arch/tile: support signal "exception-trace" hook

This change adds support for /proc/sys/debug/exception-trace to tile. Like x86 and sparc, by default it is set to "1", generating a one-line printk whenever a user process crashes. By setting it to "2", we get a much more complete userspace diagnostic at crash time, including a user-space backtrace, register dump, and memory dump around the address of the crash. Some vestiges of the Tilera-internal version of this support are removed with this patch (the show_crashinfo variable and the arch_coredump_signal function). We retain a "crashinfo" boot parameter which allows you to set the boot-time value of exception-trace. Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>

arch/tile: support signal "exception-trace" hook
This change adds support for /proc/sys/debug/exception-trace to tile. Like x86 and sparc, by default it is set to "1", generating a one-line printk whenever a user process crashes. By setting it to "2", we get a much more complete userspace diagnostic at crash time, including a user-space backtrace, register dump, and memory dump around the address of the crash. Some vestiges of the Tilera-internal version of this support are removed with this patch (the show_crashinfo variable and the arch_coredump_signal function). We retain a "crashinfo" boot parameter which allows you to set the boot-time value of exception-trace. Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
Chris Metcalf
1 parent 8aaf1dda42
Showing 8 changed files with 151 additions and 23 deletions Side-by-side Diff
arch/tile/include/asm/processor.h
arch/tile/include/asm/signal.h
arch/tile/kernel/compat_signal.c
arch/tile/kernel/signal.c
arch/tile/kernel/single_step.c
arch/tile/kernel/traps.c
arch/tile/mm/fault.c
kernel/sysctl.c
@@ -257,10 +257,6 @@
 	barrier();
 }
  
-struct siginfo;
-extern void arch_coredump_signal(struct siginfo *, struct pt_regs *);
-#define arch_coredump_signal arch_coredump_signal
-
 /* Info on this processor (see fs/proc/cpuinfo.c) */
 struct seq_operations;
 extern const struct seq_operations cpuinfo_op;
@@ -270,9 +266,6 @@
  
 /* Data on which physical memory controller corresponds to which NUMA node. */
 extern int node_controller[];
-
-/* Do we dump information to the console when a user application crashes? */
-extern int show_crashinfo;
  
 #if CHIP_HAS_CBOX_HOME_MAP()
 /* Does the heap allocator return hash-for-home pages by default? */
@@ -28,6 +28,10 @@
 int restore_sigcontext(struct pt_regs *, struct sigcontext __user *);
 int setup_sigcontext(struct sigcontext __user *, struct pt_regs *);
 void do_signal(struct pt_regs *regs);
+void signal_fault(const char *type, struct pt_regs *,
+		  void __user *frame, int sig);
+void trace_unhandled_signal(const char *type, struct pt_regs *regs,
+			    unsigned long address, int signo);
 #endif
  
 #endif /* _ASM_TILE_SIGNAL_H */
@@ -317,7 +317,7 @@
 	return 0;
  
 badframe:
-	force_sig(SIGSEGV, current);
+	signal_fault("bad sigreturn frame", regs, frame, 0);
 	return 0;
 }
  
@@ -431,7 +431,7 @@
 	return 0;
  
 give_sigsegv:
-	force_sigsegv(sig, current);
+	signal_fault("bad setup frame", regs, frame, sig);
 	return -EFAULT;
 }
@@ -39,7 +39,6 @@
  
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
  
-
 SYSCALL_DEFINE3(sigaltstack, const stack_t __user *, uss,
 		stack_t __user *, uoss, struct pt_regs *, regs)
 {
@@ -78,6 +77,13 @@
 	return err;
 }
  
+void signal_fault(const char *type, struct pt_regs *regs,
+		  void __user *frame, int sig)
+{
+	trace_unhandled_signal(type, regs, (unsigned long)frame, SIGSEGV);
+	force_sigsegv(sig, current);
+}
+
 /* The assembly shim for this function arranges to ignore the return value. */
 SYSCALL_DEFINE1(rt_sigreturn, struct pt_regs *, regs)
 {
@@ -105,7 +111,7 @@
 	return 0;
  
 badframe:
-	force_sig(SIGSEGV, current);
+	signal_fault("bad sigreturn frame", regs, frame, 0);
 	return 0;
 }
  
@@ -231,7 +237,7 @@
 	return 0;
  
 give_sigsegv:
-	force_sigsegv(sig, current);
+	signal_fault("bad setup frame", regs, frame, sig);
 	return -EFAULT;
 }
  
@@ -245,7 +251,6 @@
 {
 	int ret;
  
-
 	/* Are we from a system call? */
 	if (regs->faultnum == INT_SWINT_1) {
 		/* If so, check system call restarting.. */
@@ -362,5 +367,120 @@
 done:
 	/* Avoid double syscall restart if there are nested signals. */
 	regs->faultnum = INT_SWINT_1_SIGRETURN;
+}
+
+int show_unhandled_signals = 1;
+
+static int __init crashinfo(char *str)
+{
+	unsigned long val;
+	const char *word;
+
+	if (*str == '\0')
+		val = 2;
+	else if (*str != '=' || strict_strtoul(++str, 0, &val) != 0)
+		return 0;
+	show_unhandled_signals = val;
+	switch (show_unhandled_signals) {
+	case 0:
+		word = "No";
+		break;
+	case 1:
+		word = "One-line";
+		break;
+	default:
+		word = "Detailed";
+		break;
+	}
+	pr_info("%s crash reports will be generated on the console\n", word);
+	return 1;
+}
+__setup("crashinfo", crashinfo);
+
+static void dump_mem(void __user *address)
+{
+	void __user *addr;
+	enum { region_size = 256, bytes_per_line = 16 };
+	int i, j, k;
+	int found_readable_mem = 0;
+
+	pr_err("\n");
+	if (!access_ok(VERIFY_READ, address, 1)) {
+		pr_err("Not dumping at address 0x%lx (kernel address)\n",
+		       (unsigned long)address);
+		return;
+	}
+
+	addr = (void __user *)
+		(((unsigned long)address & -bytes_per_line) - region_size/2);
+	if (addr > address)
+		addr = NULL;
+	for (i = 0; i < region_size;
+	     addr += bytes_per_line, i += bytes_per_line) {
+		unsigned char buf[bytes_per_line];
+		char line[100];
+		if (copy_from_user(buf, addr, bytes_per_line))
+			continue;
+		if (!found_readable_mem) {
+			pr_err("Dumping memory around address 0x%lx:\n",
+			       (unsigned long)address);
+			found_readable_mem = 1;
+		}
+		j = sprintf(line, REGFMT":", (unsigned long)addr);
+		for (k = 0; k < bytes_per_line; ++k)
+			j += sprintf(&line[j], " %02x", buf[k]);
+		pr_err("%s\n", line);
+	}
+	if (!found_readable_mem)
+		pr_err("No readable memory around address 0x%lx\n",
+		       (unsigned long)address);
+}
+
+void trace_unhandled_signal(const char *type, struct pt_regs *regs,
+			    unsigned long address, int sig)
+{
+	struct task_struct *tsk = current;
+
+	if (show_unhandled_signals == 0)
+		return;
+
+	/* If the signal is handled, don't show it here. */
+	if (!is_global_init(tsk)) {
+		void __user *handler =
+			tsk->sighand->action[sig-1].sa.sa_handler;
+		if (handler != SIG_IGN && handler != SIG_DFL)
+			return;
+	}
+
+	/* Rate-limit the one-line output, not the detailed output. */
+	if (show_unhandled_signals <= 1 && !printk_ratelimit())
+		return;
+
+	printk("%s%s[%d]: %s at %lx pc "REGFMT" signal %d",
+	       task_pid_nr(tsk) > 1 ? KERN_INFO : KERN_EMERG,
+	       tsk->comm, task_pid_nr(tsk), type, address, regs->pc, sig);
+
+	print_vma_addr(KERN_CONT " in ", regs->pc);
+
+	printk(KERN_CONT "\n");
+
+	if (show_unhandled_signals > 1) {
+		switch (sig) {
+		case SIGILL:
+		case SIGFPE:
+		case SIGSEGV:
+		case SIGBUS:
+			pr_err("User crash: signal %d,"
+			       " trap %ld, address 0x%lx\n",
+			       sig, regs->faultnum, address);
+			show_regs(regs);
+			dump_mem((void __user *)address);
+			break;
+		default:
+			pr_err("User crash: signal %d, trap %ld\n",
+			       sig, regs->faultnum);
+			break;
+		}
+	}
 }
@@ -186,6 +186,8 @@
 			.si_code = SEGV_MAPERR,
 			.si_addr = addr
 		};
+		trace_unhandled_signal("segfault", regs,
+				       (unsigned long)addr, SIGSEGV);
 		force_sig_info(info.si_signo, &info, current);
 		return (tile_bundle_bits) 0;
 	}
@@ -196,6 +198,8 @@
 			.si_code = BUS_ADRALN,
 			.si_addr = addr
 		};
+		trace_unhandled_signal("unaligned trap", regs,
+				       (unsigned long)addr, SIGBUS);
 		force_sig_info(info.si_signo, &info, current);
 		return (tile_bundle_bits) 0;
 	}
@@ -308,6 +308,7 @@
 	info.si_addr = (void __user *)address;
 	if (signo == SIGILL)
 		info.si_trapno = fault_num;
+	trace_unhandled_signal("trap", regs, address, signo);
 	force_sig_info(signo, &info, current);
 }
  
@@ -43,8 +43,11 @@
  
 #include <arch/interrupts.h>
  
-static noinline void force_sig_info_fault(int si_signo, int si_code,
-	unsigned long address, int fault_num, struct task_struct *tsk)
+static noinline void force_sig_info_fault(const char *type, int si_signo,
+					  int si_code, unsigned long address,
+					  int fault_num,
+					  struct task_struct *tsk,
+					  struct pt_regs *regs)
 {
 	siginfo_t info;
  
@@ -59,6 +62,7 @@
 	info.si_code = si_code;
 	info.si_addr = (void __user *)address;
 	info.si_trapno = fault_num;
+	trace_unhandled_signal(type, regs, address, si_signo);
 	force_sig_info(si_signo, &info, tsk);
 }
  
  
@@ -71,11 +75,12 @@
 		struct pt_regs *, regs)
 {
 	if (address >= PAGE_OFFSET)
-		force_sig_info_fault(SIGSEGV, SEGV_MAPERR, address,
-				     INT_DTLB_MISS, current);
+		force_sig_info_fault("atomic segfault", SIGSEGV, SEGV_MAPERR,
+				     address, INT_DTLB_MISS, current, regs);
 	else
-		force_sig_info_fault(SIGBUS, BUS_ADRALN, address,
-				     INT_UNALIGN_DATA, current);
+		force_sig_info_fault("atomic alignment fault", SIGBUS,
+				     BUS_ADRALN, address,
+				     INT_UNALIGN_DATA, current, regs);
  
 	/*
 	 * Adjust pc to point at the actual instruction, which is unusual
@@ -471,8 +476,8 @@
 		 */
 		local_irq_enable();
  
-		force_sig_info_fault(SIGSEGV, si_code, address,
-				     fault_num, tsk);
+		force_sig_info_fault("segfault", SIGSEGV, si_code, address,
+				     fault_num, tsk, regs);
 		return 0;
 	}
  
@@ -547,7 +552,8 @@
 	if (is_kernel_mode)
 		goto no_context;
  
-	force_sig_info_fault(SIGBUS, BUS_ADRERR, address, fault_num, tsk);
+	force_sig_info_fault("bus error", SIGBUS, BUS_ADRERR, address,
+			     fault_num, tsk, regs);
 	return 0;
 }
  
@@ -1496,7 +1496,7 @@
  
 static struct ctl_table debug_table[] = {
 #if defined(CONFIG_X86) || defined(CONFIG_PPC) || defined(CONFIG_SPARC) || \
-    defined(CONFIG_S390)
+    defined(CONFIG_S390) || defined(CONFIG_TILE)
 	{
 		.procname	= "exception-trace",
 		.data		= &show_unhandled_signals,
...	...	@@ -257,10 +257,6 @@
257	257	barrier();
258	258	}
259	259
260		-struct siginfo;
261		-extern void arch_coredump_signal(struct siginfo , struct pt_regs );
262		-#define arch_coredump_signal arch_coredump_signal
263		-
264	260	/* Info on this processor (see fs/proc/cpuinfo.c) */
265	261	struct seq_operations;
266	262	extern const struct seq_operations cpuinfo_op;
...	...	@@ -270,9 +266,6 @@
270	266
271	267	/* Data on which physical memory controller corresponds to which NUMA node. */
272	268	extern int node_controller[];
273		-
274		-/* Do we dump information to the console when a user application crashes? */
275		-extern int show_crashinfo;
276	269
277	270	#if CHIP_HAS_CBOX_HOME_MAP()
278	271	/* Does the heap allocator return hash-for-home pages by default? */
...	...	@@ -28,6 +28,10 @@
28	28	int restore_sigcontext(struct pt_regs , struct sigcontext __user );
29	29	int setup_sigcontext(struct sigcontext __user , struct pt_regs );
30	30	void do_signal(struct pt_regs *regs);
	31	+void signal_fault(const char type, struct pt_regs ,
	32	+ void __user *frame, int sig);
	33	+void trace_unhandled_signal(const char type, struct pt_regs regs,
	34	+ unsigned long address, int signo);
31	35	#endif
32	36
33	37	#endif /* _ASM_TILE_SIGNAL_H */
...	...	@@ -317,7 +317,7 @@
317	317	return 0;
318	318
319	319	badframe:
320		- force_sig(SIGSEGV, current);
	320	+ signal_fault("bad sigreturn frame", regs, frame, 0);
321	321	return 0;
322	322	}
323	323
...	...	@@ -431,7 +431,7 @@
431	431	return 0;
432	432
433	433	give_sigsegv:
434		- force_sigsegv(sig, current);
	434	+ signal_fault("bad setup frame", regs, frame, sig);
435	435	return -EFAULT;
436	436	}
...	...	@@ -39,7 +39,6 @@
39	39
40	40	#define _BLOCKABLE (~(sigmask(SIGKILL) \| sigmask(SIGSTOP)))
41	41
42		-
43	42	SYSCALL_DEFINE3(sigaltstack, const stack_t __user *, uss,
44	43	stack_t __user , uoss, struct pt_regs , regs)
45	44	{
...	...	@@ -78,6 +77,13 @@
78	77	return err;
79	78	}
80	79
	80	+void signal_fault(const char type, struct pt_regs regs,
	81	+ void __user *frame, int sig)
	82	+{
	83	+ trace_unhandled_signal(type, regs, (unsigned long)frame, SIGSEGV);
	84	+ force_sigsegv(sig, current);
	85	+}
	86	+
81	87	/* The assembly shim for this function arranges to ignore the return value. */
82	88	SYSCALL_DEFINE1(rt_sigreturn, struct pt_regs *, regs)
83	89	{
...	...	@@ -105,7 +111,7 @@
105	111	return 0;
106	112
107	113	badframe:
108		- force_sig(SIGSEGV, current);
	114	+ signal_fault("bad sigreturn frame", regs, frame, 0);
109	115	return 0;
110	116	}
111	117
...	...	@@ -231,7 +237,7 @@
231	237	return 0;
232	238
233	239	give_sigsegv:
234		- force_sigsegv(sig, current);
	240	+ signal_fault("bad setup frame", regs, frame, sig);
235	241	return -EFAULT;
236	242	}
237	243
...	...	@@ -245,7 +251,6 @@
245	251	{
246	252	int ret;
247	253
248		-
249	254	/* Are we from a system call? */
250	255	if (regs->faultnum == INT_SWINT_1) {
251	256	/* If so, check system call restarting.. */
...	...	@@ -362,5 +367,120 @@
362	367	done:
363	368	/* Avoid double syscall restart if there are nested signals. */
364	369	regs->faultnum = INT_SWINT_1_SIGRETURN;
	370	+}
	371	+
	372	+int show_unhandled_signals = 1;
	373	+
	374	+static int __init crashinfo(char *str)
	375	+{
	376	+ unsigned long val;
	377	+ const char *word;
	378	+
	379	+ if (*str == '\0')
	380	+ val = 2;
	381	+ else if (*str != '=' \|\| strict_strtoul(++str, 0, &val) != 0)
	382	+ return 0;
	383	+ show_unhandled_signals = val;
	384	+ switch (show_unhandled_signals) {
	385	+ case 0:
	386	+ word = "No";
	387	+ break;
	388	+ case 1:
	389	+ word = "One-line";
	390	+ break;
	391	+ default:
	392	+ word = "Detailed";
	393	+ break;
	394	+ }
	395	+ pr_info("%s crash reports will be generated on the console\n", word);
	396	+ return 1;
	397	+}
	398	+__setup("crashinfo", crashinfo);
	399	+
	400	+static void dump_mem(void __user *address)
	401	+{
	402	+ void __user *addr;
	403	+ enum { region_size = 256, bytes_per_line = 16 };
	404	+ int i, j, k;
	405	+ int found_readable_mem = 0;
	406	+
	407	+ pr_err("\n");
	408	+ if (!access_ok(VERIFY_READ, address, 1)) {
	409	+ pr_err("Not dumping at address 0x%lx (kernel address)\n",
	410	+ (unsigned long)address);
	411	+ return;
	412	+ }
	413	+
	414	+ addr = (void __user *)
	415	+ (((unsigned long)address & -bytes_per_line) - region_size/2);
	416	+ if (addr > address)
	417	+ addr = NULL;
	418	+ for (i = 0; i < region_size;
	419	+ addr += bytes_per_line, i += bytes_per_line) {
	420	+ unsigned char buf[bytes_per_line];
	421	+ char line[100];
	422	+ if (copy_from_user(buf, addr, bytes_per_line))
	423	+ continue;
	424	+ if (!found_readable_mem) {
	425	+ pr_err("Dumping memory around address 0x%lx:\n",
	426	+ (unsigned long)address);
	427	+ found_readable_mem = 1;
	428	+ }
	429	+ j = sprintf(line, REGFMT":", (unsigned long)addr);
	430	+ for (k = 0; k < bytes_per_line; ++k)
	431	+ j += sprintf(&line[j], " %02x", buf[k]);
	432	+ pr_err("%s\n", line);
	433	+ }
	434	+ if (!found_readable_mem)
	435	+ pr_err("No readable memory around address 0x%lx\n",
	436	+ (unsigned long)address);
	437	+}
	438	+
	439	+void trace_unhandled_signal(const char type, struct pt_regs regs,
	440	+ unsigned long address, int sig)
	441	+{
	442	+ struct task_struct *tsk = current;
	443	+
	444	+ if (show_unhandled_signals == 0)
	445	+ return;
	446	+
	447	+ /* If the signal is handled, don't show it here. */
	448	+ if (!is_global_init(tsk)) {
	449	+ void __user *handler =
	450	+ tsk->sighand->action[sig-1].sa.sa_handler;
	451	+ if (handler != SIG_IGN && handler != SIG_DFL)
	452	+ return;
	453	+ }
	454	+
	455	+ /* Rate-limit the one-line output, not the detailed output. */
	456	+ if (show_unhandled_signals <= 1 && !printk_ratelimit())
	457	+ return;
	458	+
	459	+ printk("%s%s[%d]: %s at %lx pc "REGFMT" signal %d",
	460	+ task_pid_nr(tsk) > 1 ? KERN_INFO : KERN_EMERG,
	461	+ tsk->comm, task_pid_nr(tsk), type, address, regs->pc, sig);
	462	+
	463	+ print_vma_addr(KERN_CONT " in ", regs->pc);
	464	+
	465	+ printk(KERN_CONT "\n");
	466	+
	467	+ if (show_unhandled_signals > 1) {
	468	+ switch (sig) {
	469	+ case SIGILL:
	470	+ case SIGFPE:
	471	+ case SIGSEGV:
	472	+ case SIGBUS:
	473	+ pr_err("User crash: signal %d,"
	474	+ " trap %ld, address 0x%lx\n",
	475	+ sig, regs->faultnum, address);
	476	+ show_regs(regs);
	477	+ dump_mem((void __user *)address);
	478	+ break;
	479	+ default:
	480	+ pr_err("User crash: signal %d, trap %ld\n",
	481	+ sig, regs->faultnum);
	482	+ break;
	483	+ }
	484	+ }
365	485	}
...	...	@@ -186,6 +186,8 @@
186	186	.si_code = SEGV_MAPERR,
187	187	.si_addr = addr
188	188	};
	189	+ trace_unhandled_signal("segfault", regs,
	190	+ (unsigned long)addr, SIGSEGV);
189	191	force_sig_info(info.si_signo, &info, current);
190	192	return (tile_bundle_bits) 0;
191	193	}
...	...	@@ -196,6 +198,8 @@
196	198	.si_code = BUS_ADRALN,
197	199	.si_addr = addr
198	200	};
	201	+ trace_unhandled_signal("unaligned trap", regs,
	202	+ (unsigned long)addr, SIGBUS);
199	203	force_sig_info(info.si_signo, &info, current);
200	204	return (tile_bundle_bits) 0;
201	205	}
...	...	@@ -308,6 +308,7 @@
308	308	info.si_addr = (void __user *)address;
309	309	if (signo == SIGILL)
310	310	info.si_trapno = fault_num;
	311	+ trace_unhandled_signal("trap", regs, address, signo);
311	312	force_sig_info(signo, &info, current);
312	313	}
313	314
...	...	@@ -43,8 +43,11 @@
43	43
44	44	#include <arch/interrupts.h>
45	45
46		-static noinline void force_sig_info_fault(int si_signo, int si_code,
47		- unsigned long address, int fault_num, struct task_struct *tsk)
	46	+static noinline void force_sig_info_fault(const char *type, int si_signo,
	47	+ int si_code, unsigned long address,
	48	+ int fault_num,
	49	+ struct task_struct *tsk,
	50	+ struct pt_regs *regs)
48	51	{
49	52	siginfo_t info;
50	53
...	...	@@ -59,6 +62,7 @@
59	62	info.si_code = si_code;
60	63	info.si_addr = (void __user *)address;
61	64	info.si_trapno = fault_num;
	65	+ trace_unhandled_signal(type, regs, address, si_signo);
62	66	force_sig_info(si_signo, &info, tsk);
63	67	}
64	68
65	69
...	...	@@ -71,11 +75,12 @@
71	75	struct pt_regs *, regs)
72	76	{
73	77	if (address >= PAGE_OFFSET)
74		- force_sig_info_fault(SIGSEGV, SEGV_MAPERR, address,
75		- INT_DTLB_MISS, current);
	78	+ force_sig_info_fault("atomic segfault", SIGSEGV, SEGV_MAPERR,
	79	+ address, INT_DTLB_MISS, current, regs);
76	80	else
77		- force_sig_info_fault(SIGBUS, BUS_ADRALN, address,
78		- INT_UNALIGN_DATA, current);
	81	+ force_sig_info_fault("atomic alignment fault", SIGBUS,
	82	+ BUS_ADRALN, address,
	83	+ INT_UNALIGN_DATA, current, regs);
79	84
80	85	/*
81	86	* Adjust pc to point at the actual instruction, which is unusual
...	...	@@ -471,8 +476,8 @@
471	476	*/
472	477	local_irq_enable();
473	478
474		- force_sig_info_fault(SIGSEGV, si_code, address,
475		- fault_num, tsk);
	479	+ force_sig_info_fault("segfault", SIGSEGV, si_code, address,
	480	+ fault_num, tsk, regs);
476	481	return 0;
477	482	}
478	483
...	...	@@ -547,7 +552,8 @@
547	552	if (is_kernel_mode)
548	553	goto no_context;
549	554
550		- force_sig_info_fault(SIGBUS, BUS_ADRERR, address, fault_num, tsk);
	555	+ force_sig_info_fault("bus error", SIGBUS, BUS_ADRERR, address,
	556	+ fault_num, tsk, regs);
551	557	return 0;
552	558	}
553	559
...	...	@@ -1496,7 +1496,7 @@
1496	1496
1497	1497	static struct ctl_table debug_table[] = {
1498	1498	#if defined(CONFIG_X86) \|\| defined(CONFIG_PPC) \|\| defined(CONFIG_SPARC) \|\| \
1499		- defined(CONFIG_S390)
	1499	+ defined(CONFIG_S390) \|\| defined(CONFIG_TILE)
1500	1500	{
1501	1501	.procname = "exception-trace",
1502	1502	.data = &show_unhandled_signals,