Thread (25 messages) 25 messages, 6 authors, 2023-08-19

Re: [PATCH v3] tracepoint: add new `tcp:tcp_ca_event` trace event

From: Steven Rostedt <rostedt@goodmis.org>
Date: 2023-08-16 15:02:52
Also in: bpf, linux-trace-kernel, lkml

On Wed, 16 Aug 2023 14:09:06 +0800
Manjusaka [off-list ref] wrote:
quoted
+# trace include files use a completely different grammar
+		next if ($realfile =~ m{(?:include/trace/events/|/trace\.h$/)});
+
 # check multi-line statement indentation matches previous line
 		if ($perl_version_ok &&
 		    $prevline =~ /^\+([ \t]*)((?:$c90_Keywords(?:\s+if)\s*)|(?:$Declare\s*)?(?:$Ident|\(\s*\*\s*$Ident\s*\))\s*|(?:\*\s*)*$Lval\s*=\s*$Ident\s*)\(.*(\&\&|\|\||,)\s*$/) {


  
Actually, I'm not sure this is the checkpatch style issue or my code style issue.

Seems wired.
The TRACE_EVENT() macro has its own style. I need to document it, and
perhaps one day get checkpatch to understand it as well.

The TRACE_EVENT() typically looks like:


TRACE_EVENT(name,

	TP_PROTO(int arg1, struct foo *arg2, struct bar *arg3),

	TP_ARGS(arg1, arg2, arg3),

	TP_STRUCT__entry(
		__field(	int,		field1				)
		__array(	char,		mystring,	MYSTRLEN	)
		__string(	filename,	arg3->name			)
	),

	TP_fast_assign(
		__entry->field1 = arg1;
		memcpy(__entry->mystring, arg2->string);
		__assign_str(filename, arg3->name);
	),

	TP_printk("field1=%d mystring=%s filename=%s",
		__entry->field1, __entry->mystring, __get_str(filename))
);

The TP_STRUCT__entry() should be considered more of a "struct" layout than
a macro layout, and that's where checkpatch gets confused. The spacing
makes it much easier to see the fields and their types.

-- Steve
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help