core_pattern: add CPU specifier

Statistically, in a large deployment regular segfaults may indicate a CPU
issue.

Currently, it is not possible to find out what CPU the segfault happened
on. There are at least two attempts to improve segfault logging with this
regard, but they do not help in case the logs rotate.

Hence, lets make sure it is possible to permanently record a CPU the task
ran on using a new core_pattern specifier.

Link: https://lkml.kernel.org/r/[e-mail]
Signed-off-by: Oleksandr Natalenko <[e-mail]>
Suggested-by: Renaud Métrich <[e-mail]>
Reviewed-by: Oleg Nesterov <[e-mail]>
Cc: Alexander Viro <[e-mail]>
Cc: "Eric W . Biederman" <[e-mail]>
Cc: Grzegorz Halat <[e-mail]>
Cc: "Guilherme G. Piccoli" <[e-mail]>
Cc: "Huang, Ying" <[e-mail]>
Cc: Jason A. Donenfeld <[e-mail]>
Cc: Joel Savitz <[e-mail]>
Cc: Jonathan Corbet <[e-mail]>
Cc: Kees Cook <[e-mail]>
Cc: Laurent Dufour <[e-mail]>
Cc: Luis Chamberlain <[e-mail]>
Cc: Rob Herring <[e-mail]>
Cc: Stephen Kitt <[e-mail]>
Cc: Will Deacon <[e-mail]>
Cc: Xiaoming Ni <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>

Documentation/admin-guide/sysctl/kernel.rst

···
 	%f      	executable filename
 	%E		executable path
 	%c		maximum size of core file by resource limit RLIMIT_CORE
+	%C		CPU the task ran on
 	%<OTHER>	both are dropped
 	========	==========================================
 

fs/coredump.c

···
 				err = cn_printf(cn, "%lu",
 					      rlimit(RLIMIT_CORE));
 				break;
+			/* CPU the task ran on */
+			case 'C':
+				err = cn_printf(cn, "%d", cprm->cpu);
+				break;
 			default:
 				break;
 			}
···
 		 */
 		.mm_flags = mm->flags,
 		.vma_meta = NULL,
+		.cpu = raw_smp_processor_id(),
 	};
 
 	audit_core_dumps(siginfo->si_signo);

include/linux/coredump.h

···
 	struct file *file;
 	unsigned long limit;
 	unsigned long mm_flags;
+	int cpu;
 	loff_t written;
 	loff_t pos;
 	loff_t to_skip;