From 8076428f0c9f24d90270204fd39ccb11b83db71d Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 11 Sep 2017 11:24:22 +0200 Subject: s390: convert release_thread() into a static inline function release_thread() is an empty function that gets called on every task exit. Move the function to a header file and force inlining of it, so that the compiler can optimize it away instead of generating a pointless function call. Acked-by: Christian Borntraeger Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/processor.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h index c25d57e0aad3..4cc9bf53074b 100644 --- a/arch/s390/include/asm/processor.h +++ b/arch/s390/include/asm/processor.h @@ -218,7 +218,7 @@ void show_registers(struct pt_regs *regs); void show_cacheinfo(struct seq_file *m); /* Free all resources held by a thread. */ -extern void release_thread(struct task_struct *); +static inline void release_thread(struct task_struct *tsk) { } /* Free guarded storage control block for current */ void exit_thread_gs(void); -- cgit From 8d9047f8b967ce6181fd824ae922978e1b055cc0 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 11 Sep 2017 11:24:22 +0200 Subject: s390/runtime instrumentation: simplify task exit handling Free data structures required for runtime instrumentation from arch_release_task_struct(). This allows to simplify the code a bit, and also makes the semantics a bit easier: arch_release_task_struct() is never called from the task that is being removed. In addition this allows to get rid of exit_thread() in a later patch. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/runtime_instr.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/runtime_instr.h b/arch/s390/include/asm/runtime_instr.h index 402ad6df4897..c54a9310d814 100644 --- a/arch/s390/include/asm/runtime_instr.h +++ b/arch/s390/include/asm/runtime_instr.h @@ -85,6 +85,8 @@ static inline void restore_ri_cb(struct runtime_instr_cb *cb_next, load_runtime_instr_cb(&runtime_instr_empty_cb); } -void exit_thread_runtime_instr(void); +struct task_struct; + +void runtime_instr_release(struct task_struct *tsk); #endif /* _RUNTIME_INSTR_H */ -- cgit From 7b83c6297d2fc7350997e86188df84c27fd59530 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 11 Sep 2017 11:24:23 +0200 Subject: s390/guarded storage: simplify task exit handling Free data structures required for guarded storage from arch_release_task_struct(). This allows to simplify the code a bit, and also makes the semantics a bit easier: arch_release_task_struct() is never called from the task that is being removed. In addition this allows to get rid of exit_thread() in a later patch. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/processor.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h index 4cc9bf53074b..58eaaccd3cf0 100644 --- a/arch/s390/include/asm/processor.h +++ b/arch/s390/include/asm/processor.h @@ -220,8 +220,8 @@ void show_cacheinfo(struct seq_file *m); /* Free all resources held by a thread. */ static inline void release_thread(struct task_struct *tsk) { } -/* Free guarded storage control block for current */ -void exit_thread_gs(void); +/* Free guarded storage control block */ +void guarded_storage_release(struct task_struct *tsk); unsigned long get_wchan(struct task_struct *p); #define task_pt_regs(tsk) ((struct pt_regs *) \ -- cgit From 79962038dffab28094782406d8e27b095a26ce51 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Tue, 12 Sep 2017 13:49:57 +0200 Subject: s390: add support for FORTIFY_SOURCE This is the quite trivial backend for s390 which is required to enable FORTIFY_SOURCE support. See commit 6974f0c4555e ("include/linux/string.h: add the option of fortified string.h functions") for more details. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/string.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/string.h b/arch/s390/include/asm/string.h index e5f5c7074f2c..8fb43319693d 100644 --- a/arch/s390/include/asm/string.h +++ b/arch/s390/include/asm/string.h @@ -49,7 +49,7 @@ extern char *strstr(const char *, const char *); #undef __HAVE_ARCH_STRSEP #undef __HAVE_ARCH_STRSPN -#if !defined(IN_ARCH_STRING_C) +#if !defined(IN_ARCH_STRING_C) && (!defined(CONFIG_FORTIFY_SOURCE) || defined(__NO_FORTIFY)) static inline void *memchr(const void * s, int c, size_t n) { -- cgit From bb59c2da3fb4ecc83258ca6b2ecde70d80c33465 Mon Sep 17 00:00:00 2001 From: Alice Frosi Date: Thu, 14 Sep 2017 12:35:45 +0200 Subject: s390/runtime_instrumentation: clean up struct runtime_instr_cb Update runtime_instr_cb structure to be consistent with the runtime instrumentation documentation. Signed-off-by: Alice Frosi Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/runtime_instr.h | 82 +++++++++++++++++------------------ 1 file changed, 41 insertions(+), 41 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/runtime_instr.h b/arch/s390/include/asm/runtime_instr.h index c54a9310d814..f0ce4972d87c 100644 --- a/arch/s390/include/asm/runtime_instr.h +++ b/arch/s390/include/asm/runtime_instr.h @@ -5,55 +5,55 @@ #define S390_RUNTIME_INSTR_STOP 0x2 struct runtime_instr_cb { - __u64 buf_current; - __u64 buf_origin; - __u64 buf_limit; + __u64 rca; + __u64 roa; + __u64 rla; - __u32 valid : 1; - __u32 pstate : 1; - __u32 pstate_set_buf : 1; - __u32 home_space : 1; - __u32 altered : 1; - __u32 : 3; - __u32 pstate_sample : 1; - __u32 sstate_sample : 1; - __u32 pstate_collect : 1; - __u32 sstate_collect : 1; - __u32 : 1; - __u32 halted_int : 1; - __u32 int_requested : 1; - __u32 buffer_full_int : 1; + __u32 v : 1; + __u32 s : 1; + __u32 k : 1; + __u32 h : 1; + __u32 a : 1; + __u32 reserved1 : 3; + __u32 ps : 1; + __u32 qs : 1; + __u32 pc : 1; + __u32 qc : 1; + __u32 reserved2 : 1; + __u32 g : 1; + __u32 u : 1; + __u32 l : 1; __u32 key : 4; - __u32 : 9; + __u32 reserved3 : 8; + __u32 t : 1; __u32 rgs : 3; - __u32 mode : 4; - __u32 next : 1; + __u32 m : 4; + __u32 n : 1; __u32 mae : 1; - __u32 : 2; - __u32 call_type_br : 1; - __u32 return_type_br : 1; - __u32 other_type_br : 1; - __u32 bc_other_type : 1; - __u32 emit : 1; - __u32 tx_abort : 1; - __u32 : 2; - __u32 bp_xn : 1; - __u32 bp_xt : 1; - __u32 bp_ti : 1; - __u32 bp_ni : 1; - __u32 suppr_y : 1; - __u32 suppr_z : 1; + __u32 reserved4 : 2; + __u32 c : 1; + __u32 r : 1; + __u32 b : 1; + __u32 j : 1; + __u32 e : 1; + __u32 x : 1; + __u32 reserved5 : 2; + __u32 bpxn : 1; + __u32 bpxt : 1; + __u32 bpti : 1; + __u32 bpni : 1; + __u32 reserved6 : 2; - __u32 dc_miss_extra : 1; - __u32 lat_lev_ignore : 1; - __u32 ic_lat_lev : 4; - __u32 dc_lat_lev : 4; + __u32 d : 1; + __u32 f : 1; + __u32 ic : 4; + __u32 dc : 4; - __u64 reserved1; - __u64 scaling_factor; + __u64 reserved7; + __u64 sf; __u64 rsic; - __u64 reserved2; + __u64 reserved8; } __packed __aligned(8); extern struct runtime_instr_cb runtime_instr_empty_cb; -- cgit From 1887aa07b6765d345dd79f26017aa2d15d49d7af Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 22 Sep 2017 14:17:41 +0200 Subject: s390/topology: add detection of dedicated vs shared CPUs The topology information returned by STSI 15.x.x contains a flag if the CPUs of a topology-list are dedicated or shared. Make this information available if the machine provides topology information. Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/processor.h | 2 ++ arch/s390/include/asm/sysinfo.h | 3 ++- arch/s390/include/asm/topology.h | 2 ++ 3 files changed, 6 insertions(+), 1 deletion(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h index 58eaaccd3cf0..4c5fde30d741 100644 --- a/arch/s390/include/asm/processor.h +++ b/arch/s390/include/asm/processor.h @@ -21,6 +21,7 @@ #define CIF_IGNORE_IRQ 5 /* ignore interrupt (for udelay) */ #define CIF_ENABLED_WAIT 6 /* in enabled wait state */ #define CIF_MCCK_GUEST 7 /* machine check happening in guest */ +#define CIF_DEDICATED_CPU 8 /* this CPU is dedicated */ #define _CIF_MCCK_PENDING _BITUL(CIF_MCCK_PENDING) #define _CIF_ASCE_PRIMARY _BITUL(CIF_ASCE_PRIMARY) @@ -30,6 +31,7 @@ #define _CIF_IGNORE_IRQ _BITUL(CIF_IGNORE_IRQ) #define _CIF_ENABLED_WAIT _BITUL(CIF_ENABLED_WAIT) #define _CIF_MCCK_GUEST _BITUL(CIF_MCCK_GUEST) +#define _CIF_DEDICATED_CPU _BITUL(CIF_DEDICATED_CPU) #ifndef __ASSEMBLY__ diff --git a/arch/s390/include/asm/sysinfo.h b/arch/s390/include/asm/sysinfo.h index 2b498e58b914..0f09135f0ae4 100644 --- a/arch/s390/include/asm/sysinfo.h +++ b/arch/s390/include/asm/sysinfo.h @@ -156,7 +156,8 @@ static inline unsigned char topology_mnest_limit(void) struct topology_core { unsigned char nl; unsigned char reserved0[3]; - unsigned char :6; + unsigned char :5; + unsigned char d:1; unsigned char pp:2; unsigned char reserved1; unsigned short origin; diff --git a/arch/s390/include/asm/topology.h b/arch/s390/include/asm/topology.h index 5222da162b69..5108176435c1 100644 --- a/arch/s390/include/asm/topology.h +++ b/arch/s390/include/asm/topology.h @@ -16,6 +16,7 @@ struct cpu_topology_s390 { unsigned short book_id; unsigned short drawer_id; unsigned short node_id; + unsigned short dedicated : 1; cpumask_t thread_mask; cpumask_t core_mask; cpumask_t book_mask; @@ -34,6 +35,7 @@ extern cpumask_t cpus_with_topology; #define topology_book_cpumask(cpu) (&cpu_topology[cpu].book_mask) #define topology_drawer_id(cpu) (cpu_topology[cpu].drawer_id) #define topology_drawer_cpumask(cpu) (&cpu_topology[cpu].drawer_mask) +#define topology_cpu_dedicated(cpu) (cpu_topology[cpu].dedicated) #define mc_capable() 1 -- cgit From 8153380379ecc8381f6d55f6497de31a36c75aa5 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Sun, 4 Dec 2016 14:36:04 +0100 Subject: s390/spinlock: use the cpu number +1 as spinlock value The queued spinlock code will come out simpler if the encoding of the CPU that holds the spinlock is (cpu+1) instead of (~cpu). Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/spinlock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h index 8182b521c42f..6727cc30d59b 100644 --- a/arch/s390/include/asm/spinlock.h +++ b/arch/s390/include/asm/spinlock.h @@ -48,7 +48,7 @@ static inline void arch_spin_relax(arch_spinlock_t *lock) static inline u32 arch_spin_lockval(int cpu) { - return ~cpu; + return cpu + 1; } static inline int arch_spin_value_unlocked(arch_spinlock_t lock) -- cgit From b96f7d881ad94203e997cd2aa7112d4a06d121ef Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 24 Mar 2017 17:25:02 +0100 Subject: s390/spinlock: introduce spinlock wait queueing The queued spinlock code for s390 follows the principles of the common code qspinlock implementation but with a few notable differences. The format of the spinlock_t locking word differs, s390 needs to store the logical CPU number of the lock holder in the spinlock_t to be able to use the diagnose 9c directed yield hypervisor call. The inline code sequences for spin_lock and spin_unlock are nice and short. The inline portion of a spin_lock now typically looks like this: lhi %r0,0 # 0 indicates an empty lock l %r1,0x3a0 # CPU number + 1 from lowcore cs %r0,%r1, # lock operation jnz call_wait # on failure call wait function locked: ... call_wait: la %r2, brasl %r14,arch_spin_lock_wait j locked A spin_unlock is as simple as before: lhi %r0,0 sth %r0,2(%r2) # unlock operation After a CPU has queued itself it may not enable interrupts again for the arch_spin_lock_flags() variant. The arch_spin_lock_wait_flags wait function is removed. To improve performance the code implements opportunistic lock stealing. If the wait function finds a spinlock_t that indicates that the lock is free but there are queued waiters, the CPU may steal the lock up to three times without queueing itself. The lock stealing update the steal counter in the lock word to prevent more than 3 steals. The counter is reset at the time the CPU next in the queue successfully takes the lock. While the queued spinlocks improve performance in a system with dedicated CPUs, in a virtualized environment with continuously overcommitted CPUs the queued spinlocks can have a negative effect on performance. This is due to the fact that a queued CPU that is preempted by the hypervisor will block the queue at some point even without holding the lock. With the classic spinlock it does not matter if a CPU is preempted that waits for the lock. Therefore use the queued spinlock code only if the system runs with dedicated CPUs and fall back to classic spinlocks when running with shared CPUs. Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/lowcore.h | 5 +++-- arch/s390/include/asm/spinlock.h | 18 +++++++----------- 2 files changed, 10 insertions(+), 13 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h index a6870ea6ea8b..62943af36ac6 100644 --- a/arch/s390/include/asm/lowcore.h +++ b/arch/s390/include/asm/lowcore.h @@ -133,8 +133,9 @@ struct lowcore { __u8 pad_0x03b4[0x03b8-0x03b4]; /* 0x03b4 */ __u64 gmap; /* 0x03b8 */ __u32 spinlock_lockval; /* 0x03c0 */ - __u32 fpu_flags; /* 0x03c4 */ - __u8 pad_0x03c8[0x0400-0x03c8]; /* 0x03c8 */ + __u32 spinlock_index; /* 0x03c4 */ + __u32 fpu_flags; /* 0x03c8 */ + __u8 pad_0x03cc[0x0400-0x03cc]; /* 0x03cc */ /* Per cpu primary space access list */ __u32 paste[16]; /* 0x0400 */ diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h index 6727cc30d59b..2da4a6d13f54 100644 --- a/arch/s390/include/asm/spinlock.h +++ b/arch/s390/include/asm/spinlock.h @@ -36,15 +36,11 @@ bool arch_vcpu_is_preempted(int cpu); */ void arch_lock_relax(int cpu); +void arch_spin_relax(arch_spinlock_t *lock); void arch_spin_lock_wait(arch_spinlock_t *); int arch_spin_trylock_retry(arch_spinlock_t *); -void arch_spin_lock_wait_flags(arch_spinlock_t *, unsigned long flags); - -static inline void arch_spin_relax(arch_spinlock_t *lock) -{ - arch_lock_relax(lock->lock); -} +void arch_spin_lock_setup(int cpu); static inline u32 arch_spin_lockval(int cpu) { @@ -64,8 +60,7 @@ static inline int arch_spin_is_locked(arch_spinlock_t *lp) static inline int arch_spin_trylock_once(arch_spinlock_t *lp) { barrier(); - return likely(arch_spin_value_unlocked(*lp) && - __atomic_cmpxchg_bool(&lp->lock, 0, SPINLOCK_LOCKVAL)); + return likely(__atomic_cmpxchg_bool(&lp->lock, 0, SPINLOCK_LOCKVAL)); } static inline void arch_spin_lock(arch_spinlock_t *lp) @@ -78,7 +73,7 @@ static inline void arch_spin_lock_flags(arch_spinlock_t *lp, unsigned long flags) { if (!arch_spin_trylock_once(lp)) - arch_spin_lock_wait_flags(lp, flags); + arch_spin_lock_wait(lp); } static inline int arch_spin_trylock(arch_spinlock_t *lp) @@ -95,8 +90,9 @@ static inline void arch_spin_unlock(arch_spinlock_t *lp) #ifdef CONFIG_HAVE_MARCH_ZEC12_FEATURES " .long 0xb2fa0070\n" /* NIAI 7 */ #endif - " st %1,%0\n" - : "=Q" (lp->lock) : "d" (0) : "cc", "memory"); + " sth %1,%0\n" + : "=Q" (((unsigned short *) &lp->lock)[1]) + : "d" (0) : "cc", "memory"); } /* -- cgit From eb3b7b848fb3dd00f7a57d633d4ae4d194aa7865 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 24 Mar 2017 17:32:23 +0100 Subject: s390/rwlock: introduce rwlock wait queueing Like the common queued rwlock code the s390 implementation uses the queued spinlock code on a spinlock_t embedded in the rwlock_t to achieve the queueing. The encoding of the rwlock_t differs though, the counter field in the rwlock_t is split into two parts. The upper two bytes hold the write bit and the write wait counter, the lower two bytes hold the read counter. The arch_read_lock operation works exactly like the common qrwlock but the enqueue operation for a writer follows a diffent logic. After the failed inline try to get the rwlock in write, the writer first increases the write wait counter, acquires the wait spin_lock for the queueing, and then loops until there are no readers and the write bit is zero. Without the write wait counter a CPU that just released the rwlock could immediately reacquire the lock in the inline code, bypassing all outstanding read and write waiters. For s390 this would cause massive imbalances in favour of writers in case of a contended rwlock. Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/atomic_ops.h | 32 +++++--- arch/s390/include/asm/spinlock.h | 142 +++++---------------------------- arch/s390/include/asm/spinlock_types.h | 4 +- 3 files changed, 43 insertions(+), 135 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/atomic_ops.h b/arch/s390/include/asm/atomic_ops.h index ba6d29412344..daec181fa1f7 100644 --- a/arch/s390/include/asm/atomic_ops.h +++ b/arch/s390/include/asm/atomic_ops.h @@ -39,19 +39,24 @@ __ATOMIC_OPS(__atomic64_xor, long, "laxg") #undef __ATOMIC_OPS #undef __ATOMIC_OP -static inline void __atomic_add_const(int val, int *ptr) -{ - asm volatile( - " asi %[ptr],%[val]\n" - : [ptr] "+Q" (*ptr) : [val] "i" (val) : "cc"); +#define __ATOMIC_CONST_OP(op_name, op_type, op_string, op_barrier) \ +static inline void op_name(op_type val, op_type *ptr) \ +{ \ + asm volatile( \ + op_string " %[ptr],%[val]\n" \ + op_barrier \ + : [ptr] "+Q" (*ptr) : [val] "i" (val) : "cc", "memory");\ } -static inline void __atomic64_add_const(long val, long *ptr) -{ - asm volatile( - " agsi %[ptr],%[val]\n" - : [ptr] "+Q" (*ptr) : [val] "i" (val) : "cc"); -} +#define __ATOMIC_CONST_OPS(op_name, op_type, op_string) \ + __ATOMIC_CONST_OP(op_name, op_type, op_string, "\n") \ + __ATOMIC_CONST_OP(op_name##_barrier, op_type, op_string, "bcr 14,0\n") + +__ATOMIC_CONST_OPS(__atomic_add_const, int, "asi") +__ATOMIC_CONST_OPS(__atomic64_add_const, long, "agsi") + +#undef __ATOMIC_CONST_OPS +#undef __ATOMIC_CONST_OP #else /* CONFIG_HAVE_MARCH_Z196_FEATURES */ @@ -107,6 +112,11 @@ __ATOMIC64_OPS(__atomic64_xor, "xgr") #undef __ATOMIC64_OPS +#define __atomic_add_const(val, ptr) __atomic_add(val, ptr) +#define __atomic_add_const_barrier(val, ptr) __atomic_add(val, ptr) +#define __atomic64_add_const(val, ptr) __atomic64_add(val, ptr) +#define __atomic64_add_const_barrier(val, ptr) __atomic64_add(val, ptr) + #endif /* CONFIG_HAVE_MARCH_Z196_FEATURES */ static inline int __atomic_cmpxchg(int *ptr, int old, int new) diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h index 2da4a6d13f54..09e783d83d5d 100644 --- a/arch/s390/include/asm/spinlock.h +++ b/arch/s390/include/asm/spinlock.h @@ -35,7 +35,6 @@ bool arch_vcpu_is_preempted(int cpu); * (the type definitions are in asm/spinlock_types.h) */ -void arch_lock_relax(int cpu); void arch_spin_relax(arch_spinlock_t *lock); void arch_spin_lock_wait(arch_spinlock_t *); @@ -110,164 +109,63 @@ static inline void arch_spin_unlock(arch_spinlock_t *lp) * read_can_lock - would read_trylock() succeed? * @lock: the rwlock in question. */ -#define arch_read_can_lock(x) ((int)(x)->lock >= 0) +#define arch_read_can_lock(x) (((x)->cnts & 0xffff0000) == 0) /** * write_can_lock - would write_trylock() succeed? * @lock: the rwlock in question. */ -#define arch_write_can_lock(x) ((x)->lock == 0) - -extern int _raw_read_trylock_retry(arch_rwlock_t *lp); -extern int _raw_write_trylock_retry(arch_rwlock_t *lp); +#define arch_write_can_lock(x) ((x)->cnts == 0) #define arch_read_lock_flags(lock, flags) arch_read_lock(lock) #define arch_write_lock_flags(lock, flags) arch_write_lock(lock) +#define arch_read_relax(rw) barrier() +#define arch_write_relax(rw) barrier() -static inline int arch_read_trylock_once(arch_rwlock_t *rw) -{ - int old = ACCESS_ONCE(rw->lock); - return likely(old >= 0 && - __atomic_cmpxchg_bool(&rw->lock, old, old + 1)); -} - -static inline int arch_write_trylock_once(arch_rwlock_t *rw) -{ - int old = ACCESS_ONCE(rw->lock); - return likely(old == 0 && - __atomic_cmpxchg_bool(&rw->lock, 0, 0x80000000)); -} - -#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES - -#define __RAW_OP_OR "lao" -#define __RAW_OP_AND "lan" -#define __RAW_OP_ADD "laa" - -#define __RAW_LOCK(ptr, op_val, op_string) \ -({ \ - int old_val; \ - \ - typecheck(int *, ptr); \ - asm volatile( \ - op_string " %0,%2,%1\n" \ - "bcr 14,0\n" \ - : "=d" (old_val), "+Q" (*ptr) \ - : "d" (op_val) \ - : "cc", "memory"); \ - old_val; \ -}) - -#define __RAW_UNLOCK(ptr, op_val, op_string) \ -({ \ - int old_val; \ - \ - typecheck(int *, ptr); \ - asm volatile( \ - op_string " %0,%2,%1\n" \ - : "=d" (old_val), "+Q" (*ptr) \ - : "d" (op_val) \ - : "cc", "memory"); \ - old_val; \ -}) - -extern void _raw_read_lock_wait(arch_rwlock_t *lp); -extern void _raw_write_lock_wait(arch_rwlock_t *lp, int prev); +void arch_read_lock_wait(arch_rwlock_t *lp); +void arch_write_lock_wait(arch_rwlock_t *lp); static inline void arch_read_lock(arch_rwlock_t *rw) { int old; - old = __RAW_LOCK(&rw->lock, 1, __RAW_OP_ADD); - if (old < 0) - _raw_read_lock_wait(rw); + old = __atomic_add(1, &rw->cnts); + if (old & 0xffff0000) + arch_read_lock_wait(rw); } static inline void arch_read_unlock(arch_rwlock_t *rw) { - __RAW_UNLOCK(&rw->lock, -1, __RAW_OP_ADD); + __atomic_add_const_barrier(-1, &rw->cnts); } static inline void arch_write_lock(arch_rwlock_t *rw) { - int old; - - old = __RAW_LOCK(&rw->lock, 0x80000000, __RAW_OP_OR); - if (old != 0) - _raw_write_lock_wait(rw, old); - rw->owner = SPINLOCK_LOCKVAL; + if (!__atomic_cmpxchg_bool(&rw->cnts, 0, 0x30000)) + arch_write_lock_wait(rw); } static inline void arch_write_unlock(arch_rwlock_t *rw) { - rw->owner = 0; - __RAW_UNLOCK(&rw->lock, 0x7fffffff, __RAW_OP_AND); + __atomic_add_barrier(-0x30000, &rw->cnts); } -#else /* CONFIG_HAVE_MARCH_Z196_FEATURES */ - -extern void _raw_read_lock_wait(arch_rwlock_t *lp); -extern void _raw_write_lock_wait(arch_rwlock_t *lp); -static inline void arch_read_lock(arch_rwlock_t *rw) -{ - if (!arch_read_trylock_once(rw)) - _raw_read_lock_wait(rw); -} - -static inline void arch_read_unlock(arch_rwlock_t *rw) +static inline int arch_read_trylock(arch_rwlock_t *rw) { int old; - do { - old = ACCESS_ONCE(rw->lock); - } while (!__atomic_cmpxchg_bool(&rw->lock, old, old - 1)); -} - -static inline void arch_write_lock(arch_rwlock_t *rw) -{ - if (!arch_write_trylock_once(rw)) - _raw_write_lock_wait(rw); - rw->owner = SPINLOCK_LOCKVAL; -} - -static inline void arch_write_unlock(arch_rwlock_t *rw) -{ - typecheck(int, rw->lock); - - rw->owner = 0; - asm volatile( - "st %1,%0\n" - : "+Q" (rw->lock) - : "d" (0) - : "cc", "memory"); -} - -#endif /* CONFIG_HAVE_MARCH_Z196_FEATURES */ - -static inline int arch_read_trylock(arch_rwlock_t *rw) -{ - if (!arch_read_trylock_once(rw)) - return _raw_read_trylock_retry(rw); - return 1; + old = READ_ONCE(rw->cnts); + return (!(old & 0xffff0000) && + __atomic_cmpxchg_bool(&rw->cnts, old, old + 1)); } static inline int arch_write_trylock(arch_rwlock_t *rw) { - if (!arch_write_trylock_once(rw) && !_raw_write_trylock_retry(rw)) - return 0; - rw->owner = SPINLOCK_LOCKVAL; - return 1; -} - -static inline void arch_read_relax(arch_rwlock_t *rw) -{ - arch_lock_relax(rw->owner); -} + int old; -static inline void arch_write_relax(arch_rwlock_t *rw) -{ - arch_lock_relax(rw->owner); + old = READ_ONCE(rw->cnts); + return !old && __atomic_cmpxchg_bool(&rw->cnts, 0, 0x30000); } #endif /* __ASM_SPINLOCK_H */ diff --git a/arch/s390/include/asm/spinlock_types.h b/arch/s390/include/asm/spinlock_types.h index fe755eec275f..271b4da94fd0 100644 --- a/arch/s390/include/asm/spinlock_types.h +++ b/arch/s390/include/asm/spinlock_types.h @@ -12,8 +12,8 @@ typedef struct { #define __ARCH_SPIN_LOCK_UNLOCKED { .lock = 0, } typedef struct { - int lock; - int owner; + int cnts; + arch_spinlock_t wait; } arch_rwlock_t; #define __ARCH_RW_LOCK_UNLOCKED { 0 } -- cgit From eecd49c4624a2c35e74f6f4e6352edf7eba545ca Mon Sep 17 00:00:00 2001 From: Patrick Steuer Date: Mon, 18 Sep 2017 12:48:08 +0200 Subject: s390/crypto: add inline assembly for KMA instruction to cpacf.h Signed-off-by: Patrick Steuer Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/cpacf.h | 52 ++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 51 insertions(+), 1 deletion(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/cpacf.h b/arch/s390/include/asm/cpacf.h index e06f2556b316..cbf11aa3e6e9 100644 --- a/arch/s390/include/asm/cpacf.h +++ b/arch/s390/include/asm/cpacf.h @@ -1,7 +1,7 @@ /* * CP Assist for Cryptographic Functions (CPACF) * - * Copyright IBM Corp. 2003, 2016 + * Copyright IBM Corp. 2003, 2017 * Author(s): Thomas Spatzier * Jan Glauber * Harald Freudenberger (freude@de.ibm.com) @@ -133,6 +133,22 @@ #define CPACF_PRNO_TRNG_Q_R2C_RATIO 0x70 #define CPACF_PRNO_TRNG 0x72 +/* + * Function codes for the KMA (CIPHER MESSAGE WITH AUTHENTICATION) + * instruction + */ +#define CPACF_KMA_QUERY 0x00 +#define CPACF_KMA_GCM_AES_128 0x12 +#define CPACF_KMA_GCM_AES_192 0x13 +#define CPACF_KMA_GCM_AES_256 0x14 + +/* + * Flags for the KMA (CIPHER MESSAGE WITH AUTHENTICATION) instruction + */ +#define CPACF_KMA_LPC 0x100 /* Last-Plaintext/Ciphertext */ +#define CPACF_KMA_LAAD 0x200 /* Last-AAD */ +#define CPACF_KMA_HS 0x400 /* Hash-subkey Supplied */ + typedef struct { unsigned char bytes[16]; } cpacf_mask_t; /** @@ -178,6 +194,8 @@ static inline int __cpacf_check_opcode(unsigned int opcode) return test_facility(77); /* check for MSA4 */ case CPACF_PRNO: return test_facility(57); /* check for MSA5 */ + case CPACF_KMA: + return test_facility(146); /* check for MSA8 */ default: BUG(); } @@ -469,4 +487,36 @@ static inline void cpacf_pckmo(long func, void *param) : "cc", "memory"); } +/** + * cpacf_kma() - executes the KMA (CIPHER MESSAGE WITH AUTHENTICATION) + * instruction + * @func: the function code passed to KMA; see CPACF_KMA_xxx defines + * @param: address of parameter block; see POP for details on each func + * @dest: address of destination memory area + * @src: address of source memory area + * @src_len: length of src operand in bytes + * @aad: address of additional authenticated data memory area + * @aad_len: length of aad operand in bytes + */ +static inline void cpacf_kma(unsigned long func, void *param, u8 *dest, + const u8 *src, unsigned long src_len, + const u8 *aad, unsigned long aad_len) +{ + register unsigned long r0 asm("0") = (unsigned long) func; + register unsigned long r1 asm("1") = (unsigned long) param; + register unsigned long r2 asm("2") = (unsigned long) src; + register unsigned long r3 asm("3") = (unsigned long) src_len; + register unsigned long r4 asm("4") = (unsigned long) aad; + register unsigned long r5 asm("5") = (unsigned long) aad_len; + register unsigned long r6 asm("6") = (unsigned long) dest; + + asm volatile( + "0: .insn rrf,%[opc] << 16,%[dst],%[src],%[aad],0\n" + " brc 1,0b\n" /* handle partial completion */ + : [dst] "+a" (r6), [src] "+a" (r2), [slen] "+d" (r3), + [aad] "+a" (r4), [alen] "+d" (r5) + : [fc] "d" (r0), [pba] "a" (r1), [opc] "i" (CPACF_KMA) + : "cc", "memory"); +} + #endif /* _ASM_S390_CPACF_H */ -- cgit From f9a5d70cfaf3e32308de0abfcc95dafe4e36ea51 Mon Sep 17 00:00:00 2001 From: Julian Wiedmann Date: Thu, 14 Sep 2017 09:52:32 +0200 Subject: s390/ccwgroup: tie a ccwgroup driver to its ccw driver When grouping devices, the ccwgroup core only checks whether all of the devices are bound to the same ccw_driver. It has no means of checking if the requesting ccwgroup driver actually supports this device type. qeth implements its own device matching in qeth_core_probe_device(), while ctcm and lcs currently have no sanity-checking at all. Enable ccwgroup drivers to optionally defer the device type checking to the ccwgroup core, by specifying their supported ccw_driver. This allows us drop the device type matching from qeth, and improves the robustness of ctcm and lcs. Signed-off-by: Julian Wiedmann Acked-by: Sebastian Ott Reviewed-by: Peter Oberparleiter Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/ccwgroup.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/ccwgroup.h b/arch/s390/include/asm/ccwgroup.h index 057ce0ca6377..6d50e86dd0e8 100644 --- a/arch/s390/include/asm/ccwgroup.h +++ b/arch/s390/include/asm/ccwgroup.h @@ -41,6 +41,7 @@ struct ccwgroup_device { * @thaw: undo work done in @freeze * @restore: callback for restoring after hibernation * @driver: embedded driver structure + * @ccw_driver: supported ccw_driver (optional) */ struct ccwgroup_driver { int (*setup) (struct ccwgroup_device *); @@ -55,6 +56,7 @@ struct ccwgroup_driver { int (*restore)(struct ccwgroup_device *); struct device_driver driver; + struct ccw_driver *ccw_driver; }; extern int ccwgroup_driver_register (struct ccwgroup_driver *cdriver); -- cgit From 91a1fad759ffd616b836984ca0420c7ad5996eef Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Sat, 30 Sep 2017 10:54:31 +0200 Subject: s390: use generic rwsem implementation We never optimized our rwsem inline assemblies to make use of the new atomic instructions. The generic rwsem implementation implicitly makes use of the new instructions, since it implements the required rwsem primitives with atomic operations, which we did optimize. However even when compiling for old architectures the generic variant still generates better code. So it's time to simply remove our old code and switch to the generic implementation. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/Kbuild | 1 + arch/s390/include/asm/rwsem.h | 210 ------------------------------------------ 2 files changed, 1 insertion(+), 210 deletions(-) delete mode 100644 arch/s390/include/asm/rwsem.h (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/Kbuild b/arch/s390/include/asm/Kbuild index 6e2c9f7e47fa..41c211a4d8b1 100644 --- a/arch/s390/include/asm/Kbuild +++ b/arch/s390/include/asm/Kbuild @@ -15,6 +15,7 @@ generic-y += local64.h generic-y += mcs_spinlock.h generic-y += mm-arch-hooks.h generic-y += preempt.h +generic-y += rwsem.h generic-y += trace_clock.h generic-y += unaligned.h generic-y += word-at-a-time.h diff --git a/arch/s390/include/asm/rwsem.h b/arch/s390/include/asm/rwsem.h deleted file mode 100644 index 597e7e96b59e..000000000000 --- a/arch/s390/include/asm/rwsem.h +++ /dev/null @@ -1,210 +0,0 @@ -#ifndef _S390_RWSEM_H -#define _S390_RWSEM_H - -/* - * S390 version - * Copyright IBM Corp. 2002 - * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) - * - * Based on asm-alpha/semaphore.h and asm-i386/rwsem.h - */ - -/* - * - * The MSW of the count is the negated number of active writers and waiting - * lockers, and the LSW is the total number of active locks - * - * The lock count is initialized to 0 (no active and no waiting lockers). - * - * When a writer subtracts WRITE_BIAS, it'll get 0xffff0001 for the case of an - * uncontended lock. This can be determined because XADD returns the old value. - * Readers increment by 1 and see a positive value when uncontended, negative - * if there are writers (and maybe) readers waiting (in which case it goes to - * sleep). - * - * The value of WAITING_BIAS supports up to 32766 waiting processes. This can - * be extended to 65534 by manually checking the whole MSW rather than relying - * on the S flag. - * - * The value of ACTIVE_BIAS supports up to 65535 active processes. - * - * This should be totally fair - if anything is waiting, a process that wants a - * lock will go to the back of the queue. When the currently active lock is - * released, if there's a writer at the front of the queue, then that and only - * that will be woken up; if there's a bunch of consecutive readers at the - * front, then they'll all be woken up, but no other readers will be. - */ - -#ifndef _LINUX_RWSEM_H -#error "please don't include asm/rwsem.h directly, use linux/rwsem.h instead" -#endif - -#define RWSEM_UNLOCKED_VALUE 0x0000000000000000L -#define RWSEM_ACTIVE_BIAS 0x0000000000000001L -#define RWSEM_ACTIVE_MASK 0x00000000ffffffffL -#define RWSEM_WAITING_BIAS (-0x0000000100000000L) -#define RWSEM_ACTIVE_READ_BIAS RWSEM_ACTIVE_BIAS -#define RWSEM_ACTIVE_WRITE_BIAS (RWSEM_WAITING_BIAS + RWSEM_ACTIVE_BIAS) - -/* - * lock for reading - */ -static inline void __down_read(struct rw_semaphore *sem) -{ - signed long old, new; - - asm volatile( - " lg %0,%2\n" - "0: lgr %1,%0\n" - " aghi %1,%4\n" - " csg %0,%1,%2\n" - " jl 0b" - : "=&d" (old), "=&d" (new), "=Q" (sem->count) - : "Q" (sem->count), "i" (RWSEM_ACTIVE_READ_BIAS) - : "cc", "memory"); - if (old < 0) - rwsem_down_read_failed(sem); -} - -/* - * trylock for reading -- returns 1 if successful, 0 if contention - */ -static inline int __down_read_trylock(struct rw_semaphore *sem) -{ - signed long old, new; - - asm volatile( - " lg %0,%2\n" - "0: ltgr %1,%0\n" - " jm 1f\n" - " aghi %1,%4\n" - " csg %0,%1,%2\n" - " jl 0b\n" - "1:" - : "=&d" (old), "=&d" (new), "=Q" (sem->count) - : "Q" (sem->count), "i" (RWSEM_ACTIVE_READ_BIAS) - : "cc", "memory"); - return old >= 0 ? 1 : 0; -} - -/* - * lock for writing - */ -static inline long ___down_write(struct rw_semaphore *sem) -{ - signed long old, new, tmp; - - tmp = RWSEM_ACTIVE_WRITE_BIAS; - asm volatile( - " lg %0,%2\n" - "0: lgr %1,%0\n" - " ag %1,%4\n" - " csg %0,%1,%2\n" - " jl 0b" - : "=&d" (old), "=&d" (new), "=Q" (sem->count) - : "Q" (sem->count), "m" (tmp) - : "cc", "memory"); - - return old; -} - -static inline void __down_write(struct rw_semaphore *sem) -{ - if (___down_write(sem)) - rwsem_down_write_failed(sem); -} - -static inline int __down_write_killable(struct rw_semaphore *sem) -{ - if (___down_write(sem)) - if (IS_ERR(rwsem_down_write_failed_killable(sem))) - return -EINTR; - - return 0; -} - -/* - * trylock for writing -- returns 1 if successful, 0 if contention - */ -static inline int __down_write_trylock(struct rw_semaphore *sem) -{ - signed long old; - - asm volatile( - " lg %0,%1\n" - "0: ltgr %0,%0\n" - " jnz 1f\n" - " csg %0,%3,%1\n" - " jl 0b\n" - "1:" - : "=&d" (old), "=Q" (sem->count) - : "Q" (sem->count), "d" (RWSEM_ACTIVE_WRITE_BIAS) - : "cc", "memory"); - return (old == RWSEM_UNLOCKED_VALUE) ? 1 : 0; -} - -/* - * unlock after reading - */ -static inline void __up_read(struct rw_semaphore *sem) -{ - signed long old, new; - - asm volatile( - " lg %0,%2\n" - "0: lgr %1,%0\n" - " aghi %1,%4\n" - " csg %0,%1,%2\n" - " jl 0b" - : "=&d" (old), "=&d" (new), "=Q" (sem->count) - : "Q" (sem->count), "i" (-RWSEM_ACTIVE_READ_BIAS) - : "cc", "memory"); - if (new < 0) - if ((new & RWSEM_ACTIVE_MASK) == 0) - rwsem_wake(sem); -} - -/* - * unlock after writing - */ -static inline void __up_write(struct rw_semaphore *sem) -{ - signed long old, new, tmp; - - tmp = -RWSEM_ACTIVE_WRITE_BIAS; - asm volatile( - " lg %0,%2\n" - "0: lgr %1,%0\n" - " ag %1,%4\n" - " csg %0,%1,%2\n" - " jl 0b" - : "=&d" (old), "=&d" (new), "=Q" (sem->count) - : "Q" (sem->count), "m" (tmp) - : "cc", "memory"); - if (new < 0) - if ((new & RWSEM_ACTIVE_MASK) == 0) - rwsem_wake(sem); -} - -/* - * downgrade write lock to read lock - */ -static inline void __downgrade_write(struct rw_semaphore *sem) -{ - signed long old, new, tmp; - - tmp = -RWSEM_WAITING_BIAS; - asm volatile( - " lg %0,%2\n" - "0: lgr %1,%0\n" - " ag %1,%4\n" - " csg %0,%1,%2\n" - " jl 0b" - : "=&d" (old), "=&d" (new), "=Q" (sem->count) - : "Q" (sem->count), "m" (tmp) - : "cc", "memory"); - if (new > 1) - rwsem_downgrade_wake(sem); -} - -#endif /* _S390_RWSEM_H */ -- cgit From b7c92f1a4e131e459bcf53a570e7265e5ce64455 Mon Sep 17 00:00:00 2001 From: QingFeng Hao Date: Fri, 29 Sep 2017 12:41:50 +0200 Subject: s390/sthyi: reorganize sthyi implementation As we need to support sthyi instruction on LPAR too, move the common code to kernel part and kvm related code to intercept.c for better reuse. Signed-off-by: QingFeng Hao Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/sysinfo.h | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/sysinfo.h b/arch/s390/include/asm/sysinfo.h index 2b498e58b914..e4a28307bc5d 100644 --- a/arch/s390/include/asm/sysinfo.h +++ b/arch/s390/include/asm/sysinfo.h @@ -198,4 +198,5 @@ struct service_level { int register_service_level(struct service_level *); int unregister_service_level(struct service_level *); +int sthyi_fill(void *dst, u64 *rc); #endif /* __ASM_S390_SYSINFO_H */ -- cgit From 9fb6c9b3fea1b1d1c6f14178373e8f7235f3b681 Mon Sep 17 00:00:00 2001 From: QingFeng Hao Date: Fri, 29 Sep 2017 12:41:51 +0200 Subject: s390/sthyi: add cache to store hypervisor info STHYI requires extensive locking in the higher hypervisors and is very computational/memory expensive. Therefore we cache the retrieved hypervisor info whose valid period is 1s with mutex to allow concurrent access. rw semaphore can't benefit here due to cache line bounce. Signed-off-by: QingFeng Hao Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/kvm_host.h | 1 - 1 file changed, 1 deletion(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index 51375e766e90..fd006a272024 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -736,7 +736,6 @@ struct kvm_arch{ wait_queue_head_t ipte_wq; int ipte_lock_count; struct mutex ipte_mutex; - struct ratelimit_state sthyi_limit; spinlock_t start_stop_lock; struct sie_page2 *sie_page2; struct kvm_s390_cpu_model model; -- cgit From 0b77d6701cf8d4eb343a83fa8d7eca81a863bb7c Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 4 Oct 2017 19:27:05 +0200 Subject: s390: implement memset16, memset32 & memset64 Provide fast versions of the new memset variants. E.g. the generic memset64 is ten times slower than the optimized version if used on a whole page. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/string.h | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/string.h b/arch/s390/include/asm/string.h index 8fb43319693d..aa9c3a0f59ff 100644 --- a/arch/s390/include/asm/string.h +++ b/arch/s390/include/asm/string.h @@ -17,6 +17,9 @@ #define __HAVE_ARCH_MEMMOVE /* gcc builtin & arch function */ #define __HAVE_ARCH_MEMSCAN /* inline & arch function */ #define __HAVE_ARCH_MEMSET /* gcc builtin & arch function */ +#define __HAVE_ARCH_MEMSET16 /* arch function */ +#define __HAVE_ARCH_MEMSET32 /* arch function */ +#define __HAVE_ARCH_MEMSET64 /* arch function */ #define __HAVE_ARCH_STRCAT /* inline & arch function */ #define __HAVE_ARCH_STRCMP /* arch function */ #define __HAVE_ARCH_STRCPY /* inline & arch function */ @@ -49,6 +52,25 @@ extern char *strstr(const char *, const char *); #undef __HAVE_ARCH_STRSEP #undef __HAVE_ARCH_STRSPN +void *__memset16(uint16_t *s, uint16_t v, size_t count); +void *__memset32(uint32_t *s, uint32_t v, size_t count); +void *__memset64(uint64_t *s, uint64_t v, size_t count); + +static inline void *memset16(uint16_t *s, uint16_t v, size_t count) +{ + return __memset16(s, v, count * sizeof(v)); +} + +static inline void *memset32(uint32_t *s, uint32_t v, size_t count) +{ + return __memset32(s, v, count * sizeof(v)); +} + +static inline void *memset64(uint64_t *s, uint64_t v, size_t count) +{ + return __memset64(s, v, count * sizeof(v)); +} + #if !defined(IN_ARCH_STRING_C) && (!defined(CONFIG_FORTIFY_SOURCE) || defined(__NO_FORTIFY)) static inline void *memchr(const void * s, int c, size_t n) -- cgit From 41879ff65d8b025eace44610be0b07f678fb3224 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 4 Oct 2017 19:27:07 +0200 Subject: s390/mm: use memset64 instead of clear_table Use memset64 instead of the (now) open-coded variant clear_table. Performance wise there is no difference. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/pgalloc.h | 18 ++---------------- 1 file changed, 2 insertions(+), 16 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/pgalloc.h b/arch/s390/include/asm/pgalloc.h index a0d9167519b1..6b341126bebb 100644 --- a/arch/s390/include/asm/pgalloc.h +++ b/arch/s390/include/asm/pgalloc.h @@ -12,6 +12,7 @@ #define _S390_PGALLOC_H #include +#include #include #include @@ -27,24 +28,9 @@ void page_table_free_rcu(struct mmu_gather *, unsigned long *, unsigned long); void page_table_free_pgste(struct page *page); extern int page_table_allocate_pgste; -static inline void clear_table(unsigned long *s, unsigned long val, size_t n) -{ - struct addrtype { char _[256]; }; - int i; - - for (i = 0; i < n; i += 256) { - *s = val; - asm volatile( - "mvc 8(248,%[s]),0(%[s])\n" - : "+m" (*(struct addrtype *) s) - : [s] "a" (s)); - s += 256 / sizeof(long); - } -} - static inline void crst_table_init(unsigned long *crst, unsigned long entry) { - clear_table(crst, entry, _CRST_TABLE_SIZE); + memset64((u64 *)crst, entry, _CRST_ENTRIES); } static inline unsigned long pgd_entry_type(struct mm_struct *mm) -- cgit From 49913f1fd0e5ff5a484214501e84d1c73e739285 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 4 Oct 2017 19:27:09 +0200 Subject: s390: cleanup string ops prototypes Just some trivial changes like removing the extern keyword from the header file, renaming arguments to match the man pages, and whitespace removal. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/string.h | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/string.h b/arch/s390/include/asm/string.h index aa9c3a0f59ff..478767b262f3 100644 --- a/arch/s390/include/asm/string.h +++ b/arch/s390/include/asm/string.h @@ -33,17 +33,17 @@ #define __HAVE_ARCH_STRSTR /* arch function */ /* Prototypes for non-inlined arch strings functions. */ -extern int memcmp(const void *, const void *, size_t); -extern void *memcpy(void *, const void *, size_t); -extern void *memset(void *, int, size_t); -extern void *memmove(void *, const void *, size_t); -extern int strcmp(const char *,const char *); -extern size_t strlcat(char *, const char *, size_t); -extern size_t strlcpy(char *, const char *, size_t); -extern char *strncat(char *, const char *, size_t); -extern char *strncpy(char *, const char *, size_t); -extern char *strrchr(const char *, int); -extern char *strstr(const char *, const char *); +int memcmp(const void *s1, const void *s2, size_t n); +void *memcpy(void *dest, const void *src, size_t n); +void *memset(void *s, int c, size_t n); +void *memmove(void *dest, const void *src, size_t n); +int strcmp(const char *s1, const char *s2); +size_t strlcat(char *dest, const char *src, size_t n); +size_t strlcpy(char *dest, const char *src, size_t size); +char *strncat(char *dest, const char *src, size_t n); +char *strncpy(char *dest, const char *src, size_t n); +char *strrchr(const char *s, int c); +char *strstr(const char *s1, const char *s2); #undef __HAVE_ARCH_STRCHR #undef __HAVE_ARCH_STRNCHR -- cgit From df8bbd0c98bef5bcf691709437c4102e9840b0f5 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 6 Oct 2017 13:17:10 +0200 Subject: s390/kprobes: remove KPROBE_SWAP_INST state For an unknown reason the s390 kprobes instruction replacement function modifies the kprobe_status of the current CPU to KPROBE_SWAP_INST. This was supposed to catch traps that happened during instruction patching. Such a fault is not supposed to happen, and silently discarding such a fault is certainly also not what we want. In fact s390 is the only architecture which has this odd piece of code. Just remove this and behave like all other architectures. This was pointed out by Jens Remus. Reported-by: Jens Remus Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/kprobes.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/kprobes.h b/arch/s390/include/asm/kprobes.h index 28792ef82c83..921391f2341e 100644 --- a/arch/s390/include/asm/kprobes.h +++ b/arch/s390/include/asm/kprobes.h @@ -63,8 +63,6 @@ typedef u16 kprobe_opcode_t; #define kretprobe_blacklist_size 0 -#define KPROBE_SWAP_INST 0x10 - /* Architecture specific copy of original instruction */ struct arch_specific_insn { /* copy of original instruction */ -- cgit From 496da0d706a952f12b4cbbec4b9f60d3ffdf5356 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 13 Oct 2017 09:06:29 +0200 Subject: s390/debug: adjust coding style The debug feature code hasn't been touched in ages and the code also looks like this. Therefore clean up the code so it looks a bit more like current coding style. There is no functional change - actually I made also sure that the generated code with performance_defconfig is identical. A diff of old vs new with "objdump -d" is empty. The code is still not checkpatch clean, but that was not the goal. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/debug.h | 190 ++++++++++++++++++++++-------------------- 1 file changed, 98 insertions(+), 92 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/debug.h b/arch/s390/include/asm/debug.h index df7b54ea956d..4a4c6dd2585a 100644 --- a/arch/s390/include/asm/debug.h +++ b/arch/s390/include/asm/debug.h @@ -13,71 +13,71 @@ #include #include -#define DEBUG_MAX_LEVEL 6 /* debug levels range from 0 to 6 */ -#define DEBUG_OFF_LEVEL -1 /* level where debug is switched off */ -#define DEBUG_FLUSH_ALL -1 /* parameter to flush all areas */ -#define DEBUG_MAX_VIEWS 10 /* max number of views in proc fs */ -#define DEBUG_MAX_NAME_LEN 64 /* max length for a debugfs file name */ -#define DEBUG_DEFAULT_LEVEL 3 /* initial debug level */ +#define DEBUG_MAX_LEVEL 6 /* debug levels range from 0 to 6 */ +#define DEBUG_OFF_LEVEL -1 /* level where debug is switched off */ +#define DEBUG_FLUSH_ALL -1 /* parameter to flush all areas */ +#define DEBUG_MAX_VIEWS 10 /* max number of views in proc fs */ +#define DEBUG_MAX_NAME_LEN 64 /* max length for a debugfs file name */ +#define DEBUG_DEFAULT_LEVEL 3 /* initial debug level */ #define DEBUG_DIR_ROOT "s390dbf" /* name of debug root directory in proc fs */ -#define DEBUG_DATA(entry) (char*)(entry + 1) /* data is stored behind */ - /* the entry information */ +#define DEBUG_DATA(entry) (char *)(entry + 1) /* data is stored behind */ + /* the entry information */ typedef struct __debug_entry debug_entry_t; struct debug_view; -typedef struct debug_info { - struct debug_info* next; - struct debug_info* prev; +typedef struct debug_info { + struct debug_info *next; + struct debug_info *prev; refcount_t ref_count; - spinlock_t lock; + spinlock_t lock; int level; int nr_areas; int pages_per_area; int buf_size; - int entry_size; - debug_entry_t*** areas; + int entry_size; + debug_entry_t ***areas; int active_area; int *active_pages; int *active_entries; - struct dentry* debugfs_root_entry; - struct dentry* debugfs_entries[DEBUG_MAX_VIEWS]; - struct debug_view* views[DEBUG_MAX_VIEWS]; + struct dentry *debugfs_root_entry; + struct dentry *debugfs_entries[DEBUG_MAX_VIEWS]; + struct debug_view *views[DEBUG_MAX_VIEWS]; char name[DEBUG_MAX_NAME_LEN]; umode_t mode; } debug_info_t; -typedef int (debug_header_proc_t) (debug_info_t* id, - struct debug_view* view, +typedef int (debug_header_proc_t) (debug_info_t *id, + struct debug_view *view, int area, - debug_entry_t* entry, - char* out_buf); - -typedef int (debug_format_proc_t) (debug_info_t* id, - struct debug_view* view, char* out_buf, - const char* in_buf); -typedef int (debug_prolog_proc_t) (debug_info_t* id, - struct debug_view* view, - char* out_buf); -typedef int (debug_input_proc_t) (debug_info_t* id, - struct debug_view* view, - struct file* file, + debug_entry_t *entry, + char *out_buf); + +typedef int (debug_format_proc_t) (debug_info_t *id, + struct debug_view *view, char *out_buf, + const char *in_buf); +typedef int (debug_prolog_proc_t) (debug_info_t *id, + struct debug_view *view, + char *out_buf); +typedef int (debug_input_proc_t) (debug_info_t *id, + struct debug_view *view, + struct file *file, const char __user *user_buf, - size_t in_buf_size, loff_t* offset); + size_t in_buf_size, loff_t *offset); + +int debug_dflt_header_fn(debug_info_t *id, struct debug_view *view, + int area, debug_entry_t *entry, char *out_buf); -int debug_dflt_header_fn(debug_info_t* id, struct debug_view* view, - int area, debug_entry_t* entry, char* out_buf); - struct debug_view { char name[DEBUG_MAX_NAME_LEN]; - debug_prolog_proc_t* prolog_proc; - debug_header_proc_t* header_proc; - debug_format_proc_t* format_proc; - debug_input_proc_t* input_proc; - void* private_data; + debug_prolog_proc_t *prolog_proc; + debug_header_proc_t *header_proc; + debug_format_proc_t *format_proc; + debug_input_proc_t *input_proc; + void *private_data; }; extern struct debug_view debug_hex_ascii_view; @@ -86,65 +86,67 @@ extern struct debug_view debug_sprintf_view; /* do NOT use the _common functions */ -debug_entry_t* debug_event_common(debug_info_t* id, int level, - const void* data, int length); +debug_entry_t *debug_event_common(debug_info_t *id, int level, + const void *data, int length); -debug_entry_t* debug_exception_common(debug_info_t* id, int level, - const void* data, int length); +debug_entry_t *debug_exception_common(debug_info_t *id, int level, + const void *data, int length); /* Debug Feature API: */ debug_info_t *debug_register(const char *name, int pages, int nr_areas, - int buf_size); + int buf_size); debug_info_t *debug_register_mode(const char *name, int pages, int nr_areas, int buf_size, umode_t mode, uid_t uid, gid_t gid); -void debug_unregister(debug_info_t* id); +void debug_unregister(debug_info_t *id); -void debug_set_level(debug_info_t* id, int new_level); +void debug_set_level(debug_info_t *id, int new_level); void debug_set_critical(void); void debug_stop_all(void); -static inline bool debug_level_enabled(debug_info_t* id, int level) +static inline bool debug_level_enabled(debug_info_t *id, int level) { return level <= id->level; } -static inline debug_entry_t* -debug_event(debug_info_t* id, int level, void* data, int length) +static inline debug_entry_t *debug_event(debug_info_t *id, int level, + void *data, int length) { if ((!id) || (level > id->level) || (id->pages_per_area == 0)) return NULL; - return debug_event_common(id,level,data,length); + return debug_event_common(id, level, data, length); } -static inline debug_entry_t* -debug_int_event(debug_info_t* id, int level, unsigned int tag) +static inline debug_entry_t *debug_int_event(debug_info_t *id, int level, + unsigned int tag) { - unsigned int t=tag; + unsigned int t = tag; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) return NULL; - return debug_event_common(id,level,&t,sizeof(unsigned int)); + return debug_event_common(id, level, &t, sizeof(unsigned int)); } -static inline debug_entry_t * -debug_long_event (debug_info_t* id, int level, unsigned long tag) +static inline debug_entry_t *debug_long_event(debug_info_t *id, int level, + unsigned long tag) { - unsigned long t=tag; + unsigned long t = tag; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) return NULL; - return debug_event_common(id,level,&t,sizeof(unsigned long)); + return debug_event_common(id, level, &t, sizeof(unsigned long)); } -static inline debug_entry_t* -debug_text_event(debug_info_t* id, int level, const char* txt) +static inline debug_entry_t *debug_text_event(debug_info_t *id, int level, + const char *txt) { if ((!id) || (level > id->level) || (id->pages_per_area == 0)) return NULL; - return debug_event_common(id,level,txt,strlen(txt)); + return debug_event_common(id, level, txt, strlen(txt)); } /* @@ -160,6 +162,7 @@ __debug_sprintf_event(debug_info_t *id, int level, char *string, ...) debug_entry_t *__ret; \ debug_info_t *__id = _id; \ int __level = _level; \ + \ if ((!__id) || (__level > __id->level)) \ __ret = NULL; \ else \ @@ -168,38 +171,40 @@ __debug_sprintf_event(debug_info_t *id, int level, char *string, ...) __ret; \ }) -static inline debug_entry_t* -debug_exception(debug_info_t* id, int level, void* data, int length) +static inline debug_entry_t *debug_exception(debug_info_t *id, int level, + void *data, int length) { if ((!id) || (level > id->level) || (id->pages_per_area == 0)) return NULL; - return debug_exception_common(id,level,data,length); + return debug_exception_common(id, level, data, length); } -static inline debug_entry_t* -debug_int_exception(debug_info_t* id, int level, unsigned int tag) +static inline debug_entry_t *debug_int_exception(debug_info_t *id, int level, + unsigned int tag) { - unsigned int t=tag; + unsigned int t = tag; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) return NULL; - return debug_exception_common(id,level,&t,sizeof(unsigned int)); + return debug_exception_common(id, level, &t, sizeof(unsigned int)); } -static inline debug_entry_t * -debug_long_exception (debug_info_t* id, int level, unsigned long tag) +static inline debug_entry_t *debug_long_exception (debug_info_t *id, int level, + unsigned long tag) { - unsigned long t=tag; + unsigned long t = tag; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) return NULL; - return debug_exception_common(id,level,&t,sizeof(unsigned long)); + return debug_exception_common(id, level, &t, sizeof(unsigned long)); } -static inline debug_entry_t* -debug_text_exception(debug_info_t* id, int level, const char* txt) +static inline debug_entry_t *debug_text_exception(debug_info_t *id, int level, + const char *txt) { if ((!id) || (level > id->level) || (id->pages_per_area == 0)) return NULL; - return debug_exception_common(id,level,txt,strlen(txt)); + return debug_exception_common(id, level, txt, strlen(txt)); } /* @@ -215,6 +220,7 @@ __debug_sprintf_exception(debug_info_t *id, int level, char *string, ...) debug_entry_t *__ret; \ debug_info_t *__id = _id; \ int __level = _level; \ + \ if ((!__id) || (__level > __id->level)) \ __ret = NULL; \ else \ @@ -223,13 +229,13 @@ __debug_sprintf_exception(debug_info_t *id, int level, char *string, ...) __ret; \ }) -int debug_register_view(debug_info_t* id, struct debug_view* view); -int debug_unregister_view(debug_info_t* id, struct debug_view* view); +int debug_register_view(debug_info_t *id, struct debug_view *view); +int debug_unregister_view(debug_info_t *id, struct debug_view *view); /* define the debug levels: - 0 No debugging output to console or syslog - - 1 Log internal errors to syslog, ignore check conditions + - 1 Log internal errors to syslog, ignore check conditions - 2 Log internal errors and check conditions to syslog - 3 Log internal errors to console, log check conditions to syslog - 4 Log internal errors and check conditions to console @@ -247,17 +253,17 @@ int debug_unregister_view(debug_info_t* id, struct debug_view* view); #define INTERNAL_DEBMSG(x,y...) "D" __FILE__ "%d: " x, __LINE__, y #if DEBUG_LEVEL > 0 -#define PRINT_DEBUG(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) -#define PRINT_INFO(x...) printk ( KERN_INFO PRINTK_HEADER x ) -#define PRINT_WARN(x...) printk ( KERN_WARNING PRINTK_HEADER x ) -#define PRINT_ERR(x...) printk ( KERN_ERR PRINTK_HEADER x ) -#define PRINT_FATAL(x...) panic ( PRINTK_HEADER x ) +#define PRINT_DEBUG(x...) printk(KERN_DEBUG PRINTK_HEADER x) +#define PRINT_INFO(x...) printk(KERN_INFO PRINTK_HEADER x) +#define PRINT_WARN(x...) printk(KERN_WARNING PRINTK_HEADER x) +#define PRINT_ERR(x...) printk(KERN_ERR PRINTK_HEADER x) +#define PRINT_FATAL(x...) panic(PRINTK_HEADER x) #else -#define PRINT_DEBUG(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) -#define PRINT_INFO(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) -#define PRINT_WARN(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) -#define PRINT_ERR(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) -#define PRINT_FATAL(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) -#endif /* DASD_DEBUG */ - -#endif /* DEBUG_H */ +#define PRINT_DEBUG(x...) printk(KERN_DEBUG PRINTK_HEADER x) +#define PRINT_INFO(x...) printk(KERN_DEBUG PRINTK_HEADER x) +#define PRINT_WARN(x...) printk(KERN_DEBUG PRINTK_HEADER x) +#define PRINT_ERR(x...) printk(KERN_DEBUG PRINTK_HEADER x) +#define PRINT_FATAL(x...) printk(KERN_DEBUG PRINTK_HEADER x) +#endif /* DASD_DEBUG */ + +#endif /* DEBUG_H */ -- cgit From 94158e544fd60c6a94af348790dae76578ed8dae Mon Sep 17 00:00:00 2001 From: Sebastian Ott Date: Mon, 9 Oct 2017 17:49:38 +0200 Subject: s390/debug: improve debug_event debug_event currently truncates the data if used with a size larger than the buf_size of the debug feature. For lots of callers of this function, wrappers have been implemented that loop until all data is handled. Move that functionality into debug_event_common and get rid of the wrappers. Signed-off-by: Sebastian Ott Acked-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/pci_debug.h | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/pci_debug.h b/arch/s390/include/asm/pci_debug.h index ac24b26fc065..773ff1352a96 100644 --- a/arch/s390/include/asm/pci_debug.h +++ b/arch/s390/include/asm/pci_debug.h @@ -18,11 +18,7 @@ extern debug_info_t *pci_debug_err_id; static inline void zpci_err_hex(void *addr, int len) { - while (len > 0) { - debug_event(pci_debug_err_id, 0, (void *) addr, len); - len -= pci_debug_err_id->buf_size; - addr += pci_debug_err_id->buf_size; - } + debug_event(pci_debug_err_id, 0, addr, len); } #endif -- cgit From 686140a1a9c41d85a4212a1c26d671139b76404b Mon Sep 17 00:00:00 2001 From: Vasily Gorbik Date: Thu, 12 Oct 2017 13:01:47 +0200 Subject: s390: introduce CPU alternatives Implement CPU alternatives, which allows to optionally patch newer instructions at runtime, based on CPU facilities availability. A new kernel boot parameter "noaltinstr" disables patching. Current implementation is derived from x86 alternatives. Although ideal instructions padding (when altinstr is longer then oldinstr) is added at compile time, and no oldinstr nops optimization has to be done at runtime. Also couple of compile time sanity checks are done: 1. oldinstr and altinstr must be <= 254 bytes long, 2. oldinstr and altinstr must not have an odd length. alternative(oldinstr, altinstr, facility); alternative_2(oldinstr, altinstr1, facility1, altinstr2, facility2); Both compile time and runtime padding consists of either 6/4/2 bytes nop or a jump (brcl) + 2 bytes nop filler if padding is longer then 6 bytes. .altinstructions and .altinstr_replacement sections are part of __init_begin : __init_end region and are freed after initialization. Signed-off-by: Vasily Gorbik Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/alternative.h | 163 ++++++++++++++++++++++++++++++++++++ 1 file changed, 163 insertions(+) create mode 100644 arch/s390/include/asm/alternative.h (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/alternative.h b/arch/s390/include/asm/alternative.h new file mode 100644 index 000000000000..6c268f6a51d3 --- /dev/null +++ b/arch/s390/include/asm/alternative.h @@ -0,0 +1,163 @@ +#ifndef _ASM_S390_ALTERNATIVE_H +#define _ASM_S390_ALTERNATIVE_H + +#ifndef __ASSEMBLY__ + +#include +#include +#include + +struct alt_instr { + s32 instr_offset; /* original instruction */ + s32 repl_offset; /* offset to replacement instruction */ + u16 facility; /* facility bit set for replacement */ + u8 instrlen; /* length of original instruction */ + u8 replacementlen; /* length of new instruction */ +} __packed; + +#ifdef CONFIG_ALTERNATIVES +extern void apply_alternative_instructions(void); +extern void apply_alternatives(struct alt_instr *start, struct alt_instr *end); +#else +static inline void apply_alternative_instructions(void) {}; +static inline void apply_alternatives(struct alt_instr *start, + struct alt_instr *end) {}; +#endif +/* + * |661: |662: |6620 |663: + * +-----------+---------------------+ + * | oldinstr | oldinstr_padding | + * | +----------+----------+ + * | | | | + * | | >6 bytes |6/4/2 nops| + * | |6 bytes jg-----------> + * +-----------+---------------------+ + * ^^ static padding ^^ + * + * .altinstr_replacement section + * +---------------------+-----------+ + * |6641: |6651: + * | alternative instr 1 | + * +-----------+---------+- - - - - -+ + * |6642: |6652: | + * | alternative instr 2 | padding + * +---------------------+- - - - - -+ + * ^ runtime ^ + * + * .altinstructions section + * +---------------------------------+ + * | alt_instr entries for each | + * | alternative instr | + * +---------------------------------+ + */ + +#define b_altinstr(num) "664"#num +#define e_altinstr(num) "665"#num + +#define e_oldinstr_pad_end "663" +#define oldinstr_len "662b-661b" +#define oldinstr_total_len e_oldinstr_pad_end"b-661b" +#define altinstr_len(num) e_altinstr(num)"b-"b_altinstr(num)"b" +#define oldinstr_pad_len(num) \ + "-(((" altinstr_len(num) ")-(" oldinstr_len ")) > 0) * " \ + "((" altinstr_len(num) ")-(" oldinstr_len "))" + +#define INSTR_LEN_SANITY_CHECK(len) \ + ".if " len " > 254\n" \ + "\t.error \"cpu alternatives does not support instructions " \ + "blocks > 254 bytes\"\n" \ + ".endif\n" \ + ".if (" len ") %% 2\n" \ + "\t.error \"cpu alternatives instructions length is odd\"\n" \ + ".endif\n" + +#define OLDINSTR_PADDING(oldinstr, num) \ + ".if " oldinstr_pad_len(num) " > 6\n" \ + "\tjg " e_oldinstr_pad_end "f\n" \ + "6620:\n" \ + "\t.fill (" oldinstr_pad_len(num) " - (6620b-662b)) / 2, 2, 0x0700\n" \ + ".else\n" \ + "\t.fill " oldinstr_pad_len(num) " / 6, 6, 0xc0040000\n" \ + "\t.fill " oldinstr_pad_len(num) " %% 6 / 4, 4, 0x47000000\n" \ + "\t.fill " oldinstr_pad_len(num) " %% 6 %% 4 / 2, 2, 0x0700\n" \ + ".endif\n" + +#define OLDINSTR(oldinstr, num) \ + "661:\n\t" oldinstr "\n662:\n" \ + OLDINSTR_PADDING(oldinstr, num) \ + e_oldinstr_pad_end ":\n" \ + INSTR_LEN_SANITY_CHECK(oldinstr_len) + +#define OLDINSTR_2(oldinstr, num1, num2) \ + "661:\n\t" oldinstr "\n662:\n" \ + ".if " altinstr_len(num1) " < " altinstr_len(num2) "\n" \ + OLDINSTR_PADDING(oldinstr, num2) \ + ".else\n" \ + OLDINSTR_PADDING(oldinstr, num1) \ + ".endif\n" \ + e_oldinstr_pad_end ":\n" \ + INSTR_LEN_SANITY_CHECK(oldinstr_len) + +#define ALTINSTR_ENTRY(facility, num) \ + "\t.long 661b - .\n" /* old instruction */ \ + "\t.long " b_altinstr(num)"b - .\n" /* alt instruction */ \ + "\t.word " __stringify(facility) "\n" /* facility bit */ \ + "\t.byte " oldinstr_total_len "\n" /* source len */ \ + "\t.byte " altinstr_len(num) "\n" /* alt instruction len */ + +#define ALTINSTR_REPLACEMENT(altinstr, num) /* replacement */ \ + b_altinstr(num)":\n\t" altinstr "\n" e_altinstr(num) ":\n" \ + INSTR_LEN_SANITY_CHECK(altinstr_len(num)) + +#ifdef CONFIG_ALTERNATIVES +/* alternative assembly primitive: */ +#define ALTERNATIVE(oldinstr, altinstr, facility) \ + ".pushsection .altinstr_replacement, \"ax\"\n" \ + ALTINSTR_REPLACEMENT(altinstr, 1) \ + ".popsection\n" \ + OLDINSTR(oldinstr, 1) \ + ".pushsection .altinstructions,\"a\"\n" \ + ALTINSTR_ENTRY(facility, 1) \ + ".popsection\n" + +#define ALTERNATIVE_2(oldinstr, altinstr1, facility1, altinstr2, facility2)\ + ".pushsection .altinstr_replacement, \"ax\"\n" \ + ALTINSTR_REPLACEMENT(altinstr1, 1) \ + ALTINSTR_REPLACEMENT(altinstr2, 2) \ + ".popsection\n" \ + OLDINSTR_2(oldinstr, 1, 2) \ + ".pushsection .altinstructions,\"a\"\n" \ + ALTINSTR_ENTRY(facility1, 1) \ + ALTINSTR_ENTRY(facility2, 2) \ + ".popsection\n" +#else +/* Alternative instructions are disabled, let's put just oldinstr in */ +#define ALTERNATIVE(oldinstr, altinstr, facility) \ + oldinstr "\n" + +#define ALTERNATIVE_2(oldinstr, altinstr1, facility1, altinstr2, facility2) \ + oldinstr "\n" +#endif + +/* + * Alternative instructions for different CPU types or capabilities. + * + * This allows to use optimized instructions even on generic binary + * kernels. + * + * oldinstr is padded with jump and nops at compile time if altinstr is + * longer. altinstr is padded with jump and nops at run-time during patching. + * + * For non barrier like inlines please define new variants + * without volatile and memory clobber. + */ +#define alternative(oldinstr, altinstr, facility) \ + asm volatile(ALTERNATIVE(oldinstr, altinstr, facility) : : : "memory") + +#define alternative_2(oldinstr, altinstr1, facility1, altinstr2, facility2) \ + asm volatile(ALTERNATIVE_2(oldinstr, altinstr1, facility1, \ + altinstr2, facility2) ::: "memory") + +#endif /* __ASSEMBLY__ */ + +#endif /* _ASM_S390_ALTERNATIVE_H */ -- cgit From f554be42fd0fd8dd14680c67f2db26b3e7de9670 Mon Sep 17 00:00:00 2001 From: Vasily Gorbik Date: Thu, 12 Oct 2017 13:01:47 +0200 Subject: s390/spinlock: use cpu alternatives to enable niai instruction Enable niai instruction in the spinlock code at run-time for machines on which facility 49 is available (zEC12 and newer). Signed-off-by: Vasily Gorbik Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/spinlock.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h index 09e783d83d5d..709cb6c01db9 100644 --- a/arch/s390/include/asm/spinlock.h +++ b/arch/s390/include/asm/spinlock.h @@ -13,6 +13,7 @@ #include #include #include +#include #define SPINLOCK_LOCKVAL (S390_lowcore.spinlock_lockval) @@ -86,9 +87,7 @@ static inline void arch_spin_unlock(arch_spinlock_t *lp) { typecheck(int, lp->lock); asm volatile( -#ifdef CONFIG_HAVE_MARCH_ZEC12_FEATURES - " .long 0xb2fa0070\n" /* NIAI 7 */ -#endif + ALTERNATIVE("", ".long 0xb2fa0070", 49) /* NIAI 7 */ " sth %1,%0\n" : "=Q" (((unsigned short *) &lp->lock)[1]) : "d" (0) : "cc", "memory"); -- cgit From 608796ffe13855bb066bebbd58d8b86a49cb5c27 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 13 Oct 2017 12:59:22 +0200 Subject: s390/vdso: move boot_vdso_data to vdso.c The boot_vdso_data variable is related to the vdso code, the magic of the initial vdso area for the early boot and the replacement of it in vdso_init should all be put into vdso.c. Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/vdso.h | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/vdso.h b/arch/s390/include/asm/vdso.h index 88bdc477a843..ac7bf9806a9d 100644 --- a/arch/s390/include/asm/vdso.h +++ b/arch/s390/include/asm/vdso.h @@ -46,6 +46,7 @@ struct vdso_per_cpu_data { extern struct vdso_data *vdso_data; +void vdso_alloc_boot_cpu(struct lowcore *lowcore); int vdso_alloc_per_cpu(struct lowcore *lowcore); void vdso_free_per_cpu(struct lowcore *lowcore); -- cgit From 00a8f886dbdaeea1d93543d5311ddf3a2680bf2b Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 15 Sep 2017 16:24:31 +0200 Subject: s390/nmi: use smp_emergency_stop instead of smp_send_stop The smp_send_stop() function can be called from s390_handle_damage while DAT is off. This happens if a machine check indicates that kernel gprs or control registers can not be restored. The function smp_send_stop reenables DAT via __load_psw_mask. That should work for the case of lost kernel gprs and the system will do the expected stop of all CPUs. But if control registers are lost, in particular CR13 with the home space ASCE, interesting secondary crashes may occur. Make smp_emergency_stop callable from nmi.c and remove the cpumask argument. Replace the smp_send_stop call with smp_emergency_stop in the s390_handle_damage function. In addition add notrace and NOKPROBE_SYMBOL annotations for all functions required for the emergency shutdown. Reviewed-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/smp.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/smp.h b/arch/s390/include/asm/smp.h index 3deb134587b7..3470274a985c 100644 --- a/arch/s390/include/asm/smp.h +++ b/arch/s390/include/asm/smp.h @@ -27,6 +27,7 @@ extern void arch_send_call_function_ipi_mask(const struct cpumask *mask); extern void smp_call_online_cpu(void (*func)(void *), void *); extern void smp_call_ipl_cpu(void (*func)(void *), void *); +extern void smp_emergency_stop(void); extern int smp_find_processor_id(u16 address); extern int smp_store_status(int cpu); @@ -52,6 +53,10 @@ static inline void smp_call_online_cpu(void (*func)(void *), void *data) func(data); } +static inline void smp_emergency_stop(void) +{ +} + static inline int smp_find_processor_id(u16 address) { return 0; } static inline int smp_store_status(int cpu) { return 0; } static inline int smp_vcpu_scheduled(int cpu) { return 1; } -- cgit From ad3bc0ac1d2ed311ef3a9d6f2849948433a9f338 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 12 Oct 2017 13:24:45 +0200 Subject: s390/ctl_reg: use decoding unions in update_cr_regs Add a decoding union for the bits in control registers 2 and use 'union ctlreg0' and 'union ctlreg2' in update_cr_regs to improve readability. Reviewed-by: Heiko Carstens Reviewed-by: Hendrik Brueckner Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/ctl_reg.h | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/ctl_reg.h b/arch/s390/include/asm/ctl_reg.h index e508dff92535..5d23ecf4297a 100644 --- a/arch/s390/include/asm/ctl_reg.h +++ b/arch/s390/include/asm/ctl_reg.h @@ -54,7 +54,11 @@ void smp_ctl_clear_bit(int cr, int bit); union ctlreg0 { unsigned long val; struct { - unsigned long : 32; + unsigned long : 8; + unsigned long tcx : 1; /* Transactional-Execution control */ + unsigned long pifo : 1; /* Transactional-Execution Program- + Interruption-Filtering Override */ + unsigned long : 22; unsigned long : 3; unsigned long lap : 1; /* Low-address-protection control */ unsigned long : 4; @@ -70,6 +74,19 @@ union ctlreg0 { }; }; +union ctlreg2 { + unsigned long val; + struct { + unsigned long : 33; + unsigned long ducto : 25; + unsigned long : 1; + unsigned long gse : 1; + unsigned long : 1; + unsigned long tds : 1; + unsigned long tdc : 2; + }; +}; + #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) -- cgit From cc65450c8337848f97e893c5c3de973ece73aabf Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 12 Oct 2017 13:24:46 +0200 Subject: s390/ctl_reg: move control register definitions to ctl_reg.h The nmi.h header has some constant defines for control register bits. These definitions should really be located in ctl_reg.h. Move and rename the defines. Reviewed-by: Heiko Carstens Reviewed-by: Hendrik Brueckner Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/ctl_reg.h | 11 +++++++++++ arch/s390/include/asm/nmi.h | 6 ------ 2 files changed, 11 insertions(+), 6 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/ctl_reg.h b/arch/s390/include/asm/ctl_reg.h index 5d23ecf4297a..0095a40886e1 100644 --- a/arch/s390/include/asm/ctl_reg.h +++ b/arch/s390/include/asm/ctl_reg.h @@ -7,6 +7,16 @@ #ifndef __ASM_CTL_REG_H #define __ASM_CTL_REG_H +#include + +#define CR14_CHANNEL_REPORT_SUBMASK _BITUL(63 - 35) +#define CR14_RECOVERY_SUBMASK _BITUL(63 - 36) +#define CR14_DEGRADATION_SUBMASK _BITUL(63 - 37) +#define CR14_EXTERNAL_DAMAGE_SUBMASK _BITUL(63 - 38) +#define CR14_WARNING_SUBMASK _BITUL(63 - 39) + +#ifndef __ASSEMBLY__ + #include #define __ctl_load(array, low, high) do { \ @@ -95,4 +105,5 @@ union ctlreg2 { # define ctl_clear_bit(cr, bit) __ctl_clear_bit(cr, bit) #endif +#endif /* __ASSEMBLY__ */ #endif /* __ASM_CTL_REG_H */ diff --git a/arch/s390/include/asm/nmi.h b/arch/s390/include/asm/nmi.h index c8e211b9a002..77a7d9445e33 100644 --- a/arch/s390/include/asm/nmi.h +++ b/arch/s390/include/asm/nmi.h @@ -26,12 +26,6 @@ #define MCCK_CODE_PSW_MWP_VALID _BITUL(63 - 20) #define MCCK_CODE_PSW_IA_VALID _BITUL(63 - 23) -#define MCCK_CR14_CR_PENDING_SUB_MASK (1 << 28) -#define MCCK_CR14_RECOVERY_SUB_MASK (1 << 27) -#define MCCK_CR14_DEGRAD_SUB_MASK (1 << 26) -#define MCCK_CR14_EXT_DAMAGE_SUB_MASK (1 << 25) -#define MCCK_CR14_WARN_SUB_MASK (1 << 24) - #ifndef __ASSEMBLY__ union mci { -- cgit From 6c81511ca1f52a0bbe921b2b98e34319a4ca59ed Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 12 Oct 2017 13:24:47 +0200 Subject: s390/nmi: allocation of the extended save area The machine check extended save area is needed to store the vector registers and the guarded storage control block when a CPU is interrupted by a machine check. Move the slab cache allocation of the full save area to nmi.c, for early boot use a static __initdata block. Reviewed-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/nmi.h | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/nmi.h b/arch/s390/include/asm/nmi.h index 77a7d9445e33..ed41c424448d 100644 --- a/arch/s390/include/asm/nmi.h +++ b/arch/s390/include/asm/nmi.h @@ -80,6 +80,8 @@ union mci { #define MCESA_ORIGIN_MASK (~0x3ffUL) #define MCESA_LC_MASK (0xfUL) +#define MCESA_MIN_SIZE (1024) +#define MCESA_MAX_SIZE (2048) struct mcesa { u8 vector_save_area[1024]; @@ -88,8 +90,12 @@ struct mcesa { struct pt_regs; -extern void s390_handle_mcck(void); -extern void s390_do_machine_check(struct pt_regs *regs); +void nmi_alloc_boot_cpu(struct lowcore *lc); +int nmi_alloc_per_cpu(struct lowcore *lc); +void nmi_free_per_cpu(struct lowcore *lc); + +void s390_handle_mcck(void); +void s390_do_machine_check(struct pt_regs *regs); #endif /* __ASSEMBLY__ */ #endif /* _ASM_S390_NMI_H */ -- cgit From 3037a52f9846b9d6e233274453f2d4117a14f31b Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 12 Oct 2017 13:24:48 +0200 Subject: s390/nmi: do register validation as early as possible The validation of the CPU registers in the machine check handler is currently split into two parts. The first part is done at the start of the low level mcck_int_handler function, this includes the CPU timer register and the general purpose registers. The second part is done a bit later in s390_do_machine_check for all the other registers, including the control registers, floating pointer control, vector or floating pointer registers, the access registers, the guarded storage registers, the TOD programmable registers and the clock comparator. This is working fine to far but in theory a future extensions could cause the C code to use registers that are not validated yet. A better approach is to validate all CPU registers in "safe" assembler code before any C function is called. Reviewed-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/ctl_reg.h | 2 ++ arch/s390/include/asm/nmi.h | 3 +++ 2 files changed, 5 insertions(+) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/ctl_reg.h b/arch/s390/include/asm/ctl_reg.h index 0095a40886e1..4d4f35f705c7 100644 --- a/arch/s390/include/asm/ctl_reg.h +++ b/arch/s390/include/asm/ctl_reg.h @@ -9,6 +9,8 @@ #include +#define CR2_GUARDED_STORAGE _BITUL(63 - 59) + #define CR14_CHANNEL_REPORT_SUBMASK _BITUL(63 - 35) #define CR14_RECOVERY_SUBMASK _BITUL(63 - 36) #define CR14_DEGRADATION_SUBMASK _BITUL(63 - 37) diff --git a/arch/s390/include/asm/nmi.h b/arch/s390/include/asm/nmi.h index ed41c424448d..7472bf316a2f 100644 --- a/arch/s390/include/asm/nmi.h +++ b/arch/s390/include/asm/nmi.h @@ -25,6 +25,9 @@ #define MCCK_CODE_CPU_TIMER_VALID _BITUL(63 - 46) #define MCCK_CODE_PSW_MWP_VALID _BITUL(63 - 20) #define MCCK_CODE_PSW_IA_VALID _BITUL(63 - 23) +#define MCCK_CODE_CR_VALID _BITUL(63 - 29) +#define MCCK_CODE_GS_VALID _BITUL(63 - 36) +#define MCCK_CODE_FC_VALID _BITUL(63 - 43) #ifndef __ASSEMBLY__ -- cgit From 48070c73058be6de9c0d754d441ed7092dfc8f12 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Mon, 30 Oct 2017 14:38:58 +0100 Subject: s390/pci: do not require AIS facility As of today QEMU does not provide the AIS facility to its guest. This prevents Linux guests from using PCI devices as the ais facility is checked during init. As this is just a performance optimization, we can move the ais check into the code where we need it (calling the SIC instruction). This is used at initialization and on interrupt. Both places do not require any serialization, so we can simply skip the instruction. Since we will now get all interrupts, we can also avoid the 2nd scan. As we can have multiple interrupts in parallel we might trigger spurious irqs more often for the non-AIS case but the core code can handle that. Signed-off-by: Christian Borntraeger Reviewed-by: Pierre Morel Reviewed-by: Halil Pasic Acked-by: Sebastian Ott Signed-off-by: Heiko Carstens --- arch/s390/include/asm/pci_insn.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/pci_insn.h b/arch/s390/include/asm/pci_insn.h index 34abcf275799..a74efc02ad2c 100644 --- a/arch/s390/include/asm/pci_insn.h +++ b/arch/s390/include/asm/pci_insn.h @@ -81,6 +81,6 @@ int zpci_refresh_trans(u64 fn, u64 addr, u64 range); int zpci_load(u64 *data, u64 req, u64 offset); int zpci_store(u64 data, u64 req, u64 offset); int zpci_store_block(const u64 *data, u64 req, u64 offset); -void zpci_set_irq_ctrl(u16 ctl, char *unused, u8 isc); +int zpci_set_irq_ctrl(u16 ctl, char *unused, u8 isc); #endif -- cgit From f44fa88745eda1530083b361e300e1ca4e15a6c5 Mon Sep 17 00:00:00 2001 From: Harald Freudenberger Date: Fri, 27 Oct 2017 15:53:49 +0200 Subject: s390/archrandom: Reconsider s390 arch random implementation The reworked version of the random device driver now calls the arch_get_random_* functions on a very high frequency. It does about 100.000 calls to arch_get_random_long for providing 10 MB via /dev/urandom. Each invocation was fetching entropy from the hardware random generator which has a rate limit of about 4 MB/s. As the trng invocation waits until enough entropy is gathered, the random device driver is slowed down dramatically. The s390 true random generator is not designed for such a high rate. The TRNG is more designed to be used together with the arch_get_random_seed_* functions. This is similar to the way how powerpc has implemented their arch random functionality. This patch removes the invocations of the s390 TRNG for arch_get_random_long() and arch_get_random_int() but leaving the invocations for arch_get_random_seed_long() and arch_get_random_seed_int(). So the s390 arch random implementation now contributes high quality entropy to the kernel random device for reseeding. Signed-off-by: Harald Freudenberger Signed-off-by: Heiko Carstens --- arch/s390/include/asm/archrandom.h | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/archrandom.h b/arch/s390/include/asm/archrandom.h index 6033901a40b2..9695f8d09edf 100644 --- a/arch/s390/include/asm/archrandom.h +++ b/arch/s390/include/asm/archrandom.h @@ -27,42 +27,42 @@ static void s390_arch_random_generate(u8 *buf, unsigned int nbytes) static inline bool arch_has_random(void) { - if (static_branch_likely(&s390_arch_random_available)) - return true; return false; } static inline bool arch_has_random_seed(void) { - return arch_has_random(); + if (static_branch_likely(&s390_arch_random_available)) + return true; + return false; } static inline bool arch_get_random_long(unsigned long *v) { - if (static_branch_likely(&s390_arch_random_available)) { - s390_arch_random_generate((u8 *)v, sizeof(*v)); - return true; - } return false; } static inline bool arch_get_random_int(unsigned int *v) { - if (static_branch_likely(&s390_arch_random_available)) { - s390_arch_random_generate((u8 *)v, sizeof(*v)); - return true; - } return false; } static inline bool arch_get_random_seed_long(unsigned long *v) { - return arch_get_random_long(v); + if (static_branch_likely(&s390_arch_random_available)) { + s390_arch_random_generate((u8 *)v, sizeof(*v)); + return true; + } + return false; } static inline bool arch_get_random_seed_int(unsigned int *v) { - return arch_get_random_int(v); + if (static_branch_likely(&s390_arch_random_available)) { + s390_arch_random_generate((u8 *)v, sizeof(*v)); + return true; + } + return false; } #endif /* CONFIG_ARCH_RANDOM */ -- cgit From 978fa72e82e375764e6e31e7a721408c5186918f Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 2 Nov 2017 12:51:45 +0100 Subject: s390: remove named saved segment support Remove the support to create a z/VM named saved segment (NSS). This feature is not supported since quite a while in favour of jump labels, function tracing and (now) CPU alternatives. All of these features require to write to the kernel text section which is not possible if the kernel is contained within an NSS. Given that memory savings are minimal if kernel images are shared and in addition updates of shared images are painful, the NSS feature can be removed. Reviewed-by: Hendrik Brueckner Signed-off-by: Heiko Carstens --- arch/s390/include/asm/ipl.h | 3 ++- arch/s390/include/asm/sections.h | 2 +- arch/s390/include/asm/setup.h | 3 --- 3 files changed, 3 insertions(+), 5 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/ipl.h b/arch/s390/include/asm/ipl.h index 6810bd757312..c40cb348dd79 100644 --- a/arch/s390/include/asm/ipl.h +++ b/arch/s390/include/asm/ipl.h @@ -12,6 +12,8 @@ #include #include +#define NSS_NAME_SIZE 8 + #define IPL_PARMBLOCK_ORIGIN 0x2000 #define IPL_PARM_BLK_FCP_LEN (sizeof(struct ipl_list_hdr) + \ @@ -105,7 +107,6 @@ extern size_t append_ipl_scpdata(char *, size_t); enum { IPL_DEVNO_VALID = 1, IPL_PARMBLOCK_VALID = 2, - IPL_NSS_VALID = 4, }; enum ipl_type { diff --git a/arch/s390/include/asm/sections.h b/arch/s390/include/asm/sections.h index fbd9116eb17b..cd68bf115889 100644 --- a/arch/s390/include/asm/sections.h +++ b/arch/s390/include/asm/sections.h @@ -3,6 +3,6 @@ #include -extern char _eshared[], _ehead[]; +extern char _ehead[]; #endif diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h index 490e035b3716..fb3c4a138ae3 100644 --- a/arch/s390/include/asm/setup.h +++ b/arch/s390/include/asm/setup.h @@ -97,9 +97,6 @@ extern char vmpoff_cmd[]; #define SET_CONSOLE_VT220 do { console_mode = 4; } while (0) #define SET_CONSOLE_HVC do { console_mode = 5; } while (0) -#define NSS_NAME_SIZE 8 -extern char kernel_nss_name[]; - #ifdef CONFIG_PFAULT extern int pfault_init(void); extern void pfault_fini(void); -- cgit From dac6dc267d0e906fa9263462d4ebd78970a8b511 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 6 Nov 2017 13:28:38 +0100 Subject: s390/disassembler: remove insn_to_mnemonic() insn_to_mnemonic() was introduced ages ago for KVM debugging, but is unused in the meantime. Therefore remove it. Acked-by: Christian Borntraeger Signed-off-by: Heiko Carstens --- arch/s390/include/asm/dis.h | 1 - 1 file changed, 1 deletion(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/dis.h b/arch/s390/include/asm/dis.h index 37f617dfbede..793508c8fc48 100644 --- a/arch/s390/include/asm/dis.h +++ b/arch/s390/include/asm/dis.h @@ -44,7 +44,6 @@ struct pt_regs; void show_code(struct pt_regs *regs); void print_fn_code(unsigned char *code, unsigned long len); -int insn_to_mnemonic(unsigned char *instruction, char *buf, unsigned int len); struct s390_insn *find_insn(unsigned char *code); static inline int is_known_insn(unsigned char *code) -- cgit From 8bc1e4ec79c0fcd48ff8914566fe960c34f27bc2 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 6 Nov 2017 13:29:56 +0100 Subject: s390/disassembler: generate opcode tables from text file The current way of adding new instructions to the opcode tables is painful and error prone. Therefore add, similar to binutils, a text file which contains all opcodes and the corresponding mnemonics and instruction formats. A small gen_opcode_table tool then generates a header file with the required enums and opcode table initializers at the prepare step of the kernel build. This way only a simple text file has to be maintained, which can be rather easily extended. Unlike before where there were plenty of opcode tables and a large switch statement to find the correct opcode table, there is now only one opcode table left which contains all instructions. A second opcode offset table now contains offsets within the opcode table to find instructions which have the same opcode prefix. In order to save space all 1-byte opcode instructions are grouped together at the end of the opcode table. This is also quite similar to like it was before. In addition also move and change code and definitions within the disassembler. As a side effect this reduces the size required for the code and opcode tables by ~1.5k. Signed-off-by: Heiko Carstens --- arch/s390/include/asm/dis.h | 27 +-------------------------- 1 file changed, 1 insertion(+), 26 deletions(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/dis.h b/arch/s390/include/asm/dis.h index 793508c8fc48..b267287e35b6 100644 --- a/arch/s390/include/asm/dis.h +++ b/arch/s390/include/asm/dis.h @@ -8,32 +8,7 @@ #ifndef __ASM_S390_DIS_H__ #define __ASM_S390_DIS_H__ -/* Type of operand */ -#define OPERAND_GPR 0x1 /* Operand printed as %rx */ -#define OPERAND_FPR 0x2 /* Operand printed as %fx */ -#define OPERAND_AR 0x4 /* Operand printed as %ax */ -#define OPERAND_CR 0x8 /* Operand printed as %cx */ -#define OPERAND_VR 0x10 /* Operand printed as %vx */ -#define OPERAND_DISP 0x20 /* Operand printed as displacement */ -#define OPERAND_BASE 0x40 /* Operand printed as base register */ -#define OPERAND_INDEX 0x80 /* Operand printed as index register */ -#define OPERAND_PCREL 0x100 /* Operand printed as pc-relative symbol */ -#define OPERAND_SIGNED 0x200 /* Operand printed as signed value */ -#define OPERAND_LENGTH 0x400 /* Operand printed as length (+1) */ - - -struct s390_operand { - int bits; /* The number of bits in the operand. */ - int shift; /* The number of bits to shift. */ - int flags; /* One bit syntax flags. */ -}; - -struct s390_insn { - const char name[5]; - unsigned char opfrag; - unsigned char format; -}; - +#include static inline int insn_length(unsigned char code) { -- cgit From a1c5befc1c24eb9c1ee83f711e0f21ee79cbb556 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 9 Nov 2017 12:29:34 +0100 Subject: s390: fix transactional execution control register handling MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Dan Horák reported the following crash related to transactional execution: User process fault: interruption code 0013 ilc:3 in libpthread-2.26.so[3ff93c00000+1b000] CPU: 2 PID: 1 Comm: /init Not tainted 4.13.4-300.fc27.s390x #1 Hardware name: IBM 2827 H43 400 (z/VM 6.4.0) task: 00000000fafc8000 task.stack: 00000000fafc4000 User PSW : 0705200180000000 000003ff93c14e70 R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:1 AS:0 CC:2 PM:0 RI:0 EA:3 User GPRS: 0000000000000077 000003ff00000000 000003ff93144d48 000003ff93144d5e 0000000000000000 0000000000000002 0000000000000000 000003ff00000000 0000000000000000 0000000000000418 0000000000000000 000003ffcc9fe770 000003ff93d28f50 000003ff9310acf0 000003ff92b0319a 000003ffcc9fe6d0 User Code: 000003ff93c14e62: 60e0b030 std %f14,48(%r11) 000003ff93c14e66: 60f0b038 std %f15,56(%r11) #000003ff93c14e6a: e5600000ff0e tbegin 0,65294 >000003ff93c14e70: a7740006 brc 7,3ff93c14e7c 000003ff93c14e74: a7080000 lhi %r0,0 000003ff93c14e78: a7f40023 brc 15,3ff93c14ebe 000003ff93c14e7c: b2220000 ipm %r0 000003ff93c14e80: 8800001c srl %r0,28 There are several bugs with control register handling with respect to transactional execution: - on task switch update_per_regs() is only called if the next task has an mm (is not a kernel thread). This however is incorrect. This breaks e.g. for user mode helper handling, where the kernel creates a kernel thread and then execve's a user space program. Control register contents related to transactional execution won't be updated on execve. If the previous task ran with transactional execution disabled then the new task will also run with transactional execution disabled, which is incorrect. Therefore call update_per_regs() unconditionally within switch_to(). - on startup the transactional execution facility is not enabled for the idle thread. This is not really a bug, but an inconsistency to other facilities. Therefore enable the facility if it is available. - on fork the new thread's per_flags field is not cleared. This means that a child process inherits the PER_FLAG_NO_TE flag. This flag can be set with a ptrace request to disable transactional execution for the current process. It should not be inherited by new child processes in order to be consistent with the handling of all other PER related debugging options. Therefore clear the per_flags field in copy_thread_tls(). Reported-and-tested-by: Dan Horák Fixes: d35339a42dd1 ("s390: add support for transactional memory") Cc: # v3.7+ Cc: Martin Schwidefsky Reviewed-by: Christian Borntraeger Reviewed-by: Hendrik Brueckner Signed-off-by: Heiko Carstens --- arch/s390/include/asm/switch_to.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/s390/include/asm') diff --git a/arch/s390/include/asm/switch_to.h b/arch/s390/include/asm/switch_to.h index f6c2b5814ab0..8e6b07609ff4 100644 --- a/arch/s390/include/asm/switch_to.h +++ b/arch/s390/include/asm/switch_to.h @@ -36,8 +36,8 @@ static inline void restore_access_regs(unsigned int *acrs) save_ri_cb(prev->thread.ri_cb); \ save_gs_cb(prev->thread.gs_cb); \ } \ + update_cr_regs(next); \ if (next->mm) { \ - update_cr_regs(next); \ set_cpu_flag(CIF_FPU); \ restore_access_regs(&next->thread.acrs[0]); \ restore_ri_cb(next->thread.ri_cb, prev->thread.ri_cb); \ -- cgit