linux/arch/s390/kernel/smp.c

1173 lines
30 KiB
C
Raw Normal View History

/*
* SMP related functions
*
* Copyright IBM Corp. 1999, 2012
* Author(s): Denis Joseph Barrow,
* Martin Schwidefsky <schwidefsky@de.ibm.com>,
* Heiko Carstens <heiko.carstens@de.ibm.com>,
*
* based on other smp stuff by
* (c) 1995 Alan Cox, CymruNET Ltd <alan@cymru.net>
* (c) 1998 Ingo Molnar
*
* The code outside of smp.c uses logical cpu numbers, only smp.c does
* the translation of logical to physical cpu ids. All new code that
* operates on physical cpu numbers needs to go into smp.c.
*/
#define KMSG_COMPONENT "cpu"
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#include <linux/workqueue.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/mm.h>
Remove fs.h from mm.h Remove fs.h from mm.h. For this, 1) Uninline vma_wants_writenotify(). It's pretty huge anyway. 2) Add back fs.h or less bloated headers (err.h) to files that need it. As result, on x86_64 allyesconfig, fs.h dependencies cut down from 3929 files rebuilt down to 3444 (-12.3%). Cross-compile tested without regressions on my two usual configs and (sigh): alpha arm-mx1ads mips-bigsur powerpc-ebony alpha-allnoconfig arm-neponset mips-capcella powerpc-g5 alpha-defconfig arm-netwinder mips-cobalt powerpc-holly alpha-up arm-netx mips-db1000 powerpc-iseries arm arm-ns9xxx mips-db1100 powerpc-linkstation arm-assabet arm-omap_h2_1610 mips-db1200 powerpc-lite5200 arm-at91rm9200dk arm-onearm mips-db1500 powerpc-maple arm-at91rm9200ek arm-picotux200 mips-db1550 powerpc-mpc7448_hpc2 arm-at91sam9260ek arm-pleb mips-ddb5477 powerpc-mpc8272_ads arm-at91sam9261ek arm-pnx4008 mips-decstation powerpc-mpc8313_rdb arm-at91sam9263ek arm-pxa255-idp mips-e55 powerpc-mpc832x_mds arm-at91sam9rlek arm-realview mips-emma2rh powerpc-mpc832x_rdb arm-ateb9200 arm-realview-smp mips-excite powerpc-mpc834x_itx arm-badge4 arm-rpc mips-fulong powerpc-mpc834x_itxgp arm-carmeva arm-s3c2410 mips-ip22 powerpc-mpc834x_mds arm-cerfcube arm-shannon mips-ip27 powerpc-mpc836x_mds arm-clps7500 arm-shark mips-ip32 powerpc-mpc8540_ads arm-collie arm-simpad mips-jazz powerpc-mpc8544_ds arm-corgi arm-spitz mips-jmr3927 powerpc-mpc8560_ads arm-csb337 arm-trizeps4 mips-malta powerpc-mpc8568mds arm-csb637 arm-versatile mips-mipssim powerpc-mpc85xx_cds arm-ebsa110 i386 mips-mpc30x powerpc-mpc8641_hpcn arm-edb7211 i386-allnoconfig mips-msp71xx powerpc-mpc866_ads arm-em_x270 i386-defconfig mips-ocelot powerpc-mpc885_ads arm-ep93xx i386-up mips-pb1100 powerpc-pasemi arm-footbridge ia64 mips-pb1500 powerpc-pmac32 arm-fortunet ia64-allnoconfig mips-pb1550 powerpc-ppc64 arm-h3600 ia64-bigsur mips-pnx8550-jbs powerpc-prpmc2800 arm-h7201 ia64-defconfig mips-pnx8550-stb810 powerpc-ps3 arm-h7202 ia64-gensparse mips-qemu powerpc-pseries arm-hackkit ia64-sim mips-rbhma4200 powerpc-up arm-integrator ia64-sn2 mips-rbhma4500 s390 arm-iop13xx ia64-tiger mips-rm200 s390-allnoconfig arm-iop32x ia64-up mips-sb1250-swarm s390-defconfig arm-iop33x ia64-zx1 mips-sead s390-up arm-ixp2000 m68k mips-tb0219 sparc arm-ixp23xx m68k-amiga mips-tb0226 sparc-allnoconfig arm-ixp4xx m68k-apollo mips-tb0287 sparc-defconfig arm-jornada720 m68k-atari mips-workpad sparc-up arm-kafa m68k-bvme6000 mips-wrppmc sparc64 arm-kb9202 m68k-hp300 mips-yosemite sparc64-allnoconfig arm-ks8695 m68k-mac parisc sparc64-defconfig arm-lart m68k-mvme147 parisc-allnoconfig sparc64-up arm-lpd270 m68k-mvme16x parisc-defconfig um-x86_64 arm-lpd7a400 m68k-q40 parisc-up x86_64 arm-lpd7a404 m68k-sun3 powerpc x86_64-allnoconfig arm-lubbock m68k-sun3x powerpc-cell x86_64-defconfig arm-lusl7200 mips powerpc-celleb x86_64-up arm-mainstone mips-atlas powerpc-chrp32 Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-30 06:36:13 +08:00
#include <linux/err.h>
#include <linux/spinlock.h>
#include <linux/kernel_stat.h>
#include <linux/delay.h>
#include <linux/interrupt.h>
#include <linux/irqflags.h>
#include <linux/cpu.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/crash_dump.h>
#include <linux/memblock.h>
#include <asm/asm-offsets.h>
#include <asm/switch_to.h>
#include <asm/facility.h>
#include <asm/ipl.h>
#include <asm/setup.h>
#include <asm/irq.h>
#include <asm/tlbflush.h>
#include <asm/vtimer.h>
#include <asm/lowcore.h>
#include <asm/sclp.h>
#include <asm/vdso.h>
#include <asm/debug.h>
#include <asm/os_info.h>
#include <asm/sigp.h>
#include <asm/idle.h>
#include "entry.h"
enum {
ec_schedule = 0,
ec_call_function_single,
ec_stop_cpu,
};
enum {
CPU_STATE_STANDBY,
CPU_STATE_CONFIGURED,
};
static DEFINE_PER_CPU(struct cpu *, cpu_device);
struct pcpu {
struct _lowcore *lowcore; /* lowcore page(s) for the cpu */
unsigned long ec_mask; /* bit mask for ec_xxx functions */
signed char state; /* physical cpu state */
signed char polarization; /* physical polarization */
u16 address; /* physical cpu address */
};
static u8 boot_core_type;
static struct pcpu pcpu_devices[NR_CPUS];
unsigned int smp_cpu_mt_shift;
EXPORT_SYMBOL(smp_cpu_mt_shift);
unsigned int smp_cpu_mtid;
EXPORT_SYMBOL(smp_cpu_mtid);
static unsigned int smp_max_threads __initdata = -1U;
static int __init early_nosmt(char *s)
{
smp_max_threads = 1;
return 0;
}
early_param("nosmt", early_nosmt);
static int __init early_smt(char *s)
{
get_option(&s, &smp_max_threads);
return 0;
}
early_param("smt", early_smt);
/*
* The smp_cpu_state_mutex must be held when changing the state or polarization
* member of a pcpu data structure within the pcpu_devices arreay.
*/
DEFINE_MUTEX(smp_cpu_state_mutex);
/*
* Signal processor helper functions.
*/
static inline int __pcpu_sigp_relax(u16 addr, u8 order, unsigned long parm,
u32 *status)
{
int cc;
while (1) {
cc = __pcpu_sigp(addr, order, parm, NULL);
if (cc != SIGP_CC_BUSY)
return cc;
cpu_relax();
}
}
static int pcpu_sigp_retry(struct pcpu *pcpu, u8 order, u32 parm)
{
int cc, retry;
for (retry = 0; ; retry++) {
cc = __pcpu_sigp(pcpu->address, order, parm, NULL);
if (cc != SIGP_CC_BUSY)
break;
if (retry >= 3)
udelay(10);
}
return cc;
}
static inline int pcpu_stopped(struct pcpu *pcpu)
{
u32 uninitialized_var(status);
if (__pcpu_sigp(pcpu->address, SIGP_SENSE,
0, &status) != SIGP_CC_STATUS_STORED)
return 0;
return !!(status & (SIGP_STATUS_CHECK_STOP|SIGP_STATUS_STOPPED));
}
static inline int pcpu_running(struct pcpu *pcpu)
{
if (__pcpu_sigp(pcpu->address, SIGP_SENSE_RUNNING,
0, NULL) != SIGP_CC_STATUS_STORED)
return 1;
/* Status stored condition code is equivalent to cpu not running. */
return 0;
}
/*
* Find struct pcpu by cpu address.
*/
static struct pcpu *pcpu_find_address(const struct cpumask *mask, u16 address)
{
int cpu;
for_each_cpu(cpu, mask)
if (pcpu_devices[cpu].address == address)
return pcpu_devices + cpu;
return NULL;
}
static void pcpu_ec_call(struct pcpu *pcpu, int ec_bit)
{
int order;
if (test_and_set_bit(ec_bit, &pcpu->ec_mask))
return;
order = pcpu_running(pcpu) ? SIGP_EXTERNAL_CALL : SIGP_EMERGENCY_SIGNAL;
pcpu_sigp_retry(pcpu, order, 0);
}
#define ASYNC_FRAME_OFFSET (ASYNC_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE)
#define PANIC_FRAME_OFFSET (PAGE_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE)
2013-06-19 05:04:52 +08:00
static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu)
{
unsigned long async_stack, panic_stack;
struct _lowcore *lc;
if (pcpu != &pcpu_devices[0]) {
pcpu->lowcore = (struct _lowcore *)
__get_free_pages(GFP_KERNEL | GFP_DMA, LC_ORDER);
async_stack = __get_free_pages(GFP_KERNEL, ASYNC_ORDER);
panic_stack = __get_free_page(GFP_KERNEL);
if (!pcpu->lowcore || !panic_stack || !async_stack)
goto out;
} else {
async_stack = pcpu->lowcore->async_stack - ASYNC_FRAME_OFFSET;
panic_stack = pcpu->lowcore->panic_stack - PANIC_FRAME_OFFSET;
}
lc = pcpu->lowcore;
memcpy(lc, &S390_lowcore, 512);
memset((char *) lc + 512, 0, sizeof(*lc) - 512);
lc->async_stack = async_stack + ASYNC_FRAME_OFFSET;
lc->panic_stack = panic_stack + PANIC_FRAME_OFFSET;
lc->cpu_nr = cpu;
lc->spinlock_lockval = arch_spin_lockval(cpu);
if (MACHINE_HAS_VX)
lc->vector_save_area_addr =
(unsigned long) &lc->vector_save_area;
if (vdso_alloc_per_cpu(lc))
goto out;
lowcore_ptr[cpu] = lc;
pcpu_sigp_retry(pcpu, SIGP_SET_PREFIX, (u32)(unsigned long) lc);
return 0;
out:
if (pcpu != &pcpu_devices[0]) {
free_page(panic_stack);
free_pages(async_stack, ASYNC_ORDER);
free_pages((unsigned long) pcpu->lowcore, LC_ORDER);
}
return -ENOMEM;
}
#ifdef CONFIG_HOTPLUG_CPU
static void pcpu_free_lowcore(struct pcpu *pcpu)
{
pcpu_sigp_retry(pcpu, SIGP_SET_PREFIX, 0);
lowcore_ptr[pcpu - pcpu_devices] = NULL;
vdso_free_per_cpu(pcpu->lowcore);
if (pcpu == &pcpu_devices[0])
return;
free_page(pcpu->lowcore->panic_stack-PANIC_FRAME_OFFSET);
free_pages(pcpu->lowcore->async_stack-ASYNC_FRAME_OFFSET, ASYNC_ORDER);
free_pages((unsigned long) pcpu->lowcore, LC_ORDER);
}
#endif /* CONFIG_HOTPLUG_CPU */
static void pcpu_prepare_secondary(struct pcpu *pcpu, int cpu)
{
struct _lowcore *lc = pcpu->lowcore;
if (MACHINE_HAS_TLB_LC)
cpumask_set_cpu(cpu, &init_mm.context.cpu_attach_mask);
cpumask_set_cpu(cpu, mm_cpumask(&init_mm));
atomic_inc(&init_mm.context.attach_count);
lc->cpu_nr = cpu;
lc->spinlock_lockval = arch_spin_lockval(cpu);
lc->percpu_offset = __per_cpu_offset[cpu];
lc->kernel_asce = S390_lowcore.kernel_asce;
lc->machine_flags = S390_lowcore.machine_flags;
lc->user_timer = lc->system_timer = lc->steal_timer = 0;
__ctl_store(lc->cregs_save_area, 0, 15);
save_access_regs((unsigned int *) lc->access_regs_save_area);
memcpy(lc->stfle_fac_list, S390_lowcore.stfle_fac_list,
MAX_FACILITY_BIT/8);
}
static void pcpu_attach_task(struct pcpu *pcpu, struct task_struct *tsk)
{
struct _lowcore *lc = pcpu->lowcore;
struct thread_info *ti = task_thread_info(tsk);
lc->kernel_stack = (unsigned long) task_stack_page(tsk)
+ THREAD_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs);
lc->thread_info = (unsigned long) task_thread_info(tsk);
lc->current_task = (unsigned long) tsk;
lc->user_timer = ti->user_timer;
lc->system_timer = ti->system_timer;
lc->steal_timer = 0;
}
static void pcpu_start_fn(struct pcpu *pcpu, void (*func)(void *), void *data)
{
struct _lowcore *lc = pcpu->lowcore;
lc->restart_stack = lc->kernel_stack;
lc->restart_fn = (unsigned long) func;
lc->restart_data = (unsigned long) data;
lc->restart_source = -1UL;
pcpu_sigp_retry(pcpu, SIGP_RESTART, 0);
}
/*
* Call function via PSW restart on pcpu and stop the current cpu.
*/
static void pcpu_delegate(struct pcpu *pcpu, void (*func)(void *),
void *data, unsigned long stack)
{
struct _lowcore *lc = lowcore_ptr[pcpu - pcpu_devices];
unsigned long source_cpu = stap();
__load_psw_mask(PSW_KERNEL_BITS);
if (pcpu->address == source_cpu)
func(data); /* should not return */
/* Stop target cpu (if func returns this stops the current cpu). */
pcpu_sigp_retry(pcpu, SIGP_STOP, 0);
/* Restart func on the target cpu and stop the current cpu. */
mem_assign_absolute(lc->restart_stack, stack);
mem_assign_absolute(lc->restart_fn, (unsigned long) func);
mem_assign_absolute(lc->restart_data, (unsigned long) data);
mem_assign_absolute(lc->restart_source, source_cpu);
asm volatile(
"0: sigp 0,%0,%2 # sigp restart to target cpu\n"
" brc 2,0b # busy, try again\n"
"1: sigp 0,%1,%3 # sigp stop to current cpu\n"
" brc 2,1b # busy, try again\n"
: : "d" (pcpu->address), "d" (source_cpu),
"K" (SIGP_RESTART), "K" (SIGP_STOP)
: "0", "1", "cc");
for (;;) ;
}
/*
* Enable additional logical cpus for multi-threading.
*/
static int pcpu_set_smt(unsigned int mtid)
{
register unsigned long reg1 asm ("1") = (unsigned long) mtid;
int cc;
if (smp_cpu_mtid == mtid)
return 0;
asm volatile(
" sigp %1,0,%2 # sigp set multi-threading\n"
" ipm %0\n"
" srl %0,28\n"
: "=d" (cc) : "d" (reg1), "K" (SIGP_SET_MULTI_THREADING)
: "cc");
if (cc == 0) {
smp_cpu_mtid = mtid;
smp_cpu_mt_shift = 0;
while (smp_cpu_mtid >= (1U << smp_cpu_mt_shift))
smp_cpu_mt_shift++;
pcpu_devices[0].address = stap();
}
return cc;
}
/*
* Call function on an online CPU.
*/
void smp_call_online_cpu(void (*func)(void *), void *data)
{
struct pcpu *pcpu;
/* Use the current cpu if it is online. */
pcpu = pcpu_find_address(cpu_online_mask, stap());
if (!pcpu)
/* Use the first online cpu. */
pcpu = pcpu_devices + cpumask_first(cpu_online_mask);
pcpu_delegate(pcpu, func, data, (unsigned long) restart_stack);
}
/*
* Call function on the ipl CPU.
*/
void smp_call_ipl_cpu(void (*func)(void *), void *data)
{
pcpu_delegate(&pcpu_devices[0], func, data,
pcpu_devices->lowcore->panic_stack -
PANIC_FRAME_OFFSET + PAGE_SIZE);
}
int smp_find_processor_id(u16 address)
{
int cpu;
for_each_present_cpu(cpu)
if (pcpu_devices[cpu].address == address)
return cpu;
return -1;
}
int smp_vcpu_scheduled(int cpu)
{
return pcpu_running(pcpu_devices + cpu);
}
void smp_yield_cpu(int cpu)
{
if (MACHINE_HAS_DIAG9C)
asm volatile("diag %0,0,0x9c"
: : "d" (pcpu_devices[cpu].address));
else if (MACHINE_HAS_DIAG44)
asm volatile("diag 0,0,0x44");
}
/*
* Send cpus emergency shutdown signal. This gives the cpus the
* opportunity to complete outstanding interrupts.
*/
static void smp_emergency_stop(cpumask_t *cpumask)
{
u64 end;
int cpu;
end = get_tod_clock() + (1000000UL << 12);
for_each_cpu(cpu, cpumask) {
struct pcpu *pcpu = pcpu_devices + cpu;
set_bit(ec_stop_cpu, &pcpu->ec_mask);
while (__pcpu_sigp(pcpu->address, SIGP_EMERGENCY_SIGNAL,
0, NULL) == SIGP_CC_BUSY &&
get_tod_clock() < end)
cpu_relax();
}
while (get_tod_clock() < end) {
for_each_cpu(cpu, cpumask)
if (pcpu_stopped(pcpu_devices + cpu))
cpumask_clear_cpu(cpu, cpumask);
if (cpumask_empty(cpumask))
break;
cpu_relax();
}
}
/*
* Stop all cpus but the current one.
*/
void smp_send_stop(void)
{
cpumask_t cpumask;
int cpu;
/* Disable all interrupts/machine checks */
__load_psw_mask(PSW_KERNEL_BITS | PSW_MASK_DAT);
trace_hardirqs_off();
debug_set_critical();
cpumask_copy(&cpumask, cpu_online_mask);
cpumask_clear_cpu(smp_processor_id(), &cpumask);
if (oops_in_progress)
smp_emergency_stop(&cpumask);
/* stop all processors */
for_each_cpu(cpu, &cpumask) {
struct pcpu *pcpu = pcpu_devices + cpu;
pcpu_sigp_retry(pcpu, SIGP_STOP, 0);
while (!pcpu_stopped(pcpu))
cpu_relax();
}
}
/*
* This is the main routine where commands issued by other
* cpus are handled.
*/
static void smp_handle_ext_call(void)
{
unsigned long bits;
/* handle bit signal external calls */
bits = xchg(&pcpu_devices[smp_processor_id()].ec_mask, 0);
if (test_bit(ec_stop_cpu, &bits))
smp_stop_cpu();
if (test_bit(ec_schedule, &bits))
scheduler_ipi();
if (test_bit(ec_call_function_single, &bits))
generic_smp_call_function_single_interrupt();
}
static void do_ext_call_interrupt(struct ext_code ext_code,
unsigned int param32, unsigned long param64)
{
inc_irq_stat(ext_code.code == 0x1202 ? IRQEXT_EXC : IRQEXT_EMS);
smp_handle_ext_call();
}
void arch_send_call_function_ipi_mask(const struct cpumask *mask)
{
int cpu;
for_each_cpu(cpu, mask)
pcpu_ec_call(pcpu_devices + cpu, ec_call_function_single);
}
void arch_send_call_function_single_ipi(int cpu)
{
pcpu_ec_call(pcpu_devices + cpu, ec_call_function_single);
}
/*
* this function sends a 'reschedule' IPI to another CPU.
* it goes straight through and wastes no time serializing
* anything. Worst case is that we lose a reschedule ...
*/
void smp_send_reschedule(int cpu)
{
pcpu_ec_call(pcpu_devices + cpu, ec_schedule);
}
/*
* parameter area for the set/clear control bit callbacks
*/
struct ec_creg_mask_parms {
unsigned long orval;
unsigned long andval;
int cr;
};
/*
* callback for setting/clearing control bits
*/
static void smp_ctl_bit_callback(void *info)
{
struct ec_creg_mask_parms *pp = info;
unsigned long cregs[16];
__ctl_store(cregs, 0, 15);
cregs[pp->cr] = (cregs[pp->cr] & pp->andval) | pp->orval;
__ctl_load(cregs, 0, 15);
}
/*
* Set a bit in a control register of all cpus
*/
void smp_ctl_set_bit(int cr, int bit)
{
struct ec_creg_mask_parms parms = { 1UL << bit, -1UL, cr };
on_each_cpu(smp_ctl_bit_callback, &parms, 1);
}
EXPORT_SYMBOL(smp_ctl_set_bit);
/*
* Clear a bit in a control register of all cpus
*/
void smp_ctl_clear_bit(int cr, int bit)
{
struct ec_creg_mask_parms parms = { 0, ~(1UL << bit), cr };
on_each_cpu(smp_ctl_bit_callback, &parms, 1);
}
EXPORT_SYMBOL(smp_ctl_clear_bit);
#ifdef CONFIG_CRASH_DUMP
static void __smp_store_cpu_state(struct save_area_ext *sa_ext, u16 address,
int is_boot_cpu)
{
void *lc = (void *)(unsigned long) store_prefix();
unsigned long vx_sa;
if (is_boot_cpu) {
/* Copy the registers of the boot CPU. */
copy_oldmem_page(1, (void *) &sa_ext->sa, sizeof(sa_ext->sa),
SAVE_AREA_BASE - PAGE_SIZE, 0);
if (MACHINE_HAS_VX)
save_vx_regs_safe(sa_ext->vx_regs);
return;
}
/* Get the registers of a non-boot cpu. */
__pcpu_sigp_relax(address, SIGP_STOP_AND_STORE_STATUS, 0, NULL);
memcpy_real(&sa_ext->sa, lc + SAVE_AREA_BASE, sizeof(sa_ext->sa));
if (!MACHINE_HAS_VX)
return;
/* Get the VX registers */
vx_sa = memblock_alloc(PAGE_SIZE, PAGE_SIZE);
if (!vx_sa)
panic("could not allocate memory for VX save area\n");
__pcpu_sigp_relax(address, SIGP_STORE_ADDITIONAL_STATUS, vx_sa, NULL);
memcpy(sa_ext->vx_regs, (void *) vx_sa, sizeof(sa_ext->vx_regs));
memblock_free(vx_sa, PAGE_SIZE);
}
int smp_store_status(int cpu)
{
unsigned long vx_sa;
struct pcpu *pcpu;
pcpu = pcpu_devices + cpu;
if (__pcpu_sigp_relax(pcpu->address, SIGP_STOP_AND_STORE_STATUS,
0, NULL) != SIGP_CC_ORDER_CODE_ACCEPTED)
return -EIO;
if (!MACHINE_HAS_VX)
return 0;
vx_sa = __pa(pcpu->lowcore->vector_save_area_addr);
__pcpu_sigp_relax(pcpu->address, SIGP_STORE_ADDITIONAL_STATUS,
vx_sa, NULL);
return 0;
}
#endif /* CONFIG_CRASH_DUMP */
/*
* Collect CPU state of the previous, crashed system.
* There are four cases:
* 1) standard zfcp dump
* condition: OLDMEM_BASE == NULL && ipl_info.type == IPL_TYPE_FCP_DUMP
* The state for all CPUs except the boot CPU needs to be collected
* with sigp stop-and-store-status. The boot CPU state is located in
* the absolute lowcore of the memory stored in the HSA. The zcore code
* will allocate the save area and copy the boot CPU state from the HSA.
* 2) stand-alone kdump for SCSI (zfcp dump with swapped memory)
* condition: OLDMEM_BASE != NULL && ipl_info.type == IPL_TYPE_FCP_DUMP
* The state for all CPUs except the boot CPU needs to be collected
* with sigp stop-and-store-status. The firmware or the boot-loader
* stored the registers of the boot CPU in the absolute lowcore in the
* memory of the old system.
* 3) kdump and the old kernel did not store the CPU state,
* or stand-alone kdump for DASD
* condition: OLDMEM_BASE != NULL && !is_kdump_kernel()
* The state for all CPUs except the boot CPU needs to be collected
* with sigp stop-and-store-status. The kexec code or the boot-loader
* stored the registers of the boot CPU in the memory of the old system.
* 4) kdump and the old kernel stored the CPU state
* condition: OLDMEM_BASE != NULL && is_kdump_kernel()
* The state of all CPUs is stored in ELF sections in the memory of the
* old system. The ELF sections are picked up by the crash_dump code
* via elfcorehdr_addr.
*/
void __init smp_save_dump_cpus(void)
{
#ifdef CONFIG_CRASH_DUMP
int addr, cpu, boot_cpu_addr, max_cpu_addr;
struct save_area_ext *sa_ext;
bool is_boot_cpu;
if (is_kdump_kernel())
/* Previous system stored the CPU states. Nothing to do. */
return;
if (!(OLDMEM_BASE || ipl_info.type == IPL_TYPE_FCP_DUMP))
/* No previous system present, normal boot. */
return;
/* Set multi-threading state to the previous system. */
pcpu_set_smt(sclp.mtid_prev);
max_cpu_addr = SCLP_MAX_CORES << sclp.mtid_prev;
for (cpu = 0, addr = 0; addr <= max_cpu_addr; addr++) {
if (__pcpu_sigp_relax(addr, SIGP_SENSE, 0, NULL) ==
SIGP_CC_NOT_OPERATIONAL)
continue;
cpu += 1;
}
dump_save_areas.areas = (void *)memblock_alloc(sizeof(void *) * cpu, 8);
dump_save_areas.count = cpu;
boot_cpu_addr = stap();
for (cpu = 0, addr = 0; addr <= max_cpu_addr; addr++) {
if (__pcpu_sigp_relax(addr, SIGP_SENSE, 0, NULL) ==
SIGP_CC_NOT_OPERATIONAL)
continue;
sa_ext = (void *) memblock_alloc(sizeof(*sa_ext), 8);
dump_save_areas.areas[cpu] = sa_ext;
if (!sa_ext)
panic("could not allocate memory for save area\n");
is_boot_cpu = (addr == boot_cpu_addr);
cpu += 1;
if (is_boot_cpu && !OLDMEM_BASE)
/* Skip boot CPU for standard zfcp dump. */
continue;
/* Get state for this CPU. */
__smp_store_cpu_state(sa_ext, addr, is_boot_cpu);
}
diag308_reset();
pcpu_set_smt(0);
#endif /* CONFIG_CRASH_DUMP */
}
void smp_cpu_set_polarization(int cpu, int val)
{
pcpu_devices[cpu].polarization = val;
}
int smp_cpu_get_polarization(int cpu)
{
return pcpu_devices[cpu].polarization;
}
static struct sclp_core_info *smp_get_core_info(void)
{
static int use_sigp_detection;
struct sclp_core_info *info;
int address;
info = kzalloc(sizeof(*info), GFP_KERNEL);
if (info && (use_sigp_detection || sclp_get_core_info(info))) {
use_sigp_detection = 1;
for (address = 0;
address < (SCLP_MAX_CORES << smp_cpu_mt_shift);
address += (1U << smp_cpu_mt_shift)) {
if (__pcpu_sigp_relax(address, SIGP_SENSE, 0, NULL) ==
SIGP_CC_NOT_OPERATIONAL)
continue;
info->core[info->configured].core_id =
address >> smp_cpu_mt_shift;
info->configured++;
}
info->combined = info->configured;
}
return info;
}
2013-06-19 05:04:52 +08:00
static int smp_add_present_cpu(int cpu);
static int __smp_rescan_cpus(struct sclp_core_info *info, int sysfs_add)
{
struct pcpu *pcpu;
cpumask_t avail;
int cpu, nr, i, j;
u16 address;
nr = 0;
cpumask_xor(&avail, cpu_possible_mask, cpu_present_mask);
cpu = cpumask_first(&avail);
for (i = 0; (i < info->combined) && (cpu < nr_cpu_ids); i++) {
if (sclp.has_core_type && info->core[i].type != boot_core_type)
continue;
address = info->core[i].core_id << smp_cpu_mt_shift;
for (j = 0; j <= smp_cpu_mtid; j++) {
if (pcpu_find_address(cpu_present_mask, address + j))
continue;
pcpu = pcpu_devices + cpu;
pcpu->address = address + j;
pcpu->state =
(cpu >= info->configured*(smp_cpu_mtid + 1)) ?
CPU_STATE_STANDBY : CPU_STATE_CONFIGURED;
smp_cpu_set_polarization(cpu, POLARIZATION_UNKNOWN);
set_cpu_present(cpu, true);
if (sysfs_add && smp_add_present_cpu(cpu) != 0)
set_cpu_present(cpu, false);
else
nr++;
cpu = cpumask_next(cpu, &avail);
if (cpu >= nr_cpu_ids)
break;
}
}
return nr;
}
static void __init smp_detect_cpus(void)
{
unsigned int cpu, mtid, c_cpus, s_cpus;
struct sclp_core_info *info;
u16 address;
/* Get CPU information */
info = smp_get_core_info();
if (!info)
panic("smp_detect_cpus failed to allocate memory\n");
/* Find boot CPU type */
if (sclp.has_core_type) {
address = stap();
for (cpu = 0; cpu < info->combined; cpu++)
if (info->core[cpu].core_id == address) {
/* The boot cpu dictates the cpu type. */
boot_core_type = info->core[cpu].type;
break;
}
if (cpu >= info->combined)
panic("Could not find boot CPU type");
}
/* Set multi-threading state for the current system */
mtid = boot_core_type ? sclp.mtid : sclp.mtid_cp;
mtid = (mtid < smp_max_threads) ? mtid : smp_max_threads - 1;
pcpu_set_smt(mtid);
/* Print number of CPUs */
c_cpus = s_cpus = 0;
for (cpu = 0; cpu < info->combined; cpu++) {
if (sclp.has_core_type &&
info->core[cpu].type != boot_core_type)
continue;
if (cpu < info->configured)
c_cpus += smp_cpu_mtid + 1;
else
s_cpus += smp_cpu_mtid + 1;
}
pr_info("%d configured CPUs, %d standby CPUs\n", c_cpus, s_cpus);
/* Add CPUs present at boot */
get_online_cpus();
__smp_rescan_cpus(info, 0);
put_online_cpus();
kfree(info);
}
/*
* Activate a secondary processor.
*/
2013-06-19 05:04:52 +08:00
static void smp_start_secondary(void *cpuvoid)
{
S390_lowcore.last_update_clock = get_tod_clock();
S390_lowcore.restart_stack = (unsigned long) restart_stack;
S390_lowcore.restart_fn = (unsigned long) do_restart;
S390_lowcore.restart_data = 0;
S390_lowcore.restart_source = -1UL;
restore_access_regs(S390_lowcore.access_regs_save_area);
__ctl_load(S390_lowcore.cregs_save_area, 0, 15);
__load_psw_mask(PSW_KERNEL_BITS | PSW_MASK_DAT);
cpu_init();
preempt_disable();
init_cpu_timer();
vtime_init();
pfault_init();
notify_cpu_starting(smp_processor_id());
set_cpu_online(smp_processor_id(), true);
inc_irq_stat(CPU_RST);
local_irq_enable();
cpu_startup_entry(CPUHP_ONLINE);
}
/* Upping and downing of CPUs */
2013-06-19 05:04:52 +08:00
int __cpu_up(unsigned int cpu, struct task_struct *tidle)
{
struct pcpu *pcpu;
int base, i, rc;
pcpu = pcpu_devices + cpu;
if (pcpu->state != CPU_STATE_CONFIGURED)
return -EIO;
base = cpu - (cpu % (smp_cpu_mtid + 1));
for (i = 0; i <= smp_cpu_mtid; i++) {
if (base + i < nr_cpu_ids)
if (cpu_online(base + i))
break;
}
/*
* If this is the first CPU of the core to get online
* do an initial CPU reset.
*/
if (i > smp_cpu_mtid &&
pcpu_sigp_retry(pcpu_devices + base, SIGP_INITIAL_CPU_RESET, 0) !=
SIGP_CC_ORDER_CODE_ACCEPTED)
return -EIO;
rc = pcpu_alloc_lowcore(pcpu, cpu);
if (rc)
return rc;
pcpu_prepare_secondary(pcpu, cpu);
pcpu_attach_task(pcpu, tidle);
pcpu_start_fn(pcpu, smp_start_secondary, NULL);
/* Wait until cpu puts itself in the online & active maps */
while (!cpu_online(cpu) || !cpu_active(cpu))
cpu_relax();
return 0;
}
static unsigned int setup_possible_cpus __initdata;
static int __init _setup_possible_cpus(char *s)
{
get_option(&s, &setup_possible_cpus);
return 0;
}
early_param("possible_cpus", _setup_possible_cpus);
#ifdef CONFIG_HOTPLUG_CPU
int __cpu_disable(void)
{
unsigned long cregs[16];
/* Handle possible pending IPIs */
smp_handle_ext_call();
set_cpu_online(smp_processor_id(), false);
/* Disable pseudo page faults on this cpu. */
pfault_fini();
/* Disable interrupt sources via control register. */
__ctl_store(cregs, 0, 15);
cregs[0] &= ~0x0000ee70UL; /* disable all external interrupts */
cregs[6] &= ~0xff000000UL; /* disable all I/O interrupts */
cregs[14] &= ~0x1f000000UL; /* disable most machine checks */
__ctl_load(cregs, 0, 15);
clear_cpu_flag(CIF_NOHZ_DELAY);
return 0;
}
void __cpu_die(unsigned int cpu)
{
struct pcpu *pcpu;
/* Wait until target cpu is down */
pcpu = pcpu_devices + cpu;
while (!pcpu_stopped(pcpu))
cpu_relax();
pcpu_free_lowcore(pcpu);
atomic_dec(&init_mm.context.attach_count);
cpumask_clear_cpu(cpu, mm_cpumask(&init_mm));
if (MACHINE_HAS_TLB_LC)
cpumask_clear_cpu(cpu, &init_mm.context.cpu_attach_mask);
}
void __noreturn cpu_die(void)
{
idle_task_exit();
pcpu_sigp_retry(pcpu_devices + smp_processor_id(), SIGP_STOP, 0);
for (;;) ;
}
#endif /* CONFIG_HOTPLUG_CPU */
void __init smp_fill_possible_mask(void)
{
unsigned int possible, sclp_max, cpu;
sclp_max = max(sclp.mtid, sclp.mtid_cp) + 1;
sclp_max = min(smp_max_threads, sclp_max);
sclp_max = sclp.max_cores * sclp_max ?: nr_cpu_ids;
possible = setup_possible_cpus ?: nr_cpu_ids;
possible = min(possible, sclp_max);
for (cpu = 0; cpu < possible && cpu < nr_cpu_ids; cpu++)
set_cpu_possible(cpu, true);
}
void __init smp_prepare_cpus(unsigned int max_cpus)
{
/* request the 0x1201 emergency signal external interrupt */
if (register_external_irq(EXT_IRQ_EMERGENCY_SIG, do_ext_call_interrupt))
panic("Couldn't request external interrupt 0x1201");
/* request the 0x1202 external call external interrupt */
if (register_external_irq(EXT_IRQ_EXTERNAL_CALL, do_ext_call_interrupt))
panic("Couldn't request external interrupt 0x1202");
smp_detect_cpus();
}
void __init smp_prepare_boot_cpu(void)
{
struct pcpu *pcpu = pcpu_devices;
pcpu->state = CPU_STATE_CONFIGURED;
pcpu->address = stap();
pcpu->lowcore = (struct _lowcore *)(unsigned long) store_prefix();
S390_lowcore.percpu_offset = __per_cpu_offset[0];
smp_cpu_set_polarization(0, POLARIZATION_UNKNOWN);
set_cpu_present(0, true);
set_cpu_online(0, true);
}
void __init smp_cpus_done(unsigned int max_cpus)
{
}
void __init smp_setup_processor_id(void)
{
S390_lowcore.cpu_nr = 0;
S390_lowcore.spinlock_lockval = arch_spin_lockval(0);
}
/*
* the frequency of the profiling timer can be changed
* by writing a multiplier value into /proc/profile.
*
* usually you want to run this on all CPUs ;)
*/
int setup_profiling_timer(unsigned int multiplier)
{
return 0;
}
#ifdef CONFIG_HOTPLUG_CPU
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
static ssize_t cpu_configure_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
ssize_t count;
mutex_lock(&smp_cpu_state_mutex);
count = sprintf(buf, "%d\n", pcpu_devices[dev->id].state);
mutex_unlock(&smp_cpu_state_mutex);
return count;
}
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
static ssize_t cpu_configure_store(struct device *dev,
struct device_attribute *attr,
const char *buf, size_t count)
{
struct pcpu *pcpu;
int cpu, val, rc, i;
char delim;
if (sscanf(buf, "%d %c", &val, &delim) != 1)
return -EINVAL;
if (val != 0 && val != 1)
return -EINVAL;
get_online_cpus();
mutex_lock(&smp_cpu_state_mutex);
rc = -EBUSY;
/* disallow configuration changes of online cpus and cpu 0 */
cpu = dev->id;
cpu -= cpu % (smp_cpu_mtid + 1);
if (cpu == 0)
goto out;
for (i = 0; i <= smp_cpu_mtid; i++)
if (cpu_online(cpu + i))
goto out;
pcpu = pcpu_devices + cpu;
rc = 0;
switch (val) {
case 0:
if (pcpu->state != CPU_STATE_CONFIGURED)
break;
rc = sclp_core_deconfigure(pcpu->address >> smp_cpu_mt_shift);
if (rc)
break;
for (i = 0; i <= smp_cpu_mtid; i++) {
if (cpu + i >= nr_cpu_ids || !cpu_present(cpu + i))
continue;
pcpu[i].state = CPU_STATE_STANDBY;
smp_cpu_set_polarization(cpu + i,
POLARIZATION_UNKNOWN);
}
topology_expect_change();
break;
case 1:
if (pcpu->state != CPU_STATE_STANDBY)
break;
rc = sclp_core_configure(pcpu->address >> smp_cpu_mt_shift);
if (rc)
break;
for (i = 0; i <= smp_cpu_mtid; i++) {
if (cpu + i >= nr_cpu_ids || !cpu_present(cpu + i))
continue;
pcpu[i].state = CPU_STATE_CONFIGURED;
smp_cpu_set_polarization(cpu + i,
POLARIZATION_UNKNOWN);
}
topology_expect_change();
break;
default:
break;
}
out:
mutex_unlock(&smp_cpu_state_mutex);
put_online_cpus();
return rc ? rc : count;
}
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
static DEVICE_ATTR(configure, 0644, cpu_configure_show, cpu_configure_store);
#endif /* CONFIG_HOTPLUG_CPU */
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
static ssize_t show_cpu_address(struct device *dev,
struct device_attribute *attr, char *buf)
{
return sprintf(buf, "%d\n", pcpu_devices[dev->id].address);
}
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
static DEVICE_ATTR(address, 0444, show_cpu_address, NULL);
static struct attribute *cpu_common_attrs[] = {
#ifdef CONFIG_HOTPLUG_CPU
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
&dev_attr_configure.attr,
#endif
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
&dev_attr_address.attr,
NULL,
};
static struct attribute_group cpu_common_attr_group = {
.attrs = cpu_common_attrs,
};
static struct attribute *cpu_online_attrs[] = {
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
&dev_attr_idle_count.attr,
&dev_attr_idle_time_us.attr,
NULL,
};
static struct attribute_group cpu_online_attr_group = {
.attrs = cpu_online_attrs,
};
2013-06-19 05:04:52 +08:00
static int smp_cpu_notify(struct notifier_block *self, unsigned long action,
void *hcpu)
{
unsigned int cpu = (unsigned int)(long)hcpu;
struct device *s = &per_cpu(cpu_device, cpu)->dev;
int err = 0;
switch (action & ~CPU_TASKS_FROZEN) {
case CPU_ONLINE:
err = sysfs_create_group(&s->kobj, &cpu_online_attr_group);
break;
case CPU_DEAD:
sysfs_remove_group(&s->kobj, &cpu_online_attr_group);
break;
}
return notifier_from_errno(err);
}
2013-06-19 05:04:52 +08:00
static int smp_add_present_cpu(int cpu)
{
struct device *s;
struct cpu *c;
int rc;
c = kzalloc(sizeof(*c), GFP_KERNEL);
if (!c)
return -ENOMEM;
per_cpu(cpu_device, cpu) = c;
s = &c->dev;
c->hotpluggable = 1;
rc = register_cpu(c, cpu);
if (rc)
goto out;
rc = sysfs_create_group(&s->kobj, &cpu_common_attr_group);
if (rc)
goto out_cpu;
if (cpu_online(cpu)) {
rc = sysfs_create_group(&s->kobj, &cpu_online_attr_group);
if (rc)
goto out_online;
}
rc = topology_cpu_init(c);
if (rc)
goto out_topology;
return 0;
out_topology:
if (cpu_online(cpu))
sysfs_remove_group(&s->kobj, &cpu_online_attr_group);
out_online:
sysfs_remove_group(&s->kobj, &cpu_common_attr_group);
out_cpu:
#ifdef CONFIG_HOTPLUG_CPU
unregister_cpu(c);
#endif
out:
return rc;
}
#ifdef CONFIG_HOTPLUG_CPU
int __ref smp_rescan_cpus(void)
{
struct sclp_core_info *info;
int nr;
info = smp_get_core_info();
if (!info)
return -ENOMEM;
get_online_cpus();
mutex_lock(&smp_cpu_state_mutex);
nr = __smp_rescan_cpus(info, 1);
mutex_unlock(&smp_cpu_state_mutex);
put_online_cpus();
kfree(info);
if (nr)
[S390] Vertical cpu management. If vertical cpu polarization is active then the hypervisor will dispatch certain cpus for a longer time than other cpus for maximum performance. For example if a guest would have three virtual cpus, each of them with a share of 33 percent, then in case of vertical cpu polarization all of the processing time would be combined to a single cpu which would run all the time, while the other two cpus would get nearly no cpu time. There are three different types of vertical cpus: high, medium and low. Low cpus hardly get any real cpu time, while high cpus get a full real cpu. Medium cpus get something in between. In order to switch between the two possible modes (default is horizontal) a 0 for horizontal polarization or a 1 for vertical polarization must be written to the dispatching sysfs attribute: /sys/devices/system/cpu/dispatching The polarization of each single cpu can be figured out by the polarization sysfs attribute of each cpu: /sys/devices/system/cpu/cpuX/polarization horizontal, vertical:high, vertical:medium, vertical:low or unknown. When switching polarization the polarization attribute may contain the value unknown until the configuration change is done and the kernel has figured out the new polarization of each cpu. Note that running a system with different types of vertical cpus may result in significant performance regressions. If possible only one type of vertical cpus should be used. All other cpus should be offlined. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
2008-04-17 13:46:13 +08:00
topology_schedule_update();
return 0;
}
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
static ssize_t __ref rescan_store(struct device *dev,
struct device_attribute *attr,
const char *buf,
size_t count)
{
int rc;
rc = smp_rescan_cpus();
return rc ? rc : count;
}
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
static DEVICE_ATTR(rescan, 0200, NULL, rescan_store);
#endif /* CONFIG_HOTPLUG_CPU */
static int __init s390_smp_init(void)
{
int cpu, rc = 0;
#ifdef CONFIG_HOTPLUG_CPU
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
rc = device_create_file(cpu_subsys.dev_root, &dev_attr_rescan);
if (rc)
return rc;
#endif
cpu_notifier_register_begin();
for_each_present_cpu(cpu) {
rc = smp_add_present_cpu(cpu);
if (rc)
goto out;
}
__hotcpu_notifier(smp_cpu_notify, 0);
out:
cpu_notifier_register_done();
return rc;
}
subsys_initcall(s390_smp_init);