linux/arch/s390/kernel/early.c

592 lines
15 KiB
C
Raw Normal View History

/*
* Copyright IBM Corp. 2007, 2009
* Author(s): Hongjie Yang <hongjie@us.ibm.com>,
* Heiko Carstens <heiko.carstens@de.ibm.com>
*/
#define KMSG_COMPONENT "setup"
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#include <linux/compiler.h>
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/ctype.h>
#include <linux/lockdep.h>
#include <linux/extable.h>
#include <linux/pfn.h>
#include <linux/uaccess.h>
#include <linux/kernel.h>
#include <asm/diag.h>
#include <asm/ebcdic.h>
#include <asm/ipl.h>
#include <asm/lowcore.h>
#include <asm/processor.h>
#include <asm/sections.h>
#include <asm/setup.h>
#include <asm/sysinfo.h>
#include <asm/cpcmd.h>
#include <asm/sclp.h>
#include <asm/facility.h>
#include "entry.h"
/*
* Create a Kernel NSS if the SAVESYS= parameter is defined
*/
#define DEFSYS_CMD_SIZE 128
#define SAVESYS_CMD_SIZE 32
char kernel_nss_name[NSS_NAME_SIZE + 1];
static void __init setup_boot_command_line(void);
/*
* Get the TOD clock running.
*/
static void __init reset_tod_clock(void)
{
u64 time;
if (store_tod_clock(&time) == 0)
return;
/* TOD clock not running. Set the clock to Unix Epoch. */
if (set_tod_clock(TOD_UNIX_EPOCH) != 0 || store_tod_clock(&time) != 0)
disabled_wait(0);
memset(tod_clock_base, 0, 16);
*(__u64 *) &tod_clock_base[1] = TOD_UNIX_EPOCH;
S390_lowcore.last_update_clock = TOD_UNIX_EPOCH;
}
#ifdef CONFIG_SHARED_KERNEL
int __init savesys_ipl_nss(char *cmd, const int cmdlen);
asm(
" .section .init.text,\"ax\",@progbits\n"
" .align 4\n"
" .type savesys_ipl_nss, @function\n"
"savesys_ipl_nss:\n"
" stmg 6,15,48(15)\n"
" lgr 14,3\n"
" sam31\n"
" diag 2,14,0x8\n"
" sam64\n"
" lgr 2,14\n"
" lmg 6,15,48(15)\n"
" br 14\n"
" .size savesys_ipl_nss, .-savesys_ipl_nss\n"
" .previous\n");
static __initdata char upper_command_line[COMMAND_LINE_SIZE];
static noinline __init void create_kernel_nss(void)
{
unsigned int i, stext_pfn, eshared_pfn, end_pfn, min_size;
#ifdef CONFIG_BLK_DEV_INITRD
unsigned int sinitrd_pfn, einitrd_pfn;
#endif
int response;
int hlen;
size_t len;
char *savesys_ptr;
char defsys_cmd[DEFSYS_CMD_SIZE];
char savesys_cmd[SAVESYS_CMD_SIZE];
/* Do nothing if we are not running under VM */
if (!MACHINE_IS_VM)
return;
/* Convert COMMAND_LINE to upper case */
for (i = 0; i < strlen(boot_command_line); i++)
upper_command_line[i] = toupper(boot_command_line[i]);
savesys_ptr = strstr(upper_command_line, "SAVESYS=");
if (!savesys_ptr)
return;
savesys_ptr += 8; /* Point to the beginning of the NSS name */
for (i = 0; i < NSS_NAME_SIZE; i++) {
if (savesys_ptr[i] == ' ' || savesys_ptr[i] == '\0')
break;
kernel_nss_name[i] = savesys_ptr[i];
}
stext_pfn = PFN_DOWN(__pa(&_stext));
eshared_pfn = PFN_DOWN(__pa(&_eshared));
end_pfn = PFN_UP(__pa(&_end));
min_size = end_pfn << 2;
hlen = snprintf(defsys_cmd, DEFSYS_CMD_SIZE,
"DEFSYS %s 00000-%.5X EW %.5X-%.5X SR %.5X-%.5X",
kernel_nss_name, stext_pfn - 1, stext_pfn,
eshared_pfn - 1, eshared_pfn, end_pfn);
#ifdef CONFIG_BLK_DEV_INITRD
if (INITRD_START && INITRD_SIZE) {
sinitrd_pfn = PFN_DOWN(__pa(INITRD_START));
einitrd_pfn = PFN_UP(__pa(INITRD_START + INITRD_SIZE));
min_size = einitrd_pfn << 2;
hlen += snprintf(defsys_cmd + hlen, DEFSYS_CMD_SIZE - hlen,
" EW %.5X-%.5X", sinitrd_pfn, einitrd_pfn);
}
#endif
snprintf(defsys_cmd + hlen, DEFSYS_CMD_SIZE - hlen,
" EW MINSIZE=%.7iK PARMREGS=0-13", min_size);
defsys_cmd[DEFSYS_CMD_SIZE - 1] = '\0';
snprintf(savesys_cmd, SAVESYS_CMD_SIZE, "SAVESYS %s \n IPL %s",
kernel_nss_name, kernel_nss_name);
savesys_cmd[SAVESYS_CMD_SIZE - 1] = '\0';
__cpcmd(defsys_cmd, NULL, 0, &response);
if (response != 0) {
pr_err("Defining the Linux kernel NSS failed with rc=%d\n",
response);
kernel_nss_name[0] = '\0';
return;
}
len = strlen(savesys_cmd);
ASCEBC(savesys_cmd, len);
response = savesys_ipl_nss(savesys_cmd, len);
/* On success: response is equal to the command size,
* max SAVESYS_CMD_SIZE
* On error: response contains the numeric portion of cp error message.
* for SAVESYS it will be >= 263
* for missing privilege class, it will be 1
*/
if (response > SAVESYS_CMD_SIZE || response == 1) {
pr_err("Saving the Linux kernel NSS failed with rc=%d\n",
response);
kernel_nss_name[0] = '\0';
return;
}
/* re-initialize cputime accounting. */
get_tod_clock_ext(tod_clock_base);
S390_lowcore.last_update_clock = *(__u64 *) &tod_clock_base[1];
S390_lowcore.last_update_timer = 0x7fffffffffffffffULL;
S390_lowcore.user_timer = 0;
S390_lowcore.system_timer = 0;
asm volatile("SPT 0(%0)" : : "a" (&S390_lowcore.last_update_timer));
/* re-setup boot command line with new ipl vm parms */
ipl_update_parameters();
setup_boot_command_line();
ipl_flags = IPL_NSS_VALID;
}
#else /* CONFIG_SHARED_KERNEL */
static inline void create_kernel_nss(void) { }
#endif /* CONFIG_SHARED_KERNEL */
/*
* Clear bss memory
*/
static noinline __init void clear_bss_section(void)
{
memset(__bss_start, 0, __bss_stop - __bss_start);
}
/*
* Initialize storage key for kernel pages
*/
static noinline __init void init_kernel_storage_key(void)
{
#if PAGE_DEFAULT_KEY
unsigned long end_pfn, init_pfn;
end_pfn = PFN_UP(__pa(&_end));
for (init_pfn = 0 ; init_pfn < end_pfn; init_pfn++)
page_set_storage_key(init_pfn << PAGE_SHIFT,
PAGE_DEFAULT_KEY, 0);
#endif
}
static __initdata char sysinfo_page[PAGE_SIZE] __aligned(PAGE_SIZE);
static noinline __init void detect_machine_type(void)
{
struct sysinfo_3_2_2 *vmms = (struct sysinfo_3_2_2 *)&sysinfo_page;
/* Check current-configuration-level */
if (stsi(NULL, 0, 0, 0) <= 2) {
S390_lowcore.machine_flags |= MACHINE_FLAG_LPAR;
return;
}
/* Get virtual-machine cpu information. */
if (stsi(vmms, 3, 2, 2) || !vmms->count)
return;
/* Running under KVM? If not we assume z/VM */
if (!memcmp(vmms->vm[0].cpi, "\xd2\xe5\xd4", 3))
S390_lowcore.machine_flags |= MACHINE_FLAG_KVM;
else
S390_lowcore.machine_flags |= MACHINE_FLAG_VM;
}
/* Remove leading, trailing and double whitespace. */
static inline void strim_all(char *str)
{
char *s;
s = strim(str);
if (s != str)
memmove(str, s, strlen(s));
while (*str) {
if (!isspace(*str++))
continue;
if (isspace(*str)) {
s = skip_spaces(str);
memmove(str, s, strlen(s) + 1);
}
}
}
static noinline __init void setup_arch_string(void)
{
struct sysinfo_1_1_1 *mach = (struct sysinfo_1_1_1 *)&sysinfo_page;
struct sysinfo_3_2_2 *vm = (struct sysinfo_3_2_2 *)&sysinfo_page;
char mstr[80], hvstr[17];
if (stsi(mach, 1, 1, 1))
return;
EBCASC(mach->manufacturer, sizeof(mach->manufacturer));
EBCASC(mach->type, sizeof(mach->type));
EBCASC(mach->model, sizeof(mach->model));
EBCASC(mach->model_capacity, sizeof(mach->model_capacity));
sprintf(mstr, "%-16.16s %-4.4s %-16.16s %-16.16s",
mach->manufacturer, mach->type,
mach->model, mach->model_capacity);
strim_all(mstr);
if (stsi(vm, 3, 2, 2) == 0 && vm->count) {
EBCASC(vm->vm[0].cpi, sizeof(vm->vm[0].cpi));
sprintf(hvstr, "%-16.16s", vm->vm[0].cpi);
strim_all(hvstr);
} else {
sprintf(hvstr, "%s",
MACHINE_IS_LPAR ? "LPAR" :
MACHINE_IS_VM ? "z/VM" :
MACHINE_IS_KVM ? "KVM" : "unknown");
}
dump_stack_set_arch_desc("%s (%s)", mstr, hvstr);
}
static __init void setup_topology(void)
{
int max_mnest;
if (!test_facility(11))
return;
S390_lowcore.machine_flags |= MACHINE_FLAG_TOPOLOGY;
for (max_mnest = 6; max_mnest > 1; max_mnest--) {
if (stsi(&sysinfo_page, 15, 1, max_mnest) == 0)
break;
}
topology_max_mnest = max_mnest;
}
static void early_pgm_check_handler(void)
{
const struct exception_table_entry *fixup;
unsigned long cr0, cr0_new;
unsigned long addr;
addr = S390_lowcore.program_old_psw.addr;
fixup = search_exception_tables(addr);
if (!fixup)
disabled_wait(0);
/* Disable low address protection before storing into lowcore. */
__ctl_store(cr0, 0, 0);
cr0_new = cr0 & ~(1UL << 28);
__ctl_load(cr0_new, 0, 0);
S390_lowcore.program_old_psw.addr = extable_fixup(fixup);
__ctl_load(cr0, 0, 0);
}
static noinline __init void setup_lowcore_early(void)
{
psw_t psw;
psw.mask = PSW_MASK_BASE | PSW_DEFAULT_KEY | PSW_MASK_EA | PSW_MASK_BA;
psw.addr = (unsigned long) s390_base_ext_handler;
S390_lowcore.external_new_psw = psw;
psw.addr = (unsigned long) s390_base_pgm_handler;
S390_lowcore.program_new_psw = psw;
s390_base_pgm_handler_fn = early_pgm_check_handler;
S390_lowcore.preempt_count = INIT_PREEMPT_COUNT;
}
static noinline __init void setup_facility_list(void)
{
stfle(S390_lowcore.stfle_fac_list,
ARRAY_SIZE(S390_lowcore.stfle_fac_list));
}
static __init void detect_diag9c(void)
{
unsigned int cpu_address;
int rc;
cpu_address = stap();
diag_stat_inc(DIAG_STAT_X09C);
asm volatile(
" diag %2,0,0x9c\n"
"0: la %0,0\n"
"1:\n"
EX_TABLE(0b,1b)
: "=d" (rc) : "0" (-EOPNOTSUPP), "d" (cpu_address) : "cc");
if (!rc)
S390_lowcore.machine_flags |= MACHINE_FLAG_DIAG9C;
}
static __init void detect_diag44(void)
{
int rc;
diag_stat_inc(DIAG_STAT_X044);
asm volatile(
" diag 0,0,0x44\n"
"0: la %0,0\n"
"1:\n"
EX_TABLE(0b,1b)
: "=d" (rc) : "0" (-EOPNOTSUPP) : "cc");
if (!rc)
S390_lowcore.machine_flags |= MACHINE_FLAG_DIAG44;
}
static __init void detect_machine_facilities(void)
{
if (test_facility(8)) {
S390_lowcore.machine_flags |= MACHINE_FLAG_EDAT1;
__ctl_set_bit(0, 23);
}
if (test_facility(78))
S390_lowcore.machine_flags |= MACHINE_FLAG_EDAT2;
if (test_facility(3))
S390_lowcore.machine_flags |= MACHINE_FLAG_IDTE;
if (test_facility(40))
S390_lowcore.machine_flags |= MACHINE_FLAG_LPP;
if (test_facility(50) && test_facility(73))
S390_lowcore.machine_flags |= MACHINE_FLAG_TE;
if (test_facility(51))
S390_lowcore.machine_flags |= MACHINE_FLAG_TLB_LC;
if (test_facility(129)) {
S390_lowcore.machine_flags |= MACHINE_FLAG_VX;
__ctl_set_bit(0, 17);
}
if (test_facility(130)) {
S390_lowcore.machine_flags |= MACHINE_FLAG_NX;
__ctl_set_bit(0, 20);
}
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
if (test_facility(133))
S390_lowcore.machine_flags |= MACHINE_FLAG_GS;
if (test_facility(139) && (tod_clock_base[1] & 0x80)) {
/* Enabled signed clock comparator comparisons */
S390_lowcore.machine_flags |= MACHINE_FLAG_SCC;
clock_comparator_max = -1ULL >> 1;
__ctl_set_bit(0, 53);
}
}
static inline void save_vector_registers(void)
{
#ifdef CONFIG_CRASH_DUMP
if (test_facility(129))
save_vx_regs(boot_cpu_vector_save_area);
#endif
}
static int __init topology_setup(char *str)
{
bool enabled;
int rc;
rc = kstrtobool(str, &enabled);
if (!rc && !enabled)
S390_lowcore.machine_flags &= ~MACHINE_FLAG_TOPOLOGY;
return rc;
}
early_param("topology", topology_setup);
static int __init disable_vector_extension(char *str)
{
S390_lowcore.machine_flags &= ~MACHINE_FLAG_VX;
__ctl_clear_bit(0, 17);
return 1;
}
early_param("novx", disable_vector_extension);
static int __init noexec_setup(char *str)
{
bool enabled;
int rc;
rc = kstrtobool(str, &enabled);
if (!rc && !enabled) {
/* Disable no-execute support */
S390_lowcore.machine_flags &= ~MACHINE_FLAG_NX;
__ctl_clear_bit(0, 20);
}
return rc;
}
early_param("noexec", noexec_setup);
static int __init cad_setup(char *str)
{
bool enabled;
int rc;
rc = kstrtobool(str, &enabled);
if (!rc && enabled && test_facility(128))
/* Enable problem state CAD. */
__ctl_set_bit(2, 3);
return rc;
}
early_param("cad", cad_setup);
static __init void memmove_early(void *dst, const void *src, size_t n)
{
unsigned long addr;
long incr;
psw_t old;
if (!n)
return;
incr = 1;
if (dst > src) {
incr = -incr;
dst += n - 1;
src += n - 1;
}
old = S390_lowcore.program_new_psw;
S390_lowcore.program_new_psw.mask = __extract_psw();
asm volatile(
" larl %[addr],1f\n"
" stg %[addr],%[psw_pgm_addr]\n"
"0: mvc 0(1,%[dst]),0(%[src])\n"
" agr %[dst],%[incr]\n"
" agr %[src],%[incr]\n"
" brctg %[n],0b\n"
"1:\n"
: [addr] "=&d" (addr),
[psw_pgm_addr] "=Q" (S390_lowcore.program_new_psw.addr),
[dst] "+&a" (dst), [src] "+&a" (src), [n] "+d" (n)
: [incr] "d" (incr)
: "cc", "memory");
S390_lowcore.program_new_psw = old;
}
static __init noinline void ipl_save_parameters(void)
{
void *src, *dst;
src = (void *)(unsigned long) S390_lowcore.ipl_parmblock_ptr;
dst = (void *) IPL_PARMBLOCK_ORIGIN;
memmove_early(dst, src, PAGE_SIZE);
S390_lowcore.ipl_parmblock_ptr = IPL_PARMBLOCK_ORIGIN;
}
static __init noinline void rescue_initrd(void)
{
#ifdef CONFIG_BLK_DEV_INITRD
unsigned long min_initrd_addr = (unsigned long) _end + (4UL << 20);
/*
* Just like in case of IPL from VM reader we make sure there is a
* gap of 4MB between end of kernel and start of initrd.
* That way we can also be sure that saving an NSS will succeed,
* which however only requires different segments.
*/
if (!INITRD_START || !INITRD_SIZE)
return;
if (INITRD_START >= min_initrd_addr)
return;
memmove_early((void *) min_initrd_addr, (void *) INITRD_START, INITRD_SIZE);
INITRD_START = min_initrd_addr;
#endif
}
/* Set up boot command line */
static void __init append_to_cmdline(size_t (*ipl_data)(char *, size_t))
{
char *parm, *delim;
size_t rc, len;
len = strlen(boot_command_line);
delim = boot_command_line + len; /* '\0' character position */
parm = boot_command_line + len + 1; /* append right after '\0' */
rc = ipl_data(parm, COMMAND_LINE_SIZE - len - 1);
if (rc) {
if (*parm == '=')
memmove(boot_command_line, parm + 1, rc);
else
*delim = ' '; /* replace '\0' with space */
}
}
static inline int has_ebcdic_char(const char *str)
{
int i;
for (i = 0; str[i]; i++)
if (str[i] & 0x80)
return 1;
return 0;
}
static void __init setup_boot_command_line(void)
{
COMMAND_LINE[ARCH_COMMAND_LINE_SIZE - 1] = 0;
/* convert arch command line to ascii if necessary */
if (has_ebcdic_char(COMMAND_LINE))
EBCASC(COMMAND_LINE, ARCH_COMMAND_LINE_SIZE);
/* copy arch command line */
strlcpy(boot_command_line, strstrip(COMMAND_LINE),
ARCH_COMMAND_LINE_SIZE);
/* append IPL PARM data to the boot command line */
if (MACHINE_IS_VM)
append_to_cmdline(append_ipl_vmparm);
append_to_cmdline(append_ipl_scpdata);
}
/*
* Save ipl parameters, clear bss memory, initialize storage keys
* and create a kernel NSS at startup if the SAVESYS= parm is defined
*/
void __init startup_init(void)
{
reset_tod_clock();
ipl_save_parameters();
rescue_initrd();
clear_bss_section();
ipl_verify_parameters();
time_early_init();
init_kernel_storage_key();
lockdep_off();
setup_lowcore_early();
setup_facility_list();
detect_machine_type();
setup_arch_string();
ipl_update_parameters();
setup_boot_command_line();
create_kernel_nss();
detect_diag9c();
detect_diag44();
detect_machine_facilities();
save_vector_registers();
setup_topology();
sclp_early_detect();
lockdep_on();
}