forked from torvalds/linux
-
Notifications
You must be signed in to change notification settings - Fork 0
/
nv_hw.c
1687 lines (1553 loc) · 50.5 KB
/
nv_hw.c
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
/***************************************************************************\
|* *|
|* Copyright 1993-2003 NVIDIA, Corporation. All rights reserved. *|
|* *|
|* NOTICE TO USER: The source code is copyrighted under U.S. and *|
|* international laws. Users and possessors of this source code are *|
|* hereby granted a nonexclusive, royalty-free copyright license to *|
|* use this code in individual and commercial software. *|
|* *|
|* Any use of this source code must include, in the user documenta- *|
|* tion and internal comments to the code, notices to the end user *|
|* as follows: *|
|* *|
|* Copyright 1993-2003 NVIDIA, Corporation. All rights reserved. *|
|* *|
|* NVIDIA, CORPORATION MAKES NO REPRESENTATION ABOUT THE SUITABILITY *|
|* OF THIS SOURCE CODE FOR ANY PURPOSE. IT IS PROVIDED "AS IS" *|
|* WITHOUT EXPRESS OR IMPLIED WARRANTY OF ANY KIND. NVIDIA, CORPOR- *|
|* ATION DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOURCE CODE, *|
|* INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY, NONINFRINGE- *|
|* MENT, AND FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT SHALL *|
|* NVIDIA, CORPORATION BE LIABLE FOR ANY SPECIAL, INDIRECT, INCI- *|
|* DENTAL, OR CONSEQUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RE- *|
|* SULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION *|
|* OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF *|
|* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOURCE CODE. *|
|* *|
|* U.S. Government End Users. This source code is a "commercial *|
|* item," as that term is defined at 48 C.F.R. 2.101 (OCT 1995), *|
|* consisting of "commercial computer software" and "commercial *|
|* computer software documentation," as such terms are used in *|
|* 48 C.F.R. 12.212 (SEPT 1995) and is provided to the U.S. Govern- *|
|* ment only as a commercial end item. Consistent with 48 C.F.R. *|
|* 12.212 and 48 C.F.R. 227.7202-1 through 227.7202-4 (JUNE 1995), *|
|* all U.S. Government End Users acquire the source code with only *|
|* those rights set forth herein. *|
|* *|
\***************************************************************************/
/*
* GPL Licensing Note - According to Mark Vojkovich, author of the Xorg/
* XFree86 'nv' driver, this source code is provided under MIT-style licensing
* where the source code is provided "as is" without warranty of any kind.
* The only usage restriction is for the copyright notices to be retained
* whenever code is used.
*
* Antonino Daplas <[email protected]> 2005-03-11
*/
/* $XFree86: xc/programs/Xserver/hw/xfree86/drivers/nv/nv_hw.c,v 1.4 2003/11/03 05:11:25 tsi Exp $ */
#include <linux/pci.h>
#include "nv_type.h"
#include "nv_local.h"
#include "nv_proto.h"
void NVLockUnlock(struct nvidia_par *par, int Lock)
{
u8 cr11;
VGA_WR08(par->PCIO, 0x3D4, 0x1F);
VGA_WR08(par->PCIO, 0x3D5, Lock ? 0x99 : 0x57);
VGA_WR08(par->PCIO, 0x3D4, 0x11);
cr11 = VGA_RD08(par->PCIO, 0x3D5);
if (Lock)
cr11 |= 0x80;
else
cr11 &= ~0x80;
VGA_WR08(par->PCIO, 0x3D5, cr11);
}
int NVShowHideCursor(struct nvidia_par *par, int ShowHide)
{
int cur = par->CurrentState->cursor1;
par->CurrentState->cursor1 = (par->CurrentState->cursor1 & 0xFE) |
(ShowHide & 0x01);
VGA_WR08(par->PCIO, 0x3D4, 0x31);
VGA_WR08(par->PCIO, 0x3D5, par->CurrentState->cursor1);
if (par->Architecture == NV_ARCH_40)
NV_WR32(par->PRAMDAC, 0x0300, NV_RD32(par->PRAMDAC, 0x0300));
return (cur & 0x01);
}
/****************************************************************************\
* *
* The video arbitration routines calculate some "magic" numbers. Fixes *
* the snow seen when accessing the framebuffer without it. *
* It just works (I hope). *
* *
\****************************************************************************/
typedef struct {
int graphics_lwm;
int video_lwm;
int graphics_burst_size;
int video_burst_size;
int valid;
} nv4_fifo_info;
typedef struct {
int pclk_khz;
int mclk_khz;
int nvclk_khz;
char mem_page_miss;
char mem_latency;
int memory_width;
char enable_video;
char gr_during_vid;
char pix_bpp;
char mem_aligned;
char enable_mp;
} nv4_sim_state;
typedef struct {
int graphics_lwm;
int video_lwm;
int graphics_burst_size;
int video_burst_size;
int valid;
} nv10_fifo_info;
typedef struct {
int pclk_khz;
int mclk_khz;
int nvclk_khz;
char mem_page_miss;
char mem_latency;
u32 memory_type;
int memory_width;
char enable_video;
char gr_during_vid;
char pix_bpp;
char mem_aligned;
char enable_mp;
} nv10_sim_state;
static void nvGetClocks(struct nvidia_par *par, unsigned int *MClk,
unsigned int *NVClk)
{
unsigned int pll, N, M, MB, NB, P;
if (par->Architecture >= NV_ARCH_40) {
pll = NV_RD32(par->PMC, 0x4020);
P = (pll >> 16) & 0x07;
pll = NV_RD32(par->PMC, 0x4024);
M = pll & 0xFF;
N = (pll >> 8) & 0xFF;
if (((par->Chipset & 0xfff0) == 0x0290) ||
((par->Chipset & 0xfff0) == 0x0390)) {
MB = 1;
NB = 1;
} else {
MB = (pll >> 16) & 0xFF;
NB = (pll >> 24) & 0xFF;
}
*MClk = ((N * NB * par->CrystalFreqKHz) / (M * MB)) >> P;
pll = NV_RD32(par->PMC, 0x4000);
P = (pll >> 16) & 0x07;
pll = NV_RD32(par->PMC, 0x4004);
M = pll & 0xFF;
N = (pll >> 8) & 0xFF;
MB = (pll >> 16) & 0xFF;
NB = (pll >> 24) & 0xFF;
*NVClk = ((N * NB * par->CrystalFreqKHz) / (M * MB)) >> P;
} else if (par->twoStagePLL) {
pll = NV_RD32(par->PRAMDAC0, 0x0504);
M = pll & 0xFF;
N = (pll >> 8) & 0xFF;
P = (pll >> 16) & 0x0F;
pll = NV_RD32(par->PRAMDAC0, 0x0574);
if (pll & 0x80000000) {
MB = pll & 0xFF;
NB = (pll >> 8) & 0xFF;
} else {
MB = 1;
NB = 1;
}
*MClk = ((N * NB * par->CrystalFreqKHz) / (M * MB)) >> P;
pll = NV_RD32(par->PRAMDAC0, 0x0500);
M = pll & 0xFF;
N = (pll >> 8) & 0xFF;
P = (pll >> 16) & 0x0F;
pll = NV_RD32(par->PRAMDAC0, 0x0570);
if (pll & 0x80000000) {
MB = pll & 0xFF;
NB = (pll >> 8) & 0xFF;
} else {
MB = 1;
NB = 1;
}
*NVClk = ((N * NB * par->CrystalFreqKHz) / (M * MB)) >> P;
} else
if (((par->Chipset & 0x0ff0) == 0x0300) ||
((par->Chipset & 0x0ff0) == 0x0330)) {
pll = NV_RD32(par->PRAMDAC0, 0x0504);
M = pll & 0x0F;
N = (pll >> 8) & 0xFF;
P = (pll >> 16) & 0x07;
if (pll & 0x00000080) {
MB = (pll >> 4) & 0x07;
NB = (pll >> 19) & 0x1f;
} else {
MB = 1;
NB = 1;
}
*MClk = ((N * NB * par->CrystalFreqKHz) / (M * MB)) >> P;
pll = NV_RD32(par->PRAMDAC0, 0x0500);
M = pll & 0x0F;
N = (pll >> 8) & 0xFF;
P = (pll >> 16) & 0x07;
if (pll & 0x00000080) {
MB = (pll >> 4) & 0x07;
NB = (pll >> 19) & 0x1f;
} else {
MB = 1;
NB = 1;
}
*NVClk = ((N * NB * par->CrystalFreqKHz) / (M * MB)) >> P;
} else {
pll = NV_RD32(par->PRAMDAC0, 0x0504);
M = pll & 0xFF;
N = (pll >> 8) & 0xFF;
P = (pll >> 16) & 0x0F;
*MClk = (N * par->CrystalFreqKHz / M) >> P;
pll = NV_RD32(par->PRAMDAC0, 0x0500);
M = pll & 0xFF;
N = (pll >> 8) & 0xFF;
P = (pll >> 16) & 0x0F;
*NVClk = (N * par->CrystalFreqKHz / M) >> P;
}
}
static void nv4CalcArbitration(nv4_fifo_info * fifo, nv4_sim_state * arb)
{
int data, pagemiss, cas, width, video_enable, bpp;
int nvclks, mclks, pclks, vpagemiss, crtpagemiss, vbs;
int found, mclk_extra, mclk_loop, cbs, m1, p1;
int mclk_freq, pclk_freq, nvclk_freq, mp_enable;
int us_m, us_n, us_p, video_drain_rate, crtc_drain_rate;
int vpm_us, us_video, vlwm, video_fill_us, cpm_us, us_crt, clwm;
fifo->valid = 1;
pclk_freq = arb->pclk_khz;
mclk_freq = arb->mclk_khz;
nvclk_freq = arb->nvclk_khz;
pagemiss = arb->mem_page_miss;
cas = arb->mem_latency;
width = arb->memory_width >> 6;
video_enable = arb->enable_video;
bpp = arb->pix_bpp;
mp_enable = arb->enable_mp;
clwm = 0;
vlwm = 0;
cbs = 128;
pclks = 2;
nvclks = 2;
nvclks += 2;
nvclks += 1;
mclks = 5;
mclks += 3;
mclks += 1;
mclks += cas;
mclks += 1;
mclks += 1;
mclks += 1;
mclks += 1;
mclk_extra = 3;
nvclks += 2;
nvclks += 1;
nvclks += 1;
nvclks += 1;
if (mp_enable)
mclks += 4;
nvclks += 0;
pclks += 0;
found = 0;
vbs = 0;
while (found != 1) {
fifo->valid = 1;
found = 1;
mclk_loop = mclks + mclk_extra;
us_m = mclk_loop * 1000 * 1000 / mclk_freq;
us_n = nvclks * 1000 * 1000 / nvclk_freq;
us_p = nvclks * 1000 * 1000 / pclk_freq;
if (video_enable) {
video_drain_rate = pclk_freq * 2;
crtc_drain_rate = pclk_freq * bpp / 8;
vpagemiss = 2;
vpagemiss += 1;
crtpagemiss = 2;
vpm_us =
(vpagemiss * pagemiss) * 1000 * 1000 / mclk_freq;
if (nvclk_freq * 2 > mclk_freq * width)
video_fill_us =
cbs * 1000 * 1000 / 16 / nvclk_freq;
else
video_fill_us =
cbs * 1000 * 1000 / (8 * width) /
mclk_freq;
us_video = vpm_us + us_m + us_n + us_p + video_fill_us;
vlwm = us_video * video_drain_rate / (1000 * 1000);
vlwm++;
vbs = 128;
if (vlwm > 128)
vbs = 64;
if (vlwm > (256 - 64))
vbs = 32;
if (nvclk_freq * 2 > mclk_freq * width)
video_fill_us =
vbs * 1000 * 1000 / 16 / nvclk_freq;
else
video_fill_us =
vbs * 1000 * 1000 / (8 * width) /
mclk_freq;
cpm_us =
crtpagemiss * pagemiss * 1000 * 1000 / mclk_freq;
us_crt =
us_video + video_fill_us + cpm_us + us_m + us_n +
us_p;
clwm = us_crt * crtc_drain_rate / (1000 * 1000);
clwm++;
} else {
crtc_drain_rate = pclk_freq * bpp / 8;
crtpagemiss = 2;
crtpagemiss += 1;
cpm_us =
crtpagemiss * pagemiss * 1000 * 1000 / mclk_freq;
us_crt = cpm_us + us_m + us_n + us_p;
clwm = us_crt * crtc_drain_rate / (1000 * 1000);
clwm++;
}
m1 = clwm + cbs - 512;
p1 = m1 * pclk_freq / mclk_freq;
p1 = p1 * bpp / 8;
if ((p1 < m1) && (m1 > 0)) {
fifo->valid = 0;
found = 0;
if (mclk_extra == 0)
found = 1;
mclk_extra--;
} else if (video_enable) {
if ((clwm > 511) || (vlwm > 255)) {
fifo->valid = 0;
found = 0;
if (mclk_extra == 0)
found = 1;
mclk_extra--;
}
} else {
if (clwm > 519) {
fifo->valid = 0;
found = 0;
if (mclk_extra == 0)
found = 1;
mclk_extra--;
}
}
if (clwm < 384)
clwm = 384;
if (vlwm < 128)
vlwm = 128;
data = (int)(clwm);
fifo->graphics_lwm = data;
fifo->graphics_burst_size = 128;
data = (int)((vlwm + 15));
fifo->video_lwm = data;
fifo->video_burst_size = vbs;
}
}
static void nv4UpdateArbitrationSettings(unsigned VClk,
unsigned pixelDepth,
unsigned *burst,
unsigned *lwm, struct nvidia_par *par)
{
nv4_fifo_info fifo_data;
nv4_sim_state sim_data;
unsigned int MClk, NVClk, cfg1;
nvGetClocks(par, &MClk, &NVClk);
cfg1 = NV_RD32(par->PFB, 0x00000204);
sim_data.pix_bpp = (char)pixelDepth;
sim_data.enable_video = 0;
sim_data.enable_mp = 0;
sim_data.memory_width = (NV_RD32(par->PEXTDEV, 0x0000) & 0x10) ?
128 : 64;
sim_data.mem_latency = (char)cfg1 & 0x0F;
sim_data.mem_aligned = 1;
sim_data.mem_page_miss =
(char)(((cfg1 >> 4) & 0x0F) + ((cfg1 >> 31) & 0x01));
sim_data.gr_during_vid = 0;
sim_data.pclk_khz = VClk;
sim_data.mclk_khz = MClk;
sim_data.nvclk_khz = NVClk;
nv4CalcArbitration(&fifo_data, &sim_data);
if (fifo_data.valid) {
int b = fifo_data.graphics_burst_size >> 4;
*burst = 0;
while (b >>= 1)
(*burst)++;
*lwm = fifo_data.graphics_lwm >> 3;
}
}
static void nv10CalcArbitration(nv10_fifo_info * fifo, nv10_sim_state * arb)
{
int data, pagemiss, width, video_enable, bpp;
int nvclks, mclks, pclks, vpagemiss, crtpagemiss;
int nvclk_fill;
int found, mclk_extra, mclk_loop, cbs, m1;
int mclk_freq, pclk_freq, nvclk_freq, mp_enable;
int us_m, us_m_min, us_n, us_p, crtc_drain_rate;
int vus_m;
int vpm_us, us_video, cpm_us, us_crt, clwm;
int clwm_rnd_down;
int m2us, us_pipe_min, p1clk, p2;
int min_mclk_extra;
int us_min_mclk_extra;
fifo->valid = 1;
pclk_freq = arb->pclk_khz; /* freq in KHz */
mclk_freq = arb->mclk_khz;
nvclk_freq = arb->nvclk_khz;
pagemiss = arb->mem_page_miss;
width = arb->memory_width / 64;
video_enable = arb->enable_video;
bpp = arb->pix_bpp;
mp_enable = arb->enable_mp;
clwm = 0;
cbs = 512;
pclks = 4; /* lwm detect. */
nvclks = 3; /* lwm -> sync. */
nvclks += 2; /* fbi bus cycles (1 req + 1 busy) */
/* 2 edge sync. may be very close to edge so just put one. */
mclks = 1;
mclks += 1; /* arb_hp_req */
mclks += 5; /* ap_hp_req tiling pipeline */
mclks += 2; /* tc_req latency fifo */
mclks += 2; /* fb_cas_n_ memory request to fbio block */
mclks += 7; /* sm_d_rdv data returned from fbio block */
/* fb.rd.d.Put_gc need to accumulate 256 bits for read */
if (arb->memory_type == 0)
if (arb->memory_width == 64) /* 64 bit bus */
mclks += 4;
else
mclks += 2;
else if (arb->memory_width == 64) /* 64 bit bus */
mclks += 2;
else
mclks += 1;
if ((!video_enable) && (arb->memory_width == 128)) {
mclk_extra = (bpp == 32) ? 31 : 42; /* Margin of error */
min_mclk_extra = 17;
} else {
mclk_extra = (bpp == 32) ? 8 : 4; /* Margin of error */
/* mclk_extra = 4; *//* Margin of error */
min_mclk_extra = 18;
}
/* 2 edge sync. may be very close to edge so just put one. */
nvclks += 1;
nvclks += 1; /* fbi_d_rdv_n */
nvclks += 1; /* Fbi_d_rdata */
nvclks += 1; /* crtfifo load */
if (mp_enable)
mclks += 4; /* Mp can get in with a burst of 8. */
/* Extra clocks determined by heuristics */
nvclks += 0;
pclks += 0;
found = 0;
while (found != 1) {
fifo->valid = 1;
found = 1;
mclk_loop = mclks + mclk_extra;
/* Mclk latency in us */
us_m = mclk_loop * 1000 * 1000 / mclk_freq;
/* Minimum Mclk latency in us */
us_m_min = mclks * 1000 * 1000 / mclk_freq;
us_min_mclk_extra = min_mclk_extra * 1000 * 1000 / mclk_freq;
/* nvclk latency in us */
us_n = nvclks * 1000 * 1000 / nvclk_freq;
/* nvclk latency in us */
us_p = pclks * 1000 * 1000 / pclk_freq;
us_pipe_min = us_m_min + us_n + us_p;
/* Mclk latency in us */
vus_m = mclk_loop * 1000 * 1000 / mclk_freq;
if (video_enable) {
crtc_drain_rate = pclk_freq * bpp / 8; /* MB/s */
vpagemiss = 1; /* self generating page miss */
vpagemiss += 1; /* One higher priority before */
crtpagemiss = 2; /* self generating page miss */
if (mp_enable)
crtpagemiss += 1; /* if MA0 conflict */
vpm_us =
(vpagemiss * pagemiss) * 1000 * 1000 / mclk_freq;
/* Video has separate read return path */
us_video = vpm_us + vus_m;
cpm_us =
crtpagemiss * pagemiss * 1000 * 1000 / mclk_freq;
/* Wait for video */
us_crt = us_video
+ cpm_us /* CRT Page miss */
+ us_m + us_n + us_p /* other latency */
;
clwm = us_crt * crtc_drain_rate / (1000 * 1000);
/* fixed point <= float_point - 1. Fixes that */
clwm++;
} else {
/* bpp * pclk/8 */
crtc_drain_rate = pclk_freq * bpp / 8;
crtpagemiss = 1; /* self generating page miss */
crtpagemiss += 1; /* MA0 page miss */
if (mp_enable)
crtpagemiss += 1; /* if MA0 conflict */
cpm_us =
crtpagemiss * pagemiss * 1000 * 1000 / mclk_freq;
us_crt = cpm_us + us_m + us_n + us_p;
clwm = us_crt * crtc_drain_rate / (1000 * 1000);
/* fixed point <= float_point - 1. Fixes that */
clwm++;
/* Finally, a heuristic check when width == 64 bits */
if (width == 1) {
nvclk_fill = nvclk_freq * 8;
if (crtc_drain_rate * 100 >= nvclk_fill * 102)
/*Large number to fail */
clwm = 0xfff;
else if (crtc_drain_rate * 100 >=
nvclk_fill * 98) {
clwm = 1024;
cbs = 512;
}
}
}
/*
Overfill check:
*/
clwm_rnd_down = ((int)clwm / 8) * 8;
if (clwm_rnd_down < clwm)
clwm += 8;
m1 = clwm + cbs - 1024; /* Amount of overfill */
m2us = us_pipe_min + us_min_mclk_extra;
/* pclk cycles to drain */
p1clk = m2us * pclk_freq / (1000 * 1000);
p2 = p1clk * bpp / 8; /* bytes drained. */
if ((p2 < m1) && (m1 > 0)) {
fifo->valid = 0;
found = 0;
if (min_mclk_extra == 0) {
if (cbs <= 32) {
/* Can't adjust anymore! */
found = 1;
} else {
/* reduce the burst size */
cbs = cbs / 2;
}
} else {
min_mclk_extra--;
}
} else {
if (clwm > 1023) { /* Have some margin */
fifo->valid = 0;
found = 0;
if (min_mclk_extra == 0)
/* Can't adjust anymore! */
found = 1;
else
min_mclk_extra--;
}
}
if (clwm < (1024 - cbs + 8))
clwm = 1024 - cbs + 8;
data = (int)(clwm);
/* printf("CRT LWM: %f bytes, prog: 0x%x, bs: 256\n",
clwm, data ); */
fifo->graphics_lwm = data;
fifo->graphics_burst_size = cbs;
fifo->video_lwm = 1024;
fifo->video_burst_size = 512;
}
}
static void nv10UpdateArbitrationSettings(unsigned VClk,
unsigned pixelDepth,
unsigned *burst,
unsigned *lwm,
struct nvidia_par *par)
{
nv10_fifo_info fifo_data;
nv10_sim_state sim_data;
unsigned int MClk, NVClk, cfg1;
nvGetClocks(par, &MClk, &NVClk);
cfg1 = NV_RD32(par->PFB, 0x0204);
sim_data.pix_bpp = (char)pixelDepth;
sim_data.enable_video = 1;
sim_data.enable_mp = 0;
sim_data.memory_type = (NV_RD32(par->PFB, 0x0200) & 0x01) ? 1 : 0;
sim_data.memory_width = (NV_RD32(par->PEXTDEV, 0x0000) & 0x10) ?
128 : 64;
sim_data.mem_latency = (char)cfg1 & 0x0F;
sim_data.mem_aligned = 1;
sim_data.mem_page_miss =
(char)(((cfg1 >> 4) & 0x0F) + ((cfg1 >> 31) & 0x01));
sim_data.gr_during_vid = 0;
sim_data.pclk_khz = VClk;
sim_data.mclk_khz = MClk;
sim_data.nvclk_khz = NVClk;
nv10CalcArbitration(&fifo_data, &sim_data);
if (fifo_data.valid) {
int b = fifo_data.graphics_burst_size >> 4;
*burst = 0;
while (b >>= 1)
(*burst)++;
*lwm = fifo_data.graphics_lwm >> 3;
}
}
static void nv30UpdateArbitrationSettings (
struct nvidia_par *par,
unsigned int *burst,
unsigned int *lwm
)
{
unsigned int MClk, NVClk;
unsigned int fifo_size, burst_size, graphics_lwm;
fifo_size = 2048;
burst_size = 512;
graphics_lwm = fifo_size - burst_size;
nvGetClocks(par, &MClk, &NVClk);
*burst = 0;
burst_size >>= 5;
while(burst_size >>= 1) (*burst)++;
*lwm = graphics_lwm >> 3;
}
static void nForceUpdateArbitrationSettings(unsigned VClk,
unsigned pixelDepth,
unsigned *burst,
unsigned *lwm,
struct nvidia_par *par)
{
nv10_fifo_info fifo_data;
nv10_sim_state sim_data;
unsigned int M, N, P, pll, MClk, NVClk, memctrl;
struct pci_dev *dev;
if ((par->Chipset & 0x0FF0) == 0x01A0) {
unsigned int uMClkPostDiv;
dev = pci_get_bus_and_slot(0, 3);
pci_read_config_dword(dev, 0x6C, &uMClkPostDiv);
uMClkPostDiv = (uMClkPostDiv >> 8) & 0xf;
if (!uMClkPostDiv)
uMClkPostDiv = 4;
MClk = 400000 / uMClkPostDiv;
} else {
dev = pci_get_bus_and_slot(0, 5);
pci_read_config_dword(dev, 0x4c, &MClk);
MClk /= 1000;
}
pci_dev_put(dev);
pll = NV_RD32(par->PRAMDAC0, 0x0500);
M = (pll >> 0) & 0xFF;
N = (pll >> 8) & 0xFF;
P = (pll >> 16) & 0x0F;
NVClk = (N * par->CrystalFreqKHz / M) >> P;
sim_data.pix_bpp = (char)pixelDepth;
sim_data.enable_video = 0;
sim_data.enable_mp = 0;
dev = pci_get_bus_and_slot(0, 1);
pci_read_config_dword(dev, 0x7C, &sim_data.memory_type);
pci_dev_put(dev);
sim_data.memory_type = (sim_data.memory_type >> 12) & 1;
sim_data.memory_width = 64;
dev = pci_get_bus_and_slot(0, 3);
pci_read_config_dword(dev, 0, &memctrl);
pci_dev_put(dev);
memctrl >>= 16;
if ((memctrl == 0x1A9) || (memctrl == 0x1AB) || (memctrl == 0x1ED)) {
u32 dimm[3];
dev = pci_get_bus_and_slot(0, 2);
pci_read_config_dword(dev, 0x40, &dimm[0]);
dimm[0] = (dimm[0] >> 8) & 0x4f;
pci_read_config_dword(dev, 0x44, &dimm[1]);
dimm[1] = (dimm[1] >> 8) & 0x4f;
pci_read_config_dword(dev, 0x48, &dimm[2]);
dimm[2] = (dimm[2] >> 8) & 0x4f;
if ((dimm[0] + dimm[1]) != dimm[2]) {
printk("nvidiafb: your nForce DIMMs are not arranged "
"in optimal banks!\n");
}
pci_dev_put(dev);
}
sim_data.mem_latency = 3;
sim_data.mem_aligned = 1;
sim_data.mem_page_miss = 10;
sim_data.gr_during_vid = 0;
sim_data.pclk_khz = VClk;
sim_data.mclk_khz = MClk;
sim_data.nvclk_khz = NVClk;
nv10CalcArbitration(&fifo_data, &sim_data);
if (fifo_data.valid) {
int b = fifo_data.graphics_burst_size >> 4;
*burst = 0;
while (b >>= 1)
(*burst)++;
*lwm = fifo_data.graphics_lwm >> 3;
}
}
/****************************************************************************\
* *
* RIVA Mode State Routines *
* *
\****************************************************************************/
/*
* Calculate the Video Clock parameters for the PLL.
*/
static void CalcVClock(int clockIn,
int *clockOut, u32 * pllOut, struct nvidia_par *par)
{
unsigned lowM, highM;
unsigned DeltaNew, DeltaOld;
unsigned VClk, Freq;
unsigned M, N, P;
DeltaOld = 0xFFFFFFFF;
VClk = (unsigned)clockIn;
if (par->CrystalFreqKHz == 13500) {
lowM = 7;
highM = 13;
} else {
lowM = 8;
highM = 14;
}
for (P = 0; P <= 4; P++) {
Freq = VClk << P;
if ((Freq >= 128000) && (Freq <= 350000)) {
for (M = lowM; M <= highM; M++) {
N = ((VClk << P) * M) / par->CrystalFreqKHz;
if (N <= 255) {
Freq =
((par->CrystalFreqKHz * N) /
M) >> P;
if (Freq > VClk)
DeltaNew = Freq - VClk;
else
DeltaNew = VClk - Freq;
if (DeltaNew < DeltaOld) {
*pllOut =
(P << 16) | (N << 8) | M;
*clockOut = Freq;
DeltaOld = DeltaNew;
}
}
}
}
}
}
static void CalcVClock2Stage(int clockIn,
int *clockOut,
u32 * pllOut,
u32 * pllBOut, struct nvidia_par *par)
{
unsigned DeltaNew, DeltaOld;
unsigned VClk, Freq;
unsigned M, N, P;
DeltaOld = 0xFFFFFFFF;
*pllBOut = 0x80000401; /* fixed at x4 for now */
VClk = (unsigned)clockIn;
for (P = 0; P <= 6; P++) {
Freq = VClk << P;
if ((Freq >= 400000) && (Freq <= 1000000)) {
for (M = 1; M <= 13; M++) {
N = ((VClk << P) * M) /
(par->CrystalFreqKHz << 2);
if ((N >= 5) && (N <= 255)) {
Freq =
(((par->CrystalFreqKHz << 2) * N) /
M) >> P;
if (Freq > VClk)
DeltaNew = Freq - VClk;
else
DeltaNew = VClk - Freq;
if (DeltaNew < DeltaOld) {
*pllOut =
(P << 16) | (N << 8) | M;
*clockOut = Freq;
DeltaOld = DeltaNew;
}
}
}
}
}
}
/*
* Calculate extended mode parameters (SVGA) and save in a
* mode state structure.
*/
void NVCalcStateExt(struct nvidia_par *par,
RIVA_HW_STATE * state,
int bpp,
int width,
int hDisplaySize, int height, int dotClock, int flags)
{
int pixelDepth, VClk = 0;
/*
* Save mode parameters.
*/
state->bpp = bpp; /* this is not bitsPerPixel, it's 8,15,16,32 */
state->width = width;
state->height = height;
/*
* Extended RIVA registers.
*/
pixelDepth = (bpp + 1) / 8;
if (par->twoStagePLL)
CalcVClock2Stage(dotClock, &VClk, &state->pll, &state->pllB,
par);
else
CalcVClock(dotClock, &VClk, &state->pll, par);
switch (par->Architecture) {
case NV_ARCH_04:
nv4UpdateArbitrationSettings(VClk,
pixelDepth * 8,
&(state->arbitration0),
&(state->arbitration1), par);
state->cursor0 = 0x00;
state->cursor1 = 0xbC;
if (flags & FB_VMODE_DOUBLE)
state->cursor1 |= 2;
state->cursor2 = 0x00000000;
state->pllsel = 0x10000700;
state->config = 0x00001114;
state->general = bpp == 16 ? 0x00101100 : 0x00100100;
state->repaint1 = hDisplaySize < 1280 ? 0x04 : 0x00;
break;
case NV_ARCH_40:
if (!par->FlatPanel)
state->control = NV_RD32(par->PRAMDAC0, 0x0580) &
0xeffffeff;
/* fallthrough */
case NV_ARCH_10:
case NV_ARCH_20:
case NV_ARCH_30:
default:
if ((par->Chipset & 0xfff0) == 0x0240 ||
(par->Chipset & 0xfff0) == 0x03d0) {
state->arbitration0 = 256;
state->arbitration1 = 0x0480;
} else if (((par->Chipset & 0xffff) == 0x01A0) ||
((par->Chipset & 0xffff) == 0x01f0)) {
nForceUpdateArbitrationSettings(VClk,
pixelDepth * 8,
&(state->arbitration0),
&(state->arbitration1),
par);
} else if (par->Architecture < NV_ARCH_30) {
nv10UpdateArbitrationSettings(VClk,
pixelDepth * 8,
&(state->arbitration0),
&(state->arbitration1),
par);
} else {
nv30UpdateArbitrationSettings(par,
&(state->arbitration0),
&(state->arbitration1));
}
state->cursor0 = 0x80 | (par->CursorStart >> 17);
state->cursor1 = (par->CursorStart >> 11) << 2;
state->cursor2 = par->CursorStart >> 24;
if (flags & FB_VMODE_DOUBLE)
state->cursor1 |= 2;
state->pllsel = 0x10000700;
state->config = NV_RD32(par->PFB, 0x00000200);
state->general = bpp == 16 ? 0x00101100 : 0x00100100;
state->repaint1 = hDisplaySize < 1280 ? 0x04 : 0x00;
break;
}
if (bpp != 8) /* DirectColor */
state->general |= 0x00000030;
state->repaint0 = (((width / 8) * pixelDepth) & 0x700) >> 3;
state->pixel = (pixelDepth > 2) ? 3 : pixelDepth;
}
void NVLoadStateExt(struct nvidia_par *par, RIVA_HW_STATE * state)
{
int i, j;
NV_WR32(par->PMC, 0x0140, 0x00000000);
NV_WR32(par->PMC, 0x0200, 0xFFFF00FF);
NV_WR32(par->PMC, 0x0200, 0xFFFFFFFF);
NV_WR32(par->PTIMER, 0x0200 * 4, 0x00000008);
NV_WR32(par->PTIMER, 0x0210 * 4, 0x00000003);
NV_WR32(par->PTIMER, 0x0140 * 4, 0x00000000);
NV_WR32(par->PTIMER, 0x0100 * 4, 0xFFFFFFFF);
if (par->Architecture == NV_ARCH_04) {
if (state)
NV_WR32(par->PFB, 0x0200, state->config);
} else if ((par->Architecture < NV_ARCH_40) ||
(par->Chipset & 0xfff0) == 0x0040) {
for (i = 0; i < 8; i++) {
NV_WR32(par->PFB, 0x0240 + (i * 0x10), 0);
NV_WR32(par->PFB, 0x0244 + (i * 0x10),
par->FbMapSize - 1);
}
} else {
int regions = 12;
if (((par->Chipset & 0xfff0) == 0x0090) ||
((par->Chipset & 0xfff0) == 0x01D0) ||
((par->Chipset & 0xfff0) == 0x0290) ||
((par->Chipset & 0xfff0) == 0x0390) ||
((par->Chipset & 0xfff0) == 0x03D0))
regions = 15;
for(i = 0; i < regions; i++) {
NV_WR32(par->PFB, 0x0600 + (i * 0x10), 0);
NV_WR32(par->PFB, 0x0604 + (i * 0x10),
par->FbMapSize - 1);
}
}
if (par->Architecture >= NV_ARCH_40) {
NV_WR32(par->PRAMIN, 0x0000 * 4, 0x80000010);
NV_WR32(par->PRAMIN, 0x0001 * 4, 0x00101202);
NV_WR32(par->PRAMIN, 0x0002 * 4, 0x80000011);
NV_WR32(par->PRAMIN, 0x0003 * 4, 0x00101204);
NV_WR32(par->PRAMIN, 0x0004 * 4, 0x80000012);
NV_WR32(par->PRAMIN, 0x0005 * 4, 0x00101206);
NV_WR32(par->PRAMIN, 0x0006 * 4, 0x80000013);
NV_WR32(par->PRAMIN, 0x0007 * 4, 0x00101208);
NV_WR32(par->PRAMIN, 0x0008 * 4, 0x80000014);
NV_WR32(par->PRAMIN, 0x0009 * 4, 0x0010120A);
NV_WR32(par->PRAMIN, 0x000A * 4, 0x80000015);
NV_WR32(par->PRAMIN, 0x000B * 4, 0x0010120C);
NV_WR32(par->PRAMIN, 0x000C * 4, 0x80000016);
NV_WR32(par->PRAMIN, 0x000D * 4, 0x0010120E);
NV_WR32(par->PRAMIN, 0x000E * 4, 0x80000017);
NV_WR32(par->PRAMIN, 0x000F * 4, 0x00101210);