diff options
Diffstat (limited to 'arch/s390/kernel')
29 files changed, 274 insertions, 441 deletions
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index ece58f2217cb..79724d861dc9 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -13,7 +13,6 @@ #include <linux/purgatory.h> #include <linux/pgtable.h> #include <asm/idle.h> -#include <asm/vdso.h> #include <asm/gmap.h> #include <asm/nmi.h> #include <asm/stacktrace.h> @@ -48,26 +47,17 @@ int main(void) OFFSET(__PT_INT_PARM, pt_regs, int_parm); OFFSET(__PT_INT_PARM_LONG, pt_regs, int_parm_long); OFFSET(__PT_FLAGS, pt_regs, flags); + OFFSET(__PT_CR1, pt_regs, cr1); DEFINE(__PT_SIZE, sizeof(struct pt_regs)); BLANK(); /* stack_frame offsets */ OFFSET(__SF_BACKCHAIN, stack_frame, back_chain); OFFSET(__SF_GPRS, stack_frame, gprs); - OFFSET(__SF_EMPTY, stack_frame, empty1); - OFFSET(__SF_SIE_CONTROL, stack_frame, empty1[0]); - OFFSET(__SF_SIE_SAVEAREA, stack_frame, empty1[1]); - OFFSET(__SF_SIE_REASON, stack_frame, empty1[2]); - OFFSET(__SF_SIE_FLAGS, stack_frame, empty1[3]); - BLANK(); - OFFSET(__VDSO_GETCPU_VAL, vdso_per_cpu_data, getcpu_val); - BLANK(); - /* constants used by the vdso */ - DEFINE(__CLOCK_REALTIME, CLOCK_REALTIME); - DEFINE(__CLOCK_MONOTONIC, CLOCK_MONOTONIC); - DEFINE(__CLOCK_REALTIME_COARSE, CLOCK_REALTIME_COARSE); - DEFINE(__CLOCK_MONOTONIC_COARSE, CLOCK_MONOTONIC_COARSE); - DEFINE(__CLOCK_THREAD_CPUTIME_ID, CLOCK_THREAD_CPUTIME_ID); - DEFINE(__CLOCK_COARSE_RES, LOW_RES_NSEC); + OFFSET(__SF_EMPTY, stack_frame, empty1[0]); + OFFSET(__SF_SIE_CONTROL, stack_frame, empty1[1]); + OFFSET(__SF_SIE_SAVEAREA, stack_frame, empty1[2]); + OFFSET(__SF_SIE_REASON, stack_frame, empty1[3]); + OFFSET(__SF_SIE_FLAGS, stack_frame, empty1[4]); BLANK(); /* idle data offsets */ OFFSET(__CLOCK_IDLE_ENTER, s390_idle_data, clock_idle_enter); @@ -146,12 +136,11 @@ int main(void) OFFSET(__LC_RESTART_FN, lowcore, restart_fn); OFFSET(__LC_RESTART_DATA, lowcore, restart_data); OFFSET(__LC_RESTART_SOURCE, lowcore, restart_source); + OFFSET(__LC_KERNEL_ASCE, lowcore, kernel_asce); OFFSET(__LC_USER_ASCE, lowcore, user_asce); - OFFSET(__LC_VDSO_ASCE, lowcore, vdso_asce); OFFSET(__LC_LPP, lowcore, lpp); OFFSET(__LC_CURRENT_PID, lowcore, current_pid); OFFSET(__LC_PERCPU_OFFSET, lowcore, percpu_offset); - OFFSET(__LC_VDSO_PER_CPU, lowcore, vdso_per_cpu_data); OFFSET(__LC_MACHINE_FLAGS, lowcore, machine_flags); OFFSET(__LC_PREEMPT_COUNT, lowcore, preempt_count); OFFSET(__LC_GMAP, lowcore, gmap); diff --git a/arch/s390/kernel/base.S b/arch/s390/kernel/base.S index b79e0fd571f8..d255c69c1779 100644 --- a/arch/s390/kernel/base.S +++ b/arch/s390/kernel/base.S @@ -11,32 +11,10 @@ #include <asm/asm-offsets.h> #include <asm/nospec-insn.h> #include <asm/ptrace.h> -#include <asm/sigp.h> GEN_BR_THUNK %r9 GEN_BR_THUNK %r14 -ENTRY(s390_base_ext_handler) - stmg %r0,%r15,__LC_SAVE_AREA_ASYNC - basr %r13,0 -0: aghi %r15,-STACK_FRAME_OVERHEAD - larl %r1,s390_base_ext_handler_fn - lg %r9,0(%r1) - ltgr %r9,%r9 - jz 1f - BASR_EX %r14,%r9 -1: lmg %r0,%r15,__LC_SAVE_AREA_ASYNC - ni __LC_EXT_OLD_PSW+1,0xfd # clear wait state bit - lpswe __LC_EXT_OLD_PSW -ENDPROC(s390_base_ext_handler) - - .section .bss - .align 8 - .globl s390_base_ext_handler_fn -s390_base_ext_handler_fn: - .quad 0 - .previous - ENTRY(s390_base_pgm_handler) stmg %r0,%r15,__LC_SAVE_AREA_SYNC basr %r13,0 diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index 705844f73934..cc89763a4d3c 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -169,12 +169,10 @@ static noinline __init void setup_lowcore_early(void) { psw_t psw; + psw.addr = (unsigned long)s390_base_pgm_handler; psw.mask = PSW_MASK_BASE | PSW_DEFAULT_KEY | PSW_MASK_EA | PSW_MASK_BA; if (IS_ENABLED(CONFIG_KASAN)) psw.mask |= PSW_MASK_DAT; - psw.addr = (unsigned long) s390_base_ext_handler; - S390_lowcore.external_new_psw = psw; - psw.addr = (unsigned long) s390_base_pgm_handler; S390_lowcore.program_new_psw = psw; s390_base_pgm_handler_fn = early_pgm_check_handler; S390_lowcore.preempt_count = INIT_PREEMPT_COUNT; diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 86235919c2d1..f1ba197b10c0 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -52,10 +52,11 @@ 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_UPROBE | _TIF_GUARDED_STORAGE | _TIF_PATCH_PENDING) + _TIF_UPROBE | _TIF_GUARDED_STORAGE | _TIF_PATCH_PENDING | \ + _TIF_NOTIFY_SIGNAL) _TIF_TRACE = (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | \ _TIF_SYSCALL_TRACEPOINT) -_CIF_WORK = (_CIF_ASCE_PRIMARY | _CIF_ASCE_SECONDARY | _CIF_FPU) +_CIF_WORK = (_CIF_FPU) _PIF_WORK = (_PIF_PER_TRAP | _PIF_SYSCALL_RESTART) _LPP_OFFSET = __LC_LPP @@ -90,6 +91,12 @@ _LPP_OFFSET = __LC_LPP #endif .endm + .macro DEBUG_USER_ASCE +#ifdef CONFIG_DEBUG_USER_ASCE + brasl %r14,debug_user_asce +#endif + .endm + .macro CHECK_VMAP_STACK savearea,oklabel #ifdef CONFIG_VMAP_STACK lgr %r14,%r15 @@ -110,9 +117,9 @@ _LPP_OFFSET = __LC_LPP #endif .endm - .macro SWITCH_ASYNC savearea,timer + .macro SWITCH_ASYNC savearea,timer,clock tmhh %r8,0x0001 # interrupting from user ? - jnz 2f + jnz 4f #if IS_ENABLED(CONFIG_KVM) lgr %r14,%r9 larl %r13,.Lsie_gmap @@ -125,10 +132,26 @@ _LPP_OFFSET = __LC_LPP #endif 0: larl %r13,.Lpsw_idle_exit cgr %r13,%r9 - jne 1f + jne 3f - mvc __CLOCK_IDLE_EXIT(8,%r2), __LC_INT_CLOCK - mvc __TIMER_IDLE_EXIT(8,%r2), __LC_ASYNC_ENTER_TIMER + larl %r1,smp_cpu_mtid + llgf %r1,0(%r1) + ltgr %r1,%r1 + jz 2f # no SMT, skip mt_cycles calculation + .insn rsy,0xeb0000000017,%r1,5,__SF_EMPTY+80(%r15) + larl %r3,mt_cycles + ag %r3,__LC_PERCPU_OFFSET + la %r4,__SF_EMPTY+16(%r15) +1: lg %r0,0(%r3) + slg %r0,0(%r4) + alg %r0,64(%r4) + stg %r0,0(%r3) + la %r3,8(%r3) + la %r4,8(%r4) + brct %r1,1b + +2: mvc __CLOCK_IDLE_EXIT(8,%r2), \clock + mvc __TIMER_IDLE_EXIT(8,%r2), \timer # account system time going idle ni __LC_CPU_FLAGS+7,255-_CIF_ENABLED_WAIT @@ -146,17 +169,17 @@ _LPP_OFFSET = __LC_LPP mvc __LC_LAST_UPDATE_TIMER(8),__TIMER_IDLE_EXIT(%r2) nihh %r8,0xfcfd # clear wait state and irq bits -1: lg %r14,__LC_ASYNC_STACK # are we already on the target stack? +3: lg %r14,__LC_ASYNC_STACK # are we already on the target stack? slgr %r14,%r15 srag %r14,%r14,STACK_SHIFT - jnz 3f + jnz 5f CHECK_STACK \savearea aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE) - j 4f -2: UPDATE_VTIME %r14,%r15,\timer + j 6f +4: UPDATE_VTIME %r14,%r15,\timer BPENTER __TI_flags(%r12),_TIF_ISOLATE_BP -3: lg %r15,__LC_ASYNC_STACK # load async stack -4: la %r11,STACK_FRAME_OVERHEAD(%r15) +5: lg %r15,__LC_ASYNC_STACK # load async stack +6: la %r11,STACK_FRAME_OVERHEAD(%r15) .endm .macro UPDATE_VTIME w1,w2,enter_timer @@ -327,7 +350,7 @@ ENTRY(sie64a) BPENTER __SF_SIE_FLAGS(%r15),(_TIF_ISOLATE_BP|_TIF_ISOLATE_BP_GUEST) .Lsie_skip: ni __SIE_PROG0C+3(%r14),0xfe # no longer in SIE - lctlg %c1,%c1,__LC_USER_ASCE # load primary asce + lctlg %c1,%c1,__LC_KERNEL_ASCE # load primary asce .Lsie_done: # some program checks are suppressing. C code (e.g. do_protection_exception) # will rewind the PSW by the ILC, which is often 4 bytes in case of SIE. There @@ -380,6 +403,7 @@ ENTRY(system_call) lg %r12,__LC_CURRENT lghi %r14,_PIF_SYSCALL .Lsysc_per: + lctlg %c1,%c1,__LC_KERNEL_ASCE lghi %r13,__TASK_thread lg %r15,__LC_KERNEL_STACK la %r11,STACK_FRAME_OVERHEAD(%r15) # pointer to pt_regs @@ -390,6 +414,7 @@ ENTRY(system_call) mvc __PT_PSW(16,%r11),__LC_SVC_OLD_PSW mvc __PT_INT_CODE(4,%r11),__LC_SVC_ILC stg %r14,__PT_FLAGS(%r11) + xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) ENABLE_INTS .Lsysc_do_svc: # clear user controlled register to prevent speculative use @@ -406,7 +431,6 @@ ENTRY(system_call) jnl .Lsysc_nr_ok slag %r8,%r1,3 .Lsysc_nr_ok: - xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) stg %r2,__PT_ORIG_GPR2(%r11) stg %r7,STACK_FRAME_OVERHEAD(%r15) lg %r9,0(%r8,%r10) # get system call add. @@ -422,15 +446,14 @@ ENTRY(system_call) #endif LOCKDEP_SYS_EXIT .Lsysc_tif: + DISABLE_INTS TSTMSK __PT_FLAGS(%r11),_PIF_WORK jnz .Lsysc_work TSTMSK __TI_flags(%r12),_TIF_WORK jnz .Lsysc_work # check for work - TSTMSK __LC_CPU_FLAGS,(_CIF_WORK-_CIF_FPU) - jnz .Lsysc_work + DEBUG_USER_ASCE + lctlg %c1,%c1,__LC_USER_ASCE BPEXIT __TI_flags(%r12),_TIF_ISOLATE_BP -.Lsysc_restore: - DISABLE_INTS TSTMSK __LC_CPU_FLAGS, _CIF_FPU jz .Lsysc_skip_fpu brasl %r14,load_fpu_regs @@ -444,6 +467,7 @@ ENTRY(system_call) # One of the work bits is on. Find out which one. # .Lsysc_work: + ENABLE_INTS TSTMSK __TI_flags(%r12),_TIF_NEED_RESCHED jo .Lsysc_reschedule TSTMSK __PT_FLAGS(%r11),_PIF_SYSCALL_RESTART @@ -463,12 +487,10 @@ ENTRY(system_call) #endif TSTMSK __PT_FLAGS(%r11),_PIF_SYSCALL_RESTART jo .Lsysc_syscall_restart - TSTMSK __TI_flags(%r12),_TIF_SIGPENDING - jo .Lsysc_sigpending + TSTMSK __TI_flags(%r12),(_TIF_SIGPENDING|_TIF_NOTIFY_SIGNAL) + jnz .Lsysc_sigpending TSTMSK __TI_flags(%r12),_TIF_NOTIFY_RESUME jo .Lsysc_notify_resume - TSTMSK __LC_CPU_FLAGS,(_CIF_ASCE_PRIMARY|_CIF_ASCE_SECONDARY) - jnz .Lsysc_asce j .Lsysc_return # @@ -479,26 +501,6 @@ ENTRY(system_call) jg schedule # -# _CIF_ASCE_PRIMARY and/or _CIF_ASCE_SECONDARY set, load user space asce -# -.Lsysc_asce: - ni __LC_CPU_FLAGS+7,255-_CIF_ASCE_SECONDARY - lctlg %c7,%c7,__LC_VDSO_ASCE # load secondary asce - TSTMSK __LC_CPU_FLAGS,_CIF_ASCE_PRIMARY - jz .Lsysc_return -#ifndef CONFIG_HAVE_MARCH_Z10_FEATURES - tm __LC_STFLE_FAC_LIST+3,0x10 # has MVCOS ? - jnz .Lsysc_set_fs_fixup - ni __LC_CPU_FLAGS+7,255-_CIF_ASCE_PRIMARY - lctlg %c1,%c1,__LC_USER_ASCE # load primary asce - j .Lsysc_return -.Lsysc_set_fs_fixup: -#endif - larl %r14,.Lsysc_return - jg set_fs_fixup - - -# # _TIF_SIGPENDING is set, call do_signal # .Lsysc_sigpending: @@ -634,8 +636,11 @@ ENTRY(pgm_check_handler) 0: lg %r12,__LC_CURRENT lghi %r11,0 lmg %r8,%r9,__LC_PGM_OLD_PSW - tmhh %r8,0x0001 # test problem state bit - jnz 3f # -> fault in user space + tmhh %r8,0x0001 # coming from user space? + jno .Lpgm_skip_asce + lctlg %c1,%c1,__LC_KERNEL_ASCE + j 3f +.Lpgm_skip_asce: #if IS_ENABLED(CONFIG_KVM) # cleanup critical section for program checks in sie64a lgr %r14,%r9 @@ -646,7 +651,7 @@ ENTRY(pgm_check_handler) jhe 1f lg %r14,__SF_SIE_CONTROL(%r15) # get control block pointer ni __SIE_PROG0C+3(%r14),0xfe # no longer in SIE - lctlg %c1,%c1,__LC_USER_ASCE # load primary asce + lctlg %c1,%c1,__LC_KERNEL_ASCE # load primary asce larl %r9,sie_exit # skip forward to sie_exit lghi %r11,_PIF_GUEST_FAULT #endif @@ -694,8 +699,8 @@ ENTRY(pgm_check_handler) mvc __THREAD_per_address(8,%r14),__LC_PER_ADDRESS mvc __THREAD_per_cause(2,%r14),__LC_PER_CODE mvc __THREAD_per_paid(1,%r14),__LC_PER_ACCESS_ID -6: RESTORE_SM_CLEAR_PER - xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) +6: xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) + RESTORE_SM_CLEAR_PER larl %r1,pgm_check_table llgh %r10,__PT_INT_CODE+2(%r11) nill %r10,0x007f @@ -707,17 +712,27 @@ ENTRY(pgm_check_handler) .Lpgm_return: LOCKDEP_SYS_EXIT tm __PT_PSW+1(%r11),0x01 # returning to user ? - jno .Lsysc_restore + jno .Lpgm_restore TSTMSK __PT_FLAGS(%r11),_PIF_SYSCALL jo .Lsysc_do_syscall j .Lsysc_tif +.Lpgm_restore: + DISABLE_INTS + TSTMSK __LC_CPU_FLAGS, _CIF_FPU + jz .Lpgm_skip_fpu + brasl %r14,load_fpu_regs +.Lpgm_skip_fpu: + mvc __LC_RETURN_PSW(16),__PT_PSW(%r11) + stpt __LC_EXIT_TIMER + lmg %r0,%r15,__PT_R0(%r11) + b __LC_RETURN_LPSWE # # PER event in supervisor state, must be kprobes # .Lpgm_kprobe: - RESTORE_SM_CLEAR_PER xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) + RESTORE_SM_CLEAR_PER lgr %r2,%r11 # pass pointer to pt_regs brasl %r14,do_per_trap j .Lpgm_return @@ -743,7 +758,7 @@ ENTRY(io_int_handler) stmg %r8,%r15,__LC_SAVE_AREA_ASYNC lg %r12,__LC_CURRENT lmg %r8,%r9,__LC_IO_OLD_PSW - SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER + SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER,__LC_INT_CLOCK stmg %r0,%r7,__PT_R0(%r11) # clear user controlled registers to prevent speculative use xgr %r0,%r0 @@ -757,17 +772,14 @@ ENTRY(io_int_handler) xgr %r10,%r10 mvc __PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC stmg %r8,%r9,__PT_PSW(%r11) + tm __PT_PSW+1(%r11),0x01 # coming from user space? + jno .Lio_skip_asce + lctlg %c1,%c1,__LC_KERNEL_ASCE +.Lio_skip_asce: mvc __PT_INT_CODE(12,%r11),__LC_SUBCHANNEL_ID xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) - TSTMSK __LC_CPU_FLAGS,_CIF_IGNORE_IRQ - jo .Lio_restore -#if IS_ENABLED(CONFIG_TRACE_IRQFLAGS) - tmhh %r8,0x300 - jz 1f - TRACE_IRQS_OFF -1: -#endif xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) + TRACE_IRQS_OFF .Lio_loop: lgr %r2,%r11 # pass pointer to pt_regs lghi %r3,IO_INTERRUPT @@ -789,15 +801,12 @@ ENTRY(io_int_handler) TSTMSK __LC_CPU_FLAGS,_CIF_WORK jnz .Lio_work .Lio_restore: -#if IS_ENABLED(CONFIG_TRACE_IRQFLAGS) - tm __PT_PSW(%r11),3 - jno 0f TRACE_IRQS_ON -0: -#endif mvc __LC_RETURN_PSW(16),__PT_PSW(%r11) tm __PT_PSW+1(%r11),0x01 # returning to user ? jno .Lio_exit_kernel + DEBUG_USER_ASCE + lctlg %c1,%c1,__LC_USER_ASCE BPEXIT __TI_flags(%r12),_TIF_ISOLATE_BP stpt __LC_EXIT_TIMER .Lio_exit_kernel: @@ -855,38 +864,17 @@ ENTRY(io_int_handler) TSTMSK __TI_flags(%r12),_TIF_PATCH_PENDING jo .Lio_patch_pending #endif - TSTMSK __TI_flags(%r12),_TIF_SIGPENDING - jo .Lio_sigpending + TSTMSK __TI_flags(%r12),(_TIF_SIGPENDING|_TIF_NOTIFY_SIGNAL) + jnz .Lio_sigpending TSTMSK __TI_flags(%r12),_TIF_NOTIFY_RESUME jo .Lio_notify_resume TSTMSK __TI_flags(%r12),_TIF_GUARDED_STORAGE jo .Lio_guarded_storage TSTMSK __LC_CPU_FLAGS,_CIF_FPU jo .Lio_vxrs - TSTMSK __LC_CPU_FLAGS,(_CIF_ASCE_PRIMARY|_CIF_ASCE_SECONDARY) - jnz .Lio_asce j .Lio_return # -# _CIF_ASCE_PRIMARY and/or CIF_ASCE_SECONDARY set, load user space asce -# -.Lio_asce: - ni __LC_CPU_FLAGS+7,255-_CIF_ASCE_SECONDARY - lctlg %c7,%c7,__LC_VDSO_ASCE # load secondary asce - TSTMSK __LC_CPU_FLAGS,_CIF_ASCE_PRIMARY - jz .Lio_return -#ifndef CONFIG_HAVE_MARCH_Z10_FEATURES - tm __LC_STFLE_FAC_LIST+3,0x10 # has MVCOS ? - jnz .Lio_set_fs_fixup - ni __LC_CPU_FLAGS+7,255-_CIF_ASCE_PRIMARY - lctlg %c1,%c1,__LC_USER_ASCE # load primary asce - j .Lio_return -.Lio_set_fs_fixup: -#endif - larl %r14,.Lio_return - jg set_fs_fixup - -# # CIF_FPU is set, restore floating-point controls and floating-point registers. # .Lio_vxrs: @@ -953,7 +941,7 @@ ENTRY(ext_int_handler) stmg %r8,%r15,__LC_SAVE_AREA_ASYNC lg %r12,__LC_CURRENT lmg %r8,%r9,__LC_EXT_OLD_PSW - SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER + SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER,__LC_INT_CLOCK stmg %r0,%r7,__PT_R0(%r11) # clear user controlled registers to prevent speculative use xgr %r0,%r0 @@ -967,20 +955,17 @@ ENTRY(ext_int_handler) xgr %r10,%r10 mvc __PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC stmg %r8,%r9,__PT_PSW(%r11) + tm __PT_PSW+1(%r11),0x01 # coming from user space? + jno .Lext_skip_asce + lctlg %c1,%c1,__LC_KERNEL_ASCE +.Lext_skip_asce: lghi %r1,__LC_EXT_PARAMS2 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) - TSTMSK __LC_CPU_FLAGS,_CIF_IGNORE_IRQ - jo .Lio_restore -#if IS_ENABLED(CONFIG_TRACE_IRQFLAGS) - tmhh %r8,0x300 - jz 1f - TRACE_IRQS_OFF -1: -#endif xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) + TRACE_IRQS_OFF lgr %r2,%r11 # pass pointer to pt_regs lghi %r3,EXT_INTERRUPT brasl %r14,do_IRQ @@ -1066,6 +1051,7 @@ EXPORT_SYMBOL(save_fpu_regs) * %r4 */ load_fpu_regs: + stnsm __SF_EMPTY(%r15),0xfc lg %r4,__LC_CURRENT aghi %r4,__TASK_thread TSTMSK __LC_CPU_FLAGS,_CIF_FPU @@ -1097,6 +1083,7 @@ load_fpu_regs: .Lload_fpu_regs_done: ni __LC_CPU_FLAGS+7,255-_CIF_FPU .Lload_fpu_regs_exit: + ssm __SF_EMPTY(%r15) BR_EX %r14 .Lload_fpu_regs_end: ENDPROC(load_fpu_regs) @@ -1178,7 +1165,7 @@ ENTRY(mcck_int_handler) TSTMSK __LC_MCCK_CODE,MCCK_CODE_PSW_IA_VALID jno .Lmcck_panic 4: ssm __LC_PGM_NEW_PSW # turn dat on, keep irqs off - SWITCH_ASYNC __LC_GPREGS_SAVE_AREA+64,__LC_MCCK_ENTER_TIMER + SWITCH_ASYNC __LC_GPREGS_SAVE_AREA+64,__LC_MCCK_ENTER_TIMER,__LC_MCCK_CLOCK .Lmcck_skip: lghi %r14,__LC_GPREGS_SAVE_AREA+64 stmg %r0,%r7,__PT_R0(%r11) @@ -1194,6 +1181,9 @@ ENTRY(mcck_int_handler) xgr %r10,%r10 mvc __PT_R8(64,%r11),0(%r14) stmg %r8,%r9,__PT_PSW(%r11) + la %r14,4095 + mvc __PT_CR1(8,%r11),__LC_CREGS_SAVE_AREA-4095+8(%r14) + lctlg %c1,%c1,__LC_KERNEL_ASCE xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) lgr %r2,%r11 # pass pointer to pt_regs @@ -1209,6 +1199,7 @@ ENTRY(mcck_int_handler) brasl %r14,s390_handle_mcck TRACE_IRQS_ON .Lmcck_return: + lctlg %c1,%c1,__PT_CR1(%r11) lmg %r0,%r10,__PT_R0(%r11) mvc __LC_RETURN_MCCK_PSW(16),__PT_PSW(%r11) # move return PSW tm __LC_RETURN_MCCK_PSW+1,0x01 # returning to user ? @@ -1285,7 +1276,7 @@ ENDPROC(stack_overflow) 1: BPENTER __SF_SIE_FLAGS(%r15),(_TIF_ISOLATE_BP|_TIF_ISOLATE_BP_GUEST) lg %r9,__SF_SIE_CONTROL(%r15) # get control block pointer ni __SIE_PROG0C+3(%r9),0xfe # no longer in SIE - lctlg %c1,%c1,__LC_USER_ASCE # load primary asce + lctlg %c1,%c1,__LC_KERNEL_ASCE larl %r9,sie_exit # skip forward to sie_exit BR_EX %r14,%r11 diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h index d2ca3fe51f8e..a16c33b32ab0 100644 --- a/arch/s390/kernel/entry.h +++ b/arch/s390/kernel/entry.h @@ -83,7 +83,6 @@ long sys_s390_sthyi(unsigned long function_code, void __user *buffer, u64 __user DECLARE_PER_CPU(u64, mt_cycles[8]); void gs_load_bc_cb(struct pt_regs *regs); -void set_fs_fixup(void); unsigned long stack_alloc(void); void stack_free(unsigned long stack); diff --git a/arch/s390/kernel/ftrace.c b/arch/s390/kernel/ftrace.c index b388e87a08bf..c6ddeb5029b4 100644 --- a/arch/s390/kernel/ftrace.c +++ b/arch/s390/kernel/ftrace.c @@ -22,56 +22,26 @@ #include "entry.h" /* - * The mcount code looks like this: - * stg %r14,8(%r15) # offset 0 - * larl %r1,<&counter> # offset 6 - * brasl %r14,_mcount # offset 12 - * lg %r14,8(%r15) # offset 18 - * Total length is 24 bytes. Only the first instruction will be patched - * by ftrace_make_call / ftrace_make_nop. - * The enabled ftrace code block looks like this: + * To generate function prologue either gcc's hotpatch feature (since gcc 4.8) + * or a combination of -pg -mrecord-mcount -mnop-mcount -mfentry flags + * (since gcc 9 / clang 10) is used. + * In both cases the original and also the disabled function prologue contains + * only a single six byte instruction and looks like this: + * > brcl 0,0 # offset 0 + * To enable ftrace the code gets patched like above and afterwards looks + * like this: * > brasl %r0,ftrace_caller # offset 0 - * larl %r1,<&counter> # offset 6 - * brasl %r14,_mcount # offset 12 - * lg %r14,8(%r15) # offset 18 + * + * The instruction will be patched by ftrace_make_call / ftrace_make_nop. * The ftrace function gets called with a non-standard C function call ABI * where r0 contains the return address. It is also expected that the called * function only clobbers r0 and r1, but restores r2-r15. * For module code we can't directly jump to ftrace caller, but need a * trampoline (ftrace_plt), which clobbers also r1. - * The return point of the ftrace function has offset 24, so execution - * continues behind the mcount block. - * The disabled ftrace code block looks like this: - * > jg .+24 # offset 0 - * larl %r1,<&counter> # offset 6 - * brasl %r14,_mcount # offset 12 - * lg %r14,8(%r15) # offset 18 - * The jg instruction branches to offset 24 to skip as many instructions - * as possible. - * In case we use gcc's hotpatch feature the original and also the disabled - * function prologue contains only a single six byte instruction and looks - * like this: - * > brcl 0,0 # offset 0 - * To enable ftrace the code gets patched like above and afterwards looks - * like this: - * > brasl %r0,ftrace_caller # offset 0 */ unsigned long ftrace_plt; -static inline void ftrace_generate_orig_insn(struct ftrace_insn *insn) -{ -#if defined(CC_USING_HOTPATCH) || defined(CC_USING_NOP_MCOUNT) - /* brcl 0,0 */ - insn->opc = 0xc004; - insn->disp = 0; -#else - /* stg r14,8(r15) */ - insn->opc = 0xe3e0; - insn->disp = 0xf0080024; -#endif -} - int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr, unsigned long addr) { @@ -85,15 +55,10 @@ int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec, if (copy_from_kernel_nofault(&old, (void *) rec->ip, sizeof(old))) return -EFAULT; - if (addr == MCOUNT_ADDR) { - /* Initial code replacement */ - ftrace_generate_orig_insn(&orig); - ftrace_generate_nop_insn(&new); - } else { - /* Replace ftrace call with a nop. */ - ftrace_generate_call_insn(&orig, rec->ip); - ftrace_generate_nop_insn(&new); - } + /* Replace ftrace call with a nop. */ + ftrace_generate_call_insn(&orig, rec->ip); + ftrace_generate_nop_insn(&new); + /* Verify that the to be replaced code matches what we expect. */ if (memcmp(&orig, &old, sizeof(old))) return -EINVAL; @@ -198,17 +163,26 @@ int ftrace_disable_ftrace_graph_caller(void) #ifdef CONFIG_KPROBES_ON_FTRACE void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip, - struct ftrace_ops *ops, struct pt_regs *regs) + struct ftrace_ops *ops, struct ftrace_regs *fregs) { struct kprobe_ctlblk *kcb; - struct kprobe *p = get_kprobe((kprobe_opcode_t *)ip); + struct pt_regs *regs; + struct kprobe *p; + int bit; - if (unlikely(!p) || kprobe_disabled(p)) + bit = ftrace_test_recursion_trylock(ip, parent_ip); + if (bit < 0) return; + regs = ftrace_get_regs(fregs); + preempt_disable_notrace(); + p = get_kprobe((kprobe_opcode_t *)ip); + if (unlikely(!p) || kprobe_disabled(p)) + goto out; + if (kprobe_running()) { kprobes_inc_nmissed_count(p); - return; + goto out; } __this_cpu_write(current_kprobe, p); @@ -228,6 +202,9 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip, } } __this_cpu_write(current_kprobe, NULL); +out: + preempt_enable_notrace(); + ftrace_test_recursion_unlock(bit); } NOKPROBE_SYMBOL(kprobe_ftrace_handler); diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index 8b88dbbda7df..0c253886da78 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -18,12 +18,7 @@ __HEAD ENTRY(startup_continue) - tm __LC_STFLE_FAC_LIST+5,0x80 # LPP available ? - jz 0f - xc __LC_LPP+1(7,0),__LC_LPP+1 # clear lpp and current_pid - mvi __LC_LPP,0x80 # and set LPP_MAGIC - .insn s,0xb2800000,__LC_LPP # load program parameter -0: larl %r1,tod_clock_base + larl %r1,tod_clock_base mvc 0(16,%r1),__LC_BOOT_CLOCK larl %r13,.LPG1 # get base # diff --git a/arch/s390/kernel/idle.c b/arch/s390/kernel/idle.c index f7f1e64e0d98..a5d4d80d6ede 100644 --- a/arch/s390/kernel/idle.c +++ b/arch/s390/kernel/idle.c @@ -9,7 +9,6 @@ #include <linux/kernel.h> #include <linux/kernel_stat.h> -#include <linux/kprobes.h> #include <linux/notifier.h> #include <linux/init.h> #include <linux/cpu.h> @@ -21,22 +20,19 @@ static DEFINE_PER_CPU(struct s390_idle_data, s390_idle); -void enabled_wait(void) +void arch_cpu_idle(void) { struct s390_idle_data *idle = this_cpu_ptr(&s390_idle); unsigned long long idle_time; - unsigned long psw_mask, flags; - + unsigned long psw_mask; /* Wait for external, I/O or machine check interrupt. */ psw_mask = PSW_KERNEL_BITS | PSW_MASK_WAIT | PSW_MASK_DAT | PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK; clear_cpu_flag(CIF_NOHZ_DELAY); - local_irq_save(flags); - /* Call the assembler magic in entry.S */ + /* psw_idle() returns with interrupts disabled. */ psw_idle(idle, psw_mask); - local_irq_restore(flags); /* Account time spent with enabled wait psw loaded as idle time. */ raw_write_seqcount_begin(&idle->seqcount); @@ -46,8 +42,8 @@ void enabled_wait(void) idle->idle_count++; account_idle_time(cputime_to_nsecs(idle_time)); raw_write_seqcount_end(&idle->seqcount); + raw_local_irq_enable(); } -NOKPROBE_SYMBOL(enabled_wait); static ssize_t show_idle_count(struct device *dev, struct device_attribute *attr, char *buf) @@ -120,12 +116,6 @@ void arch_cpu_idle_enter(void) { } -void arch_cpu_idle(void) -{ - enabled_wait(); - local_irq_enable(); -} - void arch_cpu_idle_exit(void) { } diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 98b3aca1de8e..7a21eca498aa 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -1512,7 +1512,7 @@ static void diag308_dump(void *dump_block) while (1) { if (diag308(DIAG308_LOAD_NORMAL_DUMP, NULL) != 0x302) break; - udelay_simple(USEC_PER_SEC); + udelay(USEC_PER_SEC); } } diff --git a/arch/s390/kernel/mcount.S b/arch/s390/kernel/mcount.S index 7458dcfd6464..faf64c2f90f5 100644 --- a/arch/s390/kernel/mcount.S +++ b/arch/s390/kernel/mcount.S @@ -33,11 +33,6 @@ ENDPROC(ftrace_stub) #define TRACED_FUNC_FRAME_SIZE STACK_FRAME_OVERHEAD #endif -ENTRY(_mcount) - BR_EX %r14 -ENDPROC(_mcount) -EXPORT_SYMBOL(_mcount) - ENTRY(ftrace_caller) .globl ftrace_regs_caller .set ftrace_regs_caller,ftrace_caller @@ -46,9 +41,6 @@ ENTRY(ftrace_caller) ipm %r14 # don't put any instructions sllg %r14,%r14,16 # clobbering CC before this point lgr %r1,%r15 -#if !(defined(CC_USING_HOTPATCH) || defined(CC_USING_NOP_MCOUNT)) - aghi %r0,MCOUNT_RETURN_FIXUP -#endif # allocate stack frame for ftrace_caller to contain traced function aghi %r15,-TRACED_FUNC_FRAME_SIZE stg %r1,__SF_BACKCHAIN(%r15) diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c index 4f9e4626df55..19cd7b961c45 100644 --- a/arch/s390/kernel/perf_cpum_sf.c +++ b/arch/s390/kernel/perf_cpum_sf.c @@ -672,7 +672,7 @@ static void cpumsf_output_event_pid(struct perf_event *event, rcu_read_lock(); perf_prepare_sample(&header, data, event, regs); - if (perf_output_begin(&handle, event, header.size)) + if (perf_output_begin(&handle, data, event, header.size)) goto out; /* Update the process ID (see also kernel/events/core.c) */ @@ -2228,4 +2228,4 @@ out: } arch_initcall(init_cpum_sampling_pmu); -core_param(cpum_sfb_size, CPUM_SF_MAX_SDB, sfb_size, 0640); +core_param(cpum_sfb_size, CPUM_SF_MAX_SDB, sfb_size, 0644); diff --git a/arch/s390/kernel/perf_regs.c b/arch/s390/kernel/perf_regs.c index 4352a504f235..6e9e5d5e927e 100644 --- a/arch/s390/kernel/perf_regs.c +++ b/arch/s390/kernel/perf_regs.c @@ -53,8 +53,7 @@ u64 perf_reg_abi(struct task_struct *task) } void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { /* * Use the regs from the first interruption and let diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index ec801d3bbb37..bc3ca54edfb4 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -94,7 +94,6 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp, /* Save access registers to new thread structure. */ save_access_regs(&p->thread.acrs[0]); /* start new process with ar4 pointing to the correct address space */ - p->thread.mm_segment = get_fs(); /* Don't copy debug registers */ memset(&p->thread.per_user, 0, sizeof(p->thread.per_user)); memset(&p->thread.per_event, 0, sizeof(p->thread.per_event)); @@ -208,16 +207,3 @@ unsigned long arch_randomize_brk(struct mm_struct *mm) ret = PAGE_ALIGN(mm->brk + brk_rnd()); return (ret > mm->brk) ? ret : mm->brk; } - -void set_fs_fixup(void) -{ - struct pt_regs *regs = current_pt_regs(); - static bool warned; - - set_fs(USER_DS); - if (warned) - return; - WARN(1, "Unbalanced set_fs - int code: 0x%x\n", regs->int_code); - show_registers(regs); - warned = true; -} diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 4d843e64496f..1fbed91c73bc 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -49,6 +49,7 @@ #include <linux/memory.h> #include <linux/compat.h> #include <linux/start_kernel.h> +#include <linux/hugetlb.h> #include <asm/boot_data.h> #include <asm/ipl.h> @@ -94,10 +95,8 @@ char elf_platform[ELF_PLATFORM_SIZE]; unsigned long int_hwcap = 0; int __bootdata(noexec_disabled); -int __bootdata(memory_end_set); -unsigned long __bootdata(memory_end); +unsigned long __bootdata(ident_map_size); unsigned long __bootdata(vmalloc_size); -unsigned long __bootdata(max_physmem_end); struct mem_detect_info __bootdata(mem_detect); struct exception_table_entry *__bootdata_preserved(__start_dma_ex_table); @@ -556,24 +555,25 @@ static void __init setup_resources(void) #endif } -static void __init setup_memory_end(void) +static void __init setup_ident_map_size(void) { unsigned long vmax, tmp; /* Choose kernel address space layout: 3 or 4 levels. */ - tmp = (memory_end ?: max_physmem_end) / PAGE_SIZE; + tmp = ident_map_size / PAGE_SIZE; tmp = tmp * (sizeof(struct page) + PAGE_SIZE); if (tmp + vmalloc_size + MODULES_LEN <= _REGION2_SIZE) vmax = _REGION2_SIZE; /* 3-level kernel page table */ else vmax = _REGION1_SIZE; /* 4-level kernel page table */ + /* module area is at the end of the kernel address space. */ + MODULES_END = vmax; if (is_prot_virt_host()) - adjust_to_uv_max(&vmax); + adjust_to_uv_max(&MODULES_END); #ifdef CONFIG_KASAN - vmax = kasan_vmax; + vmax = _REGION1_SIZE; + MODULES_END = kasan_vmax; #endif - /* module area is at the end of the kernel address space. */ - MODULES_END = vmax; MODULES_VADDR = MODULES_END - MODULES_LEN; VMALLOC_END = MODULES_VADDR; VMALLOC_START = VMALLOC_END - vmalloc_size; @@ -587,22 +587,22 @@ static void __init setup_memory_end(void) tmp = min(tmp, 1UL << MAX_PHYSMEM_BITS); vmemmap = (struct page *) tmp; - /* Take care that memory_end is set and <= vmemmap */ - memory_end = min(memory_end ?: max_physmem_end, (unsigned long)vmemmap); + /* Take care that ident_map_size <= vmemmap */ + ident_map_size = min(ident_map_size, (unsigned long)vmemmap); #ifdef CONFIG_KASAN - memory_end = min(memory_end, KASAN_SHADOW_START); + ident_map_size = min(ident_map_size, KASAN_SHADOW_START); #endif - vmemmap_size = SECTION_ALIGN_UP(memory_end / PAGE_SIZE) * sizeof(struct page); + vmemmap_size = SECTION_ALIGN_UP(ident_map_size / PAGE_SIZE) * sizeof(struct page); #ifdef CONFIG_KASAN /* move vmemmap above kasan shadow only if stands in a way */ if (KASAN_SHADOW_END > (unsigned long)vmemmap && (unsigned long)vmemmap + vmemmap_size > KASAN_SHADOW_START) vmemmap = max(vmemmap, (struct page *)KASAN_SHADOW_END); #endif - max_pfn = max_low_pfn = PFN_DOWN(memory_end); - memblock_remove(memory_end, ULONG_MAX); + max_pfn = max_low_pfn = PFN_DOWN(ident_map_size); + memblock_remove(ident_map_size, ULONG_MAX); - pr_notice("The maximum memory size is %luMB\n", memory_end >> 20); + pr_notice("The maximum memory size is %luMB\n", ident_map_size >> 20); } #ifdef CONFIG_CRASH_DUMP @@ -632,12 +632,11 @@ static struct notifier_block kdump_mem_nb = { #endif /* - * Make sure that the area behind memory_end is protected + * Make sure that the area above identity mapping is protected */ -static void __init reserve_memory_end(void) +static void __init reserve_above_ident_map(void) { - if (memory_end_set) - memblock_reserve(memory_end, ULONG_MAX); + memblock_reserve(ident_map_size, ULONG_MAX); } /* @@ -674,7 +673,7 @@ static void __init reserve_crashkernel(void) phys_addr_t low, high; int rc; - rc = parse_crashkernel(boot_command_line, memory_end, &crash_size, + rc = parse_crashkernel(boot_command_line, ident_map_size, &crash_size, &crash_base); crash_base = ALIGN(crash_base, KEXEC_CRASH_MEM_ALIGN); @@ -1128,7 +1127,7 @@ void __init setup_arch(char **cmdline_p) setup_control_program_code(); /* Do some memory reservations *before* memory is added to memblock */ - reserve_memory_end(); + reserve_above_ident_map(); reserve_oldmem(); reserve_kernel(); reserve_initrd(); @@ -1143,10 +1142,12 @@ void __init setup_arch(char **cmdline_p) remove_oldmem(); setup_uv(); - setup_memory_end(); + setup_ident_map_size(); setup_memory(); - dma_contiguous_reserve(memory_end); + dma_contiguous_reserve(ident_map_size); vmcp_cma_reserve(); + if (MACHINE_HAS_EDAT2) + hugetlb_cma_reserve(PUD_SHIFT - PAGE_SHIFT); check_initrd(); reserve_crashkernel(); diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c index 9e900a8977bd..b27b6c1f058d 100644 --- a/arch/s390/kernel/signal.c +++ b/arch/s390/kernel/signal.c @@ -472,7 +472,7 @@ void do_signal(struct pt_regs *regs) current->thread.system_call = test_pt_regs_flag(regs, PIF_SYSCALL) ? regs->int_code : 0; - if (get_signal(&ksig)) { + if (test_thread_flag(TIF_SIGPENDING) && get_signal(&ksig)) { /* Whee! Actually deliver the signal. */ if (current->thread.system_call) { regs->int_code = current->thread.system_call; diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index ebfe86d097f0..27c763014114 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -47,7 +47,6 @@ #include <asm/vtimer.h> #include <asm/lowcore.h> #include <asm/sclp.h> -#include <asm/vdso.h> #include <asm/debug.h> #include <asm/os_info.h> #include <asm/sigp.h> @@ -55,6 +54,7 @@ #include <asm/nmi.h> #include <asm/stacktrace.h> #include <asm/topology.h> +#include <asm/vdso.h> #include "entry.h" enum { @@ -217,14 +217,10 @@ static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu) lc->return_mcck_lpswe = gen_lpswe(__LC_RETURN_MCCK_PSW); if (nmi_alloc_per_cpu(lc)) goto out_async; - if (vdso_alloc_per_cpu(lc)) - goto out_mcesa; lowcore_ptr[cpu] = lc; pcpu_sigp_retry(pcpu, SIGP_SET_PREFIX, (u32)(unsigned long) lc); return 0; -out_mcesa: - nmi_free_per_cpu(lc); out_async: stack_free(async_stack); out: @@ -245,7 +241,6 @@ static void pcpu_free_lowcore(struct pcpu *pcpu) pcpu_sigp_retry(pcpu, SIGP_SET_PREFIX, 0); lowcore_ptr[pcpu - pcpu_devices] = NULL; - vdso_free_per_cpu(pcpu->lowcore); nmi_free_per_cpu(pcpu->lowcore); stack_free(async_stack); if (pcpu == &pcpu_devices[0]) @@ -265,13 +260,13 @@ static void pcpu_prepare_secondary(struct pcpu *pcpu, int cpu) lc->spinlock_index = 0; lc->percpu_offset = __per_cpu_offset[cpu]; lc->kernel_asce = S390_lowcore.kernel_asce; - lc->user_asce = S390_lowcore.kernel_asce; + lc->user_asce = s390_invalid_asce; lc->machine_flags = S390_lowcore.machine_flags; lc->user_timer = lc->system_timer = lc->steal_timer = lc->avg_steal_timer = 0; __ctl_store(lc->cregs_save_area, 0, 15); lc->cregs_save_area[1] = lc->kernel_asce; - lc->cregs_save_area[7] = lc->vdso_asce; + lc->cregs_save_area[7] = lc->user_asce; save_access_regs((unsigned int *) lc->access_regs_save_area); memcpy(lc->stfle_fac_list, S390_lowcore.stfle_fac_list, sizeof(lc->stfle_fac_list)); @@ -855,16 +850,16 @@ void __init smp_detect_cpus(void) static void smp_init_secondary(void) { - int cpu = smp_processor_id(); + int cpu = raw_smp_processor_id(); S390_lowcore.last_update_clock = get_tod_clock(); restore_access_regs(S390_lowcore.access_regs_save_area); - set_cpu_flag(CIF_ASCE_PRIMARY); - set_cpu_flag(CIF_ASCE_SECONDARY); cpu_init(); + rcu_cpu_starting(cpu); preempt_disable(); init_cpu_timer(); vtime_init(); + vdso_getcpu_init(); pfault_init(); notify_cpu_starting(cpu); if (topology_cpu_dedicated(cpu)) @@ -895,24 +890,12 @@ static void __no_sanitize_address smp_start_secondary(void *cpuvoid) /* Upping and downing of CPUs */ int __cpu_up(unsigned int cpu, struct task_struct *tidle) { - struct pcpu *pcpu; - int base, i, rc; + struct pcpu *pcpu = pcpu_devices + cpu; + int rc; - pcpu = pcpu_devices + cpu; if (pcpu->state != CPU_STATE_CONFIGURED) return -EIO; - base = smp_get_base_cpu(cpu); - for (i = 0; i <= smp_cpu_mtid; i++) { - if (base + i < nr_cpu_ids) - if (cpu_online(base + i)) - break; - } - /* - * If this is the first CPU of the core to get online - * do an initial CPU reset. - */ - if (i > smp_cpu_mtid && - pcpu_sigp_retry(pcpu_devices + base, SIGP_INITIAL_CPU_RESET, 0) != + if (pcpu_sigp_retry(pcpu, SIGP_INITIAL_CPU_RESET, 0) != SIGP_CC_ORDER_CODE_ACCEPTED) return -EIO; diff --git a/arch/s390/kernel/syscalls/syscall.tbl b/arch/s390/kernel/syscalls/syscall.tbl index 28c168000483..d443423495e5 100644 --- a/arch/s390/kernel/syscalls/syscall.tbl +++ b/arch/s390/kernel/syscalls/syscall.tbl @@ -443,3 +443,4 @@ 438 common pidfd_getfd sys_pidfd_getfd sys_pidfd_getfd 439 common faccessat2 sys_faccessat2 sys_faccessat2 440 common process_madvise sys_process_madvise sys_process_madvise +441 common epoll_pwait2 sys_epoll_pwait2 compat_sys_epoll_pwait2 diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 0ac30ee2c633..c59cb44fbb7d 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -927,41 +927,25 @@ static ssize_t online_store(struct device *dev, */ static DEVICE_ATTR_RW(online); -static struct device_attribute *stp_attributes[] = { - &dev_attr_ctn_id, - &dev_attr_ctn_type, - &dev_attr_dst_offset, - &dev_attr_leap_seconds, - &dev_attr_online, - &dev_attr_leap_seconds_scheduled, - &dev_attr_stratum, - &dev_attr_time_offset, - &dev_attr_time_zone_offset, - &dev_attr_timing_mode, - &dev_attr_timing_state, +static struct attribute *stp_dev_attrs[] = { + &dev_attr_ctn_id.attr, + &dev_attr_ctn_type.attr, + &dev_attr_dst_offset.attr, + &dev_attr_leap_seconds.attr, + &dev_attr_online.attr, + &dev_attr_leap_seconds_scheduled.attr, + &dev_attr_stratum.attr, + &dev_attr_time_offset.attr, + &dev_attr_time_zone_offset.attr, + &dev_attr_timing_mode.attr, + &dev_attr_timing_state.attr, NULL }; +ATTRIBUTE_GROUPS(stp_dev); static int __init stp_init_sysfs(void) { - struct device_attribute **attr; - int rc; - - rc = subsys_system_register(&stp_subsys, NULL); - if (rc) - goto out; - for (attr = stp_attributes; *attr; attr++) { - rc = device_create_file(stp_subsys.dev_root, *attr); - if (rc) - goto out_unreg; - } - return 0; -out_unreg: - for (; attr >= stp_attributes; attr--) - device_remove_file(stp_subsys.dev_root, *attr); - bus_unregister(&stp_subsys); -out: - return rc; + return subsys_system_register(&stp_subsys, stp_dev_groups); } device_initcall(stp_init_sysfs); diff --git a/arch/s390/kernel/uv.c b/arch/s390/kernel/uv.c index 14bd9d58edc9..883bfed9f5c2 100644 --- a/arch/s390/kernel/uv.c +++ b/arch/s390/kernel/uv.c @@ -129,8 +129,15 @@ int uv_destroy_page(unsigned long paddr) .paddr = paddr }; - if (uv_call(0, (u64)&uvcb)) + if (uv_call(0, (u64)&uvcb)) { + /* + * Older firmware uses 107/d as an indication of a non secure + * page. Let us emulate the newer variant (no-op). + */ + if (uvcb.header.rc == 0x107 && uvcb.header.rrc == 0xd) + return 0; return -EINVAL; + } return 0; } diff --git a/arch/s390/kernel/vdso.c b/arch/s390/kernel/vdso.c index f9da5b149141..8bc269c55fd3 100644 --- a/arch/s390/kernel/vdso.c +++ b/arch/s390/kernel/vdso.c @@ -29,6 +29,7 @@ #include <asm/sections.h> #include <asm/vdso.h> #include <asm/facility.h> +#include <asm/timex.h> extern char vdso64_start, vdso64_end; static void *vdso64_kbase = &vdso64_start; @@ -61,17 +62,8 @@ static vm_fault_t vdso_fault(const struct vm_special_mapping *sm, static int vdso_mremap(const struct vm_special_mapping *sm, struct vm_area_struct *vma) { - unsigned long vdso_pages; - - vdso_pages = vdso64_pages; - - if ((vdso_pages << PAGE_SHIFT) != vma->vm_end - vma->vm_start) - return -EINVAL; - - if (WARN_ON_ONCE(current->mm != vma->vm_mm)) - return -EFAULT; - current->mm->context.vdso_base = vma->vm_start; + return 0; } @@ -99,60 +91,10 @@ static union { u8 page[PAGE_SIZE]; } vdso_data_store __page_aligned_data; struct vdso_data *vdso_data = (struct vdso_data *)&vdso_data_store.data; -/* - * Allocate/free per cpu vdso data. - */ -#define SEGMENT_ORDER 2 - -int vdso_alloc_per_cpu(struct lowcore *lowcore) -{ - unsigned long segment_table, page_table, page_frame; - struct vdso_per_cpu_data *vd; - segment_table = __get_free_pages(GFP_KERNEL, SEGMENT_ORDER); - page_table = get_zeroed_page(GFP_KERNEL); - page_frame = get_zeroed_page(GFP_KERNEL); - if (!segment_table || !page_table || !page_frame) - goto out; - arch_set_page_dat(virt_to_page(segment_table), SEGMENT_ORDER); - arch_set_page_dat(virt_to_page(page_table), 0); - - /* Initialize per-cpu vdso data page */ - vd = (struct vdso_per_cpu_data *) page_frame; - vd->cpu_nr = lowcore->cpu_nr; - vd->node_id = cpu_to_node(vd->cpu_nr); - - /* Set up page table for the vdso address space */ - memset64((u64 *)segment_table, _SEGMENT_ENTRY_EMPTY, _CRST_ENTRIES); - memset64((u64 *)page_table, _PAGE_INVALID, PTRS_PER_PTE); - - *(unsigned long *) segment_table = _SEGMENT_ENTRY + page_table; - *(unsigned long *) page_table = _PAGE_PROTECT + page_frame; - - lowcore->vdso_asce = segment_table + - _ASCE_TABLE_LENGTH + _ASCE_USER_BITS + _ASCE_TYPE_SEGMENT; - lowcore->vdso_per_cpu_data = page_frame; - - return 0; - -out: - free_page(page_frame); - free_page(page_table); - free_pages(segment_table, SEGMENT_ORDER); - return -ENOMEM; -} - -void vdso_free_per_cpu(struct lowcore *lowcore) +void vdso_getcpu_init(void) { - unsigned long segment_table, page_table, page_frame; - - segment_table = lowcore->vdso_asce & PAGE_MASK; - page_table = *(unsigned long *) segment_table; - page_frame = *(unsigned long *) page_table; - - free_page(page_frame); - free_page(page_table); - free_pages(segment_table, SEGMENT_ORDER); + set_tod_programmable_field(smp_processor_id()); } /* @@ -225,6 +167,7 @@ static int __init vdso_init(void) { int i; + vdso_getcpu_init(); /* Calculate the size of the 64 bit vDSO */ vdso64_pages = ((&vdso64_end - &vdso64_start + PAGE_SIZE - 1) >> PAGE_SHIFT) + 1; @@ -240,8 +183,6 @@ static int __init vdso_init(void) } vdso64_pagelist[vdso64_pages - 1] = virt_to_page(vdso_data); vdso64_pagelist[vdso64_pages] = NULL; - if (vdso_alloc_per_cpu(&S390_lowcore)) - BUG(); get_page(virt_to_page(vdso_data)); diff --git a/arch/s390/kernel/vdso64/Makefile b/arch/s390/kernel/vdso64/Makefile index 13cc5a3f9abf..a6e0fb6b91d6 100644 --- a/arch/s390/kernel/vdso64/Makefile +++ b/arch/s390/kernel/vdso64/Makefile @@ -6,8 +6,9 @@ ARCH_REL_TYPE_ABS := R_390_COPY|R_390_GLOB_DAT|R_390_JMP_SLOT|R_390_RELATIVE ARCH_REL_TYPE_ABS += R_390_GOT|R_390_PLT include $(srctree)/lib/vdso/Makefile -obj-vdso64 = vdso_user_wrapper.o note.o getcpu.o -obj-cvdso64 = vdso64_generic.o +obj-vdso64 = vdso_user_wrapper.o note.o +obj-cvdso64 = vdso64_generic.o getcpu.o +CFLAGS_REMOVE_getcpu.o = -pg $(CC_FLAGS_FTRACE) $(CC_FLAGS_EXPOLINE) CFLAGS_REMOVE_vdso64_generic.o = -pg $(CC_FLAGS_FTRACE) $(CC_FLAGS_EXPOLINE) # Build rules diff --git a/arch/s390/kernel/vdso64/getcpu.S b/arch/s390/kernel/vdso64/getcpu.S deleted file mode 100644 index 3c04f7328500..000000000000 --- a/arch/s390/kernel/vdso64/getcpu.S +++ /dev/null @@ -1,31 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -/* - * Userland implementation of getcpu() for 64 bits processes in a - * s390 kernel for use in the vDSO - * - * Copyright IBM Corp. 2016 - * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> - */ -#include <asm/vdso.h> -#include <asm/asm-offsets.h> -#include <asm/dwarf.h> - - .text - .align 4 - .globl __kernel_getcpu - .type __kernel_getcpu,@function -__kernel_getcpu: - CFI_STARTPROC - sacf 256 - lm %r4,%r5,__VDSO_GETCPU_VAL(%r0) - sacf 0 - ltgr %r2,%r2 - jz 2f - st %r5,0(%r2) -2: ltgr %r3,%r3 - jz 3f - st %r4,0(%r3) -3: lghi %r2,0 - br %r14 - CFI_ENDPROC - .size __kernel_getcpu,.-__kernel_getcpu diff --git a/arch/s390/kernel/vdso64/getcpu.c b/arch/s390/kernel/vdso64/getcpu.c new file mode 100644 index 000000000000..5b2bc7494d5b --- /dev/null +++ b/arch/s390/kernel/vdso64/getcpu.c @@ -0,0 +1,21 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright IBM Corp. 2020 */ + +#include <linux/compiler.h> +#include <linux/getcpu.h> +#include <asm/timex.h> +#include "vdso.h" + +int __s390_vdso_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *unused) +{ + __u16 todval[8]; + + /* CPU number is stored in the programmable field of the TOD clock */ + get_tod_clock_ext((char *)todval); + if (cpu) + *cpu = todval[7]; + /* NUMA node is always zero */ + if (node) + *node = 0; + return 0; +} diff --git a/arch/s390/kernel/vdso64/vdso.h b/arch/s390/kernel/vdso64/vdso.h new file mode 100644 index 000000000000..34c7a2312f9d --- /dev/null +++ b/arch/s390/kernel/vdso64/vdso.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __ARCH_S390_KERNEL_VDSO64_VDSO_H +#define __ARCH_S390_KERNEL_VDSO64_VDSO_H + +#include <vdso/datapage.h> + +struct getcpu_cache; + +int __s390_vdso_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *unused); +int __s390_vdso_gettimeofday(struct __kernel_old_timeval *tv, struct timezone *tz); +int __s390_vdso_clock_gettime(clockid_t clock, struct __kernel_timespec *ts); +int __s390_vdso_clock_getres(clockid_t clock, struct __kernel_timespec *ts); + +#endif /* __ARCH_S390_KERNEL_VDSO64_VDSO_H */ diff --git a/arch/s390/kernel/vdso64/vdso64.lds.S b/arch/s390/kernel/vdso64/vdso64.lds.S index 7ddb116b5e2e..7bde3909290f 100644 --- a/arch/s390/kernel/vdso64/vdso64.lds.S +++ b/arch/s390/kernel/vdso64/vdso64.lds.S @@ -136,7 +136,6 @@ VERSION __kernel_clock_gettime; __kernel_clock_getres; __kernel_getcpu; - local: *; }; } diff --git a/arch/s390/kernel/vdso64/vdso64_generic.c b/arch/s390/kernel/vdso64/vdso64_generic.c index a8cef7e4d137..a9aa75643c08 100644 --- a/arch/s390/kernel/vdso64/vdso64_generic.c +++ b/arch/s390/kernel/vdso64/vdso64_generic.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 #include "../../../../lib/vdso/gettimeofday.c" +#include "vdso.h" int __s390_vdso_gettimeofday(struct __kernel_old_timeval *tv, struct timezone *tz) diff --git a/arch/s390/kernel/vdso64/vdso_user_wrapper.S b/arch/s390/kernel/vdso64/vdso_user_wrapper.S index a775d7e52872..f773505c7e63 100644 --- a/arch/s390/kernel/vdso64/vdso_user_wrapper.S +++ b/arch/s390/kernel/vdso64/vdso_user_wrapper.S @@ -36,3 +36,4 @@ __kernel_\func: vdso_func gettimeofday vdso_func clock_getres vdso_func clock_gettime +vdso_func getcpu diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index 177ccfbda40a..4c0e19145cc6 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -10,7 +10,8 @@ * Put .bss..swapper_pg_dir as the first thing in .bss. This will * make sure it has 16k alignment. */ -#define BSS_FIRST_SECTIONS *(.bss..swapper_pg_dir) +#define BSS_FIRST_SECTIONS *(.bss..swapper_pg_dir) \ + *(.bss..invalid_pg_dir) /* Handle ro_after_init data on our own. */ #define RO_AFTER_INIT_DATA diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c index 8df10d3c8f6c..5aaa2ca6a928 100644 --- a/arch/s390/kernel/vtime.c +++ b/arch/s390/kernel/vtime.c @@ -222,35 +222,50 @@ void vtime_flush(struct task_struct *tsk) S390_lowcore.avg_steal_timer = avg_steal; } +static u64 vtime_delta(void) +{ + u64 timer = S390_lowcore.last_update_timer; + + S390_lowcore.last_update_timer = get_vtimer(); + + return timer - S390_lowcore.last_update_timer; +} + /* * Update process times based on virtual cpu times stored by entry.S * to the lowcore fields user_timer, system_timer & steal_clock. */ -void vtime_account_irq_enter(struct task_struct *tsk) +void vtime_account_kernel(struct task_struct *tsk) { - u64 timer; - - timer = S390_lowcore.last_update_timer; - S390_lowcore.last_update_timer = get_vtimer(); - timer -= S390_lowcore.last_update_timer; + u64 delta = vtime_delta(); - if ((tsk->flags & PF_VCPU) && (irq_count() == 0)) - S390_lowcore.guest_timer += timer; - else if (hardirq_count()) - S390_lowcore.hardirq_timer += timer; - else if (in_serving_softirq()) - S390_lowcore.softirq_timer += timer; + if (tsk->flags & PF_VCPU) + S390_lowcore.guest_timer += delta; else - S390_lowcore.system_timer += timer; + S390_lowcore.system_timer += delta; - virt_timer_forward(timer); + virt_timer_forward(delta); } -EXPORT_SYMBOL_GPL(vtime_account_irq_enter); - -void vtime_account_kernel(struct task_struct *tsk) -__attribute__((alias("vtime_account_irq_enter"))); EXPORT_SYMBOL_GPL(vtime_account_kernel); +void vtime_account_softirq(struct task_struct *tsk) +{ + u64 delta = vtime_delta(); + + S390_lowcore.softirq_timer += delta; + + virt_timer_forward(delta); +} + +void vtime_account_hardirq(struct task_struct *tsk) +{ + u64 delta = vtime_delta(); + + S390_lowcore.hardirq_timer += delta; + + virt_timer_forward(delta); +} + /* * Sorted add to a list. List is linear searched until first bigger * element is found. |