Loading...
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 | /*
* Copyright 2010 Red Hat Inc.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
* OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
* ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
* OTHER DEALINGS IN THE SOFTWARE.
*
* Authors: Ben Skeggs
*/
#include <core/gpuobj.h>
#include <core/mm.h>
#include <subdev/fb.h>
#include <subdev/vm.h>
void
nouveau_vm_map_at(struct nouveau_vma *vma, u64 delta, struct nouveau_mem *node)
{
struct nouveau_vm *vm = vma->vm;
struct nouveau_vmmgr *vmm = vm->vmm;
struct nouveau_mm_node *r;
int big = vma->node->type != vmm->spg_shift;
u32 offset = vma->node->offset + (delta >> 12);
u32 bits = vma->node->type - 12;
u32 pde = (offset >> vmm->pgt_bits) - vm->fpde;
u32 pte = (offset & ((1 << vmm->pgt_bits) - 1)) >> bits;
u32 max = 1 << (vmm->pgt_bits - bits);
u32 end, len;
delta = 0;
list_for_each_entry(r, &node->regions, rl_entry) {
u64 phys = (u64)r->offset << 12;
u32 num = r->length >> bits;
while (num) {
struct nouveau_gpuobj *pgt = vm->pgt[pde].obj[big];
end = (pte + num);
if (unlikely(end >= max))
end = max;
len = end - pte;
vmm->map(vma, pgt, node, pte, len, phys, delta);
num -= len;
pte += len;
if (unlikely(end >= max)) {
phys += len << (bits + 12);
pde++;
pte = 0;
}
delta += (u64)len << vma->node->type;
}
}
vmm->flush(vm);
}
static void
nouveau_vm_map_sg_table(struct nouveau_vma *vma, u64 delta, u64 length,
struct nouveau_mem *mem)
{
struct nouveau_vm *vm = vma->vm;
struct nouveau_vmmgr *vmm = vm->vmm;
int big = vma->node->type != vmm->spg_shift;
u32 offset = vma->node->offset + (delta >> 12);
u32 bits = vma->node->type - 12;
u32 num = length >> vma->node->type;
u32 pde = (offset >> vmm->pgt_bits) - vm->fpde;
u32 pte = (offset & ((1 << vmm->pgt_bits) - 1)) >> bits;
u32 max = 1 << (vmm->pgt_bits - bits);
unsigned m, sglen;
u32 end, len;
int i;
struct scatterlist *sg;
for_each_sg(mem->sg->sgl, sg, mem->sg->nents, i) {
struct nouveau_gpuobj *pgt = vm->pgt[pde].obj[big];
sglen = sg_dma_len(sg) >> PAGE_SHIFT;
end = pte + sglen;
if (unlikely(end >= max))
end = max;
len = end - pte;
for (m = 0; m < len; m++) {
dma_addr_t addr = sg_dma_address(sg) + (m << PAGE_SHIFT);
vmm->map_sg(vma, pgt, mem, pte, 1, &addr);
num--;
pte++;
if (num == 0)
goto finish;
}
if (unlikely(end >= max)) {
pde++;
pte = 0;
}
if (m < sglen) {
for (; m < sglen; m++) {
dma_addr_t addr = sg_dma_address(sg) + (m << PAGE_SHIFT);
vmm->map_sg(vma, pgt, mem, pte, 1, &addr);
num--;
pte++;
if (num == 0)
goto finish;
}
}
}
finish:
vmm->flush(vm);
}
static void
nouveau_vm_map_sg(struct nouveau_vma *vma, u64 delta, u64 length,
struct nouveau_mem *mem)
{
struct nouveau_vm *vm = vma->vm;
struct nouveau_vmmgr *vmm = vm->vmm;
dma_addr_t *list = mem->pages;
int big = vma->node->type != vmm->spg_shift;
u32 offset = vma->node->offset + (delta >> 12);
u32 bits = vma->node->type - 12;
u32 num = length >> vma->node->type;
u32 pde = (offset >> vmm->pgt_bits) - vm->fpde;
u32 pte = (offset & ((1 << vmm->pgt_bits) - 1)) >> bits;
u32 max = 1 << (vmm->pgt_bits - bits);
u32 end, len;
while (num) {
struct nouveau_gpuobj *pgt = vm->pgt[pde].obj[big];
end = (pte + num);
if (unlikely(end >= max))
end = max;
len = end - pte;
vmm->map_sg(vma, pgt, mem, pte, len, list);
num -= len;
pte += len;
list += len;
if (unlikely(end >= max)) {
pde++;
pte = 0;
}
}
vmm->flush(vm);
}
void
nouveau_vm_map(struct nouveau_vma *vma, struct nouveau_mem *node)
{
if (node->sg)
nouveau_vm_map_sg_table(vma, 0, node->size << 12, node);
else
if (node->pages)
nouveau_vm_map_sg(vma, 0, node->size << 12, node);
else
nouveau_vm_map_at(vma, 0, node);
}
void
nouveau_vm_unmap_at(struct nouveau_vma *vma, u64 delta, u64 length)
{
struct nouveau_vm *vm = vma->vm;
struct nouveau_vmmgr *vmm = vm->vmm;
int big = vma->node->type != vmm->spg_shift;
u32 offset = vma->node->offset + (delta >> 12);
u32 bits = vma->node->type - 12;
u32 num = length >> vma->node->type;
u32 pde = (offset >> vmm->pgt_bits) - vm->fpde;
u32 pte = (offset & ((1 << vmm->pgt_bits) - 1)) >> bits;
u32 max = 1 << (vmm->pgt_bits - bits);
u32 end, len;
while (num) {
struct nouveau_gpuobj *pgt = vm->pgt[pde].obj[big];
end = (pte + num);
if (unlikely(end >= max))
end = max;
len = end - pte;
vmm->unmap(pgt, pte, len);
num -= len;
pte += len;
if (unlikely(end >= max)) {
pde++;
pte = 0;
}
}
vmm->flush(vm);
}
void
nouveau_vm_unmap(struct nouveau_vma *vma)
{
nouveau_vm_unmap_at(vma, 0, (u64)vma->node->length << 12);
}
static void
nouveau_vm_unmap_pgt(struct nouveau_vm *vm, int big, u32 fpde, u32 lpde)
{
struct nouveau_vmmgr *vmm = vm->vmm;
struct nouveau_vm_pgd *vpgd;
struct nouveau_vm_pgt *vpgt;
struct nouveau_gpuobj *pgt;
u32 pde;
for (pde = fpde; pde <= lpde; pde++) {
vpgt = &vm->pgt[pde - vm->fpde];
if (--vpgt->refcount[big])
continue;
pgt = vpgt->obj[big];
vpgt->obj[big] = NULL;
list_for_each_entry(vpgd, &vm->pgd_list, head) {
vmm->map_pgt(vpgd->obj, pde, vpgt->obj);
}
mutex_unlock(&nv_subdev(vmm)->mutex);
nouveau_gpuobj_ref(NULL, &pgt);
mutex_lock(&nv_subdev(vmm)->mutex);
}
}
static int
nouveau_vm_map_pgt(struct nouveau_vm *vm, u32 pde, u32 type)
{
struct nouveau_vmmgr *vmm = vm->vmm;
struct nouveau_vm_pgt *vpgt = &vm->pgt[pde - vm->fpde];
struct nouveau_vm_pgd *vpgd;
struct nouveau_gpuobj *pgt;
int big = (type != vmm->spg_shift);
u32 pgt_size;
int ret;
pgt_size = (1 << (vmm->pgt_bits + 12)) >> type;
pgt_size *= 8;
mutex_unlock(&nv_subdev(vmm)->mutex);
ret = nouveau_gpuobj_new(nv_object(vm->vmm), NULL, pgt_size, 0x1000,
NVOBJ_FLAG_ZERO_ALLOC, &pgt);
mutex_lock(&nv_subdev(vmm)->mutex);
if (unlikely(ret))
return ret;
/* someone beat us to filling the PDE while we didn't have the lock */
if (unlikely(vpgt->refcount[big]++)) {
mutex_unlock(&nv_subdev(vmm)->mutex);
nouveau_gpuobj_ref(NULL, &pgt);
mutex_lock(&nv_subdev(vmm)->mutex);
return 0;
}
vpgt->obj[big] = pgt;
list_for_each_entry(vpgd, &vm->pgd_list, head) {
vmm->map_pgt(vpgd->obj, pde, vpgt->obj);
}
return 0;
}
int
nouveau_vm_get(struct nouveau_vm *vm, u64 size, u32 page_shift,
u32 access, struct nouveau_vma *vma)
{
struct nouveau_vmmgr *vmm = vm->vmm;
u32 align = (1 << page_shift) >> 12;
u32 msize = size >> 12;
u32 fpde, lpde, pde;
int ret;
mutex_lock(&nv_subdev(vmm)->mutex);
ret = nouveau_mm_head(&vm->mm, page_shift, msize, msize, align,
&vma->node);
if (unlikely(ret != 0)) {
mutex_unlock(&nv_subdev(vmm)->mutex);
return ret;
}
fpde = (vma->node->offset >> vmm->pgt_bits);
lpde = (vma->node->offset + vma->node->length - 1) >> vmm->pgt_bits;
for (pde = fpde; pde <= lpde; pde++) {
struct nouveau_vm_pgt *vpgt = &vm->pgt[pde - vm->fpde];
int big = (vma->node->type != vmm->spg_shift);
if (likely(vpgt->refcount[big])) {
vpgt->refcount[big]++;
continue;
}
ret = nouveau_vm_map_pgt(vm, pde, vma->node->type);
if (ret) {
if (pde != fpde)
nouveau_vm_unmap_pgt(vm, big, fpde, pde - 1);
nouveau_mm_free(&vm->mm, &vma->node);
mutex_unlock(&nv_subdev(vmm)->mutex);
return ret;
}
}
mutex_unlock(&nv_subdev(vmm)->mutex);
vma->vm = NULL;
nouveau_vm_ref(vm, &vma->vm, NULL);
vma->offset = (u64)vma->node->offset << 12;
vma->access = access;
return 0;
}
void
nouveau_vm_put(struct nouveau_vma *vma)
{
struct nouveau_vm *vm = vma->vm;
struct nouveau_vmmgr *vmm = vm->vmm;
u32 fpde, lpde;
if (unlikely(vma->node == NULL))
return;
fpde = (vma->node->offset >> vmm->pgt_bits);
lpde = (vma->node->offset + vma->node->length - 1) >> vmm->pgt_bits;
mutex_lock(&nv_subdev(vmm)->mutex);
nouveau_vm_unmap_pgt(vm, vma->node->type != vmm->spg_shift, fpde, lpde);
nouveau_mm_free(&vm->mm, &vma->node);
mutex_unlock(&nv_subdev(vmm)->mutex);
nouveau_vm_ref(NULL, &vma->vm, NULL);
}
int
nouveau_vm_create(struct nouveau_vmmgr *vmm, u64 offset, u64 length,
u64 mm_offset, u32 block, struct nouveau_vm **pvm)
{
struct nouveau_vm *vm;
u64 mm_length = (offset + length) - mm_offset;
int ret;
vm = kzalloc(sizeof(*vm), GFP_KERNEL);
if (!vm)
return -ENOMEM;
INIT_LIST_HEAD(&vm->pgd_list);
vm->vmm = vmm;
kref_init(&vm->refcount);
vm->fpde = offset >> (vmm->pgt_bits + 12);
vm->lpde = (offset + length - 1) >> (vmm->pgt_bits + 12);
vm->pgt = vzalloc((vm->lpde - vm->fpde + 1) * sizeof(*vm->pgt));
if (!vm->pgt) {
kfree(vm);
return -ENOMEM;
}
ret = nouveau_mm_init(&vm->mm, mm_offset >> 12, mm_length >> 12,
block >> 12);
if (ret) {
vfree(vm->pgt);
kfree(vm);
return ret;
}
*pvm = vm;
return 0;
}
int
nouveau_vm_new(struct nouveau_device *device, u64 offset, u64 length,
u64 mm_offset, struct nouveau_vm **pvm)
{
struct nouveau_vmmgr *vmm = nouveau_vmmgr(device);
return vmm->create(vmm, offset, length, mm_offset, pvm);
}
static int
nouveau_vm_link(struct nouveau_vm *vm, struct nouveau_gpuobj *pgd)
{
struct nouveau_vmmgr *vmm = vm->vmm;
struct nouveau_vm_pgd *vpgd;
int i;
if (!pgd)
return 0;
vpgd = kzalloc(sizeof(*vpgd), GFP_KERNEL);
if (!vpgd)
return -ENOMEM;
nouveau_gpuobj_ref(pgd, &vpgd->obj);
mutex_lock(&nv_subdev(vmm)->mutex);
for (i = vm->fpde; i <= vm->lpde; i++)
vmm->map_pgt(pgd, i, vm->pgt[i - vm->fpde].obj);
list_add(&vpgd->head, &vm->pgd_list);
mutex_unlock(&nv_subdev(vmm)->mutex);
return 0;
}
static void
nouveau_vm_unlink(struct nouveau_vm *vm, struct nouveau_gpuobj *mpgd)
{
struct nouveau_vmmgr *vmm = vm->vmm;
struct nouveau_vm_pgd *vpgd, *tmp;
struct nouveau_gpuobj *pgd = NULL;
if (!mpgd)
return;
mutex_lock(&nv_subdev(vmm)->mutex);
list_for_each_entry_safe(vpgd, tmp, &vm->pgd_list, head) {
if (vpgd->obj == mpgd) {
pgd = vpgd->obj;
list_del(&vpgd->head);
kfree(vpgd);
break;
}
}
mutex_unlock(&nv_subdev(vmm)->mutex);
nouveau_gpuobj_ref(NULL, &pgd);
}
static void
nouveau_vm_del(struct kref *kref)
{
struct nouveau_vm *vm = container_of(kref, typeof(*vm), refcount);
struct nouveau_vm_pgd *vpgd, *tmp;
list_for_each_entry_safe(vpgd, tmp, &vm->pgd_list, head) {
nouveau_vm_unlink(vm, vpgd->obj);
}
nouveau_mm_fini(&vm->mm);
vfree(vm->pgt);
kfree(vm);
}
int
nouveau_vm_ref(struct nouveau_vm *ref, struct nouveau_vm **ptr,
struct nouveau_gpuobj *pgd)
{
if (ref) {
int ret = nouveau_vm_link(ref, pgd);
if (ret)
return ret;
kref_get(&ref->refcount);
}
if (*ptr) {
nouveau_vm_unlink(*ptr, pgd);
kref_put(&(*ptr)->refcount, nouveau_vm_del);
}
*ptr = ref;
return 0;
}
|