1/* 2 * Copyright (c) 2012-2015 Etnaviv Project 3 * 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, sub license, 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: 10 * 11 * The above copyright notice and this permission notice (including the 12 * next paragraph) shall be included in all copies or substantial portions 13 * of the Software. 14 * 15 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL 18 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING 20 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER 21 * DEALINGS IN THE SOFTWARE. 22 * 23 * Authors: 24 * Wladimir J. van der Laan <laanwj@gmail.com> 25 */ 26 27#include "etnaviv_transfer.h" 28#include "etnaviv_clear_blit.h" 29#include "etnaviv_context.h" 30#include "etnaviv_debug.h" 31#include "etnaviv_etc2.h" 32#include "etnaviv_screen.h" 33 34#include "pipe/p_defines.h" 35#include "pipe/p_format.h" 36#include "pipe/p_screen.h" 37#include "pipe/p_state.h" 38#include "util/format/u_format.h" 39#include "util/u_inlines.h" 40#include "util/u_memory.h" 41#include "util/u_surface.h" 42#include "util/u_transfer.h" 43 44#include "hw/common_3d.xml.h" 45 46#include "drm-uapi/drm_fourcc.h" 47 48/* Compute offset into a 1D/2D/3D buffer of a certain box. 49 * This box must be aligned to the block width and height of the 50 * underlying format. */ 51static inline size_t 52etna_compute_offset(enum pipe_format format, const struct pipe_box *box, 53 size_t stride, size_t layer_stride) 54{ 55 return box->z * layer_stride + 56 box->y / util_format_get_blockheight(format) * stride + 57 box->x / util_format_get_blockwidth(format) * 58 util_format_get_blocksize(format); 59} 60 61static void etna_patch_data(void *buffer, const struct pipe_transfer *ptrans) 62{ 63 struct pipe_resource *prsc = ptrans->resource; 64 struct etna_resource *rsc = etna_resource(prsc); 65 struct etna_resource_level *level = &rsc->levels[ptrans->level]; 66 67 if (likely(!etna_etc2_needs_patching(prsc))) 68 return; 69 70 if (level->patched) 71 return; 72 73 /* do have the offsets of blocks to patch? */ 74 if (!level->patch_offsets) { 75 level->patch_offsets = CALLOC_STRUCT(util_dynarray); 76 77 etna_etc2_calculate_blocks(buffer, ptrans->stride, 78 ptrans->box.width, ptrans->box.height, 79 prsc->format, level->patch_offsets); 80 } 81 82 etna_etc2_patch(buffer, level->patch_offsets); 83 84 level->patched = true; 85} 86 87static void etna_unpatch_data(void *buffer, const struct pipe_transfer *ptrans) 88{ 89 struct pipe_resource *prsc = ptrans->resource; 90 struct etna_resource *rsc = etna_resource(prsc); 91 struct etna_resource_level *level = &rsc->levels[ptrans->level]; 92 93 if (!level->patched) 94 return; 95 96 etna_etc2_patch(buffer, level->patch_offsets); 97 98 level->patched = false; 99} 100 101static void 102etna_transfer_unmap(struct pipe_context *pctx, struct pipe_transfer *ptrans) 103{ 104 struct etna_context *ctx = etna_context(pctx); 105 struct etna_transfer *trans = etna_transfer(ptrans); 106 struct etna_resource *rsc = etna_resource(ptrans->resource); 107 108 /* XXX 109 * When writing to a resource that is already in use, replace the resource 110 * with a completely new buffer 111 * and free the old one using a fenced free. 112 * The most tricky case to implement will be: tiled or supertiled surface, 113 * partial write, target not aligned to 4/64. */ 114 assert(ptrans->level <= rsc->base.last_level); 115 116 if (rsc->texture && !etna_resource_newer(rsc, etna_resource(rsc->texture))) 117 rsc = etna_resource(rsc->texture); /* switch to using the texture resource */ 118 119 /* 120 * Temporary resources are always pulled into the CPU domain, must push them 121 * back into GPU domain before the RS execs the blit to the base resource. 122 */ 123 if (trans->rsc) 124 etna_bo_cpu_fini(etna_resource(trans->rsc)->bo); 125 126 if (ptrans->usage & PIPE_MAP_WRITE) { 127 if (trans->rsc) { 128 /* We have a temporary resource due to either tile status or 129 * tiling format. Write back the updated buffer contents. 130 * FIXME: we need to invalidate the tile status. */ 131 etna_copy_resource_box(pctx, ptrans->resource, trans->rsc, ptrans->level, &ptrans->box); 132 } else if (trans->staging) { 133 /* map buffer object */ 134 struct etna_resource_level *res_level = &rsc->levels[ptrans->level]; 135 136 if (rsc->layout == ETNA_LAYOUT_TILED) { 137 for (unsigned z = 0; z < ptrans->box.depth; z++) { 138 etna_texture_tile( 139 trans->mapped + (ptrans->box.z + z) * res_level->layer_stride, 140 trans->staging + z * ptrans->layer_stride, 141 ptrans->box.x, ptrans->box.y, 142 res_level->stride, ptrans->box.width, ptrans->box.height, 143 ptrans->stride, util_format_get_blocksize(rsc->base.format)); 144 } 145 } else if (rsc->layout == ETNA_LAYOUT_LINEAR) { 146 util_copy_box(trans->mapped, rsc->base.format, res_level->stride, 147 res_level->layer_stride, ptrans->box.x, 148 ptrans->box.y, ptrans->box.z, ptrans->box.width, 149 ptrans->box.height, ptrans->box.depth, 150 trans->staging, ptrans->stride, 151 ptrans->layer_stride, 0, 0, 0 /* src x,y,z */); 152 } else { 153 BUG("unsupported tiling %i", rsc->layout); 154 } 155 156 FREE(trans->staging); 157 } 158 159 rsc->seqno++; 160 161 if (rsc->base.bind & PIPE_BIND_SAMPLER_VIEW) { 162 ctx->dirty |= ETNA_DIRTY_TEXTURE_CACHES; 163 } 164 } 165 166 /* We need to have the patched data ready for the GPU. */ 167 etna_patch_data(trans->mapped, ptrans); 168 169 /* 170 * Transfers without a temporary are only pulled into the CPU domain if they 171 * are not mapped unsynchronized. If they are, must push them back into GPU 172 * domain after CPU access is finished. 173 */ 174 if (!trans->rsc && !(ptrans->usage & PIPE_MAP_UNSYNCHRONIZED)) 175 etna_bo_cpu_fini(rsc->bo); 176 177 if ((ptrans->resource->target == PIPE_BUFFER) && 178 (ptrans->usage & PIPE_MAP_WRITE)) { 179 util_range_add(&rsc->base, 180 &rsc->valid_buffer_range, 181 ptrans->box.x, 182 ptrans->box.x + ptrans->box.width); 183 } 184 185 pipe_resource_reference(&trans->rsc, NULL); 186 pipe_resource_reference(&ptrans->resource, NULL); 187 slab_free(&ctx->transfer_pool, trans); 188} 189 190static void * 191etna_transfer_map(struct pipe_context *pctx, struct pipe_resource *prsc, 192 unsigned level, 193 unsigned usage, 194 const struct pipe_box *box, 195 struct pipe_transfer **out_transfer) 196{ 197 struct etna_context *ctx = etna_context(pctx); 198 struct etna_screen *screen = ctx->screen; 199 struct etna_resource *rsc = etna_resource(prsc); 200 struct etna_transfer *trans; 201 struct pipe_transfer *ptrans; 202 enum pipe_format format = prsc->format; 203 204 trans = slab_zalloc(&ctx->transfer_pool); 205 if (!trans) 206 return NULL; 207 208 /* 209 * Upgrade to UNSYNCHRONIZED if target is PIPE_BUFFER and range is uninitialized. 210 */ 211 if ((usage & PIPE_MAP_WRITE) && 212 (prsc->target == PIPE_BUFFER) && 213 !util_ranges_intersect(&rsc->valid_buffer_range, 214 box->x, 215 box->x + box->width)) { 216 usage |= PIPE_MAP_UNSYNCHRONIZED; 217 } 218 219 /* Upgrade DISCARD_RANGE to WHOLE_RESOURCE if the whole resource is 220 * being mapped. If we add buffer reallocation to avoid CPU/GPU sync this 221 * check needs to be extended to coherent mappings and shared resources. 222 */ 223 if ((usage & PIPE_MAP_DISCARD_RANGE) && 224 !(usage & PIPE_MAP_UNSYNCHRONIZED) && 225 prsc->last_level == 0 && 226 prsc->width0 == box->width && 227 prsc->height0 == box->height && 228 prsc->depth0 == box->depth && 229 prsc->array_size == 1) { 230 usage |= PIPE_MAP_DISCARD_WHOLE_RESOURCE; 231 } 232 233 ptrans = &trans->base; 234 pipe_resource_reference(&ptrans->resource, prsc); 235 ptrans->level = level; 236 ptrans->usage = usage; 237 ptrans->box = *box; 238 239 assert(level <= prsc->last_level); 240 241 /* This one is a little tricky: if we have a separate render resource, which 242 * is newer than the base resource we want the transfer to target this one, 243 * to get the most up-to-date content, but only if we don't have a texture 244 * target of the same age, as transfering in/out of the texture target is 245 * generally preferred for the reasons listed below */ 246 if (rsc->render && etna_resource_newer(etna_resource(rsc->render), rsc) && 247 (!rsc->texture || etna_resource_newer(etna_resource(rsc->render), 248 etna_resource(rsc->texture)))) { 249 rsc = etna_resource(rsc->render); 250 } 251 252 if (rsc->texture && !etna_resource_newer(rsc, etna_resource(rsc->texture))) { 253 /* We have a texture resource which is the same age or newer than the 254 * render resource. Use the texture resource, which avoids bouncing 255 * pixels between the two resources, and we can de-tile it in s/w. */ 256 rsc = etna_resource(rsc->texture); 257 } else if (rsc->ts_bo || 258 (rsc->layout != ETNA_LAYOUT_LINEAR && 259 etna_resource_hw_tileable(screen->specs.use_blt, prsc) && 260 /* HALIGN 4 resources are incompatible with the resolve engine, 261 * so fall back to using software to detile this resource. */ 262 rsc->halign != TEXTURE_HALIGN_FOUR)) { 263 /* If the surface has tile status, we need to resolve it first. 264 * The strategy we implement here is to use the RS to copy the 265 * depth buffer, filling in the "holes" where the tile status 266 * indicates that it's clear. We also do this for tiled 267 * resources, but only if the RS can blit them. */ 268 if (usage & PIPE_MAP_DIRECTLY) { 269 slab_free(&ctx->transfer_pool, trans); 270 BUG("unsupported map flags %#x with tile status/tiled layout", usage); 271 return NULL; 272 } 273 274 if (prsc->depth0 > 1 && rsc->ts_bo) { 275 slab_free(&ctx->transfer_pool, trans); 276 BUG("resource has depth >1 with tile status"); 277 return NULL; 278 } 279 280 struct pipe_resource templ = *prsc; 281 templ.nr_samples = 0; 282 templ.bind = PIPE_BIND_RENDER_TARGET; 283 284 trans->rsc = etna_resource_alloc(pctx->screen, ETNA_LAYOUT_LINEAR, 285 DRM_FORMAT_MOD_LINEAR, &templ); 286 if (!trans->rsc) { 287 slab_free(&ctx->transfer_pool, trans); 288 return NULL; 289 } 290 291 if (!screen->specs.use_blt) { 292 /* Need to align the transfer region to satisfy RS restrictions, as we 293 * really want to hit the RS blit path here. 294 */ 295 unsigned w_align, h_align; 296 297 if (rsc->layout & ETNA_LAYOUT_BIT_SUPER) { 298 w_align = 64; 299 h_align = 64 * ctx->screen->specs.pixel_pipes; 300 } else { 301 w_align = ETNA_RS_WIDTH_MASK + 1; 302 h_align = ETNA_RS_HEIGHT_MASK + 1; 303 } 304 305 ptrans->box.width += ptrans->box.x & (w_align - 1); 306 ptrans->box.x = ptrans->box.x & ~(w_align - 1); 307 ptrans->box.width = align(ptrans->box.width, (ETNA_RS_WIDTH_MASK + 1)); 308 ptrans->box.height += ptrans->box.y & (h_align - 1); 309 ptrans->box.y = ptrans->box.y & ~(h_align - 1); 310 ptrans->box.height = align(ptrans->box.height, ETNA_RS_HEIGHT_MASK + 1); 311 } 312 313 if (!(usage & PIPE_MAP_DISCARD_WHOLE_RESOURCE)) 314 etna_copy_resource_box(pctx, trans->rsc, &rsc->base, level, &ptrans->box); 315 316 /* Switch to using the temporary resource instead */ 317 rsc = etna_resource(trans->rsc); 318 } 319 320 struct etna_resource_level *res_level = &rsc->levels[level]; 321 322 /* XXX we don't handle PIPE_MAP_FLUSH_EXPLICIT; this flag can be ignored 323 * when mapping in-place, 324 * but when not in place we need to fire off the copy operation in 325 * transfer_flush_region (currently 326 * a no-op) instead of unmap. Need to handle this to support 327 * ARB_map_buffer_range extension at least. 328 */ 329 /* XXX we don't take care of current operations on the resource; which can 330 be, at some point in the pipeline 331 which is not yet executed: 332 333 - bound as surface 334 - bound through vertex buffer 335 - bound through index buffer 336 - bound in sampler view 337 - used in clear_render_target / clear_depth_stencil operation 338 - used in blit 339 - used in resource_copy_region 340 341 How do other drivers record this information over course of the rendering 342 pipeline? 343 Is it necessary at all? Only in case we want to provide a fast path and 344 map the resource directly 345 (and for PIPE_MAP_DIRECTLY) and we don't want to force a sync. 346 We also need to know whether the resource is in use to determine if a sync 347 is needed (or just do it 348 always, but that comes at the expense of performance). 349 350 A conservative approximation without too much overhead would be to mark 351 all resources that have 352 been bound at some point as busy. A drawback would be that accessing 353 resources that have 354 been bound but are no longer in use for a while still carry a performance 355 penalty. On the other hand, 356 the program could be using PIPE_MAP_DISCARD_WHOLE_RESOURCE or 357 PIPE_MAP_UNSYNCHRONIZED to 358 avoid this in the first place... 359 360 A) We use an in-pipe copy engine, and queue the copy operation after unmap 361 so that the copy 362 will be performed when all current commands have been executed. 363 Using the RS is possible, not sure if always efficient. This can also 364 do any kind of tiling for us. 365 Only possible when PIPE_MAP_DISCARD_RANGE is set. 366 B) We discard the entire resource (or at least, the mipmap level) and 367 allocate new memory for it. 368 Only possible when mapping the entire resource or 369 PIPE_MAP_DISCARD_WHOLE_RESOURCE is set. 370 */ 371 372 /* 373 * Pull resources into the CPU domain. Only skipped for unsynchronized 374 * transfers without a temporary resource. 375 */ 376 if (trans->rsc || !(usage & PIPE_MAP_UNSYNCHRONIZED)) { 377 enum etna_resource_status status = etna_resource_status(ctx, rsc); 378 uint32_t prep_flags = 0; 379 380 /* 381 * Always flush if we have the temporary resource and have a copy to this 382 * outstanding. Otherwise infer flush requirement from resource access and 383 * current GPU usage (reads must wait for GPU writes, writes must have 384 * exclusive access to the buffer). 385 */ 386 if ((trans->rsc && (status & ETNA_PENDING_WRITE)) || 387 (!trans->rsc && 388 (((usage & PIPE_MAP_READ) && (status & ETNA_PENDING_WRITE)) || 389 ((usage & PIPE_MAP_WRITE) && status)))) { 390 pctx->flush(pctx, NULL, 0); 391 } 392 393 if (usage & PIPE_MAP_READ) 394 prep_flags |= DRM_ETNA_PREP_READ; 395 if (usage & PIPE_MAP_WRITE) 396 prep_flags |= DRM_ETNA_PREP_WRITE; 397 398 /* 399 * The ETC2 patching operates in-place on the resource, so the resource will 400 * get written even on read-only transfers. This blocks the GPU to sample 401 * from this resource. 402 */ 403 if ((usage & PIPE_MAP_READ) && etna_etc2_needs_patching(prsc)) 404 prep_flags |= DRM_ETNA_PREP_WRITE; 405 406 if (etna_bo_cpu_prep(rsc->bo, prep_flags)) 407 goto fail_prep; 408 } 409 410 /* map buffer object */ 411 trans->mapped = etna_bo_map(rsc->bo); 412 if (!trans->mapped) 413 goto fail; 414 415 *out_transfer = ptrans; 416 417 if (rsc->layout == ETNA_LAYOUT_LINEAR) { 418 ptrans->stride = res_level->stride; 419 ptrans->layer_stride = res_level->layer_stride; 420 421 trans->mapped += res_level->offset + 422 etna_compute_offset(prsc->format, box, res_level->stride, 423 res_level->layer_stride); 424 425 /* We need to have the unpatched data ready for the gfx stack. */ 426 if (usage & PIPE_MAP_READ) 427 etna_unpatch_data(trans->mapped, ptrans); 428 429 return trans->mapped; 430 } else { 431 unsigned divSizeX = util_format_get_blockwidth(format); 432 unsigned divSizeY = util_format_get_blockheight(format); 433 434 /* No direct mappings of tiled, since we need to manually 435 * tile/untile. 436 */ 437 if (usage & PIPE_MAP_DIRECTLY) 438 goto fail; 439 440 trans->mapped += res_level->offset; 441 ptrans->stride = align(box->width, divSizeX) * util_format_get_blocksize(format); /* row stride in bytes */ 442 ptrans->layer_stride = align(box->height, divSizeY) * ptrans->stride; 443 size_t size = ptrans->layer_stride * box->depth; 444 445 trans->staging = MALLOC(size); 446 if (!trans->staging) 447 goto fail; 448 449 if (usage & PIPE_MAP_READ) { 450 if (rsc->layout == ETNA_LAYOUT_TILED) { 451 for (unsigned z = 0; z < ptrans->box.depth; z++) { 452 etna_texture_untile(trans->staging + z * ptrans->layer_stride, 453 trans->mapped + (ptrans->box.z + z) * res_level->layer_stride, 454 ptrans->box.x, ptrans->box.y, res_level->stride, 455 ptrans->box.width, ptrans->box.height, ptrans->stride, 456 util_format_get_blocksize(rsc->base.format)); 457 } 458 } else if (rsc->layout == ETNA_LAYOUT_LINEAR) { 459 util_copy_box(trans->staging, rsc->base.format, ptrans->stride, 460 ptrans->layer_stride, 0, 0, 0, /* dst x,y,z */ 461 ptrans->box.width, ptrans->box.height, 462 ptrans->box.depth, trans->mapped, res_level->stride, 463 res_level->layer_stride, ptrans->box.x, 464 ptrans->box.y, ptrans->box.z); 465 } else { 466 /* TODO supertiling */ 467 BUG("unsupported tiling %i for reading", rsc->layout); 468 } 469 } 470 471 return trans->staging; 472 } 473 474fail: 475 etna_bo_cpu_fini(rsc->bo); 476fail_prep: 477 etna_transfer_unmap(pctx, ptrans); 478 return NULL; 479} 480 481static void 482etna_transfer_flush_region(struct pipe_context *pctx, 483 struct pipe_transfer *ptrans, 484 const struct pipe_box *box) 485{ 486 struct etna_resource *rsc = etna_resource(ptrans->resource); 487 488 if (ptrans->resource->target == PIPE_BUFFER) 489 util_range_add(&rsc->base, 490 &rsc->valid_buffer_range, 491 ptrans->box.x + box->x, 492 ptrans->box.x + box->x + box->width); 493} 494 495void 496etna_transfer_init(struct pipe_context *pctx) 497{ 498 pctx->buffer_map = etna_transfer_map; 499 pctx->texture_map = etna_transfer_map; 500 pctx->transfer_flush_region = etna_transfer_flush_region; 501 pctx->buffer_unmap = etna_transfer_unmap; 502 pctx->texture_unmap = etna_transfer_unmap; 503 pctx->buffer_subdata = u_default_buffer_subdata; 504 pctx->texture_subdata = u_default_texture_subdata; 505} 506