Free Electrons

Embedded Linux Experts

  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
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/bitops.h>
#include <asm/processor.h>
#include <asm/msr.h>
#include <asm/e820.h>
#include <asm/mtrr.h>
#include "cpu.h"

#ifdef CONFIG_X86_OOSTORE

static u32 __cpuinit power2(u32 x)
{
	u32 s=1;
	while(s<=x)
		s<<=1;
	return s>>=1;
}


/*
 *	Set up an actual MCR
 */
 
static void __cpuinit centaur_mcr_insert(int reg, u32 base, u32 size, int key)
{
	u32 lo, hi;
	
	hi = base & ~0xFFF;
	lo = ~(size-1);		/* Size is a power of 2 so this makes a mask */
	lo &= ~0xFFF;		/* Remove the ctrl value bits */
	lo |= key;		/* Attribute we wish to set */
	wrmsr(reg+MSR_IDT_MCR0, lo, hi);
	mtrr_centaur_report_mcr(reg, lo, hi);	/* Tell the mtrr driver */
}

/*
 *	Figure what we can cover with MCR's
 *
 *	Shortcut: We know you can't put 4Gig of RAM on a winchip
 */

static u32 __cpuinit ramtop(void)		/* 16388 */
{
	int i;
	u32 top = 0;
	u32 clip = 0xFFFFFFFFUL;
	
	for (i = 0; i < e820.nr_map; i++) {
		unsigned long start, end;

		if (e820.map[i].addr > 0xFFFFFFFFUL)
			continue;
		/*
		 *	Don't MCR over reserved space. Ignore the ISA hole
		 *	we frob around that catastrophy already
		 */
		 			
		if (e820.map[i].type == E820_RESERVED)
		{
			if(e820.map[i].addr >= 0x100000UL && e820.map[i].addr < clip)
				clip = e820.map[i].addr;
			continue;
		}
		start = e820.map[i].addr;
		end = e820.map[i].addr + e820.map[i].size;
		if (start >= end)
			continue;
		if (end > top)
			top = end;
	}
	/* Everything below 'top' should be RAM except for the ISA hole.
	   Because of the limited MCR's we want to map NV/ACPI into our
	   MCR range for gunk in RAM 
	   
	   Clip might cause us to MCR insufficient RAM but that is an
	   acceptable failure mode and should only bite obscure boxes with
	   a VESA hole at 15Mb
	   
	   The second case Clip sometimes kicks in is when the EBDA is marked
	   as reserved. Again we fail safe with reasonable results
	*/
	
	if(top>clip)
		top=clip;
		
	return top;
}

/*
 *	Compute a set of MCR's to give maximum coverage
 */

static int __cpuinit centaur_mcr_compute(int nr, int key)
{
	u32 mem = ramtop();
	u32 root = power2(mem);
	u32 base = root;
	u32 top = root;
	u32 floor = 0;
	int ct = 0;
	
	while(ct<nr)
	{
		u32 fspace = 0;

		/*
		 *	Find the largest block we will fill going upwards
		 */

		u32 high = power2(mem-top);	

		/*
		 *	Find the largest block we will fill going downwards
		 */

		u32 low = base/2;

		/*
		 *	Don't fill below 1Mb going downwards as there
		 *	is an ISA hole in the way.
		 */		
		 
		if(base <= 1024*1024)
			low = 0;
			
		/*
		 *	See how much space we could cover by filling below
		 *	the ISA hole
		 */
		 
		if(floor == 0)
			fspace = 512*1024;
		else if(floor ==512*1024)
			fspace = 128*1024;

		/* And forget ROM space */
		
		/*
		 *	Now install the largest coverage we get
		 */
		 
		if(fspace > high && fspace > low)
		{
			centaur_mcr_insert(ct, floor, fspace, key);
			floor += fspace;
		}
		else if(high > low)
		{
			centaur_mcr_insert(ct, top, high, key);
			top += high;
		}
		else if(low > 0)
		{
			base -= low;
			centaur_mcr_insert(ct, base, low, key);
		}
		else break;
		ct++;
	}
	/*
	 *	We loaded ct values. We now need to set the mask. The caller
	 *	must do this bit.
	 */
	 
	return ct;
}

static void __cpuinit centaur_create_optimal_mcr(void)
{
	int i;
	/*
	 *	Allocate up to 6 mcrs to mark as much of ram as possible
	 *	as write combining and weak write ordered.
	 *
	 *	To experiment with: Linux never uses stack operations for 
	 *	mmio spaces so we could globally enable stack operation wc
	 *
	 *	Load the registers with type 31 - full write combining, all
	 *	writes weakly ordered.
	 */
	int used = centaur_mcr_compute(6, 31);

	/*
	 *	Wipe unused MCRs
	 */
	 
	for(i=used;i<8;i++)
		wrmsr(MSR_IDT_MCR0+i, 0, 0);
}

static void __cpuinit winchip2_create_optimal_mcr(void)
{
	u32 lo, hi;
	int i;

	/*
	 *	Allocate up to 6 mcrs to mark as much of ram as possible
	 *	as write combining, weak store ordered.
	 *
	 *	Load the registers with type 25
	 *		8	-	weak write ordering
	 *		16	-	weak read ordering
	 *		1	-	write combining
	 */

	int used = centaur_mcr_compute(6, 25);
	
	/*
	 *	Mark the registers we are using.
	 */
	 
	rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
	for(i=0;i<used;i++)
		lo|=1<<(9+i);
	wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
	
	/*
	 *	Wipe unused MCRs
	 */
	 
	for(i=used;i<8;i++)
		wrmsr(MSR_IDT_MCR0+i, 0, 0);
}

/*
 *	Handle the MCR key on the Winchip 2.
 */

static void __cpuinit winchip2_unprotect_mcr(void)
{
	u32 lo, hi;
	u32 key;
	
	rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
	lo&=~0x1C0;	/* blank bits 8-6 */
	key = (lo>>17) & 7;
	lo |= key<<6;	/* replace with unlock key */
	wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
}

static void __cpuinit winchip2_protect_mcr(void)
{
	u32 lo, hi;
	
	rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
	lo&=~0x1C0;	/* blank bits 8-6 */
	wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
}
#endif /* CONFIG_X86_OOSTORE */

#define ACE_PRESENT	(1 << 6)
#define ACE_ENABLED	(1 << 7)
#define ACE_FCR		(1 << 28)	/* MSR_VIA_FCR */

#define RNG_PRESENT	(1 << 2)
#define RNG_ENABLED	(1 << 3)
#define RNG_ENABLE	(1 << 6)	/* MSR_VIA_RNG */

static void __cpuinit init_c3(struct cpuinfo_x86 *c)
{
	u32  lo, hi;

	/* Test for Centaur Extended Feature Flags presence */
	if (cpuid_eax(0xC0000000) >= 0xC0000001) {
		u32 tmp = cpuid_edx(0xC0000001);

		/* enable ACE unit, if present and disabled */
		if ((tmp & (ACE_PRESENT | ACE_ENABLED)) == ACE_PRESENT) {
			rdmsr (MSR_VIA_FCR, lo, hi);
			lo |= ACE_FCR;		/* enable ACE unit */
			wrmsr (MSR_VIA_FCR, lo, hi);
			printk(KERN_INFO "CPU: Enabled ACE h/w crypto\n");
		}

		/* enable RNG unit, if present and disabled */
		if ((tmp & (RNG_PRESENT | RNG_ENABLED)) == RNG_PRESENT) {
			rdmsr (MSR_VIA_RNG, lo, hi);
			lo |= RNG_ENABLE;	/* enable RNG unit */
			wrmsr (MSR_VIA_RNG, lo, hi);
			printk(KERN_INFO "CPU: Enabled h/w RNG\n");
		}

		/* store Centaur Extended Feature Flags as
		 * word 5 of the CPU capability bit array
		 */
		c->x86_capability[5] = cpuid_edx(0xC0000001);
	}

	/* Cyrix III family needs CX8 & PGE explicity enabled. */
	if (c->x86_model >=6 && c->x86_model <= 9) {
		rdmsr (MSR_VIA_FCR, lo, hi);
		lo |= (1<<1 | 1<<7);
		wrmsr (MSR_VIA_FCR, lo, hi);
		set_bit(X86_FEATURE_CX8, c->x86_capability);
	}

	/* Before Nehemiah, the C3's had 3dNOW! */
	if (c->x86_model >=6 && c->x86_model <9)
		set_bit(X86_FEATURE_3DNOW, c->x86_capability);

	get_model_name(c);
	display_cacheinfo(c);
}

static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
{
	enum {
		ECX8=1<<1,
		EIERRINT=1<<2,
		DPM=1<<3,
		DMCE=1<<4,
		DSTPCLK=1<<5,
		ELINEAR=1<<6,
		DSMC=1<<7,
		DTLOCK=1<<8,
		EDCTLB=1<<8,
		EMMX=1<<9,
		DPDC=1<<11,
		EBRPRED=1<<12,
		DIC=1<<13,
		DDC=1<<14,
		DNA=1<<15,
		ERETSTK=1<<16,
		E2MMX=1<<19,
		EAMD3D=1<<20,
	};

	char *name;
	u32  fcr_set=0;
	u32  fcr_clr=0;
	u32  lo,hi,newlo;
	u32  aa,bb,cc,dd;

	/* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
	   3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
	clear_bit(0*32+31, c->x86_capability);

	switch (c->x86) {

		case 5:
			switch(c->x86_model) {
			case 4:
				name="C6";
				fcr_set=ECX8|DSMC|EDCTLB|EMMX|ERETSTK;
				fcr_clr=DPDC;
				printk(KERN_NOTICE "Disabling bugged TSC.\n");
				clear_bit(X86_FEATURE_TSC, c->x86_capability);
#ifdef CONFIG_X86_OOSTORE
				centaur_create_optimal_mcr();
				/* Enable
					write combining on non-stack, non-string
					write combining on string, all types
					weak write ordering 
					
				   The C6 original lacks weak read order 
				   
				   Note 0x120 is write only on Winchip 1 */
				   
				wrmsr(MSR_IDT_MCR_CTRL, 0x01F0001F, 0);
#endif				
				break;
			case 8:
				switch(c->x86_mask) {
				default:
					name="2";
					break;
				case 7 ... 9:
					name="2A";
					break;
				case 10 ... 15:
					name="2B";
					break;
				}
				fcr_set=ECX8|DSMC|DTLOCK|EMMX|EBRPRED|ERETSTK|E2MMX|EAMD3D;
				fcr_clr=DPDC;
#ifdef CONFIG_X86_OOSTORE
				winchip2_unprotect_mcr();
				winchip2_create_optimal_mcr();
				rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
				/* Enable
					write combining on non-stack, non-string
					write combining on string, all types
					weak write ordering 
				*/
				lo|=31;				
				wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
				winchip2_protect_mcr();
#endif
				break;
			case 9:
				name="3";
				fcr_set=ECX8|DSMC|DTLOCK|EMMX|EBRPRED|ERETSTK|E2MMX|EAMD3D;
				fcr_clr=DPDC;
#ifdef CONFIG_X86_OOSTORE
				winchip2_unprotect_mcr();
				winchip2_create_optimal_mcr();
				rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
				/* Enable
					write combining on non-stack, non-string
					write combining on string, all types
					weak write ordering 
				*/
				lo|=31;				
				wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
				winchip2_protect_mcr();
#endif
				break;
			default:
				name="??";
			}

			rdmsr(MSR_IDT_FCR1, lo, hi);
			newlo=(lo|fcr_set) & (~fcr_clr);

			if (newlo!=lo) {
				printk(KERN_INFO "Centaur FCR was 0x%X now 0x%X\n", lo, newlo );
				wrmsr(MSR_IDT_FCR1, newlo, hi );
			} else {
				printk(KERN_INFO "Centaur FCR is 0x%X\n",lo);
			}
			/* Emulate MTRRs using Centaur's MCR. */
			set_bit(X86_FEATURE_CENTAUR_MCR, c->x86_capability);
			/* Report CX8 */
			set_bit(X86_FEATURE_CX8, c->x86_capability);
			/* Set 3DNow! on Winchip 2 and above. */
			if (c->x86_model >=8)
				set_bit(X86_FEATURE_3DNOW, c->x86_capability);
			/* See if we can find out some more. */
			if ( cpuid_eax(0x80000000) >= 0x80000005 ) {
				/* Yes, we can. */
				cpuid(0x80000005,&aa,&bb,&cc,&dd);
				/* Add L1 data and code cache sizes. */
				c->x86_cache_size = (cc>>24)+(dd>>24);
			}
			sprintf( c->x86_model_id, "WinChip %s", name );
			break;

		case 6:
			init_c3(c);
			break;
	}
}

static unsigned int __cpuinit centaur_size_cache(struct cpuinfo_x86 * c, unsigned int size)
{
	/* VIA C3 CPUs (670-68F) need further shifting. */
	if ((c->x86 == 6) && ((c->x86_model == 7) || (c->x86_model == 8)))
		size >>= 8;

	/* VIA also screwed up Nehemiah stepping 1, and made
	   it return '65KB' instead of '64KB'
	   - Note, it seems this may only be in engineering samples. */
	if ((c->x86==6) && (c->x86_model==9) && (c->x86_mask==1) && (size==65))
		size -=1;

	return size;
}

static struct cpu_dev centaur_cpu_dev __cpuinitdata = {
	.c_vendor	= "Centaur",
	.c_ident	= { "CentaurHauls" },
	.c_init		= init_centaur,
	.c_size_cache	= centaur_size_cache,
};

int __init centaur_init_cpu(void)
{
	cpu_devs[X86_VENDOR_CENTAUR] = &centaur_cpu_dev;
	return 0;
}