Linux Audio

Check our new training course

Embedded Linux Audio

Check our new training course
with Creative Commons CC-BY-SA
lecture materials

Bootlin logo

Elixir Cross Referencer

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
 484
 485
 486
 487
 488
 489
 490
 491
 492
 493
 494
 495
 496
 497
 498
 499
 500
 501
 502
 503
 504
 505
 506
 507
 508
 509
 510
 511
 512
 513
 514
 515
 516
 517
 518
 519
 520
 521
 522
 523
 524
 525
 526
 527
 528
 529
 530
 531
 532
 533
 534
 535
 536
 537
 538
 539
 540
 541
 542
 543
 544
 545
 546
 547
 548
 549
 550
 551
 552
 553
 554
 555
 556
 557
 558
 559
 560
 561
 562
 563
 564
 565
 566
 567
 568
 569
 570
 571
 572
 573
 574
 575
 576
 577
 578
 579
 580
 581
 582
 583
 584
 585
 586
 587
 588
 589
 590
 591
 592
 593
 594
 595
 596
 597
 598
 599
 600
 601
 602
 603
 604
 605
 606
 607
 608
 609
 610
 611
 612
 613
 614
 615
 616
 617
 618
 619
 620
 621
 622
 623
 624
 625
 626
 627
 628
 629
 630
 631
 632
 633
 634
 635
 636
 637
 638
 639
 640
 641
 642
 643
 644
 645
 646
 647
 648
 649
 650
 651
 652
 653
 654
 655
 656
 657
 658
 659
 660
 661
 662
 663
 664
 665
 666
 667
 668
 669
 670
 671
 672
 673
 674
 675
 676
 677
 678
 679
 680
 681
 682
 683
 684
 685
 686
 687
 688
 689
 690
 691
 692
 693
 694
 695
 696
 697
 698
 699
 700
 701
 702
 703
 704
 705
 706
 707
 708
 709
 710
 711
 712
 713
 714
 715
 716
 717
 718
 719
 720
 721
 722
 723
 724
 725
 726
 727
 728
 729
 730
 731
 732
 733
 734
 735
 736
 737
 738
 739
 740
 741
 742
 743
 744
 745
 746
 747
 748
 749
 750
 751
 752
 753
 754
 755
 756
 757
 758
 759
 760
 761
 762
 763
 764
 765
 766
 767
 768
 769
 770
 771
 772
 773
 774
 775
 776
 777
 778
 779
 780
 781
 782
 783
 784
 785
 786
 787
 788
 789
 790
 791
 792
 793
 794
 795
 796
 797
 798
 799
 800
 801
 802
 803
 804
 805
 806
 807
 808
 809
 810
 811
 812
 813
 814
 815
 816
 817
 818
 819
 820
 821
 822
 823
 824
 825
 826
 827
 828
 829
 830
 831
 832
 833
 834
 835
 836
 837
 838
 839
 840
 841
 842
 843
 844
 845
 846
 847
 848
 849
 850
 851
 852
 853
 854
 855
 856
 857
 858
 859
 860
 861
 862
 863
 864
 865
 866
 867
 868
 869
 870
 871
 872
 873
 874
 875
 876
 877
 878
 879
 880
 881
 882
 883
 884
 885
 886
 887
 888
 889
 890
 891
 892
 893
 894
 895
 896
 897
 898
 899
 900
 901
 902
 903
 904
 905
 906
 907
 908
 909
 910
 911
 912
 913
 914
 915
 916
 917
 918
 919
 920
 921
 922
 923
 924
 925
 926
 927
 928
 929
 930
 931
 932
 933
 934
 935
 936
 937
 938
 939
 940
 941
 942
 943
 944
 945
 946
 947
 948
 949
 950
 951
 952
 953
 954
 955
 956
 957
 958
 959
 960
 961
 962
 963
 964
 965
 966
 967
 968
 969
 970
 971
 972
 973
 974
 975
 976
 977
 978
 979
 980
 981
 982
 983
 984
 985
 986
 987
 988
 989
 990
 991
 992
 993
 994
 995
 996
 997
 998
 999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
/*
 * cmm.c
 *
 * DSP-BIOS Bridge driver support functions for TI OMAP processors.
 *
 * The Communication(Shared) Memory Management(CMM) module provides
 * shared memory management services for DSP/BIOS Bridge data streaming
 * and messaging.
 *
 * Multiple shared memory segments can be registered with CMM.
 * Each registered SM segment is represented by a SM "allocator" that
 * describes a block of physically contiguous shared memory used for
 * future allocations by CMM.
 *
 * Memory is coalesced back to the appropriate heap when a buffer is
 * freed.
 *
 * Notes:
 *   Va: Virtual address.
 *   Pa: Physical or kernel system address.
 *
 * Copyright (C) 2005-2006 Texas Instruments, Inc.
 *
 * This package is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 *
 * THIS PACKAGE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR
 * IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
 * WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.
 */
#include <linux/types.h>
#include <linux/list.h>

/*  ----------------------------------- DSP/BIOS Bridge */
#include <dspbridge/dbdefs.h>

/*  ----------------------------------- Trace & Debug */
#include <dspbridge/dbc.h>

/*  ----------------------------------- OS Adaptation Layer */
#include <dspbridge/sync.h>

/*  ----------------------------------- Platform Manager */
#include <dspbridge/dev.h>
#include <dspbridge/proc.h>

/*  ----------------------------------- This */
#include <dspbridge/cmm.h>

/*  ----------------------------------- Defines, Data Structures, Typedefs */
#define NEXT_PA(pnode)   (pnode->pa + pnode->size)

/* Other bus/platform translations */
#define DSPPA2GPPPA(base, x, y)  ((x)+(y))
#define GPPPA2DSPPA(base, x, y)  ((x)-(y))

/*
 *  Allocators define a block of contiguous memory used for future allocations.
 *
 *      sma - shared memory allocator.
 *      vma - virtual memory allocator.(not used).
 */
struct cmm_allocator {		/* sma */
	unsigned int shm_base;	/* Start of physical SM block */
	u32 sm_size;		/* Size of SM block in bytes */
	unsigned int vm_base;	/* Start of VM block. (Dev driver
					 * context for 'sma') */
	u32 dsp_phys_addr_offset;	/* DSP PA to GPP PA offset for this
					 * SM space */
	s8 c_factor;		/* DSPPa to GPPPa Conversion Factor */
	unsigned int dsp_base;	/* DSP virt base byte address */
	u32 dsp_size;	/* DSP seg size in bytes */
	struct cmm_object *cmm_mgr;	/* back ref to parent mgr */
	/* node list of available memory */
	struct list_head free_list;
	/* node list of memory in use */
	struct list_head in_use_list;
};

struct cmm_xlator {		/* Pa<->Va translator object */
	/* CMM object this translator associated */
	struct cmm_object *cmm_mgr;
	/*
	 *  Client process virtual base address that corresponds to phys SM
	 *  base address for translator's seg_id.
	 *  Only 1 segment ID currently supported.
	 */
	unsigned int virt_base;	/* virtual base address */
	u32 virt_size;		/* size of virt space in bytes */
	u32 seg_id;		/* Segment Id */
};

/* CMM Mgr */
struct cmm_object {
	/*
	 * Cmm Lock is used to serialize access mem manager for multi-threads.
	 */
	struct mutex cmm_lock;	/* Lock to access cmm mgr */
	struct list_head node_free_list;	/* Free list of memory nodes */
	u32 min_block_size;	/* Min SM block; default 16 bytes */
	u32 page_size;	/* Memory Page size (1k/4k) */
	/* GPP SM segment ptrs */
	struct cmm_allocator *pa_gppsm_seg_tab[CMM_MAXGPPSEGS];
};

/* Default CMM Mgr attributes */
static struct cmm_mgrattrs cmm_dfltmgrattrs = {
	/* min_block_size, min block size(bytes) allocated by cmm mgr */
	16
};

/* Default allocation attributes */
static struct cmm_attrs cmm_dfltalctattrs = {
	1		/* seg_id, default segment Id for allocator */
};

/* Address translator default attrs */
static struct cmm_xlatorattrs cmm_dfltxlatorattrs = {
	/* seg_id, does not have to match cmm_dfltalctattrs ul_seg_id */
	1,
	0,			/* dsp_bufs */
	0,			/* dsp_buf_size */
	NULL,			/* vm_base */
	0,			/* vm_size */
};

/* SM node representing a block of memory. */
struct cmm_mnode {
	struct list_head link;	/* must be 1st element */
	u32 pa;		/* Phys addr */
	u32 va;			/* Virtual address in device process context */
	u32 size;		/* SM block size in bytes */
	u32 client_proc;	/* Process that allocated this mem block */
};

/*  ----------------------------------- Globals */
static u32 refs;		/* module reference count */

/*  ----------------------------------- Function Prototypes */
static void add_to_free_list(struct cmm_allocator *allocator,
			     struct cmm_mnode *pnode);
static struct cmm_allocator *get_allocator(struct cmm_object *cmm_mgr_obj,
					   u32 ul_seg_id);
static struct cmm_mnode *get_free_block(struct cmm_allocator *allocator,
					u32 usize);
static struct cmm_mnode *get_node(struct cmm_object *cmm_mgr_obj, u32 dw_pa,
				  u32 dw_va, u32 ul_size);
/* get available slot for new allocator */
static s32 get_slot(struct cmm_object *cmm_mgr_obj);
static void un_register_gppsm_seg(struct cmm_allocator *psma);

/*
 *  ======== cmm_calloc_buf ========
 *  Purpose:
 *      Allocate a SM buffer, zero contents, and return the physical address
 *      and optional driver context virtual address(pp_buf_va).
 *
 *      The freelist is sorted in increasing size order. Get the first
 *      block that satifies the request and sort the remaining back on
 *      the freelist; if large enough. The kept block is placed on the
 *      inUseList.
 */
void *cmm_calloc_buf(struct cmm_object *hcmm_mgr, u32 usize,
		     struct cmm_attrs *pattrs, void **pp_buf_va)
{
	struct cmm_object *cmm_mgr_obj = (struct cmm_object *)hcmm_mgr;
	void *buf_pa = NULL;
	struct cmm_mnode *pnode = NULL;
	struct cmm_mnode *new_node = NULL;
	struct cmm_allocator *allocator = NULL;
	u32 delta_size;
	u8 *pbyte = NULL;
	s32 cnt;

	if (pattrs == NULL)
		pattrs = &cmm_dfltalctattrs;

	if (pp_buf_va != NULL)
		*pp_buf_va = NULL;

	if (cmm_mgr_obj && (usize != 0)) {
		if (pattrs->seg_id > 0) {
			/* SegId > 0 is SM */
			/* get the allocator object for this segment id */
			allocator =
			    get_allocator(cmm_mgr_obj, pattrs->seg_id);
			/* keep block size a multiple of min_block_size */
			usize =
			    ((usize - 1) & ~(cmm_mgr_obj->min_block_size -
					     1))
			    + cmm_mgr_obj->min_block_size;
			mutex_lock(&cmm_mgr_obj->cmm_lock);
			pnode = get_free_block(allocator, usize);
		}
		if (pnode) {
			delta_size = (pnode->size - usize);
			if (delta_size >= cmm_mgr_obj->min_block_size) {
				/* create a new block with the leftovers and
				 * add to freelist */
				new_node =
				    get_node(cmm_mgr_obj, pnode->pa + usize,
					     pnode->va + usize,
					     (u32) delta_size);
				/* leftovers go free */
				add_to_free_list(allocator, new_node);
				/* adjust our node's size */
				pnode->size = usize;
			}
			/* Tag node with client process requesting allocation
			 * We'll need to free up a process's alloc'd SM if the
			 * client process goes away.
			 */
			/* Return TGID instead of process handle */
			pnode->client_proc = current->tgid;

			/* put our node on InUse list */
			list_add_tail(&pnode->link, &allocator->in_use_list);
			buf_pa = (void *)pnode->pa;	/* physical address */
			/* clear mem */
			pbyte = (u8 *) pnode->va;
			for (cnt = 0; cnt < (s32) usize; cnt++, pbyte++)
				*pbyte = 0;

			if (pp_buf_va != NULL) {
				/* Virtual address */
				*pp_buf_va = (void *)pnode->va;
			}
		}
		mutex_unlock(&cmm_mgr_obj->cmm_lock);
	}
	return buf_pa;
}

/*
 *  ======== cmm_create ========
 *  Purpose:
 *      Create a communication memory manager object.
 */
int cmm_create(struct cmm_object **ph_cmm_mgr,
		      struct dev_object *hdev_obj,
		      const struct cmm_mgrattrs *mgr_attrts)
{
	struct cmm_object *cmm_obj = NULL;
	int status = 0;

	DBC_REQUIRE(refs > 0);
	DBC_REQUIRE(ph_cmm_mgr != NULL);

	*ph_cmm_mgr = NULL;
	/* create, zero, and tag a cmm mgr object */
	cmm_obj = kzalloc(sizeof(struct cmm_object), GFP_KERNEL);
	if (!cmm_obj)
		return -ENOMEM;

	if (mgr_attrts == NULL)
		mgr_attrts = &cmm_dfltmgrattrs;	/* set defaults */

	/* 4 bytes minimum */
	DBC_ASSERT(mgr_attrts->min_block_size >= 4);
	/* save away smallest block allocation for this cmm mgr */
	cmm_obj->min_block_size = mgr_attrts->min_block_size;
	cmm_obj->page_size = PAGE_SIZE;

	/* create node free list */
	INIT_LIST_HEAD(&cmm_obj->node_free_list);
	mutex_init(&cmm_obj->cmm_lock);
	*ph_cmm_mgr = cmm_obj;

	return status;
}

/*
 *  ======== cmm_destroy ========
 *  Purpose:
 *      Release the communication memory manager resources.
 */
int cmm_destroy(struct cmm_object *hcmm_mgr, bool force)
{
	struct cmm_object *cmm_mgr_obj = (struct cmm_object *)hcmm_mgr;
	struct cmm_info temp_info;
	int status = 0;
	s32 slot_seg;
	struct cmm_mnode *node, *tmp;

	DBC_REQUIRE(refs > 0);
	if (!hcmm_mgr) {
		status = -EFAULT;
		return status;
	}
	mutex_lock(&cmm_mgr_obj->cmm_lock);
	/* If not force then fail if outstanding allocations exist */
	if (!force) {
		/* Check for outstanding memory allocations */
		status = cmm_get_info(hcmm_mgr, &temp_info);
		if (!status) {
			if (temp_info.total_in_use_cnt > 0) {
				/* outstanding allocations */
				status = -EPERM;
			}
		}
	}
	if (!status) {
		/* UnRegister SM allocator */
		for (slot_seg = 0; slot_seg < CMM_MAXGPPSEGS; slot_seg++) {
			if (cmm_mgr_obj->pa_gppsm_seg_tab[slot_seg] != NULL) {
				un_register_gppsm_seg
				    (cmm_mgr_obj->pa_gppsm_seg_tab[slot_seg]);
				/* Set slot to NULL for future reuse */
				cmm_mgr_obj->pa_gppsm_seg_tab[slot_seg] = NULL;
			}
		}
	}
	list_for_each_entry_safe(node, tmp, &cmm_mgr_obj->node_free_list,
			link) {
		list_del(&node->link);
		kfree(node);
	}
	mutex_unlock(&cmm_mgr_obj->cmm_lock);
	if (!status) {
		/* delete CS & cmm mgr object */
		mutex_destroy(&cmm_mgr_obj->cmm_lock);
		kfree(cmm_mgr_obj);
	}
	return status;
}

/*
 *  ======== cmm_exit ========
 *  Purpose:
 *      Discontinue usage of module; free resources when reference count
 *      reaches 0.
 */
void cmm_exit(void)
{
	DBC_REQUIRE(refs > 0);

	refs--;
}

/*
 *  ======== cmm_free_buf ========
 *  Purpose:
 *      Free the given buffer.
 */
int cmm_free_buf(struct cmm_object *hcmm_mgr, void *buf_pa, u32 ul_seg_id)
{
	struct cmm_object *cmm_mgr_obj = (struct cmm_object *)hcmm_mgr;
	int status = -EFAULT;
	struct cmm_mnode *curr, *tmp;
	struct cmm_allocator *allocator;
	struct cmm_attrs *pattrs;

	DBC_REQUIRE(refs > 0);
	DBC_REQUIRE(buf_pa != NULL);

	if (ul_seg_id == 0) {
		pattrs = &cmm_dfltalctattrs;
		ul_seg_id = pattrs->seg_id;
	}
	if (!hcmm_mgr || !(ul_seg_id > 0)) {
		status = -EFAULT;
		return status;
	}

	allocator = get_allocator(cmm_mgr_obj, ul_seg_id);
	if (!allocator)
		return status;

	mutex_lock(&cmm_mgr_obj->cmm_lock);
	list_for_each_entry_safe(curr, tmp, &allocator->in_use_list, link) {
		if (curr->pa == (u32) buf_pa) {
			list_del(&curr->link);
			add_to_free_list(allocator, curr);
			status = 0;
			break;
		}
	}
	mutex_unlock(&cmm_mgr_obj->cmm_lock);

	return status;
}

/*
 *  ======== cmm_get_handle ========
 *  Purpose:
 *      Return the communication memory manager object for this device.
 *      This is typically called from the client process.
 */
int cmm_get_handle(void *hprocessor, struct cmm_object ** ph_cmm_mgr)
{
	int status = 0;
	struct dev_object *hdev_obj;

	DBC_REQUIRE(refs > 0);
	DBC_REQUIRE(ph_cmm_mgr != NULL);
	if (hprocessor != NULL)
		status = proc_get_dev_object(hprocessor, &hdev_obj);
	else
		hdev_obj = dev_get_first();	/* default */

	if (!status)
		status = dev_get_cmm_mgr(hdev_obj, ph_cmm_mgr);

	return status;
}

/*
 *  ======== cmm_get_info ========
 *  Purpose:
 *      Return the current memory utilization information.
 */
int cmm_get_info(struct cmm_object *hcmm_mgr,
			struct cmm_info *cmm_info_obj)
{
	struct cmm_object *cmm_mgr_obj = (struct cmm_object *)hcmm_mgr;
	u32 ul_seg;
	int status = 0;
	struct cmm_allocator *altr;
	struct cmm_mnode *curr;

	DBC_REQUIRE(cmm_info_obj != NULL);

	if (!hcmm_mgr) {
		status = -EFAULT;
		return status;
	}
	mutex_lock(&cmm_mgr_obj->cmm_lock);
	cmm_info_obj->num_gppsm_segs = 0;	/* # of SM segments */
	/* Total # of outstanding alloc */
	cmm_info_obj->total_in_use_cnt = 0;
	/* min block size */
	cmm_info_obj->min_block_size = cmm_mgr_obj->min_block_size;
	/* check SM memory segments */
	for (ul_seg = 1; ul_seg <= CMM_MAXGPPSEGS; ul_seg++) {
		/* get the allocator object for this segment id */
		altr = get_allocator(cmm_mgr_obj, ul_seg);
		if (!altr)
			continue;
		cmm_info_obj->num_gppsm_segs++;
		cmm_info_obj->seg_info[ul_seg - 1].seg_base_pa =
			altr->shm_base - altr->dsp_size;
		cmm_info_obj->seg_info[ul_seg - 1].total_seg_size =
			altr->dsp_size + altr->sm_size;
		cmm_info_obj->seg_info[ul_seg - 1].gpp_base_pa =
			altr->shm_base;
		cmm_info_obj->seg_info[ul_seg - 1].gpp_size =
			altr->sm_size;
		cmm_info_obj->seg_info[ul_seg - 1].dsp_base_va =
			altr->dsp_base;
		cmm_info_obj->seg_info[ul_seg - 1].dsp_size =
			altr->dsp_size;
		cmm_info_obj->seg_info[ul_seg - 1].seg_base_va =
			altr->vm_base - altr->dsp_size;
		cmm_info_obj->seg_info[ul_seg - 1].in_use_cnt = 0;

		list_for_each_entry(curr, &altr->in_use_list, link) {
			cmm_info_obj->total_in_use_cnt++;
			cmm_info_obj->seg_info[ul_seg - 1].in_use_cnt++;
		}
	}
	mutex_unlock(&cmm_mgr_obj->cmm_lock);
	return status;
}

/*
 *  ======== cmm_init ========
 *  Purpose:
 *      Initializes private state of CMM module.
 */
bool cmm_init(void)
{
	bool ret = true;

	DBC_REQUIRE(refs >= 0);
	if (ret)
		refs++;

	DBC_ENSURE((ret && (refs > 0)) || (!ret && (refs >= 0)));

	return ret;
}

/*
 *  ======== cmm_register_gppsm_seg ========
 *  Purpose:
 *      Register a block of SM with the CMM to be used for later GPP SM
 *      allocations.
 */
int cmm_register_gppsm_seg(struct cmm_object *hcmm_mgr,
				  u32 dw_gpp_base_pa, u32 ul_size,
				  u32 dsp_addr_offset, s8 c_factor,
				  u32 dw_dsp_base, u32 ul_dsp_size,
				  u32 *sgmt_id, u32 gpp_base_va)
{
	struct cmm_object *cmm_mgr_obj = (struct cmm_object *)hcmm_mgr;
	struct cmm_allocator *psma = NULL;
	int status = 0;
	struct cmm_mnode *new_node;
	s32 slot_seg;

	DBC_REQUIRE(ul_size > 0);
	DBC_REQUIRE(sgmt_id != NULL);
	DBC_REQUIRE(dw_gpp_base_pa != 0);
	DBC_REQUIRE(gpp_base_va != 0);
	DBC_REQUIRE((c_factor <= CMM_ADDTODSPPA) &&
			(c_factor >= CMM_SUBFROMDSPPA));

	dev_dbg(bridge, "%s: dw_gpp_base_pa %x ul_size %x dsp_addr_offset %x "
			"dw_dsp_base %x ul_dsp_size %x gpp_base_va %x\n",
			__func__, dw_gpp_base_pa, ul_size, dsp_addr_offset,
			dw_dsp_base, ul_dsp_size, gpp_base_va);

	if (!hcmm_mgr)
		return -EFAULT;

	/* make sure we have room for another allocator */
	mutex_lock(&cmm_mgr_obj->cmm_lock);

	slot_seg = get_slot(cmm_mgr_obj);
	if (slot_seg < 0) {
		status = -EPERM;
		goto func_end;
	}

	/* Check if input ul_size is big enough to alloc at least one block */
	if (ul_size < cmm_mgr_obj->min_block_size) {
		status = -EINVAL;
		goto func_end;
	}

	/* create, zero, and tag an SM allocator object */
	psma = kzalloc(sizeof(struct cmm_allocator), GFP_KERNEL);
	if (!psma) {
		status = -ENOMEM;
		goto func_end;
	}

	psma->cmm_mgr = hcmm_mgr;	/* ref to parent */
	psma->shm_base = dw_gpp_base_pa;	/* SM Base phys */
	psma->sm_size = ul_size;	/* SM segment size in bytes */
	psma->vm_base = gpp_base_va;
	psma->dsp_phys_addr_offset = dsp_addr_offset;
	psma->c_factor = c_factor;
	psma->dsp_base = dw_dsp_base;
	psma->dsp_size = ul_dsp_size;
	if (psma->vm_base == 0) {
		status = -EPERM;
		goto func_end;
	}
	/* return the actual segment identifier */
	*sgmt_id = (u32) slot_seg + 1;

	INIT_LIST_HEAD(&psma->free_list);
	INIT_LIST_HEAD(&psma->in_use_list);

	/* Get a mem node for this hunk-o-memory */
	new_node = get_node(cmm_mgr_obj, dw_gpp_base_pa,
			psma->vm_base, ul_size);
	/* Place node on the SM allocator's free list */
	if (new_node) {
		list_add_tail(&new_node->link, &psma->free_list);
	} else {
		status = -ENOMEM;
		goto func_end;
	}
	/* make entry */
	cmm_mgr_obj->pa_gppsm_seg_tab[slot_seg] = psma;

func_end:
	/* Cleanup allocator */
	if (status && psma)
		un_register_gppsm_seg(psma);
	mutex_unlock(&cmm_mgr_obj->cmm_lock);

	return status;
}

/*
 *  ======== cmm_un_register_gppsm_seg ========
 *  Purpose:
 *      UnRegister GPP SM segments with the CMM.
 */
int cmm_un_register_gppsm_seg(struct cmm_object *hcmm_mgr,
				     u32 ul_seg_id)
{
	struct cmm_object *cmm_mgr_obj = (struct cmm_object *)hcmm_mgr;
	int status = 0;
	struct cmm_allocator *psma;
	u32 ul_id = ul_seg_id;

	DBC_REQUIRE(ul_seg_id > 0);
	if (!hcmm_mgr)
		return -EFAULT;

	if (ul_seg_id == CMM_ALLSEGMENTS)
		ul_id = 1;

	if ((ul_id <= 0) || (ul_id > CMM_MAXGPPSEGS))
		return -EINVAL;

	/*
	 * FIXME: CMM_MAXGPPSEGS == 1. why use a while cycle? Seems to me like
	 * the ul_seg_id is not needed here. It must be always 1.
	 */
	while (ul_id <= CMM_MAXGPPSEGS) {
		mutex_lock(&cmm_mgr_obj->cmm_lock);
		/* slot = seg_id-1 */
		psma = cmm_mgr_obj->pa_gppsm_seg_tab[ul_id - 1];
		if (psma != NULL) {
			un_register_gppsm_seg(psma);
			/* Set alctr ptr to NULL for future reuse */
			cmm_mgr_obj->pa_gppsm_seg_tab[ul_id - 1] = NULL;
		} else if (ul_seg_id != CMM_ALLSEGMENTS) {
			status = -EPERM;
		}
		mutex_unlock(&cmm_mgr_obj->cmm_lock);
		if (ul_seg_id != CMM_ALLSEGMENTS)
			break;

		ul_id++;
	}	/* end while */
	return status;
}

/*
 *  ======== un_register_gppsm_seg ========
 *  Purpose:
 *      UnRegister the SM allocator by freeing all its resources and
 *      nulling cmm mgr table entry.
 *  Note:
 *      This routine is always called within cmm lock crit sect.
 */
static void un_register_gppsm_seg(struct cmm_allocator *psma)
{
	struct cmm_mnode *curr, *tmp;

	DBC_REQUIRE(psma != NULL);

	/* free nodes on free list */
	list_for_each_entry_safe(curr, tmp, &psma->free_list, link) {
		list_del(&curr->link);
		kfree(curr);
	}

	/* free nodes on InUse list */
	list_for_each_entry_safe(curr, tmp, &psma->in_use_list, link) {
		list_del(&curr->link);
		kfree(curr);
	}

	if ((void *)psma->vm_base != NULL)
		MEM_UNMAP_LINEAR_ADDRESS((void *)psma->vm_base);

	/* Free allocator itself */
	kfree(psma);
}

/*
 *  ======== get_slot ========
 *  Purpose:
 *      An available slot # is returned. Returns negative on failure.
 */
static s32 get_slot(struct cmm_object *cmm_mgr_obj)
{
	s32 slot_seg = -1;	/* neg on failure */
	DBC_REQUIRE(cmm_mgr_obj != NULL);
	/* get first available slot in cmm mgr SMSegTab[] */
	for (slot_seg = 0; slot_seg < CMM_MAXGPPSEGS; slot_seg++) {
		if (cmm_mgr_obj->pa_gppsm_seg_tab[slot_seg] == NULL)
			break;

	}
	if (slot_seg == CMM_MAXGPPSEGS)
		slot_seg = -1;	/* failed */

	return slot_seg;
}

/*
 *  ======== get_node ========
 *  Purpose:
 *      Get a memory node from freelist or create a new one.
 */
static struct cmm_mnode *get_node(struct cmm_object *cmm_mgr_obj, u32 dw_pa,
				  u32 dw_va, u32 ul_size)
{
	struct cmm_mnode *pnode;

	DBC_REQUIRE(cmm_mgr_obj != NULL);
	DBC_REQUIRE(dw_pa != 0);
	DBC_REQUIRE(dw_va != 0);
	DBC_REQUIRE(ul_size != 0);

	/* Check cmm mgr's node freelist */
	if (list_empty(&cmm_mgr_obj->node_free_list)) {
		pnode = kzalloc(sizeof(struct cmm_mnode), GFP_KERNEL);
		if (!pnode)
			return NULL;
	} else {
		/* surely a valid element */
		pnode = list_first_entry(&cmm_mgr_obj->node_free_list,
				struct cmm_mnode, link);
		list_del_init(&pnode->link);
	}

	pnode->pa = dw_pa;
	pnode->va = dw_va;
	pnode->size = ul_size;

	return pnode;
}

/*
 *  ======== delete_node ========
 *  Purpose:
 *      Put a memory node on the cmm nodelist for later use.
 *      Doesn't actually delete the node. Heap thrashing friendly.
 */
static void delete_node(struct cmm_object *cmm_mgr_obj, struct cmm_mnode *pnode)
{
	DBC_REQUIRE(pnode != NULL);
	list_add_tail(&pnode->link, &cmm_mgr_obj->node_free_list);
}

/*
 * ====== get_free_block ========
 *  Purpose:
 *      Scan the free block list and return the first block that satisfies
 *      the size.
 */
static struct cmm_mnode *get_free_block(struct cmm_allocator *allocator,
					u32 usize)
{
	struct cmm_mnode *node, *tmp;

	if (!allocator)
		return NULL;

	list_for_each_entry_safe(node, tmp, &allocator->free_list, link) {
		if (usize <= node->size) {
			list_del(&node->link);
			return node;
		}
	}

	return NULL;
}

/*
 *  ======== add_to_free_list ========
 *  Purpose:
 *      Coalesce node into the freelist in ascending size order.
 */
static void add_to_free_list(struct cmm_allocator *allocator,
			     struct cmm_mnode *node)
{
	struct cmm_mnode *curr;

	if (!node) {
		pr_err("%s: failed - node is NULL\n", __func__);
		return;
	}

	list_for_each_entry(curr, &allocator->free_list, link) {
		if (NEXT_PA(curr) == node->pa) {
			curr->size += node->size;
			delete_node(allocator->cmm_mgr, node);
			return;
		}
		if (curr->pa == NEXT_PA(node)) {
			curr->pa = node->pa;
			curr->va = node->va;
			curr->size += node->size;
			delete_node(allocator->cmm_mgr, node);
			return;
		}
	}
	list_for_each_entry(curr, &allocator->free_list, link) {
		if (curr->size >= node->size) {
			list_add_tail(&node->link, &curr->link);
			return;
		}
	}
	list_add_tail(&node->link, &allocator->free_list);
}

/*
 * ======== get_allocator ========
 *  Purpose:
 *      Return the allocator for the given SM Segid.
 *      SegIds:  1,2,3..max.
 */
static struct cmm_allocator *get_allocator(struct cmm_object *cmm_mgr_obj,
					   u32 ul_seg_id)
{
	DBC_REQUIRE(cmm_mgr_obj != NULL);
	DBC_REQUIRE((ul_seg_id > 0) && (ul_seg_id <= CMM_MAXGPPSEGS));

	return cmm_mgr_obj->pa_gppsm_seg_tab[ul_seg_id - 1];
}

/*
 *  The CMM_Xlator[xxx] routines below are used by Node and Stream
 *  to perform SM address translation to the client process address space.
 *  A "translator" object is created by a node/stream for each SM seg used.
 */

/*
 *  ======== cmm_xlator_create ========
 *  Purpose:
 *      Create an address translator object.
 */
int cmm_xlator_create(struct cmm_xlatorobject **xlator,
			     struct cmm_object *hcmm_mgr,
			     struct cmm_xlatorattrs *xlator_attrs)
{
	struct cmm_xlator *xlator_object = NULL;
	int status = 0;

	DBC_REQUIRE(refs > 0);
	DBC_REQUIRE(xlator != NULL);
	DBC_REQUIRE(hcmm_mgr != NULL);

	*xlator = NULL;
	if (xlator_attrs == NULL)
		xlator_attrs = &cmm_dfltxlatorattrs;	/* set defaults */

	xlator_object = kzalloc(sizeof(struct cmm_xlator), GFP_KERNEL);
	if (xlator_object != NULL) {
		xlator_object->cmm_mgr = hcmm_mgr;	/* ref back to CMM */
		/* SM seg_id */
		xlator_object->seg_id = xlator_attrs->seg_id;
	} else {
		status = -ENOMEM;
	}
	if (!status)
		*xlator = (struct cmm_xlatorobject *)xlator_object;

	return status;
}

/*
 *  ======== cmm_xlator_alloc_buf ========
 */
void *cmm_xlator_alloc_buf(struct cmm_xlatorobject *xlator, void *va_buf,
			   u32 pa_size)
{
	struct cmm_xlator *xlator_obj = (struct cmm_xlator *)xlator;
	void *pbuf = NULL;
	void *tmp_va_buff;
	struct cmm_attrs attrs;

	DBC_REQUIRE(refs > 0);
	DBC_REQUIRE(xlator != NULL);
	DBC_REQUIRE(xlator_obj->cmm_mgr != NULL);
	DBC_REQUIRE(va_buf != NULL);
	DBC_REQUIRE(pa_size > 0);
	DBC_REQUIRE(xlator_obj->seg_id > 0);

	if (xlator_obj) {
		attrs.seg_id = xlator_obj->seg_id;
		__raw_writel(0, va_buf);
		/* Alloc SM */
		pbuf =
		    cmm_calloc_buf(xlator_obj->cmm_mgr, pa_size, &attrs, NULL);
		if (pbuf) {
			/* convert to translator(node/strm) process Virtual
			 * address */
			 tmp_va_buff = cmm_xlator_translate(xlator,
							 pbuf, CMM_PA2VA);
			__raw_writel((u32)tmp_va_buff, va_buf);
		}
	}
	return pbuf;
}

/*
 *  ======== cmm_xlator_free_buf ========
 *  Purpose:
 *      Free the given SM buffer and descriptor.
 *      Does not free virtual memory.
 */
int cmm_xlator_free_buf(struct cmm_xlatorobject *xlator, void *buf_va)
{
	struct cmm_xlator *xlator_obj = (struct cmm_xlator *)xlator;
	int status = -EPERM;
	void *buf_pa = NULL;

	DBC_REQUIRE(refs > 0);
	DBC_REQUIRE(buf_va != NULL);
	DBC_REQUIRE(xlator_obj->seg_id > 0);

	if (xlator_obj) {
		/* convert Va to Pa so we can free it. */
		buf_pa = cmm_xlator_translate(xlator, buf_va, CMM_VA2PA);
		if (buf_pa) {
			status = cmm_free_buf(xlator_obj->cmm_mgr, buf_pa,
					      xlator_obj->seg_id);
			if (status) {
				/* Uh oh, this shouldn't happen. Descriptor
				 * gone! */
				DBC_ASSERT(false);	/* CMM is leaking mem */
			}
		}
	}
	return status;
}

/*
 *  ======== cmm_xlator_info ========
 *  Purpose:
 *      Set/Get translator info.
 */
int cmm_xlator_info(struct cmm_xlatorobject *xlator, u8 ** paddr,
			   u32 ul_size, u32 segm_id, bool set_info)
{
	struct cmm_xlator *xlator_obj = (struct cmm_xlator *)xlator;
	int status = 0;

	DBC_REQUIRE(refs > 0);
	DBC_REQUIRE(paddr != NULL);
	DBC_REQUIRE((segm_id > 0) && (segm_id <= CMM_MAXGPPSEGS));

	if (xlator_obj) {
		if (set_info) {
			/* set translators virtual address range */
			xlator_obj->virt_base = (u32) *paddr;
			xlator_obj->virt_size = ul_size;
		} else {	/* return virt base address */
			*paddr = (u8 *) xlator_obj->virt_base;
		}
	} else {
		status = -EFAULT;
	}
	return status;
}

/*
 *  ======== cmm_xlator_translate ========
 */
void *cmm_xlator_translate(struct cmm_xlatorobject *xlator, void *paddr,
			   enum cmm_xlatetype xtype)
{
	u32 dw_addr_xlate = 0;
	struct cmm_xlator *xlator_obj = (struct cmm_xlator *)xlator;
	struct cmm_object *cmm_mgr_obj = NULL;
	struct cmm_allocator *allocator = NULL;
	u32 dw_offset = 0;

	DBC_REQUIRE(refs > 0);
	DBC_REQUIRE(paddr != NULL);
	DBC_REQUIRE((xtype >= CMM_VA2PA) && (xtype <= CMM_DSPPA2PA));

	if (!xlator_obj)
		goto loop_cont;

	cmm_mgr_obj = (struct cmm_object *)xlator_obj->cmm_mgr;
	/* get this translator's default SM allocator */
	DBC_ASSERT(xlator_obj->seg_id > 0);
	allocator = cmm_mgr_obj->pa_gppsm_seg_tab[xlator_obj->seg_id - 1];
	if (!allocator)
		goto loop_cont;

	if ((xtype == CMM_VA2DSPPA) || (xtype == CMM_VA2PA) ||
	    (xtype == CMM_PA2VA)) {
		if (xtype == CMM_PA2VA) {
			/* Gpp Va = Va Base + offset */
			dw_offset = (u8 *) paddr - (u8 *) (allocator->shm_base -
							   allocator->
							   dsp_size);
			dw_addr_xlate = xlator_obj->virt_base + dw_offset;
			/* Check if translated Va base is in range */
			if ((dw_addr_xlate < xlator_obj->virt_base) ||
			    (dw_addr_xlate >=
			     (xlator_obj->virt_base +
			      xlator_obj->virt_size))) {
				dw_addr_xlate = 0;	/* bad address */
			}
		} else {
			/* Gpp PA =  Gpp Base + offset */
			dw_offset =
			    (u8 *) paddr - (u8 *) xlator_obj->virt_base;
			dw_addr_xlate =
			    allocator->shm_base - allocator->dsp_size +
			    dw_offset;
		}
	} else {
		dw_addr_xlate = (u32) paddr;
	}
	/*Now convert address to proper target physical address if needed */
	if ((xtype == CMM_VA2DSPPA) || (xtype == CMM_PA2DSPPA)) {
		/* Got Gpp Pa now, convert to DSP Pa */
		dw_addr_xlate =
		    GPPPA2DSPPA((allocator->shm_base - allocator->dsp_size),
				dw_addr_xlate,
				allocator->dsp_phys_addr_offset *
				allocator->c_factor);
	} else if (xtype == CMM_DSPPA2PA) {
		/* Got DSP Pa, convert to GPP Pa */
		dw_addr_xlate =
		    DSPPA2GPPPA(allocator->shm_base - allocator->dsp_size,
				dw_addr_xlate,
				allocator->dsp_phys_addr_offset *
				allocator->c_factor);
	}
loop_cont:
	return (void *)dw_addr_xlate;
}