsesSourceW
= wm_prog_data
->uses_src_w
;
4270 psx
.PixelShaderIsPerSample
= wm_prog_data
->persample_dispatch
;
4271 psx
.oMaskPresenttoRenderTarget
= wm_prog_data
->uses_omask
;
4274 psx
.PixelShaderPullsBary
= wm_prog_data
->pulls_bary
;
4275 psx
.PixelShaderComputesStencil
= wm_prog_data
->computed_stencil
;
4281 * Compute the size of the derived data (shader command packets).
4283 * This must match the data written by the iris_store_xs_state() functions.
4286 iris_store_cs_state(struct iris_context
*ice
,
4287 const struct gen_device_info
*devinfo
,
4288 struct iris_compiled_shader
*shader
)
4290 struct brw_stage_prog_data
*prog_data
= shader
->prog_data
;
4291 struct brw_cs_prog_data
*cs_prog_data
= (void *) shader
->prog_data
;
4292 void *map
= shader
->derived_data
;
4294 iris_pack_state(GENX(INTERFACE_DESCRIPTOR_DATA
), map
, desc
) {
4295 desc
.KernelStartPointer
= KSP(shader
);
4296 desc
.ConstantURBEntryReadLength
= cs_prog_data
->push
.per_thread
.regs
;
4297 desc
.NumberofThreadsinGPGPUThreadGroup
= cs_prog_data
->threads
;
4298 desc
.SharedLocalMemorySize
=
4299 encode_slm_size(GEN_GEN
, prog_data
->total_shared
);
4300 desc
.BarrierEnable
= cs_prog_data
->uses_barrier
;
4301 desc
.CrossThreadConstantDataReadLength
=
4302 cs_prog_data
->push
.cross_thread
.regs
;
4307 iris_derived_program_state_size(enum iris_program_cache_id cache_id
)
4309 assert(cache_id
<= IRIS_CACHE_BLORP
);
4311 static const unsigned dwords
[] = {
4312 [IRIS_CACHE_VS
] = GENX(3DSTATE_VS_length
),
4313 [IRIS_CACHE_TCS
] = GENX(3DSTATE_HS_length
),
4314 [IRIS_CACHE_TES
] = GENX(3DSTATE_TE_length
) + GENX(3DSTATE_DS_length
),
4315 [IRIS_CACHE_GS
] = GENX(3DSTATE_GS_length
),
4317 GENX(3DSTATE_PS_length
) + GENX(3DSTATE_PS_EXTRA_length
),
4318 [IRIS_CACHE_CS
] = GENX(INTERFACE_DESCRIPTOR_DATA_length
),
4319 [IRIS_CACHE_BLORP
] = 0,
4322 return sizeof(uint32_t) * dwords
[cache_id
];
4326 * Create any state packets corresponding to the given shader stage
4327 * (i.e. 3DSTATE_VS) and save them as "derived data" in the shader variant.
4328 * This means that we can look up a program in the in-memory cache and
4329 * get most of the state packet without having to reconstruct it.
4332 iris_store_derived_program_state(struct iris_context
*ice
,
4333 enum iris_program_cache_id cache_id
,
4334 struct iris_compiled_shader
*shader
)
4336 struct iris_screen
*screen
= (void *) ice
->ctx
.screen
;
4337 const struct gen_device_info
*devinfo
= &screen
->devinfo
;
4341 iris_store_vs_state(ice
, devinfo
, shader
);
4343 case IRIS_CACHE_TCS
:
4344 iris_store_tcs_state(ice
, devinfo
, shader
);
4346 case IRIS_CACHE_TES
:
4347 iris_store_tes_state(ice
, devinfo
, shader
);
4350 iris_store_gs_state(ice
, devinfo
, shader
);
4353 iris_store_fs_state(ice
, devinfo
, shader
);
4356 iris_store_cs_state(ice
, devinfo
, shader
);
4357 case IRIS_CACHE_BLORP
:
4364 /* ------------------------------------------------------------------- */
4366 static const uint32_t push_constant_opcodes
[] = {
4367 [MESA_SHADER_VERTEX
] = 21,
4368 [MESA_SHADER_TESS_CTRL
] = 25, /* HS */
4369 [MESA_SHADER_TESS_EVAL
] = 26, /* DS */
4370 [MESA_SHADER_GEOMETRY
] = 22,
4371 [MESA_SHADER_FRAGMENT
] = 23,
4372 [MESA_SHADER_COMPUTE
] = 0,
4376 use_null_surface(struct iris_batch
*batch
, struct iris_context
*ice
)
4378 struct iris_bo
*state_bo
= iris_resource_bo(ice
->state
.unbound_tex
.res
);
4380 iris_use_pinned_bo(batch
, state_bo
, false);
4382 return ice
->state
.unbound_tex
.offset
;
4386 use_null_fb_surface(struct iris_batch
*batch
, struct iris_context
*ice
)
4388 /* If set_framebuffer_state() was never called, fall back to 1x1x1 */
4389 if (!ice
->state
.null_fb
.res
)
4390 return use_null_surface(batch
, ice
);
4392 struct iris_bo
*state_bo
= iris_resource_bo(ice
->state
.null_fb
.res
);
4394 iris_use_pinned_bo(batch
, state_bo
, false);
4396 return ice
->state
.null_fb
.offset
;
4400 surf_state_offset_for_aux(struct iris_resource
*res
,
4402 enum isl_aux_usage aux_usage
)
4404 return SURFACE_STATE_ALIGNMENT
*
4405 util_bitcount(aux_modes
& ((1 << aux_usage
) - 1));
4410 surf_state_update_clear_value(struct iris_batch
*batch
,
4411 struct iris_resource
*res
,
4412 struct iris_state_ref
*state
,
4414 enum isl_aux_usage aux_usage
)
4416 struct isl_device
*isl_dev
= &batch
->screen
->isl_dev
;
4417 struct iris_bo
*state_bo
= iris_resource_bo(state
->res
);
4418 uint64_t real_offset
= state
->offset
+ IRIS_MEMZONE_BINDER_START
;
4419 uint32_t offset_into_bo
= real_offset
- state_bo
->gtt_offset
;
4420 uint32_t clear_offset
= offset_into_bo
+
4421 isl_dev
->ss
.clear_value_offset
+
4422 surf_state_offset_for_aux(res
, aux_modes
, aux_usage
);
4423 uint32_t *color
= res
->aux
.clear_color
.u32
;
4425 assert(isl_dev
->ss
.clear_value_size
== 16);
4427 if (aux_usage
== ISL_AUX_USAGE_HIZ
) {
4428 iris_emit_pipe_control_write(batch
, "update fast clear value (Z)",
4429 PIPE_CONTROL_WRITE_IMMEDIATE
,
4430 state_bo
, clear_offset
, color
[0]);
4432 iris_emit_pipe_control_write(batch
, "update fast clear color (RG__)",
4433 PIPE_CONTROL_WRITE_IMMEDIATE
,
4434 state_bo
, clear_offset
,
4435 (uint64_t) color
[0] |
4436 (uint64_t) color
[1] << 32);
4437 iris_emit_pipe_control_write(batch
, "update fast clear color (__BA)",
4438 PIPE_CONTROL_WRITE_IMMEDIATE
,
4439 state_bo
, clear_offset
+ 8,
4440 (uint64_t) color
[2] |
4441 (uint64_t) color
[3] << 32);
4444 iris_emit_pipe_control_flush(batch
,
4445 "update fast clear: state cache invalidate",
4446 PIPE_CONTROL_FLUSH_ENABLE
|
4447 PIPE_CONTROL_STATE_CACHE_INVALIDATE
);
4452 update_clear_value(struct iris_context
*ice
,
4453 struct iris_batch
*batch
,
4454 struct iris_resource
*res
,
4455 struct iris_state_ref
*state
,
4456 unsigned all_aux_modes
,
4457 struct isl_view
*view
)
4459 UNUSED
struct isl_device
*isl_dev
= &batch
->screen
->isl_dev
;
4460 UNUSED
unsigned aux_modes
= all_aux_modes
;
4462 /* We only need to update the clear color in the surface state for gen8 and
4463 * gen9. Newer gens can read it directly from the clear color state buffer.
4466 /* Skip updating the ISL_AUX_USAGE_NONE surface state */
4467 aux_modes
&= ~(1 << ISL_AUX_USAGE_NONE
);
4470 enum isl_aux_usage aux_usage
= u_bit_scan(&aux_modes
);
4472 surf_state_update_clear_value(batch
, res
, state
, all_aux_modes
,
4476 pipe_resource_reference(&state
->res
, NULL
);
4478 void *map
= alloc_surface_states(ice
->state
.surface_uploader
,
4479 state
, all_aux_modes
);
4481 enum isl_aux_usage aux_usage
= u_bit_scan(&aux_modes
);
4482 fill_surface_state(isl_dev
, map
, res
, &res
->surf
, view
, aux_usage
, 0, 0);
4483 map
+= SURFACE_STATE_ALIGNMENT
;
4489 * Add a surface to the validation list, as well as the buffer containing
4490 * the corresponding SURFACE_STATE.
4492 * Returns the binding table entry (offset to SURFACE_STATE).
4495 use_surface(struct iris_context
*ice
,
4496 struct iris_batch
*batch
,
4497 struct pipe_surface
*p_surf
,
4499 enum isl_aux_usage aux_usage
,
4500 bool is_read_surface
)
4502 struct iris_surface
*surf
= (void *) p_surf
;
4503 struct iris_resource
*res
= (void *) p_surf
->texture
;
4504 uint32_t offset
= 0;
4506 iris_use_pinned_bo(batch
, iris_resource_bo(p_surf
->texture
), writeable
);
4507 if (GEN_GEN
== 8 && is_read_surface
) {
4508 iris_use_pinned_bo(batch
, iris_resource_bo(surf
->surface_state_read
.res
), false);
4510 iris_use_pinned_bo(batch
, iris_resource_bo(surf
->surface_state
.res
), false);
4514 iris_use_pinned_bo(batch
, res
->aux
.bo
, writeable
);
4515 if (res
->aux
.clear_color_bo
)
4516 iris_use_pinned_bo(batch
, res
->aux
.clear_color_bo
, false);
4518 if (memcmp(&res
->aux
.clear_color
, &surf
->clear_color
,
4519 sizeof(surf
->clear_color
)) != 0) {
4520 update_clear_value(ice
, batch
, res
, &surf
->surface_state
,
4521 res
->aux
.possible_usages
, &surf
->view
);
4523 update_clear_value(ice
, batch
, res
, &surf
->surface_state_read
,
4524 res
->aux
.possible_usages
, &surf
->read_view
);
4526 surf
->clear_color
= res
->aux
.clear_color
;
4530 offset
= (GEN_GEN
== 8 && is_read_surface
) ? surf
->surface_state_read
.offset
4531 : surf
->surface_state
.offset
;
4534 surf_state_offset_for_aux(res
, res
->aux
.possible_usages
, aux_usage
);
4538 use_sampler_view(struct iris_context
*ice
,
4539 struct iris_batch
*batch
,
4540 struct iris_sampler_view
*isv
)
4543 enum isl_aux_usage aux_usage
=
4544 iris_resource_texture_aux_usage(ice
, isv
->res
, isv
->view
.format
, 0);
4546 iris_use_pinned_bo(batch
, isv
->res
->bo
, false);
4547 iris_use_pinned_bo(batch
, iris_resource_bo(isv
->surface_state
.res
), false);
4549 if (isv
->res
->aux
.bo
) {
4550 iris_use_pinned_bo(batch
, isv
->res
->aux
.bo
, false);
4551 if (isv
->res
->aux
.clear_color_bo
)
4552 iris_use_pinned_bo(batch
, isv
->res
->aux
.clear_color_bo
, false);
4553 if (memcmp(&isv
->res
->aux
.clear_color
, &isv
->clear_color
,
4554 sizeof(isv
->clear_color
)) != 0) {
4555 update_clear_value(ice
, batch
, isv
->res
, &isv
->surface_state
,
4556 isv
->res
->aux
.sampler_usages
, &isv
->view
);
4557 isv
->clear_color
= isv
->res
->aux
.clear_color
;
4561 return isv
->surface_state
.offset
+
4562 surf_state_offset_for_aux(isv
->res
, isv
->res
->aux
.sampler_usages
,
4567 use_ubo_ssbo(struct iris_batch
*batch
,
4568 struct iris_context
*ice
,
4569 struct pipe_shader_buffer
*buf
,
4570 struct iris_state_ref
*surf_state
,
4573 if (!buf
->buffer
|| !surf_state
->res
)
4574 return use_null_surface(batch
, ice
);
4576 iris_use_pinned_bo(batch
, iris_resource_bo(buf
->buffer
), writable
);
4577 iris_use_pinned_bo(batch
, iris_resource_bo(surf_state
->res
), false);
4579 return surf_state
->offset
;
4583 use_image(struct iris_batch
*batch
, struct iris_context
*ice
,
4584 struct iris_shader_state
*shs
, int i
)
4586 struct iris_image_view
*iv
= &shs
->image
[i
];
4587 struct iris_resource
*res
= (void *) iv
->base
.resource
;
4590 return use_null_surface(batch
, ice
);
4592 bool write
= iv
->base
.shader_access
& PIPE_IMAGE_ACCESS_WRITE
;
4594 iris_use_pinned_bo(batch
, res
->bo
, write
);
4595 iris_use_pinned_bo(batch
, iris_resource_bo(iv
->surface_state
.res
), false);
4598 iris_use_pinned_bo(batch
, res
->aux
.bo
, write
);
4600 return iv
->surface_state
.offset
;
4603 #define push_bt_entry(addr) \
4604 assert(addr >= binder_addr); \
4605 assert(s < shader->bt.size_bytes / sizeof(uint32_t)); \
4606 if (!pin_only) bt_map[s++] = (addr) - binder_addr;
4608 #define bt_assert(section) \
4609 if (!pin_only && shader->bt.used_mask[section] != 0) \
4610 assert(shader->bt.offsets[section] == s);
4613 * Populate the binding table for a given shader stage.
4615 * This fills out the table of pointers to surfaces required by the shader,
4616 * and also adds those buffers to the validation list so the kernel can make
4617 * resident before running our batch.
4620 iris_populate_binding_table(struct iris_context
*ice
,
4621 struct iris_batch
*batch
,
4622 gl_shader_stage stage
,
4625 const struct iris_binder
*binder
= &ice
->state
.binder
;
4626 struct iris_uncompiled_shader
*ish
= ice
->shaders
.uncompiled
[stage
];
4627 struct iris_compiled_shader
*shader
= ice
->shaders
.prog
[stage
];
4631 struct iris_binding_table
*bt
= &shader
->bt
;
4632 UNUSED
struct brw_stage_prog_data
*prog_data
= shader
->prog_data
;
4633 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
4634 uint32_t binder_addr
= binder
->bo
->gtt_offset
;
4636 uint32_t *bt_map
= binder
->map
+ binder
->bt_offset
[stage
];
4639 const struct shader_info
*info
= iris_get_shader_info(ice
, stage
);
4641 /* TCS passthrough doesn't need a binding table. */
4642 assert(stage
== MESA_SHADER_TESS_CTRL
);
4646 if (stage
== MESA_SHADER_COMPUTE
&&
4647 shader
->bt
.used_mask
[IRIS_SURFACE_GROUP_CS_WORK_GROUPS
]) {
4648 /* surface for gl_NumWorkGroups */
4649 struct iris_state_ref
*grid_data
= &ice
->state
.grid_size
;
4650 struct iris_state_ref
*grid_state
= &ice
->state
.grid_surf_state
;
4651 iris_use_pinned_bo(batch
, iris_resource_bo(grid_data
->res
), false);
4652 iris_use_pinned_bo(batch
, iris_resource_bo(grid_state
->res
), false);
4653 push_bt_entry(grid_state
->offset
);
4656 if (stage
== MESA_SHADER_FRAGMENT
) {
4657 struct pipe_framebuffer_state
*cso_fb
= &ice
->state
.framebuffer
;
4658 /* Note that cso_fb->nr_cbufs == fs_key->nr_color_regions. */
4659 if (cso_fb
->nr_cbufs
) {
4660 for (unsigned i
= 0; i
< cso_fb
->nr_cbufs
; i
++) {
4662 if (cso_fb
->cbufs
[i
]) {
4663 addr
= use_surface(ice
, batch
, cso_fb
->cbufs
[i
], true,
4664 ice
->state
.draw_aux_usage
[i
], false);
4666 addr
= use_null_fb_surface(batch
, ice
);
4668 push_bt_entry(addr
);
4670 } else if (GEN_GEN
< 11) {
4671 uint32_t addr
= use_null_fb_surface(batch
, ice
);
4672 push_bt_entry(addr
);
4676 #define foreach_surface_used(index, group) \
4678 for (int index = 0; index < bt->sizes[group]; index++) \
4679 if (iris_group_index_to_bti(bt, group, index) != \
4680 IRIS_SURFACE_NOT_USED)
4682 foreach_surface_used(i
, IRIS_SURFACE_GROUP_RENDER_TARGET_READ
) {
4683 struct pipe_framebuffer_state
*cso_fb
= &ice
->state
.framebuffer
;
4685 if (cso_fb
->cbufs
[i
]) {
4686 addr
= use_surface(ice
, batch
, cso_fb
->cbufs
[i
],
4687 true, ice
->state
.draw_aux_usage
[i
], true);
4688 push_bt_entry(addr
);
4692 foreach_surface_used(i
, IRIS_SURFACE_GROUP_TEXTURE
) {
4693 struct iris_sampler_view
*view
= shs
->textures
[i
];
4694 uint32_t addr
= view
? use_sampler_view(ice
, batch
, view
)
4695 : use_null_surface(batch
, ice
);
4696 push_bt_entry(addr
);
4699 foreach_surface_used(i
, IRIS_SURFACE_GROUP_IMAGE
) {
4700 uint32_t addr
= use_image(batch
, ice
, shs
, i
);
4701 push_bt_entry(addr
);
4704 foreach_surface_used(i
, IRIS_SURFACE_GROUP_UBO
) {
4707 if (i
== bt
->sizes
[IRIS_SURFACE_GROUP_UBO
] - 1) {
4708 if (ish
->const_data
) {
4709 iris_use_pinned_bo(batch
, iris_resource_bo(ish
->const_data
), false);
4710 iris_use_pinned_bo(batch
, iris_resource_bo(ish
->const_data_state
.res
),
4712 addr
= ish
->const_data_state
.offset
;
4714 /* This can only happen with INTEL_DISABLE_COMPACT_BINDING_TABLE=1. */
4715 addr
= use_null_surface(batch
, ice
);
4718 addr
= use_ubo_ssbo(batch
, ice
, &shs
->constbuf
[i
],
4719 &shs
->constbuf_surf_state
[i
], false);
4722 push_bt_entry(addr
);
4725 foreach_surface_used(i
, IRIS_SURFACE_GROUP_SSBO
) {
4727 use_ubo_ssbo(batch
, ice
, &shs
->ssbo
[i
], &shs
->ssbo_surf_state
[i
],
4728 shs
->writable_ssbos
& (1u << i
));
4729 push_bt_entry(addr
);
4733 /* XXX: YUV surfaces not implemented yet */
4734 bt_assert(plane_start
[1], ...);
4735 bt_assert(plane_start
[2], ...);
4740 iris_use_optional_res(struct iris_batch
*batch
,
4741 struct pipe_resource
*res
,
4745 struct iris_bo
*bo
= iris_resource_bo(res
);
4746 iris_use_pinned_bo(batch
, bo
, writeable
);
4751 pin_depth_and_stencil_buffers(struct iris_batch
*batch
,
4752 struct pipe_surface
*zsbuf
,
4753 struct iris_depth_stencil_alpha_state
*cso_zsa
)
4758 struct iris_resource
*zres
, *sres
;
4759 iris_get_depth_stencil_resources(zsbuf
->texture
, &zres
, &sres
);
4762 iris_use_pinned_bo(batch
, zres
->bo
, cso_zsa
->depth_writes_enabled
);
4764 iris_use_pinned_bo(batch
, zres
->aux
.bo
,
4765 cso_zsa
->depth_writes_enabled
);
4770 iris_use_pinned_bo(batch
, sres
->bo
, cso_zsa
->stencil_writes_enabled
);
4774 /* ------------------------------------------------------------------- */
4777 * Pin any BOs which were installed by a previous batch, and restored
4778 * via the hardware logical context mechanism.
4780 * We don't need to re-emit all state every batch - the hardware context
4781 * mechanism will save and restore it for us. This includes pointers to
4782 * various BOs...which won't exist unless we ask the kernel to pin them
4783 * by adding them to the validation list.
4785 * We can skip buffers if we've re-emitted those packets, as we're
4786 * overwriting those stale pointers with new ones, and don't actually
4787 * refer to the old BOs.
4790 iris_restore_render_saved_bos(struct iris_context
*ice
,
4791 struct iris_batch
*batch
,
4792 const struct pipe_draw_info
*draw
)
4794 struct iris_genx_state
*genx
= ice
->state
.genx
;
4796 const uint64_t clean
= ~ice
->state
.dirty
;
4798 if (clean
& IRIS_DIRTY_CC_VIEWPORT
) {
4799 iris_use_optional_res(batch
, ice
->state
.last_res
.cc_vp
, false);
4802 if (clean
& IRIS_DIRTY_SF_CL_VIEWPORT
) {
4803 iris_use_optional_res(batch
, ice
->state
.last_res
.sf_cl_vp
, false);
4806 if (clean
& IRIS_DIRTY_BLEND_STATE
) {
4807 iris_use_optional_res(batch
, ice
->state
.last_res
.blend
, false);
4810 if (clean
& IRIS_DIRTY_COLOR_CALC_STATE
) {
4811 iris_use_optional_res(batch
, ice
->state
.last_res
.color_calc
, false);
4814 if (clean
& IRIS_DIRTY_SCISSOR_RECT
) {
4815 iris_use_optional_res(batch
, ice
->state
.last_res
.scissor
, false);
4818 if (ice
->state
.streamout_active
&& (clean
& IRIS_DIRTY_SO_BUFFERS
)) {
4819 for (int i
= 0; i
< 4; i
++) {
4820 struct iris_stream_output_target
*tgt
=
4821 (void *) ice
->state
.so_target
[i
];
4823 iris_use_pinned_bo(batch
, iris_resource_bo(tgt
->base
.buffer
),
4825 iris_use_pinned_bo(batch
, iris_resource_bo(tgt
->offset
.res
),
4831 for (int stage
= 0; stage
<= MESA_SHADER_FRAGMENT
; stage
++) {
4832 if (!(clean
& (IRIS_DIRTY_CONSTANTS_VS
<< stage
)))
4835 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
4836 struct iris_compiled_shader
*shader
= ice
->shaders
.prog
[stage
];
4841 struct brw_stage_prog_data
*prog_data
= (void *) shader
->prog_data
;
4843 for (int i
= 0; i
< 4; i
++) {
4844 const struct brw_ubo_range
*range
= &prog_data
->ubo_ranges
[i
];
4846 if (range
->length
== 0)
4849 /* Range block is a binding table index, map back to UBO index. */
4850 unsigned block_index
= iris_bti_to_group_index(
4851 &shader
->bt
, IRIS_SURFACE_GROUP_UBO
, range
->block
);
4852 assert(block_index
!= IRIS_SURFACE_NOT_USED
);
4854 struct pipe_shader_buffer
*cbuf
= &shs
->constbuf
[block_index
];
4855 struct iris_resource
*res
= (void *) cbuf
->buffer
;
4858 iris_use_pinned_bo(batch
, res
->bo
, false);
4860 iris_use_pinned_bo(batch
, batch
->screen
->workaround_bo
, false);
4864 for (int stage
= 0; stage
<= MESA_SHADER_FRAGMENT
; stage
++) {
4865 if (clean
& (IRIS_DIRTY_BINDINGS_VS
<< stage
)) {
4866 /* Re-pin any buffers referred to by the binding table. */
4867 iris_populate_binding_table(ice
, batch
, stage
, true);
4871 for (int stage
= 0; stage
<= MESA_SHADER_FRAGMENT
; stage
++) {
4872 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
4873 struct pipe_resource
*res
= shs
->sampler_table
.res
;
4875 iris_use_pinned_bo(batch
, iris_resource_bo(res
), false);
4878 for (int stage
= 0; stage
<= MESA_SHADER_FRAGMENT
; stage
++) {
4879 if (clean
& (IRIS_DIRTY_VS
<< stage
)) {
4880 struct iris_compiled_shader
*shader
= ice
->shaders
.prog
[stage
];
4883 struct iris_bo
*bo
= iris_resource_bo(shader
->assembly
.res
);
4884 iris_use_pinned_bo(batch
, bo
, false);
4886 struct brw_stage_prog_data
*prog_data
= shader
->prog_data
;
4888 if (prog_data
->total_scratch
> 0) {
4889 struct iris_bo
*bo
=
4890 iris_get_scratch_space(ice
, prog_data
->total_scratch
, stage
);
4891 iris_use_pinned_bo(batch
, bo
, true);
4897 if ((clean
& IRIS_DIRTY_DEPTH_BUFFER
) &&
4898 (clean
& IRIS_DIRTY_WM_DEPTH_STENCIL
)) {
4899 struct pipe_framebuffer_state
*cso_fb
= &ice
->state
.framebuffer
;
4900 pin_depth_and_stencil_buffers(batch
, cso_fb
->zsbuf
, ice
->state
.cso_zsa
);
4903 iris_use_optional_res(batch
, ice
->state
.last_res
.index_buffer
, false);
4905 if (clean
& IRIS_DIRTY_VERTEX_BUFFERS
) {
4906 uint64_t bound
= ice
->state
.bound_vertex_buffers
;
4908 const int i
= u_bit_scan64(&bound
);
4909 struct pipe_resource
*res
= genx
->vertex_buffers
[i
].resource
;
4910 iris_use_pinned_bo(batch
, iris_resource_bo(res
), false);
4916 iris_restore_compute_saved_bos(struct iris_context
*ice
,
4917 struct iris_batch
*batch
,
4918 const struct pipe_grid_info
*grid
)
4920 const uint64_t clean
= ~ice
->state
.dirty
;
4922 const int stage
= MESA_SHADER_COMPUTE
;
4923 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
4925 if (clean
& IRIS_DIRTY_BINDINGS_CS
) {
4926 /* Re-pin any buffers referred to by the binding table. */
4927 iris_populate_binding_table(ice
, batch
, stage
, true);
4930 struct pipe_resource
*sampler_res
= shs
->sampler_table
.res
;
4932 iris_use_pinned_bo(batch
, iris_resource_bo(sampler_res
), false);
4934 if ((clean
& IRIS_DIRTY_SAMPLER_STATES_CS
) &&
4935 (clean
& IRIS_DIRTY_BINDINGS_CS
) &&
4936 (clean
& IRIS_DIRTY_CONSTANTS_CS
) &&
4937 (clean
& IRIS_DIRTY_CS
)) {
4938 iris_use_optional_res(batch
, ice
->state
.last_res
.cs_desc
, false);
4941 if (clean
& IRIS_DIRTY_CS
) {
4942 struct iris_compiled_shader
*shader
= ice
->shaders
.prog
[stage
];
4945 struct iris_bo
*bo
= iris_resource_bo(shader
->assembly
.res
);
4946 iris_use_pinned_bo(batch
, bo
, false);
4948 struct iris_bo
*curbe_bo
=
4949 iris_resource_bo(ice
->state
.last_res
.cs_thread_ids
);
4950 iris_use_pinned_bo(batch
, curbe_bo
, false);
4952 struct brw_stage_prog_data
*prog_data
= shader
->prog_data
;
4954 if (prog_data
->total_scratch
> 0) {
4955 struct iris_bo
*bo
=
4956 iris_get_scratch_space(ice
, prog_data
->total_scratch
, stage
);
4957 iris_use_pinned_bo(batch
, bo
, true);
4964 * Possibly emit STATE_BASE_ADDRESS to update Surface State Base Address.
4967 iris_update_surface_base_address(struct iris_batch
*batch
,
4968 struct iris_binder
*binder
)
4970 if (batch
->last_surface_base_address
== binder
->bo
->gtt_offset
)
4973 flush_before_state_base_change(batch
);
4975 iris_emit_cmd(batch
, GENX(STATE_BASE_ADDRESS
), sba
) {
4976 sba
.SurfaceStateBaseAddressModifyEnable
= true;
4977 sba
.SurfaceStateBaseAddress
= ro_bo(binder
->bo
, 0);
4979 /* The hardware appears to pay attention to the MOCS fields even
4980 * if you don't set the "Address Modify Enable" bit for the base.
4982 sba
.GeneralStateMOCS
= MOCS_WB
;
4983 sba
.StatelessDataPortAccessMOCS
= MOCS_WB
;
4984 sba
.DynamicStateMOCS
= MOCS_WB
;
4985 sba
.IndirectObjectMOCS
= MOCS_WB
;
4986 sba
.InstructionMOCS
= MOCS_WB
;
4987 sba
.SurfaceStateMOCS
= MOCS_WB
;
4989 sba
.BindlessSurfaceStateMOCS
= MOCS_WB
;
4993 flush_after_state_base_change(batch
);
4995 batch
->last_surface_base_address
= binder
->bo
->gtt_offset
;
4999 iris_viewport_zmin_zmax(const struct pipe_viewport_state
*vp
, bool halfz
,
5000 bool window_space_position
, float *zmin
, float *zmax
)
5002 if (window_space_position
) {
5007 util_viewport_zmin_zmax(vp
, halfz
, zmin
, zmax
);
5012 genX(emit_aux_map_state
)(struct iris_batch
*batch
)
5014 struct iris_screen
*screen
= batch
->screen
;
5015 void *aux_map_ctx
= iris_bufmgr_get_aux_map_context(screen
->bufmgr
);
5018 uint32_t aux_map_state_num
= gen_aux_map_get_state_num(aux_map_ctx
);
5019 if (batch
->last_aux_map_state
!= aux_map_state_num
) {
5020 /* If the aux-map state number increased, then we need to rewrite the
5021 * register. Rewriting the register is used to both set the aux-map
5022 * translation table address, and also to invalidate any previously
5023 * cached translations.
5025 uint64_t base_addr
= gen_aux_map_get_base(aux_map_ctx
);
5026 assert(base_addr
!= 0 && ALIGN(base_addr
, 32 * 1024) == base_addr
);
5027 iris_load_register_imm64(batch
, GENX(GFX_AUX_TABLE_BASE_ADDR_num
),
5029 batch
->last_aux_map_state
= aux_map_state_num
;
5035 iris_upload_dirty_render_state(struct iris_context
*ice
,
5036 struct iris_batch
*batch
,
5037 const struct pipe_draw_info
*draw
)
5039 const uint64_t dirty
= ice
->state
.dirty
;
5041 if (!(dirty
& IRIS_ALL_DIRTY_FOR_RENDER
))
5044 struct iris_genx_state
*genx
= ice
->state
.genx
;
5045 struct iris_binder
*binder
= &ice
->state
.binder
;
5046 struct brw_wm_prog_data
*wm_prog_data
= (void *)
5047 ice
->shaders
.prog
[MESA_SHADER_FRAGMENT
]->prog_data
;
5049 if (dirty
& IRIS_DIRTY_CC_VIEWPORT
) {
5050 const struct iris_rasterizer_state
*cso_rast
= ice
->state
.cso_rast
;
5051 uint32_t cc_vp_address
;
5053 /* XXX: could avoid streaming for depth_clip [0,1] case. */
5054 uint32_t *cc_vp_map
=
5055 stream_state(batch
, ice
->state
.dynamic_uploader
,
5056 &ice
->state
.last_res
.cc_vp
,
5057 4 * ice
->state
.num_viewports
*
5058 GENX(CC_VIEWPORT_length
), 32, &cc_vp_address
);
5059 for (int i
= 0; i
< ice
->state
.num_viewports
; i
++) {
5061 iris_viewport_zmin_zmax(&ice
->state
.viewports
[i
], cso_rast
->clip_halfz
,
5062 ice
->state
.window_space_position
,
5064 if (cso_rast
->depth_clip_near
)
5066 if (cso_rast
->depth_clip_far
)
5069 iris_pack_state(GENX(CC_VIEWPORT
), cc_vp_map
, ccv
) {
5070 ccv
.MinimumDepth
= zmin
;
5071 ccv
.MaximumDepth
= zmax
;
5074 cc_vp_map
+= GENX(CC_VIEWPORT_length
);
5077 iris_emit_cmd(batch
, GENX(3DSTATE_VIEWPORT_STATE_POINTERS_CC
), ptr
) {
5078 ptr
.CCViewportPointer
= cc_vp_address
;
5082 if (dirty
& IRIS_DIRTY_SF_CL_VIEWPORT
) {
5083 struct pipe_framebuffer_state
*cso_fb
= &ice
->state
.framebuffer
;
5084 uint32_t sf_cl_vp_address
;
5086 stream_state(batch
, ice
->state
.dynamic_uploader
,
5087 &ice
->state
.last_res
.sf_cl_vp
,
5088 4 * ice
->state
.num_viewports
*
5089 GENX(SF_CLIP_VIEWPORT_length
), 64, &sf_cl_vp_address
);
5091 for (unsigned i
= 0; i
< ice
->state
.num_viewports
; i
++) {
5092 const struct pipe_viewport_state
*state
= &ice
->state
.viewports
[i
];
5093 float gb_xmin
, gb_xmax
, gb_ymin
, gb_ymax
;
5095 float vp_xmin
= viewport_extent(state
, 0, -1.0f
);
5096 float vp_xmax
= viewport_extent(state
, 0, 1.0f
);
5097 float vp_ymin
= viewport_extent(state
, 1, -1.0f
);
5098 float vp_ymax
= viewport_extent(state
, 1, 1.0f
);
5100 gen_calculate_guardband_size(cso_fb
->width
, cso_fb
->height
,
5101 state
->scale
[0], state
->scale
[1],
5102 state
->translate
[0], state
->translate
[1],
5103 &gb_xmin
, &gb_xmax
, &gb_ymin
, &gb_ymax
);
5105 iris_pack_state(GENX(SF_CLIP_VIEWPORT
), vp_map
, vp
) {
5106 vp
.ViewportMatrixElementm00
= state
->scale
[0];
5107 vp
.ViewportMatrixElementm11
= state
->scale
[1];
5108 vp
.ViewportMatrixElementm22
= state
->scale
[2];
5109 vp
.ViewportMatrixElementm30
= state
->translate
[0];
5110 vp
.ViewportMatrixElementm31
= state
->translate
[1];
5111 vp
.ViewportMatrixElementm32
= state
->translate
[2];
5112 vp
.XMinClipGuardband
= gb_xmin
;
5113 vp
.XMaxClipGuardband
= gb_xmax
;
5114 vp
.YMinClipGuardband
= gb_ymin
;
5115 vp
.YMaxClipGuardband
= gb_ymax
;
5116 vp
.XMinViewPort
= MAX2(vp_xmin
, 0);
5117 vp
.XMaxViewPort
= MIN2(vp_xmax
, cso_fb
->width
) - 1;
5118 vp
.YMinViewPort
= MAX2(vp_ymin
, 0);
5119 vp
.YMaxViewPort
= MIN2(vp_ymax
, cso_fb
->height
) - 1;
5122 vp_map
+= GENX(SF_CLIP_VIEWPORT_length
);
5125 iris_emit_cmd(batch
, GENX(3DSTATE_VIEWPORT_STATE_POINTERS_SF_CLIP
), ptr
) {
5126 ptr
.SFClipViewportPointer
= sf_cl_vp_address
;
5130 if (dirty
& IRIS_DIRTY_URB
) {
5133 for (int i
= MESA_SHADER_VERTEX
; i
<= MESA_SHADER_GEOMETRY
; i
++) {
5134 if (!ice
->shaders
.prog
[i
]) {
5137 struct brw_vue_prog_data
*vue_prog_data
=
5138 (void *) ice
->shaders
.prog
[i
]->prog_data
;
5139 size
[i
] = vue_prog_data
->urb_entry_size
;
5141 assert(size
[i
] != 0);
5144 genX(emit_urb_setup
)(ice
, batch
, size
,
5145 ice
->shaders
.prog
[MESA_SHADER_TESS_EVAL
] != NULL
,
5146 ice
->shaders
.prog
[MESA_SHADER_GEOMETRY
] != NULL
);
5149 if (dirty
& IRIS_DIRTY_BLEND_STATE
) {
5150 struct iris_blend_state
*cso_blend
= ice
->state
.cso_blend
;
5151 struct pipe_framebuffer_state
*cso_fb
= &ice
->state
.framebuffer
;
5152 struct iris_depth_stencil_alpha_state
*cso_zsa
= ice
->state
.cso_zsa
;
5153 const int header_dwords
= GENX(BLEND_STATE_length
);
5155 /* Always write at least one BLEND_STATE - the final RT message will
5156 * reference BLEND_STATE[0] even if there aren't color writes. There
5157 * may still be alpha testing, computed depth, and so on.
5159 const int rt_dwords
=
5160 MAX2(cso_fb
->nr_cbufs
, 1) * GENX(BLEND_STATE_ENTRY_length
);
5162 uint32_t blend_offset
;
5163 uint32_t *blend_map
=
5164 stream_state(batch
, ice
->state
.dynamic_uploader
,
5165 &ice
->state
.last_res
.blend
,
5166 4 * (header_dwords
+ rt_dwords
), 64, &blend_offset
);
5168 uint32_t blend_state_header
;
5169 iris_pack_state(GENX(BLEND_STATE
), &blend_state_header
, bs
) {
5170 bs
.AlphaTestEnable
= cso_zsa
->alpha
.enabled
;
5171 bs
.AlphaTestFunction
= translate_compare_func(cso_zsa
->alpha
.func
);
5174 blend_map
[0] = blend_state_header
| cso_blend
->blend_state
[0];
5175 memcpy(&blend_map
[1], &cso_blend
->blend_state
[1], 4 * rt_dwords
);
5177 iris_emit_cmd(batch
, GENX(3DSTATE_BLEND_STATE_POINTERS
), ptr
) {
5178 ptr
.BlendStatePointer
= blend_offset
;
5179 ptr
.BlendStatePointerValid
= true;
5183 if (dirty
& IRIS_DIRTY_COLOR_CALC_STATE
) {
5184 struct iris_depth_stencil_alpha_state
*cso
= ice
->state
.cso_zsa
;
5186 struct pipe_stencil_ref
*p_stencil_refs
= &ice
->state
.stencil_ref
;
5190 stream_state(batch
, ice
->state
.dynamic_uploader
,
5191 &ice
->state
.last_res
.color_calc
,
5192 sizeof(uint32_t) * GENX(COLOR_CALC_STATE_length
),
5194 iris_pack_state(GENX(COLOR_CALC_STATE
), cc_map
, cc
) {
5195 cc
.AlphaTestFormat
= ALPHATEST_FLOAT32
;
5196 cc
.AlphaReferenceValueAsFLOAT32
= cso
->alpha
.ref_value
;
5197 cc
.BlendConstantColorRed
= ice
->state
.blend_color
.color
[0];
5198 cc
.BlendConstantColorGreen
= ice
->state
.blend_color
.color
[1];
5199 cc
.BlendConstantColorBlue
= ice
->state
.blend_color
.color
[2];
5200 cc
.BlendConstantColorAlpha
= ice
->state
.blend_color
.color
[3];
5202 cc
.StencilReferenceValue
= p_stencil_refs
->ref_value
[0];
5203 cc
.BackfaceStencilReferenceValue
= p_stencil_refs
->ref_value
[1];
5206 iris_emit_cmd(batch
, GENX(3DSTATE_CC_STATE_POINTERS
), ptr
) {
5207 ptr
.ColorCalcStatePointer
= cc_offset
;
5208 ptr
.ColorCalcStatePointerValid
= true;
5212 for (int stage
= 0; stage
<= MESA_SHADER_FRAGMENT
; stage
++) {
5213 if (!(dirty
& (IRIS_DIRTY_CONSTANTS_VS
<< stage
)))
5216 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
5217 struct iris_compiled_shader
*shader
= ice
->shaders
.prog
[stage
];
5222 if (shs
->sysvals_need_upload
)
5223 upload_sysvals(ice
, stage
);
5225 struct brw_stage_prog_data
*prog_data
= (void *) shader
->prog_data
;
5227 iris_emit_cmd(batch
, GENX(3DSTATE_CONSTANT_VS
), pkt
) {
5228 pkt
._3DCommandSubOpcode
= push_constant_opcodes
[stage
];
5230 /* The Skylake PRM contains the following restriction:
5232 * "The driver must ensure The following case does not occur
5233 * without a flush to the 3D engine: 3DSTATE_CONSTANT_* with
5234 * buffer 3 read length equal to zero committed followed by a
5235 * 3DSTATE_CONSTANT_* with buffer 0 read length not equal to
5238 * To avoid this, we program the buffers in the highest slots.
5239 * This way, slot 0 is only used if slot 3 is also used.
5243 for (int i
= 3; i
>= 0; i
--) {
5244 const struct brw_ubo_range
*range
= &prog_data
->ubo_ranges
[i
];
5246 if (range
->length
== 0)
5249 /* Range block is a binding table index, map back to UBO index. */
5250 unsigned block_index
= iris_bti_to_group_index(
5251 &shader
->bt
, IRIS_SURFACE_GROUP_UBO
, range
->block
);
5252 assert(block_index
!= IRIS_SURFACE_NOT_USED
);
5254 struct pipe_shader_buffer
*cbuf
= &shs
->constbuf
[block_index
];
5255 struct iris_resource
*res
= (void *) cbuf
->buffer
;
5257 assert(cbuf
->buffer_offset
% 32 == 0);
5259 pkt
.ConstantBody
.ReadLength
[n
] = range
->length
;
5260 pkt
.ConstantBody
.Buffer
[n
] =
5261 res
? ro_bo(res
->bo
, range
->start
* 32 + cbuf
->buffer_offset
)
5262 : ro_bo(batch
->screen
->workaround_bo
, 0);
5269 for (int stage
= 0; stage
<= MESA_SHADER_FRAGMENT
; stage
++) {
5270 /* Gen9 requires 3DSTATE_BINDING_TABLE_POINTERS_XS to be re-emitted
5271 * in order to commit constants. TODO: Investigate "Disable Gather
5272 * at Set Shader" to go back to legacy mode...
5274 if (dirty
& ((IRIS_DIRTY_BINDINGS_VS
|
5275 (GEN_GEN
== 9 ? IRIS_DIRTY_CONSTANTS_VS
: 0)) << stage
)) {
5276 iris_emit_cmd(batch
, GENX(3DSTATE_BINDING_TABLE_POINTERS_VS
), ptr
) {
5277 ptr
._3DCommandSubOpcode
= 38 + stage
;
5278 ptr
.PointertoVSBindingTable
= binder
->bt_offset
[stage
];
5283 if (GEN_GEN
>= 11 && (dirty
& IRIS_DIRTY_RENDER_BUFFER
)) {
5284 // XXX: we may want to flag IRIS_DIRTY_MULTISAMPLE (or SAMPLE_MASK?)
5285 // XXX: see commit 979fc1bc9bcc64027ff2cfafd285676f31b930a6
5287 /* The PIPE_CONTROL command description says:
5289 * "Whenever a Binding Table Index (BTI) used by a Render Target
5290 * Message points to a different RENDER_SURFACE_STATE, SW must issue a
5291 * Render Target Cache Flush by enabling this bit. When render target
5292 * flush is set due to new association of BTI, PS Scoreboard Stall bit
5293 * must be set in this packet."
5295 // XXX: does this need to happen at 3DSTATE_BTP_PS time?
5296 iris_emit_pipe_control_flush(batch
, "workaround: RT BTI change [draw]",
5297 PIPE_CONTROL_RENDER_TARGET_FLUSH
|
5298 PIPE_CONTROL_STALL_AT_SCOREBOARD
);
5301 for (int stage
= 0; stage
<= MESA_SHADER_FRAGMENT
; stage
++) {
5302 if (dirty
& (IRIS_DIRTY_BINDINGS_VS
<< stage
)) {
5303 iris_populate_binding_table(ice
, batch
, stage
, false);
5307 for (int stage
= 0; stage
<= MESA_SHADER_FRAGMENT
; stage
++) {
5308 if (!(dirty
& (IRIS_DIRTY_SAMPLER_STATES_VS
<< stage
)) ||
5309 !ice
->shaders
.prog
[stage
])
5312 iris_upload_sampler_states(ice
, stage
);
5314 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
5315 struct pipe_resource
*res
= shs
->sampler_table
.res
;
5317 iris_use_pinned_bo(batch
, iris_resource_bo(res
), false);
5319 iris_emit_cmd(batch
, GENX(3DSTATE_SAMPLER_STATE_POINTERS_VS
), ptr
) {
5320 ptr
._3DCommandSubOpcode
= 43 + stage
;
5321 ptr
.PointertoVSSamplerState
= shs
->sampler_table
.offset
;
5325 if (ice
->state
.need_border_colors
)
5326 iris_use_pinned_bo(batch
, ice
->state
.border_color_pool
.bo
, false);
5328 if (dirty
& IRIS_DIRTY_MULTISAMPLE
) {
5329 iris_emit_cmd(batch
, GENX(3DSTATE_MULTISAMPLE
), ms
) {
5331 ice
->state
.cso_rast
->half_pixel_center
? CENTER
: UL_CORNER
;
5332 if (ice
->state
.framebuffer
.samples
> 0)
5333 ms
.NumberofMultisamples
= ffs(ice
->state
.framebuffer
.samples
) - 1;
5337 if (dirty
& IRIS_DIRTY_SAMPLE_MASK
) {
5338 iris_emit_cmd(batch
, GENX(3DSTATE_SAMPLE_MASK
), ms
) {
5339 ms
.SampleMask
= ice
->state
.sample_mask
;
5343 for (int stage
= 0; stage
<= MESA_SHADER_FRAGMENT
; stage
++) {
5344 if (!(dirty
& (IRIS_DIRTY_VS
<< stage
)))
5347 struct iris_compiled_shader
*shader
= ice
->shaders
.prog
[stage
];
5350 struct brw_stage_prog_data
*prog_data
= shader
->prog_data
;
5351 struct iris_resource
*cache
= (void *) shader
->assembly
.res
;
5352 iris_use_pinned_bo(batch
, cache
->bo
, false);
5354 if (prog_data
->total_scratch
> 0) {
5355 struct iris_bo
*bo
=
5356 iris_get_scratch_space(ice
, prog_data
->total_scratch
, stage
);
5357 iris_use_pinned_bo(batch
, bo
, true);
5360 if (stage
== MESA_SHADER_FRAGMENT
) {
5361 UNUSED
struct iris_rasterizer_state
*cso
= ice
->state
.cso_rast
;
5362 struct pipe_framebuffer_state
*cso_fb
= &ice
->state
.framebuffer
;
5364 uint32_t ps_state
[GENX(3DSTATE_PS_length
)] = {0};
5365 iris_pack_command(GENX(3DSTATE_PS
), ps_state
, ps
) {
5366 ps
._8PixelDispatchEnable
= wm_prog_data
->dispatch_8
;
5367 ps
._16PixelDispatchEnable
= wm_prog_data
->dispatch_16
;
5368 ps
._32PixelDispatchEnable
= wm_prog_data
->dispatch_32
;
5370 /* The docs for 3DSTATE_PS::32 Pixel Dispatch Enable say:
5372 * "When NUM_MULTISAMPLES = 16 or FORCE_SAMPLE_COUNT = 16,
5373 * SIMD32 Dispatch must not be enabled for PER_PIXEL dispatch
5376 * 16x MSAA only exists on Gen9+, so we can skip this on Gen8.
5378 if (GEN_GEN
>= 9 && cso_fb
->samples
== 16 &&
5379 !wm_prog_data
->persample_dispatch
) {
5380 assert(ps
._8PixelDispatchEnable
|| ps
._16PixelDispatchEnable
);
5381 ps
._32PixelDispatchEnable
= false;
5384 ps
.DispatchGRFStartRegisterForConstantSetupData0
=
5385 brw_wm_prog_data_dispatch_grf_start_reg(wm_prog_data
, ps
, 0);
5386 ps
.DispatchGRFStartRegisterForConstantSetupData1
=
5387 brw_wm_prog_data_dispatch_grf_start_reg(wm_prog_data
, ps
, 1);
5388 ps
.DispatchGRFStartRegisterForConstantSetupData2
=
5389 brw_wm_prog_data_dispatch_grf_start_reg(wm_prog_data
, ps
, 2);
5391 ps
.KernelStartPointer0
= KSP(shader
) +
5392 brw_wm_prog_data_prog_offset(wm_prog_data
, ps
, 0);
5393 ps
.KernelStartPointer1
= KSP(shader
) +
5394 brw_wm_prog_data_prog_offset(wm_prog_data
, ps
, 1);
5395 ps
.KernelStartPointer2
= KSP(shader
) +
5396 brw_wm_prog_data_prog_offset(wm_prog_data
, ps
, 2);
5399 uint32_t psx_state
[GENX(3DSTATE_PS_EXTRA_length
)] = {0};
5400 iris_pack_command(GENX(3DSTATE_PS_EXTRA
), psx_state
, psx
) {
5402 if (!wm_prog_data
->uses_sample_mask
)
5403 psx
.InputCoverageMaskState
= ICMS_NONE
;
5404 else if (wm_prog_data
->post_depth_coverage
)
5405 psx
.InputCoverageMaskState
= ICMS_DEPTH_COVERAGE
;
5406 else if (wm_prog_data
->inner_coverage
&&
5407 cso
->conservative_rasterization
)
5408 psx
.InputCoverageMaskState
= ICMS_INNER_CONSERVATIVE
;
5410 psx
.InputCoverageMaskState
= ICMS_NORMAL
;
5412 psx
.PixelShaderUsesInputCoverageMask
=
5413 wm_prog_data
->uses_sample_mask
;
5417 uint32_t *shader_ps
= (uint32_t *) shader
->derived_data
;
5418 uint32_t *shader_psx
= shader_ps
+ GENX(3DSTATE_PS_length
);
5419 iris_emit_merge(batch
, shader_ps
, ps_state
,
5420 GENX(3DSTATE_PS_length
));
5421 iris_emit_merge(batch
, shader_psx
, psx_state
,
5422 GENX(3DSTATE_PS_EXTRA_length
));
5424 iris_batch_emit(batch
, shader
->derived_data
,
5425 iris_derived_program_state_size(stage
));
5428 if (stage
== MESA_SHADER_TESS_EVAL
) {
5429 iris_emit_cmd(batch
, GENX(3DSTATE_HS
), hs
);
5430 iris_emit_cmd(batch
, GENX(3DSTATE_TE
), te
);
5431 iris_emit_cmd(batch
, GENX(3DSTATE_DS
), ds
);
5432 } else if (stage
== MESA_SHADER_GEOMETRY
) {
5433 iris_emit_cmd(batch
, GENX(3DSTATE_GS
), gs
);
5438 if (ice
->state
.streamout_active
) {
5439 if (dirty
& IRIS_DIRTY_SO_BUFFERS
) {
5440 iris_batch_emit(batch
, genx
->so_buffers
,
5441 4 * 4 * GENX(3DSTATE_SO_BUFFER_length
));
5442 for (int i
= 0; i
< 4; i
++) {
5443 struct iris_stream_output_target
*tgt
=
5444 (void *) ice
->state
.so_target
[i
];
5447 iris_use_pinned_bo(batch
, iris_resource_bo(tgt
->base
.buffer
),
5449 iris_use_pinned_bo(batch
, iris_resource_bo(tgt
->offset
.res
),
5455 if ((dirty
& IRIS_DIRTY_SO_DECL_LIST
) && ice
->state
.streamout
) {
5456 uint32_t *decl_list
=
5457 ice
->state
.streamout
+ GENX(3DSTATE_STREAMOUT_length
);
5458 iris_batch_emit(batch
, decl_list
, 4 * ((decl_list
[0] & 0xff) + 2));
5461 if (dirty
& IRIS_DIRTY_STREAMOUT
) {
5462 const struct iris_rasterizer_state
*cso_rast
= ice
->state
.cso_rast
;
5464 uint32_t dynamic_sol
[GENX(3DSTATE_STREAMOUT_length
)];
5465 iris_pack_command(GENX(3DSTATE_STREAMOUT
), dynamic_sol
, sol
) {
5466 sol
.SOFunctionEnable
= true;
5467 sol
.SOStatisticsEnable
= true;
5469 sol
.RenderingDisable
= cso_rast
->rasterizer_discard
&&
5470 !ice
->state
.prims_generated_query_active
;
5471 sol
.ReorderMode
= cso_rast
->flatshade_first
? LEADING
: TRAILING
;
5474 assert(ice
->state
.streamout
);
5476 iris_emit_merge(batch
, ice
->state
.streamout
, dynamic_sol
,
5477 GENX(3DSTATE_STREAMOUT_length
));
5480 if (dirty
& IRIS_DIRTY_STREAMOUT
) {
5481 iris_emit_cmd(batch
, GENX(3DSTATE_STREAMOUT
), sol
);
5485 if (dirty
& IRIS_DIRTY_CLIP
) {
5486 struct iris_rasterizer_state
*cso_rast
= ice
->state
.cso_rast
;
5487 struct pipe_framebuffer_state
*cso_fb
= &ice
->state
.framebuffer
;
5489 bool gs_or_tes
= ice
->shaders
.prog
[MESA_SHADER_GEOMETRY
] ||
5490 ice
->shaders
.prog
[MESA_SHADER_TESS_EVAL
];
5491 bool points_or_lines
= cso_rast
->fill_mode_point_or_line
||
5492 (gs_or_tes
? ice
->shaders
.output_topology_is_points_or_lines
5493 : ice
->state
.prim_is_points_or_lines
);
5495 uint32_t dynamic_clip
[GENX(3DSTATE_CLIP_length
)];
5496 iris_pack_command(GENX(3DSTATE_CLIP
), &dynamic_clip
, cl
) {
5497 cl
.StatisticsEnable
= ice
->state
.statistics_counters_enabled
;
5498 if (cso_rast
->rasterizer_discard
)
5499 cl
.ClipMode
= CLIPMODE_REJECT_ALL
;
5500 else if (ice
->state
.window_space_position
)
5501 cl
.ClipMode
= CLIPMODE_ACCEPT_ALL
;
5503 cl
.ClipMode
= CLIPMODE_NORMAL
;
5505 cl
.PerspectiveDivideDisable
= ice
->state
.window_space_position
;
5506 cl
.ViewportXYClipTestEnable
= !points_or_lines
;
5508 if (wm_prog_data
->barycentric_interp_modes
&
5509 BRW_BARYCENTRIC_NONPERSPECTIVE_BITS
)
5510 cl
.NonPerspectiveBarycentricEnable
= true;
5512 cl
.ForceZeroRTAIndexEnable
= cso_fb
->layers
<= 1;
5513 cl
.MaximumVPIndex
= ice
->state
.num_viewports
- 1;
5515 iris_emit_merge(batch
, cso_rast
->clip
, dynamic_clip
,
5516 ARRAY_SIZE(cso_rast
->clip
));
5519 if (dirty
& IRIS_DIRTY_RASTER
) {
5520 struct iris_rasterizer_state
*cso
= ice
->state
.cso_rast
;
5521 iris_batch_emit(batch
, cso
->raster
, sizeof(cso
->raster
));
5523 uint32_t dynamic_sf
[GENX(3DSTATE_SF_length
)];
5524 iris_pack_command(GENX(3DSTATE_SF
), &dynamic_sf
, sf
) {
5525 sf
.ViewportTransformEnable
= !ice
->state
.window_space_position
;
5527 iris_emit_merge(batch
, cso
->sf
, dynamic_sf
,
5528 ARRAY_SIZE(dynamic_sf
));
5531 if (dirty
& IRIS_DIRTY_WM
) {
5532 struct iris_rasterizer_state
*cso
= ice
->state
.cso_rast
;
5533 uint32_t dynamic_wm
[GENX(3DSTATE_WM_length
)];
5535 iris_pack_command(GENX(3DSTATE_WM
), &dynamic_wm
, wm
) {
5536 wm
.StatisticsEnable
= ice
->state
.statistics_counters_enabled
;
5538 wm
.BarycentricInterpolationMode
=
5539 wm_prog_data
->barycentric_interp_modes
;
5541 if (wm_prog_data
->early_fragment_tests
)
5542 wm
.EarlyDepthStencilControl
= EDSC_PREPS
;
5543 else if (wm_prog_data
->has_side_effects
)
5544 wm
.EarlyDepthStencilControl
= EDSC_PSEXEC
;
5546 /* We could skip this bit if color writes are enabled. */
5547 if (wm_prog_data
->has_side_effects
|| wm_prog_data
->uses_kill
)
5548 wm
.ForceThreadDispatchEnable
= ForceON
;
5550 iris_emit_merge(batch
, cso
->wm
, dynamic_wm
, ARRAY_SIZE(cso
->wm
));
5553 if (dirty
& IRIS_DIRTY_SBE
) {
5554 iris_emit_sbe(batch
, ice
);
5557 if (dirty
& IRIS_DIRTY_PS_BLEND
) {
5558 struct iris_blend_state
*cso_blend
= ice
->state
.cso_blend
;
5559 struct iris_depth_stencil_alpha_state
*cso_zsa
= ice
->state
.cso_zsa
;
5560 const struct shader_info
*fs_info
=
5561 iris_get_shader_info(ice
, MESA_SHADER_FRAGMENT
);
5563 uint32_t dynamic_pb
[GENX(3DSTATE_PS_BLEND_length
)];
5564 iris_pack_command(GENX(3DSTATE_PS_BLEND
), &dynamic_pb
, pb
) {
5565 pb
.HasWriteableRT
= has_writeable_rt(cso_blend
, fs_info
);
5566 pb
.AlphaTestEnable
= cso_zsa
->alpha
.enabled
;
5568 /* The dual source blending docs caution against using SRC1 factors
5569 * when the shader doesn't use a dual source render target write.
5570 * Empirically, this can lead to GPU hangs, and the results are
5571 * undefined anyway, so simply disable blending to avoid the hang.
5573 pb
.ColorBufferBlendEnable
= (cso_blend
->blend_enables
& 1) &&
5574 (!cso_blend
->dual_color_blending
|| wm_prog_data
->dual_src_blend
);
5577 iris_emit_merge(batch
, cso_blend
->ps_blend
, dynamic_pb
,
5578 ARRAY_SIZE(cso_blend
->ps_blend
));
5581 if (dirty
& IRIS_DIRTY_WM_DEPTH_STENCIL
) {
5582 struct iris_depth_stencil_alpha_state
*cso
= ice
->state
.cso_zsa
;
5584 struct pipe_stencil_ref
*p_stencil_refs
= &ice
->state
.stencil_ref
;
5585 uint32_t stencil_refs
[GENX(3DSTATE_WM_DEPTH_STENCIL_length
)];
5586 iris_pack_command(GENX(3DSTATE_WM_DEPTH_STENCIL
), &stencil_refs
, wmds
) {
5587 wmds
.StencilReferenceValue
= p_stencil_refs
->ref_value
[0];
5588 wmds
.BackfaceStencilReferenceValue
= p_stencil_refs
->ref_value
[1];
5590 iris_emit_merge(batch
, cso
->wmds
, stencil_refs
, ARRAY_SIZE(cso
->wmds
));
5592 iris_batch_emit(batch
, cso
->wmds
, sizeof(cso
->wmds
));
5596 iris_batch_emit(batch
, cso
->depth_bounds
, sizeof(cso
->depth_bounds
));
5600 if (dirty
& IRIS_DIRTY_SCISSOR_RECT
) {
5601 uint32_t scissor_offset
=
5602 emit_state(batch
, ice
->state
.dynamic_uploader
,
5603 &ice
->state
.last_res
.scissor
,
5604 ice
->state
.scissors
,
5605 sizeof(struct pipe_scissor_state
) *
5606 ice
->state
.num_viewports
, 32);
5608 iris_emit_cmd(batch
, GENX(3DSTATE_SCISSOR_STATE_POINTERS
), ptr
) {
5609 ptr
.ScissorRectPointer
= scissor_offset
;
5613 if (dirty
& IRIS_DIRTY_DEPTH_BUFFER
) {
5614 struct iris_depth_buffer_state
*cso_z
= &ice
->state
.genx
->depth_buffer
;
5616 /* Do not emit the clear params yets. We need to update the clear value
5619 uint32_t clear_length
= GENX(3DSTATE_CLEAR_PARAMS_length
) * 4;
5620 uint32_t cso_z_size
= sizeof(cso_z
->packets
) - clear_length
;
5621 iris_batch_emit(batch
, cso_z
->packets
, cso_z_size
);
5623 union isl_color_value clear_value
= { .f32
= { 0, } };
5625 struct pipe_framebuffer_state
*cso_fb
= &ice
->state
.framebuffer
;
5626 if (cso_fb
->zsbuf
) {
5627 struct iris_resource
*zres
, *sres
;
5628 iris_get_depth_stencil_resources(cso_fb
->zsbuf
->texture
,
5630 if (zres
&& zres
->aux
.bo
)
5631 clear_value
= iris_resource_get_clear_color(zres
, NULL
, NULL
);
5634 uint32_t clear_params
[GENX(3DSTATE_CLEAR_PARAMS_length
)];
5635 iris_pack_command(GENX(3DSTATE_CLEAR_PARAMS
), clear_params
, clear
) {
5636 clear
.DepthClearValueValid
= true;
5637 clear
.DepthClearValue
= clear_value
.f32
[0];
5639 iris_batch_emit(batch
, clear_params
, clear_length
);
5642 if (dirty
& (IRIS_DIRTY_DEPTH_BUFFER
| IRIS_DIRTY_WM_DEPTH_STENCIL
)) {
5643 /* Listen for buffer changes, and also write enable changes. */
5644 struct pipe_framebuffer_state
*cso_fb
= &ice
->state
.framebuffer
;
5645 pin_depth_and_stencil_buffers(batch
, cso_fb
->zsbuf
, ice
->state
.cso_zsa
);
5648 if (dirty
& IRIS_DIRTY_POLYGON_STIPPLE
) {
5649 iris_emit_cmd(batch
, GENX(3DSTATE_POLY_STIPPLE_PATTERN
), poly
) {
5650 for (int i
= 0; i
< 32; i
++) {
5651 poly
.PatternRow
[i
] = ice
->state
.poly_stipple
.stipple
[i
];
5656 if (dirty
& IRIS_DIRTY_LINE_STIPPLE
) {
5657 struct iris_rasterizer_state
*cso
= ice
->state
.cso_rast
;
5658 iris_batch_emit(batch
, cso
->line_stipple
, sizeof(cso
->line_stipple
));
5661 if (dirty
& IRIS_DIRTY_VF_TOPOLOGY
) {
5662 iris_emit_cmd(batch
, GENX(3DSTATE_VF_TOPOLOGY
), topo
) {
5663 topo
.PrimitiveTopologyType
=
5664 translate_prim_type(draw
->mode
, draw
->vertices_per_patch
);
5668 if (dirty
& IRIS_DIRTY_VERTEX_BUFFERS
) {
5669 int count
= util_bitcount64(ice
->state
.bound_vertex_buffers
);
5670 int dynamic_bound
= ice
->state
.bound_vertex_buffers
;
5672 if (ice
->state
.vs_uses_draw_params
) {
5673 assert(ice
->draw
.draw_params
.res
);
5675 struct iris_vertex_buffer_state
*state
=
5676 &(ice
->state
.genx
->vertex_buffers
[count
]);
5677 pipe_resource_reference(&state
->resource
, ice
->draw
.draw_params
.res
);
5678 struct iris_resource
*res
= (void *) state
->resource
;
5680 iris_pack_state(GENX(VERTEX_BUFFER_STATE
), state
->state
, vb
) {
5681 vb
.VertexBufferIndex
= count
;
5682 vb
.AddressModifyEnable
= true;
5684 vb
.BufferSize
= res
->bo
->size
- ice
->draw
.draw_params
.offset
;
5685 vb
.BufferStartingAddress
=
5686 ro_bo(NULL
, res
->bo
->gtt_offset
+
5687 (int) ice
->draw
.draw_params
.offset
);
5688 vb
.MOCS
= mocs(res
->bo
);
5690 dynamic_bound
|= 1ull << count
;
5694 if (ice
->state
.vs_uses_derived_draw_params
) {
5695 struct iris_vertex_buffer_state
*state
=
5696 &(ice
->state
.genx
->vertex_buffers
[count
]);
5697 pipe_resource_reference(&state
->resource
,
5698 ice
->draw
.derived_draw_params
.res
);
5699 struct iris_resource
*res
= (void *) ice
->draw
.derived_draw_params
.res
;
5701 iris_pack_state(GENX(VERTEX_BUFFER_STATE
), state
->state
, vb
) {
5702 vb
.VertexBufferIndex
= count
;
5703 vb
.AddressModifyEnable
= true;
5706 res
->bo
->size
- ice
->draw
.derived_draw_params
.offset
;
5707 vb
.BufferStartingAddress
=
5708 ro_bo(NULL
, res
->bo
->gtt_offset
+
5709 (int) ice
->draw
.derived_draw_params
.offset
);
5710 vb
.MOCS
= mocs(res
->bo
);
5712 dynamic_bound
|= 1ull << count
;
5717 /* The VF cache designers cut corners, and made the cache key's
5718 * <VertexBufferIndex, Memory Address> tuple only consider the bottom
5719 * 32 bits of the address. If you have two vertex buffers which get
5720 * placed exactly 4 GiB apart and use them in back-to-back draw calls,
5721 * you can get collisions (even within a single batch).
5723 * So, we need to do a VF cache invalidate if the buffer for a VB
5724 * slot slot changes [48:32] address bits from the previous time.
5726 unsigned flush_flags
= 0;
5728 uint64_t bound
= dynamic_bound
;
5730 const int i
= u_bit_scan64(&bound
);
5731 uint16_t high_bits
= 0;
5733 struct iris_resource
*res
=
5734 (void *) genx
->vertex_buffers
[i
].resource
;
5736 iris_use_pinned_bo(batch
, res
->bo
, false);
5738 high_bits
= res
->bo
->gtt_offset
>> 32ull;
5739 if (high_bits
!= ice
->state
.last_vbo_high_bits
[i
]) {
5740 flush_flags
|= PIPE_CONTROL_VF_CACHE_INVALIDATE
|
5741 PIPE_CONTROL_CS_STALL
;
5742 ice
->state
.last_vbo_high_bits
[i
] = high_bits
;
5748 iris_emit_pipe_control_flush(batch
,
5749 "workaround: VF cache 32-bit key [VB]",
5753 const unsigned vb_dwords
= GENX(VERTEX_BUFFER_STATE_length
);
5756 iris_get_command_space(batch
, 4 * (1 + vb_dwords
* count
));
5757 _iris_pack_command(batch
, GENX(3DSTATE_VERTEX_BUFFERS
), map
, vb
) {
5758 vb
.DWordLength
= (vb_dwords
* count
+ 1) - 2;
5762 bound
= dynamic_bound
;
5764 const int i
= u_bit_scan64(&bound
);
5765 memcpy(map
, genx
->vertex_buffers
[i
].state
,
5766 sizeof(uint32_t) * vb_dwords
);
5772 if (dirty
& IRIS_DIRTY_VERTEX_ELEMENTS
) {
5773 struct iris_vertex_element_state
*cso
= ice
->state
.cso_vertex_elements
;
5774 const unsigned entries
= MAX2(cso
->count
, 1);
5775 if (!(ice
->state
.vs_needs_sgvs_element
||
5776 ice
->state
.vs_uses_derived_draw_params
||
5777 ice
->state
.vs_needs_edge_flag
)) {
5778 iris_batch_emit(batch
, cso
->vertex_elements
, sizeof(uint32_t) *
5779 (1 + entries
* GENX(VERTEX_ELEMENT_STATE_length
)));
5781 uint32_t dynamic_ves
[1 + 33 * GENX(VERTEX_ELEMENT_STATE_length
)];
5782 const unsigned dyn_count
= cso
->count
+
5783 ice
->state
.vs_needs_sgvs_element
+
5784 ice
->state
.vs_uses_derived_draw_params
;
5786 iris_pack_command(GENX(3DSTATE_VERTEX_ELEMENTS
),
5789 1 + GENX(VERTEX_ELEMENT_STATE_length
) * dyn_count
- 2;
5791 memcpy(&dynamic_ves
[1], &cso
->vertex_elements
[1],
5792 (cso
->count
- ice
->state
.vs_needs_edge_flag
) *
5793 GENX(VERTEX_ELEMENT_STATE_length
) * sizeof(uint32_t));
5794 uint32_t *ve_pack_dest
=
5795 &dynamic_ves
[1 + (cso
->count
- ice
->state
.vs_needs_edge_flag
) *
5796 GENX(VERTEX_ELEMENT_STATE_length
)];
5798 if (ice
->state
.vs_needs_sgvs_element
) {
5799 uint32_t base_ctrl
= ice
->state
.vs_uses_draw_params
?
5800 VFCOMP_STORE_SRC
: VFCOMP_STORE_0
;
5801 iris_pack_state(GENX(VERTEX_ELEMENT_STATE
), ve_pack_dest
, ve
) {
5803 ve
.VertexBufferIndex
=
5804 util_bitcount64(ice
->state
.bound_vertex_buffers
);
5805 ve
.SourceElementFormat
= ISL_FORMAT_R32G32_UINT
;
5806 ve
.Component0Control
= base_ctrl
;
5807 ve
.Component1Control
= base_ctrl
;
5808 ve
.Component2Control
= VFCOMP_STORE_0
;
5809 ve
.Component3Control
= VFCOMP_STORE_0
;
5811 ve_pack_dest
+= GENX(VERTEX_ELEMENT_STATE_length
);
5813 if (ice
->state
.vs_uses_derived_draw_params
) {
5814 iris_pack_state(GENX(VERTEX_ELEMENT_STATE
), ve_pack_dest
, ve
) {
5816 ve
.VertexBufferIndex
=
5817 util_bitcount64(ice
->state
.bound_vertex_buffers
) +
5818 ice
->state
.vs_uses_draw_params
;
5819 ve
.SourceElementFormat
= ISL_FORMAT_R32G32_UINT
;
5820 ve
.Component0Control
= VFCOMP_STORE_SRC
;
5821 ve
.Component1Control
= VFCOMP_STORE_SRC
;
5822 ve
.Component2Control
= VFCOMP_STORE_0
;
5823 ve
.Component3Control
= VFCOMP_STORE_0
;
5825 ve_pack_dest
+= GENX(VERTEX_ELEMENT_STATE_length
);
5827 if (ice
->state
.vs_needs_edge_flag
) {
5828 for (int i
= 0; i
< GENX(VERTEX_ELEMENT_STATE_length
); i
++)
5829 ve_pack_dest
[i
] = cso
->edgeflag_ve
[i
];
5832 iris_batch_emit(batch
, &dynamic_ves
, sizeof(uint32_t) *
5833 (1 + dyn_count
* GENX(VERTEX_ELEMENT_STATE_length
)));
5836 if (!ice
->state
.vs_needs_edge_flag
) {
5837 iris_batch_emit(batch
, cso
->vf_instancing
, sizeof(uint32_t) *
5838 entries
* GENX(3DSTATE_VF_INSTANCING_length
));
5840 assert(cso
->count
> 0);
5841 const unsigned edgeflag_index
= cso
->count
- 1;
5842 uint32_t dynamic_vfi
[33 * GENX(3DSTATE_VF_INSTANCING_length
)];
5843 memcpy(&dynamic_vfi
[0], cso
->vf_instancing
, edgeflag_index
*
5844 GENX(3DSTATE_VF_INSTANCING_length
) * sizeof(uint32_t));
5846 uint32_t *vfi_pack_dest
= &dynamic_vfi
[0] +
5847 edgeflag_index
* GENX(3DSTATE_VF_INSTANCING_length
);
5848 iris_pack_command(GENX(3DSTATE_VF_INSTANCING
), vfi_pack_dest
, vi
) {
5849 vi
.VertexElementIndex
= edgeflag_index
+
5850 ice
->state
.vs_needs_sgvs_element
+
5851 ice
->state
.vs_uses_derived_draw_params
;
5853 for (int i
= 0; i
< GENX(3DSTATE_VF_INSTANCING_length
); i
++)
5854 vfi_pack_dest
[i
] |= cso
->edgeflag_vfi
[i
];
5856 iris_batch_emit(batch
, &dynamic_vfi
[0], sizeof(uint32_t) *
5857 entries
* GENX(3DSTATE_VF_INSTANCING_length
));
5861 if (dirty
& IRIS_DIRTY_VF_SGVS
) {
5862 const struct brw_vs_prog_data
*vs_prog_data
= (void *)
5863 ice
->shaders
.prog
[MESA_SHADER_VERTEX
]->prog_data
;
5864 struct iris_vertex_element_state
*cso
= ice
->state
.cso_vertex_elements
;
5866 iris_emit_cmd(batch
, GENX(3DSTATE_VF_SGVS
), sgv
) {
5867 if (vs_prog_data
->uses_vertexid
) {
5868 sgv
.VertexIDEnable
= true;
5869 sgv
.VertexIDComponentNumber
= 2;
5870 sgv
.VertexIDElementOffset
=
5871 cso
->count
- ice
->state
.vs_needs_edge_flag
;
5874 if (vs_prog_data
->uses_instanceid
) {
5875 sgv
.InstanceIDEnable
= true;
5876 sgv
.InstanceIDComponentNumber
= 3;
5877 sgv
.InstanceIDElementOffset
=
5878 cso
->count
- ice
->state
.vs_needs_edge_flag
;
5883 if (dirty
& IRIS_DIRTY_VF
) {
5884 iris_emit_cmd(batch
, GENX(3DSTATE_VF
), vf
) {
5885 if (draw
->primitive_restart
) {
5886 vf
.IndexedDrawCutIndexEnable
= true;
5887 vf
.CutIndex
= draw
->restart_index
;
5892 if (dirty
& IRIS_DIRTY_VF_STATISTICS
) {
5893 iris_emit_cmd(batch
, GENX(3DSTATE_VF_STATISTICS
), vf
) {
5894 vf
.StatisticsEnable
= true;
5899 if (dirty
& IRIS_DIRTY_PMA_FIX
) {
5900 bool enable
= want_pma_fix(ice
);
5901 genX(update_pma_fix
)(ice
, batch
, enable
);
5905 if (ice
->state
.current_hash_scale
!= 1)
5906 genX(emit_hashing_mode
)(ice
, batch
, UINT_MAX
, UINT_MAX
, 1);
5909 genX(emit_aux_map_state
)(batch
);
5914 iris_upload_render_state(struct iris_context
*ice
,
5915 struct iris_batch
*batch
,
5916 const struct pipe_draw_info
*draw
)
5918 bool use_predicate
= ice
->state
.predicate
== IRIS_PREDICATE_STATE_USE_BIT
;
5920 /* Always pin the binder. If we're emitting new binding table pointers,
5921 * we need it. If not, we're probably inheriting old tables via the
5922 * context, and need it anyway. Since true zero-bindings cases are
5923 * practically non-existent, just pin it and avoid last_res tracking.
5925 iris_use_pinned_bo(batch
, ice
->state
.binder
.bo
, false);
5927 if (!batch
->contains_draw
) {
5928 iris_restore_render_saved_bos(ice
, batch
, draw
);
5929 batch
->contains_draw
= true;
5932 iris_upload_dirty_render_state(ice
, batch
, draw
);
5934 if (draw
->index_size
> 0) {
5937 if (draw
->has_user_indices
) {
5938 u_upload_data(ice
->ctx
.stream_uploader
, 0,
5939 draw
->count
* draw
->index_size
, 4, draw
->index
.user
,
5940 &offset
, &ice
->state
.last_res
.index_buffer
);
5942 struct iris_resource
*res
= (void *) draw
->index
.resource
;
5943 res
->bind_history
|= PIPE_BIND_INDEX_BUFFER
;
5945 pipe_resource_reference(&ice
->state
.last_res
.index_buffer
,
5946 draw
->index
.resource
);
5950 struct iris_genx_state
*genx
= ice
->state
.genx
;
5951 struct iris_bo
*bo
= iris_resource_bo(ice
->state
.last_res
.index_buffer
);
5953 uint32_t ib_packet
[GENX(3DSTATE_INDEX_BUFFER_length
)];
5954 iris_pack_command(GENX(3DSTATE_INDEX_BUFFER
), ib_packet
, ib
) {
5955 ib
.IndexFormat
= draw
->index_size
>> 1;
5957 ib
.BufferSize
= bo
->size
- offset
;
5958 ib
.BufferStartingAddress
= ro_bo(NULL
, bo
->gtt_offset
+ offset
);
5961 if (memcmp(genx
->last_index_buffer
, ib_packet
, sizeof(ib_packet
)) != 0) {
5962 memcpy(genx
->last_index_buffer
, ib_packet
, sizeof(ib_packet
));
5963 iris_batch_emit(batch
, ib_packet
, sizeof(ib_packet
));
5964 iris_use_pinned_bo(batch
, bo
, false);
5967 /* The VF cache key only uses 32-bits, see vertex buffer comment above */
5968 uint16_t high_bits
= bo
->gtt_offset
>> 32ull;
5969 if (high_bits
!= ice
->state
.last_index_bo_high_bits
) {
5970 iris_emit_pipe_control_flush(batch
,
5971 "workaround: VF cache 32-bit key [IB]",
5972 PIPE_CONTROL_VF_CACHE_INVALIDATE
|
5973 PIPE_CONTROL_CS_STALL
);
5974 ice
->state
.last_index_bo_high_bits
= high_bits
;
5978 #define _3DPRIM_END_OFFSET 0x2420
5979 #define _3DPRIM_START_VERTEX 0x2430
5980 #define _3DPRIM_VERTEX_COUNT 0x2434
5981 #define _3DPRIM_INSTANCE_COUNT 0x2438
5982 #define _3DPRIM_START_INSTANCE 0x243C
5983 #define _3DPRIM_BASE_VERTEX 0x2440
5985 if (draw
->indirect
) {
5986 if (draw
->indirect
->indirect_draw_count
) {
5987 use_predicate
= true;
5989 struct iris_bo
*draw_count_bo
=
5990 iris_resource_bo(draw
->indirect
->indirect_draw_count
);
5991 unsigned draw_count_offset
=
5992 draw
->indirect
->indirect_draw_count_offset
;
5994 iris_emit_pipe_control_flush(batch
,
5995 "ensure indirect draw buffer is flushed",
5996 PIPE_CONTROL_FLUSH_ENABLE
);
5998 if (ice
->state
.predicate
== IRIS_PREDICATE_STATE_USE_BIT
) {
5999 struct gen_mi_builder b
;
6000 gen_mi_builder_init(&b
, batch
);
6002 /* comparison = draw id < draw count */
6003 struct gen_mi_value comparison
=
6004 gen_mi_ult(&b
, gen_mi_imm(draw
->drawid
),
6005 gen_mi_mem32(ro_bo(draw_count_bo
,
6006 draw_count_offset
)));
6008 /* predicate = comparison & conditional rendering predicate */
6009 gen_mi_store(&b
, gen_mi_reg32(MI_PREDICATE_RESULT
),
6010 gen_mi_iand(&b
, comparison
,
6011 gen_mi_reg32(CS_GPR(15))));
6013 uint32_t mi_predicate
;
6015 /* Upload the id of the current primitive to MI_PREDICATE_SRC1. */
6016 iris_load_register_imm64(batch
, MI_PREDICATE_SRC1
, draw
->drawid
);
6017 /* Upload the current draw count from the draw parameters buffer
6018 * to MI_PREDICATE_SRC0.
6020 iris_load_register_mem32(batch
, MI_PREDICATE_SRC0
,
6021 draw_count_bo
, draw_count_offset
);
6022 /* Zero the top 32-bits of MI_PREDICATE_SRC0 */
6023 iris_load_register_imm32(batch
, MI_PREDICATE_SRC0
+ 4, 0);
6025 if (draw
->drawid
== 0) {
6026 mi_predicate
= MI_PREDICATE
| MI_PREDICATE_LOADOP_LOADINV
|
6027 MI_PREDICATE_COMBINEOP_SET
|
6028 MI_PREDICATE_COMPAREOP_SRCS_EQUAL
;
6030 /* While draw_index < draw_count the predicate's result will be
6031 * (draw_index == draw_count) ^ TRUE = TRUE
6032 * When draw_index == draw_count the result is
6033 * (TRUE) ^ TRUE = FALSE
6034 * After this all results will be:
6035 * (FALSE) ^ FALSE = FALSE
6037 mi_predicate
= MI_PREDICATE
| MI_PREDICATE_LOADOP_LOAD
|
6038 MI_PREDICATE_COMBINEOP_XOR
|
6039 MI_PREDICATE_COMPAREOP_SRCS_EQUAL
;
6041 iris_batch_emit(batch
, &mi_predicate
, sizeof(uint32_t));
6044 struct iris_bo
*bo
= iris_resource_bo(draw
->indirect
->buffer
);
6047 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_MEM
), lrm
) {
6048 lrm
.RegisterAddress
= _3DPRIM_VERTEX_COUNT
;
6049 lrm
.MemoryAddress
= ro_bo(bo
, draw
->indirect
->offset
+ 0);
6051 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_MEM
), lrm
) {
6052 lrm
.RegisterAddress
= _3DPRIM_INSTANCE_COUNT
;
6053 lrm
.MemoryAddress
= ro_bo(bo
, draw
->indirect
->offset
+ 4);
6055 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_MEM
), lrm
) {
6056 lrm
.RegisterAddress
= _3DPRIM_START_VERTEX
;
6057 lrm
.MemoryAddress
= ro_bo(bo
, draw
->indirect
->offset
+ 8);
6059 if (draw
->index_size
) {
6060 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_MEM
), lrm
) {
6061 lrm
.RegisterAddress
= _3DPRIM_BASE_VERTEX
;
6062 lrm
.MemoryAddress
= ro_bo(bo
, draw
->indirect
->offset
+ 12);
6064 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_MEM
), lrm
) {
6065 lrm
.RegisterAddress
= _3DPRIM_START_INSTANCE
;
6066 lrm
.MemoryAddress
= ro_bo(bo
, draw
->indirect
->offset
+ 16);
6069 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_MEM
), lrm
) {
6070 lrm
.RegisterAddress
= _3DPRIM_START_INSTANCE
;
6071 lrm
.MemoryAddress
= ro_bo(bo
, draw
->indirect
->offset
+ 12);
6073 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_IMM
), lri
) {
6074 lri
.RegisterOffset
= _3DPRIM_BASE_VERTEX
;
6078 } else if (draw
->count_from_stream_output
) {
6079 struct iris_stream_output_target
*so
=
6080 (void *) draw
->count_from_stream_output
;
6082 /* XXX: Replace with actual cache tracking */
6083 iris_emit_pipe_control_flush(batch
,
6084 "draw count from stream output stall",
6085 PIPE_CONTROL_CS_STALL
);
6087 struct gen_mi_builder b
;
6088 gen_mi_builder_init(&b
, batch
);
6090 struct iris_address addr
=
6091 ro_bo(iris_resource_bo(so
->offset
.res
), so
->offset
.offset
);
6092 struct gen_mi_value offset
=
6093 gen_mi_iadd_imm(&b
, gen_mi_mem32(addr
), -so
->base
.buffer_offset
);
6095 gen_mi_store(&b
, gen_mi_reg32(_3DPRIM_VERTEX_COUNT
),
6096 gen_mi_udiv32_imm(&b
, offset
, so
->stride
));
6098 _iris_emit_lri(batch
, _3DPRIM_START_VERTEX
, 0);
6099 _iris_emit_lri(batch
, _3DPRIM_BASE_VERTEX
, 0);
6100 _iris_emit_lri(batch
, _3DPRIM_START_INSTANCE
, 0);
6101 _iris_emit_lri(batch
, _3DPRIM_INSTANCE_COUNT
, draw
->instance_count
);
6104 iris_emit_cmd(batch
, GENX(3DPRIMITIVE
), prim
) {
6105 prim
.VertexAccessType
= draw
->index_size
> 0 ? RANDOM
: SEQUENTIAL
;
6106 prim
.PredicateEnable
= use_predicate
;
6108 if (draw
->indirect
|| draw
->count_from_stream_output
) {
6109 prim
.IndirectParameterEnable
= true;
6111 prim
.StartInstanceLocation
= draw
->start_instance
;
6112 prim
.InstanceCount
= draw
->instance_count
;
6113 prim
.VertexCountPerInstance
= draw
->count
;
6115 prim
.StartVertexLocation
= draw
->start
;
6117 if (draw
->index_size
) {
6118 prim
.BaseVertexLocation
+= draw
->index_bias
;
6120 prim
.StartVertexLocation
+= draw
->index_bias
;
6127 iris_upload_compute_state(struct iris_context
*ice
,
6128 struct iris_batch
*batch
,
6129 const struct pipe_grid_info
*grid
)
6131 const uint64_t dirty
= ice
->state
.dirty
;
6132 struct iris_screen
*screen
= batch
->screen
;
6133 const struct gen_device_info
*devinfo
= &screen
->devinfo
;
6134 struct iris_binder
*binder
= &ice
->state
.binder
;
6135 struct iris_shader_state
*shs
= &ice
->state
.shaders
[MESA_SHADER_COMPUTE
];
6136 struct iris_compiled_shader
*shader
=
6137 ice
->shaders
.prog
[MESA_SHADER_COMPUTE
];
6138 struct brw_stage_prog_data
*prog_data
= shader
->prog_data
;
6139 struct brw_cs_prog_data
*cs_prog_data
= (void *) prog_data
;
6141 /* Always pin the binder. If we're emitting new binding table pointers,
6142 * we need it. If not, we're probably inheriting old tables via the
6143 * context, and need it anyway. Since true zero-bindings cases are
6144 * practically non-existent, just pin it and avoid last_res tracking.
6146 iris_use_pinned_bo(batch
, ice
->state
.binder
.bo
, false);
6148 if ((dirty
& IRIS_DIRTY_CONSTANTS_CS
) && shs
->sysvals_need_upload
)
6149 upload_sysvals(ice
, MESA_SHADER_COMPUTE
);
6151 if (dirty
& IRIS_DIRTY_BINDINGS_CS
)
6152 iris_populate_binding_table(ice
, batch
, MESA_SHADER_COMPUTE
, false);
6154 if (dirty
& IRIS_DIRTY_SAMPLER_STATES_CS
)
6155 iris_upload_sampler_states(ice
, MESA_SHADER_COMPUTE
);
6157 iris_use_optional_res(batch
, shs
->sampler_table
.res
, false);
6158 iris_use_pinned_bo(batch
, iris_resource_bo(shader
->assembly
.res
), false);
6160 if (ice
->state
.need_border_colors
)
6161 iris_use_pinned_bo(batch
, ice
->state
.border_color_pool
.bo
, false);
6164 genX(emit_aux_map_state
)(batch
);
6167 if (dirty
& IRIS_DIRTY_CS
) {
6168 /* The MEDIA_VFE_STATE documentation for Gen8+ says:
6170 * "A stalling PIPE_CONTROL is required before MEDIA_VFE_STATE unless
6171 * the only bits that are changed are scoreboard related: Scoreboard
6172 * Enable, Scoreboard Type, Scoreboard Mask, Scoreboard Delta. For
6173 * these scoreboard related states, a MEDIA_STATE_FLUSH is
6176 iris_emit_pipe_control_flush(batch
,
6177 "workaround: stall before MEDIA_VFE_STATE",
6178 PIPE_CONTROL_CS_STALL
);
6180 iris_emit_cmd(batch
, GENX(MEDIA_VFE_STATE
), vfe
) {
6181 if (prog_data
->total_scratch
) {
6182 struct iris_bo
*bo
=
6183 iris_get_scratch_space(ice
, prog_data
->total_scratch
,
6184 MESA_SHADER_COMPUTE
);
6185 vfe
.PerThreadScratchSpace
= ffs(prog_data
->total_scratch
) - 11;
6186 vfe
.ScratchSpaceBasePointer
= rw_bo(bo
, 0);
6189 vfe
.MaximumNumberofThreads
=
6190 devinfo
->max_cs_threads
* screen
->subslice_total
- 1;
6192 vfe
.ResetGatewayTimer
=
6193 Resettingrelativetimerandlatchingtheglobaltimestamp
;
6196 vfe
.BypassGatewayControl
= true;
6198 vfe
.NumberofURBEntries
= 2;
6199 vfe
.URBEntryAllocationSize
= 2;
6201 vfe
.CURBEAllocationSize
=
6202 ALIGN(cs_prog_data
->push
.per_thread
.regs
* cs_prog_data
->threads
+
6203 cs_prog_data
->push
.cross_thread
.regs
, 2);
6207 /* TODO: Combine subgroup-id with cbuf0 so we can push regular uniforms */
6208 if (dirty
& IRIS_DIRTY_CS
) {
6209 uint32_t curbe_data_offset
= 0;
6210 assert(cs_prog_data
->push
.cross_thread
.dwords
== 0 &&
6211 cs_prog_data
->push
.per_thread
.dwords
== 1 &&
6212 cs_prog_data
->base
.param
[0] == BRW_PARAM_BUILTIN_SUBGROUP_ID
);
6213 uint32_t *curbe_data_map
=
6214 stream_state(batch
, ice
->state
.dynamic_uploader
,
6215 &ice
->state
.last_res
.cs_thread_ids
,
6216 ALIGN(cs_prog_data
->push
.total
.size
, 64), 64,
6217 &curbe_data_offset
);
6218 assert(curbe_data_map
);
6219 memset(curbe_data_map
, 0x5a, ALIGN(cs_prog_data
->push
.total
.size
, 64));
6220 iris_fill_cs_push_const_buffer(cs_prog_data
, curbe_data_map
);
6222 iris_emit_cmd(batch
, GENX(MEDIA_CURBE_LOAD
), curbe
) {
6223 curbe
.CURBETotalDataLength
=
6224 ALIGN(cs_prog_data
->push
.total
.size
, 64);
6225 curbe
.CURBEDataStartAddress
= curbe_data_offset
;
6229 if (dirty
& (IRIS_DIRTY_SAMPLER_STATES_CS
|
6230 IRIS_DIRTY_BINDINGS_CS
|
6231 IRIS_DIRTY_CONSTANTS_CS
|
6233 uint32_t desc
[GENX(INTERFACE_DESCRIPTOR_DATA_length
)];
6235 iris_pack_state(GENX(INTERFACE_DESCRIPTOR_DATA
), desc
, idd
) {
6236 idd
.SamplerStatePointer
= shs
->sampler_table
.offset
;
6237 idd
.BindingTablePointer
= binder
->bt_offset
[MESA_SHADER_COMPUTE
];
6240 for (int i
= 0; i
< GENX(INTERFACE_DESCRIPTOR_DATA_length
); i
++)
6241 desc
[i
] |= ((uint32_t *) shader
->derived_data
)[i
];
6243 iris_emit_cmd(batch
, GENX(MEDIA_INTERFACE_DESCRIPTOR_LOAD
), load
) {
6244 load
.InterfaceDescriptorTotalLength
=
6245 GENX(INTERFACE_DESCRIPTOR_DATA_length
) * sizeof(uint32_t);
6246 load
.InterfaceDescriptorDataStartAddress
=
6247 emit_state(batch
, ice
->state
.dynamic_uploader
,
6248 &ice
->state
.last_res
.cs_desc
, desc
, sizeof(desc
), 64);
6252 uint32_t group_size
= grid
->block
[0] * grid
->block
[1] * grid
->block
[2];
6253 uint32_t remainder
= group_size
& (cs_prog_data
->simd_size
- 1);
6254 uint32_t right_mask
;
6257 right_mask
= ~0u >> (32 - remainder
);
6259 right_mask
= ~0u >> (32 - cs_prog_data
->simd_size
);
6261 #define GPGPU_DISPATCHDIMX 0x2500
6262 #define GPGPU_DISPATCHDIMY 0x2504
6263 #define GPGPU_DISPATCHDIMZ 0x2508
6265 if (grid
->indirect
) {
6266 struct iris_state_ref
*grid_size
= &ice
->state
.grid_size
;
6267 struct iris_bo
*bo
= iris_resource_bo(grid_size
->res
);
6268 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_MEM
), lrm
) {
6269 lrm
.RegisterAddress
= GPGPU_DISPATCHDIMX
;
6270 lrm
.MemoryAddress
= ro_bo(bo
, grid_size
->offset
+ 0);
6272 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_MEM
), lrm
) {
6273 lrm
.RegisterAddress
= GPGPU_DISPATCHDIMY
;
6274 lrm
.MemoryAddress
= ro_bo(bo
, grid_size
->offset
+ 4);
6276 iris_emit_cmd(batch
, GENX(MI_LOAD_REGISTER_MEM
), lrm
) {
6277 lrm
.RegisterAddress
= GPGPU_DISPATCHDIMZ
;
6278 lrm
.MemoryAddress
= ro_bo(bo
, grid_size
->offset
+ 8);
6282 iris_emit_cmd(batch
, GENX(GPGPU_WALKER
), ggw
) {
6283 ggw
.IndirectParameterEnable
= grid
->indirect
!= NULL
;
6284 ggw
.SIMDSize
= cs_prog_data
->simd_size
/ 16;
6285 ggw
.ThreadDepthCounterMaximum
= 0;
6286 ggw
.ThreadHeightCounterMaximum
= 0;
6287 ggw
.ThreadWidthCounterMaximum
= cs_prog_data
->threads
- 1;
6288 ggw
.ThreadGroupIDXDimension
= grid
->grid
[0];
6289 ggw
.ThreadGroupIDYDimension
= grid
->grid
[1];
6290 ggw
.ThreadGroupIDZDimension
= grid
->grid
[2];
6291 ggw
.RightExecutionMask
= right_mask
;
6292 ggw
.BottomExecutionMask
= 0xffffffff;
6295 iris_emit_cmd(batch
, GENX(MEDIA_STATE_FLUSH
), msf
);
6297 if (!batch
->contains_draw
) {
6298 iris_restore_compute_saved_bos(ice
, batch
, grid
);
6299 batch
->contains_draw
= true;
6304 * State module teardown.
6307 iris_destroy_state(struct iris_context
*ice
)
6309 struct iris_genx_state
*genx
= ice
->state
.genx
;
6311 pipe_resource_reference(&ice
->draw
.draw_params
.res
, NULL
);
6312 pipe_resource_reference(&ice
->draw
.derived_draw_params
.res
, NULL
);
6314 /* Loop over all VBOs, including ones for draw parameters */
6315 for (unsigned i
= 0; i
< ARRAY_SIZE(genx
->vertex_buffers
); i
++) {
6316 pipe_resource_reference(&genx
->vertex_buffers
[i
].resource
, NULL
);
6319 free(ice
->state
.genx
);
6321 for (int i
= 0; i
< 4; i
++) {
6322 pipe_so_target_reference(&ice
->state
.so_target
[i
], NULL
);
6325 for (unsigned i
= 0; i
< ice
->state
.framebuffer
.nr_cbufs
; i
++) {
6326 pipe_surface_reference(&ice
->state
.framebuffer
.cbufs
[i
], NULL
);
6328 pipe_surface_reference(&ice
->state
.framebuffer
.zsbuf
, NULL
);
6330 for (int stage
= 0; stage
< MESA_SHADER_STAGES
; stage
++) {
6331 struct iris_shader_state
*shs
= &ice
->state
.shaders
[stage
];
6332 pipe_resource_reference(&shs
->sampler_table
.res
, NULL
);
6333 for (int i
= 0; i
< PIPE_MAX_CONSTANT_BUFFERS
; i
++) {
6334 pipe_resource_reference(&shs
->constbuf
[i
].buffer
, NULL
);
6335 pipe_resource_reference(&shs
->constbuf_surf_state
[i
].res
, NULL
);
6337 for (int i
= 0; i
< PIPE_MAX_SHADER_IMAGES
; i
++) {
6338 pipe_resource_reference(&shs
->image
[i
].base
.resource
, NULL
);
6339 pipe_resource_reference(&shs
->image
[i
].surface_state
.res
, NULL
);
6341 for (int i
= 0; i
< PIPE_MAX_SHADER_BUFFERS
; i
++) {
6342 pipe_resource_reference(&shs
->ssbo
[i
].buffer
, NULL
);
6343 pipe_resource_reference(&shs
->ssbo_surf_state
[i
].res
, NULL
);
6345 for (int i
= 0; i
< IRIS_MAX_TEXTURE_SAMPLERS
; i
++) {
6346 pipe_sampler_view_reference((struct pipe_sampler_view
**)
6347 &shs
->textures
[i
], NULL
);
6351 pipe_resource_reference(&ice
->state
.grid_size
.res
, NULL
);
6352 pipe_resource_reference(&ice
->state
.grid_surf_state
.res
, NULL
);
6354 pipe_resource_reference(&ice
->state
.null_fb
.res
, NULL
);
6355 pipe_resource_reference(&ice
->state
.unbound_tex
.res
, NULL
);
6357 pipe_resource_reference(&ice
->state
.last_res
.cc_vp
, NULL
);
6358 pipe_resource_reference(&ice
->state
.last_res
.sf_cl_vp
, NULL
);
6359 pipe_resource_reference(&ice
->state
.last_res
.color_calc
, NULL
);
6360 pipe_resource_reference(&ice
->state
.last_res
.scissor
, NULL
);
6361 pipe_resource_reference(&ice
->state
.last_res
.blend
, NULL
);
6362 pipe_resource_reference(&ice
->state
.last_res
.index_buffer
, NULL
);
6363 pipe_resource_reference(&ice
->state
.last_res
.cs_thread_ids
, NULL
);
6364 pipe_resource_reference(&ice
->state
.last_res
.cs_desc
, NULL
);
6367 /* ------------------------------------------------------------------- */
6370 iris_rebind_buffer(struct iris_context
*ice
,
6371 struct iris_resource
*res
,
6372 uint64_t old_address
)
6374 struct pipe_context
*ctx
= &ice
->ctx
;
6375 struct iris_screen
*screen
= (void *) ctx
->screen
;
6376 struct iris_genx_state
*genx
= ice
->state
.genx
;
6378 assert(res
->base
.target
== PIPE_BUFFER
);
6380 /* Buffers can't be framebuffer attachments, nor display related,
6381 * and we don't have upstream Clover support.
6383 assert(!(res
->bind_history
& (PIPE_BIND_DEPTH_STENCIL
|
6384 PIPE_BIND_RENDER_TARGET
|
6385 PIPE_BIND_BLENDABLE
|
6386 PIPE_BIND_DISPLAY_TARGET
|
6388 PIPE_BIND_COMPUTE_RESOURCE
|
6389 PIPE_BIND_GLOBAL
)));
6391 if (res
->bind_history
& PIPE_BIND_VERTEX_BUFFER
) {
6392 uint64_t bound_vbs
= ice
->state
.bound_vertex_buffers
;
6394 const int i
= u_bit_scan64(&bound_vbs
);
6395 struct iris_vertex_buffer_state
*state
= &genx
->vertex_buffers
[i
];
6397 /* Update the CPU struct */
6398 STATIC_ASSERT(GENX(VERTEX_BUFFER_STATE_BufferStartingAddress_start
) == 32);
6399 STATIC_ASSERT(GENX(VERTEX_BUFFER_STATE_BufferStartingAddress_bits
) == 64);
6400 uint64_t *addr
= (uint64_t *) &state
->state
[1];
6402 if (*addr
== old_address
+ state
->offset
) {
6403 *addr
= res
->bo
->gtt_offset
+ state
->offset
;
6404 ice
->state
.dirty
|= IRIS_DIRTY_VERTEX_BUFFERS
;
6409 /* We don't need to handle PIPE_BIND_INDEX_BUFFER here: we re-emit
6410 * the 3DSTATE_INDEX_BUFFER packet whenever the address changes.
6412 * There is also no need to handle these:
6413 * - PIPE_BIND_COMMAND_ARGS_BUFFER (emitted for every indirect draw)
6414 * - PIPE_BIND_QUERY_BUFFER (no persistent state references)
6417 if (res
->bind_history
& PIPE_BIND_STREAM_OUTPUT
) {
6418 /* XXX: be careful about resetting vs appending... */
6422 for (int s
= MESA_SHADER_VERTEX
; s
< MESA_SHADER_STAGES
; s
++) {
6423 struct iris_shader_state
*shs
= &ice
->state
.shaders
[s
];
6424 enum pipe_shader_type p_stage
= stage_to_pipe(s
);
6426 if (!(res
->bind_stages
& (1 << s
)))
6429 if (res
->bind_history
& PIPE_BIND_CONSTANT_BUFFER
) {
6430 /* Skip constant buffer 0, it's for regular uniforms, not UBOs */
6431 uint32_t bound_cbufs
= shs
->bound_cbufs
& ~1u;
6432 while (bound_cbufs
) {
6433 const int i
= u_bit_scan(&bound_cbufs
);
6434 struct pipe_shader_buffer
*cbuf
= &shs
->constbuf
[i
];
6435 struct iris_state_ref
*surf_state
= &shs
->constbuf_surf_state
[i
];
6437 if (res
->bo
== iris_resource_bo(cbuf
->buffer
)) {
6438 pipe_resource_reference(&surf_state
->res
, NULL
);
6439 ice
->state
.dirty
|= IRIS_DIRTY_CONSTANTS_VS
<< s
;
6444 if (res
->bind_history
& PIPE_BIND_SHADER_BUFFER
) {
6445 uint32_t bound_ssbos
= shs
->bound_ssbos
;
6446 while (bound_ssbos
) {
6447 const int i
= u_bit_scan(&bound_ssbos
);
6448 struct pipe_shader_buffer
*ssbo
= &shs
->ssbo
[i
];
6450 if (res
->bo
== iris_resource_bo(ssbo
->buffer
)) {
6451 struct pipe_shader_buffer buf
= {
6452 .buffer
= &res
->base
,
6453 .buffer_offset
= ssbo
->buffer_offset
,
6454 .buffer_size
= ssbo
->buffer_size
,
6456 iris_set_shader_buffers(ctx
, p_stage
, i
, 1, &buf
,
6457 (shs
->writable_ssbos
>> i
) & 1);
6462 if (res
->bind_history
& PIPE_BIND_SAMPLER_VIEW
) {
6463 uint32_t bound_sampler_views
= shs
->bound_sampler_views
;
6464 while (bound_sampler_views
) {
6465 const int i
= u_bit_scan(&bound_sampler_views
);
6466 struct iris_sampler_view
*isv
= shs
->textures
[i
];
6468 if (res
->bo
== iris_resource_bo(isv
->base
.texture
)) {
6469 void *map
= alloc_surface_states(ice
->state
.surface_uploader
,
6470 &isv
->surface_state
,
6471 isv
->res
->aux
.sampler_usages
);
6473 fill_buffer_surface_state(&screen
->isl_dev
, isv
->res
, map
,
6474 isv
->view
.format
, isv
->view
.swizzle
,
6475 isv
->base
.u
.buf
.offset
,
6476 isv
->base
.u
.buf
.size
);
6477 ice
->state
.dirty
|= IRIS_DIRTY_BINDINGS_VS
<< s
;
6482 if (res
->bind_history
& PIPE_BIND_SHADER_IMAGE
) {
6483 uint32_t bound_image_views
= shs
->bound_image_views
;
6484 while (bound_image_views
) {
6485 const int i
= u_bit_scan(&bound_image_views
);
6486 struct iris_image_view
*iv
= &shs
->image
[i
];
6488 if (res
->bo
== iris_resource_bo(iv
->base
.resource
)) {
6489 iris_set_shader_images(ctx
, p_stage
, i
, 1, &iv
->base
);
6496 /* ------------------------------------------------------------------- */
6499 flags_to_post_sync_op(uint32_t flags
)
6501 if (flags
& PIPE_CONTROL_WRITE_IMMEDIATE
)
6502 return WriteImmediateData
;
6504 if (flags
& PIPE_CONTROL_WRITE_DEPTH_COUNT
)
6505 return WritePSDepthCount
;
6507 if (flags
& PIPE_CONTROL_WRITE_TIMESTAMP
)
6508 return WriteTimestamp
;
6514 * Do the given flags have a Post Sync or LRI Post Sync operation?
6516 static enum pipe_control_flags
6517 get_post_sync_flags(enum pipe_control_flags flags
)
6519 flags
&= PIPE_CONTROL_WRITE_IMMEDIATE
|
6520 PIPE_CONTROL_WRITE_DEPTH_COUNT
|
6521 PIPE_CONTROL_WRITE_TIMESTAMP
|
6522 PIPE_CONTROL_LRI_POST_SYNC_OP
;
6524 /* Only one "Post Sync Op" is allowed, and it's mutually exclusive with
6525 * "LRI Post Sync Operation". So more than one bit set would be illegal.
6527 assert(util_bitcount(flags
) <= 1);
6532 #define IS_COMPUTE_PIPELINE(batch) (batch->name == IRIS_BATCH_COMPUTE)
6535 * Emit a series of PIPE_CONTROL commands, taking into account any
6536 * workarounds necessary to actually accomplish the caller's request.
6538 * Unless otherwise noted, spec quotations in this function come from:
6540 * Synchronization of the 3D Pipeline > PIPE_CONTROL Command > Programming
6541 * Restrictions for PIPE_CONTROL.
6543 * You should not use this function directly. Use the helpers in
6544 * iris_pipe_control.c instead, which may split the pipe control further.
6547 iris_emit_raw_pipe_control(struct iris_batch
*batch
,
6554 UNUSED
const struct gen_device_info
*devinfo
= &batch
->screen
->devinfo
;
6555 enum pipe_control_flags post_sync_flags
= get_post_sync_flags(flags
);
6556 enum pipe_control_flags non_lri_post_sync_flags
=
6557 post_sync_flags
& ~PIPE_CONTROL_LRI_POST_SYNC_OP
;
6559 /* Recursive PIPE_CONTROL workarounds --------------------------------
6560 * (http://knowyourmeme.com/memes/xzibit-yo-dawg)
6562 * We do these first because we want to look at the original operation,
6563 * rather than any workarounds we set.
6565 if (GEN_GEN
== 9 && (flags
& PIPE_CONTROL_VF_CACHE_INVALIDATE
)) {
6566 /* The PIPE_CONTROL "VF Cache Invalidation Enable" bit description
6567 * lists several workarounds:
6569 * "Project: SKL, KBL, BXT
6571 * If the VF Cache Invalidation Enable is set to a 1 in a
6572 * PIPE_CONTROL, a separate Null PIPE_CONTROL, all bitfields
6573 * sets to 0, with the VF Cache Invalidation Enable set to 0
6574 * needs to be sent prior to the PIPE_CONTROL with VF Cache
6575 * Invalidation Enable set to a 1."
6577 iris_emit_raw_pipe_control(batch
,
6578 "workaround: recursive VF cache invalidate",
6582 if (GEN_GEN
== 9 && IS_COMPUTE_PIPELINE(batch
) && post_sync_flags
) {
6583 /* Project: SKL / Argument: LRI Post Sync Operation [23]
6585 * "PIPECONTROL command with “Command Streamer Stall Enable” must be
6586 * programmed prior to programming a PIPECONTROL command with "LRI
6587 * Post Sync Operation" in GPGPU mode of operation (i.e when
6588 * PIPELINE_SELECT command is set to GPGPU mode of operation)."
6590 * The same text exists a few rows below for Post Sync Op.
6592 iris_emit_raw_pipe_control(batch
,
6593 "workaround: CS stall before gpgpu post-sync",
6594 PIPE_CONTROL_CS_STALL
, bo
, offset
, imm
);
6597 if (GEN_GEN
== 10 && (flags
& PIPE_CONTROL_RENDER_TARGET_FLUSH
)) {
6599 * "Before sending a PIPE_CONTROL command with bit 12 set, SW must issue
6600 * another PIPE_CONTROL with Render Target Cache Flush Enable (bit 12)
6601 * = 0 and Pipe Control Flush Enable (bit 7) = 1"
6603 iris_emit_raw_pipe_control(batch
,
6604 "workaround: PC flush before RT flush",
6605 PIPE_CONTROL_FLUSH_ENABLE
, bo
, offset
, imm
);
6608 /* "Flush Types" workarounds ---------------------------------------------
6609 * We do these now because they may add post-sync operations or CS stalls.
6612 if (GEN_GEN
< 11 && flags
& PIPE_CONTROL_VF_CACHE_INVALIDATE
) {
6613 /* Project: BDW, SKL+ (stopping at CNL) / Argument: VF Invalidate
6615 * "'Post Sync Operation' must be enabled to 'Write Immediate Data' or
6616 * 'Write PS Depth Count' or 'Write Timestamp'."
6619 flags
|= PIPE_CONTROL_WRITE_IMMEDIATE
;
6620 post_sync_flags
|= PIPE_CONTROL_WRITE_IMMEDIATE
;
6621 non_lri_post_sync_flags
|= PIPE_CONTROL_WRITE_IMMEDIATE
;
6622 bo
= batch
->screen
->workaround_bo
;
6626 /* #1130 from Gen10 workarounds page:
6628 * "Enable Depth Stall on every Post Sync Op if Render target Cache
6629 * Flush is not enabled in same PIPE CONTROL and Enable Pixel score
6630 * board stall if Render target cache flush is enabled."
6632 * Applicable to CNL B0 and C0 steppings only.
6634 * The wording here is unclear, and this workaround doesn't look anything
6635 * like the internal bug report recommendations, but leave it be for now...
6637 if (GEN_GEN
== 10) {
6638 if (flags
& PIPE_CONTROL_RENDER_TARGET_FLUSH
) {
6639 flags
|= PIPE_CONTROL_STALL_AT_SCOREBOARD
;
6640 } else if (flags
& non_lri_post_sync_flags
) {
6641 flags
|= PIPE_CONTROL_DEPTH_STALL
;
6645 if (flags
& PIPE_CONTROL_DEPTH_STALL
) {
6646 /* From the PIPE_CONTROL instruction table, bit 13 (Depth Stall Enable):
6648 * "This bit must be DISABLED for operations other than writing
6651 * This seems like nonsense. An Ivybridge workaround requires us to
6652 * emit a PIPE_CONTROL with a depth stall and write immediate post-sync
6653 * operation. Gen8+ requires us to emit depth stalls and depth cache
6654 * flushes together. So, it's hard to imagine this means anything other
6655 * than "we originally intended this to be used for PS_DEPTH_COUNT".
6657 * We ignore the supposed restriction and do nothing.
6661 if (flags
& (PIPE_CONTROL_RENDER_TARGET_FLUSH
|
6662 PIPE_CONTROL_STALL_AT_SCOREBOARD
)) {
6663 /* From the PIPE_CONTROL instruction table, bit 12 and bit 1:
6665 * "This bit must be DISABLED for End-of-pipe (Read) fences,
6666 * PS_DEPTH_COUNT or TIMESTAMP queries."
6668 * TODO: Implement end-of-pipe checking.
6670 assert(!(post_sync_flags
& (PIPE_CONTROL_WRITE_DEPTH_COUNT
|
6671 PIPE_CONTROL_WRITE_TIMESTAMP
)));
6674 if (GEN_GEN
< 11 && (flags
& PIPE_CONTROL_STALL_AT_SCOREBOARD
)) {
6675 /* From the PIPE_CONTROL instruction table, bit 1:
6677 * "This bit is ignored if Depth Stall Enable is set.
6678 * Further, the render cache is not flushed even if Write Cache
6679 * Flush Enable bit is set."
6681 * We assert that the caller doesn't do this combination, to try and
6682 * prevent mistakes. It shouldn't hurt the GPU, though.
6684 * We skip this check on Gen11+ as the "Stall at Pixel Scoreboard"
6685 * and "Render Target Flush" combo is explicitly required for BTI
6686 * update workarounds.
6688 assert(!(flags
& (PIPE_CONTROL_DEPTH_STALL
|
6689 PIPE_CONTROL_RENDER_TARGET_FLUSH
)));
6692 /* PIPE_CONTROL page workarounds ------------------------------------- */
6694 if (GEN_GEN
<= 8 && (flags
& PIPE_CONTROL_STATE_CACHE_INVALIDATE
)) {
6695 /* From the PIPE_CONTROL page itself:
6698 * Restriction: Pipe_control with CS-stall bit set must be issued
6699 * before a pipe-control command that has the State Cache
6700 * Invalidate bit set."
6702 flags
|= PIPE_CONTROL_CS_STALL
;
6705 if (flags
& PIPE_CONTROL_FLUSH_LLC
) {
6706 /* From the PIPE_CONTROL instruction table, bit 26 (Flush LLC):
6709 * SW must always program Post-Sync Operation to "Write Immediate
6710 * Data" when Flush LLC is set."
6712 * For now, we just require the caller to do it.
6714 assert(flags
& PIPE_CONTROL_WRITE_IMMEDIATE
);
6717 /* "Post-Sync Operation" workarounds -------------------------------- */
6719 /* Project: All / Argument: Global Snapshot Count Reset [19]
6721 * "This bit must not be exercised on any product.
6722 * Requires stall bit ([20] of DW1) set."
6724 * We don't use this, so we just assert that it isn't used. The
6725 * PIPE_CONTROL instruction page indicates that they intended this
6726 * as a debug feature and don't think it is useful in production,
6727 * but it may actually be usable, should we ever want to.
6729 assert((flags
& PIPE_CONTROL_GLOBAL_SNAPSHOT_COUNT_RESET
) == 0);
6731 if (flags
& (PIPE_CONTROL_MEDIA_STATE_CLEAR
|
6732 PIPE_CONTROL_INDIRECT_STATE_POINTERS_DISABLE
)) {
6733 /* Project: All / Arguments:
6735 * - Generic Media State Clear [16]
6736 * - Indirect State Pointers Disable [16]
6738 * "Requires stall bit ([20] of DW1) set."
6740 * Also, the PIPE_CONTROL instruction table, bit 16 (Generic Media
6741 * State Clear) says:
6743 * "PIPECONTROL command with “Command Streamer Stall Enable” must be
6744 * programmed prior to programming a PIPECONTROL command with "Media
6745 * State Clear" set in GPGPU mode of operation"
6747 * This is a subset of the earlier rule, so there's nothing to do.
6749 flags
|= PIPE_CONTROL_CS_STALL
;
6752 if (flags
& PIPE_CONTROL_STORE_DATA_INDEX
) {
6753 /* Project: All / Argument: Store Data Index
6755 * "Post-Sync Operation ([15:14] of DW1) must be set to something other
6758 * For now, we just assert that the caller does this. We might want to
6759 * automatically add a write to the workaround BO...
6761 assert(non_lri_post_sync_flags
!= 0);
6764 if (flags
& PIPE_CONTROL_SYNC_GFDT
) {
6765 /* Project: All / Argument: Sync GFDT
6767 * "Post-Sync Operation ([15:14] of DW1) must be set to something other
6768 * than '0' or 0x2520[13] must be set."
6770 * For now, we just assert that the caller does this.
6772 assert(non_lri_post_sync_flags
!= 0);
6775 if (flags
& PIPE_CONTROL_TLB_INVALIDATE
) {
6776 /* Project: IVB+ / Argument: TLB inv
6778 * "Requires stall bit ([20] of DW1) set."
6780 * Also, from the PIPE_CONTROL instruction table:
6783 * Post Sync Operation or CS stall must be set to ensure a TLB
6784 * invalidation occurs. Otherwise no cycle will occur to the TLB
6785 * cache to invalidate."
6787 * This is not a subset of the earlier rule, so there's nothing to do.
6789 flags
|= PIPE_CONTROL_CS_STALL
;
6792 if (GEN_GEN
>= 12 && ((flags
& PIPE_CONTROL_RENDER_TARGET_FLUSH
) ||
6793 (flags
& PIPE_CONTROL_DEPTH_CACHE_FLUSH
))) {
6794 /* From the PIPE_CONTROL instruction table, bit 28 (Tile Cache Flush
6797 * Unified Cache (Tile Cache Disabled):
6799 * When the Color and Depth (Z) streams are enabled to be cached in
6800 * the DC space of L2, Software must use "Render Target Cache Flush
6801 * Enable" and "Depth Cache Flush Enable" along with "Tile Cache
6802 * Flush" for getting the color and depth (Z) write data to be
6803 * globally observable. In this mode of operation it is not required
6804 * to set "CS Stall" upon setting "Tile Cache Flush" bit.
6806 flags
|= PIPE_CONTROL_TILE_CACHE_FLUSH
;
6809 if (GEN_GEN
== 9 && devinfo
->gt
== 4) {
6810 /* TODO: The big Skylake GT4 post sync op workaround */
6813 /* "GPGPU specific workarounds" (both post-sync and flush) ------------ */
6815 if (IS_COMPUTE_PIPELINE(batch
)) {
6816 if (GEN_GEN
>= 9 && (flags
& PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE
)) {
6817 /* Project: SKL+ / Argument: Tex Invalidate
6818 * "Requires stall bit ([20] of DW) set for all GPGPU Workloads."
6820 flags
|= PIPE_CONTROL_CS_STALL
;
6823 if (GEN_GEN
== 8 && (post_sync_flags
||
6824 (flags
& (PIPE_CONTROL_NOTIFY_ENABLE
|
6825 PIPE_CONTROL_DEPTH_STALL
|
6826 PIPE_CONTROL_RENDER_TARGET_FLUSH
|
6827 PIPE_CONTROL_DEPTH_CACHE_FLUSH
|
6828 PIPE_CONTROL_DATA_CACHE_FLUSH
)))) {
6829 /* Project: BDW / Arguments:
6831 * - LRI Post Sync Operation [23]
6832 * - Post Sync Op [15:14]
6834 * - Depth Stall [13]
6835 * - Render Target Cache Flush [12]
6836 * - Depth Cache Flush [0]
6837 * - DC Flush Enable [5]
6839 * "Requires stall bit ([20] of DW) set for all GPGPU and Media
6842 flags
|= PIPE_CONTROL_CS_STALL
;
6844 /* Also, from the PIPE_CONTROL instruction table, bit 20:
6847 * This bit must be always set when PIPE_CONTROL command is
6848 * programmed by GPGPU and MEDIA workloads, except for the cases
6849 * when only Read Only Cache Invalidation bits are set (State
6850 * Cache Invalidation Enable, Instruction cache Invalidation
6851 * Enable, Texture Cache Invalidation Enable, Constant Cache
6852 * Invalidation Enable). This is to WA FFDOP CG issue, this WA
6853 * need not implemented when FF_DOP_CG is disable via "Fixed
6854 * Function DOP Clock Gate Disable" bit in RC_PSMI_CTRL register."
6856 * It sounds like we could avoid CS stalls in some cases, but we
6857 * don't currently bother. This list isn't exactly the list above,
6863 /* "Stall" workarounds ----------------------------------------------
6864 * These have to come after the earlier ones because we may have added
6865 * some additional CS stalls above.
6868 if (GEN_GEN
< 9 && (flags
& PIPE_CONTROL_CS_STALL
)) {
6869 /* Project: PRE-SKL, VLV, CHV
6871 * "[All Stepping][All SKUs]:
6873 * One of the following must also be set:
6875 * - Render Target Cache Flush Enable ([12] of DW1)
6876 * - Depth Cache Flush Enable ([0] of DW1)
6877 * - Stall at Pixel Scoreboard ([1] of DW1)
6878 * - Depth Stall ([13] of DW1)
6879 * - Post-Sync Operation ([13] of DW1)
6880 * - DC Flush Enable ([5] of DW1)"
6882 * If we don't already have one of those bits set, we choose to add
6883 * "Stall at Pixel Scoreboard". Some of the other bits require a
6884 * CS stall as a workaround (see above), which would send us into
6885 * an infinite recursion of PIPE_CONTROLs. "Stall at Pixel Scoreboard"
6886 * appears to be safe, so we choose that.
6888 const uint32_t wa_bits
= PIPE_CONTROL_RENDER_TARGET_FLUSH
|
6889 PIPE_CONTROL_DEPTH_CACHE_FLUSH
|
6890 PIPE_CONTROL_WRITE_IMMEDIATE
|
6891 PIPE_CONTROL_WRITE_DEPTH_COUNT
|
6892 PIPE_CONTROL_WRITE_TIMESTAMP
|
6893 PIPE_CONTROL_STALL_AT_SCOREBOARD
|
6894 PIPE_CONTROL_DEPTH_STALL
|
6895 PIPE_CONTROL_DATA_CACHE_FLUSH
;
6896 if (!(flags
& wa_bits
))
6897 flags
|= PIPE_CONTROL_STALL_AT_SCOREBOARD
;
6900 /* Emit --------------------------------------------------------------- */
6902 if (INTEL_DEBUG
& DEBUG_PIPE_CONTROL
) {
6904 " PC [%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%"PRIx64
"]: %s\n",
6905 (flags
& PIPE_CONTROL_FLUSH_ENABLE
) ? "PipeCon " : "",
6906 (flags
& PIPE_CONTROL_CS_STALL
) ? "CS " : "",
6907 (flags
& PIPE_CONTROL_STALL_AT_SCOREBOARD
) ? "Scoreboard " : "",
6908 (flags
& PIPE_CONTROL_VF_CACHE_INVALIDATE
) ? "VF " : "",
6909 (flags
& PIPE_CONTROL_RENDER_TARGET_FLUSH
) ? "RT " : "",
6910 (flags
& PIPE_CONTROL_CONST_CACHE_INVALIDATE
) ? "Const " : "",
6911 (flags
& PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE
) ? "TC " : "",
6912 (flags
& PIPE_CONTROL_DATA_CACHE_FLUSH
) ? "DC " : "",
6913 (flags
& PIPE_CONTROL_DEPTH_CACHE_FLUSH
) ? "ZFlush " : "",
6914 (flags
& PIPE_CONTROL_DEPTH_STALL
) ? "ZStall " : "",
6915 (flags
& PIPE_CONTROL_STATE_CACHE_INVALIDATE
) ? "State " : "",
6916 (flags
& PIPE_CONTROL_TLB_INVALIDATE
) ? "TLB " : "",
6917 (flags
& PIPE_CONTROL_INSTRUCTION_INVALIDATE
) ? "Inst " : "",
6918 (flags
& PIPE_CONTROL_MEDIA_STATE_CLEAR
) ? "MediaClear " : "",
6919 (flags
& PIPE_CONTROL_NOTIFY_ENABLE
) ? "Notify " : "",
6920 (flags
& PIPE_CONTROL_GLOBAL_SNAPSHOT_COUNT_RESET
) ?
6922 (flags
& PIPE_CONTROL_INDIRECT_STATE_POINTERS_DISABLE
) ?
6924 (flags
& PIPE_CONTROL_WRITE_IMMEDIATE
) ? "WriteImm " : "",
6925 (flags
& PIPE_CONTROL_WRITE_DEPTH_COUNT
) ? "WriteZCount " : "",
6926 (flags
& PIPE_CONTROL_WRITE_TIMESTAMP
) ? "WriteTimestamp " : "",
6930 iris_emit_cmd(batch
, GENX(PIPE_CONTROL
), pc
) {
6932 pc
.TileCacheFlushEnable
= flags
& PIPE_CONTROL_TILE_CACHE_FLUSH
;
6934 pc
.LRIPostSyncOperation
= NoLRIOperation
;
6935 pc
.PipeControlFlushEnable
= flags
& PIPE_CONTROL_FLUSH_ENABLE
;
6936 pc
.DCFlushEnable
= flags
& PIPE_CONTROL_DATA_CACHE_FLUSH
;
6937 pc
.StoreDataIndex
= 0;
6938 pc
.CommandStreamerStallEnable
= flags
& PIPE_CONTROL_CS_STALL
;
6939 pc
.GlobalSnapshotCountReset
=
6940 flags
& PIPE_CONTROL_GLOBAL_SNAPSHOT_COUNT_RESET
;
6941 pc
.TLBInvalidate
= flags
& PIPE_CONTROL_TLB_INVALIDATE
;
6942 pc
.GenericMediaStateClear
= flags
& PIPE_CONTROL_MEDIA_STATE_CLEAR
;
6943 pc
.StallAtPixelScoreboard
= flags
& PIPE_CONTROL_STALL_AT_SCOREBOARD
;
6944 pc
.RenderTargetCacheFlushEnable
=
6945 flags
& PIPE_CONTROL_RENDER_TARGET_FLUSH
;
6946 pc
.DepthCacheFlushEnable
= flags
& PIPE_CONTROL_DEPTH_CACHE_FLUSH
;
6947 pc
.StateCacheInvalidationEnable
=
6948 flags
& PIPE_CONTROL_STATE_CACHE_INVALIDATE
;
6949 pc
.VFCacheInvalidationEnable
= flags
& PIPE_CONTROL_VF_CACHE_INVALIDATE
;
6950 pc
.ConstantCacheInvalidationEnable
=
6951 flags
& PIPE_CONTROL_CONST_CACHE_INVALIDATE
;
6952 pc
.PostSyncOperation
= flags_to_post_sync_op(flags
);
6953 pc
.DepthStallEnable
= flags
& PIPE_CONTROL_DEPTH_STALL
;
6954 pc
.InstructionCacheInvalidateEnable
=
6955 flags
& PIPE_CONTROL_INSTRUCTION_INVALIDATE
;
6956 pc
.NotifyEnable
= flags
& PIPE_CONTROL_NOTIFY_ENABLE
;
6957 pc
.IndirectStatePointersDisable
=
6958 flags
& PIPE_CONTROL_INDIRECT_STATE_POINTERS_DISABLE
;
6959 pc
.TextureCacheInvalidationEnable
=
6960 flags
& PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE
;
6961 pc
.Address
= rw_bo(bo
, offset
);
6962 pc
.ImmediateData
= imm
;
6967 genX(emit_urb_setup
)(struct iris_context
*ice
,
6968 struct iris_batch
*batch
,
6969 const unsigned size
[4],
6970 bool tess_present
, bool gs_present
)
6972 const struct gen_device_info
*devinfo
= &batch
->screen
->devinfo
;
6973 const unsigned push_size_kB
= 32;
6974 unsigned entries
[4];
6977 ice
->shaders
.last_vs_entry_size
= size
[MESA_SHADER_VERTEX
];
6979 gen_get_urb_config(devinfo
, 1024 * push_size_kB
,
6980 1024 * ice
->shaders
.urb_size
,
6981 tess_present
, gs_present
,
6982 size
, entries
, start
);
6984 for (int i
= MESA_SHADER_VERTEX
; i
<= MESA_SHADER_GEOMETRY
; i
++) {
6985 iris_emit_cmd(batch
, GENX(3DSTATE_URB_VS
), urb
) {
6986 urb
._3DCommandSubOpcode
+= i
;
6987 urb
.VSURBStartingAddress
= start
[i
];
6988 urb
.VSURBEntryAllocationSize
= size
[i
] - 1;
6989 urb
.VSNumberofURBEntries
= entries
[i
];
6996 * Preemption on Gen9 has to be enabled or disabled in various cases.
6998 * See these workarounds for preemption:
6999 * - WaDisableMidObjectPreemptionForGSLineStripAdj
7000 * - WaDisableMidObjectPreemptionForTrifanOrPolygon
7001 * - WaDisableMidObjectPreemptionForLineLoop
7004 * We don't put this in the vtable because it's only used on Gen9.
7007 gen9_toggle_preemption(struct iris_context
*ice
,
7008 struct iris_batch
*batch
,
7009 const struct pipe_draw_info
*draw
)
7011 struct iris_genx_state
*genx
= ice
->state
.genx
;
7012 bool object_preemption
= true;
7014 /* WaDisableMidObjectPreemptionForGSLineStripAdj
7016 * "WA: Disable mid-draw preemption when draw-call is a linestrip_adj
7017 * and GS is enabled."
7019 if (draw
->mode
== PIPE_PRIM_LINE_STRIP_ADJACENCY
&&
7020 ice
->shaders
.prog
[MESA_SHADER_GEOMETRY
])
7021 object_preemption
= false;
7023 /* WaDisableMidObjectPreemptionForTrifanOrPolygon
7025 * "TriFan miscompare in Execlist Preemption test. Cut index that is
7026 * on a previous context. End the previous, the resume another context
7027 * with a tri-fan or polygon, and the vertex count is corrupted. If we
7028 * prempt again we will cause corruption.
7030 * WA: Disable mid-draw preemption when draw-call has a tri-fan."
7032 if (draw
->mode
== PIPE_PRIM_TRIANGLE_FAN
)
7033 object_preemption
= false;
7035 /* WaDisableMidObjectPreemptionForLineLoop
7037 * "VF Stats Counters Missing a vertex when preemption enabled.
7039 * WA: Disable mid-draw preemption when the draw uses a lineloop
7042 if (draw
->mode
== PIPE_PRIM_LINE_LOOP
)
7043 object_preemption
= false;
7047 * "VF is corrupting GAFS data when preempted on an instance boundary
7048 * and replayed with instancing enabled.
7050 * WA: Disable preemption when using instanceing."
7052 if (draw
->instance_count
> 1)
7053 object_preemption
= false;
7055 if (genx
->object_preemption
!= object_preemption
) {
7056 iris_enable_obj_preemption(batch
, object_preemption
);
7057 genx
->object_preemption
= object_preemption
;
7063 iris_lost_genx_state(struct iris_context
*ice
, struct iris_batch
*batch
)
7065 struct iris_genx_state
*genx
= ice
->state
.genx
;
7067 memset(genx
->last_index_buffer
, 0, sizeof(genx
->last_index_buffer
));
7071 iris_emit_mi_report_perf_count(struct iris_batch
*batch
,
7073 uint32_t offset_in_bytes
,
7076 iris_emit_cmd(batch
, GENX(MI_REPORT_PERF_COUNT
), mi_rpc
) {
7077 mi_rpc
.MemoryAddress
= rw_bo(bo
, offset_in_bytes
);
7078 mi_rpc
.ReportID
= report_id
;
7083 * Update the pixel hashing modes that determine the balancing of PS threads
7084 * across subslices and slices.
7086 * \param width Width bound of the rendering area (already scaled down if \p
7087 * scale is greater than 1).
7088 * \param height Height bound of the rendering area (already scaled down if \p
7089 * scale is greater than 1).
7090 * \param scale The number of framebuffer samples that could potentially be
7091 * affected by an individual channel of the PS thread. This is
7092 * typically one for single-sampled rendering, but for operations
7093 * like CCS resolves and fast clears a single PS invocation may
7094 * update a huge number of pixels, in which case a finer
7095 * balancing is desirable in order to maximally utilize the
7096 * bandwidth available. UINT_MAX can be used as shorthand for
7097 * "finest hashing mode available".
7100 genX(emit_hashing_mode
)(struct iris_context
*ice
, struct iris_batch
*batch
,
7101 unsigned width
, unsigned height
, unsigned scale
)
7104 const struct gen_device_info
*devinfo
= &batch
->screen
->devinfo
;
7105 const unsigned slice_hashing
[] = {
7106 /* Because all Gen9 platforms with more than one slice require
7107 * three-way subslice hashing, a single "normal" 16x16 slice hashing
7108 * block is guaranteed to suffer from substantial imbalance, with one
7109 * subslice receiving twice as much work as the other two in the
7112 * The performance impact of that would be particularly severe when
7113 * three-way hashing is also in use for slice balancing (which is the
7114 * case for all Gen9 GT4 platforms), because one of the slices
7115 * receives one every three 16x16 blocks in either direction, which
7116 * is roughly the periodicity of the underlying subslice imbalance
7117 * pattern ("roughly" because in reality the hardware's
7118 * implementation of three-way hashing doesn't do exact modulo 3
7119 * arithmetic, which somewhat decreases the magnitude of this effect
7120 * in practice). This leads to a systematic subslice imbalance
7121 * within that slice regardless of the size of the primitive. The
7122 * 32x32 hashing mode guarantees that the subslice imbalance within a
7123 * single slice hashing block is minimal, largely eliminating this
7127 /* Finest slice hashing mode available. */
7130 const unsigned subslice_hashing
[] = {
7131 /* 16x16 would provide a slight cache locality benefit especially
7132 * visible in the sampler L1 cache efficiency of low-bandwidth
7133 * non-LLC platforms, but it comes at the cost of greater subslice
7134 * imbalance for primitives of dimensions approximately intermediate
7135 * between 16x4 and 16x16.
7138 /* Finest subslice hashing mode available. */
7141 /* Dimensions of the smallest hashing block of a given hashing mode. If
7142 * the rendering area is smaller than this there can't possibly be any
7143 * benefit from switching to this mode, so we optimize out the
7146 const unsigned min_size
[][2] = {
7150 const unsigned idx
= scale
> 1;
7152 if (width
> min_size
[idx
][0] || height
> min_size
[idx
][1]) {
7155 iris_pack_state(GENX(GT_MODE
), >_mode
, reg
) {
7156 reg
.SliceHashing
= (devinfo
->num_slices
> 1 ? slice_hashing
[idx
] : 0);
7157 reg
.SliceHashingMask
= (devinfo
->num_slices
> 1 ? -1 : 0);
7158 reg
.SubsliceHashing
= subslice_hashing
[idx
];
7159 reg
.SubsliceHashingMask
= -1;
7162 iris_emit_raw_pipe_control(batch
,
7163 "workaround: CS stall before GT_MODE LRI",
7164 PIPE_CONTROL_STALL_AT_SCOREBOARD
|
7165 PIPE_CONTROL_CS_STALL
,
7168 iris_emit_lri(batch
, GT_MODE
, gt_mode
);
7170 ice
->state
.current_hash_scale
= scale
;
7176 genX(init_state
)(struct iris_context
*ice
)
7178 struct pipe_context
*ctx
= &ice
->ctx
;
7179 struct iris_screen
*screen
= (struct iris_screen
*)ctx
->screen
;
7181 ctx
->create_blend_state
= iris_create_blend_state
;
7182 ctx
->create_depth_stencil_alpha_state
= iris_create_zsa_state
;
7183 ctx
->create_rasterizer_state
= iris_create_rasterizer_state
;
7184 ctx
->create_sampler_state
= iris_create_sampler_state
;
7185 ctx
->create_sampler_view
= iris_create_sampler_view
;
7186 ctx
->create_surface
= iris_create_surface
;
7187 ctx
->create_vertex_elements_state
= iris_create_vertex_elements
;
7188 ctx
->bind_blend_state
= iris_bind_blend_state
;
7189 ctx
->bind_depth_stencil_alpha_state
= iris_bind_zsa_state
;
7190 ctx
->bind_sampler_states
= iris_bind_sampler_states
;
7191 ctx
->bind_rasterizer_state
= iris_bind_rasterizer_state
;
7192 ctx
->bind_vertex_elements_state
= iris_bind_vertex_elements_state
;
7193 ctx
->delete_blend_state
= iris_delete_state
;
7194 ctx
->delete_depth_stencil_alpha_state
= iris_delete_state
;
7195 ctx
->delete_rasterizer_state
= iris_delete_state
;
7196 ctx
->delete_sampler_state
= iris_delete_state
;
7197 ctx
->delete_vertex_elements_state
= iris_delete_state
;
7198 ctx
->set_blend_color
= iris_set_blend_color
;
7199 ctx
->set_clip_state
= iris_set_clip_state
;
7200 ctx
->set_constant_buffer
= iris_set_constant_buffer
;
7201 ctx
->set_shader_buffers
= iris_set_shader_buffers
;
7202 ctx
->set_shader_images
= iris_set_shader_images
;
7203 ctx
->set_sampler_views
= iris_set_sampler_views
;
7204 ctx
->set_tess_state
= iris_set_tess_state
;
7205 ctx
->set_framebuffer_state
= iris_set_framebuffer_state
;
7206 ctx
->set_polygon_stipple
= iris_set_polygon_stipple
;
7207 ctx
->set_sample_mask
= iris_set_sample_mask
;
7208 ctx
->set_scissor_states
= iris_set_scissor_states
;
7209 ctx
->set_stencil_ref
= iris_set_stencil_ref
;
7210 ctx
->set_vertex_buffers
= iris_set_vertex_buffers
;
7211 ctx
->set_viewport_states
= iris_set_viewport_states
;
7212 ctx
->sampler_view_destroy
= iris_sampler_view_destroy
;
7213 ctx
->surface_destroy
= iris_surface_destroy
;
7214 ctx
->draw_vbo
= iris_draw_vbo
;
7215 ctx
->launch_grid
= iris_launch_grid
;
7216 ctx
->create_stream_output_target
= iris_create_stream_output_target
;
7217 ctx
->stream_output_target_destroy
= iris_stream_output_target_destroy
;
7218 ctx
->set_stream_output_targets
= iris_set_stream_output_targets
;
7220 ice
->vtbl
.destroy_state
= iris_destroy_state
;
7221 ice
->vtbl
.init_render_context
= iris_init_render_context
;
7222 ice
->vtbl
.init_compute_context
= iris_init_compute_context
;
7223 ice
->vtbl
.upload_render_state
= iris_upload_render_state
;
7224 ice
->vtbl
.update_surface_base_address
= iris_update_surface_base_address
;
7225 ice
->vtbl
.upload_compute_state
= iris_upload_compute_state
;
7226 ice
->vtbl
.emit_raw_pipe_control
= iris_emit_raw_pipe_control
;
7227 ice
->vtbl
.emit_mi_report_perf_count
= iris_emit_mi_report_perf_count
;
7228 ice
->vtbl
.rebind_buffer
= iris_rebind_buffer
;
7229 ice
->vtbl
.load_register_reg32
= iris_load_register_reg32
;
7230 ice
->vtbl
.load_register_reg64
= iris_load_register_reg64
;
7231 ice
->vtbl
.load_register_imm32
= iris_load_register_imm32
;
7232 ice
->vtbl
.load_register_imm64
= iris_load_register_imm64
;
7233 ice
->vtbl
.load_register_mem32
= iris_load_register_mem32
;
7234 ice
->vtbl
.load_register_mem64
= iris_load_register_mem64
;
7235 ice
->vtbl
.store_register_mem32
= iris_store_register_mem32
;
7236 ice
->vtbl
.store_register_mem64
= iris_store_register_mem64
;
7237 ice
->vtbl
.store_data_imm32
= iris_store_data_imm32
;
7238 ice
->vtbl
.store_data_imm64
= iris_store_data_imm64
;
7239 ice
->vtbl
.copy_mem_mem
= iris_copy_mem_mem
;
7240 ice
->vtbl
.derived_program_state_size
= iris_derived_program_state_size
;
7241 ice
->vtbl
.store_derived_program_state
= iris_store_derived_program_state
;
7242 ice
->vtbl
.create_so_decl_list
= iris_create_so_decl_list
;
7243 ice
->vtbl
.populate_vs_key
= iris_populate_vs_key
;
7244 ice
->vtbl
.populate_tcs_key
= iris_populate_tcs_key
;
7245 ice
->vtbl
.populate_tes_key
= iris_populate_tes_key
;
7246 ice
->vtbl
.populate_gs_key
= iris_populate_gs_key
;
7247 ice
->vtbl
.populate_fs_key
= iris_populate_fs_key
;
7248 ice
->vtbl
.populate_cs_key
= iris_populate_cs_key
;
7249 ice
->vtbl
.mocs
= mocs
;
7250 ice
->vtbl
.lost_genx_state
= iris_lost_genx_state
;
7252 ice
->state
.dirty
= ~0ull;
7254 ice
->state
.statistics_counters_enabled
= true;
7256 ice
->state
.sample_mask
= 0xffff;
7257 ice
->state
.num_viewports
= 1;
7258 ice
->state
.prim_mode
= PIPE_PRIM_MAX
;
7259 ice
->state
.genx
= calloc(1, sizeof(struct iris_genx_state
));
7260 ice
->draw
.derived_params
.drawid
= -1;
7262 /* Make a 1x1x1 null surface for unbound textures */
7263 void *null_surf_map
=
7264 upload_state(ice
->state
.surface_uploader
, &ice
->state
.unbound_tex
,
7265 4 * GENX(RENDER_SURFACE_STATE_length
), 64);
7266 isl_null_fill_state(&screen
->isl_dev
, null_surf_map
, isl_extent3d(1, 1, 1));
7267 ice
->state
.unbound_tex
.offset
+=
7268 iris_bo_offset_from_base_address(iris_resource_bo(ice
->state
.unbound_tex
.res
));
7270 /* Default all scissor rectangles to be empty regions. */
7271 for (int i
= 0; i
< IRIS_MAX_VIEWPORTS
; i
++) {
7272 ice
->state
.scissors
[i
] = (struct pipe_scissor_state
) {
7273 .minx
= 1, .maxx
= 0, .miny
= 1, .maxy
= 0,