Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760096AbYGAUUn (ORCPT ); Tue, 1 Jul 2008 16:20:43 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756606AbYGAUUY (ORCPT ); Tue, 1 Jul 2008 16:20:24 -0400 Received: from triton.rz.uni-saarland.de ([134.96.7.25]:10964 "EHLO triton.rz.uni-saarland.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754824AbYGAUUS convert rfc822-to-8bit (ORCPT ); Tue, 1 Jul 2008 16:20:18 -0400 X-Greylist: delayed 2519 seconds by postgrey-1.27 at vger.kernel.org; Tue, 01 Jul 2008 16:20:17 EDT Date: Tue, 1 Jul 2008 21:36:43 +0200 From: Alexander van Heukelum To: Ingo Molnar Cc: Thomas Gleixner , "H. Peter Anvin" , LKML Subject: [PATCH] x86: Simple changes to make traps_32.c and traps_64.c more similar Message-ID: <20080701193643.GA29695@mailshack.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: 8BIT User-Agent: Mutt/1.5.9i X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-3.0 (triton.rz.uni-saarland.de [134.96.7.25]); Tue, 01 Jul 2008 21:38:00 +0200 (CEST) X-AntiVirus: checked by AntiVir MailGate (version: 2.1.2-14; AVE: 7.8.0.59; VDF: 7.0.5.30; host: AntiVir3) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 37692 Lines: 1261 Tiny first step at unification of traps_32.c and traps_64.c. All changes are meant to be trivial. - mostly whitespace changes - chose one of obviously equivalent pieces of code: - reordering of declarations - simple renaming of local variables - if (cond) { ... } -> if (!cond) goto end_of_block ... - local caching of current, cpu, etc... This makes the diff between traps_32.c and traps_64.c smaller: from: 1474 +++++++++++++++++++++++++++++-------------------------------- 1 file changed, 722 insertions(+), 752 deletions(-) to: 1052 +++++++++++++++++++++++++++++-------------------------------- 1 file changed, 512 insertions(+), 540 deletions(-) Signed-off-by: Alexander van Heukelum --- arch/x86/kernel/traps_32.c | 141 +++++++------- arch/x86/kernel/traps_64.c | 469 ++++++++++++++++++++++---------------------- 2 files changed, 309 insertions(+), 301 deletions(-) Hi, This did not have much testing. The result compiles fine and boots to userspace on qemu, though. Patch is against recent -tip. Greetings, Alexander diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps_32.c index 18c8b8d..3338f0a 100644 --- a/arch/x86/kernel/traps_32.c +++ b/arch/x86/kernel/traps_32.c @@ -1,5 +1,6 @@ /* * Copyright (C) 1991, 1992 Linus Torvalds + * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs * * Pentium III FXSR, SSE support * Gareth Hughes , May 2000 @@ -61,8 +62,6 @@ #include "mach_traps.h" -int panic_on_unrecovered_nmi; - DECLARE_BITMAP(used_vectors, NR_VECTORS); EXPORT_SYMBOL_GPL(used_vectors); @@ -96,22 +95,26 @@ asmlinkage void page_fault(void); asmlinkage void coprocessor_error(void); asmlinkage void simd_coprocessor_error(void); asmlinkage void alignment_check(void); -asmlinkage void spurious_interrupt_bug(void); asmlinkage void machine_check(void); +asmlinkage void spurious_interrupt_bug(void); +int panic_on_unrecovered_nmi; int kstack_depth_to_print = 24; static unsigned int code_bytes = 64; +static int ignore_nmis; + +//#define MSG(txt) ops->warning(data, txt) void printk_address(unsigned long address, int reliable) { #ifdef CONFIG_KALLSYMS - char namebuf[KSYM_NAME_LEN]; unsigned long offset = 0; unsigned long symsize; const char *symname; - char reliab[4] = ""; - char *delim = ":"; char *modname; + char *delim = ":"; + char namebuf[KSYM_NAME_LEN]; + char reliab[4] = ""; symname = kallsyms_lookup(address, &symsize, &offset, &modname, namebuf); @@ -131,16 +134,17 @@ void printk_address(unsigned long address, int reliable) #endif } -static inline int valid_stack_ptr(struct thread_info *tinfo, void *p, unsigned size) +static inline int valid_stack_ptr(struct thread_info *tinfo, + void *p, unsigned int size) { - return p > (void *)tinfo && - p <= (void *)tinfo + THREAD_SIZE - size; + void *t = tinfo; + return p > t && p <= t + THREAD_SIZE - size; } /* The form of the top of the frame on the stack */ struct stack_frame { - struct stack_frame *next_frame; - unsigned long return_address; + struct stack_frame *next_frame; + unsigned long return_address; }; static inline unsigned long @@ -168,8 +172,6 @@ print_context_stack(struct thread_info *tinfo, return bp; } -#define MSG(msg) ops->warning(data, msg) - void dump_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, unsigned long bp, const struct stacktrace_ops *ops, void *data) @@ -179,7 +181,6 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, if (!stack) { unsigned long dummy; - stack = &dummy; if (task != current) stack = (unsigned long *)task->thread.sp; @@ -197,7 +198,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, } #endif - while (1) { + for (;;) { struct thread_info *context; context = (struct thread_info *) @@ -249,10 +250,10 @@ static void print_trace_address(void *data, unsigned long addr, int reliable) } static const struct stacktrace_ops print_trace_ops = { - .warning = print_trace_warning, - .warning_symbol = print_trace_warning_symbol, - .stack = print_trace_stack, - .address = print_trace_address, + .warning = print_trace_warning, + .warning_symbol = print_trace_warning_symbol, + .stack = print_trace_stack, + .address = print_trace_address, }; static void @@ -352,15 +353,14 @@ void show_registers(struct pt_regs *regs) printk(KERN_EMERG "Code: "); ip = (u8 *)regs->ip - code_prologue; - if (ip < (u8 *)PAGE_OFFSET || - probe_kernel_address(ip, c)) { + if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) { /* try starting at EIP */ ip = (u8 *)regs->ip; code_len = code_len - code_prologue + 1; } for (i = 0; i < code_len; i++, ip++) { if (ip < (u8 *)PAGE_OFFSET || - probe_kernel_address(ip, c)) { + probe_kernel_address(ip, c)) { printk(" Bad EIP value."); break; } @@ -405,24 +405,22 @@ int __kprobes __die(const char *str, struct pt_regs *regs, long err) printk("\n"); if (notify_die(DIE_OOPS, str, regs, err, - current->thread.trap_no, SIGSEGV) != NOTIFY_STOP) { - - show_registers(regs); - /* Executive summary in case the oops scrolled away */ - sp = (unsigned long) (®s->sp); - savesegment(ss, ss); - if (user_mode(regs)) { - sp = regs->sp; - ss = regs->ss & 0xffff; - } - printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip); - print_symbol("%s", regs->ip); - printk(" SS:ESP %04x:%08lx\n", ss, sp); + current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) + return 1; - return 0; + show_registers(regs); + /* Executive summary in case the oops scrolled away */ + sp = (unsigned long) (®s->sp); + savesegment(ss, ss); + if (user_mode(regs)) { + sp = regs->sp; + ss = regs->ss & 0xffff; } + printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip); + print_symbol("%s", regs->ip); + printk(" SS:ESP %04x:%08lx\n", ss, sp); - return 1; + return 0; } /* @@ -611,6 +609,7 @@ DO_ERROR_INFO(32, SIGILL, "iret exception", iret_error, ILL_BADSTK, 0, 1) void __kprobes do_general_protection(struct pt_regs *regs, long error_code) { + struct task_struct *tsk; struct thread_struct *thread; struct tss_struct *tss; int cpu; @@ -651,23 +650,24 @@ void __kprobes do_general_protection(struct pt_regs *regs, long error_code) if (regs->flags & X86_VM_MASK) goto gp_in_vm86; + tsk = current; if (!user_mode(regs)) goto gp_in_kernel; - current->thread.error_code = error_code; - current->thread.trap_no = 13; + tsk->thread.error_code = error_code; + tsk->thread.trap_no = 13; - if (show_unhandled_signals && unhandled_signal(current, SIGSEGV) && + if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) && printk_ratelimit()) { printk(KERN_INFO "%s[%d] general protection ip:%lx sp:%lx error:%lx", - current->comm, task_pid_nr(current), + tsk->comm, task_pid_nr(tsk), regs->ip, regs->sp, error_code); print_vma_addr(" in ", regs->ip); printk("\n"); } - force_sig(SIGSEGV, current); + force_sig(SIGSEGV, tsk); return; gp_in_vm86: @@ -676,14 +676,15 @@ gp_in_vm86: return; gp_in_kernel: - if (!fixup_exception(regs)) { - current->thread.error_code = error_code; - current->thread.trap_no = 13; - if (notify_die(DIE_GPF, "general protection fault", regs, + if (fixup_exception(regs)) + return; + + tsk->thread.error_code = error_code; + tsk->thread.trap_no = 13; + if (notify_die(DIE_GPF, "general protection fault", regs, error_code, 13, SIGSEGV) == NOTIFY_STOP) - return; - die("general protection fault", regs, error_code); - } + return; + die("general protection fault", regs, error_code); } static notrace __kprobes void @@ -796,14 +797,17 @@ void notrace __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic) static notrace __kprobes void default_do_nmi(struct pt_regs *regs) { unsigned char reason = 0; + int cpu; - /* Only the BSP gets external NMIs from the system: */ - if (!smp_processor_id()) + cpu = smp_processor_id(); + + /* Only the BSP gets external NMIs from the system. */ + if (!cpu) reason = get_nmi_reason(); if (!(reason & 0xc0)) { if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT) - == NOTIFY_STOP) + == NOTIFY_STOP) return; #ifdef CONFIG_X86_LOCAL_APIC /* @@ -812,7 +816,7 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) */ if (nmi_watchdog_tick(regs, reason)) return; - if (!do_nmi_callback(regs, smp_processor_id())) + if (!do_nmi_callback(regs, cpu)) unknown_nmi_error(reason, regs); #else unknown_nmi_error(reason, regs); @@ -822,6 +826,8 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) } if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) return; + + /* AK: following checks seem to be broken on modern chipsets. FIXME */ if (reason & 0x80) mem_parity_error(reason, regs); if (reason & 0x40) @@ -833,8 +839,6 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) reassert_nmi(); } -static int ignore_nmis; - notrace __kprobes void do_nmi(struct pt_regs *regs, long error_code) { int cpu; @@ -927,7 +931,7 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code) tsk->thread.debugctlmsr = 0; if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code, - SIGTRAP) == NOTIFY_STOP) + SIGTRAP) == NOTIFY_STOP) return; /* It's safe to allow irq's after DR6 has been saved */ if (regs->flags & X86_EFLAGS_IF) @@ -988,9 +992,8 @@ clear_TF_reenable: void math_error(void __user *ip) { struct task_struct *task; - unsigned short cwd; - unsigned short swd; siginfo_t info; + unsigned short cwd, swd; /* * Save the info for the exception handler and clear the error. @@ -1009,7 +1012,7 @@ void math_error(void __user *ip) * C1 reg you need in case of a stack fault, 0x040 is the stack * fault bit. We should only be taking one exception at a time, * so if this combination doesn't produce any single exception, - * then we have a bad program that isn't syncronizing its FPU usage + * then we have a bad program that isn't synchronizing its FPU usage * and it will suffer the consequences since we won't be able to * fully reproduce the context of the exception */ @@ -1018,7 +1021,7 @@ void math_error(void __user *ip) switch (swd & ~cwd & 0x3f) { case 0x000: /* No unmasked exception */ return; - default: /* Multiple exceptions */ + default: /* Multiple exceptions */ break; case 0x001: /* Invalid Op */ /* @@ -1054,8 +1057,8 @@ void do_coprocessor_error(struct pt_regs *regs, long error_code) static void simd_math_error(void __user *ip) { struct task_struct *task; - unsigned short mxcsr; siginfo_t info; + unsigned short mxcsr; /* * Save the info for the exception handler and clear the error. @@ -1213,16 +1216,16 @@ void __init trap_init(void) #ifdef CONFIG_X86_LOCAL_APIC init_apic_mappings(); #endif - set_trap_gate(0, ÷_error); - set_intr_gate(1, &debug); - set_intr_gate(2, &nmi); + set_trap_gate(0, ÷_error); + set_intr_gate(1, &debug); + set_intr_gate(2, &nmi); set_system_intr_gate(3, &int3); /* int3/4 can be called from all */ set_system_gate(4, &overflow); - set_trap_gate(5, &bounds); - set_trap_gate(6, &invalid_op); - set_trap_gate(7, &device_not_available); - set_task_gate(8, GDT_ENTRY_DOUBLEFAULT_TSS); - set_trap_gate(9, &coprocessor_segment_overrun); + set_trap_gate(5, &bounds); + set_trap_gate(6, &invalid_op); + set_trap_gate(7, &device_not_available); + set_task_gate(8, GDT_ENTRY_DOUBLEFAULT_TSS); + set_trap_gate(9, &coprocessor_segment_overrun); set_trap_gate(10, &invalid_TSS); set_trap_gate(11, &segment_not_present); set_trap_gate(12, &stack_segment); diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c index fc5b66e..1e9c8e6 100644 --- a/arch/x86/kernel/traps_64.c +++ b/arch/x86/kernel/traps_64.c @@ -10,50 +10,50 @@ * 'Traps.c' handles hardware traps and faults after we have saved some * state in 'entry.S'. */ -#include +#include +#include +#include +#include +#include +#include +#include +#include #include +#include +#include #include +#include +#include #include -#include +#include +#include #include -#include #include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include #include -#include -#include - -#include +#include +#include #if defined(CONFIG_EDAC) #include #endif -#include -#include -#include +#include +#include #include +#include +#include +#include #include #include -#include -#include +#include #include +#include +#include #include -#include #include -#include -#include -#include +#include + +#include asmlinkage void divide_error(void); asmlinkage void debug(void); @@ -77,6 +77,7 @@ asmlinkage void machine_check(void); asmlinkage void spurious_interrupt_bug(void); int panic_on_unrecovered_nmi; +int kstack_depth_to_print = 12; static unsigned int code_bytes = 64; static unsigned ignore_nmis; @@ -102,12 +103,13 @@ static inline void preempt_conditional_cli(struct pt_regs *regs) dec_preempt_count(); } -int kstack_depth_to_print = 12; +//#define MSG(txt) ops->warning(data, txt) void printk_address(unsigned long address, int reliable) { #ifdef CONFIG_KALLSYMS - unsigned long offset = 0, symsize; + unsigned long offset = 0; + unsigned long symsize; const char *symname; char *modname; char *delim = ":"; @@ -206,8 +208,6 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, return NULL; } -#define MSG(txt) ops->warning(data, txt) - /* * x86-64 can have up to three kernel stacks: * process stack @@ -235,10 +235,11 @@ struct stack_frame { }; -static inline unsigned long print_context_stack(struct thread_info *tinfo, - unsigned long *stack, unsigned long bp, - const struct stacktrace_ops *ops, void *data, - unsigned long *end) +static inline unsigned long +print_context_stack(struct thread_info *tinfo, + unsigned long *stack, unsigned long bp, + const struct stacktrace_ops *ops, void *data, + unsigned long *end) { struct stack_frame *frame = (struct stack_frame *)bp; @@ -260,7 +261,7 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo, return bp; } -void dump_trace(struct task_struct *tsk, struct pt_regs *regs, +void dump_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, unsigned long bp, const struct stacktrace_ops *ops, void *data) { @@ -269,36 +270,34 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned used = 0; struct thread_info *tinfo; - if (!tsk) - tsk = current; - tinfo = task_thread_info(tsk); + if (!task) + task = current; if (!stack) { unsigned long dummy; stack = &dummy; - if (tsk && tsk != current) - stack = (unsigned long *)tsk->thread.sp; + if (task && task != current) + stack = (unsigned long *)task->thread.sp; } #ifdef CONFIG_FRAME_POINTER if (!bp) { - if (tsk == current) { + if (task == current) { /* Grab bp right from our regs */ - asm("movq %%rbp, %0" : "=r" (bp):); + asm("movq %%rbp, %0" : "=r" (bp) :); } else { /* bp is the last reg pushed by switch_to */ - bp = *(unsigned long *) tsk->thread.sp; + bp = *(unsigned long *) task->thread.sp; } } #endif - - /* * Print function call entries in all stacks, starting at the * current stack address. If the stacks consist of nested * exceptions */ + tinfo = task_thread_info(task); for (;;) { char *id; unsigned long *estack_end; @@ -383,9 +382,8 @@ static const struct stacktrace_ops print_trace_ops = { .address = print_trace_address, }; -void -show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack, - unsigned long bp) +void show_trace(struct task_struct *tsk, struct pt_regs *regs, + unsigned long *stack, unsigned long bp) { printk("\nCall Trace:\n"); dump_trace(tsk, regs, stack, bp, &print_trace_ops, NULL); @@ -393,8 +391,8 @@ show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack, } static void -_show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *sp, - unsigned long bp) +_show_stack(struct task_struct *task, struct pt_regs *regs, + unsigned long *sp, unsigned long bp) { unsigned long *stack; int i; @@ -406,14 +404,14 @@ _show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *sp, // back trace for this cpu. if (sp == NULL) { - if (tsk) - sp = (unsigned long *)tsk->thread.sp; + if (task) + sp = (unsigned long *)task->thread.sp; else sp = (unsigned long *)&sp; } stack = sp; - for(i=0; i < kstack_depth_to_print; i++) { + for (i = 0; i < kstack_depth_to_print; i++) { if (stack >= irqstack && stack <= irqstack_end) { if (stack == irqstack_end) { stack = (unsigned long *) (irqstack_end[-1]); @@ -428,12 +426,12 @@ _show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *sp, printk(" %016lx", *stack++); touch_nmi_watchdog(); } - show_trace(tsk, regs, sp, bp); + show_trace(task, regs, sp, bp); } -void show_stack(struct task_struct *tsk, unsigned long * sp) +void show_stack(struct task_struct *task, unsigned long *sp) { - _show_stack(tsk, NULL, sp, 0); + _show_stack(task, NULL, sp, 0); } /* @@ -441,8 +439,8 @@ void show_stack(struct task_struct *tsk, unsigned long * sp) */ void dump_stack(void) { - unsigned long dummy; unsigned long bp = 0; + unsigned long stack; #ifdef CONFIG_FRAME_POINTER if (!bp) @@ -454,7 +452,8 @@ void dump_stack(void) init_utsname()->release, (int)strcspn(init_utsname()->version, " "), init_utsname()->version); - show_trace(NULL, NULL, &dummy, bp); + + show_trace(NULL, NULL, &stack, bp); } EXPORT_SYMBOL(dump_stack); @@ -465,12 +464,8 @@ void show_registers(struct pt_regs *regs) unsigned long sp; const int cpu = smp_processor_id(); struct task_struct *cur = cpu_pda(cpu)->pcurrent; - u8 *ip; - unsigned int code_prologue = code_bytes * 43 / 64; - unsigned int code_len = code_bytes; sp = regs->sp; - ip = (u8 *) regs->ip - code_prologue; printk("CPU %d ", cpu); __show_regs(regs, 1); printk("Process %s (pid: %d, threadinfo %p, task %p)\n", @@ -481,15 +476,21 @@ void show_registers(struct pt_regs *regs) * time of the fault.. */ if (!user_mode(regs)) { + unsigned int code_prologue = code_bytes * 43 / 64; + unsigned int code_len = code_bytes; unsigned char c; + u8 *ip; + printk("Stack: "); _show_stack(NULL, regs, (unsigned long *)sp, regs->bp); printk("\n"); printk(KERN_EMERG "Code: "); + + ip = (u8 *)regs->ip - code_prologue; if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) { /* try starting at RIP */ - ip = (u8 *) regs->ip; + ip = (u8 *)regs->ip; code_len = code_len - code_prologue + 1; } for (i = 0; i < code_len; i++, ip++) { @@ -505,7 +506,7 @@ void show_registers(struct pt_regs *regs) } } printk("\n"); -} +} int is_valid_bugaddr(unsigned long ip) { @@ -581,8 +582,11 @@ int __kprobes __die(const char * str, struct pt_regs * regs, long err) printk("DEBUG_PAGEALLOC"); #endif printk("\n"); - if (notify_die(DIE_OOPS, str, regs, err, current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) + + if (notify_die(DIE_OOPS, str, regs, err, + current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) return 1; + show_registers(regs); add_taint(TAINT_DIE); /* Executive summary in case the oops scrolled away */ @@ -611,8 +615,7 @@ die_nmi(char *str, struct pt_regs *regs, int do_panic) { unsigned long flags; - if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == - NOTIFY_STOP) + if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP) return; flags = oops_begin(); @@ -634,44 +637,44 @@ die_nmi(char *str, struct pt_regs *regs, int do_panic) do_exit(SIGBUS); } -static void __kprobes do_trap(int trapnr, int signr, char *str, - struct pt_regs * regs, long error_code, - siginfo_t *info) +static void __kprobes +do_trap(int trapnr, int signr, char *str, struct pt_regs * regs, + long error_code, siginfo_t *info) { struct task_struct *tsk = current; - if (user_mode(regs)) { - /* - * We want error_code and trap_no set for userspace - * faults and kernelspace faults which result in - * die(), but not kernelspace faults which are fixed - * up. die() gives the process no chance to handle - * the signal and notice the kernel fault information, - * so that won't result in polluting the information - * about previously queued, but not yet delivered, - * faults. See also do_general_protection below. - */ - tsk->thread.error_code = error_code; - tsk->thread.trap_no = trapnr; - - if (show_unhandled_signals && unhandled_signal(tsk, signr) && - printk_ratelimit()) { - printk(KERN_INFO - "%s[%d] trap %s ip:%lx sp:%lx error:%lx", - tsk->comm, tsk->pid, str, - regs->ip, regs->sp, error_code); - print_vma_addr(" in ", regs->ip); - printk("\n"); - } + if (!user_mode(regs)) + goto kernel_trap; - if (info) - force_sig_info(signr, info, tsk); - else - force_sig(signr, tsk); - return; + /* + * We want error_code and trap_no set for userspace faults and + * kernelspace faults which result in die(), but not + * kernelspace faults which are fixed up. die() gives the + * process no chance to handle the signal and notice the + * kernel fault information, so that won't result in polluting + * the information about previously queued, but not yet + * delivered, faults. See also do_general_protection below. + */ + tsk->thread.error_code = error_code; + tsk->thread.trap_no = trapnr; + + if (show_unhandled_signals && unhandled_signal(tsk, signr) && + printk_ratelimit()) { + printk(KERN_INFO + "%s[%d] trap %s ip:%lx sp:%lx error:%lx", + tsk->comm, tsk->pid, str, + regs->ip, regs->sp, error_code); + print_vma_addr(" in ", regs->ip); + printk("\n"); } + if (info) + force_sig_info(signr, info, tsk); + else + force_sig(signr, tsk); + return; +kernel_trap: if (!fixup_exception(regs)) { tsk->thread.error_code = error_code; tsk->thread.trap_no = trapnr; @@ -680,37 +683,37 @@ static void __kprobes do_trap(int trapnr, int signr, char *str, return; } -#define DO_ERROR(trapnr, signr, str, name) \ -asmlinkage void do_##name(struct pt_regs * regs, long error_code) \ -{ \ - if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ - == NOTIFY_STOP) \ - return; \ +#define DO_ERROR(trapnr, signr, str, name) \ +asmlinkage void do_##name(struct pt_regs *regs, long error_code) \ +{ \ + if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ + == NOTIFY_STOP) \ + return; \ conditional_sti(regs); \ - do_trap(trapnr, signr, str, regs, error_code, NULL); \ + do_trap(trapnr, signr, str, regs, error_code, NULL); \ } -#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ -asmlinkage void do_##name(struct pt_regs * regs, long error_code) \ -{ \ - siginfo_t info; \ - info.si_signo = signr; \ - info.si_errno = 0; \ - info.si_code = sicode; \ - info.si_addr = (void __user *)siaddr; \ - trace_hardirqs_fixup(); \ - if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ - == NOTIFY_STOP) \ - return; \ +#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ +asmlinkage void do_##name(struct pt_regs * regs, long error_code) \ +{ \ + siginfo_t info; \ + info.si_signo = signr; \ + info.si_errno = 0; \ + info.si_code = sicode; \ + info.si_addr = (void __user *)siaddr; \ + trace_hardirqs_fixup(); \ + if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \ + == NOTIFY_STOP) \ + return; \ conditional_sti(regs); \ - do_trap(trapnr, signr, str, regs, error_code, &info); \ + do_trap(trapnr, signr, str, regs, error_code, &info); \ } -DO_ERROR_INFO( 0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip) -DO_ERROR( 4, SIGSEGV, "overflow", overflow) -DO_ERROR( 5, SIGSEGV, "bounds", bounds) -DO_ERROR_INFO( 6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip) -DO_ERROR( 9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun) +DO_ERROR_INFO(0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip) +DO_ERROR(4, SIGSEGV, "overflow", overflow) +DO_ERROR(5, SIGSEGV, "bounds", bounds) +DO_ERROR_INFO(6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip) +DO_ERROR(9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun) DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS) DO_ERROR(11, SIGBUS, "segment not present", segment_not_present) DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0) @@ -746,28 +749,31 @@ asmlinkage void do_double_fault(struct pt_regs * regs, long error_code) asmlinkage void __kprobes do_general_protection(struct pt_regs * regs, long error_code) { - struct task_struct *tsk = current; + struct task_struct *tsk; conditional_sti(regs); - if (user_mode(regs)) { - tsk->thread.error_code = error_code; - tsk->thread.trap_no = 13; - - if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) && - printk_ratelimit()) { - printk(KERN_INFO - "%s[%d] general protection ip:%lx sp:%lx error:%lx", - tsk->comm, tsk->pid, - regs->ip, regs->sp, error_code); - print_vma_addr(" in ", regs->ip); - printk("\n"); - } + tsk = current; + if (!user_mode(regs)) + goto gp_in_kernel; - force_sig(SIGSEGV, tsk); - return; - } + tsk->thread.error_code = error_code; + tsk->thread.trap_no = 13; + + if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) && + printk_ratelimit()) { + printk(KERN_INFO + "%s[%d] general protection ip:%lx sp:%lx error:%lx", + tsk->comm, tsk->pid, + regs->ip, regs->sp, error_code); + print_vma_addr(" in ", regs->ip); + printk("\n"); + } + force_sig(SIGSEGV, tsk); + return; + +gp_in_kernel: if (fixup_exception(regs)) return; @@ -780,14 +786,14 @@ asmlinkage void __kprobes do_general_protection(struct pt_regs * regs, } static notrace __kprobes void -mem_parity_error(unsigned char reason, struct pt_regs * regs) +mem_parity_error(unsigned char reason, struct pt_regs *regs) { printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n", reason); printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n"); #if defined(CONFIG_EDAC) - if(edac_handler_set()) { + if (edac_handler_set()) { edac_atomic_assert_error(); return; } @@ -841,7 +847,7 @@ asmlinkage notrace __kprobes void default_do_nmi(struct pt_regs *regs) cpu = smp_processor_id(); - /* Only the BSP gets external NMIs from the system. */ + /* Only the BSP gets external NMIs from the system. */ if (!cpu) reason = get_nmi_reason(); @@ -853,18 +859,17 @@ asmlinkage notrace __kprobes void default_do_nmi(struct pt_regs *regs) * Ok, so this is none of the documented NMI sources, * so it must be the NMI watchdog. */ - if (nmi_watchdog_tick(regs,reason)) + if (nmi_watchdog_tick(regs, reason)) return; - if (!do_nmi_callback(regs,cpu)) + if (!do_nmi_callback(regs, cpu)) unknown_nmi_error(reason, regs); return; } if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) - return; + return; /* AK: following checks seem to be broken on modern chipsets. FIXME */ - if (reason & 0x80) mem_parity_error(reason, regs); if (reason & 0x40) @@ -875,9 +880,12 @@ asmlinkage notrace __kprobes void do_nmi(struct pt_regs *regs, long error_code) { nmi_enter(); + add_pda(__nmi_count, 1); + if (!ignore_nmis) default_do_nmi(regs); + nmi_exit(); } @@ -894,13 +902,14 @@ void restart_nmi(void) } /* runs on IST stack. */ -asmlinkage void __kprobes do_int3(struct pt_regs * regs, long error_code) +asmlinkage void __kprobes do_int3(struct pt_regs *regs, long error_code) { trace_hardirqs_fixup(); - if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) == NOTIFY_STOP) { + if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) + == NOTIFY_STOP) return; - } + preempt_conditional_sti(regs); do_trap(3, SIGTRAP, "int3", regs, error_code, NULL); preempt_conditional_cli(regs); @@ -928,11 +937,11 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs) } /* runs on IST stack. */ -asmlinkage void __kprobes do_debug(struct pt_regs * regs, +asmlinkage void __kprobes do_debug(struct pt_regs *regs, unsigned long error_code) { - unsigned long condition; struct task_struct *tsk = current; + unsigned long condition; siginfo_t info; trace_hardirqs_fixup(); @@ -961,21 +970,19 @@ asmlinkage void __kprobes do_debug(struct pt_regs * regs, /* Mask out spurious debug traps due to lazy DR7 setting */ if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) { - if (!tsk->thread.debugreg7) { + if (!tsk->thread.debugreg7) goto clear_dr7; - } } tsk->thread.debugreg6 = condition; - /* * Single-stepping through TF: make sure we ignore any events in * kernel space (but re-enable TF when returning to user mode). */ if (condition & DR_STEP) { - if (!user_mode(regs)) - goto clear_TF_reenable; + if (!user_mode(regs)) + goto clear_TF_reenable; } /* Ok, finally something we can handle */ @@ -988,7 +995,7 @@ asmlinkage void __kprobes do_debug(struct pt_regs * regs, force_sig_info(SIGTRAP, &info, tsk); clear_dr7: - set_debugreg(0UL, 7); + set_debugreg(0, 7); preempt_conditional_cli(regs); return; @@ -1018,7 +1025,7 @@ static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr) asmlinkage void do_coprocessor_error(struct pt_regs *regs) { void __user *ip = (void __user *)(regs->ip); - struct task_struct * task; + struct task_struct *task; siginfo_t info; unsigned short cwd, swd; @@ -1051,30 +1058,30 @@ asmlinkage void do_coprocessor_error(struct pt_regs *regs) cwd = get_fpu_cwd(task); swd = get_fpu_swd(task); switch (swd & ~cwd & 0x3f) { - case 0x000: - default: - break; - case 0x001: /* Invalid Op */ - /* - * swd & 0x240 == 0x040: Stack Underflow - * swd & 0x240 == 0x240: Stack Overflow - * User must clear the SF bit (0x40) if set - */ - info.si_code = FPE_FLTINV; - break; - case 0x002: /* Denormalize */ - case 0x010: /* Underflow */ - info.si_code = FPE_FLTUND; - break; - case 0x004: /* Zero Divide */ - info.si_code = FPE_FLTDIV; - break; - case 0x008: /* Overflow */ - info.si_code = FPE_FLTOVF; - break; - case 0x020: /* Precision */ - info.si_code = FPE_FLTRES; - break; + case 0x000: /* No unmasked exception */ + default: /* Multiple exceptions */ + break; + case 0x001: /* Invalid Op */ + /* + * swd & 0x240 == 0x040: Stack Underflow + * swd & 0x240 == 0x240: Stack Overflow + * User must clear the SF bit (0x40) if set + */ + info.si_code = FPE_FLTINV; + break; + case 0x002: /* Denormalize */ + case 0x010: /* Underflow */ + info.si_code = FPE_FLTUND; + break; + case 0x004: /* Zero Divide */ + info.si_code = FPE_FLTDIV; + break; + case 0x008: /* Overflow */ + info.si_code = FPE_FLTOVF; + break; + case 0x020: /* Precision */ + info.si_code = FPE_FLTRES; + break; } force_sig_info(SIGFPE, &info, task); } @@ -1087,7 +1094,7 @@ asmlinkage void bad_intr(void) asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs) { void __user *ip = (void __user *)(regs->ip); - struct task_struct * task; + struct task_struct *task; siginfo_t info; unsigned short mxcsr; @@ -1115,25 +1122,25 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs) */ mxcsr = get_fpu_mxcsr(task); switch (~((mxcsr & 0x1f80) >> 7) & (mxcsr & 0x3f)) { - case 0x000: - default: - break; - case 0x001: /* Invalid Op */ - info.si_code = FPE_FLTINV; - break; - case 0x002: /* Denormalize */ - case 0x010: /* Underflow */ - info.si_code = FPE_FLTUND; - break; - case 0x004: /* Zero Divide */ - info.si_code = FPE_FLTDIV; - break; - case 0x008: /* Overflow */ - info.si_code = FPE_FLTOVF; - break; - case 0x020: /* Precision */ - info.si_code = FPE_FLTRES; - break; + case 0x000: + default: + break; + case 0x001: /* Invalid Op */ + info.si_code = FPE_FLTINV; + break; + case 0x002: /* Denormalize */ + case 0x010: /* Underflow */ + info.si_code = FPE_FLTUND; + break; + case 0x004: /* Zero Divide */ + info.si_code = FPE_FLTDIV; + break; + case 0x008: /* Overflow */ + info.si_code = FPE_FLTOVF; + break; + case 0x020: /* Precision */ + info.si_code = FPE_FLTRES; + break; } force_sig_info(SIGFPE, &info, task); } @@ -1151,7 +1158,7 @@ asmlinkage void __attribute__((weak)) mce_threshold_interrupt(void) } /* - * 'math_state_restore()' saves the current math information in the + * 'math_state_restore()' saves the current math information in the * old math state array, and gets the new ones from the current task * * Careful.. There are problems with IBM-designed IRQ13 behaviour. @@ -1176,7 +1183,7 @@ asmlinkage void math_state_restore(void) local_irq_disable(); } - clts(); /* Allow maths ops (or we recurse) */ + clts(); /* Allow maths ops (or we recurse) */ restore_fpu_checking(&me->thread.xstate->fxsave); task_thread_info(me)->status |= TS_USEDFPU; me->fpu_counter++; @@ -1185,28 +1192,28 @@ EXPORT_SYMBOL_GPL(math_state_restore); void __init trap_init(void) { - set_intr_gate(0,÷_error); - set_intr_gate_ist(1,&debug,DEBUG_STACK); - set_intr_gate_ist(2,&nmi,NMI_STACK); - set_system_gate_ist(3,&int3,DEBUG_STACK); /* int3 can be called from all */ - set_system_gate(4,&overflow); /* int4 can be called from all */ - set_intr_gate(5,&bounds); - set_intr_gate(6,&invalid_op); - set_intr_gate(7,&device_not_available); - set_intr_gate_ist(8,&double_fault, DOUBLEFAULT_STACK); - set_intr_gate(9,&coprocessor_segment_overrun); - set_intr_gate(10,&invalid_TSS); - set_intr_gate(11,&segment_not_present); - set_intr_gate_ist(12,&stack_segment,STACKFAULT_STACK); - set_intr_gate(13,&general_protection); - set_intr_gate(14,&page_fault); - set_intr_gate(15,&spurious_interrupt_bug); - set_intr_gate(16,&coprocessor_error); - set_intr_gate(17,&alignment_check); + set_intr_gate(0, ÷_error); + set_intr_gate_ist(1, &debug, DEBUG_STACK); + set_intr_gate_ist(2, &nmi, NMI_STACK); + set_system_gate_ist(3, &int3, DEBUG_STACK); /* int3 can be called from all */ + set_system_gate(4, &overflow); /* int4 can be called from all */ + set_intr_gate(5, &bounds); + set_intr_gate(6, &invalid_op); + set_intr_gate(7, &device_not_available); + set_intr_gate_ist(8, &double_fault, DOUBLEFAULT_STACK); + set_intr_gate(9, &coprocessor_segment_overrun); + set_intr_gate(10, &invalid_TSS); + set_intr_gate(11, &segment_not_present); + set_intr_gate_ist(12, &stack_segment, STACKFAULT_STACK); + set_intr_gate(13, &general_protection); + set_intr_gate(14, &page_fault); + set_intr_gate(15, &spurious_interrupt_bug); + set_intr_gate(16, &coprocessor_error); + set_intr_gate(17, &alignment_check); #ifdef CONFIG_X86_MCE - set_intr_gate_ist(18,&machine_check, MCE_STACK); + set_intr_gate_ist(18, &machine_check, MCE_STACK); #endif - set_intr_gate(19,&simd_coprocessor_error); + set_intr_gate(19, &simd_coprocessor_error); #ifdef CONFIG_IA32_EMULATION set_system_gate(IA32_SYSCALL_VECTOR, ia32_syscall); @@ -1217,32 +1224,30 @@ void __init trap_init(void) */ init_thread_xstate(); /* - * Should be a barrier for any external CPU state. + * Should be a barrier for any external CPU state: */ cpu_init(); } - static int __init oops_setup(char *s) -{ +{ if (!s) return -EINVAL; if (!strcmp(s, "panic")) panic_on_oops = 1; return 0; -} +} early_param("oops", oops_setup); static int __init kstack_setup(char *s) { if (!s) return -EINVAL; - kstack_depth_to_print = simple_strtoul(s,NULL,0); + kstack_depth_to_print = simple_strtoul(s, NULL, 0); return 0; } early_param("kstack", kstack_setup); - static int __init code_bytes_setup(char *s) { code_bytes = simple_strtoul(s, NULL, 0); -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/