summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorMyles Watson <mylesgw@gmail.com>2010-04-09 04:01:55 +0000
committerMyles Watson <mylesgw@gmail.com>2010-04-09 04:01:55 +0000
commitc58f1d1df610e6fb819240919749974045d3c636 (patch)
treee3e0bf70ee9e651f63bc46078e9a52bf457cfafc
parentae60855f918dfc7f3f560528296fdd2d4b4ca791 (diff)
Indent model_fxx_init and model_10xx_init.
Signed-off-by: Myles Watson <mylesgw@gmail.com> Acked-by: Myles Watson <mylesgw@gmail.com> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5385 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
-rw-r--r--src/cpu/amd/model_10xxx/model_10xxx_init.c39
-rw-r--r--src/cpu/amd/model_fxx/model_fxx_init.c320
2 files changed, 178 insertions, 181 deletions
diff --git a/src/cpu/amd/model_10xxx/model_10xxx_init.c b/src/cpu/amd/model_10xxx/model_10xxx_init.c
index 4850ea0203..d0511a75ce 100644
--- a/src/cpu/amd/model_10xxx/model_10xxx_init.c
+++ b/src/cpu/amd/model_10xxx/model_10xxx_init.c
@@ -43,26 +43,24 @@ extern device_t get_node_pci(u32 nodeid, u32 fn);
msr_t rdmsr_amd(u32 index)
{
- msr_t result;
- __asm__ __volatile__ (
- "rdmsr"
- : "=a" (result.lo), "=d" (result.hi)
- : "c" (index), "D" (0x9c5a203a)
- );
- return result;
+ msr_t result;
+ __asm__ __volatile__(
+ "rdmsr"
+ :"=a"(result.lo), "=d"(result.hi)
+ :"c"(index), "D"(0x9c5a203a)
+ );
+ return result;
}
-
void wrmsr_amd(u32 index, msr_t msr)
{
- __asm__ __volatile__ (
+ __asm__ __volatile__(
"wrmsr"
- : /* No outputs */
- : "c" (index), "a" (msr.lo), "d" (msr.hi), "D" (0x9c5a203a)
- );
+ : /* No outputs */
+ :"c"(index), "a"(msr.lo), "d"(msr.hi), "D"(0x9c5a203a)
+ );
}
-
static void model_10xxx_init(device_t dev)
{
u8 i;
@@ -72,7 +70,7 @@ static void model_10xxx_init(device_t dev)
u32 siblings;
#endif
- id = get_node_core_id(read_nb_cfg_54()); /* nb_cfg_54 can not be set */
+ id = get_node_core_id(read_nb_cfg_54()); /* nb_cfg_54 can not be set */
printk(BIOS_DEBUG, "nodeid = %02d, coreid = %02d\n", id.nodeid, id.coreid);
/* Turn on caching if we haven't already */
@@ -85,11 +83,10 @@ static void model_10xxx_init(device_t dev)
/* zero the machine check error status registers */
msr.lo = 0;
msr.hi = 0;
- for(i=0; i < 5; i++) {
- wrmsr(MCI_STATUS + (i * 4),msr);
+ for (i = 0; i < 5; i++) {
+ wrmsr(MCI_STATUS + (i * 4), msr);
}
-
enable_cache();
/* Enable the local cpu apics */
@@ -107,7 +104,7 @@ static void model_10xxx_init(device_t dev)
wrmsr_amd(CPU_ID_FEATURES_MSR, msr);
msr = rdmsr_amd(CPU_ID_EXT_FEATURES_MSR);
- msr.hi |= 1 << (33-32);
+ msr.hi |= 1 << (33 - 32);
wrmsr_amd(CPU_ID_EXT_FEATURES_MSR, msr);
}
printk(BIOS_DEBUG, "siblings = %02d, ", siblings);
@@ -115,7 +112,7 @@ static void model_10xxx_init(device_t dev)
/* DisableCf8ExtCfg */
msr = rdmsr(NB_CFG_MSR);
- msr.hi &= ~(1 << (46-32));
+ msr.hi &= ~(1 << (46 - 32));
wrmsr(NB_CFG_MSR, msr);
/* Write protect SMM space with SMMLOCK. */
@@ -128,6 +125,7 @@ static void model_10xxx_init(device_t dev)
static struct device_operations cpu_dev_ops = {
.init = model_10xxx_init,
};
+
static struct cpu_device_id cpu_table[] = {
//AMD_GH_SUPPORT
{ X86_VENDOR_AMD, 0x100f00 }, /* SH-F0 L1 */
@@ -144,7 +142,8 @@ static struct cpu_device_id cpu_table[] = {
{ X86_VENDOR_AMD, 0x100F80 }, /* HY-D0 */
{ 0, 0 },
};
+
static const struct cpu_driver model_10xxx __cpu_driver = {
- .ops = &cpu_dev_ops,
+ .ops = &cpu_dev_ops,
.id_table = cpu_table,
};
diff --git a/src/cpu/amd/model_fxx/model_fxx_init.c b/src/cpu/amd/model_fxx/model_fxx_init.c
index ed1a168ba3..0c374859e7 100644
--- a/src/cpu/amd/model_fxx/model_fxx_init.c
+++ b/src/cpu/amd/model_fxx/model_fxx_init.c
@@ -44,66 +44,66 @@ void cpus_ready_for_init(void)
#if CONFIG_K8_REV_F_SUPPORT == 0
int is_e0_later_in_bsp(int nodeid)
{
- uint32_t val;
- uint32_t val_old;
- int e0_later;
- if(nodeid==0) { // we don't need to do that for node 0 in core0/node0
- return !is_cpu_pre_e0();
- }
- // d0 will be treated as e0 with this methods, but the d0 nb_cfg_54 always 0
- device_t dev;
- dev = dev_find_slot(0, PCI_DEVFN(0x18+nodeid,2));
- if(!dev) return 0;
- val_old = pci_read_config32(dev, 0x80);
- val = val_old;
- val |= (1<<3);
- pci_write_config32(dev, 0x80, val);
- val = pci_read_config32(dev, 0x80);
- e0_later = !!(val & (1<<3));
- if(e0_later) { // pre_e0 bit 3 always be 0 and can not be changed
- pci_write_config32(dev, 0x80, val_old); // restore it
- }
-
- return e0_later;
+ uint32_t val;
+ uint32_t val_old;
+ int e0_later;
+ if (nodeid == 0) { // we don't need to do that for node 0 in core0/node0
+ return !is_cpu_pre_e0();
+ }
+ // d0 will be treated as e0 with this methods, but the d0 nb_cfg_54 always 0
+ device_t dev;
+ dev = dev_find_slot(0, PCI_DEVFN(0x18 + nodeid, 2));
+ if (!dev)
+ return 0;
+ val_old = pci_read_config32(dev, 0x80);
+ val = val_old;
+ val |= (1 << 3);
+ pci_write_config32(dev, 0x80, val);
+ val = pci_read_config32(dev, 0x80);
+ e0_later = !!(val & (1 << 3));
+ if (e0_later) { // pre_e0 bit 3 always be 0 and can not be changed
+ pci_write_config32(dev, 0x80, val_old); // restore it
+ }
+
+ return e0_later;
}
#endif
#if CONFIG_K8_REV_F_SUPPORT == 1
int is_cpu_f0_in_bsp(int nodeid)
{
- uint32_t dword;
- device_t dev;
- dev = dev_find_slot(0, PCI_DEVFN(0x18+nodeid, 3));
- dword = pci_read_config32(dev, 0xfc);
- return (dword & 0xfff00) == 0x40f00;
+ uint32_t dword;
+ device_t dev;
+ dev = dev_find_slot(0, PCI_DEVFN(0x18 + nodeid, 3));
+ dword = pci_read_config32(dev, 0xfc);
+ return (dword & 0xfff00) == 0x40f00;
}
#endif
#define MCI_STATUS 0x401
-static inline msr_t rdmsr_amd(unsigned index)
+static inline msr_t rdmsr_amd(u32 index)
{
- msr_t result;
- __asm__ __volatile__ (
- "rdmsr"
- : "=a" (result.lo), "=d" (result.hi)
- : "c" (index), "D" (0x9c5a203a)
- );
- return result;
+ msr_t result;
+ __asm__ __volatile__(
+ "rdmsr"
+ :"=a"(result.lo), "=d"(result.hi)
+ :"c"(index), "D"(0x9c5a203a)
+ );
+ return result;
}
-static inline void wrmsr_amd(unsigned index, msr_t msr)
+static inline void wrmsr_amd(u32 index, msr_t msr)
{
- __asm__ __volatile__ (
- "wrmsr"
- : /* No outputs */
- : "c" (index), "a" (msr.lo), "d" (msr.hi), "D" (0x9c5a203a)
- );
+ __asm__ __volatile__(
+ "wrmsr"
+ : /* No outputs */
+ :"c"(index), "a"(msr.lo), "d"(msr.hi), "D"(0x9c5a203a)
+ );
}
-
#define MTRR_COUNT 8
-#define ZERO_CHUNK_KB 0x800UL /* 2M */
+#define ZERO_CHUNK_KB 0x800UL /* 2M */
#define TOLM_KB 0x400000UL
struct mtrr {
@@ -119,11 +119,11 @@ struct mtrr_state {
static void save_mtrr_state(struct mtrr_state *state)
{
int i;
- for(i = 0; i < MTRR_COUNT; i++) {
+ for (i = 0; i < MTRR_COUNT; i++) {
state->mtrrs[i].base = rdmsr(MTRRphysBase_MSR(i));
state->mtrrs[i].mask = rdmsr(MTRRphysMask_MSR(i));
}
- state->top_mem = rdmsr(TOP_MEM);
+ state->top_mem = rdmsr(TOP_MEM);
state->top_mem2 = rdmsr(TOP_MEM2);
state->def_type = rdmsr(MTRRdefType_MSR);
}
@@ -133,34 +133,33 @@ static void restore_mtrr_state(struct mtrr_state *state)
int i;
disable_cache();
- for(i = 0; i < MTRR_COUNT; i++) {
+ for (i = 0; i < MTRR_COUNT; i++) {
wrmsr(MTRRphysBase_MSR(i), state->mtrrs[i].base);
wrmsr(MTRRphysMask_MSR(i), state->mtrrs[i].mask);
}
- wrmsr(TOP_MEM, state->top_mem);
- wrmsr(TOP_MEM2, state->top_mem2);
+ wrmsr(TOP_MEM, state->top_mem);
+ wrmsr(TOP_MEM2, state->top_mem2);
wrmsr(MTRRdefType_MSR, state->def_type);
enable_cache();
}
-
#if 0
static void print_mtrr_state(struct mtrr_state *state)
{
int i;
- for(i = 0; i < MTRR_COUNT; i++) {
+ for (i = 0; i < MTRR_COUNT; i++) {
printk(BIOS_DEBUG, "var mtrr %d: %08x%08x mask: %08x%08x\n",
- i,
- state->mtrrs[i].base.hi, state->mtrrs[i].base.lo,
- state->mtrrs[i].mask.hi, state->mtrrs[i].mask.lo);
+ i,
+ state->mtrrs[i].base.hi, state->mtrrs[i].base.lo,
+ state->mtrrs[i].mask.hi, state->mtrrs[i].mask.lo);
}
printk(BIOS_DEBUG, "top_mem: %08x%08x\n",
- state->top_mem.hi, state->top_mem.lo);
+ state->top_mem.hi, state->top_mem.lo);
printk(BIOS_DEBUG, "top_mem2: %08x%08x\n",
- state->top_mem2.hi, state->top_mem2.lo);
+ state->top_mem2.hi, state->top_mem2.lo);
printk(BIOS_DEBUG, "def_type: %08x%08x\n",
- state->def_type.hi, state->def_type.lo);
+ state->def_type.hi, state->def_type.lo);
}
#endif
@@ -171,7 +170,7 @@ static void set_init_ecc_mtrrs(void)
disable_cache();
/* First clear all of the msrs to be safe */
- for(i = 0; i < MTRR_COUNT; i++) {
+ for (i = 0; i < MTRR_COUNT; i++) {
msr_t zero;
zero.lo = zero.hi = 0;
wrmsr(MTRRphysBase_MSR(i), zero);
@@ -199,46 +198,47 @@ static void set_init_ecc_mtrrs(void)
enable_cache();
}
-static inline void clear_2M_ram(unsigned long basek, struct mtrr_state *mtrr_state)
+static inline void clear_2M_ram(unsigned long basek,
+ struct mtrr_state *mtrr_state)
{
- unsigned long limitk;
- unsigned long size;
- void *addr;
+ unsigned long limitk;
+ unsigned long size;
+ void *addr;
- /* Report every 64M */
- if ((basek % (64*1024)) == 0) {
+ /* Report every 64M */
+ if ((basek % (64 * 1024)) == 0) {
- /* Restore the normal state */
- map_2M_page(0);
- restore_mtrr_state(mtrr_state);
- enable_lapic();
+ /* Restore the normal state */
+ map_2M_page(0);
+ restore_mtrr_state(mtrr_state);
+ enable_lapic();
- /* Print a status message */
- printk(BIOS_DEBUG, "%c", (basek >= TOLM_KB)?'+':'-');
+ /* Print a status message */
+ printk(BIOS_DEBUG, "%c", (basek >= TOLM_KB) ? '+' : '-');
- /* Return to the initialization state */
- set_init_ecc_mtrrs();
- disable_lapic();
+ /* Return to the initialization state */
+ set_init_ecc_mtrrs();
+ disable_lapic();
- }
+ }
- limitk = (basek + ZERO_CHUNK_KB) & ~(ZERO_CHUNK_KB - 1);
+ limitk = (basek + ZERO_CHUNK_KB) & ~(ZERO_CHUNK_KB - 1);
#if 0
- /* couldn't happen, memory must on 2M boundary */
- if(limitk>endk) {
- limitk = enk;
- }
+ /* couldn't happen, memory must on 2M boundary */
+ if (limitk > endk) {
+ limitk = enk;
+ }
#endif
- size = (limitk - basek) << 10;
- addr = map_2M_page(basek >> 11);
- if (addr == MAPPING_ERROR) {
- printk(BIOS_ERR, "Cannot map page: %lx\n", basek >> 11);
- return;
- }
-
- /* clear memory 2M (limitk - basek) */
- addr = (void *)(((uint32_t)addr) | ((basek & 0x7ff) << 10));
- memset(addr, 0, size);
+ size = (limitk - basek) << 10;
+ addr = map_2M_page(basek >> 11);
+ if (addr == MAPPING_ERROR) {
+ printk(BIOS_ERR, "Cannot map page: %lx\n", basek >> 11);
+ return;
+ }
+
+ /* clear memory 2M (limitk - basek) */
+ addr = (void *)(((uint32_t) addr) | ((basek & 0x7ff) << 10));
+ memset(addr, 0, size);
}
static void init_ecc_memory(unsigned node_id)
@@ -271,14 +271,15 @@ static void init_ecc_memory(unsigned node_id)
/* Enable cache scrubbing at the lowest possible rate */
if (enable_scrubbing) {
pci_write_config32(f3_dev, SCRUB_CONTROL,
- (SCRUB_84ms << 16) | (SCRUB_84ms << 8) | (SCRUB_NONE << 0));
+ (SCRUB_84ms << 16) | (SCRUB_84ms << 8) |
+ (SCRUB_NONE << 0));
} else {
pci_write_config32(f3_dev, SCRUB_CONTROL,
- (SCRUB_NONE << 16) | (SCRUB_NONE << 8) | (SCRUB_NONE << 0));
+ (SCRUB_NONE << 16) | (SCRUB_NONE << 8) |
+ (SCRUB_NONE << 0));
printk(BIOS_DEBUG, "Scrubbing Disabled\n");
}
-
/* If ecc support is not enabled don't touch memory */
dcl = pci_read_config32(f2_dev, DRAM_CONFIG_LOW);
if (!(dcl & DCL_DimmEccEn)) {
@@ -286,32 +287,33 @@ static void init_ecc_memory(unsigned node_id)
return;
}
- startk = (pci_read_config32(f1_dev, 0x40 + (node_id*8)) & 0xffff0000) >> 2;
- endk = ((pci_read_config32(f1_dev, 0x44 + (node_id*8)) & 0xffff0000) >> 2) + 0x4000;
+ startk =
+ (pci_read_config32(f1_dev, 0x40 + (node_id * 8)) & 0xffff0000) >> 2;
+ endk =
+ ((pci_read_config32(f1_dev, 0x44 + (node_id * 8)) & 0xffff0000) >>
+ 2) + 0x4000;
#if CONFIG_HW_MEM_HOLE_SIZEK != 0
unsigned long hole_startk = 0;
- #if CONFIG_K8_REV_F_SUPPORT == 0
- if (!is_cpu_pre_e0())
- {
- #endif
-
- uint32_t val;
- val = pci_read_config32(f1_dev, 0xf0);
- if(val & 1) {
- hole_startk = ((val & (0xff<<24)) >> 10);
- }
- #if CONFIG_K8_REV_F_SUPPORT == 0
- }
- #endif
+#if CONFIG_K8_REV_F_SUPPORT == 0
+ if (!is_cpu_pre_e0()) {
#endif
+ uint32_t val;
+ val = pci_read_config32(f1_dev, 0xf0);
+ if (val & 1) {
+ hole_startk = ((val & (0xff << 24)) >> 10);
+ }
+#if CONFIG_K8_REV_F_SUPPORT == 0
+ }
+#endif
+#endif
/* Don't start too early */
begink = startk;
if (begink < (CONFIG_RAMTOP >> 10)) {
- begink = (CONFIG_RAMTOP >>10);
+ begink = (CONFIG_RAMTOP >> 10);
}
printk(BIOS_DEBUG, "Clearing memory %luK - %luK: ", begink, endk);
@@ -326,26 +328,22 @@ static void init_ecc_memory(unsigned node_id)
/* Walk through 2M chunks and zero them */
#if CONFIG_HW_MEM_HOLE_SIZEK != 0
/* here hole_startk can not be equal to begink, never. Also hole_startk is in 2M boundary, 64M? */
- if ( (hole_startk != 0) && ((begink < hole_startk) && (endk>(4*1024*1024)))) {
- for(basek = begink; basek < hole_startk;
- basek = ((basek + ZERO_CHUNK_KB) & ~(ZERO_CHUNK_KB - 1)))
- {
- clear_2M_ram(basek, &mtrr_state);
- }
- for(basek = 4*1024*1024; basek < endk;
- basek = ((basek + ZERO_CHUNK_KB) & ~(ZERO_CHUNK_KB - 1)))
- {
- clear_2M_ram(basek, &mtrr_state);
- }
- }
- else
+ if ((hole_startk != 0)
+ && ((begink < hole_startk) && (endk > (4 * 1024 * 1024)))) {
+ for (basek = begink; basek < hole_startk;
+ basek = ((basek + ZERO_CHUNK_KB) & ~(ZERO_CHUNK_KB - 1))) {
+ clear_2M_ram(basek, &mtrr_state);
+ }
+ for (basek = 4 * 1024 * 1024; basek < endk;
+ basek = ((basek + ZERO_CHUNK_KB) & ~(ZERO_CHUNK_KB - 1))) {
+ clear_2M_ram(basek, &mtrr_state);
+ }
+ } else
#endif
- for(basek = begink; basek < endk;
- basek = ((basek + ZERO_CHUNK_KB) & ~(ZERO_CHUNK_KB - 1)))
- {
- clear_2M_ram(basek, &mtrr_state);
- }
-
+ for (basek = begink; basek < endk;
+ basek = ((basek + ZERO_CHUNK_KB) & ~(ZERO_CHUNK_KB - 1))) {
+ clear_2M_ram(basek, &mtrr_state);
+ }
/* Restore the normal state */
map_2M_page(0);
@@ -353,20 +351,20 @@ static void init_ecc_memory(unsigned node_id)
enable_lapic();
/* Set the scrub base address registers */
- pci_write_config32(f3_dev, SCRUB_ADDR_LOW, startk << 10);
+ pci_write_config32(f3_dev, SCRUB_ADDR_LOW, startk << 10);
pci_write_config32(f3_dev, SCRUB_ADDR_HIGH, startk >> 22);
/* Enable the scrubber? */
if (enable_scrubbing) {
/* Enable scrubbing at the lowest possible rate */
pci_write_config32(f3_dev, SCRUB_CONTROL,
- (SCRUB_84ms << 16) | (SCRUB_84ms << 8) | (SCRUB_84ms << 0));
+ (SCRUB_84ms << 16) | (SCRUB_84ms << 8) |
+ (SCRUB_84ms << 0));
}
printk(BIOS_DEBUG, " done\n");
}
-
static inline void k8_errata(void)
{
msr_t msr;
@@ -384,7 +382,7 @@ static inline void k8_errata(void)
/* Erratum 81... */
msr = rdmsr_amd(DC_CFG_MSR);
- msr.lo |= (1 << 10);
+ msr.lo |= (1 << 10);
wrmsr_amd(DC_CFG_MSR, msr);
}
@@ -432,12 +430,12 @@ static inline void k8_errata(void)
msr.hi |= 1 << (43 - 32);
wrmsr_amd(BU_CFG_MSR, msr);
- if(is_cpu_d0()) {
- /* Erratum 110 ...*/
+ if (is_cpu_d0()) {
+ /* Erratum 110 ... */
msr = rdmsr_amd(CPU_ID_HYPER_EXT_FEATURES);
- msr.hi |=1;
+ msr.hi |= 1;
wrmsr_amd(CPU_ID_HYPER_EXT_FEATURES, msr);
- }
+ }
#endif
#if CONFIG_K8_REV_F_SUPPORT == 0
@@ -445,9 +443,9 @@ static inline void k8_errata(void)
#endif
{
/* Erratum 110 ... */
- msr = rdmsr_amd(CPU_ID_EXT_FEATURES_MSR);
- msr.hi |=1;
- wrmsr_amd(CPU_ID_EXT_FEATURES_MSR, msr);
+ msr = rdmsr_amd(CPU_ID_EXT_FEATURES_MSR);
+ msr.hi |= 1;
+ wrmsr_amd(CPU_ID_EXT_FEATURES_MSR, msr);
}
/* Erratum 122 */
@@ -456,10 +454,10 @@ static inline void k8_errata(void)
wrmsr(HWCR_MSR, msr);
#if CONFIG_K8_REV_F_SUPPORT == 1
- /* Erratum 131... */
- msr = rdmsr(NB_CFG_MSR);
- msr.lo |= 1 << 20;
- wrmsr(NB_CFG_MSR, msr);
+ /* Erratum 131... */
+ msr = rdmsr(NB_CFG_MSR);
+ msr.lo |= 1 << 20;
+ wrmsr(NB_CFG_MSR, msr);
#endif
}
@@ -474,7 +472,7 @@ static void model_fxx_init(device_t dev)
msr_t msr;
struct node_core_id id;
#if CONFIG_LOGICAL_CPUS == 1
- unsigned siblings;
+ u32 siblings;
#endif
#if CONFIG_K8_REV_F_SUPPORT == 1
@@ -484,7 +482,7 @@ static void model_fxx_init(device_t dev)
#endif
#if CONFIG_USBDEBUG_DIRECT
- if(!ehci_debug_addr)
+ if (!ehci_debug_addr)
ehci_debug_addr = get_ehci_debug();
set_ehci_debug(0);
#endif
@@ -498,7 +496,7 @@ static void model_fxx_init(device_t dev)
set_ehci_debug(ehci_debug_addr);
#endif
- /* Update the microcode */
+ /* Update the microcode */
model_fxx_update_microcode(dev->device);
disable_cache();
@@ -506,8 +504,8 @@ static void model_fxx_init(device_t dev)
/* zero the machine check error status registers */
msr.lo = 0;
msr.hi = 0;
- for(i=0; i<5; i++) {
- wrmsr(MCI_STATUS + (i*4),msr);
+ for (i = 0; i < 5; i++) {
+ wrmsr(MCI_STATUS + (i * 4), msr);
}
k8_errata();
@@ -526,34 +524,34 @@ static void model_fxx_init(device_t dev)
setup_lapic();
#if CONFIG_LOGICAL_CPUS == 1
- siblings = cpuid_ecx(0x80000008) & 0xff;
+ siblings = cpuid_ecx(0x80000008) & 0xff;
- if(siblings>0) {
- msr = rdmsr_amd(CPU_ID_FEATURES_MSR);
- msr.lo |= 1 << 28;
- wrmsr_amd(CPU_ID_FEATURES_MSR, msr);
+ if (siblings > 0) {
+ msr = rdmsr_amd(CPU_ID_FEATURES_MSR);
+ msr.lo |= 1 << 28;
+ wrmsr_amd(CPU_ID_FEATURES_MSR, msr);
- msr = rdmsr_amd(LOGICAL_CPUS_NUM_MSR);
- msr.lo = (siblings+1)<<16;
- wrmsr_amd(LOGICAL_CPUS_NUM_MSR, msr);
+ msr = rdmsr_amd(LOGICAL_CPUS_NUM_MSR);
+ msr.lo = (siblings + 1) << 16;
+ wrmsr_amd(LOGICAL_CPUS_NUM_MSR, msr);
- msr = rdmsr_amd(CPU_ID_EXT_FEATURES_MSR);
- msr.hi |= 1<<(33-32);
- wrmsr_amd(CPU_ID_EXT_FEATURES_MSR, msr);
+ msr = rdmsr_amd(CPU_ID_EXT_FEATURES_MSR);
+ msr.hi |= 1 << (33 - 32);
+ wrmsr_amd(CPU_ID_EXT_FEATURES_MSR, msr);
}
-
#endif
- id = get_node_core_id(read_nb_cfg_54()); // pre e0 nb_cfg_54 can not be set
+ id = get_node_core_id(read_nb_cfg_54()); // pre e0 nb_cfg_54 can not be set
/* Is this a bad location? In particular can another node prefecth
* data from this node before we have initialized it?
*/
- if (id.coreid == 0) init_ecc_memory(id.nodeid); // only do it for core 0
+ if (id.coreid == 0)
+ init_ecc_memory(id.nodeid); // only do it for core 0
#if CONFIG_LOGICAL_CPUS==1
- /* Start up my cpu siblings */
-// if(id.coreid==0) amd_sibling_init(dev); // Don't need core1 is already be put in the CPU BUS in bus_cpu_scan
+ /* Start up my cpu siblings */
+// if(id.coreid==0) amd_sibling_init(dev); // Don't need core1 is already be put in the CPU BUS in bus_cpu_scan
#endif
}