blob: f4773f4aae35be4964827563d6f2843199f639b3 [file] [log] [blame]
Paul Gortmaker69c60c82011-05-26 12:22:53 -04001#include <linux/export.h>
Linus Torvalds1da177e2005-04-16 15:20:36 -07002#include <linux/init.h>
3#include <linux/bitops.h>
Stephen Rothwell5cdd1742011-08-10 11:49:56 +10004#include <linux/elf.h>
Linus Torvalds1da177e2005-04-16 15:20:36 -07005#include <linux/mm.h>
Yinghai Lu8d71a2e2008-09-07 17:58:53 -07006
Alan Cox8bdbd962009-07-04 00:35:45 +01007#include <linux/io.h>
Linus Torvalds1da177e2005-04-16 15:20:36 -07008#include <asm/processor.h>
Andi Kleend3f7eae2007-08-10 22:31:07 +02009#include <asm/apic.h>
Yinghai Lu1f442d72009-03-07 23:46:26 -080010#include <asm/cpu.h>
Andreas Herrmann42937e82009-06-08 15:55:09 +020011#include <asm/pci-direct.h>
Linus Torvalds1da177e2005-04-16 15:20:36 -070012
Yinghai Lu8d71a2e2008-09-07 17:58:53 -070013#ifdef CONFIG_X86_64
14# include <asm/numa_64.h>
15# include <asm/mmconfig.h>
16# include <asm/cacheflush.h>
17#endif
18
Linus Torvalds1da177e2005-04-16 15:20:36 -070019#include "cpu.h"
20
Yinghai Lu6c62aa42008-09-07 17:58:54 -070021#ifdef CONFIG_X86_32
Linus Torvalds1da177e2005-04-16 15:20:36 -070022/*
23 * B step AMD K6 before B 9730xxxx have hardware bugs that can cause
24 * misexecution of code under Linux. Owners of such processors should
25 * contact AMD for precise details and a CPU swap.
26 *
27 * See http://www.multimania.com/poulot/k6bug.html
28 * http://www.amd.com/K6/k6docs/revgd.html
29 *
30 * The following test is erm.. interesting. AMD neglected to up
31 * the chip setting when fixing the bug but they also tweaked some
32 * performance at the same time..
33 */
Paolo Ciarrocchifb87a292008-02-22 23:10:33 +010034
Linus Torvalds1da177e2005-04-16 15:20:36 -070035extern void vide(void);
36__asm__(".align 4\nvide: ret");
37
Yinghai Lu11fdd252008-09-07 17:58:50 -070038static void __cpuinit init_amd_k5(struct cpuinfo_x86 *c)
39{
40/*
41 * General Systems BIOSen alias the cpu frequency registers
42 * of the Elan at 0x000df000. Unfortuantly, one of the Linux
43 * drivers subsequently pokes it, and changes the CPU speed.
44 * Workaround : Remove the unneeded alias.
45 */
46#define CBAR (0xfffc) /* Configuration Base Address (32-bit) */
47#define CBAR_ENB (0x80000000)
48#define CBAR_KEY (0X000000CB)
49 if (c->x86_model == 9 || c->x86_model == 10) {
Alan Cox8bdbd962009-07-04 00:35:45 +010050 if (inl(CBAR) & CBAR_ENB)
51 outl(0 | CBAR_KEY, CBAR);
Yinghai Lu11fdd252008-09-07 17:58:50 -070052 }
53}
54
55
56static void __cpuinit init_amd_k6(struct cpuinfo_x86 *c)
57{
58 u32 l, h;
59 int mbytes = num_physpages >> (20-PAGE_SHIFT);
60
61 if (c->x86_model < 6) {
62 /* Based on AMD doc 20734R - June 2000 */
63 if (c->x86_model == 0) {
64 clear_cpu_cap(c, X86_FEATURE_APIC);
65 set_cpu_cap(c, X86_FEATURE_PGE);
66 }
67 return;
68 }
69
70 if (c->x86_model == 6 && c->x86_mask == 1) {
71 const int K6_BUG_LOOP = 1000000;
72 int n;
73 void (*f_vide)(void);
74 unsigned long d, d2;
75
76 printk(KERN_INFO "AMD K6 stepping B detected - ");
77
78 /*
79 * It looks like AMD fixed the 2.6.2 bug and improved indirect
80 * calls at the same time.
81 */
82
83 n = K6_BUG_LOOP;
84 f_vide = vide;
85 rdtscl(d);
86 while (n--)
87 f_vide();
88 rdtscl(d2);
89 d = d2-d;
90
91 if (d > 20*K6_BUG_LOOP)
Alan Cox8bdbd962009-07-04 00:35:45 +010092 printk(KERN_CONT
93 "system stability may be impaired when more than 32 MB are used.\n");
Yinghai Lu11fdd252008-09-07 17:58:50 -070094 else
Alan Cox8bdbd962009-07-04 00:35:45 +010095 printk(KERN_CONT "probably OK (after B9730xxxx).\n");
Yinghai Lu11fdd252008-09-07 17:58:50 -070096 printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n");
97 }
98
99 /* K6 with old style WHCR */
100 if (c->x86_model < 8 ||
101 (c->x86_model == 8 && c->x86_mask < 8)) {
102 /* We can only write allocate on the low 508Mb */
103 if (mbytes > 508)
104 mbytes = 508;
105
106 rdmsr(MSR_K6_WHCR, l, h);
107 if ((l&0x0000FFFF) == 0) {
108 unsigned long flags;
109 l = (1<<0)|((mbytes/4)<<1);
110 local_irq_save(flags);
111 wbinvd();
112 wrmsr(MSR_K6_WHCR, l, h);
113 local_irq_restore(flags);
114 printk(KERN_INFO "Enabling old style K6 write allocation for %d Mb\n",
115 mbytes);
116 }
117 return;
118 }
119
120 if ((c->x86_model == 8 && c->x86_mask > 7) ||
121 c->x86_model == 9 || c->x86_model == 13) {
122 /* The more serious chips .. */
123
124 if (mbytes > 4092)
125 mbytes = 4092;
126
127 rdmsr(MSR_K6_WHCR, l, h);
128 if ((l&0xFFFF0000) == 0) {
129 unsigned long flags;
130 l = ((mbytes>>2)<<22)|(1<<16);
131 local_irq_save(flags);
132 wbinvd();
133 wrmsr(MSR_K6_WHCR, l, h);
134 local_irq_restore(flags);
135 printk(KERN_INFO "Enabling new style K6 write allocation for %d Mb\n",
136 mbytes);
137 }
138
139 return;
140 }
141
142 if (c->x86_model == 10) {
143 /* AMD Geode LX is model 10 */
144 /* placeholder for any needed mods */
145 return;
146 }
147}
148
Yinghai Lu1f442d72009-03-07 23:46:26 -0800149static void __cpuinit amd_k7_smp_check(struct cpuinfo_x86 *c)
150{
Yinghai Lu1f442d72009-03-07 23:46:26 -0800151 /* calling is from identify_secondary_cpu() ? */
Robert Richterf6e9456c2010-07-21 19:03:58 +0200152 if (!c->cpu_index)
Yinghai Lu1f442d72009-03-07 23:46:26 -0800153 return;
154
155 /*
156 * Certain Athlons might work (for various values of 'work') in SMP
157 * but they are not certified as MP capable.
158 */
159 /* Athlon 660/661 is valid. */
160 if ((c->x86_model == 6) && ((c->x86_mask == 0) ||
161 (c->x86_mask == 1)))
162 goto valid_k7;
163
164 /* Duron 670 is valid */
165 if ((c->x86_model == 7) && (c->x86_mask == 0))
166 goto valid_k7;
167
168 /*
169 * Athlon 662, Duron 671, and Athlon >model 7 have capability
170 * bit. It's worth noting that the A5 stepping (662) of some
171 * Athlon XP's have the MP bit set.
172 * See http://www.heise.de/newsticker/data/jow-18.10.01-000 for
173 * more.
174 */
175 if (((c->x86_model == 6) && (c->x86_mask >= 2)) ||
176 ((c->x86_model == 7) && (c->x86_mask >= 1)) ||
177 (c->x86_model > 7))
178 if (cpu_has_mp)
179 goto valid_k7;
180
181 /* If we get here, not a certified SMP capable AMD system. */
182
183 /*
184 * Don't taint if we are running SMP kernel on a single non-MP
185 * approved Athlon
186 */
187 WARN_ONCE(1, "WARNING: This combination of AMD"
Michael Tokarev7da8b6d2009-07-22 17:50:23 +0400188 " processors is not suitable for SMP.\n");
Yinghai Lu1f442d72009-03-07 23:46:26 -0800189 if (!test_taint(TAINT_UNSAFE_SMP))
190 add_taint(TAINT_UNSAFE_SMP);
191
192valid_k7:
193 ;
Yinghai Lu1f442d72009-03-07 23:46:26 -0800194}
195
Yinghai Lu11fdd252008-09-07 17:58:50 -0700196static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c)
197{
198 u32 l, h;
199
200 /*
201 * Bit 15 of Athlon specific MSR 15, needs to be 0
202 * to enable SSE on Palomino/Morgan/Barton CPU's.
203 * If the BIOS didn't enable it already, enable it here.
204 */
205 if (c->x86_model >= 6 && c->x86_model <= 10) {
206 if (!cpu_has(c, X86_FEATURE_XMM)) {
207 printk(KERN_INFO "Enabling disabled K7/SSE Support.\n");
208 rdmsr(MSR_K7_HWCR, l, h);
209 l &= ~0x00008000;
210 wrmsr(MSR_K7_HWCR, l, h);
211 set_cpu_cap(c, X86_FEATURE_XMM);
212 }
213 }
214
215 /*
216 * It's been determined by AMD that Athlons since model 8 stepping 1
217 * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx
218 * As per AMD technical note 27212 0.2
219 */
220 if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) {
221 rdmsr(MSR_K7_CLK_CTL, l, h);
222 if ((l & 0xfff00000) != 0x20000000) {
Alan Cox8bdbd962009-07-04 00:35:45 +0100223 printk(KERN_INFO
224 "CPU: CLK_CTL MSR was %x. Reprogramming to %x\n",
225 l, ((l & 0x000fffff)|0x20000000));
Yinghai Lu11fdd252008-09-07 17:58:50 -0700226 wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h);
227 }
228 }
229
230 set_cpu_cap(c, X86_FEATURE_K7);
Yinghai Lu1f442d72009-03-07 23:46:26 -0800231
232 amd_k7_smp_check(c);
Yinghai Lu11fdd252008-09-07 17:58:50 -0700233}
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700234#endif
235
Tejun Heo645a7912011-01-23 14:37:40 +0100236#ifdef CONFIG_NUMA
Tejun Heobbc9e2f2011-01-23 14:37:39 +0100237/*
238 * To workaround broken NUMA config. Read the comment in
239 * srat_detect_node().
240 */
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700241static int __cpuinit nearby_node(int apicid)
242{
243 int i, node;
244
245 for (i = apicid - 1; i >= 0; i--) {
Tejun Heobbc9e2f2011-01-23 14:37:39 +0100246 node = __apicid_to_node[i];
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700247 if (node != NUMA_NO_NODE && node_online(node))
248 return node;
249 }
250 for (i = apicid + 1; i < MAX_LOCAL_APIC; i++) {
Tejun Heobbc9e2f2011-01-23 14:37:39 +0100251 node = __apicid_to_node[i];
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700252 if (node != NUMA_NO_NODE && node_online(node))
253 return node;
254 }
255 return first_node(node_online_map); /* Shouldn't happen */
256}
257#endif
Yinghai Lu11fdd252008-09-07 17:58:50 -0700258
259/*
Andreas Herrmann23588c32010-09-30 14:36:28 +0200260 * Fixup core topology information for
261 * (1) AMD multi-node processors
262 * Assumption: Number of cores in each internal node is the same.
Andreas Herrmann6057b4d2010-09-30 14:38:57 +0200263 * (2) AMD processors supporting compute units
Andreas Herrmann4a376ec2009-09-03 09:40:21 +0200264 */
265#ifdef CONFIG_X86_HT
Andreas Herrmann23588c32010-09-30 14:36:28 +0200266static void __cpuinit amd_get_topology(struct cpuinfo_x86 *c)
Andreas Herrmann4a376ec2009-09-03 09:40:21 +0200267{
Borislav Petkov9e815092011-02-14 18:14:51 +0100268 u32 nodes, cores_per_cu = 1;
Andreas Herrmann23588c32010-09-30 14:36:28 +0200269 u8 node_id;
Andreas Herrmann4a376ec2009-09-03 09:40:21 +0200270 int cpu = smp_processor_id();
271
Andreas Herrmann23588c32010-09-30 14:36:28 +0200272 /* get information required for multi-node processors */
273 if (cpu_has(c, X86_FEATURE_TOPOEXT)) {
Andreas Herrmann6057b4d2010-09-30 14:38:57 +0200274 u32 eax, ebx, ecx, edx;
275
276 cpuid(0x8000001e, &eax, &ebx, &ecx, &edx);
277 nodes = ((ecx >> 8) & 7) + 1;
278 node_id = ecx & 7;
279
280 /* get compute unit information */
281 smp_num_siblings = ((ebx >> 8) & 3) + 1;
282 c->compute_unit_id = ebx & 0xff;
Borislav Petkov9e815092011-02-14 18:14:51 +0100283 cores_per_cu += ((ebx >> 8) & 3);
Andreas Herrmann23588c32010-09-30 14:36:28 +0200284 } else if (cpu_has(c, X86_FEATURE_NODEID_MSR)) {
Andreas Herrmann6057b4d2010-09-30 14:38:57 +0200285 u64 value;
286
Andreas Herrmann23588c32010-09-30 14:36:28 +0200287 rdmsrl(MSR_FAM10H_NODE_ID, value);
288 nodes = ((value >> 3) & 7) + 1;
289 node_id = value & 7;
290 } else
Andreas Herrmann9d260eb2009-12-16 15:43:55 +0100291 return;
292
Andreas Herrmann23588c32010-09-30 14:36:28 +0200293 /* fixup multi-node processor information */
294 if (nodes > 1) {
Andreas Herrmann6057b4d2010-09-30 14:38:57 +0200295 u32 cores_per_node;
Andreas Herrmannd5185732011-01-24 16:05:40 +0100296 u32 cus_per_node;
Andreas Herrmann6057b4d2010-09-30 14:38:57 +0200297
Andreas Herrmann23588c32010-09-30 14:36:28 +0200298 set_cpu_cap(c, X86_FEATURE_AMD_DCM);
299 cores_per_node = c->x86_max_cores / nodes;
Andreas Herrmannd5185732011-01-24 16:05:40 +0100300 cus_per_node = cores_per_node / cores_per_cu;
Andreas Herrmann4a376ec2009-09-03 09:40:21 +0200301
Andreas Herrmann23588c32010-09-30 14:36:28 +0200302 /* store NodeID, use llc_shared_map to store sibling info */
303 per_cpu(cpu_llc_id, cpu) = node_id;
Andreas Herrmann9d260eb2009-12-16 15:43:55 +0100304
Borislav Petkov9e815092011-02-14 18:14:51 +0100305 /* core id has to be in the [0 .. cores_per_node - 1] range */
Andreas Herrmannd5185732011-01-24 16:05:40 +0100306 c->cpu_core_id %= cores_per_node;
307 c->compute_unit_id %= cus_per_node;
Andreas Herrmann23588c32010-09-30 14:36:28 +0200308 }
Andreas Herrmann4a376ec2009-09-03 09:40:21 +0200309}
310#endif
311
312/*
Yinghai Lu11fdd252008-09-07 17:58:50 -0700313 * On a AMD dual core setup the lower bits of the APIC id distingush the cores.
314 * Assumes number of cores is a power of two.
315 */
316static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
317{
318#ifdef CONFIG_X86_HT
319 unsigned bits;
Andreas Herrmann99bd0c02009-06-19 10:59:09 +0200320 int cpu = smp_processor_id();
Yinghai Lu11fdd252008-09-07 17:58:50 -0700321
322 bits = c->x86_coreid_bits;
Yinghai Lu11fdd252008-09-07 17:58:50 -0700323 /* Low order bits define the core id (index of core in socket) */
324 c->cpu_core_id = c->initial_apicid & ((1 << bits)-1);
325 /* Convert the initial APIC ID into the socket ID */
326 c->phys_proc_id = c->initial_apicid >> bits;
Andreas Herrmann99bd0c02009-06-19 10:59:09 +0200327 /* use socket ID also for last level cache */
328 per_cpu(cpu_llc_id, cpu) = c->phys_proc_id;
Andreas Herrmann23588c32010-09-30 14:36:28 +0200329 amd_get_topology(c);
Yinghai Lu11fdd252008-09-07 17:58:50 -0700330#endif
331}
332
Andreas Herrmann6a812692009-09-16 11:33:40 +0200333int amd_get_nb_id(int cpu)
334{
335 int id = 0;
336#ifdef CONFIG_SMP
337 id = per_cpu(cpu_llc_id, cpu);
338#endif
339 return id;
340}
341EXPORT_SYMBOL_GPL(amd_get_nb_id);
342
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700343static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c)
344{
Tejun Heo645a7912011-01-23 14:37:40 +0100345#ifdef CONFIG_NUMA
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700346 int cpu = smp_processor_id();
347 int node;
Yinghai Lu0d96b9f2009-08-29 13:17:14 -0700348 unsigned apicid = c->apicid;
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700349
Tejun Heobbc9e2f2011-01-23 14:37:39 +0100350 node = numa_cpu_node(cpu);
351 if (node == NUMA_NO_NODE)
352 node = per_cpu(cpu_llc_id, cpu);
Andreas Herrmann4a376ec2009-09-03 09:40:21 +0200353
Daniel J Blueman64be4c12011-12-05 16:20:37 +0800354 /*
355 * If core numbers are inconsistent, it's likely a multi-fabric platform,
356 * so invoke platform-specific handler
357 */
358 if (c->phys_proc_id != node)
359 x86_cpuinit.fixup_cpu_id(c, node);
360
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700361 if (!node_online(node)) {
Tejun Heobbc9e2f2011-01-23 14:37:39 +0100362 /*
363 * Two possibilities here:
364 *
365 * - The CPU is missing memory and no node was created. In
366 * that case try picking one from a nearby CPU.
367 *
368 * - The APIC IDs differ from the HyperTransport node IDs
369 * which the K8 northbridge parsing fills in. Assume
370 * they are all increased by a constant offset, but in
371 * the same order as the HT nodeids. If that doesn't
372 * result in a usable node fall back to the path for the
373 * previous case.
374 *
375 * This workaround operates directly on the mapping between
376 * APIC ID and NUMA node, assuming certain relationship
377 * between APIC ID, HT node ID and NUMA topology. As going
378 * through CPU mapping may alter the outcome, directly
379 * access __apicid_to_node[].
380 */
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700381 int ht_nodeid = c->initial_apicid;
382
383 if (ht_nodeid >= 0 &&
Tejun Heobbc9e2f2011-01-23 14:37:39 +0100384 __apicid_to_node[ht_nodeid] != NUMA_NO_NODE)
385 node = __apicid_to_node[ht_nodeid];
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700386 /* Pick a nearby node */
387 if (!node_online(node))
388 node = nearby_node(apicid);
389 }
390 numa_set_node(cpu, node);
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700391#endif
392}
393
Yinghai Lu11fdd252008-09-07 17:58:50 -0700394static void __cpuinit early_init_amd_mc(struct cpuinfo_x86 *c)
395{
396#ifdef CONFIG_X86_HT
397 unsigned bits, ecx;
398
399 /* Multi core CPU? */
400 if (c->extended_cpuid_level < 0x80000008)
401 return;
402
403 ecx = cpuid_ecx(0x80000008);
404
405 c->x86_max_cores = (ecx & 0xff) + 1;
406
407 /* CPU telling us the core id bits shift? */
408 bits = (ecx >> 12) & 0xF;
409
410 /* Otherwise recompute */
411 if (bits == 0) {
412 while ((1 << bits) < c->x86_max_cores)
413 bits++;
414 }
415
416 c->x86_coreid_bits = bits;
417#endif
418}
419
Borislav Petkov8fa8b032011-08-05 20:04:09 +0200420static void __cpuinit bsp_init_amd(struct cpuinfo_x86 *c)
421{
422 if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) {
423
424 if (c->x86 > 0x10 ||
425 (c->x86 == 0x10 && c->x86_model >= 0x2)) {
426 u64 val;
427
428 rdmsrl(MSR_K7_HWCR, val);
429 if (!(val & BIT(24)))
430 printk(KERN_WARNING FW_BUG "TSC doesn't count "
431 "with P0 frequency!\n");
432 }
433 }
434
435 if (c->x86 == 0x15) {
436 unsigned long upperbit;
437 u32 cpuid, assoc;
438
439 cpuid = cpuid_edx(0x80000005);
440 assoc = cpuid >> 16 & 0xff;
441 upperbit = ((cpuid >> 24) << 10) / assoc;
442
443 va_align.mask = (upperbit - 1) & PAGE_MASK;
444 va_align.flags = ALIGN_VA_32 | ALIGN_VA_64;
445 }
446}
447
Thomas Petazzoni03ae5762008-02-15 12:00:23 +0100448static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
Andi Kleen2b16a232008-01-30 13:32:40 +0100449{
Yinghai Lu11fdd252008-09-07 17:58:50 -0700450 early_init_amd_mc(c);
451
Venki Pallipadi40fb1712008-11-17 16:11:37 -0800452 /*
453 * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate
454 * with P/T states and does not stop in deep C-states
455 */
456 if (c->x86_power & (1 << 8)) {
Yinghai Lue3224232008-09-06 01:52:28 -0700457 set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
Venki Pallipadi40fb1712008-11-17 16:11:37 -0800458 set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC);
459 }
Yinghai Lu5fef55f2008-09-04 21:09:43 +0200460
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700461#ifdef CONFIG_X86_64
462 set_cpu_cap(c, X86_FEATURE_SYSCALL32);
463#else
Yinghai Lu5fef55f2008-09-04 21:09:43 +0200464 /* Set MTRR capability flag if appropriate */
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700465 if (c->x86 == 5)
466 if (c->x86_model == 13 || c->x86_model == 9 ||
467 (c->x86_model == 8 && c->x86_mask >= 8))
468 set_cpu_cap(c, X86_FEATURE_K6_MTRR);
469#endif
Andreas Herrmann42937e82009-06-08 15:55:09 +0200470#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_PCI)
471 /* check CPU config space for extended APIC ID */
Jeremy Fitzhardinge2cb07862009-07-22 09:59:35 -0700472 if (cpu_has_apic && c->x86 >= 0xf) {
Andreas Herrmann42937e82009-06-08 15:55:09 +0200473 unsigned int val;
474 val = read_pci_config(0, 24, 0, 0x68);
475 if ((val & ((1 << 17) | (1 << 18))) == ((1 << 17) | (1 << 18)))
476 set_cpu_cap(c, X86_FEATURE_EXTD_APICID);
477 }
478#endif
Andi Kleen2b16a232008-01-30 13:32:40 +0100479}
480
Magnus Dammb4af3f72006-09-26 10:52:36 +0200481static void __cpuinit init_amd(struct cpuinfo_x86 *c)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700482{
Linus Torvalds8e8da022011-12-04 11:57:09 -0800483 u32 dummy;
484
Andi Kleen7d318d72005-09-29 22:05:55 +0200485#ifdef CONFIG_SMP
Andi Kleen3c92c2b2005-10-11 01:28:33 +0200486 unsigned long long value;
Andi Kleen7d318d72005-09-29 22:05:55 +0200487
Paolo Ciarrocchifb87a292008-02-22 23:10:33 +0100488 /*
489 * Disable TLB flush filter by setting HWCR.FFDIS on K8
Andi Kleen7d318d72005-09-29 22:05:55 +0200490 * bit 6 of msr C001_0015
491 *
492 * Errata 63 for SH-B3 steppings
493 * Errata 122 for all steppings (F+ have it disabled by default)
494 */
Yinghai Lu11fdd252008-09-07 17:58:50 -0700495 if (c->x86 == 0xf) {
Andi Kleen7d318d72005-09-29 22:05:55 +0200496 rdmsrl(MSR_K7_HWCR, value);
497 value |= 1 << 6;
498 wrmsrl(MSR_K7_HWCR, value);
499 }
500#endif
501
Andi Kleen2b16a232008-01-30 13:32:40 +0100502 early_init_amd(c);
503
Linus Torvalds1da177e2005-04-16 15:20:36 -0700504 /*
Paolo Ciarrocchifb87a292008-02-22 23:10:33 +0100505 * Bit 31 in normal CPUID used for nonstandard 3DNow ID;
Ingo Molnar16282a82008-02-26 08:49:57 +0100506 * 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway
Paolo Ciarrocchifb87a292008-02-22 23:10:33 +0100507 */
Ingo Molnar16282a82008-02-26 08:49:57 +0100508 clear_cpu_cap(c, 0*32+31);
Paolo Ciarrocchifb87a292008-02-22 23:10:33 +0100509
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700510#ifdef CONFIG_X86_64
511 /* On C+ stepping K8 rep microcode works well for copy/memset */
512 if (c->x86 == 0xf) {
513 u32 level;
514
515 level = cpuid_eax(1);
Alan Cox8bdbd962009-07-04 00:35:45 +0100516 if ((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700517 set_cpu_cap(c, X86_FEATURE_REP_GOOD);
Kevin Winchesterfbd8b182009-08-10 19:56:45 -0300518
519 /*
520 * Some BIOSes incorrectly force this feature, but only K8
521 * revision D (model = 0x14) and later actually support it.
Borislav Petkov6b0f43d2009-08-31 09:50:11 +0200522 * (AMD Erratum #110, docId: 25759).
Kevin Winchesterfbd8b182009-08-10 19:56:45 -0300523 */
Borislav Petkov6b0f43d2009-08-31 09:50:11 +0200524 if (c->x86_model < 0x14 && cpu_has(c, X86_FEATURE_LAHF_LM)) {
525 u64 val;
526
Kevin Winchesterfbd8b182009-08-10 19:56:45 -0300527 clear_cpu_cap(c, X86_FEATURE_LAHF_LM);
Borislav Petkov6b0f43d2009-08-31 09:50:11 +0200528 if (!rdmsrl_amd_safe(0xc001100d, &val)) {
529 val &= ~(1ULL << 32);
530 wrmsrl_amd_safe(0xc001100d, val);
531 }
532 }
533
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700534 }
Borislav Petkov12d8a962010-06-02 20:29:21 +0200535 if (c->x86 >= 0x10)
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700536 set_cpu_cap(c, X86_FEATURE_REP_GOOD);
Yinghai Lu0d96b9f2009-08-29 13:17:14 -0700537
538 /* get apicid instead of initial apic id from cpuid */
539 c->apicid = hard_smp_processor_id();
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700540#else
541
542 /*
543 * FIXME: We should handle the K5 here. Set up the write
544 * range and also turn on MSR 83 bits 4 and 31 (write alloc,
545 * no bus pipeline)
546 */
547
Paolo Ciarrocchifb87a292008-02-22 23:10:33 +0100548 switch (c->x86) {
549 case 4:
Yinghai Lu11fdd252008-09-07 17:58:50 -0700550 init_amd_k5(c);
551 break;
Paolo Ciarrocchifb87a292008-02-22 23:10:33 +0100552 case 5:
Yinghai Lu11fdd252008-09-07 17:58:50 -0700553 init_amd_k6(c);
554 break;
Paolo Ciarrocchifb87a292008-02-22 23:10:33 +0100555 case 6: /* An Athlon/Duron */
Yinghai Lu11fdd252008-09-07 17:58:50 -0700556 init_amd_k7(c);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700557 break;
Andi Kleen67cddd92007-07-21 17:10:03 +0200558 }
Andi Kleen3556ddf2007-04-02 12:14:12 +0200559
Andi Kleenc12ceb72007-05-21 14:31:47 +0200560 /* K6s reports MCEs but don't actually have all the MSRs */
561 if (c->x86 < 6)
Ingo Molnar16282a82008-02-26 08:49:57 +0100562 clear_cpu_cap(c, X86_FEATURE_MCE);
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700563#endif
Andi Kleende421862008-01-30 13:32:37 +0100564
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700565 /* Enable workaround for FXSAVE leak */
Yinghai Lu11fdd252008-09-07 17:58:50 -0700566 if (c->x86 >= 6)
567 set_cpu_cap(c, X86_FEATURE_FXSAVE_LEAK);
568
569 if (!c->x86_model_id[0]) {
570 switch (c->x86) {
571 case 0xf:
572 /* Should distinguish Models here, but this is only
573 a fallback anyways. */
574 strcpy(c->x86_model_id, "Hammer");
575 break;
576 }
577 }
578
Borislav Petkov27c13ec2009-11-21 14:01:45 +0100579 cpu_detect_cache_sizes(c);
Yinghai Lu11fdd252008-09-07 17:58:50 -0700580
581 /* Multi core CPU? */
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700582 if (c->extended_cpuid_level >= 0x80000008) {
Yinghai Lu11fdd252008-09-07 17:58:50 -0700583 amd_detect_cmp(c);
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700584 srat_detect_node(c);
585 }
Yinghai Lu11fdd252008-09-07 17:58:50 -0700586
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700587#ifdef CONFIG_X86_32
Yinghai Lu11fdd252008-09-07 17:58:50 -0700588 detect_ht(c);
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700589#endif
Yinghai Lu11fdd252008-09-07 17:58:50 -0700590
591 if (c->extended_cpuid_level >= 0x80000006) {
Andreas Herrmannd9fadd72010-09-02 15:37:10 +0200592 if (cpuid_edx(0x80000006) & 0xf000)
Yinghai Lu11fdd252008-09-07 17:58:50 -0700593 num_cache_leaves = 4;
594 else
595 num_cache_leaves = 3;
596 }
597
Borislav Petkov12d8a962010-06-02 20:29:21 +0200598 if (c->x86 >= 0xf)
Yinghai Lu11fdd252008-09-07 17:58:50 -0700599 set_cpu_cap(c, X86_FEATURE_K8);
600
601 if (cpu_has_xmm2) {
602 /* MFENCE stops RDTSC speculation */
Ingo Molnar16282a82008-02-26 08:49:57 +0100603 set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC);
Yinghai Lu11fdd252008-09-07 17:58:50 -0700604 }
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700605
606#ifdef CONFIG_X86_64
607 if (c->x86 == 0x10) {
608 /* do this for boot cpu */
609 if (c == &boot_cpu_data)
610 check_enable_amd_mmconf_dmi();
611
612 fam10h_check_enable_mmcfg();
613 }
614
Borislav Petkov12d8a962010-06-02 20:29:21 +0200615 if (c == &boot_cpu_data && c->x86 >= 0xf) {
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700616 unsigned long long tseg;
617
618 /*
619 * Split up direct mapping around the TSEG SMM area.
620 * Don't do it for gbpages because there seems very little
621 * benefit in doing so.
622 */
623 if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg)) {
Alan Cox8bdbd962009-07-04 00:35:45 +0100624 printk(KERN_DEBUG "tseg: %010llx\n", tseg);
625 if ((tseg>>PMD_SHIFT) <
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700626 (max_low_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) ||
Alan Cox8bdbd962009-07-04 00:35:45 +0100627 ((tseg>>PMD_SHIFT) <
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700628 (max_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) &&
Alan Cox8bdbd962009-07-04 00:35:45 +0100629 (tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT))))
630 set_memory_4k((unsigned long)__va(tseg), 1);
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700631 }
632 }
633#endif
Boris Ostrovskyb87cf802011-03-15 12:13:44 -0400634
Boris Ostrovskye9cdd342011-05-26 11:19:52 -0400635 /*
636 * Family 0x12 and above processors have APIC timer
637 * running in deep C states.
638 */
639 if (c->x86 > 0x11)
Boris Ostrovskyb87cf802011-03-15 12:13:44 -0400640 set_cpu_cap(c, X86_FEATURE_ARAT);
Joerg Roedel5bbc0972011-04-15 14:47:40 +0200641
642 /*
643 * Disable GART TLB Walk Errors on Fam10h. We do this here
644 * because this is always needed when GART is enabled, even in a
645 * kernel which has no MCE support built in.
646 */
647 if (c->x86 == 0x10) {
648 /*
649 * BIOS should disable GartTlbWlk Errors themself. If
650 * it doesn't do it here as suggested by the BKDG.
651 *
652 * Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=33012
653 */
654 u64 mask;
Roedel, Joergd47cc0d2011-05-19 11:13:39 +0200655 int err;
Joerg Roedel5bbc0972011-04-15 14:47:40 +0200656
Roedel, Joergd47cc0d2011-05-19 11:13:39 +0200657 err = rdmsrl_safe(MSR_AMD64_MCx_MASK(4), &mask);
658 if (err == 0) {
659 mask |= (1 << 10);
660 checking_wrmsrl(MSR_AMD64_MCx_MASK(4), mask);
661 }
Joerg Roedel5bbc0972011-04-15 14:47:40 +0200662 }
Linus Torvalds8e8da022011-12-04 11:57:09 -0800663
664 rdmsr_safe(MSR_AMD64_PATCH_LEVEL, &c->microcode, &dummy);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700665}
666
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700667#ifdef CONFIG_X86_32
Alan Cox8bdbd962009-07-04 00:35:45 +0100668static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c,
669 unsigned int size)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700670{
671 /* AMD errata T13 (order #21922) */
672 if ((c->x86 == 6)) {
Alan Cox8bdbd962009-07-04 00:35:45 +0100673 /* Duron Rev A0 */
674 if (c->x86_model == 3 && c->x86_mask == 0)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700675 size = 64;
Alan Cox8bdbd962009-07-04 00:35:45 +0100676 /* Tbird rev A1/A2 */
Linus Torvalds1da177e2005-04-16 15:20:36 -0700677 if (c->x86_model == 4 &&
Alan Cox8bdbd962009-07-04 00:35:45 +0100678 (c->x86_mask == 0 || c->x86_mask == 1))
Linus Torvalds1da177e2005-04-16 15:20:36 -0700679 size = 256;
680 }
681 return size;
682}
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700683#endif
Linus Torvalds1da177e2005-04-16 15:20:36 -0700684
Jan Beulich02dde8b2009-03-12 12:08:49 +0000685static const struct cpu_dev __cpuinitconst amd_cpu_dev = {
Linus Torvalds1da177e2005-04-16 15:20:36 -0700686 .c_vendor = "AMD",
Paolo Ciarrocchifb87a292008-02-22 23:10:33 +0100687 .c_ident = { "AuthenticAMD" },
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700688#ifdef CONFIG_X86_32
Linus Torvalds1da177e2005-04-16 15:20:36 -0700689 .c_models = {
690 { .vendor = X86_VENDOR_AMD, .family = 4, .model_names =
691 {
692 [3] = "486 DX/2",
693 [7] = "486 DX/2-WB",
Paolo Ciarrocchifb87a292008-02-22 23:10:33 +0100694 [8] = "486 DX/4",
695 [9] = "486 DX/4-WB",
Linus Torvalds1da177e2005-04-16 15:20:36 -0700696 [14] = "Am5x86-WT",
Paolo Ciarrocchifb87a292008-02-22 23:10:33 +0100697 [15] = "Am5x86-WB"
Linus Torvalds1da177e2005-04-16 15:20:36 -0700698 }
699 },
700 },
Yinghai Lu6c62aa42008-09-07 17:58:54 -0700701 .c_size_cache = amd_size_cache,
702#endif
Thomas Petazzoni03ae5762008-02-15 12:00:23 +0100703 .c_early_init = early_init_amd,
Borislav Petkov8fa8b032011-08-05 20:04:09 +0200704 .c_bsp_init = bsp_init_amd,
Linus Torvalds1da177e2005-04-16 15:20:36 -0700705 .c_init = init_amd,
Yinghai Lu10a434f2008-09-04 21:09:45 +0200706 .c_x86_vendor = X86_VENDOR_AMD,
Linus Torvalds1da177e2005-04-16 15:20:36 -0700707};
708
Yinghai Lu10a434f2008-09-04 21:09:45 +0200709cpu_dev_register(amd_cpu_dev);
Hans Rosenfeldd78d6712010-07-28 19:09:30 +0200710
711/*
712 * AMD errata checking
713 *
714 * Errata are defined as arrays of ints using the AMD_LEGACY_ERRATUM() or
715 * AMD_OSVW_ERRATUM() macros. The latter is intended for newer errata that
716 * have an OSVW id assigned, which it takes as first argument. Both take a
717 * variable number of family-specific model-stepping ranges created by
718 * AMD_MODEL_RANGE(). Each erratum also has to be declared as extern const
719 * int[] in arch/x86/include/asm/processor.h.
720 *
721 * Example:
722 *
723 * const int amd_erratum_319[] =
724 * AMD_LEGACY_ERRATUM(AMD_MODEL_RANGE(0x10, 0x2, 0x1, 0x4, 0x2),
725 * AMD_MODEL_RANGE(0x10, 0x8, 0x0, 0x8, 0x0),
726 * AMD_MODEL_RANGE(0x10, 0x9, 0x0, 0x9, 0x0));
727 */
728
Hans Rosenfeld9d8888c2010-07-28 19:09:31 +0200729const int amd_erratum_400[] =
Borislav Petkov328935e2011-05-17 14:55:18 +0200730 AMD_OSVW_ERRATUM(1, AMD_MODEL_RANGE(0xf, 0x41, 0x2, 0xff, 0xf),
Hans Rosenfeld9d8888c2010-07-28 19:09:31 +0200731 AMD_MODEL_RANGE(0x10, 0x2, 0x1, 0xff, 0xf));
H. Peter Anvina5b91602010-07-28 16:23:20 -0700732EXPORT_SYMBOL_GPL(amd_erratum_400);
Hans Rosenfeld9d8888c2010-07-28 19:09:31 +0200733
Hans Rosenfeld1be85a62010-07-28 19:09:32 +0200734const int amd_erratum_383[] =
735 AMD_OSVW_ERRATUM(3, AMD_MODEL_RANGE(0x10, 0, 0, 0xff, 0xf));
H. Peter Anvina5b91602010-07-28 16:23:20 -0700736EXPORT_SYMBOL_GPL(amd_erratum_383);
Hans Rosenfeld9d8888c2010-07-28 19:09:31 +0200737
Hans Rosenfeldd78d6712010-07-28 19:09:30 +0200738bool cpu_has_amd_erratum(const int *erratum)
739{
Tejun Heo7b543a52010-12-18 16:30:05 +0100740 struct cpuinfo_x86 *cpu = __this_cpu_ptr(&cpu_info);
Hans Rosenfeldd78d6712010-07-28 19:09:30 +0200741 int osvw_id = *erratum++;
742 u32 range;
743 u32 ms;
744
745 /*
746 * If called early enough that current_cpu_data hasn't been initialized
747 * yet, fall back to boot_cpu_data.
748 */
749 if (cpu->x86 == 0)
750 cpu = &boot_cpu_data;
751
752 if (cpu->x86_vendor != X86_VENDOR_AMD)
753 return false;
754
755 if (osvw_id >= 0 && osvw_id < 65536 &&
756 cpu_has(cpu, X86_FEATURE_OSVW)) {
757 u64 osvw_len;
758
759 rdmsrl(MSR_AMD64_OSVW_ID_LENGTH, osvw_len);
760 if (osvw_id < osvw_len) {
761 u64 osvw_bits;
762
763 rdmsrl(MSR_AMD64_OSVW_STATUS + (osvw_id >> 6),
764 osvw_bits);
765 return osvw_bits & (1ULL << (osvw_id & 0x3f));
766 }
767 }
768
769 /* OSVW unavailable or ID unknown, match family-model-stepping range */
Hans Rosenfeld07a77952010-08-18 16:19:50 +0200770 ms = (cpu->x86_model << 4) | cpu->x86_mask;
Hans Rosenfeldd78d6712010-07-28 19:09:30 +0200771 while ((range = *erratum++))
772 if ((cpu->x86 == AMD_MODEL_RANGE_FAMILY(range)) &&
773 (ms >= AMD_MODEL_RANGE_START(range)) &&
774 (ms <= AMD_MODEL_RANGE_END(range)))
775 return true;
776
777 return false;
778}
H. Peter Anvina5b91602010-07-28 16:23:20 -0700779
780EXPORT_SYMBOL_GPL(cpu_has_amd_erratum);