diff options
Diffstat (limited to 'arch/s390')
75 files changed, 4455 insertions, 1564 deletions
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index d68fe34799b0..bb63499fc5d3 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -60,7 +60,6 @@ config PCI_QUIRKS config S390 def_bool y - select ARCH_DISCARD_MEMBLOCK select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE select ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS select ARCH_HAVE_NMI_SAFE_CMPXCHG @@ -130,6 +129,7 @@ config S390 select HAVE_KVM if 64BIT select HAVE_MEMBLOCK select HAVE_MEMBLOCK_NODE_MAP + select HAVE_MEMBLOCK_PHYS_MAP select HAVE_MOD_ARCH_SPECIFIC select HAVE_OPROFILE select HAVE_PERF_EVENTS @@ -139,6 +139,7 @@ config S390 select HAVE_VIRT_CPU_ACCOUNTING select KTIME_SCALAR if 32BIT select MODULES_USE_ELF_RELA + select NO_BOOTMEM select OLD_SIGACTION select OLD_SIGSUSPEND3 select SYSCTL_EXCEPTION_TRACE @@ -592,21 +593,14 @@ config CRASH_DUMP bool "kernel crash dumps" depends on 64BIT && SMP select KEXEC - select ZFCPDUMP help Generate crash dump after being started by kexec. Crash dump kernels are loaded in the main kernel with kexec-tools into a specially reserved region and then later executed after a crash by kdump/kexec. - For more details see Documentation/kdump/kdump.txt - -config ZFCPDUMP - def_bool n - prompt "zfcpdump support" - depends on 64BIT && SMP - help - Select this option if you want to build an zfcpdump enabled kernel. Refer to <file:Documentation/s390/zfcpdump.txt> for more details on this. + This option also enables s390 zfcpdump. + See also <file:Documentation/s390/zfcpdump.txt> endmenu diff --git a/arch/s390/appldata/appldata_mem.c b/arch/s390/appldata/appldata_mem.c index 42be53743133..edcf2a706942 100644 --- a/arch/s390/appldata/appldata_mem.c +++ b/arch/s390/appldata/appldata_mem.c @@ -13,6 +13,7 @@ #include <linux/kernel_stat.h> #include <linux/pagemap.h> #include <linux/swap.h> +#include <linux/slab.h> #include <asm/io.h> #include "appldata.h" diff --git a/arch/s390/boot/compressed/Makefile b/arch/s390/boot/compressed/Makefile index 866ecbe670e4..f90d1fc6d603 100644 --- a/arch/s390/boot/compressed/Makefile +++ b/arch/s390/boot/compressed/Makefile @@ -12,7 +12,7 @@ targets += misc.o piggy.o sizes.h head$(BITS).o KBUILD_CFLAGS := -m$(BITS) -D__KERNEL__ $(LINUX_INCLUDE) -O2 KBUILD_CFLAGS += -DDISABLE_BRANCH_PROFILING -KBUILD_CFLAGS += $(cflags-y) +KBUILD_CFLAGS += $(cflags-y) -fno-delete-null-pointer-checks KBUILD_CFLAGS += $(call cc-option,-mpacked-stack) KBUILD_CFLAGS += $(call cc-option,-ffreestanding) diff --git a/arch/s390/include/asm/atomic.h b/arch/s390/include/asm/atomic.h index 1d4706114a45..fa934fe080c1 100644 --- a/arch/s390/include/asm/atomic.h +++ b/arch/s390/include/asm/atomic.h @@ -412,9 +412,4 @@ static inline long long atomic64_dec_if_positive(atomic64_t *v) #define atomic64_dec_and_test(_v) (atomic64_sub_return(1, _v) == 0) #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0) -#define smp_mb__before_atomic_dec() smp_mb() -#define smp_mb__after_atomic_dec() smp_mb() -#define smp_mb__before_atomic_inc() smp_mb() -#define smp_mb__after_atomic_inc() smp_mb() - #endif /* __ARCH_S390_ATOMIC__ */ diff --git a/arch/s390/include/asm/barrier.h b/arch/s390/include/asm/barrier.h index 578680f6207a..19ff956b752b 100644 --- a/arch/s390/include/asm/barrier.h +++ b/arch/s390/include/asm/barrier.h @@ -27,8 +27,9 @@ #define smp_rmb() rmb() #define smp_wmb() wmb() #define smp_read_barrier_depends() read_barrier_depends() -#define smp_mb__before_clear_bit() smp_mb() -#define smp_mb__after_clear_bit() smp_mb() + +#define smp_mb__before_atomic() smp_mb() +#define smp_mb__after_atomic() smp_mb() #define set_mb(var, value) do { var = value; mb(); } while (0) diff --git a/arch/s390/include/asm/ccwdev.h b/arch/s390/include/asm/ccwdev.h index a9c2c0686177..b80e456d6428 100644 --- a/arch/s390/include/asm/ccwdev.h +++ b/arch/s390/include/asm/ccwdev.h @@ -229,5 +229,5 @@ int ccw_device_siosl(struct ccw_device *); extern void ccw_device_get_schid(struct ccw_device *, struct subchannel_id *); -extern void *ccw_device_get_chp_desc(struct ccw_device *, int); +struct channel_path_desc *ccw_device_get_chp_desc(struct ccw_device *, int); #endif /* _S390_CCWDEV_H_ */ diff --git a/arch/s390/include/asm/ccwgroup.h b/arch/s390/include/asm/ccwgroup.h index ebc2913f9ee0..057ce0ca6377 100644 --- a/arch/s390/include/asm/ccwgroup.h +++ b/arch/s390/include/asm/ccwgroup.h @@ -10,6 +10,8 @@ struct ccw_driver; * @count: number of attached slave devices * @dev: embedded device structure * @cdev: variable number of slave devices, allocated as needed + * @ungroup_work: work to be done when a ccwgroup notifier has action + * type %BUS_NOTIFY_UNBIND_DRIVER */ struct ccwgroup_device { enum { diff --git a/arch/s390/include/asm/chpid.h b/arch/s390/include/asm/chpid.h index 38c405ef89ce..7298eec98541 100644 --- a/arch/s390/include/asm/chpid.h +++ b/arch/s390/include/asm/chpid.h @@ -8,6 +8,17 @@ #include <uapi/asm/chpid.h> #include <asm/cio.h> +struct channel_path_desc { + u8 flags; + u8 lsn; + u8 desc; + u8 chpid; + u8 swla; + u8 zeroes; + u8 chla; + u8 chpp; +} __packed; + static inline void chp_id_init(struct chp_id *chpid) { memset(chpid, 0, sizeof(struct chp_id)); diff --git a/arch/s390/include/asm/ctl_reg.h b/arch/s390/include/asm/ctl_reg.h index 4e63f1a13600..31ab9f346d7e 100644 --- a/arch/s390/include/asm/ctl_reg.h +++ b/arch/s390/include/asm/ctl_reg.h @@ -57,6 +57,20 @@ static inline void __ctl_clear_bit(unsigned int cr, unsigned int bit) void smp_ctl_set_bit(int cr, int bit); void smp_ctl_clear_bit(int cr, int bit); +union ctlreg0 { + unsigned long val; + struct { +#ifdef CONFIG_64BIT + unsigned long : 32; +#endif + unsigned long : 3; + unsigned long lap : 1; /* Low-address-protection control */ + unsigned long : 4; + unsigned long edat : 1; /* Enhanced-DAT-enablement control */ + unsigned long : 23; + }; +}; + #ifdef CONFIG_SMP # define ctl_set_bit(cr, bit) smp_ctl_set_bit(cr, bit) # define ctl_clear_bit(cr, bit) smp_ctl_clear_bit(cr, bit) diff --git a/arch/s390/include/asm/futex.h b/arch/s390/include/asm/futex.h index 69cf5b5eddc9..a4811aa0304d 100644 --- a/arch/s390/include/asm/futex.h +++ b/arch/s390/include/asm/futex.h @@ -29,7 +29,7 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr) int cmparg = (encoded_op << 20) >> 20; int oldval = 0, newval, ret; - update_primary_asce(current); + load_kernel_asce(); if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; @@ -79,7 +79,7 @@ static inline int futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, { int ret; - update_primary_asce(current); + load_kernel_asce(); asm volatile( " sacf 256\n" "0: cs %1,%4,0(%5)\n" diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index 154b60089be9..4181d7baabba 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -32,16 +32,26 @@ #define KVM_NR_IRQCHIPS 1 #define KVM_IRQCHIP_NUM_PINS 4096 +#define SIGP_CTRL_C 0x00800000 + struct sca_entry { - atomic_t scn; + atomic_t ctrl; __u32 reserved; __u64 sda; __u64 reserved2[2]; } __attribute__((packed)); +union ipte_control { + unsigned long val; + struct { + unsigned long k : 1; + unsigned long kh : 31; + unsigned long kg : 32; + }; +}; struct sca_block { - __u64 ipte_control; + union ipte_control ipte_control; __u64 reserved[5]; __u64 mcn; __u64 reserved2; @@ -64,6 +74,7 @@ struct sca_block { #define CPUSTAT_ZARCH 0x00000800 #define CPUSTAT_MCDS 0x00000100 #define CPUSTAT_SM 0x00000080 +#define CPUSTAT_IBS 0x00000040 #define CPUSTAT_G 0x00000008 #define CPUSTAT_GED 0x00000004 #define CPUSTAT_J 0x00000002 @@ -71,7 +82,9 @@ struct sca_block { struct kvm_s390_sie_block { atomic_t cpuflags; /* 0x0000 */ - __u32 prefix; /* 0x0004 */ + __u32 : 1; /* 0x0004 */ + __u32 prefix : 18; + __u32 : 13; __u8 reserved08[4]; /* 0x0008 */ #define PROG_IN_SIE (1<<0) __u32 prog0c; /* 0x000c */ @@ -85,12 +98,27 @@ struct kvm_s390_sie_block { __u8 reserved40[4]; /* 0x0040 */ #define LCTL_CR0 0x8000 #define LCTL_CR6 0x0200 +#define LCTL_CR9 0x0040 +#define LCTL_CR10 0x0020 +#define LCTL_CR11 0x0010 #define LCTL_CR14 0x0002 __u16 lctl; /* 0x0044 */ __s16 icpua; /* 0x0046 */ -#define ICTL_LPSW 0x00400000 +#define ICTL_PINT 0x20000000 +#define ICTL_LPSW 0x00400000 +#define ICTL_STCTL 0x00040000 +#define ICTL_ISKE 0x00004000 +#define ICTL_SSKE 0x00002000 +#define ICTL_RRBE 0x00001000 +#define ICTL_TPROT 0x00000200 __u32 ictl; /* 0x0048 */ __u32 eca; /* 0x004c */ +#define ICPT_INST 0x04 +#define ICPT_PROGI 0x08 +#define ICPT_INSTPROGI 0x0C +#define ICPT_OPEREXC 0x2C +#define ICPT_PARTEXEC 0x38 +#define ICPT_IOINST 0x40 __u8 icptcode; /* 0x0050 */ __u8 reserved51; /* 0x0051 */ __u16 ihcpu; /* 0x0052 */ @@ -109,9 +137,24 @@ struct kvm_s390_sie_block { psw_t gpsw; /* 0x0090 */ __u64 gg14; /* 0x00a0 */ __u64 gg15; /* 0x00a8 */ - __u8 reservedb0[30]; /* 0x00b0 */ - __u16 iprcc; /* 0x00ce */ - __u8 reservedd0[48]; /* 0x00d0 */ + __u8 reservedb0[20]; /* 0x00b0 */ + __u16 extcpuaddr; /* 0x00c4 */ + __u16 eic; /* 0x00c6 */ + __u32 reservedc8; /* 0x00c8 */ + __u16 pgmilc; /* 0x00cc */ + __u16 iprcc; /* 0x00ce */ + __u32 dxc; /* 0x00d0 */ + __u16 mcn; /* 0x00d4 */ + __u8 perc; /* 0x00d6 */ + __u8 peratmid; /* 0x00d7 */ + __u64 peraddr; /* 0x00d8 */ + __u8 eai; /* 0x00e0 */ + __u8 peraid; /* 0x00e1 */ + __u8 oai; /* 0x00e2 */ + __u8 armid; /* 0x00e3 */ + __u8 reservede4[4]; /* 0x00e4 */ + __u64 tecmc; /* 0x00e8 */ + __u8 reservedf0[16]; /* 0x00f0 */ __u64 gcr[16]; /* 0x0100 */ __u64 gbea; /* 0x0180 */ __u8 reserved188[24]; /* 0x0188 */ @@ -146,6 +189,8 @@ struct kvm_vcpu_stat { u32 exit_instruction; u32 instruction_lctl; u32 instruction_lctlg; + u32 instruction_stctl; + u32 instruction_stctg; u32 exit_program_interruption; u32 exit_instr_and_program; u32 deliver_external_call; @@ -164,6 +209,7 @@ struct kvm_vcpu_stat { u32 instruction_stpx; u32 instruction_stap; u32 instruction_storage_key; + u32 instruction_ipte_interlock; u32 instruction_stsch; u32 instruction_chsc; u32 instruction_stsi; @@ -183,13 +229,58 @@ struct kvm_vcpu_stat { u32 diagnose_9c; }; -#define PGM_OPERATION 0x01 -#define PGM_PRIVILEGED_OP 0x02 -#define PGM_EXECUTE 0x03 -#define PGM_PROTECTION 0x04 -#define PGM_ADDRESSING 0x05 -#define PGM_SPECIFICATION 0x06 -#define PGM_DATA 0x07 +#define PGM_OPERATION 0x01 +#define PGM_PRIVILEGED_OP 0x02 +#define PGM_EXECUTE 0x03 +#define PGM_PROTECTION 0x04 +#define PGM_ADDRESSING 0x05 +#define PGM_SPECIFICATION 0x06 +#define PGM_DATA 0x07 +#define PGM_FIXED_POINT_OVERFLOW 0x08 +#define PGM_FIXED_POINT_DIVIDE 0x09 +#define PGM_DECIMAL_OVERFLOW 0x0a +#define PGM_DECIMAL_DIVIDE 0x0b +#define PGM_HFP_EXPONENT_OVERFLOW 0x0c +#define PGM_HFP_EXPONENT_UNDERFLOW 0x0d +#define PGM_HFP_SIGNIFICANCE 0x0e +#define PGM_HFP_DIVIDE 0x0f +#define PGM_SEGMENT_TRANSLATION 0x10 +#define PGM_PAGE_TRANSLATION 0x11 +#define PGM_TRANSLATION_SPEC 0x12 +#define PGM_SPECIAL_OPERATION 0x13 +#define PGM_OPERAND 0x15 +#define PGM_TRACE_TABEL 0x16 +#define PGM_SPACE_SWITCH 0x1c +#define PGM_HFP_SQUARE_ROOT 0x1d +#define PGM_PC_TRANSLATION_SPEC 0x1f +#define PGM_AFX_TRANSLATION 0x20 +#define PGM_ASX_TRANSLATION 0x21 +#define PGM_LX_TRANSLATION 0x22 +#define PGM_EX_TRANSLATION 0x23 +#define PGM_PRIMARY_AUTHORITY 0x24 +#define PGM_SECONDARY_AUTHORITY 0x25 +#define PGM_LFX_TRANSLATION 0x26 +#define PGM_LSX_TRANSLATION 0x27 +#define PGM_ALET_SPECIFICATION 0x28 +#define PGM_ALEN_TRANSLATION 0x29 +#define PGM_ALE_SEQUENCE 0x2a +#define PGM_ASTE_VALIDITY 0x2b +#define PGM_ASTE_SEQUENCE 0x2c +#define PGM_EXTENDED_AUTHORITY 0x2d +#define PGM_LSTE_SEQUENCE 0x2e +#define PGM_ASTE_INSTANCE 0x2f +#define PGM_STACK_FULL 0x30 +#define PGM_STACK_EMPTY 0x31 +#define PGM_STACK_SPECIFICATION 0x32 +#define PGM_STACK_TYPE 0x33 +#define PGM_STACK_OPERATION 0x34 +#define PGM_ASCE_TYPE 0x38 +#define PGM_REGION_FIRST_TRANS 0x39 +#define PGM_REGION_SECOND_TRANS 0x3a +#define PGM_REGION_THIRD_TRANS 0x3b +#define PGM_MONITOR 0x40 +#define PGM_PER 0x80 +#define PGM_CRYPTO_OPERATION 0x119 struct kvm_s390_interrupt_info { struct list_head list; @@ -229,6 +320,45 @@ struct kvm_s390_float_interrupt { unsigned int irq_count; }; +struct kvm_hw_wp_info_arch { + unsigned long addr; + unsigned long phys_addr; + int len; + char *old_data; +}; + +struct kvm_hw_bp_info_arch { + unsigned long addr; + int len; +}; + +/* + * Only the upper 16 bits of kvm_guest_debug->control are arch specific. + * Further KVM_GUESTDBG flags which an be used from userspace can be found in + * arch/s390/include/uapi/asm/kvm.h + */ +#define KVM_GUESTDBG_EXIT_PENDING 0x10000000 + +#define guestdbg_enabled(vcpu) \ + (vcpu->guest_debug & KVM_GUESTDBG_ENABLE) +#define guestdbg_sstep_enabled(vcpu) \ + (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) +#define guestdbg_hw_bp_enabled(vcpu) \ + (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) +#define guestdbg_exit_pending(vcpu) (guestdbg_enabled(vcpu) && \ + (vcpu->guest_debug & KVM_GUESTDBG_EXIT_PENDING)) + +struct kvm_guestdbg_info_arch { + unsigned long cr0; + unsigned long cr9; + unsigned long cr10; + unsigned long cr11; + struct kvm_hw_bp_info_arch *hw_bp_info; + struct kvm_hw_wp_info_arch *hw_wp_info; + int nr_hw_bp; + int nr_hw_wp; + unsigned long last_bp; +}; struct kvm_vcpu_arch { struct kvm_s390_sie_block *sie_block; @@ -238,11 +368,13 @@ struct kvm_vcpu_arch { struct kvm_s390_local_interrupt local_int; struct hrtimer ckc_timer; struct tasklet_struct tasklet; + struct kvm_s390_pgm_info pgm; union { struct cpuid cpu_id; u64 stidp_data; }; struct gmap *gmap; + struct kvm_guestdbg_info_arch guestdbg; #define KVM_S390_PFAULT_TOKEN_INVALID (-1UL) unsigned long pfault_token; unsigned long pfault_select; @@ -285,7 +417,10 @@ struct kvm_arch{ struct gmap *gmap; int css_support; int use_irqchip; + int use_cmma; struct s390_io_adapter *adapters[MAX_S390_IO_ADAPTERS]; + wait_queue_head_t ipte_wq; + spinlock_t start_stop_lock; }; #define KVM_HVA_ERR_BAD (-1UL) diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h index bbf8141408cd..4349197ab9df 100644 --- a/arch/s390/include/asm/lowcore.h +++ b/arch/s390/include/asm/lowcore.h @@ -56,13 +56,14 @@ struct _lowcore { __u16 pgm_code; /* 0x008e */ __u32 trans_exc_code; /* 0x0090 */ __u16 mon_class_num; /* 0x0094 */ - __u16 per_perc_atmid; /* 0x0096 */ + __u8 per_code; /* 0x0096 */ + __u8 per_atmid; /* 0x0097 */ __u32 per_address; /* 0x0098 */ __u32 monitor_code; /* 0x009c */ __u8 exc_access_id; /* 0x00a0 */ __u8 per_access_id; /* 0x00a1 */ __u8 op_access_id; /* 0x00a2 */ - __u8 ar_access_id; /* 0x00a3 */ + __u8 ar_mode_id; /* 0x00a3 */ __u8 pad_0x00a4[0x00b8-0x00a4]; /* 0x00a4 */ __u16 subchannel_id; /* 0x00b8 */ __u16 subchannel_nr; /* 0x00ba */ @@ -93,7 +94,9 @@ struct _lowcore { __u32 save_area_sync[8]; /* 0x0200 */ __u32 save_area_async[8]; /* 0x0220 */ __u32 save_area_restart[1]; /* 0x0240 */ - __u8 pad_0x0244[0x0248-0x0244]; /* 0x0244 */ + + /* CPU flags. */ + __u32 cpu_flags; /* 0x0244 */ /* Return psws. */ psw_t return_psw; /* 0x0248 */ @@ -139,12 +142,9 @@ struct _lowcore { __u32 percpu_offset; /* 0x02f0 */ __u32 machine_flags; /* 0x02f4 */ __u32 ftrace_func; /* 0x02f8 */ - __u8 pad_0x02fc[0x0300-0x02fc]; /* 0x02fc */ - - /* Interrupt response block */ - __u8 irb[64]; /* 0x0300 */ + __u32 spinlock_lockval; /* 0x02fc */ - __u8 pad_0x0340[0x0e00-0x0340]; /* 0x0340 */ + __u8 pad_0x0300[0x0e00-0x0300]; /* 0x0300 */ /* * 0xe00 contains the address of the IPL Parameter Information @@ -196,12 +196,13 @@ struct _lowcore { __u16 pgm_code; /* 0x008e */ __u32 data_exc_code; /* 0x0090 */ __u16 mon_class_num; /* 0x0094 */ - __u16 per_perc_atmid; /* 0x0096 */ + __u8 per_code; /* 0x0096 */ + __u8 per_atmid; /* 0x0097 */ __u64 per_address; /* 0x0098 */ __u8 exc_access_id; /* 0x00a0 */ __u8 per_access_id; /* 0x00a1 */ __u8 op_access_id; /* 0x00a2 */ - __u8 ar_access_id; /* 0x00a3 */ + __u8 ar_mode_id; /* 0x00a3 */ __u8 pad_0x00a4[0x00a8-0x00a4]; /* 0x00a4 */ __u64 trans_exc_code; /* 0x00a8 */ __u64 monitor_code; /* 0x00b0 */ @@ -237,7 +238,9 @@ struct _lowcore { __u64 save_area_sync[8]; /* 0x0200 */ __u64 save_area_async[8]; /* 0x0240 */ __u64 save_area_restart[1]; /* 0x0280 */ - __u8 pad_0x0288[0x0290-0x0288]; /* 0x0288 */ + + /* CPU flags. */ + __u64 cpu_flags; /* 0x0288 */ /* Return psws. */ psw_t return_psw; /* 0x0290 */ @@ -285,15 +288,13 @@ struct _lowcore { __u64 machine_flags; /* 0x0388 */ __u64 ftrace_func; /* 0x0390 */ __u64 gmap; /* 0x0398 */ - __u8 pad_0x03a0[0x0400-0x03a0]; /* 0x03a0 */ - - /* Interrupt response block. */ - __u8 irb[64]; /* 0x0400 */ + __u32 spinlock_lockval; /* 0x03a0 */ + __u8 pad_0x03a0[0x0400-0x03a4]; /* 0x03a4 */ /* Per cpu primary space access list */ - __u32 paste[16]; /* 0x0440 */ + __u32 paste[16]; /* 0x0400 */ - __u8 pad_0x0480[0x0e00-0x0480]; /* 0x0480 */ + __u8 pad_0x04c0[0x0e00-0x0440]; /* 0x0440 */ /* * 0xe00 contains the address of the IPL Parameter Information diff --git a/arch/s390/include/asm/mmu.h b/arch/s390/include/asm/mmu.h index f77695a82f64..a5e656260a70 100644 --- a/arch/s390/include/asm/mmu.h +++ b/arch/s390/include/asm/mmu.h @@ -16,6 +16,8 @@ typedef struct { unsigned long vdso_base; /* The mmu context has extended page tables. */ unsigned int has_pgste:1; + /* The mmu context uses storage keys. */ + unsigned int use_skey:1; } mm_context_t; #define INIT_MM_CONTEXT(name) \ diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h index 71be346d0e3c..c28f32a45af5 100644 --- a/arch/s390/include/asm/mmu_context.h +++ b/arch/s390/include/asm/mmu_context.h @@ -23,6 +23,7 @@ static inline int init_new_context(struct task_struct *tsk, mm->context.asce_bits |= _ASCE_TYPE_REGION3; #endif mm->context.has_pgste = 0; + mm->context.use_skey = 0; mm->context.asce_limit = STACK_TOP_MAX; crst_table_init((unsigned long *) mm->pgd, pgd_entry_type(mm)); return 0; @@ -30,33 +31,31 @@ static inline int init_new_context(struct task_struct *tsk, #define destroy_context(mm) do { } while (0) -static inline void update_user_asce(struct mm_struct *mm, int load_primary) +static inline void set_user_asce(struct mm_struct *mm) { pgd_t *pgd = mm->pgd; S390_lowcore.user_asce = mm->context.asce_bits | __pa(pgd); - if (load_primary) - __ctl_load(S390_lowcore.user_asce, 1, 1); set_fs(current->thread.mm_segment); + set_cpu_flag(CIF_ASCE); } -static inline void clear_user_asce(struct mm_struct *mm, int load_primary) +static inline void clear_user_asce(void) { S390_lowcore.user_asce = S390_lowcore.kernel_asce; - if (load_primary) - __ctl_load(S390_lowcore.user_asce, 1, 1); + __ctl_load(S390_lowcore.user_asce, 1, 1); __ctl_load(S390_lowcore.user_asce, 7, 7); } -static inline void update_primary_asce(struct task_struct *tsk) +static inline void load_kernel_asce(void) { unsigned long asce; __ctl_store(asce, 1, 1); if (asce != S390_lowcore.kernel_asce) __ctl_load(S390_lowcore.kernel_asce, 1, 1); - set_tsk_thread_flag(tsk, TIF_ASCE); + set_cpu_flag(CIF_ASCE); } static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, @@ -64,25 +63,17 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, { int cpu = smp_processor_id(); - update_primary_asce(tsk); if (prev == next) return; if (MACHINE_HAS_TLB_LC) cpumask_set_cpu(cpu, &next->context.cpu_attach_mask); - if (atomic_inc_return(&next->context.attach_count) >> 16) { - /* Delay update_user_asce until all TLB flushes are done. */ - set_tsk_thread_flag(tsk, TIF_TLB_WAIT); - /* Clear old ASCE by loading the kernel ASCE. */ - clear_user_asce(next, 0); - } else { - cpumask_set_cpu(cpu, mm_cpumask(next)); - update_user_asce(next, 0); - if (next->context.flush_mm) - /* Flush pending TLBs */ - __tlb_flush_mm(next); - } + /* Clear old ASCE by loading the kernel ASCE. */ + __ctl_load(S390_lowcore.kernel_asce, 1, 1); + __ctl_load(S390_lowcore.kernel_asce, 7, 7); + /* Delay loading of the new ASCE to control registers CR1 & CR7 */ + set_cpu_flag(CIF_ASCE); + atomic_inc(&next->context.attach_count); atomic_dec(&prev->context.attach_count); - WARN_ON(atomic_read(&prev->context.attach_count) < 0); if (MACHINE_HAS_TLB_LC) cpumask_clear_cpu(cpu, &prev->context.cpu_attach_mask); } @@ -93,15 +84,14 @@ static inline void finish_arch_post_lock_switch(void) struct task_struct *tsk = current; struct mm_struct *mm = tsk->mm; - if (!test_tsk_thread_flag(tsk, TIF_TLB_WAIT)) + if (!mm) return; preempt_disable(); - clear_tsk_thread_flag(tsk, TIF_TLB_WAIT); while (atomic_read(&mm->context.attach_count) >> 16) cpu_relax(); cpumask_set_cpu(smp_processor_id(), mm_cpumask(mm)); - update_user_asce(mm, 0); + set_user_asce(mm); if (mm->context.flush_mm) __tlb_flush_mm(mm); preempt_enable(); @@ -113,7 +103,9 @@ static inline void finish_arch_post_lock_switch(void) static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next) { - switch_mm(prev, next, current); + switch_mm(prev, next, current); + cpumask_set_cpu(smp_processor_id(), mm_cpumask(next)); + set_user_asce(next); } static inline void arch_dup_mmap(struct mm_struct *oldmm, diff --git a/arch/s390/include/asm/pci.h b/arch/s390/include/asm/pci.h index 2583466f576b..c030900320e0 100644 --- a/arch/s390/include/asm/pci.h +++ b/arch/s390/include/asm/pci.h @@ -78,10 +78,16 @@ struct zpci_dev { enum zpci_state state; u32 fid; /* function ID, used by sclp */ u32 fh; /* function handle, used by insn's */ + u16 vfn; /* virtual function number */ u16 pchid; /* physical channel ID */ u8 pfgid; /* function group ID */ + u8 pft; /* pci function type */ u16 domain; + u8 pfip[CLP_PFIP_NR_SEGMENTS]; /* pci function internal path */ + u32 uid; /* user defined id */ + u8 util_str[CLP_UTIL_STR_LEN]; /* utility string */ + /* IRQ stuff */ u64 msi_addr; /* MSI address */ struct airq_iv *aibv; /* adapter interrupt bit vector */ @@ -120,6 +126,8 @@ static inline bool zdev_enabled(struct zpci_dev *zdev) return (zdev->fh & (1UL << 31)) ? true : false; } +extern const struct attribute_group *zpci_attr_groups[]; + /* ----------------------------------------------------------------------------- Prototypes ----------------------------------------------------------------------------- */ @@ -166,10 +174,6 @@ static inline void zpci_exit_slot(struct zpci_dev *zdev) {} struct zpci_dev *get_zdev(struct pci_dev *); struct zpci_dev *get_zdev_by_fid(u32); -/* sysfs */ -int zpci_sysfs_add_device(struct device *); -void zpci_sysfs_remove_device(struct device *); - /* DMA */ int zpci_dma_init(void); void zpci_dma_exit(void); diff --git a/arch/s390/include/asm/pci_clp.h b/arch/s390/include/asm/pci_clp.h index d31d739f8689..dd78f92f1cce 100644 --- a/arch/s390/include/asm/pci_clp.h +++ b/arch/s390/include/asm/pci_clp.h @@ -44,6 +44,7 @@ struct clp_fh_list_entry { #define CLP_SET_DISABLE_PCI_FN 1 /* Yes, 1 disables it */ #define CLP_UTIL_STR_LEN 64 +#define CLP_PFIP_NR_SEGMENTS 4 /* List PCI functions request */ struct clp_req_list_pci { @@ -85,7 +86,7 @@ struct clp_rsp_query_pci { struct clp_rsp_hdr hdr; u32 fmt : 4; /* cmd request block format */ u32 : 28; - u64 reserved1; + u64 : 64; u16 vfn; /* virtual fn number */ u16 : 7; u16 util_str_avail : 1; /* utility string available? */ @@ -94,10 +95,13 @@ struct clp_rsp_query_pci { u8 bar_size[PCI_BAR_COUNT]; u16 pchid; u32 bar[PCI_BAR_COUNT]; - u64 reserved2; + u8 pfip[CLP_PFIP_NR_SEGMENTS]; /* pci function internal path */ + u32 : 24; + u8 pft; /* pci function type */ u64 sdma; /* start dma as */ u64 edma; /* end dma as */ - u64 reserved3[6]; + u32 reserved[11]; + u32 uid; /* user defined id */ u8 util_str[CLP_UTIL_STR_LEN]; /* utility string */ } __packed; diff --git a/arch/s390/include/asm/pgalloc.h b/arch/s390/include/asm/pgalloc.h index 884017cbfa9f..9e18a61d3df3 100644 --- a/arch/s390/include/asm/pgalloc.h +++ b/arch/s390/include/asm/pgalloc.h @@ -22,7 +22,8 @@ unsigned long *page_table_alloc(struct mm_struct *, unsigned long); void page_table_free(struct mm_struct *, unsigned long *); void page_table_free_rcu(struct mmu_gather *, unsigned long *); -void page_table_reset_pgste(struct mm_struct *, unsigned long, unsigned long); +void page_table_reset_pgste(struct mm_struct *, unsigned long, unsigned long, + bool init_skey); int set_guest_storage_key(struct mm_struct *mm, unsigned long addr, unsigned long key, bool nq); diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 12f75313e086..fcba5e03839f 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -309,7 +309,8 @@ extern unsigned long MODULES_END; #define PGSTE_HC_BIT 0x00200000UL #define PGSTE_GR_BIT 0x00040000UL #define PGSTE_GC_BIT 0x00020000UL -#define PGSTE_IN_BIT 0x00008000UL /* IPTE notify bit */ +#define PGSTE_UC_BIT 0x00008000UL /* user dirty (migration) */ +#define PGSTE_IN_BIT 0x00004000UL /* IPTE notify bit */ #else /* CONFIG_64BIT */ @@ -391,7 +392,8 @@ extern unsigned long MODULES_END; #define PGSTE_HC_BIT 0x0020000000000000UL #define PGSTE_GR_BIT 0x0004000000000000UL #define PGSTE_GC_BIT 0x0002000000000000UL -#define PGSTE_IN_BIT 0x0000800000000000UL /* IPTE notify bit */ +#define PGSTE_UC_BIT 0x0000800000000000UL /* user dirty (migration) */ +#define PGSTE_IN_BIT 0x0000400000000000UL /* IPTE notify bit */ #endif /* CONFIG_64BIT */ @@ -466,6 +468,16 @@ static inline int mm_has_pgste(struct mm_struct *mm) #endif return 0; } + +static inline int mm_use_skey(struct mm_struct *mm) +{ +#ifdef CONFIG_PGSTE + if (mm->context.use_skey) + return 1; +#endif + return 0; +} + /* * pgd/pmd/pte query functions */ @@ -699,26 +711,17 @@ static inline void pgste_set(pte_t *ptep, pgste_t pgste) #endif } -static inline pgste_t pgste_update_all(pte_t *ptep, pgste_t pgste) +static inline pgste_t pgste_update_all(pte_t *ptep, pgste_t pgste, + struct mm_struct *mm) { #ifdef CONFIG_PGSTE unsigned long address, bits, skey; - if (pte_val(*ptep) & _PAGE_INVALID) + if (!mm_use_skey(mm) || pte_val(*ptep) & _PAGE_INVALID) return pgste; address = pte_val(*ptep) & PAGE_MASK; skey = (unsigned long) page_get_storage_key(address); bits = skey & (_PAGE_CHANGED | _PAGE_REFERENCED); - if (!(pgste_val(pgste) & PGSTE_HC_BIT) && (bits & _PAGE_CHANGED)) { - /* Transfer dirty + referenced bit to host bits in pgste */ - pgste_val(pgste) |= bits << 52; - page_set_storage_key(address, skey ^ bits, 0); - } else if (!(pgste_val(pgste) & PGSTE_HR_BIT) && - (bits & _PAGE_REFERENCED)) { - /* Transfer referenced bit to host bit in pgste */ - pgste_val(pgste) |= PGSTE_HR_BIT; - page_reset_referenced(address); - } /* Transfer page changed & referenced bit to guest bits in pgste */ pgste_val(pgste) |= bits << 48; /* GR bit & GC bit */ /* Copy page access key and fetch protection bit to pgste */ @@ -729,25 +732,14 @@ static inline pgste_t pgste_update_all(pte_t *ptep, pgste_t pgste) } -static inline pgste_t pgste_update_young(pte_t *ptep, pgste_t pgste) -{ -#ifdef CONFIG_PGSTE - if (pte_val(*ptep) & _PAGE_INVALID) - return pgste; - /* Get referenced bit from storage key */ - if (page_reset_referenced(pte_val(*ptep) & PAGE_MASK)) - pgste_val(pgste) |= PGSTE_HR_BIT | PGSTE_GR_BIT; -#endif - return pgste; -} - -static inline void pgste_set_key(pte_t *ptep, pgste_t pgste, pte_t entry) +static inline void pgste_set_key(pte_t *ptep, pgste_t pgste, pte_t entry, + struct mm_struct *mm) { #ifdef CONFIG_PGSTE unsigned long address; unsigned long nkey; - if (pte_val(entry) & _PAGE_INVALID) + if (!mm_use_skey(mm) || pte_val(entry) & _PAGE_INVALID) return; VM_BUG_ON(!(pte_val(*ptep) & _PAGE_INVALID)); address = pte_val(entry) & PAGE_MASK; @@ -757,23 +749,30 @@ static inline void pgste_set_key(pte_t *ptep, pgste_t pgste, pte_t entry) * key C/R to 0. */ nkey = (pgste_val(pgste) & (PGSTE_ACC_BITS | PGSTE_FP_BIT)) >> 56; + nkey |= (pgste_val(pgste) & (PGSTE_GR_BIT | PGSTE_GC_BIT)) >> 48; page_set_storage_key(address, nkey, 0); #endif } -static inline void pgste_set_pte(pte_t *ptep, pte_t entry) +static inline pgste_t pgste_set_pte(pte_t *ptep, pgste_t pgste, pte_t entry) { - if (!MACHINE_HAS_ESOP && - (pte_val(entry) & _PAGE_PRESENT) && - (pte_val(entry) & _PAGE_WRITE)) { - /* - * Without enhanced suppression-on-protection force - * the dirty bit on for all writable ptes. - */ - pte_val(entry) |= _PAGE_DIRTY; - pte_val(entry) &= ~_PAGE_PROTECT; + if ((pte_val(entry) & _PAGE_PRESENT) && + (pte_val(entry) & _PAGE_WRITE) && + !(pte_val(entry) & _PAGE_INVALID)) { + if (!MACHINE_HAS_ESOP) { + /* + * Without enhanced suppression-on-protection force + * the dirty bit on for all writable ptes. + */ + pte_val(entry) |= _PAGE_DIRTY; + pte_val(entry) &= ~_PAGE_PROTECT; + } + if (!(pte_val(entry) & _PAGE_PROTECT)) + /* This pte allows write access, set user-dirty */ + pgste_val(pgste) |= PGSTE_UC_BIT; } *ptep = entry; + return pgste; } /** @@ -839,6 +838,8 @@ unsigned long __gmap_fault(unsigned long address, struct gmap *); unsigned long gmap_fault(unsigned long address, struct gmap *); void gmap_discard(unsigned long from, unsigned long to, struct gmap *); void __gmap_zap(unsigned long address, struct gmap *); +bool gmap_test_and_clear_dirty(unsigned long address, struct gmap *); + void gmap_register_ipte_notifier(struct gmap_notifier *); void gmap_unregister_ipte_notifier(struct gmap_notifier *); @@ -870,8 +871,8 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr, if (mm_has_pgste(mm)) { pgste = pgste_get_lock(ptep); pgste_val(pgste) &= ~_PGSTE_GPS_ZERO; - pgste_set_key(ptep, pgste, entry); - pgste_set_pte(ptep, entry); + pgste_set_key(ptep, pgste, entry, mm); + pgste = pgste_set_pte(ptep, pgste, entry); pgste_set_unlock(ptep, pgste); } else { if (!(pte_val(entry) & _PAGE_INVALID) && MACHINE_HAS_EDAT1) @@ -1017,45 +1018,6 @@ static inline pte_t pte_mkhuge(pte_t pte) } #endif -/* - * Get (and clear) the user dirty bit for a pte. - */ -static inline int ptep_test_and_clear_user_dirty(struct mm_struct *mm, - pte_t *ptep) -{ - pgste_t pgste; - int dirty = 0; - - if (mm_has_pgste(mm)) { - pgste = pgste_get_lock(ptep); - pgste = pgste_update_all(ptep, pgste); - dirty = !!(pgste_val(pgste) & PGSTE_HC_BIT); - pgste_val(pgste) &= ~PGSTE_HC_BIT; - pgste_set_unlock(ptep, pgste); - return dirty; - } - return dirty; -} - -/* - * Get (and clear) the user referenced bit for a pte. - */ -static inline int ptep_test_and_clear_user_young(struct mm_struct *mm, - pte_t *ptep) -{ - pgste_t pgste; - int young = 0; - - if (mm_has_pgste(mm)) { - pgste = pgste_get_lock(ptep); - pgste = pgste_update_young(ptep, pgste); - young = !!(pgste_val(pgste) & PGSTE_HR_BIT); - pgste_val(pgste) &= ~PGSTE_HR_BIT; - pgste_set_unlock(ptep, pgste); - } - return young; -} - static inline void __ptep_ipte(unsigned long address, pte_t *ptep) { unsigned long pto = (unsigned long) ptep; @@ -1118,6 +1080,36 @@ static inline void ptep_flush_lazy(struct mm_struct *mm, atomic_sub(0x10000, &mm->context.attach_count); } +/* + * Get (and clear) the user dirty bit for a pte. + */ +static inline int ptep_test_and_clear_user_dirty(struct mm_struct *mm, + unsigned long addr, + pte_t *ptep) +{ + pgste_t pgste; + pte_t pte; + int dirty; + + if (!mm_has_pgste(mm)) + return 0; + pgste = pgste_get_lock(ptep); + dirty = !!(pgste_val(pgste) & PGSTE_UC_BIT); + pgste_val(pgste) &= ~PGSTE_UC_BIT; + pte = *ptep; + if (dirty && (pte_val(pte) & _PAGE_PRESENT)) { + pgste = pgste_ipte_notify(mm, ptep, pgste); + __ptep_ipte(addr, ptep); + if (MACHINE_HAS_ESOP || !(pte_val(pte) & _PAGE_WRITE)) + pte_val(pte) |= _PAGE_PROTECT; + else + pte_val(pte) |= _PAGE_INVALID; + *ptep = pte; + } + pgste_set_unlock(ptep, pgste); + return dirty; +} + #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep) @@ -1137,7 +1129,7 @@ static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, pte = pte_mkold(pte); if (mm_has_pgste(vma->vm_mm)) { - pgste_set_pte(ptep, pte); + pgste = pgste_set_pte(ptep, pgste, pte); pgste_set_unlock(ptep, pgste); } else *ptep = pte; @@ -1182,7 +1174,7 @@ static inline pte_t ptep_get_and_clear(struct mm_struct *mm, pte_val(*ptep) = _PAGE_INVALID; if (mm_has_pgste(mm)) { - pgste = pgste_update_all(&pte, pgste); + pgste = pgste_update_all(&pte, pgste, mm); pgste_set_unlock(ptep, pgste); } return pte; @@ -1205,7 +1197,7 @@ static inline pte_t ptep_modify_prot_start(struct mm_struct *mm, ptep_flush_lazy(mm, address, ptep); if (mm_has_pgste(mm)) { - pgste = pgste_update_all(&pte, pgste); + pgste = pgste_update_all(&pte, pgste, mm); pgste_set(ptep, pgste); } return pte; @@ -1219,8 +1211,8 @@ static inline void ptep_modify_prot_commit(struct mm_struct *mm, if (mm_has_pgste(mm)) { pgste = pgste_get(ptep); - pgste_set_key(ptep, pgste, pte); - pgste_set_pte(ptep, pte); + pgste_set_key(ptep, pgste, pte, mm); + pgste = pgste_set_pte(ptep, pgste, pte); pgste_set_unlock(ptep, pgste); } else *ptep = pte; @@ -1246,7 +1238,7 @@ static inline pte_t ptep_clear_flush(struct vm_area_struct *vma, if ((pgste_val(pgste) & _PGSTE_GPS_USAGE_MASK) == _PGSTE_GPS_USAGE_UNUSED) pte_val(pte) |= _PAGE_UNUSED; - pgste = pgste_update_all(&pte, pgste); + pgste = pgste_update_all(&pte, pgste, vma->vm_mm); pgste_set_unlock(ptep, pgste); } return pte; @@ -1278,7 +1270,7 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, pte_val(*ptep) = _PAGE_INVALID; if (!full && mm_has_pgste(mm)) { - pgste = pgste_update_all(&pte, pgste); + pgste = pgste_update_all(&pte, pgste, mm); pgste_set_unlock(ptep, pgste); } return pte; @@ -1301,7 +1293,7 @@ static inline pte_t ptep_set_wrprotect(struct mm_struct *mm, pte = pte_wrprotect(pte); if (mm_has_pgste(mm)) { - pgste_set_pte(ptep, pte); + pgste = pgste_set_pte(ptep, pgste, pte); pgste_set_unlock(ptep, pgste); } else *ptep = pte; @@ -1326,7 +1318,7 @@ static inline int ptep_set_access_flags(struct vm_area_struct *vma, ptep_flush_direct(vma->vm_mm, address, ptep); if (mm_has_pgste(vma->vm_mm)) { - pgste_set_pte(ptep, entry); + pgste = pgste_set_pte(ptep, pgste, entry); pgste_set_unlock(ptep, pgste); } else *ptep = entry; @@ -1734,6 +1726,7 @@ static inline pte_t mk_swap_pte(unsigned long type, unsigned long offset) extern int vmem_add_mapping(unsigned long start, unsigned long size); extern int vmem_remove_mapping(unsigned long start, unsigned long size); extern int s390_enable_sie(void); +extern void s390_enable_skey(void); /* * No page table caches to initialise diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h index dc5fc4f90e52..6f02d452bbee 100644 --- a/arch/s390/include/asm/processor.h +++ b/arch/s390/include/asm/processor.h @@ -11,6 +11,13 @@ #ifndef __ASM_S390_PROCESSOR_H #define __ASM_S390_PROCESSOR_H +#define CIF_MCCK_PENDING 0 /* machine check handling is pending */ +#define CIF_ASCE 1 /* user asce needs fixup / uaccess */ + +#define _CIF_MCCK_PENDING (1<<CIF_MCCK_PENDING) +#define _CIF_ASCE (1<<CIF_ASCE) + + #ifndef __ASSEMBLY__ #include <linux/linkage.h> @@ -21,6 +28,21 @@ #include <asm/setup.h> #include <asm/runtime_instr.h> +static inline void set_cpu_flag(int flag) +{ + S390_lowcore.cpu_flags |= (1U << flag); +} + +static inline void clear_cpu_flag(int flag) +{ + S390_lowcore.cpu_flags &= ~(1U << flag); +} + +static inline int test_cpu_flag(int flag) +{ + return !!(S390_lowcore.cpu_flags & (1U << flag)); +} + /* * Default implementation of macro that returns current * instruction pointer ("program counter"). diff --git a/arch/s390/include/asm/ptrace.h b/arch/s390/include/asm/ptrace.h index f4783c0b7b43..55d69dd7473c 100644 --- a/arch/s390/include/asm/ptrace.h +++ b/arch/s390/include/asm/ptrace.h @@ -8,6 +8,12 @@ #include <uapi/asm/ptrace.h> +#define PIF_SYSCALL 0 /* inside a system call */ +#define PIF_PER_TRAP 1 /* deliver sigtrap on return to user */ + +#define _PIF_SYSCALL (1<<PIF_SYSCALL) +#define _PIF_PER_TRAP (1<<PIF_PER_TRAP) + #ifndef __ASSEMBLY__ #define PSW_KERNEL_BITS (PSW_DEFAULT_KEY | PSW_MASK_BASE | PSW_ASC_HOME | \ @@ -16,6 +22,50 @@ PSW_DEFAULT_KEY | PSW_MASK_BASE | PSW_MASK_MCHECK | \ PSW_MASK_PSTATE | PSW_ASC_PRIMARY) +struct psw_bits { + unsigned long long : 1; + unsigned long long r : 1; /* PER-Mask */ + unsigned long long : 3; + unsigned long long t : 1; /* DAT Mode */ + unsigned long long i : 1; /* Input/Output Mask */ + unsigned long long e : 1; /* External Mask */ + unsigned long long key : 4; /* PSW Key */ + unsigned long long : 1; + unsigned long long m : 1; /* Machine-Check Mask */ + unsigned long long w : 1; /* Wait State */ + unsigned long long p : 1; /* Problem State */ + unsigned long long as : 2; /* Address Space Control */ + unsigned long long cc : 2; /* Condition Code */ + unsigned long long pm : 4; /* Program Mask */ + unsigned long long ri : 1; /* Runtime Instrumentation */ + unsigned long long : 6; + unsigned long long eaba : 2; /* Addressing Mode */ +#ifdef CONFIG_64BIT + unsigned long long : 31; + unsigned long long ia : 64;/* Instruction Address */ +#else + unsigned long long ia : 31;/* Instruction Address */ +#endif +}; + +enum { + PSW_AMODE_24BIT = 0, + PSW_AMODE_31BIT = 1, + PSW_AMODE_64BIT = 3 +}; + +enum { + PSW_AS_PRIMARY = 0, + PSW_AS_ACCREG = 1, + PSW_AS_SECONDARY = 2, + PSW_AS_HOME = 3 +}; + +#define psw_bits(__psw) (*({ \ + typecheck(psw_t, __psw); \ + &(*(struct psw_bits *)(&(__psw))); \ +})) + /* * The pt_regs struct defines the way the registers are stored on * the stack during a system call. @@ -29,6 +79,7 @@ struct pt_regs unsigned int int_code; unsigned int int_parm; unsigned long int_parm_long; + unsigned long flags; }; /* @@ -79,6 +130,21 @@ struct per_struct_kernel { #define PER_CONTROL_SUSPENSION 0x00400000UL #define PER_CONTROL_ALTERATION 0x00200000UL +static inline void set_pt_regs_flag(struct pt_regs *regs, int flag) +{ + regs->flags |= (1U << flag); +} + +static inline void clear_pt_regs_flag(struct pt_regs *regs, int flag) +{ + regs->flags &= ~(1U << flag); +} + +static inline int test_pt_regs_flag(struct pt_regs *regs, int flag) +{ + return !!(regs->flags & (1U << flag)); +} + /* * These are defined as per linux/ptrace.h, which see. */ diff --git a/arch/s390/include/asm/sclp.h b/arch/s390/include/asm/sclp.h index 2f5e9932b4de..1aba89b53cb9 100644 --- a/arch/s390/include/asm/sclp.h +++ b/arch/s390/include/asm/sclp.h @@ -28,7 +28,11 @@ struct sclp_ipl_info { struct sclp_cpu_entry { u8 address; - u8 reserved0[13]; + u8 reserved0[2]; + u8 : 3; + u8 siif : 1; + u8 : 4; + u8 reserved2[10]; u8 type; u8 reserved1; } __attribute__((packed)); @@ -61,5 +65,7 @@ int sclp_pci_deconfigure(u32 fid); int memcpy_hsa(void *dest, unsigned long src, size_t count, int mode); unsigned long sclp_get_hsa_size(void); void sclp_early_detect(void); +int sclp_has_siif(void); +unsigned int sclp_get_ibc(void); #endif /* _ASM_S390_SCLP_H */ diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h index b31b22dba948..089a49814c50 100644 --- a/arch/s390/include/asm/setup.h +++ b/arch/s390/include/asm/setup.h @@ -9,7 +9,6 @@ #define PARMAREA 0x10400 -#define MEMORY_CHUNKS 256 #ifndef __ASSEMBLY__ @@ -31,22 +30,11 @@ #endif /* CONFIG_64BIT */ #define COMMAND_LINE ((char *) (0x10480)) -#define CHUNK_READ_WRITE 0 -#define CHUNK_READ_ONLY 1 - -struct mem_chunk { - unsigned long addr; - unsigned long size; - int type; -}; - -extern struct mem_chunk memory_chunk[]; extern int memory_end_set; extern unsigned long memory_end; +extern unsigned long max_physmem_end; -void detect_memory_layout(struct mem_chunk chunk[], unsigned long maxsize); -void create_mem_hole(struct mem_chunk mem_chunk[], unsigned long addr, - unsigned long size); +extern void detect_memory_memblock(void); /* * Machine features detected in head.S diff --git a/arch/s390/include/asm/smp.h b/arch/s390/include/asm/smp.h index 21703f85b48d..4f1307962a95 100644 --- a/arch/s390/include/asm/smp.h +++ b/arch/s390/include/asm/smp.h @@ -30,7 +30,6 @@ extern int smp_store_status(int cpu); extern int smp_vcpu_scheduled(int cpu); extern void smp_yield_cpu(int cpu); extern void smp_yield(void); -extern void smp_stop_cpu(void); extern void smp_cpu_set_polarization(int cpu, int val); extern int smp_cpu_get_polarization(int cpu); extern void smp_fill_possible_mask(void); @@ -54,6 +53,8 @@ static inline void smp_yield_cpu(int cpu) { } static inline void smp_yield(void) { } static inline void smp_fill_possible_mask(void) { } +#endif /* CONFIG_SMP */ + static inline void smp_stop_cpu(void) { u16 pcpu = stap(); @@ -64,8 +65,6 @@ static inline void smp_stop_cpu(void) } } -#endif /* CONFIG_SMP */ - #ifdef CONFIG_HOTPLUG_CPU extern int smp_rescan_cpus(void); extern void __noreturn cpu_die(void); diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h index 83e5d216105e..96879f7ad6da 100644 --- a/arch/s390/include/asm/spinlock.h +++ b/arch/s390/include/asm/spinlock.h @@ -11,18 +11,21 @@ #include <linux/smp.h> +#define SPINLOCK_LOCKVAL (S390_lowcore.spinlock_lockval) + extern int spin_retry; static inline int -_raw_compare_and_swap(volatile unsigned int *lock, - unsigned int old, unsigned int new) +_raw_compare_and_swap(unsigned int *lock, unsigned int old, unsigned int new) { + unsigned int old_expected = old; + asm volatile( " cs %0,%3,%1" : "=d" (old), "=Q" (*lock) : "0" (old), "d" (new), "Q" (*lock) : "cc", "memory" ); - return old; + return old == old_expected; } /* @@ -34,57 +37,69 @@ _raw_compare_and_swap(volatile unsigned int *lock, * (the type definitions are in asm/spinlock_types.h) */ -#define arch_spin_is_locked(x) ((x)->owner_cpu != 0) -#define arch_spin_unlock_wait(lock) \ - do { while (arch_spin_is_locked(lock)) \ - arch_spin_relax(lock); } while (0) +void arch_spin_lock_wait(arch_spinlock_t *); +int arch_spin_trylock_retry(arch_spinlock_t *); +void arch_spin_relax(arch_spinlock_t *); +void arch_spin_lock_wait_flags(arch_spinlock_t *, unsigned long flags); -extern void arch_spin_lock_wait(arch_spinlock_t *); -extern void arch_spin_lock_wait_flags(arch_spinlock_t *, unsigned long flags); -extern int arch_spin_trylock_retry(arch_spinlock_t *); -extern void arch_spin_relax(arch_spinlock_t *lock); +static inline u32 arch_spin_lockval(int cpu) +{ + return ~cpu; +} static inline int arch_spin_value_unlocked(arch_spinlock_t lock) { - return lock.owner_cpu == 0; + return lock.lock == 0; } -static inline void arch_spin_lock(arch_spinlock_t *lp) +static inline int arch_spin_is_locked(arch_spinlock_t *lp) { - int old; + return ACCESS_ONCE(lp->lock) != 0; +} - old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id()); - if (likely(old == 0)) - return; - arch_spin_lock_wait(lp); +static inline int arch_spin_trylock_once(arch_spinlock_t *lp) +{ + barrier(); + return likely(arch_spin_value_unlocked(*lp) && + _raw_compare_and_swap(&lp->lock, 0, SPINLOCK_LOCKVAL)); } -static inline void arch_spin_lock_flags(arch_spinlock_t *lp, - unsigned long flags) +static inline int arch_spin_tryrelease_once(arch_spinlock_t *lp) { - int old; + return _raw_compare_and_swap(&lp->lock, SPINLOCK_LOCKVAL, 0); +} - old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id()); - if (likely(old == 0)) - return; - arch_spin_lock_wait_flags(lp, flags); +static inline void arch_spin_lock(arch_spinlock_t *lp) +{ + if (!arch_spin_trylock_once(lp)) + arch_spin_lock_wait(lp); } -static inline int arch_spin_trylock(arch_spinlock_t *lp) +static inline void arch_spin_lock_flags(arch_spinlock_t *lp, + unsigned long flags) { - int old; + if (!arch_spin_trylock_once(lp)) + arch_spin_lock_wait_flags(lp, flags); +} - old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id()); - if (likely(old == 0)) - return 1; - return arch_spin_trylock_retry(lp); +static inline int arch_spin_trylock(arch_spinlock_t *lp) +{ + if (!arch_spin_trylock_once(lp)) + return arch_spin_trylock_retry(lp); + return 1; } static inline void arch_spin_unlock(arch_spinlock_t *lp) { - _raw_compare_and_swap(&lp->owner_cpu, lp->owner_cpu, 0); + arch_spin_tryrelease_once(lp); +} + +static inline void arch_spin_unlock_wait(arch_spinlock_t *lock) +{ + while (arch_spin_is_locked(lock)) + arch_spin_relax(lock); } - + /* * Read-write spinlocks, allowing multiple readers * but only one writer. @@ -115,42 +130,50 @@ extern void _raw_write_lock_wait(arch_rwlock_t *lp); extern void _raw_write_lock_wait_flags(arch_rwlock_t *lp, unsigned long flags); extern int _raw_write_trylock_retry(arch_rwlock_t *lp); +static inline int arch_read_trylock_once(arch_rwlock_t *rw) +{ + unsigned int old = ACCESS_ONCE(rw->lock); + return likely((int) old >= 0 && + _raw_compare_and_swap(&rw->lock, old, old + 1)); +} + +static inline int arch_write_trylock_once(arch_rwlock_t *rw) +{ + unsigned int old = ACCESS_ONCE(rw->lock); + return likely(old == 0 && + _raw_compare_and_swap(&rw->lock, 0, 0x80000000)); +} + static inline void arch_read_lock(arch_rwlock_t *rw) { - unsigned int old; - old = rw->lock & 0x7fffffffU; - if (_raw_compare_and_swap(&rw->lock, old, old + 1) != old) + if (!arch_read_trylock_once(rw)) _raw_read_lock_wait(rw); } static inline void arch_read_lock_flags(arch_rwlock_t *rw, unsigned long flags) { - unsigned int old; - old = rw->lock & 0x7fffffffU; - if (_raw_compare_and_swap(&rw->lock, old, old + 1) != old) + if (!arch_read_trylock_once(rw)) _raw_read_lock_wait_flags(rw, flags); } static inline void arch_read_unlock(arch_rwlock_t *rw) { - unsigned int old, cmp; + unsigned int old; - old = rw->lock; do { - cmp = old; - old = _raw_compare_and_swap(&rw->lock, old, old - 1); - } while (cmp != old); + old = ACCESS_ONCE(rw->lock); + } while (!_raw_compare_and_swap(&rw->lock, old, old - 1)); } static inline void arch_write_lock(arch_rwlock_t *rw) { - if (unlikely(_raw_compare_and_swap(&rw->lock, 0, 0x80000000) != 0)) + if (!arch_write_trylock_once(rw)) _raw_write_lock_wait(rw); } static inline void arch_write_lock_flags(arch_rwlock_t *rw, unsigned long flags) { - if (unlikely(_raw_compare_and_swap(&rw->lock, 0, 0x80000000) != 0)) + if (!arch_write_trylock_once(rw)) _raw_write_lock_wait_flags(rw, flags); } @@ -161,18 +184,16 @@ static inline void arch_write_unlock(arch_rwlock_t *rw) static inline int arch_read_trylock(arch_rwlock_t *rw) { - unsigned int old; - old = rw->lock & 0x7fffffffU; - if (likely(_raw_compare_and_swap(&rw->lock, old, old + 1) == old)) - return 1; - return _raw_read_trylock_retry(rw); + if (!arch_read_trylock_once(rw)) + return _raw_read_trylock_retry(rw); + return 1; } static inline int arch_write_trylock(arch_rwlock_t *rw) { - if (likely(_raw_compare_and_swap(&rw->lock, 0, 0x80000000) == 0)) - return 1; - return _raw_write_trylock_retry(rw); + if (!arch_write_trylock_once(rw)) + return _raw_write_trylock_retry(rw); + return 1; } #define arch_read_relax(lock) cpu_relax() diff --git a/arch/s390/include/asm/spinlock_types.h b/arch/s390/include/asm/spinlock_types.h index 9c76656a0af0..b2cd6ff7c2c5 100644 --- a/arch/s390/include/asm/spinlock_types.h +++ b/arch/s390/include/asm/spinlock_types.h @@ -6,13 +6,13 @@ #endif typedef struct { - volatile unsigned int owner_cpu; + unsigned int lock; } __attribute__ ((aligned (4))) arch_spinlock_t; -#define __ARCH_SPIN_LOCK_UNLOCKED { 0 } +#define __ARCH_SPIN_LOCK_UNLOCKED { .lock = 0, } typedef struct { - volatile unsigned int lock; + unsigned int lock; } arch_rwlock_t; #define __ARCH_RW_LOCK_UNLOCKED { 0 } diff --git a/arch/s390/include/asm/switch_to.h b/arch/s390/include/asm/switch_to.h index e759181357fc..29c81f82705e 100644 --- a/arch/s390/include/asm/switch_to.h +++ b/arch/s390/include/asm/switch_to.h @@ -132,7 +132,6 @@ static inline void restore_access_regs(unsigned int *acrs) update_cr_regs(next); \ } \ prev = __switch_to(prev,next); \ - update_primary_asce(current); \ } while (0) #define finish_arch_switch(prev) do { \ diff --git a/arch/s390/include/asm/syscall.h b/arch/s390/include/asm/syscall.h index 777687055e7b..abad78d5b10c 100644 --- a/arch/s390/include/asm/syscall.h +++ b/arch/s390/include/asm/syscall.h @@ -28,7 +28,7 @@ extern const unsigned int sys_call_table_emu[]; static inline long syscall_get_nr(struct task_struct *task, struct pt_regs *regs) { - return test_tsk_thread_flag(task, TIF_SYSCALL) ? + return test_pt_regs_flag(regs, PIF_SYSCALL) ? (regs->int_code & 0xffff) : -1; } diff --git a/arch/s390/include/asm/thread_info.h b/arch/s390/include/asm/thread_info.h index 50630e6a35de..b833e9c0bfbf 100644 --- a/arch/s390/include/asm/thread_info.h +++ b/arch/s390/include/asm/thread_info.h @@ -77,32 +77,22 @@ static inline struct thread_info *current_thread_info(void) /* * thread information flags bit numbers */ -#define TIF_SYSCALL 0 /* inside a system call */ -#define TIF_NOTIFY_RESUME 1 /* callback before returning to user */ -#define TIF_SIGPENDING 2 /* signal pending */ -#define TIF_NEED_RESCHED 3 /* rescheduling necessary */ -#define TIF_TLB_WAIT 4 /* wait for TLB flush completion */ -#define TIF_ASCE 5 /* primary asce needs fixup / uaccess */ -#define TIF_PER_TRAP 6 /* deliver sigtrap on return to user */ -#define TIF_MCCK_PENDING 7 /* machine check handling is pending */ -#define TIF_SYSCALL_TRACE 8 /* syscall trace active */ -#define TIF_SYSCALL_AUDIT 9 /* syscall auditing active */ -#define TIF_SECCOMP 10 /* secure computing */ -#define TIF_SYSCALL_TRACEPOINT 11 /* syscall tracepoint instrumentation */ -#define TIF_31BIT 17 /* 32bit process */ -#define TIF_MEMDIE 18 /* is terminating due to OOM killer */ -#define TIF_RESTORE_SIGMASK 19 /* restore signal mask in do_signal() */ -#define TIF_SINGLE_STEP 20 /* This task is single stepped */ -#define TIF_BLOCK_STEP 21 /* This task is block stepped */ +#define TIF_NOTIFY_RESUME 0 /* callback before returning to user */ +#define TIF_SIGPENDING 1 /* signal pending */ +#define TIF_NEED_RESCHED 2 /* rescheduling necessary */ +#define TIF_SYSCALL_TRACE 3 /* syscall trace active */ +#define TIF_SYSCALL_AUDIT 4 /* syscall auditing active */ +#define TIF_SECCOMP 5 /* secure computing */ +#define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ +#define TIF_31BIT 16 /* 32bit process */ +#define TIF_MEMDIE 17 /* is terminating due to OOM killer */ +#define TIF_RESTORE_SIGMASK 18 /* restore signal mask in do_signal() */ +#define TIF_SINGLE_STEP 19 /* This task is single stepped */ +#define TIF_BLOCK_STEP 20 /* This task is block stepped */ -#define _TIF_SYSCALL (1<<TIF_SYSCALL) #define _TIF_NOTIFY_RESUME (1<<TIF_NOTIFY_RESUME) #define _TIF_SIGPENDING (1<<TIF_SIGPENDING) #define _TIF_NEED_RESCHED (1<<TIF_NEED_RESCHED) -#define _TIF_TLB_WAIT (1<<TIF_TLB_WAIT) -#define _TIF_ASCE (1<<TIF_ASCE) -#define _TIF_PER_TRAP (1<<TIF_PER_TRAP) -#define _TIF_MCCK_PENDING (1<<TIF_MCCK_PENDING) #define _TIF_SYSCALL_TRACE (1<<TIF_SYSCALL_TRACE) #define _TIF_SYSCALL_AUDIT (1<<TIF_SYSCALL_AUDIT) #define _TIF_SECCOMP (1<<TIF_SECCOMP) diff --git a/arch/s390/include/asm/topology.h b/arch/s390/include/asm/topology.h index 05425b18c0aa..56af53093d24 100644 --- a/arch/s390/include/asm/topology.h +++ b/arch/s390/include/asm/topology.h @@ -26,21 +26,12 @@ extern struct cpu_topology_s390 cpu_topology[NR_CPUS]; #define mc_capable() 1 -static inline const struct cpumask *cpu_coregroup_mask(int cpu) -{ - return &cpu_topology[cpu].core_mask; -} - -static inline const struct cpumask *cpu_book_mask(int cpu) -{ - return &cpu_topology[cpu].book_mask; -} - int topology_cpu_init(struct cpu *); int topology_set_cpu_management(int fc); void topology_schedule_update(void); void store_topology(struct sysinfo_15_1_x *info); void topology_expect_change(void); +const struct cpumask *cpu_coregroup_mask(int cpu); #else /* CONFIG_SCHED_BOOK */ @@ -64,8 +55,6 @@ static inline void s390_init_cpu_topology(void) }; #endif -#define SD_BOOK_INIT SD_CPU_INIT - #include <asm-generic/topology.h> #endif /* _ASM_S390_TOPOLOGY_H */ diff --git a/arch/s390/include/asm/uaccess.h b/arch/s390/include/asm/uaccess.h index 1be64a1506d0..cd4c68e0398d 100644 --- a/arch/s390/include/asm/uaccess.h +++ b/arch/s390/include/asm/uaccess.h @@ -132,6 +132,34 @@ unsigned long __must_check __copy_to_user(void __user *to, const void *from, #define __copy_to_user_inatomic __copy_to_user #define __copy_from_user_inatomic __copy_from_user +#ifdef CONFIG_HAVE_MARCH_Z10_FEATURES + +#define __put_get_user_asm(to, from, size, spec) \ +({ \ + register unsigned long __reg0 asm("0") = spec; \ + int __rc; \ + \ + asm volatile( \ + "0: mvcos %1,%3,%2\n" \ + "1: xr %0,%0\n" \ + "2:\n" \ + ".pushsection .fixup, \"ax\"\n" \ + "3: lhi %0,%5\n" \ + " jg 2b\n" \ + ".popsection\n" \ + EX_TABLE(0b,3b) EX_TABLE(1b,3b) \ + : "=d" (__rc), "=Q" (*(to)) \ + : "d" (size), "Q" (*(from)), \ + "d" (__reg0), "K" (-EFAULT) \ + : "cc"); \ + __rc; \ +}) + +#define __put_user_fn(x, ptr, size) __put_get_user_asm(ptr, x, size, 0x810000UL) +#define __get_user_fn(x, ptr, size) __put_get_user_asm(x, ptr, size, 0x81UL) + +#else /* CONFIG_HAVE_MARCH_Z10_FEATURES */ + static inline int __put_user_fn(void *x, void __user *ptr, unsigned long size) { size = __copy_to_user(ptr, x, size); @@ -144,6 +172,8 @@ static inline int __get_user_fn(void *x, const void __user *ptr, unsigned long s return size ? -EFAULT : 0; } +#endif /* CONFIG_HAVE_MARCH_Z10_FEATURES */ + /* * These are the main single-value transfer routines. They automatically * use the right size if we just have the right pointer type. diff --git a/arch/s390/include/uapi/asm/kvm.h b/arch/s390/include/uapi/asm/kvm.h index c003c6a73b1e..0fc26430a1e5 100644 --- a/arch/s390/include/uapi/asm/kvm.h +++ b/arch/s390/include/uapi/asm/kvm.h @@ -15,6 +15,7 @@ #include <linux/types.h> #define __KVM_S390 +#define __KVM_HAVE_GUEST_DEBUG /* Device control API: s390-specific devices */ #define KVM_DEV_FLIC_GET_ALL_IRQS 1 @@ -54,6 +55,13 @@ struct kvm_s390_io_adapter_req { __u64 addr; }; +/* kvm attr_group on vm fd */ +#define KVM_S390_VM_MEM_CTRL 0 + +/* kvm attributes for mem_ctrl */ +#define KVM_S390_VM_MEM_ENABLE_CMMA 0 +#define KVM_S390_VM_MEM_CLR_CMMA 1 + /* for KVM_GET_REGS and KVM_SET_REGS */ struct kvm_regs { /* general purpose regs for s390 */ @@ -72,11 +80,31 @@ struct kvm_fpu { __u64 fprs[16]; }; +#define KVM_GUESTDBG_USE_HW_BP 0x00010000 + +#define KVM_HW_BP 1 +#define KVM_HW_WP_WRITE 2 +#define KVM_SINGLESTEP 4 + struct kvm_debug_exit_arch { + __u64 addr; + __u8 type; + __u8 pad[7]; /* Should be set to 0 */ +}; + +struct kvm_hw_breakpoint { + __u64 addr; + __u64 phys_addr; + __u64 len; + __u8 type; + __u8 pad[7]; /* Should be set to 0 */ }; /* for KVM_SET_GUEST_DEBUG */ struct kvm_guest_debug_arch { + __u32 nr_hw_bp; + __u32 pad; /* Should be set to 0 */ + struct kvm_hw_breakpoint __user *hw_bp; }; #define KVM_SYNC_PREFIX (1UL << 0) diff --git a/arch/s390/include/uapi/asm/sie.h b/arch/s390/include/uapi/asm/sie.h new file mode 100644 index 000000000000..3d97f610198d --- /dev/null +++ b/arch/s390/include/uapi/asm/sie.h @@ -0,0 +1,245 @@ +#ifndef _UAPI_ASM_S390_SIE_H +#define _UAPI_ASM_S390_SIE_H + +#include <asm/sigp.h> + +#define diagnose_codes \ + { 0x10, "DIAG (0x10) release pages" }, \ + { 0x44, "DIAG (0x44) time slice end" }, \ + { 0x9c, "DIAG (0x9c) time slice end directed" }, \ + { 0x204, "DIAG (0x204) logical-cpu utilization" }, \ + { 0x258, "DIAG (0x258) page-reference services" }, \ + { 0x308, "DIAG (0x308) ipl functions" }, \ + { 0x500, "DIAG (0x500) KVM virtio functions" }, \ + { 0x501, "DIAG (0x501) KVM breakpoint" } + +#define sigp_order_codes \ + { SIGP_SENSE, "SIGP sense" }, \ + { SIGP_EXTERNAL_CALL, "SIGP external call" }, \ + { SIGP_EMERGENCY_SIGNAL, "SIGP emergency signal" }, \ + { SIGP_STOP, "SIGP stop" }, \ + { SIGP_STOP_AND_STORE_STATUS, "SIGP stop and store status" }, \ + { SIGP_SET_ARCHITECTURE, "SIGP set architecture" }, \ + { SIGP_SET_PREFIX, "SIGP set prefix" }, \ + { SIGP_SENSE_RUNNING, "SIGP sense running" }, \ + { SIGP_RESTART, "SIGP restart" }, \ + { SIGP_INITIAL_CPU_RESET, "SIGP initial cpu reset" }, \ + { SIGP_STORE_STATUS_AT_ADDRESS, "SIGP store status at address" } + +#define icpt_prog_codes \ + { 0x0001, "Prog Operation" }, \ + { 0x0002, "Prog Privileged Operation" }, \ + { 0x0003, "Prog Execute" }, \ + { 0x0004, "Prog Protection" }, \ + { 0x0005, "Prog Addressing" }, \ + { 0x0006, "Prog Specification" }, \ + { 0x0007, "Prog Data" }, \ + { 0x0008, "Prog Fixedpoint overflow" }, \ + { 0x0009, "Prog Fixedpoint divide" }, \ + { 0x000A, "Prog Decimal overflow" }, \ + { 0x000B, "Prog Decimal divide" }, \ + { 0x000C, "Prog HFP exponent overflow" }, \ + { 0x000D, "Prog HFP exponent underflow" }, \ + { 0x000E, "Prog HFP significance" }, \ + { 0x000F, "Prog HFP divide" }, \ + { 0x0010, "Prog Segment translation" }, \ + { 0x0011, "Prog Page translation" }, \ + { 0x0012, "Prog Translation specification" }, \ + { 0x0013, "Prog Special operation" }, \ + { 0x0015, "Prog Operand" }, \ + { 0x0016, "Prog Trace table" }, \ + { 0x0017, "Prog ASNtranslation specification" }, \ + { 0x001C, "Prog Spaceswitch event" }, \ + { 0x001D, "Prog HFP square root" }, \ + { 0x001F, "Prog PCtranslation specification" }, \ + { 0x0020, "Prog AFX translation" }, \ + { 0x0021, "Prog ASX translation" }, \ + { 0x0022, "Prog LX translation" }, \ + { 0x0023, "Prog EX translation" }, \ + { 0x0024, "Prog Primary authority" }, \ + { 0x0025, "Prog Secondary authority" }, \ + { 0x0026, "Prog LFXtranslation exception" }, \ + { 0x0027, "Prog LSXtranslation exception" }, \ + { 0x0028, "Prog ALET specification" }, \ + { 0x0029, "Prog ALEN translation" }, \ + { 0x002A, "Prog ALE sequence" }, \ + { 0x002B, "Prog ASTE validity" }, \ + { 0x002C, "Prog ASTE sequence" }, \ + { 0x002D, "Prog Extended authority" }, \ + { 0x002E, "Prog LSTE sequence" }, \ + { 0x002F, "Prog ASTE instance" }, \ + { 0x0030, "Prog Stack full" }, \ + { 0x0031, "Prog Stack empty" }, \ + { 0x0032, "Prog Stack specification" }, \ + { 0x0033, "Prog Stack type" }, \ + { 0x0034, "Prog Stack operation" }, \ + { 0x0039, "Prog Region first translation" }, \ + { 0x003A, "Prog Region second translation" }, \ + { 0x003B, "Prog Region third translation" }, \ + { 0x0040, "Prog Monitor event" }, \ + { 0x0080, "Prog PER event" }, \ + { 0x0119, "Prog Crypto operation" } + +#define exit_code_ipa0(ipa0, opcode, mnemonic) \ + { (ipa0 << 8 | opcode), #ipa0 " " mnemonic } +#define exit_code(opcode, mnemonic) \ + { opcode, mnemonic } + +#define icpt_insn_codes \ + exit_code_ipa0(0x01, 0x01, "PR"), \ + exit_code_ipa0(0x01, 0x04, "PTFF"), \ + exit_code_ipa0(0x01, 0x07, "SCKPF"), \ + exit_code_ipa0(0xAA, 0x00, "RINEXT"), \ + exit_code_ipa0(0xAA, 0x01, "RION"), \ + exit_code_ipa0(0xAA, 0x02, "TRIC"), \ + exit_code_ipa0(0xAA, 0x03, "RIOFF"), \ + exit_code_ipa0(0xAA, 0x04, "RIEMIT"), \ + exit_code_ipa0(0xB2, 0x02, "STIDP"), \ + exit_code_ipa0(0xB2, 0x04, "SCK"), \ + exit_code_ipa0(0xB2, 0x05, "STCK"), \ + exit_code_ipa0(0xB2, 0x06, "SCKC"), \ + exit_code_ipa0(0xB2, 0x07, "STCKC"), \ + exit_code_ipa0(0xB2, 0x08, "SPT"), \ + exit_code_ipa0(0xB2, 0x09, "STPT"), \ + exit_code_ipa0(0xB2, 0x0d, "PTLB"), \ + exit_code_ipa0(0xB2, 0x10, "SPX"), \ + exit_code_ipa0(0xB2, 0x11, "STPX"), \ + exit_code_ipa0(0xB2, 0x12, "STAP"), \ + exit_code_ipa0(0xB2, 0x14, "SIE"), \ + exit_code_ipa0(0xB2, 0x16, "SETR"), \ + exit_code_ipa0(0xB2, 0x17, "STETR"), \ + exit_code_ipa0(0xB2, 0x18, "PC"), \ + exit_code_ipa0(0xB2, 0x20, "SERVC"), \ + exit_code_ipa0(0xB2, 0x28, "PT"), \ + exit_code_ipa0(0xB2, 0x29, "ISKE"), \ + exit_code_ipa0(0xB2, 0x2a, "RRBE"), \ + exit_code_ipa0(0xB2, 0x2b, "SSKE"), \ + exit_code_ipa0(0xB2, 0x2c, "TB"), \ + exit_code_ipa0(0xB2, 0x2e, "PGIN"), \ + exit_code_ipa0(0xB2, 0x2f, "PGOUT"), \ + exit_code_ipa0(0xB2, 0x30, "CSCH"), \ + exit_code_ipa0(0xB2, 0x31, "HSCH"), \ + exit_code_ipa0(0xB2, 0x32, "MSCH"), \ + exit_code_ipa0(0xB2, 0x33, "SSCH"), \ + exit_code_ipa0(0xB2, 0x34, "STSCH"), \ + exit_code_ipa0(0xB2, 0x35, "TSCH"), \ + exit_code_ipa0(0xB2, 0x36, "TPI"), \ + exit_code_ipa0(0xB2, 0x37, "SAL"), \ + exit_code_ipa0(0xB2, 0x38, "RSCH"), \ + exit_code_ipa0(0xB2, 0x39, "STCRW"), \ + exit_code_ipa0(0xB2, 0x3a, "STCPS"), \ + exit_code_ipa0(0xB2, 0x3b, "RCHP"), \ + exit_code_ipa0(0xB2, 0x3c, "SCHM"), \ + exit_code_ipa0(0xB2, 0x40, "BAKR"), \ + exit_code_ipa0(0xB2, 0x48, "PALB"), \ + exit_code_ipa0(0xB2, 0x4c, "TAR"), \ + exit_code_ipa0(0xB2, 0x50, "CSP"), \ + exit_code_ipa0(0xB2, 0x54, "MVPG"), \ + exit_code_ipa0(0xB2, 0x58, "BSG"), \ + exit_code_ipa0(0xB2, 0x5a, "BSA"), \ + exit_code_ipa0(0xB2, 0x5f, "CHSC"), \ + exit_code_ipa0(0xB2, 0x74, "SIGA"), \ + exit_code_ipa0(0xB2, 0x76, "XSCH"), \ + exit_code_ipa0(0xB2, 0x78, "STCKE"), \ + exit_code_ipa0(0xB2, 0x7c, "STCKF"), \ + exit_code_ipa0(0xB2, 0x7d, "STSI"), \ + exit_code_ipa0(0xB2, 0xb0, "STFLE"), \ + exit_code_ipa0(0xB2, 0xb1, "STFL"), \ + exit_code_ipa0(0xB2, 0xb2, "LPSWE"), \ + exit_code_ipa0(0xB2, 0xf8, "TEND"), \ + exit_code_ipa0(0xB2, 0xfc, "TABORT"), \ + exit_code_ipa0(0xB9, 0x1e, "KMAC"), \ + exit_code_ipa0(0xB9, 0x28, "PCKMO"), \ + exit_code_ipa0(0xB9, 0x2a, "KMF"), \ + exit_code_ipa0(0xB9, 0x2b, "KMO"), \ + exit_code_ipa0(0xB9, 0x2d, "KMCTR"), \ + exit_code_ipa0(0xB9, 0x2e, "KM"), \ + exit_code_ipa0(0xB9, 0x2f, "KMC"), \ + exit_code_ipa0(0xB9, 0x3e, "KIMD"), \ + exit_code_ipa0(0xB9, 0x3f, "KLMD"), \ + exit_code_ipa0(0xB9, 0x8a, "CSPG"), \ + exit_code_ipa0(0xB9, 0x8d, "EPSW"), \ + exit_code_ipa0(0xB9, 0x8e, "IDTE"), \ + exit_code_ipa0(0xB9, 0x8f, "CRDTE"), \ + exit_code_ipa0(0xB9, 0x9c, "EQBS"), \ + exit_code_ipa0(0xB9, 0xa2, "PTF"), \ + exit_code_ipa0(0xB9, 0xab, "ESSA"), \ + exit_code_ipa0(0xB9, 0xae, "RRBM"), \ + exit_code_ipa0(0xB9, 0xaf, "PFMF"), \ + exit_code_ipa0(0xE3, 0x03, "LRAG"), \ + exit_code_ipa0(0xE3, 0x13, "LRAY"), \ + exit_code_ipa0(0xE3, 0x25, "NTSTG"), \ + exit_code_ipa0(0xE5, 0x00, "LASP"), \ + exit_code_ipa0(0xE5, 0x01, "TPROT"), \ + exit_code_ipa0(0xE5, 0x60, "TBEGIN"), \ + exit_code_ipa0(0xE5, 0x61, "TBEGINC"), \ + exit_code_ipa0(0xEB, 0x25, "STCTG"), \ + exit_code_ipa0(0xEB, 0x2f, "LCTLG"), \ + exit_code_ipa0(0xEB, 0x60, "LRIC"), \ + exit_code_ipa0(0xEB, 0x61, "STRIC"), \ + exit_code_ipa0(0xEB, 0x62, "MRIC"), \ + exit_code_ipa0(0xEB, 0x8a, "SQBS"), \ + exit_code_ipa0(0xC8, 0x01, "ECTG"), \ + exit_code(0x0a, "SVC"), \ + exit_code(0x80, "SSM"), \ + exit_code(0x82, "LPSW"), \ + exit_code(0x83, "DIAG"), \ + exit_code(0xae, "SIGP"), \ + exit_code(0xac, "STNSM"), \ + exit_code(0xad, "STOSM"), \ + exit_code(0xb1, "LRA"), \ + exit_code(0xb6, "STCTL"), \ + exit_code(0xb7, "LCTL"), \ + exit_code(0xee, "PLO") + +#define sie_intercept_code \ + { 0x00, "Host interruption" }, \ + { 0x04, "Instruction" }, \ + { 0x08, "Program interruption" }, \ + { 0x0c, "Instruction and program interruption" }, \ + { 0x10, "External request" }, \ + { 0x14, "External interruption" }, \ + { 0x18, "I/O request" }, \ + { 0x1c, "Wait state" }, \ + { 0x20, "Validity" }, \ + { 0x28, "Stop request" }, \ + { 0x2c, "Operation exception" }, \ + { 0x38, "Partial-execution" }, \ + { 0x3c, "I/O interruption" }, \ + { 0x40, "I/O instruction" }, \ + { 0x48, "Timing subset" } + +/* + * This is the simple interceptable instructions decoder. + * + * It will be used as userspace interface and it can be used in places + * that does not allow to use general decoder functions, + * such as trace events declarations. + * + * Some userspace tools may want to parse this code + * and would be confused by switch(), if() and other statements, + * but they can understand conditional operator. + */ +#define INSN_DECODE_IPA0(ipa0, insn, rshift, mask) \ + (insn >> 56) == (ipa0) ? \ + ((ipa0 << 8) | ((insn >> rshift) & mask)) : + +#define INSN_DECODE(insn) (insn >> 56) + +/* + * The macro icpt_insn_decoder() takes an intercepted instruction + * and returns a key, which can be used to find a mnemonic name + * of the instruction in the icpt_insn_codes table. + */ +#define icpt_insn_decoder(insn) \ + INSN_DECODE_IPA0(0x01, insn, 48, 0xff) \ + INSN_DECODE_IPA0(0xaa, insn, 48, 0x0f) \ + INSN_DECODE_IPA0(0xb2, insn, 48, 0xff) \ + INSN_DECODE_IPA0(0xb9, insn, 48, 0xff) \ + INSN_DECODE_IPA0(0xe3, insn, 48, 0xff) \ + INSN_DECODE_IPA0(0xe5, insn, 48, 0xff) \ + INSN_DECODE_IPA0(0xeb, insn, 16, 0xff) \ + INSN_DECODE_IPA0(0xc8, insn, 48, 0x0f) \ + INSN_DECODE(insn) + +#endif /* _UAPI_ASM_S390_SIE_H */ diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index cc10cdd4d6a2..afe1715a4eb7 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -50,6 +50,7 @@ int main(void) DEFINE(__PT_INT_CODE, offsetof(struct pt_regs, int_code)); DEFINE(__PT_INT_PARM, offsetof(struct pt_regs, int_parm)); DEFINE(__PT_INT_PARM_LONG, offsetof(struct pt_regs, int_parm_long)); + DEFINE(__PT_FLAGS, offsetof(struct pt_regs, flags)); DEFINE(__PT_SIZE, sizeof(struct pt_regs)); BLANK(); DEFINE(__SF_BACKCHAIN, offsetof(struct stack_frame, back_chain)); @@ -89,16 +90,22 @@ int main(void) DEFINE(__LC_PGM_ILC, offsetof(struct _lowcore, pgm_ilc)); DEFINE(__LC_PGM_INT_CODE, offsetof(struct _lowcore, pgm_code)); DEFINE(__LC_TRANS_EXC_CODE, offsetof(struct _lowcore, trans_exc_code)); - DEFINE(__LC_PER_CAUSE, offsetof(struct _lowcore, per_perc_atmid)); + DEFINE(__LC_MON_CLASS_NR, offsetof(struct _lowcore, mon_class_num)); + DEFINE(__LC_PER_CODE, offsetof(struct _lowcore, per_code)); + DEFINE(__LC_PER_ATMID, offsetof(struct _lowcore, per_atmid)); DEFINE(__LC_PER_ADDRESS, offsetof(struct _lowcore, per_address)); - DEFINE(__LC_PER_PAID, offsetof(struct _lowcore, per_access_id)); - DEFINE(__LC_AR_MODE_ID, offsetof(struct _lowcore, ar_access_id)); + DEFINE(__LC_EXC_ACCESS_ID, offsetof(struct _lowcore, exc_access_id)); + DEFINE(__LC_PER_ACCESS_ID, offsetof(struct _lowcore, per_access_id)); + DEFINE(__LC_OP_ACCESS_ID, offsetof(struct _lowcore, op_access_id)); + DEFINE(__LC_AR_MODE_ID, offsetof(struct _lowcore, ar_mode_id)); + DEFINE(__LC_MON_CODE, offsetof(struct _lowcore, monitor_code)); DEFINE(__LC_SUBCHANNEL_ID, offsetof(struct _lowcore, subchannel_id)); DEFINE(__LC_SUBCHANNEL_NR, offsetof(struct _lowcore, subchannel_nr)); DEFINE(__LC_IO_INT_PARM, offsetof(struct _lowcore, io_int_parm)); DEFINE(__LC_IO_INT_WORD, offsetof(struct _lowcore, io_int_word)); DEFINE(__LC_STFL_FAC_LIST, offsetof(struct _lowcore, stfl_fac_list)); DEFINE(__LC_MCCK_CODE, offsetof(struct _lowcore, mcck_interruption_code)); + DEFINE(__LC_MCCK_EXT_DAM_CODE, offsetof(struct _lowcore, external_damage_code)); DEFINE(__LC_RST_OLD_PSW, offsetof(struct _lowcore, restart_old_psw)); DEFINE(__LC_EXT_OLD_PSW, offsetof(struct _lowcore, external_old_psw)); DEFINE(__LC_SVC_OLD_PSW, offsetof(struct _lowcore, svc_old_psw)); @@ -115,6 +122,7 @@ int main(void) DEFINE(__LC_SAVE_AREA_SYNC, offsetof(struct _lowcore, save_area_sync)); DEFINE(__LC_SAVE_AREA_ASYNC, offsetof(struct _lowcore, save_area_async)); DEFINE(__LC_SAVE_AREA_RESTART, offsetof(struct _lowcore, save_area_restart)); + DEFINE(__LC_CPU_FLAGS, offsetof(struct _lowcore, cpu_flags)); DEFINE(__LC_RETURN_PSW, offsetof(struct _lowcore, return_psw)); DEFINE(__LC_RETURN_MCCK_PSW, offsetof(struct _lowcore, return_mcck_psw)); DEFINE(__LC_SYNC_ENTER_TIMER, offsetof(struct _lowcore, sync_enter_timer)); @@ -142,7 +150,6 @@ int main(void) DEFINE(__LC_MCCK_CLOCK, offsetof(struct _lowcore, mcck_clock)); DEFINE(__LC_MACHINE_FLAGS, offsetof(struct _lowcore, machine_flags)); DEFINE(__LC_FTRACE_FUNC, offsetof(struct _lowcore, ftrace_func)); - DEFINE(__LC_IRB, offsetof(struct _lowcore, irb)); DEFINE(__LC_DUMP_REIPL, offsetof(struct _lowcore, ipib)); BLANK(); DEFINE(__LC_CPU_TIMER_SAVE_AREA, offsetof(struct _lowcore, cpu_timer_save_area)); @@ -156,6 +163,8 @@ int main(void) #ifdef CONFIG_32BIT DEFINE(SAVE_AREA_BASE, offsetof(struct _lowcore, extended_save_area_addr)); #else /* CONFIG_32BIT */ + DEFINE(__LC_DATA_EXC_CODE, offsetof(struct _lowcore, data_exc_code)); + DEFINE(__LC_MCCK_FAIL_STOR_ADDR, offsetof(struct _lowcore, failing_storage_address)); DEFINE(__LC_EXT_PARAMS2, offsetof(struct _lowcore, ext_params2)); DEFINE(SAVE_AREA_BASE, offsetof(struct _lowcore, floating_pt_save_area)); DEFINE(__LC_PASTE, offsetof(struct _lowcore, paste)); diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c index 7df5ed9f44d7..f204d6920368 100644 --- a/arch/s390/kernel/compat_signal.c +++ b/arch/s390/kernel/compat_signal.c @@ -213,7 +213,7 @@ static int restore_sigregs32(struct pt_regs *regs,_sigregs32 __user *sregs) sizeof(current->thread.fp_regs)); restore_fp_regs(current->thread.fp_regs.fprs); - clear_thread_flag(TIF_SYSCALL); /* No longer in a system call */ + clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */ return 0; } diff --git a/arch/s390/kernel/crash_dump.c b/arch/s390/kernel/crash_dump.c index d7658c4b2ed5..a3b9150e6802 100644 --- a/arch/s390/kernel/crash_dump.c +++ b/arch/s390/kernel/crash_dump.c @@ -13,6 +13,7 @@ #include <linux/slab.h> #include <linux/bootmem.h> #include <linux/elf.h> +#include <linux/memblock.h> #include <asm/os_info.h> #include <asm/elf.h> #include <asm/ipl.h> @@ -22,6 +23,24 @@ #define PTR_SUB(x, y) (((char *) (x)) - ((unsigned long) (y))) #define PTR_DIFF(x, y) ((unsigned long)(((char *) (x)) - ((unsigned long) (y)))) +static struct memblock_region oldmem_region; + +static struct memblock_type oldmem_type = { + .cnt = 1, + .max = 1, + .total_size = 0, + .regions = &oldmem_region, +}; + +#define for_each_dump_mem_range(i, nid, p_start, p_end, p_nid) \ + for (i = 0, __next_mem_range(&i, nid, &memblock.physmem, \ + &oldmem_type, p_start, \ + p_end, p_nid); \ + i != (u64)ULLONG_MAX; \ + __next_mem_range(&i, nid, &memblock.physmem, \ + &oldmem_type, \ + p_start, p_end, p_nid)) + struct dump_save_areas dump_save_areas; /* @@ -264,19 +283,6 @@ static void *kzalloc_panic(int len) } /* - * Get memory layout and create hole for oldmem - */ -static struct mem_chunk *get_memory_layout(void) -{ - struct mem_chunk *chunk_array; - - chunk_array = kzalloc_panic(MEMORY_CHUNKS * sizeof(struct mem_chunk)); - detect_memory_layout(chunk_array, 0); - create_mem_hole(chunk_array, OLDMEM_BASE, OLDMEM_SIZE); - return chunk_array; -} - -/* * Initialize ELF note */ static void *nt_init(void *buf, Elf64_Word type, void *desc, int d_len, @@ -490,52 +496,33 @@ static int get_cpu_cnt(void) */ static int get_mem_chunk_cnt(void) { - struct mem_chunk *chunk_array, *mem_chunk; - int i, cnt = 0; + int cnt = 0; + u64 idx; - chunk_array = get_memory_layout(); - for (i = 0; i < MEMORY_CHUNKS; i++) { - mem_chunk = &chunk_array[i]; - if (chunk_array[i].type != CHUNK_READ_WRITE && - chunk_array[i].type != CHUNK_READ_ONLY) - continue; - if (mem_chunk->size == 0) - continue; + for_each_dump_mem_range(idx, NUMA_NO_NODE, NULL, NULL, NULL) cnt++; - } - kfree(chunk_array); return cnt; } /* * Initialize ELF loads (new kernel) */ -static int loads_init(Elf64_Phdr *phdr, u64 loads_offset) +static void loads_init(Elf64_Phdr *phdr, u64 loads_offset) { - struct mem_chunk *chunk_array, *mem_chunk; - int i; + phys_addr_t start, end; + u64 idx; - chunk_array = get_memory_layout(); - for (i = 0; i < MEMORY_CHUNKS; i++) { - mem_chunk = &chunk_array[i]; - if (mem_chunk->size == 0) - continue; - if (chunk_array[i].type != CHUNK_READ_WRITE && - chunk_array[i].type != CHUNK_READ_ONLY) - continue; - else - phdr->p_filesz = mem_chunk->size; + for_each_dump_mem_range(idx, NUMA_NO_NODE, &start, &end, NULL) { + phdr->p_filesz = end - start; phdr->p_type = PT_LOAD; - phdr->p_offset = mem_chunk->addr; - phdr->p_vaddr = mem_chunk->addr; - phdr->p_paddr = mem_chunk->addr; - phdr->p_memsz = mem_chunk->size; + phdr->p_offset = start; + phdr->p_vaddr = start; + phdr->p_paddr = start; + phdr->p_memsz = end - start; phdr->p_flags = PF_R | PF_W | PF_X; phdr->p_align = PAGE_SIZE; phdr++; } - kfree(chunk_array); - return i; } /* @@ -584,6 +571,14 @@ int elfcorehdr_alloc(unsigned long long *addr, unsigned long long *size) /* If we cannot get HSA size for zfcpdump return error */ if (ipl_info.type == IPL_TYPE_FCP_DUMP && !sclp_get_hsa_size()) return -ENODEV; + + /* For kdump, exclude previous crashkernel memory */ + if (OLDMEM_BASE) { + oldmem_region.base = OLDMEM_BASE; + oldmem_region.size = OLDMEM_SIZE; + oldmem_type.total_size = OLDMEM_SIZE; + } + mem_chunk_cnt = get_mem_chunk_cnt(); alloc_size = 0x1000 + get_cpu_cnt() * 0x300 + diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index a734f3585ceb..0dff972a169c 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -258,13 +258,19 @@ static __init void setup_topology(void) static void early_pgm_check_handler(void) { const struct exception_table_entry *fixup; + unsigned long cr0, cr0_new; unsigned long addr; addr = S390_lowcore.program_old_psw.addr; fixup = search_exception_tables(addr & PSW_ADDR_INSN); if (!fixup) disabled_wait(0); + /* Disable low address protection before storing into lowcore. */ + __ctl_store(cr0, 0, 0); + cr0_new = cr0 & ~(1UL << 28); + __ctl_load(cr0_new, 0, 0); S390_lowcore.program_old_psw.addr = extable_fixup(fixup)|PSW_ADDR_AMODE; + __ctl_load(cr0, 0, 0); } static noinline __init void setup_lowcore_early(void) diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 1662038516c0..70203265196f 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -10,6 +10,7 @@ #include <linux/init.h> #include <linux/linkage.h> +#include <asm/processor.h> #include <asm/cache.h> #include <asm/errno.h> #include <asm/ptrace.h> @@ -37,18 +38,16 @@ __PT_R13 = __PT_GPRS + 524 __PT_R14 = __PT_GPRS + 56 __PT_R15 = __PT_GPRS + 60 -_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ - _TIF_MCCK_PENDING | _TIF_PER_TRAP | _TIF_ASCE) -_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ - _TIF_MCCK_PENDING | _TIF_ASCE) -_TIF_TRACE = (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | \ - _TIF_SYSCALL_TRACEPOINT) -_TIF_TRANSFER = (_TIF_MCCK_PENDING | _TIF_TLB_WAIT) - STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER STACK_SIZE = 1 << STACK_SHIFT STACK_INIT = STACK_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE +_TIF_WORK = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED) +_TIF_TRACE = (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | \ + _TIF_SYSCALL_TRACEPOINT) +_CIF_WORK = (_CIF_MCCK_PENDING | _CIF_ASCE) +_PIF_WORK = (_PIF_PER_TRAP) + #define BASED(name) name-system_call(%r13) .macro TRACE_IRQS_ON @@ -160,13 +159,7 @@ ENTRY(__switch_to) lctl %c4,%c4,__TASK_pid(%r3) # load pid to control reg. 4 mvc __LC_CURRENT_PID(4,%r0),__TASK_pid(%r3) # store pid of next l %r15,__THREAD_ksp(%r3) # load kernel stack of next - lhi %r6,_TIF_TRANSFER # transfer TIF bits - n %r6,__TI_flags(%r4) # isolate TIF bits - jz 0f - o %r6,__TI_flags(%r5) # set TIF bits of next - st %r6,__TI_flags(%r5) - ni __TI_flags+3(%r4),255-_TIF_TRANSFER # clear TIF bits of prev -0: lm %r6,%r15,__SF_GPRS(%r15) # load gprs of next task + lm %r6,%r15,__SF_GPRS(%r15) # load gprs of next task br %r14 __critical_start: @@ -181,6 +174,7 @@ sysc_stm: stm %r8,%r15,__LC_SAVE_AREA_SYNC l %r12,__LC_THREAD_INFO l %r13,__LC_SVC_NEW_PSW+4 + lhi %r14,_PIF_SYSCALL sysc_per: l %r15,__LC_KERNEL_STACK la %r11,STACK_FRAME_OVERHEAD(%r15) # pointer to pt_regs @@ -190,8 +184,8 @@ sysc_vtime: mvc __PT_R8(32,%r11),__LC_SAVE_AREA_SYNC mvc __PT_PSW(8,%r11),__LC_SVC_OLD_PSW mvc __PT_INT_CODE(4,%r11),__LC_SVC_ILC + st %r14,__PT_FLAGS(%r11) sysc_do_svc: - oi __TI_flags+3(%r12),_TIF_SYSCALL l %r10,__TI_sysc_table(%r12) # 31 bit system call table lh %r8,__PT_INT_CODE+2(%r11) sla %r8,2 # shift and test for svc0 @@ -207,7 +201,7 @@ sysc_nr_ok: st %r2,__PT_ORIG_GPR2(%r11) st %r7,STACK_FRAME_OVERHEAD(%r15) l %r9,0(%r8,%r10) # get system call addr. - tm __TI_flags+2(%r12),_TIF_TRACE >> 8 + tm __TI_flags+3(%r12),_TIF_TRACE jnz sysc_tracesys basr %r14,%r9 # call sys_xxxx st %r2,__PT_R2(%r11) # store return value @@ -217,9 +211,12 @@ sysc_return: sysc_tif: tm __PT_PSW+1(%r11),0x01 # returning to user ? jno sysc_restore - tm __TI_flags+3(%r12),_TIF_WORK_SVC - jnz sysc_work # check for work - ni __TI_flags+3(%r12),255-_TIF_SYSCALL + tm __PT_FLAGS+3(%r11),_PIF_WORK + jnz sysc_work + tm __TI_flags+3(%r12),_TIF_WORK + jnz sysc_work # check for thread work + tm __LC_CPU_FLAGS+3,_CIF_WORK + jnz sysc_work sysc_restore: mvc __LC_RETURN_PSW(8),__PT_PSW(%r11) stpt __LC_EXIT_TIMER @@ -231,17 +228,17 @@ sysc_done: # One of the work bits is on. Find out which one. # sysc_work: - tm __TI_flags+3(%r12),_TIF_MCCK_PENDING + tm __LC_CPU_FLAGS+3,_CIF_MCCK_PENDING jo sysc_mcck_pending tm __TI_flags+3(%r12),_TIF_NEED_RESCHED jo sysc_reschedule - tm __TI_flags+3(%r12),_TIF_PER_TRAP + tm __PT_FLAGS+3(%r11),_PIF_PER_TRAP jo sysc_singlestep tm __TI_flags+3(%r12),_TIF_SIGPENDING jo sysc_sigpending tm __TI_flags+3(%r12),_TIF_NOTIFY_RESUME jo sysc_notify_resume - tm __TI_flags+3(%r12),_TIF_ASCE + tm __LC_CPU_FLAGS+3,_CIF_ASCE jo sysc_uaccess j sysc_return # beware of critical section cleanup @@ -254,7 +251,7 @@ sysc_reschedule: br %r1 # call schedule # -# _TIF_MCCK_PENDING is set, call handler +# _CIF_MCCK_PENDING is set, call handler # sysc_mcck_pending: l %r1,BASED(.Lhandle_mcck) @@ -262,10 +259,10 @@ sysc_mcck_pending: br %r1 # TIF bit will be cleared by handler # -# _TIF_ASCE is set, load user space asce +# _CIF_ASCE is set, load user space asce # sysc_uaccess: - ni __TI_flags+3(%r12),255-_TIF_ASCE + ni __LC_CPU_FLAGS+3,255-_CIF_ASCE lctl %c1,%c1,__LC_USER_ASCE # load primary asce j sysc_return @@ -276,7 +273,7 @@ sysc_sigpending: lr %r2,%r11 # pass pointer to pt_regs l %r1,BASED(.Ldo_signal) basr %r14,%r1 # call do_signal - tm __TI_flags+3(%r12),_TIF_SYSCALL + tm __PT_FLAGS+3(%r11),_PIF_SYSCALL jno sysc_return lm %r2,%r7,__PT_R2(%r11) # load svc arguments l %r10,__TI_sysc_table(%r12) # 31 bit system call table @@ -297,10 +294,10 @@ sysc_notify_resume: br %r1 # call do_notify_resume # -# _TIF_PER_TRAP is set, call do_per_trap +# _PIF_PER_TRAP is set, call do_per_trap # sysc_singlestep: - ni __TI_flags+3(%r12),255-_TIF_PER_TRAP + ni __PT_FLAGS+3(%r11),255-_PIF_PER_TRAP lr %r2,%r11 # pass pointer to pt_regs l %r1,BASED(.Ldo_per_trap) la %r14,BASED(sysc_return) @@ -330,7 +327,7 @@ sysc_tracego: basr %r14,%r9 # call sys_xxx st %r2,__PT_R2(%r11) # store return value sysc_tracenogo: - tm __TI_flags+2(%r12),_TIF_TRACE >> 8 + tm __TI_flags+3(%r12),_TIF_TRACE jz sysc_return l %r1,BASED(.Ltrace_exit) lr %r2,%r11 # pass pointer to pt_regs @@ -384,15 +381,16 @@ ENTRY(pgm_check_handler) stm %r8,%r9,__PT_PSW(%r11) mvc __PT_INT_CODE(4,%r11),__LC_PGM_ILC mvc __PT_INT_PARM_LONG(4,%r11),__LC_TRANS_EXC_CODE + xc __PT_FLAGS(4,%r11),__PT_FLAGS(%r11) tm __LC_PGM_ILC+3,0x80 # check for per exception jz 0f l %r1,__TI_task(%r12) tmh %r8,0x0001 # kernel per event ? jz pgm_kprobe - oi __TI_flags+3(%r12),_TIF_PER_TRAP + oi __PT_FLAGS+3(%r11),_PIF_PER_TRAP mvc __THREAD_per_address(4,%r1),__LC_PER_ADDRESS - mvc __THREAD_per_cause(2,%r1),__LC_PER_CAUSE - mvc __THREAD_per_paid(1,%r1),__LC_PER_PAID + mvc __THREAD_per_cause(2,%r1),__LC_PER_CODE + mvc __THREAD_per_paid(1,%r1),__LC_PER_ACCESS_ID 0: REENABLE_IRQS xc __SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15) l %r1,BASED(.Ljump_table) @@ -420,9 +418,9 @@ pgm_kprobe: # single stepped system call # pgm_svcper: - oi __TI_flags+3(%r12),_TIF_PER_TRAP mvc __LC_RETURN_PSW(4),__LC_SVC_NEW_PSW mvc __LC_RETURN_PSW+4(4),BASED(.Lsysc_per) + lhi %r14,_PIF_SYSCALL | _PIF_PER_TRAP lpsw __LC_RETURN_PSW # branch to sysc_per and enable irqs /* @@ -445,6 +443,7 @@ io_skip: mvc __PT_R8(32,%r11),__LC_SAVE_AREA_ASYNC stm %r8,%r9,__PT_PSW(%r11) mvc __PT_INT_CODE(12,%r11),__LC_SUBCHANNEL_ID + xc __PT_FLAGS(4,%r11),__PT_FLAGS(%r11) TRACE_IRQS_OFF xc __SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15) io_loop: @@ -466,8 +465,10 @@ io_return: LOCKDEP_SYS_EXIT TRACE_IRQS_ON io_tif: - tm __TI_flags+3(%r12),_TIF_WORK_INT + tm __TI_flags+3(%r12),_TIF_WORK jnz io_work # there is work to do (signals etc.) + tm __LC_CPU_FLAGS+3,_CIF_WORK + jnz io_work io_restore: mvc __LC_RETURN_PSW(8),__PT_PSW(%r11) stpt __LC_EXIT_TIMER @@ -477,7 +478,7 @@ io_done: # # There is work todo, find out in which context we have been interrupted: -# 1) if we return to user space we can do all _TIF_WORK_INT work +# 1) if we return to user space we can do all _TIF_WORK work # 2) if we return to kernel code and preemptive scheduling is enabled check # the preemption counter and if it is zero call preempt_schedule_irq # Before any work can be done, a switch to the kernel stack is required. @@ -520,11 +521,9 @@ io_work_user: # # One of the work bits is on. Find out which one. -# Checked are: _TIF_SIGPENDING, _TIF_NOTIFY_RESUME, _TIF_NEED_RESCHED -# and _TIF_MCCK_PENDING # io_work_tif: - tm __TI_flags+3(%r12),_TIF_MCCK_PENDING + tm __LC_CPU_FLAGS+3(%r12),_CIF_MCCK_PENDING jo io_mcck_pending tm __TI_flags+3(%r12),_TIF_NEED_RESCHED jo io_reschedule @@ -532,12 +531,12 @@ io_work_tif: jo io_sigpending tm __TI_flags+3(%r12),_TIF_NOTIFY_RESUME jo io_notify_resume - tm __TI_flags+3(%r12),_TIF_ASCE + tm __LC_CPU_FLAGS+3,_CIF_ASCE jo io_uaccess j io_return # beware of critical section cleanup # -# _TIF_MCCK_PENDING is set, call handler +# _CIF_MCCK_PENDING is set, call handler # io_mcck_pending: # TRACE_IRQS_ON already done at io_return @@ -547,10 +546,10 @@ io_mcck_pending: j io_return # -# _TIF_ASCE is set, load user space asce +# _CIF_ASCE is set, load user space asce # io_uaccess: - ni __TI_flags+3(%r12),255-_TIF_ASCE + ni __LC_CPU_FLAGS+3,255-_CIF_ASCE lctl %c1,%c1,__LC_USER_ASCE # load primary asce j io_return @@ -613,6 +612,7 @@ ext_skip: stm %r8,%r9,__PT_PSW(%r11) mvc __PT_INT_CODE(4,%r11),__LC_EXT_CPU_ADDR mvc __PT_INT_PARM(4,%r11),__LC_EXT_PARAMS + xc __PT_FLAGS(4,%r11),__PT_FLAGS(%r11) TRACE_IRQS_OFF l %r1,BASED(.Ldo_IRQ) lr %r2,%r11 # pass pointer to pt_regs @@ -677,6 +677,7 @@ mcck_skip: stm %r0,%r7,__PT_R0(%r11) mvc __PT_R8(32,%r11),__LC_GPREGS_SAVE_AREA+32 stm %r8,%r9,__PT_PSW(%r11) + xc __PT_FLAGS(4,%r11),__PT_FLAGS(%r11) xc __SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15) l %r1,BASED(.Ldo_machine_check) lr %r2,%r11 # pass pointer to pt_regs @@ -689,7 +690,7 @@ mcck_skip: la %r11,STACK_FRAME_OVERHEAD(%r15) lr %r15,%r1 ssm __LC_PGM_NEW_PSW # turn dat on, keep irqs off - tm __TI_flags+3(%r12),_TIF_MCCK_PENDING + tm __LC_CPU_FLAGS+3,_CIF_MCCK_PENDING jno mcck_return TRACE_IRQS_OFF l %r1,BASED(.Lhandle_mcck) @@ -842,6 +843,8 @@ cleanup_system_call: stm %r0,%r7,__PT_R0(%r9) mvc __PT_PSW(8,%r9),__LC_SVC_OLD_PSW mvc __PT_INT_CODE(4,%r9),__LC_SVC_ILC + xc __PT_FLAGS(4,%r9),__PT_FLAGS(%r9) + mvi __PT_FLAGS+3(%r9),_PIF_SYSCALL # setup saved register 15 st %r15,28(%r11) # r15 stack pointer # set new psw address and exit diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index 5963e43618bb..f2e674c702e1 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -42,13 +42,11 @@ STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER STACK_SIZE = 1 << STACK_SHIFT STACK_INIT = STACK_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE -_TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ - _TIF_MCCK_PENDING | _TIF_PER_TRAP | _TIF_ASCE) -_TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ - _TIF_MCCK_PENDING | _TIF_ASCE) -_TIF_TRACE = (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | \ - _TIF_SYSCALL_TRACEPOINT) -_TIF_TRANSFER = (_TIF_MCCK_PENDING | _TIF_TLB_WAIT) +_TIF_WORK = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED) +_TIF_TRACE = (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | \ + _TIF_SYSCALL_TRACEPOINT) +_CIF_WORK = (_CIF_MCCK_PENDING | _CIF_ASCE) +_PIF_WORK = (_PIF_PER_TRAP) #define BASED(name) name-system_call(%r13) @@ -190,13 +188,7 @@ ENTRY(__switch_to) lctl %c4,%c4,__TASK_pid(%r3) # load pid to control reg. 4 mvc __LC_CURRENT_PID+4(4,%r0),__TASK_pid(%r3) # store pid of next lg %r15,__THREAD_ksp(%r3) # load kernel stack of next - llill %r6,_TIF_TRANSFER # transfer TIF bits - ng %r6,__TI_flags(%r4) # isolate TIF bits - jz 0f - og %r6,__TI_flags(%r5) # set TIF bits of next - stg %r6,__TI_flags(%r5) - ni __TI_flags+7(%r4),255-_TIF_TRANSFER # clear TIF bits of prev -0: lmg %r6,%r15,__SF_GPRS(%r15) # load gprs of next task + lmg %r6,%r15,__SF_GPRS(%r15) # load gprs of next task br %r14 __critical_start: @@ -211,6 +203,7 @@ sysc_stmg: stmg %r8,%r15,__LC_SAVE_AREA_SYNC lg %r10,__LC_LAST_BREAK lg %r12,__LC_THREAD_INFO + lghi %r14,_PIF_SYSCALL sysc_per: lg %r15,__LC_KERNEL_STACK la %r11,STACK_FRAME_OVERHEAD(%r15) # pointer to pt_regs @@ -221,8 +214,8 @@ sysc_vtime: mvc __PT_R8(64,%r11),__LC_SAVE_AREA_SYNC mvc __PT_PSW(16,%r11),__LC_SVC_OLD_PSW mvc __PT_INT_CODE(4,%r11),__LC_SVC_ILC + stg %r14,__PT_FLAGS(%r11) sysc_do_svc: - oi __TI_flags+7(%r12),_TIF_SYSCALL lg %r10,__TI_sysc_table(%r12) # address of system call table llgh %r8,__PT_INT_CODE+2(%r11) slag %r8,%r8,2 # shift and test for svc 0 @@ -238,7 +231,7 @@ sysc_nr_ok: stg %r2,__PT_ORIG_GPR2(%r11) stg %r7,STACK_FRAME_OVERHEAD(%r15) lgf %r9,0(%r8,%r10) # get system call add. - tm __TI_flags+6(%r12),_TIF_TRACE >> 8 + tm __TI_flags+7(%r12),_TIF_TRACE jnz sysc_tracesys basr %r14,%r9 # call sys_xxxx stg %r2,__PT_R2(%r11) # store return value @@ -248,9 +241,12 @@ sysc_return: sysc_tif: tm __PT_PSW+1(%r11),0x01 # returning to user ? jno sysc_restore - tm __TI_flags+7(%r12),_TIF_WORK_SVC + tm __PT_FLAGS+7(%r11),_PIF_WORK + jnz sysc_work + tm __TI_flags+7(%r12),_TIF_WORK jnz sysc_work # check for work - ni __TI_flags+7(%r12),255-_TIF_SYSCALL + tm __LC_CPU_FLAGS+7,_CIF_WORK + jnz sysc_work sysc_restore: lg %r14,__LC_VDSO_PER_CPU lmg %r0,%r10,__PT_R0(%r11) @@ -265,17 +261,17 @@ sysc_done: # One of the work bits is on. Find out which one. # sysc_work: - tm __TI_flags+7(%r12),_TIF_MCCK_PENDING + tm __LC_CPU_FLAGS+7,_CIF_MCCK_PENDING jo sysc_mcck_pending tm __TI_flags+7(%r12),_TIF_NEED_RESCHED jo sysc_reschedule - tm __TI_flags+7(%r12),_TIF_PER_TRAP + tm __PT_FLAGS+7(%r11),_PIF_PER_TRAP jo sysc_singlestep tm __TI_flags+7(%r12),_TIF_SIGPENDING jo sysc_sigpending tm __TI_flags+7(%r12),_TIF_NOTIFY_RESUME jo sysc_notify_resume - tm __TI_flags+7(%r12),_TIF_ASCE + tm __LC_CPU_FLAGS+7,_CIF_ASCE jo sysc_uaccess j sysc_return # beware of critical section cleanup @@ -287,17 +283,17 @@ sysc_reschedule: jg schedule # -# _TIF_MCCK_PENDING is set, call handler +# _CIF_MCCK_PENDING is set, call handler # sysc_mcck_pending: larl %r14,sysc_return jg s390_handle_mcck # TIF bit will be cleared by handler # -# _TIF_ASCE is set, load user space asce +# _CIF_ASCE is set, load user space asce # sysc_uaccess: - ni __TI_flags+7(%r12),255-_TIF_ASCE + ni __LC_CPU_FLAGS+7,255-_CIF_ASCE lctlg %c1,%c1,__LC_USER_ASCE # load primary asce j sysc_return @@ -307,7 +303,7 @@ sysc_uaccess: sysc_sigpending: lgr %r2,%r11 # pass pointer to pt_regs brasl %r14,do_signal - tm __TI_flags+7(%r12),_TIF_SYSCALL + tm __PT_FLAGS+7(%r11),_PIF_SYSCALL jno sysc_return lmg %r2,%r7,__PT_R2(%r11) # load svc arguments lg %r10,__TI_sysc_table(%r12) # address of system call table @@ -327,10 +323,10 @@ sysc_notify_resume: jg do_notify_resume # -# _TIF_PER_TRAP is set, call do_per_trap +# _PIF_PER_TRAP is set, call do_per_trap # sysc_singlestep: - ni __TI_flags+7(%r12),255-_TIF_PER_TRAP + ni __PT_FLAGS+7(%r11),255-_PIF_PER_TRAP lgr %r2,%r11 # pass pointer to pt_regs larl %r14,sysc_return jg do_per_trap @@ -357,7 +353,7 @@ sysc_tracego: basr %r14,%r9 # call sys_xxx stg %r2,__PT_R2(%r11) # store return value sysc_tracenogo: - tm __TI_flags+6(%r12),_TIF_TRACE >> 8 + tm __TI_flags+7(%r12),_TIF_TRACE jz sysc_return lgr %r2,%r11 # pass pointer to pt_regs larl %r14,sysc_return @@ -416,15 +412,16 @@ ENTRY(pgm_check_handler) stmg %r8,%r9,__PT_PSW(%r11) mvc __PT_INT_CODE(4,%r11),__LC_PGM_ILC mvc __PT_INT_PARM_LONG(8,%r11),__LC_TRANS_EXC_CODE + xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) stg %r10,__PT_ARGS(%r11) tm __LC_PGM_ILC+3,0x80 # check for per exception jz 0f tmhh %r8,0x0001 # kernel per event ? jz pgm_kprobe - oi __TI_flags+7(%r12),_TIF_PER_TRAP + oi __PT_FLAGS+7(%r11),_PIF_PER_TRAP mvc __THREAD_per_address(8,%r14),__LC_PER_ADDRESS - mvc __THREAD_per_cause(2,%r14),__LC_PER_CAUSE - mvc __THREAD_per_paid(1,%r14),__LC_PER_PAID + mvc __THREAD_per_cause(2,%r14),__LC_PER_CODE + mvc __THREAD_per_paid(1,%r14),__LC_PER_ACCESS_ID 0: REENABLE_IRQS xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) larl %r1,pgm_check_table @@ -451,10 +448,10 @@ pgm_kprobe: # single stepped system call # pgm_svcper: - oi __TI_flags+7(%r12),_TIF_PER_TRAP mvc __LC_RETURN_PSW(8),__LC_SVC_NEW_PSW larl %r14,sysc_per stg %r14,__LC_RETURN_PSW+8 + lghi %r14,_PIF_SYSCALL | _PIF_PER_TRAP lpswe __LC_RETURN_PSW # branch to sysc_per and enable irqs /* @@ -479,6 +476,7 @@ io_skip: mvc __PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC stmg %r8,%r9,__PT_PSW(%r11) mvc __PT_INT_CODE(12,%r11),__LC_SUBCHANNEL_ID + xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) TRACE_IRQS_OFF xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) io_loop: @@ -499,8 +497,10 @@ io_return: LOCKDEP_SYS_EXIT TRACE_IRQS_ON io_tif: - tm __TI_flags+7(%r12),_TIF_WORK_INT + tm __TI_flags+7(%r12),_TIF_WORK jnz io_work # there is work to do (signals etc.) + tm __LC_CPU_FLAGS+7,_CIF_WORK + jnz io_work io_restore: lg %r14,__LC_VDSO_PER_CPU lmg %r0,%r10,__PT_R0(%r11) @@ -513,7 +513,7 @@ io_done: # # There is work todo, find out in which context we have been interrupted: -# 1) if we return to user space we can do all _TIF_WORK_INT work +# 1) if we return to user space we can do all _TIF_WORK work # 2) if we return to kernel code and kvm is enabled check if we need to # modify the psw to leave SIE # 3) if we return to kernel code and preemptive scheduling is enabled check @@ -557,11 +557,9 @@ io_work_user: # # One of the work bits is on. Find out which one. -# Checked are: _TIF_SIGPENDING, _TIF_NOTIFY_RESUME, _TIF_NEED_RESCHED -# and _TIF_MCCK_PENDING # io_work_tif: - tm __TI_flags+7(%r12),_TIF_MCCK_PENDING + tm __LC_CPU_FLAGS+7,_CIF_MCCK_PENDING jo io_mcck_pending tm __TI_flags+7(%r12),_TIF_NEED_RESCHED jo io_reschedule @@ -569,12 +567,12 @@ io_work_tif: jo io_sigpending tm __TI_flags+7(%r12),_TIF_NOTIFY_RESUME jo io_notify_resume - tm __TI_flags+7(%r12),_TIF_ASCE + tm __LC_CPU_FLAGS+7,_CIF_ASCE jo io_uaccess j io_return # beware of critical section cleanup # -# _TIF_MCCK_PENDING is set, call handler +# _CIF_MCCK_PENDING is set, call handler # io_mcck_pending: # TRACE_IRQS_ON already done at io_return @@ -583,10 +581,10 @@ io_mcck_pending: j io_return # -# _TIF_ASCE is set, load user space asce +# _CIF_ASCE is set, load user space asce # io_uaccess: - ni __TI_flags+7(%r12),255-_TIF_ASCE + ni __LC_CPU_FLAGS+7,255-_CIF_ASCE lctlg %c1,%c1,__LC_USER_ASCE # load primary asce j io_return @@ -650,6 +648,7 @@ ext_skip: mvc __PT_INT_CODE(4,%r11),__LC_EXT_CPU_ADDR mvc __PT_INT_PARM(4,%r11),__LC_EXT_PARAMS mvc __PT_INT_PARM_LONG(8,%r11),0(%r1) + xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) TRACE_IRQS_OFF xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) lgr %r2,%r11 # pass pointer to pt_regs @@ -716,6 +715,7 @@ mcck_skip: stmg %r0,%r7,__PT_R0(%r11) mvc __PT_R8(64,%r11),0(%r14) stmg %r8,%r9,__PT_PSW(%r11) + xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) lgr %r2,%r11 # pass pointer to pt_regs brasl %r14,s390_do_machine_check @@ -727,7 +727,7 @@ mcck_skip: la %r11,STACK_FRAME_OVERHEAD(%r1) lgr %r15,%r1 ssm __LC_PGM_NEW_PSW # turn dat on, keep irqs off - tm __TI_flags+7(%r12),_TIF_MCCK_PENDING + tm __LC_CPU_FLAGS+7,_CIF_MCCK_PENDING jno mcck_return TRACE_IRQS_OFF brasl %r14,s390_handle_mcck @@ -884,6 +884,8 @@ cleanup_system_call: stmg %r0,%r7,__PT_R0(%r9) mvc __PT_PSW(16,%r9),__LC_SVC_OLD_PSW mvc __PT_INT_CODE(4,%r9),__LC_SVC_ILC + xc __PT_FLAGS(8,%r9),__PT_FLAGS(%r9) + mvi __PT_FLAGS+7(%r9),_PIF_SYSCALL # setup saved register r15 stg %r15,56(%r11) # r15 stack pointer # set new psw address and exit diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S index 429afcc480cb..7ba7d6784510 100644 --- a/arch/s390/kernel/head.S +++ b/arch/s390/kernel/head.S @@ -437,13 +437,13 @@ ENTRY(startup_kdump) #if defined(CONFIG_64BIT) #if defined(CONFIG_MARCH_ZEC12) - .long 3, 0xc100efe3, 0xf46ce800, 0x00400000 + .long 3, 0xc100efea, 0xf46ce800, 0x00400000 #elif defined(CONFIG_MARCH_Z196) - .long 2, 0xc100efe3, 0xf46c0000 + .long 2, 0xc100efea, 0xf46c0000 #elif defined(CONFIG_MARCH_Z10) - .long 2, 0xc100efe3, 0xf0680000 + .long 2, 0xc100efea, 0xf0680000 #elif defined(CONFIG_MARCH_Z9_109) - .long 1, 0xc100efc3 + .long 1, 0xc100efc2 #elif defined(CONFIG_MARCH_Z990) .long 1, 0xc0002000 #elif defined(CONFIG_MARCH_Z900) diff --git a/arch/s390/kernel/head31.S b/arch/s390/kernel/head31.S index 9a99856df1c9..6dbe80983a24 100644 --- a/arch/s390/kernel/head31.S +++ b/arch/s390/kernel/head31.S @@ -59,7 +59,6 @@ ENTRY(startup_continue) .long 0 # cr13: home space segment table .long 0xc0000000 # cr14: machine check handling off .long 0 # cr15: linkage stack operations -.Lmchunk:.long memory_chunk .Lbss_bgn: .long __bss_start .Lbss_end: .long _end .Lparmaddr: .long PARMAREA diff --git a/arch/s390/kernel/irq.c b/arch/s390/kernel/irq.c index c7463aa0014b..99b0b09646ca 100644 --- a/arch/s390/kernel/irq.c +++ b/arch/s390/kernel/irq.c @@ -92,7 +92,6 @@ static const struct irq_class irqclass_sub_desc[NR_ARCH_IRQS] = { void __init init_IRQ(void) { - irq_reserve_irqs(0, THIN_INTERRUPT); init_cio_interrupts(); init_airq_interrupts(); init_ext_interrupts(); @@ -151,9 +150,9 @@ out: return 0; } -int arch_show_interrupts(struct seq_file *p, int prec) +unsigned int arch_dynirq_lower_bound(unsigned int from) { - return 0; + return from < THIN_INTERRUPT ? THIN_INTERRUPT : from; } /* diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c index c4c033819879..210e1285f75a 100644 --- a/arch/s390/kernel/nmi.c +++ b/arch/s390/kernel/nmi.c @@ -55,7 +55,7 @@ void s390_handle_mcck(void) local_mcck_disable(); mcck = __get_cpu_var(cpu_mcck); memset(&__get_cpu_var(cpu_mcck), 0, sizeof(struct mcck_struct)); - clear_thread_flag(TIF_MCCK_PENDING); + clear_cpu_flag(CIF_MCCK_PENDING); local_mcck_enable(); local_irq_restore(flags); @@ -313,7 +313,7 @@ void notrace s390_do_machine_check(struct pt_regs *regs) */ mcck->kill_task = 1; mcck->mcck_code = *(unsigned long long *) mci; - set_thread_flag(TIF_MCCK_PENDING); + set_cpu_flag(CIF_MCCK_PENDING); } else { /* * Couldn't restore all register contents while in @@ -352,12 +352,12 @@ void notrace s390_do_machine_check(struct pt_regs *regs) if (mci->cp) { /* Channel report word pending */ mcck->channel_report = 1; - set_thread_flag(TIF_MCCK_PENDING); + set_cpu_flag(CIF_MCCK_PENDING); } if (mci->w) { /* Warning pending */ mcck->warning = 1; - set_thread_flag(TIF_MCCK_PENDING); + set_cpu_flag(CIF_MCCK_PENDING); } nmi_exit(); } diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index dd145321d215..93b9ca42e5c0 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -64,7 +64,7 @@ unsigned long thread_saved_pc(struct task_struct *tsk) void arch_cpu_idle(void) { local_mcck_disable(); - if (test_thread_flag(TIF_MCCK_PENDING)) { + if (test_cpu_flag(CIF_MCCK_PENDING)) { local_mcck_enable(); local_irq_enable(); return; @@ -76,7 +76,7 @@ void arch_cpu_idle(void) void arch_cpu_idle_exit(void) { - if (test_thread_flag(TIF_MCCK_PENDING)) + if (test_cpu_flag(CIF_MCCK_PENDING)) s390_handle_mcck(); } @@ -123,7 +123,6 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp, memset(&p->thread.per_user, 0, sizeof(p->thread.per_user)); memset(&p->thread.per_event, 0, sizeof(p->thread.per_event)); clear_tsk_thread_flag(p, TIF_SINGLE_STEP); - clear_tsk_thread_flag(p, TIF_PER_TRAP); /* Initialize per thread user and system timer values */ ti = task_thread_info(p); ti->user_timer = 0; @@ -152,6 +151,7 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp, } frame->childregs = *current_pt_regs(); frame->childregs.gprs[2] = 0; /* child returns 0 on fork. */ + frame->childregs.flags = 0; if (new_stackp) frame->childregs.gprs[15] = new_stackp; diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 1c82619eb4f7..2d716734b5b1 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -136,7 +136,7 @@ void ptrace_disable(struct task_struct *task) memset(&task->thread.per_user, 0, sizeof(task->thread.per_user)); memset(&task->thread.per_event, 0, sizeof(task->thread.per_event)); clear_tsk_thread_flag(task, TIF_SINGLE_STEP); - clear_tsk_thread_flag(task, TIF_PER_TRAP); + clear_pt_regs_flag(task_pt_regs(task), PIF_PER_TRAP); task->thread.per_flags = 0; } @@ -813,7 +813,7 @@ asmlinkage long do_syscall_trace_enter(struct pt_regs *regs) * debugger stored an invalid system call number. Skip * the system call and the system call restart handling. */ - clear_thread_flag(TIF_SYSCALL); + clear_pt_regs_flag(regs, PIF_SYSCALL); ret = -1; } diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 88d1ca81e2dd..1e2264b46e4c 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -78,10 +78,9 @@ EXPORT_SYMBOL(console_irq); unsigned long elf_hwcap = 0; char elf_platform[ELF_PLATFORM_SIZE]; -struct mem_chunk __initdata memory_chunk[MEMORY_CHUNKS]; - int __initdata memory_end_set; unsigned long __initdata memory_end; +unsigned long __initdata max_physmem_end; unsigned long VMALLOC_START; EXPORT_SYMBOL(VMALLOC_START); @@ -212,7 +211,7 @@ static void __init conmode_default(void) } } -#ifdef CONFIG_ZFCPDUMP +#ifdef CONFIG_CRASH_DUMP static void __init setup_zfcpdump(void) { if (ipl_info.type != IPL_TYPE_FCP_DUMP) @@ -224,7 +223,7 @@ static void __init setup_zfcpdump(void) } #else static inline void setup_zfcpdump(void) {} -#endif /* CONFIG_ZFCPDUMP */ +#endif /* CONFIG_CRASH_DUMP */ /* * Reboot, halt and power_off stubs. They just call _machine_restart, @@ -273,6 +272,7 @@ EXPORT_SYMBOL_GPL(pm_power_off); static int __init early_parse_mem(char *p) { memory_end = memparse(p, &p); + memory_end &= PAGE_MASK; memory_end_set = 1; return 0; } @@ -373,6 +373,10 @@ static void __init setup_lowcore(void) mem_assign_absolute(S390_lowcore.restart_source, lc->restart_source); mem_assign_absolute(S390_lowcore.restart_psw, lc->restart_psw); +#ifdef CONFIG_SMP + lc->spinlock_lockval = arch_spin_lockval(0); +#endif + set_prefix((u32)(unsigned long) lc); lowcore_ptr[0] = lc; } @@ -401,7 +405,8 @@ static struct resource __initdata *standard_resources[] = { static void __init setup_resources(void) { struct resource *res, *std_res, *sub_res; - int i, j; + struct memblock_region *reg; + int j; code_resource.start = (unsigned long) &_text; code_resource.end = (unsigned long) &_etext - 1; @@ -410,24 +415,13 @@ static void __init setup_resources(void) bss_resource.start = (unsigned long) &__bss_start; bss_resource.end = (unsigned long) &__bss_stop - 1; - for (i = 0; i < MEMORY_CHUNKS; i++) { - if (!memory_chunk[i].size) - continue; + for_each_memblock(memory, reg) { res = alloc_bootmem_low(sizeof(*res)); res->flags = IORESOURCE_BUSY | IORESOURCE_MEM; - switch (memory_chunk[i].type) { - case CHUNK_READ_WRITE: - res->name = "System RAM"; - break; - case CHUNK_READ_ONLY: - res->name = "System ROM"; - res->flags |= IORESOURCE_READONLY; - break; - default: - res->name = "reserved"; - } - res->start = memory_chunk[i].addr; - res->end = res->start + memory_chunk[i].size - 1; + + res->name = "System RAM"; + res->start = reg->base; + res->end = reg->base + reg->size - 1; request_resource(&iomem_resource, res); for (j = 0; j < ARRAY_SIZE(standard_resources); j++) { @@ -451,48 +445,11 @@ static void __init setup_resources(void) static void __init setup_memory_end(void) { unsigned long vmax, vmalloc_size, tmp; - unsigned long real_memory_size = 0; - int i; - - -#ifdef CONFIG_ZFCPDUMP - if (ipl_info.type == IPL_TYPE_FCP_DUMP && - !OLDMEM_BASE && sclp_get_hsa_size()) { - memory_end = sclp_get_hsa_size(); - memory_end_set = 1; - } -#endif - memory_end &= PAGE_MASK; - - /* - * Make sure all chunks are MAX_ORDER aligned so we don't need the - * extra checks that HOLES_IN_ZONE would require. - */ - for (i = 0; i < MEMORY_CHUNKS; i++) { - unsigned long start, end; - struct mem_chunk *chunk; - unsigned long align; - - chunk = &memory_chunk[i]; - if (!chunk->size) - continue; - align = 1UL << (MAX_ORDER + PAGE_SHIFT - 1); - start = (chunk->addr + align - 1) & ~(align - 1); - end = (chunk->addr + chunk->size) & ~(align - 1); - if (start >= end) - memset(chunk, 0, sizeof(*chunk)); - else { - chunk->addr = start; - chunk->size = end - start; - } - real_memory_size = max(real_memory_size, - chunk->addr + chunk->size); - } /* Choose kernel address space layout: 2, 3, or 4 levels. */ #ifdef CONFIG_64BIT vmalloc_size = VMALLOC_END ?: (128UL << 30) - MODULES_LEN; - tmp = (memory_end ?: real_memory_size) / PAGE_SIZE; + tmp = (memory_end ?: max_physmem_end) / PAGE_SIZE; tmp = tmp * (sizeof(struct page) + PAGE_SIZE) + vmalloc_size; if (tmp <= (1UL << 42)) vmax = 1UL << 42; /* 3-level kernel page table */ @@ -520,21 +477,11 @@ static void __init setup_memory_end(void) vmemmap = (struct page *) tmp; /* Take care that memory_end is set and <= vmemmap */ - memory_end = min(memory_end ?: real_memory_size, tmp); - - /* Fixup memory chunk array to fit into 0..memory_end */ - for (i = 0; i < MEMORY_CHUNKS; i++) { - struct mem_chunk *chunk = &memory_chunk[i]; + memory_end = min(memory_end ?: max_physmem_end, tmp); + max_pfn = max_low_pfn = PFN_DOWN(memory_end); + memblock_remove(memory_end, ULONG_MAX); - if (!chunk->size) - continue; - if (chunk->addr >= memory_end) { - memset(chunk, 0, sizeof(*chunk)); - continue; - } - if (chunk->addr + chunk->size > memory_end) - chunk->size = memory_end - chunk->addr; - } + pr_notice("Max memory size: %luMB\n", memory_end >> 20); } static void __init setup_vmcoreinfo(void) @@ -545,89 +492,6 @@ static void __init setup_vmcoreinfo(void) #ifdef CONFIG_CRASH_DUMP /* - * Find suitable location for crashkernel memory - */ -static unsigned long __init find_crash_base(unsigned long crash_size, - char **msg) -{ - unsigned long crash_base; - struct mem_chunk *chunk; - int i; - - if (memory_chunk[0].size < crash_size) { - *msg = "first memory chunk must be at least crashkernel size"; - return 0; - } - if (OLDMEM_BASE && crash_size == OLDMEM_SIZE) - return OLDMEM_BASE; - - for (i = MEMORY_CHUNKS - 1; i >= 0; i--) { - chunk = &memory_chunk[i]; - if (chunk->size == 0) - continue; - if (chunk->type != CHUNK_READ_WRITE) - continue; - if (chunk->size < crash_size) - continue; - crash_base = (chunk->addr + chunk->size) - crash_size; - if (crash_base < crash_size) - continue; - if (crash_base < sclp_get_hsa_size()) - continue; - if (crash_base < (unsigned long) INITRD_START + INITRD_SIZE) - continue; - return crash_base; - } - *msg = "no suitable area found"; - return 0; -} - -/* - * Check if crash_base and crash_size is valid - */ -static int __init verify_crash_base(unsigned long crash_base, - unsigned long crash_size, - char **msg) -{ - struct mem_chunk *chunk; - int i; - - /* - * Because we do the swap to zero, we must have at least 'crash_size' - * bytes free space before crash_base - */ - if (crash_size > crash_base) { - *msg = "crashkernel offset must be greater than size"; - return -EINVAL; - } - - /* First memory chunk must be at least crash_size */ - if (memory_chunk[0].size < crash_size) { - *msg = "first memory chunk must be at least crashkernel size"; - return -EINVAL; - } - /* Check if we fit into the respective memory chunk */ - for (i = 0; i < MEMORY_CHUNKS; i++) { - chunk = &memory_chunk[i]; - if (chunk->size == 0) - continue; - if (crash_base < chunk->addr) - continue; - if (crash_base >= chunk->addr + chunk->size) - continue; - /* we have found the memory chunk */ - if (crash_base + crash_size > chunk->addr + chunk->size) { - *msg = "selected memory chunk is too small for " - "crashkernel memory"; - return -EINVAL; - } - return 0; - } - *msg = "invalid memory range specified"; - return -EINVAL; -} - -/* * When kdump is enabled, we have to ensure that no memory from * the area [0 - crashkernel memory size] and * [crashk_res.start - crashk_res.end] is set offline. @@ -653,23 +517,44 @@ static struct notifier_block kdump_mem_nb = { #endif /* + * Make sure that the area behind memory_end is protected + */ +static void reserve_memory_end(void) +{ +#ifdef CONFIG_CRASH_DUMP + if (ipl_info.type == IPL_TYPE_FCP_DUMP && + !OLDMEM_BASE && sclp_get_hsa_size()) { + memory_end = sclp_get_hsa_size(); + memory_end &= PAGE_MASK; + memory_end_set = 1; + } +#endif + if (!memory_end_set) + return; + memblock_reserve(memory_end, ULONG_MAX); +} + +/* * Make sure that oldmem, where the dump is stored, is protected */ static void reserve_oldmem(void) { #ifdef CONFIG_CRASH_DUMP - unsigned long real_size = 0; - int i; - - if (!OLDMEM_BASE) - return; - for (i = 0; i < MEMORY_CHUNKS; i++) { - struct mem_chunk *chunk = &memory_chunk[i]; + if (OLDMEM_BASE) + /* Forget all memory above the running kdump system */ + memblock_reserve(OLDMEM_SIZE, (phys_addr_t)ULONG_MAX); +#endif +} - real_size = max(real_size, chunk->addr + chunk->size); - } - create_mem_hole(memory_chunk, OLDMEM_BASE, OLDMEM_SIZE); - create_mem_hole(memory_chunk, OLDMEM_SIZE, real_size - OLDMEM_SIZE); +/* + * Make sure that oldmem, where the dump is stored, is protected + */ +static void remove_oldmem(void) +{ +#ifdef CONFIG_CRASH_DUMP + if (OLDMEM_BASE) + /* Forget all memory above the running kdump system */ + memblock_remove(OLDMEM_SIZE, (phys_addr_t)ULONG_MAX); #endif } @@ -680,167 +565,132 @@ static void __init reserve_crashkernel(void) { #ifdef CONFIG_CRASH_DUMP unsigned long long crash_base, crash_size; - char *msg = NULL; + phys_addr_t low, high; int rc; rc = parse_crashkernel(boot_command_line, memory_end, &crash_size, &crash_base); - if (rc || crash_size == 0) - return; + crash_base = ALIGN(crash_base, KEXEC_CRASH_MEM_ALIGN); crash_size = ALIGN(crash_size, KEXEC_CRASH_MEM_ALIGN); - if (register_memory_notifier(&kdump_mem_nb)) + if (rc || crash_size == 0) return; - if (!crash_base) - crash_base = find_crash_base(crash_size, &msg); - if (!crash_base) { - pr_info("crashkernel reservation failed: %s\n", msg); - unregister_memory_notifier(&kdump_mem_nb); + + if (memblock.memory.regions[0].size < crash_size) { + pr_info("crashkernel reservation failed: %s\n", + "first memory chunk must be at least crashkernel size"); return; } - if (verify_crash_base(crash_base, crash_size, &msg)) { - pr_info("crashkernel reservation failed: %s\n", msg); - unregister_memory_notifier(&kdump_mem_nb); + + low = crash_base ?: OLDMEM_BASE; + high = low + crash_size; + if (low >= OLDMEM_BASE && high <= OLDMEM_BASE + OLDMEM_SIZE) { + /* The crashkernel fits into OLDMEM, reuse OLDMEM */ + crash_base = low; + } else { + /* Find suitable area in free memory */ + low = max_t(unsigned long, crash_size, sclp_get_hsa_size()); + high = crash_base ? crash_base + crash_size : ULONG_MAX; + + if (crash_base && crash_base < low) { + pr_info("crashkernel reservation failed: %s\n", + "crash_base too low"); + return; + } + low = crash_base ?: low; + crash_base = memblock_find_in_range(low, high, crash_size, + KEXEC_CRASH_MEM_ALIGN); + } + + if (!crash_base) { + pr_info("crashkernel reservation failed: %s\n", + "no suitable area found"); return; } + + if (register_memory_notifier(&kdump_mem_nb)) + return; + if (!OLDMEM_BASE && MACHINE_IS_VM) diag10_range(PFN_DOWN(crash_base), PFN_DOWN(crash_size)); crashk_res.start = crash_base; crashk_res.end = crash_base + crash_size - 1; insert_resource(&iomem_resource, &crashk_res); - create_mem_hole(memory_chunk, crash_base, crash_size); + memblock_remove(crash_base, crash_size); pr_info("Reserving %lluMB of memory at %lluMB " "for crashkernel (System RAM: %luMB)\n", - crash_size >> 20, crash_base >> 20, memory_end >> 20); + crash_size >> 20, crash_base >> 20, + (unsigned long)memblock.memory.total_size >> 20); os_info_crashkernel_add(crash_base, crash_size); #endif } -static void __init setup_memory(void) +/* + * Reserve the initrd from being used by memblock + */ +static void __init reserve_initrd(void) { - unsigned long bootmap_size; - unsigned long start_pfn, end_pfn; - int i; +#ifdef CONFIG_BLK_DEV_INITRD + initrd_start = INITRD_START; + initrd_end = initrd_start + INITRD_SIZE; + memblock_reserve(INITRD_START, INITRD_SIZE); +#endif +} - /* - * partially used pages are not usable - thus - * we are rounding upwards: - */ +/* + * Check for initrd being in usable memory + */ +static void __init check_initrd(void) +{ +#ifdef CONFIG_BLK_DEV_INITRD + if (INITRD_START && INITRD_SIZE && + !memblock_is_region_memory(INITRD_START, INITRD_SIZE)) { + pr_err("initrd does not fit memory.\n"); + memblock_free(INITRD_START, INITRD_SIZE); + initrd_start = initrd_end = 0; + } +#endif +} + +/* + * Reserve all kernel text + */ +static void __init reserve_kernel(void) +{ + unsigned long start_pfn; start_pfn = PFN_UP(__pa(&_end)); - end_pfn = max_pfn = PFN_DOWN(memory_end); -#ifdef CONFIG_BLK_DEV_INITRD /* - * Move the initrd in case the bitmap of the bootmem allocater - * would overwrite it. + * Reserve memory used for lowcore/command line/kernel image. */ + memblock_reserve(0, (unsigned long)_ehead); + memblock_reserve((unsigned long)_stext, PFN_PHYS(start_pfn) + - (unsigned long)_stext); +} - if (INITRD_START && INITRD_SIZE) { - unsigned long bmap_size; - unsigned long start; - - bmap_size = bootmem_bootmap_pages(end_pfn - start_pfn + 1); - bmap_size = PFN_PHYS(bmap_size); - - if (PFN_PHYS(start_pfn) + bmap_size > INITRD_START) { - start = PFN_PHYS(start_pfn) + bmap_size + PAGE_SIZE; - +static void __init reserve_elfcorehdr(void) +{ #ifdef CONFIG_CRASH_DUMP - if (OLDMEM_BASE) { - /* Move initrd behind kdump oldmem */ - if (start + INITRD_SIZE > OLDMEM_BASE && - start < OLDMEM_BASE + OLDMEM_SIZE) - start = OLDMEM_BASE + OLDMEM_SIZE; - } -#endif - if (start + INITRD_SIZE > memory_end) { - pr_err("initrd extends beyond end of " - "memory (0x%08lx > 0x%08lx) " - "disabling initrd\n", - start + INITRD_SIZE, memory_end); - INITRD_START = INITRD_SIZE = 0; - } else { - pr_info("Moving initrd (0x%08lx -> " - "0x%08lx, size: %ld)\n", - INITRD_START, start, INITRD_SIZE); - memmove((void *) start, (void *) INITRD_START, - INITRD_SIZE); - INITRD_START = start; - } - } - } + if (is_kdump_kernel()) + memblock_reserve(elfcorehdr_addr - OLDMEM_BASE, + PAGE_ALIGN(elfcorehdr_size)); #endif +} - /* - * Initialize the boot-time allocator - */ - bootmap_size = init_bootmem(start_pfn, end_pfn); +static void __init setup_memory(void) +{ + struct memblock_region *reg; /* - * Register RAM areas with the bootmem allocator. + * Init storage key for present memory */ - - for (i = 0; i < MEMORY_CHUNKS; i++) { - unsigned long start_chunk, end_chunk, pfn; - - if (!memory_chunk[i].size) - continue; - start_chunk = PFN_DOWN(memory_chunk[i].addr); - end_chunk = start_chunk + PFN_DOWN(memory_chunk[i].size); - end_chunk = min(end_chunk, end_pfn); - if (start_chunk >= end_chunk) - continue; - memblock_add_node(PFN_PHYS(start_chunk), - PFN_PHYS(end_chunk - start_chunk), 0); - pfn = max(start_chunk, start_pfn); - storage_key_init_range(PFN_PHYS(pfn), PFN_PHYS(end_chunk)); + for_each_memblock(memory, reg) { + storage_key_init_range(reg->base, reg->base + reg->size); } - psw_set_key(PAGE_DEFAULT_KEY); - free_bootmem_with_active_regions(0, max_pfn); - - /* - * Reserve memory used for lowcore/command line/kernel image. - */ - reserve_bootmem(0, (unsigned long)_ehead, BOOTMEM_DEFAULT); - reserve_bootmem((unsigned long)_stext, - PFN_PHYS(start_pfn) - (unsigned long)_stext, - BOOTMEM_DEFAULT); - /* - * Reserve the bootmem bitmap itself as well. We do this in two - * steps (first step was init_bootmem()) because this catches - * the (very unlikely) case of us accidentally initializing the - * bootmem allocator with an invalid RAM area. - */ - reserve_bootmem(start_pfn << PAGE_SHIFT, bootmap_size, - BOOTMEM_DEFAULT); - -#ifdef CONFIG_CRASH_DUMP - if (crashk_res.start) - reserve_bootmem(crashk_res.start, - crashk_res.end - crashk_res.start + 1, - BOOTMEM_DEFAULT); - if (is_kdump_kernel()) - reserve_bootmem(elfcorehdr_addr - OLDMEM_BASE, - PAGE_ALIGN(elfcorehdr_size), BOOTMEM_DEFAULT); -#endif -#ifdef CONFIG_BLK_DEV_INITRD - if (INITRD_START && INITRD_SIZE) { - if (INITRD_START + INITRD_SIZE <= memory_end) { - reserve_bootmem(INITRD_START, INITRD_SIZE, - BOOTMEM_DEFAULT); - initrd_start = INITRD_START; - initrd_end = initrd_start + INITRD_SIZE; - } else { - pr_err("initrd extends beyond end of " - "memory (0x%08lx > 0x%08lx) " - "disabling initrd\n", - initrd_start + INITRD_SIZE, memory_end); - initrd_start = initrd_end = 0; - } - } -#endif + /* Only cosmetics */ + memblock_enforce_memory_limit(memblock_end_of_DRAM()); } /* @@ -989,23 +839,46 @@ void __init setup_arch(char **cmdline_p) ROOT_DEV = Root_RAM0; + /* Is init_mm really needed? */ init_mm.start_code = PAGE_OFFSET; init_mm.end_code = (unsigned long) &_etext; init_mm.end_data = (unsigned long) &_edata; init_mm.brk = (unsigned long) &_end; parse_early_param(); - detect_memory_layout(memory_chunk, memory_end); os_info_init(); setup_ipl(); + + /* Do some memory reservations *before* memory is added to memblock */ + reserve_memory_end(); reserve_oldmem(); + reserve_kernel(); + reserve_initrd(); + reserve_elfcorehdr(); + memblock_allow_resize(); + + /* Get information about *all* installed memory */ + detect_memory_memblock(); + + remove_oldmem(); + + /* + * Make sure all chunks are MAX_ORDER aligned so we don't need the + * extra checks that HOLES_IN_ZONE would require. + * + * Is this still required? + */ + memblock_trim_memory(1UL << (MAX_ORDER - 1 + PAGE_SHIFT)); + setup_memory_end(); - reserve_crashkernel(); setup_memory(); + + check_initrd(); + reserve_crashkernel(); + setup_resources(); setup_vmcoreinfo(); setup_lowcore(); - smp_fill_possible_mask(); cpu_init(); s390_init_cpu_topology(); diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c index d8fd508ccd1e..42b49f9e19bf 100644 --- a/arch/s390/kernel/signal.c +++ b/arch/s390/kernel/signal.c @@ -113,7 +113,7 @@ static int restore_sigregs(struct pt_regs *regs, _sigregs __user *sregs) sizeof(current->thread.fp_regs)); restore_fp_regs(current->thread.fp_regs.fprs); - clear_thread_flag(TIF_SYSCALL); /* No longer in a system call */ + clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */ return 0; } @@ -356,7 +356,7 @@ void do_signal(struct pt_regs *regs) * call information. */ current_thread_info()->system_call = - test_thread_flag(TIF_SYSCALL) ? regs->int_code : 0; + test_pt_regs_flag(regs, PIF_SYSCALL) ? regs->int_code : 0; signr = get_signal_to_deliver(&info, &ka, regs, NULL); if (signr > 0) { @@ -384,7 +384,7 @@ void do_signal(struct pt_regs *regs) } } /* No longer in a system call */ - clear_thread_flag(TIF_SYSCALL); + clear_pt_regs_flag(regs, PIF_SYSCALL); if (is_compat_task()) handle_signal32(signr, &ka, &info, oldset, regs); @@ -394,7 +394,7 @@ void do_signal(struct pt_regs *regs) } /* No handlers present - check for system call restart */ - clear_thread_flag(TIF_SYSCALL); + clear_pt_regs_flag(regs, PIF_SYSCALL); if (current_thread_info()->system_call) { regs->int_code = current_thread_info()->system_call; switch (regs->gprs[2]) { @@ -407,9 +407,9 @@ void do_signal(struct pt_regs *regs) case -ERESTARTNOINTR: /* Restart system call with magic TIF bit. */ regs->gprs[2] = regs->orig_gpr2; - set_thread_flag(TIF_SYSCALL); + set_pt_regs_flag(regs, PIF_SYSCALL); if (test_thread_flag(TIF_SINGLE_STEP)) - set_thread_flag(TIF_PER_TRAP); + clear_pt_regs_flag(regs, PIF_PER_TRAP); break; } } diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 86e65ec3422b..243c7e512600 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -170,6 +170,7 @@ static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu) lc->panic_stack = pcpu->panic_stack + PAGE_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs); lc->cpu_nr = cpu; + lc->spinlock_lockval = arch_spin_lockval(cpu); #ifndef CONFIG_64BIT if (MACHINE_HAS_IEEE) { lc->extended_save_area_addr = get_zeroed_page(GFP_KERNEL); @@ -226,6 +227,7 @@ static void pcpu_prepare_secondary(struct pcpu *pcpu, int cpu) 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; @@ -403,15 +405,6 @@ void smp_send_stop(void) } /* - * Stop the current cpu. - */ -void smp_stop_cpu(void) -{ - pcpu_sigp_retry(pcpu_devices + smp_processor_id(), SIGP_STOP, 0); - for (;;) ; -} - -/* * This is the main routine where commands issued by other * cpus are handled. */ @@ -519,7 +512,7 @@ void smp_ctl_clear_bit(int cr, int bit) } EXPORT_SYMBOL(smp_ctl_clear_bit); -#if defined(CONFIG_ZFCPDUMP) || defined(CONFIG_CRASH_DUMP) +#ifdef CONFIG_CRASH_DUMP static void __init smp_get_save_area(int cpu, u16 address) { @@ -534,14 +527,12 @@ static void __init smp_get_save_area(int cpu, u16 address) save_area = dump_save_area_create(cpu); if (!save_area) panic("could not allocate memory for save area\n"); -#ifdef CONFIG_CRASH_DUMP if (address == boot_cpu_address) { /* Copy the registers of the boot cpu. */ copy_oldmem_page(1, (void *) save_area, sizeof(*save_area), SAVE_AREA_BASE - PAGE_SIZE, 0); return; } -#endif /* Get the registers of a non-boot cpu. */ __pcpu_sigp_relax(address, SIGP_STOP_AND_STORE_STATUS, 0, NULL); memcpy_real(save_area, lc + SAVE_AREA_BASE, sizeof(*save_area)); @@ -558,11 +549,11 @@ int smp_store_status(int cpu) return 0; } -#else /* CONFIG_ZFCPDUMP || CONFIG_CRASH_DUMP */ +#else /* CONFIG_CRASH_DUMP */ static inline void smp_get_save_area(int cpu, u16 address) { } -#endif /* CONFIG_ZFCPDUMP || CONFIG_CRASH_DUMP */ +#endif /* CONFIG_CRASH_DUMP */ void smp_cpu_set_polarization(int cpu, int val) { @@ -809,6 +800,7 @@ 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); } /* diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 386d37a228bb..0931b110c826 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -226,7 +226,7 @@ void update_vsyscall(struct timekeeper *tk) vdso_data->wtom_clock_sec = tk->xtime_sec + tk->wall_to_monotonic.tv_sec; vdso_data->wtom_clock_nsec = tk->xtime_nsec + - + (tk->wall_to_monotonic.tv_nsec << tk->shift); + + ((u64) tk->wall_to_monotonic.tv_nsec << tk->shift); nsecps = (u64) NSEC_PER_SEC << tk->shift; while (vdso_data->wtom_clock_nsec >= nsecps) { vdso_data->wtom_clock_nsec -= nsecps; diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index 6298fed11ced..355a16c55702 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -333,7 +333,9 @@ static void __init alloc_masks(struct sysinfo_15_1_x *info, nr_masks *= info->mag[TOPOLOGY_NR_MAG - offset - 1 - i]; nr_masks = max(nr_masks, 1); for (i = 0; i < nr_masks; i++) { - mask->next = alloc_bootmem(sizeof(struct mask_info)); + mask->next = alloc_bootmem_align( + roundup_pow_of_two(sizeof(struct mask_info)), + roundup_pow_of_two(sizeof(struct mask_info))); mask = mask->next; } } @@ -443,6 +445,23 @@ int topology_cpu_init(struct cpu *cpu) return sysfs_create_group(&cpu->dev.kobj, &topology_cpu_attr_group); } +const struct cpumask *cpu_coregroup_mask(int cpu) +{ + return &cpu_topology[cpu].core_mask; +} + +static const struct cpumask *cpu_book_mask(int cpu) +{ + return &cpu_topology[cpu].book_mask; +} + +static struct sched_domain_topology_level s390_topology[] = { + { cpu_coregroup_mask, cpu_core_flags, SD_INIT_NAME(MC) }, + { cpu_book_mask, SD_INIT_NAME(BOOK) }, + { cpu_cpu_mask, SD_INIT_NAME(DIE) }, + { NULL, }, +}; + static int __init topology_init(void) { if (!MACHINE_HAS_TOPOLOGY) { @@ -451,6 +470,9 @@ static int __init topology_init(void) } set_topology_timer(); out: + + set_sched_topology(s390_topology); + return device_create_file(cpu_subsys.dev_root, &dev_attr_dispatching); } device_initcall(topology_init); diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile index d3adb37e93a4..b3b553469650 100644 --- a/arch/s390/kvm/Makefile +++ b/arch/s390/kvm/Makefile @@ -11,5 +11,7 @@ common-objs = $(KVM)/kvm_main.o $(KVM)/eventfd.o $(KVM)/async_pf.o $(KVM)/irqch ccflags-y := -Ivirt/kvm -Iarch/s390/kvm -kvm-objs := $(common-objs) kvm-s390.o intercept.o interrupt.o priv.o sigp.o diag.o +kvm-objs := $(common-objs) kvm-s390.o intercept.o interrupt.o priv.o sigp.o +kvm-objs += diag.o gaccess.o guestdbg.o + obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c index 08dfc839a6cf..0161675878a2 100644 --- a/arch/s390/kvm/diag.c +++ b/arch/s390/kvm/diag.c @@ -23,7 +23,7 @@ static int diag_release_pages(struct kvm_vcpu *vcpu) { unsigned long start, end; - unsigned long prefix = vcpu->arch.sie_block->prefix; + unsigned long prefix = kvm_s390_get_prefix(vcpu); start = vcpu->run->s.regs.gprs[(vcpu->arch.sie_block->ipa & 0xf0) >> 4]; end = vcpu->run->s.regs.gprs[vcpu->arch.sie_block->ipa & 0xf] + 4096; @@ -64,12 +64,12 @@ static int __diag_page_ref_service(struct kvm_vcpu *vcpu) int rc; u16 rx = (vcpu->arch.sie_block->ipa & 0xf0) >> 4; u16 ry = (vcpu->arch.sie_block->ipa & 0x0f); - unsigned long hva_token = KVM_HVA_ERR_BAD; if (vcpu->run->s.regs.gprs[rx] & 7) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); - if (copy_from_guest(vcpu, &parm, vcpu->run->s.regs.gprs[rx], sizeof(parm))) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + rc = read_guest(vcpu, vcpu->run->s.regs.gprs[rx], &parm, sizeof(parm)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); if (parm.parm_version != 2 || parm.parm_len < 5 || parm.code != 0x258) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); @@ -89,8 +89,7 @@ static int __diag_page_ref_service(struct kvm_vcpu *vcpu) parm.token_addr & 7 || parm.zarch != 0x8000000000000000ULL) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); - hva_token = gfn_to_hva(vcpu->kvm, gpa_to_gfn(parm.token_addr)); - if (kvm_is_error_hva(hva_token)) + if (kvm_is_error_gpa(vcpu->kvm, parm.token_addr)) return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); vcpu->arch.pfault_token = parm.token_addr; @@ -167,23 +166,17 @@ static int __diag_ipl_functions(struct kvm_vcpu *vcpu) VCPU_EVENT(vcpu, 5, "diag ipl functions, subcode %lx", subcode); switch (subcode) { - case 0: - case 1: - page_table_reset_pgste(current->mm, 0, TASK_SIZE); - return -EOPNOTSUPP; case 3: vcpu->run->s390_reset_flags = KVM_S390_RESET_CLEAR; - page_table_reset_pgste(current->mm, 0, TASK_SIZE); break; case 4: vcpu->run->s390_reset_flags = 0; - page_table_reset_pgste(current->mm, 0, TASK_SIZE); break; default: return -EOPNOTSUPP; } - atomic_set_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags); + kvm_s390_vcpu_stop(vcpu); vcpu->run->s390_reset_flags |= KVM_S390_RESET_SUBSYSTEM; vcpu->run->s390_reset_flags |= KVM_S390_RESET_IPL; vcpu->run->s390_reset_flags |= KVM_S390_RESET_CPU_INIT; diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c new file mode 100644 index 000000000000..4653ac6e182b --- /dev/null +++ b/arch/s390/kvm/gaccess.c @@ -0,0 +1,726 @@ +/* + * guest access functions + * + * Copyright IBM Corp. 2014 + * + */ + +#include <linux/vmalloc.h> +#include <linux/err.h> +#include <asm/pgtable.h> +#include "kvm-s390.h" +#include "gaccess.h" + +union asce { + unsigned long val; + struct { + unsigned long origin : 52; /* Region- or Segment-Table Origin */ + unsigned long : 2; + unsigned long g : 1; /* Subspace Group Control */ + unsigned long p : 1; /* Private Space Control */ + unsigned long s : 1; /* Storage-Alteration-Event Control */ + unsigned long x : 1; /* Space-Switch-Event Control */ + unsigned long r : 1; /* Real-Space Control */ + unsigned long : 1; + unsigned long dt : 2; /* Designation-Type Control */ + unsigned long tl : 2; /* Region- or Segment-Table Length */ + }; +}; + +enum { + ASCE_TYPE_SEGMENT = 0, + ASCE_TYPE_REGION3 = 1, + ASCE_TYPE_REGION2 = 2, + ASCE_TYPE_REGION1 = 3 +}; + +union region1_table_entry { + unsigned long val; + struct { + unsigned long rto: 52;/* Region-Table Origin */ + unsigned long : 2; + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 1; + unsigned long tf : 2; /* Region-Second-Table Offset */ + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long : 1; + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long tl : 2; /* Region-Second-Table Length */ + }; +}; + +union region2_table_entry { + unsigned long val; + struct { + unsigned long rto: 52;/* Region-Table Origin */ + unsigned long : 2; + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 1; + unsigned long tf : 2; /* Region-Third-Table Offset */ + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long : 1; + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long tl : 2; /* Region-Third-Table Length */ + }; +}; + +struct region3_table_entry_fc0 { + unsigned long sto: 52;/* Segment-Table Origin */ + unsigned long : 1; + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 1; + unsigned long tf : 2; /* Segment-Table Offset */ + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long cr : 1; /* Common-Region Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long tl : 2; /* Segment-Table Length */ +}; + +struct region3_table_entry_fc1 { + unsigned long rfaa : 33; /* Region-Frame Absolute Address */ + unsigned long : 14; + unsigned long av : 1; /* ACCF-Validity Control */ + unsigned long acc: 4; /* Access-Control Bits */ + unsigned long f : 1; /* Fetch-Protection Bit */ + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long co : 1; /* Change-Recording Override */ + unsigned long : 2; + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long cr : 1; /* Common-Region Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; +}; + +union region3_table_entry { + unsigned long val; + struct region3_table_entry_fc0 fc0; + struct region3_table_entry_fc1 fc1; + struct { + unsigned long : 53; + unsigned long fc : 1; /* Format-Control */ + unsigned long : 4; + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long cr : 1; /* Common-Region Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; + }; +}; + +struct segment_entry_fc0 { + unsigned long pto: 53;/* Page-Table Origin */ + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 3; + unsigned long i : 1; /* Segment-Invalid Bit */ + unsigned long cs : 1; /* Common-Segment Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; +}; + +struct segment_entry_fc1 { + unsigned long sfaa : 44; /* Segment-Frame Absolute Address */ + unsigned long : 3; + unsigned long av : 1; /* ACCF-Validity Control */ + unsigned long acc: 4; /* Access-Control Bits */ + unsigned long f : 1; /* Fetch-Protection Bit */ + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long co : 1; /* Change-Recording Override */ + unsigned long : 2; + unsigned long i : 1; /* Segment-Invalid Bit */ + unsigned long cs : 1; /* Common-Segment Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; +}; + +union segment_table_entry { + unsigned long val; + struct segment_entry_fc0 fc0; + struct segment_entry_fc1 fc1; + struct { + unsigned long : 53; + unsigned long fc : 1; /* Format-Control */ + unsigned long : 4; + unsigned long i : 1; /* Segment-Invalid Bit */ + unsigned long cs : 1; /* Common-Segment Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; + }; +}; + +enum { + TABLE_TYPE_SEGMENT = 0, + TABLE_TYPE_REGION3 = 1, + TABLE_TYPE_REGION2 = 2, + TABLE_TYPE_REGION1 = 3 +}; + +union page_table_entry { + unsigned long val; + struct { + unsigned long pfra : 52; /* Page-Frame Real Address */ + unsigned long z : 1; /* Zero Bit */ + unsigned long i : 1; /* Page-Invalid Bit */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long co : 1; /* Change-Recording Override */ + unsigned long : 8; + }; +}; + +/* + * vaddress union in order to easily decode a virtual address into its + * region first index, region second index etc. parts. + */ +union vaddress { + unsigned long addr; + struct { + unsigned long rfx : 11; + unsigned long rsx : 11; + unsigned long rtx : 11; + unsigned long sx : 11; + unsigned long px : 8; + unsigned long bx : 12; + }; + struct { + unsigned long rfx01 : 2; + unsigned long : 9; + unsigned long rsx01 : 2; + unsigned long : 9; + unsigned long rtx01 : 2; + unsigned long : 9; + unsigned long sx01 : 2; + unsigned long : 29; + }; +}; + +/* + * raddress union which will contain the result (real or absolute address) + * after a page table walk. The rfaa, sfaa and pfra members are used to + * simply assign them the value of a region, segment or page table entry. + */ +union raddress { + unsigned long addr; + unsigned long rfaa : 33; /* Region-Frame Absolute Address */ + unsigned long sfaa : 44; /* Segment-Frame Absolute Address */ + unsigned long pfra : 52; /* Page-Frame Real Address */ +}; + +static int ipte_lock_count; +static DEFINE_MUTEX(ipte_mutex); + +int ipte_lock_held(struct kvm_vcpu *vcpu) +{ + union ipte_control *ic = &vcpu->kvm->arch.sca->ipte_control; + + if (vcpu->arch.sie_block->eca & 1) + return ic->kh != 0; + return ipte_lock_count != 0; +} + +static void ipte_lock_simple(struct kvm_vcpu *vcpu) +{ + union ipte_control old, new, *ic; + + mutex_lock(&ipte_mutex); + ipte_lock_count++; + if (ipte_lock_count > 1) + goto out; + ic = &vcpu->kvm->arch.sca->ipte_control; + do { + old = ACCESS_ONCE(*ic); + while (old.k) { + cond_resched(); + old = ACCESS_ONCE(*ic); + } + new = old; + new.k = 1; + } while (cmpxchg(&ic->val, old.val, new.val) != old.val); +out: + mutex_unlock(&ipte_mutex); +} + +static void ipte_unlock_simple(struct kvm_vcpu *vcpu) +{ + union ipte_control old, new, *ic; + + mutex_lock(&ipte_mutex); + ipte_lock_count--; + if (ipte_lock_count) + goto out; + ic = &vcpu->kvm->arch.sca->ipte_control; + do { + new = old = ACCESS_ONCE(*ic); + new.k = 0; + } while (cmpxchg(&ic->val, old.val, new.val) != old.val); + if (!ipte_lock_count) + wake_up(&vcpu->kvm->arch.ipte_wq); +out: + mutex_unlock(&ipte_mutex); +} + +static void ipte_lock_siif(struct kvm_vcpu *vcpu) +{ + union ipte_control old, new, *ic; + + ic = &vcpu->kvm->arch.sca->ipte_control; + do { + old = ACCESS_ONCE(*ic); + while (old.kg) { + cond_resched(); + old = ACCESS_ONCE(*ic); + } + new = old; + new.k = 1; + new.kh++; + } while (cmpxchg(&ic->val, old.val, new.val) != old.val); +} + +static void ipte_unlock_siif(struct kvm_vcpu *vcpu) +{ + union ipte_control old, new, *ic; + + ic = &vcpu->kvm->arch.sca->ipte_control; + do { + new = old = ACCESS_ONCE(*ic); + new.kh--; + if (!new.kh) + new.k = 0; + } while (cmpxchg(&ic->val, old.val, new.val) != old.val); + if (!new.kh) + wake_up(&vcpu->kvm->arch.ipte_wq); +} + +void ipte_lock(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.sie_block->eca & 1) + ipte_lock_siif(vcpu); + else + ipte_lock_simple(vcpu); +} + +void ipte_unlock(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.sie_block->eca & 1) + ipte_unlock_siif(vcpu); + else + ipte_unlock_simple(vcpu); +} + +static unsigned long get_vcpu_asce(struct kvm_vcpu *vcpu) +{ + switch (psw_bits(vcpu->arch.sie_block->gpsw).as) { + case PSW_AS_PRIMARY: + return vcpu->arch.sie_block->gcr[1]; + case PSW_AS_SECONDARY: + return vcpu->arch.sie_block->gcr[7]; + case PSW_AS_HOME: + return vcpu->arch.sie_block->gcr[13]; + } + return 0; +} + +static int deref_table(struct kvm *kvm, unsigned long gpa, unsigned long *val) +{ + return kvm_read_guest(kvm, gpa, val, sizeof(*val)); +} + +/** + * guest_translate - translate a guest virtual into a guest absolute address + * @vcpu: virtual cpu + * @gva: guest virtual address + * @gpa: points to where guest physical (absolute) address should be stored + * @write: indicates if access is a write access + * + * Translate a guest virtual address into a guest absolute address by means + * of dynamic address translation as specified by the architecuture. + * If the resulting absolute address is not available in the configuration + * an addressing exception is indicated and @gpa will not be changed. + * + * Returns: - zero on success; @gpa contains the resulting absolute address + * - a negative value if guest access failed due to e.g. broken + * guest mapping + * - a positve value if an access exception happened. In this case + * the returned value is the program interruption code as defined + * by the architecture + */ +static unsigned long guest_translate(struct kvm_vcpu *vcpu, unsigned long gva, + unsigned long *gpa, int write) +{ + union vaddress vaddr = {.addr = gva}; + union raddress raddr = {.addr = gva}; + union page_table_entry pte; + int dat_protection = 0; + union ctlreg0 ctlreg0; + unsigned long ptr; + int edat1, edat2; + union asce asce; + + ctlreg0.val = vcpu->arch.sie_block->gcr[0]; + edat1 = ctlreg0.edat && test_vfacility(8); + edat2 = edat1 && test_vfacility(78); + asce.val = get_vcpu_asce(vcpu); + if (asce.r) + goto real_address; + ptr = asce.origin * 4096; + switch (asce.dt) { + case ASCE_TYPE_REGION1: + if (vaddr.rfx01 > asce.tl) + return PGM_REGION_FIRST_TRANS; + ptr += vaddr.rfx * 8; + break; + case ASCE_TYPE_REGION2: + if (vaddr.rfx) + return PGM_ASCE_TYPE; + if (vaddr.rsx01 > asce.tl) + return PGM_REGION_SECOND_TRANS; + ptr += vaddr.rsx * 8; + break; + case ASCE_TYPE_REGION3: + if (vaddr.rfx || vaddr.rsx) + return PGM_ASCE_TYPE; + if (vaddr.rtx01 > asce.tl) + return PGM_REGION_THIRD_TRANS; + ptr += vaddr.rtx * 8; + break; + case ASCE_TYPE_SEGMENT: + if (vaddr.rfx || vaddr.rsx || vaddr.rtx) + return PGM_ASCE_TYPE; + if (vaddr.sx01 > asce.tl) + return PGM_SEGMENT_TRANSLATION; + ptr += vaddr.sx * 8; + break; + } + switch (asce.dt) { + case ASCE_TYPE_REGION1: { + union region1_table_entry rfte; + + if (kvm_is_error_gpa(vcpu->kvm, ptr)) + return PGM_ADDRESSING; + if (deref_table(vcpu->kvm, ptr, &rfte.val)) + return -EFAULT; + if (rfte.i) + return PGM_REGION_FIRST_TRANS; + if (rfte.tt != TABLE_TYPE_REGION1) + return PGM_TRANSLATION_SPEC; + if (vaddr.rsx01 < rfte.tf || vaddr.rsx01 > rfte.tl) + return PGM_REGION_SECOND_TRANS; + if (edat1) + dat_protection |= rfte.p; + ptr = rfte.rto * 4096 + vaddr.rsx * 8; + } + /* fallthrough */ + case ASCE_TYPE_REGION2: { + union region2_table_entry rste; + + if (kvm_is_error_gpa(vcpu->kvm, ptr)) + return PGM_ADDRESSING; + if (deref_table(vcpu->kvm, ptr, &rste.val)) + return -EFAULT; + if (rste.i) + return PGM_REGION_SECOND_TRANS; + if (rste.tt != TABLE_TYPE_REGION2) + return PGM_TRANSLATION_SPEC; + if (vaddr.rtx01 < rste.tf || vaddr.rtx01 > rste.tl) + return PGM_REGION_THIRD_TRANS; + if (edat1) + dat_protection |= rste.p; + ptr = rste.rto * 4096 + vaddr.rtx * 8; + } + /* fallthrough */ + case ASCE_TYPE_REGION3: { + union region3_table_entry rtte; + + if (kvm_is_error_gpa(vcpu->kvm, ptr)) + return PGM_ADDRESSING; + if (deref_table(vcpu->kvm, ptr, &rtte.val)) + return -EFAULT; + if (rtte.i) + return PGM_REGION_THIRD_TRANS; + if (rtte.tt != TABLE_TYPE_REGION3) + return PGM_TRANSLATION_SPEC; + if (rtte.cr && asce.p && edat2) + return PGM_TRANSLATION_SPEC; + if (rtte.fc && edat2) { + dat_protection |= rtte.fc1.p; + raddr.rfaa = rtte.fc1.rfaa; + goto absolute_address; + } + if (vaddr.sx01 < rtte.fc0.tf) + return PGM_SEGMENT_TRANSLATION; + if (vaddr.sx01 > rtte.fc0.tl) + return PGM_SEGMENT_TRANSLATION; + if (edat1) + dat_protection |= rtte.fc0.p; + ptr = rtte.fc0.sto * 4096 + vaddr.sx * 8; + } + /* fallthrough */ + case ASCE_TYPE_SEGMENT: { + union segment_table_entry ste; + + if (kvm_is_error_gpa(vcpu->kvm, ptr)) + return PGM_ADDRESSING; + if (deref_table(vcpu->kvm, ptr, &ste.val)) + return -EFAULT; + if (ste.i) + return PGM_SEGMENT_TRANSLATION; + if (ste.tt != TABLE_TYPE_SEGMENT) + return PGM_TRANSLATION_SPEC; + if (ste.cs && asce.p) + return PGM_TRANSLATION_SPEC; + if (ste.fc && edat1) { + dat_protection |= ste.fc1.p; + raddr.sfaa = ste.fc1.sfaa; + goto absolute_address; + } + dat_protection |= ste.fc0.p; + ptr = ste.fc0.pto * 2048 + vaddr.px * 8; + } + } + if (kvm_is_error_gpa(vcpu->kvm, ptr)) + return PGM_ADDRESSING; + if (deref_table(vcpu->kvm, ptr, &pte.val)) + return -EFAULT; + if (pte.i) + return PGM_PAGE_TRANSLATION; + if (pte.z) + return PGM_TRANSLATION_SPEC; + if (pte.co && !edat1) + return PGM_TRANSLATION_SPEC; + dat_protection |= pte.p; + raddr.pfra = pte.pfra; +real_address: + raddr.addr = kvm_s390_real_to_abs(vcpu, raddr.addr); +absolute_address: + if (write && dat_protection) + return PGM_PROTECTION; + if (kvm_is_error_gpa(vcpu->kvm, raddr.addr)) + return PGM_ADDRESSING; + *gpa = raddr.addr; + return 0; +} + +static inline int is_low_address(unsigned long ga) +{ + /* Check for address ranges 0..511 and 4096..4607 */ + return (ga & ~0x11fful) == 0; +} + +static int low_address_protection_enabled(struct kvm_vcpu *vcpu) +{ + union ctlreg0 ctlreg0 = {.val = vcpu->arch.sie_block->gcr[0]}; + psw_t *psw = &vcpu->arch.sie_block->gpsw; + union asce asce; + + if (!ctlreg0.lap) + return 0; + asce.val = get_vcpu_asce(vcpu); + if (psw_bits(*psw).t && asce.p) + return 0; + return 1; +} + +struct trans_exc_code_bits { + unsigned long addr : 52; /* Translation-exception Address */ + unsigned long fsi : 2; /* Access Exception Fetch/Store Indication */ + unsigned long : 7; + unsigned long b61 : 1; + unsigned long as : 2; /* ASCE Identifier */ +}; + +enum { + FSI_UNKNOWN = 0, /* Unknown wether fetch or store */ + FSI_STORE = 1, /* Exception was due to store operation */ + FSI_FETCH = 2 /* Exception was due to fetch operation */ +}; + +static int guest_page_range(struct kvm_vcpu *vcpu, unsigned long ga, + unsigned long *pages, unsigned long nr_pages, + int write) +{ + struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm; + psw_t *psw = &vcpu->arch.sie_block->gpsw; + struct trans_exc_code_bits *tec_bits; + int lap_enabled, rc; + + memset(pgm, 0, sizeof(*pgm)); + tec_bits = (struct trans_exc_code_bits *)&pgm->trans_exc_code; + tec_bits->fsi = write ? FSI_STORE : FSI_FETCH; + tec_bits->as = psw_bits(*psw).as; + lap_enabled = low_address_protection_enabled(vcpu); + while (nr_pages) { + ga = kvm_s390_logical_to_effective(vcpu, ga); + tec_bits->addr = ga >> PAGE_SHIFT; + if (write && lap_enabled && is_low_address(ga)) { + pgm->code = PGM_PROTECTION; + return pgm->code; + } + ga &= PAGE_MASK; + if (psw_bits(*psw).t) { + rc = guest_translate(vcpu, ga, pages, write); + if (rc < 0) + return rc; + if (rc == PGM_PROTECTION) + tec_bits->b61 = 1; + if (rc) + pgm->code = rc; + } else { + *pages = kvm_s390_real_to_abs(vcpu, ga); + if (kvm_is_error_gpa(vcpu->kvm, *pages)) + pgm->code = PGM_ADDRESSING; + } + if (pgm->code) + return pgm->code; + ga += PAGE_SIZE; + pages++; + nr_pages--; + } + return 0; +} + +int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, void *data, + unsigned long len, int write) +{ + psw_t *psw = &vcpu->arch.sie_block->gpsw; + unsigned long _len, nr_pages, gpa, idx; + unsigned long pages_array[2]; + unsigned long *pages; + int need_ipte_lock; + union asce asce; + int rc; + + if (!len) + return 0; + /* Access register mode is not supported yet. */ + if (psw_bits(*psw).t && psw_bits(*psw).as == PSW_AS_ACCREG) + return -EOPNOTSUPP; + nr_pages = (((ga & ~PAGE_MASK) + len - 1) >> PAGE_SHIFT) + 1; + pages = pages_array; + if (nr_pages > ARRAY_SIZE(pages_array)) + pages = vmalloc(nr_pages * sizeof(unsigned long)); + if (!pages) + return -ENOMEM; + asce.val = get_vcpu_asce(vcpu); + need_ipte_lock = psw_bits(*psw).t && !asce.r; + if (need_ipte_lock) + ipte_lock(vcpu); + rc = guest_page_range(vcpu, ga, pages, nr_pages, write); + for (idx = 0; idx < nr_pages && !rc; idx++) { + gpa = *(pages + idx) + (ga & ~PAGE_MASK); + _len = min(PAGE_SIZE - (gpa & ~PAGE_MASK), len); + if (write) + rc = kvm_write_guest(vcpu->kvm, gpa, data, _len); + else + rc = kvm_read_guest(vcpu->kvm, gpa, data, _len); + len -= _len; + ga += _len; + data += _len; + } + if (need_ipte_lock) + ipte_unlock(vcpu); + if (nr_pages > ARRAY_SIZE(pages_array)) + vfree(pages); + return rc; +} + +int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, + void *data, unsigned long len, int write) +{ + unsigned long _len, gpa; + int rc = 0; + + while (len && !rc) { + gpa = kvm_s390_real_to_abs(vcpu, gra); + _len = min(PAGE_SIZE - (gpa & ~PAGE_MASK), len); + if (write) + rc = write_guest_abs(vcpu, gpa, data, _len); + else + rc = read_guest_abs(vcpu, gpa, data, _len); + len -= _len; + gra += _len; + data += _len; + } + return rc; +} + +/** + * guest_translate_address - translate guest logical into guest absolute address + * + * Parameter semantics are the same as the ones from guest_translate. + * The memory contents at the guest address are not changed. + * + * Note: The IPTE lock is not taken during this function, so the caller + * has to take care of this. + */ +int guest_translate_address(struct kvm_vcpu *vcpu, unsigned long gva, + unsigned long *gpa, int write) +{ + struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm; + psw_t *psw = &vcpu->arch.sie_block->gpsw; + struct trans_exc_code_bits *tec; + union asce asce; + int rc; + + /* Access register mode is not supported yet. */ + if (psw_bits(*psw).t && psw_bits(*psw).as == PSW_AS_ACCREG) + return -EOPNOTSUPP; + + gva = kvm_s390_logical_to_effective(vcpu, gva); + memset(pgm, 0, sizeof(*pgm)); + tec = (struct trans_exc_code_bits *)&pgm->trans_exc_code; + tec->as = psw_bits(*psw).as; + tec->fsi = write ? FSI_STORE : FSI_FETCH; + tec->addr = gva >> PAGE_SHIFT; + if (is_low_address(gva) && low_address_protection_enabled(vcpu)) { + if (write) { + rc = pgm->code = PGM_PROTECTION; + return rc; + } + } + + asce.val = get_vcpu_asce(vcpu); + if (psw_bits(*psw).t && !asce.r) { /* Use DAT? */ + rc = guest_translate(vcpu, gva, gpa, write); + if (rc > 0) { + if (rc == PGM_PROTECTION) + tec->b61 = 1; + pgm->code = rc; + } + } else { + rc = 0; + *gpa = kvm_s390_real_to_abs(vcpu, gva); + if (kvm_is_error_gpa(vcpu->kvm, *gpa)) + rc = pgm->code = PGM_ADDRESSING; + } + + return rc; +} + +/** + * kvm_s390_check_low_addr_protection - check for low-address protection + * @ga: Guest address + * + * Checks whether an address is subject to low-address protection and set + * up vcpu->arch.pgm accordingly if necessary. + * + * Return: 0 if no protection exception, or PGM_PROTECTION if protected. + */ +int kvm_s390_check_low_addr_protection(struct kvm_vcpu *vcpu, unsigned long ga) +{ + struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm; + psw_t *psw = &vcpu->arch.sie_block->gpsw; + struct trans_exc_code_bits *tec_bits; + + if (!is_low_address(ga) || !low_address_protection_enabled(vcpu)) + return 0; + + memset(pgm, 0, sizeof(*pgm)); + tec_bits = (struct trans_exc_code_bits *)&pgm->trans_exc_code; + tec_bits->fsi = FSI_STORE; + tec_bits->as = psw_bits(*psw).as; + tec_bits->addr = ga >> PAGE_SHIFT; + pgm->code = PGM_PROTECTION; + + return pgm->code; +} diff --git a/arch/s390/kvm/gaccess.h b/arch/s390/kvm/gaccess.h index 374a439ccc60..0149cf15058a 100644 --- a/arch/s390/kvm/gaccess.h +++ b/arch/s390/kvm/gaccess.h @@ -1,7 +1,7 @@ /* * access guest memory * - * Copyright IBM Corp. 2008, 2009 + * Copyright IBM Corp. 2008, 2014 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License (version 2 only) @@ -15,100 +15,321 @@ #include <linux/compiler.h> #include <linux/kvm_host.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> +#include <linux/ptrace.h> #include "kvm-s390.h" -/* Convert real to absolute address by applying the prefix of the CPU */ +/** + * kvm_s390_real_to_abs - convert guest real address to guest absolute address + * @vcpu - guest virtual cpu + * @gra - guest real address + * + * Returns the guest absolute address that corresponds to the passed guest real + * address @gra of a virtual guest cpu by applying its prefix. + */ static inline unsigned long kvm_s390_real_to_abs(struct kvm_vcpu *vcpu, - unsigned long gaddr) + unsigned long gra) { - unsigned long prefix = vcpu->arch.sie_block->prefix; - if (gaddr < 2 * PAGE_SIZE) - gaddr += prefix; - else if (gaddr >= prefix && gaddr < prefix + 2 * PAGE_SIZE) - gaddr -= prefix; - return gaddr; + unsigned long prefix = kvm_s390_get_prefix(vcpu); + + if (gra < 2 * PAGE_SIZE) + gra += prefix; + else if (gra >= prefix && gra < prefix + 2 * PAGE_SIZE) + gra -= prefix; + return gra; } -static inline void __user *__gptr_to_uptr(struct kvm_vcpu *vcpu, - void __user *gptr, - int prefixing) +/** + * kvm_s390_logical_to_effective - convert guest logical to effective address + * @vcpu: guest virtual cpu + * @ga: guest logical address + * + * Convert a guest vcpu logical address to a guest vcpu effective address by + * applying the rules of the vcpu's addressing mode defined by PSW bits 31 + * and 32 (extendended/basic addressing mode). + * + * Depending on the vcpu's addressing mode the upper 40 bits (24 bit addressing + * mode), 33 bits (31 bit addressing mode) or no bits (64 bit addressing mode) + * of @ga will be zeroed and the remaining bits will be returned. + */ +static inline unsigned long kvm_s390_logical_to_effective(struct kvm_vcpu *vcpu, + unsigned long ga) { - unsigned long gaddr = (unsigned long) gptr; - unsigned long uaddr; - - if (prefixing) - gaddr = kvm_s390_real_to_abs(vcpu, gaddr); - uaddr = gmap_fault(gaddr, vcpu->arch.gmap); - if (IS_ERR_VALUE(uaddr)) - uaddr = -EFAULT; - return (void __user *)uaddr; + psw_t *psw = &vcpu->arch.sie_block->gpsw; + + if (psw_bits(*psw).eaba == PSW_AMODE_64BIT) + return ga; + if (psw_bits(*psw).eaba == PSW_AMODE_31BIT) + return ga & ((1UL << 31) - 1); + return ga & ((1UL << 24) - 1); } -#define get_guest(vcpu, x, gptr) \ -({ \ - __typeof__(gptr) __uptr = __gptr_to_uptr(vcpu, gptr, 1);\ - int __mask = sizeof(__typeof__(*(gptr))) - 1; \ - int __ret; \ - \ - if (IS_ERR((void __force *)__uptr)) { \ - __ret = PTR_ERR((void __force *)__uptr); \ - } else { \ - BUG_ON((unsigned long)__uptr & __mask); \ - __ret = get_user(x, __uptr); \ - } \ - __ret; \ -}) +/* + * put_guest_lc, read_guest_lc and write_guest_lc are guest access functions + * which shall only be used to access the lowcore of a vcpu. + * These functions should be used for e.g. interrupt handlers where no + * guest memory access protection facilities, like key or low address + * protection, are applicable. + * At a later point guest vcpu lowcore access should happen via pinned + * prefix pages, so that these pages can be accessed directly via the + * kernel mapping. All of these *_lc functions can be removed then. + */ -#define put_guest(vcpu, x, gptr) \ +/** + * put_guest_lc - write a simple variable to a guest vcpu's lowcore + * @vcpu: virtual cpu + * @x: value to copy to guest + * @gra: vcpu's destination guest real address + * + * Copies a simple value from kernel space to a guest vcpu's lowcore. + * The size of the variable may be 1, 2, 4 or 8 bytes. The destination + * must be located in the vcpu's lowcore. Otherwise the result is undefined. + * + * Returns zero on success or -EFAULT on error. + * + * Note: an error indicates that either the kernel is out of memory or + * the guest memory mapping is broken. In any case the best solution + * would be to terminate the guest. + * It is wrong to inject a guest exception. + */ +#define put_guest_lc(vcpu, x, gra) \ ({ \ - __typeof__(gptr) __uptr = __gptr_to_uptr(vcpu, gptr, 1);\ - int __mask = sizeof(__typeof__(*(gptr))) - 1; \ - int __ret; \ + struct kvm_vcpu *__vcpu = (vcpu); \ + __typeof__(*(gra)) __x = (x); \ + unsigned long __gpa; \ \ - if (IS_ERR((void __force *)__uptr)) { \ - __ret = PTR_ERR((void __force *)__uptr); \ - } else { \ - BUG_ON((unsigned long)__uptr & __mask); \ - __ret = put_user(x, __uptr); \ - } \ - __ret; \ + __gpa = (unsigned long)(gra); \ + __gpa += kvm_s390_get_prefix(__vcpu); \ + kvm_write_guest(__vcpu->kvm, __gpa, &__x, sizeof(__x)); \ }) -static inline int __copy_guest(struct kvm_vcpu *vcpu, unsigned long to, - unsigned long from, unsigned long len, - int to_guest, int prefixing) +/** + * write_guest_lc - copy data from kernel space to guest vcpu's lowcore + * @vcpu: virtual cpu + * @gra: vcpu's source guest real address + * @data: source address in kernel space + * @len: number of bytes to copy + * + * Copy data from kernel space to guest vcpu's lowcore. The entire range must + * be located within the vcpu's lowcore, otherwise the result is undefined. + * + * Returns zero on success or -EFAULT on error. + * + * Note: an error indicates that either the kernel is out of memory or + * the guest memory mapping is broken. In any case the best solution + * would be to terminate the guest. + * It is wrong to inject a guest exception. + */ +static inline __must_check +int write_guest_lc(struct kvm_vcpu *vcpu, unsigned long gra, void *data, + unsigned long len) +{ + unsigned long gpa = gra + kvm_s390_get_prefix(vcpu); + + return kvm_write_guest(vcpu->kvm, gpa, data, len); +} + +/** + * read_guest_lc - copy data from guest vcpu's lowcore to kernel space + * @vcpu: virtual cpu + * @gra: vcpu's source guest real address + * @data: destination address in kernel space + * @len: number of bytes to copy + * + * Copy data from guest vcpu's lowcore to kernel space. The entire range must + * be located within the vcpu's lowcore, otherwise the result is undefined. + * + * Returns zero on success or -EFAULT on error. + * + * Note: an error indicates that either the kernel is out of memory or + * the guest memory mapping is broken. In any case the best solution + * would be to terminate the guest. + * It is wrong to inject a guest exception. + */ +static inline __must_check +int read_guest_lc(struct kvm_vcpu *vcpu, unsigned long gra, void *data, + unsigned long len) +{ + unsigned long gpa = gra + kvm_s390_get_prefix(vcpu); + + return kvm_read_guest(vcpu->kvm, gpa, data, len); +} + +int guest_translate_address(struct kvm_vcpu *vcpu, unsigned long gva, + unsigned long *gpa, int write); + +int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, void *data, + unsigned long len, int write); + +int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, + void *data, unsigned long len, int write); + +/** + * write_guest - copy data from kernel space to guest space + * @vcpu: virtual cpu + * @ga: guest address + * @data: source address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @data (kernel space) to @ga (guest address). + * In order to copy data to guest space the PSW of the vcpu is inspected: + * If DAT is off data will be copied to guest real or absolute memory. + * If DAT is on data will be copied to the address space as specified by + * the address space bits of the PSW: + * Primary, secondory or home space (access register mode is currently not + * implemented). + * The addressing mode of the PSW is also inspected, so that address wrap + * around is taken into account for 24-, 31- and 64-bit addressing mode, + * if the to be copied data crosses page boundaries in guest address space. + * In addition also low address and DAT protection are inspected before + * copying any data (key protection is currently not implemented). + * + * This function modifies the 'struct kvm_s390_pgm_info pgm' member of @vcpu. + * In case of an access exception (e.g. protection exception) pgm will contain + * all data necessary so that a subsequent call to 'kvm_s390_inject_prog_vcpu()' + * will inject a correct exception into the guest. + * If no access exception happened, the contents of pgm are undefined when + * this function returns. + * + * Returns: - zero on success + * - a negative value if e.g. the guest mapping is broken or in + * case of out-of-memory. In this case the contents of pgm are + * undefined. Also parts of @data may have been copied to guest + * space. + * - a positive value if an access exception happened. In this case + * the returned value is the program interruption code and the + * contents of pgm may be used to inject an exception into the + * guest. No data has been copied to guest space. + * + * Note: in case an access exception is recognized no data has been copied to + * guest space (this is also true, if the to be copied data would cross + * one or more page boundaries in guest space). + * Therefore this function may be used for nullifying and suppressing + * instruction emulation. + * It may also be used for terminating instructions, if it is undefined + * if data has been changed in guest space in case of an exception. + */ +static inline __must_check +int write_guest(struct kvm_vcpu *vcpu, unsigned long ga, void *data, + unsigned long len) +{ + return access_guest(vcpu, ga, data, len, 1); +} + +/** + * read_guest - copy data from guest space to kernel space + * @vcpu: virtual cpu + * @ga: guest address + * @data: destination address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @ga (guest address) to @data (kernel space). + * + * The behaviour of read_guest is identical to write_guest, except that + * data will be copied from guest space to kernel space. + */ +static inline __must_check +int read_guest(struct kvm_vcpu *vcpu, unsigned long ga, void *data, + unsigned long len) +{ + return access_guest(vcpu, ga, data, len, 0); +} + +/** + * write_guest_abs - copy data from kernel space to guest space absolute + * @vcpu: virtual cpu + * @gpa: guest physical (absolute) address + * @data: source address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @data (kernel space) to @gpa (guest absolute address). + * It is up to the caller to ensure that the entire guest memory range is + * valid memory before calling this function. + * Guest low address and key protection are not checked. + * + * Returns zero on success or -EFAULT on error. + * + * If an error occurs data may have been copied partially to guest memory. + */ +static inline __must_check +int write_guest_abs(struct kvm_vcpu *vcpu, unsigned long gpa, void *data, + unsigned long len) +{ + return kvm_write_guest(vcpu->kvm, gpa, data, len); +} + +/** + * read_guest_abs - copy data from guest space absolute to kernel space + * @vcpu: virtual cpu + * @gpa: guest physical (absolute) address + * @data: destination address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @gpa (guest absolute address) to @data (kernel space). + * It is up to the caller to ensure that the entire guest memory range is + * valid memory before calling this function. + * Guest key protection is not checked. + * + * Returns zero on success or -EFAULT on error. + * + * If an error occurs data may have been copied partially to kernel space. + */ +static inline __must_check +int read_guest_abs(struct kvm_vcpu *vcpu, unsigned long gpa, void *data, + unsigned long len) +{ + return kvm_read_guest(vcpu->kvm, gpa, data, len); +} + +/** + * write_guest_real - copy data from kernel space to guest space real + * @vcpu: virtual cpu + * @gra: guest real address + * @data: source address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @data (kernel space) to @gra (guest real address). + * It is up to the caller to ensure that the entire guest memory range is + * valid memory before calling this function. + * Guest low address and key protection are not checked. + * + * Returns zero on success or -EFAULT on error. + * + * If an error occurs data may have been copied partially to guest memory. + */ +static inline __must_check +int write_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, void *data, + unsigned long len) +{ + return access_guest_real(vcpu, gra, data, len, 1); +} + +/** + * read_guest_real - copy data from guest space real to kernel space + * @vcpu: virtual cpu + * @gra: guest real address + * @data: destination address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @gra (guest real address) to @data (kernel space). + * It is up to the caller to ensure that the entire guest memory range is + * valid memory before calling this function. + * Guest key protection is not checked. + * + * Returns zero on success or -EFAULT on error. + * + * If an error occurs data may have been copied partially to kernel space. + */ +static inline __must_check +int read_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, void *data, + unsigned long len) { - unsigned long _len, rc; - void __user *uptr; - - while (len) { - uptr = to_guest ? (void __user *)to : (void __user *)from; - uptr = __gptr_to_uptr(vcpu, uptr, prefixing); - if (IS_ERR((void __force *)uptr)) - return -EFAULT; - _len = PAGE_SIZE - ((unsigned long)uptr & (PAGE_SIZE - 1)); - _len = min(_len, len); - if (to_guest) - rc = copy_to_user((void __user *) uptr, (void *)from, _len); - else - rc = copy_from_user((void *)to, (void __user *)uptr, _len); - if (rc) - return -EFAULT; - len -= _len; - from += _len; - to += _len; - } - return 0; + return access_guest_real(vcpu, gra, data, len, 0); } -#define copy_to_guest(vcpu, to, from, size) \ - __copy_guest(vcpu, to, (unsigned long)from, size, 1, 1) -#define copy_from_guest(vcpu, to, from, size) \ - __copy_guest(vcpu, (unsigned long)to, from, size, 0, 1) -#define copy_to_guest_absolute(vcpu, to, from, size) \ - __copy_guest(vcpu, to, (unsigned long)from, size, 1, 0) -#define copy_from_guest_absolute(vcpu, to, from, size) \ - __copy_guest(vcpu, (unsigned long)to, from, size, 0, 0) +void ipte_lock(struct kvm_vcpu *vcpu); +void ipte_unlock(struct kvm_vcpu *vcpu); +int ipte_lock_held(struct kvm_vcpu *vcpu); +int kvm_s390_check_low_addr_protection(struct kvm_vcpu *vcpu, unsigned long ga); #endif /* __KVM_S390_GACCESS_H */ diff --git a/arch/s390/kvm/guestdbg.c b/arch/s390/kvm/guestdbg.c new file mode 100644 index 000000000000..3e8d4092ce30 --- /dev/null +++ b/arch/s390/kvm/guestdbg.c @@ -0,0 +1,482 @@ +/* + * kvm guest debug support + * + * Copyright IBM Corp. 2014 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): David Hildenbrand <dahi@linux.vnet.ibm.com> + */ +#include <linux/kvm_host.h> +#include <linux/errno.h> +#include "kvm-s390.h" +#include "gaccess.h" + +/* + * Extends the address range given by *start and *stop to include the address + * range starting with estart and the length len. Takes care of overflowing + * intervals and tries to minimize the overall intervall size. + */ +static void extend_address_range(u64 *start, u64 *stop, u64 estart, int len) +{ + u64 estop; + + if (len > 0) + len--; + else + len = 0; + + estop = estart + len; + + /* 0-0 range represents "not set" */ + if ((*start == 0) && (*stop == 0)) { + *start = estart; + *stop = estop; + } else if (*start <= *stop) { + /* increase the existing range */ + if (estart < *start) + *start = estart; + if (estop > *stop) + *stop = estop; + } else { + /* "overflowing" interval, whereby *stop > *start */ + if (estart <= *stop) { + if (estop > *stop) + *stop = estop; + } else if (estop > *start) { + if (estart < *start) + *start = estart; + } + /* minimize the range */ + else if ((estop - *stop) < (*start - estart)) + *stop = estop; + else + *start = estart; + } +} + +#define MAX_INST_SIZE 6 + +static void enable_all_hw_bp(struct kvm_vcpu *vcpu) +{ + unsigned long start, len; + u64 *cr9 = &vcpu->arch.sie_block->gcr[9]; + u64 *cr10 = &vcpu->arch.sie_block->gcr[10]; + u64 *cr11 = &vcpu->arch.sie_block->gcr[11]; + int i; + + if (vcpu->arch.guestdbg.nr_hw_bp <= 0 || + vcpu->arch.guestdbg.hw_bp_info == NULL) + return; + + /* + * If the guest is not interrested in branching events, we can savely + * limit them to the PER address range. + */ + if (!(*cr9 & PER_EVENT_BRANCH)) + *cr9 |= PER_CONTROL_BRANCH_ADDRESS; + *cr9 |= PER_EVENT_IFETCH | PER_EVENT_BRANCH; + + for (i = 0; i < vcpu->arch.guestdbg.nr_hw_bp; i++) { + start = vcpu->arch.guestdbg.hw_bp_info[i].addr; + len = vcpu->arch.guestdbg.hw_bp_info[i].len; + + /* + * The instruction in front of the desired bp has to + * report instruction-fetching events + */ + if (start < MAX_INST_SIZE) { + len += start; + start = 0; + } else { + start -= MAX_INST_SIZE; + len += MAX_INST_SIZE; + } + + extend_address_range(cr10, cr11, start, len); + } +} + +static void enable_all_hw_wp(struct kvm_vcpu *vcpu) +{ + unsigned long start, len; + u64 *cr9 = &vcpu->arch.sie_block->gcr[9]; + u64 *cr10 = &vcpu->arch.sie_block->gcr[10]; + u64 *cr11 = &vcpu->arch.sie_block->gcr[11]; + int i; + + if (vcpu->arch.guestdbg.nr_hw_wp <= 0 || + vcpu->arch.guestdbg.hw_wp_info == NULL) + return; + + /* if host uses storage alternation for special address + * spaces, enable all events and give all to the guest */ + if (*cr9 & PER_EVENT_STORE && *cr9 & PER_CONTROL_ALTERATION) { + *cr9 &= ~PER_CONTROL_ALTERATION; + *cr10 = 0; + *cr11 = PSW_ADDR_INSN; + } else { + *cr9 &= ~PER_CONTROL_ALTERATION; + *cr9 |= PER_EVENT_STORE; + + for (i = 0; i < vcpu->arch.guestdbg.nr_hw_wp; i++) { + start = vcpu->arch.guestdbg.hw_wp_info[i].addr; + len = vcpu->arch.guestdbg.hw_wp_info[i].len; + + extend_address_range(cr10, cr11, start, len); + } + } +} + +void kvm_s390_backup_guest_per_regs(struct kvm_vcpu *vcpu) +{ + vcpu->arch.guestdbg.cr0 = vcpu->arch.sie_block->gcr[0]; + vcpu->arch.guestdbg.cr9 = vcpu->arch.sie_block->gcr[9]; + vcpu->arch.guestdbg.cr10 = vcpu->arch.sie_block->gcr[10]; + vcpu->arch.guestdbg.cr11 = vcpu->arch.sie_block->gcr[11]; +} + +void kvm_s390_restore_guest_per_regs(struct kvm_vcpu *vcpu) +{ + vcpu->arch.sie_block->gcr[0] = vcpu->arch.guestdbg.cr0; + vcpu->arch.sie_block->gcr[9] = vcpu->arch.guestdbg.cr9; + vcpu->arch.sie_block->gcr[10] = vcpu->arch.guestdbg.cr10; + vcpu->arch.sie_block->gcr[11] = vcpu->arch.guestdbg.cr11; +} + +void kvm_s390_patch_guest_per_regs(struct kvm_vcpu *vcpu) +{ + /* + * TODO: if guest psw has per enabled, otherwise 0s! + * This reduces the amount of reported events. + * Need to intercept all psw changes! + */ + + if (guestdbg_sstep_enabled(vcpu)) { + /* disable timer (clock-comparator) interrupts */ + vcpu->arch.sie_block->gcr[0] &= ~0x800ul; + vcpu->arch.sie_block->gcr[9] |= PER_EVENT_IFETCH; + vcpu->arch.sie_block->gcr[10] = 0; + vcpu->arch.sie_block->gcr[11] = PSW_ADDR_INSN; + } + + if (guestdbg_hw_bp_enabled(vcpu)) { + enable_all_hw_bp(vcpu); + enable_all_hw_wp(vcpu); + } + + /* TODO: Instruction-fetching-nullification not allowed for now */ + if (vcpu->arch.sie_block->gcr[9] & PER_EVENT_NULLIFICATION) + vcpu->arch.sie_block->gcr[9] &= ~PER_EVENT_NULLIFICATION; +} + +#define MAX_WP_SIZE 100 + +static int __import_wp_info(struct kvm_vcpu *vcpu, + struct kvm_hw_breakpoint *bp_data, + struct kvm_hw_wp_info_arch *wp_info) +{ + int ret = 0; + wp_info->len = bp_data->len; + wp_info->addr = bp_data->addr; + wp_info->phys_addr = bp_data->phys_addr; + wp_info->old_data = NULL; + + if (wp_info->len < 0 || wp_info->len > MAX_WP_SIZE) + return -EINVAL; + + wp_info->old_data = kmalloc(bp_data->len, GFP_KERNEL); + if (!wp_info->old_data) + return -ENOMEM; + /* try to backup the original value */ + ret = read_guest(vcpu, wp_info->phys_addr, wp_info->old_data, + wp_info->len); + if (ret) { + kfree(wp_info->old_data); + wp_info->old_data = NULL; + } + + return ret; +} + +#define MAX_BP_COUNT 50 + +int kvm_s390_import_bp_data(struct kvm_vcpu *vcpu, + struct kvm_guest_debug *dbg) +{ + int ret = 0, nr_wp = 0, nr_bp = 0, i, size; + struct kvm_hw_breakpoint *bp_data = NULL; + struct kvm_hw_wp_info_arch *wp_info = NULL; + struct kvm_hw_bp_info_arch *bp_info = NULL; + + if (dbg->arch.nr_hw_bp <= 0 || !dbg->arch.hw_bp) + return 0; + else if (dbg->arch.nr_hw_bp > MAX_BP_COUNT) + return -EINVAL; + + size = dbg->arch.nr_hw_bp * sizeof(struct kvm_hw_breakpoint); + bp_data = kmalloc(size, GFP_KERNEL); + if (!bp_data) { + ret = -ENOMEM; + goto error; + } + + if (copy_from_user(bp_data, dbg->arch.hw_bp, size)) { + ret = -EFAULT; + goto error; + } + + for (i = 0; i < dbg->arch.nr_hw_bp; i++) { + switch (bp_data[i].type) { + case KVM_HW_WP_WRITE: + nr_wp++; + break; + case KVM_HW_BP: + nr_bp++; + break; + default: + break; + } + } + + size = nr_wp * sizeof(struct kvm_hw_wp_info_arch); + if (size > 0) { + wp_info = kmalloc(size, GFP_KERNEL); + if (!wp_info) { + ret = -ENOMEM; + goto error; + } + } + size = nr_bp * sizeof(struct kvm_hw_bp_info_arch); + if (size > 0) { + bp_info = kmalloc(size, GFP_KERNEL); + if (!bp_info) { + ret = -ENOMEM; + goto error; + } + } + + for (nr_wp = 0, nr_bp = 0, i = 0; i < dbg->arch.nr_hw_bp; i++) { + switch (bp_data[i].type) { + case KVM_HW_WP_WRITE: + ret = __import_wp_info(vcpu, &bp_data[i], + &wp_info[nr_wp]); + if (ret) + goto error; + nr_wp++; + break; + case KVM_HW_BP: + bp_info[nr_bp].len = bp_data[i].len; + bp_info[nr_bp].addr = bp_data[i].addr; + nr_bp++; + break; + } + } + + vcpu->arch.guestdbg.nr_hw_bp = nr_bp; + vcpu->arch.guestdbg.hw_bp_info = bp_info; + vcpu->arch.guestdbg.nr_hw_wp = nr_wp; + vcpu->arch.guestdbg.hw_wp_info = wp_info; + return 0; +error: + kfree(bp_data); + kfree(wp_info); + kfree(bp_info); + return ret; +} + +void kvm_s390_clear_bp_data(struct kvm_vcpu *vcpu) +{ + int i; + struct kvm_hw_wp_info_arch *hw_wp_info = NULL; + + for (i = 0; i < vcpu->arch.guestdbg.nr_hw_wp; i++) { + hw_wp_info = &vcpu->arch.guestdbg.hw_wp_info[i]; + kfree(hw_wp_info->old_data); + hw_wp_info->old_data = NULL; + } + kfree(vcpu->arch.guestdbg.hw_wp_info); + vcpu->arch.guestdbg.hw_wp_info = NULL; + + kfree(vcpu->arch.guestdbg.hw_bp_info); + vcpu->arch.guestdbg.hw_bp_info = NULL; + + vcpu->arch.guestdbg.nr_hw_wp = 0; + vcpu->arch.guestdbg.nr_hw_bp = 0; +} + +static inline int in_addr_range(u64 addr, u64 a, u64 b) +{ + if (a <= b) + return (addr >= a) && (addr <= b); + else + /* "overflowing" interval */ + return (addr <= a) && (addr >= b); +} + +#define end_of_range(bp_info) (bp_info->addr + bp_info->len - 1) + +static struct kvm_hw_bp_info_arch *find_hw_bp(struct kvm_vcpu *vcpu, + unsigned long addr) +{ + struct kvm_hw_bp_info_arch *bp_info = vcpu->arch.guestdbg.hw_bp_info; + int i; + + if (vcpu->arch.guestdbg.nr_hw_bp == 0) + return NULL; + + for (i = 0; i < vcpu->arch.guestdbg.nr_hw_bp; i++) { + /* addr is directly the start or in the range of a bp */ + if (addr == bp_info->addr) + goto found; + if (bp_info->len > 0 && + in_addr_range(addr, bp_info->addr, end_of_range(bp_info))) + goto found; + + bp_info++; + } + + return NULL; +found: + return bp_info; +} + +static struct kvm_hw_wp_info_arch *any_wp_changed(struct kvm_vcpu *vcpu) +{ + int i; + struct kvm_hw_wp_info_arch *wp_info = NULL; + void *temp = NULL; + + if (vcpu->arch.guestdbg.nr_hw_wp == 0) + return NULL; + + for (i = 0; i < vcpu->arch.guestdbg.nr_hw_wp; i++) { + wp_info = &vcpu->arch.guestdbg.hw_wp_info[i]; + if (!wp_info || !wp_info->old_data || wp_info->len <= 0) + continue; + + temp = kmalloc(wp_info->len, GFP_KERNEL); + if (!temp) + continue; + + /* refetch the wp data and compare it to the old value */ + if (!read_guest(vcpu, wp_info->phys_addr, temp, + wp_info->len)) { + if (memcmp(temp, wp_info->old_data, wp_info->len)) { + kfree(temp); + return wp_info; + } + } + kfree(temp); + temp = NULL; + } + + return NULL; +} + +void kvm_s390_prepare_debug_exit(struct kvm_vcpu *vcpu) +{ + vcpu->run->exit_reason = KVM_EXIT_DEBUG; + vcpu->guest_debug &= ~KVM_GUESTDBG_EXIT_PENDING; +} + +#define per_bp_event(code) \ + (code & (PER_EVENT_IFETCH | PER_EVENT_BRANCH)) +#define per_write_wp_event(code) \ + (code & (PER_EVENT_STORE | PER_EVENT_STORE_REAL)) + +static int debug_exit_required(struct kvm_vcpu *vcpu) +{ + u32 perc = (vcpu->arch.sie_block->perc << 24); + struct kvm_debug_exit_arch *debug_exit = &vcpu->run->debug.arch; + struct kvm_hw_wp_info_arch *wp_info = NULL; + struct kvm_hw_bp_info_arch *bp_info = NULL; + unsigned long addr = vcpu->arch.sie_block->gpsw.addr; + unsigned long peraddr = vcpu->arch.sie_block->peraddr; + + if (guestdbg_hw_bp_enabled(vcpu)) { + if (per_write_wp_event(perc) && + vcpu->arch.guestdbg.nr_hw_wp > 0) { + wp_info = any_wp_changed(vcpu); + if (wp_info) { + debug_exit->addr = wp_info->addr; + debug_exit->type = KVM_HW_WP_WRITE; + goto exit_required; + } + } + if (per_bp_event(perc) && + vcpu->arch.guestdbg.nr_hw_bp > 0) { + bp_info = find_hw_bp(vcpu, addr); + /* remove duplicate events if PC==PER address */ + if (bp_info && (addr != peraddr)) { + debug_exit->addr = addr; + debug_exit->type = KVM_HW_BP; + vcpu->arch.guestdbg.last_bp = addr; + goto exit_required; + } + /* breakpoint missed */ + bp_info = find_hw_bp(vcpu, peraddr); + if (bp_info && vcpu->arch.guestdbg.last_bp != peraddr) { + debug_exit->addr = peraddr; + debug_exit->type = KVM_HW_BP; + goto exit_required; + } + } + } + if (guestdbg_sstep_enabled(vcpu) && per_bp_event(perc)) { + debug_exit->addr = addr; + debug_exit->type = KVM_SINGLESTEP; + goto exit_required; + } + + return 0; +exit_required: + return 1; +} + +#define guest_per_enabled(vcpu) \ + (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PER) + +static void filter_guest_per_event(struct kvm_vcpu *vcpu) +{ + u32 perc = vcpu->arch.sie_block->perc << 24; + u64 peraddr = vcpu->arch.sie_block->peraddr; + u64 addr = vcpu->arch.sie_block->gpsw.addr; + u64 cr9 = vcpu->arch.sie_block->gcr[9]; + u64 cr10 = vcpu->arch.sie_block->gcr[10]; + u64 cr11 = vcpu->arch.sie_block->gcr[11]; + /* filter all events, demanded by the guest */ + u32 guest_perc = perc & cr9 & PER_EVENT_MASK; + + if (!guest_per_enabled(vcpu)) + guest_perc = 0; + + /* filter "successful-branching" events */ + if (guest_perc & PER_EVENT_BRANCH && + cr9 & PER_CONTROL_BRANCH_ADDRESS && + !in_addr_range(addr, cr10, cr11)) + guest_perc &= ~PER_EVENT_BRANCH; + + /* filter "instruction-fetching" events */ + if (guest_perc & PER_EVENT_IFETCH && + !in_addr_range(peraddr, cr10, cr11)) + guest_perc &= ~PER_EVENT_IFETCH; + + /* All other PER events will be given to the guest */ + /* TODO: Check alterated address/address space */ + + vcpu->arch.sie_block->perc = guest_perc >> 24; + + if (!guest_perc) + vcpu->arch.sie_block->iprcc &= ~PGM_PER; +} + +void kvm_s390_handle_per_event(struct kvm_vcpu *vcpu) +{ + if (debug_exit_required(vcpu)) + vcpu->guest_debug |= KVM_GUESTDBG_EXIT_PENDING; + + filter_guest_per_event(vcpu); +} diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c index eeb1ac7d8fa4..a0b586c1913c 100644 --- a/arch/s390/kvm/intercept.c +++ b/arch/s390/kvm/intercept.c @@ -1,7 +1,7 @@ /* * in-kernel handling for sie intercepts * - * Copyright IBM Corp. 2008, 2009 + * Copyright IBM Corp. 2008, 2014 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License (version 2 only) @@ -16,6 +16,8 @@ #include <linux/pagemap.h> #include <asm/kvm_host.h> +#include <asm/asm-offsets.h> +#include <asm/irq.h> #include "kvm-s390.h" #include "gaccess.h" @@ -29,6 +31,7 @@ static const intercept_handler_t instruction_handlers[256] = { [0x83] = kvm_s390_handle_diag, [0xae] = kvm_s390_handle_sigp, [0xb2] = kvm_s390_handle_b2, + [0xb6] = kvm_s390_handle_stctl, [0xb7] = kvm_s390_handle_lctl, [0xb9] = kvm_s390_handle_b9, [0xe5] = kvm_s390_handle_e5, @@ -44,9 +47,6 @@ static int handle_noop(struct kvm_vcpu *vcpu) case 0x10: vcpu->stat.exit_external_request++; break; - case 0x14: - vcpu->stat.exit_external_interrupt++; - break; default: break; /* nothing */ } @@ -63,8 +63,7 @@ static int handle_stop(struct kvm_vcpu *vcpu) trace_kvm_s390_stop_request(vcpu->arch.local_int.action_bits); if (vcpu->arch.local_int.action_bits & ACTION_STOP_ON_STOP) { - atomic_set_mask(CPUSTAT_STOPPED, - &vcpu->arch.sie_block->cpuflags); + kvm_s390_vcpu_stop(vcpu); vcpu->arch.local_int.action_bits &= ~ACTION_STOP_ON_STOP; VCPU_EVENT(vcpu, 3, "%s", "cpu stopped"); rc = -EOPNOTSUPP; @@ -109,22 +108,120 @@ static int handle_instruction(struct kvm_vcpu *vcpu) return -EOPNOTSUPP; } +static void __extract_prog_irq(struct kvm_vcpu *vcpu, + struct kvm_s390_pgm_info *pgm_info) +{ + memset(pgm_info, 0, sizeof(struct kvm_s390_pgm_info)); + pgm_info->code = vcpu->arch.sie_block->iprcc; + + switch (vcpu->arch.sie_block->iprcc & ~PGM_PER) { + case PGM_AFX_TRANSLATION: + case PGM_ASX_TRANSLATION: + case PGM_EX_TRANSLATION: + case PGM_LFX_TRANSLATION: + case PGM_LSTE_SEQUENCE: + case PGM_LSX_TRANSLATION: + case PGM_LX_TRANSLATION: + case PGM_PRIMARY_AUTHORITY: + case PGM_SECONDARY_AUTHORITY: + case PGM_SPACE_SWITCH: + pgm_info->trans_exc_code = vcpu->arch.sie_block->tecmc; + break; + case PGM_ALEN_TRANSLATION: + case PGM_ALE_SEQUENCE: + case PGM_ASTE_INSTANCE: + case PGM_ASTE_SEQUENCE: + case PGM_ASTE_VALIDITY: + case PGM_EXTENDED_AUTHORITY: + pgm_info->exc_access_id = vcpu->arch.sie_block->eai; + break; + case PGM_ASCE_TYPE: + case PGM_PAGE_TRANSLATION: + case PGM_REGION_FIRST_TRANS: + case PGM_REGION_SECOND_TRANS: + case PGM_REGION_THIRD_TRANS: + case PGM_SEGMENT_TRANSLATION: + pgm_info->trans_exc_code = vcpu->arch.sie_block->tecmc; + pgm_info->exc_access_id = vcpu->arch.sie_block->eai; + pgm_info->op_access_id = vcpu->arch.sie_block->oai; + break; + case PGM_MONITOR: + pgm_info->mon_class_nr = vcpu->arch.sie_block->mcn; + pgm_info->mon_code = vcpu->arch.sie_block->tecmc; + break; + case PGM_DATA: + pgm_info->data_exc_code = vcpu->arch.sie_block->dxc; + break; + case PGM_PROTECTION: + pgm_info->trans_exc_code = vcpu->arch.sie_block->tecmc; + pgm_info->exc_access_id = vcpu->arch.sie_block->eai; + break; + default: + break; + } + + if (vcpu->arch.sie_block->iprcc & PGM_PER) { + pgm_info->per_code = vcpu->arch.sie_block->perc; + pgm_info->per_atmid = vcpu->arch.sie_block->peratmid; + pgm_info->per_address = vcpu->arch.sie_block->peraddr; + pgm_info->per_access_id = vcpu->arch.sie_block->peraid; + } +} + +/* + * restore ITDB to program-interruption TDB in guest lowcore + * and set TX abort indication if required +*/ +static int handle_itdb(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_itdb *itdb; + int rc; + + if (!IS_TE_ENABLED(vcpu) || !IS_ITDB_VALID(vcpu)) + return 0; + if (current->thread.per_flags & PER_FLAG_NO_TE) + return 0; + itdb = (struct kvm_s390_itdb *)vcpu->arch.sie_block->itdba; + rc = write_guest_lc(vcpu, __LC_PGM_TDB, itdb, sizeof(*itdb)); + if (rc) + return rc; + memset(itdb, 0, sizeof(*itdb)); + + return 0; +} + +#define per_event(vcpu) (vcpu->arch.sie_block->iprcc & PGM_PER) + static int handle_prog(struct kvm_vcpu *vcpu) { + struct kvm_s390_pgm_info pgm_info; + psw_t psw; + int rc; + vcpu->stat.exit_program_interruption++; - /* Restore ITDB to Program-Interruption TDB in guest memory */ - if (IS_TE_ENABLED(vcpu) && - !(current->thread.per_flags & PER_FLAG_NO_TE) && - IS_ITDB_VALID(vcpu)) { - copy_to_guest(vcpu, TDB_ADDR, vcpu->arch.sie_block->itdba, - sizeof(struct kvm_s390_itdb)); - memset((void *) vcpu->arch.sie_block->itdba, 0, - sizeof(struct kvm_s390_itdb)); + if (guestdbg_enabled(vcpu) && per_event(vcpu)) { + kvm_s390_handle_per_event(vcpu); + /* the interrupt might have been filtered out completely */ + if (vcpu->arch.sie_block->iprcc == 0) + return 0; } trace_kvm_s390_intercept_prog(vcpu, vcpu->arch.sie_block->iprcc); - return kvm_s390_inject_program_int(vcpu, vcpu->arch.sie_block->iprcc); + if (vcpu->arch.sie_block->iprcc == PGM_SPECIFICATION) { + rc = read_guest_lc(vcpu, __LC_PGM_NEW_PSW, &psw, sizeof(psw_t)); + if (rc) + return rc; + /* Avoid endless loops of specification exceptions */ + if (!is_valid_psw(&psw)) + return -EOPNOTSUPP; + } + rc = handle_itdb(vcpu); + if (rc) + return rc; + + __extract_prog_irq(vcpu, &pgm_info); + return kvm_s390_inject_prog_irq(vcpu, &pgm_info); } static int handle_instruction_and_prog(struct kvm_vcpu *vcpu) @@ -142,17 +239,110 @@ static int handle_instruction_and_prog(struct kvm_vcpu *vcpu) return rc2; } +/** + * handle_external_interrupt - used for external interruption interceptions + * + * This interception only occurs if the CPUSTAT_EXT_INT bit was set, or if + * the new PSW does not have external interrupts disabled. In the first case, + * we've got to deliver the interrupt manually, and in the second case, we + * drop to userspace to handle the situation there. + */ +static int handle_external_interrupt(struct kvm_vcpu *vcpu) +{ + u16 eic = vcpu->arch.sie_block->eic; + struct kvm_s390_interrupt irq; + psw_t newpsw; + int rc; + + vcpu->stat.exit_external_interrupt++; + + rc = read_guest_lc(vcpu, __LC_EXT_NEW_PSW, &newpsw, sizeof(psw_t)); + if (rc) + return rc; + /* We can not handle clock comparator or timer interrupt with bad PSW */ + if ((eic == EXT_IRQ_CLK_COMP || eic == EXT_IRQ_CPU_TIMER) && + (newpsw.mask & PSW_MASK_EXT)) + return -EOPNOTSUPP; + + switch (eic) { + case EXT_IRQ_CLK_COMP: + irq.type = KVM_S390_INT_CLOCK_COMP; + break; + case EXT_IRQ_CPU_TIMER: + irq.type = KVM_S390_INT_CPU_TIMER; + break; + case EXT_IRQ_EXTERNAL_CALL: + if (kvm_s390_si_ext_call_pending(vcpu)) + return 0; + irq.type = KVM_S390_INT_EXTERNAL_CALL; + irq.parm = vcpu->arch.sie_block->extcpuaddr; + break; + default: + return -EOPNOTSUPP; + } + + return kvm_s390_inject_vcpu(vcpu, &irq); +} + +/** + * Handle MOVE PAGE partial execution interception. + * + * This interception can only happen for guests with DAT disabled and + * addresses that are currently not mapped in the host. Thus we try to + * set up the mappings for the corresponding user pages here (or throw + * addressing exceptions in case of illegal guest addresses). + */ +static int handle_mvpg_pei(struct kvm_vcpu *vcpu) +{ + psw_t *psw = &vcpu->arch.sie_block->gpsw; + unsigned long srcaddr, dstaddr; + int reg1, reg2, rc; + + kvm_s390_get_regs_rre(vcpu, ®1, ®2); + + /* Make sure that the source is paged-in */ + srcaddr = kvm_s390_real_to_abs(vcpu, vcpu->run->s.regs.gprs[reg2]); + if (kvm_is_error_gpa(vcpu->kvm, srcaddr)) + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + rc = kvm_arch_fault_in_page(vcpu, srcaddr, 0); + if (rc != 0) + return rc; + + /* Make sure that the destination is paged-in */ + dstaddr = kvm_s390_real_to_abs(vcpu, vcpu->run->s.regs.gprs[reg1]); + if (kvm_is_error_gpa(vcpu->kvm, dstaddr)) + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + rc = kvm_arch_fault_in_page(vcpu, dstaddr, 1); + if (rc != 0) + return rc; + + psw->addr = __rewind_psw(*psw, 4); + + return 0; +} + +static int handle_partial_execution(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.sie_block->ipa == 0xb254) /* MVPG */ + return handle_mvpg_pei(vcpu); + if (vcpu->arch.sie_block->ipa >> 8 == 0xae) /* SIGP */ + return kvm_s390_handle_sigp_pei(vcpu); + + return -EOPNOTSUPP; +} + static const intercept_handler_t intercept_funcs[] = { [0x00 >> 2] = handle_noop, [0x04 >> 2] = handle_instruction, [0x08 >> 2] = handle_prog, [0x0C >> 2] = handle_instruction_and_prog, [0x10 >> 2] = handle_noop, - [0x14 >> 2] = handle_noop, + [0x14 >> 2] = handle_external_interrupt, [0x18 >> 2] = handle_noop, [0x1C >> 2] = kvm_s390_handle_wait, [0x20 >> 2] = handle_validity, [0x28 >> 2] = handle_stop, + [0x38 >> 2] = handle_partial_execution, }; int kvm_handle_sie_intercept(struct kvm_vcpu *vcpu) diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c index 200a8f9390b6..90c8de22a2a0 100644 --- a/arch/s390/kvm/interrupt.c +++ b/arch/s390/kvm/interrupt.c @@ -27,6 +27,8 @@ #define IOINT_CSSID_MASK 0x03fc0000 #define IOINT_AI_MASK 0x04000000 +static void deliver_ckc_interrupt(struct kvm_vcpu *vcpu); + static int is_ioint(u64 type) { return ((type & 0xfffe0000u) != 0xfffe0000u); @@ -56,6 +58,17 @@ static int psw_interrupts_disabled(struct kvm_vcpu *vcpu) return 1; } +static int ckc_interrupts_enabled(struct kvm_vcpu *vcpu) +{ + if (psw_extint_disabled(vcpu) || + !(vcpu->arch.sie_block->gcr[0] & 0x800ul)) + return 0; + if (guestdbg_enabled(vcpu) && guestdbg_sstep_enabled(vcpu)) + /* No timer interrupts when single stepping */ + return 0; + return 1; +} + static u64 int_word_to_isc_bits(u32 int_word) { u8 isc = (int_word & 0x38000000) >> 27; @@ -78,6 +91,14 @@ static int __interrupt_is_deliverable(struct kvm_vcpu *vcpu, if (vcpu->arch.sie_block->gcr[0] & 0x4000ul) return 1; return 0; + case KVM_S390_INT_CLOCK_COMP: + return ckc_interrupts_enabled(vcpu); + case KVM_S390_INT_CPU_TIMER: + if (psw_extint_disabled(vcpu)) + return 0; + if (vcpu->arch.sie_block->gcr[0] & 0x400ul) + return 1; + return 0; case KVM_S390_INT_SERVICE: case KVM_S390_INT_PFAULT_INIT: case KVM_S390_INT_PFAULT_DONE: @@ -127,11 +148,16 @@ static void __unset_cpu_idle(struct kvm_vcpu *vcpu) static void __reset_intercept_indicators(struct kvm_vcpu *vcpu) { - atomic_clear_mask(CPUSTAT_ECALL_PEND | - CPUSTAT_IO_INT | CPUSTAT_EXT_INT | CPUSTAT_STOP_INT, - &vcpu->arch.sie_block->cpuflags); + atomic_clear_mask(CPUSTAT_IO_INT | CPUSTAT_EXT_INT | CPUSTAT_STOP_INT, + &vcpu->arch.sie_block->cpuflags); vcpu->arch.sie_block->lctl = 0x0000; - vcpu->arch.sie_block->ictl &= ~ICTL_LPSW; + vcpu->arch.sie_block->ictl &= ~(ICTL_LPSW | ICTL_STCTL | ICTL_PINT); + + if (guestdbg_enabled(vcpu)) { + vcpu->arch.sie_block->lctl |= (LCTL_CR0 | LCTL_CR9 | + LCTL_CR10 | LCTL_CR11); + vcpu->arch.sie_block->ictl |= (ICTL_STCTL | ICTL_PINT); + } } static void __set_cpuflag(struct kvm_vcpu *vcpu, u32 flag) @@ -149,6 +175,8 @@ static void __set_intercept_indicator(struct kvm_vcpu *vcpu, case KVM_S390_INT_PFAULT_INIT: case KVM_S390_INT_PFAULT_DONE: case KVM_S390_INT_VIRTIO: + case KVM_S390_INT_CLOCK_COMP: + case KVM_S390_INT_CPU_TIMER: if (psw_extint_disabled(vcpu)) __set_cpuflag(vcpu, CPUSTAT_EXT_INT); else @@ -174,6 +202,106 @@ static void __set_intercept_indicator(struct kvm_vcpu *vcpu, } } +static int __deliver_prog_irq(struct kvm_vcpu *vcpu, + struct kvm_s390_pgm_info *pgm_info) +{ + const unsigned short table[] = { 2, 4, 4, 6 }; + int rc = 0; + + switch (pgm_info->code & ~PGM_PER) { + case PGM_AFX_TRANSLATION: + case PGM_ASX_TRANSLATION: + case PGM_EX_TRANSLATION: + case PGM_LFX_TRANSLATION: + case PGM_LSTE_SEQUENCE: + case PGM_LSX_TRANSLATION: + case PGM_LX_TRANSLATION: + case PGM_PRIMARY_AUTHORITY: + case PGM_SECONDARY_AUTHORITY: + case PGM_SPACE_SWITCH: + rc = put_guest_lc(vcpu, pgm_info->trans_exc_code, + (u64 *)__LC_TRANS_EXC_CODE); + break; + case PGM_ALEN_TRANSLATION: + case PGM_ALE_SEQUENCE: + case PGM_ASTE_INSTANCE: + case PGM_ASTE_SEQUENCE: + case PGM_ASTE_VALIDITY: + case PGM_EXTENDED_AUTHORITY: + rc = put_guest_lc(vcpu, pgm_info->exc_access_id, + (u8 *)__LC_EXC_ACCESS_ID); + break; + case PGM_ASCE_TYPE: + case PGM_PAGE_TRANSLATION: + case PGM_REGION_FIRST_TRANS: + case PGM_REGION_SECOND_TRANS: + case PGM_REGION_THIRD_TRANS: + case PGM_SEGMENT_TRANSLATION: + rc = put_guest_lc(vcpu, pgm_info->trans_exc_code, + (u64 *)__LC_TRANS_EXC_CODE); + rc |= put_guest_lc(vcpu, pgm_info->exc_access_id, + (u8 *)__LC_EXC_ACCESS_ID); + rc |= put_guest_lc(vcpu, pgm_info->op_access_id, + (u8 *)__LC_OP_ACCESS_ID); + break; + case PGM_MONITOR: + rc = put_guest_lc(vcpu, pgm_info->mon_class_nr, + (u64 *)__LC_MON_CLASS_NR); + rc |= put_guest_lc(vcpu, pgm_info->mon_code, + (u64 *)__LC_MON_CODE); + break; + case PGM_DATA: + rc = put_guest_lc(vcpu, pgm_info->data_exc_code, + (u32 *)__LC_DATA_EXC_CODE); + break; + case PGM_PROTECTION: + rc = put_guest_lc(vcpu, pgm_info->trans_exc_code, + (u64 *)__LC_TRANS_EXC_CODE); + rc |= put_guest_lc(vcpu, pgm_info->exc_access_id, + (u8 *)__LC_EXC_ACCESS_ID); + break; + } + + if (pgm_info->code & PGM_PER) { + rc |= put_guest_lc(vcpu, pgm_info->per_code, + (u8 *) __LC_PER_CODE); + rc |= put_guest_lc(vcpu, pgm_info->per_atmid, + (u8 *)__LC_PER_ATMID); + rc |= put_guest_lc(vcpu, pgm_info->per_address, + (u64 *) __LC_PER_ADDRESS); + rc |= put_guest_lc(vcpu, pgm_info->per_access_id, + (u8 *) __LC_PER_ACCESS_ID); + } + + switch (vcpu->arch.sie_block->icptcode) { + case ICPT_INST: + case ICPT_INSTPROGI: + case ICPT_OPEREXC: + case ICPT_PARTEXEC: + case ICPT_IOINST: + /* last instruction only stored for these icptcodes */ + rc |= put_guest_lc(vcpu, table[vcpu->arch.sie_block->ipa >> 14], + (u16 *) __LC_PGM_ILC); + break; + case ICPT_PROGI: + rc |= put_guest_lc(vcpu, vcpu->arch.sie_block->pgmilc, + (u16 *) __LC_PGM_ILC); + break; + default: + rc |= put_guest_lc(vcpu, 0, + (u16 *) __LC_PGM_ILC); + } + + rc |= put_guest_lc(vcpu, pgm_info->code, + (u16 *)__LC_PGM_INT_CODE); + rc |= write_guest_lc(vcpu, __LC_PGM_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_PGM_NEW_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + + return rc; +} + static void __do_deliver_interrupt(struct kvm_vcpu *vcpu, struct kvm_s390_interrupt_info *inti) { @@ -186,26 +314,46 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu, vcpu->stat.deliver_emergency_signal++; trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, inti->emerg.code, 0); - rc = put_guest(vcpu, 0x1201, (u16 __user *)__LC_EXT_INT_CODE); - rc |= put_guest(vcpu, inti->emerg.code, - (u16 __user *)__LC_EXT_CPU_ADDR); - rc |= copy_to_guest(vcpu, __LC_EXT_OLD_PSW, + rc = put_guest_lc(vcpu, 0x1201, (u16 *)__LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, inti->emerg.code, + (u16 *)__LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); - rc |= copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, - __LC_EXT_NEW_PSW, sizeof(psw_t)); break; case KVM_S390_INT_EXTERNAL_CALL: VCPU_EVENT(vcpu, 4, "%s", "interrupt: sigp ext call"); vcpu->stat.deliver_external_call++; trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, inti->extcall.code, 0); - rc = put_guest(vcpu, 0x1202, (u16 __user *)__LC_EXT_INT_CODE); - rc |= put_guest(vcpu, inti->extcall.code, - (u16 __user *)__LC_EXT_CPU_ADDR); - rc |= copy_to_guest(vcpu, __LC_EXT_OLD_PSW, + rc = put_guest_lc(vcpu, 0x1202, (u16 *)__LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, inti->extcall.code, + (u16 *)__LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + break; + case KVM_S390_INT_CLOCK_COMP: + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, + inti->ext.ext_params, 0); + deliver_ckc_interrupt(vcpu); + break; + case KVM_S390_INT_CPU_TIMER: + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, + inti->ext.ext_params, 0); + rc = put_guest_lc(vcpu, EXT_IRQ_CPU_TIMER, + (u16 *)__LC_EXT_INT_CODE); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); - rc |= copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, - __LC_EXT_NEW_PSW, sizeof(psw_t)); + rc |= put_guest_lc(vcpu, inti->ext.ext_params, + (u32 *)__LC_EXT_PARAMS); break; case KVM_S390_INT_SERVICE: VCPU_EVENT(vcpu, 4, "interrupt: sclp parm:%x", @@ -213,37 +361,39 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu, vcpu->stat.deliver_service_signal++; trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, inti->ext.ext_params, 0); - rc = put_guest(vcpu, 0x2401, (u16 __user *)__LC_EXT_INT_CODE); - rc |= copy_to_guest(vcpu, __LC_EXT_OLD_PSW, + rc = put_guest_lc(vcpu, 0x2401, (u16 *)__LC_EXT_INT_CODE); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); - rc |= copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, - __LC_EXT_NEW_PSW, sizeof(psw_t)); - rc |= put_guest(vcpu, inti->ext.ext_params, - (u32 __user *)__LC_EXT_PARAMS); + rc |= put_guest_lc(vcpu, inti->ext.ext_params, + (u32 *)__LC_EXT_PARAMS); break; case KVM_S390_INT_PFAULT_INIT: trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, 0, inti->ext.ext_params2); - rc = put_guest(vcpu, 0x2603, (u16 __user *) __LC_EXT_INT_CODE); - rc |= put_guest(vcpu, 0x0600, (u16 __user *) __LC_EXT_CPU_ADDR); - rc |= copy_to_guest(vcpu, __LC_EXT_OLD_PSW, + rc = put_guest_lc(vcpu, 0x2603, (u16 *) __LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, 0x0600, (u16 *) __LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); - rc |= copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, - __LC_EXT_NEW_PSW, sizeof(psw_t)); - rc |= put_guest(vcpu, inti->ext.ext_params2, - (u64 __user *) __LC_EXT_PARAMS2); + rc |= put_guest_lc(vcpu, inti->ext.ext_params2, + (u64 *) __LC_EXT_PARAMS2); break; case KVM_S390_INT_PFAULT_DONE: trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, 0, inti->ext.ext_params2); - rc = put_guest(vcpu, 0x2603, (u16 __user *) __LC_EXT_INT_CODE); - rc |= put_guest(vcpu, 0x0680, (u16 __user *) __LC_EXT_CPU_ADDR); - rc |= copy_to_guest(vcpu, __LC_EXT_OLD_PSW, + rc = put_guest_lc(vcpu, 0x2603, (u16 *)__LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, 0x0680, (u16 *)__LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); - rc |= copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, - __LC_EXT_NEW_PSW, sizeof(psw_t)); - rc |= put_guest(vcpu, inti->ext.ext_params2, - (u64 __user *) __LC_EXT_PARAMS2); + rc |= put_guest_lc(vcpu, inti->ext.ext_params2, + (u64 *)__LC_EXT_PARAMS2); break; case KVM_S390_INT_VIRTIO: VCPU_EVENT(vcpu, 4, "interrupt: virtio parm:%x,parm64:%llx", @@ -252,16 +402,17 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu, trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, inti->ext.ext_params, inti->ext.ext_params2); - rc = put_guest(vcpu, 0x2603, (u16 __user *)__LC_EXT_INT_CODE); - rc |= put_guest(vcpu, 0x0d00, (u16 __user *)__LC_EXT_CPU_ADDR); - rc |= copy_to_guest(vcpu, __LC_EXT_OLD_PSW, + rc = put_guest_lc(vcpu, 0x2603, (u16 *)__LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, 0x0d00, (u16 *)__LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); - rc |= copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, - __LC_EXT_NEW_PSW, sizeof(psw_t)); - rc |= put_guest(vcpu, inti->ext.ext_params, - (u32 __user *)__LC_EXT_PARAMS); - rc |= put_guest(vcpu, inti->ext.ext_params2, - (u64 __user *)__LC_EXT_PARAMS2); + rc |= put_guest_lc(vcpu, inti->ext.ext_params, + (u32 *)__LC_EXT_PARAMS); + rc |= put_guest_lc(vcpu, inti->ext.ext_params2, + (u64 *)__LC_EXT_PARAMS2); break; case KVM_S390_SIGP_STOP: VCPU_EVENT(vcpu, 4, "%s", "interrupt: cpu stop"); @@ -285,13 +436,12 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu, vcpu->stat.deliver_restart_signal++; trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, 0, 0); - rc = copy_to_guest(vcpu, - offsetof(struct _lowcore, restart_old_psw), - &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); - rc |= copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, - offsetof(struct _lowcore, restart_psw), - sizeof(psw_t)); - atomic_clear_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags); + rc = write_guest_lc(vcpu, + offsetof(struct _lowcore, restart_old_psw), + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, offsetof(struct _lowcore, restart_psw), + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); break; case KVM_S390_PROGRAM_INT: VCPU_EVENT(vcpu, 4, "interrupt: pgm check code:%x, ilc:%x", @@ -300,13 +450,7 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu, vcpu->stat.deliver_program_int++; trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, inti->pgm.code, 0); - rc = put_guest(vcpu, inti->pgm.code, (u16 __user *)__LC_PGM_INT_CODE); - rc |= put_guest(vcpu, table[vcpu->arch.sie_block->ipa >> 14], - (u16 __user *)__LC_PGM_ILC); - rc |= copy_to_guest(vcpu, __LC_PGM_OLD_PSW, - &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); - rc |= copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, - __LC_PGM_NEW_PSW, sizeof(psw_t)); + rc = __deliver_prog_irq(vcpu, &inti->pgm); break; case KVM_S390_MCHK: @@ -317,11 +461,12 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu, inti->mchk.mcic); rc = kvm_s390_vcpu_store_status(vcpu, KVM_S390_STORE_STATUS_PREFIXED); - rc |= put_guest(vcpu, inti->mchk.mcic, (u64 __user *) __LC_MCCK_CODE); - rc |= copy_to_guest(vcpu, __LC_MCK_OLD_PSW, + rc |= put_guest_lc(vcpu, inti->mchk.mcic, (u64 *)__LC_MCCK_CODE); + rc |= write_guest_lc(vcpu, __LC_MCK_OLD_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_MCK_NEW_PSW, &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); - rc |= copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, - __LC_MCK_NEW_PSW, sizeof(psw_t)); break; case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX: @@ -334,18 +479,20 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu, vcpu->stat.deliver_io_int++; trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type, param0, param1); - rc = put_guest(vcpu, inti->io.subchannel_id, - (u16 __user *) __LC_SUBCHANNEL_ID); - rc |= put_guest(vcpu, inti->io.subchannel_nr, - (u16 __user *) __LC_SUBCHANNEL_NR); - rc |= put_guest(vcpu, inti->io.io_int_parm, - (u32 __user *) __LC_IO_INT_PARM); - rc |= put_guest(vcpu, inti->io.io_int_word, - (u32 __user *) __LC_IO_INT_WORD); - rc |= copy_to_guest(vcpu, __LC_IO_OLD_PSW, - &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); - rc |= copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, - __LC_IO_NEW_PSW, sizeof(psw_t)); + rc = put_guest_lc(vcpu, inti->io.subchannel_id, + (u16 *)__LC_SUBCHANNEL_ID); + rc |= put_guest_lc(vcpu, inti->io.subchannel_nr, + (u16 *)__LC_SUBCHANNEL_NR); + rc |= put_guest_lc(vcpu, inti->io.io_int_parm, + (u32 *)__LC_IO_INT_PARM); + rc |= put_guest_lc(vcpu, inti->io.io_int_word, + (u32 *)__LC_IO_INT_WORD); + rc |= write_guest_lc(vcpu, __LC_IO_OLD_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_IO_NEW_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); break; } default: @@ -358,25 +505,35 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu, } } -static int __try_deliver_ckc_interrupt(struct kvm_vcpu *vcpu) +static void deliver_ckc_interrupt(struct kvm_vcpu *vcpu) { int rc; - if (psw_extint_disabled(vcpu)) - return 0; - if (!(vcpu->arch.sie_block->gcr[0] & 0x800ul)) - return 0; - rc = put_guest(vcpu, 0x1004, (u16 __user *)__LC_EXT_INT_CODE); - rc |= copy_to_guest(vcpu, __LC_EXT_OLD_PSW, - &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); - rc |= copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, - __LC_EXT_NEW_PSW, sizeof(psw_t)); + rc = put_guest_lc(vcpu, 0x1004, (u16 __user *)__LC_EXT_INT_CODE); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); if (rc) { printk("kvm: The guest lowcore is not mapped during interrupt " "delivery, killing userspace\n"); do_exit(SIGKILL); } - return 1; +} + +/* Check whether SIGP interpretation facility has an external call pending */ +int kvm_s390_si_ext_call_pending(struct kvm_vcpu *vcpu) +{ + atomic_t *sigp_ctrl = &vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].ctrl; + + if (!psw_extint_disabled(vcpu) && + (vcpu->arch.sie_block->gcr[0] & 0x2000ul) && + (atomic_read(sigp_ctrl) & SIGP_CTRL_C) && + (atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_ECALL_PEND)) + return 1; + + return 0; } int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu) @@ -406,19 +563,23 @@ int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu) spin_unlock(&fi->lock); } - if ((!rc) && (vcpu->arch.sie_block->ckc < - get_tod_clock_fast() + vcpu->arch.sie_block->epoch)) { - if ((!psw_extint_disabled(vcpu)) && - (vcpu->arch.sie_block->gcr[0] & 0x800ul)) - rc = 1; - } + if (!rc && kvm_cpu_has_pending_timer(vcpu)) + rc = 1; + + if (!rc && kvm_s390_si_ext_call_pending(vcpu)) + rc = 1; return rc; } int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu) { - return 0; + if (!(vcpu->arch.sie_block->ckc < + get_tod_clock_fast() + vcpu->arch.sie_block->epoch)) + return 0; + if (!ckc_interrupts_enabled(vcpu)) + return 0; + return 1; } int kvm_s390_handle_wait(struct kvm_vcpu *vcpu) @@ -441,8 +602,7 @@ int kvm_s390_handle_wait(struct kvm_vcpu *vcpu) return -EOPNOTSUPP; /* disabled wait */ } - if (psw_extint_disabled(vcpu) || - (!(vcpu->arch.sie_block->gcr[0] & 0x800ul))) { + if (!ckc_interrupts_enabled(vcpu)) { VCPU_EVENT(vcpu, 3, "%s", "enabled wait w/o timer"); goto no_timer; } @@ -465,7 +625,8 @@ no_timer: while (list_empty(&vcpu->arch.local_int.list) && list_empty(&vcpu->arch.local_int.float_int->list) && (!vcpu->arch.local_int.timer_due) && - !signal_pending(current)) { + !signal_pending(current) && + !kvm_s390_si_ext_call_pending(vcpu)) { set_current_state(TASK_INTERRUPTIBLE); spin_unlock_bh(&vcpu->arch.local_int.lock); spin_unlock(&vcpu->arch.local_int.float_int->lock); @@ -522,6 +683,11 @@ void kvm_s390_clear_local_irqs(struct kvm_vcpu *vcpu) } atomic_set(&li->active, 0); spin_unlock_bh(&li->lock); + + /* clear pending external calls set by sigp interpretation facility */ + atomic_clear_mask(CPUSTAT_ECALL_PEND, &vcpu->arch.sie_block->cpuflags); + atomic_clear_mask(SIGP_CTRL_C, + &vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].ctrl); } void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu) @@ -554,9 +720,8 @@ void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu) } while (deliver); } - if ((vcpu->arch.sie_block->ckc < - get_tod_clock_fast() + vcpu->arch.sie_block->epoch)) - __try_deliver_ckc_interrupt(vcpu); + if (kvm_cpu_has_pending_timer(vcpu)) + deliver_ckc_interrupt(vcpu); if (atomic_read(&fi->active)) { do { @@ -660,6 +825,31 @@ int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code) return 0; } +int kvm_s390_inject_prog_irq(struct kvm_vcpu *vcpu, + struct kvm_s390_pgm_info *pgm_info) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_interrupt_info *inti; + + inti = kzalloc(sizeof(*inti), GFP_KERNEL); + if (!inti) + return -ENOMEM; + + VCPU_EVENT(vcpu, 3, "inject: prog irq %d (from kernel)", + pgm_info->code); + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_PROGRAM_INT, + pgm_info->code, 0, 1); + + inti->type = KVM_S390_PROGRAM_INT; + memcpy(&inti->pgm, pgm_info, sizeof(inti->pgm)); + spin_lock_bh(&li->lock); + list_add(&inti->list, &li->list); + atomic_set(&li->active, 1); + BUG_ON(waitqueue_active(li->wq)); + spin_unlock_bh(&li->lock); + return 0; +} + struct kvm_s390_interrupt_info *kvm_s390_get_io_int(struct kvm *kvm, u64 cr6, u64 schid) { @@ -810,6 +1000,12 @@ int kvm_s390_inject_vm(struct kvm *kvm, return __inject_vm(kvm, inti); } +void kvm_s390_reinject_io_int(struct kvm *kvm, + struct kvm_s390_interrupt_info *inti) +{ + __inject_vm(kvm, inti); +} + int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, struct kvm_s390_interrupt *s390int) { @@ -839,6 +1035,8 @@ int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, break; case KVM_S390_SIGP_STOP: case KVM_S390_RESTART: + case KVM_S390_INT_CLOCK_COMP: + case KVM_S390_INT_CPU_TIMER: VCPU_EVENT(vcpu, 3, "inject: type %x", s390int->type); inti->type = s390int->type; break; @@ -900,7 +1098,7 @@ int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, return 0; } -static void clear_floating_interrupts(struct kvm *kvm) +void kvm_s390_clear_float_irqs(struct kvm *kvm) { struct kvm_s390_float_interrupt *fi; struct kvm_s390_interrupt_info *n, *inti = NULL; @@ -1246,7 +1444,7 @@ static int flic_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr) break; case KVM_DEV_FLIC_CLEAR_IRQS: r = 0; - clear_floating_interrupts(dev->kvm); + kvm_s390_clear_float_irqs(dev->kvm); break; case KVM_DEV_FLIC_APF_ENABLE: dev->kvm->arch.gmap->pfault_enabled = 1; diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 9ae6664ff08c..2f3e14fe91a4 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -11,6 +11,7 @@ * Christian Borntraeger <borntraeger@de.ibm.com> * Heiko Carstens <heiko.carstens@de.ibm.com> * Christian Ehrhardt <ehrhardt@de.ibm.com> + * Jason J. Herne <jjherne@us.ibm.com> */ #include <linux/compiler.h> @@ -51,6 +52,8 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "exit_instr_and_program_int", VCPU_STAT(exit_instr_and_program) }, { "instruction_lctlg", VCPU_STAT(instruction_lctlg) }, { "instruction_lctl", VCPU_STAT(instruction_lctl) }, + { "instruction_stctl", VCPU_STAT(instruction_stctl) }, + { "instruction_stctg", VCPU_STAT(instruction_stctg) }, { "deliver_emergency_signal", VCPU_STAT(deliver_emergency_signal) }, { "deliver_external_call", VCPU_STAT(deliver_external_call) }, { "deliver_service_signal", VCPU_STAT(deliver_service_signal) }, @@ -66,6 +69,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "instruction_stpx", VCPU_STAT(instruction_stpx) }, { "instruction_stap", VCPU_STAT(instruction_stap) }, { "instruction_storage_key", VCPU_STAT(instruction_storage_key) }, + { "instruction_ipte_interlock", VCPU_STAT(instruction_ipte_interlock) }, { "instruction_stsch", VCPU_STAT(instruction_stsch) }, { "instruction_chsc", VCPU_STAT(instruction_chsc) }, { "instruction_essa", VCPU_STAT(instruction_essa) }, @@ -90,7 +94,7 @@ unsigned long *vfacilities; static struct gmap_notifier gmap_notifier; /* test availability of vfacility */ -static inline int test_vfacility(unsigned long nr) +int test_vfacility(unsigned long nr) { return __test_facility(nr, (void *) vfacilities); } @@ -162,6 +166,7 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_IOEVENTFD: case KVM_CAP_DEVICE_CTRL: case KVM_CAP_ENABLE_CAP_VM: + case KVM_CAP_VM_ATTRIBUTES: r = 1; break; case KVM_CAP_NR_VCPUS: @@ -180,6 +185,25 @@ int kvm_dev_ioctl_check_extension(long ext) return r; } +static void kvm_s390_sync_dirty_log(struct kvm *kvm, + struct kvm_memory_slot *memslot) +{ + gfn_t cur_gfn, last_gfn; + unsigned long address; + struct gmap *gmap = kvm->arch.gmap; + + down_read(&gmap->mm->mmap_sem); + /* Loop over all guest pages */ + last_gfn = memslot->base_gfn + memslot->npages; + for (cur_gfn = memslot->base_gfn; cur_gfn <= last_gfn; cur_gfn++) { + address = gfn_to_hva_memslot(memslot, cur_gfn); + + if (gmap_test_and_clear_dirty(address, gmap)) + mark_page_dirty(kvm, cur_gfn); + } + up_read(&gmap->mm->mmap_sem); +} + /* Section: vm related */ /* * Get (and clear) the dirty memory log for a memory slot. @@ -187,7 +211,36 @@ int kvm_dev_ioctl_check_extension(long ext) int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log) { - return 0; + int r; + unsigned long n; + struct kvm_memory_slot *memslot; + int is_dirty = 0; + + mutex_lock(&kvm->slots_lock); + + r = -EINVAL; + if (log->slot >= KVM_USER_MEM_SLOTS) + goto out; + + memslot = id_to_memslot(kvm->memslots, log->slot); + r = -ENOENT; + if (!memslot->dirty_bitmap) + goto out; + + kvm_s390_sync_dirty_log(kvm, memslot); + r = kvm_get_dirty_log(kvm, log, &is_dirty); + if (r) + goto out; + + /* Clear the dirty log */ + if (is_dirty) { + n = kvm_dirty_bitmap_bytes(memslot); + memset(memslot->dirty_bitmap, 0, n); + } + r = 0; +out: + mutex_unlock(&kvm->slots_lock); + return r; } static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap) @@ -209,11 +262,86 @@ static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap) return r; } +static int kvm_s390_mem_control(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret; + unsigned int idx; + switch (attr->attr) { + case KVM_S390_VM_MEM_ENABLE_CMMA: + ret = -EBUSY; + mutex_lock(&kvm->lock); + if (atomic_read(&kvm->online_vcpus) == 0) { + kvm->arch.use_cmma = 1; + ret = 0; + } + mutex_unlock(&kvm->lock); + break; + case KVM_S390_VM_MEM_CLR_CMMA: + mutex_lock(&kvm->lock); + idx = srcu_read_lock(&kvm->srcu); + page_table_reset_pgste(kvm->arch.gmap->mm, 0, TASK_SIZE, false); + srcu_read_unlock(&kvm->srcu, idx); + mutex_unlock(&kvm->lock); + ret = 0; + break; + default: + ret = -ENXIO; + break; + } + return ret; +} + +static int kvm_s390_vm_set_attr(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret; + + switch (attr->group) { + case KVM_S390_VM_MEM_CTRL: + ret = kvm_s390_mem_control(kvm, attr); + break; + default: + ret = -ENXIO; + break; + } + + return ret; +} + +static int kvm_s390_vm_get_attr(struct kvm *kvm, struct kvm_device_attr *attr) +{ + return -ENXIO; +} + +static int kvm_s390_vm_has_attr(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret; + + switch (attr->group) { + case KVM_S390_VM_MEM_CTRL: + switch (attr->attr) { + case KVM_S390_VM_MEM_ENABLE_CMMA: + case KVM_S390_VM_MEM_CLR_CMMA: + ret = 0; + break; + default: + ret = -ENXIO; + break; + } + break; + default: + ret = -ENXIO; + break; + } + + return ret; +} + long kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) { struct kvm *kvm = filp->private_data; void __user *argp = (void __user *)arg; + struct kvm_device_attr attr; int r; switch (ioctl) { @@ -246,6 +374,27 @@ long kvm_arch_vm_ioctl(struct file *filp, } break; } + case KVM_SET_DEVICE_ATTR: { + r = -EFAULT; + if (copy_from_user(&attr, (void __user *)arg, sizeof(attr))) + break; + r = kvm_s390_vm_set_attr(kvm, &attr); + break; + } + case KVM_GET_DEVICE_ATTR: { + r = -EFAULT; + if (copy_from_user(&attr, (void __user *)arg, sizeof(attr))) + break; + r = kvm_s390_vm_get_attr(kvm, &attr); + break; + } + case KVM_HAS_DEVICE_ATTR: { + r = -EFAULT; + if (copy_from_user(&attr, (void __user *)arg, sizeof(attr))) + break; + r = kvm_s390_vm_has_attr(kvm, &attr); + break; + } default: r = -ENOTTY; } @@ -292,6 +441,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) spin_lock_init(&kvm->arch.float_int.lock); INIT_LIST_HEAD(&kvm->arch.float_int.list); + init_waitqueue_head(&kvm->arch.ipte_wq); debug_register_view(kvm->arch.dbf, &debug_sprintf_view); VM_EVENT(kvm, 3, "%s", "vm created"); @@ -309,6 +459,8 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) kvm->arch.css_support = 0; kvm->arch.use_irqchip = 0; + spin_lock_init(&kvm->arch.start_stop_lock); + return 0; out_nogmap: debug_unregister(kvm->arch.dbf); @@ -322,6 +474,7 @@ void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) { VCPU_EVENT(vcpu, 3, "%s", "free cpu"); trace_kvm_s390_destroy_vcpu(vcpu->vcpu_id); + kvm_s390_clear_local_irqs(vcpu); kvm_clear_async_pf_completion_queue(vcpu); if (!kvm_is_ucontrol(vcpu->kvm)) { clear_bit(63 - vcpu->vcpu_id, @@ -335,9 +488,8 @@ void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) if (kvm_is_ucontrol(vcpu->kvm)) gmap_free(vcpu->arch.gmap); - if (vcpu->arch.sie_block->cbrlo) - __free_page(__pfn_to_page( - vcpu->arch.sie_block->cbrlo >> PAGE_SHIFT)); + if (kvm_s390_cmma_enabled(vcpu->kvm)) + kvm_s390_vcpu_unsetup_cmma(vcpu); free_page((unsigned long)(vcpu->arch.sie_block)); kvm_vcpu_uninit(vcpu); @@ -372,6 +524,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm) if (!kvm_is_ucontrol(kvm)) gmap_free(kvm->arch.gmap); kvm_s390_destroy_adapters(kvm); + kvm_s390_clear_float_irqs(kvm); } /* Section: vcpu related */ @@ -442,7 +595,7 @@ static void kvm_s390_vcpu_initial_reset(struct kvm_vcpu *vcpu) vcpu->arch.sie_block->pp = 0; vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID; kvm_clear_async_pf_completion_queue(vcpu); - atomic_set_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags); + kvm_s390_vcpu_stop(vcpu); kvm_s390_clear_local_irqs(vcpu); } @@ -451,9 +604,26 @@ int kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu) return 0; } +void kvm_s390_vcpu_unsetup_cmma(struct kvm_vcpu *vcpu) +{ + free_page(vcpu->arch.sie_block->cbrlo); + vcpu->arch.sie_block->cbrlo = 0; +} + +int kvm_s390_vcpu_setup_cmma(struct kvm_vcpu *vcpu) +{ + vcpu->arch.sie_block->cbrlo = get_zeroed_page(GFP_KERNEL); + if (!vcpu->arch.sie_block->cbrlo) + return -ENOMEM; + + vcpu->arch.sie_block->ecb2 |= 0x80; + vcpu->arch.sie_block->ecb2 &= ~0x08; + return 0; +} + int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) { - struct page *cbrl; + int rc = 0; atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH | CPUSTAT_SM | @@ -464,15 +634,17 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) vcpu->arch.sie_block->ecb |= 0x10; vcpu->arch.sie_block->ecb2 = 8; - vcpu->arch.sie_block->eca = 0xC1002001U; + vcpu->arch.sie_block->eca = 0xD1002000U; + if (sclp_has_siif()) + vcpu->arch.sie_block->eca |= 1; vcpu->arch.sie_block->fac = (int) (long) vfacilities; - if (kvm_enabled_cmma()) { - cbrl = alloc_page(GFP_KERNEL | __GFP_ZERO); - if (cbrl) { - vcpu->arch.sie_block->ecb2 |= 0x80; - vcpu->arch.sie_block->ecb2 &= ~0x08; - vcpu->arch.sie_block->cbrlo = page_to_phys(cbrl); - } + vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE | + ICTL_TPROT; + + if (kvm_s390_cmma_enabled(vcpu->kvm)) { + rc = kvm_s390_vcpu_setup_cmma(vcpu); + if (rc) + return rc; } hrtimer_init(&vcpu->arch.ckc_timer, CLOCK_REALTIME, HRTIMER_MODE_ABS); tasklet_init(&vcpu->arch.tasklet, kvm_s390_tasklet, @@ -480,7 +652,7 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) vcpu->arch.ckc_timer.function = kvm_s390_idle_wakeup; get_cpu_id(&vcpu->arch.cpu_id); vcpu->arch.cpu_id.version = 0xff; - return 0; + return rc; } struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, @@ -584,7 +756,7 @@ static void kvm_gmap_notifier(struct gmap *gmap, unsigned long address) kvm_for_each_vcpu(i, vcpu, kvm) { /* match against both prefix pages */ - if (vcpu->arch.sie_block->prefix == (address & ~0x1000UL)) { + if (kvm_s390_get_prefix(vcpu) == (address & ~0x1000UL)) { VCPU_EVENT(vcpu, 2, "gmap notifier for %lx", address); kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu); exit_sie_sync(vcpu); @@ -769,10 +941,40 @@ int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu, return -EINVAL; /* not implemented yet */ } +#define VALID_GUESTDBG_FLAGS (KVM_GUESTDBG_SINGLESTEP | \ + KVM_GUESTDBG_USE_HW_BP | \ + KVM_GUESTDBG_ENABLE) + int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg) { - return -EINVAL; /* not implemented yet */ + int rc = 0; + + vcpu->guest_debug = 0; + kvm_s390_clear_bp_data(vcpu); + + if (dbg->control & ~VALID_GUESTDBG_FLAGS) + return -EINVAL; + + if (dbg->control & KVM_GUESTDBG_ENABLE) { + vcpu->guest_debug = dbg->control; + /* enforce guest PER */ + atomic_set_mask(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags); + + if (dbg->control & KVM_GUESTDBG_USE_HW_BP) + rc = kvm_s390_import_bp_data(vcpu, dbg); + } else { + atomic_clear_mask(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags); + vcpu->arch.guestdbg.last_bp = 0; + } + + if (rc) { + vcpu->guest_debug = 0; + kvm_s390_clear_bp_data(vcpu); + atomic_clear_mask(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags); + } + + return rc; } int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu, @@ -787,8 +989,27 @@ int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, return -EINVAL; /* not implemented yet */ } +bool kvm_s390_cmma_enabled(struct kvm *kvm) +{ + if (!MACHINE_IS_LPAR) + return false; + /* only enable for z10 and later */ + if (!MACHINE_HAS_EDAT1) + return false; + if (!kvm->arch.use_cmma) + return false; + return true; +} + +static bool ibs_enabled(struct kvm_vcpu *vcpu) +{ + return atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_IBS; +} + static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu) { +retry: + s390_vcpu_unblock(vcpu); /* * We use MMU_RELOAD just to re-arm the ipte notifier for the * guest prefix page. gmap_ipte_notify will wait on the ptl lock. @@ -796,27 +1017,61 @@ static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu) * already finished. We might race against a second unmapper that * wants to set the blocking bit. Lets just retry the request loop. */ - while (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) { + if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) { int rc; rc = gmap_ipte_notify(vcpu->arch.gmap, - vcpu->arch.sie_block->prefix, + kvm_s390_get_prefix(vcpu), PAGE_SIZE * 2); if (rc) return rc; - s390_vcpu_unblock(vcpu); + goto retry; + } + + if (kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu)) { + if (!ibs_enabled(vcpu)) { + trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 1); + atomic_set_mask(CPUSTAT_IBS, + &vcpu->arch.sie_block->cpuflags); + } + goto retry; } + + if (kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu)) { + if (ibs_enabled(vcpu)) { + trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 0); + atomic_clear_mask(CPUSTAT_IBS, + &vcpu->arch.sie_block->cpuflags); + } + goto retry; + } + return 0; } -static long kvm_arch_fault_in_sync(struct kvm_vcpu *vcpu) +/** + * kvm_arch_fault_in_page - fault-in guest page if necessary + * @vcpu: The corresponding virtual cpu + * @gpa: Guest physical address + * @writable: Whether the page should be writable or not + * + * Make sure that a guest page has been faulted-in on the host. + * + * Return: Zero on success, negative error code otherwise. + */ +long kvm_arch_fault_in_page(struct kvm_vcpu *vcpu, gpa_t gpa, int writable) { - long rc; - hva_t fault = gmap_fault(current->thread.gmap_addr, vcpu->arch.gmap); struct mm_struct *mm = current->mm; + hva_t hva; + long rc; + + hva = gmap_fault(gpa, vcpu->arch.gmap); + if (IS_ERR_VALUE(hva)) + return (long)hva; down_read(&mm->mmap_sem); - rc = get_user_pages(current, mm, fault, 1, 1, 0, NULL, NULL); + rc = get_user_pages(current, mm, hva, 1, writable, 0, NULL, NULL); up_read(&mm->mmap_sem); - return rc; + + return rc < 0 ? rc : 0; } static void __kvm_inject_pfault_token(struct kvm_vcpu *vcpu, bool start_token, @@ -883,8 +1138,9 @@ static int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu) if (!vcpu->arch.gmap->pfault_enabled) return 0; - hva = gmap_fault(current->thread.gmap_addr, vcpu->arch.gmap); - if (copy_from_guest(vcpu, &arch.pfault_token, vcpu->arch.pfault_token, 8)) + hva = gfn_to_hva(vcpu->kvm, gpa_to_gfn(current->thread.gmap_addr)); + hva += current->thread.gmap_addr & ~PAGE_MASK; + if (read_guest_real(vcpu, vcpu->arch.pfault_token, &arch.pfault_token, 8)) return 0; rc = kvm_setup_async_pf(vcpu, current->thread.gmap_addr, hva, &arch); @@ -907,7 +1163,7 @@ static int vcpu_pre_run(struct kvm_vcpu *vcpu) if (need_resched()) schedule(); - if (test_thread_flag(TIF_MCCK_PENDING)) + if (test_cpu_flag(CIF_MCCK_PENDING)) s390_handle_mcck(); if (!kvm_is_ucontrol(vcpu->kvm)) @@ -917,6 +1173,11 @@ static int vcpu_pre_run(struct kvm_vcpu *vcpu) if (rc) return rc; + if (guestdbg_enabled(vcpu)) { + kvm_s390_backup_guest_per_regs(vcpu); + kvm_s390_patch_guest_per_regs(vcpu); + } + vcpu->arch.sie_block->icptcode = 0; cpuflags = atomic_read(&vcpu->arch.sie_block->cpuflags); VCPU_EVENT(vcpu, 6, "entering sie flags %x", cpuflags); @@ -933,6 +1194,9 @@ static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason) vcpu->arch.sie_block->icptcode); trace_kvm_s390_sie_exit(vcpu, vcpu->arch.sie_block->icptcode); + if (guestdbg_enabled(vcpu)) + kvm_s390_restore_guest_per_regs(vcpu); + if (exit_reason >= 0) { rc = 0; } else if (kvm_is_ucontrol(vcpu->kvm)) { @@ -945,9 +1209,12 @@ static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason) } else if (current->thread.gmap_pfault) { trace_kvm_s390_major_guest_pfault(vcpu); current->thread.gmap_pfault = 0; - if (kvm_arch_setup_async_pf(vcpu) || - (kvm_arch_fault_in_sync(vcpu) >= 0)) + if (kvm_arch_setup_async_pf(vcpu)) { rc = 0; + } else { + gpa_t gpa = current->thread.gmap_addr; + rc = kvm_arch_fault_in_page(vcpu, gpa, 1); + } } if (rc == -1) { @@ -969,16 +1236,6 @@ static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason) return rc; } -bool kvm_enabled_cmma(void) -{ - if (!MACHINE_IS_LPAR) - return false; - /* only enable for z10 and later */ - if (!MACHINE_HAS_EDAT1) - return false; - return true; -} - static int __vcpu_run(struct kvm_vcpu *vcpu) { int rc, exit_reason; @@ -1008,7 +1265,7 @@ static int __vcpu_run(struct kvm_vcpu *vcpu) vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); rc = vcpu_post_run(vcpu, exit_reason); - } while (!signal_pending(current) && !rc); + } while (!signal_pending(current) && !guestdbg_exit_pending(vcpu) && !rc); srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); return rc; @@ -1019,10 +1276,15 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) int rc; sigset_t sigsaved; + if (guestdbg_exit_pending(vcpu)) { + kvm_s390_prepare_debug_exit(vcpu); + return 0; + } + if (vcpu->sigset_active) sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved); - atomic_clear_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags); + kvm_s390_vcpu_start(vcpu); switch (kvm_run->exit_reason) { case KVM_EXIT_S390_SIEIC: @@ -1031,6 +1293,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) case KVM_EXIT_S390_RESET: case KVM_EXIT_S390_UCONTROL: case KVM_EXIT_S390_TSCH: + case KVM_EXIT_DEBUG: break; default: BUG(); @@ -1056,6 +1319,11 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) rc = -EINTR; } + if (guestdbg_exit_pending(vcpu) && !rc) { + kvm_s390_prepare_debug_exit(vcpu); + rc = 0; + } + if (rc == -EOPNOTSUPP) { /* intercept cannot be handled in-kernel, prepare kvm-run */ kvm_run->exit_reason = KVM_EXIT_S390_SIEIC; @@ -1073,7 +1341,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) kvm_run->psw_mask = vcpu->arch.sie_block->gpsw.mask; kvm_run->psw_addr = vcpu->arch.sie_block->gpsw.addr; - kvm_run->s.regs.prefix = vcpu->arch.sie_block->prefix; + kvm_run->s.regs.prefix = kvm_s390_get_prefix(vcpu); memcpy(&kvm_run->s.regs.crs, &vcpu->arch.sie_block->gcr, 128); if (vcpu->sigset_active) @@ -1083,83 +1351,52 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) return rc; } -static int __guestcopy(struct kvm_vcpu *vcpu, u64 guestdest, void *from, - unsigned long n, int prefix) -{ - if (prefix) - return copy_to_guest(vcpu, guestdest, from, n); - else - return copy_to_guest_absolute(vcpu, guestdest, from, n); -} - /* * store status at address * we use have two special cases: * KVM_S390_STORE_STATUS_NOADDR: -> 0x1200 on 64 bit * KVM_S390_STORE_STATUS_PREFIXED: -> prefix */ -int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long addr) +int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long gpa) { unsigned char archmode = 1; - int prefix; + unsigned int px; u64 clkcomp; + int rc; - if (addr == KVM_S390_STORE_STATUS_NOADDR) { - if (copy_to_guest_absolute(vcpu, 163ul, &archmode, 1)) + if (gpa == KVM_S390_STORE_STATUS_NOADDR) { + if (write_guest_abs(vcpu, 163, &archmode, 1)) return -EFAULT; - addr = SAVE_AREA_BASE; - prefix = 0; - } else if (addr == KVM_S390_STORE_STATUS_PREFIXED) { - if (copy_to_guest(vcpu, 163ul, &archmode, 1)) + gpa = SAVE_AREA_BASE; + } else if (gpa == KVM_S390_STORE_STATUS_PREFIXED) { + if (write_guest_real(vcpu, 163, &archmode, 1)) return -EFAULT; - addr = SAVE_AREA_BASE; - prefix = 1; - } else - prefix = 0; - - if (__guestcopy(vcpu, addr + offsetof(struct save_area, fp_regs), - vcpu->arch.guest_fpregs.fprs, 128, prefix)) - return -EFAULT; - - if (__guestcopy(vcpu, addr + offsetof(struct save_area, gp_regs), - vcpu->run->s.regs.gprs, 128, prefix)) - return -EFAULT; - - if (__guestcopy(vcpu, addr + offsetof(struct save_area, psw), - &vcpu->arch.sie_block->gpsw, 16, prefix)) - return -EFAULT; - - if (__guestcopy(vcpu, addr + offsetof(struct save_area, pref_reg), - &vcpu->arch.sie_block->prefix, 4, prefix)) - return -EFAULT; - - if (__guestcopy(vcpu, - addr + offsetof(struct save_area, fp_ctrl_reg), - &vcpu->arch.guest_fpregs.fpc, 4, prefix)) - return -EFAULT; - - if (__guestcopy(vcpu, addr + offsetof(struct save_area, tod_reg), - &vcpu->arch.sie_block->todpr, 4, prefix)) - return -EFAULT; - - if (__guestcopy(vcpu, addr + offsetof(struct save_area, timer), - &vcpu->arch.sie_block->cputm, 8, prefix)) - return -EFAULT; - + gpa = kvm_s390_real_to_abs(vcpu, SAVE_AREA_BASE); + } + rc = write_guest_abs(vcpu, gpa + offsetof(struct save_area, fp_regs), + vcpu->arch.guest_fpregs.fprs, 128); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, gp_regs), + vcpu->run->s.regs.gprs, 128); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, psw), + &vcpu->arch.sie_block->gpsw, 16); + px = kvm_s390_get_prefix(vcpu); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, pref_reg), + &px, 4); + rc |= write_guest_abs(vcpu, + gpa + offsetof(struct save_area, fp_ctrl_reg), + &vcpu->arch.guest_fpregs.fpc, 4); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, tod_reg), + &vcpu->arch.sie_block->todpr, 4); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, timer), + &vcpu->arch.sie_block->cputm, 8); clkcomp = vcpu->arch.sie_block->ckc >> 8; - if (__guestcopy(vcpu, addr + offsetof(struct save_area, clk_cmp), - &clkcomp, 8, prefix)) - return -EFAULT; - - if (__guestcopy(vcpu, addr + offsetof(struct save_area, acc_regs), - &vcpu->run->s.regs.acrs, 64, prefix)) - return -EFAULT; - - if (__guestcopy(vcpu, - addr + offsetof(struct save_area, ctrl_regs), - &vcpu->arch.sie_block->gcr, 128, prefix)) - return -EFAULT; - return 0; + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, clk_cmp), + &clkcomp, 8); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, acc_regs), + &vcpu->run->s.regs.acrs, 64); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, ctrl_regs), + &vcpu->arch.sie_block->gcr, 128); + return rc ? -EFAULT : 0; } int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr) @@ -1176,6 +1413,109 @@ int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr) return kvm_s390_store_status_unloaded(vcpu, addr); } +static inline int is_vcpu_stopped(struct kvm_vcpu *vcpu) +{ + return atomic_read(&(vcpu)->arch.sie_block->cpuflags) & CPUSTAT_STOPPED; +} + +static void __disable_ibs_on_vcpu(struct kvm_vcpu *vcpu) +{ + kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu); + kvm_make_request(KVM_REQ_DISABLE_IBS, vcpu); + exit_sie_sync(vcpu); +} + +static void __disable_ibs_on_all_vcpus(struct kvm *kvm) +{ + unsigned int i; + struct kvm_vcpu *vcpu; + + kvm_for_each_vcpu(i, vcpu, kvm) { + __disable_ibs_on_vcpu(vcpu); + } +} + +static void __enable_ibs_on_vcpu(struct kvm_vcpu *vcpu) +{ + kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu); + kvm_make_request(KVM_REQ_ENABLE_IBS, vcpu); + exit_sie_sync(vcpu); +} + +void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu) +{ + int i, online_vcpus, started_vcpus = 0; + + if (!is_vcpu_stopped(vcpu)) + return; + + trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 1); + /* Only one cpu at a time may enter/leave the STOPPED state. */ + spin_lock_bh(&vcpu->kvm->arch.start_stop_lock); + online_vcpus = atomic_read(&vcpu->kvm->online_vcpus); + + for (i = 0; i < online_vcpus; i++) { + if (!is_vcpu_stopped(vcpu->kvm->vcpus[i])) + started_vcpus++; + } + + if (started_vcpus == 0) { + /* we're the only active VCPU -> speed it up */ + __enable_ibs_on_vcpu(vcpu); + } else if (started_vcpus == 1) { + /* + * As we are starting a second VCPU, we have to disable + * the IBS facility on all VCPUs to remove potentially + * oustanding ENABLE requests. + */ + __disable_ibs_on_all_vcpus(vcpu->kvm); + } + + atomic_clear_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags); + /* + * Another VCPU might have used IBS while we were offline. + * Let's play safe and flush the VCPU at startup. + */ + vcpu->arch.sie_block->ihcpu = 0xffff; + spin_unlock_bh(&vcpu->kvm->arch.start_stop_lock); + return; +} + +void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu) +{ + int i, online_vcpus, started_vcpus = 0; + struct kvm_vcpu *started_vcpu = NULL; + + if (is_vcpu_stopped(vcpu)) + return; + + trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 0); + /* Only one cpu at a time may enter/leave the STOPPED state. */ + spin_lock_bh(&vcpu->kvm->arch.start_stop_lock); + online_vcpus = atomic_read(&vcpu->kvm->online_vcpus); + + atomic_set_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags); + __disable_ibs_on_vcpu(vcpu); + + for (i = 0; i < online_vcpus; i++) { + if (!is_vcpu_stopped(vcpu->kvm->vcpus[i])) { + started_vcpus++; + started_vcpu = vcpu->kvm->vcpus[i]; + } + } + + if (started_vcpus == 1) { + /* + * As we only have one VCPU left, we want to enable the + * IBS facility for that VCPU to speed it up. + */ + __enable_ibs_on_vcpu(started_vcpu); + } + + spin_unlock_bh(&vcpu->kvm->arch.start_stop_lock); + return; +} + static int kvm_vcpu_ioctl_enable_cap(struct kvm_vcpu *vcpu, struct kvm_enable_cap *cap) { diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index 3c1e2274d9ea..a8655ed31616 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -28,7 +28,6 @@ int kvm_handle_sie_intercept(struct kvm_vcpu *vcpu); /* Transactional Memory Execution related macros */ #define IS_TE_ENABLED(vcpu) ((vcpu->arch.sie_block->ecb & 0x10)) -#define TDB_ADDR 0x1800UL #define TDB_FORMAT1 1 #define IS_ITDB_VALID(vcpu) ((*(char *)vcpu->arch.sie_block->itdba == TDB_FORMAT1)) @@ -62,9 +61,15 @@ static inline int kvm_is_ucontrol(struct kvm *kvm) #endif } +#define GUEST_PREFIX_SHIFT 13 +static inline u32 kvm_s390_get_prefix(struct kvm_vcpu *vcpu) +{ + return vcpu->arch.sie_block->prefix << GUEST_PREFIX_SHIFT; +} + static inline void kvm_s390_set_prefix(struct kvm_vcpu *vcpu, u32 prefix) { - vcpu->arch.sie_block->prefix = prefix & 0x7fffe000u; + vcpu->arch.sie_block->prefix = prefix >> GUEST_PREFIX_SHIFT; vcpu->arch.sie_block->ihcpu = 0xffff; kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu); } @@ -130,6 +135,7 @@ void kvm_s390_tasklet(unsigned long parm); void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu); void kvm_s390_deliver_pending_machine_checks(struct kvm_vcpu *vcpu); void kvm_s390_clear_local_irqs(struct kvm_vcpu *vcpu); +void kvm_s390_clear_float_irqs(struct kvm *kvm); int __must_check kvm_s390_inject_vm(struct kvm *kvm, struct kvm_s390_interrupt *s390int); int __must_check kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, @@ -137,35 +143,94 @@ int __must_check kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, int __must_check kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code); struct kvm_s390_interrupt_info *kvm_s390_get_io_int(struct kvm *kvm, u64 cr6, u64 schid); +void kvm_s390_reinject_io_int(struct kvm *kvm, + struct kvm_s390_interrupt_info *inti); int kvm_s390_mask_adapter(struct kvm *kvm, unsigned int id, bool masked); /* implemented in priv.c */ +int is_valid_psw(psw_t *psw); int kvm_s390_handle_b2(struct kvm_vcpu *vcpu); int kvm_s390_handle_e5(struct kvm_vcpu *vcpu); int kvm_s390_handle_01(struct kvm_vcpu *vcpu); int kvm_s390_handle_b9(struct kvm_vcpu *vcpu); int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu); +int kvm_s390_handle_stctl(struct kvm_vcpu *vcpu); int kvm_s390_handle_lctl(struct kvm_vcpu *vcpu); int kvm_s390_handle_eb(struct kvm_vcpu *vcpu); /* implemented in sigp.c */ int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu); +int kvm_s390_handle_sigp_pei(struct kvm_vcpu *vcpu); /* implemented in kvm-s390.c */ +long kvm_arch_fault_in_page(struct kvm_vcpu *vcpu, gpa_t gpa, int writable); int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long addr); int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr); +void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu); +void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu); void s390_vcpu_block(struct kvm_vcpu *vcpu); void s390_vcpu_unblock(struct kvm_vcpu *vcpu); void exit_sie(struct kvm_vcpu *vcpu); void exit_sie_sync(struct kvm_vcpu *vcpu); -/* are we going to support cmma? */ -bool kvm_enabled_cmma(void); +int kvm_s390_vcpu_setup_cmma(struct kvm_vcpu *vcpu); +void kvm_s390_vcpu_unsetup_cmma(struct kvm_vcpu *vcpu); +/* is cmma enabled */ +bool kvm_s390_cmma_enabled(struct kvm *kvm); +int test_vfacility(unsigned long nr); + /* implemented in diag.c */ int kvm_s390_handle_diag(struct kvm_vcpu *vcpu); +/* implemented in interrupt.c */ +int kvm_s390_inject_prog_irq(struct kvm_vcpu *vcpu, + struct kvm_s390_pgm_info *pgm_info); + +/** + * kvm_s390_inject_prog_cond - conditionally inject a program check + * @vcpu: virtual cpu + * @rc: original return/error code + * + * This function is supposed to be used after regular guest access functions + * failed, to conditionally inject a program check to a vcpu. The typical + * pattern would look like + * + * rc = write_guest(vcpu, addr, data, len); + * if (rc) + * return kvm_s390_inject_prog_cond(vcpu, rc); + * + * A negative return code from guest access functions implies an internal error + * like e.g. out of memory. In these cases no program check should be injected + * to the guest. + * A positive value implies that an exception happened while accessing a guest's + * memory. In this case all data belonging to the corresponding program check + * has been stored in vcpu->arch.pgm and can be injected with + * kvm_s390_inject_prog_irq(). + * + * Returns: - the original @rc value if @rc was negative (internal error) + * - zero if @rc was already zero + * - zero or error code from injecting if @rc was positive + * (program check injected to @vcpu) + */ +static inline int kvm_s390_inject_prog_cond(struct kvm_vcpu *vcpu, int rc) +{ + if (rc <= 0) + return rc; + return kvm_s390_inject_prog_irq(vcpu, &vcpu->arch.pgm); +} /* implemented in interrupt.c */ int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu); int psw_extint_disabled(struct kvm_vcpu *vcpu); void kvm_s390_destroy_adapters(struct kvm *kvm); +int kvm_s390_si_ext_call_pending(struct kvm_vcpu *vcpu); + +/* implemented in guestdbg.c */ +void kvm_s390_backup_guest_per_regs(struct kvm_vcpu *vcpu); +void kvm_s390_restore_guest_per_regs(struct kvm_vcpu *vcpu); +void kvm_s390_patch_guest_per_regs(struct kvm_vcpu *vcpu); +int kvm_s390_import_bp_data(struct kvm_vcpu *vcpu, + struct kvm_guest_debug *dbg); +void kvm_s390_clear_bp_data(struct kvm_vcpu *vcpu); +void kvm_s390_prepare_debug_exit(struct kvm_vcpu *vcpu); +void kvm_s390_handle_per_event(struct kvm_vcpu *vcpu); #endif diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c index 476e9e218f43..f89c1cd67751 100644 --- a/arch/s390/kvm/priv.c +++ b/arch/s390/kvm/priv.c @@ -35,8 +35,8 @@ static int handle_set_clock(struct kvm_vcpu *vcpu) { struct kvm_vcpu *cpup; s64 hostclk, val; + int i, rc; u64 op2; - int i; if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); @@ -44,8 +44,9 @@ static int handle_set_clock(struct kvm_vcpu *vcpu) op2 = kvm_s390_get_base_disp_s(vcpu); if (op2 & 7) /* Operand must be on a doubleword boundary */ return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); - if (get_guest(vcpu, val, (u64 __user *) op2)) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + rc = read_guest(vcpu, op2, &val, sizeof(val)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); if (store_tod_clock(&hostclk)) { kvm_s390_set_psw_cc(vcpu, 3); @@ -65,8 +66,8 @@ static int handle_set_clock(struct kvm_vcpu *vcpu) static int handle_set_prefix(struct kvm_vcpu *vcpu) { u64 operand2; - u32 address = 0; - u8 tmp; + u32 address; + int rc; vcpu->stat.instruction_spx++; @@ -80,14 +81,18 @@ static int handle_set_prefix(struct kvm_vcpu *vcpu) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); /* get the value */ - if (get_guest(vcpu, address, (u32 __user *) operand2)) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + rc = read_guest(vcpu, operand2, &address, sizeof(address)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); - address = address & 0x7fffe000u; + address &= 0x7fffe000u; - /* make sure that the new value is valid memory */ - if (copy_from_guest_absolute(vcpu, &tmp, address, 1) || - (copy_from_guest_absolute(vcpu, &tmp, address + PAGE_SIZE, 1))) + /* + * Make sure the new value is valid memory. We only need to check the + * first page, since address is 8k aligned and memory pieces are always + * at least 1MB aligned and have at least a size of 1MB. + */ + if (kvm_is_error_gpa(vcpu->kvm, address)) return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); kvm_s390_set_prefix(vcpu, address); @@ -101,6 +106,7 @@ static int handle_store_prefix(struct kvm_vcpu *vcpu) { u64 operand2; u32 address; + int rc; vcpu->stat.instruction_stpx++; @@ -113,12 +119,12 @@ static int handle_store_prefix(struct kvm_vcpu *vcpu) if (operand2 & 3) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); - address = vcpu->arch.sie_block->prefix; - address = address & 0x7fffe000u; + address = kvm_s390_get_prefix(vcpu); /* get the value */ - if (put_guest(vcpu, address, (u32 __user *)operand2)) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + rc = write_guest(vcpu, operand2, &address, sizeof(address)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); VCPU_EVENT(vcpu, 5, "storing prefix to %x", address); trace_kvm_s390_handle_prefix(vcpu, 0, address); @@ -127,28 +133,44 @@ static int handle_store_prefix(struct kvm_vcpu *vcpu) static int handle_store_cpu_address(struct kvm_vcpu *vcpu) { - u64 useraddr; + u16 vcpu_id = vcpu->vcpu_id; + u64 ga; + int rc; vcpu->stat.instruction_stap++; if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); - useraddr = kvm_s390_get_base_disp_s(vcpu); + ga = kvm_s390_get_base_disp_s(vcpu); - if (useraddr & 1) + if (ga & 1) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); - if (put_guest(vcpu, vcpu->vcpu_id, (u16 __user *)useraddr)) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + rc = write_guest(vcpu, ga, &vcpu_id, sizeof(vcpu_id)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); - VCPU_EVENT(vcpu, 5, "storing cpu address to %llx", useraddr); - trace_kvm_s390_handle_stap(vcpu, useraddr); + VCPU_EVENT(vcpu, 5, "storing cpu address to %llx", ga); + trace_kvm_s390_handle_stap(vcpu, ga); return 0; } +static void __skey_check_enable(struct kvm_vcpu *vcpu) +{ + if (!(vcpu->arch.sie_block->ictl & (ICTL_ISKE | ICTL_SSKE | ICTL_RRBE))) + return; + + s390_enable_skey(); + trace_kvm_s390_skey_related_inst(vcpu); + vcpu->arch.sie_block->ictl &= ~(ICTL_ISKE | ICTL_SSKE | ICTL_RRBE); +} + + static int handle_skey(struct kvm_vcpu *vcpu) { + __skey_check_enable(vcpu); + vcpu->stat.instruction_storage_key++; if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) @@ -160,9 +182,21 @@ static int handle_skey(struct kvm_vcpu *vcpu) return 0; } +static int handle_ipte_interlock(struct kvm_vcpu *vcpu) +{ + psw_t *psw = &vcpu->arch.sie_block->gpsw; + + vcpu->stat.instruction_ipte_interlock++; + if (psw_bits(*psw).p) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + wait_event(vcpu->kvm->arch.ipte_wq, !ipte_lock_held(vcpu)); + psw->addr = __rewind_psw(*psw, 4); + VCPU_EVENT(vcpu, 4, "%s", "retrying ipte interlock operation"); + return 0; +} + static int handle_test_block(struct kvm_vcpu *vcpu) { - unsigned long hva; gpa_t addr; int reg2; @@ -171,16 +205,18 @@ static int handle_test_block(struct kvm_vcpu *vcpu) kvm_s390_get_regs_rre(vcpu, NULL, ®2); addr = vcpu->run->s.regs.gprs[reg2] & PAGE_MASK; + addr = kvm_s390_logical_to_effective(vcpu, addr); + if (kvm_s390_check_low_addr_protection(vcpu, addr)) + return kvm_s390_inject_prog_irq(vcpu, &vcpu->arch.pgm); addr = kvm_s390_real_to_abs(vcpu, addr); - hva = gfn_to_hva(vcpu->kvm, gpa_to_gfn(addr)); - if (kvm_is_error_hva(hva)) + if (kvm_is_error_gpa(vcpu->kvm, addr)) return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); /* * We don't expect errors on modern systems, and do not care * about storage keys (yet), so let's just clear the page. */ - if (clear_user((void __user *)hva, PAGE_SIZE) != 0) + if (kvm_clear_guest(vcpu->kvm, addr, PAGE_SIZE)) return -EFAULT; kvm_s390_set_psw_cc(vcpu, 0); vcpu->run->s.regs.gprs[0] = 0; @@ -190,9 +226,12 @@ static int handle_test_block(struct kvm_vcpu *vcpu) static int handle_tpi(struct kvm_vcpu *vcpu) { struct kvm_s390_interrupt_info *inti; + unsigned long len; + u32 tpi_data[3]; + int cc, rc; u64 addr; - int cc; + rc = 0; addr = kvm_s390_get_base_disp_s(vcpu); if (addr & 3) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); @@ -201,30 +240,41 @@ static int handle_tpi(struct kvm_vcpu *vcpu) if (!inti) goto no_interrupt; cc = 1; + tpi_data[0] = inti->io.subchannel_id << 16 | inti->io.subchannel_nr; + tpi_data[1] = inti->io.io_int_parm; + tpi_data[2] = inti->io.io_int_word; if (addr) { /* * Store the two-word I/O interruption code into the * provided area. */ - if (put_guest(vcpu, inti->io.subchannel_id, (u16 __user *)addr) - || put_guest(vcpu, inti->io.subchannel_nr, (u16 __user *)(addr + 2)) - || put_guest(vcpu, inti->io.io_int_parm, (u32 __user *)(addr + 4))) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + len = sizeof(tpi_data) - 4; + rc = write_guest(vcpu, addr, &tpi_data, len); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); } else { /* * Store the three-word I/O interruption code into * the appropriate lowcore area. */ - put_guest(vcpu, inti->io.subchannel_id, (u16 __user *) __LC_SUBCHANNEL_ID); - put_guest(vcpu, inti->io.subchannel_nr, (u16 __user *) __LC_SUBCHANNEL_NR); - put_guest(vcpu, inti->io.io_int_parm, (u32 __user *) __LC_IO_INT_PARM); - put_guest(vcpu, inti->io.io_int_word, (u32 __user *) __LC_IO_INT_WORD); + len = sizeof(tpi_data); + if (write_guest_lc(vcpu, __LC_SUBCHANNEL_ID, &tpi_data, len)) + rc = -EFAULT; } - kfree(inti); + /* + * If we encounter a problem storing the interruption code, the + * instruction is suppressed from the guest's view: reinject the + * interrupt. + */ + if (!rc) + kfree(inti); + else + kvm_s390_reinject_io_int(vcpu->kvm, inti); no_interrupt: /* Set condition code and we're done. */ - kvm_s390_set_psw_cc(vcpu, cc); - return 0; + if (!rc) + kvm_s390_set_psw_cc(vcpu, cc); + return rc ? -EFAULT : 0; } static int handle_tsch(struct kvm_vcpu *vcpu) @@ -292,10 +342,10 @@ static int handle_stfl(struct kvm_vcpu *vcpu) if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); - rc = copy_to_guest(vcpu, offsetof(struct _lowcore, stfl_fac_list), - vfacilities, 4); + rc = write_guest_lc(vcpu, offsetof(struct _lowcore, stfl_fac_list), + vfacilities, 4); if (rc) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + return rc; VCPU_EVENT(vcpu, 5, "store facility list value %x", *(unsigned int *) vfacilities); trace_kvm_s390_handle_stfl(vcpu, *(unsigned int *) vfacilities); @@ -314,7 +364,8 @@ static void handle_new_psw(struct kvm_vcpu *vcpu) #define PSW_ADDR_24 0x0000000000ffffffUL #define PSW_ADDR_31 0x000000007fffffffUL -static int is_valid_psw(psw_t *psw) { +int is_valid_psw(psw_t *psw) +{ if (psw->mask & PSW_MASK_UNASSIGNED) return 0; if ((psw->mask & PSW_MASK_ADDR_MODE) == PSW_MASK_BA) { @@ -325,6 +376,8 @@ static int is_valid_psw(psw_t *psw) { return 0; if ((psw->mask & PSW_MASK_ADDR_MODE) == PSW_MASK_EA) return 0; + if (psw->addr & 1) + return 0; return 1; } @@ -333,6 +386,7 @@ int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu) psw_t *gpsw = &vcpu->arch.sie_block->gpsw; psw_compat_t new_psw; u64 addr; + int rc; if (gpsw->mask & PSW_MASK_PSTATE) return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); @@ -340,8 +394,10 @@ int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu) addr = kvm_s390_get_base_disp_s(vcpu); if (addr & 7) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); - if (copy_from_guest(vcpu, &new_psw, addr, sizeof(new_psw))) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + + rc = read_guest(vcpu, addr, &new_psw, sizeof(new_psw)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); if (!(new_psw.mask & PSW32_MASK_BASE)) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); gpsw->mask = (new_psw.mask & ~PSW32_MASK_BASE) << 32; @@ -357,6 +413,7 @@ static int handle_lpswe(struct kvm_vcpu *vcpu) { psw_t new_psw; u64 addr; + int rc; if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); @@ -364,8 +421,9 @@ static int handle_lpswe(struct kvm_vcpu *vcpu) addr = kvm_s390_get_base_disp_s(vcpu); if (addr & 7) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); - if (copy_from_guest(vcpu, &new_psw, addr, sizeof(new_psw))) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + rc = read_guest(vcpu, addr, &new_psw, sizeof(new_psw)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); vcpu->arch.sie_block->gpsw = new_psw; if (!is_valid_psw(&vcpu->arch.sie_block->gpsw)) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); @@ -375,7 +433,9 @@ static int handle_lpswe(struct kvm_vcpu *vcpu) static int handle_stidp(struct kvm_vcpu *vcpu) { + u64 stidp_data = vcpu->arch.stidp_data; u64 operand2; + int rc; vcpu->stat.instruction_stidp++; @@ -387,8 +447,9 @@ static int handle_stidp(struct kvm_vcpu *vcpu) if (operand2 & 7) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); - if (put_guest(vcpu, vcpu->arch.stidp_data, (u64 __user *)operand2)) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + rc = write_guest(vcpu, operand2, &stidp_data, sizeof(stidp_data)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); VCPU_EVENT(vcpu, 5, "%s", "store cpu id"); return 0; @@ -474,9 +535,10 @@ static int handle_stsi(struct kvm_vcpu *vcpu) break; } - if (copy_to_guest_absolute(vcpu, operand2, (void *) mem, PAGE_SIZE)) { - rc = kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); - goto out_exception; + rc = write_guest(vcpu, operand2, (void *)mem, PAGE_SIZE); + if (rc) { + rc = kvm_s390_inject_prog_cond(vcpu, rc); + goto out; } trace_kvm_s390_handle_stsi(vcpu, fc, sel1, sel2, operand2); free_page(mem); @@ -485,7 +547,7 @@ static int handle_stsi(struct kvm_vcpu *vcpu) return 0; out_no_data: kvm_s390_set_psw_cc(vcpu, 3); -out_exception: +out: free_page(mem); return rc; } @@ -496,6 +558,7 @@ static const intercept_handler_t b2_handlers[256] = { [0x10] = handle_set_prefix, [0x11] = handle_store_prefix, [0x12] = handle_store_cpu_address, + [0x21] = handle_ipte_interlock, [0x29] = handle_skey, [0x2a] = handle_skey, [0x2b] = handle_skey, @@ -513,6 +576,7 @@ static const intercept_handler_t b2_handlers[256] = { [0x3a] = handle_io_inst, [0x3b] = handle_io_inst, [0x3c] = handle_io_inst, + [0x50] = handle_ipte_interlock, [0x5f] = handle_io_inst, [0x74] = handle_io_inst, [0x76] = handle_io_inst, @@ -591,6 +655,11 @@ static int handle_pfmf(struct kvm_vcpu *vcpu) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); start = vcpu->run->s.regs.gprs[reg2] & PAGE_MASK; + if (vcpu->run->s.regs.gprs[reg1] & PFMF_CF) { + if (kvm_s390_check_low_addr_protection(vcpu, start)) + return kvm_s390_inject_prog_irq(vcpu, &vcpu->arch.pgm); + } + switch (vcpu->run->s.regs.gprs[reg1] & PFMF_FSC) { case 0x00000000: end = (start + (1UL << 12)) & ~((1UL << 12) - 1); @@ -606,10 +675,15 @@ static int handle_pfmf(struct kvm_vcpu *vcpu) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); } while (start < end) { - unsigned long useraddr; - - useraddr = gmap_translate(start, vcpu->arch.gmap); - if (IS_ERR((void *)useraddr)) + unsigned long useraddr, abs_addr; + + /* Translate guest address to host address */ + if ((vcpu->run->s.regs.gprs[reg1] & PFMF_FSC) == 0) + abs_addr = kvm_s390_real_to_abs(vcpu, start); + else + abs_addr = start; + useraddr = gfn_to_hva(vcpu->kvm, gpa_to_gfn(abs_addr)); + if (kvm_is_error_hva(useraddr)) return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); if (vcpu->run->s.regs.gprs[reg1] & PFMF_CF) { @@ -618,6 +692,7 @@ static int handle_pfmf(struct kvm_vcpu *vcpu) } if (vcpu->run->s.regs.gprs[reg1] & PFMF_SK) { + __skey_check_enable(vcpu); if (set_guest_storage_key(current->mm, useraddr, vcpu->run->s.regs.gprs[reg1] & PFMF_KEY, vcpu->run->s.regs.gprs[reg1] & PFMF_NQ)) @@ -642,7 +717,7 @@ static int handle_essa(struct kvm_vcpu *vcpu) VCPU_EVENT(vcpu, 5, "cmma release %d pages", entries); gmap = vcpu->arch.gmap; vcpu->stat.instruction_essa++; - if (!kvm_enabled_cmma() || !vcpu->arch.sie_block->cbrlo) + if (!kvm_s390_cmma_enabled(vcpu->kvm)) return kvm_s390_inject_program_int(vcpu, PGM_OPERATION); if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) @@ -672,7 +747,10 @@ static int handle_essa(struct kvm_vcpu *vcpu) } static const intercept_handler_t b9_handlers[256] = { + [0x8a] = handle_ipte_interlock, [0x8d] = handle_epsw, + [0x8e] = handle_ipte_interlock, + [0x8f] = handle_ipte_interlock, [0xab] = handle_essa, [0xaf] = handle_pfmf, }; @@ -693,32 +771,67 @@ int kvm_s390_handle_lctl(struct kvm_vcpu *vcpu) { int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; int reg3 = vcpu->arch.sie_block->ipa & 0x000f; - u64 useraddr; u32 val = 0; int reg, rc; + u64 ga; vcpu->stat.instruction_lctl++; if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); - useraddr = kvm_s390_get_base_disp_rs(vcpu); + ga = kvm_s390_get_base_disp_rs(vcpu); - if (useraddr & 3) + if (ga & 3) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); - VCPU_EVENT(vcpu, 5, "lctl r1:%x, r3:%x, addr:%llx", reg1, reg3, - useraddr); - trace_kvm_s390_handle_lctl(vcpu, 0, reg1, reg3, useraddr); + VCPU_EVENT(vcpu, 5, "lctl r1:%x, r3:%x, addr:%llx", reg1, reg3, ga); + trace_kvm_s390_handle_lctl(vcpu, 0, reg1, reg3, ga); reg = reg1; do { - rc = get_guest(vcpu, val, (u32 __user *) useraddr); + rc = read_guest(vcpu, ga, &val, sizeof(val)); if (rc) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + return kvm_s390_inject_prog_cond(vcpu, rc); vcpu->arch.sie_block->gcr[reg] &= 0xffffffff00000000ul; vcpu->arch.sie_block->gcr[reg] |= val; - useraddr += 4; + ga += 4; + if (reg == reg3) + break; + reg = (reg + 1) % 16; + } while (1); + + return 0; +} + +int kvm_s390_handle_stctl(struct kvm_vcpu *vcpu) +{ + int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; + int reg3 = vcpu->arch.sie_block->ipa & 0x000f; + u64 ga; + u32 val; + int reg, rc; + + vcpu->stat.instruction_stctl++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + ga = kvm_s390_get_base_disp_rs(vcpu); + + if (ga & 3) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + VCPU_EVENT(vcpu, 5, "stctl r1:%x, r3:%x, addr:%llx", reg1, reg3, ga); + trace_kvm_s390_handle_stctl(vcpu, 0, reg1, reg3, ga); + + reg = reg1; + do { + val = vcpu->arch.sie_block->gcr[reg] & 0x00000000fffffffful; + rc = write_guest(vcpu, ga, &val, sizeof(val)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + ga += 4; if (reg == reg3) break; reg = (reg + 1) % 16; @@ -731,7 +844,7 @@ static int handle_lctlg(struct kvm_vcpu *vcpu) { int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; int reg3 = vcpu->arch.sie_block->ipa & 0x000f; - u64 useraddr; + u64 ga, val; int reg, rc; vcpu->stat.instruction_lctlg++; @@ -739,23 +852,58 @@ static int handle_lctlg(struct kvm_vcpu *vcpu) if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); - useraddr = kvm_s390_get_base_disp_rsy(vcpu); + ga = kvm_s390_get_base_disp_rsy(vcpu); - if (useraddr & 7) + if (ga & 7) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); reg = reg1; - VCPU_EVENT(vcpu, 5, "lctlg r1:%x, r3:%x, addr:%llx", reg1, reg3, - useraddr); - trace_kvm_s390_handle_lctl(vcpu, 1, reg1, reg3, useraddr); + VCPU_EVENT(vcpu, 5, "lctlg r1:%x, r3:%x, addr:%llx", reg1, reg3, ga); + trace_kvm_s390_handle_lctl(vcpu, 1, reg1, reg3, ga); do { - rc = get_guest(vcpu, vcpu->arch.sie_block->gcr[reg], - (u64 __user *) useraddr); + rc = read_guest(vcpu, ga, &val, sizeof(val)); if (rc) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); - useraddr += 8; + return kvm_s390_inject_prog_cond(vcpu, rc); + vcpu->arch.sie_block->gcr[reg] = val; + ga += 8; + if (reg == reg3) + break; + reg = (reg + 1) % 16; + } while (1); + + return 0; +} + +static int handle_stctg(struct kvm_vcpu *vcpu) +{ + int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; + int reg3 = vcpu->arch.sie_block->ipa & 0x000f; + u64 ga, val; + int reg, rc; + + vcpu->stat.instruction_stctg++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + ga = kvm_s390_get_base_disp_rsy(vcpu); + + if (ga & 7) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + reg = reg1; + + VCPU_EVENT(vcpu, 5, "stctg r1:%x, r3:%x, addr:%llx", reg1, reg3, ga); + trace_kvm_s390_handle_stctl(vcpu, 1, reg1, reg3, ga); + + do { + val = vcpu->arch.sie_block->gcr[reg]; + rc = write_guest(vcpu, ga, &val, sizeof(val)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + ga += 8; if (reg == reg3) break; reg = (reg + 1) % 16; @@ -766,6 +914,7 @@ static int handle_lctlg(struct kvm_vcpu *vcpu) static const intercept_handler_t eb_handlers[256] = { [0x2f] = handle_lctlg, + [0x25] = handle_stctg, }; int kvm_s390_handle_eb(struct kvm_vcpu *vcpu) @@ -781,8 +930,9 @@ int kvm_s390_handle_eb(struct kvm_vcpu *vcpu) static int handle_tprot(struct kvm_vcpu *vcpu) { u64 address1, address2; - struct vm_area_struct *vma; - unsigned long user_address; + unsigned long hva, gpa; + int ret = 0, cc = 0; + bool writable; vcpu->stat.instruction_tprot++; @@ -793,32 +943,41 @@ static int handle_tprot(struct kvm_vcpu *vcpu) /* we only handle the Linux memory detection case: * access key == 0 - * guest DAT == off * everything else goes to userspace. */ if (address2 & 0xf0) return -EOPNOTSUPP; if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_DAT) - return -EOPNOTSUPP; - - down_read(¤t->mm->mmap_sem); - user_address = __gmap_translate(address1, vcpu->arch.gmap); - if (IS_ERR_VALUE(user_address)) - goto out_inject; - vma = find_vma(current->mm, user_address); - if (!vma) - goto out_inject; - vcpu->arch.sie_block->gpsw.mask &= ~(3ul << 44); - if (!(vma->vm_flags & VM_WRITE) && (vma->vm_flags & VM_READ)) - vcpu->arch.sie_block->gpsw.mask |= (1ul << 44); - if (!(vma->vm_flags & VM_WRITE) && !(vma->vm_flags & VM_READ)) - vcpu->arch.sie_block->gpsw.mask |= (2ul << 44); - - up_read(¤t->mm->mmap_sem); - return 0; + ipte_lock(vcpu); + ret = guest_translate_address(vcpu, address1, &gpa, 1); + if (ret == PGM_PROTECTION) { + /* Write protected? Try again with read-only... */ + cc = 1; + ret = guest_translate_address(vcpu, address1, &gpa, 0); + } + if (ret) { + if (ret == PGM_ADDRESSING || ret == PGM_TRANSLATION_SPEC) { + ret = kvm_s390_inject_program_int(vcpu, ret); + } else if (ret > 0) { + /* Translation not available */ + kvm_s390_set_psw_cc(vcpu, 3); + ret = 0; + } + goto out_unlock; + } -out_inject: - up_read(¤t->mm->mmap_sem); - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + hva = gfn_to_hva_prot(vcpu->kvm, gpa_to_gfn(gpa), &writable); + if (kvm_is_error_hva(hva)) { + ret = kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + } else { + if (!writable) + cc = 1; /* Write not permitted ==> read-only */ + kvm_s390_set_psw_cc(vcpu, cc); + /* Note: CC2 only occurs for storage keys (not supported yet) */ + } +out_unlock: + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_DAT) + ipte_unlock(vcpu); + return ret; } int kvm_s390_handle_e5(struct kvm_vcpu *vcpu) diff --git a/arch/s390/kvm/sigp.c b/arch/s390/kvm/sigp.c index 26caeb530a78..43079a48cc98 100644 --- a/arch/s390/kvm/sigp.c +++ b/arch/s390/kvm/sigp.c @@ -54,33 +54,23 @@ static int __sigp_sense(struct kvm_vcpu *vcpu, u16 cpu_addr, static int __sigp_emergency(struct kvm_vcpu *vcpu, u16 cpu_addr) { - struct kvm_s390_local_interrupt *li; - struct kvm_s390_interrupt_info *inti; + struct kvm_s390_interrupt s390int = { + .type = KVM_S390_INT_EMERGENCY, + .parm = vcpu->vcpu_id, + }; struct kvm_vcpu *dst_vcpu = NULL; + int rc = 0; if (cpu_addr < KVM_MAX_VCPUS) dst_vcpu = kvm_get_vcpu(vcpu->kvm, cpu_addr); if (!dst_vcpu) return SIGP_CC_NOT_OPERATIONAL; - inti = kzalloc(sizeof(*inti), GFP_KERNEL); - if (!inti) - return -ENOMEM; - - inti->type = KVM_S390_INT_EMERGENCY; - inti->emerg.code = vcpu->vcpu_id; - - li = &dst_vcpu->arch.local_int; - spin_lock_bh(&li->lock); - list_add_tail(&inti->list, &li->list); - atomic_set(&li->active, 1); - atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags); - if (waitqueue_active(li->wq)) - wake_up_interruptible(li->wq); - spin_unlock_bh(&li->lock); - VCPU_EVENT(vcpu, 4, "sent sigp emerg to cpu %x", cpu_addr); + rc = kvm_s390_inject_vcpu(dst_vcpu, &s390int); + if (!rc) + VCPU_EVENT(vcpu, 4, "sent sigp emerg to cpu %x", cpu_addr); - return SIGP_CC_ORDER_CODE_ACCEPTED; + return rc ? rc : SIGP_CC_ORDER_CODE_ACCEPTED; } static int __sigp_conditional_emergency(struct kvm_vcpu *vcpu, u16 cpu_addr, @@ -116,33 +106,23 @@ static int __sigp_conditional_emergency(struct kvm_vcpu *vcpu, u16 cpu_addr, static int __sigp_external_call(struct kvm_vcpu *vcpu, u16 cpu_addr) { - struct kvm_s390_local_interrupt *li; - struct kvm_s390_interrupt_info *inti; + struct kvm_s390_interrupt s390int = { + .type = KVM_S390_INT_EXTERNAL_CALL, + .parm = vcpu->vcpu_id, + }; struct kvm_vcpu *dst_vcpu = NULL; + int rc; if (cpu_addr < KVM_MAX_VCPUS) dst_vcpu = kvm_get_vcpu(vcpu->kvm, cpu_addr); if (!dst_vcpu) return SIGP_CC_NOT_OPERATIONAL; - inti = kzalloc(sizeof(*inti), GFP_KERNEL); - if (!inti) - return -ENOMEM; + rc = kvm_s390_inject_vcpu(dst_vcpu, &s390int); + if (!rc) + VCPU_EVENT(vcpu, 4, "sent sigp ext call to cpu %x", cpu_addr); - inti->type = KVM_S390_INT_EXTERNAL_CALL; - inti->extcall.code = vcpu->vcpu_id; - - li = &dst_vcpu->arch.local_int; - spin_lock_bh(&li->lock); - list_add_tail(&inti->list, &li->list); - atomic_set(&li->active, 1); - atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags); - if (waitqueue_active(li->wq)) - wake_up_interruptible(li->wq); - spin_unlock_bh(&li->lock); - VCPU_EVENT(vcpu, 4, "sent sigp ext call to cpu %x", cpu_addr); - - return SIGP_CC_ORDER_CODE_ACCEPTED; + return rc ? rc : SIGP_CC_ORDER_CODE_ACCEPTED; } static int __inject_sigp_stop(struct kvm_s390_local_interrupt *li, int action) @@ -235,7 +215,6 @@ static int __sigp_set_prefix(struct kvm_vcpu *vcpu, u16 cpu_addr, u32 address, struct kvm_vcpu *dst_vcpu = NULL; struct kvm_s390_interrupt_info *inti; int rc; - u8 tmp; if (cpu_addr < KVM_MAX_VCPUS) dst_vcpu = kvm_get_vcpu(vcpu->kvm, cpu_addr); @@ -243,10 +222,13 @@ static int __sigp_set_prefix(struct kvm_vcpu *vcpu, u16 cpu_addr, u32 address, return SIGP_CC_NOT_OPERATIONAL; li = &dst_vcpu->arch.local_int; - /* make sure that the new value is valid memory */ - address = address & 0x7fffe000u; - if (copy_from_guest_absolute(vcpu, &tmp, address, 1) || - copy_from_guest_absolute(vcpu, &tmp, address + PAGE_SIZE, 1)) { + /* + * Make sure the new value is valid memory. We only need to check the + * first page, since address is 8k aligned and memory pieces are always + * at least 1MB aligned and have at least a size of 1MB. + */ + address &= 0x7fffe000u; + if (kvm_is_error_gpa(vcpu->kvm, address)) { *reg &= 0xffffffff00000000UL; *reg |= SIGP_STATUS_INVALID_PARAMETER; return SIGP_CC_STATUS_STORED; @@ -456,3 +438,38 @@ int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu) kvm_s390_set_psw_cc(vcpu, rc); return 0; } + +/* + * Handle SIGP partial execution interception. + * + * This interception will occur at the source cpu when a source cpu sends an + * external call to a target cpu and the target cpu has the WAIT bit set in + * its cpuflags. Interception will occurr after the interrupt indicator bits at + * the target cpu have been set. All error cases will lead to instruction + * interception, therefore nothing is to be checked or prepared. + */ +int kvm_s390_handle_sigp_pei(struct kvm_vcpu *vcpu) +{ + int r3 = vcpu->arch.sie_block->ipa & 0x000f; + u16 cpu_addr = vcpu->run->s.regs.gprs[r3]; + struct kvm_vcpu *dest_vcpu; + u8 order_code = kvm_s390_get_base_disp_rs(vcpu); + + trace_kvm_s390_handle_sigp_pei(vcpu, order_code, cpu_addr); + + if (order_code == SIGP_EXTERNAL_CALL) { + dest_vcpu = kvm_get_vcpu(vcpu->kvm, cpu_addr); + BUG_ON(dest_vcpu == NULL); + + spin_lock_bh(&dest_vcpu->arch.local_int.lock); + if (waitqueue_active(&dest_vcpu->wq)) + wake_up_interruptible(&dest_vcpu->wq); + dest_vcpu->preempted = true; + spin_unlock_bh(&dest_vcpu->arch.local_int.lock); + + kvm_s390_set_psw_cc(vcpu, SIGP_CC_ORDER_CODE_ACCEPTED); + return 0; + } + + return -EOPNOTSUPP; +} diff --git a/arch/s390/kvm/trace-s390.h b/arch/s390/kvm/trace-s390.h index 13f30f58a2df..647e9d6a4818 100644 --- a/arch/s390/kvm/trace-s390.h +++ b/arch/s390/kvm/trace-s390.h @@ -68,6 +68,27 @@ TRACE_EVENT(kvm_s390_destroy_vcpu, ); /* + * Trace point for start and stop of vpcus. + */ +TRACE_EVENT(kvm_s390_vcpu_start_stop, + TP_PROTO(unsigned int id, int state), + TP_ARGS(id, state), + + TP_STRUCT__entry( + __field(unsigned int, id) + __field(int, state) + ), + + TP_fast_assign( + __entry->id = id; + __entry->state = state; + ), + + TP_printk("%s cpu %d", __entry->state ? "starting" : "stopping", + __entry->id) + ); + +/* * Trace points for injection of interrupts, either per machine or * per vcpu. */ @@ -223,6 +244,28 @@ TRACE_EVENT(kvm_s390_enable_css, __entry->kvm) ); +/* + * Trace point for enabling and disabling interlocking-and-broadcasting + * suppression. + */ +TRACE_EVENT(kvm_s390_enable_disable_ibs, + TP_PROTO(unsigned int id, int state), + TP_ARGS(id, state), + + TP_STRUCT__entry( + __field(unsigned int, id) + __field(int, state) + ), + + TP_fast_assign( + __entry->id = id; + __entry->state = state; + ), + + TP_printk("%s ibs on cpu %d", + __entry->state ? "enabling" : "disabling", __entry->id) + ); + #endif /* _TRACE_KVMS390_H */ diff --git a/arch/s390/kvm/trace.h b/arch/s390/kvm/trace.h index e8e7213d4cc5..916834d7a73a 100644 --- a/arch/s390/kvm/trace.h +++ b/arch/s390/kvm/trace.h @@ -2,7 +2,7 @@ #define _TRACE_KVM_H #include <linux/tracepoint.h> -#include <asm/sigp.h> +#include <asm/sie.h> #include <asm/debug.h> #include <asm/dis.h> @@ -30,6 +30,20 @@ TP_printk("%02d[%016lx-%016lx]: " p_str, __entry->id, \ __entry->pswmask, __entry->pswaddr, p_args) +TRACE_EVENT(kvm_s390_skey_related_inst, + TP_PROTO(VCPU_PROTO_COMMON), + TP_ARGS(VCPU_ARGS_COMMON), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + ), + VCPU_TP_PRINTK("%s", "first instruction related to skeys on vcpu") + ); + TRACE_EVENT(kvm_s390_major_guest_pfault, TP_PROTO(VCPU_PROTO_COMMON), TP_ARGS(VCPU_ARGS_COMMON), @@ -111,17 +125,6 @@ TRACE_EVENT(kvm_s390_sie_fault, VCPU_TP_PRINTK("%s", "fault in sie instruction") ); -#define sie_intercept_code \ - {0x04, "Instruction"}, \ - {0x08, "Program interruption"}, \ - {0x0C, "Instruction and program interruption"}, \ - {0x10, "External request"}, \ - {0x14, "External interruption"}, \ - {0x18, "I/O request"}, \ - {0x1C, "Wait state"}, \ - {0x20, "Validity"}, \ - {0x28, "Stop request"} - TRACE_EVENT(kvm_s390_sie_exit, TP_PROTO(VCPU_PROTO_COMMON, u8 icptcode), TP_ARGS(VCPU_ARGS_COMMON, icptcode), @@ -151,7 +154,6 @@ TRACE_EVENT(kvm_s390_intercept_instruction, TP_STRUCT__entry( VCPU_FIELD_COMMON __field(__u64, instruction) - __field(char, insn[8]) ), TP_fast_assign( @@ -162,10 +164,8 @@ TRACE_EVENT(kvm_s390_intercept_instruction, VCPU_TP_PRINTK("intercepted instruction %016llx (%s)", __entry->instruction, - insn_to_mnemonic((unsigned char *) - &__entry->instruction, - __entry->insn, sizeof(__entry->insn)) ? - "unknown" : __entry->insn) + __print_symbolic(icpt_insn_decoder(__entry->instruction), + icpt_insn_codes)) ); /* @@ -213,18 +213,6 @@ TRACE_EVENT(kvm_s390_intercept_validity, * Trace points for instructions that are of special interest. */ -#define sigp_order_codes \ - {SIGP_SENSE, "sense"}, \ - {SIGP_EXTERNAL_CALL, "external call"}, \ - {SIGP_EMERGENCY_SIGNAL, "emergency signal"}, \ - {SIGP_STOP, "stop"}, \ - {SIGP_STOP_AND_STORE_STATUS, "stop and store status"}, \ - {SIGP_SET_ARCHITECTURE, "set architecture"}, \ - {SIGP_SET_PREFIX, "set prefix"}, \ - {SIGP_STORE_STATUS_AT_ADDRESS, "store status at addr"}, \ - {SIGP_SENSE_RUNNING, "sense running"}, \ - {SIGP_RESTART, "restart"} - TRACE_EVENT(kvm_s390_handle_sigp, TP_PROTO(VCPU_PROTO_COMMON, __u8 order_code, __u16 cpu_addr, \ __u32 parameter), @@ -251,12 +239,28 @@ TRACE_EVENT(kvm_s390_handle_sigp, __entry->cpu_addr, __entry->parameter) ); -#define diagnose_codes \ - {0x10, "release pages"}, \ - {0x44, "time slice end"}, \ - {0x308, "ipl functions"}, \ - {0x500, "kvm hypercall"}, \ - {0x501, "kvm breakpoint"} +TRACE_EVENT(kvm_s390_handle_sigp_pei, + TP_PROTO(VCPU_PROTO_COMMON, __u8 order_code, __u16 cpu_addr), + TP_ARGS(VCPU_ARGS_COMMON, order_code, cpu_addr), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(__u8, order_code) + __field(__u16, cpu_addr) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->order_code = order_code; + __entry->cpu_addr = cpu_addr; + ), + + VCPU_TP_PRINTK("handle sigp pei order %02x (%s), cpu address %04x", + __entry->order_code, + __print_symbolic(__entry->order_code, + sigp_order_codes), + __entry->cpu_addr) + ); TRACE_EVENT(kvm_s390_handle_diag, TP_PROTO(VCPU_PROTO_COMMON, __u16 code), @@ -301,6 +305,31 @@ TRACE_EVENT(kvm_s390_handle_lctl, __entry->reg1, __entry->reg3, __entry->addr) ); +TRACE_EVENT(kvm_s390_handle_stctl, + TP_PROTO(VCPU_PROTO_COMMON, int g, int reg1, int reg3, u64 addr), + TP_ARGS(VCPU_ARGS_COMMON, g, reg1, reg3, addr), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(int, g) + __field(int, reg1) + __field(int, reg3) + __field(u64, addr) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->g = g; + __entry->reg1 = reg1; + __entry->reg3 = reg3; + __entry->addr = addr; + ), + + VCPU_TP_PRINTK("%s: storing cr %x-%x to %016llx", + __entry->g ? "stctg" : "stctl", + __entry->reg1, __entry->reg3, __entry->addr) + ); + TRACE_EVENT(kvm_s390_handle_prefix, TP_PROTO(VCPU_PROTO_COMMON, int set, u32 address), TP_ARGS(VCPU_ARGS_COMMON, set, address), diff --git a/arch/s390/lib/spinlock.c b/arch/s390/lib/spinlock.c index f709983f41f8..5b0e445bc3f3 100644 --- a/arch/s390/lib/spinlock.c +++ b/arch/s390/lib/spinlock.c @@ -26,83 +26,81 @@ __setup("spin_retry=", spin_retry_setup); void arch_spin_lock_wait(arch_spinlock_t *lp) { - int count = spin_retry; - unsigned int cpu = ~smp_processor_id(); + unsigned int cpu = SPINLOCK_LOCKVAL; unsigned int owner; + int count; while (1) { - owner = lp->owner_cpu; - if (!owner || smp_vcpu_scheduled(~owner)) { - for (count = spin_retry; count > 0; count--) { - if (arch_spin_is_locked(lp)) - continue; - if (_raw_compare_and_swap(&lp->owner_cpu, 0, - cpu) == 0) - return; - } - if (MACHINE_IS_LPAR) - continue; + owner = ACCESS_ONCE(lp->lock); + /* Try to get the lock if it is free. */ + if (!owner) { + if (_raw_compare_and_swap(&lp->lock, 0, cpu)) + return; + continue; } - owner = lp->owner_cpu; - if (owner) + /* Check if the lock owner is running. */ + if (!smp_vcpu_scheduled(~owner)) { + smp_yield_cpu(~owner); + continue; + } + /* Loop for a while on the lock value. */ + count = spin_retry; + do { + owner = ACCESS_ONCE(lp->lock); + } while (owner && count-- > 0); + if (!owner) + continue; + /* + * For multiple layers of hypervisors, e.g. z/VM + LPAR + * yield the CPU if the lock is still unavailable. + */ + if (!MACHINE_IS_LPAR) smp_yield_cpu(~owner); - if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0) - return; } } EXPORT_SYMBOL(arch_spin_lock_wait); void arch_spin_lock_wait_flags(arch_spinlock_t *lp, unsigned long flags) { - int count = spin_retry; - unsigned int cpu = ~smp_processor_id(); + unsigned int cpu = SPINLOCK_LOCKVAL; unsigned int owner; + int count; local_irq_restore(flags); while (1) { - owner = lp->owner_cpu; - if (!owner || smp_vcpu_scheduled(~owner)) { - for (count = spin_retry; count > 0; count--) { - if (arch_spin_is_locked(lp)) - continue; - local_irq_disable(); - if (_raw_compare_and_swap(&lp->owner_cpu, 0, - cpu) == 0) - return; - local_irq_restore(flags); - } - if (MACHINE_IS_LPAR) - continue; + owner = ACCESS_ONCE(lp->lock); + /* Try to get the lock if it is free. */ + if (!owner) { + local_irq_disable(); + if (_raw_compare_and_swap(&lp->lock, 0, cpu)) + return; + local_irq_restore(flags); } - owner = lp->owner_cpu; - if (owner) + /* Check if the lock owner is running. */ + if (!smp_vcpu_scheduled(~owner)) { smp_yield_cpu(~owner); - local_irq_disable(); - if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0) - return; - local_irq_restore(flags); - } -} -EXPORT_SYMBOL(arch_spin_lock_wait_flags); - -int arch_spin_trylock_retry(arch_spinlock_t *lp) -{ - unsigned int cpu = ~smp_processor_id(); - int count; - - for (count = spin_retry; count > 0; count--) { - if (arch_spin_is_locked(lp)) continue; - if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0) - return 1; + } + /* Loop for a while on the lock value. */ + count = spin_retry; + do { + owner = ACCESS_ONCE(lp->lock); + } while (owner && count-- > 0); + if (!owner) + continue; + /* + * For multiple layers of hypervisors, e.g. z/VM + LPAR + * yield the CPU if the lock is still unavailable. + */ + if (!MACHINE_IS_LPAR) + smp_yield_cpu(~owner); } - return 0; } -EXPORT_SYMBOL(arch_spin_trylock_retry); +EXPORT_SYMBOL(arch_spin_lock_wait_flags); -void arch_spin_relax(arch_spinlock_t *lock) +void arch_spin_relax(arch_spinlock_t *lp) { - unsigned int cpu = lock->owner_cpu; + unsigned int cpu = lp->lock; if (cpu != 0) { if (MACHINE_IS_VM || MACHINE_IS_KVM || !smp_vcpu_scheduled(~cpu)) @@ -111,6 +109,17 @@ void arch_spin_relax(arch_spinlock_t *lock) } EXPORT_SYMBOL(arch_spin_relax); +int arch_spin_trylock_retry(arch_spinlock_t *lp) +{ + int count; + + for (count = spin_retry; count > 0; count--) + if (arch_spin_trylock_once(lp)) + return 1; + return 0; +} +EXPORT_SYMBOL(arch_spin_trylock_retry); + void _raw_read_lock_wait(arch_rwlock_t *rw) { unsigned int old; @@ -121,10 +130,10 @@ void _raw_read_lock_wait(arch_rwlock_t *rw) smp_yield(); count = spin_retry; } - if (!arch_read_can_lock(rw)) + old = ACCESS_ONCE(rw->lock); + if ((int) old < 0) continue; - old = rw->lock & 0x7fffffffU; - if (_raw_compare_and_swap(&rw->lock, old, old + 1) == old) + if (_raw_compare_and_swap(&rw->lock, old, old + 1)) return; } } @@ -141,12 +150,13 @@ void _raw_read_lock_wait_flags(arch_rwlock_t *rw, unsigned long flags) smp_yield(); count = spin_retry; } - if (!arch_read_can_lock(rw)) + old = ACCESS_ONCE(rw->lock); + if ((int) old < 0) continue; - old = rw->lock & 0x7fffffffU; local_irq_disable(); - if (_raw_compare_and_swap(&rw->lock, old, old + 1) == old) + if (_raw_compare_and_swap(&rw->lock, old, old + 1)) return; + local_irq_restore(flags); } } EXPORT_SYMBOL(_raw_read_lock_wait_flags); @@ -157,10 +167,10 @@ int _raw_read_trylock_retry(arch_rwlock_t *rw) int count = spin_retry; while (count-- > 0) { - if (!arch_read_can_lock(rw)) + old = ACCESS_ONCE(rw->lock); + if ((int) old < 0) continue; - old = rw->lock & 0x7fffffffU; - if (_raw_compare_and_swap(&rw->lock, old, old + 1) == old) + if (_raw_compare_and_swap(&rw->lock, old, old + 1)) return 1; } return 0; @@ -169,6 +179,7 @@ EXPORT_SYMBOL(_raw_read_trylock_retry); void _raw_write_lock_wait(arch_rwlock_t *rw) { + unsigned int old; int count = spin_retry; while (1) { @@ -176,9 +187,10 @@ void _raw_write_lock_wait(arch_rwlock_t *rw) smp_yield(); count = spin_retry; } - if (!arch_write_can_lock(rw)) + old = ACCESS_ONCE(rw->lock); + if (old) continue; - if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000) == 0) + if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000)) return; } } @@ -186,6 +198,7 @@ EXPORT_SYMBOL(_raw_write_lock_wait); void _raw_write_lock_wait_flags(arch_rwlock_t *rw, unsigned long flags) { + unsigned int old; int count = spin_retry; local_irq_restore(flags); @@ -194,23 +207,27 @@ void _raw_write_lock_wait_flags(arch_rwlock_t *rw, unsigned long flags) smp_yield(); count = spin_retry; } - if (!arch_write_can_lock(rw)) + old = ACCESS_ONCE(rw->lock); + if (old) continue; local_irq_disable(); - if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000) == 0) + if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000)) return; + local_irq_restore(flags); } } EXPORT_SYMBOL(_raw_write_lock_wait_flags); int _raw_write_trylock_retry(arch_rwlock_t *rw) { + unsigned int old; int count = spin_retry; while (count-- > 0) { - if (!arch_write_can_lock(rw)) + old = ACCESS_ONCE(rw->lock); + if (old) continue; - if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000) == 0) + if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000)) return 1; } return 0; diff --git a/arch/s390/lib/uaccess.c b/arch/s390/lib/uaccess.c index 7416efe8eae4..53dd5d7a0c96 100644 --- a/arch/s390/lib/uaccess.c +++ b/arch/s390/lib/uaccess.c @@ -76,7 +76,7 @@ static inline unsigned long copy_from_user_mvcp(void *x, const void __user *ptr, { unsigned long tmp1, tmp2; - update_primary_asce(current); + load_kernel_asce(); tmp1 = -256UL; asm volatile( " sacf 0\n" @@ -159,7 +159,7 @@ static inline unsigned long copy_to_user_mvcs(void __user *ptr, const void *x, { unsigned long tmp1, tmp2; - update_primary_asce(current); + load_kernel_asce(); tmp1 = -256UL; asm volatile( " sacf 0\n" @@ -225,7 +225,7 @@ static inline unsigned long copy_in_user_mvc(void __user *to, const void __user { unsigned long tmp1; - update_primary_asce(current); + load_kernel_asce(); asm volatile( " sacf 256\n" " "AHI" %0,-1\n" @@ -292,7 +292,7 @@ static inline unsigned long clear_user_xc(void __user *to, unsigned long size) { unsigned long tmp1, tmp2; - update_primary_asce(current); + load_kernel_asce(); asm volatile( " sacf 256\n" " "AHI" %0,-1\n" @@ -358,7 +358,7 @@ unsigned long __strnlen_user(const char __user *src, unsigned long size) { if (unlikely(!size)) return 0; - update_primary_asce(current); + load_kernel_asce(); return strnlen_user_srst(src, size); } EXPORT_SYMBOL(__strnlen_user); diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c index 2f51a998a67e..3f3b35403d0a 100644 --- a/arch/s390/mm/fault.c +++ b/arch/s390/mm/fault.c @@ -415,7 +415,7 @@ static inline int do_exception(struct pt_regs *regs, int access) * The instruction that caused the program check has * been nullified. Don't signal single step via SIGTRAP. */ - clear_tsk_thread_flag(tsk, TIF_PER_TRAP); + clear_pt_regs_flag(regs, PIF_PER_TRAP); if (notify_page_fault(regs)) return 0; diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c index 0727a55d87d9..0ff66a7e29bb 100644 --- a/arch/s390/mm/hugetlbpage.c +++ b/arch/s390/mm/hugetlbpage.c @@ -220,11 +220,6 @@ int pud_huge(pud_t pud) return 0; } -int pmd_huge_support(void) -{ - return 1; -} - struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmdp, int write) { diff --git a/arch/s390/mm/mem_detect.c b/arch/s390/mm/mem_detect.c index cca388253a39..5535cfe0ee11 100644 --- a/arch/s390/mm/mem_detect.c +++ b/arch/s390/mm/mem_detect.c @@ -6,130 +6,60 @@ #include <linux/kernel.h> #include <linux/module.h> +#include <linux/memblock.h> +#include <linux/init.h> +#include <linux/debugfs.h> +#include <linux/seq_file.h> #include <asm/ipl.h> #include <asm/sclp.h> #include <asm/setup.h> #define ADDR2G (1ULL << 31) -static void find_memory_chunks(struct mem_chunk chunk[], unsigned long maxsize) +#define CHUNK_READ_WRITE 0 +#define CHUNK_READ_ONLY 1 + +static inline void memblock_physmem_add(phys_addr_t start, phys_addr_t size) +{ + memblock_add_range(&memblock.memory, start, size, 0, 0); + memblock_add_range(&memblock.physmem, start, size, 0, 0); +} + +void __init detect_memory_memblock(void) { unsigned long long memsize, rnmax, rzm; - unsigned long addr = 0, size; - int i = 0, type; + unsigned long addr, size; + int type; rzm = sclp_get_rzm(); rnmax = sclp_get_rnmax(); memsize = rzm * rnmax; if (!rzm) rzm = 1ULL << 17; - if (sizeof(long) == 4) { + if (IS_ENABLED(CONFIG_32BIT)) { rzm = min(ADDR2G, rzm); - memsize = memsize ? min(ADDR2G, memsize) : ADDR2G; + memsize = min(ADDR2G, memsize); } - if (maxsize) - memsize = memsize ? min((unsigned long)memsize, maxsize) : maxsize; + max_physmem_end = memsize; + addr = 0; + /* keep memblock lists close to the kernel */ + memblock_set_bottom_up(true); do { size = 0; type = tprot(addr); do { size += rzm; - if (memsize && addr + size >= memsize) + if (max_physmem_end && addr + size >= max_physmem_end) break; } while (type == tprot(addr + size)); if (type == CHUNK_READ_WRITE || type == CHUNK_READ_ONLY) { - if (memsize && (addr + size > memsize)) - size = memsize - addr; - chunk[i].addr = addr; - chunk[i].size = size; - chunk[i].type = type; - i++; + if (max_physmem_end && (addr + size > max_physmem_end)) + size = max_physmem_end - addr; + memblock_physmem_add(addr, size); } addr += size; - } while (addr < memsize && i < MEMORY_CHUNKS); -} - -/** - * detect_memory_layout - fill mem_chunk array with memory layout data - * @chunk: mem_chunk array to be filled - * @maxsize: maximum address where memory detection should stop - * - * Fills the passed in memory chunk array with the memory layout of the - * machine. The array must have a size of at least MEMORY_CHUNKS and will - * be fully initialized afterwards. - * If the maxsize paramater has a value > 0 memory detection will stop at - * that address. It is guaranteed that all chunks have an ending address - * that is smaller than maxsize. - * If maxsize is 0 all memory will be detected. - */ -void detect_memory_layout(struct mem_chunk chunk[], unsigned long maxsize) -{ - unsigned long flags, flags_dat, cr0; - - memset(chunk, 0, MEMORY_CHUNKS * sizeof(struct mem_chunk)); - /* - * Disable IRQs, DAT and low address protection so tprot does the - * right thing and we don't get scheduled away with low address - * protection disabled. - */ - local_irq_save(flags); - flags_dat = __arch_local_irq_stnsm(0xfb); - /* - * In case DAT was enabled, make sure chunk doesn't reside in vmalloc - * space. We have disabled DAT and any access to vmalloc area will - * cause an exception. - * If DAT was disabled we are called from early ipl code. - */ - if (test_bit(5, &flags_dat)) { - if (WARN_ON_ONCE(is_vmalloc_or_module_addr(chunk))) - goto out; - } - __ctl_store(cr0, 0, 0); - __ctl_clear_bit(0, 28); - find_memory_chunks(chunk, maxsize); - __ctl_load(cr0, 0, 0); -out: - __arch_local_irq_ssm(flags_dat); - local_irq_restore(flags); -} -EXPORT_SYMBOL(detect_memory_layout); - -/* - * Create memory hole with given address and size. - */ -void create_mem_hole(struct mem_chunk mem_chunk[], unsigned long addr, - unsigned long size) -{ - int i; - - for (i = 0; i < MEMORY_CHUNKS; i++) { - struct mem_chunk *chunk = &mem_chunk[i]; - - if (chunk->size == 0) - continue; - if (addr > chunk->addr + chunk->size) - continue; - if (addr + size <= chunk->addr) - continue; - /* Split */ - if ((addr > chunk->addr) && - (addr + size < chunk->addr + chunk->size)) { - struct mem_chunk *new = chunk + 1; - - memmove(new, chunk, (MEMORY_CHUNKS-i-1) * sizeof(*new)); - new->addr = addr + size; - new->size = chunk->addr + chunk->size - new->addr; - chunk->size = addr - chunk->addr; - continue; - } else if ((addr <= chunk->addr) && - (addr + size >= chunk->addr + chunk->size)) { - memmove(chunk, chunk + 1, (MEMORY_CHUNKS-i-1) * sizeof(*chunk)); - memset(&mem_chunk[MEMORY_CHUNKS-1], 0, sizeof(*chunk)); - } else if (addr + size < chunk->addr + chunk->size) { - chunk->size = chunk->addr + chunk->size - addr - size; - chunk->addr = addr + size; - } else if (addr > chunk->addr) { - chunk->size = addr - chunk->addr; - } - } + } while (addr < max_physmem_end); + memblock_set_bottom_up(false); + if (!max_physmem_end) + max_physmem_end = memblock_end_of_DRAM(); } diff --git a/arch/s390/mm/page-states.c b/arch/s390/mm/page-states.c index 27c50f4d90cb..a90d45e9dfb0 100644 --- a/arch/s390/mm/page-states.c +++ b/arch/s390/mm/page-states.c @@ -12,8 +12,6 @@ #include <linux/mm.h> #include <linux/gfp.h> #include <linux/init.h> -#include <asm/setup.h> -#include <asm/ipl.h> #define ESSA_SET_STABLE 1 #define ESSA_SET_UNUSED 2 @@ -43,14 +41,6 @@ void __init cmma_init(void) if (!cmma_flag) return; - /* - * Disable CMM for dump, otherwise the tprot based memory - * detection can fail because of unstable pages. - */ - if (OLDMEM_BASE || ipl_info.type == IPL_TYPE_FCP_DUMP) { - cmma_flag = 0; - return; - } asm volatile( " .insn rrf,0xb9ab0000,%1,%1,0,0\n" "0: la %0,0\n" diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c index d7cfd57815fb..37b8241ec784 100644 --- a/arch/s390/mm/pgtable.c +++ b/arch/s390/mm/pgtable.c @@ -53,8 +53,10 @@ static void __crst_table_upgrade(void *arg) { struct mm_struct *mm = arg; - if (current->active_mm == mm) - update_user_asce(mm, 1); + if (current->active_mm == mm) { + clear_user_asce(); + set_user_asce(mm); + } __tlb_flush_local(); } @@ -108,7 +110,7 @@ void crst_table_downgrade(struct mm_struct *mm, unsigned long limit) pgd_t *pgd; if (current->active_mm == mm) { - clear_user_asce(mm, 1); + clear_user_asce(); __tlb_flush_mm(mm); } while (mm->context.asce_limit > limit) { @@ -134,7 +136,7 @@ void crst_table_downgrade(struct mm_struct *mm, unsigned long limit) crst_table_free(mm, (unsigned long *) pgd); } if (current->active_mm == mm) - update_user_asce(mm, 1); + set_user_asce(mm); } #endif @@ -832,6 +834,7 @@ void gmap_do_ipte_notify(struct mm_struct *mm, pte_t *pte) } spin_unlock(&gmap_notifier_lock); } +EXPORT_SYMBOL_GPL(gmap_do_ipte_notify); static inline int page_table_with_pgste(struct page *page) { @@ -864,8 +867,7 @@ static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm, atomic_set(&page->_mapcount, 0); table = (unsigned long *) page_to_phys(page); clear_table(table, _PAGE_INVALID, PAGE_SIZE/2); - clear_table(table + PTRS_PER_PTE, PGSTE_HR_BIT | PGSTE_HC_BIT, - PAGE_SIZE/2); + clear_table(table + PTRS_PER_PTE, 0, PAGE_SIZE/2); return table; } @@ -883,8 +885,8 @@ static inline void page_table_free_pgste(unsigned long *table) __free_page(page); } -static inline unsigned long page_table_reset_pte(struct mm_struct *mm, - pmd_t *pmd, unsigned long addr, unsigned long end) +static inline unsigned long page_table_reset_pte(struct mm_struct *mm, pmd_t *pmd, + unsigned long addr, unsigned long end, bool init_skey) { pte_t *start_pte, *pte; spinlock_t *ptl; @@ -895,6 +897,22 @@ static inline unsigned long page_table_reset_pte(struct mm_struct *mm, do { pgste = pgste_get_lock(pte); pgste_val(pgste) &= ~_PGSTE_GPS_USAGE_MASK; + if (init_skey) { + unsigned long address; + + pgste_val(pgste) &= ~(PGSTE_ACC_BITS | PGSTE_FP_BIT | + PGSTE_GR_BIT | PGSTE_GC_BIT); + + /* skip invalid and not writable pages */ + if (pte_val(*pte) & _PAGE_INVALID || + !(pte_val(*pte) & _PAGE_WRITE)) { + pgste_set_unlock(pte, pgste); + continue; + } + + address = pte_val(*pte) & PAGE_MASK; + page_set_storage_key(address, PAGE_DEFAULT_KEY, 1); + } pgste_set_unlock(pte, pgste); } while (pte++, addr += PAGE_SIZE, addr != end); pte_unmap_unlock(start_pte, ptl); @@ -902,8 +920,8 @@ static inline unsigned long page_table_reset_pte(struct mm_struct *mm, return addr; } -static inline unsigned long page_table_reset_pmd(struct mm_struct *mm, - pud_t *pud, unsigned long addr, unsigned long end) +static inline unsigned long page_table_reset_pmd(struct mm_struct *mm, pud_t *pud, + unsigned long addr, unsigned long end, bool init_skey) { unsigned long next; pmd_t *pmd; @@ -913,14 +931,14 @@ static inline unsigned long page_table_reset_pmd(struct mm_struct *mm, next = pmd_addr_end(addr, end); if (pmd_none_or_clear_bad(pmd)) continue; - next = page_table_reset_pte(mm, pmd, addr, next); + next = page_table_reset_pte(mm, pmd, addr, next, init_skey); } while (pmd++, addr = next, addr != end); return addr; } -static inline unsigned long page_table_reset_pud(struct mm_struct *mm, - pgd_t *pgd, unsigned long addr, unsigned long end) +static inline unsigned long page_table_reset_pud(struct mm_struct *mm, pgd_t *pgd, + unsigned long addr, unsigned long end, bool init_skey) { unsigned long next; pud_t *pud; @@ -930,28 +948,33 @@ static inline unsigned long page_table_reset_pud(struct mm_struct *mm, next = pud_addr_end(addr, end); if (pud_none_or_clear_bad(pud)) continue; - next = page_table_reset_pmd(mm, pud, addr, next); + next = page_table_reset_pmd(mm, pud, addr, next, init_skey); } while (pud++, addr = next, addr != end); return addr; } -void page_table_reset_pgste(struct mm_struct *mm, - unsigned long start, unsigned long end) +void page_table_reset_pgste(struct mm_struct *mm, unsigned long start, + unsigned long end, bool init_skey) { unsigned long addr, next; pgd_t *pgd; + down_write(&mm->mmap_sem); + if (init_skey && mm_use_skey(mm)) + goto out_up; addr = start; - down_read(&mm->mmap_sem); pgd = pgd_offset(mm, addr); do { next = pgd_addr_end(addr, end); if (pgd_none_or_clear_bad(pgd)) continue; - next = page_table_reset_pud(mm, pgd, addr, next); + next = page_table_reset_pud(mm, pgd, addr, next, init_skey); } while (pgd++, addr = next, addr != end); - up_read(&mm->mmap_sem); + if (init_skey) + current->mm->context.use_skey = 1; +out_up: + up_write(&mm->mmap_sem); } EXPORT_SYMBOL(page_table_reset_pgste); @@ -989,7 +1012,7 @@ int set_guest_storage_key(struct mm_struct *mm, unsigned long addr, /* changing the guest storage key is considered a change of the page */ if ((pgste_val(new) ^ pgste_val(old)) & (PGSTE_ACC_BITS | PGSTE_FP_BIT | PGSTE_GR_BIT | PGSTE_GC_BIT)) - pgste_val(new) |= PGSTE_HC_BIT; + pgste_val(new) |= PGSTE_UC_BIT; pgste_set_unlock(ptep, new); pte_unmap_unlock(*ptep, ptl); @@ -1011,6 +1034,11 @@ static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm, return NULL; } +void page_table_reset_pgste(struct mm_struct *mm, unsigned long start, + unsigned long end, bool init_skey) +{ +} + static inline void page_table_free_pgste(unsigned long *table) { } @@ -1357,6 +1385,37 @@ int s390_enable_sie(void) } EXPORT_SYMBOL_GPL(s390_enable_sie); +/* + * Enable storage key handling from now on and initialize the storage + * keys with the default key. + */ +void s390_enable_skey(void) +{ + page_table_reset_pgste(current->mm, 0, TASK_SIZE, true); +} +EXPORT_SYMBOL_GPL(s390_enable_skey); + +/* + * Test and reset if a guest page is dirty + */ +bool gmap_test_and_clear_dirty(unsigned long address, struct gmap *gmap) +{ + pte_t *pte; + spinlock_t *ptl; + bool dirty = false; + + pte = get_locked_pte(gmap->mm, address, &ptl); + if (unlikely(!pte)) + return false; + + if (ptep_test_and_clear_user_dirty(gmap->mm, address, pte)) + dirty = true; + + spin_unlock(ptl); + return dirty; +} +EXPORT_SYMBOL_GPL(gmap_test_and_clear_dirty); + #ifdef CONFIG_TRANSPARENT_HUGEPAGE int pmdp_clear_flush_young(struct vm_area_struct *vma, unsigned long address, pmd_t *pmdp) diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c index 72b04de18283..fe9012a49aa5 100644 --- a/arch/s390/mm/vmem.c +++ b/arch/s390/mm/vmem.c @@ -10,6 +10,7 @@ #include <linux/list.h> #include <linux/hugetlb.h> #include <linux/slab.h> +#include <linux/memblock.h> #include <asm/pgalloc.h> #include <asm/pgtable.h> #include <asm/setup.h> @@ -66,7 +67,8 @@ static pte_t __ref *vmem_pte_alloc(unsigned long address) if (slab_is_available()) pte = (pte_t *) page_table_alloc(&init_mm, address); else - pte = alloc_bootmem(PTRS_PER_PTE * sizeof(pte_t)); + pte = alloc_bootmem_align(PTRS_PER_PTE * sizeof(pte_t), + PTRS_PER_PTE * sizeof(pte_t)); if (!pte) return NULL; clear_table((unsigned long *) pte, _PAGE_INVALID, @@ -371,16 +373,14 @@ out: void __init vmem_map_init(void) { unsigned long ro_start, ro_end; - unsigned long start, end; - int i; + struct memblock_region *reg; + phys_addr_t start, end; ro_start = PFN_ALIGN((unsigned long)&_stext); ro_end = (unsigned long)&_eshared & PAGE_MASK; - for (i = 0; i < MEMORY_CHUNKS; i++) { - if (!memory_chunk[i].size) - continue; - start = memory_chunk[i].addr; - end = memory_chunk[i].addr + memory_chunk[i].size; + for_each_memblock(memory, reg) { + start = reg->base; + end = reg->base + reg->size - 1; if (start >= ro_end || end <= ro_start) vmem_add_mem(start, end - start, 0); else if (start >= ro_start && end <= ro_end) @@ -400,23 +400,21 @@ void __init vmem_map_init(void) } /* - * Convert memory chunk array to a memory segment list so there is a single - * list that contains both r/w memory and shared memory segments. + * Convert memblock.memory to a memory segment list so there is a single + * list that contains all memory segments. */ static int __init vmem_convert_memory_chunk(void) { + struct memblock_region *reg; struct memory_segment *seg; - int i; mutex_lock(&vmem_mutex); - for (i = 0; i < MEMORY_CHUNKS; i++) { - if (!memory_chunk[i].size) - continue; + for_each_memblock(memory, reg) { seg = kzalloc(sizeof(*seg), GFP_KERNEL); if (!seg) panic("Out of memory...\n"); - seg->start = memory_chunk[i].addr; - seg->size = memory_chunk[i].size; + seg->start = reg->base; + seg->size = reg->size; insert_memory_segment(seg); } mutex_unlock(&vmem_mutex); diff --git a/arch/s390/oprofile/hwsampler.c b/arch/s390/oprofile/hwsampler.c index 276f2e26c761..e53c6f268807 100644 --- a/arch/s390/oprofile/hwsampler.c +++ b/arch/s390/oprofile/hwsampler.c @@ -209,13 +209,11 @@ static void init_all_cpu_buffers(void) } } -static int prepare_cpu_buffers(void) +static void prepare_cpu_buffers(void) { - int cpu; - int rc; struct hws_cpu_buffer *cb; + int cpu; - rc = 0; for_each_online_cpu(cpu) { cb = &per_cpu(sampler_cpu_buffer, cpu); atomic_set(&cb->ext_params, 0); @@ -230,8 +228,6 @@ static int prepare_cpu_buffers(void) cb->oom = 0; cb->stop_mode = 0; } - - return rc; } /* @@ -1107,9 +1103,7 @@ int hwsampler_start_all(unsigned long rate) if (rc) goto start_all_exit; - rc = prepare_cpu_buffers(); - if (rc) - goto start_all_exit; + prepare_cpu_buffers(); for_each_online_cpu(cpu) { rc = start_sampling(cpu); @@ -1156,7 +1150,7 @@ int hwsampler_stop_all(void) rc = 0; if (hws_state == HWS_INIT) { mutex_unlock(&hws_sem); - return rc; + return 0; } hws_state = HWS_STOPPING; mutex_unlock(&hws_sem); diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c index 1df1d29ac81d..9ddc51eeb8d6 100644 --- a/arch/s390/pci/pci.c +++ b/arch/s390/pci/pci.c @@ -401,11 +401,11 @@ static void zpci_irq_handler(struct airq_struct *airq) int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) { struct zpci_dev *zdev = get_zdev(pdev); - unsigned int hwirq, irq, msi_vecs; + unsigned int hwirq, msi_vecs; unsigned long aisb; struct msi_desc *msi; struct msi_msg msg; - int rc; + int rc, irq; if (type == PCI_CAP_ID_MSI && nvec > 1) return 1; @@ -433,7 +433,7 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) list_for_each_entry(msi, &pdev->msi_list, list) { rc = -EIO; irq = irq_alloc_desc(0); /* Alloc irq on node 0 */ - if (irq == NO_IRQ) + if (irq < 0) goto out_msi; rc = irq_set_msi_desc(irq, msi); if (rc) @@ -530,11 +530,6 @@ static void zpci_unmap_resources(struct zpci_dev *zdev) } } -int pcibios_add_platform_entries(struct pci_dev *pdev) -{ - return zpci_sysfs_add_device(&pdev->dev); -} - static int __init zpci_irq_init(void) { int rc; @@ -671,6 +666,7 @@ int pcibios_add_device(struct pci_dev *pdev) int i; zdev->pdev = pdev; + pdev->dev.groups = zpci_attr_groups; zpci_map_resources(zdev); for (i = 0; i < PCI_BAR_COUNT; i++) { diff --git a/arch/s390/pci/pci_clp.c b/arch/s390/pci/pci_clp.c index c747394029ee..96545d7659fd 100644 --- a/arch/s390/pci/pci_clp.c +++ b/arch/s390/pci/pci_clp.c @@ -114,6 +114,16 @@ static int clp_store_query_pci_fn(struct zpci_dev *zdev, zdev->end_dma = response->edma; zdev->pchid = response->pchid; zdev->pfgid = response->pfgid; + zdev->pft = response->pft; + zdev->vfn = response->vfn; + zdev->uid = response->uid; + + memcpy(zdev->pfip, response->pfip, sizeof(zdev->pfip)); + if (response->util_str_avail) { + memcpy(zdev->util_str, response->util_str, + sizeof(zdev->util_str)); + } + return 0; } diff --git a/arch/s390/pci/pci_event.c b/arch/s390/pci/pci_event.c index 01e251b1da0c..6d7f5a3016ca 100644 --- a/arch/s390/pci/pci_event.c +++ b/arch/s390/pci/pci_event.c @@ -76,7 +76,7 @@ static void __zpci_event_availability(struct zpci_ccdf_avail *ccdf) switch (ccdf->pec) { case 0x0301: /* Standby -> Configured */ - if (!zdev || zdev->state == ZPCI_FN_STATE_CONFIGURED) + if (!zdev || zdev->state != ZPCI_FN_STATE_STANDBY) break; zdev->state = ZPCI_FN_STATE_CONFIGURED; zdev->fh = ccdf->fh; @@ -86,7 +86,8 @@ static void __zpci_event_availability(struct zpci_ccdf_avail *ccdf) pci_rescan_bus(zdev->bus); break; case 0x0302: /* Reserved -> Standby */ - clp_add_pci_device(ccdf->fid, ccdf->fh, 0); + if (!zdev) + clp_add_pci_device(ccdf->fid, ccdf->fh, 0); break; case 0x0303: /* Deconfiguration requested */ if (pdev) diff --git a/arch/s390/pci/pci_sysfs.c b/arch/s390/pci/pci_sysfs.c index ab4a91393005..9190214b8702 100644 --- a/arch/s390/pci/pci_sysfs.c +++ b/arch/s390/pci/pci_sysfs.c @@ -12,43 +12,29 @@ #include <linux/stat.h> #include <linux/pci.h> -static ssize_t show_fid(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct zpci_dev *zdev = get_zdev(to_pci_dev(dev)); - - return sprintf(buf, "0x%08x\n", zdev->fid); -} -static DEVICE_ATTR(function_id, S_IRUGO, show_fid, NULL); - -static ssize_t show_fh(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct zpci_dev *zdev = get_zdev(to_pci_dev(dev)); - - return sprintf(buf, "0x%08x\n", zdev->fh); -} -static DEVICE_ATTR(function_handle, S_IRUGO, show_fh, NULL); - -static ssize_t show_pchid(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct zpci_dev *zdev = get_zdev(to_pci_dev(dev)); - - return sprintf(buf, "0x%04x\n", zdev->pchid); -} -static DEVICE_ATTR(pchid, S_IRUGO, show_pchid, NULL); - -static ssize_t show_pfgid(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct zpci_dev *zdev = get_zdev(to_pci_dev(dev)); - - return sprintf(buf, "0x%02x\n", zdev->pfgid); -} -static DEVICE_ATTR(pfgid, S_IRUGO, show_pfgid, NULL); - -static ssize_t store_recover(struct device *dev, struct device_attribute *attr, +#define zpci_attr(name, fmt, member) \ +static ssize_t name##_show(struct device *dev, \ + struct device_attribute *attr, char *buf) \ +{ \ + struct zpci_dev *zdev = get_zdev(to_pci_dev(dev)); \ + \ + return sprintf(buf, fmt, zdev->member); \ +} \ +static DEVICE_ATTR_RO(name) + +zpci_attr(function_id, "0x%08x\n", fid); +zpci_attr(function_handle, "0x%08x\n", fh); +zpci_attr(pchid, "0x%04x\n", pchid); +zpci_attr(pfgid, "0x%02x\n", pfgid); +zpci_attr(vfn, "0x%04x\n", vfn); +zpci_attr(pft, "0x%02x\n", pft); +zpci_attr(uid, "0x%x\n", uid); +zpci_attr(segment0, "0x%02x\n", pfip[0]); +zpci_attr(segment1, "0x%02x\n", pfip[1]); +zpci_attr(segment2, "0x%02x\n", pfip[2]); +zpci_attr(segment3, "0x%02x\n", pfip[3]); + +static ssize_t recover_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct pci_dev *pdev = to_pci_dev(dev); @@ -70,38 +56,55 @@ static ssize_t store_recover(struct device *dev, struct device_attribute *attr, pci_rescan_bus(zdev->bus); return count; } -static DEVICE_ATTR(recover, S_IWUSR, NULL, store_recover); +static DEVICE_ATTR_WO(recover); -static struct device_attribute *zpci_dev_attrs[] = { - &dev_attr_function_id, - &dev_attr_function_handle, - &dev_attr_pchid, - &dev_attr_pfgid, - &dev_attr_recover, - NULL, -}; - -int zpci_sysfs_add_device(struct device *dev) +static ssize_t util_string_read(struct file *filp, struct kobject *kobj, + struct bin_attribute *attr, char *buf, + loff_t off, size_t count) { - int i, rc = 0; - - for (i = 0; zpci_dev_attrs[i]; i++) { - rc = device_create_file(dev, zpci_dev_attrs[i]); - if (rc) - goto error; - } - return 0; + struct device *dev = kobj_to_dev(kobj); + struct pci_dev *pdev = to_pci_dev(dev); + struct zpci_dev *zdev = get_zdev(pdev); -error: - while (--i >= 0) - device_remove_file(dev, zpci_dev_attrs[i]); - return rc; + return memory_read_from_buffer(buf, count, &off, zdev->util_str, + sizeof(zdev->util_str)); } +static BIN_ATTR_RO(util_string, CLP_UTIL_STR_LEN); +static struct bin_attribute *zpci_bin_attrs[] = { + &bin_attr_util_string, + NULL, +}; -void zpci_sysfs_remove_device(struct device *dev) -{ - int i; +static struct attribute *zpci_dev_attrs[] = { + &dev_attr_function_id.attr, + &dev_attr_function_handle.attr, + &dev_attr_pchid.attr, + &dev_attr_pfgid.attr, + &dev_attr_pft.attr, + &dev_attr_vfn.attr, + &dev_attr_uid.attr, + &dev_attr_recover.attr, + NULL, +}; +static struct attribute_group zpci_attr_group = { + .attrs = zpci_dev_attrs, + .bin_attrs = zpci_bin_attrs, +}; - for (i = 0; zpci_dev_attrs[i]; i++) - device_remove_file(dev, zpci_dev_attrs[i]); -} +static struct attribute *pfip_attrs[] = { + &dev_attr_segment0.attr, + &dev_attr_segment1.attr, + &dev_attr_segment2.attr, + &dev_attr_segment3.attr, + NULL, +}; +static struct attribute_group pfip_attr_group = { + .name = "pfip", + .attrs = pfip_attrs, +}; + +const struct attribute_group *zpci_attr_groups[] = { + &zpci_attr_group, + &pfip_attr_group, + NULL, +}; |