linux_dsm_epyc7002/arch/x86/include/asm/apic.h

671 lines
16 KiB
C
Raw Normal View History

#ifndef _ASM_X86_APIC_H
#define _ASM_X86_APIC_H
#include <linux/cpumask.h>
#include <linux/pm.h>
x86: APIC: remove apic_write_around(); use alternatives Use alternatives to select the workaround for the 11AP Pentium erratum for the affected steppings on the fly rather than build time. Remove the X86_GOOD_APIC configuration option and replace all the calls to apic_write_around() with plain apic_write(), protecting accesses to the ESR as appropriate due to the 3AP Pentium erratum. Remove apic_read_around() and all its invocations altogether as not needed. Remove apic_write_atomic() and all its implementing backends. The use of ASM_OUTPUT2() is not strictly needed for input constraints, but I have used it for readability's sake. I had the feeling no one else was brave enough to do it, so I went ahead and here it is. Verified by checking the generated assembly and tested with both a 32-bit and a 64-bit configuration, also with the 11AP "feature" forced on and verified with gdb on /proc/kcore to work as expected (as an 11AP machines are quite hard to get hands on these days). Some script complained about the use of "volatile", but apic_write() needs it for the same reason and is effectively a replacement for writel(), so I have disregarded it. I am not sure what the policy wrt defconfig files is, they are generated and there is risk of a conflict resulting from an unrelated change, so I have left changes to them out. The option will get removed from them at the next run. Some testing with machines other than mine will be needed to avoid some stupid mistake, but despite its volume, the change is not really that intrusive, so I am fairly confident that because it works for me, it will everywhere. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-07-17 01:15:30 +07:00
#include <asm/alternative.h>
#include <asm/cpufeature.h>
#include <asm/apicdef.h>
#include <linux/atomic.h>
#include <asm/fixmap.h>
#include <asm/mpspec.h>
#include <asm/msr.h>
x86, trace: Introduce entering/exiting_irq() When implementing tracepoints in interrupt handers, if the tracepoints are simply added in the performance sensitive path of interrupt handers, it may cause potential performance problem due to the time penalty. To solve the problem, an idea is to prepare non-trace/trace irq handers and switch their IDTs at the enabling/disabling time. So, let's introduce entering_irq()/exiting_irq() for pre/post- processing of each irq handler. A way to use them is as follows. Non-trace irq handler: smp_irq_handler() { entering_irq(); /* pre-processing of this handler */ __smp_irq_handler(); /* * common logic between non-trace and trace handlers * in a vector. */ exiting_irq(); /* post-processing of this handler */ } Trace irq_handler: smp_trace_irq_handler() { entering_irq(); /* pre-processing of this handler */ trace_irq_entry(); /* tracepoint for irq entry */ __smp_irq_handler(); /* * common logic between non-trace and trace handlers * in a vector. */ trace_irq_exit(); /* tracepoint for irq exit */ exiting_irq(); /* post-processing of this handler */ } If tracepoints can place outside entering_irq()/exiting_irq() as follows, it looks cleaner. smp_trace_irq_handler() { trace_irq_entry(); smp_irq_handler(); trace_irq_exit(); } But it doesn't work. The problem is with irq_enter/exit() being called. They must be called before trace_irq_enter/exit(), because of the rcu_irq_enter() must be called before any tracepoints are used, as tracepoints use rcu to synchronize. As a possible alternative, we may be able to call irq_enter() first as follows if irq_enter() can nest. smp_trace_irq_hander() { irq_entry(); trace_irq_entry(); smp_irq_handler(); trace_irq_exit(); irq_exit(); } But it doesn't work, either. If irq_enter() is nested, it may have a time penalty because it has to check if it was already called or not. The time penalty is not desired in performance sensitive paths even if it is tiny. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Link: http://lkml.kernel.org/r/51C3238D.9040706@hds.com Signed-off-by: H. Peter Anvin <hpa@linux.intel.com> Cc: Steven Rostedt <rostedt@goodmis.org>
2013-06-20 22:45:17 +07:00
#include <asm/idle.h>
#define ARCH_APICTIMER_STOPS_ON_C3 1
/*
* Debugging macros
*/
#define APIC_QUIET 0
#define APIC_VERBOSE 1
#define APIC_DEBUG 2
x86/apic: Introduce apic_extnmi command line parameter This patch introduces a command line parameter apic_extnmi: apic_extnmi=( bsp|all|none ) The default value is "bsp" and this is the current behavior: only the Boot-Strapping Processor receives an external NMI. "all" allows external NMIs to be broadcast to all CPUs. This would raise the success rate of panic on NMI when BSP hangs in NMI context or the external NMI is swallowed by other NMI handlers on the BSP. If you specify "none", no CPUs receive external NMIs. This is useful for the dump capture kernel so that it cannot be shot down by accidentally pressing the external NMI button (on platforms which have it) while saving a crash dump. Signed-off-by: Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@kernel.org> Cc: Bandan Das <bsd@redhat.com> Cc: Baoquan He <bhe@redhat.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiang Liu <jiang.liu@linux.intel.com> Cc: Joerg Roedel <joro@8bytes.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: kexec@lists.infradead.org Cc: linux-doc@vger.kernel.org Cc: "Maciej W. Rozycki" <macro@linux-mips.org> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Viresh Kumar <viresh.kumar@linaro.org> Cc: Vivek Goyal <vgoyal@redhat.com> Cc: x86-ml <x86@kernel.org> Link: http://lkml.kernel.org/r/20151210014632.25437.43778.stgit@softrs Signed-off-by: Borislav Petkov <bp@suse.de> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2015-12-14 17:19:12 +07:00
/* Macros for apic_extnmi which controls external NMI masking */
#define APIC_EXTNMI_BSP 0 /* Default */
#define APIC_EXTNMI_ALL 1
#define APIC_EXTNMI_NONE 2
/*
* Define the default level of output to be very little
* This can be turned up by using apic=verbose for more
* information and apic=debug for _lots_ of information.
* apic_verbosity is defined in apic.c
*/
#define apic_printk(v, s, a...) do { \
if ((v) <= apic_verbosity) \
printk(s, ##a); \
} while (0)
#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86_32)
extern void generic_apic_probe(void);
#else
static inline void generic_apic_probe(void)
{
}
#endif
#ifdef CONFIG_X86_LOCAL_APIC
extern unsigned int apic_verbosity;
extern int local_apic_timer_c2_ok;
extern int disable_apic;
extern unsigned int lapic_timer_frequency;
#ifdef CONFIG_SMP
extern void __inquire_remote_apic(int apicid);
#else /* CONFIG_SMP */
static inline void __inquire_remote_apic(int apicid)
{
}
#endif /* CONFIG_SMP */
static inline void default_inquire_remote_apic(int apicid)
{
if (apic_verbosity >= APIC_DEBUG)
__inquire_remote_apic(apicid);
}
/*
* With 82489DX we can't rely on apic feature bit
* retrieved via cpuid but still have to deal with
* such an apic chip so we assume that SMP configuration
* is found from MP table (64bit case uses ACPI mostly
* which set smp presence flag as well so we are safe
* to use this helper too).
*/
static inline bool apic_from_smp_config(void)
{
return smp_found_config && !disable_apic;
}
/*
* Basic functions accessing APICs.
*/
#ifdef CONFIG_PARAVIRT
#include <asm/paravirt.h>
#endif
extern int setup_profiling_timer(unsigned int);
static inline void native_apic_mem_write(u32 reg, u32 v)
{
x86: APIC: remove apic_write_around(); use alternatives Use alternatives to select the workaround for the 11AP Pentium erratum for the affected steppings on the fly rather than build time. Remove the X86_GOOD_APIC configuration option and replace all the calls to apic_write_around() with plain apic_write(), protecting accesses to the ESR as appropriate due to the 3AP Pentium erratum. Remove apic_read_around() and all its invocations altogether as not needed. Remove apic_write_atomic() and all its implementing backends. The use of ASM_OUTPUT2() is not strictly needed for input constraints, but I have used it for readability's sake. I had the feeling no one else was brave enough to do it, so I went ahead and here it is. Verified by checking the generated assembly and tested with both a 32-bit and a 64-bit configuration, also with the 11AP "feature" forced on and verified with gdb on /proc/kcore to work as expected (as an 11AP machines are quite hard to get hands on these days). Some script complained about the use of "volatile", but apic_write() needs it for the same reason and is effectively a replacement for writel(), so I have disregarded it. I am not sure what the policy wrt defconfig files is, they are generated and there is risk of a conflict resulting from an unrelated change, so I have left changes to them out. The option will get removed from them at the next run. Some testing with machines other than mine will be needed to avoid some stupid mistake, but despite its volume, the change is not really that intrusive, so I am fairly confident that because it works for me, it will everywhere. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-07-17 01:15:30 +07:00
volatile u32 *addr = (volatile u32 *)(APIC_BASE + reg);
alternative_io("movl %0, %P1", "xchgl %0, %P1", X86_BUG_11AP,
x86: APIC: remove apic_write_around(); use alternatives Use alternatives to select the workaround for the 11AP Pentium erratum for the affected steppings on the fly rather than build time. Remove the X86_GOOD_APIC configuration option and replace all the calls to apic_write_around() with plain apic_write(), protecting accesses to the ESR as appropriate due to the 3AP Pentium erratum. Remove apic_read_around() and all its invocations altogether as not needed. Remove apic_write_atomic() and all its implementing backends. The use of ASM_OUTPUT2() is not strictly needed for input constraints, but I have used it for readability's sake. I had the feeling no one else was brave enough to do it, so I went ahead and here it is. Verified by checking the generated assembly and tested with both a 32-bit and a 64-bit configuration, also with the 11AP "feature" forced on and verified with gdb on /proc/kcore to work as expected (as an 11AP machines are quite hard to get hands on these days). Some script complained about the use of "volatile", but apic_write() needs it for the same reason and is effectively a replacement for writel(), so I have disregarded it. I am not sure what the policy wrt defconfig files is, they are generated and there is risk of a conflict resulting from an unrelated change, so I have left changes to them out. The option will get removed from them at the next run. Some testing with machines other than mine will be needed to avoid some stupid mistake, but despite its volume, the change is not really that intrusive, so I am fairly confident that because it works for me, it will everywhere. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-07-17 01:15:30 +07:00
ASM_OUTPUT2("=r" (v), "=m" (*addr)),
ASM_OUTPUT2("0" (v), "m" (*addr)));
}
static inline u32 native_apic_mem_read(u32 reg)
{
return *((volatile u32 *)(APIC_BASE + reg));
}
extern void native_apic_wait_icr_idle(void);
extern u32 native_safe_apic_wait_icr_idle(void);
extern void native_apic_icr_write(u32 low, u32 id);
extern u64 native_apic_icr_read(void);
static inline bool apic_is_x2apic_enabled(void)
{
u64 msr;
if (rdmsrl_safe(MSR_IA32_APICBASE, &msr))
return false;
return msr & X2APIC_ENABLE;
}
extern void enable_IR_x2apic(void);
extern int get_physical_broadcast(void);
extern int lapic_get_maxlvt(void);
extern void clear_local_APIC(void);
extern void disconnect_bsp_APIC(int virt_wire_setup);
extern void disable_local_APIC(void);
extern void lapic_shutdown(void);
extern void sync_Arb_IDs(void);
extern void init_bsp_APIC(void);
extern void setup_local_APIC(void);
extern void init_apic_mappings(void);
void register_lapic_address(unsigned long address);
extern void setup_boot_APIC_clock(void);
extern void setup_secondary_APIC_clock(void);
extern int APIC_init_uniprocessor(void);
#ifdef CONFIG_X86_64
static inline int apic_force_enable(unsigned long addr)
{
return -1;
}
#else
extern int apic_force_enable(unsigned long addr);
#endif
extern int apic_bsp_setup(bool upmode);
extern void apic_ap_setup(void);
/*
* On 32bit this is mach-xxx local
*/
#ifdef CONFIG_X86_64
extern int apic_is_clustered_box(void);
#else
static inline int apic_is_clustered_box(void)
{
return 0;
}
#endif
extern int setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask);
#else /* !CONFIG_X86_LOCAL_APIC */
static inline void lapic_shutdown(void) { }
#define local_apic_timer_c2_ok 1
static inline void init_apic_mappings(void) { }
static inline void disable_local_APIC(void) { }
# define setup_boot_APIC_clock x86_init_noop
# define setup_secondary_APIC_clock x86_init_noop
#endif /* !CONFIG_X86_LOCAL_APIC */
#ifdef CONFIG_X86_X2APIC
/*
* Make previous memory operations globally visible before
* sending the IPI through x2apic wrmsr. We need a serializing instruction or
* mfence for this.
*/
static inline void x2apic_wrmsr_fence(void)
{
asm volatile("mfence" : : : "memory");
}
static inline void native_apic_msr_write(u32 reg, u32 v)
{
if (reg == APIC_DFR || reg == APIC_ID || reg == APIC_LDR ||
reg == APIC_LVR)
return;
wrmsr(APIC_BASE_MSR + (reg >> 4), v, 0);
}
static inline void native_apic_msr_eoi_write(u32 reg, u32 v)
{
wrmsr(APIC_BASE_MSR + (APIC_EOI >> 4), APIC_EOI_ACK, 0);
}
static inline u32 native_apic_msr_read(u32 reg)
{
u64 msr;
if (reg == APIC_DFR)
return -1;
rdmsrl(APIC_BASE_MSR + (reg >> 4), msr);
return (u32)msr;
}
static inline void native_x2apic_wait_icr_idle(void)
{
/* no need to wait for icr idle in x2apic */
return;
}
static inline u32 native_safe_x2apic_wait_icr_idle(void)
{
/* no need to wait for icr idle in x2apic */
return 0;
}
static inline void native_x2apic_icr_write(u32 low, u32 id)
{
wrmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), ((__u64) id) << 32 | low);
}
static inline u64 native_x2apic_icr_read(void)
{
unsigned long val;
rdmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), val);
return val;
}
extern int x2apic_mode;
extern int x2apic_phys;
extern void __init check_x2apic(void);
extern void x2apic_setup(void);
static inline int x2apic_enabled(void)
{
return boot_cpu_has(X86_FEATURE_X2APIC) && apic_is_x2apic_enabled();
}
#define x2apic_supported() (boot_cpu_has(X86_FEATURE_X2APIC))
#else /* !CONFIG_X86_X2APIC */
static inline void check_x2apic(void) { }
static inline void x2apic_setup(void) { }
static inline int x2apic_enabled(void) { return 0; }
#define x2apic_mode (0)
#define x2apic_supported() (0)
#endif /* !CONFIG_X86_X2APIC */
#ifdef CONFIG_X86_64
#define SET_APIC_ID(x) (apic->set_apic_id(x))
#else
#endif
/*
* Copyright 2004 James Cleverdon, IBM.
* Subject to the GNU Public License, v.2
*
* Generic APIC sub-arch data struct.
*
* Hacked for x86-64 by James Cleverdon from i386 architecture code by
* Martin Bligh, Andi Kleen, James Bottomley, John Stultz, and
* James Cleverdon.
*/
struct apic {
char *name;
int (*probe)(void);
int (*acpi_madt_oem_check)(char *oem_id, char *oem_table_id);
int (*apic_id_valid)(int apicid);
int (*apic_id_registered)(void);
u32 irq_delivery_mode;
u32 irq_dest_mode;
const struct cpumask *(*target_cpus)(void);
int disable_esr;
int dest_logical;
unsigned long (*check_apicid_used)(physid_mask_t *map, int apicid);
void (*vector_allocation_domain)(int cpu, struct cpumask *retmask,
const struct cpumask *mask);
void (*init_apic_ldr)(void);
void (*ioapic_phys_id_map)(physid_mask_t *phys_map, physid_mask_t *retmap);
void (*setup_apic_routing)(void);
int (*cpu_present_to_apicid)(int mps_cpu);
void (*apicid_to_cpu_present)(int phys_apicid, physid_mask_t *retmap);
int (*check_phys_apicid_present)(int phys_apicid);
int (*phys_pkg_id)(int cpuid_apic, int index_msb);
unsigned int (*get_apic_id)(unsigned long x);
unsigned long (*set_apic_id)(unsigned int id);
unsigned long apic_id_mask;
x86/apic: Make cpu_mask_to_apicid() operations return error code Current cpu_mask_to_apicid() and cpu_mask_to_apicid_and() implementations have few shortcomings: 1. A value returned by cpu_mask_to_apicid() is written to hardware registers unconditionally. Should BAD_APICID get ever returned it will be written to a hardware too. But the value of BAD_APICID is not universal across all hardware in all modes and might cause unexpected results, i.e. interrupts might get routed to CPUs that are not configured to receive it. 2. Because the value of BAD_APICID is not universal it is counter- intuitive to return it for a hardware where it does not make sense (i.e. x2apic). 3. cpu_mask_to_apicid_and() operation is thought as an complement to cpu_mask_to_apicid() that only applies a AND mask on top of a cpumask being passed. Yet, as consequence of 18374d8 commit the two operations are inconsistent in that of: cpu_mask_to_apicid() should not get a offline CPU with the cpumask cpu_mask_to_apicid_and() should not fail and return BAD_APICID These limitations are impossible to realize just from looking at the operations prototypes. Most of these shortcomings are resolved by returning a error code instead of BAD_APICID. As the result, faults are reported back early rather than possibilities to cause a unexpected behaviour exist (in case of [1]). The only exception is setup_timer_IRQ0_pin() routine. Although obviously controversial to this fix, its existing behaviour is preserved to not break the fragile check_timer() and would better addressed in a separate fix. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Acked-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Yinghai Lu <yinghai@kernel.org> Link: http://lkml.kernel.org/r/20120607131559.GF4759@dhcp-26-207.brq.redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
2012-06-07 20:15:59 +07:00
int (*cpu_mask_to_apicid_and)(const struct cpumask *cpumask,
const struct cpumask *andmask,
unsigned int *apicid);
/* ipi */
void (*send_IPI)(int cpu, int vector);
void (*send_IPI_mask)(const struct cpumask *mask, int vector);
void (*send_IPI_mask_allbutself)(const struct cpumask *mask,
int vector);
void (*send_IPI_allbutself)(int vector);
void (*send_IPI_all)(int vector);
void (*send_IPI_self)(int vector);
/* wakeup_secondary_cpu */
int (*wakeup_secondary_cpu)(int apicid, unsigned long start_eip);
void (*inquire_remote_apic)(int apicid);
/* apic ops */
u32 (*read)(u32 reg);
void (*write)(u32 reg, u32 v);
/*
* ->eoi_write() has the same signature as ->write().
*
* Drivers can support both ->eoi_write() and ->write() by passing the same
* callback value. Kernel can override ->eoi_write() and fall back
* on write for EOI.
*/
void (*eoi_write)(u32 reg, u32 v);
u64 (*icr_read)(void);
void (*icr_write)(u32 low, u32 high);
void (*wait_icr_idle)(void);
u32 (*safe_wait_icr_idle)(void);
#ifdef CONFIG_X86_32
/*
* Called very early during boot from get_smp_config(). It should
* return the logical apicid. x86_[bios]_cpu_to_apicid is
* initialized before this function is called.
*
* If logical apicid can't be determined that early, the function
* may return BAD_APICID. Logical apicid will be configured after
* init_apic_ldr() while bringing up CPUs. Note that NUMA affinity
* won't be applied properly during early boot in this case.
*/
int (*x86_32_early_logical_apicid)(int cpu);
#endif
};
/*
* Pointer to the local APIC driver in use on this system (there's
* always just one such driver in use - the kernel decides via an
* early probing process which one it picks - and then sticks to it):
*/
extern struct apic *apic;
/*
* APIC drivers are probed based on how they are listed in the .apicdrivers
* section. So the order is important and enforced by the ordering
* of different apic driver files in the Makefile.
*
* For the files having two apic drivers, we use apic_drivers()
* to enforce the order with in them.
*/
#define apic_driver(sym) \
static const struct apic *__apicdrivers_##sym __used \
__aligned(sizeof(struct apic *)) \
__section(.apicdrivers) = { &sym }
#define apic_drivers(sym1, sym2) \
static struct apic *__apicdrivers_##sym1##sym2[2] __used \
__aligned(sizeof(struct apic *)) \
__section(.apicdrivers) = { &sym1, &sym2 }
extern struct apic *__apicdrivers[], *__apicdrivers_end[];
/*
* APIC functionality to boot other CPUs - only used on SMP:
*/
#ifdef CONFIG_SMP
extern int wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip);
#endif
#ifdef CONFIG_X86_LOCAL_APIC
static inline u32 apic_read(u32 reg)
{
return apic->read(reg);
}
static inline void apic_write(u32 reg, u32 val)
{
apic->write(reg, val);
}
static inline void apic_eoi(void)
{
apic->eoi_write(APIC_EOI, APIC_EOI_ACK);
}
static inline u64 apic_icr_read(void)
{
return apic->icr_read();
}
static inline void apic_icr_write(u32 low, u32 high)
{
apic->icr_write(low, high);
}
static inline void apic_wait_icr_idle(void)
{
apic->wait_icr_idle();
}
static inline u32 safe_apic_wait_icr_idle(void)
{
return apic->safe_wait_icr_idle();
}
extern void __init apic_set_eoi_write(void (*eoi_write)(u32 reg, u32 v));
#else /* CONFIG_X86_LOCAL_APIC */
static inline u32 apic_read(u32 reg) { return 0; }
static inline void apic_write(u32 reg, u32 val) { }
static inline void apic_eoi(void) { }
static inline u64 apic_icr_read(void) { return 0; }
static inline void apic_icr_write(u32 low, u32 high) { }
static inline void apic_wait_icr_idle(void) { }
static inline u32 safe_apic_wait_icr_idle(void) { return 0; }
static inline void apic_set_eoi_write(void (*eoi_write)(u32 reg, u32 v)) {}
#endif /* CONFIG_X86_LOCAL_APIC */
static inline void ack_APIC_irq(void)
{
/*
* ack_APIC_irq() actually gets compiled as a single instruction
* ... yummie.
*/
apic_eoi();
}
static inline unsigned default_get_apic_id(unsigned long x)
{
unsigned int ver = GET_APIC_VERSION(apic_read(APIC_LVR));
if (APIC_XAPIC(ver) || boot_cpu_has(X86_FEATURE_EXTD_APICID))
return (x >> 24) & 0xFF;
else
return (x >> 24) & 0x0F;
}
/*
* Warm reset vector position:
*/
#define TRAMPOLINE_PHYS_LOW 0x467
#define TRAMPOLINE_PHYS_HIGH 0x469
#ifdef CONFIG_X86_64
extern void apic_send_IPI_self(int vector);
DECLARE_PER_CPU(int, x2apic_extra_bits);
extern int default_cpu_present_to_apicid(int mps_cpu);
extern int default_check_phys_apicid_present(int phys_apicid);
#endif
extern void generic_bigsmp_probe(void);
#ifdef CONFIG_X86_LOCAL_APIC
#include <asm/smp.h>
#define APIC_DFR_VALUE (APIC_DFR_FLAT)
static inline const struct cpumask *default_target_cpus(void)
{
#ifdef CONFIG_SMP
return cpu_online_mask;
#else
return cpumask_of(0);
#endif
}
static inline const struct cpumask *online_target_cpus(void)
{
return cpu_online_mask;
}
2012-06-11 16:56:52 +07:00
DECLARE_EARLY_PER_CPU_READ_MOSTLY(u16, x86_bios_cpu_apicid);
static inline unsigned int read_apic_id(void)
{
unsigned int reg;
reg = apic_read(APIC_ID);
return apic->get_apic_id(reg);
}
static inline int default_apic_id_valid(int apicid)
{
return (apicid < 255);
}
x86, mpparse: Simplify arch/x86/include/asm/mpspec.h Simplify arch/x86/include/asm/mpspec.h by 1) Change max_physical_apicid to static as it's only used in apic.c. 2) Kill declaration of mpc_default_type, it's never defined. 3) Delete default_acpi_madt_oem_check(), it has already been declared in apic.h. 4) Make default_acpi_madt_oem_check() depends on CONFIG_X86_LOCAL_APIC instead of CONFIG_X86_64 to support i386. 5) Change mp_override_legacy_irq(), mp_config_acpi_legacy_irqs() and mp_register_gsi() as static because they are only used in acpi/boot.c. Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Acked-by: David Rientjes <rientjes@google.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Joerg Roedel <joro@8bytes.org> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: H. Peter Anvin <hpa@linux.intel.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Grant Likely <grant.likely@linaro.org> Cc: Rafael J. Wysocki <rjw@rjwysocki.net> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: Randy Dunlap <rdunlap@infradead.org> Cc: Yinghai Lu <yinghai@kernel.org> Cc: Len Brown <len.brown@intel.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: Seiji Aguchi <seiji.aguchi@hds.com> Cc: HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com> Cc: Paul Gortmaker <paul.gortmaker@windriver.com> Cc: Richard Weinberger <richard@nod.at> Cc: Andi Kleen <ak@linux.intel.com> Link: http://lkml.kernel.org/r/1402302011-23642-4-git-send-email-jiang.liu@linux.intel.com Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2014-06-09 15:19:32 +07:00
extern int default_acpi_madt_oem_check(char *, char *);
extern void default_setup_apic_routing(void);
extern struct apic apic_noop;
#ifdef CONFIG_X86_32
static inline int noop_x86_32_early_logical_apicid(int cpu)
{
return BAD_APICID;
}
/*
* Set up the logical destination ID.
*
* Intel recommends to set DFR, LDR and TPR before enabling
* an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel
* document number 292116). So here it goes...
*/
extern void default_init_apic_ldr(void);
static inline int default_apic_id_registered(void)
{
return physid_isset(read_apic_id(), phys_cpu_present_map);
}
static inline int default_phys_pkg_id(int cpuid_apic, int index_msb)
{
return cpuid_apic >> index_msb;
}
#endif
x86/apic: Make cpu_mask_to_apicid() operations return error code Current cpu_mask_to_apicid() and cpu_mask_to_apicid_and() implementations have few shortcomings: 1. A value returned by cpu_mask_to_apicid() is written to hardware registers unconditionally. Should BAD_APICID get ever returned it will be written to a hardware too. But the value of BAD_APICID is not universal across all hardware in all modes and might cause unexpected results, i.e. interrupts might get routed to CPUs that are not configured to receive it. 2. Because the value of BAD_APICID is not universal it is counter- intuitive to return it for a hardware where it does not make sense (i.e. x2apic). 3. cpu_mask_to_apicid_and() operation is thought as an complement to cpu_mask_to_apicid() that only applies a AND mask on top of a cpumask being passed. Yet, as consequence of 18374d8 commit the two operations are inconsistent in that of: cpu_mask_to_apicid() should not get a offline CPU with the cpumask cpu_mask_to_apicid_and() should not fail and return BAD_APICID These limitations are impossible to realize just from looking at the operations prototypes. Most of these shortcomings are resolved by returning a error code instead of BAD_APICID. As the result, faults are reported back early rather than possibilities to cause a unexpected behaviour exist (in case of [1]). The only exception is setup_timer_IRQ0_pin() routine. Although obviously controversial to this fix, its existing behaviour is preserved to not break the fragile check_timer() and would better addressed in a separate fix. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Acked-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Yinghai Lu <yinghai@kernel.org> Link: http://lkml.kernel.org/r/20120607131559.GF4759@dhcp-26-207.brq.redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
2012-06-07 20:15:59 +07:00
static inline int
flat_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
const struct cpumask *andmask,
unsigned int *apicid)
{
unsigned long cpu_mask = cpumask_bits(cpumask)[0] &
cpumask_bits(andmask)[0] &
cpumask_bits(cpu_online_mask)[0] &
APIC_ALL_CPUS;
x86/apic: Make cpu_mask_to_apicid() operations return error code Current cpu_mask_to_apicid() and cpu_mask_to_apicid_and() implementations have few shortcomings: 1. A value returned by cpu_mask_to_apicid() is written to hardware registers unconditionally. Should BAD_APICID get ever returned it will be written to a hardware too. But the value of BAD_APICID is not universal across all hardware in all modes and might cause unexpected results, i.e. interrupts might get routed to CPUs that are not configured to receive it. 2. Because the value of BAD_APICID is not universal it is counter- intuitive to return it for a hardware where it does not make sense (i.e. x2apic). 3. cpu_mask_to_apicid_and() operation is thought as an complement to cpu_mask_to_apicid() that only applies a AND mask on top of a cpumask being passed. Yet, as consequence of 18374d8 commit the two operations are inconsistent in that of: cpu_mask_to_apicid() should not get a offline CPU with the cpumask cpu_mask_to_apicid_and() should not fail and return BAD_APICID These limitations are impossible to realize just from looking at the operations prototypes. Most of these shortcomings are resolved by returning a error code instead of BAD_APICID. As the result, faults are reported back early rather than possibilities to cause a unexpected behaviour exist (in case of [1]). The only exception is setup_timer_IRQ0_pin() routine. Although obviously controversial to this fix, its existing behaviour is preserved to not break the fragile check_timer() and would better addressed in a separate fix. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Acked-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Yinghai Lu <yinghai@kernel.org> Link: http://lkml.kernel.org/r/20120607131559.GF4759@dhcp-26-207.brq.redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
2012-06-07 20:15:59 +07:00
if (likely(cpu_mask)) {
*apicid = (unsigned int)cpu_mask;
return 0;
} else {
return -EINVAL;
}
}
extern int
default_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
x86/apic: Make cpu_mask_to_apicid() operations return error code Current cpu_mask_to_apicid() and cpu_mask_to_apicid_and() implementations have few shortcomings: 1. A value returned by cpu_mask_to_apicid() is written to hardware registers unconditionally. Should BAD_APICID get ever returned it will be written to a hardware too. But the value of BAD_APICID is not universal across all hardware in all modes and might cause unexpected results, i.e. interrupts might get routed to CPUs that are not configured to receive it. 2. Because the value of BAD_APICID is not universal it is counter- intuitive to return it for a hardware where it does not make sense (i.e. x2apic). 3. cpu_mask_to_apicid_and() operation is thought as an complement to cpu_mask_to_apicid() that only applies a AND mask on top of a cpumask being passed. Yet, as consequence of 18374d8 commit the two operations are inconsistent in that of: cpu_mask_to_apicid() should not get a offline CPU with the cpumask cpu_mask_to_apicid_and() should not fail and return BAD_APICID These limitations are impossible to realize just from looking at the operations prototypes. Most of these shortcomings are resolved by returning a error code instead of BAD_APICID. As the result, faults are reported back early rather than possibilities to cause a unexpected behaviour exist (in case of [1]). The only exception is setup_timer_IRQ0_pin() routine. Although obviously controversial to this fix, its existing behaviour is preserved to not break the fragile check_timer() and would better addressed in a separate fix. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Acked-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Yinghai Lu <yinghai@kernel.org> Link: http://lkml.kernel.org/r/20120607131559.GF4759@dhcp-26-207.brq.redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
2012-06-07 20:15:59 +07:00
const struct cpumask *andmask,
unsigned int *apicid);
static inline void
flat_vector_allocation_domain(int cpu, struct cpumask *retmask,
const struct cpumask *mask)
{
/* Careful. Some cpus do not strictly honor the set of cpus
* specified in the interrupt destination when using lowest
* priority interrupt delivery mode.
*
* In particular there was a hyperthreading cpu observed to
* deliver interrupts to the wrong hyperthread when only one
* hyperthread was specified in the interrupt desitination.
*/
cpumask_clear(retmask);
cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
}
static inline void
default_vector_allocation_domain(int cpu, struct cpumask *retmask,
const struct cpumask *mask)
{
cpumask_copy(retmask, cpumask_of(cpu));
}
static inline unsigned long default_check_apicid_used(physid_mask_t *map, int apicid)
{
return physid_isset(apicid, *map);
}
static inline void default_ioapic_phys_id_map(physid_mask_t *phys_map, physid_mask_t *retmap)
{
*retmap = *phys_map;
}
static inline int __default_cpu_present_to_apicid(int mps_cpu)
{
if (mps_cpu < nr_cpu_ids && cpu_present(mps_cpu))
return (int)per_cpu(x86_bios_cpu_apicid, mps_cpu);
else
return BAD_APICID;
}
static inline int
__default_check_phys_apicid_present(int phys_apicid)
{
return physid_isset(phys_apicid, phys_cpu_present_map);
}
#ifdef CONFIG_X86_32
static inline int default_cpu_present_to_apicid(int mps_cpu)
{
return __default_cpu_present_to_apicid(mps_cpu);
}
static inline int
default_check_phys_apicid_present(int phys_apicid)
{
return __default_check_phys_apicid_present(phys_apicid);
}
#else
extern int default_cpu_present_to_apicid(int mps_cpu);
extern int default_check_phys_apicid_present(int phys_apicid);
#endif
#endif /* CONFIG_X86_LOCAL_APIC */
x86, trace: Introduce entering/exiting_irq() When implementing tracepoints in interrupt handers, if the tracepoints are simply added in the performance sensitive path of interrupt handers, it may cause potential performance problem due to the time penalty. To solve the problem, an idea is to prepare non-trace/trace irq handers and switch their IDTs at the enabling/disabling time. So, let's introduce entering_irq()/exiting_irq() for pre/post- processing of each irq handler. A way to use them is as follows. Non-trace irq handler: smp_irq_handler() { entering_irq(); /* pre-processing of this handler */ __smp_irq_handler(); /* * common logic between non-trace and trace handlers * in a vector. */ exiting_irq(); /* post-processing of this handler */ } Trace irq_handler: smp_trace_irq_handler() { entering_irq(); /* pre-processing of this handler */ trace_irq_entry(); /* tracepoint for irq entry */ __smp_irq_handler(); /* * common logic between non-trace and trace handlers * in a vector. */ trace_irq_exit(); /* tracepoint for irq exit */ exiting_irq(); /* post-processing of this handler */ } If tracepoints can place outside entering_irq()/exiting_irq() as follows, it looks cleaner. smp_trace_irq_handler() { trace_irq_entry(); smp_irq_handler(); trace_irq_exit(); } But it doesn't work. The problem is with irq_enter/exit() being called. They must be called before trace_irq_enter/exit(), because of the rcu_irq_enter() must be called before any tracepoints are used, as tracepoints use rcu to synchronize. As a possible alternative, we may be able to call irq_enter() first as follows if irq_enter() can nest. smp_trace_irq_hander() { irq_entry(); trace_irq_entry(); smp_irq_handler(); trace_irq_exit(); irq_exit(); } But it doesn't work, either. If irq_enter() is nested, it may have a time penalty because it has to check if it was already called or not. The time penalty is not desired in performance sensitive paths even if it is tiny. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Link: http://lkml.kernel.org/r/51C3238D.9040706@hds.com Signed-off-by: H. Peter Anvin <hpa@linux.intel.com> Cc: Steven Rostedt <rostedt@goodmis.org>
2013-06-20 22:45:17 +07:00
extern void irq_enter(void);
extern void irq_exit(void);
static inline void entering_irq(void)
{
irq_enter();
exit_idle();
}
static inline void entering_ack_irq(void)
{
entering_irq();
x86/apic: Fix suspicious RCU usage in smp_trace_call_function_interrupt() Since 4.4, I've been able to trigger this occasionally: =============================== [ INFO: suspicious RCU usage. ] 4.5.0-rc7-think+ #3 Not tainted Cc: Andi Kleen <ak@linux.intel.com> Link: http://lkml.kernel.org/r/20160315012054.GA17765@codemonkey.org.uk Signed-off-by: Thomas Gleixner <tglx@linutronix.de> ------------------------------- ./arch/x86/include/asm/msr-trace.h:47 suspicious rcu_dereference_check() usage! other info that might help us debug this: RCU used illegally from idle CPU! rcu_scheduler_active = 1, debug_locks = 1 RCU used illegally from extended quiescent state! no locks held by swapper/3/0. stack backtrace: CPU: 3 PID: 0 Comm: swapper/3 Not tainted 4.5.0-rc7-think+ #3 ffffffff92f821e0 1f3e5c340597d7fc ffff880468e07f10 ffffffff92560c2a ffff880462145280 0000000000000001 ffff880468e07f40 ffffffff921376a6 ffffffff93665ea0 0000cc7c876d28da 0000000000000005 ffffffff9383dd60 Call Trace: <IRQ> [<ffffffff92560c2a>] dump_stack+0x67/0x9d [<ffffffff921376a6>] lockdep_rcu_suspicious+0xe6/0x100 [<ffffffff925ae7a7>] do_trace_write_msr+0x127/0x1a0 [<ffffffff92061c83>] native_apic_msr_eoi_write+0x23/0x30 [<ffffffff92054408>] smp_trace_call_function_interrupt+0x38/0x360 [<ffffffff92d1ca60>] trace_call_function_interrupt+0x90/0xa0 <EOI> [<ffffffff92ac5124>] ? cpuidle_enter_state+0x1b4/0x520 Move the entering_irq() call before ack_APIC_irq(), because entering_irq() tells the RCU susbstems to end the extended quiescent state, so that the following trace call in ack_APIC_irq() works correctly. Suggested-by: Andi Kleen <ak@linux.intel.com> Fixes: 4787c368a9bc "x86/tracing: Add irq_enter/exit() in smp_trace_reschedule_interrupt()" Signed-off-by: Dave Jones <davej@codemonkey.org.uk> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: stable@vger.kernel.org
2016-03-15 08:20:54 +07:00
ack_APIC_irq();
x86, trace: Introduce entering/exiting_irq() When implementing tracepoints in interrupt handers, if the tracepoints are simply added in the performance sensitive path of interrupt handers, it may cause potential performance problem due to the time penalty. To solve the problem, an idea is to prepare non-trace/trace irq handers and switch their IDTs at the enabling/disabling time. So, let's introduce entering_irq()/exiting_irq() for pre/post- processing of each irq handler. A way to use them is as follows. Non-trace irq handler: smp_irq_handler() { entering_irq(); /* pre-processing of this handler */ __smp_irq_handler(); /* * common logic between non-trace and trace handlers * in a vector. */ exiting_irq(); /* post-processing of this handler */ } Trace irq_handler: smp_trace_irq_handler() { entering_irq(); /* pre-processing of this handler */ trace_irq_entry(); /* tracepoint for irq entry */ __smp_irq_handler(); /* * common logic between non-trace and trace handlers * in a vector. */ trace_irq_exit(); /* tracepoint for irq exit */ exiting_irq(); /* post-processing of this handler */ } If tracepoints can place outside entering_irq()/exiting_irq() as follows, it looks cleaner. smp_trace_irq_handler() { trace_irq_entry(); smp_irq_handler(); trace_irq_exit(); } But it doesn't work. The problem is with irq_enter/exit() being called. They must be called before trace_irq_enter/exit(), because of the rcu_irq_enter() must be called before any tracepoints are used, as tracepoints use rcu to synchronize. As a possible alternative, we may be able to call irq_enter() first as follows if irq_enter() can nest. smp_trace_irq_hander() { irq_entry(); trace_irq_entry(); smp_irq_handler(); trace_irq_exit(); irq_exit(); } But it doesn't work, either. If irq_enter() is nested, it may have a time penalty because it has to check if it was already called or not. The time penalty is not desired in performance sensitive paths even if it is tiny. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Link: http://lkml.kernel.org/r/51C3238D.9040706@hds.com Signed-off-by: H. Peter Anvin <hpa@linux.intel.com> Cc: Steven Rostedt <rostedt@goodmis.org>
2013-06-20 22:45:17 +07:00
}
static inline void ipi_entering_ack_irq(void)
{
ack_APIC_irq();
irq_enter();
}
x86, trace: Introduce entering/exiting_irq() When implementing tracepoints in interrupt handers, if the tracepoints are simply added in the performance sensitive path of interrupt handers, it may cause potential performance problem due to the time penalty. To solve the problem, an idea is to prepare non-trace/trace irq handers and switch their IDTs at the enabling/disabling time. So, let's introduce entering_irq()/exiting_irq() for pre/post- processing of each irq handler. A way to use them is as follows. Non-trace irq handler: smp_irq_handler() { entering_irq(); /* pre-processing of this handler */ __smp_irq_handler(); /* * common logic between non-trace and trace handlers * in a vector. */ exiting_irq(); /* post-processing of this handler */ } Trace irq_handler: smp_trace_irq_handler() { entering_irq(); /* pre-processing of this handler */ trace_irq_entry(); /* tracepoint for irq entry */ __smp_irq_handler(); /* * common logic between non-trace and trace handlers * in a vector. */ trace_irq_exit(); /* tracepoint for irq exit */ exiting_irq(); /* post-processing of this handler */ } If tracepoints can place outside entering_irq()/exiting_irq() as follows, it looks cleaner. smp_trace_irq_handler() { trace_irq_entry(); smp_irq_handler(); trace_irq_exit(); } But it doesn't work. The problem is with irq_enter/exit() being called. They must be called before trace_irq_enter/exit(), because of the rcu_irq_enter() must be called before any tracepoints are used, as tracepoints use rcu to synchronize. As a possible alternative, we may be able to call irq_enter() first as follows if irq_enter() can nest. smp_trace_irq_hander() { irq_entry(); trace_irq_entry(); smp_irq_handler(); trace_irq_exit(); irq_exit(); } But it doesn't work, either. If irq_enter() is nested, it may have a time penalty because it has to check if it was already called or not. The time penalty is not desired in performance sensitive paths even if it is tiny. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Link: http://lkml.kernel.org/r/51C3238D.9040706@hds.com Signed-off-by: H. Peter Anvin <hpa@linux.intel.com> Cc: Steven Rostedt <rostedt@goodmis.org>
2013-06-20 22:45:17 +07:00
static inline void exiting_irq(void)
{
irq_exit();
}
static inline void exiting_ack_irq(void)
{
irq_exit();
/* Ack only at the end to avoid potential reentry */
ack_APIC_irq();
}
x86/ioapic/kcrash: Prevent crash_kexec() from deadlocking on ioapic_lock Prevent crash_kexec() from deadlocking on ioapic_lock. When crash_kexec() is executed on a CPU, the CPU will take ioapic_lock in disable_IO_APIC(). So if the cpu gets an NMI while locking ioapic_lock, a deadlock will happen. In this patch, ioapic_lock is zapped/initialized before disable_IO_APIC(). You can reproduce this deadlock the following way: 1. Add mdelay(1000) after raw_spin_lock_irqsave() in native_ioapic_set_affinity()@arch/x86/kernel/apic/io_apic.c Although the deadlock can occur without this modification, it will increase the potential of the deadlock problem. 2. Build and install the kernel 3. Set up the OS which will run panic() and kexec when NMI is injected # echo "kernel.unknown_nmi_panic=1" >> /etc/sysctl.conf # vim /etc/default/grub add "nmi_watchdog=0 crashkernel=256M" in GRUB_CMDLINE_LINUX line # grub2-mkconfig 4. Reboot the OS 5. Run following command for each vcpu on the guest # while true; do echo <CPU num> > /proc/irq/<IO-APIC-edge or IO-APIC-fasteoi>/smp_affinitity; done; By running this command, cpus will get ioapic_lock for setting affinity. 6. Inject NMI (push a dump button or execute 'virsh inject-nmi <domain>' if you use VM). After injecting NMI, panic() is called in an nmi-handler context. Then, kexec will normally run in panic(), but the operation will be stopped by deadlock on ioapic_lock in crash_kexec()->machine_crash_shutdown()-> native_machine_crash_shutdown()->disable_IO_APIC()->clear_IO_APIC()-> clear_IO_APIC_pin()->ioapic_read_entry(). Signed-off-by: Yoshihiro YUNOMAE <yoshihiro.yunomae.ez@hitachi.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Gleb Natapov <gleb@redhat.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Joerg Roedel <joro@8bytes.org> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com> Cc: Sebastian Andrzej Siewior <sebastian@breakpoint.cc> Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Cc: yrl.pp-manager.tt@hitachi.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Seiji Aguchi <seiji.aguchi@hds.com> Link: http://lkml.kernel.org/r/20130820070107.28245.83806.stgit@yunodevel Signed-off-by: Ingo Molnar <mingo@kernel.org>
2013-08-20 14:01:07 +07:00
extern void ioapic_zap_locks(void);
#endif /* _ASM_X86_APIC_H */