diff options
Diffstat (limited to 'arch/s390/kernel')
32 files changed, 487 insertions, 640 deletions
diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c index b376f0377a2c..a85e0c3e7027 100644 --- a/arch/s390/kernel/debug.c +++ b/arch/s390/kernel/debug.c @@ -60,7 +60,7 @@ typedef struct { * except of floats, and long long (32 bit) * */ - long args[0]; + long args[]; } debug_sprintf_entry_t; /* internal function prototyes */ @@ -981,16 +981,6 @@ static struct ctl_table s390dbf_table[] = { { } }; -static struct ctl_table s390dbf_dir_table[] = { - { - .procname = "s390dbf", - .maxlen = 0, - .mode = S_IRUGO | S_IXUGO, - .child = s390dbf_table, - }, - { } -}; - static struct ctl_table_header *s390dbf_sysctl_header; /** @@ -1574,7 +1564,7 @@ out: */ static int __init debug_init(void) { - s390dbf_sysctl_header = register_sysctl_table(s390dbf_dir_table); + s390dbf_sysctl_header = register_sysctl("s390dbf", s390dbf_table); mutex_lock(&debug_mutex); debug_debugfs_root_entry = debugfs_create_dir(DEBUG_DIR_ROOT, NULL); initialized = 1; diff --git a/arch/s390/kernel/dumpstack.c b/arch/s390/kernel/dumpstack.c index 1e3233eb510a..d2012635b093 100644 --- a/arch/s390/kernel/dumpstack.c +++ b/arch/s390/kernel/dumpstack.c @@ -41,60 +41,50 @@ const char *stack_type_name(enum stack_type type) EXPORT_SYMBOL_GPL(stack_type_name); static inline bool in_stack(unsigned long sp, struct stack_info *info, - enum stack_type type, unsigned long low, - unsigned long high) + enum stack_type type, unsigned long stack) { - if (sp < low || sp >= high) + if (sp < stack || sp >= stack + THREAD_SIZE) return false; info->type = type; - info->begin = low; - info->end = high; + info->begin = stack; + info->end = stack + THREAD_SIZE; return true; } static bool in_task_stack(unsigned long sp, struct task_struct *task, struct stack_info *info) { - unsigned long stack; + unsigned long stack = (unsigned long)task_stack_page(task); - stack = (unsigned long) task_stack_page(task); - return in_stack(sp, info, STACK_TYPE_TASK, stack, stack + THREAD_SIZE); + return in_stack(sp, info, STACK_TYPE_TASK, stack); } static bool in_irq_stack(unsigned long sp, struct stack_info *info) { - unsigned long frame_size, top; + unsigned long stack = S390_lowcore.async_stack - STACK_INIT_OFFSET; - frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs); - top = S390_lowcore.async_stack + frame_size; - return in_stack(sp, info, STACK_TYPE_IRQ, top - THREAD_SIZE, top); + return in_stack(sp, info, STACK_TYPE_IRQ, stack); } static bool in_nodat_stack(unsigned long sp, struct stack_info *info) { - unsigned long frame_size, top; + unsigned long stack = S390_lowcore.nodat_stack - STACK_INIT_OFFSET; - frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs); - top = S390_lowcore.nodat_stack + frame_size; - return in_stack(sp, info, STACK_TYPE_NODAT, top - THREAD_SIZE, top); + return in_stack(sp, info, STACK_TYPE_NODAT, stack); } static bool in_mcck_stack(unsigned long sp, struct stack_info *info) { - unsigned long frame_size, top; + unsigned long stack = S390_lowcore.mcck_stack - STACK_INIT_OFFSET; - frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs); - top = S390_lowcore.mcck_stack + frame_size; - return in_stack(sp, info, STACK_TYPE_MCCK, top - THREAD_SIZE, top); + return in_stack(sp, info, STACK_TYPE_MCCK, stack); } static bool in_restart_stack(unsigned long sp, struct stack_info *info) { - unsigned long frame_size, top; + unsigned long stack = S390_lowcore.restart_stack - STACK_INIT_OFFSET; - frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs); - top = S390_lowcore.restart_stack + frame_size; - return in_stack(sp, info, STACK_TYPE_RESTART, top - THREAD_SIZE, top); + return in_stack(sp, info, STACK_TYPE_RESTART, stack); } int get_stack_info(unsigned long sp, struct task_struct *task, @@ -152,7 +142,13 @@ void show_stack(struct task_struct *task, unsigned long *stack, static void show_last_breaking_event(struct pt_regs *regs) { printk("Last Breaking-Event-Address:\n"); - printk(" [<%016lx>] %pSR\n", regs->last_break, (void *)regs->last_break); + printk(" [<%016lx>] ", regs->last_break); + if (user_mode(regs)) { + print_vma_addr(KERN_CONT, regs->last_break); + pr_cont("\n"); + } else { + pr_cont("%pSR\n", (void *)regs->last_break); + } } void show_registers(struct pt_regs *regs) diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index d26f02495636..2dd5976a55ac 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -34,8 +34,6 @@ #include <asm/switch_to.h> #include "entry.h" -int __bootdata(is_full_image); - #define decompressor_handled_param(param) \ static int __init ignore_decompressor_param_##param(char *s) \ { \ @@ -53,6 +51,14 @@ decompressor_handled_param(nokaslr); decompressor_handled_param(prot_virt); #endif +static void __init kasan_early_init(void) +{ +#ifdef CONFIG_KASAN + init_task.kasan_depth = 0; + sclp_early_printk("KernelAddressSanitizer initialized\n"); +#endif +} + static void __init reset_tod_clock(void) { union tod_clock clk; @@ -288,17 +294,6 @@ static void __init setup_boot_command_line(void) strscpy(boot_command_line, early_command_line, COMMAND_LINE_SIZE); } -static void __init check_image_bootable(void) -{ - if (is_full_image) - return; - - sclp_early_printk("Linux kernel boot failure: An attempt to boot a vmlinux ELF image failed.\n"); - sclp_early_printk("This image does not contain all parts necessary for starting up. Use\n"); - sclp_early_printk("bzImage or arch/s390/boot/compressed/vmlinux instead.\n"); - disabled_wait(); -} - static void __init sort_amode31_extable(void) { sort_extable(__start_amode31_ex_table, __stop_amode31_ex_table); @@ -306,8 +301,8 @@ static void __init sort_amode31_extable(void) void __init startup_init(void) { + kasan_early_init(); reset_tod_clock(); - check_image_bootable(); time_early_init(); init_kernel_storage_key(); lockdep_off(); diff --git a/arch/s390/kernel/earlypgm.S b/arch/s390/kernel/earlypgm.S index f521c6da37b8..c634871f0d90 100644 --- a/arch/s390/kernel/earlypgm.S +++ b/arch/s390/kernel/earlypgm.S @@ -7,7 +7,7 @@ #include <linux/linkage.h> #include <asm/asm-offsets.h> -ENTRY(early_pgm_check_handler) +SYM_CODE_START(early_pgm_check_handler) stmg %r8,%r15,__LC_SAVE_AREA_SYNC aghi %r15,-(STACK_FRAME_OVERHEAD+__PT_SIZE) la %r11,STACK_FRAME_OVERHEAD(%r15) @@ -20,4 +20,4 @@ ENTRY(early_pgm_check_handler) mvc __LC_RETURN_PSW(16),STACK_FRAME_OVERHEAD+__PT_PSW(%r15) lmg %r0,%r15,STACK_FRAME_OVERHEAD+__PT_R0(%r15) lpswe __LC_RETURN_PSW -ENDPROC(early_pgm_check_handler) +SYM_CODE_END(early_pgm_check_handler) diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 76a06f3d3671..e5b6c1369e8e 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -29,10 +29,6 @@ #include <asm/export.h> #include <asm/nospec-insn.h> -STACK_SHIFT = PAGE_SHIFT + THREAD_SIZE_ORDER -STACK_SIZE = 1 << STACK_SHIFT -STACK_INIT = STACK_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE - _LPP_OFFSET = __LC_LPP .macro STBEAR address @@ -53,7 +49,7 @@ _LPP_OFFSET = __LC_LPP .macro CHECK_STACK savearea #ifdef CONFIG_CHECK_STACK - tml %r15,STACK_SIZE - CONFIG_STACK_GUARD + tml %r15,THREAD_SIZE - CONFIG_STACK_GUARD lghi %r14,\savearea jz stack_overflow #endif @@ -62,8 +58,8 @@ _LPP_OFFSET = __LC_LPP .macro CHECK_VMAP_STACK savearea,oklabel #ifdef CONFIG_VMAP_STACK lgr %r14,%r15 - nill %r14,0x10000 - STACK_SIZE - oill %r14,STACK_INIT + nill %r14,0x10000 - THREAD_SIZE + oill %r14,STACK_INIT_OFFSET clg %r14,__LC_KERNEL_STACK je \oklabel clg %r14,__LC_ASYNC_STACK @@ -154,26 +150,26 @@ _LPP_OFFSET = __LC_LPP .endm #endif + .macro STACKLEAK_ERASE +#ifdef CONFIG_GCC_PLUGIN_STACKLEAK + brasl %r14,stackleak_erase_on_task_stack +#endif + .endm + GEN_BR_THUNK %r14 .section .kprobes.text, "ax" .Ldummy: /* - * This nop exists only in order to avoid that __bpon starts at - * the beginning of the kprobes text section. In that case we would - * have several symbols at the same address. E.g. objdump would take - * an arbitrary symbol name when disassembling this code. - * With the added nop in between the __bpon symbol is unique - * again. + * The following nop exists only in order to avoid that the next + * symbol starts at the beginning of the kprobes text section. + * In that case there would be several symbols at the same address. + * E.g. objdump would take an arbitrary symbol when disassembling + * the code. + * With the added nop in between this cannot happen. */ nop 0 -ENTRY(__bpon) - .globl __bpon - BPON - BR_EX %r14 -ENDPROC(__bpon) - /* * Scheduler resume function, called by switch_to * gpr2 = (task_struct *) prev @@ -181,11 +177,11 @@ ENDPROC(__bpon) * Returns: * gpr2 = prev */ -ENTRY(__switch_to) +SYM_FUNC_START(__switch_to) stmg %r6,%r15,__SF_GPRS(%r15) # store gprs of prev task lghi %r4,__TASK_stack lghi %r1,__TASK_thread - llill %r5,STACK_INIT + llill %r5,STACK_INIT_OFFSET stg %r15,__THREAD_ksp(%r1,%r2) # store kernel stack of prev lg %r15,0(%r4,%r3) # start of kernel stack of next agr %r15,%r5 # end of kernel stack of next @@ -197,7 +193,7 @@ ENTRY(__switch_to) lmg %r6,%r15,__SF_GPRS(%r15) # load gprs of next task ALTERNATIVE "nop", "lpp _LPP_OFFSET", 40 BR_EX %r14 -ENDPROC(__switch_to) +SYM_FUNC_END(__switch_to) #if IS_ENABLED(CONFIG_KVM) /* @@ -206,7 +202,7 @@ ENDPROC(__switch_to) * %r3 pointer to sie control block virt * %r4 guest register save area */ -ENTRY(__sie64a) +SYM_FUNC_START(__sie64a) stmg %r6,%r14,__SF_GPRS(%r15) # save kernel registers lg %r12,__LC_CURRENT stg %r2,__SF_SIE_CONTROL_PHYS(%r15) # save sie block physical.. @@ -227,7 +223,7 @@ ENTRY(__sie64a) TSTMSK __LC_CPU_FLAGS,_CIF_FPU jo .Lsie_skip # exit if fp/vx regs changed lg %r14,__SF_SIE_CONTROL_PHYS(%r15) # get sie block phys addr - BPEXIT __SF_SIE_FLAGS(%r15),(_TIF_ISOLATE_BP|_TIF_ISOLATE_BP_GUEST) + BPEXIT __SF_SIE_FLAGS(%r15),_TIF_ISOLATE_BP_GUEST .Lsie_entry: sie 0(%r14) # Let the next instruction be NOP to avoid triggering a machine check @@ -235,7 +231,7 @@ ENTRY(__sie64a) nopr 7 .Lsie_leave: BPOFF - BPENTER __SF_SIE_FLAGS(%r15),(_TIF_ISOLATE_BP|_TIF_ISOLATE_BP_GUEST) + BPENTER __SF_SIE_FLAGS(%r15),_TIF_ISOLATE_BP_GUEST .Lsie_skip: lg %r14,__SF_SIE_CONTROL(%r15) # get control block pointer ni __SIE_PROG0C+3(%r14),0xfe # no longer in SIE @@ -252,8 +248,7 @@ ENTRY(__sie64a) nopr 7 .Lrewind_pad2: nopr 7 - .globl sie_exit -sie_exit: +SYM_INNER_LABEL(sie_exit, SYM_L_GLOBAL) lg %r14,__SF_SIE_SAVEAREA(%r15) # load guest register save area stmg %r0,%r13,0(%r14) # save guest gprs 0-13 xgr %r0,%r0 # clear guest registers to @@ -273,7 +268,7 @@ sie_exit: EX_TABLE(.Lrewind_pad4,.Lsie_fault) EX_TABLE(.Lrewind_pad2,.Lsie_fault) EX_TABLE(sie_exit,.Lsie_fault) -ENDPROC(__sie64a) +SYM_FUNC_END(__sie64a) EXPORT_SYMBOL(__sie64a) EXPORT_SYMBOL(sie_exit) #endif @@ -283,7 +278,7 @@ EXPORT_SYMBOL(sie_exit) * are entered with interrupts disabled. */ -ENTRY(system_call) +SYM_CODE_START(system_call) stpt __LC_SYS_ENTER_TIMER stmg %r8,%r15,__LC_SAVE_AREA_SYNC BPOFF @@ -291,11 +286,9 @@ ENTRY(system_call) .Lsysc_per: STBEAR __LC_LAST_BREAK lctlg %c1,%c1,__LC_KERNEL_ASCE - lg %r12,__LC_CURRENT lg %r15,__LC_KERNEL_STACK xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) stmg %r0,%r7,STACK_FRAME_OVERHEAD+__PT_R0(%r15) - BPENTER __TI_flags(%r12),_TIF_ISOLATE_BP # clear user controlled register to prevent speculative use xgr %r0,%r0 xgr %r1,%r1 @@ -312,39 +305,40 @@ ENTRY(system_call) MBEAR %r2 lgr %r3,%r14 brasl %r14,__do_syscall + STACKLEAK_ERASE lctlg %c1,%c1,__LC_USER_ASCE mvc __LC_RETURN_PSW(16),STACK_FRAME_OVERHEAD+__PT_PSW(%r15) - BPEXIT __TI_flags(%r12),_TIF_ISOLATE_BP + BPON LBEAR STACK_FRAME_OVERHEAD+__PT_LAST_BREAK(%r15) lmg %r0,%r15,STACK_FRAME_OVERHEAD+__PT_R0(%r15) stpt __LC_EXIT_TIMER LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE -ENDPROC(system_call) +SYM_CODE_END(system_call) # # a new process exits the kernel with ret_from_fork # -ENTRY(ret_from_fork) +SYM_CODE_START(ret_from_fork) lgr %r3,%r11 brasl %r14,__ret_from_fork + STACKLEAK_ERASE lctlg %c1,%c1,__LC_USER_ASCE mvc __LC_RETURN_PSW(16),STACK_FRAME_OVERHEAD+__PT_PSW(%r15) - BPEXIT __TI_flags(%r12),_TIF_ISOLATE_BP + BPON LBEAR STACK_FRAME_OVERHEAD+__PT_LAST_BREAK(%r15) lmg %r0,%r15,STACK_FRAME_OVERHEAD+__PT_R0(%r15) stpt __LC_EXIT_TIMER LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE -ENDPROC(ret_from_fork) +SYM_CODE_END(ret_from_fork) /* * Program check handler routine */ -ENTRY(pgm_check_handler) +SYM_CODE_START(pgm_check_handler) stpt __LC_SYS_ENTER_TIMER BPOFF stmg %r8,%r15,__LC_SAVE_AREA_SYNC - lg %r12,__LC_CURRENT lghi %r10,0 lmg %r8,%r9,__LC_PGM_OLD_PSW tmhh %r8,0x0001 # coming from user space? @@ -355,6 +349,7 @@ ENTRY(pgm_check_handler) #if IS_ENABLED(CONFIG_KVM) # cleanup critical section for program checks in __sie64a OUTSIDE %r9,.Lsie_gmap,.Lsie_done,1f + BPENTER __SF_SIE_FLAGS(%r15),_TIF_ISOLATE_BP_GUEST SIEEXIT lghi %r10,_PIF_GUEST_FAULT #endif @@ -366,8 +361,7 @@ ENTRY(pgm_check_handler) aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE) # CHECK_VMAP_STACK branches to stack_overflow or 4f CHECK_VMAP_STACK __LC_SAVE_AREA_SYNC,4f -3: BPENTER __TI_flags(%r12),_TIF_ISOLATE_BP - lg %r15,__LC_KERNEL_STACK +3: lg %r15,__LC_KERNEL_STACK 4: la %r11,STACK_FRAME_OVERHEAD(%r15) stg %r10,__PT_FLAGS(%r11) xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) @@ -388,8 +382,9 @@ ENTRY(pgm_check_handler) brasl %r14,__do_pgm_check tmhh %r8,0x0001 # returning to user space? jno .Lpgm_exit_kernel + STACKLEAK_ERASE lctlg %c1,%c1,__LC_USER_ASCE - BPEXIT __TI_flags(%r12),_TIF_ISOLATE_BP + BPON stpt __LC_EXIT_TIMER .Lpgm_exit_kernel: mvc __LC_RETURN_PSW(16),STACK_FRAME_OVERHEAD+__PT_PSW(%r15) @@ -407,32 +402,30 @@ ENTRY(pgm_check_handler) lghi %r14,1 LBEAR __LC_PGM_LAST_BREAK LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE # branch to .Lsysc_per -ENDPROC(pgm_check_handler) +SYM_CODE_END(pgm_check_handler) /* * Interrupt handler macro used for external and IO interrupts. */ .macro INT_HANDLER name,lc_old_psw,handler -ENTRY(\name) +SYM_CODE_START(\name) stckf __LC_INT_CLOCK stpt __LC_SYS_ENTER_TIMER STBEAR __LC_LAST_BREAK BPOFF stmg %r8,%r15,__LC_SAVE_AREA_ASYNC - lg %r12,__LC_CURRENT lmg %r8,%r9,\lc_old_psw tmhh %r8,0x0001 # interrupting from user ? jnz 1f #if IS_ENABLED(CONFIG_KVM) OUTSIDE %r9,.Lsie_gmap,.Lsie_done,0f - BPENTER __SF_SIE_FLAGS(%r15),(_TIF_ISOLATE_BP|_TIF_ISOLATE_BP_GUEST) + BPENTER __SF_SIE_FLAGS(%r15),_TIF_ISOLATE_BP_GUEST SIEEXIT #endif 0: CHECK_STACK __LC_SAVE_AREA_ASYNC aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE) j 2f -1: BPENTER __TI_flags(%r12),_TIF_ISOLATE_BP - lctlg %c1,%c1,__LC_KERNEL_ASCE +1: lctlg %c1,%c1,__LC_KERNEL_ASCE lg %r15,__LC_KERNEL_STACK 2: xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) la %r11,STACK_FRAME_OVERHEAD(%r15) @@ -455,13 +448,14 @@ ENTRY(\name) mvc __LC_RETURN_PSW(16),__PT_PSW(%r11) tmhh %r8,0x0001 # returning to user ? jno 2f + STACKLEAK_ERASE lctlg %c1,%c1,__LC_USER_ASCE - BPEXIT __TI_flags(%r12),_TIF_ISOLATE_BP + BPON stpt __LC_EXIT_TIMER 2: LBEAR __PT_LAST_BREAK(%r11) lmg %r0,%r15,__PT_R0(%r11) LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE -ENDPROC(\name) +SYM_CODE_END(\name) .endm INT_HANDLER ext_int_handler,__LC_EXT_OLD_PSW,do_ext_irq @@ -470,7 +464,7 @@ INT_HANDLER io_int_handler,__LC_IO_OLD_PSW,do_io_irq /* * Load idle PSW. */ -ENTRY(psw_idle) +SYM_FUNC_START(psw_idle) stg %r14,(__SF_GPRS+8*8)(%r15) stg %r3,__SF_EMPTY(%r15) larl %r1,psw_idle_exit @@ -486,29 +480,26 @@ ENTRY(psw_idle) stckf __CLOCK_IDLE_ENTER(%r2) stpt __TIMER_IDLE_ENTER(%r2) lpswe __SF_EMPTY(%r15) -.globl psw_idle_exit -psw_idle_exit: +SYM_INNER_LABEL(psw_idle_exit, SYM_L_GLOBAL) BR_EX %r14 -ENDPROC(psw_idle) +SYM_FUNC_END(psw_idle) /* * Machine check handler routines */ -ENTRY(mcck_int_handler) +SYM_CODE_START(mcck_int_handler) stckf __LC_MCCK_CLOCK BPOFF la %r1,4095 # validate r1 spt __LC_CPU_TIMER_SAVE_AREA-4095(%r1) # validate cpu timer LBEAR __LC_LAST_BREAK_SAVE_AREA-4095(%r1) # validate bear - lmg %r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r1)# validate gprs - lg %r12,__LC_CURRENT + lmg %r0,%r15,__LC_GPREGS_SAVE_AREA # validate gprs lmg %r8,%r9,__LC_MCK_OLD_PSW TSTMSK __LC_MCCK_CODE,MCCK_CODE_SYSTEM_DAMAGE jo .Lmcck_panic # yes -> rest of mcck code invalid TSTMSK __LC_MCCK_CODE,MCCK_CODE_CR_VALID jno .Lmcck_panic # control registers invalid -> panic - la %r14,4095 - lctlg %c0,%c15,__LC_CREGS_SAVE_AREA-4095(%r14) # validate ctl regs + lctlg %c0,%c15,__LC_CREGS_SAVE_AREA # validate ctl regs ptlb lghi %r14,__LC_CPU_TIMER_SAVE_AREA mvc __LC_MCCK_ENTER_TIMER(8),0(%r14) @@ -530,16 +521,13 @@ ENTRY(mcck_int_handler) TSTMSK __LC_MCCK_CODE,MCCK_CODE_PSW_IA_VALID jno .Lmcck_panic #if IS_ENABLED(CONFIG_KVM) - OUTSIDE %r9,.Lsie_gmap,.Lsie_done,.Lmcck_stack + OUTSIDE %r9,.Lsie_gmap,.Lsie_done,.Lmcck_user OUTSIDE %r9,.Lsie_entry,.Lsie_leave,4f oi __LC_CPU_FLAGS+7, _CIF_MCCK_GUEST -4: BPENTER __SF_SIE_FLAGS(%r15),(_TIF_ISOLATE_BP|_TIF_ISOLATE_BP_GUEST) +4: BPENTER __SF_SIE_FLAGS(%r15),_TIF_ISOLATE_BP_GUEST SIEEXIT - j .Lmcck_stack #endif .Lmcck_user: - BPENTER __TI_flags(%r12),_TIF_ISOLATE_BP -.Lmcck_stack: lg %r15,__LC_MCCK_STACK la %r11,STACK_FRAME_OVERHEAD(%r15) stctg %c1,%c1,__PT_CR1(%r11) @@ -567,7 +555,7 @@ ENTRY(mcck_int_handler) mvc __LC_RETURN_MCCK_PSW(16),__PT_PSW(%r11) # move return PSW tm __LC_RETURN_MCCK_PSW+1,0x01 # returning to user ? jno 0f - BPEXIT __TI_flags(%r12),_TIF_ISOLATE_BP + BPON stpt __LC_EXIT_TIMER 0: ALTERNATIVE "nop", __stringify(lghi %r12,__LC_LAST_BREAK_SAVE_AREA),193 LBEAR 0(%r12) @@ -583,10 +571,10 @@ ENTRY(mcck_int_handler) */ lhi %r5,0 lhi %r6,1 - larl %r7,.Lstop_lock + larl %r7,stop_lock cs %r5,%r6,0(%r7) # single CPU-stopper only jnz 4f - larl %r7,.Lthis_cpu + larl %r7,this_cpu stap 0(%r7) # this CPU address lh %r4,0(%r7) nilh %r4,0 @@ -602,16 +590,15 @@ ENTRY(mcck_int_handler) 3: sigp %r1,%r4,SIGP_STOP # stop this CPU brc SIGP_CC_BUSY,3b 4: j 4b -ENDPROC(mcck_int_handler) +SYM_CODE_END(mcck_int_handler) -ENTRY(restart_int_handler) +SYM_CODE_START(restart_int_handler) ALTERNATIVE "nop", "lpp _LPP_OFFSET", 40 stg %r15,__LC_SAVE_AREA_RESTART TSTMSK __LC_RESTART_FLAGS,RESTART_FLAG_CTLREGS,4 jz 0f - la %r15,4095 - lctlg %c0,%c15,__LC_CREGS_SAVE_AREA-4095(%r15) -0: larl %r15,.Lstosm_tmp + lctlg %c0,%c15,__LC_CREGS_SAVE_AREA +0: larl %r15,stosm_tmp stosm 0(%r15),0x04 # turn dat on, keep irqs off lg %r15,__LC_RESTART_STACK xc STACK_FRAME_OVERHEAD(__PT_SIZE,%r15),STACK_FRAME_OVERHEAD(%r15) @@ -632,7 +619,7 @@ ENTRY(restart_int_handler) 2: sigp %r4,%r3,SIGP_STOP # sigp stop to current cpu brc 2,2b 3: j 3b -ENDPROC(restart_int_handler) +SYM_CODE_END(restart_int_handler) .section .kprobes.text, "ax" @@ -642,7 +629,7 @@ ENDPROC(restart_int_handler) * No need to properly save the registers, we are going to panic anyway. * Setup a pt_regs so that show_trace can provide a good call trace. */ -ENTRY(stack_overflow) +SYM_CODE_START(stack_overflow) lg %r15,__LC_NODAT_STACK # change to panic stack la %r11,STACK_FRAME_OVERHEAD(%r15) stmg %r0,%r7,__PT_R0(%r11) @@ -652,26 +639,27 @@ ENTRY(stack_overflow) xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) lgr %r2,%r11 # pass pointer to pt_regs jg kernel_stack_overflow -ENDPROC(stack_overflow) +SYM_CODE_END(stack_overflow) #endif .section .data, "aw" - .align 4 -.Lstop_lock: .long 0 -.Lthis_cpu: .short 0 -.Lstosm_tmp: .byte 0 + .balign 4 +SYM_DATA_LOCAL(stop_lock, .long 0) +SYM_DATA_LOCAL(this_cpu, .short 0) +SYM_DATA_LOCAL(stosm_tmp, .byte 0) + .section .rodata, "a" #define SYSCALL(esame,emu) .quad __s390x_ ## esame - .globl sys_call_table -sys_call_table: +SYM_DATA_START(sys_call_table) #include "asm/syscall_table.h" +SYM_DATA_END(sys_call_table) #undef SYSCALL #ifdef CONFIG_COMPAT #define SYSCALL(esame,emu) .quad __s390_ ## emu - .globl sys_call_table_emu -sys_call_table_emu: +SYM_DATA_START(sys_call_table_emu) #include "asm/syscall_table.h" +SYM_DATA_END(sys_call_table_emu) #undef SYSCALL #endif diff --git a/arch/s390/kernel/ftrace.c b/arch/s390/kernel/ftrace.c index 416b5a94353d..c46381ea04ec 100644 --- a/arch/s390/kernel/ftrace.c +++ b/arch/s390/kernel/ftrace.c @@ -49,26 +49,6 @@ struct ftrace_insn { s32 disp; } __packed; -asm( - " .align 16\n" - "ftrace_shared_hotpatch_trampoline_br:\n" - " lmg %r0,%r1,2(%r1)\n" - " br %r1\n" - "ftrace_shared_hotpatch_trampoline_br_end:\n" -); - -#ifdef CONFIG_EXPOLINE -asm( - " .align 16\n" - "ftrace_shared_hotpatch_trampoline_exrl:\n" - " lmg %r0,%r1,2(%r1)\n" - " exrl %r0,0f\n" - " j .\n" - "0: br %r1\n" - "ftrace_shared_hotpatch_trampoline_exrl_end:\n" -); -#endif /* CONFIG_EXPOLINE */ - #ifdef CONFIG_MODULES static char *ftrace_plt; #endif /* CONFIG_MODULES */ @@ -246,7 +226,7 @@ static int __init ftrace_plt_init(void) start = ftrace_shared_hotpatch_trampoline(&end); memcpy(ftrace_plt, start, end - start); - set_memory_ro((unsigned long)ftrace_plt, 1); + set_memory_rox((unsigned long)ftrace_plt, 1); return 0; } device_initcall(ftrace_plt_init); diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index 3b3bf8329e6c..df77ba102096 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -16,7 +16,7 @@ #include <asm/ptrace.h> __HEAD -ENTRY(startup_continue) +SYM_CODE_START(startup_continue) larl %r1,tod_clock_base mvc 0(16,%r1),__LC_BOOT_CLOCK # @@ -24,19 +24,17 @@ ENTRY(startup_continue) # larl %r14,init_task stg %r14,__LC_CURRENT - larl %r15,init_thread_union+THREAD_SIZE-STACK_FRAME_OVERHEAD-__PT_SIZE + larl %r15,init_thread_union+STACK_INIT_OFFSET + stg %r15,__LC_KERNEL_STACK brasl %r14,sclp_early_adjust_va # allow sclp_early_printk -#ifdef CONFIG_KASAN - brasl %r14,kasan_early_init -#endif brasl %r14,startup_init # s390 specific early init brasl %r14,start_kernel # common init code # # We returned from start_kernel ?!? PANIK # basr %r13,0 - lpswe .Ldw-.(%r13) # load disabled wait psw + lpswe dw_psw-.(%r13) # load disabled wait psw +SYM_CODE_END(startup_continue) .align 16 -.LPG1: -.Ldw: .quad 0x0002000180000000,0x0000000000000000 +SYM_DATA_LOCAL(dw_psw, .quad 0x0002000180000000,0x0000000000000000) diff --git a/arch/s390/kernel/idle.c b/arch/s390/kernel/idle.c index 38e267c7bff7..e7239aaf428b 100644 --- a/arch/s390/kernel/idle.c +++ b/arch/s390/kernel/idle.c @@ -88,7 +88,7 @@ void arch_cpu_idle_exit(void) { } -void arch_cpu_idle_dead(void) +void __noreturn arch_cpu_idle_dead(void) { cpu_die(); } diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 5f0f5c86963a..43de939b7af1 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -176,11 +176,11 @@ static bool reipl_fcp_clear; static bool reipl_ccw_clear; static bool reipl_eckd_clear; -static inline int __diag308(unsigned long subcode, void *addr) +static inline int __diag308(unsigned long subcode, unsigned long addr) { union register_pair r1; - r1.even = (unsigned long) addr; + r1.even = addr; r1.odd = 0; asm volatile( " diag %[r1],%[subcode],0x308\n" @@ -195,7 +195,7 @@ static inline int __diag308(unsigned long subcode, void *addr) int diag308(unsigned long subcode, void *addr) { diag_stat_inc(DIAG_STAT_X308); - return __diag308(subcode, addr); + return __diag308(subcode, addr ? virt_to_phys(addr) : 0); } EXPORT_SYMBOL_GPL(diag308); @@ -649,7 +649,6 @@ static struct kset *ipl_kset; static void __ipl_run(void *unused) { - __bpon(); diag308(DIAG308_LOAD_CLEAR, NULL); } diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c index 7b41ceecbb25..d4b863ed0aa7 100644 --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -41,7 +41,7 @@ void *alloc_insn_page(void) page = module_alloc(PAGE_SIZE); if (!page) return NULL; - __set_memory((unsigned long) page, 1, SET_MEMORY_RO | SET_MEMORY_X); + set_memory_rox((unsigned long)page, 1); return page; } diff --git a/arch/s390/kernel/kprobes_insn_page.S b/arch/s390/kernel/kprobes_insn_page.S index f6cb022ef8c8..b6335296dcd8 100644 --- a/arch/s390/kernel/kprobes_insn_page.S +++ b/arch/s390/kernel/kprobes_insn_page.S @@ -14,9 +14,9 @@ */ .section .kprobes.text, "ax" .align 4096 -ENTRY(kprobes_insn_page) +SYM_CODE_START(kprobes_insn_page) .rept 2048 .word 0x07fe .endr -ENDPROC(kprobes_insn_page) +SYM_CODE_END(kprobes_insn_page) .previous diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c index 2a8e73266428..6d9276c096a6 100644 --- a/arch/s390/kernel/machine_kexec.c +++ b/arch/s390/kernel/machine_kexec.c @@ -29,8 +29,8 @@ #include <asm/nmi.h> #include <asm/sclp.h> -typedef void (*relocate_kernel_t)(kimage_entry_t *, unsigned long, - unsigned long); +typedef void (*relocate_kernel_t)(unsigned long, unsigned long, unsigned long); +typedef int (*purgatory_t)(int); extern const unsigned char relocate_kernel[]; extern const unsigned long long relocate_kernel_len; @@ -41,11 +41,14 @@ extern const unsigned long long relocate_kernel_len; * Reset the system, copy boot CPU registers to absolute zero, * and jump to the kdump image */ -static void __do_machine_kdump(void *image) +static void __do_machine_kdump(void *data) { - int (*start_kdump)(int); + struct kimage *image = data; + purgatory_t purgatory; unsigned long prefix; + purgatory = (purgatory_t)image->start; + /* store_status() saved the prefix register to lowcore */ prefix = (unsigned long) S390_lowcore.prefixreg_save_area; @@ -58,13 +61,11 @@ static void __do_machine_kdump(void *image) * prefix register of this CPU to zero */ memcpy(absolute_pointer(__LC_FPREGS_SAVE_AREA), - (void *)(prefix + __LC_FPREGS_SAVE_AREA), 512); + phys_to_virt(prefix + __LC_FPREGS_SAVE_AREA), 512); - __load_psw_mask(PSW_MASK_BASE | PSW_DEFAULT_KEY | PSW_MASK_EA | PSW_MASK_BA); - start_kdump = (void *)((struct kimage *) image)->start; - start_kdump(1); + call_nodat(1, int, purgatory, int, 1); - /* Die if start_kdump returns */ + /* Die if kdump returns */ disabled_wait(); } @@ -111,18 +112,6 @@ static noinline void __machine_kdump(void *image) store_status(__do_machine_kdump, image); } -static unsigned long do_start_kdump(unsigned long addr) -{ - struct kimage *image = (struct kimage *) addr; - int (*start_kdump)(int) = (void *)image->start; - int rc; - - __arch_local_irq_stnsm(0xfb); /* disable DAT */ - rc = start_kdump(0); - __arch_local_irq_stosm(0x04); /* enable DAT */ - return rc; -} - #endif /* CONFIG_CRASH_DUMP */ /* @@ -131,12 +120,10 @@ static unsigned long do_start_kdump(unsigned long addr) static bool kdump_csum_valid(struct kimage *image) { #ifdef CONFIG_CRASH_DUMP + purgatory_t purgatory = (purgatory_t)image->start; int rc; - preempt_disable(); - rc = call_on_stack(1, S390_lowcore.nodat_stack, unsigned long, do_start_kdump, - unsigned long, (unsigned long)image); - preempt_enable(); + rc = call_nodat(1, int, purgatory, int, 0); return rc == 0; #else return false; @@ -210,7 +197,7 @@ int machine_kexec_prepare(struct kimage *image) return -EINVAL; /* Get the destination where the assembler code should be copied to.*/ - reboot_code_buffer = (void *) page_to_phys(image->control_code_page); + reboot_code_buffer = page_to_virt(image->control_code_page); /* Then copy it */ memcpy(reboot_code_buffer, relocate_kernel, relocate_kernel_len); @@ -250,19 +237,20 @@ void machine_crash_shutdown(struct pt_regs *regs) */ static void __do_machine_kexec(void *data) { - unsigned long diag308_subcode; - relocate_kernel_t data_mover; + unsigned long data_mover, entry, diag308_subcode; struct kimage *image = data; - s390_reset_system(); - data_mover = (relocate_kernel_t) page_to_phys(image->control_code_page); - - __arch_local_irq_stnsm(0xfb); /* disable DAT - avoid no-execute */ - /* Call the moving routine */ + data_mover = page_to_phys(image->control_code_page); + entry = virt_to_phys(&image->head); diag308_subcode = DIAG308_CLEAR_RESET; if (sclp.has_iplcc) diag308_subcode |= DIAG308_FLAG_EI; - (*data_mover)(&image->head, image->start, diag308_subcode); + s390_reset_system(); + + call_nodat(3, void, (relocate_kernel_t)data_mover, + unsigned long, entry, + unsigned long, image->start, + unsigned long, diag308_subcode); /* Die if kexec returns */ disabled_wait(); diff --git a/arch/s390/kernel/mcount.S b/arch/s390/kernel/mcount.S index 43ff91073d2a..dbece2803c50 100644 --- a/arch/s390/kernel/mcount.S +++ b/arch/s390/kernel/mcount.S @@ -28,9 +28,14 @@ .section .kprobes.text, "ax" -ENTRY(ftrace_stub) +SYM_FUNC_START(ftrace_stub) BR_EX %r14 -ENDPROC(ftrace_stub) +SYM_FUNC_END(ftrace_stub) + +SYM_CODE_START(ftrace_stub_direct_tramp) + lgr %r1, %r0 + BR_EX %r1 +SYM_CODE_END(ftrace_stub_direct_tramp) .macro ftrace_regs_entry, allregs=0 stg %r14,(__SF_GPRS+8*8)(%r15) # save traced function caller @@ -135,10 +140,25 @@ SYM_FUNC_END(return_to_handler) #endif #endif /* CONFIG_FUNCTION_TRACER */ -#ifdef CONFIG_RETHOOK +SYM_CODE_START(ftrace_shared_hotpatch_trampoline_br) + lmg %r0,%r1,2(%r1) + br %r1 +SYM_INNER_LABEL(ftrace_shared_hotpatch_trampoline_br_end, SYM_L_GLOBAL) +SYM_CODE_END(ftrace_shared_hotpatch_trampoline_br) + +#ifdef CONFIG_EXPOLINE +SYM_CODE_START(ftrace_shared_hotpatch_trampoline_exrl) + lmg %r0,%r1,2(%r1) + exrl %r0,0f + j . +0: br %r1 +SYM_INNER_LABEL(ftrace_shared_hotpatch_trampoline_exrl_end, SYM_L_GLOBAL) +SYM_CODE_END(ftrace_shared_hotpatch_trampoline_exrl) +#endif /* CONFIG_EXPOLINE */ -SYM_FUNC_START(arch_rethook_trampoline) +#ifdef CONFIG_RETHOOK +SYM_CODE_START(arch_rethook_trampoline) stg %r14,(__SF_GPRS+8*8)(%r15) lay %r15,-STACK_FRAME_SIZE(%r15) stmg %r0,%r14,STACK_PTREGS_GPRS(%r15) @@ -161,7 +181,6 @@ SYM_FUNC_START(arch_rethook_trampoline) mvc __SF_EMPTY(16,%r7),STACK_PTREGS_PSW(%r15) lmg %r0,%r15,STACK_PTREGS_GPRS(%r15) lpswe __SF_EMPTY(%r15) - -SYM_FUNC_END(arch_rethook_trampoline) +SYM_CODE_END(arch_rethook_trampoline) #endif /* CONFIG_RETHOOK */ diff --git a/arch/s390/kernel/module.c b/arch/s390/kernel/module.c index 2d159b32885b..f1b35dcdf3eb 100644 --- a/arch/s390/kernel/module.c +++ b/arch/s390/kernel/module.c @@ -26,6 +26,7 @@ #include <asm/facility.h> #include <asm/ftrace.lds.h> #include <asm/set_memory.h> +#include <asm/setup.h> #if 0 #define DEBUGP printk @@ -35,6 +36,24 @@ #define PLT_ENTRY_SIZE 22 +static unsigned long get_module_load_offset(void) +{ + static DEFINE_MUTEX(module_kaslr_mutex); + static unsigned long module_load_offset; + + if (!kaslr_enabled()) + return 0; + /* + * Calculate the module_load_offset the first time this code + * is called. Once calculated it stays the same until reboot. + */ + mutex_lock(&module_kaslr_mutex); + if (!module_load_offset) + module_load_offset = get_random_u32_inclusive(1, 1024) * PAGE_SIZE; + mutex_unlock(&module_kaslr_mutex); + return module_load_offset; +} + void *module_alloc(unsigned long size) { gfp_t gfp_mask = GFP_KERNEL; @@ -42,9 +61,11 @@ void *module_alloc(unsigned long size) if (PAGE_ALIGN(size) > MODULES_LEN) return NULL; - p = __vmalloc_node_range(size, MODULE_ALIGN, MODULES_VADDR, MODULES_END, - gfp_mask, PAGE_KERNEL_EXEC, VM_DEFER_KMEMLEAK, NUMA_NO_NODE, - __builtin_return_address(0)); + p = __vmalloc_node_range(size, MODULE_ALIGN, + MODULES_VADDR + get_module_load_offset(), + MODULES_END, gfp_mask, PAGE_KERNEL, + VM_FLUSH_RESET_PERMS | VM_DEFER_KMEMLEAK, + NUMA_NO_NODE, __builtin_return_address(0)); if (p && (kasan_alloc_module_shadow(p, size, gfp_mask) < 0)) { vfree(p); return NULL; @@ -126,6 +147,7 @@ int module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs, Elf_Rela *rela; char *strings; int nrela, i, j; + struct module_memory *mod_mem; /* Find symbol table and string table. */ symtab = NULL; @@ -173,14 +195,15 @@ int module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs, /* Increase core size by size of got & plt and set start offsets for got and plt. */ - me->core_layout.size = ALIGN(me->core_layout.size, 4); - me->arch.got_offset = me->core_layout.size; - me->core_layout.size += me->arch.got_size; - me->arch.plt_offset = me->core_layout.size; + mod_mem = &me->mem[MOD_TEXT]; + mod_mem->size = ALIGN(mod_mem->size, 4); + me->arch.got_offset = mod_mem->size; + mod_mem->size += me->arch.got_size; + me->arch.plt_offset = mod_mem->size; if (me->arch.plt_size) { if (IS_ENABLED(CONFIG_EXPOLINE) && !nospec_disable) me->arch.plt_size += PLT_ENTRY_SIZE; - me->core_layout.size += me->arch.plt_size; + mod_mem->size += me->arch.plt_size; } return 0; } @@ -304,7 +327,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab, case R_390_GOTPLT64: /* 64 bit offset to jump slot. */ case R_390_GOTPLTENT: /* 32 bit rel. offset to jump slot >> 1. */ if (info->got_initialized == 0) { - Elf_Addr *gotent = me->core_layout.base + + Elf_Addr *gotent = me->mem[MOD_TEXT].base + me->arch.got_offset + info->got_offset; @@ -329,7 +352,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab, rc = apply_rela_bits(loc, val, 0, 64, 0, write); else if (r_type == R_390_GOTENT || r_type == R_390_GOTPLTENT) { - val += (Elf_Addr) me->core_layout.base - loc; + val += (Elf_Addr) me->mem[MOD_TEXT].base - loc; rc = apply_rela_bits(loc, val, 1, 32, 1, write); } break; @@ -345,7 +368,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab, char *plt_base; char *ip; - plt_base = me->core_layout.base + me->arch.plt_offset; + plt_base = me->mem[MOD_TEXT].base + me->arch.plt_offset; ip = plt_base + info->plt_offset; *(int *)insn = 0x0d10e310; /* basr 1,0 */ *(int *)&insn[4] = 0x100c0004; /* lg 1,12(1) */ @@ -375,7 +398,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab, val - loc + 0xffffUL < 0x1ffffeUL) || (r_type == R_390_PLT32DBL && val - loc + 0xffffffffULL < 0x1fffffffeULL))) - val = (Elf_Addr) me->core_layout.base + + val = (Elf_Addr) me->mem[MOD_TEXT].base + me->arch.plt_offset + info->plt_offset; val += rela->r_addend - loc; @@ -397,7 +420,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab, case R_390_GOTOFF32: /* 32 bit offset to GOT. */ case R_390_GOTOFF64: /* 64 bit offset to GOT. */ val = val + rela->r_addend - - ((Elf_Addr) me->core_layout.base + me->arch.got_offset); + ((Elf_Addr) me->mem[MOD_TEXT].base + me->arch.got_offset); if (r_type == R_390_GOTOFF16) rc = apply_rela_bits(loc, val, 0, 16, 0, write); else if (r_type == R_390_GOTOFF32) @@ -407,7 +430,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab, break; case R_390_GOTPC: /* 32 bit PC relative offset to GOT. */ case R_390_GOTPCDBL: /* 32 bit PC rel. off. to GOT shifted by 1. */ - val = (Elf_Addr) me->core_layout.base + me->arch.got_offset + + val = (Elf_Addr) me->mem[MOD_TEXT].base + me->arch.got_offset + rela->r_addend - loc; if (r_type == R_390_GOTPC) rc = apply_rela_bits(loc, val, 1, 32, 0, write); @@ -489,7 +512,7 @@ static int module_alloc_ftrace_hotpatch_trampolines(struct module *me, start = module_alloc(numpages * PAGE_SIZE); if (!start) return -ENOMEM; - set_memory_ro((unsigned long)start, numpages); + set_memory_rox((unsigned long)start, numpages); end = start + size; me->arch.trampolines_start = (struct ftrace_hotpatch_trampoline *)start; @@ -515,7 +538,7 @@ int module_finalize(const Elf_Ehdr *hdr, !nospec_disable && me->arch.plt_size) { unsigned int *ij; - ij = me->core_layout.base + me->arch.plt_offset + + ij = me->mem[MOD_TEXT].base + me->arch.plt_offset + me->arch.plt_size - PLT_ENTRY_SIZE; ij[0] = 0xc6000000; /* exrl %r0,.+10 */ ij[1] = 0x0005a7f4; /* j . */ diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c index c9ab971498d6..cf1b6e8a708d 100644 --- a/arch/s390/kernel/perf_cpum_cf.c +++ b/arch/s390/kernel/perf_cpum_cf.c @@ -76,7 +76,6 @@ static inline int ctr_stcctm(enum cpumf_ctr_set set, u64 range, u64 *dest) } struct cpu_cf_events { - struct cpumf_ctr_info info; atomic_t ctr_set[CPUMF_CTR_SET_MAX]; u64 state; /* For perf_event_open SVC */ u64 dev_state; /* For /dev/hwctr */ @@ -95,6 +94,15 @@ static DEFINE_PER_CPU(struct cpu_cf_events, cpu_cf_events); static unsigned int cfdiag_cpu_speed; /* CPU speed for CF_DIAG trailer */ static debug_info_t *cf_dbg; +/* + * The CPU Measurement query counter information instruction contains + * information which varies per machine generation, but is constant and + * does not change when running on a particular machine, such as counter + * first and second version number. This is needed to determine the size + * of counter sets. Extract this information at device driver initialization. + */ +static struct cpumf_ctr_info cpumf_ctr_info; + #define CF_DIAG_CTRSET_DEF 0xfeef /* Counter set header mark */ /* interval in seconds */ @@ -167,11 +175,10 @@ struct cf_trailer_entry { /* CPU-M CF_DIAG trailer (64 byte) */ /* Create the trailer data at the end of a page. */ static void cfdiag_trailer(struct cf_trailer_entry *te) { - struct cpu_cf_events *cpuhw = this_cpu_ptr(&cpu_cf_events); struct cpuid cpuid; - te->cfvn = cpuhw->info.cfvn; /* Counter version numbers */ - te->csvn = cpuhw->info.csvn; + te->cfvn = cpumf_ctr_info.cfvn; /* Counter version numbers */ + te->csvn = cpumf_ctr_info.csvn; get_cpu_id(&cpuid); /* Machine type */ te->mach_type = cpuid.machine; @@ -184,50 +191,60 @@ static void cfdiag_trailer(struct cf_trailer_entry *te) } /* - * Return the maximum possible counter set size (in number of 8 byte counters) - * depending on type and model number. + * The number of counters per counter set varies between machine generations, + * but is constant when running on a particular machine generation. + * Determine each counter set size at device driver initialization and + * retrieve it later. */ -static size_t cpum_cf_ctrset_size(enum cpumf_ctr_set ctrset, - struct cpumf_ctr_info *info) +static size_t cpumf_ctr_setsizes[CPUMF_CTR_SET_MAX]; +static void cpum_cf_make_setsize(enum cpumf_ctr_set ctrset) { size_t ctrset_size = 0; switch (ctrset) { case CPUMF_CTR_SET_BASIC: - if (info->cfvn >= 1) + if (cpumf_ctr_info.cfvn >= 1) ctrset_size = 6; break; case CPUMF_CTR_SET_USER: - if (info->cfvn == 1) + if (cpumf_ctr_info.cfvn == 1) ctrset_size = 6; - else if (info->cfvn >= 3) + else if (cpumf_ctr_info.cfvn >= 3) ctrset_size = 2; break; case CPUMF_CTR_SET_CRYPTO: - if (info->csvn >= 1 && info->csvn <= 5) + if (cpumf_ctr_info.csvn >= 1 && cpumf_ctr_info.csvn <= 5) ctrset_size = 16; - else if (info->csvn == 6 || info->csvn == 7) + else if (cpumf_ctr_info.csvn == 6 || cpumf_ctr_info.csvn == 7) ctrset_size = 20; break; case CPUMF_CTR_SET_EXT: - if (info->csvn == 1) + if (cpumf_ctr_info.csvn == 1) ctrset_size = 32; - else if (info->csvn == 2) + else if (cpumf_ctr_info.csvn == 2) ctrset_size = 48; - else if (info->csvn >= 3 && info->csvn <= 5) + else if (cpumf_ctr_info.csvn >= 3 && cpumf_ctr_info.csvn <= 5) ctrset_size = 128; - else if (info->csvn == 6 || info->csvn == 7) + else if (cpumf_ctr_info.csvn == 6 || cpumf_ctr_info.csvn == 7) ctrset_size = 160; break; case CPUMF_CTR_SET_MT_DIAG: - if (info->csvn > 3) + if (cpumf_ctr_info.csvn > 3) ctrset_size = 48; break; case CPUMF_CTR_SET_MAX: break; } + cpumf_ctr_setsizes[ctrset] = ctrset_size; +} - return ctrset_size; +/* + * Return the maximum possible counter set size (in number of 8 byte counters) + * depending on type and model number. + */ +static size_t cpum_cf_read_setsize(enum cpumf_ctr_set ctrset) +{ + return cpumf_ctr_setsizes[ctrset]; } /* Read a counter set. The counter set number determines the counter set and @@ -248,14 +265,13 @@ static size_t cpum_cf_ctrset_size(enum cpumf_ctr_set ctrset, static size_t cfdiag_getctrset(struct cf_ctrset_entry *ctrdata, int ctrset, size_t room, bool error_ok) { - struct cpu_cf_events *cpuhw = this_cpu_ptr(&cpu_cf_events); size_t ctrset_size, need = 0; int rc = 3; /* Assume write failure */ ctrdata->def = CF_DIAG_CTRSET_DEF; ctrdata->set = ctrset; ctrdata->res1 = 0; - ctrset_size = cpum_cf_ctrset_size(ctrset, &cpuhw->info); + ctrset_size = cpum_cf_read_setsize(ctrset); if (ctrset_size) { /* Save data */ need = ctrset_size * sizeof(u64) + sizeof(*ctrdata); @@ -269,10 +285,6 @@ static size_t cfdiag_getctrset(struct cf_ctrset_entry *ctrdata, int ctrset, need = 0; } - debug_sprintf_event(cf_dbg, 3, - "%s ctrset %d ctrset_size %zu cfvn %d csvn %d" - " need %zd rc %d\n", __func__, ctrset, ctrset_size, - cpuhw->info.cfvn, cpuhw->info.csvn, need, rc); return need; } @@ -377,40 +389,35 @@ static enum cpumf_ctr_set get_counter_set(u64 event) return set; } -static int validate_ctr_version(const struct hw_perf_event *hwc, - enum cpumf_ctr_set set) +static int validate_ctr_version(const u64 config, enum cpumf_ctr_set set) { - struct cpu_cf_events *cpuhw; - int err = 0; u16 mtdiag_ctl; - - cpuhw = &get_cpu_var(cpu_cf_events); + int err = 0; /* check required version for counter sets */ switch (set) { case CPUMF_CTR_SET_BASIC: case CPUMF_CTR_SET_USER: - if (cpuhw->info.cfvn < 1) + if (cpumf_ctr_info.cfvn < 1) err = -EOPNOTSUPP; break; case CPUMF_CTR_SET_CRYPTO: - if ((cpuhw->info.csvn >= 1 && cpuhw->info.csvn <= 5 && - hwc->config > 79) || - (cpuhw->info.csvn >= 6 && hwc->config > 83)) + if ((cpumf_ctr_info.csvn >= 1 && cpumf_ctr_info.csvn <= 5 && + config > 79) || (cpumf_ctr_info.csvn >= 6 && config > 83)) err = -EOPNOTSUPP; break; case CPUMF_CTR_SET_EXT: - if (cpuhw->info.csvn < 1) + if (cpumf_ctr_info.csvn < 1) err = -EOPNOTSUPP; - if ((cpuhw->info.csvn == 1 && hwc->config > 159) || - (cpuhw->info.csvn == 2 && hwc->config > 175) || - (cpuhw->info.csvn >= 3 && cpuhw->info.csvn <= 5 - && hwc->config > 255) || - (cpuhw->info.csvn >= 6 && hwc->config > 287)) + if ((cpumf_ctr_info.csvn == 1 && config > 159) || + (cpumf_ctr_info.csvn == 2 && config > 175) || + (cpumf_ctr_info.csvn >= 3 && cpumf_ctr_info.csvn <= 5 && + config > 255) || + (cpumf_ctr_info.csvn >= 6 && config > 287)) err = -EOPNOTSUPP; break; case CPUMF_CTR_SET_MT_DIAG: - if (cpuhw->info.csvn <= 3) + if (cpumf_ctr_info.csvn <= 3) err = -EOPNOTSUPP; /* * MT-diagnostic counters are read-only. The counter set @@ -425,35 +432,15 @@ static int validate_ctr_version(const struct hw_perf_event *hwc, * counter set is enabled and active. */ mtdiag_ctl = cpumf_ctr_ctl[CPUMF_CTR_SET_MT_DIAG]; - if (!((cpuhw->info.auth_ctl & mtdiag_ctl) && - (cpuhw->info.enable_ctl & mtdiag_ctl) && - (cpuhw->info.act_ctl & mtdiag_ctl))) + if (!((cpumf_ctr_info.auth_ctl & mtdiag_ctl) && + (cpumf_ctr_info.enable_ctl & mtdiag_ctl) && + (cpumf_ctr_info.act_ctl & mtdiag_ctl))) err = -EOPNOTSUPP; break; case CPUMF_CTR_SET_MAX: err = -EOPNOTSUPP; } - put_cpu_var(cpu_cf_events); - return err; -} - -static int validate_ctr_auth(const struct hw_perf_event *hwc) -{ - struct cpu_cf_events *cpuhw; - int err = 0; - - cpuhw = &get_cpu_var(cpu_cf_events); - - /* Check authorization for cpu counter sets. - * If the particular CPU counter set is not authorized, - * return with -ENOENT in order to fall back to other - * PMUs that might suffice the event request. - */ - if (!(hwc->config_base & cpuhw->info.auth_ctl)) - err = -ENOENT; - - put_cpu_var(cpu_cf_events); return err; } @@ -471,13 +458,10 @@ static void cpumf_pmu_enable(struct pmu *pmu) return; err = lcctl(cpuhw->state | cpuhw->dev_state); - if (err) { - pr_err("Enabling the performance measuring unit " - "failed with rc=%x\n", err); - return; - } - - cpuhw->flags |= PMU_F_ENABLED; + if (err) + pr_err("Enabling the performance measuring unit failed with rc=%x\n", err); + else + cpuhw->flags |= PMU_F_ENABLED; } /* @@ -497,13 +481,10 @@ static void cpumf_pmu_disable(struct pmu *pmu) inactive = cpuhw->state & ~((1 << CPUMF_LCCTL_ENABLE_SHIFT) - 1); inactive |= cpuhw->dev_state; err = lcctl(inactive); - if (err) { - pr_err("Disabling the performance measuring unit " - "failed with rc=%x\n", err); - return; - } - - cpuhw->flags &= ~PMU_F_ENABLED; + if (err) + pr_err("Disabling the performance measuring unit failed with rc=%x\n", err); + else + cpuhw->flags &= ~PMU_F_ENABLED; } #define PMC_INIT 0UL @@ -515,8 +496,6 @@ static void cpum_cf_setup_cpu(void *flags) switch ((unsigned long)flags) { case PMC_INIT: - memset(&cpuhw->info, 0, sizeof(cpuhw->info)); - qctri(&cpuhw->info); cpuhw->flags |= PMU_F_RESERVED; break; @@ -602,7 +581,6 @@ static int __hw_perf_event_init(struct perf_event *event, unsigned int type) struct perf_event_attr *attr = &event->attr; struct hw_perf_event *hwc = &event->hw; enum cpumf_ctr_set set; - int err = 0; u64 ev; switch (type) { @@ -678,12 +656,15 @@ static int __hw_perf_event_init(struct perf_event *event, unsigned int type) cpumf_hw_inuse(); event->destroy = hw_perf_event_destroy; - /* Finally, validate version and authorization of the counter set */ - err = validate_ctr_auth(hwc); - if (!err) - err = validate_ctr_version(hwc, set); - - return err; + /* + * Finally, validate version and authorization of the counter set. + * If the particular CPU counter set is not authorized, + * return with -ENOENT in order to fall back to other + * PMUs that might suffice the event request. + */ + if (!(hwc->config_base & cpumf_ctr_info.auth_ctl)) + return -ENOENT; + return validate_ctr_version(hwc->config, set); } /* Events CPU_CYLCES and INSTRUCTIONS can be submitted with two different @@ -983,7 +964,7 @@ static void cpumf_measurement_alert(struct ext_code ext_code, /* counter authorization change alert */ if (alert & CPU_MF_INT_CF_CACA) - qctri(&cpuhw->info); + qctri(&cpumf_ctr_info); /* loss of counter data alert */ if (alert & CPU_MF_INT_CF_LCDA) @@ -1000,9 +981,14 @@ static int __init cpumf_pmu_init(void) { int rc; - if (!cpum_cf_avail()) + /* Extract counter measurement facility information */ + if (!cpum_cf_avail() || qctri(&cpumf_ctr_info)) return -ENODEV; + /* Determine and store counter set sizes for later reference */ + for (rc = CPUMF_CTR_SET_BASIC; rc < CPUMF_CTR_SET_MAX; ++rc) + cpum_cf_make_setsize(rc); + /* * Clear bit 15 of cr0 to unauthorize problem-state to * extract measurement counters @@ -1269,28 +1255,26 @@ static int cfset_all_start(struct cfset_request *req) */ static size_t cfset_needspace(unsigned int sets) { - struct cpu_cf_events *cpuhw = get_cpu_ptr(&cpu_cf_events); size_t bytes = 0; int i; for (i = CPUMF_CTR_SET_BASIC; i < CPUMF_CTR_SET_MAX; ++i) { if (!(sets & cpumf_ctr_ctl[i])) continue; - bytes += cpum_cf_ctrset_size(i, &cpuhw->info) * sizeof(u64) + + bytes += cpum_cf_read_setsize(i) * sizeof(u64) + sizeof(((struct s390_ctrset_setdata *)0)->set) + sizeof(((struct s390_ctrset_setdata *)0)->no_cnts); } bytes = sizeof(((struct s390_ctrset_read *)0)->no_cpus) + nr_cpu_ids * (bytes + sizeof(((struct s390_ctrset_cpudata *)0)->cpu_nr) + sizeof(((struct s390_ctrset_cpudata *)0)->no_sets)); - put_cpu_ptr(&cpu_cf_events); return bytes; } static int cfset_all_copy(unsigned long arg, cpumask_t *mask) { struct s390_ctrset_read __user *ctrset_read; - unsigned int cpu, cpus, rc; + unsigned int cpu, cpus, rc = 0; void __user *uptr; ctrset_read = (struct s390_ctrset_read __user *)arg; @@ -1304,17 +1288,20 @@ static int cfset_all_copy(unsigned long arg, cpumask_t *mask) rc |= put_user(cpuhw->sets, &ctrset_cpudata->no_sets); rc |= copy_to_user(ctrset_cpudata->data, cpuhw->data, cpuhw->used); - if (rc) - return -EFAULT; + if (rc) { + rc = -EFAULT; + goto out; + } uptr += sizeof(struct s390_ctrset_cpudata) + cpuhw->used; cond_resched(); } cpus = cpumask_weight(mask); if (put_user(cpus, &ctrset_read->no_cpus)) - return -EFAULT; - debug_sprintf_event(cf_dbg, 4, "%s copied %ld\n", __func__, + rc = -EFAULT; +out: + debug_sprintf_event(cf_dbg, 4, "%s rc %d copied %ld\n", __func__, rc, uptr - (void __user *)ctrset_read->data); - return 0; + return rc; } static size_t cfset_cpuset_read(struct s390_ctrset_setdata *p, int ctrset, @@ -1354,7 +1341,7 @@ static void cfset_cpu_read(void *parm) if (!(p->sets & cpumf_ctr_ctl[set])) continue; /* Counter set not in list */ - set_size = cpum_cf_ctrset_size(set, &cpuhw->info); + set_size = cpum_cf_read_setsize(set); space = sizeof(cpuhw->data) - cpuhw->used; space = cfset_cpuset_read(sp, set, set_size, space); if (space) { @@ -1385,14 +1372,10 @@ static int cfset_all_read(unsigned long arg, struct cfset_request *req) static long cfset_ioctl_read(unsigned long arg, struct cfset_request *req) { - struct s390_ctrset_read read; int ret = -ENODATA; - if (req && req->ctrset) { - if (copy_from_user(&read, (char __user *)arg, sizeof(read))) - return -EFAULT; + if (req && req->ctrset) ret = cfset_all_read(arg, req); - } return ret; } @@ -1569,16 +1552,13 @@ static void cfdiag_read(struct perf_event *event) static int get_authctrsets(void) { - struct cpu_cf_events *cpuhw; unsigned long auth = 0; enum cpumf_ctr_set i; - cpuhw = &get_cpu_var(cpu_cf_events); for (i = CPUMF_CTR_SET_BASIC; i < CPUMF_CTR_SET_MAX; ++i) { - if (cpuhw->info.auth_ctl & cpumf_ctr_ctl[i]) + if (cpumf_ctr_info.auth_ctl & cpumf_ctr_ctl[i]) auth |= cpumf_ctr_ctl[i]; } - put_cpu_var(cpu_cf_events); return auth; } @@ -1716,7 +1696,7 @@ static size_t cfdiag_maxsize(struct cpumf_ctr_info *info) enum cpumf_ctr_set i; for (i = CPUMF_CTR_SET_BASIC; i < CPUMF_CTR_SET_MAX; ++i) { - size_t size = cpum_cf_ctrset_size(i, info); + size_t size = cpum_cf_read_setsize(i); if (size) max_size += size * sizeof(u64) + @@ -1750,16 +1730,12 @@ static void cfdiag_get_cpu_speed(void) static int cfset_init(void) { - struct cpumf_ctr_info info; size_t need; int rc; - if (qctri(&info)) - return -ENODEV; - cfdiag_get_cpu_speed(); /* Make sure the counter set data fits into predefined buffer. */ - need = cfdiag_maxsize(&info); + need = cfdiag_maxsize(&cpumf_ctr_info); if (need > sizeof(((struct cpu_cf_events *)0)->start)) { pr_err("Insufficient memory for PMU(cpum_cf_diag) need=%zu\n", need); diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c index e7b867e2f73f..7ef72f5ff52e 100644 --- a/arch/s390/kernel/perf_cpum_sf.c +++ b/arch/s390/kernel/perf_cpum_sf.c @@ -882,10 +882,6 @@ static int __hw_perf_event_init(struct perf_event *event) SAMPL_FLAGS(hwc) |= PERF_CPUM_SF_DIAG_MODE; } - /* Check and set other sampling flags */ - if (attr->config1 & PERF_CPUM_SF_FULL_BLOCKS) - SAMPL_FLAGS(hwc) |= PERF_CPUM_SF_FULL_BLOCKS; - err = __hw_perf_event_init_rate(event, &si); if (err) goto out; @@ -1293,11 +1289,8 @@ static inline __uint128_t __cdsg(__uint128_t *ptr, __uint128_t old, __uint128_t * The sampling buffer position are retrieved and saved in the TEAR_REG * register of the specified perf event. * - * Only full sample-data-blocks are processed. Specify the flash_all flag - * to also walk through partially filled sample-data-blocks. It is ignored - * if PERF_CPUM_SF_FULL_BLOCKS is set. The PERF_CPUM_SF_FULL_BLOCKS flag - * enforces the processing of full sample-data-blocks only (trailer entries - * with the block-full-indicator bit set). + * Only full sample-data-blocks are processed. Specify the flush_all flag + * to also walk through partially filled sample-data-blocks. */ static void hw_perf_event_update(struct perf_event *event, int flush_all) { @@ -1315,9 +1308,6 @@ static void hw_perf_event_update(struct perf_event *event, int flush_all) if (SAMPL_DIAG_MODE(&event->hw)) return; - if (flush_all && SDB_FULL_BLOCKS(hwc)) - flush_all = 0; - sdbt = (unsigned long *) TEAR_REG(hwc); done = event_overflow = sampl_overflow = num_sdb = 0; while (!done) { diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index 67df64ef4839..87ca3a727604 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -136,12 +136,12 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args) p->thread.last_break = 1; frame->sf.back_chain = 0; - frame->sf.gprs[5] = (unsigned long)frame + sizeof(struct stack_frame); - frame->sf.gprs[6] = (unsigned long)p; + frame->sf.gprs[11 - 6] = (unsigned long)&frame->childregs; + frame->sf.gprs[12 - 6] = (unsigned long)p; /* new return point is ret_from_fork */ - frame->sf.gprs[8] = (unsigned long)ret_from_fork; + frame->sf.gprs[14 - 6] = (unsigned long)ret_from_fork; /* fake return stack for resume(), don't go back to schedule */ - frame->sf.gprs[9] = (unsigned long)frame; + frame->sf.gprs[15 - 6] = (unsigned long)frame; /* Store access registers to kernel stack of new process. */ if (unlikely(args->fn)) { @@ -149,8 +149,6 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args) memset(&frame->childregs, 0, sizeof(struct pt_regs)); frame->childregs.psw.mask = PSW_KERNEL_BITS | PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK; - frame->childregs.psw.addr = - (unsigned long)__ret_from_fork; frame->childregs.gprs[9] = (unsigned long)args->fn; frame->childregs.gprs[10] = (unsigned long)args->fn_arg; frame->childregs.orig_gpr2 = -1; diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c index a194611ba88c..0a999c8226d7 100644 --- a/arch/s390/kernel/processor.c +++ b/arch/s390/kernel/processor.c @@ -364,21 +364,3 @@ const struct seq_operations cpuinfo_op = { .stop = c_stop, .show = show_cpuinfo, }; - -int s390_isolate_bp(void) -{ - if (!test_facility(82)) - return -EOPNOTSUPP; - set_thread_flag(TIF_ISOLATE_BP); - return 0; -} -EXPORT_SYMBOL(s390_isolate_bp); - -int s390_isolate_bp_guest(void) -{ - if (!test_facility(82)) - return -EOPNOTSUPP; - set_thread_flag(TIF_ISOLATE_BP_GUEST); - return 0; -} -EXPORT_SYMBOL(s390_isolate_bp_guest); diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index cf9659e13f03..ea244a73efad 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -474,9 +474,7 @@ long arch_ptrace(struct task_struct *child, long request, } return 0; case PTRACE_GET_LAST_BREAK: - put_user(child->thread.last_break, - (unsigned long __user *) data); - return 0; + return put_user(child->thread.last_break, (unsigned long __user *)data); case PTRACE_ENABLE_TE: if (!MACHINE_HAS_TE) return -EIO; @@ -824,9 +822,7 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, } return 0; case PTRACE_GET_LAST_BREAK: - put_user(child->thread.last_break, - (unsigned int __user *) data); - return 0; + return put_user(child->thread.last_break, (unsigned int __user *)data); } return compat_ptrace_request(child, request, addr, data); } diff --git a/arch/s390/kernel/reipl.S b/arch/s390/kernel/reipl.S index 4a22163962eb..88087a32ebc6 100644 --- a/arch/s390/kernel/reipl.S +++ b/arch/s390/kernel/reipl.S @@ -19,7 +19,7 @@ # r2 = Function to be called after store status # r3 = Parameter for function # -ENTRY(store_status) +SYM_CODE_START(store_status) /* Save register one and load save area base */ stg %r1,__LC_SAVE_AREA_RESTART /* General purpose registers */ @@ -61,7 +61,7 @@ ENTRY(store_status) stpx 0(%r1) /* Clock comparator - seven bytes */ lghi %r1,__LC_CLOCK_COMP_SAVE_AREA - larl %r4,.Lclkcmp + larl %r4,clkcmp stckc 0(%r4) mvc 1(7,%r1),1(%r4) /* Program status word */ @@ -73,9 +73,9 @@ ENTRY(store_status) lgr %r9,%r2 lgr %r2,%r3 BR_EX %r9 -ENDPROC(store_status) +SYM_CODE_END(store_status) .section .bss - .align 8 -.Lclkcmp: .quad 0x0000000000000000 + .balign 8 +SYM_DATA_LOCAL(clkcmp, .quad 0x0000000000000000) .previous diff --git a/arch/s390/kernel/relocate_kernel.S b/arch/s390/kernel/relocate_kernel.S index a9a1a6f45375..0ae297c82afd 100644 --- a/arch/s390/kernel/relocate_kernel.S +++ b/arch/s390/kernel/relocate_kernel.S @@ -26,53 +26,51 @@ */ .text -ENTRY(relocate_kernel) - basr %r13,0 # base address - .base: - lghi %r7,PAGE_SIZE # load PAGE_SIZE in r7 - lghi %r9,PAGE_SIZE # load PAGE_SIZE in r9 - lg %r5,0(%r2) # read another word for indirection page - aghi %r2,8 # increment pointer - tml %r5,0x1 # is it a destination page? - je .indir_check # NO, goto "indir_check" - lgr %r6,%r5 # r6 = r5 - nill %r6,0xf000 # mask it out and... - j .base # ...next iteration - .indir_check: - tml %r5,0x2 # is it a indirection page? - je .done_test # NO, goto "done_test" - nill %r5,0xf000 # YES, mask out, - lgr %r2,%r5 # move it into the right register, - j .base # and read next... - .done_test: - tml %r5,0x4 # is it the done indicator? - je .source_test # NO! Well, then it should be the source indicator... - j .done # ok, lets finish it here... - .source_test: - tml %r5,0x8 # it should be a source indicator... - je .base # NO, ignore it... - lgr %r8,%r5 # r8 = r5 - nill %r8,0xf000 # masking - 0: mvcle %r6,%r8,0x0 # copy PAGE_SIZE bytes from r8 to r6 - pad with 0 - jo 0b - j .base - .done: - lgr %r0,%r4 # subcode - cghi %r3,0 - je .diag - la %r4,load_psw-.base(%r13) # load psw-address into the register - o %r3,4(%r4) # or load address into psw - st %r3,4(%r4) - mvc 0(8,%r0),0(%r4) # copy psw to absolute address 0 - .diag: - diag %r0,%r0,0x308 -ENDPROC(relocate_kernel) +SYM_CODE_START(relocate_kernel) + basr %r13,0 # base address +.base: + lghi %r7,PAGE_SIZE # load PAGE_SIZE in r7 + lghi %r9,PAGE_SIZE # load PAGE_SIZE in r9 + lg %r5,0(%r2) # read another word for indirection page + aghi %r2,8 # increment pointer + tml %r5,0x1 # is it a destination page? + je .indir_check # NO, goto "indir_check" + lgr %r6,%r5 # r6 = r5 + nill %r6,0xf000 # mask it out and... + j .base # ...next iteration +.indir_check: + tml %r5,0x2 # is it a indirection page? + je .done_test # NO, goto "done_test" + nill %r5,0xf000 # YES, mask out, + lgr %r2,%r5 # move it into the right register, + j .base # and read next... +.done_test: + tml %r5,0x4 # is it the done indicator? + je .source_test # NO! Well, then it should be the source indicator... + j .done # ok, lets finish it here... +.source_test: + tml %r5,0x8 # it should be a source indicator... + je .base # NO, ignore it... + lgr %r8,%r5 # r8 = r5 + nill %r8,0xf000 # masking +0: mvcle %r6,%r8,0x0 # copy PAGE_SIZE bytes from r8 to r6 - pad with 0 + jo 0b + j .base +.done: + lgr %r0,%r4 # subcode + cghi %r3,0 + je .diag + la %r4,load_psw-.base(%r13) # load psw-address into the register + o %r3,4(%r4) # or load address into psw + st %r3,4(%r4) + mvc 0(8,%r0),0(%r4) # copy psw to absolute address 0 +.diag: + diag %r0,%r0,0x308 +SYM_CODE_END(relocate_kernel) - .align 8 - load_psw: - .long 0x00080000,0x80000000 - relocate_kernel_end: - .align 8 - .globl relocate_kernel_len - relocate_kernel_len: - .quad relocate_kernel_end - relocate_kernel + .balign 8 +SYM_DATA_START_LOCAL(load_psw) + .long 0x00080000,0x80000000 +SYM_DATA_END_LABEL(load_psw, SYM_L_LOCAL, relocate_kernel_end) + .balign 8 +SYM_DATA(relocate_kernel_len, .quad relocate_kernel_end - relocate_kernel) diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 8ec5cdf9dadc..fe10da1a271e 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -74,7 +74,7 @@ #include <asm/numa.h> #include <asm/alternative.h> #include <asm/nospec-branch.h> -#include <asm/mem_detect.h> +#include <asm/physmem_info.h> #include <asm/maccess.h> #include <asm/uv.h> #include <asm/asm-offsets.h> @@ -147,14 +147,10 @@ static u32 __amode31_ref *__ctl_duct = __ctl_duct_amode31; int __bootdata(noexec_disabled); unsigned long __bootdata(ident_map_size); -struct mem_detect_info __bootdata(mem_detect); -struct initrd_data __bootdata(initrd_data); -unsigned long __bootdata(pgalloc_pos); -unsigned long __bootdata(pgalloc_end); -unsigned long __bootdata(pgalloc_low); +struct physmem_info __bootdata(physmem_info); unsigned long __bootdata_preserved(__kaslr_offset); -unsigned long __bootdata(__amode31_base); +int __bootdata_preserved(__kaslr_enabled); unsigned int __bootdata_preserved(zlib_dfltcc_support); EXPORT_SYMBOL(zlib_dfltcc_support); u64 __bootdata_preserved(stfle_fac_list[16]); @@ -385,39 +381,27 @@ void stack_free(unsigned long stack) #endif } -int __init arch_early_irq_init(void) +void __init __noreturn arch_call_rest_init(void) { - unsigned long stack; - - stack = __get_free_pages(GFP_KERNEL, THREAD_SIZE_ORDER); - if (!stack) - panic("Couldn't allocate async stack"); - S390_lowcore.async_stack = stack + STACK_INIT_OFFSET; - return 0; + smp_reinit_ipl_cpu(); + rest_init(); } -void __init arch_call_rest_init(void) +static unsigned long __init stack_alloc_early(void) { unsigned long stack; - smp_reinit_ipl_cpu(); - stack = stack_alloc(); - if (!stack) - panic("Couldn't allocate kernel stack"); - current->stack = (void *) stack; -#ifdef CONFIG_VMAP_STACK - current->stack_vm_area = (void *) stack; -#endif - set_task_stack_end_magic(current); - stack += STACK_INIT_OFFSET; - S390_lowcore.kernel_stack = stack; - call_on_stack_noreturn(rest_init, stack); + stack = (unsigned long)memblock_alloc(THREAD_SIZE, THREAD_SIZE); + if (!stack) { + panic("%s: Failed to allocate %lu bytes align=0x%lx\n", + __func__, THREAD_SIZE, THREAD_SIZE); + } + return stack; } static void __init setup_lowcore(void) { struct lowcore *lc, *abs_lc; - unsigned long mcck_stack; /* * Setup lowcore for boot cpu @@ -441,8 +425,6 @@ static void __init setup_lowcore(void) lc->io_new_psw.mask = PSW_KERNEL_BITS | PSW_MASK_MCHECK; lc->io_new_psw.addr = (unsigned long) io_int_handler; lc->clock_comparator = clock_comparator_max; - lc->nodat_stack = ((unsigned long) &init_thread_union) - + THREAD_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs); lc->current_task = (unsigned long)&init_task; lc->lpp = LPP_MAGIC; lc->machine_flags = S390_lowcore.machine_flags; @@ -455,17 +437,15 @@ static void __init setup_lowcore(void) lc->steal_timer = S390_lowcore.steal_timer; lc->last_update_timer = S390_lowcore.last_update_timer; lc->last_update_clock = S390_lowcore.last_update_clock; - /* * Allocate the global restart stack which is the same for - * all CPUs in cast *one* of them does a PSW restart. + * all CPUs in case *one* of them does a PSW restart. */ - restart_stack = memblock_alloc(THREAD_SIZE, THREAD_SIZE); - if (!restart_stack) - panic("%s: Failed to allocate %lu bytes align=0x%lx\n", - __func__, THREAD_SIZE, THREAD_SIZE); - restart_stack += STACK_INIT_OFFSET; - + restart_stack = (void *)(stack_alloc_early() + STACK_INIT_OFFSET); + lc->mcck_stack = stack_alloc_early() + STACK_INIT_OFFSET; + lc->async_stack = stack_alloc_early() + STACK_INIT_OFFSET; + lc->nodat_stack = stack_alloc_early() + STACK_INIT_OFFSET; + lc->kernel_stack = S390_lowcore.kernel_stack; /* * Set up PSW restart to call ipl.c:do_restart(). Copy the relevant * restart data to the absolute zero lowcore. This is necessary if @@ -476,13 +456,6 @@ static void __init setup_lowcore(void) lc->restart_data = 0; lc->restart_source = -1U; __ctl_store(lc->cregs_save_area, 0, 15); - - mcck_stack = (unsigned long)memblock_alloc(THREAD_SIZE, THREAD_SIZE); - if (!mcck_stack) - panic("%s: Failed to allocate %lu bytes align=0x%lx\n", - __func__, THREAD_SIZE, THREAD_SIZE); - lc->mcck_stack = mcck_stack + STACK_INIT_OFFSET; - lc->spinlock_lockval = arch_spin_lockval(0); lc->spinlock_index = 0; arch_spin_lock_setup(0); @@ -635,7 +608,11 @@ static struct notifier_block kdump_mem_nb = { */ static void __init reserve_pgtables(void) { - memblock_reserve(pgalloc_pos, pgalloc_end - pgalloc_pos); + unsigned long start, end; + struct reserved_range *range; + + for_each_physmem_reserved_type_range(RR_VMEM, range, &start, &end) + memblock_reserve(start, end - start); } /* @@ -712,13 +689,13 @@ static void __init reserve_crashkernel(void) */ static void __init reserve_initrd(void) { -#ifdef CONFIG_BLK_DEV_INITRD - if (!initrd_data.start || !initrd_data.size) + unsigned long addr, size; + + if (!IS_ENABLED(CONFIG_BLK_DEV_INITRD) || !get_physmem_reserved(RR_INITRD, &addr, &size)) return; - initrd_start = (unsigned long)__va(initrd_data.start); - initrd_end = initrd_start + initrd_data.size; - memblock_reserve(initrd_data.start, initrd_data.size); -#endif + initrd_start = (unsigned long)__va(addr); + initrd_end = initrd_start + size; + memblock_reserve(addr, size); } /* @@ -730,72 +707,40 @@ static void __init reserve_certificate_list(void) memblock_reserve(ipl_cert_list_addr, ipl_cert_list_size); } -static void __init reserve_mem_detect_info(void) +static void __init reserve_physmem_info(void) { - unsigned long start, size; + unsigned long addr, size; - get_mem_detect_reserved(&start, &size); - if (size) - memblock_reserve(start, size); + if (get_physmem_reserved(RR_MEM_DETECT_EXTENDED, &addr, &size)) + memblock_reserve(addr, size); } -static void __init free_mem_detect_info(void) +static void __init free_physmem_info(void) { - unsigned long start, size; - - get_mem_detect_reserved(&start, &size); - if (size) - memblock_phys_free(start, size); -} + unsigned long addr, size; -static const char * __init get_mem_info_source(void) -{ - switch (mem_detect.info_source) { - case MEM_DETECT_SCLP_STOR_INFO: - return "sclp storage info"; - case MEM_DETECT_DIAG260: - return "diag260"; - case MEM_DETECT_SCLP_READ_INFO: - return "sclp read info"; - case MEM_DETECT_BIN_SEARCH: - return "binary search"; - } - return "none"; + if (get_physmem_reserved(RR_MEM_DETECT_EXTENDED, &addr, &size)) + memblock_phys_free(addr, size); } -static void __init memblock_add_mem_detect_info(void) +static void __init memblock_add_physmem_info(void) { unsigned long start, end; int i; pr_debug("physmem info source: %s (%hhd)\n", - get_mem_info_source(), mem_detect.info_source); + get_physmem_info_source(), physmem_info.info_source); /* keep memblock lists close to the kernel */ memblock_set_bottom_up(true); - for_each_mem_detect_usable_block(i, &start, &end) + for_each_physmem_usable_range(i, &start, &end) memblock_add(start, end - start); - for_each_mem_detect_block(i, &start, &end) + for_each_physmem_online_range(i, &start, &end) memblock_physmem_add(start, end - start); memblock_set_bottom_up(false); memblock_set_node(0, ULONG_MAX, &memblock.memory, 0); } /* - * Check for initrd being in usable memory - */ -static void __init check_initrd(void) -{ -#ifdef CONFIG_BLK_DEV_INITRD - if (initrd_data.start && initrd_data.size && - !memblock_is_region_memory(initrd_data.start, initrd_data.size)) { - pr_err("The initial RAM disk does not fit into the memory\n"); - memblock_phys_free(initrd_data.start, initrd_data.size); - initrd_start = initrd_end = 0; - } -#endif -} - -/* * Reserve memory used for lowcore/command line/kernel image. */ static void __init reserve_kernel(void) @@ -803,7 +748,7 @@ static void __init reserve_kernel(void) memblock_reserve(0, STARTUP_NORMAL_OFFSET); memblock_reserve(OLDMEM_BASE, sizeof(unsigned long)); memblock_reserve(OLDMEM_SIZE, sizeof(unsigned long)); - memblock_reserve(__amode31_base, __eamode31 - __samode31); + memblock_reserve(physmem_info.reserved[RR_AMODE31].start, __eamode31 - __samode31); memblock_reserve(__pa(sclp_early_sccb), EXT_SCCB_READ_SCP); memblock_reserve(__pa(_stext), _end - _stext); } @@ -825,13 +770,13 @@ static void __init setup_memory(void) static void __init relocate_amode31_section(void) { unsigned long amode31_size = __eamode31 - __samode31; - long amode31_offset = __amode31_base - __samode31; + long amode31_offset = physmem_info.reserved[RR_AMODE31].start - __samode31; long *ptr; pr_info("Relocating AMODE31 section of size 0x%08lx\n", amode31_size); /* Move original AMODE31 section to the new one */ - memmove((void *)__amode31_base, (void *)__samode31, amode31_size); + memmove((void *)physmem_info.reserved[RR_AMODE31].start, (void *)__samode31, amode31_size); /* Zero out the old AMODE31 section to catch invalid accesses within it */ memset((void *)__samode31, 0, amode31_size); @@ -997,14 +942,14 @@ void __init setup_arch(char **cmdline_p) reserve_kernel(); reserve_initrd(); reserve_certificate_list(); - reserve_mem_detect_info(); + reserve_physmem_info(); memblock_set_current_limit(ident_map_size); memblock_allow_resize(); /* Get information about *all* installed memory */ - memblock_add_mem_detect_info(); + memblock_add_physmem_info(); - free_mem_detect_info(); + free_physmem_info(); setup_memory_end(); memblock_dump_all(); setup_memory(); @@ -1017,7 +962,6 @@ void __init setup_arch(char **cmdline_p) if (MACHINE_HAS_EDAT2) hugetlb_cma_reserve(PUD_SHIFT - PAGE_SHIFT); - check_initrd(); reserve_crashkernel(); #ifdef CONFIG_CRASH_DUMP /* diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index d4888453bbf8..726de4f4df01 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -280,9 +280,8 @@ static void pcpu_attach_task(struct pcpu *pcpu, struct task_struct *tsk) cpu = pcpu - pcpu_devices; lc = lowcore_ptr[cpu]; - lc->kernel_stack = (unsigned long) task_stack_page(tsk) - + THREAD_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs); - lc->current_task = (unsigned long) tsk; + lc->kernel_stack = (unsigned long)task_stack_page(tsk) + STACK_INIT_OFFSET; + lc->current_task = (unsigned long)tsk; lc->lpp = LPP_MAGIC; lc->current_pid = tsk->pid; lc->user_timer = tsk->thread.user_timer; @@ -348,7 +347,6 @@ static void pcpu_delegate(struct pcpu *pcpu, abs_lc->restart_source = source_cpu; put_abs_lowcore(abs_lc); } - __bpon(); asm volatile( "0: sigp 0,%0,%2 # sigp restart to target cpu\n" " brc 2,0b # busy, try again\n" @@ -553,7 +551,7 @@ void arch_send_call_function_single_ipi(int cpu) * it goes straight through and wastes no time serializing * anything. Worst case is that we lose a reschedule ... */ -void smp_send_reschedule(int cpu) +void arch_smp_send_reschedule(int cpu) { pcpu_ec_call(pcpu_devices + cpu, ec_schedule); } @@ -986,7 +984,6 @@ void __cpu_die(unsigned int cpu) void __noreturn cpu_die(void) { idle_task_exit(); - __bpon(); pcpu_sigp_retry(pcpu_devices + smp_processor_id(), SIGP_STOP, 0); for (;;) ; } @@ -1227,11 +1224,17 @@ static DEVICE_ATTR_WO(rescan); static int __init s390_smp_init(void) { + struct device *dev_root; int cpu, rc = 0; - rc = device_create_file(cpu_subsys.dev_root, &dev_attr_rescan); - if (rc) - return rc; + dev_root = bus_get_dev_root(&cpu_subsys); + if (dev_root) { + rc = device_create_file(dev_root, &dev_attr_rescan); + put_device(dev_root); + if (rc) + return rc; + } + for_each_present_cpu(cpu) { rc = smp_add_present_cpu(cpu); if (rc) @@ -1296,9 +1299,9 @@ int __init smp_reinit_ipl_cpu(void) local_mcck_enable(); local_irq_restore(flags); - free_pages(lc_ipl->async_stack - STACK_INIT_OFFSET, THREAD_SIZE_ORDER); memblock_free_late(__pa(lc_ipl->mcck_stack - STACK_INIT_OFFSET), THREAD_SIZE); + memblock_free_late(__pa(lc_ipl->async_stack - STACK_INIT_OFFSET), THREAD_SIZE); + memblock_free_late(__pa(lc_ipl->nodat_stack - STACK_INIT_OFFSET), THREAD_SIZE); memblock_free_late(__pa(lc_ipl), sizeof(*lc_ipl)); - return 0; } diff --git a/arch/s390/kernel/syscalls/syscall.tbl b/arch/s390/kernel/syscalls/syscall.tbl index 799147658dee..b68f47541169 100644 --- a/arch/s390/kernel/syscalls/syscall.tbl +++ b/arch/s390/kernel/syscalls/syscall.tbl @@ -449,7 +449,7 @@ 444 common landlock_create_ruleset sys_landlock_create_ruleset sys_landlock_create_ruleset 445 common landlock_add_rule sys_landlock_add_rule sys_landlock_add_rule 446 common landlock_restrict_self sys_landlock_restrict_self sys_landlock_restrict_self -# 447 reserved for memfd_secret +447 common memfd_secret sys_memfd_secret sys_memfd_secret 448 common process_mrelease sys_process_mrelease sys_process_mrelease 449 common futex_waitv sys_futex_waitv sys_futex_waitv 450 common set_mempolicy_home_node sys_set_mempolicy_home_node sys_set_mempolicy_home_node diff --git a/arch/s390/kernel/text_amode31.S b/arch/s390/kernel/text_amode31.S index e0f01ce251f5..14c6d25c035f 100644 --- a/arch/s390/kernel/text_amode31.S +++ b/arch/s390/kernel/text_amode31.S @@ -27,7 +27,7 @@ /* * int _diag14_amode31(unsigned long rx, unsigned long ry1, unsigned long subcode) */ -ENTRY(_diag14_amode31) +SYM_FUNC_START(_diag14_amode31) lgr %r1,%r2 lgr %r2,%r3 lgr %r3,%r4 @@ -42,12 +42,12 @@ ENTRY(_diag14_amode31) lgfr %r2,%r5 BR_EX_AMODE31_r14 EX_TABLE_AMODE31(.Ldiag14_ex, .Ldiag14_fault) -ENDPROC(_diag14_amode31) +SYM_FUNC_END(_diag14_amode31) /* * int _diag210_amode31(struct diag210 *addr) */ -ENTRY(_diag210_amode31) +SYM_FUNC_START(_diag210_amode31) lgr %r1,%r2 lhi %r2,-1 sam31 @@ -60,12 +60,12 @@ ENTRY(_diag210_amode31) lgfr %r2,%r2 BR_EX_AMODE31_r14 EX_TABLE_AMODE31(.Ldiag210_ex, .Ldiag210_fault) -ENDPROC(_diag210_amode31) +SYM_FUNC_END(_diag210_amode31) /* * int diag8c(struct diag8c *addr, struct ccw_dev_id *devno, size_t len) */ -ENTRY(_diag8c_amode31) +SYM_FUNC_START(_diag8c_amode31) llgf %r3,0(%r3) sam31 diag %r2,%r4,0x8c @@ -74,11 +74,11 @@ ENTRY(_diag8c_amode31) lgfr %r2,%r3 BR_EX_AMODE31_r14 EX_TABLE_AMODE31(.Ldiag8c_ex, .Ldiag8c_ex) -ENDPROC(_diag8c_amode31) +SYM_FUNC_END(_diag8c_amode31) /* * int _diag26c_amode31(void *req, void *resp, enum diag26c_sc subcode) */ -ENTRY(_diag26c_amode31) +SYM_FUNC_START(_diag26c_amode31) lghi %r5,-EOPNOTSUPP sam31 diag %r2,%r4,0x26c @@ -87,42 +87,42 @@ ENTRY(_diag26c_amode31) lgfr %r2,%r5 BR_EX_AMODE31_r14 EX_TABLE_AMODE31(.Ldiag26c_ex, .Ldiag26c_ex) -ENDPROC(_diag26c_amode31) +SYM_FUNC_END(_diag26c_amode31) /* * void _diag0c_amode31(struct hypfs_diag0c_entry *entry) */ -ENTRY(_diag0c_amode31) +SYM_FUNC_START(_diag0c_amode31) sam31 diag %r2,%r2,0x0c sam64 BR_EX_AMODE31_r14 -ENDPROC(_diag0c_amode31) +SYM_FUNC_END(_diag0c_amode31) /* * void _diag308_reset_amode31(void) * * Calls diag 308 subcode 1 and continues execution */ -ENTRY(_diag308_reset_amode31) - larl %r4,.Lctlregs # Save control registers +SYM_FUNC_START(_diag308_reset_amode31) + larl %r4,ctlregs # Save control registers stctg %c0,%c15,0(%r4) lg %r2,0(%r4) # Disable lowcore protection nilh %r2,0xefff - larl %r4,.Lctlreg0 + larl %r4,ctlreg0 stg %r2,0(%r4) lctlg %c0,%c0,0(%r4) - larl %r4,.Lfpctl # Floating point control register + larl %r4,fpctl # Floating point control register stfpc 0(%r4) - larl %r4,.Lprefix # Save prefix register + larl %r4,prefix # Save prefix register stpx 0(%r4) - larl %r4,.Lprefix_zero # Set prefix register to 0 + larl %r4,prefix_zero # Set prefix register to 0 spx 0(%r4) - larl %r4,.Lcontinue_psw # Save PSW flags + larl %r4,continue_psw # Save PSW flags epsw %r2,%r3 stm %r2,%r3,0(%r4) larl %r4,.Lrestart_part2 # Setup restart PSW at absolute 0 - larl %r3,.Lrestart_diag308_psw + larl %r3,restart_diag308_psw og %r4,0(%r3) # Save PSW lghi %r3,0 sturg %r4,%r3 # Use sturg, because of large pages @@ -134,39 +134,26 @@ ENTRY(_diag308_reset_amode31) lhi %r1,2 # Use mode 2 = ESAME (dump) sigp %r1,%r0,SIGP_SET_ARCHITECTURE # Switch to ESAME mode sam64 # Switch to 64 bit addressing mode - larl %r4,.Lctlregs # Restore control registers + larl %r4,ctlregs # Restore control registers lctlg %c0,%c15,0(%r4) - larl %r4,.Lfpctl # Restore floating point ctl register + larl %r4,fpctl # Restore floating point ctl register lfpc 0(%r4) - larl %r4,.Lprefix # Restore prefix register + larl %r4,prefix # Restore prefix register spx 0(%r4) - larl %r4,.Lcontinue_psw # Restore PSW flags + larl %r4,continue_psw # Restore PSW flags larl %r2,.Lcontinue stg %r2,8(%r4) lpswe 0(%r4) .Lcontinue: BR_EX_AMODE31_r14 -ENDPROC(_diag308_reset_amode31) +SYM_FUNC_END(_diag308_reset_amode31) .section .amode31.data,"aw",@progbits -.align 8 -.Lrestart_diag308_psw: - .long 0x00080000,0x80000000 - -.align 8 -.Lcontinue_psw: - .quad 0,0 - -.align 8 -.Lctlreg0: - .quad 0 -.Lctlregs: - .rept 16 - .quad 0 - .endr -.Lfpctl: - .long 0 -.Lprefix: - .long 0 -.Lprefix_zero: - .long 0 + .balign 8 +SYM_DATA_LOCAL(restart_diag308_psw, .long 0x00080000,0x80000000) +SYM_DATA_LOCAL(continue_psw, .quad 0,0) +SYM_DATA_LOCAL(ctlreg0, .quad 0) +SYM_DATA_LOCAL(ctlregs, .fill 16,8,0) +SYM_DATA_LOCAL(fpctl, .long 0) +SYM_DATA_LOCAL(prefix, .long 0) +SYM_DATA_LOCAL(prefix_zero, .long 0) diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index c6eecd4a5302..9fd19530c9a5 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -637,24 +637,23 @@ static struct ctl_table topology_ctl_table[] = { { }, }; -static struct ctl_table topology_dir_table[] = { - { - .procname = "s390", - .maxlen = 0, - .mode = 0555, - .child = topology_ctl_table, - }, - { }, -}; - static int __init topology_init(void) { + struct device *dev_root; + int rc = 0; + timer_setup(&topology_timer, topology_timer_fn, TIMER_DEFERRABLE); if (MACHINE_HAS_TOPOLOGY) set_topology_timer(); else topology_update_polarization_simple(); - register_sysctl_table(topology_dir_table); - return device_create_file(cpu_subsys.dev_root, &dev_attr_dispatching); + register_sysctl("s390", topology_ctl_table); + + dev_root = bus_get_dev_root(&cpu_subsys); + if (dev_root) { + rc = device_create_file(dev_root, &dev_attr_dispatching); + put_device(dev_root); + } + return rc; } device_initcall(topology_init); diff --git a/arch/s390/kernel/uv.c b/arch/s390/kernel/uv.c index 9f18a4af9c13..cb2ee06df286 100644 --- a/arch/s390/kernel/uv.c +++ b/arch/s390/kernel/uv.c @@ -192,21 +192,10 @@ static int expected_page_refs(struct page *page) return res; } -static int make_secure_pte(pte_t *ptep, unsigned long addr, - struct page *exp_page, struct uv_cb_header *uvcb) +static int make_page_secure(struct page *page, struct uv_cb_header *uvcb) { - pte_t entry = READ_ONCE(*ptep); - struct page *page; int expected, cc = 0; - if (!pte_present(entry)) - return -ENXIO; - if (pte_val(entry) & _PAGE_INVALID) - return -ENXIO; - - page = pte_page(entry); - if (page != exp_page) - return -ENXIO; if (PageWriteback(page)) return -EAGAIN; expected = expected_page_refs(page); @@ -304,17 +293,18 @@ again: goto out; rc = -ENXIO; - page = follow_page(vma, uaddr, FOLL_WRITE); - if (IS_ERR_OR_NULL(page)) - goto out; - - lock_page(page); ptep = get_locked_pte(gmap->mm, uaddr, &ptelock); - if (should_export_before_import(uvcb, gmap->mm)) - uv_convert_from_secure(page_to_phys(page)); - rc = make_secure_pte(ptep, uaddr, page, uvcb); + if (pte_present(*ptep) && !(pte_val(*ptep) & _PAGE_INVALID) && pte_write(*ptep)) { + page = pte_page(*ptep); + rc = -EAGAIN; + if (trylock_page(page)) { + if (should_export_before_import(uvcb, gmap->mm)) + uv_convert_from_secure(page_to_phys(page)); + rc = make_page_secure(page, uvcb); + unlock_page(page); + } + } pte_unmap_unlock(ptep, ptelock); - unlock_page(page); out: mmap_read_unlock(gmap->mm); diff --git a/arch/s390/kernel/vdso32/Makefile b/arch/s390/kernel/vdso32/Makefile index 245bddfe9bc0..bafd3147eb4e 100644 --- a/arch/s390/kernel/vdso32/Makefile +++ b/arch/s390/kernel/vdso32/Makefile @@ -2,9 +2,8 @@ # List of files in the vdso KCOV_INSTRUMENT := n -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 the generic Makefile to check the built vdso. include $(srctree)/lib/vdso/Makefile obj-vdso32 = vdso_user_wrapper-32.o note-32.o diff --git a/arch/s390/kernel/vdso32/vdso_user_wrapper.S b/arch/s390/kernel/vdso32/vdso_user_wrapper.S index 3f42f27f978c..2e645003fdaf 100644 --- a/arch/s390/kernel/vdso32/vdso_user_wrapper.S +++ b/arch/s390/kernel/vdso32/vdso_user_wrapper.S @@ -1,12 +1,13 @@ /* SPDX-License-Identifier: GPL-2.0 */ +#include <linux/linkage.h> #include <asm/unistd.h> #include <asm/dwarf.h> .macro vdso_syscall func,syscall .globl __kernel_compat_\func .type __kernel_compat_\func,@function - .align 8 + __ALIGN __kernel_compat_\func: CFI_STARTPROC svc \syscall diff --git a/arch/s390/kernel/vdso64/Makefile b/arch/s390/kernel/vdso64/Makefile index 34f9542636e9..a766d286e15f 100644 --- a/arch/s390/kernel/vdso64/Makefile +++ b/arch/s390/kernel/vdso64/Makefile @@ -2,9 +2,8 @@ # List of files in the vdso KCOV_INSTRUMENT := n -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 the generic Makefile to check the built vdso. include $(srctree)/lib/vdso/Makefile obj-vdso64 = vdso_user_wrapper.o note.o obj-cvdso64 = vdso64_generic.o getcpu.o diff --git a/arch/s390/kernel/vdso64/vdso_user_wrapper.S b/arch/s390/kernel/vdso64/vdso_user_wrapper.S index 97f0c0a669a5..57f62596e53b 100644 --- a/arch/s390/kernel/vdso64/vdso_user_wrapper.S +++ b/arch/s390/kernel/vdso64/vdso_user_wrapper.S @@ -1,4 +1,5 @@ /* SPDX-License-Identifier: GPL-2.0 */ +#include <linux/linkage.h> #include <asm/vdso.h> #include <asm/unistd.h> #include <asm/asm-offsets.h> @@ -16,7 +17,7 @@ .macro vdso_func func .globl __kernel_\func .type __kernel_\func,@function - .align 8 + __ALIGN __kernel_\func: CFI_STARTPROC aghi %r15,-WRAPPER_FRAME_SIZE @@ -41,7 +42,7 @@ vdso_func getcpu .macro vdso_syscall func,syscall .globl __kernel_\func .type __kernel_\func,@function - .align 8 + __ALIGN __kernel_\func: CFI_STARTPROC svc \syscall diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index b653ba8d51e6..2ae201ebf90b 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -14,6 +14,8 @@ #define BSS_FIRST_SECTIONS *(.bss..swapper_pg_dir) \ *(.bss..invalid_pg_dir) +#define RO_EXCEPTION_TABLE_ALIGN 16 + /* Handle ro_after_init data on our own. */ #define RO_AFTER_INIT_DATA @@ -66,7 +68,6 @@ SECTIONS *(.data..ro_after_init) JUMP_TABLE_DATA } :data - EXCEPTION_TABLE(16) . = ALIGN(PAGE_SIZE); __end_ro_after_init = .; @@ -219,6 +220,13 @@ SECTIONS QUAD(init_mm) QUAD(swapper_pg_dir) QUAD(invalid_pg_dir) +#ifdef CONFIG_KASAN + QUAD(kasan_early_shadow_page) + QUAD(kasan_early_shadow_pte) + QUAD(kasan_early_shadow_pmd) + QUAD(kasan_early_shadow_pud) + QUAD(kasan_early_shadow_p4d) +#endif } :NONE /* Debugging sections. */ |